- , m ^^ 
 
 ; -- .... -.,. 
 
 -'-,. . ' :,:. ;, .-. . 
 
 '*-^}^^ : && 
 
''^^^^^^'^'- 
 
 
 '.:. . 
 
 The D. Van NoSlrand Company 
 
 intend this book to be sold to the Public 
 at the advertised price, and supply 1 
 the Trade on terms which will not allow 
 of reduction. 
 
A TEXT-BOOK OF 
 
 QUANTITATIVE 
 CHEMICAL ANALYSIS 
 
 BY 
 
 GRAVIMETRIC, ELECTROLYTIC, VOLUMETRIC 
 AND GASOMETRIC METHODS 
 
 WITH SEVENTY-FOUR LABORATORY EXERCISES 
 
 GIVING 
 
 THE ANALYSIS OF PURE SALTS, ALLOYS, MINERALS 
 AND TECHNICAL PRODUCTS 
 
 BY 
 
 J. C. OLSEN, A.M., PH.D. 
 
 Professor of Analytical Chemistry in the Polytechnic Institute of Brooklyn 
 
 formerly Fellow of the Johns Hopkins University 
 Member American Institute of Chemical Engineers 
 
 FOURTH EDITION, REVISED AND ENLARGED 
 
 NEW YORK 
 
 D. VAN NOSTRAND COMPANY 
 
 25 PARK PLACE 
 1910. 
 
Engineering 
 Library 
 
 Copyright, 1908 
 
 BY 
 
 D. VAN NOSTRAiND COMPANY 
 
 CAMBLOT PRESS, l8-2O OAK STREET, NEW YORK 
 
PEEFACE. 
 
 IN writing the present book the author has endeavored in the 
 first place to produce a text-book on Quantitative Analysis which 
 shall meet his own needs in presenting the subject to his students. 
 The text-books available did not give as thorough and at the 
 same time as comprehensive a view of the subject as seemed 
 desirable. In order to present the subject from the theoretical 
 as well as from the practical standpoint, reference by the student 
 to a very considerable number of text-books and journals seemed 
 necessary. This was largely due to the fact that each author 
 has given special prominence to a particular branch of the subject, 
 such as gravimetric, electrolytic, volumetric, or gas analysis. In 
 the present text-book the endeavor has been made to accord each 
 of these subjects the relative prominence which is justified by 
 the extent to which the methods concerned are actually used. 
 Obsolete methods and new methods which have not come into 
 general use have generally been excluded. 
 
 In the arrangement and presentation of the subject-matter the 
 needs of the student rather than the experienced analyst have 
 been kept continually in view The needs of the student have 
 been taken to be the acquisition of a thorough comprehension 
 of the reasons for each step in an analysis as well as the develop- 
 ment of the skill in manipulation which is necessary in rapid and 
 accurate work. It is believed that by this method the require- 
 ments of the professional chemist will also be best served when 
 a reference book is needed. 
 
 The order in which the various general methods are taken up 
 is that used by the author in instructing his classes. The instruc- 
 tion on the manipulation of the balance and general operations 
 
 iii 
 
 258752 
 
iv PREFACE. 
 
 is first given. Then the determination of metals and acids in 
 pure salts is taken up, each student making one determination 
 under each of the general methods given. Instruction is then 
 given in the separation of the elements in connection with the 
 analysis of alloys and minerals The list of alloys and minerals 
 whose analyses are given was selected with the object of affording 
 practice in the separation and determination of the metals and 
 acids which are commonly found associated together. It was 
 thought undesirable to spend much time studying the separation 
 of elements which rarely occur together either in nature or in 
 the arts. 
 
 Although the chapters on Electrolytic and Volumetric Methods 
 follow those on the analysis of minerals and alloys, electrolytic 
 and volumetric methods have been given in these analyses for 
 the determination of individual elements when these methods 
 are decidedly better than the gravimetric methods. Such analyses 
 are most advantageously made while studying electrolytic or 
 volumetric methods. 
 
 The analyses of iron and steel, coal, water, fats, and oils are 
 given at the end of the course to afford practice in general ana- 
 lytical work and also to illustrate methods of determining elements 
 present in very small amounts. 
 
 The chapter on Stoichiometry was written to help overcome 
 the shortcoming so often met with in otherwise excellent chemists, 
 namely, the inability to make the necessary analytical calcula- 
 tions rapidly and accurately. The author realizes that only a 
 large amount of practice will give the desired facility, and there- 
 fore would urge the student as well as the instructor to pass over 
 no problem which is not thoroughly understood. Most of the 
 methods of calculation have been given through the book in the 
 exercises to which they apply 
 
 Free use has been made of the text-books published on the 
 various branches of analytical chemistry. The author has aimed 
 to acknowledge in each case the source from which methods 
 liave been obtained. 
 
 The author takes great pleasure in acknowledging the obliga- 
 tions he is under to those who have so cheerfully assisted him in 
 preparing this work. He is especially grateful to Dr. A. C. Lang- 
 
PREFACE. V 
 
 muir, who has read the entire manuscript as well as proof, and who 
 has made numerous valuable suggestions at all stages of the 
 work. The author is also under special obligation to Mr. Albert 
 Seeker, who read the entire proof with great care and calculated 
 the factors given in Table II, page 470. To Professors I. W. Fay 
 and Samuel Sheldon of the Polytechnic Institute, and to Professor 
 Chas. M. Allen of the Pratt Institute, as well as to Messrs. Geo. C. 
 Whipple, Lincoln Burrows, and Geo. M. S. Neustadt, the author 
 is under obligations for suggestions and corrections on portions 
 of the book. 
 
 The drawings for the cuts were made under the supervision 
 of Professor Constantine Hertzberg of this Institute by Messrs. 
 Walter Rapelje, C. A. Benoit, and Albert Seeker, to whom the 
 author wishes to express his obligation. Several cuts were kindly 
 loaned by Messrs. Eimer & Amend and by the Edison Manu- 
 facturing Company. 
 
 J. 0. OLSEN. 
 
 POLYTECHNIC INSTITUTE OF BROOKLYN, 
 May, 1904. 
 
PREFACE TO THIED EDITION. 
 
 Two editions of this text-book having been exhausted since 
 its publication, four years ago, the author has taken advantage of 
 the printing of the third edition, to make such corrections and 
 additions as his own experience and that of a number of its users, 
 who have been kind enough to suggest improvements, has shown 
 to be advisable. The author has endeavored in this revision to 
 adhere to the original plan of the work, namely, to describe no 
 method which has not been thoroughly tried in his own as well as 
 other laboratories and found to be satisfactory, believing that only 
 after continued use by many workers are all the conditions neces- 
 sary to the successful carrying out of a method discovered, so that 
 it can be accurately described. 
 
 The new methods which have been introduced in this edition 
 are principally the following. Penfield's method of determining 
 water has been described because it is believed that this simple 
 and accurate method should be more widely known. The precipi- 
 tation of nitric acid by means of nitron is important as the only 
 gravimetric method for determining this common acid. The rota- 
 tion of the anode or cathode in electrolytic determinations has 
 proved so advantageous that its use is fully described, as well as 
 the use of electric and water motors for this purpose. The 
 methods for analyzing water have been revised and brought into 
 better accord with the standard methods adopted by the Ameri- 
 can Public Health Association. The simple bacteria count in 
 water and the test for coli have also been described, as the prep- 
 aration of the culture media requires considerable chemical skill. 
 A description of various methods of determining specific gravity 
 as well as the determination of the calorific value of fuels has also 
 
 vii 
 
viii PREFACE. 
 
 been introduced, since chemists are so often called upon to make 
 such determinations. A number of typical chemical problems 
 have been introduced throughout the book in order to develop 
 the ability to make chemical calculations, which seems so difficult 
 for many students to acquire. 
 
 A number of less important additions have been made through- 
 out the book, among which may be mentioned : Rules for using 
 and cleaning platinum ware; volumetric methods for determining 
 antimony, copper, and testing shellac for rosin. The method of 
 determining the Polenske value, the flash point, and the viscosity 
 of fats and oils, as well as the separation of mineral from animal 
 and vegetable fats and oils. All atomic and molecular weights as 
 well as the table of factors have been recalculated by the 1908 
 table of atomic weights. The description of methods has been 
 improved wherever experience has shown the necessity. 
 
 While the author has given the precautions necessary to 
 obtain correct results with all methods described, he has endeav- 
 ored to avoid producing a book which can be followed mechanic- 
 ally with the production of good analytical results, but poor chem- 
 ists. In writing the exercises he has assumed that the student 
 has studied the text-book and knows the reason for each step in 
 an analysis. The author is of the opinion that the student can- 
 not successfully perform these exercises unless he has carefully 
 studied and understood the text. It is undoubtedly for this rea- 
 son that the book has not been found adapted for use in large 
 laboratory classes where a sufficient number of competent instruc- 
 tors has not been provided. 
 
 The author has been very much encouraged by the favorable 
 reception . accorded the book in the past and hopes for a still 
 larger field of usefulness in the future. He takes this opportunity 
 of thanking all those who have been kind enough to point out 
 errors or suggest improvements. He desires especially to express 
 his obligation to Mr. Albert Seeker for the revision of the chapter 
 on oils and fats and to Mr. Andrew Mayer, Jr. chemist with the 
 National Lead Co. for the revision of the chapters on the analysis 
 of alloys. 
 
 T. C. OLSEN. 
 
 POLYTECHNIC INSTITUTE OF BROOKLYN, 
 September 4, 1908. 
 
CONTENTS. 
 
 INTRODUCTION. 
 
 PAGE 
 
 Definition and Comparison of Gravimetric and Volumetric Methods 1 
 
 Personal Qualities Essential to the Analytical Chemist 2 
 
 Limit of Accuracy 4 
 
 Utility and Importance of Quantitative Analysis 6 
 
 CHAPTER I. 
 THE BALANCE. 
 
 Construction 7 
 
 Weighing 7 
 
 Determination of the Sensibility of a Balance (Ex. 1) 15 
 
 Determination of the Relative Length of the Arms of a Balance (Ex. 2). ... 16 
 
 Calibration of Weights (Ex. 3) 16 
 
 Weighing Substances Containing Water 19 
 
 Rules to be Observed in Using the Balance 20 
 
 CHAPTER II. 
 GENERAL OPERATIONS. 
 
 Theory of Precipitation 21 
 
 Filtration 24 
 
 Washing Precipitates 26 
 
 Preparation of Pure Salts 26 
 
 (a) By Recrystallization 27 
 
 (6) Precipitation by Change of Solvent 30 
 
 (c) Precipitation by Double Decomposition 30 
 
 Table of Compounds of the Metals and Acids which can be Prepared in Pure 
 
 Condition 31 
 
 Preparation of Pure Copper Sulphate (Ex. 4) 36 
 
 Preparation of Potassium Magnesium Sulphate (Ex. 5) 36 
 
 Preparation of Pure Sodium Chloride (Ex. 6) 37 
 
 ix 
 
x CONTENTS. 
 
 CHAPTER III. 
 DETERMINATION OF WATER. 
 
 PAGE 
 
 By Loss in Weight 39 
 
 Determination of Water of Crystallization in Copper Sulphate (Ex. 7) 41 
 
 Determination of Water of Crystallization in Barium Chloride (Ex. 8) 43 
 
 Penfield's Method of Determining Water 43 
 
 Absorption and Weighing of Water 44 
 
 Drying Properties of Fused Calcium Chloride 45 
 
 Drying Properties of Concentrated Sulphuric Acid 46 
 
 Drying Properties of Phosphorus Pentoxide 46 
 
 Determination of Water of Crystallization in Magnesium Sulphate (Ex. 9). 47 
 
 DETERMINATION OF METALS. 
 
 CHAPTER IV. 
 DETERMINATION OF METALS AS OXIDE. 
 
 General Properties of the Oxides of the Metals 50 
 
 Precipitation of Iron, Aluminium, and Chromium by Ammonia and Ammo- 
 nium Chloride 51 
 
 Contamination and Determination of Silica in the Precipitates 53 
 
 Determination of Aluminium in Potash Alum (Ex. 10) 54 
 
 Determination of Iron in Soft-Iron Wire (Ex. 11) 57 
 
 Determination of Copper, Manganese, Nickel, and Cobalt by Precipitation 
 
 with Caustic Alkali 60 
 
 Determination of Copper in Crystallized Copper Sulphate (Ex. 12) 62 
 
 Determination of Nickel in Nickel Ammonium Sulphate (Ex. 13) 62 
 
 CHAPTER V. 
 DETERMINATION OF METALS AS OXIDE. (Continued.-) 
 
 Precipitation of Lead, Bismuth, Calcium, Barium, and Strontium by Ammo- 
 nium Carbonate 64 
 
 Determination of Strontium in Strontium Carbonate (Ex. 14) 66 
 
 Precipitation of Calcium 66 
 
 Determination of Calcium as Oxalate in Calcium Carbonate (Ex. 15) 67 
 
 Determination of Zinc, Manganese, and Cadmium by Precipitation with 
 
 Sodium Carbonate 68 
 
 Determination of Zinc in Zinc Ammonium Sulphate (Ex. 16) 70 
 
 Ignition of Salts of Volatile Inorganic Acids 70 
 
 Ignition of Salts of Organic Acids 71 
 
 Determination of Lead by Ignition of Lead Nitrate (Ex. 17) 72 
 
CONTENTS. xi 
 
 CHAPTER VI. 
 DETERMINATION OF METALS AS SULPHATE AND AS SULPHIDE. 
 
 PAGE 
 
 Precipitation of Barium, Strontium, Calcium, and Lead as Sulphate 73 
 
 Determination of Barium in Barium Chloride (Ex. 18) 75 
 
 Determination of Metals by Evaporation with Sulphuric Acid 76 
 
 Determination of Magnesium in Magnesium Carbonate (Ex. 19) 77 
 
 Determination of Metals as Sulphide 78 
 
 Determination of Copper, Mercury, Lead, Cadmium, Silver, Arsenic, and 
 
 Tin by Precipitation with Hydrogen Sulphide 78 
 
 Removal of Free Sulphur and Drying at 100 79 
 
 Determination of Copper in Copper Sulphate (Ex. 20) 80 
 
 Determination of Arsenic in Arsenious Oxide (Ex. 21) 81 
 
 Precipitation of Zinc, Manganese, and Iron with Ammonium Sulphide 81 
 
 Determination of Manganese in Potassium Permanganate (Ex. 22) 83 
 
 CHAPTER VII. 
 
 DETERMINATION OF METALS AS PHOSPHATE, CHROMATE, AND 
 
 CHLORIDE. 
 
 Determination of Manganese, Zinc, and Magnesium as Phosphate 84 
 
 Removal of Ammonium Salts 85 
 
 Determination of Magnesium in Magnesium Sulphate (Ex. 23) 87 
 
 Determination of Arsenic as Magnesium Pyro-Arsenate 88 
 
 Determination of Lead, Barium, and Chromium as Chromate 89 
 
 Determination of Potassium as Plat ino-Chlo ride 89 
 
 Lindo-Gladding Method for Determining Potassium 89 
 
 Determination of Silver as Chloride 90 
 
 Determination of Mercury as Mercurous Chloride 92 
 
 Determination of Bismuth as Oxychloride 92 
 
 Determination of Silver in Silver Nitrate (Ex. 24) 93 
 
 DETERMINATION OF ACIDS. 
 
 CHAPTER VIIT. 
 DETERMINATION OF THE HALOGENS, SULPHUR, AND NITROGEN. 
 
 Determination of the Halogens 94 
 
 Separation of Iodine as Palladous Iodide from Chlorine and Bromine 96 
 
 Separation of Iodine from Chlorine as Thallous Iodide 97 
 
 Separation of Chlorine, Bromine, and Iodine by the Method of Jannasch 
 
 and Aschoff 97 
 
 Determination of Chloric Acid 100 
 
xii CONTENTS. 
 
 PAGE 
 
 Determination of Hydrocyanic Acid 100 
 
 Separation of Hydrocyanic Acid from Chlorides, Bromides, and Iodides. . . . 10Q 
 Determination of Sulphur. Oxidation of Sulphur Compounds by 
 
 (a) Fusion with Alkali Carbonates and Nitrates 101 
 
 (6) Fusion with Sodium Peroxide 102 
 
 (c) Digestion with Fuming Nitric Acid 102 
 
 (</) Digestion with Liquid Bromine, Aqua Regia, etc 104 
 
 (?) Digestion with Chlorine 104 
 
 Determination of Tellurium and Selenium 105 
 
 Determination of Nitrogen 
 
 (a) As Ammonium Chloride 105 
 
 (6) As Ammonium Platino-Chloride 106 
 
 (c) As Nitron Nitrate 106 
 
 (d) By Fusion of Nitrates with Silica or Potassium Chromate 107 
 
 CHAPTER IX. 
 DETERMINATION OF CARBONIC, BORIC, AND PHOSPHORIC ACIDS. 
 
 Determination of Carbon Dioxide 
 
 (a) By Loss after Decomposition of Carbonate with Acid 1 09 
 
 (6) By Loss after Fusion with Borax or Microcosrnic Salt Ill 
 
 (c) By Direct Weighing 112 
 
 Determination of Boric Acid by the Method of Gooch 115 
 
 Determination of Phosphoric Acid 
 
 (a) By Precipitation as Magnesium Ammonium Phosphate 117 
 
 (6) By Precipitation as Phosphomolybdate 118 
 
 1. Direct Weighing of the Phosphomolybdate 119 
 
 2. Reprecipitation as Ammonium Magnesium Phosphate 120 
 
 ANALYSIS OF ALLOYS. 
 
 CHAPTER X. 
 ALLOYS OF SILVER, COPPER, LEAD, BISMUTH, CADMIUM, AND TIN. 
 
 Errors in Complete Analyses 121 
 
 Separation of Silver from Other Metals 123 
 
 Analysis of a Silver Coin (Ex. 25) 123 
 
 Separation of Tin as Stannic Oxide 125 
 
 Separation of Lead from Other Metals 120 
 
 Analysis of Soft Solder (Ex. 26) 128 
 
 Analysis of Rose's Metals (Ex. 27) 129 
 
 Separation of Copper and Cadmium 130 
 
 Analysis of Wood's Metal (Ex. 28) 131 
 
CONTENTS. xiii 
 
 CHAPTER XI. 
 ANALYSIS OF ALLOYS CONTAINING ARSENIC, ANTIMONY, AND TIN. 
 
 PAGE 
 
 Separation of Arsenic, Antimony, and Tin 134 
 
 F. W. Clarke's Method of Separation of Arsenic from Antimony and Tin. . . 134 
 
 Other Methods of Separation of Arsenic from Antimony and Tin 135 
 
 Separation of Antimony and Tin 136 
 
 H. Rose's Method of Separation of Antimony from Arsenic and Tin 136 
 
 Classen's Electrolytic Separation of Antimony from Arsenic and Tin 137 
 
 Separation of Arsenic as Trichloride from Antimony and Tin 138 
 
 Analysis of Type-Metal (Ex. 29) 140 
 
 Analysis of Britannia Metal (Ex. 30) 142 
 
 CHAPTER XII. 
 ANALYSIS OF ALLOYS CONTAINING IRON, NICKEL, AND ZINC. 
 
 Separation of Zinc from Other Metals 145 
 
 Analysis of Brass or Bronze (Ex. 31 ) 146 
 
 Analysis of German Silver (Ex. 32) 148 
 
 Analysis of Manganese-Phosphorus-Bronze (Ex. 33) 149 
 
 ANALYSIS OF MINERALS. 
 
 CHAPTER XIII. 
 MINERALS CONTAINING IRON, ALUMINIUM, AND CHROMIUM. 
 
 Selection and Preparation of Sample, Pulverizing 152 
 
 Separation of Iron, Aluminium, and Chromium as Hydroxides 156 
 
 Separation of Aluminium and Iron as Basic Acetates 157 
 
 Separation of Aluminium and Iron as Basic Carbonates 159 
 
 Separation of Chromium as Chromate 159 
 
 Rothe's Ether Separation of Iron 160 
 
 Volumetric Separation of Iron from Aluminium 162 
 
 Separation of Iron from Aluminium as Aluminate 162 
 
 Analysis of Chromite (Ex. 34) 163 
 
 CHAPTER XIV. 
 
 ANALYSIS OF SULPHIDES CONTAINING MANGANESE, NICKEL, 
 COBALT, AND MERCURY. 
 
 Separation of Nickel and Cobalt as Sulphides 166 
 
 Electrolytic Separation of Manganese from Cobalt and Nickel 167 
 
 Separation of Cobalt as Tripotassium Cobaltic Nitrite 167 
 
xiv CONTENTS. 
 
 Separation of Cobalt and Nickel by Means of Nitroso-/?-Naphthol 168 
 
 Analysis of Smaltite (Ex. 35) 169 
 
 Analysis of Pyrite, Arsenopyrite, or Chalcopyrite (Ex. 36) 172 
 
 Analysis of Cinnabar (Ex. 37) 174 
 
 CHAPTER XV. 
 
 ANALYSIS OF CARBONATES CONTAINING CALCIUM, BARIUM, 
 STRONTIUM, AND MANGANESE. 
 
 Separation of Calcium, Barium, and Strontium from Magnesium and the 
 
 Alkali Metals 170 
 
 Separation of Calcium as Oxalate 176 
 
 Separation of Calcium, Barium, and Strontium as Sulphates 177 
 
 Separation of Barium as Chromate from Strontium and Barium 178 
 
 Separation of Calcium and Strontium as Nitrates 179 
 
 Analysis of Dolomite (Ex. 38) 180 
 
 CHAPTER XVI. 
 
 ANALYSIS OF SILICATES AND SEPARATION OF SODIUM AND 
 
 POTASSIUM. 
 
 Separation of Magnesium from the Alkalies 186 
 
 Separation of the Alkalies 187 
 
 Decomposition of Silicates 188 
 
 (a) By Fusion with Alkali Carbonates 188 
 
 (6) By the Method of J. Lawrence Smith 190 
 
 (c) By Fusion with Boric Oxide According to Jannasch 190 
 
 Analysis of Feldspar (Ex. 39) 192 
 
 ELECTROLYTIC METHODS. 
 
 CHAPTER XVII. 
 
 THE IONIC THEORY; ELECTROLYTIC APPARATUS AND 
 MANIPULATION. 
 
 The Ionic Theory of Electrolysis 197 
 
 Significance of Current Density 200 
 
 Secondary Electrolytic Reactions 200 
 
 Production and Control of the Electric Current 202 
 
 Electrolytic Apparatus 205 
 
 Electrolysis of Warm Solutions 208 
 
 Washing and Drying of Deposited Metals 208 
 
 Rotation of Electrodes . 209 
 
CONTENTS. XV 
 
 CHAPTER XVIII. 
 ELECTROLYTIC DETERMINATION OF METALS. 
 
 PAGE 
 
 Determination of Copper by Deposition from a Nitric or Sulphuric Acid 
 
 Solution 213 
 
 Separation of Copper from Other Metals 214 
 
 Decomposition of Copper Ores 215 
 
 Determination of Copper by Deposition from an Ammonium-Oxalate Solu- 
 tion 215 
 
 Determination of Copper in Copper Sulphate by the Ammonium-Oxalate 
 
 Method (Ex. 40) 215 
 
 Determination of Copper in the Refined Metal by Nitric-Acid Method 
 
 (Ex. 41) 217 
 
 Determination and Separation of Silver from Other Metals 217 
 
 Determination of Silver in Silver Nitrate (Ex. 42) 218 
 
 Analysis of a Silver Coin (Ex. 43) 218 
 
 Determination of Nickel and Cobalt 219 
 
 (a) By Ammonium-Oxalate Method 219 
 
 (6) By Ammonia Method 220 
 
 Analysis of a Nickel Coin (Ex. 44) 220 
 
 Determination of Tin by the Ammonium-Oxalate Method 221 
 
 Determination of Lead as the Peroxide 222 
 
 Decomposition of Lead Ores 223 
 
 Determination of Zinc 224 
 
 (a) By Ammonium-Oxalate Method 224 
 
 (6) By Acetic- Acid Method 224 
 
 Determination of Zinc in Ores 225 
 
 VOLUMETRIC ANALYSIS. 
 
 CHAPTER XIX. 
 CALIBRATION OF APPARATUS. 
 
 Volumetric Apparatus 227 
 
 Errors in Reading Burettes and Pipettes 228 
 
 Standard Temperatures 231 
 
 Standards of Volume 232 
 
 Calibration of Flasks and Burettes 233 
 
 (a) By Weighing Water 233 
 
 (6) By Morse and Blalock Bulbs 235 
 
 Calibration of Morse and Blalock Bulbs (Ex. 45) 242 
 
 Calibration of Flasks by Means of Morse and Blalock Bulbs (Ex. 46) 244 
 
 Calibration of Burettes by Means of Morse and Blalock Bulbs (Ex. 47). . . . 244 
 
 Problems. . 245 
 
xvi CONTENTS. 
 
 CHAPTER XX. 
 
 ACIDIMETRY. 
 
 PAGE 
 
 Standard and Normal Solutions 247 
 
 Percentage Given by Number of Cubic Centimeters of Acid Used 249 
 
 Effect of Carbon Dioxide on Indicators 250 
 
 Basicity of Acids with Various Indicators 251 
 
 Properties of Litmus, Cochineal, Methyl Orange, and Phenolphthalein 253 
 
 Strength and Nature of Standard Acid Most Desirable for General Use. . . . 255 
 
 Problem 257 
 
 CHAPTER XXI. 
 STANDARD ACIDS. 
 
 Methods of Standardizing Acids 258 
 
 Preparation and Standardization of Fifth-Normal Hydrochloric Acid (Ex. 
 
 48) 263 
 
 Determination of Sodium Hydroxide, Carbonate, and Bicarbonate 265 
 
 Analysis of Caustic Soda (Ex. 49) 267 
 
 Determination of Temporary and Permanent Hardness of Water (Ex. 50). . 269 
 Preparation and Standardization of Fifth-Normal Sulphuric Acid (Ex. 51). 271 
 
 Determination of Specific Gravity 272 
 
 Problems 277 
 
 CHAPTER XXII. 
 STANDARD ALKALIES. 
 
 Standard Alkaline Solutions 279 
 
 (a) Caustic Soda 279 
 
 (6) Alcoholic Solutions of Caustic Potash 281 
 
 (c) Barium Hydroxide 281 
 
 (d) Ammonia 282 
 
 Preparation of n/5 Caustic Soda and Determination of Ammonia in Ammo- 
 nium Chloride (Ex. 52) 282 
 
 The Kjeldahl Method of Determining Nitrogen 283 
 
 (a) Digestion 284 
 
 (6) Distillation 285 
 
 (c) Wilfarth's Modification 287 
 
 (d) Gunning's Modification 287 
 
 (e) Gunning-Jodlbauer Modification 288 
 
 (/) Forster's Modification 288 
 
 Determination of Nitrogen in Milk by the Kjeldahl-Gunning Method 
 
 (Ex. 53) 289 
 
 Determination of Nitrogen in Potassium Nitrate by the Kjeldahl-Forster 
 
 Method (Ex. 54) 291 
 
CONTENTS. 
 
 CHAPTER XXIII. 
 TITRATION OF BORIC AND CARBONIC ACIDS. 
 
 PAGE 
 
 Titration of Boric Acid 293 
 
 (a) By Thompson's Glycerine Method 293 
 
 \b) By Jones' Mannitol Method 294 
 
 Determination of Carbon Dioxide in Water 296 
 
 (a) Free 296 
 
 (6) Semi-Combined 297 
 
 (c) Total 298 
 
 Determination of Carbon Dioxide in Gases by Pettenkofer's Method 300 
 
 Determination of Carbon Dioxide in the Air (Ex. 55) 304 
 
 OXIDATION AND REDUCTION METHODS. 
 
 CHAPTER XXIV. 
 POTASSIUM PERMANGANATE AND DICHROMATE SOLUTIONS. 
 
 Normal Oxidizing Solutions 306 
 
 Indicate rs 307 
 
 Potassium- Permanganate Solution 308 
 
 Standardization 309 
 
 (a) By Iron. . , 309 
 
 (6) By Oxalic Acid and the Oxalates 313 
 
 (c) By Direct Measurement of Oxygen 313 
 
 Preparation and Standardization of n/5 Potassium-Permanganate Solution 
 
 (Ex. 56) 315 
 
 Determination of Iron in Iron Ores 318 
 
 Determination of Ferrous and Total Iron in an Iron Ore (Ex. 57) 320 
 
 Analysis of Manganese Ores 322 
 
 Determination of Manganese by Volhard's Method 323 
 
 Determination of Available Oxygen in Pyrolusite (Ex. 58) 325 
 
 Determination of Manganese in Pyrolusite by Volhard's Method (Ex. 59). . 326 
 
 Determination of Tin ; 327 
 
 Analysis of Stannous Chloride (Ex. 60) 328 
 
 Determination of Calcium 328 
 
 Potassium-Dichromate Solution 330 
 
 Standardization by Means of Iron 330 
 
 Preparation of n/5 Potassium-Dichromate Solution (Ex. 61) 332 
 
 Determination of Iron in an Iron Ore (Ex. 62) 333 
 
 Determination of Chromium hi Chrome Iron Ore (Ex. 63) 333 
 
 Problems... . 334 
 
xviii CONTENTS. 
 
 CHAPTER XXV. 
 IODOMETRIC METHODS. 
 
 PAGE 
 
 Preparation and Standardization of Iodine and Sodium Thiosulphate Solu- 
 tions 335 
 
 Preparation and Standardization of n/10 Iodine and Sodium Thiosulphate 
 
 Solutions (Ex. 64) 339 
 
 Determination of Oxidizing and Reducing Substances 342 
 
 Titration' of Acids by lodometric Methods , 343 
 
 Determination of Available Oxygen in Pyrolusite (Ex. 65) 345 
 
 Determination of Sulphur Dioxide hi Sodium Sulphite (Ex. 66) 346 
 
 Analysis of Bleaching Powder 346 
 
 Determination of Available Chlorine in Bleaching Powder (Ex. 67) 347 
 
 Problems. . . 348 
 
 PRECIPITATION METHODS. 
 
 CHAPTER XXVI. 
 DETERMINATION OF CHLORIDES, CYANIDES, AND SILVER. 
 
 Theory of Indicators 349 
 
 (a) In Acidimetry and Alkalimetry 349 
 
 (6) In Precipitation Methods 350 
 
 Titration of Chlorides 350 
 
 Titration of Cyanides 351 
 
 Standard Silver-Nitrate Solution 352 
 
 Determinations by Means of Silver Nitrate 353 
 
 Determination of Cyanogen in Potassium Cyanide (Ex. 68) 354 
 
 Determination of Silver in Alloys 355 
 
 Determination of Silver in a Coin (Ex. 69) 355 
 
 Problems 356 
 
 CHAPTER XXVII. 
 DETERMINATION OF PHOSPHORIC ACID. 
 
 Determination of Phosphoric Acid 357 
 
 (a) By the Uranium Method 357 
 
 (6) By Pemberton's Alkalimetric Titration of the Phosphomolybdate . . 360 
 (c) By the Reduction of the Molybdate Precipitate and Titration 
 
 with Permanganate Solution . 362 
 
CONTENTS. xix 
 
 TECHNICAL ANALYSIS. 
 
 CHAPTER XXVIII. 
 ANALYSIS OF IRON, STEEL, AND COAL. 
 
 PAGE 
 
 Determination of Silicon 365 
 
 Determination of Sulphur 367 
 
 (a) By Oxidation and Precipitation as Barium Sulphate 367 
 
 (6) By Evolution Methods 369 
 
 Determination of Phosphorus 372 
 
 Determination of Total Carbon 374 
 
 (a) By Oxidation with Chromic Acid 374 
 
 (6) By Oxidation after Dissolving Iron in Copper Solution 376 
 
 1 . Combustion in Oxygen 377 
 
 2. Combustion with Chromic Acid 380 
 
 (c) Determination of Carbon after Volatilization of Iron in a Stream 
 
 of Chlorine 381 
 
 Determination of Graphitic Carbon 382 
 
 Determination of Total Carbon by Eggertz's Method 383 
 
 Determination of Manganese 385 
 
 (a) By Ford's Method 385 
 
 (6) By Volhard's Method 387 
 
 (c) By Deshay's Method 388 
 
 Proximate Analysis of Coal 390 
 
 Determination of Moisture 390 
 
 Determination of Volatile Combustible Matter 391 
 
 Determination of Ash 391 
 
 Determination of Sulphur 392 
 
 Proximate Analysis of Coal (Ex. 70) 392 
 
 Determination of the Calorific value of Fuels 395 
 
 The Parr Calorimeter 397 
 
 Determination of the Calorific Value of Coal by Means of the Parr Calorim- 
 eter (Ex. 71) 403 
 
 CHAPTER XXIX. 
 WATER ANALYSIS. 
 
 Sanitary Analysis 406 
 
 Physical Examination 408 
 
 Chemical Examination 410 
 
 Determination of Free and Albuminoid Ammonia 41 1 
 
 Determination of Nitrites 416 
 
 Determination of Nitrates. 418 
 
 Determination of Oxygen Required 419 
 
 Determination of Chlorides 421 
 
 Determination of Total Solids . 422 
 
xx CONTENTS. 
 
 PAQB 
 
 Bacteriological Examination of Water 423 
 
 Bacterial Count (Ex. 72) 425 
 
 Test for B. Coli (Ex. 73) 428 
 
 Interpretation of the Results of a Sanitary Water Analysis 428 
 
 Analysis of Water for Use in Boilers 430 
 
 Calculation of Results . 433 
 
 CHAPTER XXX. 
 ANALYSIS OF OILS AND FATS. 
 
 Chemical Composition of Oils and Fats 436 
 
 Acidity .- 437 
 
 Kottstorfer Value 439 
 
 Reichert and Reichert-Meissl Value 440 
 
 The Polenske Value 441 
 
 Hehner Value 443 
 
 Iodine Value 443 
 
 Determination of Rosin in Shellac 446 
 
 Acetyl Value 447 
 
 Specific Gravity 448 
 
 Index of Refraction 449 
 
 Maumene" Number and Specific Temperature Reaction 449 
 
 Melting-Point of Fatty Acids 451 
 
 Detection of Phytosterol and Cholesterol 451 
 
 Flash Point 452 
 
 Oil Analysis 454 
 
 Table of Iodine Values 457 
 
 Table of Reichert-Meissl Values 459 
 
 Table of Polenske Values 459 
 
 Table of Kottstorfer Saponification Values 459 
 
 Table of Hehner Values 459 
 
 Table of MaumenS Number 460 
 
 Table of Specific Temperature Number 461 
 
 Table of Acetyl Values 461 
 
 CHAPTER XXXI. 
 GAS ANALYSIS. 
 
 Absorbents for Oxygen 462 
 
 Absorbents for Carbon Monoxide 464 
 
 Absorbents for Illuminants 466 
 
 Determination of Hydrogen 467 
 
 Determination of Methane 470 
 
 Hempel's Gas Apparatus 472 
 
 Analysis of Illuminating-Gas (Ex. 74) 478 
 
CONTENTS. xxi 
 
 PAGE 
 
 The Orsat Apparatus 4{ 1 
 
 Analysis of Flue Gases 4c4 
 
 Problems 485 
 
 CHAPTER XXXII. 
 STOICHIOMETRY. 
 
 Atomic Weights 486 
 
 Factors 487 
 
 Use of Logarithms 487 
 
 Indirect Gravimetric Analyses 491 
 
 Factor Weights 492 
 
 Calculation of Formula? of Salts and Minerals 493 
 
 Balancing Equations 496 
 
 Calculation of Volumetric Determinations 499 
 
 APPENDIX. 
 
 Reagents 505 
 
 Tables: 
 
 International Atomic Weights 509 
 
 Chemical Factors and Their Logarithms 510 
 
 Logarithms 514 
 
 Antilogarithms 516 
 
 Specific Gravity of Ammonia Solutions 518 
 
 " " " Hydrochloric Acid 519 
 
 " " " Nitric Acid 520 
 
 " " " Sulphuric Acid 523 
 
 Vapor Tension of Water 526 
 
 Density of Water 527 
 
QUANTITATIVE ANALYSIS. 
 
 INTRODUCTION. 
 
 1. Object of Quantitative Analysis. A quantitative analysis 
 has for its purpose the determination of the proportion in which 
 one or more constituents exist in a compound substance. While 
 in a qualitative analysis the object is to prove the presence or 
 absence of a given constituent, in the quantitative analysis the 
 amount of the constituent must be determined. Unless the 
 constituents of the substance examined are already known, the 
 qualitative must precede the quantitative analysis. The student 
 of the latter art must therefore be familiar with the former. In 
 quantitative analysis most of the methods employed may be 
 grouped in two classes designated as GRAVIMETRIC AND VOLU- 
 METRIC methods. 
 
 2. Gravimetric Methods. In a gravimetric determination the 
 substance to be determined is separated from the other material 
 present and weighed. It may be separated either as an element 
 or as part of a compound which is of definite composition, and 
 which can be obtained pure and be weighed accurately. Copper, 
 for instance, by means of an electric current, may be deposited 
 from a solution of a copper salt in a bright coherent form so that 
 it may be readily washed, dried, and weighed. It may also be 
 precipitated as sulphide, which can also be washed, dried, and 
 weighed. As the percentage of copper in copper sulphide is 
 known, the amount of copper present may be readily calculated. 
 
 3. Volumetric Methods. In a volumetric determination a solu- 
 tion is used which contains a known amount of a substance 
 
2 QUANTITATIVE ANALYSIS. 
 
 which can react in a definite manner with the constituent to be 
 determined. The volume of this so-called standard solution 
 which is just sufficient to complete the reaction is measured and 
 the weight of the cdnstituent to be determined is calculated. 
 The volumetric determination of silver may be carried out as 
 follows- 
 
 A pure chloride, such as sodium chloride, may be weighed out 
 and dissolved in a measured amount of water. This solution is 
 added in small quantities to the silver solution, until the metal 
 is completely precipitated. The volume added is measured by 
 means of appropriate instruments, and the amount of sodium 
 chloride necessary to precipitate the silver is computed. Since 
 one molecule of sodium chloride, or 58.5 parts, is necessary to pre- 
 cipitate one atom of silver, or 107.9 parts, the weight of the silver 
 may be readily calculated. 
 
 4. Relative Advantage of the Two Methods. Equally accurate 
 results in general may be obtained by either gravimetric or volu- 
 metric methods, though in the case of a given substance the best 
 method known may belong to one or the other class. Gravi- 
 metric processes and manipulations are generally simpler, while 
 volumetric determinations usually require less time, provided the 
 standard solutions are once prepared. Where many determina- 
 tions must be made volumetric methods are preferable, while 
 single determinations may usually be conducted more rapidly by 
 gravimetric methods. 
 
 5. Skill and Knowledge Required. The successful carrying out 
 of a quantitative determination by either of these general methods 
 requires a large amount of skill in manipulation as well as chem- 
 ical knowledge. This is evident from the consideration that a 
 correct quantitative analysis requires that both solutions and 
 solids be carried through numerous operations without loss. The 
 operator must also have a very complete knowledge of the chem- 
 ical reactions taking place, so that he can be certain that the 
 products finally weighed are absolutely pure and contain all of a 
 given constituent. This is rendered the more difficult as the 
 complexity of the substance analyzed increases. The precipi- 
 tates obtained from complex substances are very apt to be con- 
 taminated with some of the elements present which alone would 
 
INTRODUCTION. 3 
 
 not be precipitated. Only careful testing and laborious purifica- 
 tion of such precipitates will insure correct results. 
 
 6. Patience and Honesty Necessary. The ways in which errors 
 may be made, especially by the inexperienced worker, are so 
 numerous that unless the greatest care is taken in carrying out 
 the work the results will certainly be worthless. Persons who 
 are not conscientious enough to reject determinations in which 
 avoidable errors are known to be present, such as spilling a part 
 of a solution, should not undertake quantitative analysis. To 
 the beginner the numerous precautions seem to require almost 
 endless patience and expenditure of time. If a sufficient amount 
 of laboratory practice is taken, the worker will ultimately acquire 
 the skill which enables him to carry on rapidly and without error 
 the most difficult operations. A skilful worker may also operate 
 on many determinations at once, having learned to economize 
 his time. A reputation for producing absolutely trustworthy 
 results fully repays the analyst for the time and labor expended. 
 Absolute truthfulness and patience are therefore personal quali- 
 ties indispensable to the successful analyst. 
 
 7. Neatness. In order to be absolutely certain of one's analyti- 
 cal results a number of habits must be acquired by rigidly adher- 
 ing to fixed rules of procedure. All apparatus must be kept per- 
 fectly clean, and the working-table and surroundings in a neat 
 and orderly condition. No assurance can be given of the accuracy 
 of an analysis which has been carried out in vessels from which 
 material remaining from former work may not have been re- 
 moved. It does not help matters to say that one does not think 
 there could have been enough impurity to affect the result. Abso- 
 lute certainty on this point is required of the chemist. 
 
 8. Records. One of the most common sources of error is found 
 in the careless recording of results. Weights are frequently 
 recorded on loose pieces of paper, backs of envelopes, etc., which 
 may obviously be easily lost. Frequently the figures are accom- 
 panied with no indication of the nature of the substance weighed 
 or the analysis carried out. The only safe method consists in 
 always recording the weights or other observations in a laboratory 
 note-book kept for the purpose, and keeping this original record 
 for future reference. The page should be distinctly and fully 
 
4 QUANTITATIVE ANALYSIS. 
 
 headed with the date, the nature of the determination, and num- 
 ber of the analysis or other mark of identification if several similar 
 determinations are being carried out. The weights or other observa- 
 tions should be carefully verified when the record has been made. 
 All subsequent records or observations on this determination 
 should be made on this page. A brief outline of the method of 
 analysis should be written on the page opposite that on which 
 the weights and other figures are recorded. This outline should 
 be written before the analysis is begun. The student should 
 study the process until the reason for each step is clear. The 
 mechanical operation of carrying out an analysis according to 
 directions without understanding the process is of very little value. 
 
 9. Calculation of Results. When the accuracy of the record of 
 the work has been assured, the most common general source of 
 error is found in the calculation of the result. Many chemists 
 who are otherwise very capable find the calculation of their 
 results very difficult. The student is therefore advised to pay 
 special attention to the method of calculating the results of his 
 determinations in the exercises given in this book as well as 
 carefully solving the problems given for practice. Unless these 
 rather simple calculations are thoroughly understood, he will 
 find himself utterly unable to interpret the results in the many 
 complicated analyses the chemist is called upon to perform. It 
 need hardly be said that mere multiplication and division should 
 be carried out correctly, and yet as a matter of fact many re- 
 ported results are erroneous for this reason alone. 
 
 10. Limit of Accuracy. While the analyst must be certain of 
 the accuracy of his records and calculations, and of the cleanliness 
 of his apparatus and surroundings, he must not make the mis- 
 take of doing part of his work with a far greater degree of accu- 
 racy than the remainder. If he weighs a precipitate to the fourth 
 place of decimals and multiplies this weight by the percentage to 
 which the element sought is present, the result need not be carried 
 out to the sixth or eighth decimal place. If, for instance, a pre- 
 cipitate of barium sulphate weighed .4325 gram, on multiplying 
 this number by .5886, which represents the percentage of barium in 
 barium sulphate, the figure .25456950 is obtained. The last four 
 figures of this number are absolutely without significance and must 
 
INTRODUCTION. 5 
 
 be discarded. As the weight of the barium sulphate was taken to 
 but four places after the decimal point, the amount of barium 
 present cannot possibly be calculated to more than four places 
 and should be given as .2545 gram. Similarly, if an analysis is 
 carried out by a process or for a purpose in which an error of 
 one per cent may be present, no pains need be taken to secure 
 much greater accuracy than this in any part of the process. 
 
 That unavoidable errors exist in all analytical methods, no 
 matter how carefully carried out, is beyond dispute. The expe- 
 rience of all workers proves that it is impossible to carry through 
 an analysis twice, and obtain exactly identical results. With 
 the greatest possible care, results will differ by an amount which 
 for a given method can be made* to approach a certain minimum 
 value. For a considerable number of quantitative methods this 
 f3rror is .1%. Some duplicates may agree more closely than 
 this, while others will differ somewhat more widely. The aver- 
 age error in a series of results will be found to be- at least .1% . 
 Most quantitative methods have a tendency to give a high or a 
 low result because of impurities in the substance weighed or 
 measured, or failure to obtain all of it. The quantitative result 
 is therefore always a more or less close approximation to the 
 truth, with a general minimum difference from the true result of 
 .1%. Results are therefore generally computed to hundredths 
 of per cent, though in most cases the number in the second place 
 after the decimal point is of very little significance. It simply 
 helps to establish the correct number in the tenths place. 
 
 There is therefore very little need of making measurements 
 or weighings closer than 1 part in 1000. The ordinary analytical 
 balance gives the true weight of a substance within one or two 
 tenths of a milligram. If the amount weighed, therefore, is 200 
 milligrams, the error of weighing is about the same as in the 
 remainder of the work. If a gram is weighed out, the error in 
 weighing is 1 or 2 parts hi 10,000, which is usually a much smaller 
 amount than can be estimated in the remainder of the work. 
 The analytical chemist, therefore, need generally spend no time 
 or effort in overcoming an error which is less than .1% or which 
 in the ordinary analysis amounts to less than a few tenths of a 
 milligram. The larger the amount of the substance taken for 
 
6 QUANTITATIVE ANALYSIS. 
 
 analysis, the more accurate will be the result if due care is taken. 
 The time consumed in handling increasing amounts of chemicals 
 soon becomes excessive, so that a limit of from J to 1 gram in 
 the weight of the precipitate is found advisable in practice. If 
 a constituent is present in very small amount, a very large 
 amount of the substance to be analyzed must be taken to obtain 
 enough of the constituent to work with. 
 
 ii. Utility and Importance of Quantitative Analysis. The art 
 and science of quantitative analysis is of fundamental impor- 
 tance to most branches of chemistry. Historically, rapid strides 
 in the theory of the subject were made only after the necessity 
 was recognized of taking into account the proportions in which 
 elements combine by weight. The most fundamental concep- 
 tion of the atom is that of a portion of matter possessing a defi- 
 nite weight. The chemical formulas which are so generally used 
 and seem to express so much are primarily expressions of the 
 results of a quantitative analysis of the compounds concerned. 
 In the industrial world we find one manufacturing process after 
 another coming under the control of the analytical chemist, who 
 tests the purity of the raw materials and the finished products 
 and reduces the loss of valuable material to a minimum. Andrew 
 Carnegie, the famous iron and steel master, tells us that he was 
 the first man in America to employ a chemist for the iron and 
 steel industry. To-day every part of the process is guided by 
 the chemist's analysis of the material, and every plant is equipped 
 with a chemical laboratory. Many other industries are similarly 
 controlled by the chemist. In the commercial world an increas- 
 ing number of substances are daily bought and sold on the value 
 assigned by the chemical analysis. 
 
 In view of the fundamental importance of the subject to the 
 science and art of chemistry and to its industrial and commercial 
 applications, the student is urged to spare no pains to acquire 
 the most thorough knowledge of the subject and the greatest 
 possible skill in its application. 
 
CHAPTER I. 
 THE BALANCE. 
 
 12 Construction. The chemical balance consists essentially of 
 three parts, namely, the beam, post, and pans. An agate or 
 steel triangular prism, called a knife-edge, is securely fastened to 
 the centre of the beam, and a similar one to each end. An agate 
 or steel plane is fastened to the post, and on this plane the beam 
 rests, the contact being formed by the central knife-edge. The 
 pans are suspended from the ends of the beam by means of 
 agate or steel planes set in pieces of metal called stirrups. These 
 planes rest on the terminal knife-edges of the beam. The object 
 of this method of supporting the beam and pans is to enable the 
 system to oscillate with the least possible friction. A long, slen- 
 der pointer is attached to the centre of the beam. The oscilla- 
 tions of the beam can be observed by means of the movements of 
 the end of this pointer across the front of a scale attached to 
 the bottom of the post. In this manner very slight movements 
 may be observed. By means of a thumb-screw a second beam 
 can be raised or lowered, so as to lift the real beam of the balance 
 off its knife-edges when the balance is not in use, or lower it into 
 place when it is desired to make a weighing. Adjustable sup- 
 ports are also provided for the pans. A levelling device is attached 
 to the post, so that by means of thumb-screws under the balance 
 its position may be adjusted so as to place the post in an absolutely 
 vertical position. The instrument is enclosed in a glass case in 
 order to protect the parts from dust and mechanical injury, as 
 well as to prevent air-currents and sudden heating and cooling 
 of the parts during the weighing. 
 
 13. Weighing. The weight of a given substance is obtained by 
 placing it on one of the pans and placing weights on the other 
 pan until the pointer indicates the same position of the beam as 
 
8 
 
 QUANTITATIVE ANALYSIS. 
 
 it occupied when the pans were empty. The smallest fractional * 
 weights are obtained by moving a bent wire called a rider along 
 the beam and placing it on the various divisions of the beam 
 until an exact counterbalance is obtained. The rider must 
 
 FIG. 1. 
 
 weigh as many milligrams as the total number of divisions of 
 the beam, the last division being directly over the terminal knife- 
 edge, and the zero-mark being directly over the central knife- 
 edge. Balances are usually constructed for riders of 6 or 12 
 milligrams. As the 6 or 12 divisions of the beam are each divided 
 into 10 parts, the smallest weight indicated by the rider will be 
 the tenth of a milligram. 
 
 * All weights of a denomination less than one gram are called fractional 
 weights. 
 
THE BALANCE. 
 
 9 
 
 It will be observed that the chemical balance gives the equal- 
 ity in weight between the substance to be weighed and the weights 
 used. We assume that if the earth attracts two substances 
 equally, they are equal in mass. When a substance has been 
 weighed in a balance, it can be stated that its mass is equal to the 
 
 FIG. 2. 
 
 mass of the weights used to counterbalance it. If the force of the 
 earth's attraction should vary, it would act upon the masses on 
 the pans equally; therefore the weights found are independent 
 of the force of gravity at the time and place of weighing. 
 
 14. Sensibility. The beam, pans, weights, etc., constitute a sys- 
 tem which operates in many respects like a pendulum, of which 
 the central knife-edge is the point of support about which the 
 system oscillates. The centre of gravity of the system must be 
 below the central knife-edge, for if it were at the knife-edge the 
 system would remain at rest wherever placed; while if it were 
 above the knife-edge, the beam would be in unstable equilibrium. 
 The nearer the centre of gravity is to the central knife-edge, the more 
 sensitive the balance becomes. A small weight is fastened to the 
 pointer by means of a screw. By raising this weight the centre 
 of gravity may be brought nearer and nearer to the central knife- 
 edge. As this is done, a smaller and smaller weight will be found 
 necessary to displace the beam a given distance and move the 
 
10 QUANTITATIVE ANALYSIS. 
 
 pointer through a given angle or over a given number of divisions 
 on the scale; that is, the balance becomes more and more sensi- 
 tive. By inspection of Fig. 3 it is evident that if the centre of 
 D gravity were at A, a displacement of the oscillating 
 body through a given angle, (7, would raise the centre 
 of gravity a greater distance than if it were located 
 at some higher point, as at B. The oscillations be- 
 come slower and slower, however, so that a balance 
 should not be made more sensitive than is required for 
 the work at hand. 
 
 When the weights are placed on the pans of a 
 balance, the position of the centre of gravity is not changed if 
 the terminal knife-edges of the beam are in the same straight 
 line with the central knife-edge, because the weights act as if 
 placed on the terminal knife-edges. However, the rate of oscilla- 
 tion decreases with increasing load. If the beam bends, the 
 terminal knife-edges are lowered, and consequently the centre of 
 gravity of the system is lowered and the sensibility of the balance 
 decreased. Even if the beam does not bend, the increased fric- 
 tion with increased load decreases the sensibility. Some balance- 
 makers place the terminal knife-edges slightly above the one in 
 the centre, so that when the balance is loaded the sensibility 
 increases until the beam bends sufficiently to bring the terminal 
 into the same plane with the central knife-edge, after which the 
 sensibility decreases with increasing load. 
 
 The sensibility of a balance for a given load is obtained by 
 loading the pans equally and noting the number of divisions of 
 the scale the pointer is moved by adding a one-milligram weight 
 to one of the pans. Sensibility is defined therefore as the number 
 of divisions of the scale through which the pointer is displaced by a 
 load of one milligram. 
 
 15. Zero-point. When a weighing is to be made it is found 
 more convenient and more accurate if, instead of waiting until the 
 beam and pointer come to rest, one ascertains where they would 
 come to rest. This is done by noting the points on the scale to 
 which the pointer swings, and then calculating the point of rest. 
 When the beam with empty pans is set swinging, it will seldom 
 be found that the pointer tends to come to rest with the point at 
 
THE BALANCE. 11 
 
 the zero-point of the scale. An adjustable weight is attached 
 to the end of the beam, so that by moving it from or toward the 
 central knife-edge the pointer may come to rest very near the 
 zero-point of the scale. In practice it will be found impossible to 
 keep the balance adjusted so that the pointer shall indicate exactly 
 zero. Unequal changes in temperature and other causes occasion 
 variations in the point of rest from day to day, and also some- 
 what regular changes during each day. The point of rest is 
 brought to the centre of the scale by adjusting the weight on 
 the beam only when the pointer comes to rest more than one 
 division from the zero-point of the scale. When a weighing is 
 to be taken the balance is set swinging with empty pans, and 
 .the point of rest calculated from the movements of the pointer. 
 This point of rest is called the zero-point. The readings are 
 recorded in the following manner and the calculation made as 
 indicated, the central division of the scale being designated 10 : 
 
 5.3 14.0 
 
 5.6 13.6 
 
 6.0 
 
 Average 5.6 13.8 
 
 5.6+13.8 
 Zero-point = ~ = 9.7 
 
 16. Reducing Weights to Vacuo. While the balance gives the 
 true weight of a substance independent of variations in the force 
 of gravity, the weight of the atmosphere introduces an error 
 which is in some cases appreciable. A given substance weighs 
 less in the air than in a vacuum because it displaces a volume 
 of air equal to its own volume. The weights also are affected by 
 the buoyant force of the air, but if the specific gravity of the 
 weights and of the substance to be weighed is the same the two 
 errors counterbalance each other. If the specific gravity of the 
 weights be denoted by w, and that of the substance by w', and 
 the weight of the substance in grams by a, then 
 
 = number of c.c. of air displaced by the weights, and 
 7= number of c.c. of air displaced by the substance weighed; 
 
12 QUANTITATIVE ANALYSIS. 
 
 therefore 
 , multiplied by the weight of a c.c. of air at the time of 
 
 weighing gives the amount to be added to obtain the weight in 
 vacuo. 
 
 17. Methods of Weighing. The simplest method of weighirg 
 consists in adding or removing weights until the pointer does 
 not move when the pans are released by a steady pressure on 
 the button connected with the spring supporting the pans. This 
 method will give the true weight within three-tenths of a milligram 
 on a good balance. When the weight must be obtained more 
 accurately, the balance is set swinging and the point of rest found 
 by the method used for finding the zero-point. By adjusting the 
 weights the point of rest may be made to coincide with the zero- 
 point of the empty pans. When the highest possible accuracy 
 is desired, the weights are adjusted until the point of rest nearly 
 coincides with the zero-point of the empty pans. The point of 
 rest is accurately found by reading the swings. The weight 
 required to make the point of rest coincide with the zero-point 
 is then calculated by means of the sensibility already found. If 
 this sensibility is 2 divisions of the scale and the zero-point 10.1, 
 while the point of rest with weights on the pan is 10.7, an addi- 
 tional weight of .3 milligram ( : ~ ) would be required to bal- 
 ance the object being weighed. This method is called weighing 
 by means of the sensibility. 
 
 If the distances between the central and the terminal knife- 
 edges are not equal, the true equality between the substances in 
 the pans will not be obtained by the methods of weighing already 
 given. The substance suspended from the longer arm will have 
 the greater leverage, and must therefore be counterbalanced by 
 a larger mass in the other pan. In practice it is found that bal- 
 ance-makers frequently fail to attain the ideal of a balance with 
 arms equal within the limit of error of weighing. In ordinary 
 work this source of error may be avoided by always placing the 
 substance to be weighed on the same side. As the same error is 
 in this manner introduced into the weight of the original substance 
 and of the separated constituent, the percentage is not affected. 
 
THE BALANCE. 13 
 
 Even if constructed equal, unequal heating of the two sides of the 
 balance may change the relative lengths of the arms of the beam. 
 If the true weight of a substance is desired, it may be found by 
 counterbalancing it exactly with sand, shot, or other convenient 
 material and then replacing the unknown substance in the pan 
 by weights until the counterbalance is exactly restored. This 
 method of weighing is called the method of substitution. 
 
 Another method is to weigh a substance in one pan and then 
 exchange weights and substance and weigh again. The mean 
 of the weights obtained is taken as the weight of the substance. 
 The true weight will be equal to the square root of the product of 
 these two weights. As in practice these two weights are very 
 nearly equal, their mean differs from the square root of their prod- 
 uct by an amount far less than the experimental error. 
 
 Let R = length of the right arm, and 
 L= " " " left arm; 
 w = true weight ; 
 
 w f = apparent weight in left pan; 
 u/'= " " " right pan. 
 
 From the principle of the lever, 
 
 w'L=wR and wL=w"R, 
 
 whence 
 
 L ___ 
 
 R ~~w'~ w' 
 
 Therefore 
 
 w =\/w'w" . 
 
 If the arms are very nearly equal, w' and w" will have such 
 
 w' +w" 
 nearly equal values that one may place w = ^ 
 
 18. Testing Balance for Equality of Arms. By this method a 
 balance can be tested for equality of arms; for if they are equal, a 
 substance should weigh the same in either pan. The relative 
 lengths can be ascertained by weighing a ten-gram weight first 
 on one pan and then on the other. A small weight will have to be 
 
14 QUANTITATIVE ANALYSIS. 
 
 added or subtracted each time to produce equilibrium. Let 
 these small weights be designated as m and n. 
 
 Let A =one of the 10-gram weights; 
 
 B= the second " weight; 
 
 A be the weight which is being weighed, and when it is 
 placed in the right-hand pan its weight is found to be B+m, 
 while when it is placed in the left-hand pan its weight is equal 
 to B +n. If the weights as well as the arms of the balance are 
 equal, m and n will each equal zero. If A and B are unequal or 
 equal, and the arms of the balance unequal, m and n will be 
 unequal. By the principle of the lever as above, 
 
 AR =L(B +ra), 
 AL=R(B+n). 
 
 Solving these equations, we obtain 
 
 ? mn 
 L~ ~~2B~' 
 
 19. Errors in Weighing. The weighing of large objects is 
 attended by considerable error, because the surfaces condense air 
 and water vapor, forming a layer whose thickness varies with the 
 temperature and the degree of humidity of the atmosphere. This 
 error may be avoided by using a counterpoise, which is identical 
 hi shape and size with the substance weighed. Another source 
 of error arises from efforts which are made to clean objects very 
 thoroughly, especially glass. For this purpose a vigorous rub- 
 bing is frequently given. This may give rise to a static electrifi- 
 cation of the surface, which produces attraction or repulsion 
 between the object to be weighed and neighboring bodies. 
 Changes in weight to the extent of 6 milligrams have been pro- 
 duced in this manner. Under favorable conditions such a charge 
 may remain from 20 minutes to half an hour before being dissi- 
 pated. It is therefore advisable to allow glass vessels to stand 
 in the balance-case at least 20 minutes after a thorough cleaning 
 before weighing. 
 
THE BALANCE. 15 
 
 EXERCISE i. 
 Determination of the Sensibility of a Balance. 
 
 Ascertain the sensibility of a balance by first finding the zero-point with 
 empty pans. Then place a one-milligram weight on the right-hand pan or 
 place the rider on the one-milligram division and again find the zero-point. 
 The number of divisions on the scale between these two points is the 
 sensibility for zero-load. In taking the zero-point the balance is brought 
 into vibration by opening the door and directing a current of air against 
 one of the pans by a quick downward movement of the hand. Aftei 
 closing the door, note the distance the pointer moves to the right and to 
 the left across the scale. Read to one-tenth of the scale-divisions, taking 
 two readings on one side and three on the other. If the readings are irregu- 
 lar, more must be taken until the difference between successive readings 
 is approximately constant. Average the readings on each side and com- 
 pute the zero-point or place where the pointer would stop. The computa- 
 tion is made as follows: 
 
 7.7 13.8 
 
 7.9 13.6 7.9 + 13.7 
 
 8.2 Zero-point = 10.8 
 
 Av 7.9 13.7 
 
 After adding a one-milligram weight to the right-hand pan the zero-point 
 is determined again in the same manner. If, for instance, 7.9 was found 
 as the zero-point with one milligram on the right-hand pan, then the sensi- 
 bility will be 10.8 7.9 = 2.9 divisions. The sensibility of the balance with 
 loads of 5, 10, 20, 30, 40, 50, 70, and 90 grams is now obtained. For this 
 purpose a 5-gram weight is placed in each pan and the zero-point taken. 
 A one-milligram weight is added to the right-hand pan and the zero point 
 again taken. The number of divisions of the scale between these two points 
 is the sensibility of the balance with a load of five grams. If, when the 
 weights of equal denomination are placed on the opposite pans, the swings 
 of the pointer indicate that the zero-point is off the scale or very far from 
 its centre, a small fractional weight of suitable size must be added to one 
 pan so as to bring the zero-point near the centre of the scale. This is most 
 conveniently done by placing the lighter weight on the right-hand pan and 
 moving the rider until equality is nearly reached. The zero-point is then 
 carefully taken and the rider moved one division further to the right or to 
 the left and the zero-point taken again. The number of divisions between 
 these two points gives the sensibility with the load used. 
 
 20. Maximum Load of Balance. If the balance shows a marked decrease 
 in sensibility at any given load, the limit of its safe use has been reached 
 and no increase of the load should be made. A decrease of 50 per cent of 
 the maximum sensibility indicates overloading of the balance. The sensi- 
 bilities obtained together with the corresponding loads should be recorded 
 in the laboratory note-book in tabular form for future use. 
 
16 QUANTITATIVE ANALYSIS. 
 
 EXERCISE 2. 
 Determination of the Relative Length of the Arms of a Balance. 
 
 Determine the zero-point of the balance with empty pans. Ha\ing 
 marked one of the 10-gram weights so as to distinguish it from the other, and 
 calling this weight A , place it on the right-hand pan. Place the other 10-gram 
 weight, which is det>nated B, on the left-hand pan. If the zero-point 
 has not changed materially, locate it accurately by taking a set of readings. 
 The weight in milligrams corresponding to a change in the position of the 
 zero-point is found by dividing the number of divisions the zero-point has 
 been displaced by the sensibility at the given load. For example, if the 
 zero-point with empty pans was 11.1, and after A and B had been placed 
 on the pans it was 7.9 and the sensibility with 10 grams load is 2.9 divi- 
 sions, then I 29 ) = 1 * milligrams must be added to B to make it 
 
 weigh as much as A. This small weight is m. Interchanging the weights and 
 again taking the zero-point and calculating, we obtain another small weight 
 to be added or subtracted from B, which is designated n. For example, 
 let it be 2.1 nig. The relative length of the arm is obtained from the 
 formula 
 
 In the illustration given, 
 
 In this case the right arm is ^nger than the left, nd the true weight of a 
 substance would be obtained by multiplying its apparent weight by the 
 factor 1 00016. 
 
 EXERCISE 3. 
 Calibration of Weights. 
 
 For identification after calibration, one of the duplicate weights is in 
 each case marked by a sharp steel point. In the case of the two 100 
 milligram weights, the unmarked one is recorded as 100 mg., while the 
 marked one is indicated by 100 mg. If three weights of equal denomi- 
 nation are found in the same set, the third one is marked twice and two 
 horizontal bars over its denomination serve to distinguish it from the 
 remainder. If the arms of the balance have been found to be equal, 
 the weights may be compared by the ordinary process of weighing. If, as 
 is frequently the case, the arms are unequal, one of the methods already 
 given for obtaining the true weight of a substance must be used. The 
 method of substitution is very convenient for this purpose. A second set 
 of weights is necessary in using this method. The weights of this set are 
 placed on the left-hand pan only and are used merely as counterpoise. 
 For comparison of the 10-milligram weights the 10-milligram counterpoise 
 
THE BALANCE. 1? 
 
 is placed on the left-hand pan and one of the two 10-milligram weights 
 placed on the right-hand pan and the zero-point taken. The other 10- 
 milligram weight is then substituted for the one on the right-hand pan 
 and the zero-point again taken. If these two zero-points are identical, 
 the two weights are equal. If they are not identical, the number of divisions 
 of the scale between them is divided by the sensibility. This gives the 
 difference in weight of the weights compared, the heavier one being the 
 one which gave the zero-point farthest to the left. This relation is recorded 
 as in the following instance: 10 mg. = 10 mg.+ .1 mg. 
 
 In comparing the gram weights the differences may be greater than 
 one milligram. The rider or other fractional weights must then be used to 
 obtain equality between the counterpoise and the weights compared. 
 
 If the rider is 10 or 12 milligrams it is also compared with one of the 
 10-milligram weights. If the rider is 5 or 6 milligrams, it must be com- 
 pared with the 5-milligram weight and these two compared with one of the 
 10-milligram weights. The following comparisons are also made: 
 
 10 mg.+ 1 J mg with 20 mg. 
 
 20 " + 10mg.+ 10mg. + rider (orrider+5 mg.) " 50 " 
 
 50 " + 20mg.+ 10mg.+ 10mg.+rider " 100 " 
 
 100. " . .^ " 100 " 
 
 100 " +100mg. .. J _ 1 _. " 200 " 
 
 200 " +100 mg.+ lOO mg.+ 50 mg.+20 mg.+ lO mg.+ 
 
 lOmg. + rider ^ " 500 " 
 
 500 " +200 mg.+ lOO mg.+ lOO mg.+50 mg.+20 mg.+ 
 
 10 mg., etc " 1 gram 
 
 1 gram+ fractionals " 2 gram 
 
 2 grams with 2 grams 
 
 2 " + 2~ grams +J gram " 5 " 
 
 5 '* +2 grams +2 grams +1 gram " 10 " 
 
 10 " ._ " 10 " 
 
 10 u +10 grams. ..._._ ^_ " 20 " 
 
 20 " +10 grams+10 grams+5 grams+2 grams+2 
 
 grams+ 1 gram " 50 " 
 
 21. Calculation of the Results of the Calibration. The method of calcu- 
 lating the results is shown by the following example of the calibration 
 of a set of weights. The following comparisons were made with the results 
 indicated : 
 
 5 mg. = rider + . 1 mg. 
 
 10 " -5mg.+rider -.1 " 
 
 10 " -lOmg. ._ -.1 " 
 
 20 " =10mg.+ 10mg. . ._._ +.1 " 
 
 50 " =20mg.+ 10mg.+ 10mg.+^_mg.+rider +.1 " 
 
 100 " =50mg. + 20mg. + 10mg.+ 10mg. + 5mg.+rider +.1 " 
 
18 QUANTITATIVE ANALYSIS. 
 
 100 mg. = 100 mg. - .1 mg. 
 
 200 " = 100mg. + iOOmg + .2 " 
 
 500 " = fractional weights + . 2 " 
 
 1 gram = fractional weights . 2 " 
 
 2 grams = fractional weights+ 1 gram + . 2 " 
 
 2 " = 2" grams. .^ 
 
 5 " =2 grams +2 grams +1 gram +.5 " 
 
 10 " 5 grams +2 grams +2 grams +1 gram + 1.0 " 
 
 10 " = 10 grams. .^ 
 
 20 " = 10 grams+ To grams. . ._._ + .4 " 
 
 50 " = 20 grams+10 grams+10 grams+5 grams+2 grams 
 
 + 2 grams+ 1 gram + 2.0 " 
 
 Using the 5-milligram weight as the basis of comparison and assuming 
 for this comparison that it weighs exactly 5 milligrams, we would obtain 
 the following values for the fractional weights, where the values to the left 
 of the equality sign merely designate the weights being calibrated, while 
 the figures to the right give the comparative values of these weights in 
 terms of the 5-milligram weight: 
 
 5 mg. = .005 gram 50 mg. = .0491 gram 
 
 Rider =.0049 " 100 " =.0982 " 
 
 10mg. = .0098 " 100 " =.0982 " 
 
 10 " =.0097 " 200 " =.1963 " 
 
 20 " =.0196 " 500 " =.4910 " 
 
 Sum of fractional weights =. 9818 " 
 
 The one-gram weight is taken as the standard for the entire set of weights 
 and is assumed to be correct or equal to 1.0000 gram. As the sum of the 
 fractional weights have been found to be .0002 gram heavier than the 1-gram 
 weight, the sum of these weights should be 1.0002 grams. As by assuming 
 the 5-mg. weight to be correct, we obtained a total of .9818 gram, we must 
 increase the values assigned by .0184 gram (1.0002 -.9818 gram). Half 
 of this increase must be added to the 500-mg. weight, one-fifth to the value 
 given to the 200-mg. weight, etc. The following values will result from 
 this calculation : 
 
 500 mg. = .5002 gram (.4910+ .0092) 
 
 200 " =.2000 " (.1963+. 0037) 
 
 100 " =.1000 '' (.0982+. 0018) 
 
 100 " =.1000 " (.0982+. 0018) 
 
 50 " =.0500 " (.0491+. 0009) 
 
 20 " =.0200 " (.0196+. 0004) 
 
 10 " =.0099 " (.0097+. 0002) 
 
 10 " =.0100 " (.0098+. 0002) 
 
 5 " =.0051 " (.0050+. 0001) 
 
 Rider =.0050 " (.0049+ .0001) 
 
THE BALANCE. 19 
 
 Assuming the 1-gram weight to be correct, the following values are 
 obtained for the larger weights: 
 
 1 gram = 1 . 0000 gram 10 grams = 10 . 0031 grams 
 
 2 grams =2. 0004 grams 10 " =10.0031 " 
 2 " =2.0004 " 20 " =20.0066 " 
 5 " =5.0013 " 50 " =50.0169 " 
 
 When careful work is being done the weight of a substance is obtained 
 by using the values found by the calibration instead of those marked on 
 the weights. 
 
 22. Weighing Substances Containing Water. Most substances 
 submitted to analysis may be weighed on watch-glasses without 
 any appreciable change in weight during weighing due to loss or 
 gain of water. If any doubt exists on this point, the substance 
 after being exactly counterbalanced may be left on the scale-pan 
 for about the same length rf time as was required for weighing 
 and its weight again taken. If any change has occurred, the 
 substance must be weighed in a weighing-bottle. These bottles 
 are made of thin glass with ground-glass stoppers, and may be 
 obtained of any suitable size. The substance to be analyzed is 
 quickly transferred to the bottle after having been brought to 
 the condition, so far as moisture is concerned, in which it is desired 
 to analyze it. The bottle is then weighed and a sufficient quan- 
 tity of the contents carefully tipped out into a beaker or other 
 vessel in which to operate on the compound. The stopper is 
 quickly replaced and the bottle weighed. The difference is the 
 amount taken out for analysis. 
 
 23. Use of Desiccator. After a substance has been heated, and 
 while it is cooling, preparatory to being weighed, it should be 
 placed in a desiccator, which is a glass vessel containing a dehy- 
 drating agent, such as concentrated sulphuric acid or fused cal- 
 cium chloride. A cover is ground to fit the desiccator, and the 
 joint is made air-tight by greasing the ground surface with vasel- 
 ine. In the dry air of the desiccator most substances cool without 
 absorbing moisture. It is best to weigh a substance as soon as 
 cool, usually in ten to twenty minutes. Some substances begin 
 to absorb moisture as soon as placed on the scale-pan. Such 
 substances should be weighed as quickly as possible, reheated, 
 cooled, and weighed. In taking the second weight, however, 
 
20 QUANTITATIVE ANALYSIS. 
 
 the weights should be placed on the scale-pan before the sub- 
 stance to be weighed is taken out of the desiccator, so that as 
 soon as it is placed on the scale-pan the weight may be ascer- 
 tained by taking the swings of the pointer or by moving the rider. 
 
 24. Rules to be Observed in Using the Balance. The pans 
 should be brushed off with a earner s-hair brush before weighing. 
 
 Care should be taken not to spill any substance whatever on 
 the floor of the balance-case. It should always be kept clean. 
 
 Students should not adjust or attempt to repair a balance. 
 Report any irregularities to the instructor. 
 
 All volatile liquids and acids must be weighed in closed weigh- 
 ing-bottles. 
 
 Only metals, alloys, and pieces of minerals should be placed 
 directly on the pans of the balance. All other substances should 
 be weighed on watch-crystals or in a weighing-bottle. The 
 beam and pans must always be supported when substances which 
 weigh more than a gram are taken off or placed on the pans. 
 
 The substance to be weighed must always be placed on the 
 left-hand pan, and the weights on the right-hand pan. 
 
 Always count the weights twice first, by noting the ones 
 missing from the box; second, by noting the weights when taking 
 them off the pan. An error in noting the weight of the sub- 
 stance to be analyzed frequently involves serious loss of time. 
 
 Crucibles and substances which are to be weighed must be 
 allowed to become perfectly cold before being placed on the scale- 
 pan. A warm substance produces a current of air which by. its 
 buoyant effect apparently lessens the weight. The amount of 
 air and moisture condensed on the surface is always less in the 
 case of a warm body. Allow at least fifteen minutes for a platinum 
 crucible and precipitate to cool in a desiccator, while twenty 
 minutes should be allowed for the cooling of a porcelain crucible 
 under the same conditions. 
 
 Lower the support of the beam to start the balance swinging 
 with a slow even motion so as not to damage the delicate knife- 
 edges by suddenly bringing them against the planes. When 
 through weighing, be sure not to leave the balance swinging, and 
 remove the rider from the beam. Note all weights in a book. 
 Do not use scraps of paper. 
 
CHAPTER II. 
 GENERAL OPERATIONS. 
 
 A NUMBER of the operations of gravimetric analysis are com- 
 mon to most determinations, no matter what element is concerned 
 or employed. The element to be determined must first be PRE- 
 CIPITATED by the addition of the reagent necessary to form an 
 insoluble compound. The precipitate must be separated from 
 the solution by FILTRATION and WASHING, and finally the precipi- 
 tate must be DRIED and WEIGHED. 
 
 25. Precipitation. A careful study of the solubility of the various 
 compounds of an element must be made before the most suitable 
 form for precipitation can be selected. The most insoluble pre- 
 cipitate is generally the best form. The solubility in both hot 
 and cold water as well as in a solution of the salts present under 
 the conditions of precipitation must be studied. In many cases 
 the solubility of a precipitate is very much reduced by the pres- 
 ence of an excess of the precipitating reagent in the solution. 
 This follows from the law of ionization of a saturated solution. 
 All inorganic salts, acids, and bases dissociate in water solution 
 into parts called ions, which in the case of salts are, on the one 
 hand, the atoms of the metal, and on the other hand the acid 
 radical. Barium sulphate, for instance, breaks up into Ba and 
 S0 4 . Acids dissociate, giving the acid hydrogen atom H as one 
 ion, and the remainder of the acid as the other ion. Bases dis- 
 sociate, giving OH as one ion and the metallic atom as the other 
 ion. If a and b represent the number of ions in unit volume of a 
 solution to be precipitated, and ab the number of undissociated 
 
 molecules, then the equation r-= c, where c is a constant, ex- 
 presses the relation between these quantities. Taking barium sul- 
 phate as an example, in a saturated solution of this salt there will 
 be present some undissociated molecules, some Ba ions and some S0 4 
 
 21 
 
22 QUANTITATIVE ANALYSIS. 
 
 ions. If, now, some barium chloride molecules are added to the 
 solution, this salt will dissociate into Ba and Cl ions. The relation 
 of the number of ions and molecules expressed by the equation 
 
 T~=C no longer holds for the solution, since a or the number of 
 
 barium ions has been increased by the barium ions produced by 
 the dissociation of the barium chloride. As c is found by experi- 
 ment to remain constant, a change in the values of b and ab must 
 take place. If some of the sulphate ions (b) unite with an equal 
 number of barium ions (a), there will be produced some mole- 
 cules of barium sulphate (ab), the numerator of the fraction will 
 have been decreased, and the denominator increased. This proc- 
 ess will continue until the former value of c will have been reached 
 and equilibrium restored. As the solution was already saturated 
 with barium sulphate, the added number of molecules of this salt 
 must be precipitated. If, therefore, the salt used in precipitation is 
 quite insoluble, an excess of the reagent containing one of the ions 
 of this salt will dimmish the solubility to a negligible quantity, by 
 
 increasing the values of a or b in the general equation, jp =c - 
 
 In some cases an excess of the reagent used in precipitation is 
 to be avoided, because the precipitate redissolves in the excess of 
 the reagent. Aluminium hydroxide, for instance, redissolves to 
 an appreciable extent in an excess of ammonia, the aluminium 
 probably acting as an acid, and forming an aluminate with the 
 base. Silver chloride dissolves in an excess of sodium chloride 
 or hydrochloric acid, forming a double chloride. In such cases as 
 these, care must betaken to add only a very slight excess. 
 
 26. Contamination of the Precipitate with Soluble Salts. 
 Another difficulty met with in precipitation is the tendency of 
 many salts to carry down other salts which are themselves very 
 soluble. It is frequently impossible to wash such a precipitate 
 free from the contamination In many cases the precipitate 
 must be redissolved and reprecipitated. As the solution contain- 
 ing the bulk of the soluble salt is separated from the insoluble 
 salt by the first precipitation, during the second precipitation only 
 a small amount of the soluble salt is present, and the rather small 
 fraction of this amount which is carried down may be neglected. 
 
GENERAL OPERATIONS. 23 
 
 Various theories have been given to account for this phenomenon. 
 If the precipitant is added rapidly, particles of the solid will, no 
 doubt, surround and enclose portions of the solution. For this 
 reason it is advisable to add the precipitating reagent in a fine 
 stream, with vigorous stirring of the solution. Heating the solu- 
 tion also increases the solubility of the soluble salts. 
 
 Besides the mechanical action of a precipitate, it is also 
 undoubtedly true that more or less insoluble chemical compounds 
 are formed by the union of soluble and insoluble salts. Barium 
 sulphate has been shown to unite with ferric sulphate, forming a 
 double sulphate of barium and iron which loses S0 3 on ignition, 
 leaving the sulphate of barium colored red with ferric oxide. 
 
 27. Digestion of Precipitates. Many very insoluble precipi- 
 tates, such as barium sulphate and calcium oxalate, come down 
 in such a fine state of subdivision, that in attempting to filter 
 them, they pass through the pores of the paper. Such precipitates 
 should be allowed to stand for a considerable time in contact with 
 the mother-liquor, as the size of the particles will then increase. 
 The same object is more quickly accomplished if the solution is 
 boiling during the addition of the precipitant, and is then allowed 
 to digest at a temperature near the boiling-point. This change 
 in the size of the particles is due to the well-known fact that the 
 solubility of a finely divided substance is greater than that of 
 large particles of the same substance. If both large and small 
 particles are present in contact with the solution, the smaller 
 ones will tend to dissolve and produce a solution which is super- 
 saturated with respect to the large particles. This process is 
 quite rapid in the case of moderately soluble salts, but with the 
 insoluble compounds used in quantitative analysis, the action is 
 very slow. Heating the solution increases the solubility,* and 
 therefore accelerates the solution of the fine powder. The final 
 result of the process is that the small particles dissolve and the 
 large particles grow in size by the addition of the small ones. 
 
 As the solubility of the small crystals is greater than that of 
 the large ones, complete precipitation is not obtained unless the 
 
 * Though this is generally true, there are a few substances which are more 
 soluble in cold than in hot water. Among these are lithium and strontium sul- 
 phates and calcium hydroxide. 
 
24 QUANTITATIVE ANALYSIS. 
 
 precipitate is allowed to stand for some time in contact with the 
 solution. Digesting the precipitate hot hastens this process, as 
 it removes the small crystals. Shaking the precipitate thor- 
 oughly accomplishes the same object, as it hastens the solution 
 of the small particles by bringing them into contact with the 
 solvent and prevents supersaturation by bringing the large par- 
 ticles into contact with the solution. 
 
 28. Filtration. The separation of the precipitate from the 
 solution is accomplished by nitration. The difficulties encoun- 
 tered at this point are of two kinds: Finely divided precipitates 
 pass through the pores of the paper or other filtering medium, 
 while gelatinous solids. prevent the solution from passing through. 
 The former difficulty is overcome by digesting the precipitate 
 with the solution in a warm place. The filter-paper should first 
 be moistened with water, and if the precipitate comes through, 
 the first portions of the turbid filtrate should be repeatedly passed 
 through the filter-paper until it becomes clear. The pores of the 
 paper will then be filled with the precipitate. The filtration is 
 most rapidly accomplished if the solution is hot, boiling water 
 passing through the pores of paper about six times as fast as cold 
 water. If the precipitate is gelatinous, so that filtration is slow 
 and the precipitate does not pass through, recourse may be had 
 to filtration by suction. The most convenient way of producing 
 a vacuum is by the Bunsen filter-pump. A great many forms of 
 such pumps are on the market, varying in efficiency and durabil- 
 ity. The paper must be fitted perfectly to the funnel, and the 
 point supported by means of a perforated platinum cone or a 
 small square of linen or some similar device. The piece of linen, 
 which should be about 3 cm. square, is neatly and evenly folded 
 on the centre of the filter-paper, which is then folded again in 
 the usual manner. The paper must be carefully fitted to the 
 funnel, and pressed firmly down into the apex and then held there 
 by the index finger of the left hand while it is moistened with a little 
 distilled water from the wash-bottle. All parts of the paper must 
 be in contact with the glass. If the sides of the funnel are uneven, 
 this is impossible and another funnel must be selected. A funnel 
 whose sides curve down to the stem must always be rejected. 
 
 Very few funnels are made with the correct angle, so that the 
 
GENERAL OPERATIONS. 
 
 25 
 
 filter-paper must generally be folded with the edges uneven, as 
 
 shown at c in Fig. 4. If the paper as 
 
 folded does not fit the funnel, it must 
 
 be folded again, allowing a larger or 
 
 smaller lap. When the paper has been 
 
 placed in the funnel, the latter should 
 
 be inserted into the rubber stopper of 
 
 the filter-flask, and after filling with 
 
 water the suction applied. If the paper 
 
 does not break while the water is being 
 
 sucked through, it has been properly 
 
 fitted to the funnel and the solution to be filtered may be poured 
 
 on it. 
 
 FIG. 5. 
 
 A much simpler method of filtering by suction is by the use 
 of the GOOCH CRUCIBLE, as shown in Fig. 5. The bottom of this 
 crucible consists of a plate which is perforated with a large num- 
 
26 QUANTITATIVE ANALYSIS. 
 
 ber of small holes. A layer of asbestos in water is floated over 
 the bottom and sucked down to a compact mass, the thickness of 
 which is varied according to the fineness of the precipitate. The 
 crucible is then connected, by means of a rubber band, with a 
 funnel which is connected as usual with the filter-flask. Solu- 
 tions which attack paper may be filtered in this manner. The 
 Gooch crucible is especially well adapted for precipitates which 
 must be dried at temperatures below ignition. 
 
 The asbestos must be purified by digestion with strong aqua 
 regia for several hours and washing with water until free from 
 chlorides. The asbestos is then washed into a glass-stoppered 
 stock bottle. After the layer of asbestos has been formed in the 
 crucible, it must be washed with water until fine particles of 
 asbestos are no longer washed out. 
 
 29. Washing Precipitates. When a filtering medium has been 
 secured by means of which the precipitate can be rapidly and 
 completely separated from the solution, the next object of the 
 analyst is to secure thorough washing of the precipitate. Water 
 is the most common reagent used for this purpose, and when pos- 
 sible, it should be hot. In some cases precipitates are insoluble 
 in water only when other salts are present. Volatile salts, such 
 as ammonium chloride, are used for this purpose, as the subse- 
 quent ignition expels them. When non-volatile salts must be 
 used, they must finally be washed out with the least amount of 
 pure water. In some cases alcohol or ammonia serves as the 
 washing liquid. The most rapid and thorough way of washing 
 is to leave the precipitate in the beaker and simply decant the 
 wash-water through the paper. The precipitate may then be 
 thoroughly mixed with the successive portions of wash- water 
 and finally transferred to the paper and the washing completed. 
 This process is called washing by decantation. When the precipi- 
 tate is immediately transferred to the funnel the amount of wash- 
 water used is less, but the time required is generally greater and 
 the washing is less thorough. Stirring up the precipitate for 
 this reason with the stream of water from the wash- bottle is 
 advantageous and should be done whenever possible, and more 
 especially when the precipitate is large. To avoid spattering, 
 the stream of water is first directed against a portion of the paper 
 not covered with the precipitate. In all cases the filtrate must 
 
GENERAL OPERATIONS. 27 
 
 be tested for the constituent to be washed out, and for the final 
 test several cubic centimeters should be taken. 
 
 The filtration and washing of gelatinous precipitates is very 
 much facilitated by adding the paper pulp from 2 or 3 quanti- 
 tative filter-papers before precipitation. As the precipitate is 
 intimately mixed with the paper pulp the wash-water comes into 
 intimate contact with the precipitate. On igniting such a pre- 
 cipitate it is left in a finely powdered state and its re-solution is 
 comparatively easy. The filter-paper may be pulped by mois- 
 tening with a little concentrated hydrochloric acid. 
 
 30. Preparation of Pure Salts. Before applying a method of 
 analysis to an unknown substance it is necessary to analyze some 
 substance of known composition in order to test the method and 
 the individual's skill in using it. Owirg to personal peculiarities, 
 individual chemists sometimes utterly fail to get satisfactory 
 results by using methods which in the hands of others give uni- 
 formly satisfactory results. If, therefore, a method is first applied 
 to material of known composition, the chance of failure on an 
 unknown substance is avoided. For this purpose, pure recrys- 
 tallized and carefully dried salts are most frequently used. The 
 method of preparing a number of these salts will therefore be 
 given. 
 
 31. Recrystallization is the method most commonly used for 
 purification. By this process the impurities are gotten rid of in 
 one of two ways. If insoluble, they are filtered off; if soluble, 
 they remain in the solution when the salt which composes the 
 bulk of the substance crystallizes out. Recrystallization also 
 serves the important purpose of securing a product containing 
 the theoretical amount of water of crystallization, since even a 
 salt which is correctly called chemically pure, inasmuch as it con- 
 tains no foreign material, may contain varying amounts of water. 
 
 In most cases the so-called commercial salts, instead of the 
 more expensive chemically pure quality, may be used in the 
 preparation of the pure recrystallized salt. If the salt is in large 
 crystals, it should be coarsely powdered in a porcelain mortar. 
 It is then dissolved in hot water, so as to obtain a nearly satu- 
 rated solution. Any insoluble material must be filtered off. 
 During this process, the salt will crystallize out in the filter-paper 
 unless the funnel is kept warm. This is best accomplished by 
 
28 
 
 QUANTITATIVE ANALYSIS. 
 
 means of some form of the hot-water funnel, which is a hollow 
 metallic funnel which can be filled with wp.ter and heat applied 
 from a Bunsen burner, as shown in Fig. 6. A folded filter-paper 
 should be employed. If in spite of the hot-water funnel the salt 
 crystallizes out on the paper, the solution must be diluted. The 
 filtered solution must be stirred vigorously while the salt is crys- 
 tallizing out. This prevents the formation of large crystals or 
 aggregations of small crystals, which are very apt to enclose por- 
 tions of the mother-liquor in the interstices. As soon as the solu- 
 tion is cold the crystals should be filtered off and as much of 
 the mother-liquor as possible sucked out, while with a porcelain 
 spatula or spoon the same object is accomplished by pressing 
 the crystals into a compact mass. The remainder of the mother- 
 liquor may be washed out by ice-cold distilled water, unless the 
 
 salt is very soluble. The salt 
 should be redissolved in distilled 
 water and recrystallized. Its 
 purity is increased by each re- 
 crystallization, but ordinarily 
 two crystallizations are suffi- 
 cient; and if a C. P. salt is used 
 in the first instance one crystal- 
 lization is frequently sufficient. 
 After being freed from the 
 mother-liquor by suction, press- 
 ing, and washing, the salt is 
 removed from the funnel and 
 spread out on large sheets of 
 absorbent or filter-paper, or, still 
 better, on a porous unglazed 
 porcelain plate. The crystals are 
 pressed between the paper or 
 rubbed on the plate with a porce- 
 lain spatula until no more liquor is absorbed. It is then allowed 
 to stand in the air with occasional stirrirg to break up large 
 lumps and to expose moist crystals to the air until it is dry, 
 which can be determined by ascertaining if the crystals adhere to 
 a clean and dry glass or porcelain surlace. This test must be 
 made from time to time so that the salt shall not be exposed 
 
 FIG. 6. 
 
GENERAL OPERATIONS. 29 
 
 for too long a time to the air, as in that case water of crystalliza- 
 tion may be lost. When dry, the crystals must be transferred 
 to a clean dry glass-stoppered bottle. 
 
 32. Preparation of Double Salts. The process of recrystalliza- 
 tion will by no means secure in the case of every salt a degree 
 of purity sufficient to permit its use as a standard. Many salts 
 are not easily obtained of constant composition so far as the per- 
 centage of acid and base are concerned, while others cannot be 
 dried so as to retain a definite percentage of water. Many salts 
 which alone cannot be prepared pure can be purified by the for- 
 mation of double salts, especially of ammonium. The alums 
 belong to this class as well as many double sulphates which are 
 not alums, such as the double ammonium sulphates of zinc, nickel, 
 cobalt, ferrous iron, as well as the double chlorides of copper, 
 platinum, etc. Some of these double salts are so stable that 
 they are not decomposed by solution in water, so that they can 
 be recrystallized in the same manner as the simple salts. The 
 alums can be purified in this manner. The less stable double 
 salts decompose in water solution, so that part of the less soluble 
 constituent crystallizes out first, then crystals of the double salt 
 appear. The double sulphate of magnesium and potassium 
 behaves in this manner. This difficulty may be overcome by 
 having an excess of the more soluble constituent present. For 
 this purpose and in the preparation of any of the double salts, 
 the constituents of the double salt may be dissolved separately in 
 hot distilled water and then the solutions mixed and vigorously 
 stirred while the salt crystallizes out. Ordinarily the single salts 
 may be taken in the proportion of their molecular weights. Fer- 
 rous ammonium sulphate [FeS0 4 .(NH 4 ) 2 S0 4 .6H 2 O] may be made 
 by dissolving the constituents separately in the proportions 278 to 
 132 [FeS0 4 .7H.O=278 and (NH 4 ) 2 S0 4 = 132] and then pouring 
 the solutions together. As the iron oxidizes easily, the tempera- 
 ture of the water must not be allowed to exceed 40, and the 
 process must be carried out rapidly. If an unstable double salt, 
 like potassium magnesium sulphate (MgS0 4 .K,S0 4 .6H 2 0), must be 
 prepared, the magnesium sulphate must be present in excess to 
 the extent of J to J of its molecular weight. MgS0 4 .7H 2 having 
 a molecular weight of 246, we must use (246 +80) 326 parts of 
 this salt to 174 parts of the potassium sulphate (K 2 S0 4 =174). 
 
30 QUANTITATIVE ANALYSIS. 
 
 33. Precipitation by Change of Solvent. For various reasons 
 jecrystallization is not always applicable, and recourse to other 
 means must be made. Common salt or sodium chloride differs 
 so little in its solubility in hot and cold water that it cannot con- 
 veniently be recrystallized. Advantage is here taken of the fact 
 that this salt is much less soluble in concentrated hydrochloric 
 acid than in water. This property is based on the fact already 
 explained on p. 21, that an increase in the number of one of the 
 ions of a salt diminishes the solubility of that salt. As chlorine 
 is one of the ions of sodium chloride and also of hydrochloric 
 acid, the addition of the latter to a concentrated solution of the 
 former precipitates it in large quantities. As the impurities are 
 present in only very small quantities, the solution is very far 
 from being saturated for these salts, even when they are chlorides 
 of other metals than sodium. In order not to dilute the solu- 
 tion the hydrochloric acid is conducted into it in the gaseous form. 
 
 Sodium carbonate may be purified in a similar manner by 
 conducting carbon dioxide into a saturated solution of the sodium 
 salt. In this case, however, the precipitate is the more insoluble 
 sodium bicarbonate which is reconverted into the carbonate by 
 gentle heat. Another illustration of this method of purification is 
 found in the preparation of pure ferrous sulphate. Recrystalli- 
 zation is inadvisable in this case, because ferrous salts oxidize 
 very rapidly when their solutions are heated in contact with the 
 air. A cold saturated solution of this salt is made and crystallized 
 out by the addition of an equal volume of alcohol. As ferrous 
 sulphate is quite insoluble in 50% alcohol, a good crop of crystals 
 is obtained by stirring this mixture. The crystals are washed 
 with alcohol and dried in the usual manner. As the moist crys- 
 tals oxidize quite readily, the presence of alcohol is here advan- 
 tageous, since it facilitates drying by its volatility and by dis- 
 placing the water. 
 
 34. Precipitation by Double Decomposition. Insoluble salts 
 must be prepared by precipitation under conditions in which no 
 other substance comes down. The carbonates of the alkaline 
 earths are prepared in this manner. The chlorides and nitrates 
 of these metals can be easily obtained in the necessary state of 
 purity. If other metals are present besides the alkali metals, 
 they may be removed from the solution of the alkaline-earth 
 
PURE SALTS. 
 
 "31 
 
 COMPOUNDS OF THE METALS AND ACIDS WHICH CAN BE PRE- 
 PARED IN A PURE CONDITION. 
 
 METALS. 
 
 COMPOUNDS. 
 
 Method of 
 Purification. 
 
 Parts of the Salt Dissolved by 100 Parts 
 of Water at 
 
 
 
 
 20 
 
 
 100 
 
 Aluminium. . 
 Ammonium 
 
 A1 2 (SO 4 ) 3 .K 2 S0 4 .24H 2 O 
 NH 4 C1 
 
 R. 
 R 
 
 3.9 
 29.7 
 
 
 
 15.0 
 37.2 
 
 
 
 357.5 
 77.3 
 103.3 
 
 35.7 
 
 11.5 
 
 68.8 
 
 Antimony. . . 
 
 Arsenic 
 
 (NH 4 ) 2 S0 4 
 
 KSbOC 4 H 4 6 4H 2 O. ... 
 
 As 2 O 3 
 
 R. 
 
 R. 
 
 S 
 
 70.6 
 37 
 
 7 
 
 5.26 
 
 75.4 
 
 7.75 
 
 
 
 
 Barium 
 
 BaCl 2 .2H 2 O 
 
 R 
 
 36 2 
 
 
 41 8 
 
 
 Bismuth. 
 
 Bi 2 0,. . 
 
 T 
 
 
 
 
 
 Cadmium 
 
 CdI 2 ... 
 
 R. 
 P 
 
 86 
 
 
 92 6 
 
 
 133.3 
 
 Calcium 
 
 CaCO 3 
 
 
 
 
 Chromium. . . 
 Cobalt 
 
 Iceland spar 
 
 
 
 
 
 
 
 K 2 Cr 2 O 7 
 
 R. 
 R 
 
 5 
 25 4 
 
 
 13.1 
 
 
 102 
 
 CoS0 4 K 2 SO 4 6H 2 0. ... 
 
 
 52 4 
 
 49 
 
 108 1 
 
 Copper. . . . 
 
 CuSO 4 .5H 2 O 
 
 R. 
 
 24.2 
 
 
 36.5 
 
 
 204.5 
 
 Iron 
 
 Metal 
 
 FeS0 4 .(NH 4 ) 2 S0 4 .6H 2 O 
 Metal . . 
 
 R. 
 
 18 
 
 
 29.8 
 
 75 
 
 78.2 
 
 
 Lead . 
 
 
 
 Pb(N0 8 ) 2 
 
 R 
 
 36.5 
 
 
 52 3 
 
 
 127 
 671.2 
 
 Magnesium. . 
 
 Manganese. . . 
 Mercury 
 
 Nickel 
 
 MgSO 4 7H 2 O. . 
 
 R 
 
 76 9 
 
 
 119.7 
 
 37.2 
 25 
 
 51.3 
 
 
 K 2 S0 4 .MgS0 4 .6H 2 0. . . 
 
 MnSO 4 .(NH 4 ) 2 S0 4 .6H 2 
 HgCl 2 
 Metal 
 
 R. 
 
 R. 
 R. 
 D 
 
 20.3 
 
 
 
 75 
 
 104.6 
 
 5.73 
 
 
 
 7.39 
 
 
 53.96 
 
 NiSO 4 .(NH 4 ) 2 SO 4 .6H 2 O 
 KNO 3 
 
 R. 
 R 
 
 
 35 
 
 2.5 
 
 8.1 
 31.2 
 
 85 
 
 39.5 
 
 
 Potassium . . . 
 Silver . . . 
 
 13 3 
 
 247 
 56.6 
 940 
 
 KC1 
 AgNO,. . 
 
 R. 
 R. 
 
 28.5 
 122 
 
 
 
 34.7 
 227 
 
 
 
 Sodium. . . . 
 
 Metal 
 
 
 
 NaCl 
 
 P 
 
 35 7 
 
 
 36.0 
 
 
 39.8 
 
 Strontium. . . 
 Tin 
 
 SrCO 3 
 
 P 
 
 
 
 
 
 SnCl 4 2NH 4 C1 
 
 R 
 
 
 14-5 
 
 33 
 
 
 
 
 Zinc 
 
 ZnSO 4 .(NH 4 ) 2 S0 4 .6H 2 O 
 
 R. 
 
 10 
 
 17.3 
 
 80 
 
 67.8 
 
 
 
 
 
 In the column headed "Method of Purification," the letters have the 
 following signification: R., recrystallization ; D., distillation; S., sublimation; 
 I., ignition of nitrate, and P., precipitation. 
 
32 GENERAL OPERATIONS. 
 
 COMPOUNDS OF THE METALS AND ACIDS (Continued). 
 
 ACIDS. 
 
 COMPOUNDS. 
 
 Method of 
 Purification. 
 
 Parts of the Salt Dissolved by 100 Parts 
 of Water at 
 
 Q*- 
 
 
 20 
 
 
 100 
 
 Hydriodic. . . 
 
 I.. 
 
 s 
 
 
 
 
 
 
 Hydrobromic 
 Hydrochloric. 
 Nitric 
 
 KI 
 KH(IO 3 \. . 
 
 11. 
 
 R 
 
 127.9 
 
 15 
 
 1.53 
 
 144.2 
 
 17 
 
 5.4 
 
 64.5 
 36.0 
 
 
 
 209 
 
 KBr 
 NaCl 
 
 R. 
 P. 
 
 53.3 
 35.7 
 
 
 102 
 39.8 
 
 HNO 3 
 
 Oxalic 
 Phosphoric. . 
 
 KNO 3 
 H 2 C 2 O 4 .2H 2 
 
 (NH,),C O. . 
 
 R. 
 R. 
 
 R, 
 R 
 
 13.3 
 2.2 
 
 15 
 
 4.2 
 
 31.2 
 11.1 
 
 
 
 247 
 350 
 
 
 KH 3 (C,6 4 ) 2 .2H 2 O . 
 
 
 
 
 Na 2 HPO 4 .12H 2 0. . . 
 
 R 
 
 6.3 
 
 
 23.4 
 
 
 236.8 
 100 
 
 Sulphuric. . . . 
 
 NaNH 4 HPO 4 .4H 2 O. 
 KH 2 PO 4 
 
 R. 
 R 
 
 
 
 16 
 
 
 
 
 
 H 2 SO 4 
 
 
 
 
 
 
 
 K 2 S0 4 
 
 R. 
 
 8.5 
 
 
 
 10.9 
 
 
 
 26.2 
 
 In the column headed "Method of Purification," the letters have the 
 following signification: R., recrystallization; D., distillation; S., sublima- 
 tion; I., ignition of nitrate, and P., precipitation. 
 
 metal by passing hydrogen sulphide through the acid solution, 
 filtering off any precipitate, and then adding ammonia and again 
 filtering. To the solution thus purified, ammonium carbonate 
 solution is added with vigorous stirring. The precipitated car- 
 bonate is then thoroughly washed and dried. This method can- 
 not be used for the preparation of the pure carbonates of the 
 other metals, because these elements do not form stable carbonates 
 of definite composition. 
 
 NOTES. 
 
 Ammonium Chloride as purchased varies greatly in purity. Phosphoric 
 arsenic, sulphuric, and sulphocyanic acids may be present as well as 
 heavy metals and earths and aniline derivatives. The pure salt is best 
 prepared by neutralizing C.P. hydrochloric acid by a stream of ammonia- 
 gas produced by heating pure concentrated ammonium hydrate. When the 
 acid is neutralized the solution is made slightly acid with hydrochloric acid 
 and evaporated to dryness in a porcelain dieh and dried on the water-bath. 
 
 Ammon'um Oxalate is examined in the same manner as oxalic acid. 
 It does not lose water of crystallization as readily as the acid. 
 
PURE SALTS 33 
 
 Ammonium Sulphate may be made by the method given for ammonium 
 chloride, but must be dried at 120 and the solution must be neutral or 
 alkaline before evaporating to dryness. 
 
 Arsenious Oxide may be tested for sulphide by heating 1 gram in a 
 test-tube. The first sublimate which forms must be pure white. At the 
 end of the operation no residue must remain in the tube. 
 
 Barium Carbonate must be prepared by precipitation of the C.P. chloride 
 with ammonia and ammonium carbonate. The washed precipitate is 
 gently ignited in a platinum dish. 
 
 Barium Chloride may be tested for metals by passing hydrogen sulphide 
 through the acid solution and then rendering it alkaline with ammonia. 
 Five grams of the salt are dissolved in 200 c.c. of water, .acidified with a 
 drop or two of dilute hydrochloric acid, heated to boiling, and the barium 
 precipitated with sulphuric acid. After digesting for some time, the 
 precipitate is filtered off and the filtrate evaporated to dryness in a platinum 
 dish. The alkalies, lime, etc., will be found in the residue, which should be 
 less than .1% in the C.P. salt. 
 
 Bismuth Oxide may be prepared pure by dissolving the commercial 
 article in nitric acid, diluting largely with water, filtering and washing the 
 precipitate, which may be dried and ignited in a platinum dish. A still 
 purer product may be obtained by dissolving the basic nitrate in pure 
 nitric acid, precipitating with ammonia and ammonium carbonate, washing 
 and igniting. 
 
 Cadmium Iodide may be tested for tin, lead, copper, zinc, and other 
 metals by dissolving 2 grams in water and adding nitric acid. A white pre- 
 cipitate indicates tin. To one-half of the solution a large excess of ammonia 
 is added. The solution must remain clear and colorless. The other part 
 of the solution is diluted with water and filtered after adding excess of 
 potash solution. Pass hydrogen sulphide through the filtrate and acidify. 
 No precipitate must be produced either in the alkaline or acid solution. 
 
 Calcium and Strontium Carbonates are prepared by the method given 
 for barium carbonate. During the drying, especially of the calcium car- 
 bonate, the platinum dish must not be allowed to become even faintly red. 
 It is advisable to support the dish about an inch above a wire gauze, which 
 is heated by the Bunsen burner. A very pure form of calcium carbonate 
 may be purchased in the form of Iceland spar. 
 
 Copper prepared electrolytically and of a high percentage of purity 
 may be readily obtained. 
 
 Copper Sulphate. This salt must almost invariably be recrystallized 
 and carefully dried in order to insure the presence of the theoretical amount 
 of water. The dry crystals on exposure to the air effloresce. The loss 
 of water is indicated by the appearance of white spots in place of the clear 
 blue of the hydrated salt. 
 
 Ferrous Ammonium Sulphate, frequently called MOHR'S SALT, may be 
 purchased remarkably pure. The crystals should be small and free from 
 
34 GENERAL OPERATIONS. 
 
 ferric iron. A gram of the salt dissolved in boiled water which has been 
 allowed to cool out of contact with the air must show only a faint red with 
 thiocyanate. 
 
 Hydrochloric, Sulphuric, and Nitric Acids may be easily obtained almost 
 absolutely free from impurities. The percentage of acid present may be 
 ascertained from the tables pp. 479-486 after carefully taking the specific 
 gravity. 
 
 Iodine must always be purified by sublimation. For this purpose the 
 commercial article is mixed with one-fourth or one-fifth of its weight of 
 potassium iodide by grinding them together in a porcelain mortar. The 
 material is placed in a watch-crystal or crystallizing-dish, which is placed 
 on a piece of asbestos board having a hole cut in the centre. The iodine 
 is sublimed by gentle heat from a Bunsen burner, the first portion of the 
 vapor being allowed to escape as the material is frequently quite moist. 
 Another watch-crystal or crystallizing-dish is now placed over the first one, 
 the upper one being kept cold by means of moist filter-paper or cloth. The 
 upper dish is removed before all of the iodine is sublimed. It is best to 
 sublime the iodine immediately before use as it is very hygroscopic. 
 
 Iron. The soft iron wire, which is sold for standardizing purposes, con- 
 tains from .2 to .4% of carbon. For careful work the percentage of iron 
 or carbon must be determined gravimetrically. 
 
 The DOUBLE AMMONIUM SALTS of MANGANESE, COBALT, NICKEL, and ZINC 
 
 must be prepared by the general methods given. 
 
 Mercury. On being shaken in a clean dry bottle the surface must 
 remain perfectly bright and the metal must not adhere to the surface of 
 the glass. 
 
 Mercuric Chloride may be tested for mercurous chloride by treating a 
 gram of the finely ground sample with 15 c.c. of pure ether. It must dis- 
 solve completely. The filtrate from the hydrogen-sulphide precipitate 
 must leave no weighable residue. On testing this precipitate with dilute 
 ammonia, filtering and acidifying the filtrate with hydrochloric acid, no 
 yellow precipitate or color must be produced indicating absence of arsenic 
 
 Oxalic Acid must leave no residue on ignition. An alkaline residue 
 indicates the presence of sodium or potassium. Sulphuric acid is tested 
 for by dissolving 5 grams in 100 c.c. of water, adding a few drops of dilute 
 hydrochloric acid and barium chloride. No precipitate must form after 
 standing several hours in a hot place. Ammonia and ammonium sulphide 
 must produce no coloration. On heating 2 grams in a test-tube with caustic- 
 soda solution no ammonia must be given off. A very pure acid may be 
 made by dissolving the impure material in a mixture of equal parts of 
 alcohol and ether, filtering, evaporating off the alcohol and ether after the 
 addition of water, and recrystallizing the product. 
 
 Potash Alum. Recrystallization and carefully drying may be necessary 
 to insure the correct percentage of water. 
 
PURE SALTS. 35 
 
 Potassium-acid lodate may be purchased remarkably pure. Meineke 
 found two samples 100.001 and 100.010% pure. 
 
 Potassium Bromide and Iodide can usually be obtained quite pure. 
 The presence of bromates or iodates is shown by the blue color produced 
 by added starch solution and a little dilute hydrochloric acid to the dilute 
 solution and in the case of the bromide a little potassium iodide. The 
 heavy metals and sulphuric acid are tested for in the usual manner. The 
 salts must be dried at 100. 
 
 Potassium Bichromate is generally obtained very pure. The com- 
 mercial samples frequently contain sulphates or chlorides. The C.P. or 
 recrystallized salt may be dried by heating until the salt is fused. It should 
 not be heated higher, and all organic matter must be carefully excluded. 
 
 Potassium Nitrate is also sold in a very pure condition. Manufacturers 
 generally guarantee a maximum impurity of sodium chloride of about 1 
 part in 10,000 or .01%. 
 
 Potassium Permanganate, which is very nearly 100% pure, may be 
 readily obtained. Chlorides, sulphates, and nitrates, which are present in 
 the commercial article, are absent from the C.P. salt. The latter, however, 
 invariably contains small quantities of manganese dioxide. 
 
 Potassium Sulphate must give a clear neutral solution (1 to 20) and 
 must give no reaction with hydrogen-sulphide water, ammonium oxalate, 
 potassium carbonate, or silver nitrate. It may be dried at 100. 
 
 Potassium Tetroxalate may be prepared by making a hot saturated 
 solution of oxalic acid, neutralizing one-fourth of it with pure potassium 
 carbonate and adding this with stirring to the remainder of the solution. 
 The product is still further purified by recrystallization. 
 
 Silver Nitrate. Good commercial samples have been found to contain 
 from .01 to .03% of impurity. The crystals should be pure white. 
 
 Metallic Silver may be obtained of a high state of purity guaranteed by 
 government assay. 
 
 Sodium Phosphate and Sodium-ammonium Phosphate must be neutral 
 to phenolphthalein and free from sulphates, chlorides, and nitrates. If the 
 crystals are not transparent, but show signs of efflorescence, the salt must 
 be recrystallized and carefully dried. 
 
 Sodium and Potassium Chlorides may be purchased pure enough for most 
 quantitative work. The method of purification of sodium chloride given in 
 Experiment 6 yields a very pure product. 
 
 Zinc and Magnesium Sulphates must always be recrystallized and care- 
 fully dried. The double sulphate of magnesium and potassium is pre- 
 pared as directed in Exercise V. 
 
 Many of the tests given in this section have been quoted from "Testing 
 of Chemical Reagents," by Dr. C. Krauch. The student is referred to this 
 book or to " Chemical Reagents, their Purity and Tests," by E. Merck, for the 
 testing of other salts. 
 
36 QUANTITATIVE ANALYSIS. 
 
 EXERCISE 4. 
 Preparation of Pure Copper Sulphate, CuS0 4 .sH 2 0. 
 
 Weigh out about 150 grams of crystallized copper sulphate. Crush the 
 large crystals in a porcelain mortar. Dissolve in about 150 c.c. of hot distilled 
 water. If the solution is not perfectly clear, filter, using a folded filter-paper 
 and a hot-water funnel. The clear filtrate must be vigorously stirred as soon 
 as crystals begin to separate. Cool the solution by placing the beaker in cold 
 tap-water or in ice-water. Continue the stirring until no more crystals sepa- 
 rate out. Filter off the crystals immediately, using a funnel of suitable size 
 with a platinum cone or a perforated porcelain dish in the bottom of the 
 funnel. Connect the funnel by means of a rubber stopper to a filter-flask 
 and remove as much of the mother-liquor as possible by means of strong 
 suction from the filter-pump. Press the crystals into a compact mass by 
 means of a spatula. Pour about 50 c.c. of ice-water over the crystals and 
 suck it through as rapidly as possible with the suction. When no more 
 liquid comes through, spread out the crystals by means of a spatula on an 
 unglazed porcelain plate or on several folds of filter-paper. When no 
 more liquid seems to be absorbed , spread the crystals out in an even layer. 
 Cover them with a piece of filter-paper and allow them to dry in the air for 
 several hours with occasional stirring. Test for moisture from time to time 
 by placing a few crystals on a clean and dry porcelain or glass surface. Dry 
 crystals will not adhere to such a surface. As soon as dry, transfer to a 
 clean dry bottle with ground-glass stopper. Care must be taken in removing 
 from the filter-paper not to detach shreds of filter-paper with the crystals. 
 If the crystals have been exposed too long to the air, portions will appear 
 white from loss of water of crystallization. If the work has been carried on 
 in a room containing hydrogen sulphide or ammonium sulphide in the atmos- 
 phere, black spots will appear on the crystals. If such crystals cannot be 
 removed the work must be repeated. 
 
 EXERCISE 5. 
 Preparation of Potassium-magnesium Sulphate, K 2 S0 4 .MgS0 4 .6H 2 0. 
 
 Weigh out about 163 grams of magnesium sulphate and dissolve in 115 c.c. 
 of water. Dissolve 87 grams of potassium sulphate in 350 c.c. of water. 
 If necessary, filter the solutions separately, using the hot-water funnel and 
 folded filter-papers. Pour the hot solution of potassium sulphate rapidly 
 and with vigorous stirring into the hot magnesium solution. Stir con- 
 stantly while the double salt crystallizes out, and finally cool the solution by 
 surrounding the beaker with cracked ice. When no more salt crystallizes out, 
 filter the crystals off in a funnel with a platinum cone or perforated porce- 
 lain plate in the bottom. Suck the mother-liquor off by means of the 
 filter-pump, pressing the crystals together with a porcelain spatula. Wash 
 the crystals with 50 c.c. of ice-cold distilled water, sucking them ory with the 
 
GENERAL OPERATIONS. 
 
 37 
 
 pump. Transfer the crystals to an unglazed porcelain plate or several folds 
 of filter-paper. Remove the mother-liquor by pressing the crystals on the 
 porcelain plate with a porcelain spatula or by pressing them with the hand 
 between the folds of the filter-paper. Finally, spread them out to dry in 
 the air, testing them for moisture from time to time by placing them on a 
 clean and dry porcelain or glass surface. The dry crystals will not adhere 
 to such a surface. When dry, transfer to a clean and dry bottle. 
 
 EXERCISE 6. 
 Preparation of Pure Sodium Chloride. 
 
 About 70 grams of common salt are weighed out and dissolved in 200 c.c. 
 of distilled water. If necessary, the solution is filtered. Gaseous hydrochloric 
 acid is now passed into the solution. It is generated in the flask A, which 
 
 FIG. 7. 
 
 contains about 150 grams of common salt. 300 grams of concentrated sul- 
 phuric acid are poured into 220 c.c. of distilled water and allowed to cool 
 somewhat. This acid is transferred to the dropping-funnel and allowed to 
 drop slowly on the salt. B is a wash-bottle containing concentrated hydro- 
 chloric acid to cat2h any sodium sulphate which might spatter over. The 
 
38 QUANTITATIVE ANALYSIS. 
 
 thistle-tube C touches the salt solution in the beaker. Toward the end of 
 the operation the flask may be heated so as to expel the hydrochloric acid. 
 The stream of hydrochloric acid is allowed to pass until no more precipita- 
 tion is noticed in the salt solution. The crystals are filtered off and washed 
 with concentrated c.p. hydrochloric acid. When no more liquid can be 
 sucked out by the pump, the sodium chloride is transferred to a porcelain 
 dish and heated with the Bunsen burner until dry and free from hydro- 
 chloric acid. The pure salt is then transferred to a clean and dry glass- 
 stoppered bottle. 
 
 RULES FOR USING AND CLEANING PLATINUM UTENSILS. 
 
 Platinum when hot easily forms alloys with other metals. 
 For this reason metals must never be heated in platinum. The 
 same rule applies to all compounds of easily reducible metals, 
 such as silver, copper, lead, and bismuth. Even stannic oxide is 
 sometimes reduced to metallic tin, which alloys with the platinum. 
 
 Heat platinum only with a clear blue flame. A deposit of 
 carbon is injurious. The vessel should also be placed beyond the 
 inner cone of the flame, as reducing gases may pass through the 
 hot platinum. 
 
 As phosphorus and sulphur combine with platinum, making 
 it brittle, compounds of these elements should not be heated in 
 platinum under reducing conditions. 
 
 Caustic alkalies should not be fused in platinum as it is oxi- 
 dized and dissolved. Nickel, iron, silver, or copper dishes may 
 be used for this purpose. 
 
 The carbonates of the alkalies may be fused in platinum, 
 although a little platinum is dissolved, especially when potassium 
 nitrate is present. 
 
 Platinum vessels may be cleaned by digestion with hydro- 
 chloric or nitric acids, but not with a mixture of these acids. 
 Basic impurities not removed in this manner maybe dissolved by 
 fused acid potassium sulphate, while acid impurities such as silica 
 may be removed by fused sodium carbonate. The surface of 
 platinum should be kept smooth and bright by polishing with 
 sea sand which is free from sharp particles. 
 
 Platinum vessels change weight very slowly, losing a few tenths 
 of a milligram after strong ignition, a little more after a fusion, 
 especially with an alkaline oxidizing mixture, and still more on 
 being polished. 
 
CHAPTER III. 
 
 DETERMINATION OF WATER. 
 
 35. Conditions in which Water is Held. Water is present in 
 the great majority of substances submitted to chemical analysis, 
 and the amount present must frequently be determined. It may 
 be held MECHANICALLY or combined CHEMICALLY as water of 
 crystallization, or it may be an inherent part of the molecule. In 
 considering its determination, the manner of separating it from 
 the other constituents present must be studied as well as the 
 method of weighing it. The simplest method of determining it 
 involves heating the compound to the temperature at which the 
 water volatilizes. This temperature differs greatly, ranging from 
 the ordinary temperature of the air to that of the blast-lamp. 
 Where the water is held mechanically, as in most minerals, as 
 well as metals and non-crystalline substances, a temperature of 
 100 or a few degrees above that point is sufficient to dry the 
 substance completely. In the case of salts which crystallize 
 with water, the temperature at which it is completely given off 
 varies from 100 centigrade to a red heat. In some cases a definite 
 number of molecules of water of crystallization will be driven 
 off at one temperature, while a higher temperature is required 
 to completely dehydrate the substance. The silicates, as a class, 
 hold water with great persistence. Glass ma}^ be heated to a 
 red heat for months, and still give off a trace of water. In this 
 case the water is probably combined with the silica, forming a 
 more or less complicated silicic acid. 
 
 36. Methods of Determining Water. Whenever a substance 
 can be dried completely by simply heating it to a temperature at 
 which no other constituent is volatilized, the amount of water 
 may be determined by weighing the substance before and after 
 drying it. In this case the amount of water present may also 
 
40 QUANTITATIVE ANALYSIS. 
 
 be determined by allowing the water to be absorbed by a weighed 
 amount of some hygroscopic substance, like calcium chloride or 
 concentrated sulphuric acid. Such a determination would give 
 the total amount of water present. This water may be present 
 either as HYGROSCOPIC water or as water of CRYSTALLIZATION or 
 
 Of CONSTITUTION. 
 
 If it is necessary to determine the amount of each, the sub- 
 stance must first be heated to a temperature at which only the 
 hygroscopic water is given off. At a temperature of 105 the 
 hygroscopic water will certainly be expelled in every case, but 
 as at that temperature many crystals lose a part or all of their 
 water of crystallization, a lower temperature must sometimes be 
 used. In such cases, the substance is placed in a desiccator in 
 which the air is dried by sulphuric acid or calcium chloride. Even 
 under these circumstances many crystals lose water. By diluting 
 the sulphuric acid, however, an atmosphere may be obtained 
 having a tension of water vapor equal to or greater than that 
 produced by the water of crystallization. Long standing in such 
 an atmosphere will free the crystal of adherent or hygroscopic 
 water. 
 
 37. Sources of Error. When substances lose other constitu- 
 ents than water on heating, the error may frequently be avoided 
 by adding an anhydrous substance which is capable of retaining 
 the volatile substance. The oxides of calcium, lead, and bis- 
 muth have been used as retainers for fluorine, chlorine, sulphur, 
 etc. 
 
 Not only must loss of volatile constituents on heating be 
 guarded against, but a possible increase in weight from the ab- 
 sorption of oxygen, carbon dioxide or air must be kept in mind. 
 Metals in their lower state of oxidation may absorb oxygen, and 
 pass to the higher state of oxidation. A familiar example of this 
 is ferrous sulphate, which will absorb oxygen even at the ordinary 
 temperature. Sulphides of some metals easily take up oxygen, 
 becoming sulphates. Some basic oxides of the metals absorb 
 carbon dioxide, forming carbonates. Some minerals, such as 
 the zeolites, have been found to absorb air after the water has 
 been expelled. 
 
DETERMINATION OF WATER. 41 
 
 EXERCISE 7. 
 
 Determination of Water of Crystallization in Copper Sulphate, 
 CuS0 4 .5H 2 O. 
 
 38. Weighing the Salt. A watch crystal of convenient size is weighed, 
 a one-gram weight is added to the right-hand pan and recrystallized copper 
 sulphate is placed on the watch-crystal by means of a spatula until equi- 
 librium is restored. Exactly one gram of copper sulphate will then have 
 been weighed out. In placing the salt on the watch-crystal by means of 
 the spatula, the latter should not be allowed to touch the watch-crystal 
 nor the scale-pan. The spatula with a moderate amount of the salt in it 
 is held in the right hand so that by tapping it gently with the left hand the 
 crystals will fall on the watch-glass. The arms supporting the beam are 
 slightly lowered by the central thumb-screw, so that, by releasing by means 
 of the left hand the spring supporting the scale-pans, the pointer will indi- 
 cate when enough salt has been added by remaining at zero or moving to 
 the right. After a little experience the quickness with which the pointer 
 moves will indicate the amount of salt to be added or taken away. When 
 the pointer moves slowly to the left, only a few small crystals at a time 
 are allowed to fall on the watch-crystal until the pointer ceases to move. 
 The swings may then be taken if necessary and the final adjustment made. 
 If too much has been added, a considerable amount should be very care- 
 fully taken off, and added in small amounts as before. When exactly a 
 gram is weighed out, the calculation of percentage is very simple, as the number 
 of milligrams of water lost is equal to the tenths of per cent. At first it may 
 seem easier to calculate from an amount weighed out which is only approxi- 
 mately one gram than to endeavor to obtain exactly one gram, but it is 
 found that an experienced worker can weigh more readily than make the 
 required calculations. 
 
 39. Determination of Four Molecules of Water. The watch-glass with 
 copper sulphate is now transferred to an air-oven, heated to 115. The 
 temperature is regulated by adjusting the flame of the Bunsen burner 
 from time to time until the thermometer indicates a constant temperature. 
 After about one hour the copper sulphate is transferred to a desiccator, 
 and after ten minutes weighed. It is again heated in the air-bath one- half 
 hour and cooled and weighed. This is repeated until the weight is con- 
 stant. In most analytical work a weight is considered constant if the change 
 is not greater than 0.3 of a milligram, since errors of weighing may be as 
 great as this. Frequently a change of weight of 1 milligram warrants a 
 discontinuation of the heating, since the next heating would probably 
 produce a much smaller change in weight. As the watch-crystal may 
 have lost weight on heating, the copper sulphate is brushed off and the 
 watch-crystal weighed. The difference between this weight and the last 
 weight of the dried copper sulphate and the watch-crystal gives the weight 
 
42 
 
 QUANTITATIVE ANALYSIS. 
 
 of the dried copper sulphate. The difference between the weight of the 
 dried copper sulphate and of the crystals taken is the weight of water 
 expelled. The percentage of water lost in this manner will be found to be 
 very nearly that calculated for 4 molecules of water or 28.85%. The 
 correct weight of the watch-crystal may also be obtained by drying it in 
 the air-bath at 115 until its weight is constant before placing the copper 
 sulphate on it. 
 
 3oa. Determination of the Fifth Molecule of Water. If the air-bath can 
 be heated to 210-215, the fifth molecule of water may be determined 
 in the same manner as the first four. For this purpose either another gram 
 of the crystals may be weighed out and the total amount of water deter- 
 mined or the portion from which the 4 molecules have been expelled may 
 be heated to constant weight at 210 to 215 and the percentage correspond- 
 ing to the fifth molecule determined. If the air-bath cannot be heated 
 to 215, the watch-crystal containing the copper sulphate may be placed 
 on a sand-bath which consists of an iron plate filled with sand and placed 
 on a tripod so as to be heated by means of a Bunsen burner. Until the 
 bulk of the water has been expelled, the thermometer-bulb should be 
 immersed in the sand so as to touch the watch-crystal, which should be 
 
 FIG. 8. 
 
 pressed down firmly in the sand. When the copper sulphate has been 
 heated about one hour in this manner, the bulb of the thermometer, after 
 being freed from sand, should be placed on the copper sulphate in a slanting 
 position. The watch-crystal should be covered with another and the 
 Bunsen burner regulated until the thermometer indicates 210 to 215 
 and the heating continued for one-half hour, when the copper sulphate is 
 carefully brushed off the thermometer and the former cooled in the desicca- 
 tor and weighed. The heating with the thermometer on the copper sulphate 
 is repeated until the weight is constant. The watch-crystal is now weighed 
 
DETERMINATION OF WATER. 43 
 
 after removing the copper sulphate and the percentage of water calculated. 
 The theoretical percentage for 5 molecules is 36.07. 
 
 EXERCISE 8. 
 Determination of Water of Crystallization in Barium Chloride, BaCl 2 .2H 2 O. 
 
 A crucible, preferably of platinum, with its lid, is placed on a pipe-stem 
 triangle and heated with the Bunsen burner for a few minutes, cooled in a 
 desiccator, and weighed. Two grams of pure crystallized barium chloride 
 are weighed out and transferred to the crucible. The crucible is placed 
 on the pipe-stem triangle and heated gently by a small Bunsen burner 
 flame, the lid being on the crucible. The temperature is gradually raised 
 until the crucible attains a low red heat, at which it is maintained for about 
 ten minutes. It is then cooled in a desiccator and weighed. The heating 
 and weighing is repeated until the weight is constant. Theoretical per- 
 centage of water of crystallization in crystallized barium chloride is 14.74. 
 
 40. Penfield's Method of Determining Water. A very simple 
 and accurate method of determining water, especially when it is 
 present in small amount, is that of Pen field.* The apparatus 
 used consists simply of a glass tube closed at one end and having 
 one or two bulbs blown on the middle of the tube If a high tem- 
 perature is required to expel the water, a hard glass tube must be 
 used, otherwise ordinary soft glass tubing may be used. Tubes 
 20 cm. long and 6 mm. internal diameter will be found a convenient 
 size. When the amount of water given off is large, a bulb should 
 be blown on the middle of the tube. The tube is first dried by 
 placing in an air-bath heated to a little above 100 and sucking 
 out the moist air by means of a smaller tube. The tube is then 
 cooled and weighed and the substance to be analyzed is carefully 
 transferred to the bulb at the closed end of the tube, which is again 
 weighed, the difference between this weight and the weight of the 
 empty tube giving the weight of the substance analyzed. 
 
 A short piece of glass tubing drawn out to a capillary is attached 
 to the Penfield tube by means of a short piece of rubber tubing. 
 This simple device prevents the loss of water by circulating air 
 currents. Moist filter-paper or cloth is wrapped around the bulb 
 in the middle of the tube, and the bulb at the end containing the 
 
 * Am. Jour. Sci., 3d series, Vol. XLVIII, p. 31. 1894. 
 
44 QUANTITATIVE ANALYSIS. 
 
 substance being analyzed is heated high enough to expel the 
 moisture. This may generally be accomplished with the Bunsen 
 burner, although some substances require the blast-lamp. The 
 water condenses on the cold middle portion of the tube and col- 
 lects in the bulb, which prevents it from running back to the 
 heated portion and cracking the glass. When all of the water 
 has been expelled the flame is moved so as to melt off the bulb 
 containing the substance being analyzed; when this has been 
 accomplished the tube is cooled and weighed after removing the 
 capillary tube at the end. The tube is now placed in the air- 
 bath and dried as in the beginning of the experiment. It is then 
 cooled and again weighed, the loss of weight being equal to the, 
 amount of water obtained. 
 
 If other volatile substances besides water are given off, a 
 retainer must be used. The oxides of calcium, lead, or bismuth 
 have been used for this purpose. The retainer need not be 
 weighed, but it must be thoroughly dried before being placed in the 
 tube, and enough should be added after the weighed substance 
 to be analyzed has been introduced to fill the bulb and form a 
 layer about two centimeters deep. The retainer must be heated 
 first and then the flame is increased so as to also heat the bulb. 
 When all of the water has been expelled the part of the tube con- 
 taining the retainer is melted off and the determination com- 
 pleted as already described, when no retainer is required. 
 
 The Penfield method is well adapted to the determination of 
 water in minerals and substances which require high heat in order 
 to expel the moisture. The results are accurate if the operation 
 is carefully carried out, although only moderate amounts of water 
 can be weighed in this manner. 
 
 41. Efficiency of Various Drying-agents for Gases. When the 
 water is to be weighed directly, it must be absorbed in a weighed 
 amount of some dehydrating agent. The substances commonly 
 used for this purpose are PHOSPHORUS PENTOXIDE, CONCEN- 
 TRATED SULPHURIC ACID, and FUSED CALCIUM CHLORIDE. PhoS- 
 
 phorus pentoxide is the most efficient drying-agent for gases 
 known. It is doubtful if any water at all remains in a gas which 
 has been dried over this reagent. On the contrary, gases dried 
 over concentrated sulphuric acid give up moisture to phosphorus 
 
DETERMINATION OF WATER. 45 
 
 pentoxide, while gases dried with fused calcium chloride give up 
 moisture both to concentrated sulphuric acid and to phosphorus 
 pentoxide. One litre of air dried with calcium chloride at 15 C. 
 gives up 1 milligram of water to concentrated sulphuric acid, 
 while phosphorus pentoxide is capable of absorbing .002 milli- 
 gram of water per litre after the air has been dried over con- 
 centrated sulphuric acid. 
 
 Equally correct results, however, may be obtained by the use of 
 the less efficient dehydrating agents if care is taken to dry the air 
 when it enters the apparatus to the same extent as when it leaves. 
 This object is best secured by using the same dehydrating agent 
 for both purposes. Some moisture will be present in the air 
 which passes over the hydrated substance, but as exactly the 
 same amount of water is present in the air on leaving the second 
 set of drying-tubes the water actually absorbed by these tubes will 
 be that taken up within the apparatus only. 
 
 42. Drying Properties of Fused Calcium Chloride. When cal- 
 cium chloride is used care must be taken to keep all of the 
 tubes in a series of the same temperature, as this salt differs in its 
 efficiency at different temperatures. At centigrade, 1 litre 
 of air after being dried over calcium chloride contains .3 milli- 
 gram of water; if dried at 15 it contains 1 milligram of water, 
 while if dried at 30, 3.3 milligrams of water remain. In one ex- 
 periment where air was passed through a series of U-tubes 
 (A, B, C, D, E) filled with calcium chloride, which were main- 
 tained at different temperatures, the gains and losses in the 
 weights noted below were observed. Before entering the first 
 tube, A, the air was dried by passing through a tube of calcium 
 chloride maintained at the temperature of A. 
 
 Tubes. Temparatures. Changes in Weight. 
 
 A 16.9 .0000 gram 
 
 B 0.0 .0452 " 
 
 C 30.2 .1565 " 
 
 D 0.0 .1569 " 
 
 E 16.9 .0442 " 
 
 Total change in weight . 0014 gram 
 
 It is extremely difficult to obtain fused calcium chloride which 
 is entirely free from calcium oxide. It will, therefore, be found 
 to absorb a small amount of carbon dioxide. To neutralize the 
 
46 QUANTITATIVE ANALYSIS. 
 
 calcium oxide the calcium chloride should be kept in an atmo- 
 sphere of carbon dioxide or hydrochloric acid for some time and 
 then the acid gas displaced by air. This is best accomplished by 
 filling tubes with it and passing a slow stream of the acid gas 
 through them, and finally drawing air through. When moist air 
 passes over calcium chloride some of it dissolves in the water 
 absorbed, and this liquid mass on the surface of the lumps after- 
 ward solidifies. The so-called Marshand tube is, therefore, best 
 adapted for use with calcium chloride. This U-tube has a bulb 
 blown in the exit tube on one side. The moist air should be 
 led in at this point, so that the water may condense in the bulb, 
 which may be kept cold by moist filter-paper wrapped around it. 
 The calcium chloride in the half of the tube nearest the bulb 
 should be in large lumps, so that the tube shall not be clogged 
 by the dissolved calcium chloride. The other half of the U-tube 
 should be filled with finer material, so as to dry the gas thoroughly. 
 A plug of glass wool should be inserted above the fine calcium 
 chloride, so as to prevent small particles from being carried out 
 by the current of air. 
 
 43. Drying Properties of Concentrated Sulphuric Acid. Con- 
 centrated sulphuric acid is about equally efficient at and 30 
 centigrade. It should be free from sulphur dioxide. If pres- 
 ent, this gas may be expelled by passing a current of air through 
 the acid. If the acid is at all dark or brown in color from the 
 presence of organic matter it should not be used, as sulphur diox- 
 ide will result from the slow oxidation of the organic material. 
 It absorbs carbon dioxide to an appreciable extent, but this is 
 wholly expelled by passing air through it. U-tubes should first 
 be partly filled with glass beads and then sulphuric acid added 
 until the beads are thoroughly moistened and the liquid just 
 forms a seal at the bottom of the tube. Care must be taken that 
 sulphuric-acid U-tubes are always held upright, so that the acid 
 shall not come in contact with the stoppers or enter the exit tubes. 
 It is therefore most convenient to attach a platinum wire, so that 
 they may be suspended during the weighing and at other times. 
 This wire should be left on the tube. 
 
 44. Drying Properties of Phosphorus Pentoxide. Phosphorus 
 pentoxide is seldom used, because of the difficulty of manipu- 
 
DETERMINATION OF WATER. 47 
 
 lating tubes containing it. When moist air comes in contact with 
 it, a very viscuous liquid is produced which very quickly prevents 
 the further passage of gases through it. U-tubes are therefore 
 filled with alternate layers of glass wool and the drying material. 
 The phosphorus pentoxide must not contain any of the trioxide. 
 If the latter is present, the material must be distilled in a stream 
 of oxygen. 
 
 45. Making Impervious Connections. Either cork or rubber 
 stoppers may be used in closing U-tubes after being filled, unless 
 they are provided with glass stoppers. Both cork and rubber are 
 by no means impervious to moisture and carbon dioxide. The ex- 
 posed surface must therefore be coated with sealing wax, paraffine, 
 or shellac varnish. If the latter is used, a sufficient time should be 
 allowed for the escape of the alcohol before weighing the tube. 
 
 When the tubes are connected, short pieces of rubber tubing 
 should be used, but the glass parts of the apparatus should be 
 brought into contact, and the rubber tubing wired or tied to the 
 glass tube. The portion of the rubber tube between the wire 
 or string should be shellacked. Only in this manner can an abso- 
 lutely tight and impervious joint be secured. When weighed 
 U-tubes are disconnected from the other tubes, the ends should 
 be closed with short pieces of rubber tubing closed at one end by 
 glass plugs. These stoppers should be removed while weighing 
 the tubes. 
 
 EXERCISE 9. 
 
 Determination of Water of Crystallization in Magnesium Sulphate, 
 
 MgS0 4 . 7 H 2 0. 
 
 Fill three Marshand tubes with calcium chloride as directed in the pre- 
 ceding section. Shellac the exposed parts of the rubber or cork stoppers. 
 Connect these tubes with a hard-glass bulb as shown in Fig. 9, and support 
 the series so that the bulb-tube may be heated by means of a Bunsen burner. 
 A Bunsen filter-pump or a large bottle filled with water having an exit at 
 the bottom for the water, or a siphon extending to the bottom of the bottle 
 and a second tube passing through the stopper at the top for entrance of 
 air, should be provided to draw air through the apparatus. A wash-bottle 
 containing concentrated sulphuric acid should be inserted between the 
 last U-tube and the aspirator. All joints made by rubber tubing should 
 have the glass tubes in contact and the rubber tubes should be wired or 
 tied to the glass and the part between should be shellacked. After the 
 
48 
 
 QUANTITATIVE ANALYSIS. 
 
 apparatus has been set up the first calcium-chloride tube should be closed 
 by a rubber tube and glass plug. Gentle suction should be applied so 
 
 FIG. 9. 
 
 that on disconnecting with the aspirator, the acid in the wash-bottle shall 
 rise in the long glass tube, showing a partial vacuum in the apparatus. 
 The height of this acid should be carefully noted, and if an appreciable 
 change occurs in five minutes, a leak exists hi some of the joints, which 
 must be found and closed before proceeding with the analysis. When the 
 apparatus has been made air-tight, light the Bunsen burner and heat the 
 bulb-tube to redness and draw air slowly through the apparatus for about 
 one-half hour. The U-tubes to the ri^ht of the bulb-tube are now discon- 
 nected and weighed after being carefully cleaned. In the meantime, the 
 bulb-tube must be closed with a rubber stopper. 
 
 The magnesium sulphate should be weighed out in a glass tube of about 
 5 mm. internal diameter and about 8 cm. long, which has been closed at 
 one end. This tube is weighed, about one-half gram of the salt placed in 
 it, and again carefully weighed. Another piece of glass tubing of the same 
 diameter and of sufficient length is taken, and the weighing-tube containing 
 the salt fastened to it by means of a short piece of rubber tubing of suitable 
 size. By means of this handle, the weighed tube is introduced into the 
 bulb, and by tapping the glass-tube the salt is deposited in the bulb, as 
 shown in Fig. 10. The tube is carefully withdrawn so as not to drop any 
 
 FIG. 10. 
 
 adhering particles of salt, the weighed tube carefully detached from the 
 rubber connector and again accurately weighed with whatever magnesium 
 sulphate is left adhering to it. The difference between the last two weights 
 gives the weight of magnesium sulphate deposited in the bulb-tube. The 
 
DETERMINATION OF WATER. 49 
 
 weighed U-tubes are connected in position again and the apparatus once 
 more tested for leaks. If none are found, the bulb-tube is gently heated 
 by means of the Bunsen burner, while a slow current of air is drawn through 
 the tubes. If the U-tubes become appreciably warm from the heat of the 
 Bunsen burners, they must be shielded from the heat by means of asbestos 
 boards or other non-conducting material. Finally, when the moisture 
 which condenses in the bulb has been removed by the current of ah*, the 
 bulb is heated to dull redness for about fifteen minutes. The burner is 
 then removed, the weighed U-tubes disconnected and weighed again. As 
 soon as the U-tube is removed, the bulb-tube is closed with a rubber stopper. 
 It is then replaced and the heating continued for about half an hour and 
 the U-tubes again weighed. This process is repeated until the U-tubes 
 no longer increase in weight. The percentage of water is calculated from 
 the weight of magnesium sulphate taken and the increase in weight of the 
 U-tubes. 
 
DETERMINATION OF METALS. 
 
 CHAPTER IV. 
 DETERMINATION OF METALS AS OXIDE. 
 
 46. Properties of Weighable Precipitates. In Chapter II mention 
 was made of some of the properties of a salt or other combination 
 of an element which determine its availability for the separation 
 of that element from the solution. The discussion was limited to 
 the requirements of complete precipitation and thorough wash- 
 ing. In other words, all of the element to be determined must 
 be present in the precipitate, and all other substances must be 
 removed from the precipitate. It is then in condition to be 
 weighed as such. Generally the element forms part of a more 
 or less complex chemical compound oxides, chlorides, and sul- 
 phates being common forms in which elements are weighed. This 
 compound must be of definite composition, so that the percentage 
 of the element to be determined is fixed and invariable. The 
 weighed substance must be non-volatile, not strongly hygroscopic, 
 and in general unacted on by the atmosphere or the fumes liable 
 to be present in the laboratory. Other things being equal, a 
 precipitate of simple composition is apt to be of constant compo- 
 sition, and therefore suitable for weighing. 
 
 47. Properties of the Metallic Oxides. The oxides of the 
 metals have been found in many cases to possess all the properties 
 essential to a weighable compound. The affinity of most ele- 
 ments for oxygen is so strong that even on ignition this element is 
 not expelled. The oxygen of the air will even reoxidize most 
 metals if they have become reduced by carbon or reducing-gases. 
 The almost universal solvent in which precipitations are made is 
 water which is composed of oxygen to the extent of about 89%. 
 If a metal is to be weighed as an OXIDE, it may be precipitated as 
 
 50 
 
DETERMINATION OF METALS AS OXIDE. 51 
 
 HYDROXIDE, CARBONATE, or OXALATE. In the case of many metals 
 the salts of volatile acids may be ignited, leaving the metal as 
 oxide. This method is especially applicable to nitrates and 
 organic salts. In some cases the salts of non-volatile acids may 
 be converted into oxides by heating with ammonium carbonate 
 or mercuric oxide, the acid being driven off hi combination as 
 the volatile mercury or ammonium salt. These various methods 
 will now be studied in detail, the metals being considered in groups 
 in which the method of analysis is nearly identical for each group. 
 
 IRON, ALUMINIUM, AND CHROMIUM GROUP. 
 
 METAL PRECIPITATED AS HYDROXIDE BY AMMONIUM HYDROXIDE 
 AND WEIGHED AS OXIDE. 
 
 48. Precipitation of the Hydroxides of Iron, Aluminium, and 
 Chromium. The hydroxides of these three metals are almost 
 entirely insoluble in water, either hot or cold. They differ mark- 
 edly in their solubilities in alkali. Iron is almost absolutely 
 insoluble in ammonia or the fixed alkalies, while aluminium is 
 easily soluble in the fixed alkalies and slightly so in ammonia. 
 Chromium occupies an intermediate position in this respect 
 between iron and aluminium, being almost insoluble in ammonia 
 and moderately soluble in the fixed alkalies. 
 
 Iron may, therefore, be completely precipitated by sodium, 
 potassium, or ammonium hydroxide. It is found, however, that 
 ferric hydrate carries down some of the potassium or sodium 
 which cannot be washed out. Ammonium hydroxide is, there- 
 fore, always used in precipitating iron, which for this purpose 
 must be in the ferric condition. Any ferrous iron may be converted 
 into ferric by boiling the solution after the addition of a few drops 
 of nitric acid or bromine-water. 
 
 As chromium and aluminium are soluble in sodium hydroxide, 
 these metals are also precipitated by ammonium hydroxide. As 
 aluminium hydroxide is soluble in even a slight excess of ammo- 
 nia, the solution must be made as nearly neutral as possible. 
 Ammonium chloride greatly weakens the alkaline properties of 
 ammonia, and for this reason a considerable amount of this salt 
 
52 QUANTITATIVE ANALYSIS. 
 
 is added to the solution of aluminium. This is unnecessary in 
 the case of chromium and iron, because of their insolubility in 
 ammonium hydroxide. 
 
 Chromium hydroxide dissolves slightly in ammonium hydrox- 
 ide in the cold, but is completely precipitated on boiling. Chro- 
 mium must be in the chromic or basic condition. Chromates are 
 reduced to chromic compounds by adding sodium sulphite to 
 the solution acidified with hydrochloric acid, or, still better, by 
 passing in gaseous sulphur dioxide. 
 
 49. Washing. The hydroxides of these three metals are gelat- 
 inous, and therefore there is no danger of their passing through 
 the paper. They clog the pores, however, so that rapid filtration 
 and washing can be accomplished only by using suction or a hot- 
 water funnel. The washing must be thorough, as any ammo- 
 nium chloride left in the precipitate reacts with the hydroxides, 
 liberating ammonia and forming a chloride of the metal which 
 volatilizes so that the result obtained is low. When these hydrox- 
 ides dry they shrivel and crack so that if the washing is discon- 
 tinued for a short time fissures are produced in the precipitate, 
 through which the water passes without taking the impurities 
 out of the precipitate. If the washing must be discontinued, 
 the stem of the funnel should be closed with a piece of rubber 
 tubing and a glass plug and the funnel filled with water. Many 
 of these difficulties may be overcome by using paper pulp, as 
 described on page 27. 
 
 50. Ignition of the Filter-paper. When the filter-paper is 
 burned neither the aluminium nor the chromium hydroxides are 
 reduced by the hot carbon. Ferric hydroxide, however, is partly 
 reduced to the magnetic oxide of iron, Fe 3 4 , while the remainder 
 is converted into ferric oxide, Fe 2 3 . As much of the precipitate 
 as possible should therefore be detached from the paper, so that 
 the error from this source may be negligible. If the filter-paper, 
 with whatever ferric hydroxide cannot be removed, is burned in 
 the platinum crucible, the iron which is reduced is apt to alloy 
 with the platinum and be difficult of removal. It should, there- 
 fore, be burned in the flame of the Bunsen burner while held 
 by a platinum wire. If a more open platinum dish, such as is 
 used in milk analysis, is at hand, the moist precipitate may be 
 
DETERMINATION OF METAL AS OXIDE. 53 
 
 placed in the dish with the apex of the filter-paper on top. The 
 side of the dish is now heated so that the paper dries and finally 
 burns. The access of the air to the paper is so good that little 
 iron is reduced. 
 
 51. Contamination of the Precipitate with Silica. It is very 
 difficult to obtain precipitates of these three metals which are 
 free from silica. The ammonia which is used as a precipitant 
 almost invariably contains silica which has been derived from the 
 reagent bottle, most of which is precipitated and can be filtered 
 off. Appreciable amounts of silica are obtained by the disinte- 
 gration of the glass of the beaker in which the metal is precipi- 
 tated. This silica is filtered off and weighed with the precipitate. 
 The digestion of the alkaline solution should, therefore, be as brief as 
 possible, and the excess of ammonia as small as possible. If too 
 much ammonia is added it should be neutralized with dilute 
 hydrochloric acid. It is also better to conduct the precipitation in 
 a porcelain dish, since porcelain is less easily attacked by alkalies 
 than glass. If a platinum dish is available, this source of error is 
 entirely avoided. If glass is used the error from this source may 
 be reduced, by working rapidly, to almost negligible quantities of 
 .1 or .2%. If Jena beakers are used the error is still less. 
 
 If the highest degree of accuracy is desired the percentage of 
 silica must be determined and a correction made on the weight of 
 the original precipitate. This determination is carried out by 
 dissolving the precipitate in acid, evaporating to dryness on the 
 water-bath, redissolving in acid and water, and filtering off the 
 silica, which is washed, ignited, and weighed. For this purpose 
 the iron precipitate may be dissolved by digesting the powdered 
 material with hydrochloric acid. Aluminium oxide may be dis- 
 solved in the same manner, but the action is very slow unless paper 
 pulp is used during the precipitation, as directed on page 27. It dis- 
 solves more easily if first heated with eight parts of concentrated 
 sulphuric acid and three parts of water, and then water added. 
 It is also rendered soluble by fusion with potassium bisulphate. 
 This method is especially to be recommended for both iron and 
 aluminium, because the silica may be rendered insoluble by 
 dissolving the melt in dilute sulphuric acid and evaporating until 
 fumes of sulphuric acid are evolved. Chromium oxide is best 
 dissolved by fusion in the platinum crucible with sodium carbo- 
 
54 QUANTITATIVE ANALYSIS. 
 
 nate and potassium nitrate, potassium chromate being formed. 
 This solution must be acidified with hydrochloric acid and evap- 
 orated to dryness and the silica determined as in the other oxides. 
 52. Removal of Organic Material. These metals are not 
 completely precipitated if organic matter is present in the solu- 
 tion. It may be detected and also removed by evaporating the 
 solution to dryness and igniting. This is not applicable if chlo- 
 rides are present. In that case boiling with concentrated nitric 
 acid or potassium chlorate and hydrochloric acid must be resorted 
 to. The solution may also be treated with excess of sulphuric acid 
 and evaporated to dryness and ignited. If the amount of organic 
 matter is small, the ignition will frequently be unnecessary, as 
 the charred material will be entirely destroyed by the concentrated 
 sulphuric acid produced by the evaporation. A few drops of 
 nitric acid followed by heating the solution will render it clear 
 more quickly. 
 
 EXERCISE 10. 
 
 Determination of Aluminium in Potash Alum, 
 K 2 S0 4 .A1 2 (S0 4 ) 3 .2 4 H 2 0. 
 
 Weigh carefully 2 grams of pure recrystallized potash alum. If bal- 
 anced watch-crystals are used, place them on the scale-pans and ascertain 
 if the pointer indicates zero when the beam and pans are released. If it 
 does not, balance the watch-crystals by means of a small weight or the 
 rider. Place a 2-gram weight on the right-hand watch-crystal and weigh 
 out 2 grams of the alum as directed in Exercise 4. Transfer to a 500-c c. 
 beaker, brushing off the watch-crystal with a clean camel's-hair brush, and 
 dissolve in 200 c.c. of water. Add about 50 c.c. of ammonium chloride 
 solution, which should be filtered if not perfectly clear. The ammonium 
 chloride may also be introduced together with paper pulp by adding 15 c.c. 
 of concentrated hydrochloric acid to two or three ashless filter-papers * and 
 25 c.c. of dilute ammonia. Warm on a hot plate or an asbestos board or 
 wire gauze over a Bunsen burner. Add filtered dilute ammonia with con- 
 stant and vigorous stirring of the solution until the odor of ammonia can 
 be detected after vigorous stirring. Red litmus-paper should be only 
 slowly turned blue when held over the solution. If too much ammonia has 
 been added it should be nearly neutralized with dilute hydrochloric acid. 
 
 * The weight of the ash of this paper is stated on the package obtained from 
 the dealers in chemical apparatus and is usually small enough to be neglected. 
 The paper should be kept in a box or other receptacle where it can be kept per- 
 fectly clean. 
 
DETERMINATION OF METAL AS OXIDE. 
 
 55 
 
 53. Washing and Transferring the Precipitate to the Filter-paper. 
 After the solution has thus been made faintly alkaline, heat it nearly to 
 the boiling-point for a few minutes. Fit a funnel with an ashless filter-paper 
 preparatory to filtering by suction as directed in Chapter II. An 11 or 
 I2 cm. filter-paper should be used and the funnel should be of such a size 
 that a space at least a quarter of an inch is left free above the paper. Allow 
 the precipitate to settle a few minutes and then decant the clear liquid 
 through the paper, using suction. Guide the stream of water by holding 
 the stirring-rod against the lip of the beaker. Add about 75 c.c. of hot 
 water from the wash-bottle,* which for this purpose is kept on a steam-bath 
 or hot plate. Stir the precipitate well with the stream of water. For 
 this purpose the jet of water must be quite large and care must be taken 
 not to spatter the precipitate out of the beaker. Allow the precipitate to 
 settle and decant through the funnel as before. Repeat the washing by 
 decantation twice aud then transfer the main precipitate to the funnel. 
 Holding the stirring-rod against the beaker with the left hand, sweep the 
 precipitate from the beaker by means of a stream of water from the wash- 
 bottle, which is held in the right hand as shown in Fig. 11. When the 
 
 FIG. 11. 
 
 bulk of the precipitate has been transferred in this manner, the particles 
 adhering firmly to the glass must be removed by means of a so-called police- 
 man, which is made by inserting the end of a rather large-sized stirring-rod 
 
 * A so-called hot-water wash-bottle is prepared by fastening asbestos paper, 
 a sheet of cork, or other insulating material around the neck of an ordinary wash- 
 bottle by means of pieces of wire or string. A thick cord closely wound around 
 the neck of the flask will serve the same purpose. The water should not be 
 allowed to boil or steam will enter the mouth when blowing out the water. To 
 overcome this difficulty, a piece of rubber tubing should be placed on the glass 
 tube which enters the mouth. By closing this tube with the teeth when not 
 blowing, steam is kept out of the mouth. 
 
56 QUANTITATIVE ANALYSIS. 
 
 into a short piece of rubber tubing. The rubber tube should be left pro- 
 jecting slightly beyond the end of the glass tube and sealed together with 
 a little bicycle cement. 
 
 The bottom as well as the sides of the beaker must be rubbed clean 
 with this policeman. Both the stirring-rod and the policeman must finally 
 be freed from particles of the precipitate. If portions of the precipitate 
 cannot be rubbed off the glass they should be dissolved in a few drops of 
 dilute hydrochloric acid and reprecipitated by neutralizing with ammonia 
 and warming the solution. When all of the precipitate has been transferred 
 to the funnel, it is washed with hot distilled water until free from chlorides. 
 The exposed portion of the filter-paper must also be well washed. Test 
 the wash-water from time to time with nitric acid and silver nitrate, finally 
 using several cubic centimeters. In washing this precipitate, care must 
 be taken that all of the water is not sucked off the precipitate so that it 
 cracks, producing fissures through which the wash-water passes without 
 dissolving the ammonium chloride. If for any reason the washing must 
 be discontinued, the stem of the funnel should be closed with a short piece 
 of rubber tubing and a glass plug and the funnel filled with water. 
 
 54. Ignition and Weighing of the Precipitate. During the washing of the 
 precipitate a platinum crucible with the lid is supported on a clean pipe- 
 stem or platinum triangle resting on a tripod and heated to redness with a 
 Bunsen burner. The air-vents of the Bunsen burner are adjusted so that 
 the flame is entirely colorless. The crucible must be kept well above the 
 inner cone. While still red-hot the crucible is seized with a clean pair of 
 forceps or crucible tongs and placed in a desiccator, which, for this purpose, 
 is provided with a pipe-stem or glass triangle. After fifteen minutes it is 
 weighed, the weight being recorded on the right-hand side of the NOTE- 
 BOOK, the page being fully labelled and dated. The left-hand page is 
 reserved for a brief description of the process of analysis, and especially 
 any procedure which led to disaster or failure and the remedy adopted. 
 All of this work should be done during the intervals when the liquid is 
 running through the funnel. It will run just as fast if not watched. 
 
 The moist precipitate with the paper is taken out of the funnel and 
 transferred to the weighed platinum crucible. If there is any delay, so 
 that the precipitate is partially dried, it must be placed in the steam-oven 
 and completely dried. It then shrinks to a very small bulk. The lid of 
 the crucible must in this case be very carefully placed on the crucible, as a 
 little moisture left in the hard lumps of the precipitate forms steam and 
 scatters the precipitate with considerable force. When burning the moist 
 precipitate, the paper is placed in the crucible with the apex up, so that 
 spattering during drying will not throw the precipitate out of the crucible. 
 Any portions of the precipitate left on the funnel are wiped off with small 
 pieces of quantitative filter-paper and placed in the crucible. The covered 
 crucible is then placed on the triangle and heated with a very small flame 
 
DETERMINATION OF METAL AS OXIDE. 57 
 
 until steam and volatile matter cease to escape. The heat is gradually 
 increased until the water and most of the volatile matter are expelled from 
 the precipitate and the paper. The lid is now taken off the crucible, which 
 is placed on its side and heated with the full flame of the Bunsen burner. 
 As the carbon burns, the crucible is turned to expose fresh portions to the 
 action of the air. The crucible is finally heated with the blast-lamp until 
 the aluminium oxide is perfectly white. While still red-hot, the crucible 
 is transferred to the desiccator and after fifteen minutes, weighed. The 
 crucible is again heated in the blast-lamp, cooled in the desiccator, and 
 weighed. This is repeated until the weight is constant within .2 or .3 
 milligram. 
 
 The percentage of A1 2 3 found is calculated by dividing the weight of 
 the precipitate by the weight of alum taken and multiplying the result by 
 100. The theoretical percentage of A1 2 O 3 in potash alum is 10.76. If a 
 result much higher than this is obtained, the precipitate may be contami- 
 nated with silica, which is determined as directed in the next exercise, with 
 the exception that grinding the precipitate will be unnecessary if paper 
 pulp has been used during the precipitation. 
 
 EXERCISE ii. 
 Determination of Iron in Soft-iron Wire.* 
 
 In this exercise the determination should be carried out in duplicate, 
 the two analyses being carried on side by side. 
 
 Clean the wire thoroughly by rubbing with emery-cloth or sandpaper 
 until all rust is removed and finally wiping with filter-paper. The wire 
 should be wound in a spiral around a lead-pencil and should not be touched 
 with the fingers after being wiped with the filter-paper. Weigh out from 
 200 to 300 milligrams of the wire and transfer to a 200-c.c. beaker. Add 
 5 c.c. of concentrated nitric acid and a little water. Cover the beaker with 
 a watch-crystal and warm on the hot plate under the hood. When the 
 iron is all dissolved, rinse off the watch-crystal with a little distilled water 
 and evaporate to dryness on the hot plate. Add 10 c.c. of dilute hydro- 
 chloric acid and again evaporate to dryness on the water-bath. After the 
 ferric chloride is dry, heat for one-half hour on the water-bath, dissolve in 
 a few cubic centimeters of dilute hydrochloric acid, add a little water, and 
 filter off the silica on a small filter-paper. Wash the paper free from iron 
 with hot distilled water to which a little hydrochloric acid has been added. 
 The volume of the filtrate should be about 200 c.c. 
 
 It is best to precipitate the iron in a platinum dish. If such a dish is 
 not available, an evaporating-dish of Berlin porcelain should be used or a 
 
 *This wire is such as is used in standardizing potassium permanganate or 
 other oxidizing solutions and is sold for use as piano-strings, and to florists for 
 making wreaths. 
 
58 DETERMINATION OF METALS. 
 
 Jena beaker. The solution is neutralized with filtered ammonia, and 
 brought to a boil. After allowing the precipitate to settle for a few minutes, 
 the clear liquid is decanted through a 12^ or 15-cm. ashless filter-paper 
 arranged for suction as described in Chapter II. The precipitate is washed 
 twice by decantation and then transferred to the funnel and washed with 
 hot water, observing the precautions given in Exercise 10. Toward the 
 end of the washing the stream of water from the wash-bottle should be so 
 directed as to loosen the precipitate from the sides of the paper and wash 
 it down into the point of the funnel. When the precipitate has been 
 washed free from chlorides it is dried in the steam- or air-oven. The tem- 
 perature of the latter should not be allowed to rise much above 100 Centi- 
 grade. 
 
 55. Ignition of the Paper on Platinum Wire. A piece of glazed paper, 
 preferably of a light color, about 8X10 inches is spread out on the desk. 
 A small camel's-hair brush, a piece of platinum wire about 4 inches long 
 fastened to a glass or other handle, a Bunsen burner with a long blue flame, 
 the ignited and weighed platinum crucible and the dried precipitate in the 
 funnel are brought together. The glazed paper is carefully brushed clean 
 with the camel's-hair brush. The crucible is placed in the center of the 
 paper. The paper is taken out of the funnel and the precipitate loosened 
 with the platinum wire and dropped into the crucible. Very little of the 
 precipitate should be left on the paper. Rubbing the two inner surfaces 
 of the filter-paper together will sometimes loosen a considerable amount 
 of the precipitate. If any portion of the precipitate is found adhering to 
 the funnel it should be rubbed off with a small piece of ashless filter-paper 
 which has been moistened with a little dilute nitric acid. The filter-paper 
 is now folded as if to be inserted into a funnel. It is rolled up tightly on 
 one of its edges and the platinum wire is wrapped around the upper part 
 of the paper which is always freest from portions of the precipitate. It is 
 now held by means of the handle of the platinum wire over the crucible 
 and the point touched to the Bunsen burner flame and ignited. The charred 
 paper is completely burned by holding it in the oxidizing flame of the Bunsen 
 burner and then exposing the red-hot carbon to the oxygen of the air. 
 The ash is allowed to drop into the crucible. The wire is now brushed off, 
 the crucible is taken in the left hand and any particles which have fallen 
 on the paper are brushed into it. The crucible is placed on its side on a 
 pipe-stem triangle and heated with the Bunsen burner until any unburned 
 carbon of the filter-paper is oxidized. 
 
 If a platinum dish is available the moist precipitate with the apex up 
 may be placed in the dish which is gently heated on one side with the Bunsen 
 burner. When the paper is completely charred the dish may be heated 
 to redness until the carbon is completely burned. The precipitate is then 
 heated with the blast lamp for a few minutes, cooled in the desiccator and 
 weighed. It is re-ignited and weighed until the weight is constant. 
 
DETERMINATION OF METALS AS OXIDE. 59 
 
 56. Calculation. The weight of Fe 2 O 3 multiplied by the percentage 
 of iron in Fe 2 3 (70 per cent) gives the weight of iron found. This 
 number multiplied by 100 and divided by the weight of iron taken gives 
 the percentage of iron in the wire which is usually 99.6 to 99.8 per cent. 
 The result may also be computed by logarithms. The log. of the weight 
 of the precipitate is added to the log. of the percentage of iron in Fe 2 3 . 
 From this sum is subtracted the log. of the weight of iron wire taken. This 
 log. +2.0 gives the log. of the percentage of iron in the wire. The dupli- 
 cates in this analysis should agree within .1 to .2 per cent. If they vary 
 more than this amount a third or even a fourth analysis must be made. 
 
 57. Determination of Silica. If the percentage found is greater than 
 99.8 per cent, the error is probably due to silica, especially if the precipita- 
 tion was carried out in a beaker. Instead of making another determination, 
 the percentage of silica should be determined and a correction made on the 
 percentage found. For this purpose the precipitate is ground quite fine in 
 an agate mortar. Loss by spattering should be avoided by covering the 
 mortar with a piece of paper through which the pestle passes. When as 
 much of the precipitate as can be readily removed from the crucible has been 
 ground, the pulverized material is returned to the crucible which is then 
 weighed. The difference between this weight and the weight of the crucible 
 gives the weight of the iron oxide taken for the silica determination. The 
 loose material is transferred to a 100-c.c. beaker, 15 c.c. concentrated hydro- 
 chloric acid and 10 c.c. water added. The beaker is digested on the hot 
 plate until the iron is dissolved. A little acid is also added to the crucible 
 which is also heated until the iron is dissolved. This solution is added to 
 that in the beaker. The beaker is now placed on the water-bath and the 
 solution evaporated to dryness and heated dry for one-half hour. Dilute 
 hydrochloric acid is added and the heating continued until the iron is again 
 dissolved. 
 
 The powdered precipitate, after being weighed in the crucible, may 
 also be fused with acid potassium sulphate. The burner should be turned 
 down until the heat is just sufficient to keep the material fused. When 
 the iron is all dissolved the crucible is allowed to cool, transferred to a 
 beaker and the melt extracted with dilute sulphuric acid and the solution 
 evaporated until fum.es of sulphuric acid are evolved. 
 
 The solution obtained by either method is filtered through an ashless 
 filter-paper, the silica in the beaker being transferred to the paper by means 
 of a policeman as described in Exercise 10. The silica is washed and, 
 after burning the paper, is ignited, finally for a few minutes with the 
 blast-lamp. The crucible is cooled in the desiccator and weighed. The 
 contents of the crucible should be pure white, any tinge of red indi- 
 cating iron due to imperfect washing or undissolved ferric oxide. On 
 adding a little pure hydrofluoric acid the precipitate should be entirely 
 volatilized on heating. Any residue which, on repeated treatment with 
 
60 DETERMINATION OF METALS. 
 
 hydrofluoric acid, fails to volatilize is not silica. As only part of the 
 original iron precipitate was taken for this analysis the amount of silica 
 in the entire precipitate must be calculated and deducted to give the 
 true weight of Fe 2 3 from which the percentage of iron in the iron wire 
 must be re-calculated, as in the following illustration. 0.3350 gram of t he 
 iron wire was weighed out. 0.4820 gram of the ferric oxide was obtained. 
 The percentage of iron calculated from this weight of oxide by the method 
 already given is 100.7. After grinding the precipitate for the silica deter- 
 mination, 0.4153 gram remained in which .0043 gram Si0 2 was found. 
 The amount present in the entire precipitate was obtained from the pro- 
 portion 0.4135 : .0043 :: 0.482 : x where a; = .0050. On subtracting the 
 silica from the weight of the impure precipitate we obtain 0.4770 as the 
 weight of ferric oxide. On recalculating the percentage of iron we obtain 
 99.68. 
 
 DETERMINATION OF COPPER, MANGANESE, NICKEL, 
 AND COBALT BY PRECIPITATION WITH CAUSTIC 
 ALKALI. 
 
 58. Precipitation and Washing. A number of metals cannot 
 be precipitated as hydroxide with ammonia, because of their 
 solubility in ammonium salts. Some of these metals form hydrox- 
 ides which are insoluble in sodium and potassium hydroxides, 
 and may be precipitated by either of these reagents. Consider- 
 able difficulty is experienced in completely washing out the excess 
 of alkali from the precipitate and, unlike ammonia, it is not 
 expelled by ignition. On digesting the ignited precipitate with 
 water the alkali can frequently be detected by its alkaline reac- 
 tion. By filtering, washing, and reigniting, a considerable loss 
 in weight is frequently noted. 
 
 The solution should be dilute and only a slight excess of alkali 
 added. The liability of contamination of the precipitate with 
 silica from the action of the alkali on the glass, which was met 
 with in the determination of iron, aluminium, and chromium is 
 still more pronounced when the stronger alkalies are used. The 
 precipitation should therefore never be carried out in glass. Por- 
 celain or, wherever possible, platinum dishes should be used. 
 The precipitate should be tested for the presence of silica by solu- 
 tion in hydrochloric acid, evaporation to dry ness, again dissolv- 
 
DETERMINATION OF METALS AS OXIDE. 61 
 
 ing in acid, filtering, washing, and weighing the residue as directed 
 under iron. These oxides dissolve with ease in acid. 
 
 59. Composition of the Ignited Oxides. Considerable diffi- 
 culty is experienced in obtaining precipitates of constant compo- 
 sition. CUPRIC HYDROXIDE is converted on ignition into cupric 
 oxide, CuO, but if reduced by the filter-paper or reducing gases 
 from the flame to metallic copper it reoxidizes to cuprous oxide, 
 Cu 2 0. This is converted into cupric oxide by moistening with 
 nitric acid, evaporating to dryness and igniting. 
 
 MANGANESE HYDROXIDE is oxidized by the oxygen of the air 
 even at the ordinary temperature. It is not reduced either by 
 the carbon of the filter-paper or by the reducing gases of the 
 flame. On ignition in presence of air it is converted into man- 
 ganoso-manganic oxide, Mn 3 4 . It is impossible, however, to 
 obtain a precipitate in which the percentage of manganese cor- 
 responds to the theoretical for Mn 3 4 . Pickering has shown that 
 the amount of manganese in the ignited oxide varies from 69.688 
 to 74.997%, according to the temperature to which the oxide 
 has been heated and other undetermined conditions. For this 
 reason only small amounts of manganese should be weighed as 
 Mn 3 4 . 
 
 When COBALTOUS HYDOXIDE is heated in contact with the 
 solution from which it was precipitated, it is. gradually converted 
 by the oxygen of the air into cobaltoso-cobaltic hydroxide. On 
 drying, it absorbs more oxygen, but on strong ignition oxygen is 
 driven off, leaving cobaltous oxide CoO. On cooling in contact 
 with the air, oxygen is again absorbed, the light brown cobaltous 
 oxide changing more or less completely to the black cobaltoso- 
 cobaltic oxide Co 3 4 . If the air is excluded by a current of car- 
 bon dioxide this action is prevented, and the pure cobaltous 
 oxide is obtained. This oxide is readily reduced to metallic 
 cobalt by ignition in a stream of hydrogen and the alkali which 
 could not be completely removed by washing the hydroxide can 
 now be quite readily extracted with hot water. 
 
 NICKELOUS HYDROXIDE is converted on ignition into nickelous 
 oxide NiO, which is unaltered on heating in the air. It is readily 
 reduced to metallic nickel on ignition in a stream of hydrogen or 
 of carbon monoxide. 
 
62 DETERMINATION OF METALS. 
 
 EXERCISE 12. 
 
 Determination of Copper in Crystallized Copper Sulphate, 
 CuS0 4 .5H 2 O. 
 
 Weigh out 1 gram of the pure salt, transfer to a platinum or porcelain 
 dish, and dissolve in about 200 c.c. of distilled water. Make a dilute solution 
 of caustic soda by dissolving about 1 gram in 25 c.c. distilled water. Heat 
 the copper solution nearly to boiling and add the caustic-soda solution with 
 constant stirring until no further precipitation occurs. Allow to settle 
 for a few minutes, then decant through a paper fitted into a funnel for 
 suction. Wash the precipitate twice by decantation with hot water, then 
 transfer to the funnel and complete the washing with hot water until the 
 wash-water is perfectly neutral. Dry the precipitate in the funnel at 
 100 C. Remove the precipitate from the paper as directed in Exercise 11 
 and place it on a watch-crystal. Burn the paper on the platinum wire, 
 being careful to wrap the wire around a portion of the paper on which there 
 is the least amount of precipitate. Let the ash fall into a porcelain cruci- 
 ble. Add a few drops of concentrated nitric acid, place the lid on the 
 crucible, and evaporate off the excess of acid with a very small flame 
 of the Bunsen burner, finally heating quite strongly. If any carbon from 
 the filter-paper remains unburned, repeat the treatment with nitric acid. 
 When the crucible is cold set it on the glazed paper and transfer the main 
 portion of the precipitate to the crucible, brushing the watch-crystal and 
 paper thoroughly. The moist paper with the precipitate may also be placed 
 in the crucible and the paper burned as directed in Exercise 10. The 
 precipitate must be moistened with nitric acid as directed for the ash of 
 the filter-paper. Finally heat the crucible with the full flame of the Bunsen 
 burner for ten or fifteen minutes, cool in the desiccator, and weigh. As 
 copper oxide is hygroscopic, the weight should be taken rapidly. The 
 crucible is reheated and when weighed the second time the weights should 
 be placed on the right-hand pan before the crucible is placed on the left-hand 
 pan so as to make the final adjustment with the rider or by taking the 
 swings of the pointer before the precipitate gains appreciably in weight. 
 The percentage of copper found in copper sulphate should be within .1 or 
 .2 per cent of the theoretical 25.46 per cent. The method gives high results. 
 
 EXERCISE 13. 
 
 Determination of Nickel in Nickel-ammonium Sulphate, 
 (NH 4 ) 2 S0 4 .NiS0 4 .6H 2 0. 
 
 Weigh out 2 grams of crystallized nickel-ammonium sulphate and trans- 
 fer to a porcelain dish. Dissolve in about 200 c.c. distilled water. Heat 
 nearly to boiling and add, with stirring, a dilute but freshly made solution 
 of caustic soda until precipitation is complete. Continue heating the solu- 
 tion for ten or fifteen minutes. Wash by decantation two or three times, 
 then wash thorouehlv with hot water on the filter-paper. Dry the pre- 
 
DETERMINATION OF METALS AS OXIDE. 63 
 
 cipitate, detach from the paper, and place it on a watch-crystal. Burn 
 the filter-paper over the porcelain crucible. Treat the residue with a drop 
 of concentrated nitric acid, evaporating off the excess. Add the remainder 
 of the precipitate and heat with the Bunsen burner to constant weight. 
 The precipitate should now be examined for alkali by digesting with hot 
 water and testing with litmus. If alkali is present it is removed by 
 repeated extraction with hot water, the precipitate being finally heated to 
 constant weight again. Calculate the percentage of nickel found in the 
 nickel-ammonium sulphate. Theoretical percentage is 14.86. 
 
 The nickel oxide may also be converted into metallic nickel by heating 
 in a stream of hydrogen. For this purpose the perforated lid and porce- 
 lain stem cf the Rose crucible are most convenient. In place of this an 
 ordinary clay pipe may be used. The hydrogen from a Kipp generator 
 should be freed from arsine by passing it through a mercuric chloride solu- 
 tion and dried by passing through calcium chloride or concentrated sul- 
 phuric acid. The apparatus is set up as shown in Fig. 12. The stream 
 of hydrogen should be passed for a few minutes to displace the air from 
 the wash-bottle and the U-tube. An explosion will not pass through the 
 
 FIG. 12. 
 
 apparatus if a plug of cotton or glass wool is piaced in the glass tube or 
 limb of the U-tube nearest the crucible. The crucible is then gently heated 
 with a small flame from a Bunsen burner, which is turned on full after a 
 few minutes. The crucible is finally allowed to become nearly cold in the 
 stream of hydrogen. It is then transferred to the desiccator for a few 
 minutes before weighing. The ignition in the stream of hydrogen is 
 repeated until the weight is constant. The precipitate is then tested for 
 silica by dissolving the metal in a little dilute nitric acid. If any silica is 
 found, it is filtered off, washed, and weighed. 
 
CHAPTER V. 
 DETERMINATION OF METALS AS OXIDE. 
 
 PRECIPITATION OF LEAD, BISMUTH, CALCIUM, BARIUM, 
 AND STRONTIUM BY AMMONIUM CARBONATE. 
 
 60. Precipitation by Ammonium Carbonate. Some of those 
 metals whose oxides or hydroxides are soluble in water or in 
 solutions of the alkali hydroxides may be precipitated by ammo- 
 nium or sodium carbonate. Ammonium carbonate cannot be 
 used with some of the metals because the precipitate dissolves, 
 forming a double salt with the ammonium compounds produced 
 by the neutralization of the mineral acid with which the base 
 was combined. Ammonium carbonate is used in preference to 
 sodium or potassium carbonate, because of the greater ease of 
 freeing the precipitate from the ammonium salts by washing or 
 volatilization during ignition. On the other hand, some carbon- 
 ates of the metals which can be precipitated by ammonium car- 
 bonate form acid carbonates which are slightly soluble. Ammo- 
 nium carbonate as obtained either in the solid form or in solu- 
 tion almost invariably contains acid ammonium carbonate. Free 
 ammonia must therefore always be added to the solution to be 
 precipitated. The precipitation is therefore usually said \to be 
 made by ammonia and ammonium carbonate, by the latter term 
 being designated the commercial article. . 
 
 61. Conversion of the Carbonate into the Oxide. The ease of 
 converting the carbonates of these metals into oxides varies 
 greatly. Lead carbonate loses all of its carbon dioxide below red 
 heat ; indeed at the ordinary temperature part of it passes off so that 
 the precipitate formed by ammonia and ammonium carbonate is a 
 basic lead carbonate. In the case of the alkaline-earth metals 
 the precipitate formed in each case is the normal carbonate, and 
 it may be dried and weighed as such. Calcium carbonate may be 
 
 64 
 
DETERMINATION OF METALS AS OXIDE. 65 
 
 completely converted by the heat of the blast-lamp into calcium 
 oxide, so that this metal may be weighed either as the oxide or 
 as the carbonate. It is impossible to completely convert stron- 
 tium or barium carbonates into oxides. These metals are, there- 
 fore, weighed as carbonates.* 
 
 As LEAD SALTS are very easily reduced to the metallic form, 
 the bulk of the lead precipitate must be removed from the filter- 
 paper before the latter is burned. The lead which remains on 
 the paper is reduced to metal, and must be reconverted to oxide 
 by treatment with concentrated nitric acid and ignition of the 
 lead nitrate formed. The bulk of the lead precipitate is then 
 placed in the crucible and by ignition converted into the monox- 
 ide of lead. 
 
 If it is desired to weigh the CALCIUM as OXIDE, the precipitate 
 need not be dried or removed from the filter-paper, but is intro- 
 duced into the crucible, dried by gentle heat, ignited with access 
 of air to burn the paper, and finally heated with the blast-lamp 
 to constant weight. This is, therefore, the simplest and most 
 rapid method of weighing calcium. It has the disadvantage, 
 however, that the calcium oxide is very hygroscopic, and when 
 moist absorbs carbon dioxide from the air. The precipitate 
 must, therefore, be weighed as quickly as possible. 
 
 CALCIUM CARBONATE is stable in the air and may, therefore, 
 be weighed with the greatest accuracy. If the precipitate is to 
 be weighed as carbonate, it must be dried and removed from the 
 paper. This is not difficult, as it forms a loose powder on dry- 
 ing. It is placed on a watch-crystal, while the paper is burned in 
 the crucible or on a platinum wire. By this treatment the cal- 
 cium remaining on the paper is converted into oxide, which must 
 be changed to carbonate. This is accomplished by sprinkling 
 some dry powdered ammonium carbonate on the calcium oxide, 
 moistening with a drop or two of water, and heating very gently 
 to drive off the excess of ammonium carbonate and water. This 
 operation is repeated, then the bulk of the precipitate is added, 
 which is also heated gently with the Bunsen burner. After 
 weighing the precipitate, it must be treated with ammonium 
 
 * Calcium carbonate decomposes at 825, strontium carbonate at 1155, 
 and barium carbonate at 1450. 
 
66 DETERMINATION OF METALS. 
 
 carbonate and a little water and again heated and weighed, to 
 ascertain if the weight is constant. 
 
 The precipitate of BARIUM or STRONTIUM may be thrown 
 moist into the platinum crucible and heated with the Bunsen 
 burner until dry and the paper burned. The hot carbon reacts 
 with the carbonates of these metals, forming carbon monoxide 
 and the oxide of the metal. After the complete combustion of 
 the filter-paper, the precipitate should be treated with ammonium 
 carbonate and water and heated to expel the excess of the 
 reagent, finally to dull redness with the Bunsen burner. 
 
 EXERCISE 14. 
 Determination of Strontium in Strontium Carbonate, SrC0 3 . 
 
 Weigh out 1 gram of pure strontium carbonate. Transfer to a 400-c.c. 
 beaker. Add 50 c.c. distilled water and dilute hydrochloric acid drop by 
 drop with stirring or agitation of the solution until the strontium is dissolved. 
 Dilute to 150 c.c., add a few cubic centimeters dilute filtered ammonia, then 
 ammonium carbonate solution with constant stirring until no further pre- 
 cipitation occurs. Allow the beaker to stand in a warm place for several hours. 
 Filter and wash the precipitate with water containing a little ammonia. 
 Transfer the moist precipitate to a weighed platinum crucible. With 
 the lid on the crucible, heat gently until the precipitate is dry. Remove 
 the lid, place the crucible on its side, and heat strongly until the paper is 
 burned. Cool, sprinkle some powdered ammonium carbonate over the 
 precipitate, moisten with a little water, heat gently with the lid on the 
 crucible until the precipitate is dry, then more strongly, but not to red- 
 ness. Cool in a desiccator and weigh. Repeat until constant weight is 
 obtained. The theoretical percentage of strontium in strontium carbonate 
 is 59.37. 
 
 PRECIPITATION OF CALCIUM BY AMMONIUM OXALATE. 
 
 62. Calcium is almost invariably precipitated as oxalate, since 
 this salt of calcium is more insoluble than the carbonate, while on 
 ignition it can be converted into the carbonate or the oxide as 
 desired. It forms a very finely divided powder when first pre- 
 cipitated, which gives considerable trouble by passing through 
 the pores of the filter-paper. By digesting the precipitate with 
 the hot solution from which it has been separated, the crystals 
 increase in size so that after two or three hours it can be filtered 
 without any difficulty. The same object is more quickly accom- 
 plished if the solution of calcium is brought to a boil before adding 
 
DETERMINATION OF METALS AS OXIDE. 67 
 
 the ammonium oxalate, which should also be hot. The solution 
 should also be vigorously stirred during the precipitation. If 
 the precipitate is heated very gently, it decomposes almost 
 completely, according to the equation CaC 2 4 = CaC0 3 + CO. 
 Almost invariably, however, the calcium carbonate will be dark- 
 colored because of the formation of some free carbon. On heat- 
 ing the precipitate strongly enough to cause the combustion of 
 this carbon and thus to give it a pure-white color, a large portion 
 of the precipitate will be converted into oxide. That portion 
 which remains with the filter-paper will also be converted into 
 oxide. Instead of attempting to produce the carbonate in the 
 first instance, it is simpler to heat the precipitate with the paper 
 from the beginning quite strongly with the Bunsen burner or the 
 blast-lamp until it is perfectly white. It may then be brought to 
 constant weight as calcium oxide, or it may be converted into 
 carbonate by the addition of at least an equal bulk of ammonium 
 carbonate, moistening with water and heating gently. This is 
 repeated until constant weight is obtained. By weighing the 
 calcium as oxide, the result is undoubtedly obtained more quickly, 
 and if the weighing is done rapidly it is quite as accurate as when 
 the calcium is weighed as carbonate. The weight of the calcium 
 oxide may also be verified by converting it into sulphate. A 
 few drops of sulphuric acid are added cautiously to the oxide. 
 The excess is evaporated off on the hot plate and the calcium 
 sulphate finally brought to constant weight by heating to red- 
 ness with the Bunsen burner. 
 
 EXERCISE 15. 
 Determination of Calcium in Calcium Carbonate. 
 
 Weigh out 0.5 gram of pure calcium carbonate. Transfer to a 500-c.c. 
 beaker, add a little water, and while covering the beaker with a watch- 
 crystal add 5 c.c. dilute hydrochloric acid. When the calcium is dissolved, 
 rinse the watch-crystal and the sides of the beaker, bringing the bulk of 
 the solution to about 250 c.c. Neutralize with filtered ammonia, heat 
 nearly to boiling on the hot plate, and add ammonium oxalate solution 
 slowly with vigorous stirring of the solution until no further precipitation 
 occurs. Keep the solution on the hot plate, so that it nearly boils, for several 
 hours. Decant the clear solution through the funnel, wash with hot water 
 by decantation several times, then transfer the precipitate to the paper 
 
68 DETERMINATION OF METALS. 
 
 and complete the washing with hot water, testing the wash-water for chlo- 
 rides. Transfer the moist precipitate to the weighed platinum crucible, dry, 
 and burn the paper in the usual manner, finally heating with the blast- lamp 
 for about ten minutes. Cool in a desiccator and weigh. Heat with the 
 blast-lamp again for ten minutes and weigh, first placing the weight found 
 necessary by the first weighing on the pan and finally taking the crucible 
 out of the desiccator and making the final adjustment as rapidly as possible. 
 Contjnue heating and weighing until constant weight is obtained. 
 
 CONVERT the OXIDE into the CARBONATE by 
 adding at least an equal bulk of powdered ammo- 
 nium carbonate * and moistening with water. Place 
 the cover on the crucible and evaporate off the 
 water by gentle heat, then place the crucible on 
 a pipe-stem triangle which is supported about 1 
 inch above a wire gauze which is heated by a 
 Bunsen burner. Repeat this treatment, cool in the 
 desiccator, and weigh. Continue this operation until 
 constant weight is obtained. This weight should 
 equal that of the carbonate weighed out for the 
 analysis. The theoretical percentage of calcium oxide 
 in calcium carbonate is 56. 
 
 DETERMINATION OF ZINC, MANGANESE, AND CADMIUM 
 BY PRECIPITATION WITH SODIUM CARBONATE. 
 
 The salts of these metals readily form double salts with ammo 
 nium compounds. They cannot, therefore, be completely pre- 
 cipitated by sodium carbonate in the presence of considerable 
 amounts of ammonium salts. The latter are readily removed 
 by evaporating the solution to dryness and volatilizing the ammo- 
 nium compound by gently heating the dry residue. 
 
 63. Cadmium Carbonate is almost insoluble in ammonium salts 
 and completely insoluble in the fixed alkali carbonates and in 
 water. On burning the filter-paper, the cadmium carbonate left 
 upon it is reduced to metallic cadmium, which is very volatile at 
 the temperature of the Bunsen burner. The precipitate should 
 therefore be dried and removed as completely as possible from 
 the paper. The loss of cadmium may be almost completely pre- 
 vented by saturating the paper with a strong solution of ammo- 
 nium nitrate, drying and igniting. The oxygen from the nitrate 
 serves to reoxidize any cadmium which may have been reduced. 
 
 * If on volatilizing one or two grams of the ammonium carbonate a weighable 
 residue is left, it must be purified by sublimation. 
 
DETERMINATION OF METALS AS OXIDE. 69 
 
 The paper during the ignition should also be held in the oxidizing 
 zone of the Bunsen burner, for the same reason. The error may 
 be still further reduced by first washing the paper with a little 
 dilute nitric acid and water, and evaporating the solution to 
 dryness in the weighed crucible. The paper may then be burned 
 as directed previously. The precipitate is converted by ignition 
 into oxide, which is not decomposed nor volatilized at white heat. 
 If the precipitate is small it need not be dried, but may be imme- 
 diately dissolved in nitric acid and the solution evaporated to 
 dryness in the porcelain crucible. 
 
 64. Manganess Carbonate is quite soluble in ammonium salts. 
 The recently precipitated carbonate is white, but on standing it 
 becomes brown, being oxidized by the oxygen of the air. It is 
 insoluble in solutions of potassium or sodium carbonate. On 
 ignition in the presence of air the carbonate is converted into 
 protosesquioxide of manganese which approximates the formula 
 Mn 3 4 . Pickering has shown that the percentage of manganese 
 in the ignited oxide varies from 69.69 to 75, according to the 
 temperature to which the oxide has been heated. The theoretical 
 percentage of manganese in Mn 3 4 is 72.05. Only small amounts 
 of manganese should be weighed in this form if accurate results 
 are desired. .No manganese is lost on burning the filter-paper. 
 If the precipitate is large and not removed from the paper some 
 difficulty will be found in securing complete combustion of the 
 paper. 
 
 65. Zinc is precipitated as the basic carbonate. Carbon diox- 
 ide is, therefore, liberated during the reaction. The bicarbonate 
 produced holds some of the zinc in solution, but this is completely 
 precipitated on boiling. It is, therefore, best to precipitate the 
 zinc from a boiling solution and keep it at this temperature for 
 some time. If even a small amount of ammonium salts are pres- 
 ent, the precipitation is not complete. As the boiling alkaline 
 solution acts on glass quite energetically, it is best to use a porce- 
 lain or platinum dish. The precipitate is washed with hot water. 
 On ignition, the carbonate is converted into oxide. In the pres- 
 ence of carbon or other reducing substances, both the oxide and 
 carbonate are reduced on ignition to metallic zinc which volatilizes. 
 The precipitate must, therefore, be treated in the same manner 
 
70 DETERMINATION OF METALS. 
 
 as the cadmium precipitate. The zinc oxide is stable when 
 ignited in the air in the absence of reducing agents. 
 
 EXERCISE 16. 
 
 Determination of Zinc in Zinc-ammonium Sulphate, 
 (NH 4 ) 2 SO 4 .ZnSO 4 .6H 2 0. 
 
 Weigh out 2 grams of crystallized zinc-ammonium sulphate. Transfer 
 to a platinum dish or crucible. Heat very gently with a small Bunsen burner 
 flame or on the hot plate until the water is expelled. Keep the dish covered 
 to prevent spattering. Increase the heat somewhat and volatilize the ammo- 
 nium sulphate. When no more fumes come off, allow to cool and dissolve 
 the zinc sulphate in water and a little hydrochloric acid if necessary. If a 
 crucible has been used, transfer the solution to a platinum or porcelain 
 dish, heat nearly to boiling and add with stirring a freshly made sodium 
 carbonate solution until the zinc is entirely precipitated. Continue the 
 heating for about fifteen minutes, allow to settle, wash several times by 
 decantation, finally transfer to the funnel and complete the washing with 
 hot water. Dry the precipitate, detach it from the paper and place on a 
 watch-crystal. Replace the paper in the funnel and wash with 25 c.c. of water 
 to which a little nitric acid has been added. Evaporate the wash-water to 
 dryness in the weighed platinum crucible. In the meantime moisten the 
 paper with a saturated solution of ammonium nitrate and dry in the steam- 
 bath. Burn the paper on the platinum wire, being careful to touch it only 
 with the oxidizing portion of a Bunsen-burner flame which is entirely free 
 from a luminous zone. The precipitate is now transferred to the crucible 
 and ignited with the Bunsen burner and finally with the blast-lamp. Theo- 
 retical percentage of zinc in zinc-ammonium sulphate is 16.28. 
 
 IGNITION OF SALTS OF VOLATILE ACIDS. 
 
 66. Determination of the Metals by Ignition of the Salts of 
 Volatile Acids. As has already been observed, it is possible to 
 ignite the salts of many metals and obtain oxides of definite com- 
 position, from the weight of which the percentage of the metal 
 present may be calculated. Calcium oxalate and carbonate, 
 cadmium, manganese, zinc, and lead carbonates are examples of 
 such salts which are obtained by precipitation. In many cases 
 the nitrates and nitrites may be treated in this manner, but not 
 the chlorides since they are quite volatile and the oxygen is not 
 present to unite with the metal to form the oxide. 
 
 In some cases the oxygen may be furnished and the chlorine 
 driven off, as a volatile compound by igniting the salt with mer- 
 
DETERMINATION OF METALS AS OXIDE. 71, 
 
 curie oxide, nitric acid, or ammonium carbonate or nitrate. Zinc 
 chloride may be converted into oxide by treatment with excess 
 of mercuric oxide or nitric acid. Magnesium chloride may be 
 converted into oxide by ignition with mercuric oxide, ammonium 
 carbonate, or nitrate. 
 
 In the absence of substances of this character, the acid with 
 which the metal is combined must contain oxygen and be readily 
 decomposed and volatilized by heat. The sulphates and phos- 
 phates, though containing the necessary oxygen, are usually very 
 stable, and the acid is with difficulty volatilized by heat. 
 
 The metals whose salts of volatile oxygen acids may be decom- 
 posed into oxides are LEAD, BISMUTH, COPPER, TIN, IRON, ALUMIN- 
 IUM, CHROMIUM, MANGANESE, NICKEL, ZINC, and MAGNESIUM. 
 
 No other volatile substance except the metal may be present. 
 
 67. Ignition of Salts of Organic Acids. Many of the metallic 
 salts of organic acids may be ignited so as to leave the metal in a 
 weighable form. The carbon, hydrogen, nitrogen, and some of 
 the oxygen of the organic compound are either volatilized in the 
 free condition or combined with each other, or they unite with 
 the oxygen of the air to form simple oxygen compounds. Most 
 metals retain at least some of the oxygen of the organic com- 
 pound, or are oxidized by the oxygen of the air. Under these 
 conditions BARIUM, STRONTIUM, and SODIUM remain as CARBONATES 
 with a partial reduction to oxide, which is converted to carbonate 
 by treatment with ammonium carbonate, so that these metals 
 are weighed as carbonates. Sodium compounds must be heated 
 only to very low redness, because of the volatility of this metal. 
 It is best not to attempt the complete combustion of the carbon 
 by heat. The charred mass should be extracted with water and 
 filtered, the solution of the sodium carbonate being evaporated 
 to dry ness in a weighed dish. 
 
 ORGANIC COMPOUNDS of CALCIUM, MAGNESIUM, ALUMINIUM, 
 
 CHROMIUM, MANGANESE, NICKEL, COPPER, LEAD, and BISMUTH 
 
 may be ignited in the air until the organic matter is burned off 
 and the oxide of the metal remains. Many organic salts froth on 
 being heated. The crucible should therefore have a tight lid, and 
 the heat should be applied gently at first. If this confinement is 
 not sufficient, the crucible should be set in a larger covered cruci- 
 
72 DETERMINATION OF METALS. 
 
 ble, since the error due to loss of metal by the mechanical carry- 
 ing away of fine particles by the products of the combustion may 
 be considerable. The ignition of the salts containing lead or 
 bismuth must be conducted at as low a heat as possible, as these 
 metals are readily reduced and are volatile at red heat. When 
 the organic material is charred and the carbon ceases to burn 
 readily the flame is removed from the crucible and ammonium 
 nitrate added. The crucible is covered and gently heated. The 
 ammonium nitrate fuses and oxidizes the remaining carbon and 
 converts the metallic lead or bismuth into nitrate which is con- 
 verted into oxide by red heat. Concentrated nitric acid may be 
 used for the same purpose. When copper salts are ignited, some 
 of the copper is left as cuprous oxide, which it is most advanta- 
 geous to oxidize by igniting with mercuric oxide. 
 
 EXERCISE 17. 
 Determination of Lead by Ignition of Lead Nitrate, Pb(N0 3 ) 2 . 
 
 Heat a porcelain, crucible to redness with the Bunsen burner for ten 
 minutes. Cool in the desiccator and weigh. Weigh out one-half gram of 
 lead nitrate and transfer to the crucible. If the salt is in large crystals, 
 it should be reduced to a coarse powder by grinding in a clean and dry 
 porcelain mortar. It should be weighed out as soon as powdered. Heat 
 gently, at first with the lid on the crucible. When decrepitation ceases 
 and copious nitrous fumes cease to be evolved the lid may be removed. 
 Heat to dull redness until no more fumes are evolved. Cool in the desiccator 
 and weigh as PbO. Repeat until the weight is constant. Calculate the 
 percentage of lead in the lead nitrate. 
 
CHAPTER VI. 
 
 DETERMINATION OF METALS AS SULPHATE 
 AND AS SULPHIDE. 
 
 A LARGE number of the metals form sulphates which are of 
 definite composition and which can be heated high enough to 
 expel all water and volatile matter without suffering decomposi- 
 tion. Most of the sulphates, however, are soluble. Only three, 
 those of BARIUM, STRONTIUM, and LEAD, are insoluble enough for 
 separation of the metal from water solution by precipitation. 
 
 68. Barium Sulphate is the most insoluble, requiring about 
 400,000 parts of water to dissolve one part of the salt. It is 
 somewhat more soluble in even dilute solutions of hydrochloric 
 acid and of ammonium chloride. It readily dissolves in concen- 
 trated sulphuric acid, especially when hot, from which it is com- 
 pletely precipitated by the addition of water. Barium sulphate 
 has the very peculiar though by no means uncommon property 
 of rendering insoluble some very soluble salts, notably the nitrates, 
 chlorides, sulphates, and chlorates of the alkaline earths and 
 alkali metals, ferric, aluminium, and chromic salts. Potassium 
 salts give more trouble than sodium salts. Once carried down 
 by the barium sulphate, these salts cannot be entirely washed 
 out. The solution must be freed from nitrates and chlorates by 
 evaporation with hydrochloric acid. If ferric salts are reduced 
 to ferrous salts, iron will not be carried down by the barium sul- 
 phate. A precipitate contaminated by any of these impurities 
 may be dissolved in concentrated sulphuric acid and reprecipi- 
 tated by diluting with much water. It may also be fused with 
 sodium carbonate, which converts the sulphate of barium into 
 the carbonate, which may be filtered off, washed, redissolved, and 
 reprecipitated with sulphuric acid. 
 
 It has been shown that the contamination of the precipitate 
 
 73 
 
74 DETERMINATION OF METALS. 
 
 takes place at the moment of its formation. The most favorable 
 conditions for avoiding the contamination have been found to be a 
 boiling-hot, slightly acid solution of large bulk, and the addition 
 of the precipitant in very dilute solution drop by drop, and with the 
 most vigorous stirring. If from failure to observe any of these 
 conditions salts are carried down, washing will not remove them. 
 On account of its slight solubility, barium sulphate tends to come 
 down in such a finely divided condition that it passes through the 
 pores of the filter-paper. By precipitating as directed above 
 and keeping the solution near the boiling-point for some time 
 the precipitate settles out, leaving a clear solution and does not 
 pass through the filter-paper. 
 
 On burning the paper some of the sulphate is reduced to sul- 
 phide of barium. It is customary to add a drop or two of sul- 
 phuric acid in order to reconvert the sulphide into sulphate, and 
 then to ignite the precipitate and volatilize the excess of sulphuric 
 acid. By this treatment barium pyrosulphate is produced, 
 which is decomposed slowly and only at a high temperature. It 
 has been shown that the addition of sulphuric acid is unnecessary 
 since the barium sulphide is oxidized to sulphate by ignition in 
 the air. 
 
 69. Strontium Sulphate is much more soluble in water and 
 acids than barium sulphate, 1 part of the former being dissolved 
 by about 7000 parts of cold water, and 10,000 parts of boiling 
 water, while if the water contains sulphuric acid, from 11,000 to 
 12,000 parts are required. One part of the precipitate is dis- 
 solved by about 500 parts of dilute nitric or hydrochloric acid, 
 while it is practically insoluble in 95% or absolute alcohol and in 
 boiling ammonium sulphate solution (1 to 4). The precipitate 
 should therefore be washed by one of the two latter solutions, or 
 by dilute sulphuric acid. On ignition in presence of carbon or 
 reducing gases it is converted into sulphide. This is volatile in 
 the Bunsen burner flame, coloring it red, so that the precipitate 
 must be detached as completely as possible from the paper. The 
 subsequent treatment is similar to that of barium sulphate. 
 
 70. Lead Sulphate is quite insoluble in water, requiring for 
 its solution about 23,000 parts of cold water, but if the water con- 
 tains sulphuric acid 36,500 parts are required. The presence of 
 
DETERMINATION OF METALS AS SULPHATE. 75 
 
 alcohol reduces this solubility. Solutions of ammonium salts, 
 especially the nitrate, tartrate, and acetate dissolve lead sulphate 
 quite readily. The two latter should be rendered strongly alka- 
 line with ammonia. It is also dissolved by hot solutions of caus- 
 tic soda or potash. Dilute nitric and hydrochloric acids dissolve 
 considerable quantities of lead sulphate which is reprecipitated on 
 the addition of dilute sulphuric acid. These acids are readily 
 removed by evaporating the solution after the addition of sul- 
 phuric acid, until fumes of the latter appear. On diluting the 
 solution with considerable water and adding alcohol, the lead is 
 completely precipitated. 
 
 The lead precipitate should be washed by alcohol or dilute 
 sulphuric acid, preferably by the former as the paper would char 
 on drying if sulphuric acid were left in it, and if washed out by 
 water some lead would be lost. On burning the paper some of 
 the lead sulphate is reduced to metallic lead. The precipitate 
 should therefore be removed from the paper, the latter placed in 
 a porcelain crucible, and ignited until all carbon is burned. A 
 drop or two of nitric acid and the same amount of sulphuric acid 
 is added and evaporated off, care being taken to avoid spattering. 
 The remainder of the precipitate should then be added and heated 
 with a Bunsen burner. Sulphate of lead may very advantage- 
 ously be filtered off on a Gooch crucible and dried on the hot plate. 
 
 EXERCISE 18. 
 Determination of Barium in Barium Chloride, BaCl 2 .2H 2 0. 
 
 Weigh out about one-half gram of crystallized barium chloride. Transfer 
 to a 500-c.c. beaker and dissolve in about 250 c.c. of distilled water. Heat 
 nearly to boiling on the hot plate and add, while stirring vigorously, dilute 
 sulphuric acid drop by drop until no further precipitate is produced. The 
 acid may be drawn up in a glass tube or pipette and the flow of acid regu- 
 lated by pressing the index finger against the upper end of the tube. After 
 allowing the precipitate to settle a little, it can easily be seen if a drop 
 of acid produces any precipitation. The heating of the solution is con- 
 tinued until the precipitate has settled completely, leaving a clear solution. 
 About an hour is usually sufficient. It is washed several times by decanta- 
 tion, using about 75 c.c. of hot water each time and stirring up the precipi- 
 tate thoroughly. The precipitate is transferred to the funnel and the 
 washing with hot water continued until the wash-water is free from chlorides. 
 
73 DETERMINATION OF METALS. 
 
 The moist precipitate with the paper may be transferred to a weighed 
 platinum crucible or dish and ignited over the Bunsen burner with free 
 access of air and finally over the blast-lamp. The theoretical percentage 
 of barium in barium chloride is 56.25. 
 
 71. Determination of Metals by Evaporation with Sulphuric 
 Acid. Many metals form sulphates which are stable enough to 
 be ignited and weighed, but which are soluble and therefore can- 
 not be precipitated and washed. POTASSIUM, SODIUM, CALCIUM, 
 
 MAGNESIUM, COBALT, NICKEL, and MANGANESE form SUCh SUl- 
 
 phates. A salt or compound of one of these metals which con- 
 tains no other non-volatile metal and which contains only volatile 
 acids may be analyzed by ignition with excess of sulphuric acid. 
 The sulphate of the metal will remain, from the weight of which 
 the percentage of the metal present may be calculated. In this 
 manner the salts of nitric and nitrous acids, the halogen acids, 
 organic acids, etc., may be analyzed. 
 
 ORGANIC SALTS of SODIUM and POTASSIUM are treated with a 
 moderate excess of dilute sulphuric acid in a covered platinum 
 crucible and gently heated to dull redness. The evolution of 
 white fumes indicates an excess of sulphuric acid. The readily 
 fusible acid sulphate of the alkali remains. Small lumps of 
 ammonium carbonate which leave no residue on ignition are 
 thrown into the hot crucible and the heating continued. This is 
 repeated until the alkali sulphate no longer fuses. The crucible 
 is cooled and weighed, and the addition of ammonium carbonate 
 and the heating repeated until constant weight is obtained. The 
 alkali-acid sulphate may be converted into the neutral sulphate 
 by the application of a white heat, but some loss of the metal 
 results. 
 
 BARIUM, STRONTIUM, and CALCIUM may be determined in 
 salts of volatile acids by the addition of an excess of sulphuric 
 acid, evaporating to dryness and igniting, at first gently and 
 finally to full redness. To remove the excess of sulphuric acid, 
 a little ammonium carbonate is sprinkled on the precipitate, 
 which is again heated to redness. MAGNESIUM may be deter- 
 mined in the same manner, but a large excess of sulphuric acid 
 should be avoided and the residue may be exposed to a moderate 
 
DETERMINATION OF METALS AS SULPHATE. 77 
 
 red heat only or loss of sulphuric acid will result. Magnesium 
 sulphate is very hygroscopic and must be weighed rapidly. 
 
 NICKEL, COBALT, and MANGANESE may be weighed as sul- 
 phates, but care must be exercised not to overheat, as sulphuric 
 acid may be lost. If this has occurred, as shown by blackening 
 of the material, dilute sulphuric acid must be added, and after 
 evaporating off the excess the ignition is repeated. The platinum 
 dish must not be heated above dull redness. 
 
 LEAD may be determined in organic salts, and those of volatile 
 acids by treatment in a weighed porcelain dish or crucible with 
 excess of dilute sulphuric acid. Loss by spattering is avoided by 
 keeping the vessel well covered and evaporating with very gentle 
 heat. If the residue from the ignition of an organic compound is 
 not perfectly white, the addition of sulphuric acid and the ignition 
 must be repeated. Lead sulphate is stable when exposed to a 
 red heat, but reducing gases must not be allowed to come in con- 
 tact with it. 
 
 EXERCISE 19. 
 Determination of Magnesium in Magnesium Carbonate. 
 
 Weigh one-half gram of magnesium carbonate and transfer to a weighed 
 platinum crucible. To prevent loss by spattering expel most of the car- 
 bonic acid by heating the crucible at first gently, then strongly with the 
 Bunsen burner, the whole operation taking about ten minutes. Allow 
 the crucible to cool quite completely and then add about 3 c.c. dilute sul- 
 phuric acid, at first cautiously, to prevent spattering in case much carbonic 
 acid stiil remains. Evaporate the solution by first heating on the hot 
 plate and then very gently with the Bunsen burner until acid fumes cease 
 to be evolved. If white fumes do not escape enough acid has not been 
 added. Heat for a few minutes to duil redness, cool in the desiccator, and 
 weigh quickly, as magnesium sulphate is hygroscopic. Repeat the ignition 
 with the Bunsen burner until the weight is constant. The residue should 
 dissolve completely in water and should form a neutral solution, otherwise 
 it has been too highly heated. 
 
 Calculate the percentage of magnesium carbonate in the sample tested. 
 Ordinary magnesium carbonate approximates the formula 
 
 4MgC0 3 .Mg(OH) 2 .5H 2 0. 
 
78 DETERMINATION OF METALS. 
 
 DETERMINATION OF METALS AS SULPHIDE. 
 
 A large number of the metals form sulphides which are almost 
 entirely insoluble in water, especially when it contains an excess 
 of the precipitant. The separation of metals as sulphides from 
 their solution is therefore very complete, and as in some case* 
 the precipitates can be ignited as sulphides, giving very pure 
 products, the determination of the metal as sulphide is in many 
 cases of the highest accuracy. This is true of COPPER, ZINC, and 
 
 MANGANESE. 
 
 COPPER, MERCURY, LEAD, SILVER, BISMUTH, CADMIUM, ARSENIC, 
 
 ANTIMONY, and TIN may be precipitated by hydrogen sulphide 
 from acid solutions. SILVER, CADMIUM, LEAD, and BISMUTH may 
 also be precipitated by ammonium sulphide in alkaline solution, 
 though the precipitation by hydrogen sulphide in moderately 
 acid solution is generally used. LEAD and SILVER may also be 
 precipitated by hydrogen sulphide or ammonium sulphide in 
 neutral solution. CADMIUM may be precipitated from a potas- 
 sium cyanide solution by hydrogen sulphide. 
 
 72. Copper may be precipitated from a neutral or slightly 
 acid solution by ammonium sulphocyanate after reduction by 
 sulphurous acid, ammonium sulphite, or hypophosphorous acid. 
 The copper is precipitated as cuprous sulphocyanate and may be 
 dried at 100 and weighed as such, or it may be converted into 
 cuprous sulphide by ignition with sulphur in a stream of hydrogen. 
 Many copper compounds containing no non-volatile constituent 
 except copper may be determined by heating a weighed portion 
 with sulphur in a stream of hydrogen, thus converting the copper 
 into cuprous sulphide. The oxides of copper, the sulphate, nitrate, 
 etc., may be analyzed in this manner. If the copper solution is 
 free from oxidizing substances, it is advisable to heat it nearly to 
 boiling before passing hydrogen sulphide, as under these condi- 
 tions the precipitate collects and settles better. 
 
 73. Washing. Some of the sulphides form so-called colloidal 
 solutions with water, while others are readily oxidized, forming 
 sulphates which dissolve in the wash-water. For this reason 
 the wash-water must contain hydrogen sulphide, and in the case 
 
DETERMINATION OF METALS AS SULPHIDE. 79 
 
 of cadmium and arsenic hydrochloric acid should be present, more 
 being necessary for arsenic sulphide than for cadmium sulphide. 
 
 74. Removal of Free Sulphur and Drying at 100. The sul- 
 phides of SILVER, MERCURY, CADMIUM, and ARSENIC cannot be 
 ignited, but must be dried at 100 before weighing. A Gooch 
 crucible is most advantageous for this purpose, though a weighed 
 filter-paper may be used. As hydrogen sulphide is very readily 
 oxidized even by the oxygen of the air, the precipitated sulphides 
 will almost invariably contain sulphur. When the precipitate 
 can be ignited, this contaminant does not interfere with the accuracy 
 of the determination. When the sulphide is dried at 100 the 
 removal of the free sulphur presents some difficulties. The solu- 
 tion from which the metal is to be precipitated should be freed as 
 far as possible from oxidizing agents. The precipitation should 
 be carried out in a cold solution, which should be protected from 
 the air as much as possible by placing it in an Erlenmeyer flask 
 and filtering and washing immediately after precipitation. Even 
 under these circumstances more or less sulphur is apt to be formed, 
 so that the precipitate should always be treated for the removal 
 of this element. Carbon disulphide and a moderately strong 
 solution of sodium sulphite have been used for this purpose. If 
 carbon disulphide is used, the washed precipitate is first dried, 
 then digested with small quantities of carbon disulphide until 
 no sulphur is obtained by evaporating a portion of the carbon 
 disulphide. If a Gooch crucible is used, it may be placed in a 
 small beaker and the carbon disulphide poured over the precipi- 
 tate. The precipitate may also be freed from water by washing 
 with alcohol, after which the carbon disulphide may be added. 
 If the sodium sulphite solution is used, the precipitate need not be 
 dried, but must be washed thoroughly after the sulphur has been 
 removed. The sodium sulphite solution must be heated very 
 nearly to the boiling-point. The solution may be tested for 
 sulphur by acidifying and allowing to stand for a few moments. 
 A milky precipitate indicates sulphur. The removal of the sul- 
 phur by either of these solutions is incomplete. 
 
 On drying at 100 BISMUTH SULPHIDE first loses water and then 
 begins to oxidize and gain in weight. This precipitate should 
 therefore be weighed after drying for half an hour. The drying 
 
80 DETERMINATION OF METALS 
 
 for periods of half an hour and weighing is continued until the 
 weight is constant or begins to increase. The lowest weight 
 found is that of the dry sulphide. 
 
 75. The Sulphides of Tin cannot be weighed as such, but are con- 
 verted into the dioxide by ignition in the air. The heat must be 
 applied gently at first or some of the tin will be volatilized as sul- 
 phide. The stannic sulphide is more volatile than the stannous 
 sulphide. When sulphur dioxide is no longer given off copiously 
 the precipitate is heated more intensely. To completely expel 
 the oxides of sulphur a little powdered ammonium carbonate is 
 sprinkled over the precipitate, which is again ignited. 
 
 Antimony sulphide is ignited in the same manner and converted 
 into the tetroxide Sb 2 4 . 
 
 EXERCISE 20. 
 Determination of Copper in Copper Sulphate, CuS0 4 .sH 2 O. 
 
 Weigh out 1 gram of pure crystallized copper sulphate. Transfer to a 3CKV 
 c.c. beaker and dissolve in about 200 c.c. of water. Add a few cubic centimeters 
 of hydrochloric acid, heat nearly to boiling, and pass hydrogen sulphide until 
 precipitation is complete. Allow the precipitate to settle for a few minutes, 
 decant the solution through the filter, and transfer the precipitate. Wash 
 with hydrogen-sulphide water. Before discarding any portion of the 
 filtrate, test for copper with hydrogen sulphide. Copper sulphide is very 
 readily oxidized by the oxygen of the air to copper sulphate, which passes 
 through the filter-paper. The hydrogen sulphide of the wash-water is 
 present to reprecipitate this dissolved copper. The precipitate should be 
 exposed to the air as little as possible. For this purpose it should be 
 washed down into the point of the funnel and kept well covered with the 
 wash-water containing hydrogen sulphide. 
 
 76. Ignition of the Precipitate with Sulphur in a Stream of Hydrogen. 
 The precipitate is dried in the steam-oven, detached from the paper and 
 placed in a weighed Rose crucible. The paper is then burned and the 
 ash added to the precipitate. If the paper crumbles considerably, it may 
 'be placed in a porcelain dish or crucible and burned by heating the dish or 
 crucible. Unburned particles of carbon sometimes remain unless the 
 dish is heated strongly for some time. The residue is transferred to the 
 crucible, the contents of which are mixed with an equal bulk of powdered 
 sulphur which leaves no residue on ignition. A gentle stream of dry hydro- 
 gen is now led in through the porcelain tube which passes through the lid 
 of the crucible. The hydrogen is freed from arsine by passing through a 
 solution of mercuric chloride. The arrangement of the apparatus is shown 
 in Fig. 12, page 63. The hydrogen must be allowed to pass until the air 
 has been displaced from the wash-bottles as well as from the crucible. 
 
DETERMINATION OF METALS AS SULPHIDE. 81 
 
 About ten minutes is generally sufficient if the stream of hydrogen passes 
 quite rapidly. The danger of explosion is greatly reduced if a plug of 
 glass wool or cotton is loosely inserted in the glass tube nearest to the 
 crucible. The crucible is at first heated gently with the Bunsen burner, 
 and when the excess of sulphur is nearly all volatilized it is heated to redness 
 with the full flame of the burner. After a few minutes' strong heating 
 the burner is removed and the crucible is allowed to become very nearly 
 cold in the stream of hydrogen. It is then placed in the desiccator and 
 after a few minutes is weighed. 
 
 The precipitate is again mixed with sulphur, heated in the stream of 
 .hydrogen and weighed. This is repeated until constant weight is ob- 
 tained. The percentage of copper in crystallized copper sulphate is 25.47. 
 The ignited precipitate is cuprous sulphide, Cu 2 S. 
 
 EXERCISE 21. 
 Determination of Arsenic in Arsenious Oxide, As 2 0g. 
 
 Weigh out about 0.3 gram of pure resublimed arsenous oxide. Transfer to 
 a small beaker and dissolve in a little freshly made dilute solution of caustic 
 soda. Transfer the solution to a 250-c.c. Erlenmeyer flask. Acidify with 
 hydrochloric acid and then add about 30 c.c. of the dilute acid. Dilute 
 to about 200 c.c. or until the flask is nearly full and pass hydrogen sulphide 
 until precipitation is complete. Filter immediately through a Gooch crucible 
 which has been dried at 100 to constant weight. Wash with about 250 c.c. 
 of water containing a few cubic centimeters of dilute hydrochloric acid* 
 and a considerable amount of hydrogen sulphide. Suck the water out as com- 
 pletely as possible with the filter-pump, wash two or three times with alco- 
 hol, place the crucible in a small beaker and pour in carbon disulphide until 
 the precipitate is covered. The carbon disulphide used for this purpose 
 should leave no residue on evaporating a few cubic centimeters on a watch- 
 crystal. If a residue is left, the carbon disulphide must be redistilled. 
 Allow to stand about fifteen minutes. Pour off the carbon disulphide and 
 extract again with a fresh portion. Continue this operation until a portion 
 of the carbon disulphide evaporated on a watch-crystal leaves no deposit 
 of sulphur. Dry the precipitate in the steam-oven and weigh. Theoreti- 
 cal percentage of arsenic in arsenious oxide is 75.75. 
 
 77. Precipitation of Zinc as Sulphide. Zinc as well as man- 
 ganese and iron are precipitated by ammonium sulphide in alka- 
 line solution. Zinc may be precipitated by hydrogen sulphide in 
 a solution which is only very slightly acid with a mineral acid or 
 moderately acid with acetic acid. This sulphide is so finely 
 divided that it passes through the filter-paper, making it very 
 
 * The acid prevents the formation of a "colloidal " solution 
 
82 DETERMINATION OF METALS. 
 
 difficult to filter and weigh. If some mercuric chloride is added 
 to the zinc solution, a precipitate is obtained which can be filtered 
 and washed with ease. In the subsequent ignition the mercury 
 is completely volatilized. The zinc sulphide obtained by adding 
 ammonium sulphide to a solution of zinc made strongly alkaline 
 with caustic soda is also easily washed, but it generally is con- 
 taminated with alkali and silica. 
 
 78. The Two Sulphides of Manganese. Manganese forms two 
 sulphides, one of which is green, while the other is pink. The 
 pink sulphide resembles the sulphide of zinc in being so finely 
 divided that filtration is difficult. The 'green sulphide on the 
 other hand offers no difficulty of this kind in manipulation. It is, 
 therefore, highly desirable to obtain this variety in determining 
 manganese. This is accomplished by placing an excess of color- 
 less ammonium sulphide together with some ammonium chloride 
 in an Erlenmeyer flask, diluting to about 100 c.c. and heating 
 nearly to boiling. The concentrated Lnd hot manganese solution 
 is then added all at once, and the flask vigorously shaken. The 
 solution is kept hot and occasionally vigorously agitated, until 
 the sulphide changes from pink to green and settles readily, leav- 
 ing a clear supernatant liquid. Failure to obtain this result is 
 generally due to the presence of an insufficient amount of ammo- 
 nium sulphide. The precipitate is filtered immediately, and 
 washed with water containing ammonium sulphide and ammonium 
 chloride. 
 
 79. Precipitation of Iron as Sulphide. Iron is precipitated as 
 sulphide from solutions containing organic matter, which prevents 
 the complete precipitation of the metal in the usual manner as 
 hydroxide. Ammonium chloride must be present, and either 
 the yellow or colorless ammonium sulphide may be used for pre- 
 cipitation. The solution must stand in a warm place until the 
 precipitate has completely settled, leaving a clear supernatant 
 fluid. Ammonium chloride and sulphide must be present in the 
 wash- water, as the sulphide oxidizes to sulphate, which passes 
 through the filter-paper unless it is reprecipitated by the ammo- 
 nium sulphide. The bulk of the dried precipitate should be 
 removed from the paper before burning the latter, though there 
 is no danger of losing iron. The oxide formed during the com- 
 
DETERMINATION OF METALS AS SULPHIDE. 83 
 
 bustion of the paper is reconverted to sulphide by ignition in 
 hydrogen after the addition of sulphur. 
 
 EXERCISE 22. 
 Determination of Manganese in Potassium Permanganate, KMn0 4 . 
 
 Weigh out one-half gram of pure potassium permanganate. Transfer 
 to a small beaker, add a few cubic centimeters of cone, hydrochloric acid, 
 cover with a watch-crystal, and evaporate to dryness on the water-bath. 
 Dissolve in a few c.c. of water and neutralize the solution with ammonia. 
 
 Prepare some colorless ammonium sulphide by adding 100 c.c. of water 
 to 50 c.c. concentrated ammonia (sp. gr. 0.90) , saturating half of this solu- 
 tion with hydrogen sulphide and then adding the other half of the solution. 
 In an Erlenmeyer flask of about 250 c.c. capacity is placed 25 c.c. of the 
 ammonium sulphide solution, 10 c.c. of a five times normal solution of 
 ammonium chloride or about 3 grams of the dry salt. The solution is 
 diluted to about 100 c.c. and warmed on the hot plate or with a Bunsen 
 burner. As soon as it comes to a boil the hot manganese solution is added, 
 the beaker being rinsed with a little water. The flask is shaken vigorously 
 and the solution kept nearly at the boiling-point. After alternate shaking 
 and heating for a few minutes the manganese sulphide which came down 
 pink turns green and settles readily, leaving a clear supernatant liquid. 
 Failure to secure this result is due to the absence of a sufficient amount of 
 ammonium sulphide, the reagent being weak or the salt was boiled away 
 before adding the manganese solution. If the addition of more ammonium 
 sulphide and heating fails to produce the green precipitate, the operation 
 should be repeated, using more ammonium sulphide. 
 
 The green sulphide is allowed to settle for half an hour, the solution being 
 kept warm. The precipitate is then filtered off and washed with water con- 
 taining ammonium sulphide and ammonium chloride. It is dried and de- 
 tached from the paper, which is burned, the ash being placed on top of the 
 precipitate in a weighed porcelain crucible. The precipitate is mixed with 
 two or three times its bulk of sulphur which is free from non-volatile matter. 
 It is ignited in a stream of hydrogen which has been passed through a solu- 
 tion of mercuric chloride to free it from arsine and then dried over calcium 
 chloride or concentrated sulphuric acid. The crucible is heated, at first 
 gently, until the sulphur begins to burn, and then with the full flame of a 
 Bunsen burner, giving a flame at least 15 cm. long when burnmg free. 
 The crucible is allowed to become nearly cold in the stream of hydrogen. 
 If brown particles of oxides of manganese are still visible, the precipitate 
 should be again ignited after mixing with sulphur, and after cooling in 
 the stream of hydrogen is transferred to the desiccator, completely cooled 
 and weighed. No further change in weight should be observed on again 
 heating with sulphur in the stream of hydrogen. 
 
CHAPTER VII. 
 
 DETERMINATION OF METALS AS PHOSPHATE, ARSE- 
 NATE, CHROMATE, AND CHLORIDE. 
 
 MANGANESE, ZINC, and MAGNESIUM may be precipitated and 
 weighed as phosphates. Although most of the metals form phos- 
 phates which are quite insoluble, only those of the three metals 
 mentioned are sufficiently insoluble or stable on ignition for 
 quantitative determination. Phosphoric acid is one of the most 
 non-volatile acids, but on account of its weak affinity and ten- 
 dency to form acid salts it is difficult to procure a precipitate of 
 constant composition. It may also be reduced more or less com- 
 pletely during the combustion of the filter-paper. 
 
 80. Precipitation of Manganese and Zinc. With zinc and 
 manganese an amorphous precipitate is first formed. This is 
 the neutral metallic phosphate M 3 (P0 4 ) 2 ,* and must be converted 
 into the double ammonium phosphate MNH 4 P0 4 . This is ac- 
 complished by heating the amorphous precipitate with the solu- 
 tion containing a large amount of ammonium chloride. 20 grams 
 of the dry salt should be present for 0.2 gram of the metal. 5 to 
 10 c.c. of a cold saturated solution of microcosmic salt is then 
 added and the solution diluted to 200 c.c. The solution is then 
 cautiously neutralized with ammonia, and on the first appearance 
 of a precipitate it is heated nearly to boiling and vigorously stirred 
 until the amorphous phosphate has become crystalline. Ammo- 
 nia is added drop by drop and the stirring continued after each 
 addition of ammonia until the precipitate is crystalline. The 
 addition of ammonia and the stirring is continued until no fur- 
 ther precipitate forms. With manganese a large excess of ammo- 
 nia should be avoided. The zinc solution should be made exactly 
 neutral. The heating of the solution must be continued during 
 the addition of ammonia until the phosphate is completely pre- 
 
 * In this formula M may be either Mn or Zn. 
 
 84 
 
DETERMINATION OF METALS AS PHOSPHATE. 85 
 
 cipitated and converted into the crystalline form. A platinum 
 dish is most suitable for this purpose. A porcelain dish is prefer- 
 able to a beaker, as the latter is attacked by the hot alkaline solu- 
 tion, contaminating the precipitate with silica. The solution 
 should contain no other metals than the alkalies. The precipitate 
 is washed with water containing ammonium nitrate. 
 
 81. Magnesium forms a crystalline precipitate much more read- 
 ily than zinc and manganese. The precipitation is carried out 
 in a cold solution which must be allowed to stand for at least 
 twelve hours in order to secure complete precipitation. If the 
 solution is stirred or vigorously shaken, the precipitation is com- 
 plete in an hour or two. If the amount of ammonium salts pres- 
 ent is considerable, the magnesium is precipitated more slowly, 
 requiring at least twenty-four hours' standing. In this case the 
 composition of the precipitate is also different, containing rela- 
 tively more ammonia. Such a precipitate is not suitable for 
 weighing, but must be dissolved in the least amount of hydro- 
 chloric acid and reprecipitated by the addition of a little alkali- 
 phosphate solution and ammonia drop by drop with vigorous 
 stirring. The volume of the solution should be kept small on 
 account of the solubility of the precipitate. It should not exceed 
 100 c.c. 
 
 82. Removal of Ammonium Salts. If one does not wish to 
 make two precipitations in this manner, the ammonium salts 
 must first be removed. A very convenient method of accom- 
 plishing this object is that suggested by J. L. Smith. The solu- 
 tion is evaporated in a casserole or porcelain dish over the free 
 flame to a sirupy consistency. Dilute nitric acid is then added 
 in small portions, and the heating continued until nitrous fumes 
 are no longer evolved. A little hydrochloric acid is then added 
 in small portions and the heating continued until the excess of 
 nitric acid is decomposed and chlorine is no longer evolved. 
 
 The ammonium salts may also be removed by evaporating 
 the solution to dryness in a porcelain dish and gently heating 
 the residue. The magnesium remains as oxide and is dissolved 
 in hydrochloric acid. The solution is then made alkaline with 
 ammonia. If a precipitate of magnesium hydroxide is formed, 
 the solution should be acidified with hydrochloric acid and 
 
86 DETERMINATION OF METALS. 
 
 again made alkaline with ammonia. This operation should &e 
 repeated until no precipitate occurs on making the solution 
 alkaline. Finally the solution is acidified and a solution of di- 
 sodium phosphate added in sufficient amount to precipitate all of 
 the magnesium. It is then made alkaline by the addition of 
 ammonia with constant stirring until no further precipitate is 
 formed. The solution must be allowed to stand twelve hours 
 before filtering off the precipitate. A few drops of the clear 
 liquid should give a distinct test for phosphoric acid with mag- 
 nesia mixture, otherwise enough of the precipitant has not been 
 added. The precipitate is washed with dilute ammonia, one part 
 of strong ammonia to ten parts of water. 
 
 83. Ignition of Phosphate Precipitates. All of the phosphate 
 precipitates lose ammonia and water on ignition and are con- 
 verted into pyrophosphates according to the equation 
 
 2NH 4 MP0 4 =2NH 3 +11,0 +Mff> r 
 
 If too much ammonium salts have been present in the solution, the 
 precipitate will contain more ammonia than corresponds to the 
 formula NH 4 MP0 4 . Some metaphosphate of the metal, M(P0 3 ) 2 , 
 will be formed on the ignition of such a precipitate. The meta- 
 phosphate loses phosphorous pentoxide, and is converted into 
 pyrophosphate only after very prolonged and intense ignition. 
 Such a precipitate could be considered constant only when it does 
 not lose weight after a half-hour's ignition with the blast-lamp. 
 For this reason the excess of ammonium salts is removed before 
 precipitation, as already directed. The evolution of ammonia 
 and water is quite rapid if the precipitate is heated strongly, 
 resulting hi appreciable loss of material due to spattering. The 
 precipitate should therefore at first be heated very gently with 
 the Bunsen burner with the lid on the crucible. When no more 
 ammonia is evolved the crucible may be heated with the blast- 
 lamp. 
 
 As the phosphates fuse readily, the complete combustion of the 
 filter-paper is secured with some difficulty. It is best to dry the 
 precipitate and remove it as completely as possible from the 
 paper, and burn the latter on the platinum wire, using the oxid- 
 
DETERMINATION OF METALS AS PHOSPHATE. 87 
 
 izing flame of the Bunsen burner to assist the combustion. If 
 the ash which falls into the crucible contains carbon which cannot 
 be burned by simple ignition, it may be moistened with a drop or 
 two of concentrated nitric acid which is volatilized by gentle heat 
 and the ignition continued. The difficulty experienced in burn- 
 ing the carbon is frequently due to insufficient washing of the 
 precipitate, so that it still contains sodium acid phosphate. 
 
 If the platinum crucible is used for the ignition of phosphates, 
 it is sometimes damaged by the reduction of the phosphates and 
 the union of the phosphorus with the platinum. The likelihood 
 of damage seems to be much reduced if the moist precipitate and 
 paper are introduced into the crucible and the ignition conducted 
 by gradually bringing the crucible to redness with the Bunsen 
 burner. The reducing gases from the flame must be absolutely 
 excluded. Ignition of the precipitate in a porcelain crucible is 
 entirely satisfactory. 
 
 EXERCISE 23. 
 Determination of Magnesium in Magnesium Sulphate, MgS0 4 .7H 2 O. 
 
 Weigh out 1 gram of the pure recrystallized salt. Transfer to a small 
 beaker and dissolve in 50 c.c. of water. Add 5 c.c. dilute hydrochloric 
 acid and 15 c.c. of disodium phosphate solution. Add dilute filtered ammo- 
 nia slowly with vigorous stirring, being careful to avoid rubbing the sides 
 of the beaker with the stirring-rod, as very firmly adhering crystals would 
 form where the beaker is rubbed. Add a few c.c of ammonia in excess, 
 allow it to stand for twelve hours, then transfer to a funnel containing a 
 small paper and wash with dilute filtered ammonia (1 part of strong ammonia 
 to 10 parts of water). The filtrate may be tested for chlorides by acidify- 
 ing with nitric acid and adding a drop of silver nitrate solution. As the 
 ammonia frequently contains chlorides, the washing should be continued 
 until the filtrate gives no more turbidity with silver nitrate than the dilute 
 ammonia. 
 
 The precipitate is dried in the steam-oven, removed from the paper 
 quite completely by means of a brush and transferred to a watch-crystal 
 The paper is burned on the platinum wire, heating the charred portions 
 from tune to time in the oxidizing flame of the Bunsen burner. The ash 
 is allowed to drop into a weighed porcelain crucible. If unburned carbon 
 still remains, the open crucible should be placed on its side and heated 
 with the Bunsen burner. If necessary, the combustion is completed by 
 allowing the crucible to cool, adding a few drops of concentrated nitric 
 
88 DETERMINATION OF METALS. 
 
 acid and placing the lid on the crucible and heating gently until the nitric 
 acid is volatilized. When the carbon has been . completely burned the 
 main portion of the precipitate is added and the covered crucible very 
 gently heated until no more ammonia fumes escape. It is then heated with 
 the full flame of the Bunsen burner and finally for five or ten minutes with 
 the blast-lamp. It is then weighed and again heated with the blast-lamp 
 and weighed. This is repeated until constant weight is obtained. Theo- 
 retical percentage of magnesium in crystallized magnesium sulphate is 9.88. 
 
 84. Determination of Arsenic as Magnesium Pyroarsenate. 
 When present in its higher state of oxidation arsenic may be 
 precipitated as magnesium ammonium arsenate, which is ignited 
 and weighed as magnesium pyroarsenate, Mg 2 As 2 7 . Solution 
 of compounds of arsenic in hydrochloric acid and potassium 
 chlorate, aqua regia, or strong nitric acid converts the arsenic 
 into arsenic acid.* The solution is filtered if necessary, evapo- 
 rated to a small bulk, neutralized with ammonia, and magnesia 
 mixture added with stirring. Alcohol is added to the extent of 
 one-fourth of the volume of the solution. After standing twelve 
 or, still better, as long as forty-eight hours the precipitate is 
 filtered off and washed with a mixture of 2 volumes of strong 
 ammonia, 2 volumes of water, and 1 volume of alcohol. 
 
 The precipitate is dried, detached from the paper and placed 
 on a watch-crystal. The paper is returned to the funnel and the 
 remaining portions of the precipitate dissolved in hot dilute nitric 
 acid, and the paper washed with hot water. If it is not large, the 
 entire precipitate may be dissolved in nitric acid. The solution 
 is evaporated to dryness in a weighed porcelain crucible and the 
 remainder of the precipitate added. With the lid on, the crucible 
 is heated very gently until the ammonia is expelled. There is 
 great danger of loss of arsenic before the ammonia is expelled 
 because of its reducing action on the arsenic. The presence of 
 nitric acid tends to prevent this, while by heating in a stream of 
 oxygen the time required to expel the ammonia may be reduced 
 to ten minutes. Moistening the precipitate with nitric acid is 
 almost as efficient. When the ammonia is completely expelled 
 the crucible is heated to redness with the Bunsen burner. The 
 
 * Note for student: Write the equations for these reactions. 
 
DETERMINATION OF METALS AS CHLORIDE. 89 
 
 precipitate is then magnesium pyroarsenate, Mg 2 As 2 7 . The pre- 
 cipitate may also be collected on a Gooch crucible and ignited 
 with the precautions given. 
 
 85. Determination of Lead, Barium, and Chromium as Chro- 
 mates. Lead and barium form very insoluble chromates which 
 are of definite composition and may be completely dried without 
 suffering decomposition. These precipitates may therefore serve 
 to determine not only lead and barium, but also chromium. The 
 precipitation is carried out in a solution acid with acetic acid. 
 In determinations of lead and barium, potassium dichromate is 
 used as the precipitant, wiiile chromic acid is precipitated by lead 
 acetate or barium chloride. Solutions acid with hydrochloric or 
 nitric acid are rendered acid with acetic acid by addition of sodium 
 acetate. The precipitate is washed with water, filtered on a 
 Gooch crucible, and dried on the hot plate, or it may be gently 
 ignited with the Bunsen burner, care being taken not to allow it 
 to fuse. 
 
 86. Determination of Potassium as Platino-chloride. Potas- 
 sium may be precipitated as potassium platino-chloride, collected 
 on a Gooch crucible which has been dried at 100 and weighed, 
 or it may be collected on a filter-paper, washed with 80% alcohol, 
 dissolved in hot water, and the solution evaporated to dryness on 
 the water-bath in a dish and weighed. The precipitate may 
 also be decomposed by ignition with a reducing agent, such as 
 oxalic acid, or in a stream of hydrogen. The potassium then 
 remains as chloride and is washed out of the reduced spongy 
 platinum by hot water. The platinum may then be gently ignited 
 and weighed. Potassium platino-chloride is soluble in 100 parts 
 of cold water and 20 parts of hot water, in 40,000 parts of absolute 
 alcohol and 25,000 parts of 80% alcohol. The presence of sodium 
 platino-chloride and excess of the precipitant, platinum tetra- 
 chloride, or, more correctly, chlorplatinic acid, decreases the solu- 
 bility. As the precipitate is washed with alcohol, salts insoluble in 
 this liquid must be absent. Practically the solution may contain 
 only chlorides of sodium and potassium but sufficient platino- 
 chloride solution must be added to convert both of these salts 
 into the double platinum salt. 
 
 87. Lindo-Gladding Method. If the Lindo-Gladding method 
 
90 DETERMINATION OF METALS. 
 
 of determining potassium is used, sulphates may be present, but 
 the heavy metals should be absent. If chlorides are not present, 
 sodium chloride should be added, and after the addition of an 
 excess of platino-chloride solution the solution is evaporated to a 
 sirupy consistency. The sodium platino-chloride and the excess 
 of the precipitant is washed out by means of 80% alcohol. The 
 sulphates are washed out with a solution of ammonium chloride 
 made by dissolving 100 grams in 500 c.c. of water and shaking 
 up the solution with 10 grams of potassium platino-chloride. The 
 ammonium chloride is now washed out with 80% alcohol. While 
 the Lindo-Gladding method has been extensively used by Ameri- 
 can chemists, the German workers still prefer to remove sulphates 
 by precipitation with barium chloride. 
 
 88. Determination of Silver as Chloride. Silver as well as 
 mercury and bismuth may be precipitated as chloride and weighed 
 as such. The neutral or slightly acid solution of silver is precipi- 
 tated by the addition of hydrochloric acid or sodium chloride. It 
 should be heated nearly to boiling before precipitation and the 
 chloride added while stirring the solution vigorously. For this 
 purpose it is convenient to place the solution in an Erlenmeyer 
 flask. The addition of a considerable excess of chloride should be 
 avoided, as silver chloride is appreciably soluble both in hydro- 
 chloric acid and in sodium chloride. After each addition of the 
 reagent, and vigorous shaking or stirring of the solution, the pre- 
 cipitate should be allowed to settle and a drop of chloride added to 
 the clear liquid. If a precipitate appears, more of the precipitant 
 must be added. When the silver has been completely precipi- 
 tated the solution should be heated on the hot plate or the water- 
 bath with occasional stirring until the solution is clear. The pre- 
 cipitate is filtered off on a Gooch crucible which has been dried on 
 the hot plate or in an air-bath at a temperature above 120. Silver 
 chloride loses chlorine when exposed to light, acquiring a blue 
 color. It should therefore be protected as much as possible during 
 the manipulation from the action of strong light. 
 
 It may also be filtered off on paper and washed, as already de- 
 scribed. The precipitate is dried and removed from the paper 
 as completely as possible and placed on a watch-crystal. The 
 paper is now burned in a weighed porcelain crucible. The silver 
 
DETERMINATION OF METALS AS CHLORIDE. 91 
 
 chloride left on the paper is reduced to metallic silver. It is 
 reconverted to chloride by the addition of a drop of nitric acid 
 and one of hydrochloric and evaporating to dryness. As silver 
 chloride is quite volatile, the amount left on the paper should be 
 very small. The main portion of the precipitate is now trans- 
 ferred to the crucible and dried by gently heating with the Bunsen 
 burner. The precipitate must not be allowed to fuse. The Bun- 
 sen burner, having a small flame, should be held in the hand, and 
 immediately removed when the precipitate begins to fuse around 
 the edge. The results are very accurate if the determination is 
 properly carried out. 
 
 89. Washing the Precipitate without Filtration. The silver 
 chloride may also be washed and transferred to a crucible without 
 
 FIG. 14. 
 
 the use of filter-paper. The precipitation is carried out in an 
 Erlenmeyer flask. When all of the silver has been precipitated 
 the flask is shaken vigorously, until the liquid is perfectly clear 
 and the precipitate has collected into a compact mass. The 
 liquid is carefully decanted, and the precipitate washed by decan- 
 tation. The flask is then filled with water, and a weighed porce- 
 lain crucible inverted over the mouth of the flask. By quickly 
 inverting the flask the precipitate falls into the crucible, which 
 is placed on the desk while the flask is held upright with the 
 mouth in 1 the bottom of the crucible. By gently tapping the 
 flask any adhering particles of the precipitate fall into the crucible. 
 By tilting the flask slightly, bubbles of air enter and the water 
 flows quietly into the crucible. When the latter is full the flask 
 may be removed by quickly passing it to one side of the crucible. 
 
92 DETERMINATION OF METALS. 
 
 The water in the crucible is decanted as much as possible, the 
 portion remaining being evaporated on the water-bath. The 
 precipitate is then heated in the usual manner and weighed. 
 
 90. Determination of Mercury as Mercurous Chloride. For 
 precipitation as chloride, mercury must be in the mercurous con- 
 dition. The cold and highly dilute solution is mixed with sodium 
 chloride solution until precipitation is complete. After allowing 
 the precipitate to settle, filter on a Gooch crucible dried at 100. 
 Wash with cold water. Weigh after drying at 100. If the solu- 
 tion contains much nitric acid it should be nearly neutralized 
 with -sodium carbonate before precipitation of the mercury. If 
 the metal is present in the solution as mercuric mercury, it 
 may be reduced by means of phosphorous acid. Hydrochloric 
 acid together with excess of phosphorous acid is added to the 
 cold solution, which is allowed to stand for twelve hours cold or 
 heated not higher than 60. The precipitate of mercurous chlo- 
 ride is filtered off, washed, dried, and weighed as already directed. 
 
 91. Determination of Bismuth as Oxychloride. Bismuth is 
 precipitated and weighed as the oxychloride, BiOCl. If the 
 solution is strongly acid, potassium, sodium, or ammonium hydrox- 
 ides are added until not over J per cent of free acid remains. If 
 chlorides are not present, ammonium chloride is added and the 
 solution is then largely diluted by the addition of water. After 
 standing some time water is added to a portion of the clear liquid. 
 If a precipitate forms, more water must be added to the entire 
 solution. When all of the bismuth has been precipitated the 
 oxychloride is filtered off on a Gooch crucible which has been 
 dried at 100. It is washed with water containing a few drops 
 of hydrochloric acid, dried at 100, and weighed. If sulphuric 
 or phosphorous acids are present, small quantities of these acids 
 are apt to contaminate the precipitate. The solution should 
 contain only nitric and hydrochloric acids. Because of the ten- 
 dency of the precipitate to lose chlorine, the results are not quite 
 as reliable as when the bismuth is weighed as oxide. A little 
 hydrochloric acid is added to the wash-water to prevent this loss. 
 
DETERMINATION OF METALS AS CHLORIDE. 93 
 
 EXERCISE 24. 
 Determination of Silver in Silver Nitrate, AgNO,. 
 
 Prepare a Gooch crucible with a layer of asbestos about cm. thick. 
 Wash until no fine particles of asbestos come through with the wash-water, 
 Dry on the hot plate or in an air-bath at 120 to 150 and weigh. Usually 
 two hours' drying is sufficient. 
 
 Weigh out ^ gram of silver nitrate and transfer to a 250-c.c. Erlen- 
 meyer flask. Dissolve in 100 to 150 c.c. of water and heat nearly to 
 boiling. Add dilute hydrochloric acid, a few drops at a time, shaking the 
 flask vigorously after each addition. To avoid adding a large excess of 
 the acid notice after each addition if a precipitate is produced. When 
 sufficient hydrochloric acid has been added the flask should be wanned 
 and vigorously shaken until the curdy precipitate is well collected and the 
 solution is very nearly clear. Decant the clear liquid on the weighed 
 Gooch crucible, wash two or three times by decantation, using hot water 
 which contains a few drops of concentrated nitric acid. Finally transfer the 
 precipitate to the crucible, using the " policeman " to clean the flask if 
 necessary. Dry the Gooch crucible as before and weigh. The silver chlo- 
 ride should be exposed to the light as little as possible. The theoretical 
 percentage of silver in silver nitrate is 63.50. 
 
 REVIEW. At this point the student is advised to review the deter- 
 minations of the metals by tabulating the methods as follows: In the first 
 column write the name or symbol of each metal studied. In the next 
 column headed "Precipitated as" write the formulas of the precipitates 
 in which each element may be separated from solution. In the next 
 column headed "Remarks, sources of error, etc.," indicate briefly the 
 special precautions to be observed in each precipitation. In the next 
 column headed " Weighed as " give the formulas of the dried or ignited 
 precipitates. In the next column headed "Methods of drying" state 
 whether the precipitate is dried at 100 on the hot-plate or by ignition. 
 
DETERMINATION OF ACIDS. 
 
 CHAPTER VIII. 
 
 DETERMINATION OF THE HALOGENS, SUL- 
 PHUR, AND NITROGEN. 
 
 THE gravimetric determination of acids is more difficult than 
 the determination of metals because the acid radicals are as a rule 
 more volatile and unstable than the metallic compounds. In 
 the determination of metals various acid radicals have formed a 
 part of precipitates which can be washed and weighed. Among 
 these are the carbonates of the alkaline-earth metals; the sul- 
 phates of lead and barium, the sulphides of mercury, cadmium, 
 silver, and arsenic, the phosphates of magnesium, zinc, and manga- 
 nese, the chromates of barium and lead, and the chlorides of silver, 
 mercury, and bismuth. In a good many cases it is found possible 
 to use a given precipitate for the determination of either the metal 
 or the acid radical. 
 
 DETERMINATION OF THE HALOGENS. 
 
 92. Precipitation as the Silver Salt. Where a given acid 
 forms a weighable precipitate with several metals, choice is made 
 of the most insoluble and most stable precipitate. The CHLORIDE 
 OF SILVER has been found to be the most advantageous form in 
 which to precipitate and weigh either silver or the chlorine radical. 
 In the determination of a chloride or of hydrochloric acid the 
 process is very nearly identical with the determination of silver 
 as given on page 90. The chloride solution must be slightly 
 acid. For this purpose the neutral or alkaline solution is acidified 
 with nitric acid, adding a drop or two in excess to the neutral 
 solution. The solution is heated to about 60 and silver nitrate 
 solution added with constant stirring until no more precipitate is 
 formed. Th solution is then heated nearly to boiling and stirred 
 vigorously until the precipitate coagulates, leaving a clear superna- 
 
 94 
 
DETERMINATION OF THE HALOGENS. 95 
 
 tant liquid. Failure to secure this result may be due to the absence 
 of an excess of silver nitrate. The precipitate is filtered off on a 
 weighed Gooch crucible and washed with hot water containing 
 a little silver nitrate. The presence of silver nitrate reduces very 
 much the solubility of silver chloride in water. When the other 
 impurities have been washed out the silver nitrate is removed 
 by washing with as little distilled water as possible. The pre- 
 cipitate is then dried on the hot plate, or in an air-bath at about 
 150. Silver chloride should be protected from the light as much 
 as possible, as it turns dark, losing chlorine. The method gives 
 very accurate results. IODIDES and BROMIDES as well as the 
 free acids may be determined in the same manner. 
 
 93. Ignition of Paper. If a Gooch crucible is not at hand, the 
 silver halides may be filtered off on paper, washed in the manner 
 already described, and dried. The precipitate is then separated 
 from the paper as completely as possible and placed on a watch- 
 crystal. The paper is burned on the platinum wire or in a weighed 
 porcelain crucible. The SILVER CHLORIDE or BROMIDE remaining 
 on the paper is reduced to metallic silver. The ash of the paper 
 in the crucible is treated with a drop or two of nitric acid and a 
 drop of hydrochloric or hydrobromic acid to reconvert the silver 
 into the proper halide salt. The excess of acid is evaporated off 
 by gentle heat and the remainder of the precipitate added. The 
 crucible is now heated very gently with the Bunsen burner, care 
 being taken not to fuse the precipitate, which volatilizes slightly 
 when fused. The SILVER IODIDE which remains on the filter-paper 
 is not reduced when the latter is burned, but on account of its 
 volatility it should be removed as completely as possible. The 
 silver precipitate may also be washed without filtration as described 
 in Chapter VII, page 91. 
 
 94. Determination of the Halogens in Metallic Salts. The 
 chlorine in MERCURIC CHLORIDE, STANNOUS CHLORIDE, PLATINIC 
 
 CHLORIDE, the CHLORIDE of ANTIMONY, and the GREEN CHLORIDE 
 
 of CHROMIUM cannot be determined by precipitation with silver 
 nitrate. The precipitate is contaminated in the case of the stannous 
 salt with stannic oxide and silver oxide, in the case of the mer- 
 curic and antimony salts with metallic mercury or antimony. 
 All of the chlorine is not precipitated from the chromium solution. 
 
96 DETERMINATION OF ACIDS. 
 
 Platinous chloride contaminates the precipitate of silver chloride 
 from platinic chloride. The tin in stannous chloride is therefore 
 precipitated by boiling the concentrated neutral solution with 
 ammonium nitrate. The mercury and antimony are precipitctul 
 with hydrogen sulphide, tartaric acid having been added to the anti- 
 mony solution. The chlorine in platinic chloride may be determined 
 by fusing the salt with sodium carbonate, filtering off the platinum 
 from the solution of the fusion and determining the chlorine in 
 the filtrate. The chromium in the green chloride is precipitated 
 by means of ammonia, filtered off and washed, and the chlorine 
 precipitated with silver nitrate in the filtrate. The corresponding 
 salts of BROMINE and IODINE are treated in a similar manner. 
 
 The INSOLUBLE CHLORIDES, of LEAD, SILVER, and MERCURY 
 
 must be decomposed by alkalies for the separation of the metal. 
 Silver chloride may be fused in a porcelain crucible with sodium 
 and potassium carbonates. Lead chloride may be treated in 
 the same manner or digested with alkali bicarbonates and water, 
 while mercurous chloride may be decomposed by digestion with a 
 solution of sodium or potassium hydroxide. The INSOLUBLE 
 BROMIDES and IODIDES may be treated in a similar manner. 
 
 SEPARATION OF CHLORINE, BROMINE, AND IODINE. 
 
 95. Separation of Iodine as Fallacious Iodide from Chlorine and 
 Bromine. Very excellent methods have recently been worked 
 out for the separation of chlorine, bromine, and iodine. Iodine 
 may be precipitated either as palladous or thallous iodide from 
 solutions which contain chlorine or bromine. For the precipita- 
 tion as palladous iodide the solution is made slightly acid with 
 nitric acid and palladous nitrate added in slight excess. The 
 solution is allowed to stand for twenty-four to forty-eight hours. 
 The brownish-black precipitate of palladous iodide, PdI 2 , is fil- 
 tered off on a Gooch crucible or on a tared paper, washed with 
 warm water, and dried at 100 to constant weight. The paper 
 may also be burned, and the precipitate ignited in a Rose crucible 
 in a stream of hydrogen. It is then reduced to metallic palla- 
 dium, and may be weighed as such. 
 
 If only hydrobromic or hydrochloric acid is present in the 
 filtrate, it may be precipitated with silver nitrate, after removal 
 of the excess of palladium with hydrogen sulphide. The excess 
 
SEPARATION OF THE HALOGENS. 97 
 
 of hydrogen sulphide must be removed by means of ferric sul- 
 phate or hydrogen peroxide, and the precipitated sulphur filtered 
 off after digesting some time on the water-bath. If sufficient 
 material is at hand, the iodine may be determined by means of 
 palladous nitrate or chloride in one portion, and the iodide 
 together with the chloride or bromide precipitated and weighed as 
 the silver salt in another equal portion. The weight of silver 
 iodide is then calculated from the weight of the palladium pre- 
 cipitate and subtracted from the weight of the combined silver 
 precipitate. 
 
 96. Separation of Iodine from Chlorine as Thallous Iodide. 
 Iodine may also be separated from chlorine by precipitation as 
 thallous iodide, according to Jannasch and Aschoff. Both thallous 
 iodide and chloride are slightly soluble in cold water, the iodide 
 less than the chloride, however. The presence of ammonium 
 sulphate increases the solubility of the chloride and decreases the 
 solubility of the iodide, which is almost absolutely insoluble if 
 alcohol is present. To the solution of the chloride and iodide, 
 which should have a volume of about 50 c.c., the same volume of 
 20% ammonium sulphate solution is added and 30 c.c. of alcohol. 
 A 4% solution of thallous sulphate is now added until the iodide 
 is precipitated. The solution is gently warmed, well stirred, and 
 allowed to stand for twelve hours. The precipitate of thallous 
 iodide, Til, is filtered off on a Gooch crucible, and washed with a 
 mixture of 5 parts ammonium sulphate, 70 parts of water, and 30 
 parts of alcohol. It is dried at 100. After expelling the alcohol 
 from the filtrate the chlorine may be precipitated with silver 
 nitrate after diluting considerably to prevent the precipitation of 
 silver sulphate. For the same reason the solution should be 
 digested hot for some time. 
 
 97. Separation of Chlorine, Bromine, and Iodine. Jannasch 
 and Aschoff have also worked out a method of separating the 
 three halogens which depends on the fact that iodine is liberated 
 from hydriodic acid by nitrous acid, while neither hydrochloric 
 nor hydrobromic acid is oxidized under these conditions. The 
 liberated iodine may be removed from the solution by means of 
 carbon disulphide, or it may be volatilized with steam after dilut- 
 ing the solution so that neither hydrobromic nor hydrochloric 
 
98 
 
 DETERMINATION OF ACIDS. 
 
 acid is expelled. Ferric sulphate and a number of other reagents 
 have been found to liberate iodine and not chlorine or bromine. 
 The hydrobromic acid may be oxidized by potassium permangan- 
 ate and acetic acid, and the bromine driven out of the solution by 
 means of steam. The chlorine may then be precipitated with sil- 
 ver nitrate. 
 
 A suitable arrangement of apparatus is shown in Fig. 15. The 
 ground-glass joint of the tubes entering the flask K is very essen- 
 
 FIG. 15. 
 
 tiaJ, as neither cork nor rubber is unacted on by the halogens. 
 The most satisfactory substitute for a ground-glass joint is a 
 cork stopper coated with paraffine. The stoppers in the absorp- 
 tion-flask as well as the Peligot tube are of this character. The 
 distilling-flask K should have a capacity of at least 1 liter, and 
 the absorption-flask a capacity of 500 c.c. W is a tin or metallic* 
 vessel for generating steam. 
 
 The neutral solution of the halogens is placed in the flask K] 
 diluted to about 750 c.c. with water, 5 c.c. of dilute sulphuric 
 acid added, as well as 1 gram of sodium nitrite dissolved in a little 
 water. The flask is closed immediately. The absorption-flask 
 should contain a mixture of 50 c.c. of caustic soda * solution and 
 
 * If caustic soda free from the halogens is not at hand, it may be made by dis- 
 solving metallic sodium in distilled water. 
 
SEPARATION OF THE HALOGENS. 99 
 
 an equal volume of hydrogen peroxide solution. The bulbs of 
 the Peligot tube should be nearly filled with a similar solution, 
 and a little dilute caustic soda solution should be placed in the 
 Erlenmeyer flask. 
 
 98. Distillation of the Iodine. On heating the contents of the 
 flask K to boiling and passing a current of steam, the iodine is 
 driven over into the absorption-flask, which is kept cold by means 
 of the cold water in the beaker. The distillation is continued 
 until the solution in K is colorless, which requires about fifteen 
 minutes. All of the iodine will then be driven out. On being 
 absorbed by the caustic soda, part of the iodine is converted into 
 iodate according to the equation 
 
 GNaOH +3I 2 =NaI0 3 +5NaI +3H 2 0. 
 
 The hydrogen peroxide present converts the iodate into iodide: 
 NaI0 3 +3H 2 2 =NaI +3H 2 +30 2 . 
 
 99. Determination of the Iodine as Silver Iodide. While the 
 steam is still passing, the delivery-tube is withdrawn from the 
 absorption-flask and rinsed with a little water. The contents of 
 the flask as well as the Peligot tube are transferred to a porcelain 
 dish, 50 c.c. of hydrogen peroxide added, and the solution warmed 
 for some time on the water-bath. Silver nitrate is added until a 
 permanent precipitate of silver hydroxide is formed. On digesting 
 the precipitate the brown silver hydroxide is converted into the 
 yellow iodide. If the brown color disappears entirely, more silver 
 nitrate must be added. Finally the solution is acidified with 
 nitric acid, warmed, and the silver iodide filtered off on a Gooch 
 crucible, washed with hot water, dried on the hot plate or in the 
 air-oven at 150 and weighed. 
 
 100. Distillation of the Bromine. The contents of the dis- 
 tilling flask are made faintly alkaline with caustic soda and con- 
 centrated to a volume of 500 c.c. The solution is cooled, 60 c.c. 
 of 33% acetic acid is added, and 1 to 1J grams of potassium per- 
 manganate dissolved in a little water. The absorption flask and 
 Peligot tube are filled, and the distillation conducted exactly as 
 directed for the iodine. The bromine is expelled from the solution 
 with more difficulty than in the case of the iodine, 45 to 75 minutes 
 
100 DETERMINATION OF ACIDS. 
 
 being generally required. The distillation must be continued foi 
 some time after bromine vapors are no longer visible. 
 
 The solution of the bromine is transferred to a porcelain dish, 
 50 c.c. of hydrogen peroxide solution added, and warmed on the 
 water-bath. The bromine is then precipitated, washed, and 
 weighed in the same manner as the iodine. 
 
 10 1. Determination of the Chlorine. The potassium per- 
 manganate in the distilling-flask is decomposed by neutralizing 
 the solution with caustic soda, adding a little methyl alcohol and 
 warming. The manganese dioxide is filtered off and washed with 
 warm water. The chlorine in the filtrate is precipitated in the 
 usual manner. 
 
 102. Determination of Chloric Acid. Chloric acid may be 
 determined by precipitation with silver nitrate after reduction 
 with pure zinc and sulphuric acid. 
 
 103. Determination of Hydrocyanic Acid. Hydrocyanic acid 
 may be precipitated and weighed as silver cyanide. The solution 
 should be dilute and slightly acid with nitric acid. Excess of 
 silver nitrate is added to the cold solution with vigorous stirring. 
 The precipitate may be collected on a weighed filter-paper or, still 
 better, on a Gooch crucible and dried at 100. It may also be col- 
 lected on an unweighed filter-paper and converted into metallic 
 silver by simple ignition until the weight is constant. This opera- 
 tion should be carried out in a porcelain crucible. Heating to 
 redness for one-quarter hour is generally sufficient to completely 
 decompose the cyanide. 
 
 104. Separation of Hydrocyanic Acid from Chlorides, Bromides, 
 and Iodides. If chlorides, bromides, or iodides are present in the 
 solution of the cyanide, the precipitated silver salt of the halogen 
 must be separated from the silver cyanide. For this purpose 
 the precipitate is digested with a solution of mercuric oxide in 
 dilute acetic acid. The silver cyanide is converted into soluble 
 mercuric cyanide and silver acetate: 
 
 2AgCN +Hg(C 2 H 3 2 ) 2 =Hg(CN) 2 +2AgC 2 H 3 2 . 
 
 A solution of mercuric oxide in 100 c.c. of water and 5 c.c. dilute 
 acetic acid is sufficient to dissolve the silver cyanide from 0. 1 gram 
 of potassium cyanide. The halogen salts of silver are not decom- 
 
DETERMINATION OF SULPHUR. 101 
 
 posed by the mercury solution, which is filtered off and the dis- 
 solved silver precipitated as chloride after the addition of nitric 
 acid. The silver chloride must be ignited in a stream of hydrogen 
 in a porcelain crucible. Some mercury is carried down with the 
 silver and cannot be entirely removed by drying the silver chloride. 
 The presence of the halogens do not interfere in the volumetric 
 determination of cyanogen. 
 
 DETERMINATION OF SULPHUR. 
 
 105. Determination of Sulphur by Fusion with Alkali Carbon- 
 ates and Nitrates. Sulphur in sulphides and in most of its com- 
 pounds is usually converted into sulphuric acid and precipitated 
 and weighed as barium sulphate. Many methods of oxidation 
 have been proposed for converting the various sulphur compounds 
 into sulphuric acid or soluble sulphates. Perhaps the most gen- 
 erally applicable method is fusion with sodium carbonate and 
 potassium nitrate. The finely powdered and weighed material 
 is intimately mixed with six parts of dry sodium carbonate and 
 four parts of potassium nitrate. The proportion of potassium 
 nitrate should be reduced if the percentage of sulphur is small, as 
 when much nitrate is present the action on the platinum crucible 
 is considerable. This mixture is transferred to the crucible, and 
 gradually heated until fused. The illuminating-gas used for the 
 Bunsen burner frequently contains sulphur, which may be ab- 
 sorbed by the fusion mixture. To prevent this it is advisable to 
 place the crucible in a hole cut in a piece of asbestos board.* The 
 crucible should not be heated higher than necessary to keep the 
 contents fused. When cool, dissolve the residue in hot water, 
 filter, and wash with water containing a little sodium carbonate. 
 Acidify the filtrate with hydrochloric acid and evaporate to dry- 
 ness in a porcelain dish. Add a little dilute hydrochloric acid 
 and evaporate to dryness again to completely remove the nitric 
 acid. Dissolve the chlorides in water and a drop or two of dilute 
 hydrochloric acid. Filter and wash the paper free from chlorides. 
 The filtrate should have a bulk of about 250 c.c. Heat to boiling, 
 add barium chloride solution slowly with constant stirring, digest 
 
 * As alcohol is free from sulphur the crucible is frequently heated with a 
 spirit lamp. 
 
102 DETERMINATION OF ACIDS. 
 
 hot until the solution is clear; filter, wash free from chlorides, 
 ignite and weigh. 
 
 This method has been largely used for determining sulphur hi 
 sulphides, either minerals or laboratory products. If the sul- 
 phides lose sulphur on heating, the fusion mixture should consist 
 of 4 parts of sodium carbonate, 8 parts of potassium nitrate, and 
 24 parts of pure and perfectly dry sodium chloride. As the 
 reagents used may contain sulphur, a blank determination is nec- 
 essary in careful work. For this purpose the fusion and precipi- 
 tation must be carried out exactly as in the actual determination. 
 The amount of barium sulphate obtained in this manner is deducted 
 from that obtained in the analysis of the unknown. 
 
 106. Fusion of Sulphur Compounds with Sodium Peroxide. 
 Sulphur may also be oxidized to sulphuric acid by fusion with 
 sodium peroxide. The finely ground material is mixed with five 
 or six times its weight of powdered sodium peroxide, and the 
 mixture fused in a nickel or copper crucible. The crucible must 
 be covered with a closely fitting lid, and must be very cautiously 
 heated, as the action is quite violent in the beginning. The flame 
 of the Bunsen burner must be small and not touching the crucible. 
 After a few minutes the heat is increased just sufficiently to keep 
 the contents of the crucible in the state of fusion, which is main- 
 tained for a few minutes. The crucible is then allowed to cool, 
 placed on its side in a beaker, and water added from a wash-bottle 
 while a watch-crystal is held over the beaker. When the fused 
 mass is dissolved the crucible and lid are taken out with a clean 
 pair of tongs and washed with water. The solution is filtered, 
 and the residue washed with water containing a little sodium 
 carbonate. If the residue is small, and especially if it is free from 
 iron, this filtration may be omitted. The filtrate is acidified with 
 hydrochloric acid, and if silica was present in the original material 
 the solution is evaporated to dryness finally on the water-bath, 
 where it is heated for half an hour. The residue is treated with 
 water and a few drops of hydrochloric acid and filtered. The 
 sulphuric acid in the filtrate is precipitated and weighed as barium 
 sulphate. As the sodium peroxide is not always free from sulphur, 
 a blank determination must be made as in the preceding method 
 
 107. Oxidation of Sulphur Compounds with Fuming Nitric 
 Acid. Digestion with red fuming nitric acid is one of the best of 
 
DETERMINATION OF SULPHUR. 103 
 
 the methods in which a liquid oxidizing substance is used. Most 
 sulphides and sulphur compounds may be oxidized in an Erlen- 
 meyer flask in the neck of which a small funnel is placed to pre- 
 vent the too rapid escape of the fumes. With substances which 
 are readily decomposed, giving off sulphuretted hydrogen, it is 
 advisable to conduct the operation in a glass-stoppered bottle. 
 A small, flat-bottomed glass tube or weighing-bottle, which may 
 easily be inserted into the bottle, is provided. The material to 
 be analyzed is weighed out and placed in this small tube, which is 
 then lowered into the bottle so as not to spill the contents. The 
 fuming nitric t acid is now added from a graduate. 30 c.c. is 
 sufficient for quantities of material which will yield not more 
 than 1 gram of barium sulphate. The stopper is then inserted in 
 the bottle and held firmly with the hand, while the bottle is tilted 
 so that the acid comes in contact with the sulphide. When the 
 first violent reaction has ceased shake the bottle a little, cool the 
 bottle and contents by holding it under a tap of running water 
 for a few minutes, and then cautiously remove the stopper so as 
 to relieve the pressure without loss of material by spattering. 
 The stopper is replaced in a slanting position to allow exit for the 
 fumes, and the bottle is placed on the water-bath. If after half 
 an hour's digestion particles of sulphur remain floating on the 
 liquid, strong hydrochloric acid, potassium chlorate, or liquid 
 bromine may be added in small portions, the potassium chlorate 
 and liquid bromine being the most efficient. Care should be 
 taken to prevent the fuming nitric acid or bromine from coming 
 in contact with the hands, as serious burns are very quickly 
 produced. The digestion on the water-bath is then continued 
 until the sulphur is oxidized. 
 
 108. Decomposition of Insoluble Sulphates. The complete 
 disintegration of the material is frequently indicated by a change 
 in color or physical appearance. The nitrates of the metals are 
 generally insoluble in the concentrated acid. Insoluble sulphates, 
 of such metals as barium and lead, may also be produced. In the 
 case of minerals, gangue, quartz, etc., will be left. The concentrated 
 acid is diluted with water and digested for a few minutes to dis- 
 solve soluble salts. If BARIUM is present, the insoluble material 
 must be filtered off, washed thoroughly, and fused with five parts 
 of mixed sodium and potassium carbonates. The melt is dis- 
 
104 DETERMINATION OF ACIDS. 
 
 solved in water, and the insoluble carbonates filtered off and thor- 
 oughly washed. The filtrate is acidified with hydrochloric acid 
 and added to the nitric acid solution. If LEAD is present, the sul- 
 phate of this metal may be decomposed by digesting with a solu- 
 tion of acid sodium or potassium carbonate, filtering, washing, 
 and treating the filtrate as before. 
 
 109. Removal of the Nitric Acid. The nitric acid solution, 
 now containing all of the sulphur as sulphuric acid, is evaporated 
 to dryness after the addition of a little sodium chloride, if alkalies 
 have not already been introduced. The evaporation is repeated 
 after the addition of dilute hydrochloric acid, until- the nitric acid 
 is entirely removed. The residue is dissolved in water, a few 
 drops of hydrochloric acid are added, the solution is filtered if 
 necessary, and the sulphuric acid determined as barium sulphate. 
 
 no. Other Oxidizing Solutions Used for Sulphur Compounds. 
 The digestion with LIQUID BROMINE or AQUA REGIA made by mixing 
 one part of strong hydrochloric acid with three parts of strong 
 nitric acid is carried out in a similar manner. The bromine may 
 either be added gradually to the material mixed with a little 
 nitric acid or to the dry material. These methods are largely used 
 for the determination of sulphur in pyrites or crude sulphur. 
 (See Exercise 36, p. 172.) The dry material may also be mixed 
 with powdered POTASSIUM CHLORATE in an Erlenmeyer flask, 
 and concentrated hydrochloric acid added in small portions. 
 Finally the flask is gently heated on the water-bath. The sub- 
 sequent treatment is identical with that given after treatment 
 with nitric acid. A very powerful oxidizing agent consists of a 
 saturated solution of potassium chlorate in concentrated nitric 
 acid. 
 
 in. Oxidation of Sulphur Compounds by Means of Chlorine. 
 If lead is present, the oxidation of the sulphur by means of chlo- 
 rine in alkaline solution is advantageous, because the lead is pre- 
 cipitated as peroxide. Iron is also precipitated as ferric hydrox- 
 ide. On passing the chlorine for a considerable time, however, 
 the iron passes into solution as ferrate, producing a red tint. Heat- 
 ing the liquid for a few minutes with powdered quartz serves to 
 precipitate the iron. The finely powdered SULPHIDE or CRUDE 
 SULPHUR is heated for some time with a dilute solution of caustic 
 potash which is free from sulphuric acid. Free sulphur as well as 
 
DETERMINATION OF NITROGEN. 105 
 
 the sulphides of arsenic and antimony dissolve in the caustic 
 potash. Chlorine-gas is now led into the warm solution for some 
 time. Most of the metals together with the gangue remain in the 
 precipitate, which is filtered off and well washed. The filtrate is 
 acidified with hydrochloric acid, again filtered if necessary to 
 remove silica, and the sulphuric acid determined as barium sul- 
 phate in the usual manner. 
 
 112. Determination of Tellurium and Selenium. These ele- 
 ments are precipitated from acid solutions by means of sulphur 
 dioxide or sodium or potassium sulphite. If the solution contains 
 nitric acid it must be evaporated to dryness after the addition of 
 hydrochloric acid and sodium or potassium chloride. The solu- 
 tion should be heated nearly to boiling, and kept at this temp- 
 erature for a considerable time to insure complete precipitation. 
 If the solution is dilute or a small amount of the element is 
 present, it must be allowed to stand for several days in a warm 
 place. The precipitate is washed by decantation with water 
 containing sulphur dioxide, and finally transferred to a weighed 
 filter-paper, or to a Gooch crucible, and dried at 100. 
 
 DETERMINATION OF NITROGEN. 
 
 Nitrogen unites with hydrogen to form a fairly strong base, 
 and with oxygen and hydrogen to form one of the strongest acids. 
 The methods used for determining the nitrogen when it exists in 
 one of these forms are not generally applicable to the determina- 
 tion of nitrogen in the other form. The methods most commonly 
 used are volumetric. In this chapter the gravimetric methods, 
 which are at times convenient, will be given. 
 
 113. Determination of Ammonia as Ammonium Chloride. 
 Ammonia may be weighed as ammonium chloride, which can be 
 dried at 100. A solution which contains nothing but free ammo- 
 nia, ammonium chloride, or the ammonium salt of a very readily 
 volatilized acid, such as carbonic acid or hydrogen sulphide, may 
 be acidified with pure hydrochloric acid and evaporated to dryness 
 on the water-bath in a weighed platinum dish. Many impurities, 
 such as silica and the alkaline earth metals and the alkalies, do not 
 interfere, since after weighing the residue in the dish the ammo- 
 nium chloride may be volatilized by gentle heat, and the non- 
 volatile residue weighed. Carried out in this manner the deter- 
 
106 DETERMINATION OF ACIDS. 
 
 mination of ammonia is quicKly and accurately made. It is 
 a convenient method when so few determinations are required 
 that the preparation of volumetric solutions is not warranted. 
 
 1 14. Determination of Ammonia as Ammonium Platmo-chloride. 
 Ammonia may also be precipitated as ammonium platino- 
 chloride which is washed with alcohol and dried at 100, as de- 
 scribed for potassium. The precipitate may also be decomposed 
 by ignition into metallic platinum. The heat must be applied 
 cautiously to the covered crucible, otherwise the escaping ammo- 
 nium chloride will carry away platinum. 
 
 115. Determination of Nitric Acid by Means of Nitron. No 
 inorganic compound of nitric acid is known which is sufficiently 
 insoluble to be used for the quantitative determination of this 
 acid. Recently an organic base has been made which forms an 
 insoluble nitrate.* This base is 1.4 diphenyl-3.5 endanilodihy- 
 drotriazol, having the formula C 20 H 16 N 4 . For convenience in 
 ordinary use and because it precipitates nitric acid, the name 
 " nitron" has been given to the base. For use as a reagent a 10 
 per cent solution in 5 per cent acetic acid is made. 
 
 At ordinary temperatures nitron precipitates nitric acid from 
 solutions of 1 in 60,000, while at C it will precipitate 1 part in 
 80,000. Unfortunately nitron also precipitates a number of other 
 common acids, such as hydrobromic when present to the amount 
 of 1 part in 800 of the solution, hydriodic acid (1 in 20,000), 
 chromic acid (1 in 6000), chloric acid (1 in 4000), perchloric acid 
 (1 hi 50,000), sulphocyanic acid (1 in 15,000), and nitrous acid 
 (1 in 4000). These acids must therefore be first removed from 
 the solution. Hydrobromic and hydriodic acids may be removed 
 by adding chlorine water and boiling out the liberated halogen, 
 or in the case of hydriodic acid it may be oxidized to the iodate. 
 These acids may also be removed by means of silver sulphate. 
 Nitrous and chromic acids may be reduced by means of hydrazine 
 sulphate. To remove the nitrous acid the neutral concentrated 
 solution (5-6 c.c.) is dropped slowly on one-fourth gram of finely 
 powdered hydrazine sulphate. It is advisable to place the hydra- 
 zine sulphate in a small flask in order to avoid loss by spattering 
 
 * M. Busch Ber., 38, 861 (1905). J. Chem. Soc., 88, II, 282, 418 (1905). 
 
DETERMINATION OF NITROGEN. 107 
 
 and to facilitate cooling the mixture by means of cold water, as 
 a small amount of nitric acid is formed if the mixture is allowed 
 to become warm. 
 
 The volume of the solution of the nitrate or nitric acid should 
 be small because of the slight solubility of the precipitate. About 
 0.1 gram of nitric acid should be present in a volume of 80 to 100 c.c. 
 which has been acidified with 10 drops of sulphuric acid. The 
 solution is brought almost to the boiling-point and 10 to 12 c.c. of 
 the nitron reagent is added. The hot solution gives a crystal- 
 line precipitate. The beaker is then immersed in ice-water, where 
 it is allowed to stand for one and one-half to two hours. The pre- 
 cipitate is filtered off on a Gooch crucible which has been dried 
 at 110. It is transferred to the crucible by means of the mother- 
 liquor and is washed with small portions of ice-water, using not 
 more than 10 to 12 c.c., the precipitate being sucked dry after 
 each addition of ice-water. The precipitate is then dried at 110 
 until constant, which requires about three-fourths of an hour. 
 The precipitate has the formula C 20 H ltJ N 4 .HN0 3 , and contains 
 16.8% HN0 3 or 14.4% N 2 5 . 
 
 116. Volatilization of N 2 5 by Fusion of Nitrates with Silica or 
 Potassium Chromate. Some very excellent methods of determin- 
 ing nitric acid depend on its conversion into ammonia, nitric oxide, 
 or nitrogen. Most of these methods will be described in the 
 chapters on volumetric and gas analysis. It may also be deter- 
 mined indirectly by heating the dry nitrate with a weighed amount 
 of silica or potassium-acid chromate until all of the nitric acid is 
 expelled. The loss in weight is due to expulsion of N 2 5 . The 
 method can obviously not be applied to material which ma}^ lose 
 anything but nitric acid. It has been extensively applied to 
 sodium and potassium nitrates which can be dried at 130 and 100 
 respectively. The silica used must be finely powdered and ignited 
 before being weighed. About seven times as much silica as nitrate 
 must be taken. The material is intimately mixed and heated 
 to redness in a platinum crucible for two to four hours or until 
 constant weight is obtained. 
 
 Instead of the silica a mixture of equal parts of neutral and acid 
 potassium chromate may be used. This material is fused in a plat- 
 inum crucible, allowed to cool, and finely powdered. About 3 grams 
 
108 DETERMINATION OF ACIDS. 
 
 of the chromate mixture are weighed out and placed in a platinum 
 crucible. .8000 gram of the dried nitrate is added and thoroughly 
 mixed with the chrornate by means of a platinum or glass rod. 
 The crucible is heated gently so as to gradually bring the contents 
 to a state of fusion. When no more acid fumes are evolved, the 
 crucible is cooled in the desiccator and weighed. It is reheated 
 until constant weight is obtained. 
 
CHAPTER IX. 
 
 DETERMINATION OF CARBONIC, BORIC, AND 
 PHOSPHORIC ACIDS. 
 
 DETERMINATION OF CARBON DIOXIDE BY LOSS. 
 
 Two general methods are in use for the determination of 
 carbon dioxide. By the first method the carbonate is decom- 
 posed by means of a strong mineral acid or other means, and the 
 carbon dioxide expelled. The amount present is ascertained by 
 weighing the apparatus before and after the expulsion of the 
 carbon dioxide. By the second method the gas evolved by the 
 decomposition of the carbonate* is led into a caustic-potash solu- 
 tion, which is weighed before and after the absorption of the car- 
 bon dioxide. The gain in weight represents the amount of carbon 
 dioxide present. 
 
 Various forms of apparatus have been devised for the deter- 
 mination of carbon dioxide by loss when 
 the carbonate is decomposed by means of 
 acid. Two of these will be described, one 
 of which may be purchased complete, 
 while the other rnay be put together 
 from the usual laboratory apparatus. 
 
 117. The Schrotter Apparatus shown 
 in Fig. 16 may be purchased ready for use. 
 A weighed amount of the carbonate is 
 placed in the small bulb e and the stopper 
 a firmly inserted. The tube c is filled with 
 dilute nitric or 10% hydrochloric acid, 
 and b is about half filled with concen- 
 trated sulphuric acid. The entire appa- 
 ratus is then carefully weighed. By open- 
 ing the stop-cock d, acid is allowed to 
 flow into e at such a rate that the bubbles of gas can readily 
 be counted while passing through the concentrated sulphuric 
 
 109 
 
 FIG. 16. 
 
110 
 
 DETERMINATION OF ACIDS. 
 
 acid in 6. When all of the acid has been allowed to flow out of 
 c, and carbon dioxide is no longer evolved, the stop-cock d is 
 closed and the solution hi e is gently warmed on the water-bath 
 or a hot plate to expel the carbon dioxide. The stop-cock d is 
 now opened and air is sucked through the apparatus at a moderate 
 rate until the carbon dioxide has been displaced. After being 
 allowed to come to the atmospheric temperature, the apparatus 
 is again weighed. The loss in weight represents the amount 
 of carbon dioxide evolved. 
 
 118. Sources of Error in the Determination of Carbon Dioxide 
 by the Schrotter Apparatus. Though the manipulation of this 
 apparatus is very simple, it is impossible to obtain results accu- 
 rate within more than J%. The error involved in weighing a 
 glass apparatus of large surface is considerable. The gas which 
 leaves the apparatus is dried over concentrated sulphuric acid, 
 while that which enters is not dry. It is customary to ascertain 
 when all of the carbon dioxide has been displaced by air by the 
 taste of the gas which is sucked through by the mouth. The 
 ordinary chemical tests for carbon dioxide cannot be used 
 unless the air drawn through is freed from this gas. The accu- 
 racy of the method would undoubtedly be increased if the air 
 drawn through were first passed through 
 tubes containing soda-lime and concen- 
 trated sulphuric acid. Care must also be 
 taken to have the apparatus at the same 
 temperature each time it is weighed. 
 
 119. Simple Laboratory Apparatus for 
 Determining Carbon Dioxide. Various 
 forms of apparatus which can be made 
 from the usual laboratory material have 
 been suggested. A very simple device is 
 shown in Fig. 17. The small flask A is 
 closed with a two-holed rubber stopper, 
 through which the thistle-tube passes as 
 well as the bent-glass tube 6, the latter 
 also passing to the bottom of the short test- 
 The thistle-tube is closed with a glass rod, 
 piece of rubber tube has been 
 
 FIG. 17. 
 
 tube or small bottle B. 
 
 over the end of which a short 
 
DETERMINATION OF CARBON DIOXIDE. Ill 
 
 slipped. The weighed amount of the carbonate is placed in the 
 flask, and covered with a little water. The rubber stopper is 
 firmly inserted, and a few cubic centimeters of dilute hydrochloric 
 acid placed in the thistle-tube. If the carbonate is completely 
 decomposed by sulphuric acid, it may be used instead of the dilute 
 nitric or 1C% hydrochloric acid. The tube B is half filled with con- 
 centrated sulphuric acid. The apparatus is tested for gas leaks 
 by drawing out a little air from the bottle A, by applying suction 
 to the tube a. The sulphuric acid rises in the tube 6, and should 
 remain at the same height for several minutes. The apparatus 
 is now carefully weighed, and the carbonate decomposed by allow- 
 ing the acid in the thistle-tube to flow into the flask in small por- 
 tions. When all of the acid has been introduced, the flask A is 
 gently warmed to expel the carbon dioxide. The carbon dioxide 
 is displaced with air by applying suction at a. In careful work 
 the air must be purified before entering the apparatus by passing 
 through soda-lime and concentrated sulphuric acid, these absorp- 
 tion-tubes being connected to the thistle-tube by means of a one- 
 holed rubber stopper. 
 
 120. Determination of Carbon Dioxide by Fusion of Carbonates 
 with Anhydrous Borax or Microcosmic Salt. The carbon dioxide 
 in all anhydrous carbonates may be determined by fusion with 
 anhydrous borax. The carbon dioxide is completely expelled 
 and its amount computed from the loss in weight. About 4 
 grams of vitrified borax are taken for 1 gram of the carbonate. 
 The borax is placed in a platinum crucible, and heated with 
 the Bunsen burner until the material is fused. It is allowed 
 to cool in the desiccator and then weighed. As there is dan- 
 ger of loss by the cracking of the borax if cooled suddenly, the 
 flame of the Bunsen burner is turned down gradually before 
 placing the crucible in the desiccator. As the borax retains 
 water somewhat persistently, it is advisable to fuse and weigh it 
 again. The anhydrous material may be kept in a state of fusion 
 for one-quarter to one-half hour without loss in weight. The 
 borax will be volatilized if heated with the blast-lamp. The weighed 
 quantity of the dry carbonate is placed on top of the borax, and 
 the latter again fused. As there is danger of loss by the cracking 
 of the borax on being heated, the crucible should be covered until 
 the borax is melted. It must be kept melted until the carbonate 
 
112 DETERMINATION OF ACIDS. 
 
 is dissolved and no more bubbles escape. A few generally remain 
 in the fused mass, and cannot be expelled except by very pro- 
 longed heating. After being again cooied in the desiccator, the 
 crucible is weighed. The loss in weight is du3 to carbon dioxide. 
 The borax glass may generally be removed by pressing the crucible 
 between the thumb and fingers, thus loosening the fused mass, 
 which will drop out on gently tapping the inverted crucible. 
 
 Microcosmic salt may be substituted for borax in this deter- 
 mination.* This salt melts very easily, needirg only a small 
 flame of the Bunsen burner. The heating is continued until all 
 of the ammonia and water has been expelled. It is cooled and 
 weighed and reheated until the weight is constant. The weighed 
 carbonate is then added and the contents of the crucible heated 
 very cautiously to avoid loss by the foaming caused by the escape 
 of carbon dioxide. When no more carbon dioxide escapes, the 
 crucible is cooled and weighed. It is again heated until the 
 weight is constant. 
 
 The action of the borax on the carbonate produces metaborates 
 from the pyrob orate, while the fused microcosmic salt is changed 
 from the metaphosphate to the normal phosphate, according to 
 the following equations : 
 
 Na 2 B 4 7 + CaC0 3 = 2NaB0 2 + Ca (B0 2 ) 2 + C0 2 ; 
 NaP0 3 +CaC0 3 = NaCaP0 4 +C0 2 . 
 
 DIRECT WEIGHING OF CARBON DIOXIDE. 
 
 The most reliable and accurate method of determining carbon 
 dioxide is that in which the gas is absorbed and weighed. The 
 carbonate is decomposed by sulphuric or hydrochloric acid. 
 The gas evolved is dried, freed from hydrochloric acid, and 
 absorbed in soda-lime or caustic-potash solution. The last por- 
 tions of carbon dioxide are swept out of the apparatus by means of 
 a stream of air which has been freed from carbon dioxide. 
 
 121. Apparatus. Many forms of apparatus have been proposed 
 
 and used for this purpose. The simplest and most serviceable 
 
 form is that given by FRESENIUS. The characteristic part of this 
 
 apparatus consists of a long glass tube which acts as a condenser, 
 
 and also holds the drying and purifying material. In other forms 
 
 * O. Lutz and A. Tschischikow, Jour. russ. phys.-chem. Gesellsch., 36, 1274. 
 
DETERMINATION OF CARBON DIOXIDE. 
 
 113 
 
 of apparatus the drying materials are placed in a series of U-tubes 
 which are connected by rubber tubing, stoppers, etc. The diffi- 
 culties in making absolutely gas-tight joints are numerous, espe- 
 cially when the permeability of rubber to carbon dioxide is consid- 
 ered. The small number of joints in the apparatus of Fresenius 
 gives it a decided advantage over other forms. In all forms of 
 apparatus the carbonate is placed in a small flask closed with a 
 two-holed stopper. Through one hole a dropping-funnel is passed 
 by means of which the acid is introduced, and through the other 
 the exit tube for the gas is passed. The end of the dropping- 
 funnel is drawn out to a capillary, which is bent upwards to pre- 
 vent loss of carbon dioxide. 
 
 The arrangement of the apparatus of Fresenius is shown in Fig. 
 18. The Erlenmeyer flask should be of about 150 c.c. capacity. 
 The glass tube should be about 70 cm. long and at least 1.2 cm. 
 in diameter. It is inclined slightly so as to allow any water 
 which condenses in its lower half to flow back into the flask. A 
 
 FIG. 18. 
 
 wad of glass wool or cotton is pushed down to the centre of the 
 tube, the upper half of which is then filled with granulated cal- 
 cium chloride, large lumps being first introduced and then the 
 finer material. If this calcium chloride has not already been 
 
114 DETERMINATION OF ACIDS. 
 
 saturated with carbon dioxide, a slow stream of the dry gas is 
 passed through the tube for some time and then completely dis- 
 placed with air. 
 
 122. Drying and Absorption of the Carbon Dioxide. The 
 GEISSLER CAUSTIC-POTASH BULBS are filled two-thirds full with a 
 solution of 1 part of caustic potash and 2 parts of water. The 
 straight calcium- chloride tube is filled with soda-lime and fused 
 calcium chloride, so .that the gas leaves the apparatus after pass- 
 ing over the calcium chloride. SODA-LIME is sometimes used foi 
 absorbing the carbon dioxide. It is placed in U-tubes, which 
 are weighed before and after absorption of the carbon dioxide. 
 Unless the soda-lime is freshly prepared and somewhat moist the 
 absorption of large amounts of carbon dioxide is somewhat slow, 
 so that two U-tubes are usually required. A calcium-chloride 
 tube must also be attached and weighed. Whether caustic potash 
 or soda-lime is used, an unweighed guard tube filled with calcium 
 chloride must be attached. A small tube filled with soda-lime is 
 also fitted to the dropping-funnel to remove carbon dioxide 
 from the air which is drawn through the apparatus to sweep out 
 the carbon dioxide. 
 
 123. Decomposition of the Carbonate. Dilute sulphuric acid 
 should be used for decomposing the carbonate, provided insoluble 
 sulphates are not formed. When sulphuric acid cannot be used, 
 dilute hydrochloric or nitric acids may be employed. If sul- 
 phides or sulphites are present, chromic acid or a chromate is 
 added to prevent the evolution of sulphuretted hydrogen or sul- 
 phur dioxide and sulphuric or nitric acid must be used to decom- 
 pose the carbonate. When hydrochloric acid is used or chlorides are 
 present, the carrying of the acid to the absorption bulbs may be 
 prevented by substituting for part of the calcium chloride in the 
 long drying-tube pieces of pumice-stone which have been saturated 
 with copper-sulphate solution and dried at 125. The method of 
 setting up the apparatus and the details of manipulation are 
 given in Chapter XVI, p. 183, in the determination of carbon diox- 
 ide in dolomite. 
 
DETERMINATION OF BORIC ACID. 
 
 115 
 
 DETERMINATION OF BORIC ACID BY THE METHOD 
 
 OF GOOCH. 
 
 Considerable difficulty has been met with hi determining boric 
 acid, since no insoluble salt of this acid is known which can be 
 washed and weighed accurately without the expenditure of a 
 great deal of labor. The acid can be readily and completely 
 separated from all other acids and bases by distillation with 
 methyl alcohol. The distillate may then be evaporated to dryness 
 and ignited with a known amount of calcium oxide. No boric 
 acid is lost and the amount present is given by the increase hi 
 weight of the lime. The details of this method were first pub- 
 lished by Gooch.* 
 
 124. Apparatus. The apparatus recommended by Gooch con- 
 sists of a retort made of a 150-c.c. or 200-c.c. pipette. One of 
 the stems of the pipette, which should have an internal diameter 
 of at least 0.7 cm., is bent into the form of a gooseneck, and is con- 
 
 FIG. 19. 
 
 nected by means of a rubber stopper with an upright condenser, 
 
 which in turn is connected by means of a loosely fitting stopper 
 
 * Am. Chem. Jour., vol. 9, p. 23. 
 
116 DETERMINATION OF ACIDS. 
 
 with an Erlenmeyer flask of about 250 c.c. capacity. The other 
 stem of the pipette is bent at right angles and is connected by 
 means of a short piece of rubber tubing with a small dropping- 
 funnel. An oil- or paraffine-bath is provided of such a size that 
 the retort may be immersed in the liquid. 
 
 125. The Solution of the Borate, which should not contain 
 more than 0.2 gram of B 2 3 , is acidified with nitric or acetic acid.* 
 Hydrochloric acid should be absent, as a considerable amount of 
 this acid would pass over into the lime and be weighed with it 
 unless the solution is neutralized and then acidified with acetic 
 acid. It is better, however, to remove the chlorides with silver 
 nitrate. The filtrate may be distilled directly or the excess of 
 silver nitrate removed with sodium hydroxide or carbonate. 
 Substances insoluble in nitric acid are fused with sodium carbon- 
 ate. If fluorine is present, it is removed by precipitation as 
 calcium fluoride in the water solution of the fused material. 
 
 126. Distillation of the Boric Acid with Methyl Alcohol. From 
 1 to 2 grams of calcium oxide are placed in a platinum dish and 
 heated over the blast-lamp until the weight is constant. It 
 is then rinsed with a little water into the Erlenmeyer flask. 
 A large excess of the acetic or nitric acid used to acidify 
 the solution of the borate should be avoided, since these acids 
 volatilize and neutralize the lime. Their salts also form viscous 
 liquids on evaporation. A drop or two of phenolphthalein should 
 therefore be added to the solution and a drop or two of the acid 
 added after the alkaline color of the indicator has been removed. 
 The smaller amounts of calcium oxide may be used when acetic 
 acid is employed. The solution of the borate is transferred to 
 the retort, which is lowered so that at first only the bottom touches 
 the oil or parafnne, which is heated to 130-140. As the liquid 
 distils off the retort is lowered until half of it is immersed and 
 the distillation continued until the water is entirely removed. 
 10 c.c. of methyl alcohol is then introduced through the dropping- 
 funnel and the solution distilled to dryness. This is repeated 
 five times. If nitric acid has been used to acidify the solution, 
 introduce into the retort 2 c.c. of water between the second and 
 third and between the fourth and fifth additions of methyl alco- 
 
 *The small amounts of these acids which distil over into the lime are ex- 
 pelled during the subsequent ignition. 
 
DETERMINATION OF PHOSPHORIC ACID. 117 
 
 hol and distil to dryness. If acetic acid has been used, add a 
 few drops of this acid with the fourth portion of methyl alcohol. 
 
 127. Weighing the Boric Acid. After the sixth distillation 
 with the methyl alcohol transfer the solution hi the Erlenmeyer 
 flask to the platinum dish in which the lime was ignited and 
 weighed. Evaporate the solution to dryness, being careful to 
 avoid spattering. Finally heat the dish with the blast-lamp until 
 constant weight is obtained. The increase in weight is due to B 2 3 . 
 The residue in the retort may be tested for boric acid by means 
 of turmeric-paper. If nitric acid has been used the nitrous acid 
 should first be oxidized by means of bromine and the excess of 
 the latter expelled. 
 
 DETERMINATION OF PHOSPHORIC ACID. 
 
 128. Precipitation of Phosphoric Acid as Magnesium-arnmon- 
 ium Phosphat3. Phosphoric acid may be most accurately 
 weighed as magnesium pyrophosphate. The properties of 
 magnesium-ammonium phosphate as well as the magnesium 
 pyrophosphate produced by ignition of this precipitate, have 
 already been discussed under the Determination of Magnesium 
 in Chapter VII, page 85. Only the alkali metals may be pres- 
 ent in the solution, as the phosphates of the other metals are in- 
 soluble in ammonia and would contaminate the precipitate. Only a 
 moderate amount of ammonium salts should be present, and the 
 solution should be slightly acid with hydrochloric acid. The 
 phosphoric acid is precipitated by adding with constant stirring 
 magnesia mixture in slight excess. This mixture is made by 
 dissolving 55 grams of crystallized magnesium chloride and 70 
 grains of ammonium chloride in water, adding 300 c.c. of dilute 
 ammonia or 88 c.c. of concentrated ammonia (sp. gr. 0.90) and 
 diluting to 1 liter. After standing for several days, the solution 
 is filtered as required for use. 10 c.c. of this solution is sufficient 
 to precipitate 0.15 gram of P 2 5 . If the phosphate solution is 
 not distinctly alkaline after the addition of the magnesia mixture, 
 it should be made so -by the addition of ammonia. After stand- 
 ing for at least six hours the precipitate is filtered off, washed 
 with water containing one-tenth its volume of strong ammonia, 
 and the precipitate ignited as directed for the determination of 
 magnesium. 
 
118 DETERMINATION OF ACIDS. 
 
 129. Separation of Phosphoric Acid as Phosphomolybdate. If 
 
 metals other than the alkalies are present, the phosphoric acid must 
 be precipitated from a nitric-acid solution as phosphomolybdate. 
 All of the phosphoric acid is precipitated while the metals are 
 held in solution by the nitric acid, in which the phosphorous pre- 
 cipitate is insoluble. The percentage of phosphoric acid in the 
 precipitate is not constant, but varies within quite wide limits 
 with the conditions of the precipitation. Although by rigidly 
 fixing these conditions a precipitate of sufficiently constant com- 
 position for direct weighing can be obtained, more reliable results 
 are obtained by dissolving the precipitate in ammonia, repre- 
 cipitating with magnesia mixture and weighing as magnesium 
 pyrophosphate. 
 
 130. Removal of Silicic and Arsenic Acids. Before precipi- 
 tating the phosphoric acid as molybdate, silicic acid must be 
 removed from the solution by evaporating to dryness and heating 
 at 100 for some time. The dried residue is dissolved in water 
 and nitric acid. Arsenic acid must also be absent except in 
 traces, as this acid forms a precipitate with molybdic acid, which 
 is very similar to the phosphomolybdate. It is also reprecipi- 
 tated with the magnesia mixture. It may readily be separated 
 from phosphoric acid by passing hydrogen sulphide through the 
 hot hydrochloric-acid solution. As HYDROCHLORIC ACID inter- 
 feres with the precipitation of the phosphoric acid as molybdate, 
 it must be expelled by evaporation with nitric acid after precipita- 
 tion of the arsenic, or the separation of the arsenic acid may be 
 delayed until the phosphoric acid has been precipitated by means 
 of magnesia mixture. This precipitate, which will generally con- 
 tain only a part of the arsenic acid, is dissolved in a little hydro- 
 chloric acid, the solution warmed to 70, and hydrogen sulphide 
 passed for a considerable time. The precipitate is filtered off 
 and washed and the nitrate evaporated to a volume of 50 to 75 c.c. 
 The solution is allowed to cool, a few drops of magnesia mixture 
 are added, and then ammonia, with constant stirring until a slight 
 excess is present. This precipitate is filtered off, washed, ignited, 
 and weighed as usual. 
 
 131. Precautions. The precautions * to be observed in pre- 
 cipitating the phosphoric acid as molybdate are as follows: 
 
 * Zeit. Anal. Ch., 28, 141 . 
 
DETERMINATION OF PHOSPHORIC ACID. 119 
 
 Hydrochloric, sulphuric, and boric acids should be absent or 
 present only in small amounts. 
 
 Free nitric acid should be present only to the extent of 26 
 molecules to one molecule of the phosphoric acid. If more than 
 80 molecules of free nitric acid are present to one molecule of 
 phosphoric acid, the precipitation is incomplete. 
 
 The presence of ammonium nitrate hastens the precipitation. 
 
 Twice as much molybdic acid must be present as is necessary 
 to form the precipitate; that is, 24 molecules to 1 molecule of 
 phosphoric acid. 
 
 132. Precipitation of the Phosphomolybdate. The solution 
 of the phosphate should be concentrated so that about 0.1 gram 
 P 2 5 is present in 25 c.c., 0.2 gram P 2 5 being sufficient for a 
 determination. Enough of a concentrated ammonium-nitrate 
 solution (750 gr. per liter) is added to constitute 15% of the total 
 volume after sufficient molybdate solution has been added to 
 have 80 c.c. present for each 0.1 gram of P 2 5 . The solution is 
 then heated on the water-bath to 80-90 for about ten minutes, 
 and is then allowed to stand for about one hour, when the precipi- 
 tate may be filtered off and washed. The addition of the ammo- 
 nium nitrate may be omitted, but more of the molybdate solution 
 must be added, and the solution heated on the water-bath to 60 
 for from four to six hours. In either case the filtrate should be 
 tested for phosphoric acid by adding more molybdate solution 
 and warming. A white precipitate of molybdic acid must not be 
 mistaken for the yellow phosphate precipitate. 
 
 133. Washing the Precipitate. Various solutions have been 
 used for washing the precipitate, which is not absolutely insoluble, 
 so that the minimum amount of wash-water should be used. 
 Wagner recommends a solution made by dissolving 150 grams of 
 ammonium nitrate in water, adding 10 c.c. concentrated nitric 
 acid, and diluting the solution to one liter. Fresenius recom- 
 mends the use of a solution made by adding to 100 c.c. molybdate 
 solution 20 c.c. nitric acid (sp. gr. 1.2) and 80 c.c. water. A solu- 
 tion made by diluting the molybdate solution with an equal vol- 
 ume of water is also frequently used. 
 
 134. Direct Weighing of the Phosphomolybdate. When it is 
 desired to weigh the molybdate precipitate, the following slight 
 variations in the method of procedure already given must be ob- 
 
120 DETERMINATION OF ACIDS. 
 
 served: Silica must first be completely removed from the solution 
 of the phosphoric acid. After the molybdate has been added, 
 the solution should be heated to about 40 to effect complete 
 precipitation. Serious error results if it is heated to a tempera- 
 ture near 100. About four hours' digestion at 40 is generally 
 sufficient. The precipitate is filtered off on a Gooch crucible and 
 washed, first with a dilute solution of molybdate until the metals 
 present are removed, and then with water containing 1% of nitric 
 acid until the molybdic acid is removed. The precipitate is 
 then dried in an air-bath heated to 120. It contains 1.63% of 
 phosphorus. 
 
 135. Reprecipitation of the Phosphoric Acid as Ammonium- 
 magnesium Phosphate. When it is desired to weigh the phos- 
 phorus as magnesium pyrophosphate, the molybdate precipitate 
 is washed with the solutions given in par. 133 until the metals 
 present are removed. The beaker in which the precipitation was 
 made, and from which the yellow precipitate need not be wholly 
 removed, is placed under the funnel, the paper is pierced, and the 
 precipitate washed into the beaker with a fine jet of water. The 
 paper is moistened with a little ammonia and washed with hot 
 water, a few drops of ammonia being added if necessary to dis- 
 solve the precipitate. Not more than 50 c.c. of wash-water should 
 be used. If the precipitate does not dissolve on stirring the solu- 
 tion, a little more ammonia should be added. Neutralize the 
 solution with strong hydrochloric acid, and if the yellow phos- 
 phomolybdate begins to precipitate, add ammonia until dis- 
 solved. A white flocculent precipitate insoluble in ammonia is 
 probably silica, and should be filtered off. The solution is again 
 rendered slightly acid with hydrochloric acid and enough mag- 
 nesia mixture added to precipitate the phosphoric acid present, 
 using 10 c.c. for 0.1 gram P 2 5 . The solution is neutralized by 
 adding ammonia while stirring constantly. A moderate excess of 
 ammonia is added and the solution is allowed to stand for at least 
 four hours. The precipitate is filtered off, washed with dilute 
 ammonia, ignited, and weighed as usual. Heating with the 
 blast-lamp for ten minutes is advisable, in order to expel any 
 molybdic acid which may have been carried down with the mag- 
 nesia precipitate. 
 
ANALYSIS OF ALLOYS.* 
 
 CHAPTER X. 
 
 ALLOYS OF SILVER, LEAD, COPPER, BISMUTH, 
 CADMIUM, AND TIN. 
 
 136. Reasons for Analysis of Pure Salts. In the determina- 
 tions given in the preceding chapters pure compounds were 
 taken for analysis. This method of procedure serves two 
 purposes. In the first place the exact composition of the 
 compounds being known, the error in a given determination 
 can easily be ascertained. In the second place no error can 
 arise from the contamination of a precipitate with foreign 
 material other than the reagents used. When the student has 
 acquired proficiency in the comparatively simple processes of 
 analyzing pure compounds, he must undertake the determination 
 of elements when existing in combination with each other. It 
 is evident that the determination of iron by precipitation with 
 ammonia is impracticable if aluminium or chromic salts are present. 
 
 137. Difficulty of Complete Separation of Elements. Although 
 calcium is the only metal which can be quantitatively precipitated 
 as oxalate, yet it is by no means true that this element can be 
 separated as oxalate from a solution containing all other metals 
 without carrying many of these down with it. This is due to 
 the fact that most of the oxalates are soluble with difficulty. 
 All of the sulphates except those of barium, strontium, and lead 
 are quite soluble, and yet when such soluble sulphates as those of 
 iron or aluminium are present the determination of sulphuric 
 acid by precipitation with barium chloride presents difficulties 
 
 * The alloys and minerals whose analyses are given in this and the following 
 chapters have been selected so as to afford practice in the separations required 
 in the analysis of commonly occurring substances. The elements mentioned as 
 occurring in a given alloy or mineral are those commonly found. Some of these 
 are at times absent, while others not listed may be present. 
 
 121 
 
122 ANALYSIS OF ALLOYS. 
 
 which have been the subject of investigation by many ingenious 
 workers for more than half a century. As the great majority of 
 the substances with which the chemist is called upon to deal are 
 complex rather than simple, a careful study of the separation of 
 the elements is of the greatest importance. Only by the closest 
 attention to details can success be attained hi the analysis of 
 complex substances. The importance of testing precipitates for 
 impurities and the solution for unprecipitated portions of an 
 element cannot be too strongly urged. Only in this manner can 
 the accuracy of an analysis be assured. 
 
 138. Limit of Accuracy in Analysis. If a complete analysis 
 is made the sum of all the constituents must be very close to 
 100%. A summation which is within .5% can generally be ob- 
 tained if the analysis is conducted with care and reliable methods 
 are used. In general the analysis of an unknown substance 
 should be conducted in duplicate. If the duplicate results do 
 not agree within .2 or at most .3%, a third analysis should be 
 made. As the error of most determinations is at least .1%, it is 
 unnecessary to calculate results to more than hundredths of per 
 cent. As the error in each determination of the analysis of a 
 given substance may be either plus or minus, the practice of 
 dividing the difference between the summation and 100$ among 
 the various determinations is not justifiable. 
 
 It is in some cases possible to analyze a substance in such a 
 manner that the results are accurate to the hundredth of a per 
 cent. Such results may be computed to the .001 of a per cent. 
 This practice is common in the analysis of metals. Large quan- 
 tities of the metal are taken, so that considerable quantities of 
 the impurities which are present in small amounts are obtained 
 for determination. The results may then be accurate to the hun- 
 dredth of a per cent. This does not imply a higher degree of 
 accuracy in the determination of a given element than .1 of a per 
 cent. For example ; if iron were present in copper to the extent 
 of .5%, a determination of the iron which is accurate to .01% of 
 the impure copper would represent an error of ^ of the amount 
 of iron present in the copper. In giving the results of such analy- 
 ses the percentage of the main constituent is obtained by difference, 
 so that the summation is exactly 100%. 
 
SEPARATION OF SILVER. 123 
 
 139. Separation of Silver from Other Metals. Silver is readily 
 separated from other metals by precipitation as chloride. Silver 
 chloride being more insoluble in dilute nitric acid than in water 
 the precipitate is washed with dilute nitric acid. This also serves 
 to dissolve compounds of metals such as bismuth and antimony, 
 which tend to form insoluble basic salts. As silver chloride is 
 soluble in mercuric nitrate enough hydrochloric acid must be added 
 to convert all of the mercury present into chloride. This metal 
 must be present in the mercuric condition. The small amount 
 of silver remaining in solution is subsequently precipitated with 
 the mercury as sulphide and may be separated by volatilization 
 of the mercury. If lead is present the solution must be dilute 
 and hot, and no more hydrochloric acid added than just sufficient 
 to precipitate the silver. The silver chloride must be thoroughly 
 washed with hot water. If sodium acetate is added the lead is 
 more easily kept in solution. 
 
 EXERCISE 25. 
 Analysis of a Silver Coin. 
 
 Alloy of Copper and Silver. 
 
 Prepare two Gooch crucibles for weighing silver chloride. They may 
 be marked 1 and 2 with a blue pencil * and the marks made permanent by 
 heating to redness with the blast-lamp. While the crucibles are drying 
 on the hot plate the following operations should be carried out: 
 
 140. Dissolving the Alloy. Clean a dime thoroughly by rubbing with 
 sand or " Sapolio." Cut into pieces and weigh carefully a piece of about one- 
 half gram. Place in a 200-c.c. beaker and add about 5 c.c. concentrated 
 nitric acid and an equal bulk of water. Cover the beaker with a watch- 
 crystal and warm on the water-bath until the metal is converted into nitrate. 
 Dissolve the residue in water. A little gold is sometimes present and will 
 remain undissolved at this point. The solution of silver and copper is 
 decanted into a 300-c.c. Erlenmeyer flask and the gold washed by decanta- 
 tion five or six times, using about 25 c.c. of water each time. The gold is 
 now transferred to a small filter-paper, which is washed with hot water 
 until free from silver. The moist paper is transferred to a weighed porcelain 
 crucible, burned in the usual manner, and the gold weighed. As the amount of 
 gold is always small, it is best to collect on the same paper the gold from all of 
 the weighed portions of the dime, which are dissolved separately in nitric acid 
 to obtain duplicate determinations of the silver and copper. The paper is 
 washed free from silver after each precipitate has been transferred to it. 
 
 * A pencil marie for marking on glass, porcelain, etc , is used for this purpose 
 
124 ANALYSIS OF ALLOYS. 
 
 141. Silver. The volumes of the solutions of silver and copper in the 
 Erlenmeyer flasks should not much exceed 200 c.c. By proper manipula- 
 tion this result can easily be secured. Heat nearly to boiling and precipi- 
 tate the silver by adding hydrochloric acid, drop by drop, with vigorous 
 shaking of the flask. Add only a slight excess of hydrochloric acid and 
 digest the precipitate on the water-bath with occasional vigorous shaking 
 until the solution is very nearly clear. Decant the clear liquid through the 
 weighed Gooch crucible and wash the precipitate two or three times by 
 decantation with about 50 c.c. of hot water to which a few drops of nitric 
 acid have been added. The precipitate should be digested each time for a 
 few minutes on the water-bath. Finally transfer the precipitate to the 
 crucible and wash with hot water containing a little nitric acid until the 
 wash-water gives no precipitate with hydrogen sulphide. Dry the pre- 
 cipitate on the hot plate and weigh. 
 
 142. Copper. Ignite and weigh a Rose crucible. Heat the copper 
 solution, the volume of which should be about 400 c.c., nearly to boiling and 
 saturate with hydrogen sulphide. Filter immediately and wash with 
 water containing hydrogen sulphide. Test the filtrate and wash -water 
 for copper by passing hydrogen sulphide. Dry the precipitate, detach 
 from the paper, burn the latter, and ignite the precipitate in the Rose cruci- 
 ble in a stream of dry hydrogen after the addition of sulphur. 
 
 143. Division of Coin Volumetrically. If there is available a 250-c.c. 
 flask and a 50-c.c. pipette which have been carefully calibrated against 
 each other, the division and weighing of the coin may be carried out in the 
 following manner. The coin is weighed after thorough cleansing. It is 
 then placed in a 200-c.c. beaker and treated with 10 c.c. concentrated nitric 
 acid and an equal bulk of water. The 'beaker is covered with a watch- 
 crystal and warmed on the water-bath until the metal is converted into 
 nitrate. The watch-crystal is rinsed off and the solution is evaporated to 
 dryness on the water-bath. The residue is dissolved in water, the gold, if 
 present, filtered off and weighed, the filtrate and washings being allowed 
 to flow into the 250-c.c. flask. Finally the solution in the flask is diluted 
 to the mark with water, thoroughly shaken, and 50-c.c. portions taken out 
 with the pipette. The analysis of these portions is carried out as already 
 directed. 
 
 The duplicate determinations of silver and copper should agree within 
 about .2%. American silver coins contain very nearly 90% of silver and 
 10% of copper. 
 
 144. Economizing Time. In carrying out this and subse' 
 quent exercises the work should be carefully planned to econo- 
 mize time. Crucibles should be prepared and weighed during 
 the intervals when precipitates are digesting, filtering, or drying, 
 so that the latter may be transferred and weighed without delay. 
 
SEPARATION OF TIN. 125 
 
 Two precipitations and nitrations can be carried out at the same 
 time just as readily as one. As experience is gained, a greater 
 and greater number of analyses can be carried on simultaneously. 
 Careful and systematic marking of beakers, funnels, etc., is abso- 
 lutely necessary. All weights must be recorded in the note-book 
 under systematic headings and numbering of duplicate analyses. 
 
 145. Separation of Tin as Stannic Oxide. Tin is usually sepa- 
 rated from other metals as the dioxide Sn0 2 . On treating an 
 alloy containing tin with nitric acid, most of the metals form 
 nitrates, while the tin is converted into the dioxide, which remains 
 undissolved on treating the residue with water. If ANTIMONY or 
 ARSENIC is present more or less of these elements, depending on 
 the amount present, will remain with the tin. Besides these ele- 
 ments, which one would expect from the nature of their oxides to 
 remain undissolved by water or dilute acids, small portions of 
 copper, lead, bismuth, iron, manganese, and zinc remain with the stannic 
 oxide and cannot be removed by digestion with dilute nitric 
 acid. In the absence of more than traces of arsenic and antimony 
 the stannic oxide is filtered from the nitric-acid solution, washed 
 with dilute nitric acid, ignited and weighed. 
 
 146. Purification of the Stannic Oxide. The weighed precip- 
 itate is then fused in a porcelain crucible with six times its 
 weight of equal parts of sodium carbonate and sulphur, or with 
 the same weight of dry sodium thiosulphate. By this treatment 
 the metals present are converted into sulphides. On treating 
 the fused mass with hot water the sulphides of tin, arsenic, and 
 antimony and a small amount of copper if this element is present 
 dissolve as thio salts of sodium, while the sulphides of the 
 other metals remain and may be filtered off. On treating 
 the insoluble sulphides with warm dilute hydrochloric acid, cop- 
 per sulphide remains undissolved and may be ignited and weighed 
 as CuO. In the filtrate the lead may be precipitated with sul- 
 phuric acid and weighed as sulphate. The iron is precipitated 
 with ammonia, while the zinc is separated from the manganese 
 by passing hydrogen sulphide through the slightly acid solution. 
 The manganese may be precipitated with hydrogen peroxide and 
 weighed as Mn 3 4 . The weight of any or all of these metals 
 computed as oxide is deducted from the weight of the impure 
 stannic oxide. 
 
126 ANALYSIS OF ALLOYS. 
 
 If several of these metals are present in the alloy it is perhaps 
 easier and fully as accurate to omit weighing the impure stannic 
 oxide, but fuse it immediately as already described for its purifi- 
 cation. The insoluble sulphides are dissolved in a little nitric 
 acid and added to the nitric acid solution of the alloy. The solu- 
 tion of sodium sulphostannate is nearly decolorized by boiling 
 with the addition of pure caustic soda and hydrogen peroxide. 
 The tin is precipitated as sulphide by acidifying with hydro- 
 chloric acid. The precipitate is filtered off and washed with 
 water containing ammonium acetate and a little acetic acid. It 
 is then treated with a little concentrated ammonium carbonate 
 solution to dissolve arsenic sulphide. After washing with water the 
 sulphides of tin and antimony are dissolved in warm concentrated 
 hydrochloric acid. Any antimony which is present may be pre- 
 cipitated by digesting the diluted solution with metallic iron. The 
 tin in the filtrate from the antimony is precipitated with hydrogen 
 sulphide and washed. The precipitate is dried and detached 
 from the paper, which is burned. It is then placed in a porcelain 
 crucible and gently heated until sulphur dioxide ceases to be 
 given off. It is then strongly heated after the addition of ammo- 
 nium carbonate and weighed as stannic oxide, Sn0 2 
 
 147. Separation of Lead as Sulphate. Lead is most frequently 
 separated from other metals by precipitation as sulphate. The 
 nitric acid solution is evaporated with excess of sulphuric acid 
 until the nitric acid is completely expelled, which is indicated by 
 the evolution of dense white fumes of sulphuric acid. On dilu- 
 tion with water the lead separates out almost completely. In 
 the absence of much bismuth or iron, a more complete precipita- 
 tion of the lead may be effected by the addition of an equal bulk 
 of alcohol to the dilute sulphuric acid solution. Besides BARIUM 
 and STRONTIUM sulphates which may contaminate the precipitate, 
 BISMUTH may be present as an insoluble basic sulphate. To keep 
 the bismuth in solution considerable sulphuric acid must be present 
 and the precipitate must be washed with dilute (10%; sulphuric 
 acid, which is finally removed by washing with dilute alcohol 
 
 148. Separation of Lead as Chloride from Bismuth. The 
 precipitate of lead sulphate may be tested for bismuth by warming 
 with concentrated hydrochloric acid. After allowing to cool, 
 
SEPARATION OF LEAD. 127 
 
 50 c.c. absolute alcohol is added to the solution of lead chloride, 
 the volume of which should be about 5 c.c. The bismuth chlo- 
 ride dissolves in the alcohol and may be filtered off from the pre- 
 cipitated lead chloride. A small amount of the lead dissolves 
 in the alcohol with the bismuth. The latter may be precipitated 
 by largely diluting the alcoholic solution. This method may 
 also be used for separating the lead and bismuth in the nitric- 
 acid solution of lead and bismuth alloys. Before the addition of 
 the alcohol the nitric acid is expelled by evaporating twice with 
 20 c.c. concentrated hydrochloric acid. The lead chloride may 
 be collected on a Gooch crucible, washed with alcohol, and dried 
 for three hours at 150 and weighed. About 1 mg. of lead chloride 
 remains in the alcoholic solution. The bismuth is precipitated 
 as BiOCl by largely diluting the alcoholic solution and neutralizing 
 most of the hydrochloric a3i.l with ammonia if a large excess is 
 present. It may be washed with water containing a few drops of 
 hydrochloric acid, dried at 110 and weighed. 
 
 149. Separation of Lead from Antimony and Barium. If con- 
 siderable antimony is present the alloy is dissolved in a rather 
 small amount of nitric acid (4 c.c.) with the addition of tartaric 
 acid (10 grams) and water (5 c.c.). The lead is precipitated by 
 the addition of 4 c.c. concentrated sulphuric acid and diluting 
 the solution to about 250 c.c. 
 
 If barium is present, as when minerals are analyzed, so that 
 the sulphate of lead is contaminated with barium sulphate a 
 separation may be effected by digesting the precipitate with 
 ammonium carbonate solution. The sulphate of lead is converted 
 into carbonate, while the barium sulphate is unaffected. The 
 precipitate is filtered off and washed with ammonium carbonate 
 solution, then with water until the wash-water no longer con- 
 tains sulphates. The lead carbonate is then dissolved with dilute 
 acetic or nitric acid. In this method of separation a little lead is 
 lost on account of the slight solubility of lead carbonate in solu- 
 tions of ammonium salts. It may be recovered by passing 
 hydrogen sulphide through the ammonium carbonate solution. 
 
128 ANALYSIS OF ALLOYS. 
 
 EXERCISE 26. 
 Analysis of Soft Solder. 
 
 AUoy of Lead and Tin, Generally containing Small Amounts of Arsenic^ 
 Antimony, Iron, and Zinc. 
 
 150. Solution of the Alloy. One gram of the alloy is weighed out and 
 transferred to a beaker of about 500 c.c. capacity. 10 c.c. of concentrated 
 nitric acid and 5 c.c. of water are added. The beaker is covered with a 
 watch-crystal and heated on the water-bath until the alloy is completely 
 decomposed and the nitrous fumes are entirely expelled. 100 c.c. of water 
 is added and the solution boiled for five minutes and allowed to settle for 
 one hour. The stannic oxide is filtered off and washed with hot water. 
 The moist precipitate may be introduced into a weighed porcelain crucible, 
 the paper burned in the usual manner, and finally heated to redness for ten 
 minutes. 
 
 151. Tin. When the precipitate has been brought to constant weight, 
 it is fused with six times its weight of a mixture of equal parts of sulphur 
 and sodium carbonate. The fused mass is dissolved in hot water and the 
 solution filtered. The insoluble sulphides are washed with hot water and 
 treated with a little dilute hydrochloric acid and the paper washed with water. 
 If COPPER is present it will remain on the paper and the small amount 
 present may be weighed as CuO after burning the paper in a porcelain cruci- 
 ble and igniting the precipitate. The LEAD is precipitated by the addition 
 of a few drops of sulphuric acid and 25 c.c. of alcohol to the solution, which 
 should not exceed 50 c.c. After standing one hour, the precipitate is 
 filtered off on a Gooch crucible, washed with alcohol, dried on the hot plate, 
 and weighed. The filtrate is evaporated until the alcohol is completely 
 expelled. Any IRON present is precipitated with ammonia and weighed. 
 Hydrogen sulphide is passed through the filtrate to precipitate any ZINC 
 present, which is filtered off. The filtrate from the insoluble sulphides 
 will contain the tin as a thiostannate and part of the ANTIMONY present 
 in the alloy as a thioantimonate. The solution is boiled after the addi- 
 tion of caustic soda and hydrogen peroxide until it is nearly decolorized. 
 On acidifying and passing hydrogen sulphide both metals are precipi- 
 tated as sulphides. If antimony is present the metals should be separated 
 by the method given in section 178, page 141. The weight of the im- 
 purities found, computed as oxides, is deducted from the weight of the 
 stannic oxide. 
 
 152. Lead. The filtrate from the stannic oxide is transferred to a porce- 
 lain dish, 5 c.c. concentrated sulphuric acid added, and evaporated until 
 fumes of sulphuric acid are evolved. Cool the dish by floating it in cold 
 water and add cautiously 75 c.c. of water. Stir thoroughly and add 25 c.c. 
 of alcohol. Allow the solution to stand for at least one hour, filter off the 
 lead sulphate on a weighed Gooch crucible, wash with alcohol until free 
 from acid, dry on the hot plate, and weigh. 
 
ANALYSIS OF ROSE'S METAL. 129 
 
 153. Arsenic and Antimony. The alcohol is completely expelled from 
 the filtrate by evaporation and any arsenic present precipitated by passing 
 hydrogen sulphide. If this precipitate is of an orange color instead of pure 
 yellow, antimony is present. It should be filtered off and washed with 
 water containing a little hydrochloric acid until free from iron and hydrogen 
 sulphide. It is then washed with small portions of concentrated ammo- 
 nium carbonate solution until the arsenic sulphide is entirely dissolved. 
 The arsenic is reprecipitated by acidifying the solution with hydrochloric 
 acid and passing hydrogen sulphide. It is filtered off on a Gooch crucible 
 and washed with water containing hydrogen sulphide and a little hydro- 
 chloric acid. The water is removed by alcohol and the precipitate 
 digested with carbon disulphide until sulphur is entirely removed. The 
 arsenic sulphide is dried at 100 and weighed. If antimony is absent the 
 treatment with ammonium carbonate is omitted, the precipitate being 
 filtered off on a Gooch crucible, washed, dried, and weighed. If antimony 
 is present it is ignited and weighed as directed in section 179, page 141. 
 
 154. Iron. A few drops of bromine water are added and the solution is 
 boiled to oxidize the iron and to expel the hydrogen sulphide. The iron is 
 then precipitated by making the filtrate alkaline with filtered ammonia 
 and warming for a few minutes. It is filtered off on a small paper 
 and dissolved by adding a few drops of dilute hydrochloric acid. The 
 paper is washed with about 75 c. c. of water in small portions. The 
 iron is reprecipitated and filtered on the same paper after moistening 
 with a few drops of ammonia. After washing free from chlorides, the 
 moist paper is transferred to the weighed platinum crucible and ignited. 
 
 155. Zinc. Hydrogen sulphide is passed into the combined filtrates to 
 precipitate any zinc present, which is filtered off, washed, and weighed as 
 sulohide. 
 
 EXERCISE 27. 
 
 Analysis of Rose's Metal. 
 
 Alloy of Lead } Bismuth, and Tin, Generally containing Small Amounts of 
 Copper, Arsenic, Antimony, Iron, and Zinc. 
 
 One gram of the metal is weighed out and decomposed with nitric 
 acid as directed in Exercise 26. The STANNIC OXIDE is weighed and the 
 impurities determined as directed in the same exercise. 
 
 156. Lead. To the filtrate from the stannic oxide containing the ni- 
 trates of lead and bismuth, 5 c.c. concentrated sulphuric acid is added. 
 The solution is evaporated in a porcelain dish until sulphuric-acid fumes are 
 given off. The dish may be placed on the hot plate, sand-bath, or wire 
 gauze and the liquid heated to just below the boiling-point to avoid spatter- 
 ing. When the acid becomes concentrated, the heat may be somewhat 
 increased. The hot concentrated solution is diluted by slowly pouring it with 
 constant stirring into about 100 c.c. of water and digested hot for about half 
 an hour with occasional stirring. The lead sulphate is then filtered off on a 
 
130 ANALYSIS OF ALLOYS. 
 
 Gooch crucible, washed with 10% sulphuric acid until the wash-wafer no longer 
 gives a precipitate on making it alkaline with ammonia, adding ammonium 
 carbonate, and warming. The sulphuric acid is then washed out with alcohol. 
 The precipitate is dried and weighed. It is tested fora possible contamination 
 with bismuth as follows : It is dissolved in 5 to 10 c.c. of warm concentrated 
 hydrochloric acid and 50 c. c. of absolute alcohol are added to the solution. 
 After standing for a few moments, the solution, containing the bismuth as 
 chloride, is filtered off. By nearly neutralizing with ammonia, and largely 
 diluting with water, the bismuth is precipitated as oxychloride and may be 
 washed with water containing a few drops of hydrochloric acid, dried, and 
 weighed. 
 
 157. Bismuth. In the filtrate from the lead sulphate, the bismuth is 
 precipitated by just neutralizing with filtered ammonia, adding a few drops 
 of ammonium carbonate, and warming the solution gently for about fifteen 
 minutes. The precipitate is filtered off and washed a few times with water. 
 To free the precipitate from a small amount of basic sulphate it is dissolved 
 in a small amount of dilute nitric acid and reprecipitated. The precipitate 
 is washed with water containing a little ammonium nitrate and dried. It 
 is removed from the paper as completely as possible and placed on a 
 watch-crystal. The paper is replaced in the funnel, moistened with a 
 few drops of dilute nitric acid, and washed with small amounts of warm 
 water. The wash-water is evaporated to dryness in a fairly large weighed 
 porcelain crucible, and the residue ignited until the nitric acid is com- 
 pletely expelled. The main portion of the precipitate is now added, 
 heated with the Bunsen burner and weighed as Bi,,O. r 
 
 158. Copper. If copper is present in the alloy, it will be contained in 
 the two filtrates from the bismuth precipitate. Combine these nitrates, 
 acidify with hydrochloric acid, and concentrate to a convenient bulk. Pass 
 hydrogen sulphide through the warm solution, filter, and wash with water 
 containing hydrogen sulphide. Even if copper is absent, a small black 
 precipitate of bismuth sulphide will be obtained at this point because of the 
 slight solubility of the bismuth hydroxide or carbonate. The precipitate 
 may be tested for bismuth by treating with a little dilute hydrochloric acid 
 and diluting the filtrate. A white precipitate indicates bismuth. The 
 copper sulphide being insoluble in dilute hydrochloric acid remains on 
 the paper and may be ignited together with the paper and weighed as 
 oxide. If ARSENIC, ANTIMONY, IRON, or ZINC are present they are separated 
 and determined by the methods given in Exercises 26 or 28. 
 
 159. Separation of Copper and Cadmium. Cadmium may be 
 separated from copper by passing hydrogen sulphide through or 
 adding ammonium sulphide to a solution of the two metals in 
 potassium cyanide. In such a solution copper sulphide is soluble, 
 while cadmium sulphide is not. Silver, bismuth, and lead must 
 be absent, as these metals would be precipitated with the cad- 
 
ANALYSIS OF WOOD'S METAL. 131 
 
 mium. Cadmium may also be separated electrolytically from 
 copper. The latter metal may be deposited from a solution con- 
 taining 5% of acid by a current having a tension not exceeding 
 1.85 volts. Under these conditions cadmium is not precipitated. 
 
 EXERCISE 28. 
 Analysis of Wood's Metal. 
 
 Alloy of Lead, Bismuth, Tin, and Cadmium, Generally containing Small 
 Amounts of Copper, Arsenic, Antimony, Iron, and Zinc. 
 
 One gram of the metal is weighed, dissolved in nitric acid, and the STANNIC 
 OXIDE weighed and purified by the method given in Exercise 26. The 
 filtrate from the tin is evaporated to dryness on a water-bath. The nitrates 
 are converted into chlorides by evaporating twice on the water-bath to a 
 small bulk after the addition of 20 c.c. of concentrated hydrochloric acid. 
 
 160. Lead. After coding, 25 c.c. absolute alcohol are added. The 
 mixture is stirred and after standing some time the chloride of lead is fil- 
 tered off on a Gooch crucible, and washed with an ice-cold mixture of 
 4 parts of 95% alcohol and 1 part of concentrated hydrochloric acid. 
 It is dried on the hot plate or at 150 for three hours and weighed. 
 
 161. Bismuth. The filtrate is diluted with about one-half liter of water 
 and nearly neutralized with ammonia (about 40 c.c. of dilute ammonia will 
 be required). After standing twenty-four hours the bismuth oxychloride 
 is filtered off on a Gooch crucible, washed with water containing a few drops 
 of dilute hydrochloric acid, dried at 110, and weighed as BiOCl. 
 
 The bismuth may also be precipitated as bismuth hydroxide by vola- 
 tilizing most of the alcohol, neutralizing with ammonia, and warming gently. 
 If IRON is present this precipitate will be reddish. In that case it is best 
 to dissolve it in hydrochloric acid and precipitate the bismuth as oxy- 
 chloride. The bismuth hydroxide is ignited and weighed as oxide, Bi 2 3 . 
 
 162. Cadmium. The filtrate from the bismuth oxychloride is evapo- 
 rated to a bulk of 200 or 300 c.c. If the bismuth has been precipitated by 
 means of ammonia, the filtrate is first acidified with hydrochloric acid and 
 evaporated to a moderate bulk. The solution is saturated with hydrogen 
 sulphide and the precipitate filtered off and washed with water containing 
 hydrogen sulphide. If the cadmium sulphide is dark colored or black, 
 traces of lead or bismuth sulphides may be present because of incomplete 
 separations, or copper may have been present in the alloy. Any arsenic 
 which may have been in the alloy or a trace of tin or antimony will also 
 be present in this precipitate. 
 
 163. Arsenic, Antimony, and Tin. It should be tested for these three 
 elements by pouring over it a few drops of warm potassium or sodium 
 sulphide and washing two or three times with warm water, being careful 
 
132 ANALYSIS OF ALLOYS. 
 
 to stir up the precipitate with the stream of water from the wash-bottle. 
 A precipitate formed on acidifying the filtrate indicates the presence of 
 arsenic, antimony, or tin. If the characteristic orange color of antimony 
 is absent, the supernatant liquid should be decanted and the precipitate 
 warmed with a little concentrated hydrochloric acid. If it dissolves com- 
 pletely, arsenic is absent and the tin may be reprecipitated by diluting and 
 passing hydrogen sulphide. After washing, the moist precipitate with the 
 paper may be burned and the sulphide of tin converted into oxide by ignition. 
 If arsenic or antimony is present, it may be determined as directed in the 
 following exercise. 
 
 164. Separation of Copper and Cadmium. To dissolve out any copper 
 which may be present with the cadmium sulphide, a few drops of potassium 
 cyanide should be poured over the precipitate. It should be thoroughly 
 stirred up with water and washed a few times. If a considerable amount 
 of copper is present, the bulk of the precipitate should be transferred to a 
 beaker by washing out the paper while still in the funnel with a stream cf 
 water. The remainder of the precipitate on the paper is dissolved by 
 washing with a little warm dilute nitric acid. The paper is then thor- 
 oughly washed with small portions of hot water. The washings are allowed 
 to flow into the beaker containing the main portion of the precipitate. The 
 beaker is warmed and more nitric acid is added if necessary to dissolve 
 the precipitate. The solution is neutralized with sodium carbonate and a 
 slight excess of potassium cyanide added. A small white precipitate at this 
 point may be lead or bismuth carbonates, which should be filtered off 
 and determined. On passing hydrogen sulphide through the filtrate, the 
 cadmium is precipitated as sulphide and may be filtered off on a Gooch 
 crucible and washed with water containing a little hydrogen sulphide. It 
 is finally washed with pure water and the free sulphur extracted by washing 
 with alcohol and then with carbon disulphide. The precipitate is dried at 
 100 and weighed. 
 
 165. Copper. The filtrate from the cadmium sulphide contains the copper 
 and is acidified* with sulphuric acid and a little nitric acid and evaporated 
 to fumes. The residue is dissolved in water, filtered if necessary, and the 
 copper precipitated as sulphide. If it is small in amount it may be ignited 
 and weighed as oxide. If considerable copper is present, it must be ignited 
 with sulphur in a stream of hydrogen and weighed as cuprous sulphide, 
 Cu-jS. When much copper is present, it is better to determine it electro- 
 lytically. 
 
 166. Separation of Iron and Zinc. The filtrate from the first precipita- 
 tion with hydrogen sulphide ( 162) contains any zinc or iron which may 
 have been present. These metals may be determined as described for the 
 
 * This should be done under a hood with good draught to avoid any 
 bility of inhaling the very poisonous hydrocyanic-acid fumes. 
 
ANALYSIS OF WOOD'S METAL. 133 
 
 similar filtrate in Exercise 26. They may also be separated in the follow- 
 ing manner : 
 
 The solution is boiled to expel hydrogen sulphide, neutralized with ammo- 
 nia, and acidified with acetic acid. Hydrogen sulphide is passed for some 
 time and the solution allowed to stand for several hours. The clear liquid 
 is carefully decanted through a filter-paper, and after replacing the beaker 
 containing the clear filtrate with another beaker, the sulphide of zinc is 
 brought on the paper and washed with water containing ammonium acetate 
 and acetic acid. The precipitate is dissolved in a little dilute nitric acid 
 and the paper washed with hot water. The solution of the zinc is evaporated 
 to dryness in a weighed porcelain crucible, ignited finally over the blast-lamp 
 to decompose any zinc sulphate which may have been formed, and weighed 
 as oxide. The filtrate is boiled to expel the hydrogen sulphide. A little 
 nitric acid is then added to oxidize the iron, which is precipitated with 
 ammoni# and weighed as oxide. A very convenient method of oxidizing the 
 iron and removing the hydrogen sulphide is by the use of bromine water. 
 The bromine should be added until the solution is colored, indicating com- 
 plete oxidation of the iron and the presence of an excess of bromine. If a 
 solution of bromine in concentrated hydrochloric acid is used a few drops 
 will suffice and the solution will not be diluted to any extent. If manganese 
 is to be removed together with iron, the presence of an excess of bromine 
 is advantageous ; otherwise it must be boiled out. As the bromine oxidizes 
 hydrogen sulphide in the cold, the excess of the latter need not be boiled 
 out. 
 
CHAPTER XI. 
 
 ANALYSIS OF ALLOYS CONTAINING ARSENIC, 
 ANTIMONY, AND TIN. 
 
 SEPARATION OF ARSENIC, ANTIMONY, AND TIN. 
 
 167. F. W. Clarke's Method. A large number of methods 
 have been proposed for the separation of arsenic from antimony 
 and tin. One of the best is that of F. W. Clarke.* It is based 
 on the fact that stannic sulphide is not precipitated by hydrogen sul- 
 phide from a boiling solution of oxalic acid. The sulphides of 
 arsenic dissolve slightly, and the sulphide of antimony still more 
 so, in boiling oxalic-acid solutions, but both of these elements are 
 completely reprecipitated by hydrogen sulphide. Free mineral 
 acids must be absent from the solution. At least 20 parts of 
 crystallized oxalic acid must be present for 1 part of tin, and 
 the solution should be diluted to 125 c.c. for each 0.1 gram of anti- 
 mony present. Hydrogen sulphide is passed through the boiling 
 solution for at least one-half hour. The precipitate must be 
 filtered off immediately, as stannic sulphide separates out on 
 standing. The precipitate generally contains a little stannic 
 sulphide. After filtering off and washing two or three times it is 
 dissolved in a little ammonium sulphide and the solution poured 
 into excess of hot strong solution of oxalic acid. Hydrogen sul- 
 phide is passed through the boiling solution for ten minutes. 
 The precipitate is now free from tin, and is filtered off and washed 
 with hot water. The oxalic-acid solution of the tin is evaporated 
 down with the addition of sulphuric acid until the oxalic acid is 
 decomposed. The tin may then be precipitated with hydrogen 
 sulphide and weighed as oxide. 
 
 1 68. Separation of Arsenic and Antimony. The sulphides of 
 arsenic and antimony may be separated by treatment with a 
 saturated ammonium-carbonate solution which dissolves the arsenic 
 
 *Chem. News, 21, 124. 
 
 134 
 
SEPARATION OF ARSENIC, ANTIMONY, AND TIN. 135 
 
 easily and the antimony only slightly. No more ammonium- 
 carbonate solution than necessary to dissolve the arsenic sulphide 
 should be used. The precipitate, which must first be washed 
 entirely free from hydrogen sulphide, should be treated with small 
 portions of the carbonate solution until no more arsenic is dis- 
 solved, as indicated by the absence of a precipitate on acidifying 
 the filtrate. The arsenic is completely precipitated by acidifying 
 with hydrochloric acid, warming, and passing hydrogen sulphide. 
 This method of separating arsenic and antimony is especially 
 applicable if the amount of arsenic present is small. 
 
 A better separation of the arsenic and antimony may be effected 
 by dissolving the sulphides in concentrated hydrochloric acid with 
 the addition of small amounts of potassium chlorate. The solution 
 is warmed to expel the chlorine, but not to boiling, so as not to 
 melt the sulphur which is generally liberated. It is filtered through 
 asbestos, which is washed with concentrated hydrochloric acid. 
 It is cooled in ice-water and hydrogen sulphide passed for one hour. 
 The arsenic is precipitated as As 2 S 5 and is washed with strong 
 hydrochloric acid and then with hot water. Antimony remains 
 in solution as trichloride. This method also serves for the SEPA- 
 RATION of ARSENIC and TIN, the latter behaving like antimony. 
 According to Neher,* for this separation the solution should 
 contain 1 volume of water to 2 volumes of concentrated hydro- 
 chloric acid (sp. gr. 1.20). If a stronger acid is used, the precipi- 
 tation of the arsenic is hindered. 
 
 The arsenic is best dissolved and reprecipitated as magnesium 
 ammonium arsenate. If hydrogen peroxide is at hand which is free 
 from phosphoric acid, aluminium, etc., the arsenic sulphide may be 
 dissolved in a warm mixture of ammonia and hydrogen peroxide. 
 A very excellent solvent is a solution of sodium peroxide in water. 
 The arsenic acid may then be precipitated by means of magnesia 
 mixture. The arsenic sulphide may also be dissolved in strong 
 hydrochloric acid and potassium chlorate or in fuming nitric 
 acid. These oxidizing solutions should be warmed, but not high 
 enough to melt the sulphur which usually separates out. The 
 latter becomes very nearly white when all of the arsenic has been 
 dissolved out. If the sulphur is large in amount or otherwise 
 
 * Zeit. f. Anal. Chem., 1893, p. 45. 
 
136 ANALYSIS OF ALLOYS. 
 
 troublesome, liquid bromine may be added in small portions to 
 dissolve it. This method of precipitation also serves to separate 
 ARSENIC from ANTIMONY. Tartaric acid is added to prevent the 
 precipitation of the antimony. 
 
 169. The Separation of the Antimony and Tin in the hydro- 
 chloric-acid filtrate from the arsenic is effected as follows: The 
 hydrogen sulphide is decomposed by the addition of a little potas- 
 sium chlorate or bromine. The antimony is precipitated by 
 digesting the somewhat diluted solution on the water-bath with 
 the addition of pure iron. Iron produced by reduction in hydro- 
 gen is best, and in any event it must be free from phosphorus. 
 The precipitation of the antimony is complete in about one-half 
 hour. It forms a black powder. The tin is reduced to the stan- 
 nous- condition. The solution is heated until the iron is nearly 
 dissolved. The antimony is then filtered off on an asbestos filter 
 over which a little iron has been sprinkled, and is washed with 
 boiled water containing considerable hydrochloric acid. 
 
 The TIN is precipitated by passing hydrogen sulphide through 
 the filtrate, which for this purpose is diluted, and the bulk of the 
 acid neutralized with ammonia. The stannous sulphide is washed 
 with water containing hydrogen sulphide and a few grams of 
 ammonium sulphate. It is dried, transferred to a weighed cruci- 
 ble, ignited, and weighed as stannic oxide. 
 
 The ANTIMONY is dissolved in hydrochloric acid with the addi- 
 tion of a little potassium chlorate. The solution is diluted after 
 the addition of tartaric acid and the antimony precipitated as 
 sulphide and weighed as oxide. 
 
 170. H. Rose's Separation of Antimony from Arsenic and Tin. 
 This method of separation is based on the insolubility of sodium 
 metantimonate. If the substance is metallic, it may be con- 
 veniently oxidized by treatment with concentrated nitric acid 
 in a porcelain evaporat ing-dish or crucible. The material is 
 dried on the water-bath and transferred to a silver crucible, the 
 porcelain dish being rinsed with caustic-soda solution. The con- 
 tents of the crucible are evaporated to dryness, eight times the 
 bulk of the mass of solid caustic soda added, heated cautiously to 
 avoid spattering until all water is expelled and then fused for some 
 time. The mass is allowed to cool, and is treated with hot water 
 until the fused material is entirely disintegrated. One-third the 
 
SEPARATION OF ARSENIC, ANTIMONY, AND TIN. 137 
 
 volume of alcohol is added and the solution allowed to stand for 
 twenty-four hours with frequent stirring. 
 
 If the arsenic, antimony, and tin are obtained by separation 
 as sulphides, they may be dissolved in solution of sodium sulphide 
 and the cold-saturated solution treated with small portions of 
 sodium peroxide until the solution is colorless and oxygen is 
 copiously evolved on the further addition of the peroxide. The 
 solution is boiled and after cooling one-third the volume of alco- 
 hol is added. The solution is then allowed to stand for twenty- 
 four hours. 
 
 The precipitate obtained by either method is washed, first 
 with dilute alcohol composed of equal volumes of alcohol and 
 water and then with a mixture of one volume of water and three 
 volumes of alcohol. A few drops of sodium carbonate solution are 
 added to each of the washing fluids. If much tin is present, it is 
 advisable to fuse the metantimonate a second time with soda. 
 The antimony is dissolved in hydrochloric acid containing a little 
 tartaric acid, precipitated with hydrogen sulphide, and weighed 
 as oxide. 
 
 The filtrate containing the arsenic and tin is acidified with 
 hydrochloric acid and evaporated until all of the alcohol is ex- 
 pelled. The solution is warmed and hydrogen sulphide passed 
 until precipitation is complete. The arsenic and tin may then be 
 separated by one of the methods already given. 
 
 171. Electrolytic Separation of Antimony from Arsenic and 
 Tin. Antimony may also be separated electrolytically from 
 arsenic and tin, according to Classen.* The arsenic must be 
 present as arsenic acid, otherwise a portion of it will be precipi- 
 tated with the antimony, while the remainder will be oxidized to 
 arsenic acid. A saturated solution of sodium sulphide must first 
 be prepared. If the commercial sodium sulphide is used it must 
 be dissolved in water and the solution saturated with hydrogen 
 sulphide which has been washed with water and then passed 
 through several tubes filled with cotton wool. While passing 
 hydrogen sulphide the air should be excluded. The solution is 
 filtered and rapidly evaporated in a large platinum or porcelain 
 
 * Ber., 17, 2245; 13, 1110; 23, 2160. 
 
138 ANALYSIS OF ALLOYS. 
 
 dish over the free flame until a thin skin begins to form on top. 
 The hot solution is poured into glass-stoppered bottles, which 
 must be completely filled and immediately well stoppered. The 
 sodium sulphide may also be prepared by completely saturating 
 a solution of pure caustic soda with hydrogen sulphide, filtering, 
 and evaporating down, as already directed. 
 
 The sulphides of arsenic, antimony, and tin are transferred to 
 a weighed platinum dish of about 150 c.c. capacity and dissolved 
 in 80 c.c. of the sodium-sulphide solution. Add enough concen- 
 trated solution of pure caustic soda to furnish 1 to 2 grams of 
 NaOH. The caustic-soda solution must be clear and give no 
 precipitate with the sodium-sulphide solution. The dish is gently 
 warmed to aid solution. It is electrolyzed at a temperature of 
 50 to 60 with a current of 0.5 ampere. Precipitation of the 
 antimony is complete in two hours. The watch-crystal is washed 
 with a little water, which is allowed to run down the positive 
 electrode. The solution is removed without interrupting the 
 current, the antimony washed with water and then with alcohol, 
 dried at 80 to 90, and weighed. 
 
 The TIN may be determined by acidifying the filtrate with 
 hydrochloric acid, concentrating to a bulk of 200 or 300 c.c., 
 filtering, if ne'cessary, and passing hydrogen sulphide. The pre- 
 cipitate is washed with water containing hydrogen sulphide and a 
 few grams of ammonium sulphate. It is weighed as stannic oxide. 
 
 172. Separation of Arsenic. Arsenic is very readily separated 
 from other metals as arsenious chloride, AsCl 3 . This compound is 
 very volatile. When pure it boils at 130. It is rapidly volatilized 
 when its solution in hydrochloric acid is boiled. Even at tem- 
 peratures considerably below the boiling-point the arsenic slowly 
 volatilizes. When it is present in its higher state of oxidation, 
 the arsenic is not volatilized on boiling its hydrochloric-acid 
 solution. Various reducing agents have been used to convert the 
 arsenic into the arsenious chloride. The most common among 
 these are ferrous an.d cuprous salts and hydrogen sulphide. If 
 small amounts of arsenic are present the solution may be distilled 
 after the addition of concentrated hydrochloric acid. On repeat- 
 ing the distillation once or twice after the addition of concentrated 
 hydrochloric acid all of the arsenic will be carried over. If much 
 
SEPARATION OF ARSENIC, ANTIMONY, AND TIN. 139 
 
 arsenic is present, the solution must be saturated with hydrochloric- 
 acid gas and the distillation carried out in a current of this gas. 
 
 173. Dissolving Arsenic Compounds. Various methods of de- 
 composing the ore or other material have been proposed, the 
 objects sought being to prevent the loss of arsenic as trichloride 
 and to leave the hydrochloric-acid solution as free as possible from 
 oxidizing material. If the material is soluble in hydrochloric 
 acid, it may be digested cold with concentrated hydrochloric acid 
 for some time. Liquid bromine may then be added and the solu- 
 tion warmed, at first gently and then more strongly, until the excess 
 of bromine is expelled. The material may also be brought into 
 solution by treatment with concentrated nitric acid, the excess 
 being expelled by evaporation after the addition of a little sulphuric 
 acid. The residue may then be dis- 
 
 solved in concentrated hydrochloric 
 acid. 
 
 174. Distillation of the Arsenic. If 
 the amount of arsenic present is small, 
 the solution may be decanted into the 
 distilling-fl ask of the apparatus shown in 
 Fig. 20 and the insoluble residue washed 
 with concentrated hydrochloric acid. 
 About 5 grams of ferrous sulphate, or, 
 still better, ferrous chloride or cuprous 
 chloride, are added and 50 to 75 c.c. 
 concentrated hydrochloric acid. The 
 solution is heated to boiling, and when 
 the liquid has boiled down to a volume 
 of about 40 c.c., 50 c.c. concentrated 
 hydrochloric acid is poured through 
 the funnel and the solution again 
 boiled down to 40 c.c. The distillate 
 is received in a beaker or flask con- 
 taining about 250 c.c. of hydrogen- 
 sulphide water. After the second dis- 
 
 tillation this beaker is removed and replaced by another contain- 
 ing hydrogen-sulphide water. The distillation is repeated after 
 the addition of 50 c.c. of concentrated hydrochloric acid to the 
 
 FIG. 20. 
 
140 ANALYSIS OF ALLOYS. 
 
 flask. A trace of arsenic is sometimes obtained by a fourth 
 distillation. 
 
 When LARGE AMOUNTS of ARSENIC must be distilled, the drop- 
 ping-funnel is replaced by a glass tube leading to a flask in which 
 hydrochloric-acid gas is generated. This gas is very readily 
 obtained by allowing concentrated sulphuric acid to drop into 
 concentrated hydrochloric acid. After transferring the arsenic 
 solution to the distillation flask it is saturated with hydrochloric 
 acid by passing the gas into the solution, which has been cooled to 
 zero by immersing the flask in a vessel filled with cracked ice. 
 When the gas is no longer absorbed, but bubbles through the 
 solution, the ice is removed and the solution brought to a boil. 
 The stream of hydrochloric-acid gas is allowed to flow and the 
 distillation is continued until the volume of the solution is reduced 
 to about 40 c.c. The arsenic may be absorbed in hydrogen-sul- 
 phide water and the precipitate of arsenious sulphide filtered off 
 and weighed as such. 
 
 EXERCISE 29. 
 Analysis of Type Metal. 
 
 Alloy of Copper, Lead, Antimony, Tin, with Small Amounts of Iron and 
 
 Arsenic. 
 
 175. Solution of the Alloy. To 1 gram of the alloy, which has been 
 cut into small shavings with a clean knife, or sampled by means of a 
 clean hack saw producing fine "sawings," is added 15 c.c. concentrated 
 hydrochloric acid. The solution is gently warmed on the water-bath and 
 a drop or two of concentrated nitric acid is added occasionally until 
 solution is affected. All of the metals will be converted into chlorides 
 which will remain in solution with the possible exception of lead chloride. 
 An excess of nitric acid is to be avoided as it tends to form insoluble metas- 
 tannic acid, which can be readily distinguished from the crystalline lead 
 chloride. If metastannic acid forms, the operation must be repeated, using 
 less nitric acid or adding it less frequently. After a few trials the correct 
 method of adding the nitric acid is soon acquired. 
 
 176. Lead. The solution is allowed to cool and then stand at least 
 hour or better over night to allow the lead chloride to crystallize out. 
 Ten times the volume of absolute alcohol is then added in several por- 
 tions. After standing for about half an hour, the lead chloride is 
 filtered off on a Gooch crucible, washed with a mixture of 4 parts of 95% 
 alcohol and 1 part of concentrated hydrochloric acid, and finally with 
 pure alcohol. It is dried for three hours at 150 and weighed. The great 
 advantage of this method of separating the lead is that the very trouble- 
 some treatment of the sulphides of the metals present with sodium or 
 potassium sulphide is avoided. The most difficult part of the operation 
 is the solution of the alloy. 
 
ANALYSIS OF TYPE METAL. 141 
 
 177. Copper and Iron The filtrate from the lead chloride is heated 
 until the alcohol is expelled. Two grams of tartaric acid and an excess 
 of ammonia are added and the solution warmed until the precipitate 
 dissolves. By the addition of five c.c. of saturated hydrogen-sulphide 
 water, the copper and the small amount of lead still unprecipitated 
 as well as a trace of iron which may be present may be precipitated with- 
 out bringing down any of the tin and antimony. The solution is 
 warmed and when the dark colored precipitate has settled, one c.c. of 
 the hydrogen-sulphide water is added to the clear supernatant liquid. 
 If no further precipitate is produced, the solution is filtered and the 
 precipitate washed with water containing hydrogen sulphide. 
 
 The precipitate is dissolved in a little warm dilute nitric acid and the 
 lead separated as sulphate, the nitric acid being expelled by evaporation- 
 after the addition of sulphuric acid. The copper is precipitated from the 
 filtrate as sulphide and if small in amount may be ignited and weighed 
 as oxide. If considerable copper is present it must be weighed as 
 sulphide or without precipitation as sulphide may be separated electro- 
 lytically from the iron. One or two c.c. concentrated nitric acid are added 
 and a current of one-half ampere passed until all the copper is precipitated. 
 The iron may then be precipitated with ammonia and weighed as oxide. 
 
 178. Separation of Antimony and Tin. The solution of antimony and 
 tin is acidified with hydrochloric acid, hydrogen sulphide passed, and the 
 precipitate filtered off and washed two or three times. A hole is made in 
 the point of the filter-paper by means of a glass rod and the bulk of the 
 precipitate washed into a beaker with a little water. Warm dilute hydro- 
 chloric acid is poured over the paper to dissolve the portion of the precipi- 
 tate still adhering to the paper. The precipitate in the beaker is dissolved 
 by warming and adding concentrated hydrochloric acid. The hydrogen 
 sulphide is decomposed by the addition of a crystal of potassium chlorate 
 and warming. Some pure metallic iron is added and the solution heated 
 on the water-bath for about one-half hour or until the iron is nearly dissolved. 
 The precipitated antimony is filtered off on a Gooch crucible, a little iron 
 having been sprinkled on the asbestos. The precipitate is washed with 
 boiled water to which considerable hydrochloric acid has been added. 
 
 179. The antimony is dissolved in hydrochloric acid to which a little potas- 
 sium chlorate has been added. The solution is warmed to expel chlorine and, 
 after the addition of tartaric acid and water, hydrogen sulphide is passed. 
 The antimony sulphide is filtered off and washed with water containing a little 
 hydrogen sulphide. The moist precipitate is rinsed into a capacious porce- 
 lain crucible with water. The small portion still adhering to the paper 
 is dissolved in a little warm ammonium sulphide and the solution allowed 
 to flow into the crucible. The solution is evaporated on the water-bath 
 after the addition of a few cubic centimeters of concentrated nitric acid. If 
 sulphur separates, a little liquid bromine is added when the solution has 
 become quite concentrated. When the globule of sulphur has disappeared, 
 expel the excess of nitric acid by heating on the hot plate or with the Bunsen 
 
142 ANALYSIS OF ALLOYS. 
 
 burner, finally heating to full redness. Cool a little, sprinkle some ammo- 
 nium carbonate over the precipitate, and ignite again to completely expel 
 sulphuric acid and weigh as antimony tetroxide, Sb 2 O 4 . 
 
 1 80. Tin. To precipitate the tin in the filtrate from the antimony the 
 excess of hydrochloric acid is neutralized with ammonia, the solution diluted 
 somewhat, warmed, and hydrogen sulphide passed until the tin is entirely 
 precipitated. The stannous sulphide is washed with water containing 
 hydrogen sulphide and a few grams of ammonium sulphate. It is dried 
 and detached from the paper which is burned. The precipitate and the 
 ash are placed in a weighed porcelain crucible and heated very gently 
 with free access of air until sulphur dioxide ceases to be given off. The 
 oxidation may be assisted by the addition of a few drops of nitric acid. 
 Finally the precipitate is strongly heated to expel sulphuric acid, which 
 is completely removed by the addition of a little ammonium carbonate 
 and again igniting. It is weighed as stannic oxide SnO 2 . 
 
 181. Arsenic. As only a trace of arsenic is present, a 5- or 10-gram portion 
 of the alloy should be taken for its determination. Dissolve in hydrochloric 
 acid and potassium chlorate and warm to expel the chlorine. Filter off the 
 lead chloride on asbestos and wash a few times with dilute hydrochloric acid. 
 Add one-third the volume of concentrated hydrochloric acid and pass hydrogen 
 sulphide. Filter off the precipitate consisting of the sulphides of copper and 
 arsenic on asbestos, wash with hot water containing hydrogen sulphide and 
 a little hydrochloric acid. Dissolve the arsenic sulphide by washing the pre- 
 cipitate with a little warm dilute ammonia. Evaporate the solution nearly to 
 dryness in a porcelain dish. Oxidize the arsenic by warming with concen> 
 trated nitric acid, dilute the solution somewhat, neutralize with filtered 
 ammonia, and add magnesia mixture. After standing twenty-four hours 
 filter, wash, ignite, and weigh as magnesium pyroarsenate according to the 
 directions given in Chapter VII, page 88. 
 
 EXERCISE 30. 
 Analysis of Britannia Metal. 
 
 Alloy of Tin, Antimony, and Copper, with Small Amounts of Bismuth, 
 Lead, and Iron. 
 
 182. Decomposition of the Alloy by Means of Chlorine. Alloys containing 
 a large percentage of tin are best decomposed by a stream of chlorine. 
 The method is applicable to alloys containing less than 15% of lead and 
 copper. 
 
 A hard-glass combustion-tube 70 cm. long is taken and one end drawn 
 out, making a small tube 20 cm. long, which is bent at right angles. 
 This small tube is connected by means of a cork stopper with a Peligot 
 tube the bulbs of which are nearly filled with dilute hydrochloric acid (1 : 3) 
 
ANALYSIS OF BRITANNIA METAL. 143 
 
 containing about 1 gram of tartaric acid. A second Peligot tube is con- 
 nected with the first and contains a solution of caustic soda (1 : 3). The 
 chlorine is evolved in a 2-liter flask containing pieces of pyrolusite, over 
 which concentrated hydrochloric acid is poured. The flask is heated on a 
 water-bath. The chlorine is passed through a wash-bottle containing 
 water and then through two wash-bottles containing sulphuric acid. It 
 is then passed into the combustion-tube, connection being made by means 
 of a cork stopper. Wherever rubber is used for making connections, it 
 must be well coated with paraffine. This is also advisable for the cork 
 stoppers. The chlorine is not allowed to pass into the combustion-tube 
 until all of the air has been displaced from the flask and the wash-bottles. 
 All escaping chlorine should be absorbed in caustic-soda solution. 
 
 One gram of the alloy in fine turnings is weighed out and placed in a 
 porcelain boat which is placed in the middle of the combustion-tube. The 
 chlorine is first allowed to act on the alloy in the cold. When no further 
 action is observed, the part of the tube in which the boat is situated is 
 heated gently with the Bunsen burner, and then more strongly until the 
 contents of the boat fuse. The CHLORIDES of MERCURY, BISMUTH, ARSENIC, 
 ANTIMONY, and TIN VOLATILIZE and are driven out of the tube by heating 
 it gently from the boat to the end which is drawn out. These chlorides 
 are absorbed in the hydrochloric acid contained in the first Peligot tube, 
 while the excess of chlorine is absorbed in the caustic-soda solution con- 
 tained in the second Peligot tube. 
 
 The chlorine in the apparatus is then displaced by means of a stream 
 of dry air or carbon dioxide, the chlorine generator having been removed. 
 The apparatus is disconnected, the boat containing the CHLORIDES of 
 COPPER, LEAD, and IRON is placed in a porcelain dish, and the tube washed 
 out with hot water which is allowed to flow into the dish containing the boat. 
 Hydrochloric acid is added and the dish warmed until the contents of the 
 bout are dissolved. The latter is removed and washed. 
 
 183. Lead, Copper, and Iron. The lead is precipitated by evaporation 
 with sulphuric acid and diluting and is filtered off and weighed as sulphate. 
 The copper is precipitated by means of hydrogen sulphide and weighed as 
 sulphide or determined electrolytically from a nitric acid solution. The 
 iron is precipitated by means of ammonia and weighed as oxide. 
 
 The contents of the first Peligot tube are poured into a beaker and the 
 Peligot tube well washed out with water to which hydrochloric acid is 
 added if necessary. The solution is warmed and hydrogen sulphide passed 
 until precipitation is complete. The filtrate should be heated to boiling, 
 strong hydrochloric acid added, and hydrogen sulphide passed again to 
 insure complete precipitation of the arsenic. 
 
 184. Bismuth. If the sulphide precipitate is dark colored, bismuth is 
 present. The precipitate is washed into a beaker, ammonium sulphide 
 added, and the solution warmed. The solution is filtered through the same 
 paper and the precipitate washed with warm water containing a little 
 
144 ANALYSIS OF ALLOYS. 
 
 ammonium sulphide. The bismuth sulphide is dissolved in a little warm 
 dilute nitric acid and the paper washed. The bismuth is precipitated with 
 ammonia and ammonium carbonate, ignited, and weighed as oxide, Bi 2 3 . 
 
 185. Separation of Tin from Arsenic and Antimony. The ammonium- 
 sulphide solution of arsenic, antimony, and tin is poured with vigorous 
 stirring into a hot solution of 25 grams of oxalic acid in 200 c.c. of water. 
 The solution is heated to boiling and hydrogen sulphide passed for about 
 fifteen minutes. The precipitate is filtered off immediately and washed 
 with hot water containing hydrogen sulphide. It is dissolved in ammo- 
 nium sulphide and the treatment with hot oxalic acid and hydrogen sulphide 
 repeated. 
 
 186. Tin. The oxalic-acid solution of tin is evaporated down, with the 
 addition of 5 c.c. concentrated sulphuric acid, to fumes. The solution 
 is cooled, cautiously diluted with water, and hydrogen sulphide passed to 
 insure complete precipitation of the tin. Wash the precipitate with water 
 containing ammonium acetate and a little acetic acid, dry, ignite, and 
 weigh as stannic oxide SnO 9 . 
 
 187. Arsenic and Antimony. The precipitate of arsenic and antimony 
 sulphides is treated with a little concentrated ammonium-carbonate solu- 
 tion and washed to remove arsenic. The antimony is then weighed as 
 oxide according to the directions given in Exercise 29, page 141. The 
 arsenic is determined according to the directions given in the same 
 exercise. 
 
CHAPTER XII. 
 
 ANALYSIS OF ALLOYS CONTAINING IRON, 
 NICKEL, AND ZINC. 
 
 188. Separation of Zinc as Sulphide. Zinc may be separated 
 very completely from most metals as sulphide. Hydrogen sul- 
 phide passed through the moderately acid solution separates 
 silver, lead, mercury, bismuth, copper, cadmium, arsenic, antimony, 
 and tin from zinc. On neutralizing the solution almost com- 
 pletely, or making it acid with acetic or citric acid, or a drop or 
 two of a mineral acid, and passing hydrogen sulphide, zinc is pre- 
 cipitated as sulphide, while cobalt, nickel, manganese, aluminium, 
 and iron remain in solution as well as the alkaline-earth metals and 
 the alkalies. Zinc may be separated from chromium and aluminium 
 by adding sodium sulphide to a solution of these metals containing 
 tartaric acid made alkaline with ammonia. 
 
 Although the separation of zinc as sulphide from other metals 
 is very perfect and the metal can be weighed with the greatest 
 accuracy as sulphide, the method is not very largely used. This 
 arises from the fact that zinc sulphide is very difficult to filter 
 and wash because of its tendency to pass through the filter-paper. 
 Ignition of a precipitate with sulphur in a stream of hydrogen is 
 also somewhat troublesome. The latter difficulty may be obviated 
 by dissolving the sulphide in hydrochloric or nitric acid and pre- 
 cipitating the zinc as carbonate and weighing as oxide or precipi- 
 tating as zinc-ammonium phosphate and weighing as zinc pyrophos- 
 phate. The zinc sulphide is washed more readily if hydrogen 
 sulphide and ammonium sulphate or acetate are present in the 
 wash-water. 
 
 189. Saparation of Zinc as Zincate. Zinc may also be sepa- 
 rated from cobalt, nickel, iron, and manganese by pouring the acid 
 solution of these metals into excess of caustic-soda solution. The 
 zinc hydroxide first precipitated redissolves in the excess of alkali, 
 
 145 
 
146 ANALYSIS OF ALLOYS. 
 
 forming sodium zincate, while the hydroxides of the other metals 
 are insoluble in the caustic alkali and may be filtered off. Before 
 filtering, the solution should be diluted to precipitate a small 
 amount of cobalt, which dissolves in concentrated caustic-soda 
 solution. The concentrated alkali is also liable to ruin the 
 filter-paper. 
 
 The zinc may be precipitated as sulphide by passing hydrogen 
 sulphide through the filtrate. , The sulphide obtained in this 
 manner from an alkaline solution coagulates and settles readily 
 and does not tend to pass through the filter-paper. If the hydro- 
 gen sulphide is passed too long the zinc sulphide redissolves.* 
 The presence of sodium or ammonium chloride tends to prevent 
 this action. The danger may be avoided by adding hydrogen 
 sulphide water until no more precipitate forms. The alkali is 
 washed out of the precipitate with considerable difficulty, how- 
 ever, and the precipitate also carries down silica. After igniting 
 with sulphur in a stream of hydrogen the alkali may be extracted 
 with hot water and the zinc sulphide again ignited and weighed 
 as sulphide. For determining silica, the zinc sulphide is dissolved 
 in hydrochloric acid and the solution evaporated to dry ness. The 
 zinc chloride is dissolved hi water and the silica filtered off, ignited, 
 and weighed. 
 
 EXERCISE 31. 
 Analysis of Brass or Bronze. 
 
 Attoy of Lead, Copper, Tin, and Zinc, Small Amounts of Arsenic and Iron. 
 
 190. Solution of the Alloy. Weigh out 1 gram of the alloy and place in a 
 300-c.c. beaker, add 10 c.c. concentrated nitric acid and 5 c.c. water. Cover 
 the beaker with a watch-crystal and place in a dish of cold water. After 
 one-half hour place the beaker on the water-bath and evaporate the solu- 
 tion to dryness. 100 c.c. of boiling water and a few drops of nitric acid are 
 added and the solution boiled for five minutes. 
 
 191. Tin. The stannic oxide is filtered off and washed with hot water. 
 The moist precipitate is introduced into a weighed porcelain crucible and 
 the paper burned in the usual manner. If the amount of tin is small (less 
 than l%)it is weighed at this point, otherwise it is fused with six times its 
 weight of a mixture of equal parts of sulphur and sodium carbonate. The 
 fused mass is dissolved in hot water and the solution filtered. The copper, 
 lead, and iron which were carried down with the stannic oxide will remain 
 on the paper as sulphides, while the filtrate will contain all of the tin and 
 
 * L. W. McCay, Jour. Am. Chem. Soc., 30. 376. 
 
ANALYSIS OF BRASS OR BRONZE. 147 
 
 any arsenic or antimony which may have been present. The insoluble 
 sulphides are dissolved in a little nitric acid, the paper washed, and the 
 solution added to the nitrate from the stannic oxide. 
 
 If arsenic and antimony are absent, the tin may be precipitated out of 
 the sodium sulphide solution and weighed. The excess of sulphur should 
 first be removed frcm the solution by heating to boiling after the addition of 
 caustic soda and then adding hydrogen peroxide in small quantities until 
 the solution is nearly decolorized. It is then acidified with hydrochloric 
 acid while stirring constantly, heated, and hydrogen sulphide passed. The 
 stannic sulphide is washed with hot water containing ammonium acetate 
 and a little acetic acid. It is ignited and weighed as stannic oxide in the 
 usual manner. 
 
 192. Arsenic and Antimony. If arsenic is present in the alloy, a small 
 amount of this element will be present in the sodium sulphide solution of 
 the tin and will be precipitated with the stannic sulphide. It may be 
 removed by treating the precipitate with a little concentrated solution 
 of ammonium carbonate and washing. The solution of arsenic should be 
 added to the nitric acid solution of the alloy. 
 
 If antimony is also present in the alloy, the sulphides of arsenic, anti- 
 mony, and tin must be separated by one of the methods given in 
 Chapter XI, page 141. 
 
 193. Lead. To the filtrate from the stannic oxide 5 c.c. concentrated 
 sulphuric acid are added and the solution evaporated in a porcelain dish 
 until the nitric acid is entirely expelled and white fumes of sulphuric acid 
 are given off. The solution is cooled by floating the dish on cold water 
 and diluted with 75 c.c. of water. After stirring thoroughly, 25 c.c. alcohol 
 are added and the solution allowed to stand for at least one hour. The lead 
 sulphate is filtered off on a Gooch crucible, washed with water containing 
 about 1% of sulphuric acid and 25% of alcohol and then with pure alcohol 
 until free from acid. It is dried on the hot plate and weighed. 
 
 194. The copper is best determined electrolytically. The filtrate from 
 the lead sulphate is heated on the hot plate until most of the alcohol is 
 expelled. Two c.c. concentrated nitric acid are added and the warm solu- 
 tion (about 60) electrolyzed with a current of one-half to one ampere for 
 about six hours. 
 
 Hydrogen sulphide is passed through the acid filtrate from the copper 
 to precipitate traces of arsenic, antimony, or unseparated tin which may be 
 present. If more than traces are found, the metals must be separated and 
 determined by the methods given in the preceding exercises. When the 
 amount of copper is large, as is generally the case, it is advisable to divide 
 the solution into two portions for the electrolysis, as about 300 mg. of 
 copper is generally sufficient for a good determination. The solution may 
 be divided by weighing it and then pouring out about half of it and again 
 weighing or the solution may be diluted to a known volume as 250 or 500 c.c. 
 and a portion measured out. The copper may be determined in each 
 portion and the filtrates combined fcr the zinc determination. For the 
 
148 ANALYSIS OF ALLOYS. 
 
 duplicate zinc determination the copper may be precipitated as sulphide, 
 which is filtered off, well washed, and discarded. 
 
 195. Iron. The filtrate from the copper is boiled to expel hydrogen 
 sulphide and a little nitric acid added to oxidize the iron, which is precipi- 
 tated with ammonia and weighed as oxide. If more than a small amount 
 of iron is present, the precipitate must be redissolved and reprecipitated to 
 completely separate it from the zinc. 
 
 196. Zinc. The filtrate from the ircn is evaporated to small bulk and 
 the zinc precipitated and weighed as pyrophosphate. The zinc may also be 
 precipitated and weighed as sulphide. 
 
 EXERCISE 32. 
 
 Analysis of German Silver. 
 
 AUoy of Copper, Zinc, and Nickel, with Small Amounts of Lead, Iron, and Tin 
 
 One gram of the alloy is weighed out and dissolved in nitric acid as 
 directed in the preceding exercise. The tin, lead, and copper are deter- 
 mined as directed in the same exercise. 
 
 Hydrogen sulphide is passed through the acid filtrate from the copper 
 to precipitate traces of arsenic, antimony, tin, or unseparated copper which 
 rr ay be present. If more than traces are found, the metals must be sepa- 
 rated and determined by the methods given in the preceding exercises. 
 
 197. Zinc. The filtrate is boiled until the hydrogen sulphide is expelled 
 and the solution concentrated to a small bulk and the acid nearly neutral- 
 ized with caustic soda. Five to ten grams of caustic soda are dissolved in 
 about 50 c.c. of water and the solution of zinc and nickel added slowly 
 with constant stirring. The solution is diluted with an equal bulk of water 
 and the precipitate filtered off and washed. The zinc in the filtrate is pre- 
 cipitated with hydrogen sulphide, filtered off, and washed free from alkali. 
 The zinc sulphide is dried and detached from the paper as completely as 
 rozsible. 
 
 The portion still adhering to the paper is dissolved in nitric acid and the 
 solution evaporated to dryness in a porcelain crucible. The remainder 
 of the precipitate is added and the whole ignited with sulphur in a stream 
 of hydrogen. If the precipitate is small it need not be dried, but is imme- 
 diately dissolved in nitric acid and after evaporation converted into sul- 
 phide. The sulphide is tested for alkali by digestion with hot water. If 
 alkali is found it must be completely extracted and the sulphide again 
 weighed after ignition with sulphur in hydrogen. The precipitate is then 
 dissolved in nitric acid and the solution evaporated to dryness. The 
 zinc nitrate is dissolved in water and the silica filtered off, washed, ignited, 
 and weighed. The zinc sulphide may also be dissolved in hydrochloric 
 acid, the zinc precipitated as zinc ammonium phosphate and weighed as 
 pyrophosphate. 
 
 198. Iron and Nickel. If iron is absent, the nickel hydroxide may be 
 washed and after transferring the precipitate to a weighed porcelain crucible 
 
ANALYSIS OF GERMAN SILVER. 149 
 
 and burning the paper it may be reduced to metallic nickel by heating in a 
 stream of hydrogen and weighed. If iron is present, the precipitate is 
 dissolved in hydrochloric acid and the iron precipitated with ammonia. 
 Unless a very small amount is present it must be redissolved and repre- 
 cipitated, and after washing is ignited and weighed as oxide. The nickel 
 is then reprecipitated as hydroxide by means of an excess of caustic soda, 
 reduced in a stream of hydrogen and weighed as the metal. 
 
 EXERCISE 33. 
 Analysis of Manganese-Phosphorus-Bronze. 
 
 Alloy of Copper, Lead, Tin, Zinc, Manganese, Phosphorus (less than 1%), 
 
 Traces of Iron. 
 
 199. Solution. One gram of the alloy is weighed out and dissolved in 
 nitric acid as directed in Exercise 31, page 146. Nearly all of the phosphorus 
 remains with the stannic oxide as a phosphate. After fusing the impure 
 precipitate and separating the impurities as given in Exercise 31, and pre- 
 cipitating the tin as sulphide, the solution containing only the phosphorus 
 as phosphoric acid is discarded, as this element is determined in a separate 
 portion of the alloy. 
 
 200. Lead, Copper, and Zinc are determined as given in Exercise 31. 
 The phosphoric acid which did not rcirain with the stannic oxide will be 
 present in the alkaline solution of the zinc. This element should therefore 
 be precipitated and weighed as pyrophosphate. 
 
 201. Iron. In order to separate manganese and iron from zinc, bromine 
 or hydrogen peroxide is added to the nitrate from the copper. The solu- 
 tion is boiled and excess of ammonium added to redissolve any zinc phos- 
 phate which may be precipitated. The precipitate consisting of ferric 
 hydroxide and manganese dioxide is filtered off and washed. It is dis- 
 solved in a little hydrochloric acid and the paper well washed. The solu- 
 tion is boiled until the chlorine is completely expelled, then neutralized 
 with ammonia, warmed, and the trace of iron filtered off immediately. 
 Unless the precipitate is very small it is redissolved in hydrochloric acid 
 and again precipitated with ammonia and quickly filtered off and washed. 
 It is ignited and weighed as oxide. 
 
 202. Manganese. The combined filtrates from the iron contain all of 
 the manganese unless the amount of iron present is considerable. The 
 solution should be evaporated to dryness in a porcelain dish and the ammo- 
 nium chloride volatilized by gently heating with the Bunsen burner. The 
 residue is dissolved in a few c.c. water and a few drops of hydrochloric 
 acid and the manganese precipitated and weighed as sulphide as directed 
 in Exercise 22. 
 
 203. Volumetric Determination of Iron and Manganese. If considerable 
 iron is present, the method of separation given is not applicable. In this 
 case the simplest methods of determining the two metals are volumetric. 
 The ammonium precipitate should be dissolved in sulphuric acid with 
 
150 ANALYSIS OF ALLOYS. ,^ 
 
 expelled by boiling. The solution must be made up to a definite vol- 
 ume and divided into two equal portions. For this purpose a 100-c.c. 
 flask should be used which has been calibrated with a 50-c.c. pipette 
 by emptying the pipette twice into the dry flask and making a mark on 
 the stem. The solution of iron and manganese is evaporated to small 
 bulk, transferred to the flask, made up to the mark and thoroughly 
 mixed. 50 c.c. are withdrawn with the dry pipette. The solution 
 adhering to the walls of the pipette is rinsed out with distilled water and 
 added to the portion remaining in the flask. One of these portions is re- 
 duced with zinc and the iron titrated with standard permanganate. (See 
 p. 316.) The other portion is shaken up with zinc oxide until the free 
 acid is neutralized. One gram of zinc sulphate and a drop or two of 
 dilute nitric acid are added and the solution diluted to several hundred 
 cubic centimeters. The manganese is titrated with standard potassium 
 permanganate according to Volhard. (See p. 327.) 
 
 204. Phosphorus. For the determination of phosphorus a 5-gram por- 
 tion of the alloy is taken, as the percentage of this element is usually small 
 (seldom more than 0.2%). The material is placed in a 200-c.c. beaker and 
 20 to 30 c.c. concentrated nitric acid added. The beaker is covered with a 
 watch-crystal and after the first violent action of the acid has ceased it is 
 placed on the water-bath and heated until the alloy is completely decom- 
 posed and the residue is pure white. All of the phosphoric acid will remain 
 with the tin provided a sufficient amount of the latter is present in the alloy. 
 From six to eight times as much tin as P 2 5 must be present. Unless at least 
 5% of tin has been found, a preliminary test should be made by dissolving 
 about a gram of the alloy in concentrated nitric acid, filtering, and testing 
 the filtrate for phosphoric acid with molybdate mixture. If phosphoric 
 acid is found in the filtrate, metallic tin must be added before dissolving 
 the alloy in nitric acid. From one-half to one gram will usually be found 
 sufficient. 
 
 The nitric acid solution of the alloy is diluted and the stannic oxide 
 containing the phosphoric acid is filtered off and washed a few times. After 
 drying, the precipitate is transferred to a porcelain crucible, the paper is 
 burned, and the ash added. After adding three times its weight of potassium 
 cyanide, cover the crucible and fuse for a few minutes at a red heat. The 
 stannic oxide is reduced to metallic tin and the phosphoric acid forms potas- 
 sium phosphate. After cooling, extract the fused mass with hot water, filter, 
 and wash the paper with hot water. 
 
 Expel the hydrocyanic and cyanic acids by boiling with concentrated 
 hydrochloric acid. This operation must be conducted under a hood with 
 good draught. Evaporate to dryness to dehydrate the silicic acid which 
 has been dissolved from the porcelain by the action of the potassium cyanide. 
 Dissolve the dry residue in a little hydrochloric acid and pass hydrogen 
 sulphide to precipitate a small amount of tin and copper which is present. 
 Filter, wash the precipitate, and destroy the hydrogen sulphide in the fil- 
 
ANALYSIS OF MANGANESE-PHOSPHORUS-BRONZE. 151 
 
 trate by adding bromine water and boiling. If the volume of the solution 
 exceeds 50 c.c., concentrate by boiling. Cool and precipitate the phos- 
 phoric acid by adding about one-half gram of crystallized magnesium 
 chloride or sulphate dissolved in a little water and then neutralizing the 
 solution with filtered ammonia while stirring vigorously. Add a small 
 excess of ammonia. Assure yourself that the phosphoric acid is all pre- 
 cipitated by adding a little magnesia mixture to the clear supernatant 
 liquid. After standing several hours, filter, wash with dilute ammonia, 
 ignite in a porcelain crucible, and weigh as magnesium pyrophosphate. 
 
 The precipitation of the metals present with hydrogen sulphide may 
 be omitted and the separation effected by precipitating the phosphoric 
 acid as molybdate. The dry residue should then be dissolved in nitric 
 acid, and after filtering off the silica, the phosphoric acid is precipitated as 
 directed in Chapter IX, page 118. 
 
ANALYSIS OF MINERALS. 
 
 CHAPTER XIII. 
 
 MINERALS CONTAINING IRON, ALUMINIUM, 
 AND CHROMIUM. 
 
 SELECTION AND PREPARATION OF SAMPLE. 
 
 205. Taking the Sample for Analysis. One of the most impor- 
 tant operations preliminary to many chemical analyses is the 
 selection of the sample to be analyzed. When the material to be 
 analyzed is a chemical, such as the various salts made for commercial 
 purposes, the task is comparatively simple since the product is 
 usually quite uniform and small grained. Samples are taken 
 from a number of the barrels or other receptacles. In many cases 
 it is advisable to take samples from various parts of the barrel, 
 such as the end, middle, etc. This is most readily accomplished 
 by means of a " butter-tryer " or " thief " or some similar apparatus. 
 These various samples are brought together, thoroughly mixed, 
 and a smaller sample taken out. This may be done by taking 
 small portions from various parts of the pile or by dividing it 
 into quarters. One quarter or two diagonal quarters are taken, 
 well mixed, and the sampling repeated in the same manner until 
 a convenient quantity for analysis is obtained. 
 
 Care in sampling dry and apparently uniform products is 
 necessary for a good many reasons. Many substances when 
 exposed to the air, as in the outer portions of a barrel, absorb cr 
 lose moisture, oxidize, lose volatile constituents, or in other ways 
 change in composition. Even in the case of a mixture of two 
 solid substances of different specific gravity the heavier constituent 
 tends to settle to the bottom, so that after the most thorough 
 mixing the material will not be uniform after standing for some 
 time. 
 
 If LIQUIDS are to be sampled they must be thoroughly shaken 
 or parts of the first, middle, and last runnings of a cask or drum 
 
 152 
 
SELECTION AND PREPARATION OF SAMPLE. 153 
 
 must be taken. These portions should be thoroughly mixed before 
 taking the sample for analysis. 
 
 METALS and ALLOYS are by no means uniform, since, on solidi- 
 fying from the molten state, the heavier constituents are apt to 
 settle to the bottom, as do also the constituents which crystallize 
 out or solidify first. In the case of a bar the outer portions are 
 not always of the same composition as the interior of the metal. 
 Drilling through a piece of metal with a clean tool is therefore 
 apt to give the best sample. If the metal is in the form of a sheet, 
 portions should be punched or cut from the outer edges as well as 
 from the centre. If car-toads of metals &s"pigs" are to be sam- 
 pled, one or more "pigs" in each car should be punched or drilled 
 and the samples combined by melting or mixing before the labora- 
 tory sample is taken. The turnings of the more brittle metals 
 may be pulverized by grinding in a mortar or a coffee-mill. 
 
 If a ROCK is to be analyzed, the sample is usually selected 
 because of its apparent homogeneity. The physical properties 
 usually give sufficient evidence of the presence of foreign material. 
 If the rock or mineral is to be analyzed with a view to its use as 
 the source of a given metal or other constituent, the sample must 
 include all foreign material which would be taken out in mining 
 operations. 
 
 The cars or shiploads of ORE are usually sampled by taking 
 out shovelfuls of the material from various parts of the load. 
 These portions are thrown together on a floor or other convenient 
 place. The larger pieces are broken with a hammer and the material 
 well mixed. ;JThe pile is quartered and the quarter taken is still 
 further powdered and quartered. This process is continued until 
 a sample of convenient size is obtained. About 25 grams will 
 usually be found sufficient, though in some cases as much as a 
 kilogram will be required for the determination of elements present 
 in very small amount. 
 
 206. Pulverizing. The tools used in breaking off and pow- 
 dering ores or rocks include, in the first place, HAMMERS with 
 hardened surfaces. These are made in various shapes and sizes 
 to suit the geologist prospecting in the field or the laboratory 
 assistant in breaking up the large specimens of rock. A chilled 
 IRON PLATE with a hardened steel MULLER to which a rocking 
 
154 
 
 ANALYSIS OF MINERALS. 
 
 motion may be given by means of a long handle may be used to 
 reduce many ores and rocks to powder fine enough for analysis. 
 
 FIG. 21. 
 
 Hardened STEEL MORTARS are also useful in crushing moderately 
 small pieces of ore or rock. The most convenient method of 
 
 ^ J 
 
 FIG. 22. 
 
 crushing large samples of rock or ore is by the use of ORE-CRUSHERS. 
 The jaws between which the ore is placed are actuated by means 
 
 FIG. 23. 
 
 of a long handle so connected with levers that a very great crush- 
 ing pressure may be brought to bear on the material. It is cus- 
 tomary to remove by means of a magnet particles of iron which 
 
SELECTION AND PREPARATION OF SAMPLE. 155 
 
 have been detached from the tools during the process of pulverizing. 
 Very few rocks or ores are entirely free from magnetic particles, so 
 that this method of procedure is open to objection. 
 
 Hard rocks and ores, as well as those known to contain iron, 
 should be pulverized as far as possible in AGATE MORTAES. These, 
 as well as the iron mortars, are frequently arranged to be oper- 
 ated by mechanical means. Many chemists, however, still prefer 
 the hand grinding. 
 
 207. Sifting. The requisite degree of fineness is frequently 
 secured by passing the powdered material through sieves of the 
 requisite size of mesh. Eighty-mesh BRASS SIEVES are suitable for 
 most ores. In special cases where the sample is very irregular, 
 sieves of 100 to 120 mesh to the square inch should be used. Even 
 when kept scrupulously clean a small amount of the metal of the 
 sieve is apt to contaminate the sifted material. This error may 
 be entirely disregarded if a constituent present in very small 
 amount is to be determined. When a complete analysis is to be 
 made, many chemists prefer to use BOLTING-CLOTH or FINE LINEN. 
 Where the state of oxidation of the iron is to be determined, the 
 shreds of the cloth, which almost necessarily contaminate the 
 sifted material, undoubtedly influence the result to an appreciable 
 degree. For this reason some chemists advocate grinding the 
 material without sifting at all. 
 
 In sifting through cloth large particles are readily forced 
 through if the powder is rubbed over the cloth. A wide-mouthed 
 bottle should be selected, thoroughly cleaned and dried. The 
 cloth should be stretched over the mouth of the bottle and 
 fastened securely by winding a string around the neck. The 
 material to be sifted, after having been ground in small portions 
 in an agate mortar, is placed on the cloth and a piece of sheet 
 rubber or leather stretched over the mouth of the bottle and 
 held securely with the hand or by a rubber band placed around 
 the neck. The rubber or leather is now lightly tapped with 
 a spatula until no more fine material sifts into the bottle. 
 The coarser material is again ground hi the mortar and sifted. 
 In no case should the coarse material be discarded. Frequently 
 the more friable portions of a rock or ore are different in compo- 
 sition from the particles which resist grinding longest. In pul- 
 
156 ANALYSIS OF MINERALS. 
 
 verizing a sample the rule must be strictly adhered to that the 
 whole of a portion taken for powdering must be ground and passed 
 through the sieve. In sampling, the large pieces must be broken 
 completely so as to mingle with the finer material and constitute a 
 fair portion of the sample finally taken. 
 
 208. Drying. After being reduced to a sufficient state of 
 fineness, the material to be analyzed is frequently dried at a tem- 
 perature of about 100 to remove hygroscopic moisture before 
 being weighed out. Other workers weigh out the air-dried mate- 
 rial and determine the percentage of water lost at 100. It is 
 held that the amount of hygroscopic water present is a very char- 
 acteristic property of some rocks and minerals, and serves to 
 distinguish minerals which are in other respects identical. Hygro- 
 scopic material can also be weighed out much more accurately 
 when air-dried than when entirely desiccated. In the latter case 
 it gradually absorbs moisture, as it is exposed to the air from time 
 to time in the course of the analysis, so that the material weighed 
 out the first day is not identical in composition with that weighed 
 out later* 
 
 SEPARATION OF IRON, ALUMINIUM, AND CHROMIUM. 
 
 Iron and aluminium occur very frequently together, and are 
 generally precipitated together from solutions containing man- 
 ganese, cobalt, nickel, and zinc. The methods suitable for pre- 
 cipitating these two metals are not generally applicable to the 
 complete precipitation of chromium. The separation of iron and 
 aluminium will therefore be first considered, and then that of 
 chromium. 
 
 209. Separation as Hydroxide. In the presence of excess of 
 ammonium chloride and ammonia, nickel, cobalt, and zinc are 
 not at all, while manganese is only slowly precipitated by being 
 oxidized to the hydrated sesquioxide, while ferric iron, aluminium 
 and chromium are very nearly completely precipitated. If all 
 of these metals are present in solution, the iron, aluminium, and 
 chromium may be completely precipitated by means of ammonia 
 and ammonium chloride, but the precipitate will be contaminated 
 with considerable amounts of manganese and varying amounts 
 of zinc, nickel, and cobalt. In the absence of chromium, by 
 
SEPARATION OF IRON, ALUMINIUM, AJfD CHROMIUM. 157 
 
 redissolving and reprecipitating the iron and aluminium once or 
 twice, the separation from nickel and cobalt may be made com- 
 plete and very nearly so from zin3. 
 
 Only a small part of the other metals present in the solution 
 will be carried down each time with the iron and aluminium pre- 
 cipitate. On dissolving the first precipitate only this small part 
 of the contaminating metals will be present in the solution; the 
 bulk of these metals will be in the filtrate. On making a second 
 precipitation the fraction carried down will be very small indeed, 
 so that on a third precipitation the amount carried down may 
 be wholly neglected. This process will not free the precipitate 
 from manganese, because this metal is oxidized very readily 
 hi alkaline solution to the insoluble dioxide. 
 
 The solution should be heated, excess of ammonia added, and 
 if aluminium is present ammonium carbonate added. After 
 digesting hot for some time the precipitate is filtered off, washed 
 a few times, redissolved in hydrochloric acid, and the operation 
 repeated. If the amount of nickel and cobalt is considerable, a 
 third precipitation is necessary. Manganese must be absent. 
 The precipitate may be ignited and weighed, and one of the con- 
 stituents determined, the other one being obtained by differ- 
 ence. 
 
 210. Separation as Basic Acetates. If manganese and zinc 
 are present, the aluminium and iron must be precipitated as basic 
 acetates. This method of separation is based on the fact that 
 iron and aluminium may be precipitated as basic acetates by 
 boiling a solution of these metals which contains ammonium or 
 sodium acetate and free acetic acid. The acid present prevents 
 the precipitation of manganese, zinc, cobalt, and nickel. The 
 separation is therefore much sharper than when the iron and 
 aluminium are precipitated from alkaline solution. Oxidizing 
 agents must be absent or manganese will be precipitated, but the 
 iron must be in the ferric condition. 
 
 The proportion of free acetic acid to ammonium acetate must 
 be carefully regulated. They should be present in gram-molecular 
 proportion. As the commercial ammonium acetate is not always 
 neutral it should be tested before use and neutralized. The sim- 
 plest way of securing the proper proportion of free acid and neu- 
 
158 ANALYSIS OF MINERALS. 
 
 tral salt is to make the ammonium acetate by neutralization of 
 acetic acid. If the same volume of acetic acid is neutralized as 
 has already been added to the solution the correct amount of 
 ammonium acetate will be secured. The volume of the solution 
 must also be carefully regulated with reference to the amount of 
 iron present. 
 
 The perfectly cold hydrochloric acid solution must be made 
 very nearly neutral by the addition of concentrated and then 
 dilute ammonium carbonate solution until the slightest turbidity 
 is noticed. If a precipitate forms it must be dissolved in hydro- 
 chloric acid, and the solution again cautiously neutralized. One c.c. 
 of 5 N acetic acid is added for each 0.1 gram of iron. Water is 
 added until at least 150 c.c. is present for each 0.1 gram of iron. 
 Heat to boiling hi a porcelain dish, and add 1 c.c. of 5 N acetic 
 acid which has been exactly neutralized with ammonia for each 
 0.1 gram of iron present. Continue the boiling for two or three 
 minutes. After allowing the precipitate to settle, decant the 
 liquid and wash by decantation a few times, bringing the wash- 
 water to a boil each time before decantation. The precipitate is 
 free from zinc and all but a trace of manganese, but if nickel and 
 cobalt are present it is apt to contain a little of these metals, 
 more of the former than of the latter. It may be freed from 
 these metals by redissolving and reprecipitating. If the filtrate 
 has a slight yellow color the iron has not been completely precipi- 
 tated because of the presence of too much acetic acid. A few 
 drops of ammonia may be added and the solution warmed until 
 the precipitate collects. It is washed with hot water containing 
 a little acetic acid and ammonium acetate. 
 
 This method gives better results with iron than with alu- 
 minium, unless iron is also present in considerable amount. Small 
 amounts of chromium may also be precipitated by this method 
 if considerable iron is present. The precipitate cannot be ignited 
 and weighed . if alkalies were present in the solution because it 
 carries down alkalies which cannot be washed out. It must be 
 dissolved in hydrochloric acid and reprecipitated with ammonia. 
 
 The method has the disadvantage that the precipitate is very 
 bulky, and is frequently slimy and difficult to wash. For this 
 reason it is not used when a large amount of iron must be handled. 
 
SEPARATION OF IRON, ALUMINIUM, AND CHROMIUM. 159 
 
 211. Separation of Iron and Aluminium as Basic Carbonates.^ 
 Iron and aluminium may also be separated from manganese, 
 cobalt, nickel, and zinc, by precipitation of the former as basic 
 carbonates. The hydrochloric acid solution is carefully neutral- 
 ized with ammonium carbonate, until a slight turbidity appears. 
 A drop or two of acetic acid is added. Ammonium chloride must 
 be present to the extent of about 1 gram for each 0.1 gram of iron. 
 The solution is gradually brought to a boil and the heating con- 
 tinued until the carbon dioxide is completely expelled. The pre- 
 cipitate is washed by decantation, a few drops of ammonia being 
 added to the wash-water. A slight yellow color in the filtrate 
 indicates incomplete precipitation of iron. A few drops of ammo- 
 nia may be added, but care must be taken not to make the solution 
 alkaline or manganese will be precipitated. 
 
 212. Separation of Chromium as Chromic Acid. Chromium 
 is best separated from other metals by methods which are based 
 on the acid properties of chromium trioxide. It is converted into 
 the acid oxide by treatment with an alkaline oxidizing agent. A 
 mixture of sodium carbonate and potassium nitrate has been largely 
 used for this purpose. On extracting the fused mass with water 
 the oxides of zinc, cobalt, nickel, and iron remain undissolved. 
 Part of the manganese remains undissolved while part goes into 
 solution as manganate. On boiling the solution after the addi- 
 tion of a little alcohol the manganese is completely precipitated 
 as the dioxide. Part of the aluminium remains with the insoluble 
 portion, while part passes into solution with the sodium chromate 
 as sodium alumina te. The aluminium may be separated from 
 the chromium by precipitation as aluminium hydroxide. The 
 solution must be acidified with hydrochloric acid to decompose 
 the sodium aluminate. On adding ammonium chloride and 
 ammonia until the solution is faintly alkaline and warming, the 
 aluminium is completely precipitated as hydroxide. The chro- 
 mium in the condition of chromic acid is not precipitated by the 
 ammonia. 
 
 The potassium nitrate used to oxidize the chromium is reduced 
 during the fusion to nitrite which is present in the water solution 
 of the melt. On acidifying this solution, nitrous acid is liberated 
 which tends to reduce the chromic acid. To destroy the nitrous 
 
160 ANALYSIS OF MINERALS. 
 
 acid potassium chlorate or still better liquid bromine is added to 
 the alkaline solution. On acidifying and warming gently, the 
 nitrous acid is oxidized and the excess of bromine or chlorine 
 expelled. The chromium in the filtrate from the aluminium may 
 be precipitated and weighed as lead or barium chromate. It 
 may also be reduced by sulphurous acid, precipitated with ammo- 
 nia and weighed as oxide, or it may be determined volumetrically 
 by means of ferrous sulphate and a standard chromate solution. 
 
 Recently fusion with SODIUM PEROXIDE has come into exten- 
 sive use. This flux seems to be especially adapted to decompose 
 the refractory chrome iron ore. A silver, nickel, or copper cruci- 
 ble must be used, as platinum is strongly attacked by the alkaline 
 flux. A mixture of 5 to 6 parts of caustic soda with 3 to 4 parts 
 of sodium peroxide has also been considerably used. 
 
 If the chromium is in solution with other metals the separation 
 is not so simple. The acid solution is nearly neutralized with 
 sodium carbonate, sodium acetate added, the solution heated, 
 and chlorine passed or bromine added. The chromium is oxidized 
 to chromic acid and remains in solution. Iron and aluminium 
 are precipitated during the boiling by the sodium acetate. Any 
 manganese present is precipitated as the peroxide together with 
 part of the cobalt, nickel, and zinc. 
 
 213. Rothe's Ether Separation of Iron. The method pro- 
 posed by J. W. Rothe for the separation of iron from chromium, 
 aluminium, manganese, cobalt, nickel, and copper has been much 
 used, and overcomes many difficulties met with in analyzing 
 compounds containing much iron. It depends on the fact that 
 ferric chloride may be almost completely extracted by means of 
 ether from a hydrochloric acid solution of sp. gr. 1.100 to 1.105 
 containing the metals mentioned. Nitric acid, chlorine, and 
 more than small amounts of sulphuric acid must be absent, and 
 the solution must be kept absolutely cold, as in a warm solution 
 the ferric chloride is reduced by the ether. All suspended matter 
 such as silica, carbon, etc., must be removed by filtration The 
 apparatus designed for this separation is shown in Fig 24. The 
 two bulbs A and B are connected by a three-way stop-cock at E^ 
 so that liquid may flow from one bulb to the other or from either 
 bulb through the exit-tube to a receptacle placed below II the 
 
SEPARATION OF IRON, ALUMINIUM, AND CHROMIUM. 161 
 
 hydrochloric acid solution of the iron is placed hi A, about an 
 equal volume of ether is placed hi B. A slight pressure is pro- 
 duced over the ether by means of a rubber bulb. By cautiously 
 turning the stop-cock E into the position 2, the ether is allowed to 
 flow into the bulb A. The liquids are well mixed by vigorous 
 
 FIG. 24. 
 
 shaking. During this operation it is well to keep the bulb A well 
 cooled by holding it in a stream of cold water. After allowing a 
 few minutes for the liquids to separate, the stop-cock E is turned so 
 that the lower heavy liquid may flow into B. Leave just enough of 
 
162 ANALYSIS OF MINERALS. 
 
 this liquid to fill the capillary tube (7, shake down any heavy liquid 
 adhering to the sides of A, and allow to stand until the ethereal 
 layer is clear. Again connect A and B by turning the stop-cock C, 
 and transfer the remainder of the heavy liquid to B together 
 with a little of the ether solution. Turn the stop-cock E into the 
 position 3, and allow the ether solution to flow out into a beaker. 
 Rinse the walls of the bulb A with a little ether, and after stand- 
 ing a few minutes observe if any of the heavy solution collects in 
 C. If so, transfer as before to B. Repeat the extraction with 
 50 c.c. of ether. (See Exercise 35, page 170.) 
 
 214. Volumetric Separation of Iron and Aluminium. The 
 iron and aluminium precipitate obtained in separations from 
 other metals is best ignited and weighed as Al 2 3 +Fe 2 3 and one 
 of the metals determined, the other being obtained by differ- 
 ence. For this purpose it is simplest to determine the iron volu- 
 me trically. The precipitate is fused with acid potassium sul- 
 phate until dissolved, the fused mass dissolved in water, and the 
 solution filtered to separate a small amount of silica which may 
 be present. This is ignited and weighed, and the amount deducted 
 from the weight of the oxides. The solution of the iron and alu- 
 minium is treated with zinc to reduce the iron, which is titrated 
 with standard potassium permanganate solution as directed in 
 Chapter XXIV, page 316. If the precipitate is large it may be 
 dissolved in hydrochloric acid before weighing and divided into 
 two portions. The iron is determined in one portion while in the 
 other both metals are precipitated as hydroxides, ignited, and 
 weighed as oxides. 
 
 215. Separation of Iron from Aluminium as Aluminate. The 
 two metals may also be separated by means of boiling caustic 
 potash or soda solution. The hydrochloric acid or potassium 
 bisulphate solution of the two metals is nearly neutralized, and 
 poured slowly with constant stirring into excess of caustic potash or 
 soda solution, heated nearly to boiling in a platinum or silver dish. 
 The iron is precipitated as hydroxide, while the aluminium remains 
 in solution. The iron precipitate is washed more readily if part 
 of it is reduced by adding a little sodium sulphite and heating 
 before pouring into the boiling alkaline solution. Before weighing, 
 the iron precipitate must be dissolved in hydrochloric acid and re- 
 
ANALYSIS OF CHROMITE. 163 
 
 precipitated with ammonia to free it from the alkali which is 
 always present. 
 
 The aluminium in the filtrate is precipitated by acidifying 
 with hydrochloric acid and just neutralizing with ammonia and 
 warming. The caustic potash or soda must be free from impuri- 
 ties which are precipitated by ammonia and ammonium chloride. 
 If pure caustic cannot be obtained by solution of the metal in 
 water in a platinum or silver dish or other method, a blank deter- 
 mination of the amount of impurities present must be made. 
 
 EXERCISE 34. 
 
 Analysis of Chromite (FeO,MgO)(Cr 2 O 3 ,Al 2 O 8 ). 
 Silica and Small Amounts of Manganese are Generally Present. 
 
 This ore is chiefly of value for its content of chromium. The deter- 
 mination of the amount of this element is the most important and frequently 
 the only determination required. For this reason, and also for convenience 
 in determining the amount of the other elements present, a separate portion 
 is taken for the determination of chromium. 
 
 216. Solution of Ore. Weigh one-half gram of the finely pulverized ore 
 and mix in a small copper or nickel crucible with 2 to 5 grams of powdered 
 sodium peroxide. Brush off carefully the platinum or glass rod used for 
 mixing the material. If the sodium peroxide is not pure yellow and fresh, 
 the larger amount should be taken, as it decomposes rapidly, forming sodium 
 carbonate. Heat the crucible with a small flame of the Bunsen burner, 
 so that the contents are completely fused only after several minutes. Keep 
 the material fused for about ten minutes. 
 
 When cold, place the crucible in a porcelain dish and dissolve the con- 
 tents in 50 to 100 c.c. of hot water. Boil for fifteen to twenty minutes to 
 decompose the sodium peroxide. 
 
 217. Chromium. Filter off and wash the precipitate consisting of iron, 
 magnesium, manganese and some of the silica, as well as a little nickel or 
 copper from the crucible. The filtrate will contain all of the chromium 
 and aluminium and part of the silica. Acidify the filtrate with acetic acid, 
 warm and filter off any alumina and silica which separates out. Precipi- 
 tate the chromic acid by adding lead acetate or barium chloride and weigh as 
 lead or barium chromate. The chromium may also be determined volu- 
 metrically as directed in Exercise 63, page 333. This volumetric method 
 will usually be found superior to the gravimetric method. 
 
 218. Iron. The insoluble material filtered off from the solution of the 
 sodium peroxide fusion may be used for the determination of iron. It is 
 dissolved in a little sulphuric acid, the material adhering to the crucible 
 
164 ANALYSIS OF MINERALS. 
 
 being dissolved in the same manner. Particles of undecomposed chromite 
 may be distinguished from the gelatinous silicic acid as a fine heavy powder 
 which quickly settles to the bottom of the beaker en stirring the solution. 
 If material of this kind is present it must be filtered off and again fused with 
 sodium peroxide. The solution of the melt must be filtered and the filtrate 
 added to the bulk of the chromium solution. The insoluble residue is 
 dissolved in hydrochloric acid and added to the solution of the insoluble 
 material from the first fusion. The iron may then be titrated as directed in 
 Exercise 57, page 320. 
 
 219. Silica. For the determination of the elements present other than 
 chromium, a separate portion of one-half to one gram is fused with sodium 
 peroxide, in the same manner as directed for the determination of chromium. 
 The fused mass is dissolved in water, boiled to decompose the sodium per- 
 oxide and the insoluble matter filtered off and washed free from chromium. 
 The filtrate is acidified with hydrochloric acid, evaporated to dryness and 
 heated one-half hour on the water-bath to dehydrate the silicic acid, which 
 is filtrated off and washed. 
 
 220. Aluminium. The filtrate will contain the aluminium and the 
 chromium. It is neutralized with ammonium carbonate and a little hydro- 
 gen peroxide or bromine added, and the solution boiled to reoxidize the 
 chromium which has been reduced during the evaporation with hydro- 
 chloric acid. The solution should be boiled in a porcelain or platinum 
 dish to avoid solution of the glass of the beaker in the alkaline liquid. The 
 aluminium hydroxide is filtered off, washed free from chromium, ignited, 
 and weighed as oxide. 
 
 The chromium and aluminium may also be precipitated together and 
 weighed as oxides, the amount of aluminium being found by deducting 
 the weight of chromium oxide as obtained by the previous determination 
 of this element. The filtrate from the silica is warmed after the addition 
 of alcohol or sulphurous acid to completely reduce the chromium. The 
 aluminium and chromium are then precipitated with ammonia and weighed 
 as oxides. 
 
 221. Silica. The material insoluble in the sodium peroxide solution, 
 which contains the manganese, magnesium, and part of the silica, is dis- 
 solved in warm hydrochloric acid and any undecomposed chromite filtered 
 off and again fused with sodium peroxide. The acid solution of the 
 insoluble matter is evaporated to dryness on the water-bath to dehydrate 
 the silica which is filtered off, washed, and weighed, the weight being added 
 to that of the silica already found, unless the same filter-paper was used 
 for both precipitates. 
 
 222. Manganese. If a copper crucible was used, hydrogen sulphide is 
 passed through the acid filtrate from the silica to precipitate the copper dis- 
 solved, which is filtered off and washed. Bromine water or liquid bromine 
 dissolved in hydrochloric acid is added to the filtrate to oxidize the iron and 
 to decompose the excess of hydrogen sulphide. Sufficient bromine is added 
 
ANALYSIS OF CHROMITE. 165 
 
 to color the solution distinctly. Ammonia is added until the reaction is 
 distinctly alkaline. The solution is warmed and the precipitate consisting 
 of the iron and any manganese which may be present is filtered off and 
 washed. It may be ignited and weighed, the manganese being obtained by 
 difference, the amount of iron being known. The manganese will be present 
 as Mn 3 O 4 . If the amount of manganese i^ small it may be determined 
 more accurately by Ford's method. The precipitate is dissolved in a 
 little nitric acid and the determination carried out as directed in Chapter 
 XXVIII, page 385. 
 
 223. The magnesium will be found in the filtrate from the iron precipi- 
 tate. It is evaporated to small bulk, precipitated as magnesium ammo- 
 nium phosphate, and weighed as pyrophosphate. 
 
CHAPTER XIV. 
 
 ANALYSIS OF SULPHIDES CONTAINING MAN- 
 GANESE, NICKEL, COBALT, AND MERCURY. 
 
 SEPARATION OF MANGANESE, NICKEL, AND COBALT. 
 
 224. Separation of Nickel and Cobalt as Sulphides. One of 
 the best methods of separating manganese from nickel and cobalt 
 depends on the properties of the sulphides of these metals. The 
 sulphide of manganese is readily soluble in dilute acids, including 
 acetic acid. The sulphides of nickel and cobalt after precipitation 
 are almost insoluble in all dilute acids. These two metals are 
 not precipitated, however, from solutions acid with mineral acids, 
 but do come down from warm acetic-acid solution. 
 
 The acid solution of manganese, cobalt, and nickel from which 
 zinc, iron, aluminium, etc., have been separated by methods 
 already given, is concentrated to a small bulk (30 to 50 c.c.). 
 The solution is made alkaline with sodium carbonate, acidified 
 with acetic acid and 3 to 4 grams of sodium acetate dissolved in 
 a little water are added. The solution is heated to 70, and hydro- 
 gen sulphide passed until saturated. The sulphides of nickel and 
 cobalt are filtered off and washed. The filtrate and washings 
 are concentrated, a little ammonium sulphide and then acetic 
 acid are added, and hydrogen sulphide again passed. A little 
 more of the cobalt and nickel sulphides is frequently obtained. 
 A second test of the filtrate in this manner sometimes yields a 
 small precipitate. 
 
 Another method of carrying out this separation which requires 
 but one precipitation is as follows: The sulphate solution of man- 
 ganese, cobalt, and nickel is transferred to a pressure-bottle of 
 about one-half liter capacity. The solution is neutralized with 
 ammonia, 30 c.c. ammonium-acetate solution (1 to 10), and 20 c.c. 
 50% acetic acid added. The solution is diluted to 300 to 400 c.c., 
 and hydrogen sulphide passed for one to two hours. The bottle 
 
 is closed, placed hi a water-bath containing cold water which is 
 
 166 
 
SEPARATION OF MANGANESE, NICKEL, AND COBALT 167 
 
 brought to a boil in one hour. The bottle is allowed to cool in 
 the water-bath to 50. The precipitated sulphides are filtered 
 off and washed with water containing acetic acid and hydrogen 
 sulphide. 
 
 225. Electrolytic Separation of Manganese from Cobalt and 
 Nickel. Small amounts of manganese can readily be separated 
 from cobalt and nickel electrolytically. The sulphate solution of 
 the three metals should be treated with 30 c.c. of a cold-saturated 
 ammonium-sulphate solution, and 30 to 50 c.c. of ammonia 
 (sp. gr. 0.96). The solution is diluted to about 150 c.c., and elec- 
 trolyzed with a current of about 0.7 ampere at the ordinary tem- 
 perature. The manganese separates as dioxide on the anode. 
 This precipitate carries down small amounts of nickel and cobalt 
 unless these metals are deposited before the manganese dioxide 
 forms. Small amounts of nitrates and chlorides hinder the pre- 
 cipitation of nickel and cobalt. The deposited metal is washed 
 with water without interrupting the current. Finally it is washed 
 with alcohol, dried in the air-bath, and weighed. 
 
 SEPARATION OF NICKEL AND COBALT. 
 
 226. Precipitation of Cobalt as Tripotassium Cobaltic Nitrite, 
 Co(N0 2 ) 3 .3KN0 2 , separates it from nickel as well as from man- 
 ganese and zinc. The alkaline earth-metals and iron must be 
 absent. The method is satisfactory for mixtures of nickel and 
 cobalt in all proportions, but is especially suited for the separation 
 of a little cobalt from much nickel. The solution of the metals is 
 evaporated to small bulk, and neutralized with potassium hydrox- 
 ide. About 5 grams of potassium nitrite dissolved hi water are 
 added. Acetic acid is added until the solution is acid and nitrous 
 fumes are evolved. It is allowed to stand for at least twenty- 
 four hours. A portion of the supernatant liquid is taken out and 
 tested for cobalt by adding more potassium nitrite. If a precipi- 
 tate forms after long standing, more potassium nitrite must be 
 added to the mam solution. The precipitate is filtered off and 
 washed with a solution of potassium acetate (1 to 10) containing 
 a little potassium nitrite. The precipitate may be dissolved in 
 hydrochloric acid and the cobalt reprecipitated with caustic soda 
 
168 ANALYSIS OF MINERALS. 
 
 and weighed as oxide. It may also be dissolved in dilute sul- 
 phuric acid and the solution evaporated to dryness in a weighed 
 platinum dish or crucible. The residue consisting of the sulphates 
 of cobalt and potassium is heated to dull redness and weighed. 
 If weighed in this manner the potassium acetate and nitrite must 
 finally be washed out with a little distilled water. The precipi- 
 tate may also be dissolved in hot dilute sulphuric acid, the solution 
 concentrated, treated with excess of ammonia, ammonium sulphate 
 added, and the cobalt determined electrolytically. 
 
 If the NICKEL has previously been weighed with the cobalt, it 
 may be obtained by difference. Otherwise the nickel may be 
 determined in the filtrate from the cobalt. For this purpose the 
 solution is treated with hydrochloric acid and the nickel precipi- 
 tated with caustic soda and bromine water. The nickelic hydrox- 
 ide may be filtered off, washed, and reduced to metallic nickel by 
 heating in a stream of hydrogen, or it may be dissolved in sulphuric 
 acid and a little sulphurous acid and determined electrolytically. 
 
 227. Separation of Cobalt and Nickel by Means of Nitroso-/?- 
 Naphthol. The cobalt compound Co(C 10 H 6 0(NO)) 3 is insoluble 
 in hydrochloric acid, while the corresponding nickel compound is 
 soluble. Nitric acid is expelled by evaporation with sulphuric 
 acid. The solution is diluted, 5 c.c. hydrochloric acid and 
 a freshly prepared hot solution of nitroso-/?-naphthol in 50% 
 acetic acid added, until further addition produces no precipitate. 
 The solution is digested in a warm place for a few hours. 
 The precipitate, consisting of cobalt nitroso-/?-naphthol as well as 
 considerable of the reagent, is filtered off and washed, first with 
 cold and then with warm 12% hydrochloric acid until all the 
 nickel has been washed out. It is then washed with hot water 
 until free from acid. The precipitate is dried and placed in a 
 weighed Rose crucible. A little pure recrystallized oxalic acid is 
 added and the paper burned, at first with low heat and then with 
 the full flame of the Bunsen burner. When the carbon is com- 
 pletely burned the precipitate is reduced to metallic cobalt by 
 heating in a stream of hydrogen. 
 
 The filtrate is evaporated after the addition of sulphuric acid. 
 The NICKEL may then be precipitated as hydroxide and weighed 
 as oxide, or it may be determined electrolytically. 
 
ANALYSIS OF SMALTITB. 169 
 
 EXERCISE 35. 
 
 Analysis of Smaltite (Co,Ni,Fe)(S,As) ? . 
 Silica and Small Amounts of Manganese Are Usually Present. 
 
 228. Solution. One gram of the finely powdered material is weighed 
 out and transferred to an Erlenmeyer flask of about 250 c.c. capacity. 
 This is best done by means of a small tube sealed at one end. The tube 
 containing about one gram of the smaltite is carefully weighed. It is then 
 inserted into the neck of the flask which is tilted. On raising the flask up- 
 right and tapping the tube gently the smaltite is deposited on the bottom 
 of the flask. The tube is carefully withdrawn and weighed. Twenty c.c. of 
 aqua regia, made by adding to 3 volumes of concentrated nitric acid 1 volume 
 of concentrated hydrochloric acid, are added. A small funnel is placed in 
 the neck of the flask and, after being shaken and standing until the violent 
 action has ceased, the flask is warmed gently on the water-bath until the 
 mineral is entirely decomposed. If sulphur floats on top of the solution, 
 liquid bromine should be added, a few drops at a time, and the solution 
 warmed until the sulphur is entirely oxidized. 
 
 229. Silica. The solution is poured into a porcelain dish, the flask and 
 funnel rinsed with water and after the addition of concentrated sulphuric 
 acid, evaporated on the hot plate until fumes of sulphuric acid are copiously 
 evolved. The solution is cooled and diluted with water. The siliceous 
 residue is immediately filtered off, washed, and weighed. 
 
 230. Arsenic, Antimony, and Tin. A little sulphur dioxide is added 
 to the filtrate, which is warmed to reduce arsenic to arsenous acid. The 
 excess of sulphur dioxide is expelled by boiling. Hydrogen sulphide is 
 passed through the warm solution until no more precipitate forms. The 
 arsenious sulphide is filtered off on a Gooch crucible and the filtrate 
 again treated with sulphur dioxide and hydrogen sulphide to insure com- 
 plete precipitation of the arsenic. After washing the arsenious sulphide 
 the sulphur is extracted by treatment with alcohol and carbon disulphide. 
 The precipitate is then dried and weighed. The color of the precipitate 
 indicates the absence of lead, bismuth, and copper and all but a trace of 
 antimony. The absence of the latter element as well as tin may be shown 
 by treating the weighed precipitate with a concentrated solution of ammo- 
 nium carbonate. The sulphides of antimony and tin remain on the 
 filter and may be dissolved in hydrochloric acid and separated by the 
 methods given in Chapter XI, page 134. The arsenic must also be repre- 
 cipitated and weighed. 
 
 231. Iron. The filtrate from the hydrogen sulphide precipitate is 
 evaporated with the addition of nitric acid to oxidize the iron, The solution 
 is made alkaline with ammonia, warmed, and the precipitate filtered off 
 
170 
 
 ANALYSIS OF MINERALS. 
 
 and washed two or three times. The filtrate will contain most of the cobalt 
 and nickel, while the precipitate will contain all of the iron and a small 
 amount of the nickel and cobalt. It is dissolved in dilute 
 hydrochloric acid and the paper well washed. The solu- 
 tion is evaporated to a syrupy consistency and transferred 
 with hydrochloric acid of sp. gr. 1.124 to the bulb A of the 
 apparatus shown in Fig. 24, on p. 161, and the extraction 
 with ether carried out as there described. The apparatus 
 shown in Fig. 25 may also be used. Before transferring 
 the solution to the bulb D, a few cubic centimeters of ether 
 are poured into the bulb and allowed to flow into B. By 
 warming the bulb B with the hand and opening and closing 
 the stop-cock C, a partial vacuum is created in B. A and 
 C are closed and the solution transferred to D, rinsing the 
 beaker with hydrochloric acid of sp. gr. 1.124 until the bulb 
 is filled to the mark on D, making the volume 60 c.c. Place 
 the bulb B in cold water, open the stop-cock A, and allow the 
 liquid in D to flow into B. Close A and fill D with ether 
 (100 c.c.) and allow it to flow slowly into B. Mix the 
 liquids gradually, placing the bulb in cold water from time 
 to time to prevent a large rise in temperature. Finally 
 close A. Shake well, opening A occasionally to relieve 
 the pressure, then place the apparatus in cold water for about 
 five minutes to allow the liquids to separate. 
 Allow the lower liquid, which is very nearly free from iron, to flow into 
 a beaker. Introduce about 10 c.c. of hydrochloric acid (sp. gr. 1.1) into 
 the bulb D. Allow it to flow into B, shake, and after standing a few minutes 
 allow the lower liquid to flow into the beaker. This liquid, together with 
 the filtrate from the ferric hydroxide, will contain all of the cobalt, nickel, 
 manganese, and aluminium and a little of the iron. Draw off the ether 
 solution of the iron into another beaker and repeat the ether extraction 
 of the iron, the nickel-cobalt solution being poured back again into the 
 bulb D. Rinsing the beaker will be unnecessary. 
 
 The ether solution of the ferric chloride should be poured into a Florence 
 or distilling flask connected with a condenser by means of cork stoppers 
 and the beaker rinsed with water. The ether is distilled off on the water- 
 bath, care being taken that the ether is not placed near a Bunsen burner 
 or other flame. The ferric chloride is transferred to a beaker and the iron 
 precipitated with ammonia and weighed as oxide, or reduced with zinc and 
 determined volumetrically. 
 
 232. Aluminium and Chromium. The combined cobalt nickel solu- 
 tion is warmed to expel the ether and a little bromine or hydrogen 
 peroxide added. The solution is neutralized with ammonia, a little 
 ammonium carbonate added, and boiled. A small precipitate of iron is 
 usually obtained and any aluminium or chromium present is precipitated. 
 
 FIG. 25. 
 
ANALYSIS OF SMALTITE. 171 
 
 The precipitate, which is usually quite small, is filtered off, redissolved, 
 and reprecipitated. In the absence of aluminium and chromium, it is 
 added to the main portion of the iron and weighed. If aluminium and 
 chromium are present, the metals must be separated by methods already 
 given. 
 
 233. Manganese, Cobalt, and Nickel. About 10 c.c. of concentrated sul- 
 phuric acid are added to the filtrate, which is evaporated on the hot 
 plate until the hydrochloric acid is entirely expelled, as indicated by the 
 evolution of dense fumes of sulphuric acid. After cooling, water is added 
 and the solution is neutralized with ammonia. 30 to 50 c.c. of ammonia 
 (sp. gr. 0.96) is then added and the nickel and cobalt determined electro- 
 lytically. The trace of manganese which may be present will be con- 
 verted into the dioxide and deposited on the anode. It may be ignited 
 and weighed as Mn 3 O 4 . 
 
 The deposit of nickel and cobalt after being weighed is dissolved hi 
 warm dilute nitric acid and the solution evaporated to about 50 c.c. 
 It is neutralized with caustic potash and 5 grams of potassium nitrite dis- 
 solved in water are added. Acetic acid is added until the solution is acid 
 and the slight precipitate produced by the caustic potash dissolved. After 
 standing for twenty-four hours, the clear liquid is tested for cobalt by 
 adding more potassium nitrite. If cobalt is present, the entire solution 
 must be allowed to stand for some time after the addition of potassium 
 nitrite. The precipitate is filtered off and washed with a 10% solution of 
 potassium acetate containing a little potassium nitrite. It is then dis- 
 solved in hot dilute sulphuric acid, excess of ammonia added, and the cobalt 
 determined electrolytically. 
 
 334. Sulphur is determined in a separate portion of 0.5 gram, which 
 is dissolved in aqua regia as already directed, or in fuming nitric acid. 
 The reagents must be absolutely free from sulphur. If necessary, a 
 blank determination should be made side by side with the determination 
 of the mineral. Instead of the smaltite, a little pure sodium or potassium 
 chloride should be added to the blank. 
 
 After the expulsion of the nitric acid with hydrochloric acid and filter- 
 ing off the silica, the solution is ready for the precipitation of the sulphuric 
 acid by means of barium chloride. If much sulphur is present, it is well 
 to add a little sodium or potassium chloride to the nitric acid solution 
 before evaporation to prevent loss of sulphuric acid. The evaporation 
 should be conducted on a hot plate or water-bath which is wide enough 
 to keep the products of combustion of the illuminating-gas away from 
 the solution, otherwise sulphur dioxide may be absorbed. Sulphuretted 
 hydrogen or ammonium sulphide must also be absent from the atmosphere. 
 
 The solution is diluted to 300 or 400 c.c., heated to boiling, and barium 
 chloride solution added slowly with constant and vigorous stirring until 
 no more precipitate forms on adding another drop of the reagent. After 
 digesting the precipitate on the hot plate until the solution is clear, it is 
 
172 ANALYSIS OF MINERALS. 
 
 filtered cff, washed with hot water, ignited, and weighed. If the color of 
 the ignited precipitate is not pure white, but reddish, iron has been carried 
 down. Several grams of sodium carbonate are added to the precipitate 
 and after fusion the melt is dissolved in water and the barium carbonate 
 filtered off and washed. The sulphuric acid in the filtrate is precipitated' 
 by barium chloride after acidifying with hydrochloric acid. The error 
 due to a slight reddish color of the precipitate may generally be neglected. 
 
 EXERCISE 36. 
 Analysis of Pyrite, FeS 2 , Arsenopyrite, Fe(S,As) 2 , or Chalcopyrite, CuFeS 2 . 
 
 Small Amounts of Silica, Arsenic, and Antimony May be Present. 
 
 235. Sulphur Available for Sulphuric-acid Manufacture. As the natural 
 sulphides are frequently used as a source of sulphur, especially in the 
 manufacture of sulphuric acid, the determination of this element is 
 frequently carried out for the purpose of ascertaining the amount of 
 sulphur available under the conditions of manufacture of sulphuric acid. 
 For this purpose the determination is carried out as follows: 
 
 One-half gram of the powdered sulphide is treated in an Erlenmeyer 
 flask of about 250 c.c. capacity with 10 c.c. of a mixture of 3 volumes 
 of concentrated nitric acid and 1 volume of concentrated hydrochloric 
 acid. The mouth of the flask is closed with a small funnel and the solution 
 is gently heated on the water-bath until the material is decomposed. If 
 unoxidized sulphur separates, a few crystals of potassium chlorate are added 
 from time to time and the heating continued until the globules of sulphur 
 have disappeared. Liquid bromine may be added for the same purpose. 
 
 The solution is transferred to a porcelain dish and evaporated to dry- 
 ness after the addition of one-half gram of pure sodium or potassium chloride. 
 If the evaporation is conducted on the water-bath only, the addition of 
 the alkali chlorides may be omitted. Five c.c. of concentrated hydro- 
 chloric acid are added and the solution again evaporated to dryness. The 
 residue is dissolved in a few cubic centimeters of dilute hydrochloric acid 
 and 100 c.c. of hot water. The insoluble residue is filtered off on a small 
 paper and well washed with hot water. This residue consists of silica, 
 undecomposed silicates, and the sulphates of lead, barium, and calcium, if 
 these elements are present. This material is reserved for analysis to obtain 
 the percentage of total sulphur present. 
 
 The filtrate is neutralized with ammonia and a considerable excess added. 
 It is warmed for ten to fifteen minutes at 60-7Q, but not to boiling. I/ 
 the solution does not smell strongly of ammonia more must be added. 
 The FERRIC HYDROXIDE is filtered off and washed twice by decantation with 
 very hot water. The precipitate is transferred to the filter-paper, which 
 should not be completely filled, so that it may be well stirred up with the 
 stream of hot water. The precipitate must be washed until the wash- 
 
ANALYSIS OF PYRITE. 173 
 
 water no longer gives a precipitate with barium chloride. As basic sulphates 
 of iron may be present in the ferric hydroxide, the precipitate is dried and 
 fused with sodium carbonate. The melt is dissolved in water and the solu- 
 tion tested for sulphuric acid. 
 
 The nitrate from the iron is acidified with hydrochloric acid, evaporated 
 to a bulk of about 300 c.c., heated to boiling, and barium chloride solution 
 added slowly with constant stirring until a considerable excess is present. 
 Twenty c.c. of a 10% solution is usually more than sufficient for 0.5 gram 
 of pyrite. After digesting hot until the solution is clear, the barium sul- 
 phate is filtered off, ignited, and weighed. 
 
 236. Total Sulphur. To obtain the total percentage of sulphur the 
 material insoluble in aqua regia is fused with sodium carbonate. If lead is 
 absent, the fusion may be conducted in a platinum crucible, otherwise a 
 porcelain crucible must be used. The melt is dissolved in hot water, fil- 
 tered, the residue well washed, and the sulphuric acid in the filtrate pre- 
 cipitated with barium chloride and weighed. 
 
 The total percentage of sulphur may also be determined as follows: 
 Fuse 0.5 gram of the sulphide with 10 grams of a mixture of two parts sodium 
 carbonate and one part potassium nitrate. The material should be well 
 mixed by stirring with a platinum or glass rod, which is then cleaned with 
 a camel's-hair brush. The material is gradually brought to fusion and 
 maintained in the fluid state for some time. The fused mass is extracted 
 with hot water. If lead is present it is precipitated by passing carbon 
 dioxide through the solution. The insoluble material is filtered off and 
 washed with hot water containing a little sodium carbonate, first by decanta- 
 tion and then on the filter-paper. 
 
 The filtrate is acidified with hydrochloric acid and evaporated to dry- 
 ness to remove nitric acid. The residue is dissolved in hot water to which 
 a little hydrochloric acid has been added. Any insoluble material present 
 is filtered off and well washed. The sulphuric acid in the filtrate is deter- 
 mined by precipitation with barium chloride. 
 
 237. Arsenic. For the determination of ARSENIC, ANTIMONY, and COPPER, 
 a portion weighing from 0.5 to 2 grams is taken. If the percentage 
 of these elements is small, the larger amount is taken. The material is 
 treated cold in a beaker with 20 to 30 c.c. concentrated hydrochloric acid. 
 The solution is agitated from time to time. One or two c.c. of liquid 
 bromine or concentrated nitric acid are then added and the solution is 
 heated, gently at first and then more strongly, until the bromine or nitrous 
 fumes are expelled. The arsenic is then distilled and determined as directed 
 in Chapter XI, p. 133. Ferrous sulphate or chloride is used as the reducing- 
 agent. 
 
 238. Antimony. The contents of the flask are rinsed into a beaker, 
 the solution diluted to 200 to 300 c.c., warmed, and hydrogen sulphide 
 passed. The precipitate, consisting of the sulphides of copper and antimony, 
 is filtered off and washed with water containing hydrogen sulphide. The 
 
174 ANALYSIS OF MINERALS. 
 
 antimony sulphide is dissolved in a little sodium-sulphide solution and the 
 residue washed with water containing a little sodium sulphide. The sodium 
 sulphide may be made by saturating with hydrogen sulphide one-half of a 
 caustic soda solution and then adding the other half. The antimony is 
 precipitated by neutralizing the sodium sulphide solution with hydrochloric 
 acid and passing hydrogen sulphide. It may be ignited and weighed as the 
 tetroxide after treatment with nitric acid. 
 
 239. Copper. The copper sulphide is dissolved in a little nitric acid 
 and determined electrolytically. 
 
 240. Iron. Iron is usually determined volumetrically. The ammonia 
 precipitate from the aqua-regia solution made for determining sulphur 
 contains nearly all of the iron. A small amount is frequently left in the 
 insoluble residue as ferrous silicate. On fusing this residue with sodium 
 carbonate the ferrous silicate is decomposed and the insoluble ferric hydrate 
 may be filtered off. These two precipitates may be dissolved in hydrochloric 
 acid and the iron titrated, as given in Chapter XXIV, page 320. 
 
 Both iron and silica may be determined in the portion fused with sodium 
 carbonate and potassium nitrate for the determination of total sulphur. 
 All of the iron will be present in the precipitate filtered off from the water 
 solution of the fused material. It may be dissolved in hydrochloric acid, 
 precipitated with ammonia and weighed as oxide, or it may be determined 
 volumetrically. 
 
 241. Silica. The silica constitutes the insoluble material filtered off 
 after acidifying the sodium carbonate solution, evaporating to dryness 
 and dissolving in hydrochloric acid and water. It is ignited and weighed. 
 
 EXERCISE 37. 
 Analysis of Cinnabar, HgS. 
 
 Iron, Manganese, Copper, Aluminium, and Calcium may also be Present. 
 Sulphur is determined by the method given on p. 167. 
 
 242. Mercury. One gram of the mineral is treated with 20 to 30 c.c. 
 concentrated hydrochloric acid to which 40 c.c. water is added. One to 
 two grams of potassium chlorate are added in small portions and the solu- 
 tion warmed until the cinnabar is entirely decomposed. The solution is 
 evaporated to dryness on the water-bath and the residue dissolved in water 
 and hydrochloric acid. Any insoluble material is filtered off, ignited, and 
 weighed. A little sulphurous acid is added to reduce any iron present to 
 the ferrous condition. The solution is warmed to expel the excess of 
 sulphurous acid and hydrogen sulphide passed until no more precipitate is 
 formed. If copper is absent, the mercuric sulphide is filtered off on a 
 Gooch crucible which has been dried at 100. Sulphur is extracted by 
 washing the precipitate with alcohol and carbon disulphide. The precipi- 
 tate is then dried at 100 and weighed. 
 
 If copper is present the precipitate, after being washed, is dissolved in 
 aqua regia and the chlorine expelled by boiling after the addition of dilute 
 
ANALYSIS OF CINNABAR. 175 
 
 hydrochloric acid. When entirely free from chlorine or nitric acid, phos- 
 phorous acid is ad led and the solution is allowed to stand in a warm place 
 for at least twelve hours. The precipitated mercurous chloride is filtered 
 off on a Gooch crucible, washed with hot water, dried at 100, and weighed. 
 
 243. Copper. The nitrate is evaporated to small bulk after the addition 
 of a few cubic centimeters of dilute sulphuric acid to expel the hydrochloric 
 acid, diluted with water, and the copper determined electrolytically. The 
 copper may also be precipitated as sulphide which is dissolved in 2 or 3 c.c. 
 of concentrated nitric acid, the solution diluted and electrolyzed. 
 
 244. Mercury may also be determined very accurately in a separate 
 portion as follows: A combustion-tube 30 to 45 cm. long and 10 to 15 mm. 
 wide is taken. The end is sealed, then a layer of magnesite in large pieces, 
 chalk, or sodium bicarbonate is introduced until the tube is filled for about 
 10 cm. Then a layer of recently ignited lime is introduced, occupying 
 about 5 cm., then an intimate mixture of the ore with lime for about 6 cm. 
 This mixture should be made by grinding the constituents together in a 
 mortar. The mortar should be rinsed with lime, which is next introduced, 
 occupying about 5 cm. Finally a layer of about 5 cm. of lime is introduced, 
 and a loose plug of asbestos inserted. The tube is then drawn out and 
 bent at right angles. It is then tapped gently so as to produce a free 
 passage for the gases the entire length of the tube, and is then placed in a 
 combustion-furnace with the drawn-out end dipping into a flask. 
 
 The burners of the furnace are lighted so that the lime near the drawn-out 
 end of the tube is first heated to redness. The burners are then lighted in 
 order until the magnesite in the closed end of the tube evolves carbon dioxide, 
 which sweeps out the last traces of mercury vapor. While the tube is still 
 hot it is cut off at the point where it was drawn out and bent and the globules 
 of mercury washed out of the portion cut off by means of a jet of water. 
 By shaking the flask the mercury is brought together into a single drop. 
 The water is decanted and the mercury transferred to a weighed porcelain 
 crucible. The water is absorbed in filter-paper and the mercury dried in a 
 desiccator and weighed. 
 
 245. The Iron, Aluminium, Manganese, and Calcium will be present in 
 the filtrate from the mercuric sulphide. If the amount of iron and alaminium 
 is small, these metals may be precipitated as hydroxides and weighed as 
 oxides. The separation from manganese is effected in the manner described 
 in Exercise 33, p. 149, except that an excess of ammonia must be avoided 
 or the aluminium will be redissolved. The manganese is determined as 
 directed in the same exercise. The filtrate from the manganese is acidified 
 with hydrochloric acid and evaporated to a bulk of 200 or 300 c.c. It is 
 neutralized with ammonia, warmed nearly to boiling, and a little ammonium 
 oxalate added. After digesting for some time, the calcium oxalate is 
 filtered off and washed with hot water. The moist precipitate is placed 
 in the platinum crucible, the paper burned in the usual manner, and the 
 precipitate ignited, finally, over the blast-lamp and weighed as oxide. 
 
CHAPTER XV. 
 
 ANALYSIS OF CARBONATES CONTAINING CAL- 
 CIUM, BARIUM, STRONTIUM, AND MAGNESIUM. 
 
 SEPARATION OF CALCIUM, BARIUM, AND STRONTIUM 
 FROM THE ALKALI METALS AND FROM EACH OTHER. 
 
 246. Calcium, Barium, and Strontium may be separated from 
 magnesium and the alkali metals by precipitation as carbonates, 
 by ammonia and ammonium carbonate. A sufficient amount of 
 ammonium salts must be present to prevent the precipitation of 
 magnesium. It is very seldom that this method of separation is 
 used because the occurrence of these three elements together is 
 very rare, although recent work has shown that traces of barium 
 and strontium occur in rocks far more frequently than was for- 
 merly supposed. Excellent methods of separation of each of the 
 alkaline-earth metals from magnesium and the alkalies are avail- 
 able, by which the metal is left in the best condition for weighing. 
 
 SEPARATION OF CALCIUM. 
 
 247. Separation of Calcium as Oxalate. Calcium is usually 
 separated from magnesium and the alkali metals by means of 
 ammonium oxalate. The precipitation is carried out in a hot 
 solution. Sufficient ammonium oxalate must be present to con- 
 vert both calcium and magnesium into oxalates, since calcium 
 oxalate is slightly soluble in magnesium chloride. Enough 
 ammonium chloride must be present to prevent the precipitation 
 of magnesium hydroxide when the solution is made alkaline. A 
 sufficient amount is usually present from the precipitation of iron 
 and aluminium. If an excessive amount has been introduced, it 
 must be removed by evaporating the solution to dryness and vola- 
 tilizing the ammonium salts. The solution may also be boiled 
 down to small bulk and the ammonium salts decomposed by the 
 addition of nitric acid. It is generally sufficient to treat in this 
 
 176 
 
SEPARATION OF CALCIUM, BARIUM, AND STRONTIUM. Ill 
 
 manner the filtrate from the second, third, and subsequent precipi- 
 tations of the iron and aluminium. If the ammonium salts from 
 the entire solution have been expelled, the residue is dissolved in 
 hydrochloric acid and the solution neutralized with ammonia. If 
 a precipitate forms it is dissolved in acid, and the solution again 
 neutralized. This is repeated until a precipitate of magnesium 
 hydroxide is no longer formed on making the solution alkaline. 
 
 The first precipitate of calcium oxalate will invariably contain 
 some magnesium and also some sodium, which cannot be washed 
 out. It must therefore be redissolved in hydrochloric acid, and 
 reprecipitated by neutralizing with ammonia and adding a little 
 ammonium oxalate. The second precipitate will be free from 
 magnesium and sodium. If STRONTIUM is present it will be almost 
 completely precipitated with the calcium. The two metals may 
 be weighed together as oxides after ignition of the oxalate, and 
 separated by methods given later. If BARIUM is present to the 
 extent of 3 or 4 mg., it will not contaminate the calcium after 
 the second precipitation. If more than 3 or 4 mg. of barium are 
 present, the calcium cannot be separated as oxalate. 
 
 248. Separation of Barium and Strontium as Sulphates. 
 Barium and strontium may be separated from magnesium and 
 the alkali metals by precipitation as sulphate. If strontium is 
 present an equal volume of alcohol must be added to insure com- 
 plete precipitation. The precipitate must also be washed with 
 alcohol. Barium sulphate frequently carries down a small amount 
 of the alkalies which cannot be washed out. It must, therefore, 
 in very exact analyses, be redissolved in a little warm concen- 
 trated sulphuric acid and reprecipitated by diluting with water. 
 
 249. Separation of Calcium as Sulphate. Calcium may also 
 be separated from magnesium and the alkalies by precipitation as 
 the sulphate. The solution must be evaporated until free hydro- 
 chloric acid and water are expelled. The residue is dissolved in 
 alcohol, and a slight excess of concentrated sulphuric acid added. 
 After digesting for some time the solution is filtered and the pre- 
 cipitate is washed with absolute alcohol until free from acid. A 
 small amount of magnesium sulphate is then washed out by means 
 of 35 to 40% alcohol. The alcohol may be expelled from the 
 filtrate and the magnesium precipitated as phosphate. If barium 
 
178 ANALYSIS OF MINERALS. 
 
 and strontium are present they will be precipitated with the 
 calcium. 
 
 SEPARATION OF BARIUM, STRONTIUM, AND CALCIUM. 
 
 250. Precipitation of Barium as Chromate. Two absolutely 
 reliable methods for the separation of barium, strontium, and cal- 
 cium have been devised by Fresenius. The metals are precipi- 
 tated as carbonates with ammonia and ammonium carbonate. 
 If the separation from the magnesium has been effected by sulphuric 
 acid the precipitate is fused with sodium carbonate, the melt 
 treated with hot water, and the carbonates of the alkaline-earth 
 metals filtered off and well washed. The precipitate is dissolved 
 in the least amount of dilute hydrochloric acid, and the excess 
 removed by evaporation. The chlorides are dissolved in 300 c.c. 
 of water, 6 drops of acetic acid (sp. gr. 1.065) and a few drops of 
 ammonium acetate added. The solution is heated, and 10 c.c. 
 of a neutral 10% ammonium chromate solution added. After 
 allowing the precipitate to settle and the solution to cool for one 
 hour, decant on a small filter and wash by decantation two to 
 three times with water containing ammonium chromate. Finally 
 transfer the precipitate to the paper and wash until the filtrate 
 gives no precipitate with ammonia and ammonium carbonate. 
 About 100 c.c. of wash-water should be used. The washing is 
 continued with pure warm water, until the filtrate gives only a 
 slight reddish coloration with silver nitrate. About 110 c.c. will 
 be required. 
 
 The precipitate which contains all of the barium and a little of 
 the strontium is washed into the precipitating-dish, the portion 
 adhering to the filter-paper dissolved in a little warm dilute nitric 
 acid, and the paper well washed. About 2 c.c. nitric acid (sp. gr. 
 1.20) should be used. The solution is diluted to 200 c.c. and 
 heated. 5 c.c. of a 30% ammonium acetate solution is added 
 very gradually. Ammonium chromate solution (10%) is then 
 added until the odor of acetic acid has entirely disappeared. 
 After standing one hour, the supernatant liquid is decanted 
 through the filter or on a Gooch crucible, and the precipitate is 
 digested with hot water which is allowed to cool. The precipitate 
 
SEPARATION OF CALCIUM, BARIUM, AND STRONTIU&. 179 
 
 is then brought on the filter-paper or crucible and washed with 
 cold water until the filtrate gives a scarcely perceptible test for 
 chromic acid with silver nitrate. 
 
 251. Weighing Barium as Chromate. The now pure barium 
 chromate is removed from the paper, the latter burned, and the 
 whole gently ignited in a platinum or porcelain crucible. If the 
 heat is too great the chromic acid will be reduced and the precipitate 
 will become green. The portions of the precipitate adhering to 
 the filter-paper will also be reduced during the incineration of 
 the latter. On gently heating the precipitate, the chromium will 
 be reoxidized. If a Gooch crucible has been used the precipitate 
 may finally be washed with alcohol, dried on the hot plate, and 
 weighed. 
 
 252. Separation of Calcium and Strontium as Nitrates. 
 One c.c. of nitric acid is added to the filtrate from the barium, 
 which is concentrated and the calcium and strontium precipitated 
 by ammonia and ammonium carbonate. If only strontium is 
 present the precipitate of strontium carbonate may be weighed 
 or it may be dissolved ha hydrochloric acid, precipitated by the 
 addition of sulphuric acid and alcohol, and weighed as sulphate. 
 If only calcium is present the precipitation as carbonate may 
 be omitted and the calcium immediately precipitated as oxalate 
 and weighed as oxide. 
 
 If both strontium and calcium are present the mixed carbo- 
 nates are dissolved in nitric acid and the solution evaporated to dry- 
 ness in a small porcelain dish and heated for some time in an air- 
 bath at 130. The temperature may rise to 180 without decom- 
 posing the nitrates. The dried material is pulverized and treated 
 five times with 5-c.c. portions of a mixture of equal parts of abso- 
 lute alcohol and ether and the solution decanted each time into a 
 small flask. The calcium nitrate is dissolved by the alcohol and 
 ether in which the strontium nitrate is insoluble. After expelling 
 the ether and alcohol from the residue, it is dissolved in water, 
 evaporated to dryness, and again heated for some time at 130 hi 
 the air-bath. The mass is pulverized and transferred to the flask 
 with about 15 c.c. of the ether-alcohol. The flask is corked and 
 allowed to stand with occasional shaking for twenty-four hours. 
 The solution is then filtered through a small filter, the strontium 
 
180 ANALYSIS OF MINERALS. 
 
 nitrate washed by decantation and then on the filter with 5-c.c. 
 portions of the ether-alcohol about ten or twelve times. The stron- 
 tium nitrate in the flask, on the paper, and in the dish, if it was 
 not all transferred, is then dissolved in a little hot water and the 
 strontium precipitated as sulphate by the addition of sulphuric 
 acid and alcohol. The calcium in the ether-alcohol solution may 
 be precipitated as sulphate and washed with absolute alcohol. 
 It is ignited at a low red heat and weighed as calcium sulphate. 
 The ether and alcohol may also be distilled off, the calcium nitrate 
 dissolved in water, and the calcium precipitated as oxalate and 
 weighed as oxide. 
 
 As BARIUM as well as STRONTIUM nitrate is insoluble in the 
 ether-alcohol mixture, the carbonates of the three metals may be 
 converted into nitrates and treated in the manner directed for 
 the separation of calcium and strontium. The nitrates of barium 
 and strontium may then be dissolved in water and the barium 
 separated from the strontium by two precipitations of the barium 
 as chromate. Gooch and Browning have suggested the use of 
 amyl alcohol in place of the mixture of absolute alcohol and 
 ether. As the amyl alcohol boils at 138, it may be rendered 
 anhydrous by boiling out the water. The preparation of abso- 
 lute alcohol or ether is quite laborious. 
 
 EXERCISE 38. 
 Analysis of Dolomite, (CaO,MgO)C0 2 . 
 
 Small Amounts of Silica, Iron, Aluminium, and Manganese Are 
 Usually Present. 
 
 253. Solution of the Dolomite. One gram of the powdered material is 
 weighed out, placed in a beaker (about 100 c.c.), and dissolved in about 
 15 c.c. dilute hydrochloric acid. The beaker should be covered with a 
 watch-crystal and the acid poured slowly down the lip to prevent loss 
 by spattering. The beaker is warmed on the hot-plate or water-bath 
 until carbon dioxide ceases to be evolved. The watch-crystal is rinsed with 
 a little water and the open beaker heated on the water-bath until the solu- 
 tion is evaporated to a syrupy consistency. It is then stirred with a glass 
 rod until dry. The residue is heated at least one-half hour on the water- 
 bath and then dissolved in water and a little hydrochloric acid and the 
 solution heated for a few minutes. 
 
ANALYSIS OF DOLOMITE. 181 
 
 254. Silica. The siliceous residue is filtered off on a small paper and 
 washed with hot water until free from chlorides. The moist paper is placed 
 in a weighed platinum crucible, burned in the usual manner, and the siliceous 
 residue finally ignited for a few minutes over the blast-lamp and weighed. 
 The silica is then volatilized by moistening the weighed precipitate with 
 water, adding 2 or 3 c.c. of hydrofluoric acid and a drop of sulphuric acid, 
 warming until the precipitate is dissolved, and then evaporating to dryness 
 on the hot-plate and igniting. The hydrofluoric acid must leave no residue 
 on evaporation. Finally the residue in the crucible is ignited over the blast- 
 lamp and brought to constant weight. The weight of the residue must be 
 subtracted from the weight of the siliceous residue to obtain the weight of 
 silica. The residue is dissolved in hydrochloric acid and added to the 
 filtrate from the silica. 
 
 255. Iron and Aluminium. Ten c.c. of concentrated hydrochloric acid 
 are added to the acid solution, which should not exceed 200 c.c. It is 
 neutralized with filtered ammonia w r hile stirring vigorously and heated 
 nearly to boiling on the hot-plate. If an excess of ammonia is present, 
 very dilute hydrochloric acid is added until the faintest odor of ammonia 
 can be detected over the hot solution or a moist piece of red litmus-paper 
 held over the solution is slowly turned blue. The neutralization and diges- 
 tion should not occupy more than one-half hour and a Jena beaker should 
 be used. The precipitate of iron and aluminium is filtered off and if more 
 than a few flakes are present, it is dissolved in a little warm dilute hydro- 
 chloric acid, the paper washed a few times with hot water, and the solution 
 again neutralized with ammonia, digested on the hot-plate for a short time, 
 and filtered. The precipitate is washed free from chlorides with hot water, 
 the moist paper transferred to the platinum crucible and burned. The 
 precipitate is heated with the blast-lamp for a few minutes and weighed. 
 As this precipitate seldom weighs more than a few milligrams, the separa- 
 tion of the iron and aluminium is not necessary. 
 
 256. Manganese. If the combined filtrates from the iron and aluminium 
 exceed 300 c.c., the solution should be acidified with hydrochloric acid 
 and concentrated on the hot-plate. To precipitate the manganese the solu- 
 tion is made alkaline with ammonia and bromine water or hydrogen per- 
 oxide is added. A few cubic centimeters of hydrogen peroxide are sufficient, 
 while enough bromine water must be added to produce a yellow color in 
 the solution. It is heated nearly to boiling, ammonia added if necessary 
 to make it alkaline, the precipitate filtered off on a small paper and 
 washed with hot water. The moist paper is burned in the crucible and 
 the manganese weighed as Mn 3 4 . 
 
 257. Calcium. The filtrate is again concentrated on the hot-plate to 
 about 250 c.c. after acidifying with hydrochloric acid. The calcium is 
 precipitated by making the solution alkaline with ammonia and adding 
 to the hot solution with vigorous stirring 30 to 40 c.c. of a 5% solution of 
 
182 ANALYSIS OF MINERALS. 
 
 ammonium oxalate. The solution is digested on the water-bath for about 
 one hour or until the precipitate has settled, leaving a clear supernatant 
 liquid. This is decanted through a filter-paper and the precipitate in the 
 beaker is washed two or three times by decantation with hot water. About 
 50 c.c. of water are used each time. The precipitate is well stirred up each 
 time and allowed to settle. A little warm dilute hydrochloric acid is poured 
 over the precipitate on the paper and allowed to flow into the beaker con- 
 taining the main portion of the precipitate. The calcium is reprecipitated 
 by neutralizing the hot solution with ammonia while stirring vigorously. 
 A few cubic centimeters of ammonium oxalate are added and the precipi- 
 tate digested on the hot-plate as before. Finally the calcium oxalate 
 is filtered off and washed with hot water until free from chlorides. The 
 moist precipitate may be placed in the platinum crucible, gently heated 
 until dry, and the paper burned, then heated more strongly until the carbon 
 is completely incinerated, and finally with the blast-lamp for five to ten 
 minutes. After weighing, the heating with the blast-lamp and weighing are 
 repeated until the weight is constant. 
 
 258. Magnesium. The filtrates from the calcium are combined and, 
 after acidifying with hydrochloric acid, evaporated on the hot-plate to 
 200 or 300 c.c. Excess of disodium phosphate or sodium ammonium 
 phosphate solution and a moderate excess of ammonia are added with stir- 
 ring. After being allowed to stand for twenty-four hours the solution 
 is decanted quite completely from the precipitate, which is dissolved in a 
 few cubic centimeters of dilute hydrochloric acid. A few drops of phos- 
 phate solution are added, then ammonia with stirring, until a moderate excess 
 is present. The volume of the solution should now not be greater than 50 
 to 75 c.c. After standing from twelve to twenty-four hours, the precipitate 
 is filtered off and washed with dilute ammonia until free from chlorides. 
 It is then dried, detached from the paper, and the latter burned on the 
 platinum wire. The precipitate may also be placed in the crucible fol- 
 lowed by the folded paper. The precipitate is first heated gently, the lid 
 being on the crucible. When ammonia is no longer evolved, the full flame 
 of the Bunsen burner is applied and finally the blast-lamp for a few minutes. 
 
 259. Carbon Dioxide by Loss. If the dolomite is free from organic 
 matter, carbon dioxide may be determined by loss after fusing with borax.* 
 For this purpose 5 grams of vitrified borax are fused in a platinum crucible 
 with the Bunsen burner. If considerable water is present, it may be more 
 quickly expelled by heating over the blast-lamp. As borax is slightly vola- 
 tile at this temperature, the material must be brought to constant weight 
 by heating to fusion with the Bunsen burner. One gram of the dolomite is 
 placed on the borax, the lid is placed on the crucible, and the contents are 
 
 * Microcosmic salt may also be used for this determination, as described on 
 page 111. 
 
ANALYSIS OF DOLOMITE. 183 
 
 brought to fusion with the Bunsen burner. The lid must not be removed 
 until the borax is fused, as it is apt to crack so that small pieces may be 
 thrown out of the crucible. When the dolomite has dissolved in the melted 
 borax and bubbles of gas escape only at considerable intervals, the crucible 
 is cooled in the desiccator and weighed. When the weight is constant, 
 the amount of carbon dioxide is found from the loss in weight of the borax 
 and dolomite. The borax-glass may be removed from the crucible by 
 pressing it between the thumb and fingers on all sides. This generally loosens 
 the melt so that it drops out on gently tapping the crucible. 
 
 DIRECT WEIGHING OF CARBON DIOXIDE. 
 
 260. Apparatus. Carbon dioxide may also be determined by absorp- 
 tion in caustic potash and ascertaining the increase in weight. For this 
 purpose the apparatus shown in Fig. 18, p. 113, is very convenient and 
 easily made. The glass tube is about 70 cm. long and not less than 12 mm. 
 in diameter. It is drawn out so as to pass through the rubber stopper in 
 the flask. The upper half is filled with granulated calcium chloride, a 
 plug of asbestos or glass wool being first inserted and pushed to the middle 
 of the tube. A few lumps of pumice-stone which have been saturated with 
 copper sulphate solution and dried at about 125 are first dropped in to absorb 
 any hydrochloric acid gas which may pass over. Large lumps of calcium 
 chloride are then dropped in and then smaller ones, the fine material being 
 at the end where another plug of cotton or glass wool is placed. After 
 being filled in this manner, each end is connected with a calcium chloride 
 tube or other drying apparatus and a slow stream of carbon dioxide or 
 hydrochloric acid gas passed through the tube. This is maintained for twenty- 
 four hours, when the acid gas is displaced by a stream of air. The upper end 
 of the tube must be carefully protected from exposure to the atmosphere 
 or moisture. The flask may have a capacity of 100 or 200 c.c. The end 
 of the dropping-funnel should be drawn out and the end of the small tube 
 turned up to prevent the escape of carbon dioxide through the dropping- 
 funnel. A straight calcium chloride tube filled with soda-lime is fitted 
 to the dropping-funnel by means of a small perforated rubber stopper. 
 
 261. For the Geissler caustic potash bulb a solution of caustic potash 
 is made by dissolving one part of the potash in two parts of water. The 
 bulbs are filled not more than two-thirds full with this solution. The cal- 
 cium chloride tube is filled with soda-lime and calcium chloride, the soda- 
 lime being nearest the caustic potash. The calcium chloride should be 
 moderately fine, so that the gases may leave the apparatus well dried. Before 
 placing the soda-lime in the tube a small plug of glass wool or cotton should 
 be inserted. If a caustic potash bulb without the attached calcium chloride 
 tube is used a separate U-tube or straight calcium chloride tube must be 
 filled as directed and weighed separately. 
 
184 ANALYSIS OF MINERALS. 
 
 262. Testing for Air Leaks. When setting up the apparatus, it is well 
 to shellac the small end of the rubber stoppers, as rubber is pervious to 
 carbon dioxide. The small entrance-tube of the Geissler bulb can generally 
 be inserted directly into the rubber stopper in the end of the glass tube, 
 thus saving an extra rubber joint. When the apparatus has been set up. 
 it is tested for leaks by closing the stop-cock in the dropping-funnel and suck- 
 ing out a little air through the caustic potash bulb.. Enough air should be 
 sucked out to cause the solution to rise an inch or two in the limb of the 
 Geissler bulb nearest the long glass tube. This column of liquid forms an 
 index of the partial vacuum in the apparatus. The position of the liquid 
 is carefully noted. If the liquid slowly falls, a leak exists which must be 
 found and closed. The most delicate index is formed by the small drops of 
 liquid that generally remain in one end of the small tubes connecting the 
 bulbs of the absorption apparatus. If the apparatus is absolutely gas- 
 tight, a drop of this kind will remain motionless for ten to fifteen minutes. 
 
 263. Determination of the Carbon Dioxide. When the apparatus 
 has been made tight, 1 gram of the dolomite is weighed out and trans- 
 ferred to the flask. After replacing the flask the apparatus is again tested 
 for leaks. Air is now slowly drawn through the apparatus for about half 
 an hour to remove any carbon dioxide which may be present. The caustic 
 potash bulb is detached, thoroughly cleaned, and after standing in the bal- 
 ance-case for twenty to thirty minutes, is weighed. Whenever this bulb is 
 detached, the hole in the stopper in the end of the long glass tube should 
 be closed with a glass plug. 
 
 After replacing the weighed caustic potash bulb, introduce from 30 to 50 
 c.c. of dilute hydrochloric acid into the dropping-funnel. Open the stop- 
 cock and allow the acid to flow into the flask at such a rate that not more 
 than two bubbles of air per second pass through the caustic potash solution. 
 After all of the acid has been introduced, close the stop-cock in the funnel 
 and warm the flask gently, finally to boiling. A very small flame must be 
 used at first to prevent the gases from passing too rapidly through the 
 caustic potash solution. It must at all times pass slowly enough to permit 
 the bubbles to be readily counted. When water begins to condense in 
 the lower part of the long glass tube, the heating of the liquid should be 
 stopped, the stop-cock quickly opened, and suction applied to the absorption- 
 bulb. When the first violent flow of air into the flask has ceased, air is 
 drawn through the apparatus at the rate of two bubbles per second for 
 about three-quarters of an hour. For this purpose a Bunsen filter-pump 
 may be used. The flow of air may be regulated by a screw pinch-cock. A 
 better device consists of a large bottle filled with water which is led out 
 through a siphon extending to the bottom of the bottle and passing through 
 a doubly perforated stopper. Through the second perforation passes the 
 tube connected to the carbon dioxide apparatus. The flow of water is 
 controlled by a pinch-cock placed on a rubber tube at the end of the siphon. 
 
ANALYSIS OF DOLOMITE. 185 
 
 The caustic potash bulb is disconnected, carefully cleaned, placed in 
 the balance-case,, and weighed. After replacing the caustic potash bulb, 
 the solution of the dolomite is gradually heated to boiling again and air 
 drawn through the apparatus in the same manner as before, after which 
 the caustic-potash bulb is again disconnected and weighed to ascertain 
 if all of the carbon dioxide has been swept out of the apparatus. If the 
 operation has been properly conducted, not more than 2 or 3 mg. will be 
 obtained on the second weighing. When sufficient experience has been 
 gained the operation may be so conducted that only one weighing will be 
 necessary. 
 
CHAPTER XVI. 
 
 ANALYSIS OF SILICATES AND SEPARATION OF 
 SODIUM AND POTASSIUM. 
 
 SEPARATION OF MAGNESIUM FROM THE ALKALIES. 
 
 IN the course of an analysis the other metals are first precipi- 
 tated, leaving magnesium and the alkalies in solution. Magnesium 
 may then be precipitated as phosphate and weighed. During the 
 course of this determination alkalies are introduced into the solu- 
 tion so that it is not available for the determination of these metals. 
 A different method of precipitation of the magnesium must there- 
 fore be adopted or a fresh portion of the unknown must be taken 
 and the magnesium as well as the other heavy metals separated 
 from the alkalies. The latter is by far the most satisfactory method 
 of procedure, and is invariably adopted when sufficient material 
 for analysis is at hand. 
 
 264. Separation of Magnesium by Means of Mercuric Oxide. 
 The magnesium may be removed without introducing alkalies 
 by means of mercuric oxide. The solution of the chlorides 
 from which other acids must be absent is treated with mercuric 
 oxide which is free from alkalies. The operation should be con- 
 ducted in a platinum dish. It is best to use freshly precipitated 
 moist mercuric oxide. If much ammonium chloride is present, 
 the solution should first be evaporated to dryness in the platinum 
 dish and the residue gently ignited until ammonium salts are 
 expelled. The residue is then dissolved in a little water and 
 excess of the mercuric oxide added and the material well stirred. 
 The solution is evaporated to dryness with frequent stirring. 
 The residue is heated, at first gently and then more strongly, until 
 mercuric chloride is no longer evolved. All of the mercuric oxide 
 need not be volatilized. The residue is extracted repeatedly 
 with small quantities of hot water and the solution rapidly fil- 
 
 186 
 
SEPARATION OF MAGNESIUM FROM THE ALKALIES. 187 
 
 tered. The washing must be discontinued as soon as the residue 
 is free from chlorides. 
 
 By this operation the magnesium is converted into oxide accord- 
 ing to the following equation: MgCl 2 +HgO=MgO + HgCl 2 . The 
 chlorides of the alkalies are not decomposed in this manner. As 
 the magnesium oxide is slightly soluble, the solution of the alkalies 
 will contain a little magnesium. Unless the amount of magnesium 
 present is less than 1%, the treatment with mercuric oxide must be 
 repeated. The insoluble mixture of magnesium and mercuric 
 oxides may be ignited, the mercury volatilized, and the magnesium 
 weighed as oxide. The magnesium may also be dissolved in 
 hydrochloric acid after volatilization cf the mercury and deter- 
 mined as phosphate. The mercuric oxide used must not leave 
 a residue on volatilizing a portion by heating it to redness in a 
 platinum dish or crucible. Alkalies are frequently present and 
 remain as an alkaline^ residue after volatilization of the mercury. 
 If mercuric oxide free from alkalies cannot be obtained the amount 
 used must be weighed and a correction made for the amount of 
 alkali present. 
 
 265. Separation of Magnesium by Means of Barium Hydroxide. 
 The magnesium may also be removed by treating the concen- 
 trated solution with a solution of barium hydroxide until it is 
 strongly alkaline. After filtering off and washing the magne- 
 sium hydroxide the excess of barium must be removed from the 
 solution of the alkalies by precipitation with ammonia and am- 
 monium carbonate. 
 
 266. Separation of the Alkalies. The solution of potassium 
 and sodium chlorides is evaporated to dryness in a weighed plati- 
 num dish, finally on the water-bath. The residue must be heated 
 to low redness to completely dry the chlorides before weighing. 
 Two difficulties are met with in this process. The crystals of 
 sodium and potassium chloride tend to enclose water so that on 
 drying the material decrepitates with considerable violence, tending 
 to throw the material out of the dish. The chlorides of sodium 
 and potassium are volatile at a very low temperature. It is 
 therefore necessary after evaporating on the water-bath to cover 
 the platinum dish with a watch-crystal and transfer it to the hot- 
 plate and dry as completely as possible before heating with the 
 
188 ANALYSIS OF MINERALS. 
 
 Bunsen burner. The burner must be held in the hand and the 
 flame waved under the dish and removed as 
 soon as any portion of the dish becomes red- 
 hot. If the decrepitation is excessive, the 
 watch-crystal on the dish may be inverted 
 and the platinum dish placed in another of 
 larger size. Any material which is thrown out 
 of the first dish is then deflected downwards into the second one. 
 
 The weighed chlorides are dissolved in water. A slight amount 
 of insoluble matter which is usually present is filtered off on a 
 small filter-paper and well washed. The paper is thrown into the 
 platinum dish and ignited. The dish with the residue is weighed. 
 This weight is subtracted from the weight of the dish and the dry 
 chlorides. To the solution of the alkali chlorides enough platinic 
 chloride is added to convert both the sodium and potassium chlor- 
 ides into the chlorplatinic salts. The subsequent manipulation 
 is the same as that given for potassium in Chapter VII. The 
 sodium is found by difference. 
 
 DECOMPOSITION OF SILICATES. 
 
 The first step in the analysis of a silicate has for its object the 
 breaking of the bond between the silica and the basic elements 
 present, so that the latter may be brought into solution. As the 
 silicates are among the most stable bodies found in nature, the 
 accomplishment of this object is not always easy. When com- 
 pletely decomposed, or opened up, as the operation is sometimes 
 termed, the analysis of the silicate involves only the ordinary 
 problems of separation and determination. 
 
 267. Fusion with Alkali Carbonates. By far the most largely 
 used and perhaps the best method of accomplishing this object 
 consists in fusing the very finely ground material with sodium 
 carbonate or the so-called fusion mixture, which consists of sodium 
 and potassium carbonates in the proportion of their molecular 
 weights. This mixture fuses at a somewhat lower temperature 
 than sodium carbonate alone. The result of this method of 
 decomposition is that the silica is largely converted into sodium 
 silicate, the silicic acid displacing the volatile carbonic acid in 
 
DECOMPOSITION OF SILICATES. 189 
 
 the alkali carbonate. The bases remain as carbonates, or if a 
 given base forms a carbonate unstable at the temperature of 
 the fusion it will remain as oxide. The bases which have marked 
 acid properties, such as aluminium, will combine more or less 
 completely with the alkali. 
 
 268. Determination of Silica. On treating the fused mass 
 with water and acidifying with hydrochloric acid the alkali sili- 
 cate is decomposed with liberation of silicic acid. The remaining 
 alkali salts which are formed are also decomposed. On evapo- 
 rating to dryness, to dehydrate the silicic acid, and again taking up 
 with water and hydrochloric acid, all of the elements present go 
 into solution with the exception of silica and part of the TITANIUM 
 which is frequently present in silicates. It is seldom that the 
 decomposition of the silicate and the solution of the bases in 
 hydrochloric acid is so complete that the silica is absolutely pure. 
 On the other hand, some of the silica goes into solution even after 
 very prolonged drying. 
 
 To recover the dissolved silica, the filtrate is evaporated to 
 dryness and the residue again taken up with water and hydro- 
 chloric acid. The most complete separation of silica is effected 
 by heating the dry residue on the water-bath for one-h&if to 
 one hour. If the residue is dried on the hot-plate or in tt j air- 
 oven at a temperature above 100, more of the silica tends to go 
 into solution again. The residue after the first evaporation may 
 be dried on the hot-plate. This is a more rapid method, and any 
 silica which goes into solution will be recovered by the evaporation 
 of the filtrate, the residue of which must be dried at 100. A small 
 amount of silica (1 to 2 mg.) goes into solution even after the 
 second evaporation. This is precipitated with the iron and alu- 
 minium and may be recovered by dissolving these elements in 
 acid-potassium sulphate, dissolving the melt in sulphuric acid, 
 evaporating till fumes appear, diluting with water, and filtering. 
 
 269. Determination of Aluminium, Iron, and Titanium. The 
 aluminium and titanium oxides which are usually present with the 
 silica may be recovered by treating the ignite< 1 and weighed pre- 
 cipitate with hydrofluoric and sulphuric acids and evaporating. 
 In this manner the silica may be completely volatilized. Unless 
 
190 ANALYSIS OF MINERALS. 
 
 sulphuric acid is present, some of the titanium will also be volatil- 
 ized. After blasting, the residue is weighed and the correction 
 applied to the weight of the silica. As the bulk of this residue is 
 generally composed of titanium and aluminium oxides, it is advis- 
 able to combine it with the precipitate of aluminium, iron, and 
 titanium hydroxides. After igniting and weighing, the combined 
 precipitate is dissolved by fusing with acid-potassium sulphate. 
 The iron is reduced by passing hydrogen sulphide through the 
 water solution of the fusion. It is then determined volume trically. 
 The titanium is then oxidized by means of hydrogen peroxide 
 and the amount determined by comparing the intensity of color 
 with the color of a standard titanium solution. (See page 194.) 
 
 The other metals are separated and determined by methods 
 which have already been described. 
 
 270. Decomposition of Silicates for the Determination of the Al- 
 kalies. As alkalies have been introduced into the fusion mixture, 
 these metals must be determined in a separate portion, which must 
 be decomposed by a different method. One of the best of these 
 methods is that of J. LAWRENCE SMITH. The silicate is decom- 
 posed by gently heating an intimate mixture of the silicate with 
 
 . 1 part of ammonium chloride and 8 parts of pure calcium carbonate. 
 The alkalies as well as some of the calcium are converted into 
 chlorides. Ammonia, the excess of the ammonium chloride, and 
 carbon dioxide are volatilized. On treating with water, the chlo- 
 rides of the alkalies dissolve, while the other bases together with 
 most of the calcium remain undibsolved. One advantage of the 
 method is found in the fact that all of the magnesium remains 
 in the insoluble residue, thus separating it from the alkalies. The 
 calcium in the nitrate is precipitated with ammonia and ammo- 
 nium carbonate and filtered off. The nitrate is evaporated to 
 small bulk and a little more calcium precipitated with ammonium 
 oxalate and filtered off. The alkalies in the filtrate, which now 
 contains nothing but these metals as chlorides, are separated and 
 determined by methods already given. 
 
 271. Decomposition of Silicates by Fusion with Boric Oxide. 
 Another method for decomposing silicates by which a solution is 
 ob tamed in which all of the bases, including the alkalies, may be 
 
DECOMPOSITION OF SILICATES. 191 
 
 determined, has been devised by JANNASCH and HEIDENREICH.* 
 The flux is powdered boric oxide. It is prepared by recrystallizing 
 two or three times the commercial article to free it from traces of 
 alkali. The purified crystals are dried and fused in a large plati- 
 num crucible until free from water. By suddenly cooling the 
 fused mass it cracks into pieces of a convenient size for powdering. 
 The powdered material is kept in well-stoppered bottles, as it is 
 hygroscopic. 
 
 Nearly all silicates are readily decomposed when mixed with 3 
 to 8 or more parts of this flux and heated, at first gently, and finally 
 with the blast-lamp. The silicate must be very finely powdered, 
 from one-half to one hour being taken for the grinding of 0.5- to 
 1-gram portions. A platinum crucible holding 45 to 60 c.c. 
 should be used. After intimately mixing the silicate and the 
 flux, gentle heat from a Bunsen burner is applied. Frothing 
 and rising in the crucible is prevented as far as possible by stirring 
 with a short platinum wire which does not reach above the edge 
 of the crucible. When the material has been in quiet fusion for 
 some time in the covered crucible, the flame of the blast-lamp is 
 applied. From twenty to thirty minutes are generally required for 
 the entire operation. 
 
 Some silicates like andolusite, cyanite, and topaz are not fully 
 decomposed when heated with the flux by means of the blast- 
 lamp, For these minerals more of the flux must be used, as high 
 as 30 parts to 1 of the mineral being taken. After the preliminary 
 heating with the Bunsen burner, a few additional grams of boric 
 oxide are added and a flame which is fed by oxygen in place of air 
 is applied. An ordinary blast-lamp is used with an opening 2J mm. 
 wide. It is supplied with gas from at least 5 or 6 ordinary gas- 
 cocks and the flame is made broad and free from luminosity. The 
 heating is continued until the fusion is as transparent as glass. 
 
 272. Volatilization of the Boric Acid. After fusion, the hot 
 covered crucible is immersed in cold water until cold. The con- 
 tents are then placed in a large porcelain or platinum dish and a 
 saturated solution of hydrochloric acid hi methyl alcohol is added 
 
 * Zeit. fur anorg. Chem., Vol. XII, p. 208, 1896. 
 
192 ANALYSIS OF MINERALS. 
 
 while the dish is covered with a watch-crystal. The crucible is 
 rinsed with more of the methyl chloride and the solution in the 
 dish is gently heated until, with occasional additions of the methyl 
 chloride, solution is complete. This requires from ten to fifteen 
 minutes. The solution is then boiled down to small bulk and 
 evaporated to dryness on a water-bath in which the water is not 
 quite boiling, so that the temperature in the dish is 80 to 85. 
 More methyl chloride is added and the heating at this temperature 
 is continued until the boric acid is completely expelled. All of 
 the metals will then be in solution as chlorides and may be sepa- 
 rated and determined by methods already given. 
 
 This method of decomposing silicates has the advantage that 
 a solution of the mineral is obtained which is small in bulk and 
 free from the large amount of alkali salts which are introduced 
 by the carbonate fusion. The separations of many of the metals 
 are much cleaner and sharper in the absence of a large amount 
 of the alkali salts. All of the metals, including the alkalies, may 
 be determined in one portion. This is a decided advantage 
 when only a small amount of the mineral is at hand. On the other 
 hand, the purification of the boric acid, and more especially the 
 uncertainty as to the complete decomposition of the mineral, are 
 disadvantageous.* 
 
 EXERCISE 39. 
 
 Analysis of Feldspar, (K 2 0,Al 2 3 )(Si0 2 ) fl . 
 Iron, Calcium, Magnesium, Sodium, and Titanium are also Usually Present. 
 
 273. Decomposition of the Mineral. One gram of the finely powdered 
 material is placed in a rather large platinum crucible and mixed with 5 
 grams of a mixture of 5 parts of sodium and 7 parts of potassium carbonate. 
 The crucible is heated with the Bunsen burner until the mass is in quiet 
 fusion and no more carbon dioxide is evolved. About one-half hour will be 
 required. Seize the crucible with the tongs and cause the contents to solidify 
 in a thin film around the sides of the crucible by giving it a rotary motion 
 while the contents are cooling. Transfer the crucible to a beaker just large 
 enough to allow the crucible to be placed on its side. Add enough water 
 to cover the crucible and heat on the water-bath until the melt is 
 disintegrated. Take out the crucible with a pair of tongs and rinse it well. 
 The lid must also be well washed. 
 
 * For common errors in the determination of silica, see Jour. Am. Chem. Soc., 
 J4 (1902), p. 262. 
 
ANALYSIS OF FELDSPAR. 193 
 
 The material in the beaker should be examined for undecomposed sili- 
 cate by rubbing the bottom with a glass rod. Any gritty substance is 
 undecomposed feldspar. The main portion of the solution may be decanted 
 and a little water added. When the material in the water is stirred .up, 
 undecomposed silicate settles quickly, while separated silica rises readily 
 and settles slowly. When strong hydrochloric acid is added and the beakei 
 is heated, undecomposed silicate does not dissolve. Filter it off on a small 
 paper, wash, and add the nitrate to the main solution. Burn the paper 
 and fuse the insoluble material with five times its weight of the mixture of 
 sodium and potassium carbonate. The second fusion is treated exactly 
 as the first. Failure to secure complete decomposition the first time is 
 generally due to insufficient grinding of the silicate. 
 
 273. Silica. Transfer the alkaline solution to a platinum or porcelain dish, 
 co^ r er with a watch-crystal and add excess of hydrochloric acid. The fused 
 material may also be dissolved by the following method. The melt is 
 poured into the crucible cover. As soon as redness has gone, and while 
 still hot, slip the crucible and cover into a covered beaker containing enough 
 dilute hydrochloric acid to neutralize the sodium carbonate. Solution is 
 complete in a few minutes and most of the silicic acid remains in solution. 
 Decant the solution if there is any undecomposed silicate which is filtered 
 off and again fused. 
 
 The solution is evaporated to dryness on the water-bath and heated 
 for at least one-half hour. Add 20 c.c. dilute hydrochloric acid and then 
 heat on the water-bath until nothing more dissolves. About one-half hour 
 is generally sufficient. Add 50 c.c. hot water, filter off the silica and wash 
 a few times. Evaporate the filtrate to dryness and heat one-half hour on 
 the water-bath. Add 20 c.c. dilute hydrochloric acid to the residue and 
 heat on the water-bath until nothing more dissolves. Add 50 c.c. hot 
 water and filter off the silica on the paper containing the main portion, 
 and wash with hot water containing hydrochloric acid, and finally with 
 pure water, until the precipitate is free from chlorides. Particular care 
 should be taken to wash this silica very thoroughly. If it should turn black 
 on ignition it indicates insufficient washing. 
 
 Drjr the precipitate, transfer it to the weighed platinum crucible, fold 
 the paper and place it on the precipitate. Cover the crucible and heat 
 gently with the Bunsen burner until volatile matter is expelled, finally 
 with the full flame of the Bunsen burner. Remove the cover, incline the 
 crucible, and continue heating until the carbon is completely burned and the 
 precipitate is pure white. Replace the cover and heat with the blast-lamp 
 for fifteen to twenty minutes, cool, and weigh. Moisten the precipitate with 
 a little water, add a few cubic centimeters of hydrofluoric acid and a drop 
 or two of sulphuric acid. Evaporate the acid and ignite the residue with 
 the blast-lamp. The weight of the residue must be subtracted from the 
 weight of the silica. The aluminium iron precipitate is ignited in this plat- 
 inum crucible together with the small residue from the silica precipitate. 
 
194 ANALYSIS OF MINERALS. 
 
 275. Aluminium. In the absence of manganese, the IRON, TITANIUM, 
 and ALUMINIUM may be precipitated with ammonia and ammonium chloride. 
 The precipitate must be redissolved and reprecipitated. If MANGANESE is 
 present, the alkali fusion of the silicate will generally have a bluish-green 
 color. Absence of this color is not proof of the absence of manganese. In 
 the presence of manganese, the aluminium must be precipitated as basic 
 acetate. The second precipitation, however, may be made by ammonia 
 and ammonium chloride. These precipitations are carried out as directed 
 on p. 157, Chapter XIII. On concentrating the two nitrates separately to 
 i bulk of 75 or 100 c.c., a small precipitate of aluminium hydroxide some- 
 times separates out in one or both of the solutions. This is filtered off on 
 a small filter-paper, the first filtrate being first passed through the paper, 
 the second one being used to wash the first beaker and the filter-paper. 
 The small precipitate is then redissolved in a little hydrochloric acid and 
 reprecipitated, filtered off on the same paper, washed, and added to the main 
 precipitate. 
 
 276. Iron. After igniting and weighing the aluminium-iron precipitate, 
 it is dissolved in acid-potassium sulphate, 3 or 4 grams of which are placed 
 in the crucible, which is heated just high enough to melt the acid sulphate. 
 Several hours' heating may be required to dissolve a large precipitate. 
 When everything is dissolved, except possibly a little silica, the crucible 
 is cooled and the contents dissolved in dilute sulphuric acid and the solu- 
 tion evaporated until fumes of sulphuric acid are evolved. Hot water is 
 added and the silica is filtered off, ignited, and weighed. Its weight is to 
 be subtracted from the weight of the alumina and ferric oxide. The solu- 
 tion is placed in an Erlenmeyer flask of about 250 c.c. capacity. The flask 
 is fitted with a two-holed rubber stopper, a glass tube extending to the 
 bottom of the flask being passed through one hole while a short glass tube 
 is inserted in the other. The solution is heated nearly to boiling and hydro- 
 gen sulphide passed to reduce the iron. The excess of hydrogen sulphide is 
 expelled by boiling the solution and passing a stream of carbon dioxide. 
 When the hydrogen sulphide is entirely expelled, as indicated by lead- 
 acetate paper, the solution is cooled, while the stream of carbon dioxide is 
 passing and titrated with standard potassium-permanganate solution. The 
 iron cannot be reduced by zinc, since the titanium would then be reduced 
 also. After titrating the iron, the solution is suitable for the determina- 
 tion of titanium as described in the following paragraph. 
 
 277. Titanium. The titanium is estimated by comparing the color 
 of the unknown solution, after oxidation with hydrogen peroxide, with 
 the color of a standard solution of titanium. The standard solution of 
 titanium sulphate should contain 1 gram of Ti0 2 per liter or the equivalent 
 amount of titanium sulphate. 5% or more of sulphuric acid must be 
 present in this solution. 10 c.c. are carefully measured out and mixed 
 with 2 c.c. of 5% hydrogen peroxide free from fluorides and diluted to 100 c.c. 
 in a measuring-flask. The solution to be tested is evaporated to about 
 
ANALYSIS OF FELDSPAR. 195 
 
 75 c.c., transferred to a 100-c.c. flask, hydrogen peroxide added as long as 
 the color is intensified, and the solution finally made up to 100 c.c. with dilute 
 sulphuric acid. If the color of the solution is much more intense than 
 that of the standard solution, it should be transferred to a larger flask and 
 diluted with water and sulphuric acid. 50 c.c. of the standard solution is 
 placed in a Xessler tube. The unknown solution is introduced from a 
 burette into a second Nessler tube until the color seems equal to that of the 
 standard. It is then diluted to 50 c.c. After shaking, the color is again 
 compared with that of the standard. If the colors are not identical, a third 
 Xessler tube is taken and a little more or less of the unknown solution is in- 
 troduced as indicated by the first test. After diluting to 50 c.c., the colors 
 are again compared. If the unknown solution is more dilute than the 
 standard, the process is reversed. As 50 c.c. of the standard solution con- 
 tains 5 mg. of TiO 2 , the volume of the unknown solution which gives the 
 same depth of color will also contain 5 mg. of titanium oxide. The total 
 amount may then be easily computed. The amount of aluminium oxide 
 is found by subtracting the weight of iron found computed as Fe 2 O 3 plus 
 the weight of Ti0 2 from the total weight of the precipitate. 
 
 278. The Manganese, Calcium, and Magnesium are separated and deter- 
 mined as given in Exercise 38, page 181. 
 
 DETERMINATION OF THE ALKALIES. 
 
 279. Decomposition of the Mineral. For the estimation of the alkalies, 
 1 gram of the finely ground mineral is intimately mixed in a platinum 
 crucible having a closely fitting cover with about 1 gram of resublimed 
 ammonium chloride and 8 grams of calcium carbonate which is free from 
 alkalies. The thorough mixing required is best effected by grinding together 
 in a large agate or porcelain mortar the mineral and the ammonium chloride. 
 The carbonate of lime is divided into four parts. Three of the portions 
 are successively added and thoroughly mixed. The material is then trans- 
 ferred to the platinum crucible. The mortar is rinsed with the fourth 
 portion of the carbonate of lime. The glazed paper, which should from the 
 beginning be under the mortar, is then brushed off. The contents of the 
 crucible are settled down by gently tapping it on the table. If a large 
 enough crucible is not at hand, half the quantities may be used. 
 
 The crucible is covered and placed in an inclined position on a pipe-stem 
 triangle. It is at first heated gently with the small flame of a Bunsen 
 burner, placed so that the flame does not touch the crucible. As soon as 
 the odor of ammonia is no longer perceptible, which should be after 
 about ten minutes' heating, the crucible is heated to dull redness for not 
 more than two-fifths of its height for forty or fifty minutes. 
 
 280. Removal of Calcium. After cooling, the sintered cake, which is gen- 
 erally detached very readily from the crucible, is placed in a porcelain dish 
 and 60 to 80 c.c. of hot water added. The crucible is washed out with 
 
196 ANALYSIS OF MINERALS. 
 
 hot water and the cake is broken up with a blunt rod. The water is brought 
 to a boil and after being allowed to settle, the solution is decanted through 
 a filter-paper. The material in the dish is washed two or three times with 
 hot water by decantation, finally brought on the filter-paper, and washed 
 free from chlorides. The residue may be tested for unattacked mineral 
 by dissolving in hydrochloric acid. 
 
 The calcium in the filtrate is precipitated by neutralizing with ammonia 
 and adding ammonium carbonate and ammonia. The precipitation should 
 be carried out in a platinum or porcelain dish and the solution should be 
 heated nearly to boiling. After digesting some time, the precipitate is 
 filtered off and well washed with hot water. The filtrate is evaporated 
 to dryness in a platinum dish and the ammonium sa^.ts expelled. The 
 residue is dissolved in a few cubic centimeters of water, a few drops of 
 ammonium-oxalate solution added, and the small precipitate of calcium 
 oxalate filtered off and washed. 
 
 281. Potassium and Sodium. The filtrate, which now contains only 
 sodium and potassium as chlorides and a little ammonium oxalate and 
 chloride, is evaporated to dryness in a weighed platinum dish. The ammo- 
 nium salts are expelled by gently heating the dish, covered with a watch- 
 crystal, first on the hot-plate and then with the flame of the Bunsen burner. 
 The dish must not be heated higher than to dull redness, and that only for 
 a moment. The burner is held in the hand and waved under the dish. 
 (See p. 187, 266.) The residue is moistened with hydrochloric acid to 
 convert any alkali carbonate into chloride. It is again evaporated to 
 dryness, ignited and weighed. 
 
 When the alkali chlorides have been brought to constant weight they 
 are dissolved in a few cubic centimeters of water and enough 10% platino- 
 chloride solution added to convert all of the alkali chlorides present into 
 the double platinum salt. In this calculation, the amount of platinum 
 solution necessary to convert the precipitate into the double chloride, if 
 it were all sodium chloride, should be found and a little more than this 
 amount added. The solution is evaporated to a syrupy consistency and 
 about 50 c.c. of 80% alcohol, added. The evaporation and subsequent 
 manipulation must be carried out in an atmosphere which is free from 
 ammonia fumes. The solid material is occasionally well stirred with a 
 glass rod until no more solvent action can be observed. The solution is 
 decanted through a Gooch crucible, which has been dried at 110 to 115. 
 More 80% alcohol is added, the precipitate digested and stirred, and 
 the alcohol decanted. When the alcohol no longer becomes strongly col- 
 ored, the precipitate is transferred to the crucible and washed with small 
 portions of the 80% alcohol until it comes through colorless. Dry the 
 crucible at 110 to 115 and weigh. The amount of potassium chloride is 
 now computed and subtracted from the weight of the two chlorides to obtain 
 the amount of sodium chloride present. 
 
ELECTROLYTIC METHODS. 
 | CHAPTER XVII. 
 
 THE IONIC THEORY; ELECTROLYTIC APPARATUS AND 
 MANIPULATIONS. 
 
 THE separation of metals from their solutions and their 
 electrolytic deposition in films which can be dried and weighed 
 offers in the case of many metals rapid and accurate methods of 
 determination, and while in many cases the time necessary for 
 complete deposition of the metal is considerable, little or no atten- 
 tion from the operator is required. The methods in use have 
 generally been empirically discovered without reference to the 
 prevailing theories of the nature of electrolysis. The subject is 
 made more intelligible when studied in the light of theories which 
 serve to combine and harmonize all of the facts known. 
 
 282. The Ionic Theory of Electrolysis. It is now generally 
 believed that when the salt of a metal is dissolved in water, the 
 molecules are separated by the water into at least two parts ; which 
 are electrically charged and are called ions. In the case of copper 
 sulphate, the copper atom is separated as an ion from the remainder 
 of the molecule, the sulphur and oxygen remaining in combination 
 and constituting a second ion. The molecule of cupric chloride 
 separates into three ions, one of which is the copper atom, while 
 the other two are the chloride atoms. When an electrical current 
 is passed through the solution, the metallic ions move with the 
 current and tend to accumulate around the cathode, and are 
 therefore called CATHIONS, while the ions composed of the acid 
 radicles move in a direction opposite to that in which the current 
 is moving and tend to accumulate around the anode, and are 
 therefore called ANIONS. This separation of the constituent parts 
 of a metallic salt is called electrolysis. 
 
 An ION may therefore be defined as a metallic atom, or an acid 
 radicle, which is charged with positive or negative electricity and 
 
 197 
 
198 ELECTROLYTIC METHODS. 
 
 exists in the solution of salts, bases, and acids, in water and a few 
 other solvents. 
 
 283. Amount of the Electrical Charge on Ions. According to 
 Faraday's law the amount of a given metal which is deposited 
 from a solution is proportional to the amount of current passed, 
 provided no other decomposition is effected by the current. For 
 example, a current of one ampere passing for one hour will deposit 
 4.026 grams of silver, any change in the amount of current passed 
 producing a proportionate change in the amount of silver depos- t 
 ited. This would be the case if each metallic atom or ion bore a 
 fixed and definite amount of electricity, since the current is carried 
 through the solution only by means of the ions and when a given 
 metallic ion gives up its load of electricity at the cathode it is 
 deposited in metallic form. 
 
 As, still further, the amount of a given metal deposited by a 
 current is inversely proportional to the valence of the metal, the 
 charge on each ion must be directly proportional to the valence. A 
 current of one ampere passing for one hour through a solution of 
 cupric copper will deposit 1.184 grams of copper, while the same 
 current passing for one hour through a solution of cuprous copper 
 will deposit twice as much, or 2.368 grams of copper. An atom 
 of copper present as cuprous chloride must therefore carry only 
 half as much electricity as an atom of copper existing as cupric 
 chloride. As the valence of the chlorine atom in these com- 
 pounds is the same, and it is charged with negative electricity, 
 the amount of this negative charge on each of the chlorine atoms 
 must be the same since the solutions of both cuprous and cupric 
 chloride are uncharged; that is, they contain the same amount 
 of positive and negative electricity. 
 
 While the amount of a given metal deposited is proportional 
 to the current used, the amounts of different metals deposited by 
 the same current are directly proportional to the atomic weights 
 and inversely proportional to the valence of the metals. The 
 
 M M' M" 
 
 quantities , , 77, etc., express the relative amounts of dif- 
 ferent metals deposited by the same current, where M, M', M" ', 
 etc., represent the atomic weights of the metals used and n, n', n" 
 represent the valences of the metals as they existed in the solu- 
 
THE IONIC THEORY. 199 
 
 tions used. Taking silver, cupric copper, and gold as illustrations, 
 the relative amounts of the metals deposited will be in the ratio of 
 
 />O 1O I Q >J 
 
 107.66, 2~-, and ^-, or 107.66, 31.59, and 65.57. These 
 
 numbers are in the same ratios as 4.026, 1.184, and 2.458, the 
 amounts deposited by one ampere-hour of current. 
 
 In the language of the present electrolytic theory, an atom of 
 any metal in the ionic condition carries a positive charge of 
 electricity the amount of which is proportional to the valence of 
 the metal and is the same for all metals of equal valence. All 
 acid atoms or radicles carry a charge of negative electricity 
 which is proportional to the valence and independent of the 
 nature of the acid. Atoms of sodium, potassium, silver, and 
 ammonium carry the same amount of positive electricity, while 
 monovalent acid radicles such as chlorine, bromine, iodine, N0 3 , 
 C10 3 , etc., carry the same amount of negative electricity. 
 
 284. Potential of the Electrical Charge on Ions. While the 
 amount of the charge on all univalent atoms is the same, the 
 potential is characteristic of the metal concerned. This may 
 readily be shown by electrolyzing solutions of the various metals 
 in which the same acid anion is present. The minimum differ- 
 ence in potential which must be impressed on the electrodes 
 before a given metal will be deposited will be found to be character- 
 istic of that metal. 
 
 The following minimum decomposition tensions were deter- 
 mined by LeBlanc for normal solutions : 
 
 Volts. Volts. Volts. 
 
 ZnS0 4 2.35 NiCl 2 1.85 Cd(N0 3 ) 2 . . . 1.98 
 
 NiS0 4 2.09 CdCl 2 . 1.88 Pb(N0 3 ) 2 . . 1.52 
 
 CdS0 4 2.03 CoCl 2 1.78 AgN0 3 0.70 
 
 CoS0 4 1.92 
 
 This peculiarity of the metals has been used to effect their 
 separation. The metal which separates at the lowest potential 
 is taken out first. On increasing the electromotive force of the 
 current other metals may be deposited. 
 
 In a sulphuric-acid solution of cobalt and zinc, for example, 
 the cobalt could be deposited by a current of an electromotive 
 force less than 2.35 volts and greater than 1.92. After all of the 
 
200 ELECTROLYTIC METHODS. 
 
 cobalt had been deposited the zinc could be deposited by increas- 
 ing the electromotive force of the current to more than 2.35 volts. 
 
 285. Significance of Current Density. Not only must the poten- 
 tial of the current used exceed a definite quantity before a given 
 metal will be deposited, but the amount also of current used 
 must be sufficiently large. The necessity for this is found in 
 the fact that the deposited metal tends to redissolve in most of 
 the solutions from which it is deposited. The rate at which the 
 metal is deposited must exceed that at which it redissolve s. As the 
 rate at which the metal dissolves is proportional to the amount 
 of surface exposed, the amount of current necessary to deposit the 
 metal will depend on the area of the cathode surface. The amount 
 of current per unit area, known as the current density, is there- 
 fore of importance in electrolytic determinations. The area 
 generally used as the unit is 100 sq. cm., and statements of the 
 amount of current necessary refer to the amount needed over 
 such an area. The symbol ND 100 is used for such a current den- 
 sity. The expression ND 1QO = 1.5 amperes means that 1.5 amperes 
 of current should be used for each 100 sq. cm., of cathode area. 
 
 As metals differ greatly in the ease with which they dissolve 
 in a given acid, it is possible to so choose the strength of current 
 to be used that with a given concentration of acid in a solution 
 of two metals one will be deposited while the other will remain 
 in solution. Copper and iron are generally separated in this 
 manner. 
 
 While the tendency of the metal to redissolve in the solution 
 fixes a lower limit to the amount of current which may be used, 
 the tendency of most metals to form deposits which are spongy 
 and non-coherent when rapidly deposited prevents the use of an 
 indefinitely large current. As the condition of the deposited 
 metal is affected by the solution from which it is deposited, espe- 
 cially by the amount of acid present, and the temperature, the 
 amount of current which may be used must be determined by 
 experiment, the temperature,- concentration, acidity, etc., of the 
 solution being fixed. 
 
 286. Secondary Electrolytic Reactions of Acid Radicles. The 
 secondary reactions produced by the current frequently influence 
 the electrolysis to a marked extent. Nitric acid is so rapidly 
 
SECONDARY ELECTROLYTIC REACTIONS. 201 
 
 reduced to ammonia by a current of moderate strength, that the 
 reaction may become strongly alkaline during the course of an 
 ordinary electrolytic determination if a too dilute acid is used. 
 Solutions of oxalic acid are entirely decomposed, hydrogen being 
 given off at one electrode, while carbon dioxide is liberated at the 
 other, the decomposition taking place according to the following 
 equation: 
 
 H 2 C 2 4 = H 2 (cathode) + 2C0 2 (anode). 
 
 If an oxalate of one of the alkali metals is present, the carbon 
 dioxide is not evolved, but remains in combination with the alkali, 
 the transformation from oxalate to carbonate being ultimately 
 complete. The reaction with ammonium oxalate is as follows: 
 
 2H 2 + (NH 4 ) 2 C A - 2NH 4 HC0 3 + H 2 . 
 
 The chlorine set free during the electrolysis of hydrochloric 
 acid or a chloride reacts with the water to form hypochlorous, 
 chloric, and perchloric acids, which combine with the bases present 
 to form salts. No secondary products are formed by the electroly- 
 sis of sulphuric acid and sulphates. 
 
 287. The Secondary Reactions Produced by the Metals are less 
 numerous and complicated. The alkali- and alkaline-earth metals 
 when deposited by the current react with the water present to 
 form hydroxides, which combine with any acids present. Silver 
 is almost always deposited from a solution of the double cyanide 
 KAgCy 2 . In this compound the silver undoubtedly forms a 
 part of the acid or negative radicle which is carried by the current 
 to the anode. The silver, however, is found deposited on the 
 cathode. This is explained on the assumption that the potassium 
 which is liberated at the cathode displaces the silver in the cyanide 
 radicle, one atom of potassium thus liberating one atom of silver. 
 The deposits of some of the heavy metals, such as copper, are 
 found contaminated with oxygen, while with others, as well as 
 copper, hydrogen is occluded under certain conditions. Some 
 of the metals, notably lead and manganese, are found deposited 
 on the anode. These metals form negative ions which are depos- 
 ited on the anode as most metals are on the cathode, or are oxidized 
 
202 ELECTROLYTIC METHODS. 
 
 by the nascent oxygen liberated at the anode. The oxides formed 
 in the case of manganese and lead, being insoluble in the nitric acid 
 present, are deposited on the anode. The peroxides of manganese 
 and lead formed in this manner may be dried and weighed as 
 such. This property of manganese and lead offers an excellent 
 method of separating these from many other metals. In this 
 way the determination of two metals may be carried on simul- 
 taneously, one being deposited on the anode and the other on the 
 cathode. 
 
 288. Reducing Current Strength by Means of Incandescent 
 Lamps. The electric current for quantitative work may be 
 obtained in various ways. The ordinary commercial direct- 
 current circuit of 110 volts is undoubtedly the most convenient, 
 and where available the work is accomplished by such a current 
 most satisfactorily and with the least labor. The current must, 
 of course, be available the entire twenty-four hours of the day, 
 as it is often most convenient to allow the electrolysis to continue 
 overnight. The voltage mentioned is much greater than is ever 
 required, and, as in reducing it part of the energy is wasted, a cur- 
 rent of smaller voltage is more economical. The reduction of the 
 current is very conveniently accomplished by means of incandes- 
 cent lamps. If the current is passed through a 16-candle-power 
 lamp about 0.4 ampere will be obtained. A 32-candle-power lamp 
 will give a current of 0.8 ampere, while a 50-candle-power lamp will 
 give a current of 1.2 amperes. The resistance of most solutions 
 prepared for electrolysis is so small that when placed in series 
 with the lamp the current is not materially 
 reduced. A very convenient arrangement 
 of the lamp-sockets is shown in Fig. 27. 
 The lamp-sockets L and U are connected 
 to binding-posts a, 6, and c, as shown in 
 the figure, the various parts being screwed 
 to a convenient sized board. By connecting 
 at a and b, or at a and c, the current is 
 .p passed through one lamp, while by con- 
 
 necting at b and c the current passes 
 
 through both lamps in series, and by connecting b and c with a 
 short piece of wire and passing the current from a to 6, it passes 
 
REGULATION OF CURRENT. 203 
 
 through the lamps in parallel. If 16-candle-power lamps are 
 used, the single lamps give 0.4 ampere, the lamps in series give 0.2 
 ampere, while the two lamps in parallel give 0.8 ampere. The 32- 
 candle-power lamps connected in the same manner give 0.8, 0.4, 
 and 1.6 amperes respectively, while the 50-candle-power lamps give 
 currents of 1.2, 0.6, and 2.4 amperes. Most electrolytic work 
 can be done by these strengths of current. 
 
 289. Reduction of the Voltage of a Current. If very small cur- 
 rents are desired they may most easily be obtained by making 
 the electrolytic solution a shunt. If a 16- 
 candle-power lamp is connected by means of 
 
 a Gorman-silver or other wire of several ohms 
 
 resistance to the main circuit of 110 volts, 
 
 any small number of volts desired may be 
 
 sent through the solution by attaching the p IG< 28. 
 
 wires to two points on the resistance wire. 
 
 If the resistance between the two points A and B is about 2J ohms 
 
 the difference of potential between these points will be about 1 
 
 volt. If about 5 ohms are enclosed between A and B, the difference 
 
 in potential will be about 2 volts. In this manner any desired small 
 
 voltage may be obtained. As OHM'S LAW holds for the solution, the 
 
 current produced will be inversely proportional to the resistance. 
 
 If, for instance, the resistance of the solution is 30 ohms and a 
 
 difference of potential of 2 volts is desired at the electrodes the cur- 
 
 1 f 1 
 
 rent obtained will be ampere according to Ohm's law, C=#, 
 lo . J\ 
 
 C being the current, E the voltage, and R the resistance. If it 
 is desired to increase the current while maintaining the difference 
 of potential at 2 volts, the resistance of the solution must be 
 decreased by warming it, or adding salts or acids or bringing 
 the electrodes closer together or increasing their size. 
 
 290. Primary and Storage . Cells. A very convenient method 
 of producing small currents is by the use of primary cells. Any 
 desired voltage may be obtained by this means if a sufficient num- 
 ber of cells are provided. The cells used for this purpose should be 
 those that give a current which is constant for a long period of 
 time. The ordinary DANIELL gravity cell is cheap and gives a 
 very steady current. The most satisfactory primary cell on the 
 
204 
 
 ELECTROLYTIC METHODS. 
 
 market is the EDISON PRIMARY CELL, formerly known as the 
 EDISON LALANDE CELL. The form known as type "S" is shown 
 in Fig. 29. It has an electromotive force of 0.667 volt, and a ca- 
 pacity of 300 ampere-hours. The positive 
 plate is made of copper oxide and is 
 placed between two zinc plates. The 
 liquid is a solution of caustic soda which 
 is protected from the carbon dioxide of 
 the air by a layer of heavy paraffine-oil, 
 which also prevents the solution from 
 " creeping." The jar is made of porce- 
 lain, and is entirely unacted on by the 
 alkaline liquid. No gases of any kind 
 are evolved. The chemical reactions 
 taking place may be represented as fol- 
 lows: Zn+2NaOH=Na 2 Zn0 2 +H 2 . The 
 hydrogen is oxidized by the copper oxide 
 as follows: CuO +H 2 =Cu+H 2 0. The 
 current produced is very constant, 
 although the potential rises slowly to 
 ab >ut one volt as the copper oxide be- 
 comes reduced to metallic copper. 
 When a dynamo current is available which is not generated 
 continuously STORAGE BATTERIES must be provided. If prop- 
 erly cared for, these cells give a very steady current, the small 
 drop from about 2 volts when charged to 1.8 volts when discharged 
 not affecting most electrolytic work, especially since adjustment 
 may be made by resistance coils. Portable storage-cells may be 
 purchased for use in laboratories where a dynamo current is not 
 available. These cells may be carried to a power-station for charging. 
 When either primary or storage cells are used the strength of 
 the current may be modified by varying the number of cells used, 
 the voltage in every case being equal to the sum of the voltage 
 given by the individual cells which are connected in series. The 
 current may be still further regulated by inserting or removing 
 resistance from the circuit. This is most easily done by means 
 of a resistance-box connected in series with the solution to be 
 electrolyzed. 
 
 FIG. 29. 
 
ELECTROLYTIC APPARATUS. 205 
 
 A simple form of AMMETER and VOLT-METER are indispensable 
 for much electrolytic work, especially in separations where a 
 slight change of the potential may spoil a determination. 
 
 291. Use of Platinum Dishes as Cathodes. Various forms of 
 electrodes are in use upon which to deposit the metals determined. 
 The most commonly used is an ordinary platinum evaporating- 
 dish, such as is kept in most laboratories for other uses. Dishes 
 of a capacity of about 150 c.c. and weighing 35 to 40 grams are 
 most convenient. They should be free from scratches or dents 
 produced by wear. Old dishes should be sent to the manufacturer 
 to be reformed. The deposits of manganese and lead peroxides 
 adhere more firmly to a platinum surface which has been rough- 
 ened with a sand-blast. Such a rough surface may also be pro- 
 duced by depositing platinum electrolytically from a solution of 
 potassium-platino-chloride by a rather large current. The deposit 
 does not adhere very firmly unless, after washing, the dish is heated 
 to redness for some time with the blast-lamp. Such a roughened 
 dish is equally available for use with other metallic deposits than 
 lead and manganese. As the dish cannot be employed for general 
 laboratory purposes, it is advisable to roughen the surface of a 
 platinum cone which is available only for electrolytic work. 
 
 The ANODES used with the platinum dishes are either disks 
 of platinum to which a rod of the same metal has been securely 
 fastened, or a stiff platinum wire the end of which has been made 
 into a flat coil at right angles to the wire. 
 
 292. Electrolytic Stands. The dish and anode are held by an 
 electrolytic stand which is shown in Fig. 30, and consists of an 
 iron base into which a stout glass rod has been fastened. A brass 
 ring to which several platinum points have been soldered is 
 attached to a clamp so that it may be fastened to the glass rod at 
 any convenient height. A small binding-post is also fastened to 
 the clamp for connection with the source of the electric current; 
 A brass rod is fitted with a similar clamp for attachment to the 
 glass rod. A binding-post for the second wire from the battery, 
 or other source of electricity, is attached to the clamp. Another 
 binding-post which can be moved along the brass rod holds the 
 platinum anode. 
 
 For this rather expensive electrolytic stand various devices 
 
206 
 
 ELECTROLYTIC METHODS. 
 
 made from ordinary laboratory apparatus may be substituted. 
 An ordinary iron ring-stand is easily adapted to the purpose. A 
 clean copper wire is wound around an iron 'ring of suitable size. 
 The platinum dish is placed on this ring, the electric current being 
 led to the dish by means of the copper wire. The stiff wire forming 
 
 FIG. 30. 
 
 part of the anode is forced through a cork which is held in a clamp 
 which in turn is fastened to the iron rod of the ring stand. 
 
 As considerable loss may result from the spattering of the solu- 
 tion due to the gases liberated by the electrolysis, it is advisable 
 to cover the platinum dish with a watch-crystal through the 
 centre of which a hole has been drilled. The watch-crystal is then 
 split so that it may be placed on the dish with the anode passing 
 between the halves where the hole was drilled. This watch- 
 crystal must be rinsed off into the dish a short time before inter- 
 rupting the electrolysis. A sheet of mica of suitable size is also 
 convenient for this purpose. 
 
 293. Platinum Cylinders as Electrodes. Platinum cylinders 
 
ELECTROLYTIC APPARATUS. 
 
 207 
 
 made of thin foil and fastened to a stiff platinum wire are very 
 largely used in laboratories where many electrolytic determinations 
 must be conducted. The cylinders are made of two sizes, so that 
 the one of smaller diameter may be placed within the other. The 
 cylinders are of the same height. By carefully centering the cylin- 
 ders, the space between them becomes of uniform width at all 
 points, thus distributing the current so that the metal is deposited 
 very uniformly. The larger cylinder is always connected so as to 
 receive the metallic deposit. Instead of the inner cylinder, a 
 
 FIG. 31. 
 
 platinum wire either straight, or twisted into a coil, may be used. 
 In place of the outer cylinder a truncated cone with the smaller 
 end up has been used. Eecently, cylinders made of platinum 
 gauze have been found to offer many advantages over the forms 
 already described. Not only is the time necessary for depositing 
 the metal greatly shortened, but a much more firmly adhering 
 deposit is obtained. The rapid deposition is undoubtedly due to 
 the free circulation of the electrolyte permitted by the wire gauze. 
 The firm adherence of the deposit is probably due to its cylindrical 
 form, which does not give any point for the peeling process to start. 
 Rotation of one of the electrodes, especially that on which the 
 
208 ELECTROLYTIC METHODS. 
 
 metal is being deposited, has been found to improve greatly the 
 quality of the deposit and shorten the time for complete deposi- 
 tion. 
 
 Supports for the cylindrical electrodes are made with an iron 
 foot and glass standard, exactly like those used for the platinum 
 dishes. Two brass rods similar to the one used for holding the 
 anode for the dish are provided for holding the two cylinders. 
 The use of cylinders has the advantage that the solution to be 
 electrolyzed may be prepared in beakers, and in many cases 
 insoluble matter need not be filtered off, nor the volume reduced 
 by evaporation to that of the platinum dish. 
 
 294. Electrolysis of Warm Solutions. The time required for 
 many electrolytic determinations is greatly reduced by warming 
 the solution. Temperatures from 60 to 80 are generally found 
 suitable. This temperature need not generally be taken with a 
 thermometer, since most chemists are able to estimate the tempera- 
 ture closely enough by touching the beaker with the hand. The 
 small flame needed for this purpose may most easily be obtained 
 by screwing off the tube of the Bunsen burner, and using the small 
 white flame which is then produced by lighting the burner, and 
 turning the gas nearly off. This white flame must not be allowed 
 to come in contact with the bottom of the platinum dish. Starting 
 the electrolysis with the solution warm will frequently insure a 
 better deposit than can be obtained from a cold ; solution. The 
 solution should be tested for the metal before interrupting the 
 electrolysis unless the time required for similar determinations 
 has been carefully noted. It is usually not advisable to allow the 
 current to pass for a longer time than necessary to completely 
 deposit the metal. 
 
 295. Washing and Drying Deposited Metals. In many cases 
 the deposited metal redissolves quite rapidly in the solution from 
 which it has been deposited. In most cases the loss is very slight 
 if the cylinders are removed from the solution while the current is 
 passing and the acid is removed by instantly dipping them into 
 distilled water or washing them in a stream of water. More accu- 
 rate results are obtained, especially by the beginner, if the acid 
 solution is replaced by water while the current is still passing. 
 For this purpose, the acid solution is removed by a siphon, while a 
 
WASHING AND DRYING DEPOSITED METALS. 209 
 
 stream of distilled water is poured in from the wash-bottle. If an 
 incandescent lamp is used as resistance in circuit with the electrodes, 
 the removal of the acid solution is indicated by the disappearance 
 of light in the lamp. When the acid solution has been removed in 
 this manner, the deposited metal is washed with d. stilled water 
 and then with alcohol to remove the water. Ether is sometimes 
 used for the same purpose, but it is much more liable to contain 
 fatty substances in solution, which remain after evaporation of 
 the ether. If ether is used, the metal will dry much more rapidly. 
 The removal of the water is necessary, as moist metals oxidize 
 rapidly and increase in weight. After washing with alcohol or 
 ether, the metal is dried for a few minutes in the steam-bath, or 
 in the air-bath kept at 100, and after cooling, is weighed. The 
 removal of the alcohol and the drying may be accomplished much 
 more rapidly by burning the portion adhering to the surface. After 
 cooling, the metal is weighed. A trifling amount of oxidation takes 
 place when this method is used. 
 
 296. Rotation of the Anode or the Cathode offers several very 
 decided advantages. The metal is generally deposited in a much 
 smoother and more coherent film, so that a number of metals 
 which cannot be deposited in coherent films with stationary 
 electrodes can be successfully determined by rotating one of 
 the electrodes. This character of the film. also allows the use of 
 a much larger current, so that the deposition is much more 
 rapid. The agitation of the solution also increases the rate of 
 deposition because the 'layer of liquid near the cathode ife very 
 soon deprived of its metal by the electric current when the solu- 
 tion is allowed to remain at rest, while when it is kept in constant 
 motion, more metal is constantly being brought up to the electrode. 
 For the same reason, heating the solution increases the rate of 
 deposition because convection currents are set up in the liquid. 
 Rotation of the cathode or anode produces a much more vigor- 
 ous circulation and consequently hastens the deposition, so that 
 15 minutes is in many cases a sufficient time for the complete 
 deposition of the metal. 
 
 297. Electrodes for Rotation. Cylinders are not suitable 
 electrodes for rotation. Platinum dishes, however, are admir- 
 ably adapted for this purpose. The platinum spiral used as anode 
 
210 
 
 ELECTROLYTIC METHODS. 
 
 can be very conveniently rotated, care being taken to accurately 
 center the spiral and not to rotate it so fast that the solution 
 is thrown out of the dish. Ordinary platinum crucibles may 
 also be used for this purpose. A rubber or cork stopper is 
 selected which fits tightly into the crucible. The shaft of the 
 rotator is passed through a hole in the middle of the stopper, 
 electrical contact being made between the shaft and the crucible. 
 For this purpose a copper wire is wound around the shaft and 
 passes through the hole in the stopper, projecting far enough to 
 
 FIG. 32 
 
 touch the bottom of the crucible. Contact is made between 
 the shaft and a binding post which is connected with the nega- 
 tive terminal. The positive terminal is connected with the anode, 
 which may consist of platinum foil or wire. This arrangement 
 is shown in Fig. 32. The metal is deposited on the outside of the 
 crucible to within about a quarter of an inch of the top. When 
 being weighed the crucible is inverted on the scale pan. 
 
 298. The Motors for rotation may be electric or water motors. 
 A velocity of 300 to 600 revolutior.s per minute will be required. 
 
MOTORS FOR ROTATION. 
 
 211 
 
 FIG. 33. 
 
212 
 
 ELECTROLYTIC METHODS. 
 
 A system of pulleys of different sizes is convenient for regulating 
 the speed of rotation. In addition to this means of regulation, 
 the current supplied to electric motors may be varied by chang- 
 ing the resistance in the circuit, 
 and with water motors the amount 
 of water supplied is easily con- 
 trolled. An apparatus designed 
 for utilizing a water motor is 
 shown in Fig. 32. The upright 
 shaft of the motor rotates on a 
 needle point, while the shaft hold- 
 ing the crucible rotates on the 
 ball-bearings of a bicycle hub. 
 An apparatus designed for using 
 an electric motor is shown in 
 Fig. 33. R is the resistance in 
 circuit with the solution being 
 electrolyzed and controls the 
 amount of current used, as indi- 
 cated by the ammeter A, which is 
 in the same circuit. The fuses F 
 protect the ammeter from being 
 burned out. The motor M is 
 attached by means of the flexible 
 chord C at the socket T. The 
 motor may be instantly stopped 
 by pulling the chord out of the 
 socket T. The motor requires no 
 outside resistance. The switch S 
 controls both the motor circuit 
 and the electrolytic circuit, so 
 that the electrolysis and the rota- 
 tion may be started or interrupted 
 simultaneously. The wiring of this board is shown in Fig. 34, 
 which is a diagrammatic representation of the reverse of the slate 
 board upon which the apparatus is mounted. A volt meter has 
 not been permanently connected with the other apparatus because 
 in ordinary work it is not needed. The resistance R is in series 
 with the electrolytic circuit, but not with the motor circuit. 
 
 FIG. 34. 
 
CHAPTER XVIII. 
 ELECTROLYTIC DETERMINATION OF METALS. 
 
 DETERMINATION OF COPPER. 
 
 299. Deposition from a Nitric or Sulphuric Acid Solution. The 
 electrolytic determination of copper is, no doubt, the most accurate 
 as well as the most rapid method of determining this metal. If the 
 solution of copper is free from other metals, a great many of the 
 methods which have been proposed may be used. If, as most gener- 
 ally occurs in practice, the copper must be determined in the pres- 
 ence of other metals but one or two methods are available. By 
 these methods the copper is deposited from a solution containing 
 FREE NITRIC or SULPHURIC ACID. As the copper is apt to be 
 deposited in a spongy form when only sulphuric acid is present, 
 solutions containing only nitric acid, or nitric and sulphuric acids 
 are generally used. Not more than 5% of concentrated nitric acid 
 may be present, and if the same amount of concentrated sulphuric 
 is present, J% of concentrated nitric acid is sufficient. The sul- 
 phuric acid solution is used when the presence of the nitric acid or 
 the ammonium salts formed by its decomposition would interfere 
 with subsequent operations, such as the precipitation of zinc as 
 sulphide. 
 
 The nitric acid may be entirely omitted if 1 gram of urea is 
 added to the solution containing 7% to 10% of concentrated 
 sulphuric acid. By using a current of about one ampere, the 
 decomposition is then effected in less than two hours.* 
 
 300. Separation and Determination of Lead. The sulphuric 
 acid solution is used for copper when much lead is present. This 
 metal is then separated as sulphate, in which form it may be fil- 
 tered off and weighed. If only a small amount of lead is present, 
 it may be deposited on the anode from a nitric acid solution, and 
 
 * Classen, Quan. Chein. Analysis by Electrolysis, page 160. 
 
 213 
 
214 ELECTROLYTIC METHODS. 
 
 the peroxide washed, dried at 180, and weighed. If the lead is to 
 be separated and determined in this manner, sulphuric acid should 
 not be introduced, as the lead peroxide will be contaminated with 
 this acid. 
 
 301. Separation of Copper from Other Metals. ZINC, COBALT, 
 NICKEL, CADMIUM, MANGANESE, and moderate amounts of IRON 
 may be present in the nitric acid solution without interfering with 
 the accuracy of the determination of copper. Large amounts of 
 iron are objectionable because the ferric salts tend to redissolve 
 the precipitated copper. The difficulty is partially overcome by 
 diluting the solution and increasing the current density. The 
 copper may also be completely separated from iron by precipita- 
 tion with hydrogen sulphide or sodium thiosulphate. The precipi- 
 tate is filtered off, washed, and redissolved in nitric acid. From 
 this solution it may be precipitated electrolytically and weighed. 
 
 SILVER, MERCURY, BISMUTH, ARSENIC, ANTIMONY, and TIN are 
 deposited with the copper. Silver should be precipitated as chlo- 
 ride before electrolyzing the solution, an excess of hydrochloric 
 acid being avoided. Mercury may be expelled from the ore by roast- 
 ing. When alloys or ores of copper are dissolved in nitric acid 
 the tin and almost all of the antimony remains undissolved while 
 most of the arsenic goes into solution. The removal of these 
 metals may be accomplished by passing hydrogen sulphide through 
 the solution until no more precipitate is formed, and then dissolving 
 the sulphides of arsenic, antimony, and tin in sodium sulphide 
 solution. The copper sulphide may then be dissolved in nitric 
 acid and the copper deposited and weighed. 
 
 Copper may also be readily and completely separated from 
 BISMUTH, ARSENIC, ANTIMONY, and IRON by precipitation as 
 cuprous sulphocyanate. The solution of the copper salt is made 
 slightly acid with hydrochloric acid, and the copper reduced by 
 saturating the solution with sulphur dioxide. A solution of 
 ammonium thiocyanate is then added until precipitation is com- 
 plete, and the mixture heated nearly to the boiling-point for a 
 few minutes. The copper may also be precipitated by the 
 addition of a solution of equal parts of acid ammonium sulphite 
 and ammonium thiocyanate. The precipitate, which is colored 
 when first formed, quickly becomes white, and settles very readily. 
 
DETERMINATION OF COPPER. 215 
 
 It is washed with warm water until free from the other metals 
 present. The copper is dissolved in a little warm concentrated 
 nitric acid, and then deposited as usual. 
 
 COPPER ORES containing ARSENIC, ANTIMONY, and TIN may 
 be decomposed by fusion with six times the weight of a mixture 
 of equal parts of sulphur and sodium carbonate or the same amount 
 of dry sodium thiosulphate. The ore must be very finely pow- 
 dered and the fusion conducted in a porcelain crucible. After 
 heating with a small Bunsen-burner flame until sulphur is no 
 longer evolved, the melt is allowed to cool, and then extracted 
 with hot water, and the residue washed with water containing a 
 little ammonium sulphide. By this treatment the arsenic, anti- 
 mony, and tin are dissolved while the copper remains as sulphide, 
 which is dissolved in nitric acid, and the copper in the solution 
 determined. 
 
 302. Deposition of Copper from an Ammonium-oxalate Solu- 
 tion. A very excellent method of determining copper is based 
 on its separation from a solution containing excess of ammonium 
 oxalate. Classen* has improved this method so that the determina- 
 tion may be carried out in two hours. As most of the metals are 
 precipitated from an oxalate solution, this method cannot be used 
 unless the copper is first separated from the other metals, or the 
 amount present in a pure salt is to be determined. 
 
 EXERCISE 40. 
 Determination of Copper in Copper Sulphate^ CuSO 4 -5H O. 
 
 Clean, dry at 100, and weigh a platinum dish or platinum cone or cylin- 
 der. Weigh out 1 gram of pure recrystallized copper sulphate. If the 
 cone or cylinder is used, transfer the copper sulphate to a 250-c.c. beaker, 
 otherwise place it in the weighed dish. Dissolve in a few cubic centimeters 
 of water, add a solution of 4 grams of ammonium oxalate, and dilute the 
 solution to about 120 c.c. Make a saturated solution of oxalic acid by 
 dissolving 4 grams of the crystals in 50 c.c. of water, and filter if necessary. 
 Place the dish on the ring of the electrolytic stand, and adjust the anode 
 so that it is about inch below the surface of the liquid. If a beaker is 
 used, it must be placed on a tripod or other support, so that the solution 
 can be heated. The cylinder or cone should be adjusted so that it very 
 nearly touches the bottom of the beaker. Fasten a burette by means of a 
 
 * Classen, Quant. Anal, by Electrolysis, page 154. 
 
216 
 
 ELECTROLYTIC METHODS. 
 
 clamp and ring-stand, so that the solution of oxalic acid which is placed 
 in the burette will drop on the split watch-crystal which covers the dish 
 or beaker. A thermometer is also suspended with the bulb in the copper 
 solution, which is heated with a small flame from a Bunsen burner to 80. 
 The current of electricity used should be so adjusted that about 1 ampere 
 passes through the solution, and the tension at the electrodes is from 2.5 
 to 3.2 volts. 
 
 As soon as a film of copper is seen to cover the cathode, the stop-cock 
 of the burette is so adjusted that the oxalic-acid solution falls at the rate 
 
 FIG. 35. 
 
 of 10 drops per minute. The copper comes down as a bright-red deposit. It 
 it becomes dark-colored or spongy, the oxalic acid should be allowed to 
 drop faster. The electrolysis will be completed within two hours. When 
 the solution has become colorless, portions may be taken out and tested 
 for copper by acidifying with hydrochloric acid and adding potassium 
 ferrocyanide. When all of the copper has been deposited, siphon off the 
 acid liquid while water from the inverted wash-bottle is allowed to flow 
 over the copper deposit. The copper is then washed with alcohol, dried 
 at 100, and weighed. The theoretical percentage of copper in copper 
 sulphate is 25.41. 
 
DETERMINATION OF SILVER. 217 
 
 EXERCISE 41. 
 Determination of Copper from a Nitric-acid Solution. 
 
 The copper in copper sulphate may also be precipitated from a nitric- ' 
 acid solution. This method is well adapted to the determination of 
 copper in the refined metal. About 0.3 gram of the metal is carefully weighed 
 out, and dissolved in 3 c.c. concentrated nitric acid and a little water. The 
 solution is diluted to about 100 c.c., and warmed to 50 or 60, and the 
 copper precipitated with a current of 0.5 to 1 ampere. No arrangement 
 need be made to keep the solution warm, as in the preceding method. The 
 deposition of the metal is more rapid, and a better deposit is obtained if 
 the solution is warm at first. After four to five hours, all of the copper 
 will be deposited. If a cylinder made of wire gauze is used, about half this 
 time will be required. If the rotating cathode or anode is used, the amount 
 of nitric acid may be reduced to 1 c.c., while the current may be increased 
 to from 3 to 5 amperes. Most of the copper will be deposited in 10 min- 
 utes. The small amount still remaining in solution will be very rapidly 
 deposited if the solution is nearly neutralized by the addition of ammonia. 
 When all of the copper is deposited, as shown by the ferrocyanide test, 
 the solution is displaced with distilled water while the current is passing, 
 and the deposit washed with alcohol, dried, and weighed exactly as directed 
 in the first method. 
 
 303. Deposition of Silver from a Potassium-cyanide Solution. 
 Silver is determined electrolytically almost exclusively from a 
 cyanide solution. The potassium cyanide used must be pure. 
 Failure to obtain good results is frequently due to the use of 
 impure cyanide. Another sample of cyanide should therefore be 
 obtained and used. The current density should be less than 0.5 
 ampere. The electrolysis may be conducted at the ordinary 
 temperature, though when heated to about 65, the time re- 
 quired for a determination is very much reduced, so that two 
 or three hours is generally sufficient. 3 grams of potassium cyanide 
 are sufficient for 0.5 gram of silver. If other metals are present, 
 the silver may be precipitated as chloride, which is then dissolved 
 in the potassium cyanide. It is doubtful, however, if in this case 
 the result would be more accurate or obtained more quickly than 
 if the silver chloride were directly weighed. 
 
 304. Separation of Silver from Lead and Copper. Silver may 
 be separated from lead in consequence of the deposition of the lat- 
 
218 ELECTROLYTIC METHODS. 
 
 ter on the anode as peroxide, when a solution containing 20% of 
 concentrated nitric acid is electrolyzed. The separation of silver 
 from copper is effected by electrolyzirg a solution of the two metals 
 in potassium cyanide by a current giving a tension between the 
 electrodes of about 1.4 volts. When all of the silver has been 
 deposited, the copper may be determined by adding nitric acid to 
 the solution, and increasing the tension between the electrodes. 
 
 EXERCISE 42. 
 Determination of Silver in Silver Nitrate. 
 
 Weigh out 0.5 gram of pure dry silver nitrate, and dissolve in a few cubic 
 centimeters of water. Dissolve about 2 grams of potassium cyanide in a 
 small amount of water, and pour the silver solution into the cyanide solution. 
 Either the platinum dish or the cylinders may be used as described under 
 the determination of copper. Three Edison-Lalande cells connected in 
 series or other source of electricity giving about 3 volts, should be used 
 The electrolysis requires about six hours, but the time necessary may be 
 reduced by warming the solution to 65, which is about the temperature 
 the hand can bear for a moment without much discomfort. If the anode 
 is rotated, a current of 2 to 3 amperes may be used, and the deposition will 
 be complete in 10 to 15 minutes. The end of the operation is ascertained 
 by testing a few drops of the solution by adding hydrochloric acid. The 
 test should be made under the hood. Siphon off the solution while add- 
 ing distilled water, rinse with alcohol, dry at 100, and weigh. If the 
 cyanide solution becomes quite dark on passing the current, the potassium 
 cyanide is impure and the result is not apt to be correct. The determina- 
 tion should be repeated with another sample of the cyanide. The theo- 
 retical percentage of silver in silver nitrate is 63.51. 
 
 EXERCISE 43. 
 Analysis of a Silver Coin, 
 
 305. Solution of Coin. When results within 0.1 or 0.2% of the theo- 
 retical have been obtained with silver nitrate, a coin is analyzed. Thor- 
 oughly clean a dime or other silver coin by rubbing with alcohol, ammonia, 
 etc. Cut it into four or five pieces so that each piece weighs about 0.4 
 gram. Weigh one of the pieces carefully, and dissolve in a few cubic 
 centimeters of a mixture of equal parts of water and concentrated nitric 
 acid. This should be done in a small covered beaker. Warm the beaker 
 on the water-bath, and when solution is complete remove the watch- 
 crystal, rinsing it if necessary, and let the solution go to dryness. 
 
DETERMINATION OF SILVER, 219 
 
 306. Silver. In the mean time weigh out roughly and dissolve in 
 water two grams of potassium cyanide. Dissolve the copper and silver 
 nitrates in a little water, and pour the solution into the cyanide solution. 
 If difficulty is experienced in dissolving the salts of copper and silver which 
 have dried on the sides of the beaker, a little potassium cyanide solution 
 may be added after the water solution has been removed. The silver 
 may be deposited on a platinum dish or on a cylinder. The solution should 
 be warmed to about 65, and electrolyzed with the current from two Edison- 
 Lalande or similar cells connected in series. The difference of potential 
 between the two platinum electrodes must be measured with a volt-meter 
 and must not exceed 1.5 volts. If the two cells give a higher voltage than 
 this, the potential must be reduced by introducing resistance. A resistance- 
 box may be used, or the connection between the cells and one of the electrodes 
 may be made by means of a few feet of German-silver wire. By drawing 
 more or less of this wire through one of the binding-posts the resistance 
 may be changed until the potential between the two electrodes is a little 
 less than 1.5 volts. 
 
 The dish or beaker is covered with a split watch-crystal. As the evapo- 
 ration at 65 is considerable, water must be added from time to time. In 
 about five hours the electrolysis will be completed, as may be ascertained 
 by adding water until an unplated portion of the platinum dish is covered. 
 If no silver is deposited on this portion of the dish in half an hour, the 
 deposition of the silver is complete. The solution is now siphoned off 
 into a beaker and replaced with water, care being taken to lose none. The 
 silver is rinsed with alcohol, dried at 100, and weighed. 
 
 307. Copper. The copper in the filtrate is deposited on a cone or cylin- 
 der after the addition of 5 c.c. concentrated nitric acid. The operation is 
 repeated until analytical duplicates are obtained. The rotating anode may 
 be used in this determination as well as in the determination of the silver. 
 
 DETERMINATION OF NICKEL AND COBALT. 
 
 Both nickel and cobalt are most easily determined electro- 
 lytically. The same methods serve for both metals. Where the 
 two metals occur together, they are separated from the other 
 metals present by the methods given in Chapter XIV. Both 
 metals are then precipitated electrolytically, and, after weighing, 
 the deposit is dissolved in nitric acid, one of the metals being 
 precipitated and weighed by methods given in the chapter 
 mentioned. 
 
 308. Ammonium-oxalate Method. According to Classen * 
 these metals may be readily deposited from a solution of the 
 
 * Classen, Quan. Anal, hy Electrolysis, page 141. 
 
220 ELECTROLYTIC METHODS. 
 
 double-ammonium oxalate. Nitrates and chlorides should be 
 removed by evaporation with sulphuric acid. The excess of 
 acid is neutralized with ammonia, and 4 to 5 grams of ammonium 
 oxalate added, which is dissolved by warming the solution. 
 Water is then added until the volume is from ICO to 120 c.c. If 
 the temperature of the solution is kept at 60 to 70, the elec- 
 trolysis will be complete in three to four hours, a current of 
 about 1 ampere and an electrode tension of 3 to 4 volts being used. 
 309. Ammonia Method. By the method of Fresenius and 
 Bergmann * chlorides and nitrates are removed by evaporation 
 with sulphuric acid. The excess of acid is then neutralized with 
 ammonia, and unless at least 6 c.c. of strong ammonia (sp. gr. 0.90) 
 are used for this purpose, 6 grams of ammonium sulphate should 
 be added. 15 c.c. of strong ammonia are then added, or if more 
 than 0.5 gram of metal is present, 20 c.c. should be used. The 
 solution is then diluted to 150-170 c.c., and electrolyzed with a 
 current of 0.7 ampere. The separation is not hastened by warm- 
 ing the solution. The deposition is complete in about six hours. 
 
 EXERCISE 44. 
 Analysis of a Nickel Coin. 
 
 310. Solution of Coin. Clean a nickel coin thoroughly by rubbing with 
 sand or " Sapolio." Cut it into small pieces weighing about 0.4 gram. 
 After weighing one of these pieces place it in a small beaker (about 150 c.c.), 
 and add 10 c.c. dilute nitric acid. Warm the covered beaker on the water- 
 bath until the metal is dissolved. Rinse off the watch-crystal, add 16 c.c. 
 dilute sulphuric acid, and evaporate on the water-bath until no more nitrous 
 fumes come off. 
 
 311. Copper. If a cylinder is available, dilute the solution untii the 
 cylinder is nearly covered when immersed so as to very nearly touch the 
 bottom of the beaker. Electrolyze with a current of about 0.8 ampere. 
 If the solution is warmed to about 60, the copper will be deposited in from 
 four to six hours. If it is convenient to conduct the electrolysis overnight, 
 a current of about 0.4 ampere should be used, and the electrolysis begun 
 with % warm solution, which is then allowed to cool off. The end of the 
 operation is ascertained by testing a small portion of the solution for copper 
 by passing hydrogen sulphide. After boiling for a few minutes to expel 
 
 *Zeit. Anal. Chem., vol. 19, page 329. 
 
DETERMINATION OF TIN. 221 
 
 the hydrogen sulphide, this portion may be returned to the beaker. The 
 solution may also be tested for copper by adding water so that a clean por- 
 tion of the cylinder is immersed. If no copper is deposited on this portion 
 after half an hour, all of the metal has been deposited. When all of the 
 copper has been deposited, the solution containing the nickel is siphoned 
 off and replaced with water. The copper is washed with water and alcohol, 
 dried at 100, and weighed. The anode or cathode may be rotated, and 
 the current largely increased, and the time required for the deposition 
 reduced to less than half an hour. 
 
 312. Nickel. The solution containing the nickel should be evaporated 
 if the volume is more than 250 c.c. It is neutralized with ammonia, 15 c.c. 
 of concentrated ammonia added, and the nickel deposited with a current of 
 about 0.7 ampere. About five to six hours will be required. The solution 
 is tested for nickel by adding to a small portion a little colorless ammo- 
 nium sulphide, or passing hydrogen sulphide. If the nickel is deposited 
 slowly, enough ammonia is not present and more must be added. The solu- 
 tion need not be siphoned off. The nickel is washed with water and alcohol, 
 dried at 100, and weighed. The determination is repeated with other por- 
 tions of the coin until duplicates are obtained. The rotating anode or 
 cathode may be used for the deposition of the nickel. The current may 
 then be increased to 5 amperes, when the deposition will be complete in 
 less than half an hour. 
 
 DETERMINATION OF TIN. 
 
 Tin may be deposited in a bright, crystalline, closely adherent 
 form from solutions of the double-ammonium oxalate containing 
 free oxalic acid. If the tin is obtained as the dioxide, it may be 
 dissolved by fusing with 8 times its weight of caustic soda in a 
 silver dish. The melt is dissolved in water, an excess of oxalic 
 acid added and the solution heated until clear. Hydrogen sulphide 
 is then passed to precipitate the silver which is filtered off, after which 
 the solution is boiled to expel the hydrogen sulphide. Ammonium 
 oxalate is then added, the solution warmed to about 65, and 
 electrolyzed with a current of 4 to 5 volts and from 0.5 to 1.5 
 amperes. For 0.3 gram of tin 9 to 10 grams oxalic acid and 
 4 grams of ammonium oxalate should be used. The tin is com- 
 pletely deposited in about 4J hours. It is washed as usual with 
 water and alcohol, and dried in the steam-bath. Excellent results 
 are obtained with the rotating ai.ode. 
 
 313- Dissolving Tin Deposits on Platinum. Some difficulty 
 is experienced in dissolving the l!n which has been deposited on 
 
222 ELECTROLYTIC METHODS. 
 
 the platinum. This metal tends to form an alloy with the platinum 
 so that acids do not readily attack the tin. The difficulty may be 
 overcome by first coating the dish or other electrode with silver or 
 copper or a silver dish may be used for these determinations. 
 The silver is deposited from a solution of silver nitrate in potassium 
 cyanide, while the copper may be deposited from a solution of 
 copper sulphate acidified with nitric acid. After a sufficiently thick 
 coating has been obtained by either method the solution is de- 
 canted and the deposited metal washed and dried as usual for the 
 determination of these metals. The best solvent for the tin is 
 prepared by diluting concentrated nitric acid with an equal volume 
 of water and then saturating with oxalic acid. After most of the 
 tin has been dissolved by this solution the nitric acid is washed off 
 with water and the cylinder or dish immersed in concentrated C. 
 P. hydrochloric and allowed to remain for 5 to 6 hours. 
 
 It is advisable to keep cylinders used for the determination of 
 tin immersed in concentrated C. P. hydrochloric acid when not in 
 use. This acid will dissolve all the tin if allowed to act for a 
 sufficient time. 
 
 DETERMINATION OF LEAD. 
 
 This element cannot be determined by deposition in the metal- 
 lic form, because during the process of washing and drying it 
 invariably oxidizes to a greater or less extent. Advantage is 
 therefore taken of the fact that when a current of electricity is 
 passed through a nitric acid solution of lead, it is converted into 
 the peroxide which is deposited on the anode, and may be 
 dried by heating to 180-200. The platinum anode should be 
 roughened according to the suggestion of A. Classen by means of 
 a sand-blast, otherwise the deposit readily flakes off unless depos- 
 ited very slowly and in small amount. By this method the lead is 
 separated from zinc, cobalt, nickel, iron, aluminium, copper, gold, 
 mercury, antimony, and cadmium. Traces of silver and bismuth, if 
 present, are deposited as peroxide with the lead. 
 
 314. Conditions of Deposition of Lead. The amount of nitric 
 acid which must be present depends on the temperature of the solu- 
 tion and the amount of current used. If the electrolysis is carried 
 out at the ordinary temperature, about 10% by volume of nitric 
 
DETERMINATION OF LEAD. 223 
 
 acid of sp. gr. 1.38 must be present, and a current of .05 amperes 
 used, while if a current of 0.5 amperes is employed 20% by vol- 
 ume of the nitric acid must be present. The depositon of the lead 
 is much more rapid at a temperature of 60 to 65, but the amount 
 of acid necessary at this temperature for a current of .05 amperes 
 is 2% by volume; while for a current of 0.5 amperes 7%, and for 
 1.5 to 1.7 amperes 20% is necessary. The small current must be 
 used when only smooth dishes or electrodes are available. If too 
 little nitric acid is present, metallic lead will be deposited on the 
 cathode. Hydrochloric acid should be absent from the solution, 
 and only small amounts of sulphuric acid should be present, as it 
 is partly precipitated with the lead peroxide. The rotating anode 
 or cathode may be used to advantage. 
 
 The end of the operation is ascertained by adding water, so 
 as to expose a fresh portion of the electrode. If lead is still 
 present, a dark-brown deposit will be formed after a quarter to 
 half an hour. The solution need not be siphoned off. The lead 
 peroxide is washed with hot water and then with alcohol, and is 
 dried for half an hour in an air-bath heated to 200. The lead 
 dioxide may be dissolved in hot dilute nitric acid to which a little 
 oxalic acid has been added, or in dilute hydrochloric acid which 
 has been saturated with sulphur dioxide. It may also be dis- 
 solved in dilute nitric acid after being converted into litharge by 
 heating for a few minutes in the oxidizing flame of the blast-lamp 
 or of a Bunsen burner. 
 
 315. Determination of Lead in Ores. The electrolytic method 
 is well suited to the determination of lead in ores. If possible 
 the ore should be decomposed without the use of nitric acid, 
 which converts any sulphur present into sulphuric acid. As has 
 already been said, the presence of this acid in the solution intro- 
 duces a slight error in the determination. A method of treating 
 the ore, which is suitable for galena, is given by Medicus.* The 
 finely powdered ore is dissolved in concentrated hydrochloric 
 acid. Excess of caustic-potash solution (1:3), and if antimony is 
 present, 1 to 2 grams of tartaric acid are added. The solution is 
 warmed for a few minutes at 100, 3,nd, after cooling and diluting 
 with water, the lead is precipitated by passing carbon dioxide. 
 
 *Ber. d. Chem. Ges., 1892, p. 2490. 
 
224 ELECTROLYTIC METHODS. 
 
 This will require about one and one-half hours. The lead car- 
 bonate is washed with hot water until free from chlorine, and is 
 then dissolved in dilute nitric acid and the paper well washed. 
 The lead in the solution is then deposited as the peroxide, as 
 already directed. 
 
 DETERMINATION OF ZINC. 
 
 A large number of electrolytic methods have been proposed 
 for zinc, very few of which have been found practicable. The 
 difficulties attending the gravimetric determination of this metal 
 have led to continued effort in this direction, until methods which 
 appear to be successful have been devised which depend on the 
 deposition of the zinc from a solution kept slightly acid with a 
 weak organic acid. 
 
 316. Difficulty of Dissolving Zinc. In the electrolytic deter- 
 minations of zinc, the same inconvenience is met with as in the 
 determination of tin, that the metal is dissolved with difficulty 
 from the platinum electrode. The same remedy is adopted: 
 namely, to coat the platinum with silver or copper, or to use silver 
 dishes. The zinc may be dissolved, leaving the copper or silver, 
 by warming with dilute sulphuric acid. 
 
 317. Ammonium-oxalate Method. In the method proposed by 
 Classen,* the zinc is deposited from a solution of the double oxa- 
 late of zinc and potassium or ammonium. To the neutral and 
 concentrated solution of zinc about 4 grams of potassium or ammo- 
 nium oxalate are added. Solution is effected by warming, and, 
 if necessary, adding a little water. The solution is diluted to 
 about 120 c.c., warmed to 50 or 60, and electrolyzed with a cur- 
 rent of 0.5 to 1 ampere with an electrode tension of 3.5 to 4.8 volts. 
 After the current has passed for three to five minutes, a cold- 
 saturated solution of oxalic acid, or better, a 6% solution of tar- 
 taric acid is allowed to flow from a burette at the rate of about 
 10 drops per minute upon the watch-crystal which covers the 
 dish or beaker. The end of the operation is ascertained by test- 
 ing small portions of the solution with potassium ferrocyanide. 
 
 * Classen, Quan. Analysis by Electrolysis, p. 146. 
 
DETERMINATION OF ZINC. 225 
 
 The time required is about two hours. The deposit must be 
 washed without interrupting the current. 
 
 318. Acetic-acid Methods. A method proposed by Riche and 
 modified by Smith * involves the use of a solution of zinc acidified 
 with acetic acid. To the, neutral solution of the zinc, from which 
 ammonium salts should be absent, 1 gram of sodium acetate, J to 
 1 gram of ammonium oxalate, and 0.3 c.c. of 99% acetic acid are 
 added. The solution is heated to 65, and a current of about 0.36 
 ampere and 4 volts passed, and the anode or cathode rotated. 
 Add acetic acid occasionally, as zinc forms the best deposits from 
 a nearly neutral solution. At the end of an hour the current is 
 increased to about 0.7 ampere and 5 volts. Water is added so as 
 to cover a clean silver surface with the solution. If no more zinc 
 is deposited and the solution is acid, it is carefully neutralized by 
 the addition of ammonia and the current passed for about fifty 
 minutes longer. The condition of the solution which requires the 
 addition of ammonia is usually indicated by the solution appearing 
 as if full of bubbles. The complete precipitation of the zinc is 
 ascertained by testing a portion of the solution with potassium 
 ferrocyanide. The solution is then siphoned off and the depos- 
 ited metal washed with water and alcohol, dried at 100, and 
 weighed. 
 
 319. Determination of Zinc in Ores. In carrying out the 
 determination of zinc in zinc ores, the finely powdered material is 
 weighed out and dissolved in hydrochloric acid, or, if necessary, in 
 nitric acid. In the latter case the nitric acid is removed by evapo- 
 ration with a little sulphuric acid and the residue dissolved in 
 water, : If considerable quantities of copper, arsenic, antimony, 
 etc., are present, hydrogen sulphide is passed through the solution 
 and the precipitate washed with water containing hydrogen sul- 
 phide. If iron, cobalt, and nickel are absent, the zinc may be 
 determined m the filtrate by the method of Classen. Iron does 
 not interfere if the method of Smith is used, since small quantities 
 are not carried down with the zinc, and large amounts would be 
 precipitated as basic acetate on warming the solution. 
 
 * Jour. Am. Chem. Soc., Vol. XXIV, p. 1073, 
 
VOLUMETRIC METHODS. 
 
 CHAPTER XIX. 
 
 CALIBRATION OF APPARATUS. 
 
 320. Definition. Volumetric are distinguished from gravi- 
 metric methods by the fact that in the former the amount of a 
 given constituent is ascertained by MEASURING the VOLUME of a 
 LIQUID, while in gravimetric methods, the element to be determined 
 is brought to a condition where it may be placed on the balance- 
 pan, and WEIGHED. Solutions of reagents which have been made 
 to contain an accurately determined amount of a given chemical 
 substance are called STANDARD SOLUTIONS. The volume of such 
 a solution, which will just complete the reaction with a given 
 constituent in a known amount of a substance to be analyzed, will 
 give the amount of the constituent to be determined. 
 
 We may, for instance, weigh out 58.50 grams of pure sodium 
 chloride and dissolve in a liter of water. The amount of silver 
 contained in a given solution may be ascertained by adding the 
 salt solution until all of the silver has been precipitated as silver 
 chloride, and further addition of the salt solution gives no further 
 precipitate. The volume of the salt solution having been carefully 
 noted, the amount of silver present may be computed from the 
 known amount of salt solution used. From the equation, 
 
 Nad + AgN0 3 = AgCl + NaN0 3 , 
 
 (23.05 + 35.45 = 58.50) (Ag = 107.93) 
 
 we know that 58.50 parts of sodium chloride can precipitate 
 107.93 parts of silver. As the salt solution contained 58.50 
 grams of sodium chloride per liter, if 100 c.c. were used to pre- 
 cipitate the silver, then 10.793 grams of silver were present in 
 the unknown solution. 
 
 226 
 
VOLUMETRIC APPARATUS. 
 
 227 
 
 VOLUMETRIC APPARATUS. 
 
 321. Flasks. The apparatus which has been devised for meas- 
 uring the solutions consists in the first place of so-called standard 
 flasks. These flasks are made of such a capacity that when filled to 
 a mark made on the stem, they contain definite volumes of liquid, 
 such as 1 liter or 1000 c.c., J liter or 500 c.c., \ liter or 250 c.c., ect. 
 
 322. Pipettes are tubes constructed to deliver definite volumes 
 having, generally, a bulb at the centre and a mark on the upper 
 part of the stem to indicate the point to which the pipette must be 
 filled to deliver the volume marked on the instrument. The liquid 
 
 s. / 
 
 FIG. 36. 
 
 is sucked up into the pipette, the upper end closed with the finger, 
 and after allowing the liquid to flow down to the mark, the contents 
 of the pipette are allowed to flow into the desired vessel. 
 
228 
 
 VOLUMETRIC METHODS. 
 
 323. Burettes are glass tubes of uniform bore, having a stop-cock 
 a t the lower end, and marked so that any desired number of cubic 
 centimeters may be measured out. The total capacity is usually 
 
 FIG. 37. 
 
 50 c.c., and the smallest divisions indicate either the fifth or the 
 tenth of a cubic centimeter. It is customary to estimate the tenth 
 of the smallest division so that volumes may; be measured with 
 burettes to the -fa or the y^ of a cubic centimeter. 
 
 324. Reading Burettes. Unices special care is taken in making 
 readings, errors of tV f a cubic centimeter may easily be made. 
 Generally the lowest part of the meniscus is the point at which 
 readings are made. As this point is in the centre of the tube, a 
 different reading will be made according to the position of the 
 eye, as will be seen from Fig. 38. One of the simplest devices for 
 securing the correct position of the eye, corresponding to the 
 point b, consists of a piece of paper or card with a perfectly straight 
 edge folded double so as to completely surround the tube. The 
 card is held so that the folded upper edges meet exactly, as shown 
 
READING BURETTES. 
 
 229 
 
 in Fig. 39. The eye is placed in such a position that the edge of the 
 card on the front of the burette, the bottom of the meniscus, and 
 the edge on the back of the burette are in the same line. The 
 reading of the burette is then taken at the upper edge of the card. 
 The eye may also be held in the same straight line with the 
 upper edges of the meniscus. The lowest point of the meniscus 
 
 FIG. 38. 
 
 FIG. 39. 
 
 is then read. This method will give a reading which is slightly 
 too high, but if the burette is always read in this manner, no 
 error will result, as the volumes measured out are always found by 
 taking the difference between two readings. 
 
 325. Floats. Another method of securing accurate readings is 
 by the use of floats, which are small glass tubes sealed at both ends 
 and containing enough mercury to cause them to be very nearly 
 immersed in dilute water solutions. A circle is etched on the 
 float at right angles to its length, and the readings are made 
 by the line which is produced by holding the eye in the plane 
 of this circle. While this line gives very exact readings, serious 
 error is often occasioned by the tendency of floats to go down 
 irregularly by a series of jumps instead of following the level of 
 the liquid. Before being used a float should be carefully tested 
 by placing it in the burette filled with water. The water is 
 allowed to flow out at a moderate rate, while the float is closely 
 observed to see if it descends at a perfectly regular rate. 
 
 326. Errors in Reading Burettes. When skill has been ac- 
 quired in using one or the other of these devices and in estimating 
 
230 
 
 VOLUMETRIC METHODS. 
 
 the tenth of one of the smallest divisions on the burette, the accu* 
 racy acquired in -practice is still never greater than T ^ of a cubic 
 centimeter; that is, the reading will be doubtful to the tenth of 
 the smallest division of the burette. As an error of this magnitude 
 is possible at both of the readings, in measuring a given volume of 
 liquid the total error may be at least two-hundredths of a cubic 
 centimeter. Careful experiments show that in careful work 
 
 FIG. 40. 
 
 errors of three-hundredths of a cubic centimeter occur. The 
 percentage error of a measurement is greater, therefore, the 
 smaller the volume of liquid measured out, being 3% for 1 c.c., 
 0.3% for 10 c.c., and 0.1% for 30 c.c. As the possible errors in the 
 other parts of volumetric analysis are not less than 0.1%, about 
 25 c.c. is the minimum volume of liquid which should be used in 
 a volumetric analysis. Greater accuracy is seldom obtained by 
 using a larger volume of liquid. 
 
CORRECT USE OF PIPETTES. 231 
 
 327. Method of Using Pipettes. If a pipette is properly used, 
 liquids may be measured with greater accuracy than is possible 
 with a burette. This arises from the fact that the tube in which 
 the initial and final readings are made is so much smaller than in 
 the case of a burette. The errors in using a pipette arise from 
 allowing the liquid to drain for varying intervals of time, and also 
 from leaving a drop of liquid on the lower end of the tube. This 
 drop should be taken off, when the liquid has been brought to the 
 mark on the upper part of the stem, by drawing the point of the 
 pipette along the glass surface of the vessel from which the liquid 
 has been drawn. The liquid should be allowed to flow out of the 
 pipette at its maximum speed, and the point should be touched 
 against a glass surface when the pipette is empty. If it has a 
 mark on the lower stem, the liquid must be allowed to flow to this 
 mark ; otherwise the pipette is completely emptied with the excep- 
 tion of the drop which always remains in the point. Some chem- 
 ists blow this drop out of the pipette. This is unnecessary, but 
 the amounts of liquid delivered will be identical if this practice is 
 uniformly adhered to. More uniform results will be obtained if 
 the hole at the point of the pipette is not over 2 mm. in diameter, 
 so that the contents of the pipette may flow out in about ten seconds. 
 If the opening is too large, it may be diminished by heating the 
 point to redness in the flame of a Bunsen burner. The upper end 
 of the stem as well as the finger used in closing it must be dry or the 
 flow of liquid cannot be properly controlled. 
 
 328. Standard Temperature. The volumes of the glass- 
 measuring apparatus vary appreciably with changes in temper- 
 ature, while the changes in the volumes of liquids is very marked 
 indeed. It is necessary, therefore, in volumetric work to keep 
 solutions and apparatus at some standard temperature. The 
 most convenient temperature is that of the air in the laboratory. 
 This changes within quite wide limits and varies in different coun- 
 tries. The German chemists have chosen 15 Centigrade as the 
 average temperature of the laboratory. American laboratories 
 are kept much warmer than this, so that 20 is a much better 
 standard for this country. This will be used as the standard 
 temperature for the work described in this book, the glassware 
 being calibrated for this temperature. 
 
232 VOLUMETRIC METHODS. 
 
 329. Expansion of Glass, Water, etc. The change in volume 
 of a liter flask due to a change in temperature of 5 C. is 0.15 c.c. 
 As this is .015% of the total volume, it is much less than other 
 experimental errors and may, therefore, be entirely disregarded. 
 The change in volume of water is much greater. A liter of water 
 measured at 20 will contract 0.9 c.c. if cooled to 15, while if 
 heated to 25 it will expand 1.144 c.c. The error in this case for 
 a change in temperature of 5 will therefore be about 0.1%. The 
 change in volume of solutions is generally much greater, the more 
 concentrated the solution the greater being the coefficient of 
 expansion. 
 
 330. Errors of Graduation. A very common source of error 
 in volumetric analysis is the inaccuracy of the graduations of the 
 measuring apparatus used. It is by no means uncommon to find 
 errors amounting to 1% in apparatus bought of the dealers, while 
 J% would, perhaps, fairly represent the error in using most un- 
 tested apparatus. The errors are due to several causes, but chiefly 
 to the fact that various standards of volume are in use. 
 
 331. Definition of Units of Volume. Volumetric apparatus 
 is usually constructed on the basis of the metric system, the units 
 of volume being the LITER and the CUBIC CENTIMETER. In this 
 system the unit of length was first selected and prepared. The 
 unit of volume was obtained from the unit of length, a liter being 
 defined as the cube whose sides are a decimeter in length. A 
 cubic centimeter is the one-thousandth part of this volume. The 
 unit of weight was obtained by preparing pure water and calling 
 the weight of one liter weighed in a vacuum at 4 C. a kilogram. 
 
 332. Various Standard Liters in Use. On account of the ease 
 of preparing pure water, measuring its temperature correctly 
 and weighing it with accuracy, the custom has arisen of using 
 accurately weighed water as the standard of volume. This cus- 
 tom is still further justified by the fact that it is comparatively 
 easy to obtain weights which are correct far beyond the accuracy 
 attainable in measuring volumes. The errors found in commer- 
 cial volumetric apparatus have arisen from a desire to avoid the 
 trouble of making the weighings of the water under the standard 
 conditions of temperature and absence of an atmosphere. In 
 many instances the suggestion made by Mohr has been adopted. 
 
STANDARDS OF VOLUME. 233 
 
 He proposed to adopt ~as a liter the volume of 1 kilogram of water 
 at 4 C., weighed in air. He adopted 17.5 C. as the standard 
 temperature. This liter would differ from the true liter by 1.2 
 cubic centimeters. Other workers have used as a liter the vol- 
 ume of one kilogram of water weighed in the air at 15 C., while 
 still others weigh at 17.5 C. or 20 C. 
 
 With these various standards in use it is not surprising, espe- 
 cially when taken in connection with the possible errors of work- 
 manship, that the use of uncalibrated apparatus frequently pro- 
 duces very erroneous results. The possibility of error is consider- 
 ably reduced if all of the volumetric apparatus in a laboratory 
 has been made by the same manufacturer, since no error will 
 result from the use of a standard other than the true liter if all of 
 the apparatus is graduated according to the same standard, the 
 cubic centimeter being exactly the one-thousandth part of the 
 liter. Apparatus used for gas analysis must, however, be grad- 
 uated according to the true liter. All confusion would be avoided 
 if the true liter were generally adopted. The German chemists 
 have taken the lead in this direction. The German Imperial 
 Standards Commission has made it illegal to use any other than 
 the true liter for official purposes. The directions given for cali- 
 bration in this book will be based on the true liter. 
 
 CALIBRATION. 
 
 There are in use two general methods of calibrating volumetric 
 apparatus. By the first method the weight and temperature of 
 the water which just fills the apparatus to be calibrated is found 
 and the volume calculated from the known density of water. By 
 the second method standard bulbs whose capacity has been care- 
 fully determined are used. The water which just fills the bulb is 
 allowed to flow into the flask to be calibrated. Where many pieces 
 of apparatus must be calibrated this method is to be preferred. 
 
 333. Calibration of Flasks by Weighing Water. Flasks which 
 are to be calibrated are first cleaned and dried, then carefully 
 counterpoised or weighed. Distilled water at 20 C. is poured in 
 until the flask is filled to the mark. The additional weights neces- 
 sary to again counterbalance the flask give the weight of the 
 
234 VOLUMETRIC METHODS. 
 
 water. The weight of the water in vacuo must now be calculated. 
 The air displaced by a kilogram of water may be taken as a liter, 
 which under ordinary temperatures and pressures weighs 1.2 
 grams. When in the air the water is lighter by this amount. 
 The weights, however, are also lighter in the air than in a vacuum. 
 The specific gravity of brass weights is 8.4. A kilogram of brass 
 weights would displace -| of a liter of air which would weigh 
 .14 gram. As the buoyant force on the water is 1.2 grams and 
 on the weights .14 gram, the diminution hi weight in the air is 
 1 .06 grams (1 .20 . 14) . A kilogram of water, therefore, weighed in 
 the air with brass weights would weigh 998.94 grams (1000 1.06), 
 the weight in a vacuum being exactly 1000 grams. The general 
 correction for air displacement is therefore +.106%. The weight 
 of the water in a vacuum divided by the density at 20 gives the 
 volume. In this manner the volume of the flask when filled to 
 the mark is obtained. If a flask is to be calibrated to deliver a 
 definite volume it should be rinsed with distilled water and 
 allowed to drain for the same length of time employed in ordinary 
 use. It is then weighed and the calibration carried out as already 
 described. 
 
 If it is desired to ascertain the point on the stem to which the 
 flask must be filled in order to contain one liter, it should be dried 
 and counterbalanced on the scale-pan, and then weights equal to 
 997.2 * grams placed on the opposite pan and distilled water at 
 20 C. poured into the flask until the weights are again counter- 
 balanced. A mark may now be made with a file on the neck of 
 the flask opposite the lowest point of the meniscus, and then a 
 circular mark may be etched on by means of hydrofluoric acid. 
 For this purpose a thin layer of paraffine or beeswax should be 
 melted on the neck of the flask and a circle traced with a sharp 
 steel point while the flask is rotated in a lathe or some similar 
 device. The acid may be applied by means of a piece of filter- 
 paper and should be allowed to remain from three to five minutes. 
 
 334. Calibration of Burettes by Weighing Water. The burette 
 is filled with distilled water at about 20 C. A small flask or 
 
 * 1000 c.c. water at 20 in a vacuum weigh 998.26 grams. From this 
 weight the loss in weight of 1.06 grams due to air displacement must be sub- 
 tracted. 
 
CALIBRATION BY WEIGHING WATER. 235 
 
 weighing-tube capable of holding 50 c.c. is weighed. The water 
 in the burette is now brought exactly to the zero-point and then 
 about 5 c.c. is allowed to flow into the flask. The exact reading 
 of the burette should be taken and the flask weighed. Another 
 5-c.c. portion of water is then allowed to flow into the flask and 
 the burette again read and the flask weighed. This process is 
 repeated until the last reading on the burette is exactly 50 c.c. 
 and the flask contains the water delivered by the burette between 
 zero and the 50-c.c. mark. The water need be weighed only to 
 the hundredth of a gram, as volumes smaller than .01 c.c. cannot 
 be measured on ordinary volumetric apparatus. 
 
 The volumes corresponding to the weights of water are now 
 calculated by the method given for the calibration of a flask, 
 .997 gram of water being equal to 1 c.c. The difference between 
 the volume indicated on the burette and that calculated from 
 the weight of the water is the correction to be applied at that 
 point. The correction for intermediate points is found by plot- 
 ting a curve, as shown in Fig. 44, in which the ordinates are the 
 numbers 1 to 50 representing the readings on the burette, while 
 the abscissas are the corrections found by the calibration. The 
 assumption is made that the variations in volume are uniform 
 between the points calibrated. Generally this does not lead to 
 serious error. If the curve is very irregular, the burette 'should 
 be calibrated for every 2 or 3 c.c. 
 
 CALIBRATION BY MEANS OF THE MORSE-BLALOCK 
 
 BULBS. 
 
 335. Description of Bulbs. If many pieces of apparatus are to 
 be calibrated the Morse-Blalock calibrating bulbs should be used. 
 These authors recommend the use of 6 bulbs of different sizes 
 combined into 3 pieces of apparatus, as shown in Fig. 41. Three 
 bulbs of capacities 2 c.c., 3 c.c., and 50 c.c., respectively, are all 
 blown on one stem, various parts of which are graduated. Two 
 bulbs of 50 c.c. and 200 c.c. capacity are connected on a single 
 stem, which is also graduated, while the 500-c.c. bulb is on a sepa- 
 rate graduated stem. Bulbs of other sizes could be made as 
 desired. The intention of Morse and Blalock being to adapt the 
 bulbs for use at a considerable range of temperature, they were 
 
236 
 
 VOLUMETRIC METHODS. 
 
 made of such a capacity that the volume of a bulb at 30 should not 
 be greater than that marked on it, while at the combined vol- 
 ume of the stem and bulb should not be less than that indicated 
 on the bulb. 
 
 336. Calibration of the Bulbs. The exact capacity of each bulb 
 
 FIG. 41. 
 
 is ascertained by filling it with distilled water and weighing the 
 amount delivered on emptying it. The temperature of the 
 water is noted and the volume of the bulb for this temperature 
 is calculated. The temperature of the water need not be that at 
 which the bulb is to be used, since the volume of the bulb having 
 
MORSE-BLALOCK CALIBRATING BULBS. 237 
 
 been ascertained for a given temperature, by using the coeffi- 
 cient of expansion of the glass of which volumetric apparatus is 
 made, .000026, the volume at any other temperature can be cal- 
 culated. The volume of the graduated stem is ascertained in 
 the same manner. In using a bulb for calibration it is filled with 
 distilled water, which is allowed to flow into the dry and empty 
 flask until the bulb and a sufficient number of divisions on the 
 stem are emptied to give the required volume. An example will 
 make this clearer. 
 
 Weight of Water from Bulb. Calculated Volume. 
 
 At 17.5. At 14.3. At 20. At 15. 
 
 498.340 gr 499.51 c.c. 
 
 498. 456 gr. 499.43 c.c. 
 
 Av. 499 . 47 c.c. 499 . 40 c.c. 
 
 Weight of Water from Stem. 
 At 14.6. At 14.7. 
 
 3 . 177 gr 3 . 182 c.c. 
 
 3.180gr. 3. 185 c.c. 
 
 Av. 3. 184 c.c. 
 
 Volume of 1 division of stem = .032 c.c. 
 
 Bulb +17 divisions of stem =500.00 c.c. at 20 (499.47+17 
 X.032). 
 
 Bulb+19 divisions of stem =500.00 c.c. at 15 (499.40+19 
 X.032). 
 
 337. Temperature of Water Used in Calibration. If it is de- 
 sired to calibrate a flask to hold 500 c.c. at 20, the water con- 
 tained in the bulb and 17 divisions of the stem is allowed to flow 
 into the flask and a mark made on the stem opposite the bottom 
 of the meniscus. The temperature of the water need not be 20. 
 If the temperature of the water and consequently of the bulb is 
 greater than 20, the volume of , the bulb and 17 divisions of the 
 stem will be greater than 500 c.c., but on cooling to 20 it will 
 be exactly 500 c.c. The temperature of the flask will be raised 
 to that of the water, so that the volume of the flask will be greater 
 than 500 c.c., but .the flask also will contract on cooling, and as 
 the bulbs are made of the same kind of glass as the flasks, the con- 
 traction of the flask will be exactly equal to the contraction of the bulb. 
 
238 
 
 VOLUMETRIC METHODS. 
 
 The same line of reasoning will show that the temperature of 
 the water may be below the standard temperature without affect- 
 ing the accuracy of the calibration. 
 
 338. Calibration of Flasks. By means of the 50-c.c. and 
 200-c.c. bulbs, flasks of capacities which are multiples of 50 c.c. 
 may be calibrated. For flasks whose capacities are multiples of 
 500 c.c., the large bulb is used. 
 
 FIG. 42. 
 
 339. Calibration of Burettes. The 50-c.c. bulb with 2-c.c. 
 and 3-c.c. bulbs are used for calibrating burettes. For this work 
 
CALIBRATION OF BURETTES. 
 
 239 
 
 as well as for calibrating flasks it is necessary to have a bottle 
 holding several liters placed on a shelf which is above the top of 
 the burette or bulb when fixed in clamps for use. A Friedrich- 
 
 FIG. 43. 
 
 Greiner two-way stop-cock is attached to the bulb so that water 
 may be withdrawn from the bulb by two openings in the stop- 
 cock. One of these tubes is connected with one arm of a T-tube, 
 the second arm being connected by a tube to the large bottle on 
 
240 VOLUMETRIC METHODS. 
 
 9 
 
 the shelf, and the third one is connected with the burette. The 
 latter must be clamped in such a position that the point giving 
 the last reading is above the highest point of the calibrating bulb. 
 All joints must be absolutely water-tight, and the connecting 
 tubes should be of glass as far as possible. A pinch-cock should 
 be placed on the tube leading to the large bottle, which is filled 
 with distilled water. By opening this pinch-cock water may be 
 admitted either to the burette or to the bulb. All bubbles of air 
 must be displaced from the connecting tubes. 
 
 340. Cleaning Mixture. The burette and calibrating bulb 
 must be cleaned so that a continuous film of water is left on 
 emptying them. A most excellent cleaning mixture is made by 
 adding a few grams of potassium dichromate to a half liter of 
 concentrated commercial sulphuric acid. This mixture should be 
 kept in a glass-stoppered bottle to which portions which have 
 been used may be returned unless largely diluted with water. 
 Apparatus filled with this mixture and allowed to stand for some 
 time will be perfectly clean on pouring out the cleaning mixture 
 and washing with water. The cleaning mixture is much more 
 efficient if warmed. When it has become green because of reduced 
 chromium, it must be discarded or more chromate added. Burettes 
 and pipettes must always be kept clean or errors arise in their 
 use because of the drops of liquid which adhere to the glass. 
 If burettes are filled with distilled water when not in use, they 
 keep cleaner than if left empty. 
 
 341. Total Capacity. The bulb and burette are filled with 
 water which is allowed to flow out at a moderate rate. The burette 
 is again filled with water, and by cautiously turning the two-way 
 stop-cock, water is allowed to flow from the burette into the bulb 
 until the reading in the former is exactly zero. The stop-cock 
 is now turned through 180, and the 50 c.c. bulb emptied until the 
 meniscus is opposite the mark on the stem. By again turning the 
 stop-cock the water from the burette is allowed to flow into the 
 bulb until the reading on the burette is exactly 50 c.c. The 
 number of divisions of the stem of the bulb filled by the water is 
 
 "now noted. From this the total capacity of the burette can be 
 calculated.* Duplicates should agree within 1 or 2 divisions of 
 the stem. The 50-c,c. bulb should be so constructed that its 
 
CALIBRATION OF BURETTES. 241 
 
 volume is not greater than 49.75 c.c., while the capacity of the 
 stem should be at least .75 C.G., so that the total capacity of the 
 stem and bulb shall be about 50.50 c.c. If the capacity of the 
 burette is greater than 50.50 c.c., the stem may be emptied, and 
 water again allowed to flow in from the burette. 
 
 342. Irregularities of Bore. The burette is once more filled 
 with water which is brought to the zero-mark. It is then 
 allowed to fill either the 2-c.c., 3-c.c., or both bulbs, depending 
 upon the accuracy desired in the calibration. The reading of the 
 burette is carefully taken, the small bulb emptied, and the opera- 
 tion repeated until the burette has been emptied. The capacity 
 of the small bulbs should be a trifle less than that indicated, so 
 that the last reading on the burette may always be less than 50 c.c. 
 The method of calculating the results can best be given by the 
 results of an actual calibration. 
 
 Total capacity =50 c.c. bulb + 78 divisions of stem. 
 
 Total capacity =50 c.c. bulb + 80 divisions of stem. 
 
 Volume of 50 c.c. bulb =49.62 c.c. 
 
 Volume of stem (100 divisions) =.7807 c.c. 
 
 Total capacity of burette =50.24 c.c. (49.62 +.0078X79). 
 
 Readings. 
 
 0.00 
 
 True Capacity. 
 
 0.00 
 
 Corrections. 
 
 0.00 
 
 1.99 
 
 1.98 
 
 -0.01 
 
 3.98 
 
 3.96 
 
 -0.02 
 
 5.92 
 
 5.94 
 
 + 0.02 
 
 7.90 
 
 7.92 
 
 + 0.02 
 
 9.89 
 
 9.89 
 
 0.00 
 
 11.87 
 
 11.87 
 
 0.00 
 
 13.83 
 
 13.85 
 
 + 0.02 
 
 15.81 
 
 15.83 
 
 + 0.02 
 
 17.79 
 
 17.81 
 
 + 0.02 
 
 19.77 
 
 19.79 
 
 + 0.02 
 
 21.71 
 
 21.77 
 
 + 0.06 
 
 23.68 
 
 23.75 
 
 + 0.07 
 
 25.64 
 
 25.72 
 
 + 0.08 
 
 27.60 
 
 27.70 
 
 + 0.10 
 
 29.58 
 
 29.68 
 
 + 0.10 
 
242 VOLUMETRIC METHODS. 
 
 Readings. True Capacity. Corrections. 
 
 31.55 31.66 +0.11 
 
 33.50 33.64 +0.14 
 
 35.48 35.62 +0.14 
 
 37.45 37.60 +0.15 
 
 39.40 39.58 +0.18 
 
 41.38 41.56 +0.18 
 
 43.32 43.53 +0.21 
 
 45.29 45.51 +0.22 
 
 47.28 47.49 +0.21 
 
 49.24 49.47 +0.23 
 
 50.00 50.24 +0.24 
 
 The last reading being 49.24 and the true capacity at 50.00 
 being 50.24, the true capacity at 49.24 is found by the proportion 
 
 50 : 50.24 : : 49.24 : x, where x =49.47. 
 
 As the 2-c.c. bulb has been filled twenty-five times with water, 
 the volume of which is 49.47 c.c., the capacity of the bulb must- 
 be ^5- of 49.47, or 1.979. As at each successive reading of the 
 burette one bulb-full of water has been taken out, the true volumes, 
 at these points are found by multiplying 1.979 by the numbers, 
 from 1 to 25. These values are given in the second column. The 
 differences between these values and the readings of the burette 
 are the corrections to be applied at these points. The correc- 
 tions are plotted on co-ordinate paper as abscissa, while the cor- 
 responding burette readings are the ordinates, as shown in Fig. 44. 
 
 EXERCISE 45. 
 Calibration of Morse and Blalock Bulbs. 
 
 Clean one of the bulbs by filling with bichromate solution and, after ten 
 minutes, empty it and wash with water. Clamp it in position with the 
 Greiner-Friedrich stop-cock, which is connected with the large bottle of 
 distilled water, as shown in Fig. 42. Fill with water and allow it to 
 flow out at a moderate rate. If the film of water left on the glass breaks, 
 leaving drops, the cleaning with the bichromate solution must be repeated. 
 When the bulb is clean, fill again with distilled water and bring the lower 
 part of the meniscus exactly on a line with the upper mark on the bulb. 
 
CALIBRATION CURVE. 
 
 243 
 
 CORRECTIONS '" CUBIC CENTIMETERS 
 -Q.I go -t-ai +0.2 0.0 +0.1 +02 +03 go +01 
 
 FIG. 44. 
 
 
244 VOLUMETRIC METHODS. 
 
 Let the water flow at a moderate rate into a weighed flask until the menis- 
 cus is at the lower mark on the bulb. Weigh the water to milligrams. 
 Draw out another portion of the water into a flask and take its tempera- 
 ture. Repeat the determination until duplicates are obtained which are 
 identical to at least 1 part in 1000. Calculate the volume using the specific 
 gravity of water at the temperature noted and a correction of +.106% for 
 air displacement. Calculate the volume at 20. In the same manner ascer- 
 tain the volume of the graduated stem attached to the bulb. Calculate 
 the number of divisions of the stem which must be used to make the volume 
 equal to that marked on the bulb. 
 
 EXERCISE 46. 
 Calibration of Flasks. 
 
 Clean and dry the standard flasks you have. Clean all calibrating 
 bulbs to be used according to the directions given in the preceding exer- 
 cise. Clamp the ^-liter bulb in position as shown in Fig. 42. Fill the 
 bulb to the upper mark and allow the water to flow into the liter flask 
 until the bulb and the number of divisions on the stem giving 500 c.c. 
 have been emptied. Repeat the operation and immediately make a 
 mark on the neck of the flask opposite the bottom of the meniscus. Cali- 
 brate the other flasks in a similar manner. A thin layer of beeswax or par- 
 affine should be melted on the neck and a circular line passing through 
 the file-mark made with a sharp steel point. The flask should be held in a 
 lathe for this purpose. The mark is etched on by dipping a piece of filter- 
 paper into hydrofluoric acid and moistening the mark with the acid. After 
 about three minutes the acid is washed off with water. After wiping the 
 flask dry, the paraffine is melted by waving the neck of the flask in the 
 Bunsen-burner flame. The neck of the flask may then be cleaned by wiping 
 with dry filter-paper. 
 
 EXERCISE 47. 
 Calibration of Burettes. 
 
 After cleaning your burette, set it up with the 50-c.c. bulb as shown in 
 Fig. 43, p. 239. Test the apparatus for leaks by filling with water and wiping 
 all joints dry with filter-paper. No drops of water should be visible after 
 fifteen minutes. All joints except a and b may be made tight by painting 
 the glass tubing with shellac varnish before the rubber tubing is put on. 
 The total capacity of the burette is first found by filling it with water by 
 opening the stop-cock c and closing d. The water is carefully brought 
 to the zero-mark by opening a (which should now remain open) and d. 
 
* 
 
 CALIBRATION OF BURETTES. 245 
 
 The stop-cock d is now turned so that water can flow* into the beakei 
 until it is brought to the lower mark on the 50-c.c. bulb, d is again turned 
 so that water is slowly admitted from the burette until it reads exactly 
 50 c.c. The number of divisions on the stem of the 50-c.c. bulb is now 
 read. The determination is repeated until duplicates^agreeing within 2 
 or 3 cU visions of the stem are obtained. From the values already found 
 for the 50-c.c. bulb, and for one division of the stem, the total capa^fty of 
 the burette is calculated. 
 
 The burette is""aga"m filled with water and the reading brought to the 
 
 .zero-mark. If the total capacity is not more than 0.2 c.c. over 50 c.c., the 
 burette may be calibrated every 5 c.c. For thisr- purpose the meniscus 
 of the water in the calibrating bulb is brought to the mark below the 2-c.c. 
 bulb and water allowed to flow in from the burette until the 3-c.c. bulb is 
 filled to the mark on the stem. The reading on the burette is now taken 
 and the operation repeated until the burette is empty. Repeat the calibra- 
 tion until duplicates are- obtained for each reading agreeing within less 
 than .02 c.c. 10 portions of approximately 5 c.c., but exactly equal, 
 will now have been withdrawn from the burette. The method of calcu- 
 lating the re'sult is best given by the following example: A burette 
 whose total capacity was found to be 50.15 c.c. gave as the last reading 
 with the 5-c.c. portions 49.90 c.c. The volume at this point is found by 
 the following proportion: 50.00 : 49.90 : : 50.15 :x, from which x, OP the 
 volume of 10 of the 5-c.c. portions withdrawn, equals 50.05 c.c. Each 
 5-c.c. portion is therefore equal to 5.005 c.c. This number multiplied by 
 the numbers from 1 to 10 gives the true volumes at the burette readings 
 
 * taken. The differences between the readings and the true volumes give the 
 corrections to be applied at the points calibrated. To find the corrections 
 at intermediate points, plot a curve with the corrections as the abscissas, 
 and the readings on the burette as ordinates, as shown in Fig. 44. If 
 the difference in the corrections between any two consecutive readings is 
 more than .05 c.c:, this portion of the burette should be recalibrated with 
 the 2-c.c. or the 3-c.c. bulb. 
 
 PROBLEMS. 
 
 Problem i. The weight of water at 20 C., which just fills a flask, is 
 249.2 grams. Calculate (a) the volume of the flask at 20, (6) its volume 
 at 1, (c) its volume at 25. 
 
 Problem 2. A burette was calibrated with a Morse and Blalock bulb, 
 of which the 50-c.c. bulb had a capacity of 49.55, and the stem of 100 divi- 
 sions had a capacity of 0.73 c.c. When obtaining the total capacity of the 
 burette, the bulb was filled and 67 divisions of the stem. When calibrating 
 the bore of the burette, 5-c.c. portions were withdrawn from the burette by 
 means of the 2- and 3-c.c. bulbs. The following readings were obtained: 
 5.05, 10.09, 15.09, 20.19, 25.19, 30.19, 35.15, 40.19, 45.19, 50.10. Calculate 
 the total capacity of the burette and the corrections at the readings given. 
 

 
 246 VOLUMETRIC METHODS. 
 
 Problem 3. Another burette was calibrated with the same bulb. For 
 the total capacity, the bulb full plus 70 divisions of the stem was obtained. 
 When calibrating the bore of the burette, the following readings were taken: 
 5.00, 10.11, 15.09, 20.09, 25.08, 30.09, 35.05, 40.09, 45.10, 50.10. Calculate 
 the total capacity of the burette and the corrections at the readings given. 
 
 The accuracy of the calibration of these burettes may be judged by a 
 comparison of the volumes calculated for the 5-c.c. portions withdrawn 
 from the burettes. These values should not differ by more than .03 c.c. 
 
 > 
 
 - 
 
 
CHAPTER XX. 
 ACIDIMETRY. 
 
 343. A Standard Solution has already been defined as one 
 which has been made so as to contain an accurately determined 
 amount of a given chemical substance in a definite volume. It 
 has been found that the acidity of a solution depends on the 
 amount of replaceable or acid hydrogen present, irrespective of 
 the amount of other substances present. A standard solution of 
 an acid may therefore be defined as one which contains a known 
 amount of acid or replaceable hydrogen in a definite volume. In 
 practice it has been found very convenient to use standard 
 solutions of acids which contain 1 gram * of replaceable hydro- 
 gen per liter. Such a standard solution is called a NORMAL solu- 
 tion. The strength of any standard solution may be given in 
 terms of a normal solution. It may for instance be J, J, fa 
 etc., normal, or 3, 3J, 1.362, etc., times normal. 
 
 344. Normal Acids. The amount of a given acid to be dissolved 
 in a liter to make a normal solution will vary directly as its molec- 
 ular weight, and inversely as the number of acid or replaceable 
 hydrogens in the molecule. HYDROCHLORIC ACID having a molec- 
 ular weight of 36.458 (Cl =35.45, H = 1.008), a liter of a normal 
 solution of this acid must contain 36.458 grams of the acid. The 
 molecular weight of SULPHURIC ACID is 98.076 (S =32.06, 4 =64.00, 
 H 2 =2.016). As the molecule, however, contains two replaceable 
 hydrogen atoms, one-half of the molecular weight, or 49.038 grams, 
 must be dissolved in a liter to make a normal solution. Normal 
 solutions of all monobasic acids must therefore contain per liter 
 the weight of acid equal to the molecular weight in grams, while 
 
 * Since the system of atomic weights having oxygen as 16.00 as the basis 
 has come into use, 1.008 grams of hydrogen per liter has become the standard. 
 
 247 
 
248 VOLUMETRIC METHODS. 
 
 one-half the molecular weight in grams must be taken per liter for 
 dibasic acids. 
 
 As the acid value of a solution depends wholly on the amount 
 of replaceable hydrogen present, and as the designation of the 
 strength of a standard acid in terms of normal gives the amount 
 of hydrogen per liter, the weight of the acid radicle need not be 
 known for calculating the results of a titration. If caustic soda 
 is titrated with normal hydrochloric acid, one liter of the acid 
 will neutralize exactly 40.058 grams of NaOH(Na =23.05, = 16.000, 
 H=1.008), according to the equation HCl+NaOH=NaCl+H 2 0. 
 If normal sulphuric acid were used the reaction would take place 
 according to the equation H 2 S0 4 + 2NaOH = Na 2 S0 4 + 2H 2 0. 
 Two molecules of caustic soda or 80.116 grams per liter would be 
 neutralized by one molecule of sulphuric acid or 98.076 grams per 
 liter. But half of the latter amount having been taken for a liter 
 of normal sulphuric acid, only one molecule of the caustic soda 
 or 40.058 grams per liter of acid will be neutralized. It may be 
 said then in general that a liter of a normal acid will neutralize 
 40.058 grams of caustic soda. 
 
 If sodium carbonate were being titrated only half of the molec- 
 ular weight must be taken, as one molecule of this base is capable 
 of replacing two atoms of acid hydrogen, the molecular weight 
 in grams, 106.10, being able to replace 2 grams of acid hydrogen. 
 One liter of normal acid is therefore equal to 53.05 grams of sodium 
 carbonate. One- thousandth of this weight or the amount per 
 cubic centimeter is more convenient for use in calculation, being 
 .040058 gram of caustic soda and .05305 gram of sodium carbonate. 
 If the strength of the acid must be expressed in fractions or deci- 
 mal parts of normal, the value per cubic centimeter will be found 
 by multiply ing the value per cubic centimeter of a normal acid by 
 the fractions or decimals. The value in caustic soda per cubic 
 centimeter of T V normal acid will be T V of .040058 or .0040058. 
 
 When many titrations of a given substance must be made the- 
 calculation is simpler if the acid is so made that 1 c.c. is equal to 
 an amount of base, which can be expressed by a simple number 
 such as .050 gram of sodium carbonate. If such an acid is used 
 for titrating any other substance, the calculation is not as simple 
 as when the value of the acid is expressed in terms of normal. 
 
ACIDIMETRY. 249 
 
 345. Percentage Given by Number of Cubic Centimeters of Acid 
 Used. The calculation of the result for a single substance may 
 be made very simple indeed if the acid is made of such a strength 
 that the percentage is given by the number of cubic centimeters 
 used in titrating a given weight of the substance. For instance, 
 1 gram of sodium carbonate requires for its neutralization .6872 
 gram of hydrochloric acid. This follows from the fact that 1 
 molecule of sodium carbonate (106.10) is neutralized by 2 mole- 
 cules of hydrochloric acid (72.916). If a standard solution of 
 hydrochloric acid is made of such a strength that it contains 6.872 
 grams of acid per liter, then 100 c.c. of this acid will just neutral- 
 ize 1 gram of pure sodium carbonate. If mixtures of sodium 
 carbonate and neutral salts are analyzed, the percentage of sodium 
 carbonate will be given by the number of cubic centimeters of 
 the acid necessary to neutralize 1 gram of the mixture. The 
 strength of the acid in terms of normal may be found from the 
 ratio of 6.872 to 36.458, which gives .1885 normal. As burettes 
 holding 50 c.c. are commonly used, it will generally be found 
 more convenient when mixtures containing more than 50% of 
 sodium carbonate are to be analyzed to make the acid twice 
 as strong. Then the percentage will be twice the number of cubic 
 centimeters used. The same object will also be accomplished if 
 acid of the same strength is used and gram of the material is 
 weighed out for each determination. 
 
 The percentage may also be given by the number of cubic 
 centimeters of a normal acid, provided a definite weight of the 
 material is taken. In the case of sodium carbonate, 100 c.c. of a 
 normal acid will be equal to 5.305 grams of the pure material, since, 
 as has already been shown, one liter of the normal acid is equal to 
 53.05 grams of sodium carbonate. If 5.305 grams were weighed, 
 the number of cubic centimeters of normal acid used would be 
 equal to the percentage of sodium carbonate. It would be more 
 convenient to weigh out 2.652 grams and multiply the number of 
 cubic centimeters used by 2. If a fifth-normal acid were used, 
 one-fifth or one-tenth of 5.305 grams would be taken. 
 
250 VOLUMETRIC METHODS. 
 
 INDICATORS. 
 
 346. Effect of Carbon Dioxide on Indicators. The results 
 obtained by titrating a given base with a given standard acid 
 will differ according to the indicator used to give the end-point. 
 This difference is due to the varying degree of sensitiveness of 
 the indicators. For instance, a given amount of sodium carbo- 
 nate will require about half as much standard hydrochloric acid 
 for its neutralization if phenolphthalein is used as the indicator as 
 when methyl orange is employed. This arises from the fact that 
 when hydrochloric acid is added to a solution of sodium carbonate 
 sodium chloride is produced and carbonic acid liberated. As 
 carbonic acid is too weak an acid to change the color of methyl 
 orange, the addition of the strong hydrochloric acid is continued 
 until a slight excess is present, which gives the acid color to the 
 indicator. Phenolphthalein, on the other hand, is sensitive to 
 carbonic acid. Sodium carbonate reacts alkaline with phenol- 
 phthalein until, by the addition of a strong acid like hydrochloric, 
 one-half of the sodium has been converted into sodium chloride, 
 and the remainder is present as the bicarbonate. Another drop 
 of hydrochloric acid liberates a little carbonic acid, which changes 
 the color of the phenolphthalein. If the solution is concentrated 
 or hot, carbon dioxide will escape, so that if the solution is 
 boiled the titration of sodium carbonate will give the same result 
 with either indicator. 
 
 As carbon dioxide is present in the air, in ordinary distilled 
 water, and in most solutions of acids as well as alkalies, unless 
 special precautions are taken to exclude it, the analyst must be 
 thoroughly familiar with the behavior towards this acid of all 
 indicators used. As has already been stated, methyl orange 
 is almost entirely unaffected by it, while phenolphthalein is sen- 
 sitive to one of the two hydrogen atoms of the acid. Carbonic 
 acid may therefore be said to be monobasic to phenolphthalein 
 in cold dilute solution. Its basicity is zero in boiling solution 
 with phenolphthalein, as well as with methyl orange both hot and 
 cold. Much valuable work on indicators has been carried out 
 by R. T. Thomson,* who has divided them into three classes. 
 
 * C. N., v. 47, p. 123; 185, v. 49, pp. 32, 119. J. S. C. I., v. 6, pT 195. 
 
IN DIC A TORS. 251 
 
 The first class, of which methyl orange is the best example and 
 the one most commonly used, comprises lacmoid, dimethyl amido- 
 benzene, cochineal, and congo red. This class is most sensitive to 
 alkalies and can be used with strong acids only, being unaffected 
 by weak acids such as carbonic, silicic, etc. The second class, 
 of which phenolphthalein is the type and the one most commonly 
 used, includes also turmeric. These indicators are extremely sensi- 
 tive to acids, and are therefore especially adapted to the titration 
 of weak acids among which most organic acids must be placed. 
 Strong bases must be used with these indicators, and carbon 
 dioxide must be carefully excluded. The third class, of which 
 litmus is the type, includes rosolic acid and phenacetolin. This 
 class of indicators is about midway between the other two 
 in sensitiveness, being fairly sensitive to both weak acids and 
 weak bases. 
 
 347. Basicity of Acids with Various Indicators. The table on 
 page 252 prepared by Thomson gives the number of molecules 
 of a univalent base, such as caustic soda or potash, neutralized 
 by a molecule of each acid in the presence of the indicator at the 
 head of the column. A blank indicates the absence of a sharp 
 end-point. The same results are obtained with calcium or barium 
 hydrate where insoluble compounds are not formed. 
 
 348. Litmus, although one of the earliest indicators used, has 
 by no means been superseded by the many indicators recently 
 introduced, especially by the workers in organic chemistry. It 
 is considerably more sensitive than methyl orange, not only being 
 more sensitive to weak acids, but giving an end-point more readily 
 distinguished by most workers. Carbon dioxide interferes seri- 
 ously, so that to" obtain the best results this acid must be quite 
 completely excluded. This may be readily done by boiling the 
 solution. This operation does not destroy the indicator, but as 
 it is considerably more sensitive cold than hot, the solution should 
 be cooled before the titration is completed. For this reason 
 methyl orange is more commonly used when considerable carbon 
 dioxide is present. 
 
 As obtained in commerce litmus consists of small cubes com- 
 posed of powdered chalk saturated with the coloring-matter, 
 which is a vegetable extract. The coloring-matter is not a single 
 
252 
 
 VOLUMETRIC METHODS. 
 
 Acids. 
 
 Methyl 
 Orange. 
 
 Phenolphthalein. 
 
 Litmus. 
 
 Name. 
 
 Formula. 
 
 Cold. 
 
 Cold. 
 
 Boiling. 
 
 Cold. 
 
 Boiling. 
 
 Sulphuric. . . 
 
 H 2 S0 4 
 HC1 
 HN0 3 
 H 2 S 2 3 
 H 2 CO 3 
 H 2 S0 3 ' 
 H 2 S 
 H 3 P0 4 
 H 3 AsO 4 
 H 3 AsO, 
 HN0 2 
 
 H 4 Si0 4 
 H 3 B0 3 
 
 H 2 Cr(V 
 H 2 C 2 4 
 HC 2 H 3 2 
 HC 4 H 7 2 
 H 2 C 4 H 4 4 
 HC,H 8 O f 
 H 2 C 4 H 4 6 
 H 3 C 6 H 5 7 
 
 2 
 1 
 1* 
 
 2 
 
 1 
 
 
 1 
 1 
 
 indicator 
 destroyed 
 
 
 1 
 
 2 
 1 
 1 
 2 
 1 dilute 
 2 
 1 dilute 
 2 
 2 
 
 1 
 
 t 
 2 
 
 2 
 
 1 
 1 
 2 
 1 
 2 
 3 
 
 2 
 1 
 1 
 2 
 
 
 
 
 2 
 2 
 
 2 
 1 
 1 
 2 
 
 
 1 
 
 
 
 2 
 1 nearly 
 1 nearly 
 2 nearly 
 1 
 2 
 
 2 
 1 
 1 
 2 
 
 
 
 
 'o 
 
 
 
 . i 
 2 
 
 Hydrochloric 
 Nitric 
 
 Thiosulphuric .... 
 Carbonic 
 
 Sulphurous 
 Hydrosulphuric. . 
 Phosphoric. . . . 
 
 Arsenic 
 
 Arsenious .... 
 
 Nitrous 
 
 Silicic 
 
 Boric 
 
 Chromic 
 
 Oxalic 
 
 Acetic 
 
 Butyric 
 
 Succinic 
 
 Lactic 
 
 Tartaric 
 
 Citric 
 
 
 * Concentrated nitric acid sometimes contains oxides of nitrogen, producing 
 on dilution nitrous acid which destroys methyl orange. 
 
 f 1 in a solution containing 30% of glycerine, or a few grams of mannitol. 
 
 substance. Various methods have been proposed for obtaining 
 the pure coloring-matter for use as a sensitive indicator. One of 
 the simplest and best methods consists in extracting the cubes 
 two or three times with hot 85% alcohol. By this treatment a 
 violet coloring-matter is removed. The residue is then extracted 
 several times with water, the first portion being rejected, as it 
 contains the bulk of the alkali carbonates. The remainder of 
 the water-extract is acidified with dilute sulphuric acid and boiled 
 for some time to completely expel carbonic acid. The excess of 
 sulphuric acid is then neutralized with barium hydrate. The 
 neutral color is violet, which a drop of alkali turns to a decided 
 blue, and a drop of acid to a decided red. The solution should be 
 
INDICATORS. 
 
 preserved from mould by the addition of a few drops of chloroform. 
 If kept in a closely stoppered bottle, the color disappears, but 
 is restored on exposure to the air. The solution is best pre- 
 served in a wide-mouthed bottle partially filled, and protected 
 from dust by a cover of filter-paper, or a loose cork stopper, through 
 which passes a glass tube with which to withdraw the solution as 
 needed. 
 
 349. Cochineal solution is very similar to litmus solution. 
 When acidified the color is red, which is turned to violet by alkalies. 
 It is not as sensitive to carbonic and weak organic acids as litmus. 
 This gives it an advantage over the latter, as carbon dioxide need 
 not be so rigidly excluded. It cannot be used, however, in the 
 presence of even traces of compounds of iron or aluminium 0** 
 acetates. It is used largely in the titration of ammonia in the 
 Kjeldahl method for determining nitrogen. The solution is made 
 by digesting 1 part of the crushed cochineal with 10 parts of 25% 
 alcohol. 
 
 350. Methyl Orange is the sodium or ammonium salt of para- 
 dimethyl-aniline-azo-benzene-sulphonic acid. It is a synthetic 
 product, and can generally be purchased in a sufficiently pure 
 condition for use. It is dissolved in water, 1 gram per liter giving 
 a solution of which 1 drop is sufficient for most titrations. A 
 large amount of this indicator decreases the sharpness of the end- 
 point. An amount just sufficient to give a faint yellow color to 
 the alkaline solution should be used. The solution should be kept 
 in a bottle provided with a stopper, through which a drawn-out 
 glass tube passes so that the indicator may be introduced by 
 drops. 
 
 Methyl orange is very largely used because it is comparatively 
 unaffected by carbonic acid as well as by sulphuretted hydrogen, 
 hydrocyanic, silicic, boric, arsenious, oleic, stearic, palmitic, and 
 carbolic acids, etc. The base in the salts of these acids may be 
 titrated without removal of the acid. It has found a very ex- 
 tended use in the standardization of acids by means of pure sodium 
 carbonate. In the presence of carbonic acid the change from 
 the yellow alkaline to the red acid color is somewhat gradual. It 
 is difficult to remember the exact shade corresponding to the 
 neutral point, especially as a slight reddish tint is imparted by 
 
254 VOLUMETRIC METHODS. 
 
 the carbon dioxide. It is therefore advisable to prepare a solu- 
 tion for comparison which is given this tint by using distilled 
 water saturated with carbon dioxide which has been freed from 
 mineral acid by passing through a solution of sodium bicarbonate. 
 The beaker containing this solution should be of the same size 
 and contain the same volume of water as the beaker in which 
 the titration is made. The same amount of indicator is also 
 added. At the end-point of the titration a color a little deeper 
 than that of the carbon dioxide solution must be obtained. The 
 indicator is most sensitive in a cold solution. The presence of 
 alcohol greatly decreases its sensitiveness. It cannot be used in 
 solutions containing nitrous acid, as this acid destroys the indi- 
 cator. If only small amounts are present, the titration may be 
 completed before the indicator is destroyed. If large amounts 
 are present, excess of standard alkali must be added, and the 
 solution titrated back with standard acid. 
 
 351. Phenolphthalein is a synthetic organic product which 
 may readily be purchased in a pure condition. It is dissolved in 
 90 to 95% alcohol, 1 or 2 grams per liter. It is advisable to first 
 distil the alcohol after adding a little caustic soda or potash, as 
 the commercial product is seldom free from organic acids. One 
 or two drops of this solution are used. In neutral or acid 
 solutions it is colorless, but the faintest excess of alkali gives a 
 sudden change to purple-red. Carbon dioxide must be entirely 
 absent, as in its presence the indicator reacts acid until it is con- 
 verted ^o the bicarbonate by the alkali. Small amounts of 
 carbonataBm presence of alkali hydrates may be titrated accu- 
 rately by means of phenolphthalein. Acid is added until the 
 pink color has disappeared. All of the alkali hydrate will then 
 be neutralized, and the alkali carbonate converted into bicarbo- 
 nate. On adding methyl orange and standard acid until the 
 red acid color of this indicator is developed the bicarbonate will 
 be decomposed. From the amount of acid added for this purpose 
 the amount of carbonate originally present may be calculated. 
 The amount of carbonate must be small, as even the bicarbonate 
 gives a slight color to the phenolphthalein.* Ammonia must 
 
 * Kiister, Zeit. f. anorg. Ch., 13, 127. 
 
INDICATORS. 255 
 
 also be absent, as in its presence the end-point is not sharp. Phe- 
 nolphthalein is especially valuable for the titration of weak acids, 
 especially the organic acids. Many of these acids being insoluble 
 in water are dissolved in alcohol and then titrated. This does 
 not seem to be quite as accurate as titration in water solution. 
 A strong base must always be used with phenolphthalein. As 
 water solutions of caustic soda and potash always contain carbon 
 dioxide, an alcoholic solution of the latter is sometimes used, and 
 more frequently a solution of barium hydrate. 
 
 352. Best Acid for General Use. As has already been said, 
 the acidity of a solution depends only on the amount of replace- 
 able hydrogen present in a given volume, irrespective of the 
 nature of the acid radicle. As has been shown, the acid value 
 of the different hydrogen atoms differs when two or more are 
 present in the same molecule, so that the acidity of such a solu- 
 tion may differ widely with different indicators. For general use, 
 therefore, a strong acid should be chosen so that the strength 
 of the solution may be the same with all indicators. A stable 
 non-volatile acid is also desirable so that a solution once stand- 
 ardized may be permanent. These considerations limit us to 
 the three commonly used strong acids, hydrochloric, sulphuric, 
 and nitric. Of these three only the first two have found any 
 very general use. The standardization of nitric acid offers some 
 difficulties 'because it forms no insoluble weighable salt. It is 
 also an oxidizing agent, so that titrations may be disturbed by 
 secondary reactions. On the other hand, both sulphuric and 
 hydrochloric acids are very stable and may also be precipitated 
 and weighed with great accuracy. Choice must therefore be made 
 between these two acids for a carefully standardized perma- 
 nent acid to be used in the standardization of alkalies and other 
 acids. For general use hydrochloric acid offers several decided 
 advantages over sulphuric acid. It is appreciably stronger, so 
 that the end-point with most indicators is considerably sharper 
 than with sulphuric acid. This is partly due to the fact that there 
 are two atoms of hydrogen in the sulphuric acid molecule, so 
 that when the solution is partly neutralized an acid sulphate is 
 formed whose hydrogen atom is not so strongly acid as the hydro- 
 gen atom first displaced. The use of sulphuric acid is also lin> 
 
256 VOLUMETRIC METHODS. 
 
 ited by the fact that it forms insoluble sulphates with the alkaline- 
 earth metals. The gravimetric standardization of hydrochloric 
 acid by precipitation as silver chloride is subject to fewer sources 
 of error than the standardization of sulphuric acid by precipita- 
 tion as barium sulphate. 
 
 353. Most Desirable Strength of Standard Acids. Various 
 strengths of acid are in use among chemists, many using tenth- or 
 fifth-normal acids, while among commercial chemists the use of 
 half- or full-normal solutions is quite common. The use of such 
 concentrated solutions is defended by the argument that a sharp 
 end-point may always be obtained with one drop of the acid. 
 This is advantageous in commercial laboratories, where titrations 
 must be made rapidly, and as large amounts of material must be 
 weighed out to use a proper volume of acid, the weighings may 
 be more quickly made. This is evident from the consideration 
 that an error of 5 mg. on 5 grams is the same percentage as an 
 error of 1 mg. on 1 gram. As methyl orange must frequently 
 be used in the presence of carbon dioxide, the strong acid gives 
 a sharp end-point even with one drop of the standard solution. 
 
 The use of the more dilute solutions is defended by the argu- 
 ment that a larger volume of the solution must be used for titrat- 
 ing a given weight of material, and therefore the reading of the 
 Volume on the burette is more accurate. The indefinite dilution 
 of the solution does not lead to greater and greater accuracy, 
 since a point is soon reached where a greater and greater volume 
 of the solution must be taken to give a distinct end-point with the 
 indicator. This limit is usually reached with fifth- or tenth-normal 
 solutions. With moderate care a distinct end-point may be 
 obtained even with methyl orange in the presence of carbon 
 dioxide by means of one drop of fifth-normal acid. Another 
 decided advantage of the fifth-normal acid is found in the mod- 
 erately small coefficient of expansion with change in temperature, 
 which with stronger acids becomes quite considerable. In gen- 
 eral, therefore, more accurate work can be done with the more 
 dilute acid, while time is saved at the expense of accuracy by the 
 use of strong acids. The final decision must therefore be made 
 on the requirements of the work at hand. For the exercises 
 given in this book fifth-normal solutions will be uniformly used. 
 
PROBLEM. 257 
 
 PROBLEM. 
 
 Problem 4. Calculate the number of grams of impure potassium car- 
 bonate which must be weighed out, so that when the solution is titrated 
 with N/5 acid, the number of cubic centimeters used shall be equal to 
 the per cent of potassium carbonate. Calculate the strength of acid in 
 terms of normal to be made up, so that if 1 gram of impure potassium car- 
 bonate is weighed out, the per cent of potassium carbonate will be equal 
 to the number of cubic centimeters used multiplied by two. 
 
CHAPTER XXI. 
 
 STANDARD ACIDS. 
 STANDARDIZATION. 
 
 ONE of the most important points in volumetric work is the 
 accurate standardization of the solutions. No subsequent care 
 can overcome an error in this part of the work. Although many 
 excellent methods of standardizing acids have been proposed, 
 considerable skill and experience are necessary in order to secure 
 absolutely reliable results. 
 
 354- Sodium Carbonate. One of the oldest as well as most 
 reliable methods of standardization is by means of sodium car- 
 bonate. It may be purchased in a high state of purity, so that 
 it frequently needs only to be heated to completely dehydrate 
 it and decompose any bicarbonate which may be present. For 
 this purpose it need not be heated over 300.* It is better to heat 
 to this temperature than, as has long been the custom, to heat 
 to dull redness and below the fusing-point of the carbonate, as 
 small amounts of the carbonate are apt to be converted into the 
 hydroxide by this treatment. 
 
 The carbonate must first be tested for impurities. It should 
 dissolve completely in water. A portion of 2 or 3 grams is tested 
 for chlorides by dissolving in water, acidifying with dilute nitric 
 acid, and adding silver nitrate. Only a slight opalescence is per- 
 missible. Another 2- or 3-gram portion is dissolved in water, 
 acidified with hydrochloric acid, and tested for sulphuric acid by 
 means of barium chloride. 
 
 A very pure sodium carbonate may be obtained by heating 
 PURE SODIUM BICARBONATE to 270-300 for one-half to one hour. 
 The bicarbonate must be tested for impurities by the methods 
 
 * Lunge, Zeit. f. angew. Chem., 1897, 522. 
 
 258 
 
STANDARD ACIDS. 259 
 
 given for the carbonate. Small amounts of chlorides and sulphates 
 may be washed out by repeated treatment with small amounts of 
 cold distilled water. The bicarbonate may also be recrystallized. 
 Distilled water heated to 80 in a Jena beaker is saturated with 
 the bicarbonate by adding small amounts and stirring. The 
 saturated solution is filtered through a folded filter in a hot-water 
 funnel, and the solution cooled and vigorously stirred as the salt 
 crystallizes out. It is filtered off in a funnel closed with a 
 platinum cone or porcelain plate, washed with small portions of 
 cold water and dried in a porcelain dish. Impure sodium carbon- 
 ate may be purified by making a cold saturated solution, filtering 
 if necessary, and saturating with carbon dioxide which has been 
 washed by passing through a solution of sodium bicarbonate. 
 The precipitated bicarbonate is filtered off and washed as already 
 directed. It is advisable to preserve the bicarbonate as such 
 and convert suitable portions into carbonate immediately before 
 weighing it out for use. Better results will also be obtained if 
 separate small portions are weighed out for each titration instead 
 of making a standard solution, portions of which are measured 
 out. The errors of the volumetric apparatus are thus avoided. 
 
 355. Crystallized Oxalic Acid. Extended use has also been 
 made of crystallized oxalic acid. A solution of an alkali, such as 
 caustic soda or potash, is first standardized by means of the pure 
 oxalic acid. The acid solution to be standardized is then care- 
 fully compared with the standardized alkaline solution. This 
 method has the disadvantage that the errors of two titrations are 
 introduced into the standardization. Carbon dioxide must also be 
 absent as litmus or phenolphthalein must be used as the indicator. 
 
 Two difficulties are met with in preparing pure oxalic acid 
 the removal of traces of the alkalies and the drying of the crystal- 
 lized product so as not to lose water of crystallization. The 
 first object is attained by dissolving the commercial article in a 
 mixture of equal parts of alcohol and ether, in which the alkali 
 oxalates as well as other impurities are insoluble. After filtering 
 the solution, the bulk of the alcohol and ether are distilled off, 
 water is added, and the distillation continued. Finally the 
 water solution is transferred to a porcelain dish and kept near 
 the boiling-point until the last traces of the alcohol and ether 88 
 
260 VOLUMETRIC METHODS. 
 
 well as any ethereal salts which may have been formed are vola- 
 tilized. The steam will then have no odor. If the oxalic acid 
 begins to crystallize out of the hot solution during this process, 
 small quantities of distilled water are added from time to time. 
 When the volatile matter has been expelled, the solution is allowed 
 to cool with constant stirring. The crystals are filtered off in the 
 usual manner and washed with small quantities of cold distilled 
 water. They are then spread out on unglazed porcelain and 
 exposed to the air until dry. The material is preserved in glass- 
 stoppered bottles. The crystals must not be exposed to the air 
 after they are dry, as shown by their non-adherence to a 
 clean dry glass surface. The crystals then have the formula 
 H 2 C 2 4 .2H 2 0. 
 
 356. Anhydrous Oxalic Acid. To obviate the uncertainty of 
 the degree of hydration, anhydrous oxalic acid has been somewhat 
 used. Drying at 60 to 80 for several hours is said to secure an 
 anhydrous product. At 100 considerable oxalic acid is volatil- 
 ized. As the anhydrous material is quite hydroscopic, it must be 
 cooled in a desiccator and weighed as soon as cool. The pure 
 recrystallized product prepared as already directed must be used 
 for this purpose. Solutions of oxalic acid are not entirely stable, 
 especially if dilute. For standardizing purposes it is therefore 
 advisable to weigh out convenient amounts, and, after drying 
 and weighing, to dissolve in water and titrate immediately. 
 
 357. Potassium Tetroxalate. In place of the crystallized dehy- 
 drated oxalic acid, the use of crystallized potassium tetroxalate 
 KHC 2 4 .H 2 C 2 4 .2H 2 has been recommended. This salt may 
 easily be prepared by making a saturated solution of oxalic acid 
 and filtering if necessary. One-fourth of this solution is neutral- 
 ized with pure potassium carbonate and added with stirring to 
 the remainder of the oxalic acid solution. The crystals which 
 separate out are recrystallized from hot water several times and 
 dried as directed for crystallized oxalic acid. The salt is pre- 
 
 * served in well-stoppered bottles. 
 
 358. Potassium Bichromate. The suggestion of Richter * to 
 use pure potassium dichromate which may be titrated against a 
 
 *2eit.f.anal.Ch..21.205 
 
STANDARD ACIDS. 261 
 
 solution of caustic potash, using phenolphthalein as the indicator, 
 would seem to be of value. Potassium dichromate may readily 
 be prepared pure and anhydrous. 
 
 359. Gravimetric Standardization. The precipitation and 
 gravimetric estimation of the acid radicle offers a most excellent 
 method of standardizing acids, especially sulphuric and hydro- 
 chloric. The details of the precipitation and weighing of sul- 
 phuric acid as barium sulphate, and of hydrochloric acid as silver 
 chloride, have already been given. A very simple gravimetric 
 method consists in evaporating to dryness in a weighed platinum 
 dish a measured volume of the acid solution which has been 
 neutralized with ammonia. Both ammonium chloride and ammo- 
 nium sulphate may be dried at 100 without appreciable volatili- 
 zation. Very pure ammonia may be obtained for this purpose 
 by distilling the concentrated solution and absorbing the gas in 
 distilled water. Test-tubes may be used for this purpose, and 
 the distilled ammonia should be used immediately. The weighed 
 ammonium salt may be still further tested for impurities by vola- 
 tilizing the weighed salt by gently heating the dish with the 
 Bunsen burner and weighing the residue. 
 
 360. Standardization by Specific Gravity. Standard solutions 
 of acids, especially sulphuric acid, may be made with almost as 
 high a degree of accuracy as by the methods already given, by 
 carefully taking the specific gravity of the concentrated acid and 
 diluting a measured or weighed amount to a definite volume. 
 The C. P. concentrated sulphuric acid should first be diluted to 
 about 30%. Acid of this concentration is not hydroscopic, and 
 the change in specific gravity with change in concentration is 
 greater than with the stronger acid. The specific gravity is 
 carefully taken at the standard temperature, 15 C., by means 
 of a Westphal balance or carefully calibrated pycnometer. The 
 greatest care should be taken to bring the temperature to exactly 
 15 C. with a tested thermometer. If the acid is to be measured 
 out it must be kept at this temperature as carefully as when the 
 specific gravity is being taken. This difficulty may be obviated 
 by weighing the acid. This method has the additional advantage 
 that the weight can be taken far more accurately than the volume, 
 especially if it is corrected for air displacement. The weighing 
 
262 VOLUMETRIC METHODS. 
 
 need not be made closer than a few milligrams, as an error of 
 10 mg. is negligible. The calculation is also very simple. The 
 number of grams of acid required per liter for the solution to be 
 made is divided by th^ percentage given in the table for the spe- 
 cific gravity found. The quotient multiplied by 100 is the num- 
 ber of grams of the concentrated acid to be weighed out and 
 diluted to a liter. 
 
 As the measuring out of concentrated hydrochloric acid is 
 accompanied by much more uncertainty, it is advisable to measure 
 out a little more of this acid than the calculated amount, so that 
 after standardization it may be diluted to the exact strength. 
 A solution of hydrochloric acid of approximately constant strength 
 may be obtained by boiling either the dilute or concentrated 
 acid. Equilibrium is soon reached so that, when boiling at about 
 760-mm. pressure, both acid and water are volatilized in equal 
 amount from the residue, which contains 20.2% of hydrochloric 
 acid and has a specific gravity of 1.10.* 180 grams of this acid 
 diluted to 1 liter give an acid slightly stronger than normal. 
 
 361. Dilution of Acid to Exact Strength. If the sulphuric 
 acid is not made by diluting a strong acid whose specific gravity 
 has been very carefully taken, it should also be made so as to be 
 somewhat stronger than the acid required. At least 1 liters 
 should be made, placed in a bottle of suitable size and well 
 mixed by shaking. After carefully standardizing the acid by 
 two of the methods already given, it should be diluted to the 
 exact strength. Perhaps the simplest method of calculating the 
 results of each standardization is to find the number of grams of 
 acid present per liter. The average of the standardizations, 
 which agree within at least 0.2%, is taken and divided into the 
 number of grams per liter required. For normal hydrochloric 
 this will be 36.46 grams and for normal sulphuric 49.04 grams. 
 The quotient multiplied by 1000 gives the number of cubic centi- 
 meters to be diluted to 1 liter. This number subtracted from 1000 
 gives the number of cubic centimeters of water to be measured 
 out into the liter-flask, which is then filled to the mark with the 
 acid to be diluted. The strength of the diluted acid is verified 
 
 * Roscoe and Dittmar, Jour. Chem. Soc., XII, 128, 1860. 
 
STANDARD ACIDS. 263 
 
 by one of the methods already used. This is especially necessary 
 if the amount of water added per liter is large. Frequently the 
 most satisfactory method of standardization is by the use of 
 sodium carbonate, this being a volumetric method, and is carried 
 out under the same conditions that will be met in regular work. 
 
 '" : ' ' ' -T u - : ' 
 
 EXERCISE 48. 
 Preparation of Standard N/s Hydrochloric Acid. 
 
 To 75 c.c. of concentrated hydrochloric acid an equal volume of water 
 is added and the solution boiled in a beaker or a porcelain dish for a few 
 minutes. About 36 grams of this acid are weighed out and transferred 
 to the liter flask, which should be clean, but need not be dry. The beaker 
 or other vessel is rinsed out into the flask. Distilled water is added until 
 the flask is filled to the mark. The solution is poured into a large bottle,* 
 which, if not dry, should first be rinsed with a little of the acid. The flask 
 is rinsed with a little distilled water, and a second liter of acid made and 
 poured into the large bottle. The bottle is filled from a third liter of the 
 diluted acid. After thoroughly mixing the solution by shaking, the acid 
 is standardized by two of the following methods, duplicate determinations 
 being made by each method. 
 
 STANDARDIZATION. 
 
 362. First Method. By Evaporation with Ammonia. Measure out with 
 a burette or a calibrated pipette 50 c.c. of the acid into a weighed plati- 
 num dish. Fit a small flask or a test-tube with a glass delivery-tube bent 
 twice at right angles. Place 10 to 15 c.c. of strong ammonia in the flask 
 and distil off the ammonia, using a small flask or test-tube containing 15 
 to 20 c.c. distilled water as a receiver. The end of the delivery-tube should 
 dip into the distilled water. Neutralize the acid with this ammonia, evapo- 
 rate to dryness on the water-bath, and weigh the ammonium chloride. 
 Bring to constant weight by heating on the water-bath for fifteen to twenty 
 minutes and weighing. When the weight is constant volatilize the ammo- 
 nium chloride by gently heating the dish with the Bunsen burner, and 
 weigh again. Deduct this weight from the weight of the dish and the 
 residue dried on the water-bath to obtain the weight of the ammonium 
 chloride. 
 
 Calculate the weight of hydrochoric acid present in 50 c.c. by the 
 proportion, 
 
 Mol. wt. NH 4 C1 : Mol. wt. HC1 : : weight NH 4 C1 : weight HC1. 
 
 * The 2Hiter glass-stoppered bottles in which the concentrated C. P. acids 
 are purchased are very convenient for this purpose. 
 
264 VOLUMETRIC METHODS. 
 
 363. Second Method. By Titration with Sodium Carbonate. Place about 
 4J grams of pure sodium bicarbonate in a weighed platinum crucible. 
 Place it in a sand-bath so that the crucible is nearly immersed in the sand. 
 Insert a thermometer in the bicarbonate and heat the sand-bath until the 
 thermometer registers 270 to 300 for one-half to one hour. If an air-bath 
 is at hand which can be raised to this temperature, the bicarbonate may 
 be heated in it for the same length of time. It may then be placed on a 
 weighed watch-crystal. The carbonate is cooled in the desiccator and 
 weighed. The heating is repeated for half-hour periods until constant 
 weight is obtained. The weight of the carbonate is then brought to exactly 
 2.650 grams by taking out small portions with a spatula. Transfer the 
 carbonate to a beaker, rinse the crucible with water, dissolve the material 
 in water, and transfer to a 250-c.c. flask. After diluting to the mark and 
 shaking thoroughly, fill a burette with the solution and titrate against the 
 hydrochloric acid solution, using methyl orange as the indicator. The titra- 
 tion should be made the same day the carbonate is dissolved, as the solution is 
 not entirely stable because of action on the glass. 
 
 Measure out 25 c.c. of the alkaline solution into a 100-c.c. beaker, and 
 add one drop of the indicator. 50 c.c. of distilled water are placed in a similar 
 beaker, one drop of methyl orange added, and the solution saturated with 
 carbon dioxide which has been passed through a solution of sodium bicar- 
 bonate. To the beaker containing the alkali hydrochloric acid is added 
 until, when compared with the neutral solution, a pink tinge is noticeable. 
 A drop of alkali is then added, and if the neutral tint is not restored, the 
 solution is made alkaline and again neutralized by cautious additions of the 
 acid. This process is repeated until no difficulty is experienced in so closely 
 judging the acid tint that one drop of the alkali will restore the neutral 
 tint. When a satisfactory end-point has been obtained both burettes are 
 carefully read. The titration is repeated until duplicates agreeing within 
 less than 0.1 c.c. are obtained. 
 
 When skill has been attained in making the titration, more accurate 
 results can be obtained by weighing out portions of the carbonate just suffi- 
 cient for one titration. A considerable amount of the bicarbonate is con- 
 verted into carbonate and brought to constant weight. Portions weighing 
 from 0.4 gram to not over .5265 gram are weighed out and transferred to 
 beakers. After dissolving in about 50 c.c. of distilled water and adding a 
 drop of methyl orange the solutions are titrated with the hydrochloric acid. 
 
 Calculation. By dividing the weight of sodium carbonate by the num- 
 ber of cubic centimeters of acid used in each case the weight of sodium 
 carbonate per cubic centimeter will be obtained, which for an exactly N/5 
 normal solution is .01061 gram. The amount found per cubic centimeter 
 should be divided into .01061, and the quotient multiplied by 1000 to 
 find the number of cubic centimeters to be diluted to one liter to make 
 the acid exactly fifth-normal. 
 
STANDARD ACIDS. 265 
 
 364. Third Method. By Determination of Chlorine as Silver Chloride. 
 Prepare two Gooch crucibles and dry on the hot plate. Measure out two 
 25-c.c. portions of the acid into Erlenmeyer flasks and dilute to 200 c.c. 
 Add silver nitrate solution with constant agitation until all the chlorine 
 is precipitated. Heat nearly to boiling, and shake vigorously until the 
 supernatant liquid is clear. Decant the clear liquid through the Gooch 
 crucibles, wash two or three times by decantation with hot water, transfer 
 the precipitates to the crucibles, and wash with hot water until free from 
 silver. Dry on the hot plate and weigh. If the duplicates differ from 
 each other by more than J mg., repeat the determination, using the same 
 crucibles without removing the precipitates. 
 
 Calculation and Dilution of Acid. From the average weight of silver 
 chloride obtained from 25 c.c. ol arid the strength of the acid is obtained 
 by the following proportion : 
 
 Mol. wt. AgCl : mol. wt. HCl : weight AgCl : X, 
 
 where X is equal to the number ol grams of hydrochloric acid in 25 c.c. 
 This number divided into 7.292 and the quotient multiplied by 25 
 
 l~7 292 
 
 (1000) gives the number of cubic centimeters to be diluted to 1000 
 
 [_ 40A 
 
 to give a fifth-normal solution. This number subtracted from 1000 gives 
 the number of cubic centimeters of water required. If the liter-flask is 
 wet, it should be rinsed two or three times with a little of the acid to be 
 diluted. The necessary amount of water is carefully measured out with a 
 burette and allowed to flow into the flask, which is then filled to the mark 
 with the acid and thoroughly shaken. The diluted acid is emptied into a 
 bottle capable of holding all of the acid to be diluted. If it is not dry, it is 
 rinsed with a little of the diluted acid. The remainder of the acid is diluted 
 in the same manner and poured into the large bottle. The strength of the 
 diluted acid must be verified by one of the methods already used. This 
 is especially necessary if the amount of water added is large. 
 
 DETERMINATION OF SODIUM HYDROXIDE, CARBON- 
 ATE, AND BICARBONATE. 
 
 These three compounds of sodium evidently cannot occur 
 together, because the hydroxide and bicarbonate would react to 
 form the normal carbonate. We have therefore to consider the 
 determination of sodium hydroxide and carbonate when occurring 
 together, as well as the analysis of mixtures of the bicarbonate and 
 the carbonate. 
 
 365. Titration of Carbonic Acid. One method of analyzing 
 the first combination has already been outlined in the discussion 
 
266 VOLUMETRIC METHODS. 
 
 of the properties of indicators. Using phenolphthalein as the 
 indicator, all of the hydroxide will be neutralized by standard 
 acid, while the carbonate will be converted into bicarbonate. 
 On adding methyl orange and continuing the addition of acid 
 until the solution is pink the bicarbonate formed will be decom- 
 posed. The amount of acid added during the second part of the 
 titration will be half that required to completely decompose the 
 carbonate. The difference between the two amounts of acid 
 used will be the amount necessary to neutralize the sodium hydrox- 
 ide. This method will give correct results only if the amount of 
 carbonate is small and the solution is dilute and cold. The escape 
 of carbon dioxide because of a local excess of acid must be abso- 
 lutely prevented by vigorous stirring. The method is well adapted 
 to the analysis of commercial caustic soda. 
 
 366. Precipitation of Carbonic Acid. If a large amount of 
 sodium carbonate is present, the total amount of alkali present is 
 determined by titration with standard acid, using methyl orange 
 as the indicator. From another portion the carbon dioxide is 
 precipitated by the addition of BARIUM CHLORIDE solution. The 
 alkali is in this manner converted into chloride, and on titrating 
 the solution only the alkali hydroxide remains to react with the 
 acid. Any excess of barium chloride will remain as such or react 
 with the sodium hydroxide, giving sodium chloride and barium 
 hydroxide. It has been shown, however, that the excess of 
 barium chloride must be small, as otherwise the percentage of 
 caustic soda found will be low. Using phenolphthalein as the 
 indicator, the alkali hydroxide may be titrated in the presence of 
 the barium carbonate, if care is exercised to stir the solution 
 thoroughly so as not to have, at any time, an excess of acid pres- 
 ent in any part of the solution. The acid reaction of the solution 
 is not permanent because of the slight solubility of the barium 
 carbonate. The titration is therefore conducted quite rapidly 
 and a momentary disappearance of the red color is taken as the 
 end-point. 
 
 The solution, after the addition of the barium chloride, may 
 also be diluted to a definite volume, and after allowing the pre- 
 cipitate to settle, a measured portion of the clear liquid may be 
 withdrawn and titrated. The amount of acid necessary to neu- 
 
TIT RAT ION OF THE ALKALIES. 267 
 
 tralize the carbonate present is found by subtracting the amount 
 of acid necessary to neutralize the sodium hydroxide from the 
 amount used in titrating the total alkali. 
 
 367. Determination of Bicarbonates. If a mixture of sodium 
 carbonate and bicarbonate is to be analyzed, the total alkali may 
 be found by tit ration with acid, using methyl orange as the indi- 
 cator. As the precipitation of the carbon dioxide by barium 
 chloride would tend to produce free acid according to the equation 
 
 2NaHC0 3 +2BaCl 2 =2BaC0 3 +2NaCl +2HC1, 
 
 thus preventing the completion of the reaction, a measured 
 excess of standard alkali solution must first be added. After 
 the precipitation of the carbon dioxide by the addition of 
 barium chloride the alkali present is titrated with standard acid. 
 The difference between the amount of alkali added and that 
 found after the precipitation of the carbon dioxide is that neces- 
 sary to convert the bicarbonate present into carbonate or to neu- 
 tralize the equivalent amount of hydrochloric acid liberated from 
 the barium chloride. The standard alkali used in this case may 
 be either ammonia or caustic soda. The former is disadvanta- 
 geous because of its disturbing effect on phenolphthalein. while 
 it is difficult to obtain either alkaline solution free from carbon 
 dioxide. Perhaps the simplest way of overcoming this difficulty 
 is to conduct a blank determination by precipitating the carbonate 
 from a measured volume of the alkaline solution and titrating as 
 in the actual determination. 
 
 EXERCISE 49. 
 
 Determination of Sodium Hydroxide, Sodium Carbonate, and Water in 
 
 Caustic Soda. 
 
 Weigh out in a weighing-bottle from 3 to 5 grams of caustic soda. Do 
 not take the sample from the top of the bottle, where it has been exposed 
 to the moisture and carbon dioxide of the air. Transfer to the weighing- 
 bottle with as little exposure as possible and stopper quickly. When 
 weighed, transfer to a beaker, rinse the bottle with water, and dissolve the 
 soda. Transfer to a 250-c.c. flask and dilute to the mark with boiled water. 
 
 368. First Method. Titration of the Carbonic Acid. Take out 23 c.c. 
 with a pipette and dilute to about 100 c.c. with boiled water which has been 
 thoroughly cooled. Titrate, while stirring vigorously, with N /5 acid after add- 
 ing 2 or 3 drops of phenolphthalein until the pink color has just disappeared. 
 
268 VOLUMETRIC METHODS. 
 
 Add I or 2 drops of methyl orange and continue the titration until a faint 
 pink appears. Repeat until duplicates are obtained. Twice the amount 
 of acid used in the methyl-orange tritation is the amount necessary to 
 neutralize the sodium carbonate. 1 c.c. of N/5 acid equals .01061 gram 
 Na 2 C0 3 . The difference between the amounts of acid used in the phenol- 
 phthalein and the methyl orange titrations is the amount necessary to neu- 
 tralize the sodium hydroxide. 1 c.c. of N/5 acid equals .008012 gram NaOH. 
 Calculate the amount of Na 2 CO 3 and of NaOH present in 250 c.c. The 
 difference between the sum of these weights and the weight of the caustic 
 soda gives the weight of water present. Calculate the per cent of each 
 constituent. 
 
 369. Second Method. Precipitation of the Carbonic Acid. Measure out 
 25 c.c. of the soda solution with a pipette into a beaker, add 1 or 2 drops of 
 methyl orange, and titrate to faint pink with N/5 acid. 
 
 Measure out 50 c.c. into a 100-c.c. flask and add with shaking a little 
 dilute solution of barium chloride drop by drop as long as a white precipi- 
 tate forms. Dilute to the mark with boiled and cooled .water, shake thor- 
 oughly, and allow to settle. Measure out 50 c.c. 'and titrate with the N/5 
 acid, using phenolphthalein as the indicator. Repeat both titrations until 
 duplicates are obtained. The second titration gives the amount of acid 
 necessary to neutralize the caustic soda, while the difference between the 
 first and second gives the amount of acid necessary to neutralize the sodium 
 carbonate present in 50 c.c. of the solution. Calculate the percentage of 
 NaOH, Na 2 C0 3 , and H 2 O as directed under the first method. 
 
 DETERMINATION OF HARDNESS OF WATER. 
 
 370. Temporary Hardness or Alkalinity. The hardness of 
 water is caused by the presence of salts of calcium and magnesium 
 in solution. If these metals are in solution as bicarbonates, they 
 may be precipitated as carbonates by boiling the water, the excess 
 of the carbon dioxide being expelled. The hardness of water 
 which may be removed in this manner is called temporary hard- 
 ness. The amount of calcium and magnesium held in solution in 
 this manner may be ascertained by titrating the solution with a 
 standard acid, using methyl orange or erythrosene and chloroform 
 as the indicator. The result is given in parts- of calcium carbon- 
 ate per 100,000, giving the so-called degrees of hardness. As the 
 bicarbonates of calcium and magnesium react alkaline to methyl 
 orange and some other indicators, this titration also gives the 
 alkalinity of the water. If sodium or potassium carbonates are 
 present, they will also react alkaline, but a correction for this 
 error will be obtained in determining permanent hardness. 
 
I 
 
 DETERMINATION OF HARDNESS OF WATER. 269 
 
 371. Permanent Hardness or Incrustants. The presence of 
 chlorides and sulphates of calcium and magnesium produce per- 
 manent hardness, so called because these salts cannot be removed 
 by boiling the water. On adding excess of sodium carbonate, and 
 Waporating to dryness, the sulphates and chlorides of calcium and 
 magnesium are converted into the carbonates of these two metals 
 and the sulphate and chloride of sodium. The bicarbonates of 
 calcium and magnesium are also decomposed. On filtering and 
 washing, the carbonates of calcium and magnesium, being insol- 
 uble, remain on the paper, while the excess of the sodium car- 
 bonate as well as the sodium chloride and sulphate pass into solu- 
 tion. The excess of sodium carbonate having been ascertained by 
 titration, the amount needed to decompose the chlorides and sul- 
 phates of calcium and magnesium is found ly difference. If the 
 so-called soda reagent is used, the magnesium will be precipitated 
 as hydroxide. The amount of permanent hardness is also calcu- 
 lated as parts of calcium carbonate per 100,000 or per million. 
 As -the principal salt producing permanent hardness is calcium 
 sulphate, which tends to produce a very hard scale in boilers, the 
 term incrustants is also used to designate these constituents of the 
 water. If the alkalinity of the filtrate is greater than that due to 
 the sodium carbonate added, alkali carbonates must have been 
 present in the water. Permanent hardness is 'then absent, and 
 the amount of alkali carbonates present computed as calcium car- 
 bonate must be subtracted f rom the temporary hardness already 
 found. 
 
 EXERCISE 50. 
 Determination GJ. flardness of Water, Temporary and Permanent. 
 
 Make an N/50 solution of hydrochloric acid by measuring out 100 c.c. 
 of the IS 5 acid into a liter flask and diluting to the mark. Make an N/50 
 solution of sodium carbonate by diluting 100 c.c. of the N/5 alkali to a liter, 
 or weigh out 1 .061 grams of the pure dry salt, dissolve in water, and make 
 up to a liter. 
 
 372. Temporary Hardness or Alkalinity. Measure out 100 c.c. of the 
 water to be tested into a porcelain dish and add one drop of methyl orange. 
 Measure out 100 c.c. of distilled water into a similar porcelain dish and add 
 one drop of the indicator. Add the N/50 acid drop by drop untj] a faint 
 pink color is obtained. Titrate the water in the other dish with the acid 
 until the color matches that in the dish containing the distilled water. From 
 tae amount of acid used, less the amount added to the distilled water (.1 to 
 
270 VOLUMETRIC METHODS. 
 
 .3 c.c.), the alkalinity or hardness is calculated in parts per 100,000 or parts 
 per million. 
 
 Erythrosene and chloroform may also be used as the indicator in this 
 titration. The end-point with this indicator is sharper and its use is espe- 
 cially to be advised when alum is likely to be present in the water. O.I 
 gram of the erythrosene is dissolved in 1000 c.c. of water. 100 c.c. of the 
 water to be tested is placed in a 250-c.c. glass-stoppered bottle; 2.5 c.c. of 
 the indicator* and 5 c.c. of chloroform are added. The bottle is shaken 
 thoroughly, so that the chloroform forms an emulsion. The standard acid 
 is added and the shaking continued until the color is discharged. The 
 alkalinity is calculated from the amount of acid used. 
 
 373. Permanent Hardness or Incrustants. 100 c.c. of the water is meas- 
 ured out into a porcelain evaporating-dish. Add a measured amount of 
 the N/50 sodium carbonate solution sufficient to make the water strongly 
 alkaline. Evaporate to dryness, take up with a small amount of water, 
 filter, and wash the dish and filter-paper until free from alkali. Titrate 
 the excess of the alkali with N/50 acid, using methyl orange as the indicator- 
 The difference between the amount of alkali added and that titrated back 
 is the amount used in precipitating the calcium and magnesium, which 
 produced the permanent hardness. The calculation is made in the same 
 manner as that of temporary hardness. If the amount of alkali titrated 
 back should be greater than the amount added, the water contains alkali 
 carbonates which have been reckoned as temporary hardness. The amount 
 of excess should therefore be deducted from the temporary hardness. Per- 
 manent hardness is absent. 
 
 The so-called soda reagent gives somewhat more accurate results because 
 the salts of calcium and magnesium are rendered less soluble by this reagent. 
 It is composed of equal parts of N/10 sodium carbonate and sodium hydrox- 
 ide. It may be prepared by weighing out 2.653 grams of pure sodium 
 carbonate or measuring out 250 c.c. of N/5 sodium carbonate solution and 
 250 c.c. of N/5 sodium hydroxide solution (page 282) , and diluting to one 
 liter. 200 c.c. of the water to be tested is measured into a Jena Erlen- 
 meyer flask. After boiling 10 minutes to expel free carbonic acid, 25 c.c. 
 of the soda reagent is added. It is again boiled down to a volume of 100 
 c.c., cooled and rinsed into a 200-c.c. graduated flask and the volume 
 made up to 200 c.c. with boiled distilled water and thoroughly mixed by 
 shaking the flask. The solution is filtered through a dry paper, and after 
 rejecting the first 50 c.c., 100 c.c. is titrated with the N/50 acid, using ery- 
 throsine as the indicator. The calculation is made in the same manner 
 as before except that the 25 c.c. of tenth, normal soda reagent must be 
 considered equal to 125 c.c. of the N/50 acid. 
 
 When' the permanent hardness is small, better results are obtained by 
 determining total hardness by means of the soda reagent, and getting per- 
 
 *When erythrosene is referred to as the indicator, the .01% solution of the 
 sodium salt of erythrosene is implied. 
 
STANDARD SULPHURIC ACID. 
 
 271 
 
 manent hardness by difference. For this purpose the exact amount of 
 acid used in titrating the alkalinity is added to 100 c.c. of the water being 
 tested, thus converting the temporary hardness to permanent hardness, 
 which is determined by the method already described. 
 
 EXERCISE 51. 
 Standard N/s Sulphuric Acid. 
 
 374. Specific Gravity. To 100 c.c. distilled water add 40 grams con- 
 centrated sulphuric acid. Cool to 15 C. and take the specific gravity with 
 the Westphal balance. In setting up this balance, the screw A is adjusted 
 so that the two points at C are exactly on a level when the thermometer D 
 
 FIG. 45. 
 
 is suspended in the air. By the thumb-screw B the height of the beam is 
 regulated so that the thermometer does net touch the bottom of the small 
 cylinder. The weights are of four sizes. The largest size hung on the hook 
 on the end of the beam gives units, while on any other space gives tenths, 
 the next size gives hundredths, the next size thousandths, and the smallest 
 ten-thousandths of specific gravity. The instrument in Fig. 45 reads 0.0885, 
 the largest size weight being absent. The sulphuric acid is poured into the 
 cylinder and the weights placed on the various divisions until the two points 
 at C are again exactly on a level. If the thermometer now reads 15, the 
 
272 VOLUMETRIC METHODS. 
 
 specific gravity is given by the weights on the beam. The percentage of 
 HjjSO^ corresponding to the specific gravity found, is obtained from the 
 table of specific gravity of sulphuric acid given on p. 523. The specific 
 gravity times the per cent of H 2 SO 4 present gives the number of grams of 
 HsSC^ per cubic centimeter. The number of cubic centimeters of acid 
 to be measured out to give 9808 grams of H 2 SO 4 is now computed. 
 The acid, still being at 15, is transferred to a burette, measured out into a 
 liter flask, diluted to the mark and thoroughly mixed. Another liter is 
 prepared in the same manner and mixed with the first in a 2-liter bottle.^ 
 Instead of measuring out the concentrated 'acid, it may be weighed out.' 
 For this purpose it need not be kept at 15. The amount to be weighed 
 out for one liter is found by dividing 9.808 by the per cent found in the 
 table and multiplying the quotient by 100. If this solution has been care- 
 fully prepared, it will be very near the calculated strength, 
 
 STANDARDIZATION. 
 
 First Method. Same as first method in Exercise 48. 
 Second Method. Same as second method in Exercise 48. 
 
 375. Third Method. Precipitation as Barium Sulphate. Measure out by 
 a burette 50 c.c. of the sulphuric acid into a beaker. Dilute to 400 c.c., 
 heat to boiling, and add barium chloride solution drop by drop while stirring 
 the solution vigorously until all of the sulphuric acid is precipitated. Digest 
 until the solution is clear. Wash by decantation, dry, ignite, and weigh. 
 Repeat the determination until duplicates are obtained agreeing within 
 .2%. Compute the weight of H 2 SO 4 , multiply by 20, and divide 9.808 by 
 the product. This gives the number of cubic centimeters of the acid to be 
 diluted to 1000 c.c. The acid is diluted as directed in Exercise 48. 
 
 DETERMINATION OF SPECIFIC GRAVITY. 
 
 The strength of acids as well as of a great variety of solutions 
 is commonly ascertained by a determination of the specific grav- 
 ity. This method is especially advantageous when dealing with 
 solutions of pure substances. 
 
 376. Specific Gravity may be defined as the ratio between the 
 weight of a given volume of a substance and the weight of an equal 
 volume of some standard substance. When dealing with liquids or 
 solids, water is almost invariably used as the standard of com- 
 parison. In terms of the metric system of weights and measures, 
 specific gravity may be defined as the ratio of the weight of a 
 cubic centimeter of a substance to the weight of a cubic centi- 
 meter of water. As the weight of the latter is 1 gram, specific 
 
DETERMINATION OF SPECIFIC GRAVITY. 273 
 
 gravity may be defined as the weight in grams of 1 cubic centi- 
 meter of a given substance. When using the metric system of 
 weights and measures, specific gravity is therefore identical with 
 density, which is defined as the weight of unit volume of a sub- 
 stance. 
 
 377. Standard Temperatures. With the English units of weights 
 and measures, specific gravity and density are not identical. The 
 identity is not exact even with the metric system, except when 
 the water is used at its maximum density (4 C.) and is weighed 
 in vacuo. Considerable confusion has arisen because in deter- 
 mi: ing specific gravity the weight of water in the air and at 
 other temperatures than 4 C. has been assumed to be 1 gram 
 per cubic centimeter. Even when the exact weight of the water 
 has been ascertained, confusion may arise if the conditions used 
 are not given. The temperature at which both the substance 
 and the water are weighed should always be given. This is ordi- 
 narily done by placing a fraction after the figure giving the specific 
 gravity. The numerator of the fraction gives the temperature of 
 the substance, while the denominator gives the temperature of 
 the water. 1.563^, for instance, means that a given volume of 
 the substance at 15 weighs 1.563 times as much as the same 
 volume of water at 4 C. If these weights have been reduced to 
 vacuo, we may also state that the density of the substance at 15 
 is 1.563; that is, the weight of 1 cubic centimeter is 1.563 grams. 
 For substances whose specific gravity is small, the correction 
 obtained by reducing the weight to vacuo is very small, being 
 about one-tenth per cent. The error introduced by considering 
 the weight of 1 c.c. of water at 15 to be 1 gram is .08%, while 
 at 20 the error is .17%, and at 25 it is .29%. As both of these 
 corrections are positive, the error introduced by neglecting both 
 of them is the sum of the two. The simplest and most conve- 
 nient method of determining specific gravity consists in weighing 
 both the substance and the water in the air at ordinary temper- 
 tures, such as 15 or 20. When the specific-gravity tables have 
 been calculated from determinations made in the same manner, 
 no error is introduced unless the volume of solutions are calcu- 
 lated from their weight or weights from volumes. Such calcula- 
 tions require true densities. 
 
274 VOLUMETRIC METHODS. 
 
 378. Pycnometers. The most reliable method of determining 
 specific gravity involves the use of some form of pycnometer. The 
 simplest form of this instrument consists of a small bottle closed 
 with a glass stopper containing a capillary opening through which 
 the excess of liquid may escape when inserting the stopper after 
 filling the bottle. It is first weighed empty and dry, again weighed 
 after filling with water, and finally the weight of the bottle filled 
 with the substance being tested is taken. The latter as well as 
 the water must be brought to the standard temperature before 
 filling the bottle. As most liquids transmit heat quite readily 
 and have a fairly high coefficient of expansion, there is consid- 
 erable difficulty in getting the pycnometer filled at the desired 
 temperature. After being filled the bottle may be placed in a 
 vessel of water which is maintained at the standard temperature. 
 If allowed to remain sufficiently long in the water, the bottle and 
 contents ultimately attain the same temperature as the water. 
 The time necessary for this purpose varies with the viscosity and 
 other properties of the liquid. To obviate uncertainty in this 
 respect, pycnometers are made with a thermometer ground so as 
 to serve as a stopper, the bulb being in the liquid while the stem 
 projects out of the bottle. A capillary tube is fused into the 
 side of the pycnometer and is provided with a hollow glass ground 
 cap. The liquid is cooled down below the standard temperature 
 before filling the bottle, which is then placed in water maintained 
 at this temperature. As the liquid expands the excess escapes 
 through the capillary tube. When the thermometer in the pyc- 
 nometer indicates the desired temperature, the excess of liquid on 
 the end of the capillary tube is removed and the cap put on. The 
 bottle is removed from the water and wiped clean and set aside 
 until the contents have reached room temperature, when it is 
 weighed. The weight of distilled water which just fills the bottle 
 at the standard temperature is obtained in the same manner. If 
 the true density is desired, the weight of the water is corrected 
 for air displacement, as explained on page 11, and the volume in 
 cubic centimeters calculated from the density of the water at the 
 temperature used. The weight of an equal volume of water in 
 grams at 4 C. in vacuo is numerically equal to the number of 
 cubic centimeters found by the calculation given. This weight, 
 
DETERMINATION OF SPECIFIC GRAVITY. 275 
 
 divided into the weight of the bottle full of the liquid being 
 tested, gives the density or the specific gravity compared with 
 water at 4 C. 
 
 379. The Westphal Balance. The manipulation of this instru- 
 ment is very simple, and the results are almost as accurate and 
 reliable as those obtained with the pycnometer. It consists of a 
 beam suspended on a knife-edge and so adjusted that when a glass 
 plummet is suspended from a hook at one end, the weight on the 
 other end is just counterbalanced, as indicated by the two points 
 at C, Fig. 45, which are then exactly on a level. If they are not, 
 the balance may be adjusted by means of the thumb -screw A. 
 As commonly made, the plummet weighs 15 grams and displaces 
 5 grams of water at the standard temperature. The arm from 
 which the plummet is suspended is divided into 10 equal divi- 
 sions, and a set of weights in the form of riders is provided, such 
 that the largest is just equal to the weight of the water displaced 
 by the plummet, generally 5 grams. If the plummet is suspended 
 in water, it will lose 5 grams in weight, which will te exactly 
 counterbalanced by the large weight when hung from the hook at 
 the end of the beam. This indicates a specific gravity of one. 
 When the plummet is suspended in other liquids, the large weight 
 indicates tenths when placed on other divisions of the beam. 
 Similarly, riders weighing 0.5, 0.05, and 0.005 gram indicate hun- 
 dredths, thousandths, and ten thousandths of specific gravity 
 when they are hung on the beam so as to counterbalance the 
 buoyant force of the liquid being tested on the glass plummet. 
 The temperature of the liquid is taken by means of a thermometer 
 within the plummet. 
 
 380. Hydrometers, or specific-gravity spindles, are instruments 
 the construction of which is based on the principle that the buoyant 
 force of a liquid is directly proportional to its density or specific 
 gravity. The hydrometer is essentially a long, closed, glass tube, 
 more or less enlarged at the lower end and so weighted with shot 
 or mercury that it will assume a perpendicular position in a liquid 
 in which it does not entirely sink. As the weight of the instru- 
 ment is constant, the depth to which the hydrometer sinks in a 
 given liquid may be used as an index of the specific gravity of 
 the liquid. A paper scale is inserted in the stem of the instru- 
 
276 VOLUMETRIC METHODS. 
 
 merit and fixed in such a position that the specific gravity may 
 be read directly on the scale at the surface of the liquid. A 
 thermometer is sometimes made a part of the instrument, so that 
 the temperature at which the readings are taken may be noted. 
 These instruments are very convenient and rapid in use, but are 
 not well adapted for accurate work. Specific gravity is often 
 designated by degrees of one of the several scales which have 
 been devised and which are almost invariably made use of in the 
 construction and graduation of hydrometers. The degrees 
 Twaddell are the simplest and most scientific of these scales. 
 On this scale a degree corresponds to .005 specific gravity above 
 unity. 7 Twaddell is therefore equal to 1.035 sp. gr.; 30 is 
 equal to 1.150, etc. The spindle is usually made quite large, with 
 a very small stem, so that determinations of specific gravity may 
 be made with very great accuracy. The entire scale of 150 is 
 usually divided equally among 6 spindles. This instrument is 
 very generally used in Great Britain. 
 
 381. The Baume Scale is generally used in other countries. This 
 instrument was originally graduated as follows: The point to 
 which the spindle sank in water at 17J C. was marked 0, and 
 the point to which it sank in a 10 per cent solution of common 
 salt was marked 10. This space was divided into 10 equal divi- 
 sions, and the remainder of the stem marked off in spaces of the 
 same length. This gave the scale for liquids heavier than water. 
 For liquids lighter than water, the point to which the spindle sank 
 in a solution of 1 part of salt in 9 parts of water was marked 0, 
 and the point to which it sank in water was marked 10. One- 
 tenth of this space was taken as a degree, and the remainder of 
 the scale divided into degrees of this size. This method gave 
 degrees of unequal value in terms of specific gravity. On account 
 of the varying purity of the salt used, the degrees on different 
 instruments varied greatly in value, so that 66 Baume might 
 indicate in the case of sulphuric acid a minimum strength of 
 89.5% H 2 S0 4 , or a maximum of 95% H 2 S0 4 . As a remedy for 
 this confusion, several formulas have been proposed by which the 
 degree Baume may be calculated from the specific gravity, or the 
 latter calculated from the former. The so-called rational formula 
 
 144 3 
 
 is d= ' , in which d = specific gravity and B = the degree 
 
 JL44.O Jt5 
 
DETERMINATION OF SPECIFIC GRAVITY. 277 
 
 144 
 Baume. In Holland the formula d = and in the United 
 
 JtS 
 
 145 
 
 States the formula d = has been adopted as the standard 
 14o 13 
 
 Baume scale for liquids heavier than water, and for liquids lighter 
 
 140 
 than water, the formula d = ~ has been adopted as the 
 
 J.OU ~p _D 
 
 standard American scale. Under these circumstances it is evident 
 that the Baume scale is the worst which could have been adopted, 
 and that degrees Baume is by no means a definite statement of 
 specific gravity unless it is known what particular scale has been 
 used. The formula giving the relation between specific gravity 
 and degrees Baume should be printed on the scale, and only the 
 standard American scale should be used in the United States. 
 
 PROBLEMS. 
 
 Problem 5. How many cubic centimeters of sulphuric acid having a 
 specific gravity of 1.3567 at 15 must be diluted to a liter to make N/5 
 acid? How many grams of the same acid must be used for making a liter 
 of N/5 acid? 
 
 Problem 6. Four grams of caustic soda were weighed out, dissolved in 
 water, and diluted to 500 c.c. Fifty c. c. of this solution required, 
 when titrated with N/5 acid and phenolphthalein, 46.36 c.c. of acid, and when 
 methyl orange was added, 0.94 c.c. was required to give the end-point with 
 this indicator. Calculate the percentage of sodium hydroxide, sodium car- 
 bonate, and water in the caustic soda. 
 
 Problem 7. One gram of sodium phosphate was dissolved in water and 
 titrated with N/5 acid, using phenolphthalein as the indicator, 0.67 c.c. of 
 the acid being used. Methyl orange was then introduced and acid again 
 added until the end-point was reached, 13.92 c.c. of the N/5 acid being 
 used in the second titration. Calculate the percentage of crystallized tri- 
 sodium and disodium phosphate present. 
 
 Problem 8. Ten grams of a mixture of sodium carbonate and sodium 
 bicarbonate was dissolved in water and diluted to a liter. To 50 c.c. of 
 this solution, 25 c.c. of N/5 sodium hydroxide was added, and then enough 
 barium chloride to precipitate the carbonate. On titrating with N/5 hydro- 
 chloric acid, 15 c.c. was used. Another 50-c.c. portion required 40 c.c. 
 when titrated with the N/5 acid and methyl orange. Calculate the per- 
 centage of sodium carbonate and bicarbonate present. 
 
 Problem 9. In titrating 100 c.c. of a sample of water, 5 c.c. of N/20 
 sodium hydroxide was required with phenolphthalein. Another 100-c.c. 
 
278 VOLUMETRIC METHODS. 
 
 portion required 15 c.c. of N/20 hydrochloric acid with methyl orange. 
 Calculate (a) the amount of carbon dioxide present and the alkalinity in 
 parts per million; (6) calculate the amount of calcium oxide necessary 
 to soften 1000 gallons of the water; (c) calculate the amount of lime- 
 water necessary to soften the same quantity of water. 
 
 Problem 10. To 100 c.c. of the same water, 50 c.c. of N/50 sodium-car- 
 bonate solution was added, and after evaporation, 35 c.c. of N/50 hydro- 
 chloric acid was required to titrate the excess. Calculate (a) the permanent 
 hardness in parts per million ; (b) the amount of sodium carbonate neces- 
 sary to soften 1000 gallons of the water. 
 
 After adding the softening agent and allowing the water to settle, or 
 filtering it, how could (a) the presence of an excess of lime be found by 
 titration, (6) a deficiency of lime, (c) an excess or deficiency of the soda- 
 ash? 
 
CHAPTER XXII. 
 STANDARD ALKALIES. 
 
 382. Standard Alkaline Solutions not Permanent. While 
 standard solutions of hydrochloric and sulphuric acids may be 
 kept for months and even years without change in strength, no 
 suitable alkaline solution has been found which in practice may 
 be preserved without deterioration. This arises mainly from two 
 causes, the absorption of carbon dioxide from the air and the action 
 of the alkaline solutions on the glass of the containing bottles. In 
 the case of AMMONIA neither of these causes of deterioration is 
 serious, but another difficulty is met with in the volatility of the 
 base. Moreover, it is a weak base and cannot be used with 
 phenol phthaiein. The strength of all basic solutions must there- 
 fore be frequently checked by titration with standard acid. The 
 presence of carbon dioxide may in many cases be disregarded 
 by using methyl orange as the indicator in titrations. As this is 
 not always possible on account of the nature of the acid, standard 
 alkaline solutions must frequently be prepared and kept free 
 from carbon dioxide by methods described later. 
 
 383. Caustic Soda has been found to be the alkali best suited 
 for general work. It is a strong base, forms soluble salts with 
 all acids, and may be prepared of any desired strength, while 
 the pure commercial article is cheaper than caustic potash. A 
 solution of the highest purity may be made by dissolving metallic 
 sodium in pure water. This metal may now be readily obtained 
 in pure condition. Caustic soda made from the metal may also 
 be purchased. The pure commercial caustic soda almost invari- 
 ably contains small amounts of chlorides, sulphates, silica, and 
 alumina, besides considerable amounts of carbon dioxide and 
 water. About 20% more than the calculated amount of NaOH 
 should therefore be weighed out, and dissolved in the required 
 
 279 
 
280 
 
 VOLUMETRIC METHODS. 
 
 amount of water. After titrating this solution against standard 
 acid it is diluted to the required strength. On account of the 
 presence of carbon dioxide, methyl orange must be used as the 
 indicator, or litmus accompanied by boiling to expel the carbon 
 dioxide. 
 
 384. Removal of Carbon Dioxide. The alkaline solution 
 must be kept in a bottle closed with a rubber stopper or a glass 
 stopper coated with paraffine or vaseline. If the solution is to 
 be used for titrating organic acids with phenolphthalein as the 
 indicator, the carbon dioxide must be removed by means of barium 
 chloride or hydroxide, and the solution subsequently protected from 
 the carbon dioxide of the air. This may be accomplished either 
 by pouring a layer of kerosene over the solution or by passing the 
 air which enters the bottle through a soda-lime tube. In either 
 case the soda solution must be withdrawn by means of a siphon. 
 The arrangement shown in Fig. 46 is very convenient. The caus- 
 
 FIG. 46. 
 
 tic-soda solution is made up in a bottle of suitable size and, after 
 the addition of enough barium chloride or barium hydroxide solu- 
 tion to precipitate all of the carbon dioxide, is allowed to stand 
 until the precipitate has settled. The clear solution is then 
 siphoned off into the bottle A, and if a layer of petroleum-oil is not 
 
STANDARD ALKALIES. 281 
 
 placed on top, the tube B is filled with soda-lime and inserted into 
 the stopper. The long arm of the siphon C is pushed through 
 the second hole of the stopper so that a space of about one inch 
 is left between the end and the bottom of the bottle. The alkali 
 may then be withdrawn, leaving undisturbed any barium carbon- 
 ate which settles to the bottom. The siphon is filled by apply- 
 ing suction to the tube D, which is filled with soda-lime, or pressure 
 may be applied at B. The long arm of the siphon may be adjusted 
 so that the end is opposite the zero-mark on the burette, and any 
 excess of alkali drawn over will flow back into the stock-bottle. 
 
 385. Alcoholic Caustic-potash Solution is frequently used foi 
 the titration of ORGANIC ACIDS with phenolphthalein as the indi- 
 cator. Such a solution is of twofold advantage. Many of the 
 organic acids are insoluble in water, but dissolve readily in alco- 
 hol. Alcoholic solutions of caustic potash are free from carbon 
 dioxide, because potassium carbonate is almost absolutely insolu- 
 ble in alcohol. On dissolving the caustic potash in the alcohol, 
 the potassium carbonate remains undissolved. After standing 
 twenty-four hours the clear solution may readily be siphoned 
 off from the potassium carbonate, which adheres quite firmly to 
 the bottle. The alcohol used must first be freed from aldehyde 
 or organic material extracted from the oak-barrel, which turns 
 yellow on the addition of alkali and interferes with the titration. 
 For this purpose a stick of caustic potash is added, and after stand- 
 ing for some time the alcohol is distilled off. The caustic-potash 
 solution must be protected from the carbon dioxide of the air 
 by the method suggested for caustic soda or some similar device. 
 
 386. Water Solutions of Barium Hydroxide have been largely 
 used when the presence of carbon dioxide would be objectionable. 
 Any carbon dioxide which enters the solution is immediately 
 precipitated as the insoluble barium salt. A solution saturated 
 with barium hydroxide at the ordinary temperature is about two- 
 fifths normal. After shaking the solution with excess of the com- 
 mercial barium hydroxide until no more dissolves, the clear liquid 
 should be siphoned off and diluted a little before standardizing 
 so as to prevent the crystallizing out of the barium hydroxide 
 during occasional periods of low temperature. This alkali gives 
 a very sharp end -reaction with phenolphthalein. 
 
282 VOLUMETRIC METHODS. 
 
 387. Ammonia. On account of its volatility strong solutions 
 of ammonia cannot be used. Even with tenth- or fifth-normal 
 solutions the loss of alkali is considerable if the solution is 
 kept in a half-filled bottle and is poured out into the burette. 
 During this process the air above the liquid is largely replaced, 
 and as the pressure of the ammonia vapor is proportional to the 
 concentration of the ammonia in solution, the loss in any case is 
 considerable. The action of the ammonia on the glass is com- 
 paratively slight, especially if the same bottle is used continually. 
 The amount of carbon dioxide absorbed is also inconsiderable. 
 If therefore the ammonia solution is siphoned out of the bottle, 
 a fifth- or tenth-normal solution will show only inappreciable 
 changes in strength from day to day, and if kept in an old bottle 
 may be used for considerable periods without change of titre. 
 
 EXERCISE 52. 
 
 Preparation of N/5 Caustic Soda and Determination of Ammonia in Ammonium 
 
 Chloride. 
 
 Weigh out roughly about 10 grams of pure caustic soda, dissolve in 
 water, and dilute to 1 liter. Place in a bottle provided with a rubber stopper. 
 Standardize by titrating against standard hydrochloric or sulphuric acid, 
 using methyl orange as the indicator. Use at least 25 c.c. of alkali for 
 each titration. If the acid is exactly N/5 normal, divide the number of 
 cubic centimeters of alkali by the number of cubic centimeters of acid used. 
 The quotient multiplied by 1000 gives the number of cubic centimeters of the 
 alkali to be diluted to 1000 to make a N/5 solution. Verify the strength 
 of the diluted solution by titrating against the standard acid. 
 
 Weigh out 250 mg. of ammonium chloride and transfer to a 300-c.c. 
 Jena beaker. Dissolve in 50 c.c. of water and add 50 c.c. of the caustic- 
 soda solution. In a similar beaker place 50 c.c. of water and 50 c.c. of the 
 caustic-soda solution. Place both beakers on the hot plate and warm, 
 finally to boiling, until no more ammonia is evolved. Cool both beakers, 
 and titrate the alkali remaining in each with the standard acid, using methyl 
 orange as the indicator. The difference between the amounts of acid used 
 to neutralize the alkali in the beakers is the amount necessary to neu- 
 tralize the ammonia in the ammonium chloride. 1 c.c. of N/5 acid =.003407 
 gram of ammonia or .0107 gram of ammonium chloride. Compute the 
 percentage of ammonia as well as of ammonium chloride in the sample 
 analyzed. 
 
KJELDAHL METHOD FOR NITROGEN. 283 
 
 THE KJELDAHL METHOD OF DETERMINING NITROGEN. 
 
 In 1883 Kjeldahl described a method for determining nitro- 
 gen in organic substances which, with the numerous modifications 
 which have since been suggested, is now the method most largely 
 used for the determination of nitrogen, not only in organic com- 
 pounds of the greatest diversity of composition, but also in nitrates 
 and other inorganic compounds. There are very few nitrogen 
 compounds in which by one or the other modification the nitro- 
 gen cannot be determined with a degree of accuracy and with a 
 rapidity almost unequalled by any other quantitative method. 
 
 388. Oxidation of Organic Matter and Conversion of Nitrogen 
 into Ammonium Sulphate. By the original method the organic 
 material was digested hot with concentrated sulphuric acid, by 
 which the carbon was oxidized to carbon dioxide, while the nitrogen 
 remained as ammonium sulphate. The oxidation was completed 
 by the addition of powdered potassium permanganate. The 
 concentrated acid was then diluted somewhat, and neutralized 
 with caustic soda. The ammonia liberated was distilled off, and 
 determined by means of standard sulphuric acid. In many cases 
 the combustion of the carbon was very slow. Various modifica- 
 tions have been introduced to overcome this difficulty. The addi- 
 'tion of mercury, mercuric oxide, or copper oxide or sulphate greatly 
 shortens the time of combustion, probably by acting as carriers 
 of oxygen, the metal being alternately oxidized by the sulphuric 
 acid and reduced by the carbon. The addition of phosphorus 
 pentoxide or potassium sulphate serves the same purpose by raising 
 the temperature to which the solution may be heated. When 
 mercury has been added to the sulphuric acid it must be re- 
 moved by means of sodium sulphide before the ammonia is dis- 
 tilled off, otherwise some of the ammonia combines with the 
 mercuric oxide and cannot be expelled by distillation. 
 
 389. Reduction of Nitrates. By the original Kjeldahl method 
 the conversion of nitrates to ammonia was very uncertain. To 
 overcome this difficulty, salicylic acid or phenol was dissolved in 
 the sulphuric acid. This solution first converts the nitrate into 
 nitro-phenol. This compound is readily reduced to amido-phenol, 
 
284 
 
 VOLUMETRIC METHODS. 
 
 which is completely hydrolyzed by caustic soda with liberation 
 of ammonia. These reactions take place according to the follow- 
 ing equations: * 
 
 C 6 H 5 .OH +HN0 3 =C 6 H 4 .OH.N0 2 +H 2 ; 
 C 6 H 4 .OH.N0 2 +3H 2 =C 6 H 4 .OH.NH 2 +2H 2 0; 
 C 6 H 4 .OH.NH 2 +NaOH =C 6 H 4 .OH.ONa+ NH 3 . 
 
 The reduction is greatly assisted by the addition of sodium thio- 
 sulphate or zinc. The addition of pure cane-sugar is also advan- 
 tageous when substances of high nitrogen content are oxidized. 
 
 390. Kjeldahl Digestion-flasks. The digestion of the nitrog- 
 enous substance with sulphuric acid is carried out in pear- 
 shaped, long-necked flasks of Bohemian or Jena glass which are 
 made for this purpose. The flasks are made of various sizes, 
 since some material foams considerably on being heated with the 
 sulphuric acid, and therefore requires the use of a larger flask. 
 Some workers also desire to obviate the transference of the sul- 
 phuric-acid solution from the digestion-flask to a special distilla- 
 tion-flask. If this method is pursued, a digestion-flask capable 
 
 tf-2 
 
 FIG. 47. 
 
 of holding about 500 c.c. must be used. If a separate distillation- 
 flask is employed, the digestion-flask need not ordinarily have a 
 capacity of more than 200 c.c. 
 
 391. Digestion. During the digestion the flask is placed so 
 as to be inclined at an angle of 45, in order that there shall be no 
 loss of material by spattering, nor any contamination by the falling 
 
 * These reactions are in reality more complicated than represented, as several 
 sulphuric acid molecules are combined with the benzene radicle. The reactions 
 as given simply show the transformation of the nitric-acid radicle into ammonia. ' 
 
KJELDAHL METHOD FOR NITROGEN. 285 
 
 in of foreign material. As an additional precaution some 
 workers close the mouth of the flask by means of a stopper 
 made by drawing out a glass tube which is somewhat larger than 
 the neck of the digestion-flask. The drawn-out end is sealed, 
 and after cutting off about an inch of the glass tube the stopper 
 is inserted with the drawn-out end in the neck of the flask. 
 Racks for holding six or eight flasks in an inclined position 
 are used where much work must be done. These racks are pro- 
 vided with a Bunsen burner under each flask. If the material 
 froths much, the flask is at first heated with a small flame and over 
 a wire gauze. The heat is gradually increased until finally the 
 flask is heated with the bare flame, until the acid is colorless or a 
 light yellow. The digestion requires from one-half to several 
 hours. As fumes of sulphuric acid and sulphur dioxide are copi- 
 ously evolved, the digestion must be conducted under a hood. 
 
 392. Distillation. After cooling, the acid is diluted somewhat 
 and transferred to the distillation-flask, if a separate flask is used 
 for this purpose. The distillation-flasks should have 
 a capacity of 500 to 600 c.c. and should be made of 
 hard Bohemian or Jena glass. They may be round- 
 bottomed or of the Erlenmeyer form. To ' prevent 
 the carrying over of the caustic alkali in the form of 
 spray, a trap of some kind must be interposed between 
 the distillation-flask and the condenser. The simplest 
 form consists of a glass bulb 4 to 5 cm. in diameter 
 filled with glass wool. In another trap shown in Fig. 48, 
 which is very much used, the upper exit tube extends 
 into and is somewhat curved towards the side of the 
 bulb. The rate of distillation is much increased by 
 covering the bulb with thin asbestos paper, which may 
 be made to adhere by simply moistening with water. It is then 
 pressed with the hand around the bulb. 
 
 The trap is connected to the distillation-flask by means of a 
 rubber stopper and to the condenser-tube by means of a piece of 
 thick- walled rubber tubing. The tube of the condenser must be of 
 block tin. Glass cannot be used, as it is dissolved by the steam 
 and ammonia vapor, increasing the alkalinity of the distillate. 
 The condenser-tube is curved downwards, where it is joined to 
 
286 
 
 VOLUMETRIC METHODS. 
 
 the trap. A number of these block-tin tubes are frequently 
 passed through the same condenser, which consists of a copper or 
 iron tank of convenient size, as shown in Fig. 49. 
 
 The DISTILLATE is received in an Erlenmeyer flask of 300- to 
 400-c.c. capacity, in which a measured amount of standard sul- 
 phuric acid has been placed. If the condenser is supplied with a 
 sufficient amount of cold water, it appears to be unnecessary for 
 the end of the condenser-tube to dip into the standard acid. It 
 is customary, however, to attach a large-sized glass tube to the exit 
 of the tin condenser-tube by means of a piece of rubber tubing. 
 
 FIG. 49. 
 
 The end of this glass tube should touch the surface of the acid. 
 All of the ammonia is usually carried over when 150 c.c. of liquid 
 has distilled over. Until experience is gained it is advisable to 
 continue the distillation after the Erlenmeyer flask containing 
 about 150 c.c. of distillate has been removed. The distillate is 
 then allowed to flow into a small beaker containing a drop of the 
 indicator. If the distillate is still alkaline, the contents of the 
 beaker are poured into the Erlenmeyer flask. The bumping 
 which so frequently accompanies the boiling of an alkaline solu- 
 
KJELDAHL METHOD FOR NITROGEN. 287 
 
 tion may generally be prevented by the addition of a few grams 
 of granulated zinc. Bumping is also very much reduced if caus- 
 tic soda free from carbonate is used. A small piece of paraffine 
 will prevent foaming of the liquid. 
 
 393- Titration of Ammonia. The excess of acid in the distil- 
 late is titrated back with standard alkali, which may be caustic 
 soda, potash, or ammonia. The indicator used may be methyl 
 orange, litmus, or cochineal. As the various reagents used may 
 contain more or less nitrogen, a blank must be conducted when- 
 ever one or more of the reagents is used for the first time. For 
 the blank determination pure cane-sugar is digested, and care 
 must be taken to add each of the reagents in the amount used hi 
 the actual analysis. The amount of nitrogen found hi this deter- 
 mination is deducted from each result subsequently obtained 
 with the reagents tested. 
 
 WILFARTH'S MODIFICATION 
 
 APPLICABLE TO SUBSTANCES FREE FROM NITRATES. 
 
 One gram of the finely powdered substance is placed in the 
 digestion-flask, 1 gram of metallic mercury or about 0.7 gram of 
 mercuric oxide is added, and 25 c.c. of a solution of 200 grams of 
 phosphorus pentoxide in 1 liter of concentrated sulphuric acid 
 (sp. gr. 1.84). After digesting and transferring the acid to the 
 distillation-flask with about 200 c.c. of water the mercury is pre- 
 cipitated by the addition of 25 c.c. of a solution of 40 grams of 
 potassium sulphide in 1 liter of water. The acid is then neutral- 
 ized by the addition of a saturated solution of caustic soda, of 
 which about 50 c.c. will be required. After the addition of a few 
 grams of granulated zinc the ammonia is distilled into the sul- 
 phuric acid, of which 50 c.c. of N/5 acid is usually ample. The 
 excess of acid is titrated back with N/5 or N/10 alkali. 
 
 GUNNING'S MODIFICATION 
 
 APPLICABLE TO SUBSTANCES FREE FROM NITRATES. 
 
 Gunning's improvement consisted in the addition of potassium 
 sulphate to the sulphuric acid, by which the time of digestion is 
 considerably reduced. Sodium sulphate may also be used as 
 
288 VOLUMETRIC METHODS. 
 
 well as sodium pyrophosphate. A very excellent digestion mix- 
 ture consists of 20 c.c. of the mixture of phosphorus pentoxide 
 and sulphuric acid, given in Wilfarth's modification, to which 10 to 
 15 grams of potassium sulphate are added, either in the beginning 
 or after the first violent oxidation is ended. Excellent results 
 may be obtained with this mixture without the addition of 
 mercury, though the oxidation is considerably hastened by the 
 addition of the usual amount of the latter. The decomposition 
 is frequently complete in thirty minutes and rarely requires more 
 than an hour. The precipitation of the mercury, neutraliza- 
 tion, distillation, etc., are carried out as given in the Wilfarth 
 modification. 
 
 GUNNING-JODLBAUER MODIFICATION 
 
 APPLICABLE TO NITRATES. 
 
 In this modification the digestion is carried out with sulphuric 
 acid which contains 40 grams of phenol or salicylic acid per liter. 
 1 gram of the nitrogenous substance is placed in the digestion- 
 flask and dissolved in 30 c.c. of the phenol-sulphuric acid. The 
 solution should be kept cold and thoroughly shaken to hasten 
 the solution. 2 to 3 grams of zinc-dust and 1 gram of mercury 
 are then added while the flask is shaken and cooled. After stand- 
 ing one to two hours the flask is heated, at first gently, then more 
 strongly. 10 to 15 grams of potassium sulphate may then be 
 added, and the flask heated until the solution is colorless or nearly 
 so. The remainder of the process is conducted as directed for 
 the other modifications. 
 
 FORSTER'S MODIFICATION 
 
 APPLICABLE TO NITRATES. 
 
 Having failed to obtain all of the nitrogen from nitrates by 
 the previous modification, Forster introduced the use of sodium 
 thiosulphate as a reducing agent. He treated J gram of the 
 nitrate in the digestion-flask with 15 c.c. of a 6% phenol-sulphuric 
 acid, and after the nitrate was completely dissolved 1 to 2 grams 
 of sodium thiosulphate were added. When the decomposition of 
 the thiosulphate is complete 10 c.c. of concentrated sulphuric 
 
KJELDAHL METHOD FOR NITROGEN. 289 
 
 acid and 1 gram of mercury are added. The decomposition is 
 completed in the usual manner. The thiosulphate must not be 
 added before the addition of the phenol-sulphuric acid, or nitrogen 
 will be lost. The solution need not be cooled as hi the 
 Jodlbauer modification. The oxidation is generally complete hi 
 1 J hours. 
 
 394. Substances Decomposed with Difficulty. In regard to 
 the substances which cannot be determined by the Kjeldahl 
 method and its modifications Dyer* states that the nitrogen in 
 aromatic nitro-compounds cannot be determined by simple reduc- 
 tion with zinc. Phenol or salicylic acid must be present. The 
 same thing is true of many of the azo compounds. Hydroxyl- 
 amine and acetaldoxime required the addition of sugar before 
 all of the nitrogen was converted into ammonia by the Jodlbauer 
 method. The cyanide and ferrocyanide of potassium could be 
 analyzed by the Gunning modification, while the ferricyanide 
 required the Forster modification. Phenylhydrazine could not 
 be correctly analyzed by any modification. H. C. Sherman f 
 states that where large proportions of chlorides and nitrates 
 exist together all modifications fail. 
 
 EXERCISE 53. 
 
 Determination of Nitrogen in Milk by the Kjeldahl-Gunning Method. 
 
 395. Standard Solutions. For this determination a fifth normal acid 
 and alkali will be needed. The fifth normal sulphuric- or hydrochloric-acid 
 solution already prepared may be used. The fifth-normal solution of caustic 
 soda may also be used, or a fifth-normal solution of ammonia may be pre- 
 pared as follows: About 15 c.c. of concentrated ammonia (sp. gr. 0.90) are 
 diluted to 1 liter. A second liter is made in the same manner as the first. 
 The entire solution is well mixed in a large bottle, and then titrated against 
 the N/5 standard acid, using cochineal or methyl orange as the indicator. 
 The solution of ammonia, which is somewhat stronger than fifth-normal, 
 is then diluted to the exact strength, and placed in a bottle connected 
 with the burette by means of a siphon, as described on p. 280. The 
 soda-lime tube may be omitted. A simpler arrangement, which answers 
 fully as well for this purpose, consists of a suitable-sized bottle, closed 
 
 * Chem. Centr., Ill, XVII, 433. 
 t J. Am. Chem. Soc.. XVII. 567. 
 
290 
 
 VOLUMETRIC METHODS. 
 
 by a two-holed rubber stopper through which a siphon passes, the long 
 arm of which is outside the bottle. To the end of the long arm is 
 attached, by means of a short piece of rubber tubing, a piece of glass 
 tubing. A pinch-cock is placed on the rubber 
 tube. By opening this pinch-cock and forcing 
 air through the second opening in the stopper, 
 the siphon may be filled. The burette is filled 
 by opening the pinch-cock and allowing the so- 
 lution to flow out. 
 
 3Qb. Weighing the Sample. The sample of milk 
 to be analyzed must be thoroughly mixed by shak- 
 ing the bottle. 10 c.c. is withdrawn with a pipette 
 and allowed to flow over the asbestos contained in 
 a Hofmeister capsule. The asbestos must be first 
 ignited, and after placing in the capsule, the 
 whole is weighed. After the milk has been added 
 the whole is again weighed. Neither of these 
 weights need be taken closer than several milli- 
 grams. A second or even a third sample of milk 
 is weighed out in the same manner. The milk 
 FIG. 50. j s evaporated to dryness on the water-bath, 
 
 the capsule crushed in a clean porcelain mortar and transferred to the 
 digestion-flask. 
 
 The milk may also be evaporated in the flasks, the weight then being 
 obtained from the specific gravity, or the average weight of 10 c.c. may be 
 obtained by first weighing several portions in small beakers. The evapora- 
 tion in the flasks is slow unless the steam is displaced by a stream of air. 
 The water may also be boiled out after the addition of the digestion- 
 mixture. 
 
 397. Digestion. Weigh out 20 grams of phosphorous pentoxide in a 
 beaker, and add 500 c.c. of concentrated C. P. sulphuric acid. Stir until 
 solution is complete, and transfer to a glass-stoppered bottle. Transfer 
 20 c.c. portions of this solution to the digestion-flasks containing the milk, 
 and also to two other flasks in each of which about a gram of granulated 
 sugar has been placed. Add 10 grams of potassium sulphate to each flask, 
 and heat gently over the wire gauze with the Bunsen burner. The heating 
 is conducted under the hood. Increase the heat as the frothing ceases, 
 finally remove the wire gauze, and heat strongly until the solution is color- 
 less or only a light yellow. 
 
 While the solution is digesting a saturated solution of caustic soda is 
 made by dissolving 200 grams in twice this amount of water. 40 or 50 c.c. 
 of the standard acid is measured out into each of the Erlenmeyer flasks 
 to be used as receivers. If the condensers of the distilling apparatus have 
 not been recently used, they are cleaned by passing through them the 
 steam from a flask of water strongly acidified with sulphuric acid until 
 
KJELDAHL METHOD FOR NITROGEN. 291 
 
 the distillate is no longer alkaline to cochineal solution. If 100 c.c. of the 
 laboratory supply of distilled water reacts alkaline with the cochineal 
 solution, ammonia-free water should be made by condensing the steam 
 from a flask of water acidified with sulphuric acid. This steam should be 
 condensed by passing it through one of the block-tin tubes of the Kjeldahl 
 condenser. This distilled water should then be used for diluting the con- 
 centrated acid and whenever needed during the titrations. 
 
 398. Distillation. When the digestion-flasks have cooled, about 50 c.c. of 
 water are added to each, and if the small-sized digestion-flasks have been 
 employed, the solutions are transferred to the distillation-flasks, using about 
 150 c.c. of the ammonia-free water to rinse out the digestion-flasks. The 
 acid is neutralized with the concentrated caustic-soda solution, of which 
 about 50 c.c. will be required. 2 or 3 grams of granulated zinc are added 
 and the flask is immediately connected with the condenser to prevent loss 
 of ammonia from the hot alkaline solution. The distillation of the ammonia 
 in the first flask is begun immediately, and the acid in the other flasks is 
 diluted, transferred, and neutralized in rotation. 
 
 399- Titration. When about 150 c.c. of distillate has passed into the 
 first Erlenmeyer flask, it is replaced by a small beaker containing a little of the 
 indicator. If 10 or 15 c.c. of the distillate does not react alkaline with 
 the indicator, 2 or 3 drops of the cochineal are added to the solution in the 
 Erlenmeyer flask and the excess of acid titrated with the standard alkaline 
 solution. If the latter has not been compared within a day or two with 
 the standard acid, a measured volume must be titrated against the 
 acid. The other Erlenmeyer flasks should be removed and the solu- 
 tions titrated in the order in which the acid solutions were transferred to 
 the distillation-flasks. After deducting the average amount of acid neu- 
 tralized by the ammonia obtained in the blank determinations, the amount 
 of nitrogen found is computed by the factor 1 c.c. of N/5 acid =.002802 
 gram of nitrogen. To obtain the percentage of casein, the percentage of 
 nitrogen is multiplied by the commonly used factor 6.25 or by the more 
 recently obtained and probably more correct factor 6.43. 
 
 EXERCISE 54. 
 
 
 
 Determination of Nitrogen in Potassium Nitrate by the Forster Modification of 
 
 the Kjeldahl Method. 
 
 For this determination it is advisable to use a sample of pure potas- 
 sium nitrate as a check on the process. A sample of commercial sodium 
 or potassium nitrate may then be analyzed. 
 
 Twenty grams of phenol are dissolved in 340 grams of concentrated 
 sulphuric acid. A solution of potassium sulphide is made by dissolving 
 20 grams of the commercial article in 500 c.c. of water. Jf-gram portions of 
 the nitrate are weighed out and transferred to digestion-flasks and dis- 
 
292 VOLUMETRIC METHODS. 
 
 solved in 15 c.c. of the phenol-sulphuric-acid solution. After the nitrate 
 is completely dissolved, 1 to 2 grams of sodium thiosulphate are added. 
 When the reaction due to the latter is ended, 10 c.c. of concentrated sul- 
 phuric acid is added and about 1 gram of metallic mercury. The mercury 
 may be dropped from a glass tube the end of which is partially closed by 
 fusion or drawing out. The size of this opening may readily be so adjusted 
 that one drop shall weigh about 1 gram. A blank determination is made 
 by treating about 1 gram of sugar with the reagents as already given. The 
 digestion and the subsequent operations are carried out exactly as given in 
 the preceding exercise, except that after the acid solution has been trans- 
 ferred to the distillation-flask, 25 c.c. of the potassium -sulphide solution are 
 added. 1 c.c. of N/5 acid is equal to one five-thousandth of the molecular 
 weight of potassium nitrate or the same proportion of the molecular weight 
 of sodium nitrate. 
 
CHAPTER XXIII. 
 TITRATION OF BORIC AND CARBONIC ACIDS. 
 
 ESTIMATION OF BORIC ACID. 
 
 THE titration of boric acid cannot be carried out by means of 
 any indicator in the ordinary manner of titrating an acid, on account 
 of the very weak acid properties of this substance. Because of 
 this peculiarity, the base in many of the salts of boric acid may 
 be readily titrated by means of a strong mineral acid and methyl 
 orange as the indicator without the removal of the boric acid, 
 which is entirely neutral towards methyl orange. Because of 
 the difficulty of carrying out the gravimetric methods which have 
 been devised, the discovery of a method of increasing the acid 
 properties of this acid so that it can be titrated with phenol- 
 phthalein as the indicator has been very welcome to chemists. 
 
 400. Thompson's Glycerine Method. The method proposed 
 by R. T. Thompson * consists in the addition of enough glycerine 
 to constitute 30% of the solution when the titration is complete. 
 In such a solution the boric acid acts as a monobasic acid. To 
 secure a solution which shall contain no free alkali nor acid except 
 boric acid, the borate is dissolved in water and the solution neu- 
 tralized with sulphuric acid after the addition of a drop of methyl 
 orange. A drop or two of acid is then added and the solution 
 boiled in an open dish for a few minutes to expel carbon dioxide. 
 A dilute solution may be boiled in this manner for fifteen minutes 
 without the loss of more than a faint trace of boric acid.f Dilute 
 caustic-soda solution is now added until the pink color of the 
 methyl orange just disappears. Glycerine is then added in such 
 amount that after completion of the titration at least 30% shall be 
 
 * Jour. Soc. Chem. Ind., XII, 432. 
 
 f L. de Koningh, Jour. Am. Chem. Soc., 1897, 385. 
 
 293 
 
294 VOLUMETRIC METHODS. 
 
 present. After the addition of a drop of phenolphthalein the solu- 
 tion is titrated with fifth-normal caustic soda which is free from 
 carbon dioxide. The usual pink end-point of the phenolphthalein 
 is obtained when a slight excess of the alkali has been added. 
 More glycerine should then be added, and if the pink color of the 
 phenolphthalein fades, more alkali must be added. A blank deter- 
 mination should be carried out to determine the acidity of the 
 glycerine. For this purpose the same amount of glycerine should 
 be diluted with an equal volume of water and the solution tit- 
 rated with the standard alkali. 
 
 If ammonia is present, it must first be removed by adding a 
 slight excess of sodium carbonate and evaporating the solution 
 down to half its bulk. By this treatment any iron present will be 
 precipitated and should be filtered off. The neutralization of 
 the boric acid with caustic soda takes place according to the fol- 
 lowing equation: 
 
 HB0 2 +NaOH =NaB0 2 +H 2 0. 
 
 1 c.c. of normal caustic alkali is therefore equal to .035 gram of B 2 3 . 
 
 401. Jones's Mannitol Method. Another method of titrating 
 boric acid has been based on the fact that in the presence of manni- 
 tol the acidity is sufficiently increased to allow of its titration 
 with phenolphthalein as the indicator.* The solution may be neu- 
 tralized in the same manner as given in the glycerine method. 
 Phenolphthalein is then added and standard alkali introduced 
 until the solution is red. A pinch of mannitol is then thrown in, 
 which bleaches the red color. Enough alkali is added to produce 
 a slight alkaline reaction, followed by mannitol until a permanent 
 red color is produced which cannot be discharged by further addi- 
 tion of mannitol. 1 or 2 grams of mannitol is usually sufficient. 
 
 The author of this method, instead of using methyl orange as 
 the indicator to secure a solution free from acids except boric, 
 utilizes the fact that a mixture of potassium iodide and iodate is 
 entirely unaffected by free boric acid, while strong mineral acids 
 liberate iodine, thus destroying the free mineral acid according to 
 the following equation : 
 
 6HC1 +5KI +KI0 3 =6KC1 + 3I 2 +3H 2 0. 
 
 * L. C. Jones, Am. J. Sci., 1898, pp. 147-153. 
 
TITRATION OF BORIC ACID. 295 
 
 The free iodine is removed by adding sodium thiosulphate solution, 
 which combines with the iodine, producing only neutral compounds 
 as follows: 
 
 2Na 2 SA +1, =2NaI + Na 2 S 4 6 . 
 
 The amount of free mineral acid present may be estimated by de- 
 termining the amount of free iodine, as will be shown in the next 
 chapter. 
 
 The solution of the borate, the volume of which should be 
 about 50 c.c. and which should contain about 0.1 gram of boric 
 acid, is made slightly acid with hydrochloric acid. To precipitate 
 carbonic acid 5 c.c. of a 10% solution of barium chloride is added. 
 A mixture of a solution of potassium iodide and iodate is made 
 in a separate beaker. About 2 grams of the iodide and J gram 
 of the iodate are generally sufficient to destroy all the mineral 
 acid present, unless the amount is excessive. Starch solution is 
 added to the iodate solution, and if a blue color develops, it is re- 
 moved by the cautious addition, of dilute sodium thiosulphate solu- 
 tion. A drop of the borate solution added to the iodate should 
 produce a blue color, indicating the presence of free hydrochloric 
 acid, and the certainty of the presence of all boric acid in the free 
 state. The two solutions are then mixed, and the iodine liberated 
 is removed by the addition of sodium thiosulphate. The colorless 
 solution now contains only starch, neutral chlorides, potassium 
 tetrathionate, iodide, and iodate, and the total amount of the 
 boric acid to be determined in the free condition. Phenolphthalein 
 is now added and the boric acid titrated, the mannitol being 
 added as already directed. The end-reaction by this method is 
 sharper than that in Thompson's method. 
 
 402. Decomposition of Organic Matter containing Boric Acid. 
 Boric acid must frequently be determined in food products, and 
 the separation of the organic matter is first necessary. Thompson 
 proceeds as follows: 100 grams of the sample is made strongly 
 alkaline with lime-water, evaporated to dryness in a platinum 
 dish, and cautiously heated until the organic matter is thoroughly 
 charred so as to give a colorless solution. 20 c.c. of water is 
 added, and the solution acidified with hydrochloric acid, when all 
 but carbon will be in solution. The whole is transferred to a 
 100-c.c. flask and 0.5 gram calcium chloride added. A few drops of 
 
296 VOLUMETRIC METHODS. 
 
 phenolphthalein are introduced, then caustic soda until a slight 
 pink color is produced, and finally 25 c.c. of lime-water. By this 
 treatment the carbon dioxide as well as other weak acids, such as 
 phosphoric acid, which may be present are precipitated as calcium 
 salts. This solution is made up to 100 c.c., thoroughly shaken, 
 and filtered through a dry paper. To 50 c.c. of the filtrate normal 
 sulphuric acid is added until the pink color disappears, then 
 a drop or two of methyl orange is introduced and acid is again 
 added until the yellow color is just changed to pink. Fifth- 
 normal caustic soda is then added drop by drop until the liquid 
 assumes the yellow tinge, excess of soda being carefully avoided. 
 The solution is boiled until the carbon dioxide is entirely expelled. 
 It is then cooled and the boric acid titrated, using phenolphthalein 
 and glycerine or mannitol as the indicator. 
 
 DETERMINATION OF CARBONIC ACID IN WATER. 
 
 The determination of the percentage of carbonates and bicar- 
 bonates when existing as alkali salts, and in the case of the former 
 in the presence of caustic alkali, has already been given. The 
 determination of carbonic acid when present as an insoluble car- 
 bonate is best carried out gravimetrically, as given in Chapter 
 IX. Carbon dioxide existing in solution, as hi natural or 
 artificial mineral waters, is generally determined volumetrically. 
 It exists in three conditions in such waters: as FREE CARBONIC 
 ACID; as SEMICOMBINED ACID, or BICARBONATES of the alkali or 
 alkaline-earth metals; and as COMBINED ACID, or CARBONATES of 
 the alkali metals. 
 
 403. The Percentage of Free Carbon Dioxide may be deter- 
 mined by titrating 100 c.c. of the water with a standard solution 
 of caustic soda or sodium carbonate after the addition of a few 
 drops of phenolphthalein. The alkali should be dilute, tenth or 
 twentieth normal being convenient. When sufficient alkali has 
 been added to give a faint pink to the indicator, all the carbon 
 dioxide has been converted into bicarbonate and a slight excess of 
 carbonate is present, which produces the color with the indicator. 
 The titration must be repeated, adding at once nearly all the 
 alkali used in the first titration as some carbon dioxide may 
 have been lost during the stirring of the acid solution. 
 
TITRATION OF CARBONIC ACID. 297 
 
 If free carbon dioxide is present, normal carbonates must 
 be absent, since such salts would react with the free acid to 
 form bicarbonates as follows: 
 
 Na 2 C0 3 + H 2 C0 3 =2NaHC0 3 , 
 
 Only bicarbonates, therefore, can exist in a solution containing 
 free carbonic acid. On the addition of caustic soda the following 
 reaction takes place: 
 
 NaOH+ H 2 C0 3 =NaHCO,+ H 2 0. 
 
 40.058 parts of caustic soda are therefore equivalent to 44 parts 
 of carbon dioxide. 1 c.c. of N/10 caustic soda will therefore be 
 equal to .004400 gram of C0 2 . 
 
 With sodium carbonate solution the following reaction takes 
 place : 
 
 Na 2 C0 3 +H 2 C0 3 =2NaHC0 3 . 
 
 106.1 parts of sodium carbonate are therefore equal to 44 parts 
 of carbon dioxide. 1 c.c. of N/10 sodium carbonate solution will 
 therefore be equal to .0022 gram of C0 2 . The use of an alkaline 
 solution of which 1 c.c. is equal to 1 mg. of carbon dioxide is 
 very convenient. 2.412 g. of sodium carbonate are equal to 1 
 
 g. of carbon dioxide f .. = 2.412 j. Probably the most con- 
 venient method of procedure is to dissolve 2.412 g. of pure 
 anhydrous sodium carbonate in 1 liter of distilled water from 
 which the carbon dioxide has been expelled by boiling. 1 c.c. of 
 this solution will be equal to 1 mg. of carbon dioxide. 
 
 If the water is neutral to phenolphthalein only bicarbonates 
 can be present. The water may be tested for the presence of 
 FREE MINERAL ACIDS by titration with the standard solution of 
 alkali, using methyl orange as the indicator. The difference 
 between the amounts of alkali used in this titration and the one 
 in which phenolphthalein is used as the indicator gives the 
 amount of alkali necessary to neutralize the free carbonic acid. 
 In the presence of free mineral acid semi-combined or combined 
 carbonic acid would be absent. 
 
 404. The Amount of Semi-combined Carbon Dioxide may be 
 determined by titrating 100 c.c. of the water with dilute standard 
 
298 VOLUMETRIC METHODS. 
 
 acid, using methyl orange as the indicator. The titration with acid 
 and methyl orange gives the amount of semi-combined carbon di- 
 oxide only when the water is acid or neutral to phenolphthalein, as 
 normal carbonates are then absent. If the water is alkaline to 
 phenolphthalein free carbon dioxide is absent and only carbon- 
 ates and bicarbonates can be present. By titrating 100 c.c. of such 
 a water with phenolphthalein and dilute (N/20) acid until the pink 
 color is removed and then adding methyl orange and continuing 
 the addition of the acid until the pink color of the methyl orange 
 is developed, the amount of combined and semi-combined acid 
 may be calculated. The combined carbon dioxide is calculated 
 from the number of cubic centimeters of acid used in the first 
 stage of the titration, 1 c.c. of N/20 acid being equal to .0022 gram 
 
 /44 00 \ 
 of carbon dioxide f ' ), tho following reaction taking place: 
 
 HC1 +Na 2 C0 3 =NaHC0 3 +NaCL 
 
 The number of cubic centimeters of acid used in the second stage 
 of the titration minus the amount used in the first stage gives the 
 amount of semi-combined carbon dioxide, 1 c.c. of N/20 acid 
 being equal, as before, to .0022 gram of carbon dioxide. 
 
 405. The Total Amount of Carbon Dioxide may be obtained by 
 adding together the amount of free carbon dioxide to that semi- 
 combined; or when free carbon dioxide is absent, by adding the 
 amount of semi-combined to the amount combined. The amount 
 of free and semi-combined carbon dioxide is frequently deter- 
 mined by PETTENKOFER'S METHOD. In this process a measured 
 amount of a standard solution of calcium or barium hydroxide 
 is added to a measured volume, generally 100 c.c., of the water to 
 be tested. The free and semi-combined carbon dioxide is precipi- 
 tated as barium or calcium carbonate with consequent decrease of 
 alkalinity of the solution. The excess of alkali is titrated back with 
 dilute acid. The total amount of carbon dioxide is calculated 
 from the excess of barium hydroxide plus the initial alkalinity of 
 the water : 
 
 Ba(OH) 2 + H 2 C0 3 - BaC0 3 + 2H 2 0; 
 Ba(OH) 2 + Ca(HC0 3 ) 2 = BaC0 3 + CaC0 3 + 2H 2 0. 
 
 406. If Sulphates or Carbonates of the Alkalies are present, the 
 precipitated barium salts carry down some of the alkalies unless 
 
TITRATION OF CARBONIC ACID. 299 
 
 the barium was present as chloride, so as to leave the alkali metals 
 as chlorides. For this reason a little neutral barium chloride 
 solution should be added. 
 
 407. Determination of Total Amount of Carbon Dioxide in the 
 Presence of Magnesium. If magnesium bicarbonate is present, 
 an error is introduced into the determination as shown by the 
 following equations: 
 
 +Ba(OH) 2 =BaC0 3 +MgC0 3 +2H 2 0; 
 MgC0 3 +Ba(OH) 2 =BaC0 3 +Mg(OH) 2 . 
 
 As the magnesium hydroxide is insoluble the alkalinity of the 
 solution has been decreased by 2 molecules of barium hydroxide, 
 whereas with other bicarbonates but 1 molecule is neutralized. 
 To remedy this difficulty Pettenkofer added 2 c.c. of a saturated 
 solution of ammonium chloride so that the magnesium hydroxide 
 should remain in solution. According to Trillion* this is only 
 a partial remedy of the difficulty. He proposes to omit the addi- 
 tion of the ammonium chloride and apply a correction to the 
 result from a separate determination of magnesium. 
 
 The DETERMINATION WAS CARRIED OUT according to Petten- 
 kofer as follows: 100 c.c. of the water are placed in a flask and to 
 it are added 3 c.c. of a 10% solution of calcium or barium chloride 
 and 2 c.c. of a saturated solution of ammonium chloride, as well 
 as 45 c.c. of a solution of barium hydroxide of such a strength 
 that 1 c.c. =1 mg. of carbon dioxide. The flask is then corked 
 and allowed to stand for twelve hours. 50 c.c. of the clear liquid 
 are withdrawn and the excess of alkali titrated with standard 
 acid. Trillich omits the addition of ammonium chloride, so that 
 the magnesium hydroxide is entirely precipitated. The amount 
 of magnesium hydroxide is calculated from a gravimetric deter- 
 mination, and its equivalent in barium hydroxide subtracted 
 from the amount of barium hydroxide neutralized by the water. 
 
 408. Precipitation of Carbon Dioxide as Calcium Carbonate. 
 When water contains a considerable amount of carbon dioxide 
 so that there is danger of loss during the manipulation, a sample 
 should be collected in a measuring-flask which contains a measured 
 volume of an alkaline solution of calcium chloride made as follows: 
 * Zeit. f. angew. Chem., 1889, pp. 117, 337. 
 
300 VOLUMETRIC METHODS. 
 
 To one part of crystallized calcium chloride 5 parts of water are 
 added and 10 parts of ammonium hydroxide sp. gr. 0.96. This 
 solution is placed in well-stoppered bottles and allowed to stand 
 several days. The clear supernatant liquid is drawn off as needed. 
 On the addition of this reagent to carbonated water all of the 
 carbonic acid present is precipitated as calcium carbonate. The 
 precipitate is at first amorphous, but on standing for at least 
 twelve hours it becomes crystalline. This change may be hastened 
 by warming on the water-bath. The precipitate must not be 
 filtered off until it has been digested in this manner or stood at 
 least 12 hours. It is then filtered off and washed free from alkali. 
 The carbon dioxide may now be determined in the calcium car- 
 bonate by various methods. A very simple method consists 
 in dissolving the carbonate in excess of standard acid and titrating 
 back the excess with standard alkali. 
 
 409. If Bottled Carbonated Waters are to be Examined, the 
 bottles may be opened so as not to lose carbon dioxide by a very 
 simple device. An ordinary brass cork-borer of suitable size is 
 taken, and several small holes are drilled a little farther from the 
 cutting end than the length of the cork to be pierced. A rubber 
 tube is securely fastened to the upper end of the borer, and con- 
 nected to a convenient absorption apparatus containing some of 
 the alkaline calcium chloride solution. The cork of the bottle 
 is pierced with the cork-borer by holding the latter firmly in the 
 hand and turning the bottle around. When one of the small 
 holes in the borer has been brought below the cork the carbon 
 dioxide will pass out through the cork-borer and into the absorp- 
 tion apparatus. When no more carbon dioxide is evolved the 
 bottle is warmed by placing it in lukewarm water, which is then 
 brought nearly to a boil. The absorption apparatus is then dis- 
 connected, and the carbon dioxide swept out of the connecting 
 tube by means of air freed from carbon dioxide. The contents of 
 the bottle are added to the liquid in the absorption apparatus, 
 the solution digested on the water-bath for some time, and finally 
 the calcium carbonate is filtered off and washed. The calcium 
 carbonate may then be titrated with standard acid and alkali. 
 
 410. Determination of Carbon Dioxide in Gases. When car- 
 bon dioxide occurs in considerable amount in mixtures of gases 
 
TITRATION OF CARBONIC ACID. 301 
 
 it is determined by gasometric methods. When it occurs in 
 amounts considerably less than 1% it is customary to determine 
 it gravimetrically by passing the gas through an appropriate 
 series of absorption-tubes so that the carbon dioxide may finally 
 be weighed. More frequently, however, the carbon dioxide is 
 determined volumetrically. 
 
 411. Pettenkofer's Method. Many methods of procedure and 
 forms of apparatus have been proposed, but nearly all of them 
 follow in general the method suggested by Pettenkofer, which 
 consists in absorbing the carbon dioxide in a measured volume 
 of standard alkali and titrating back the excess. Pettenkofer 
 used barium hydroxide and oxalic acid as the standard solutions. 
 He took a large-sized bottle, aspirated the air to be examined 
 through it until the air already present was completely displaced, 
 and added a measured amount of the standard alkali. On shaking 
 the bottle vigorously for some time the carbon dioxide was com- 
 pletely absorbed. The alkali was poured into a cylinder or bottle, 
 the precipitate allowed to settle, a measured volume with- 
 drawn, and the excess of alkali titrated with the oxalic acid, using 
 phenolphthalein as the indicator. 
 
 412. Several Difficulties and Errors are met with in this 
 method. The surface of the glass bottle is capable of holding 
 barium hydroxide so that more alkali is apparently removed from 
 the solution than corresponds to the carbon dioxide present. To 
 overcome this difficulty various remedies have been proposed. 
 The bottle may be rinsed with some of the barium hydroxide 
 solution, washed free from alkali with distilled water, and then 
 dried. A better method much used consists in coating the inside 
 of the bottle with a thin layer of paraffine. This method is suc- 
 cessful where a coat of paraffine can be obtained which does not 
 peel off. Still another method consists in shaking the bottle with 
 moist freshly-precipitated barium carbonate. A thin coating of 
 this salt adheres quite firmly to the glass, and acts quite efficiently 
 as a protective agent. 
 
 413. Apparatus. Another source of error is found in the 
 contamination of the barium hydroxide solution with carbon 
 dioxide while being poured from the large bottle into a suitable 
 vessel for allowing the precipitate to settle. In this case the 
 
302 
 
 VOLUMETRIC METHODS. 
 
 most serious contaminating influence is the breath of the operator. 
 A form of apparatus recently described by Sewaschen * seems 
 well adapted to overcome this difficulty. The large bottle A has 
 a capacity of about 6 liters. The exact volume is carefully deter- 
 mined, most conveniently by the Morse-Blalock bulbs. The vol- 
 ume is taken to a mark c on the neck, to which the rubber stopper 
 B must always be inserted. The small bottle C should have a 
 capacity of very nearly 100 c.c., the exact capacity having been 
 
 6000 cc 
 
 FIG. 51. 
 
 carefully determined when the stopper D is inserted to the mark 
 d on the neck, the bottle being filled so that no air-bubbles are 
 left under the stopper and the glass rod b inserted. The volume 
 of the neck from the point d must also be determined and added 
 to that of the large bottle. The neck of the small bottle must be 
 free from any projecting ridge, and must fit quite snugly into the 
 opening in the stopper B. If a suitable 100-c.c. bottle is not at 
 hand, a stout 100-c.c. Florence flask may be used. The projecting 
 rim on the neck may be quite readily cut off by means of a dia- 
 mond, a small flame, or other glass-cutting device. 
 
 414. Manipulation. To use this apparatus, the bottle A is 
 cleaned and dried and the inner surface treated by one of the 
 methods already given. The air to be examined is aspirated 
 through the bottle after the stopper B is inserted. While the 
 large bottle is being filled with air the standard solution of barium 
 hydroxide is measured out into the small bottle, the stopper D 
 
 * Hygienische Rundschau, No. 9, 1897. 
 
TITRATION OF CARBONIC ACID. 303 
 
 inserted, and finally the rod b. The large bottle is turned on its 
 side and the neck of the small bottle inserted into the stopper B 
 after removing the stopper D. On turning the large bottle upright 
 the barium hydroxide solution flows out of the small bottle and 
 comes in contact with the air in the large bottle. The latter is 
 shaken occasionally for about two hours, when the absorption of 
 the carbon dioxide will be complete. By inverting the large 
 bottle the barium hydroxide solution will flow into the small 
 bottle, which is then disconnected from the large bottle, the 
 stopper D inserted, and the barium carbonate allowed to settle, 
 which requires from twelve to twenty-four hours. Measured 
 volumes are then withdrawn and titrated with the standard 
 acid. 
 
 415. Barium Hydroxide has been used most largely as the 
 alkaline solution in which to absorb the carbon dioxide. The 
 barium hydroxide is seldom free from traces of the alkalies, which 
 exert a disturbing influence on the titration when oxalic acid is 
 used and barium carbonate is present. The potassium or sodium 
 oxalate formed in the solution reacts with the barium carbonate 
 as follows: 
 
 K 2 C 2 4 + BaC0 3 =BaC 2 4 + K 2 C0 3 . 
 
 The potassium carbonate is converted into oxalate by the further 
 addition of oxalic acid. Carbon dioxide is liberated, and the 
 potassium oxalate reacts with more of the barium carbonate, 
 according to the equation given. More oxalic acid is therefore used 
 than corresponds to the amount of free alkali originally present. 
 This action is entirely prevented by the addition of a little barium 
 chloride to the solution, which converts any potassium or sodium 
 salts present into chlorides. The solution must be carefully pro- 
 tected from carbon dioxide by one of the methods already given. 
 It is standardized by titration against standard acid. 
 
 416. Oxalic Acid has been most largely used to titrate the 
 excess of the alkali. 2.8636 * grams of the pure recrystallized acid 
 dissolved in one liter of distilleJ water gives a solution 1 c.c. of 
 which is equal to 1 mg. of carbon dioxide. As dilute solutions 
 of oxalic acid are not stable, a fresh solution must be made as 
 needed by weighing out the crystallized acid. 
 
 * Note to student. Show how this number is calculated. 
 
304 VOLUMETRIC METHODS. 
 
 417. Sulphuric Acid of equivalent strength may also be used 
 with advantage. The presence of barium carbonate does not inter- 
 fere so much with the titration when sulphuric acid is used. Some 
 workers who use this acid titrate the excess of the barium hy- 
 droxide without removing any of the barium carbonate. Alkalies, 
 if present, do not interfere with the titration if sulphuric acid is 
 used. It may be made by diluting a stronger solution. 45.45 
 c.c. of a normal solution or 227.27* c.c. of a fifth-normal acid 
 diluted to a liter will give an acid 1 c.c. of which is equal to 
 1 mg. of carbon dioxide. The acid may also be made by diluting 
 concentrated acid so as to make an approximate solution, which is 
 then standardized by titration with the barium hydroxide solu- 
 tion, which in turn is standardized by titration against weighed 
 amounts of pure crystallized oxrJic acid. Phenolphthalein is used 
 as the indicator in all titrations. 
 
 418. The Results of the Analysis must be expressed in per 
 cent by volume, the amount of carbon dioxide in the air being 
 given in parts by volume in 10,000. The temperature of the air 
 and the barometric pressure must be taken at the time the large 
 bottle is filled. In very careful work the tension of water vapor in 
 the air must also be taken. The volume of the air at and 760 mm. 
 pressure must then be calculated according to the formula 
 
 V(P v) 
 
 7 o- 760 = (l + 0.00366670760' where V is the V lume f the krge 
 bottle used, P is the barometric pressure in millimeters, and t the 
 temperature when the sample of air is taken, p is the tension of 
 water vapor in millimeters of mercury. The amount of carbon 
 dioxide, being obtained in milligrams, must be reduced to cubic 
 centimeters by multiplying the number of milligrams found by .508. 
 
 EXERCISE 55. 
 Determination of Carbon Dioxide in the Air. 
 
 419. Standard Solutions. Prepare a saturated solution of barium hy- 
 droxide by treating 15 to 20 grams of barium hydroxide with warm dis- 
 tilled water. Place in a well-stoppered bottle, and when the insoluble 
 matter has settled, draw off by a siphon or large pipette about 250 c.c. 
 and dilute to 1 liter. Protect the solution from the carbon dioxide of the 
 air by placing the solution in a bottle arranged as shown in Fig. 50, p. 290. 
 
 * Note to student. Show how this number is calculated. 
 
TITRATION OF CARBONIC ACID. 305 
 
 Measure out 227.27 c.c. of N/5 sulphuric acid by means of a burette 
 and dilute to 1 liter. If standard sulphuric acid is not at hand, weigh out 
 2.8647 grams of oxalic acid and dilute to a liter. If the oxalic acid is used, 
 about 0.2 gram of barium chloride must be added to the barium hydroxide 
 solution. 
 
 The bottles described on p. 302 are prepared : 
 
 420. Collecting the Sample. The rubber stopper is inserted in the large 
 bottle A, and into the hole of the stopper is inserted a smaller two-holed 
 rubber stopper through which two glass tubes pass, one of which extends 
 to the bottom of the bottle, while the other passes through the stopper 
 only. The tube extending to the bottom of the bottle is connected to a 
 long tube passing out of a window. The smaller glass tube is connected 
 to a Bunsen suction-pump, which is allowed to draw air through the bottle 
 for half an hour. The small bottle is filled with the barium hydroxide 
 solution, the rubber stopper inserted so as not to enclose any air-bubbles, 
 and after pressing it down to the mark, the excess of barium solution is 
 removed and the glass plug inserted. 
 
 When the large bottle has been filled with the outside air, the stopper 
 containing the glass tubes is removed. The stopper is also removed from the 
 small bottle, which is inserted into the hole in the stopper of the large bottle, 
 which for this purpose is placed on its side. The temperature is taken from 
 a thermometer which has been hung by the large bottle. The barometer is 
 also read. The large bottle is shaken at intervals for about two hours. 
 It is then inverted, the small bottle containing the barium hydroxide 
 solution is removed, and the stopper inserted. 
 
 421. Titration. In the meantime the standard solutions are compared, 
 the titrations being carried out rapidly to prevent absorption of carbonic 
 acid. Phenolphthalein is used as the indicator. If oxalic acid is used, the 
 solution in the small bottle must stand overnight before being titrated. 
 If sulphuric acid is used, it may be titrated when convenient, the glass plug 
 being withdrawn from the stopper, and 25 c.c. of the solution withdrawn 
 by means of a pipette and titrated immediately. Two or three titrations 
 should be made with 25-c.c. portions. The difference between the amount 
 of acid used for these 25-c.c. portions and the amount used in the stand- 
 ardization for 25 c.c. is found and multiplied by the ratio of the volume of 
 the small bottle to 25. This gives the number of cubic centimeters of 
 acid equivalent to the carbon dioxide in the sample of air taken. Calculate 
 the number of cubic centimeters of carbon dioxide by multiplying the num- 
 ber of milligrams by .508. Calculate the volume of air taken at 760 mm. 
 and 0. Divide the former by the latter and multiply by 10,000 to find 
 the number of parts of carbon dioxide per 10,000. 
 
OXIDATION AND REDUCTION METHODS. 
 
 CHAPTER XXIV. 
 
 POTASSIUM PERMANGANATE AND BICHROMATE 
 SOLUTIONS. 
 
 422. Oxidation. A number of very excellent volumetric 
 methods are based on the fact that many substances exist in two 
 states of oxidation. In the case of arsenic, we have arsenious 
 oxide, As 2 3 , and arsenic oxide, As 2 5 . Sulphur forms several 
 compounds of varying oxygen content, as sulphurous acid, H 2 S0 3 , 
 and sulphuric acid, H 2 S0 4 . Substances capable of giving up oxy- 
 gen, such as POTASSIUM PERMANGANATE, KMn0 4 , convert the former 
 into the latter. The oxidizing substance itself may not contain 
 oxygen, but may be able to oxidize by acting on water or some inter- 
 mediary substance. IODINE is such a substance which is capable 
 of converting sodium arsenite into sodium arsenate, as follows: 
 
 Na 3 As0 3 + 1,+ H 2 = NasAs0 4 + 2HI. 
 
 An alkali must be present to unite with the hydriodic acid so 
 that the action might be represented as follows: 
 
 Na 3 As0 3 + 1 2 + 2NaOH = Na 3 As0 4 + 2NaI + H 2 0. 
 
 The mechanism of the reaction is probably quite complicated, 
 as many reactions are possible under the circumstances. It may, 
 however, be carried out so that 2 atoms of iodine are equal to 
 1 atom of oxygen. 
 
 The result of the oxidation may not even be the addition of 
 oxygen, as when ferrous chloride, FeCl 2 , is converted into ferric 
 chloride, FeCl 3 . Perhaps the simplest DEFINITION of OXIDATION 
 is the change in valence of an element by the addition of negative 
 atoms or radicles or the removal of positive atoms or radicles. The 
 change in valence may not always be an increase, though it 
 generally is. The oxidation of ammonia may result in a decrease 
 
 306 
 
AVAILABLE OXYGEN. 307 
 
 of valence. It is apparent that the reactions taking place during 
 oxidation and reduction are more complicated than those involved 
 in acidimetry and alkalimetry. 
 
 423. Normal Oxidizing and Reducing Solutions. The strength 
 of all oxidizing and reducing solutions may nevertheless be 
 expressed in as simple terms as the acid and alkaline solutions. 
 The basis selected is oxygen, the strength of all standard oxidizing 
 solutions being expressed in terms of AVAILABLE OXYGEN or its 
 equivalent in chlorine, iodine, etc. A normal oxidizing solution is 
 defined as one which contains 8 grams of available oxygen or its 
 equivalent per liter. HYDROGEN has been selected as the basis of 
 reducing solutions, 1 liter of a normal reducing solution containing 
 1 gram of available hydrogen or its equivalent. 
 
 424. Available Oxygen. The distinction between available 
 and non-available oxygen must be carefully made, just as the 
 acid hydrogens of a molecule must be distinguished from the 
 non-acid atoms in acidimetry. Of the 4 atoms of oxygen in the 
 molecule of POTASSIUM PERMANGANATE but 2J are available. 
 The decomposition of 2 molecules of this compound may be 
 represented as follows: 
 
 2KMn0 4 = K 2 + 2MnO + 50. 
 
 The oxidation is generally carried out in acid solution, the potas- 
 sium oxide and manganous oxide combining with the acid as 
 follows : 
 
 K 2 0+ 2MnO +3H 2 S0 4 =K 2 S0 4 + 2MnS0 4 +3H 2 0. 
 
 Five atoms of oxygen, or 80 parts by weight, being free for oxida- 
 tion purposes, one-tenth of the gram molecular weight of 2 mole- 
 cules of potassium permanganate, or 31.63 grams must be dis- 
 solved in 1 liter to make a normal solution. 
 
 The decomposition of POTASSIUM BICHROMATE is similar: 
 
 K 2 Cr 2 7 =K 2 +Cr 2 3 +30. 
 
 This substance is also used in acid solution, the potassium and 
 chromium oxides reacting with sulphuric acid as follows: 
 
 K 2 +Cr 2 3 + 4H 2 S0 4 =K 2 S0 4 + Cr 2 (S0 4 ) 3 +4H 2 0. 
 
 As 3 atoms of oxygen, or 48 grams, are available, one-sixth of 
 the gram molecular weight of potassium dichromate must be dis- 
 
308 VOLUMETRIC METHODS. 
 
 solved in a liter to obtain a solution containing 8 grams of avail- 
 able oxygen per liter. IODINE reacts as follows in the presence 
 of a substance capable of absorbing oxygen: 
 
 I 2 +H 2 0=2HI-K). 
 
 Two atoms of iodine liberate 1 atom of oxygen. The weight of 
 1 atom of iodine in grams must therefore be dissolved in a liter 
 to give a normal solution. 
 
 425. Indicators. No indicator is known which directly shows 
 the presence or absence of available oxygen or hydrogen in a 
 solution. POTASSIUM PERMANGANATE serves as its own indicator. 
 As it has an intense reddish-purple color in water solution, and 
 when reduced in acid solution forms the colorless potassium 
 and manganous sulphates, the least excess of the oxidizing solu- 
 tion is indicated by the characteristic permanganate color. As 
 long as even a trace of a reducing substance is present the color 
 fades out by reduction of the permanganate. 
 
 As POTASSIUM DICHROMATE is yellow and on being reduced 
 gives the green chromic salts it is impossible to detect the presence 
 of a small amount of the yellow dichromate in the presence of 
 the large amount of green chromic salts. The dichromate solu- 
 tion is generally used with a ferrous iron solution as the reducing 
 substance. In acid solution potassium dichromate instantly 
 converts any ferrous iron present into ferric iron. A very delicate 
 test for ferrous iron is found in the blue color which even very 
 small amounts produce in a solution of POTASSIUM FERRIC YANIDE. 
 A solution of ferrous iron to which dichromate solution has 
 been added may be tested for the presence of the former by 
 taking out a drop of the solution and adding it to a drop of the 
 indicator. When, after the successive addition of small amounts 
 of dichromate solution, a test for ferrous iron is no longer obtained, 
 the tit ration of the ferrous iron is complete. 
 
 Solutions of IODINE are deep red to yellow according to the 
 concentration. After reduction the iodine forms colorless salts. 
 If the substance oxidized also forms colorless salts the titration 
 may be conducted by using the IODINE COLOR as the indicator. 
 This gives fairly accurate results, but a much more delicate indi- 
 cator for iodometric work is found in STARCH SOLUTION, which 
 
POTASSIUM PERMANGANATE. 309 
 
 gives an intense blue color when the least excess of free iodine 
 is present. Because of the delicacy of this method of obtaining 
 the end-point, iodometric titrations are among the most accurate 
 known to the chemist. 
 
 POTASSIUM PERMANGANATE. 
 
 426. Decomposition of the Permanganate by Manganese Diox- 
 ide. This salt may now be purchased in a high state of purity. 
 As it is not hygroscopic, solutions made by weighing out the C.P. 
 article and dissolving in water are accurate within less than J%. 
 Sulphates, nitrates, and chlorides are the common impurities 
 which must be absent from the C.P. article. It is almost impossi- 
 ble, however, to prepare a sample of this salt which is free from 
 small amounts of manganese dioxide. The injurious effect of 
 this impurity on solutions of potassium permanganate has been 
 clearly shown by Morse and his pupils.* He has proved that 
 the permanganate is slowly decomposed by even traces of the 
 dioxide with liberation of oxygen and formation of more of the 
 dioxide. As the rate of decomposition is proportional to the 
 amount of the dioxide, the decomposition is accelerated with the 
 age of the solution. The remedy consists in the complete removal 
 of the dioxide. This is easily accomplished by filtering the per- 
 manganate solution through asbestos which has been digested with 
 strong aqua regia and well washed. Organic matter, dust, etc., 
 which may reduce the permanganate with formation of the diox- 
 ide must be rigorously excluded from the solution after nitration. 
 All bottles and glassware with which the solution may come in 
 contact must be freed from organic matter by washing with 
 dichr ornate cleaning mixture. As distilled water usually contains 
 ammonia and organic matter, it is well to allow the permanganate 
 solution to stand some time before filtration so as to entirely 
 decompose this organic matter. The distilled water may also be 
 purified by adding a very little potassium permanganate, boiling 
 until the permanganate is decomposed, and then filtering off the 
 manganese dioxide through asbestos. If the permanganate solution 
 is made up in this manner and kept out of strong sunlight it is very 
 
 *Am. Chem. Jour., 18, 411; 23, 313. 
 
VOLUMETRIC METHODS. 
 
 permanent, readily keeping its strength for months. Portions of the 
 solution withdrawn from the bottle should not be returned to it. 
 
 STANDARDIZATION OF POTASSIUM PERMANGANATE 
 
 SOLUTIONS. 
 
 427. By Pure Iron Wire. As has been said, permanganate 
 solutions made by weighing out the pure salt are within at least 
 J% of the calculated strength. A higher degree of accuracy is 
 obtained by carefully standardizing the solution. The most largely 
 used and perhaps the most reliable method consists in the use of 
 known amounts of iron. Soft-iron wire containirg from 99.6 to 
 99.8% of iron may readily be obtained. The exact percentage of 
 iron may be obtained by a gravimetric determination of the iron as 
 given in Exercise 11, or of the total carbon, as given on page 374, the 
 iron being found by difference. Formerly the practice was com- 
 mon of dissolving the iron in dilute sulphuric acid, the air having 
 been excluded by means of carbon dioxide. Under these circum- 
 stances the iron is converted into ferrous sulphate, which was 
 then titrated with the permanganate. As one of the impurities 
 in the iron is carbon which is combined to a greater or less extent 
 as a carbide of iron, the solution of the wire in acid leads to the 
 formation of hydrocarbons, which are oxidized by the permangan- 
 ate. As one part of carbon may reduce ten and one-half parts of 
 potassium permanganate, and if the carbon is present as a hydro- 
 carbon it may reduce twice this amount of the permanganate, 
 the error from this source, even if the amount of carbon present 
 is very small, may be very considerable. Titrations of solutions 
 of iron wire have been made which indicate as much as 104% of 
 metal in the wire if calculated on the assumption that the only 
 reducing substance present is ferrous iron. Blair recommends 
 the oxidation of the hydrocarbons by means of potassium chlo- 
 rate * or potassium permanganate. f If this method is pursued 
 the iron need not be dissolved in an atmosphere free from oxygen, 
 as it must be reduced after oxidation of the organic material. If 
 the chlorate is used the iron is dissolved in hydrochloric acid, a 
 few crystals of potassium chlorate added, and the excess of 
 
 * Blair, The Chemical Analysis of Iron, 4th Ed., p. 220. 
 t Ibid., p. 95. 
 
POTASSIUM PERMANGANTE. 
 
 311 
 
 chlorine expelled by boiling. If permanganate is used the iron 
 may be dissolved in sulphuric acid. 
 
 428. By Standard Iron Solution. A very excellent reagent 
 for standardizing permanganate' solutions is recommended by 
 Blair.* 100 grams of pure wrought iron free from manganese and 
 arsenic, and in which the phosphorus has been determined, is 
 weighed out and dissolved in nitric acid. The solution is evap- 
 orated to dryness and the residue 
 
 heated in a platinum crucible or dish 
 as hot as possible with the blast-lamp. 
 The ignited material is ground in an 
 agate mortar, dissolved in hydro- 
 chloric acid, evaporated to dryness 
 to dehydrate the silica, taken up 
 with hydrochloric acid, and the silica 
 filtered off. The filtrate is diluted to 
 about 4 liters. Portions of this solu- 
 tion of 15 to 25 c.c. are weighed in 
 small flasks and the iron precipitated 
 and weighed as ferric oxide. The 
 weight of the phosphoric acid must 
 be subtracted from the weight of the 
 oxide. The percentage of phosphorus 
 in the iron multiplied by 1.6f will give 
 the percentage by which the weight of 
 the ferric oxide should be reduced. After determining in this man- 
 ner the exact amount of iron in a weighed portion of the solution, 
 the permanganate may be standardized by taking weighed por- 
 tions of the ferric chloride solution, reducing the iron, and titrating. 
 
 429. Reduction of Iron Solutions. Various methods of reduc- 
 ing the solutions of iron are in use. The most common substance 
 used is GRANULATED MOSSY ZINC. The solution of the iron is placed 
 in a flask of about 250 c.c. capacity. It may be closed with a 
 Bunsen valve, made by cutting a slit a quarter of an inch long with 
 a sharp knife in a small rubber tube which is then closed at one 
 end with a glass plug, while the other end is passed over a short 
 piece of glass tubing which passes through the rubber stopper 
 
 * Blair, The Chemical Analysis of Iron, 4th Ed., p. 218. 
 f Note to student. How is this number obtained ? 
 
 FIG. 52. 
 
312 
 
 VOLUMETRIC METHODS. 
 
 into the flask, as shown in Fig. 52. The gases within the flask can 
 readily pass out through this valve, while the air cannot enter. 
 The air in the flask is expelled by adding a little sodium carbonate. 
 A few grams of zinc are then added, and the solution warmed 
 nearly to boiling. The sides of the flask must not be heated with 
 the flame of the burner above the point filled by the solution. 
 If hydrochloric acid is present, the complete reduction of the iron 
 is indicated by the change in color from the yellow of the ferric 
 chloride to the light green of ferrous chloride. Sulphuric acid is then 
 added and the solution heated until all of the zinc is dissolved. 
 Cold, recently boiled water is then added and the solution allowed 
 to cool. When the stopper is removed from the flask it should 
 be replaced as soon as possible, a little sodium carbonate having 
 been added to expel the air. 
 
 430. Exclusion of the Air. Another very convenient method 
 of excluding the air consists of inserting into the flask a one-holed 
 
 rubber stopper through which passes 
 a glass tube bent to a slightly acute 
 angle, so that when the flask is in- 
 clined the long end of the glass tube 
 may be vertical, as shown in Fig. 53. 
 This end dips into a saturated solu- 
 tion of sodium bicarbonate. On heat- 
 ing the flask the air is expelled, and 
 on cooling a partial vacuum is formed 
 so that the bicarbonate solution is 
 drawn up into the flask. On meeting 
 the acid solution, carbon dioxide is 
 evolved, which prevents the remain- 
 der of the solution from enterirg the 
 flask. After the addition of zinc and 
 FIG. 53. reduction of the iron, sulphuric acid 
 
 is introduced to dissolve the remainder of the zinc. The carbon- 
 ate solution is removed, boiled and cooled distilled water is added, 
 and the iron solution titrated. 
 
 431. Titration of Iron in Zinc. As it is difficult to obtain zinc 
 which is entirely free from iron, the amount of zinc used must be 
 roughly weighed. The same amount of zinc is weighed out and 
 
POTASSIUM PERMANGANATE. 313 
 
 treated with sulphuric acid in a flask exactly as in reducing the 
 iron. After all of the zinc has been dissolved the solution is 
 titrated with the permanganate. The amount of solution used 
 in this blank determination is subtracted from the volume of 
 permanganate used in titrating the iron. 
 
 The REDUCTION of the iron solution by means of STANNOUS 
 
 CHLORIDE, AMMONIUM SULPHITE, and HYDROGEN SULPHIDE are 
 
 more especially adapted to the reduction of iron in iron ores, and 
 will be described under that heading on page 318. These 
 methods are equally well adapted to the reduction of iron solutions 
 for the purpose of standardization. 
 
 432. Ferrous Ammonium Sulphate may also be used for 
 standardizing permanganate solutions. This salt is remarkably 
 permanent. When dry it may be kept for a long time without 
 any appreciable oxidation of the iron. It also retains its water of 
 crystallization with considerable persistence. If a sample of this 
 salt is at hand which is free from ferric iron and has been com- 
 pared with a standard of known purity, it offers a very convenient 
 method of standardizing permanganate solutions, as the salt may 
 be weighed out, dissolved in water, and, after the addition of 
 dilute sulphuric acid, may be titrated with the permanganate 
 immediately. The iron begins to oxidize very soon after being 
 dissolved, so that the solution must be titrated immediately. As 
 this salt may be titrated just as readily with bichromate solutions, 
 its purity may be tested by titrating a weighed portion with a 
 standard solution of potassium dichr ornate. 
 
 433. Oxalic Acid and the Oxalates, such as potassium tetrox- 
 alate prepared and tested exactly as given in Chapter XXI for 
 standardizing alkalies and acids, may be used for standardizing 
 permanganate solutions. It is evident that this class of sub- 
 stances affords a means of comparing standard acids with standard 
 oxidizing solutions. Testing the purity of the oxalic acid by 
 titrating it with a standard acid is of value only when the absence 
 of alkalies from the oxalic acid is assured. This is readily accom- 
 plished by igniting a portion of the acid. It should volatilize 
 completely. The titration of oxalic acid with permanganate 
 solution is generally carried out in a warm solution acidified with 
 sulphuric acid. In the cold the oxidation of the oxalic acid is 
 very slow. It may also be accelerated by the addition of man- 
 
314 VOLUMETRIC METHODS. 
 
 ganous sulphate solution. If the solution is warmed, the temper- 
 ature must not be allowed to exceed 60, as at a higher temperature 
 small amounts of oxalic acid may be volatilized. At this temper- 
 ature the beaker can be taken in the hand without serious discom- 
 fort. The permanganate solution must be added slowly and with 
 vigorous stirring to the oxalic acid solution. Much manganese 
 dioxide must not be allowed to form in the solution or an excess 
 of permanganate will be required, as the manganese dioxide decom- 
 poses some of the permanganate with evolution of gaseous oxygen. 
 
 434. Direct Measurement of Oxygen. An excellent method of 
 standardization which does, not require the preparation of a pure 
 substance has been given by G. Lunge.* In this method a measured 
 volume of the permanganate solution is decomposed by means 
 of hydrogen peroxide and sulphuric acid according to the follow- 
 ing reaction: 
 
 2KMnO 4 + 3H 2 S0 4 -h 5H 2 2 =K 2 S0 4 + 2MnS0 4 + 8H 2 + 50 2 . 
 Twice the amount of available oxygen present in the permanga- 
 nate is evolved in the gaseous form. This gas is measured 
 and the weight of KMn0 4 calculated. LUNGE'S NITROMETER is 
 used for carrying out the determination. The accuracy of the 
 determination depends largely on the accuracy with which the 
 nitrometer has been graduated or calibrated. As the solutions 
 used are already saturated with air the evolution of the oxygen 
 is very quickly complete. The gas is measured in the nitrometer 
 over mercury. The details of the process and a diagram of the 
 nitrometer are given in Exercise 56, page 315. 
 
 435. Standard Sulphuric Acid may be used for standardizing a 
 permanganate solution. A measured amount of the standard acid 
 is placed in a beaker and a sufficient amount of hydrogen per- 
 oxide to reduce about 25 c.c. of the permanganate solution is 
 added. The peroxide is titrated with the permanganate solution. 
 The faint red color giving the end-point is removed with a drop 
 of the peroxide. Methyl orange is added and the excess of sul- 
 phuric acid titrated with standard alkali. 
 
 The difference between the amount of acid originally added and 
 that found by titration gives the amount of acid required for the 
 reaction given m the preceding paragraph. It is evident from 
 *Chem. Ind., 1885, 168; Ber., 18, 1072; Zeit. angew. Ch., 1890, 10. 
 
POTASSIUM PERMANGANATE. 
 
 315 
 
 this equation that 25 c.c. of N/5 potassium permanganate will 
 require 15 c.c. of N/5 sulphuric acid. From this relation the 
 strength of the potassium permanganate solution can be calculated. 
 The methods of standardization in which a known amount of 
 iron is used are preferred for obvious reasons wheb^he perman- 
 ganate solution is to be employed for the determinatfon of iron. 
 
 EXERCISE 56. 
 Preparation and Standardization of N/5 Potassium Permanganate Solution. 
 
 436. Preparation and Preservation of the Solution. Weigh out 6.35 grams 
 of KMn0 4 . Warm about half a liter of distilled water to about 50. Add 
 the permanganate with stirring to the warm water. 
 Prepare two carbon filters, using asbestos which has 
 been digested with warm concentrated aqua regia for 
 about an hour and then well washed with water and 
 finally transferred to a stoppered bottle with distilled 
 water. These filters, as well as all vessels with which 
 the permanganate solution is to come in contact, should 
 be freed from organic matter by digestion with cleaning 
 mixture, made by dissolving a few crystals of potassium 
 dichromate in concentrated sulphuric acid. Suck the 
 asbestos down with the filter-pump, being careful to 
 leave a film projecting upwards around the edge. By 
 means of rubber stoppers arrange the filters as shown 
 in Fig. 54 Filter the permanganate solution, being 
 careful not to apply too strong suction for fear of break- 
 ing the liter flask. Rinse the beaker carefully and 
 wash the asbestos until the water comes through color- 
 less. After being washed, the asbestos in the second 
 filter should be very nearly pure white. If much brown 
 manganese dioxide is seen, the solution must be filtered 
 again, the filters being interchanged, and clean asbestos 
 put in what is now the second filter. After cooling, the 
 solution is diluted to the mark with distilled water. This 
 solution should be kept in the dark and out of con- 
 tact with organic matter. Portions taken out of the 
 bottle should not be returned to it. 
 
 T 
 
 STANDARDIZATION. 
 
 437. First Method. By Means of Pure Iron Wire. Clean thoroughly by 
 rubbing with sandpaper and filter-paper some soft iron wire in which the 
 percentage of iron has been determined. When cleaned, wind in a spiral 
 on a lead-pencil with as little contact with the fingers as possible. Weigh 
 carefully about 0.250 gram, place it in a 250-c.c. Florence flask, and dissolve 
 
316 VOLUMETRIC METHODS. 
 
 with gentle heat in 10 c.c. concentrated hydrochloric acid and 20 c.c. (A 
 water. When the iron is dissolved, throw in a few crystals of potassium 
 chlorate and boil gently a few minutes with a small Bunsen-burner flame. 
 Carefully avoid heating with the flame the sides of the flask above the 
 liquid. Add about 2 grams of granulated zinc which is free from iron. 
 Close the mouth of the flask with a rubber stopper through which passes a 
 bent glass tube dipping into a solution of sodium bicarbonate, as shown in 
 Fig. 53. Heat nearly to boiling till the .solution is of a pure green color, free 
 from the slightest tinge of yellow. If the zinc is all dissolved before this 
 point is reached, add another gram and continue heating until all of the iron 
 is reduced. Allow the flask to cool somewhat. The bicarbonate solution 
 will rise in the glass tube and enter the flask only to be expelled by the 
 carbon dioxide evolved when it reaches the acid solution. Add a mixture 
 of 10 c.c. of concentrated sulphuric acid and 20 c.c. of water and again heat 
 until no particles of undissolved zinc remain in the solution or on the sides 
 of the flask. Allow the solution to cool and then add cold but recently 
 boiled distilled water until the flask is about two-thirds full. Titrate 
 immediately with the permanganate solution until a rose-tint is produced 
 which is permanent for two minutes. Repeat the determination until 
 duplicates are obtained agreeing within less than 0.1 c.c. 
 
 438. Second Method. By Means of Mohr's Salt. Weigh out 19.62 grams 
 of pure recrystallized ferrous ammonium sulphate (FeSO 4 .(NH 4 ) 2 S0 4 .6H 2 O). 
 Transfer to a beaker, dissolve in about 100 c.c. of water, and transfer to a 
 250-c.c. flask. Add 40 c.c. dilute sulphuric acid, shake, cool, and 
 dilute to the mark. This solution will be exactly N/5 normal. Titrate imme- 
 diately 25-c.c. portions with the permanganate solution until duplicates 
 are obtained and calculate the strength of the permanganate solution in 
 terms of normal. 
 
 439. Third Method. By Means of Potassium Tetroxalate. Weigh out be- 
 tween 0.3 and 0.35 gram of recrystallized potassium tetroxalate and transfer 
 to a beaker. Add about 50 c.c. of water and a few cubic centimeters of 
 dilute sulphuric acid and dissolve by gently heating the solution to 50 or 
 60. Be very careful not to heat any higher. Add the permanganate 
 solution slowly with constant stirring. If the solution does not clear 
 readily, warm gently again. When nearly all of the oxalic acid has been 
 oxidized, the solution will decolorize the permanganate almost instantly. 
 At this point add the permanganate cautiously until a faint permanent 
 pink is obtained. If a permanent brown coloration or precipitate is 
 obtained, too much permanganate has been added. Repeat until 
 duplicates have been obtained. One part of the oxalate equals 0.4978 part 
 KMnO 4 . 
 
 440. Fourth Method. Determination of Oxygen Evolved with Hydrogen 
 Peroxide. Set up the Lunge nitrometer as shown in Fig. 55. The box 
 is designed to prevent loss of mercury if any is accidentally spilled. 
 Suspend a thermometer near the nitrometer. Fill with mercury and 
 
POTASSIUM PERMANGANATE. 
 
 317 
 
 test the joints for leaks by lowering the level-tube A so as to produce 
 a vacuum in B and noticing if the height of the mercury column in B 
 remains constant for ten minutes. Now raise the level-tube and pro- 
 duce pressure in the apparatus, and again wait ten minutes to see if the 
 mercury column moves. Measure out into a beaker 15 c.c. of the perman- 
 ganate solution, and, after acidifying with 10 c.c. of dilute sulphuric acid, 
 add from a burette hydrogen peroxide quite rapidly with stirring until the 
 
 FIG. 55. 
 
 solution is clear. Repeat the determination, measuring out the hydrogen 
 peroxide solution first. 
 
 Measure out into the bottle C the amount of hydrogen peroxide solution 
 found to be equivalent to 15 c.c. of the permanganate solution, adding a 
 slight excess. Add about 10 c.c. dilute sulphuric acid. Measure into the 
 inner tube 15 c.c. of the permanganate solution. Insert the stopper firmly 
 and bring the mercury in the two tubes to the same level. Turn the stop- 
 cock of the nitrometer through 180 and bring the mercury in B just up to 
 the key of the stop-cock, thus expelling the air through the funnel-tube. 
 
318 VOLUMETRIC METHODS. 
 
 After clamping the level-tube, reverse the key of the stop-cock and observe 
 if the mercury falls or rises on connecting the tube B with the bottle. If 
 it falls a little, air should be drawn out of C by lowering the level-tube. 
 The stop-cock is then reversed, and the air expelled as before. This 
 operation must be repeated until the mercury column in B neither falls nor 
 rises when the key of the stop-cock is reversed. If the mercury in B rises, 
 air must of course be admitted through the funnel-tube. 
 
 The barometer and the thermometer suspended near the gas burette 
 should now be read. The apparatus should then be allowed to stand undis- 
 turbed for five minutes to see if the air in the bottle is at the room tem- 
 perature. Any change in temperature will be indicated by movement of 
 the mercury in B up or down, in which case the amount of air in C must be 
 again adjusted. The bottle is now tilted slightly and gently shaken, so 
 that the two liquids come in contact. The level-tube is lowered so that no 
 considerable pressure is developed in B. The shaking is continued until all 
 of the permanganate is decomposed. The mercury columns are brought to 
 the same level and the apparatus allowed to stand for five minutes. If a 
 change in the volume of the gas has taken place, the bottle is again shaken 
 and the test repeated. When no more gas is evolved, the final adjustment of 
 pressure is made. The mercury surfaces are brought as nearly on a level as 
 possible with the eye. A drop of water is introduced into the funnel-tube 
 and the stop-cock slowly turned. If the pressure within the tube B is not 
 identical with the atmospheric pressure, the drop of water will start to 
 move, the direction indicating in which tube the pressure is greater. By 
 raising or lowering the level-tube slightly, the inequality can generally be 
 overcome in a few minutes. 
 
 When atmospheric pressure has been secured within the tube B, the 
 volume of the gas is carefully read, and after five minutes read again. If 
 no change has occurred, the thermometer and barometer are again read. 
 As the entire determination need not consume more than one hour, neither 
 the temperature nor the pressure usually change an appreciable amount. 
 Compute the volume, at and 760 mm. pressure, of the oxygen evolved. 
 In this computation, the pressure of water vapor must be taken into 
 account by deducting from the observed barometric pressure the vapor 
 tension of water at the observed temperature. 1 c.c. of dry oxygen at 
 and 760 mm. pressure weighs .0014298 gram. 1 gram of oxygen equals 1.977 
 grams KMnO 4 . Repeat the determination until duplicates are obtained. 
 
 DETERMINATION OF IRON IN IRON ORES. 
 
 441. Dissolving the Ore. By digesting the powdered ore on 
 the hot-plate with a mixture of equal parts of water and con- 
 centrated hydrochloric acid the iron in some ores is completely 
 dissolved; but most iron ores contain a small amount of ferrous 
 
DETERMINATION OF IRON 319 
 
 silicate or titanate which is not decomposed by hydrochloric acid. 
 After digesting the ore until the residue is white or unaffected 
 by further digestion with hydrochloric acid, it is filtered off and 
 washed free from iron. After burning the paper the silica is 
 volatilized by treatment with hydrofluoric and sulphuric acids. 
 The iron remains in the crucible and may be dissolved by digestion 
 with hydrochloric acid, or it may be fused with acid potassium 
 sulphate. The insoluble material may also be decomposed by 
 fusion with sodium carbonate. If the ore contains organic mater- 
 ial a little potassium chlorate must be added to the hydrochloric 
 acid. 
 
 442. Reduction of the Iron by Zinc. The solution of the ore 
 may be deoxidized by means of zinc as given for the standardiza- 
 tion of the permanganate solution. This method introduces an 
 error in the determination if titanium is present, as it is partially 
 reduced by the zinc. The reduction by ammonium bisulphite 
 or hydrogen sulphide does not affect the titanium. It is carried 
 out as directed in Exercise 57. 
 
 443. The Reduction of Iron by Means of Stannous Chloride is 
 very rapid and with the addition of phosphoric acid to the solu- 
 tion to remove the color of the ferric chloride the titration is very 
 exact. The hydrochloric acid solution of the iron is heated 
 nearly to boiling and stannous chloride solution added cautiously 
 until the yellow color of the ferric chloride is replaced by the light 
 green of ferrous chloride. A slight excess of slannous chloride 
 is added and then all at once with vigorous shaking of the flask 60 
 c.c. of mercuric chloride are added. 60 c.c. of an acid manganous 
 sulphate solution containing phosphoric acid is added and after 
 diluting the solution with cold water it is titrated immediately 
 with the permanganate solution. The manganous sulphate is 
 introduced to prevent reduction of the permanganate by the 
 hydrochloric acid. 
 
 The stannous chloride solution is made by dissolving 30 grams 
 of tin in 200 to 300 c.c. of strong hydrochloric acid and diluting 
 to 1 liter after filtering through asbestos. 
 
 The mercuric chloride solution is made by dissolving 50 grams 
 of mercuric chloride in 1 liter of water. 
 
 One liter of the manganous sulphate solution should contain 66$- 
 
320 VOLUMETRIC METHODS. 
 
 grams of crystallized manganous sulphate, 333J c.c. of phosphoric 
 acid (sp. gr. 1.3), and 133 c.c. of concentrated sulphuric acid. 
 
 444. Reduction of Iron by Hydrogen Sulphide. The iron may 
 also be reduced with hydrogen sulphide. For this purpose the 
 solution is placed in a flask closed with a two-holed rubber stopper, 
 through which passes a glass tube extending nearly to the bottom 
 of the flask. The solution is heated nearly to boiling and hydrogen 
 sulphide passed through the long glass tube until the iron is com- 
 pletely reduced. The stream of hydrogen sulphide is replaced 
 by a stream of carbon dioxide and the hydrogen sulphide expelled 
 by boiling the solution for a few minutes. The flask is cooled 
 while the stream of carbon dioxide is passing and the iron tit- 
 rated immediately with the permanganate solution. The sulphur 
 is not oxidized by permanganate in the cold.* 
 
 EXERCISE 57. 
 Determination of Ferrous and Total Iron in an Iron Ore. 
 
 445. Solution of the Ore. Weigh out 0.5 gram of the finely powdered ore, 
 transfer to a 100-c.c. beaker, and add 10 c.c. of concentrated hydrochloric 
 acid and 10 c.c. of water. Cover the beaker with a watch-crystal and 
 digest on the hot-plate until the residue is colorless or unaffected by further 
 digestion with the acid. The solution must not be allowed to go to dry- 
 ness. If necessary add a little dilute hydrochloric acid as the solution 
 evaporates. 
 
 Decant the solution into a 250-c.c. flask, transfer the insoluble material 
 to a small filter-paper, and wash with small portions of hot water. Throw 
 the moist paper into a platinum crucible and burn the paper. Add 20 to 
 30 drops of concentrated sulphuric acid and as much hydrofluoric acid 
 and warm gently. If the material dissolves, continue heating the crucible 
 until the hydrofluoric acid is entirely expelled. If solution is not com- 
 plete add, after expulsion of the hydrofluoric acid, about ^ gram of acid 
 potassium sulphate and heat the covered crucible sufficiently to fuse the 
 sulphate. When the material is entirely dissolved, let the crucible cool 
 and dissolve the contents in hot water, if necessary adding a little hydro- 
 chloric acid. 
 
 446. Reduction of the Iron by Means of Sulphurous Acid. In the mean- 
 time the main portion of the ore is reduced by means of ammonium bisul- 
 phite. Ammonia is added until a small permanent precipitate of ferric 
 hydrate is produced. 5 c.c. of a solution of ammonium bisulphite, made by 
 
 * Jour. Am. Chem. Soc., Vol. XVII, p. 78, 1895; Chem. News, Vol. LXXIII, 
 p. 123, 1896. 
 
ANALYSIS OF IRON ORES. 
 
 321 
 
 saturating strong ammonia with sulphur dioxide, is added and, after shaking 
 vigorously, the solution is gently heated until it is colorless. The solution 
 of the material insoluble in hydrochloric acid is now added, as well as a mix- 
 ture of 10 c.c. of concentrated sulphuric acid and 20 c.c. of water. The 
 solution is then boiled until the steam no longer smells of sulphur dioxide. 
 The flask is filled with boiled distilled water without mixing with the solu- 
 tion of iron. The flask may be covered with a small watch-crystal and 
 cooled by placing it in cold water. The solution may then be poured into a 
 large beaker or porcelain dish and titrated with the standard permanganate 
 solution. Calculate the percentage of metallic iron present in the ore. 
 
 DETERMINATION OF FERROUS IRON. 
 
 447. Solution of the Ore. Weigh out ? gram of the ore and transfer 
 to the flask A, which should have a capacity of about 250 c.c. The simplest 
 method of weighing the ore is to use a glass tube about 10 cm. long, having 
 a diameter of about ^ cm. and sealed at 
 one end. Weigh the tube roughly, intro- 
 duce about gram of the ore and weigh 
 carefully. Insert the tube in the neck of 
 the flask, and by gently tapping the tube, 
 deposit the ore on the bottom of the flask. 
 Withdraw the tube and weigh again. 
 The difference between the last two 
 weights is the weight of the ore in the 
 flask. 
 
 Insert the stopper with the tubes, as 
 shown in Fig. 56. Place a solution of 
 sodium bicarbonate in the beaker and 
 pass a stream of carbon dioxide from a 
 Kipp generator through the tube B for 
 about ten minutes. Disconnect the tube B 
 at the joint C and attach a funnel. Pour 
 in 30 c.c. of dilute hydrochloric acid, raising 
 the tube D out of the bicarbonate solution. 
 
 FIG. 56. 
 
 Remove the funnel and again connect the tube B at C and continue passing 
 the current of carbon dioxide. Dissolve the ore in the acid by means of 
 gentle heat from the Bunsen burner. 
 
 448. Titration. When no black particles remain in the insoluble por- 
 tion remove the burner, replace the beaker of bicarbonate solution with a 
 beaker of boiled and cooled distilled water, disconnect the tube B and close C 
 with a clamp. Dissolve 3 grams of zinc in 10 c.c. of concentrated sulphuric 
 acid and 20 c.c. of water. When the solution of the iron is cold, add the acid 
 solution of zinc and titrate at once. 
 
322 VOLUMETRIC METHODS. 
 
 449. Calculation. Compute the percentage of iron present as ferrous 
 oxide. 1 c.c. N/5 potassium permanganate is equal to .01438 gram of fer- 
 rous oxide. Compute the percentage of ferric oxide present. For this 
 purpose subtract the number of cubic centimeters of permanganate solu- 
 tion reduced by the ferrous iron in gram of the ore from the number of 
 cubic centimeters required to titrate the total amount of iron present. The 
 difference gives the number of cubic centimeters of permanganate solution 
 required to titrate the iron present as ferric oxide in ^ gram of ore. 1 c.c. of 
 N/5 permanganate solution is equal to .01598 gram of ferric oxide. 
 
 ANALYSIS OF MANGANESE ORES. 
 
 The only ores of manganese which are of commercial importance 
 are PYROLUSITE arid several related ores which consist essentially 
 of manganese dioxide combined with small amounts cf the lower 
 oxides of manganese together with more or less iron, silica, and 
 other impurities. Formerly the manganese ore was of value 
 chiefly on account of the oxygen present by which free chlorine, 
 bleaching powder, etc., could be produced. Recently an ex- 
 tended use for manganese has been found in connection with 
 the production of iron and steel, so that nine-tenths of all man- 
 ganese ores are said to. be used for this purpose. The valuation of 
 the manganese ore may therefore require the determination of 
 the available oxygen or the percentage of manganese present. 
 
 450. Determination of Available Oxygen. As the oxygen 
 present is used to evolve chlorine, the most rational method of 
 determination would seem to be to treat the ore with hydrochloric 
 acid and determine the amount of chlorine evolved. This may 
 be done with great accuracy, and the process will be described 
 in the chapter on iodometric methods. This method is not so 
 largely used at present as those in which the ore is dissolved 
 in acid ; n the presence of a known amount of a reducing agent, 
 and the excess of the latter determined by means of a standard 
 solution of an oxidizing agent. Oxalic acid and ferrous iron have 
 been most largely used as reducing agents with potassium per- 
 manganate and potassium dichromate respectively as the standard 
 oxidizing solutions. 
 
 The finely powdered material is dried by spreading it out on 
 a watch-crystal and heating for from four to six hours on a water- 
 
DETERMINATION OF MANGANESE. 323 
 
 bath, or in an air-bath maintained at 100. The dried material 
 must be weighed quickly, as it is hygroscopic. If oxalic acid is 
 used the ore may be dissolved in an open beaker, as this reducing 
 agent is not affected by the oxygen of the air. If ferrous iron 
 is used the air must be excluded by one of the methods used during 
 the reduction of iron with zinc. It is advisable, both when using 
 oxalic acid and ferrous iron, which is most conveniently introduced 
 as ferrous sulphate or ferrous ammonium sulphate, to weigh out 
 the dry salts instead of making a standard solution of these re- 
 ducing agents. The reducing solutions, especially of the iron, 
 would rquire standardization every day, while one titration of 
 the dry salt if preserved in a well-stoppered bottle would be suffi- 
 cient. The errors of measuring out portions of a solution are also 
 considerably greater than the errors of weighing a salt. 
 
 DETERMINATION OF MANGANESE BY VOLHARD'S 
 
 METHOD. 
 
 Manganese may be determined very quickly and accurately 
 by the method of Volhard. The manganese must be in neutral 
 solution as a manganous salt. On adding potassium perman- 
 ganate to such a solution all of the manganese present, including 
 that which was added as permanganate, will be precipitated as 
 the dioxide. The precipitate has a fairly strong acid character, 
 so that varying amounts of the bases present in the solution will 
 be carried down with the precipitate. The manganous manga- 
 nese present is apt to be carried down in this manner, forming 
 salts of the general character (MnO^MnO. Less permanganate 
 solution will be used in this case than theory requires. This 
 difficulty is almost entirely overcome by having a considerable 
 amount of zinc sulphate present in the solution. The zinc will 
 then be carried down with the manganese dioxide in preference 
 to the manganous manganese. 
 
 451. Determination of Amount of Error. The extent of the 
 error from this source may be ascertained by a blank determination 
 as follows: A given volume of the permanganate solution, 30 c.c. 
 for instance, is carefully measured out into a beaker and reduced 
 by acidifying with sulphuric acid and adding hydrogen peroxide 
 
324 VOLUMETRIC METHODS. 
 
 in small portions, carefully avoiding excess. The solution is then 
 transferred to a liter flask and neutralized by the addition of 
 zinc oxide. If several grams of zinc have , not been used some 
 neutral zinc sulphate should be added. The solution is then 
 diluted to 400 c.c. and warmed to 80. 19 c.c. of the perman- 
 ganate solution are then added rapidly and, after vigorous shak- 
 ing of the flask, the permanganate is added in 0.1-c.c. portions, 
 until a slight excess is present as shown by the pink color of the 
 solution. The solution is vigorously shaken after each addition 
 of permanganate. On standing a few minutes the precipitate 
 settles so that the color of the solution can be seen on holding 
 the flask up to the light. 
 
 452. Calculation. If the reaction proceeded wholly according 
 to the equation 
 
 2KMn0 4 + 2H 2 +3MnS0 4 =2H 2 S0 4 +K 2 S0 4 +5Mn0 2 , 
 
 exactly 20 c.c. of the permanganate solution would be required 
 to titrate the manganese obtained from the reduction of 30 
 c.c. of the same solution. Generally, a little less than 20 c.c. will 
 be used, the deficiency averaging about 1% of the total volume. 
 The volume of the permanganate solution used in a given titration 
 or the value of 1 c.c. in manganese must be increased in the pro- 
 portion found by the blank determination. 
 
 453. Method of Adding the Permanganate Solution. If the per- 
 manganate solution is added in small portions to the solution of 
 the manganous salt the error will be greater than when the total 
 amount of permanganate is added all at once. If the amount of 
 manganese present is entirely unknown one or more preliminary 
 titrations must be made. In the first the permanganate solution 
 is added several cubic centimeters at a time until the end-point has 
 been reached or passed. In the second nearly the total amount of 
 permanganate is added all at once and the end-point reached by the 
 addition of amounts less than a cubic centimeter. In the third 
 titration the total amount of the permanganate solution less a few 
 tenths of a cubic centimeter is added at once and the end-point 
 reached by adding portions of a tenth of a cubic centimeter or less. 
 
 454. Interfering Metals. Of the other metals which occur in 
 manganese ores IRON is most common, but it does not interfere 
 
DETERMINATION OF MANGANESE. 325 
 
 with the titration of the manganese since it is precipitated on 
 neutralizing the solution with zinc oxide. COPPER is also precipi- 
 tated by means of the zinc oxide. NICKEL and LEAD, if present, 
 produce high results. The lead may be removed by treating the 
 solution with sulphuric acid or may be removed with cobalt and 
 nickel as sulphides. The solution is neutralized with ammonia 
 and ammonium sulphide added. It is then acidified with hydro- 
 chloric acid and filtered from the sulphides of cobalt, nickel, and 
 lead. The hydrogen sulphide is expelled by boiling and the 
 manganese titrated after neutralization of the solution with zinc 
 oxide. CHROMIUM as well as COBALT interferes if present even in 
 small amounts. Manganese may be separated from chromium by 
 precipitation as dioxide from a strong nitric acid solution by 
 means of potassium chlorate. It is filtered off, washed, and, after 
 solution in hydrochloric acid and neutralization with zinc oxide, 
 titrated with the permanganate solution. 
 
 455. Strength of Solution. As the Volhard reaction may be 
 represented as follows, MnO + = Mn0 2 , one liter of a fifth-normal 
 permanganate solution would be er-ual to 1/10 of the molecular 
 weight of Mn0 2 in grams if the oxygen were evolved from the 
 potassium permanganate in the usual manner. As in the Volhard 
 reaction it decomposes as follows, 
 
 2KMn0 4 = K 2 + 2Mn0 2 + 30, 
 
 a fifth-normal solution has only three-fifths of its usual strength. 
 One liter of the fifth-normal solution is therefore equal to 3/50 of 
 the molecular weight of Mn0 2 , i.e., 1 c.c. is equal to .00522 gram 
 of manganese dioxide. 
 
 EXERCISE 58. 
 Determination of Available Oxygen in Pyrolusite. 
 
 (a) By Oxalic Acid. 
 
 The ore is finely powdered, spread out on a watch-crystal, and dried four 
 to six hours on the water-bath or in an air-bath kept at 100. Portions 
 weighing 0.870 gram are quickly weighed out and transferred to beakers. 
 Portions of pure recrystallized oxalic acid weighing 1.2605 grams are added 
 to the beakers containing the ore. About 50 c.c. of water and a few cubic 
 centimeters of dilute sulphuric acid are added to each beaker and the solu- 
 tion heated gently until the ore is dissolved. The temperature must not 
 
326 VOLUMETRIC METHODS. 
 
 be allowed to rise above 60. Without filtering off the insoluble residue, 
 which is generally white, the warm solution is titrated with standard per- 
 manganate solution. More dilute sulphuric acid is added if the permanganate 
 color fades slowly or manganese dioxide is precipitated. The percentage 
 of Mn0 2 present is found by subtracting the number of cubic centimeters 
 of N/5 potassium permanganate used from 100. 
 
 If pure recrystallized oxalic acid is not at hand, the C.P. article may be 
 used. It is tested as follows: One-gram portions of the oxalic acid are 
 weighed out and titrated against the permanganate solution. The amount 
 of oxalic acid which will reduce exactly 100 c.c. of the permanganate solution 
 is then calculated. This amount is then weighed out instead of 1.2605 grams 
 of the pure acid. It is advisable to test even the most carefully prepared 
 article by titration against the permanganate solution used. 
 
 (6) By a Ferrous Salt. 
 
 Determine the percentage of MnO 2 in the same or another sample of 
 pyrolusite as follows: 0.870 gram of the finely powdered and dried material 
 is weighed out and transferred to a 250-c.c. flask. Add a little distilled 
 water and some dilute sulphuric acid and then a gram or two of sodium 
 bicarbonate. Insert a rubber stopper fitted with a bent glass tube dipping 
 into a solution of sodium bicarbonate exactly as used in the standardization 
 of the permanganate solution with iron wire. An amount of ferrous sul- 
 phate or of ferrous ammonium sulphate which has been found by titration 
 to be exactly equal to 100 c.c. of the permanganate solution is now added. 
 The solution is warmed gently until no more black particles of the pyrolusite 
 remain undissolved. The excess of ferrous sulphate is titrated with the 
 permanganate solution, a red color permanent for one-half minute being 
 taken as the end-point, both in this titration and in testing the purity of 
 the ferrous salt. The number of cubic centimeters of N/5 permanganate 
 used, subtracted from 100 gives the percentage of MnO 2 present. 
 
 EXERCISE 59. 
 Determination of Manganese in Pyrolusite. 
 
 456. Solution of the Ore. 1.5 grams of the dried ore are weighed out 
 and digested with 30 c.c. of concentrated hydrochloric acid until no black 
 particles remain or until no further solvent action is noticed. The insoluble 
 residue is filtered off on a small paper, the solution being allowed to flow 
 into a 250-c.c. flask. After washing the paper free from manganese, the 
 residue is fused with sodium and potassium carbonates unless the absence 
 of manganese has been shown by a previous test. The moist paper is 
 thrown into a platinum crucible and burned. Four to six times as much 
 sodium and potassium carbonates are added and the contents of the crucible 
 
DETERMINATION OF TIN. 327 
 
 fused. The presence of manganese is generally indicated by the charac- 
 teristic green color. Whether the fusion is green or not the fused material 
 is dissolved in water, the solution acidified with hydrochloric acid and 
 added to the main solution of the ore. 
 
 457. Neutralization of the Solution. The solution is diluted to the mark 
 and 50-c.c. portions withdrawn with a calibrated pipette. These portions 
 are placed in Florence flasks of 750- to 1000-c.c. capacity. Zinc oxide is 
 added in small portions with vigorous shaking of the flask until the solution 
 is neutral. At this point the iron which is almost invariably present begins 
 to be precipitated as the hydroxide. The precipitation of the iron is hastened 
 by warming the solution. The zinc oxide should be add?d cautiously until 
 {he solution is colorless, indicating the complete precipitation of the iron. 
 A large excess of the zinc oxide should be avoided. If too much has been 
 added, so that the solution is turbid or the precipitate whitish instead of 
 red, it may be removed by the cautious addition of hydrochloric acid. A 
 slight excess of zinc oxide is not disadvantageous. 
 
 458. Titration. The solution is diluted to about 400 c.c., warmed to 
 80, and titrated with the permanganate solution, adding 4- or 5-c.c. por- 
 tions at once and shaking vigorously. After allowing the precipitate to 
 settle a little, the color of the solution can easily be observed. When the 
 characteristic permanganate color has been given to the solution, the titra- 
 tion is interrupted, and to the second flask, after neutralizing, diluting to 
 400 c.c., and warming, is added the total amount of permanganate solution 
 which was added to the first flask less the final portion which gave the end- 
 point. After thorough shaking, 1-c.c. portions of the permanganate are 
 added until the end-point is reached. To the third flask the total amount of 
 permanganate solution within 1 c.c. is added, and after shaking, the end- 
 point is reached by adding 0.1 -c.c. portions of the permanganate solution. 
 The fourth and fifth portions of the solutions are also titrated. 
 
 Calculate the per cent of manganese in the ore. 
 
 459. Determination of Tin. Tin may readily be determined 
 volumetrically by titrating the ferrous iron produced by the 
 reduction of ferric chloride by stannous chloride or metallic 
 tin. If the tin is obtained as the stannous salt, it is treated 
 with an excess of ferric chloride, when the following reaction 
 takes place: 
 
 2FeCl 3 + SnCl 2 = 2FeCl 2 + SnCl 4 . 
 
 By determining the amount of ferrous iron produced, the amount 
 of tin present in the stannous condition is found. This determina- 
 tion is of value because the tin salt is frequently used as a reducing 
 agent. Cupric chloride may be used in place of the ferric chloride. 
 If the tin is in the metallic condition it may be dissolved 
 
328 VOLUMETRIC METHODS. 
 
 in hydrochloric acid with exclusion of air and the solution of 
 stannous chloride treated with ferric chloride. The metallic tin 
 may also be treated directly with ferric chloride solution, in which 
 it dissolves readily, especially if it is finely divided. No hydrogen 
 is evolved, but the tin dissolves according to the following equa- 
 tion: 
 
 Sn+4FeCl 3 =4FeCl 2 +SnCl 4 . 
 
 If iron is present in the tin the solution in hydrochloric acid 
 must be allowed to stand twelve hours after the addition of metal- 
 lic zinc. The tin is completely precipitated and may be removed 
 from the zinc with a brush, washed, dissolved in ferric chloride 
 and the ferrous iron determined. The tin in stannic compounds 
 may be determined in the same manner, the solution of the stannic 
 compound in hydrochloric acid being treated with zinc as directed 
 above. 
 
 EXERCISE 60. 
 Analysis of Stannous Chloride. 
 
 Make a solution of ferric chloride by dissolving 60 grams of the com- 
 mercial salt in a liter of water. One gram of the stannous chloride is weighed 
 out, placed in a beaker, and 50 c.c. of the ferric chloride solution added together 
 with a few cubic centimeters of dilute hydrochloric acid. 50 c.c. of the ferric 
 chloride solution is placed in a similar beaker and the same amount of 
 hydrochloric acid added. When the stannous chloride is dissolved, add a few 
 grams of sodium sulphate and at least 200 c.c. of boiled and cooled distilled 
 water to each beaker and titrate the tin solution with standard permanganate 
 solution until a faint rose color permanent for one-half minute is obtained. 
 Add permanganate to the ferric chloride solution drop by drop until the 
 same color is obtained. The amount of permanganate added to this 
 solution must be subtracted from the amount added to the solution of the 
 tin. Calculate the amount of crystallized stannous chloride, SnCl 2 .2H 2 O, 
 present. 1 c.c. N/5 permanganate equals .02259 gram of this salt. 
 
 DETERMINATION OF CALCIUM. 
 
 460. Titration of Calcium Oxalate. Calcium may be deter- 
 mined very rapidly and accurately by titrating the oxalic acid in 
 the calcium oxalate precipitate with standard permanganate 
 solution. The calcium may be precipitated in the usual manner 
 by means of ammonium oxalate and digesting the precipitate 
 
DETERMINATION OF CALCIUM. 329 
 
 until the solution is clear. The precipitate is thoroughly washed 
 with hot water. If it is large it is advisable to wash entirely by 
 decantation. The filter-paper through which the wash-water is 
 passed is treated with a little warm dilute sulphuric acid and 
 then washed with water, allowing the liquid to flow into the beaker 
 containing the bulk of the precipitate. When the precipitate 
 on the filter-paper has been entirely dissolved and washed out, 
 the portion in the beaker is dissolved by gently warming the 
 solution and adding more dilute sulphuric acid if necessary. The 
 oxalic acid is then titrated in the usual manner. 
 
 461. Titration of the Excess of Oxalic Acid. The washing 
 and dissolving of the precipitate may be obviated, if the calcium 
 is precipitated hi a measuring-flask of convenient size, by neutral- 
 izing with ammonia and adding a measured excess of oxalic acid. 
 For this purpose a standard solution may be used, or the pure 
 crystals may be weighed out. After digesting the precipitate 
 until the solution is clear it is diluted to the mark on the flask 
 with water and after thorough shaking is allowed to settle a few 
 minutes. As the precipitate occupies a small volume a few drops 
 of water should be added after the solution is up to the mark. 
 The exact amount to be added may be found by dividing the 
 weight of the precipitate by its specific gravity. 
 
 The solution is filtered through a dry paper and the oxalic 
 acid in a measured volume of the filtrate titrated with standard 
 permanganate solution after acidifying with sulphuric acid. The 
 solution need not be cooled to the room temperature if the portion 
 of the filtrate to be titrated is measured out as soon as filtered. 
 A very convenient method of procedure is to precipitate the 
 calcium in a 500-c.c. flask and after digestion allow to cool 
 somewhat, dilute, and shake. After standing a few minutes the 
 solution is filtered through a dry paper into a 250-c.c. measuring- 
 flask. The solution is allowed to flow into this flask until it is 
 filled to a little above the mark. The excess is taken out 'with a 
 glass tube used like a pipette. The solution is then poured into a 
 beaker, the flask rinsed out, and the oxalic acid titrated. 
 
 When a large number of calcium determinations must be 
 made the volumetric method is convenient and rapid. If only a 
 few need be made it is doubtful if time would be saved. 
 
330 VOLUMETRIC METHODS. 
 
 POTASSIUM BICHROMATE SOLUTION. 
 
 462. Conditions of Use. The use of this substance as an oxi- 
 dizing solution is limited by the fact that it does not give up its 
 oxygen to many substances which can be oxidized by potassium 
 permanganate or iodine. The titration must always be so con- 
 ducted that the end-point is obtained by titrating ferrous iron, 
 using potassium ferricyanide as the outside indicator. On the 
 other hand the solution has the advantage that it is very stable, 
 being unaffected by dust or the organic matter which might come 
 in contact with it. It may be used in the Mohr burette, which is 
 closed with a rubber tube and a pinch-cock. Dilute hydrochloric 
 acid is not oxidized as it is by potassium permanganate. The 
 presence of this acid in the solution does not interfere with the 
 titration when carried out cold. Potassium dichromate may 
 also be easily obtained very pure, so that the solution may be made 
 by weighing out the pure salt. Its oxidizing power is consider- 
 ably increased when used in hot concentrated solution. It differs 
 from potassium permanganate in that under these conditions it 
 does not lose an appreciable amount of oxygen in the free con- 
 dition. 
 
 463. Pure Potassium Dichromate may generally be obtained 
 commercially. Sulphates and chlorides are the common impuri- 
 ties. If the pure salt is not at hand, it may be made by recrys- 
 tallizing the impure substance. The salt contains no water of 
 crystallization and is not hygroscopic. It may be fused without 
 loss of oxygen, and this method of drying is frequently used. If 
 dust or organic matter is present, the fused salt is reduced. If 
 heated higher than necessary to fuse it, it may lose oxygen. As 
 the salt dried at 100 loses only an inappreciable amount when 
 fused, .this method of drying seems to be unnecessary. 9.817 
 grams of the pure salt are weighed out, dissolved in water and 
 diluted to a liter to make a fifth-normal solution. 
 
 464. Standardization by Means of Iron Wire. The strength of 
 the solution may be verified by standardization with a known 
 amount of iron as directed for the standardization of potassium 
 
POTASSIUM BICHROMATE. 331 
 
 permanganate solution. The oxidation of the organic matter in 
 the hydrochloric acid solution of the iron wire by means of potas- 
 sium chlorate is unnecessary, as the dichromate is not reduced by 
 the hydrocarbons present. The iron wire may therefore be dis- 
 solved in dilute sulphuric acid, the air being excluded by means 
 of carbon dioxide by the methods already given. If the standard- 
 ized solution of ferric chloride is used, it is not advisable to reduce 
 it by means of zinc, as the zinc salts react with the ferricyanide 
 used as the indicator, forming the white zinc salt which obscures 
 the end-point. The reduction by means of stannous chloride 
 solution is well adapted for use when the iron is to be titrated with 
 dichromate solution. The stannous chloride is added cautiously 
 to the hot ferric chloride solution until the characteristic color 
 of the ferric chloride has disappeared. The slight excess of stan- 
 nous chloride is removed by adding a considerable amount of 
 mercuric chloride solution. After diluting the solution somewhat 
 and adding dilute sulphuric acid the ferrous iron is immediately 
 titrated with the dichromate solution. The reduction with ammo- 
 nium bisulphite and hydrogen sulphide is carried out exactly as 
 directed for titration with the permanganate solution. 
 
 465. Standardization by Means of Pure Ferrous Ammonium 
 Sulphate is carried out in the same manner as when a permanga- 
 nate solution is standardized with this salt, the end-point of the 
 titration being ascertained by means of potassium ferricyanide. 
 The standardization by means of oxalic acid or oxalates is not 
 available for dichromate solutions, nor is the decomposition by 
 means of hydrogen peroxide and measuring the gas evolved. A 
 potassium permanganate solution which has been carefully stand- 
 ardized by any of these methods may be used as a standard for 
 a dichromate solution, using a ferrous salt to make the comparison. 
 In this case the errors of three titrations would be included in the 
 final result. 
 
 466. Analysis of Iron Ores. When a dichromate solution is 
 used for the determination of the iron in an iron ore, it is decom- 
 posed in exactly the same manner as when the iron is titrated 
 with a permanganate solution. Zinc is not used for the reduction, 
 .and when stannous chloride is employed, the addition of the 
 phosphoric acid and the manganous sulphate is unnecessary. As 
 
332 VOLUMETRIC METHODS 
 
 hydrochloric acid is not readily oxidized by potassium dichromate, 
 the solution of iron need not be so thoroughly cooled as when 
 permanganate is used. 
 
 EXERCISE 61. 
 Preparation of Standard N/5 Potassium Bichromate Solution. 
 
 Weigh out 9.817 grams of pure dry potassium dichromate. Dissolve in 
 warm water, transfer the solution to a liter flask, cool, and dilute to the mark. 
 
 467. Standardization by Ferrous Ammonium Sulphate. 1.962 grams of the 
 pure salt are weighed out and dissolved in water, dilute sulphuric acid 
 is added, and the solution titrated immediately with the dichromate 
 solution. 
 
 468. Indicator. The potassium ferricyanide solution to be used as the 
 indicator must be made immediately before it is required for use. A crystal 
 about the size of a pea is placed in a small beaker and 75 to 100 c.c. of dis- 
 tilled water added. The solution is most readily withdrawn by means of a 
 short glass tube. A row of single drops are placed by means of this glass tube 
 on a white porcelain plate. On touching one of these drops with a rod dipped 
 in a solution of ferric chloride only a light-brown color should be produced. 
 If any blue color develops, the ferric chloride should be boiled after the addi- 
 tion of a drop or two of nitric acid and the test repeated. If a blue color is 
 still developed, the ferricyanide is contaminated with ferrocyanide and 
 cannot be used. When an indicator has been obtained which is satisfactory, 
 the progress of the titration is ascertained by touching drops of the indicator 
 with drops of the solution taken out on the stirring-rod. 
 
 469. Titration. To avoid loss of ferrous iron, nearly all of the calcu- 
 lated amount of the dichromate solution should be added to the iron solu- 
 tion before a test is made. Two or three drops of the dichromate solution 
 are then added at a time and the te3t made until only a faint blue color is 
 developed. A drop at a tune is then added until the spot test shows no blue 
 color. No change in the color of the spot test can then be noticed on adding 
 more dichromate to the iron solution. 
 
 470. The Standardization by Means of Soft-iron Wire in which the per- 
 centage of iron has been determined is carried out as follows: Prepare a 
 Bunsen valve as directed on p. 311 and a 250-c.c. Florence flask. Clean 
 some soft-iron wire, in which the percentage of iron has been determined, by 
 rubbing with sandpaper and filter-paper. Make it into a spiral by wind- 
 ing on a lead-pencil. After cleaning do not touch the wire with the fingers. 
 Weigh from 0.25 to 0.3 gram and transfer to the flask. Add about 15 c.c. of 
 dilute sulphuric acid and a gram or two of sodium bicarbonate. Close the 
 flask with the rubber stopper and the Bunsen valve and warm gently until 
 the iron is dissolved. Add cold boiled water until the flask is about half full, 
 
DETERMINATION OF CHROMIUM. 333 
 
 then throw in a gram or two of sodium bicarbonate, and if the solution is not 
 strongly acid, add a few cubic centimeters of dilute sulphuric acid and titrate 
 immediately with the dichromate solution. 
 
 EXERCISE 62. 
 Determination of Iron in an Iron Ore. 
 
 To determine the iron in an iron ore, ^ gram of the powdered ore is 
 weighed and dissolved exactly as directed in Exercise 57, page 320. The 
 solution of the iron is reduced by means of stannous chloride. It is heated 
 nearly to boiling and the stannous chloride solution added drop by drop until 
 the color of the solution is a pure green and a slight excess of the stannous 
 chloride is present. Cool the solution by allowing cold water to flow over 
 the flask. c.e. of mercuric-chloride solution is ndded all at once and 
 the solution vigorously agitated by rotating the flask with the hand. The 
 iron is immediately titrated with the dichromate solution. 1 c.c. of N/5 
 dichromate solution is equal to .01118 gram of iron, or .01598 gram of ferric 
 oxide. Compute the percentage of metallic iron and also of ferric oxide in 
 the ore. 
 
 471. Determination of Chromium in Chrome Iron Ores. 
 These ores are most readily decomposed by fusion with an alkaline 
 oxidizing mixture by which the chromium is converted into chro- 
 mate, which on treatment with water dissolves, leaving the iron 
 as oxide. After destroying the excess of the oxidizing agent, the 
 chromate may be determined by means of a known amount of 
 ferrous iron. The best fusion mixture for this purpose seems to 
 be 4 to 5 parts of sodium hydroxide and 3 to 4 parts of sodium 
 peroxide, or 8 to 10 parts of sodium peroxide alone. The fusion 
 must be conducted in a silver, nickel, or copper crucible, as plati- 
 num is strongly attacked by this mixture. 
 
 EXERCISE 63. 
 Determination of Chromium in Chrome Iron Ore. 
 
 472. Decomposition of the Ore. Grind the ore in an agate mortar until 
 it is reduced to the finest powder. Weigh out gram and transfer to a 
 nickel or copper crucible. Add 4 to 5 grams of sodium peroxide, being care- 
 ful to select the yellow material. The white crust on top is sodium carbonate 
 and oxide resulting from decomposition of the peroxide. Mix the material 
 
334 VOLUMETRIC METHODS. 
 
 well with a platium wire or glass rod. This can be more efficiently done if 
 the peroxide is placed in the crucible first and the ore on top. 
 
 Heat at first with a very small flame. After about ten minutes the 
 material should be entirely liquid. It should be kept in this condition for 
 ten or fifteen minutes. Allow to cool until a crust forms on top, add 1 gram 
 of sodium peroxide, and fuse again for about five minutes. After cooling, 
 transfer the crucible to a porcelain dish and dissolve the fused mass in hot 
 water. Take out and rinse the crucible. If the solution is purple, add a 
 little more sodium peroxide. Boil the solution for ten minutes to decom- 
 pose the sodium peroxide. Filter off and wash thoroughly the insoluble 
 material. Any portion of this material which is not soluble in hydrochloric 
 acid must be fused with a little more sodium peroxide, as it may be unde- 
 composed portions of the ore. 
 
 473. Titration of the Chromic Acid. Acidify the filtrate with dilute sul- 
 phuric acid and add an excess of about 15 c.c. Add a weighed amount of 
 ferrous ammonium sulphate which is more than sufficient to reduce the 
 chromium present. The presence of an excess of the ferrous salt may be 
 ascertained by touching a drop of the ferricyanide indicator with a glass 
 rod which has been dipped into the solution. A sufficient amount of the 
 ferrous salt should be placed in a weighing-bottle and carefully weighed. 
 After adding an excess to the chromate solution the residue is weighed. 
 The amount added is found by difference. 5 grams of the ferrous ammo- 
 nium sulphate will generally be found to be a sufficient amount. The excess 
 of ferrous iron is titrated with standard dichromate solution. 
 
 The value of the ferrous salt is ascertained by titrating 2-gram portions 
 with the dichromate solution. Calculate the volume of the dichromate 
 solution equivalent to 5 grams of the ferrous salt or the amount added to the 
 solution of the chrome ore. The difference between this number and the 
 number of cubic centimeters used in titrating the excess of ferrous iron 
 gives the amount of chromate present in the solution of the ore in cubic 
 centimeters of N/5 dichromate solution. Calculate the percentage of 
 O 2 3 in the ore, 1 c.c. of the N/5 dichromate solution being equal to .005073 
 gram of Cr 2 3 . 
 
 PROBLEMS. 
 
 Problem n. The calcium in \ grain of a sample of dolomite was pre- 
 cipitated with 1 gram of crystallized oxalic acid and the excess titrated 
 with 34 c.c. of N/5 potassium permanganate solution. Calculate the per 
 cent of calcium oxide in the dolomite. How much dolomite and oxalic 
 acid must be taken so that the per cent of calcium oxide will be given by 
 the number of cubic centimeters of N/5 potassium permanganate? 
 
 Problem 12. In the analysis of a sample of stannous chloride, a half- 
 gram portion was treated with ferric chloride solution and titrated with 
 N/5 potassium permanganate solution requiring 20 c.c. The tin in another 
 half-gram portion was weighed as stannic oxide, .330 grams being obtaL,. 1. 
 Calculate the per cent of SnCl 2 .2H 2 O and SnCl 4 present. 
 
CHAPTER XXV. 
 IODOMETHIC METHODS. 
 
 474. Conditions of Use of Iodine Solutions. lodometric meth- 
 ods are largely used, and are very accurate on account of the 
 great delicacy of the indicator available, starch solution giving a 
 distinct blue color with a very small amount of free iodine. Iodine 
 acts as an oxidizing agent in acid and neutral solutions. The caustic 
 alkalies, as well as solutions of the alkali carbonates, combine 
 with the iodine, rendering titration in such solutions impossible. 
 Sodium or potassium bicarbonate is riot acted on by iodine, and 
 as solutions of these salts are alkaline toward all acids stronger 
 than carbonic acid, by working with solutions of these salts, iodine 
 may be used to oxidize many substances which require an alkaline 
 solution. It is therefore possible to use iodine to oxidize a great 
 variety of chemical substances. The solubility of iodine in water 
 solutions of potassium iodide, as well as in alcohol, chloroform, 
 glacial acetic acid, and other organic solvents, still further enlarges 
 its use, since both the iodine and the substance acted on must be 
 in solution. 
 
 The use of iodine is somewhat restricted because its solutions 
 are not stable, while many of the methods of standardization are 
 indirect. The volatility of the iodine also introduces difficulties 
 in its use. Iodine solutions must not come in contact with rubber, 
 cork, or other organic material. Paraffine is unacted on. 
 
 475. Solvents of Iodine. The most commonly used solution 
 of iodine is made by dissolving the element in a water solution of 
 potassium iodide. An unstable compound of the formula, KI 3 is 
 formed under these circumstances. A considerable excess of 
 potassium iodide must be present in order to dissolve the iodine. 
 Alcoholic solutions of iodine have been largely used for the deter- 
 mination of the so-called iodine number of oils. For -this. purpose 
 
 335 
 
336 VOLUMETRIC METHODS. 
 
 a solution in glacial acetic acid has recently superseded the 
 alcoholic solutions formerly used. 
 
 476. Reducing Solutions. As many of the methods of stand- 
 ardizing iodine solutions consist in first standardizing a reducing 
 solution, which is then compared with the iodine solution, the 
 reducing solution must always be at hand in iodometric work. 
 The solution most commonly used for this purpose is SODIUM 
 THIOSULPHATE. The iodine is oxidized according to the following 
 equation : 
 
 2Na 2 S 2 3 +I 2 =Na 2 S 4 6 +2NaI. 
 
 A solution of SODIUM ARSENITE is sometimes used, especially for 
 the titration of bleaching-powder. It reacts with the iodine as 
 follows : 
 
 Na 2 HAs0 3 + 1 3 +2NaHC0 3 = Na 2 HAs0 4 + 2NaI + H 2 + 2C0 2 . 
 
 Neither of these solutions is absolutely stable, though more so than 
 the iodine solutions. The most active decomposing influence on 
 the sodium thiosulphate solution seems to be the action of carbon 
 dioxide in the presence of oxygen and sunlight.* The solution 
 should therefore be made by dissolving the thiosulphate in 
 water from which the carbon dioxide has been expelled by boil- 
 ing. The carbon dioxide may then be excluded from the bottle 
 by means of a soda-lime tube or the solution may be protected 
 from the air by pouring a layer of petroleum-oil over it and 
 siphoning out the solution. Made and protected in one of these 
 ways, the solution will remain unchanged for months. 
 
 477. Preparation of Solutions. Although the commercial re- 
 sublimed iodine may frequently be obtained free from all impuri- 
 ties except moisture, which may be removed by drying in a desicca- 
 tor over sulphuric acid, a given sample cannot be relied on until 
 tested. For this purpose a portion is sublimed according to 
 the direction given in Chapter II, p. 34. A solution made from 
 this resublimed iodine may be titrated against a thiosulphate solu- 
 tion, which is then titrated against a solution of a weighed portion 
 of the unpurified iodine dissolved in potassium iodide solution. 
 
 The sodium thiosulphate cannot be weighed out exactly because 
 of uncertain hydration, unless it has been recrystallized and care- 
 
 * Topf. Zeit. anal. Chem., 26, 150. 
 
STANDARDIZA TION. 337 
 
 fully dried. A salt may sometimes be obtained in this manner 
 which gives solutions of exact strength. 
 
 STANDARDIZATION. 
 
 Neither the iodine nor the sodium thiosulphate solution should 
 be relied on without standardization, although if both solutions 
 are made from carefully purified material, and on comparison by 
 titration are found to be of the strength calculated from the weight 
 of material used, they may be employed without further verifica- 
 tion, except in the most careful work. 
 
 478. Resublimed Iodine. The thiosulphate solution may be 
 standardized by means of small weighed portions of resublimed 
 iodine. As the iodine is hygroscopic, it should be resublimed 
 immediately before use. The loss by volatilization may also be 
 considerable unless the weighed portions are immediately dis- 
 solved in potassium iodide and titrated without delay. 
 
 479. Potassium Bichromate, as well as various other oxidizing 
 substances, have been recommended for use in standardizing the 
 thiosulphate solutions. Potassium dichromate can readily be 
 obtained pure. It will liberate a definite amount of iodine from a 
 potassium iodide solution acidified with hydrochloric acid. The 
 reaction takes place according to the following equation: 
 
 K 2 Cr 2 7 + 6KI + 14HC1 = 8KC1 + 2CrCl 3 + 3I 2 + 7H 2 0. 
 
 A sufficient amount of potassium iodide must be present to dis- 
 solve the iodine liberated, which is then titrated with the thio- 
 sulphate solution. The decomposition of the potassium dichro- 
 mate is not complete unless the potassium iodide, as well as the 
 hydrochloric acid, is present in considerable excess. If a suffi- 
 cient amount of hydrochloric acid is not present, the blue starch 
 iodide color will reappear on adding more hydrochloric acid 
 after titrating the iodine with the thiosulphate solution. The fol- 
 lowing proportions will generally be found satisfactory: To 10 c.c. 
 of fifth-normal potassium-dichromate solution, 10 c.c. of a 10 per 
 cent potassium-iodide solution and 5 c.c. of strong hydrochloric 
 acid is added. On account of the volatility of the iodine, it is 
 advisable to carry out the operation in a glass-stoppered bottle. 
 
 480. Potassium Permanganate reacts more readily with the 
 hydriodic acid. If a standard solution is at hand, a measured 
 
338 VOLUMETRIC METHODS. 
 
 volume may be added to a solution of potassium iodide acidified 
 with hydrochloric acid and the iodine liberated titrated at once. 
 The decomposition takes place according to the following equation: 
 
 KMn0 4 +5KI f 8HC1 = 6KC1 +MnCl 2 +4H 2 +51. 
 
 481. The lodates of Potassium and Sodium have also been 
 recommended for use in a similar manner. They are decomposed 
 by hydriodic acid instantly and completely. The acid potassium 
 iodate is very readily obtained in a high state of purity. This 
 salt has the disadvantage of being sparingly soluble in water, and 
 on account of the large percentage of available oxygen the quanti- 
 ties to be weighed out are small, necessitating very careful weigh- 
 ing. If a large amount is weighed out, the solution must be made 
 up to a definite volume and portions measured out. The decom- 
 position takes place according to the following equation: 
 
 KH(T0 3 ) 2 +10KI+11HC1 = 11KCH-6H 2 0+6I 2 . 
 
 482. Barium Thiosulphate. The iodine solution is standardized 
 by titration with a sodium thiosulphate solution which has been 
 standardized by one of the methods already given. It may be 
 directly standardized by the titration of a weighed amount of pure 
 barium thiosulphate. This salt may be made by mixing together 
 a warm solution of 50 grams of sodium thiosulphate in 300 c.c. of 
 water and 40 grams of barium chloride dissolved in the same 
 amount of water. The solution should be stirred vigorously while 
 the salt crystallizes out. It is filtered off, washed with water until 
 free from chlorides, and dried in the air. It contains one molecule 
 of water of crystallization, its formula being BaS 2 3 .H 2 0. It is 
 very sparingly soluble in water, a solution made by shaking an 
 excess of the salt with water at 17.5 for fifteen minutes being 
 exactly N/100 in strength. Because of its high molecular weight 
 careful weighing is unnecessary. Quantities weighed out for 
 the standardization of iodine solutions are treated with water in 
 a beaker. Because of the slight solubility of the salt, the iodine 
 solution is added to the beaker while undissolved thiosulphate still 
 remains. When all or nearly all of the salt is dissolved, starch 
 solution is added and the addition of the iodine solution continued 
 until a permanent blue color is produced. 
 
 
IODINE SOLUTION. 339 
 
 483. Pure Arsenious Oxide may also be used for the standard- 
 ization of iodine solutions. Weighed portions are dissolved in 
 sodium bicarbonate solution, and after the addition of starch 
 solution the iodine is added until the end-point is reached. 
 
 484. The Starch Solution is made by placing a gram or two 
 of starch in a porcelain mortar, adding a little water, and grind- 
 ing to a smooth paste. 150 or 200 c.c. of boiling water are 
 then poured into the mortar while stirring continually with the 
 pestle. The solution is then poured into a beaker containing an 
 equal volume of cold distilled water. After the solution has 
 been allowed to stand and settle for some time, the clear liquid is 
 decanted for use. This starch solution is sensitive only when 
 fresh and must be prepared each day. If saturated with sodium 
 chloride, or a few drops of chloroform added, it keeps much longer. 
 Various other methods of making a permanent starch solution 
 have been suggested, but the solution made fresh each day seems 
 to be the most satisfactory. 
 
 If the starch is added to a solution containing a considerable 
 amount of free iodine, a compound of starch and iodine is pro- 
 duced which is decomposed very slowly by the thiosulphate. 
 The indicator is therefore added only when nearly all of the free 
 iodine has been removed by the sodium thiosulphate. Potassium 
 iodide as well as free iodine must be present to produce the blue 
 color with starch. As solutions of potassium iodide are not stable, 
 this salt must be kept in the solid form and only freshly made 
 solutions used. 
 
 EXERCISE 64. 
 
 Preparation and Standardization of N/io Iodine and Sodium 
 Thiosulphate Solutions. 
 
 485- Sodium Thiosulphate Solution. If recrystallized and carefully dried 
 sodium thiosulphate is at hand, weigh out carefully 24.83 grams, dissolve 
 in water, and dilute the solution to a liter. If the ordinary C.P. salt is used, 
 weigh out about 25 grams for a liter of solution. In either case use distilled 
 water which has been thoroughly boiled to expel the carbon dioxide and 
 place the solution in a bottle, through the stopper of which a siphon passes 
 for withdrawing the solution. Pour enough naphtha over the solution to 
 form a layer about 1 cm. deep, or put a straight calcium chloride tube in the 
 second hole of the stopper. Fill the tube with fresh soda-lime. 
 
340 VOLUMETRIC METHODS. 
 
 486. Iodine Solution. Weigh 12.697 grams of recently sublimed iodine in 
 a weighing-bottle. Keep the stopper tightly closed, except when putting 
 in or taking out iodine. Only glass, porcelain, or platinum should come in 
 contact with the iodine. Be careful not to spill crystals on the balance-pan 
 or in the case. About 18 grams of pure potassium iodide are weighed out 
 and dissolved in about 150 c.c. of water. The potassium iodide should be 
 tested by dissolving about 1 gram in 10 c.c. of water and acidifying with 
 dilute hydrochloric acid. The solution must be colorless and remain so on 
 the addition of a few cubic centimeters of starch solution. 
 
 The iodine is transferred to a liter flask by removing the stopper of the 
 weighing-bottle and inserting the bottle into the neck of the flask, which 
 for this purpose is held in a horizontal position. On raising the flask to an 
 upright position and tapping the weighing-bottle, the bulk of the iodine will 
 be transferred to the flask. The few remaining crystals are removed by 
 adding a little of the potassium iodide solution and shaking the stoppered 
 weighing-bottle until dissolved. The solution is poured into the flask and 
 the weighing-bottle and stopper are thoroughly rinsed with the potassium 
 iodide solution. The remainder of this solution is added to the liter flask, 
 which is stoppered, and shaken at intervals until all of the iodine is dissolved. 
 If the last of the iodine crystals dissolve slowly, a few crystals of potassium 
 iodide may be dropped into the flask and the concentrated solution formed 
 left undisturbed in contact with the iodine crystals. The solution may also 
 be very gently warmed on the water-bath. 
 
 487. Comparison of Iodine and Sodium Thiosulphate Solutions. When all 
 of the iodine has been dissolved, the solution is diluted to the mark and 
 transferred to a glass-stoppered bottle, which is best kept in a cool, dark place. 
 The solution is immediately compared with the sodium thiosulphate solu- 
 tion. 25 c.c. of iodine solution is measured out into a beaker and the thiosul- 
 phate solution added until the color has been almost entirely removed, 
 leaving the solution a light yellow. A few cubic centimeters of the starch 
 solution are then added and the addition of the thiosulphate solution 
 continued until the intense-blue color is just removed. Until experience is 
 gained, it is advisable to add iodine solution drop by drop until the blue 
 color is restored before reading the burette. There is some danger of over- 
 stepping the mark when the end-point is obtained by removing the blue 
 color with the thiosulphate solution. 
 
 STANDARDIZATION OF THE SOLUTIONS. 
 
 488. First Method. Barium Thiosulphate. Weigh out several portions 
 of pure barium thiosulphate of about gram. If exactly 0.6688-gram por- 
 tions are weighed, each will be equal to 25 c.c. of tenth-normal iodine solu- 
 tion. Transfer the portions to beakers and add from 50 to 100 c.c. of water. 
 Without waiting for the salt to dissolve completely, add the iodine solution 
 
IODINE SOLUTION. 341 
 
 slowly with constant stirring until nearly all of the thiosulphate is dissolved. 
 Starch solution is then added and then iodine solution until a slight perma- 
 nent blue color is produced. 
 
 489. Second Method. Arsenious Oxide. Weigh out 4.95 grams of pure 
 arsenious oxide. Place in a beaker and add about 13 grams of caustic 
 soda. Dissolve in 50 c.c. of distilled water, transfer to a liter flask, and 
 saturate the solution with carbon dioxide. At first the delivery-tube must 
 not be immersed in the liquid, as the latter might be sucked up the tube by 
 the complete absorption of the gas. When the gas-bubbles pass through 
 the liquid undiminished in size, the latter is saturated. Rinse the delivery- 
 tube and dilute the solution to the mark. This solution will be exactly 
 decinormal. Titrate 25-c.c. portions with the iodine solution. 
 
 490. Third Method. Acid Potassium lodate. Weigh out 0.3251 gram of 
 acid potassium iodate. Dissolve in 50 c.c. of water, heating gently if 
 necessary. Transfer the solution to a 100-c.c. flask, rinsing the beaker care- 
 fully, and dilute to the mark. This solution will be exactly decinormal. 
 Measure out 25 c.c. into a beaker, add about 1 gram of potassium iodide 
 dissolved in a little water and a few cubic centimeters of dilute hydro- 
 chloric acid. Titrate immediately with the thiosulphate solution. 
 
 491. Fourth Method. Resublimed Iodine. Weigh out about 0.3 gram of 
 recently sublimed iodine, using a tightly stoppered weighing-bottle. Add 
 10 c.c. of a 10% solution of potassium iodide and shake until the iodine is 
 completely dissolved. Transfer the solution to a beaker and rinse out the 
 weighing-bottle with water and, if necessary, a little potassium iodide solu- 
 tion. Titrate immediately with the thiosulphate solution. 
 
 492. Calculation of the Results. The iodine solution, having been made 
 from pure iodine, must be considered exactly tenth-normal when made, 
 unless some iodine was known to be lost. The titration of the thiosulphate 
 solution immediately after the iodine solution was made is considered as a 
 standardization of the former and is so calculated. The average of the titra- 
 tions which agreed within less than 0.2 or at most 0.3% is taken. If duplicate 
 titrations agreeing within this limit are not readily obtained, some error is 
 being made in the work, which must be discovered by varying the method 
 of procedure. If, for instance, 24.05 c.c. of the thiosulphate solution is 
 found to be equal to 25 c.c. of the iodine solution, the strength of the solu- 
 tions being inversely as the number of cubic centimeters of each used, the 
 strength of the thiosulphate solution will be given by the following pro- 
 portion: 24.05 : 25.00 : : 0.1N : x. z = .10395N. 
 
 By the first method of standardization the iodine solution is compared 
 with a weighed amount of barium thiosulphate, which is exactly equal to 
 25 c.c. of tenth-normal solution. By the same proportion as given above, 
 the strength of the iodine solution may be computed. If, for instance, an 
 average of 24.95 c.c. of iodine solution were used, the iodine solution is 
 .1002N by the proportion 24.95 : 25.00 : : 0.1N : x. As this value differs 
 
342 VOLUMETRIC METHODS. 
 
 from tenth-normal by only 0.2%, the iodine solution is considered tenth- 
 normal, since the errors of titration may amount to 0.2%. 
 
 The results of the second method of standardization are calculated in the 
 same manner. The third as well as the fourth methods being direct 
 standardizations of the thiosulphate solution, its strength is calculated from 
 the results. 
 
 If several days have elapsed between the standardization of the thiosul- 
 phate solution and its comparison with the iodine solution this comparison 
 must be repeated. The average of the values found for the thiosulphate 
 solution is taken and used in the proportion when calculating the strength 
 of the iodine solution. If, for instance, 25.18 c.c. of the iodine solution 
 was found to be equal to 24.12 c.c. of the thiosulphate solution, and the 
 average strength of this solution to be .1039N, the proportion will be 
 25.18 : 24.12 : : ;1039 : x, from which x, or the strength of the iodine 
 solution, will be .09952N. 
 
 493. Analysis of Reducing Agents. All substances which may 
 be analyzed by iodometric methods may be divided into two 
 classes, according to whether they act as reducing agents towards 
 iodine or as oxidizing agents towards hydriodic acid. The reducing 
 agents are analyzed either by titrating directly with standard 
 iodine solution, or by adding an excess of the iodine solution and 
 titrating back with standard thiosulphate solution. The method 
 of titrating a THIOSULPHATE or ARSENIOUS OXIDE has already 
 been given in the methods of standardization. Antimonons oxide 
 may be titrated in exactly the same manner as arsenious oxide, 
 tartaric acid or cream of tartar being used to dissolve the anti- 
 monous oxide. The titration of a SULPHITE will be given in 
 Exercise 66, page 346. The SULPHIDES, which may be completely 
 decomposed with liberation of hydrogen sulphide, may be 
 analyzed in a similar manner, the hydrogen sulphide being 
 passed through excess of standard iodine solution. The follow- 
 ing reaction takes place: H 2 S + I 2 -2HI + S. The amount of 
 unreduced iodine is then determined. Easily decomposed sul- 
 phides, such as those of cadmium and zinc, the alkali .sulphides, 
 and hydrogen sulphide water, are completely decomposed by 
 adding dilute acid and iodine solution. Other reducing sub- 
 stances which can be determined iodometrically are ANTIMONY 
 
 OXIDE, ALKALI CYANIDES, STANNOUS CHLORIDE, and OILS and FATS, 
 
 as will be given in the chapter on this subject. 
 
IODOMETRIC METHODS. 343 
 
 494. Oxidizing Substances, on the contrary, are treated with 
 an excess of potassium iodide, and the solution acidified. The 
 hydriodic acid formed is oxidized to iodine, which combines with 
 the excess of potassium iodide and is then titrated with standard 
 thiosulphate solution. The standardization of the thiosulphate 
 solution by means of ACID POTASSIUM IODATE is an illustration of 
 this method. In this manner FREE CHLORINE, BROMINE, HYPO- 
 
 CHLORITES, BROMATES, FERRIC CHLORIDE, NITROUS ACID, HYDROGEN 
 
 PEROXIDE, and PERSULPHURIC ACID may be analyzed. Oxidizing 
 substances decomposed with difficulty, such as CHLORATES, CHRO- 
 MATES, and minerals containing MANGANESE DIOXIDE, may be 
 analyzed by treatment with concentrated hydrochloric acid and 
 distillation of the chlorine which is absorbed in potassium iodide 
 solution. The iodine liberated is then titrated with standard 
 thiosulphate solution. 
 
 495. Antimony may be determined by first oxidizing with potas- 
 sium chlorate (1 gram) in strong hydrochloric-acid solution. The 
 excess of chlorine is boiled out, the solution cooled, and excess 
 of potassium iodide added (1 gram). The antimony is reduced 
 from Sb 2 Cl 5 to Sb 2 Cl 3 with liberation of iodine which is titrated 
 with thiosulphate solution. Copper and iron must be absent 
 from the solution. 
 
 496. Copper may be determined in a similar manner. The 
 nitric-acid solution of the copper is neutralized with ammonia or 
 sodium carbonate and then acidified with acetic acid. Ten c.c. 
 of a 10 per cent solution of potassium iodide is added to the 
 copper solution, which should not exceed 15 c.c. in volume. 
 Iodine is liberated according to the following reaction: 
 
 Cu (C 2 H 3 2 ) 2 + 2KI = Cul + 1 + 2KC 2 H 3 2 . 
 
 The iodine liberated is titrated with sodium-thiosulphate solu- 
 tion. Lead, bismuth, arsenic, antimony, and iron should be 
 absent from the solution. 
 
 497. Acids may also be determined iodometrically by using a 
 solution of potassium iodate and potassium iodide in the propor- 
 tion of 1 atom of the former to 5 atoms of the latter. On acidi- 
 fying such a solution, the following reaction takes place: 
 
344 
 
 VOLUMETRIC METHODS. 
 
 By titrating the iodine liberated, the amount of acid added may 
 be computed, since 1 atom of iodine is liberated by 1 molecule of 
 a monobasic acid. Bases may also be determined by adding an 
 excess of acid and titrating the iodine liberated after addition of 
 the iodate and iodide solution. In this manner the extremely 
 sensitive indicator of iodometric titrations may be used in acid 
 titrations. Although in this manner a much sharper end-point 
 may be obtained than by the use of the ordinary indicators for 
 acids and bases, the accuracy of a given titration will be no greater 
 unless the measurement of volumes and the purification of reagents 
 is carried out with the same high degree of accuracy. 
 
 498. Bunsen's Distillation Apparatus. For carrying on the 
 distillation of the chlorine from oxidizing agents, such as manga- 
 nese dioxide, various forms of apparatus have been devised. One 
 of the best is that of Bunsen, shown in Fig. 57. The oxidizing 
 
 FIG. 57. 
 
 substance to be analyzed is placed in the small bulb a, which is 
 connected with the tube b by means of a ground-glass joint over 
 which a rubber tube of suitable size is stretched. This joint must 
 be tight, so as to prevent loss of chlorine, which must not, however, 
 be allowed to come in contact with the rubber. The receiver c 
 is half filled with 10% potassium iodide solution, which is decom- 
 posed by the chlorine with liberation of iodine. The chlorine is 
 driven out of the flask a and the tube b with the excess of hydro- 
 chloric acid and the steam produced by boiling the solution. A 
 
ANALYSIS OF MANGANESE DIOXIDE. 
 
 345 
 
 slow stream of carbon dioxide may be produced by placing a few 
 pieces of magnesite in the flask a. This is not necessary, however. 
 499. Mohr's Distillation Apparatus. Another form of apparatus 
 which was devised by Mohr is shown in Fig. 58. In this apparatus 
 the joints are made by cork stoppers coated with paraffine, which 
 
 FIG. 58. 
 
 is not acted on by chlorine. The receiver is a tube about 30 cm. 
 long and 3 cm. wide, which is immersed in a cylinder filled with 
 cold water. The distillation-flask 6 should have a capacity of 
 50 to 75 c.c. 
 
 EXERCISE 65. 
 Determination of Available Oxygen in Pyrolusite. 
 
 Weigh out between 0.2 and 0.25 gram of the finely powdered pyrolusite 
 and introduce into the distillation-flask of either of the forms of apparatus 
 already described. Use a small glass tube for introducing the ore into the 
 flask, obtaining the weight of the material by weighing the tube before and 
 after introducing the ore. About 1.5 grams of potassium iodide are dissolved 
 in 50 c.c. of distilled water and placed in the receiver, which must be kept 
 cool by immersion in ice-water. 20 to 30 e.c. of concentrated hydrochloric 
 acid are introduced into the distillation-flask, which is immediately con- 
 nected with the delivery-tube and the latter immersed in the potassium 
 
346 VOLUMETRIC METHODS. 
 
 iodide solution. The paraffine-coated cork at a is loosely inserted into the 
 receiver. 
 
 The distillation-flask 6 is now gently heated until the manganese dioxide 
 is completely dissolved, leaving a white residue which is not affected by 
 further digestion. The solution is finally brought to a boil, and while the 
 Eunsen burner is still under the distillation-flask, the delivery-tube is with- 
 drawn from the receiver and rinsed with distilled water. The iodine solu- 
 tion is transferred to an Erlenmeyer flask and the receiver washed with 
 water. A few cubic centimeters of potassium iodide solution are placed in 
 the receiver, into which the delivery-tube is again passed, and the contents 
 of the distillation-flask again boiled for a few minutes. If no more iodine 
 is liberated in the potassium iodide solution, the decomposition of the 
 manganese dioxide is complete, otherwise the distillation must be continued 
 for some time, the contents of the receiver being finally added to the main 
 solution. 
 
 The iodine liberated is titrated with standard thiosulphate solution. 
 1 c.c. of N/10 sodium thiosulphate is equal to .004350 gram of Mn0 2 or 
 .003545 gram of chlorine. Compute the percentage of MnO 2 as well as 
 the percentage of chlorine liberated. 
 
 EXERCISE 66. 
 Determination of Sulphur Dioxide in Sodium Sulphite. 
 
 Weigh out 1 gram of the sulphite and placg in^an Erlenmeyer flask. 
 Add 100 c.c. of N/10 iodine solution all at once. It is advisable to measure 
 out the iodine solution with a pipette or a burette into a small beaker. The 
 solution may then be poured from the beaker into the flask. Shake the 
 flask until the salt is dissolved. Rinse the remainder of the iodine solu- 
 tion in the beaker into the flask and titrate the excess with standard thio- 
 sulphate solution. 1 c.c. of N/10 iodine solution is equal to .003203 gram 
 of SO 2 or .01261 gram of Na 2 SO 3 .7H 2 O. Compute the percentage of sulphur 
 dioxide and also of the crystallized salt. 
 
 500. The Active Substance in Bleaching-powder is calcium 
 oxychloride Cl-Ca-0-Cl. On adding water, this compound is 
 decomposed into calcium chloride and calcium hypochlorite, 
 Ca(C10) 2 . The value of the bleaching-powder depends on the 
 amount of hypochlorite present. The result of an analysis is 
 usually expressed in percentage by weight of available chlorine, 
 although in France it is customary to report the analysis in terms 
 of the number of liters of chlorine gas measured at and 760 
 mm., which may be obtained from 1 kilogram of the bleaching- 
 powder. 
 
ANALYSIS OF BLEACHING-POWDER. 347 
 
 501. Sample. The analysis may be carried out by a large 
 number of methods, of which two of the most largely used will be 
 given. In order to secure a fair sample, a considerable amount 
 is weighed out and dissolved in a definite volume of water. Meas- 
 ured portions of this solution are taken for analysis. While calcium 
 hypochlorite is quite soluble, bleach ing-powder contains a con- 
 siderable amount of insoluble material. Even when the bleaching- 
 pow T der is ground to a smooth paste with water, the insoluble 
 material always contains some bleaching material. The insoluble 
 material is, therefore, not filtered off, but is shaken up when por- 
 tions are measured out. A pipette should be used for this pur- 
 pose, so as to obtain a fair proportion of the insoluble material. 
 
 502. lodometric Titration. The amount of available chlorine 
 may be estimated by adding potassium iodide to a portion of 
 the solution of the bleaching-powder and acidifying. The follow- 
 ing reaction takes place: 
 
 Ca(C10) 2 + 4KI + 4HC1 = CaCl 2 + 4KC1 + 2I 2 + 2H 2 0. 
 
 As some calcium chlorate is usually present, the solution is not 
 made strongly acid, as chlorine would in that case be produced by 
 the decomposition of this salt. For this reason, if hydrochloric 
 acid is employed, it is added in very slight excess. The difficulty 
 is entirely obviated by the use of acetic acid. The iodine liber- 
 ated is then titrated by means of sodium thiosulphate solution. 
 
 503. Titration with Sodium Arsenite. The hypochlorite may 
 also be titrated directly with sodium arsenite solution. To ascer- 
 tain the end-point drops of the solution are touched to starch- 
 iodide paper. When the hypochlorite is entirely decomposed, 
 the starch-iodide paper is no longer made blue. 
 
 EXERCISE 67. 
 Determination of Available Chlorine in Bleaching-powder. 
 
 504. Solution. Weigh out 7.091 grams of bleaching-powder. As the 
 material is hygroscopic, and also loses chlorine, a weighing-bottle must be 
 used. Transfer the material to a clean porcelain mortar and grind to a 
 smooth paste with the addition of a little water. Add about 75 c.c. of 
 water, mix thoroughly with the pestle, allow to settle a few minutes, and 
 
348 VOLUMETRIC METHODS. 
 
 decant into a liter flask, leaving the bulk of the bleaching-powder in the 
 mortar. Repeat this operation until all of the material has been trans- 
 ferred to the liter flask, which is then filled to the mark with distilled water. 
 
 505. lo do me trie Titration. The solution is thoroughly shaken and 
 50-c.c. portions of the milky liquid are measured out with a pipette and 
 transferred to beakers. About 1 gram of potassium iodide dissolved in a 
 little water is added to one of the beakers. The solution is acidified with 
 acetic acid and the iodine liberated titrated with N/10 sodium thiosulphate 
 solution. Duplicate determinations with the other portions are made in 
 the same manner. 1 c.c. of the tenth-normal thiosulphate solution is equal 
 to .003545 gram of chlorine. If exactly 7.091 * grams of the bleaching- 
 powder was weighed out, the number of cubic centimeters of N/10 thiosul- 
 phate solution used will be equal to the percentage of available chlorine 
 in the bleaching-powder. 
 
 506. Titration with Sodium Arsenite. 50-c.c. portions of the solution of 
 bleaching-powder may also be titrated with the sodium arsenite solution 
 made for standardizing the iodine solution (page 341). STARCH-IODIDE 
 PAPER may be made as follows : One gram of starch is ground to a paste witn 
 a little water and 100 c.c. of boiling water added with stirring. The solution 
 is filtered and to the filtrate 0.1 gram of potassium iodide is added. Strips 
 of filter-paper are dipped in this solution and dried at 40 to 50. Strips 
 of this paper are moistened with distilled water and laid on a porcelain or 
 glass plate. Nearly all the sodium arsenite solution thought necessary is 
 added to the solution of the bleaching-powder. After stirring thoroughly, 
 a drop is taken out and touched to the test paper. The depth of the color 
 indicates the amount of arsenite solution still to be added. When consid- 
 erable hypochlorite is still present, the spot becomes almost black and 
 then colorless again as the iodine is converted into the trichloride IC1 3 . 
 Several cubic centimeters of the arsenite solution should then be added. 
 When the spot test is only light blue, the arsenite solution should be added 
 by drops until the blue color has entirely disappeared. The calculation 
 is made in the same manner as with the iodine solution. 
 
 PROBLEM. 
 
 Problem 13. A mixture of crystallized sodium thiosulphate, sodium sul- 
 phite, and sodium sulphate was analyzed in the following manner : A gram 
 portion was dissolved in water and titrated with N/10 iodine solution, 49 c.c. 
 being required. The hydriodic acid in this solution was then titrated 
 with N/5 sodium-hydroxide solution, 9.5 c.c. being required. The total 
 amount of sulphur in a half-gram portion was then determined as barium 
 sulphate, of which .821 grams was obtained. Calculate the amount of 
 HA and Na 2 S0 4 .10H :2 O present in the mixture. 
 
 * Note for student: Show how this figure is obtained. 
 
PRECIPITATION METHODS. 
 
 CHAPTER XXVI. 
 
 DETERMINATION OF CHLORIDES, CYANIDES, 
 AND SILVER. 
 
 507. Theory of Indicators. In the volumetric methods already 
 discussed the products of the reactions have been soluble, and the 
 end-point has been obtained by the action of a slight excess of one 
 of the reagents on the indicator. 
 
 In ACIDIMETRY and ALKALIMETRY the indicator is either an add 
 or a base, but possessing such a weak chemical affinity that it cannot 
 remain combined until an excess of the appropriate reagent (base 
 or acid) is present. METHYL ORANGE, for instance, is a weak acid 
 having a bright-red color, while its salts are light yellow. If methyl 
 orange is added to a solution containing a free base, the salt of 
 methyl orange and the base will be formed and the yellow color 
 of the salt will be given to the solution. If carbon dioxide in 
 excess is added to this solution, the color remains yellow because 
 the carbonic acid formed is weaker than the methyl-orange acid r 
 and is therefore unable to displace the latter from its combination 
 with the base. If a strong acid like hydrochloric is added to the 
 solution, it remains yellow as long as there is more base present 
 than can combine with the hydrochloric acid added. When a 
 slight excess of the strong acid has been added, the base combined 
 with the weakly acidic methyl orange is taken by the strong acid 
 The free methyl orange then gives the reddish color to the solution. 
 
 In PRECIPITATION METHODS, on the other hand, advantage is 
 most frequently taken of the relative solubilities of the salts of a 
 metal to obtain the end-point. The colored salt which indicates 
 the end of a reaction must be a soluble compound, while the ele- 
 ment to be determined must form an insoluble compound. 
 
 349 
 
350 VOLUMETRIC METHODS. 
 
 508. Silver Chromate as Indicator. In the determination of 
 chlorides by means of silver nitrate solution, a little neutral potas- 
 sium chromate is added to the solution. On adding the silver 
 nitrate solution the silver is precipitated as silver chloride. When 
 enough silver nitrate has been added to precipitate all of the 
 chlorine, the red silver chromate is formed, which indicates the 
 end of the reaction. As long as chlorides exist in the solution, 
 any silver chromate formed will be decomposed, since if any of 
 this salt is present, it will tend to keep the solution saturated, 
 and any silver which goes into solution as chromate will be imme- 
 diately precipitated as chloride because of the insolubility of the 
 latter salt. Only when chlorides are absent from the solution 
 can silver chromate be permanently formed. As the silver chro- 
 mate is decomposed by even a small amount of free acid, the 
 solution must be neutral or slightly alkaline with sodium car- 
 bonate. 
 
 509. Ferric Sulphocyanate as Indicator. Silver also forms an 
 insoluble salt with sulphocyanic acid. This acid produces an in- 
 tensely red color with solutions of ferric iron. The titration of silver 
 is conducted by adding a standard solution of the sulphocyanate 
 to the silver solution containing some ferric sulphate or nitrate. 
 As long as silver is present in the solution the sulphocyanic acid 
 is precipitated, so that the iron compound which is soluble cannot 
 exist. When all of the silver has been precipitated, the next drop 
 of the sulphocyanate solution forms a little of the red ferric com- 
 pound which indicates the end of the reaction. As the silver 
 sulphocyanate is insoluble in dilute acids, and the ferric sulpho- 
 cyanate is readily formed under these conditions, the titration 
 may be carried on in acid solutions. The titration of chlorides 
 in acid solution may also be carried out by this method which is 
 known as VOLHARD'S. A measured amount of silver nitrate solution 
 which is more than sufficient to precipitate the chloride present is 
 added to the solution. After the addition of some ferric iron as 
 indicator, the excess of silver is titrated back with standard sul- 
 phocyanate solution. Because of the solubility of silver chloride 
 in solutions of the thiocyanates, this method does not give abso- 
 lutely correct results unless the silver chloride is filtered off and 
 the excess of silver determined in the filtrate. 
 
TITRATION OF CYANIDES. 351 
 
 510. Silver Chloride as Indicator. Chlorides may also be 
 titrated with silver nitrate solution without the addition of an 
 indicator. After each addition of silver nitrate the solution is 
 thoroughly shaken. The silver chloride readily collects so as to 
 leave a clear solution in which the formation of a white precipitate 
 on the addition of another drop of silver nitrate may readily be 
 observed. The silver chloride does not collect readily unless a 
 considerable amount of chloride is present and the volume of the 
 solution is not too large. 
 
 511. Titration of Cyanides. The alkali cyanides may be 
 titrated with a silver nitrate solution in the same manner as chlo- 
 rides, the end-point being obtained when no more precipitate is 
 formed. The reaction takes place in two stages as represented 
 by the following equations: 
 
 2KCN + AgN0 3 = KCN.AgCN + KN0 3 ; 
 KCN.AgCN + AgN0 3 = 2AgCN + KN0 3 . 
 
 Until one-half of the cyanide has been converted into the silver 
 salt, any precipitate of silver cyanide which forms is dissolved, 
 by the undecomposed cyanide on stirring the solution. Whem 
 the reaction represented by the first equation is completed, further 
 addition of silver nitrate forms a permanent precipitate of silver 
 cyanide according to the second equation. The precipitate readily 
 collects in flocks on vigorously stirring the solution, leaving a 
 clear liquid in which the absence of a precipitate on the addition 
 of a drop of silver nitrate may be readily observed and serves to 
 indicate the end-point. 
 
 This titration may also be conducted so that the reaction 
 represented by the first equation is carried out. The silver solu- 
 tion is added until a slight permanent precipitate is formed. If 
 CHLORIDES are present in the alkali cyanides, as is frequently the 
 case, silver chloride will be precipitated in preference to the silver 
 cyanide because of the greater insolubility of the former. The 
 silver chloride reacts with the alkali cyanide as follows: 
 
 AgCl-f 2KCN = KCN.AgCN-}- KCL 
 
352 VOLUMETRIC METHODS. 
 
 When all of the cyanide has been converted into the double cyanide 
 of silver and potassium, the next drop of silver nitrate produces a 
 permanent white precipitate. If chlorides are present, this precipi- 
 tate will be silver chloride, thus forming a more delicate indicator 
 than the more soluble silver cyanide. The amount of cyanide 
 present will also be correctly determined, as is evident from the 
 equation given. If the titration is carried to the second stage, 
 the presence of the chloride introduces an error, as all the chloride 
 combines with silver before the end-point is reached. The amount 
 of chloride present may be determined by noting the volume of 
 silver nitrate solution required for the completion of each stage 
 of the titration. The excess of the silver nitrate required for the 
 second part of the titration is that necessary to precipitate the 
 chloride. 
 
 512. Strength of Solutions. The standard solutions used in 
 precipitation methods may be made according to the normal 
 system, a gram molecular weight of silver nitrate, sodium chloride, 
 etc., being dissolved in a liter of a normal solution. Such solutions 
 are convenient for use when several substances must be titrated 
 with the same standard solution. When a solution is to be used 
 exclusively for the titration of a given substance, it is more con- 
 venient to make it of such strength that 1 c.c. is equal to an amount 
 of the given substance which can be represented by a simple 
 number. If a silver nitrate solution is made of such a strength that 
 1 c.c. is equal to 1 mg. of chlorine, the number of cubic centi- 
 meters of silver nitrate solution used to titrate the chlorine in 1 
 gram of a given substance will give the per cent of chlorine if 
 divided by 10. 
 
 513. Standard Silver Nitrate Solutions are made by dissolving 
 weighed amounts of the crystals and diluting to the required 
 volume. The salt may be purchased in very pure condition, 
 seldom requiring drying even. As pure metallic silver may also 
 be readily obtained the standard solutions are sometimes made 
 by weighing out the pure metal and dissolving in nitric acid. If 
 a neutral solution is required, the excess of nitric acid may be 
 removed by evaporation to dryness on the water-bath. The last 
 traces of acid are removed by moistening with water several times 
 and evaporating to dryness on the water-bath. If protected from 
 
TITRATION OF CHLORIDES. 353 
 
 the sunlight and contact with organic matter, solutions of silver 
 nitrate retain their strength for a considerable time unless very 
 dilute. 
 
 514. Standardization. Solutions of silver nitrate may be 
 standardized by titration against weighed amounts of pure sodium 
 chloride prepared as directed in Exercise 6. A saturated solution 
 of potassium chromate is used as the indicator. 4 to 5 drops of 
 fihe chromate solution are added to the solution. to be titrated, the 
 volume of which should not be large. The detection of the first 
 faint tinge of red in the yellow chromate solution offers some 
 difficulties. It is more readily done by gaslight than by sun- 
 light because of the yellow tinge of the former. By daylight the 
 red tinge is more readily detected by a comparison of the color of 
 the solution being titrated, with the color of a solution previously 
 titrated from which the red color has been removed by the addi- 
 tion of a pinch of sodium chloride. The red color is also made 
 more pronounced by looking at the solution through a layer of 
 water tinted with chromate to the same color as the solution to be 
 titrated. It is also advantageous to carry out the titration in a 
 porcelain dish, thus giving a white surface for comparison. 
 
 515. Indirect Determinations by Means of Standard Silver 
 Nitrate. Besides the determination of chlorides, this method may 
 be used to determine indirectly all substances which can be con- 
 verted into neutral chlorides by treatment with hydrochloric acid 
 and evaporation to dryness to expel the excess of hydrochloric 
 acid. The carbonates of the alkalies and the alkaline earths may 
 be treated in this manner to determine the base, and in cases where 
 only the normal carbonate can be present, carbon dioxide may be 
 determined in this manner. The nitrates also of these metals may 
 be converted into chlorides by a similar treatment and determined. 
 
 516. Determination of Sodium and Potassium in the Mixed 
 Chlorides. The amount of potassium and sodium present in a 
 mixture of the chlorides of these metals may be ascertained by a 
 determination of the amount of chlorine present, the weight of 
 the mixed chlorides having first been found. An indirect deter- 
 mination of this kind gives good results only when there is a con- 
 siderable difference in the atomic weights of the elements, and a 
 large precipitate is weighed. The method is not accurate for a 
 
354 VOLUMETRIC METHODS. 
 
 small amount of one element in the presence of a large amount of the 
 other. The weight of sodium chloride in the mixture is calculated 
 by multiplying the amount of chlorine found by 2.1032, 
 
 /Molecular weight KC1\ 
 \ Atomic weight Cl /' 
 
 deducting from the product the combined weight of the chlorides 
 and multiplying the remainder by 3.6426: 
 
 Mol. wt. Nad 
 
 Mol. wt. KCl-Mol. wt. NaCl 
 
 In multiplying the weight of chlorine found by 2.1032, the amount 
 of potassium chloride is found which would be obtained if this 
 salt alone were present. If this product is the same as the weight 
 of the mixed chlorides, only potassium can be present. 
 
 EXERCISE 68. 
 Determination of Cyanogen in Potassium Cyanide. 
 
 In order to secure a fair sample of the material, weigh out 5 grams of 
 the potassium cyanide selected from various parts of the stock to be exam- 
 ined. Dissolve in water and dilute to 500 c.c. Withdraw 50-c.c. portions 
 for titration. If a pipette is used for this purpose, a loose plug of cotton 
 moistened with silver nitrate solution should be placed in the upper part of 
 the stem, and great care should be taken not to draw any of the solution 
 into the mouth. 
 
 The silver nitrate solution is made by weighing out 17 grams of the pure 
 dry crystals, dissolving in water, and diluting to 1 liter. It is advisable to 
 test the distilled water for chloride before adding it to the silver nitrate. 
 Pour the solution into a glass-stoppered bottle and keep in the dark. The 
 50-c.c. portions of the cyanide solution are diluted with an equal volume of 
 water and the silver-nitrate solution added with constant stirring until a 
 slight permanent opalescence is obtained. 1 c.c. N/10 silver nitrate solu- 
 tion is equal to .01304 gram of potassium cyanide. As commerical potas- 
 sium cyanide is frequently contaminated with sodium cyanide, the percentage 
 of potassium cyanide found may be greater than 100%. This will not occur 
 if the result is given in percentage of cyanogen. 1 c.c. of N/10 silver nitrate 
 solution is equal to .005208 gram of cyanogen. 
 
DETERMINATION OF SILVER. 355 
 
 517. Determination of Silver in Alloys. The silver in alloys 
 can very readily be titrated by the VOLHARD METHOD, since the 
 presence of a considerable amount of free nitric acid does not dis- 
 solve the silver sulphocyanate. Nitrous acid must be absent, 
 however, and that produced by dissolving the alloy must be entirely 
 expelled by boiling the solution. The amount of copper present 
 must not exceed 70%. Alloys in which the percentage of copper 
 is higher, may be analyzed by adding an accurately measured 
 volume of standard silver solution or an accurately weighed 
 amount of the pure metal, so that the ratio of copper to silver 
 shall not exceed 7 to 3. Mercury must be removed by ignition. 
 Arsenic, antimony, cadmium, lead, bismuth, tin, and zinc do not 
 interfere with the tit rat ion. Cobalt and nickel interfere because 
 of the color of their nitrates. 
 
 EXERCISE 69. 
 Determination of Silver in a Coin by Volhard's Method. 
 
 Weigh out 7.5 to 8 grams of ammonium sulphocyanate, dissolve in 
 water, and dilute to 1 liter. Standardize the solution by titration against 
 the tenth-normal silver solution prepared for Exercise 68. Measure out 
 into a beaker 30 to 40 c.c. of the standard silver solution, dilute with 
 100 to 200 c.c. of water, and add 5 c.c. of a cold saturated solution of 
 ferric alum. The ferric iron solution must be free from chlorine, and if 
 turbid, a few cubic centimeters of dilute nitric acid may be added. Place 
 the beaker containing the silver on a white surface and add the sulpho- 
 cyanate solution with constant stirring until a faint permanent reddish 
 tinge is given to the solution. Repeat the titration until duplicates are 
 obtained. As the sulphocyanate solution is permanent, it is advisable to 
 dilute it to exact strength. When it is used to determine silver, it is most 
 convenient to dilute it to such a strength that 1 c.c. will be equal to exactly 
 .010 gram of silver. Having found, for instance, that 30 c.c. of the sulpho- 
 cyanate solution is equal to 32 c.c. of the N/10 silver nitrate solution, the 
 amount of silver equivalent to the 30 c.c. of the sulphocyanate solution is 
 found by multiplying 32 by .010793, which gives .3454 gram of silver. As 
 it is desired to dilute the solution so that 30 c.c. shall be equal to .300 gram, 
 the volume to be diluted to 1000 c.c. is found by the proportion 
 
 .300 : .3454 : : x : 1000. 
 
356 VOLUMETRIC METHODS. 
 
 As x = 868.58, this number of cubic centimeters is diluted to 1 liter. The 
 strength of the solution after dilution is verified by again titrating against 
 the standard silver solution. 
 
 Thoroughly clean a dime or other silver coin by rubbing with alcohol 
 ammonia, etc. Weigh it carefully and dissolve in 20 c.c. of a mixture of 
 equal parts of nitric acid and water. For this purpose place the coin in a 
 beaker, cover with a watch-crystal, and warm on the water-bath until it is 
 entirely dissolved. Remove the watch-crystal after rinsing it with water 
 and continue warming the solution until the nitrous fumes are entirely 
 expelled. Transfer the solution to a 250-c.c. flask, rinsing the beaker thor- 
 oughly. Dilute the solution to the mark and shake thoroughly. Take 
 out 50-c.c. portions with a pipette, dilute with 100 to 200 c.c. of water, 
 add 5 c.c. of the ferric iron solution, and titrate with the sulphocyanate. 
 Repeat the titration until duplicates are obtained. Calculate the per cent 
 of silver in the coin. 
 
 PROBLEMS. 
 
 Problem 14. The combined weight of sodium and potassium chlorides 
 was 2.5685 grams. 1.432 grams of chlorine was present. Calculate the 
 percentage of potassium and sodium chloride present. 
 
 Problem 15. The combined weight of the sodium and potassium chlorides 
 was 2.032 grams. When converted into sulphates, the weight was 2.385 
 grams. Calculate the percentage of sodium and potassium present. 
 
CHAPTER XXVII. 
 
 DETERMINATION OF PHOSPHORIC ACID. 
 
 A GOOD many methods of determining phosphoric acid volu- 
 me trically have been devised. None of these methods give as 
 reliable and accurate results as the gravimetric method. The 
 great amount of labor necessary and the great length of time 
 which must elapse from the beginning to the end of the gravimetric 
 determination render the use of the volumetric methods very 
 desirable. Three of the best of these methods will be described. 
 
 THE URANIUM METHOD. 
 
 This method is applicable to the titration of the phosphoric 
 acid in the alkali and alkaline earth phosphates. IRON, and espe- 
 cially ALUMINIUM, if present in considerable amount, interfere 
 seriously with the titration. The method is applicable to most 
 of the material used for fertilizers, such as bones, bone-ash, soluble 
 phosphates, phosphorite, etc. The method is based on the fact 
 that when uranium nitrate or acetate is added to a neutral solution 
 of a phosphate all of the phosphoric acid is thrown down as yellow 
 uranyl phosphate, U0 2 HP0 4 . Free mineral acid must be absent, 
 but the solution may be moderately acid with acetic acid. 
 
 A very delicate INDICATOR for the excess of the uranium is 
 found in the reaction with potassium ferrocyanide, which gives a 
 decided brown tinge with very small amounts of uranium. Unfor- 
 tunately, the method is not an absolute one, but gives results 
 which vary with the volume of the solution, the salts present, etc., 
 so that the standardization of the solution must be made under 
 as nearly as possible the same conditions as the determination to be 
 carried out. The standardization and determination should be 
 
 357 
 
358 VOLUMETRIC METHODS. 
 
 carried out by the same individual. To complete the precipita- 
 tion of the phosphoric acid, the solution must be boiled after each 
 addition of the standard solution. Some calcium phosphate is 
 carried down with the uranium phosphate, so that if a solution 
 containing calcium is to be titrated, the uranium solution must be 
 standardized by calcium phosphate, as the alkali phosphates are 
 not carried down in this manner. Several titrations must also 
 be made, in the first adding the uranium solution in small amounts 
 until the end-point is reached. In the second tit ration nearly 
 all of the uranium solution is added at once. 
 
 518. Standard Solution. The uranium solution is made by 
 dissolving about 35 grams of uranium nitrate or acetate in water 
 and diluting to a liter. 3J grams of ammonium or sodium acetate 
 are added to the nitrate solution to neutralize any free acid present. 
 The addition of 50 c.c. of pure glacial acetic acid to either the 
 nitrate or the acetate solution renders it more stable on exposure 
 to light. The nitrate may usually be obtained in a purer condi- 
 tion than the acetate, while the addition of an alkali acetate to 
 the latter is unnecessary. The solution is allowed to stand for 
 several days and is then filtered, or the clear liquid drawn off. 
 
 519. Standardization. For standardizing the solution a pure 
 salt or a solution of a phosphate in which the phosphoric acid has 
 been accurately determined must be available. DISODIUM PHOS- 
 PHATE and MICROCOSMIC SALT may be recrystallized until absolutely 
 pure, although some uncertainty always exists as to the amount 
 of water present in the dried crystals. DIHYDRIC POTASSIUM 
 PHOSPHATE has the advantage that it contains no water of crystal- 
 lization. If these pure salts are not at hand, a solution may be 
 made containing approximately 0.1 gram P 2 5 in 50 c.c. 100-c.c. 
 portions of this solution may be measured out and the phosphoric 
 acid present determined by precipitation with magnesia mixture 
 and weighing as magnesium pyrophosphate. 
 
 520. Titration. 50 c.c. of the standard phosphate solution is 
 measured out into a beaker and heated to 90 or 100. The 
 uranium solution is added until a drop taken out and placed on a 
 white porcelain surface gives a brown ring when the centre of 
 the drop is touched with a glass rod which has been dipped in a 
 freshly made solution of potassium ferrocyanide. This solution 
 
DETERMINATION OF PHOSPHORIC ACID. 359 
 
 is made by dissolving about J gram of the salt in 20 c.c. of water. 
 After each addition of uranium the solution is brought to a boil. 
 The color of the spot test increases slowly on standing. The worker 
 must therefore note the depth of color after a fixed interval of 
 time. Uniformity in this respect is most readily secured by always 
 observing the color after an additional drop has been added to 
 the solution being titrated and the solution brought to a boil. 
 When the end-point is reached, this drop will be in excess, but a cor- 
 rection may easily be made to the reading after noting the change 
 in the reading of the burette produced by letting one drop fall 
 from the burette. A second and third tit rat ion are made, adding 
 nearly the entire amount of uranium solution at once. The ura- 
 nium solution is then diluted so that 1 c.c. is equal to .005 gram 
 of P 2 5 . 
 
 521. Standardization with Calcium Phosphate. When the ura- 
 nium solution is to be used for the titration of phosphoric acid in 
 the presence of calcium, it must be standardized by means of a 
 solution of tricalcium phosphate. About 5.5 grams of the pure 
 material is weighed out and dissolved in a little dilute nitric or 
 hydrochloric acid. It is precipitated by adding a slight excess of 
 ammonia and redissolved in a moderate excess of acetic acid. 
 The solution is then diluted to a liter. The phosphoric acid in 50- 
 c.c. portions of this solution is determined by precipitation as 
 molybdate, dissolving in ammonia, and reprecipitating with mag- 
 nesia mixture and weighing as magnesium pyrophosphate. The 
 titration of 50-c.c. portions of this solution with the uranium 
 solution is conducted as already given. 
 
 522. Titration of Phosphates. The analysis of phosphates 
 which are free from more than traces of iron and aluminium is 
 made by dissolving weighed portions, so that 50 c.c. shall con- 
 tain about 0.1 gram P 2 5 . If the material is dissolved in the strong 
 mineral acids, the solution must be neutralized with ammonia and 
 then acidified with acetic acid. 50-c.c. portions are then titrated 
 exactly as directed for the standardization. 
 
 When considerable iron or aluminium is present, the phosphoric 
 acid must be separated from these elements by precipitation as 
 phosphomolybdate, which is dissolved in ammonia, and the phos- 
 phoric acid again precipitated with magnesia mixture. The mag- 
 
360 VOLUMETRIC METHODS. 
 
 nesia precipitate is then dissolved in a little nitric or hydrochloric 
 acid, the solution neutralized with ammonia and then acidified 
 with acetic acid. The phosphoric acid is then titrated with the 
 uranium solution. 
 
 TITRATION OF THE PHOSPHOMOLYBDATE. 
 
 Two volumetric methods of determining phosphoric acid have 
 been based on the properties of the phosphomolybdate precipi- 
 tate. By one of these methods, that of Pemberton,* the molybdic 
 acid in this precipitate is titrated by means of standard caustic soda 
 or potash with phenolphthalein as the indicator according to the 
 following equation : 
 
 2(NH 4 ) 3 P0 4 .12Mo0 3 +46KOH= 
 
 2(NH 4 ) 2 HP0 4 + (NH 4 ) 2 Mo0 4 + 23K 2 Mo0 4 + 22H 2 0. 
 
 By the other methodf the molybdic acid is reduced by means of 
 zinc to molybdic oxide according to the equation: 
 
 2Mo0 3 + 6H=Mo 2 3 + 3H 2 0. 
 
 The molybdic oxide is then oxidized by means of standard potas- 
 sium permanganate solution. 
 
 523. Pemberton's Alkalimetric Method. This method is appli- 
 cable to the titration of the phosphates of the alkalies and the alka- 
 line earths, but the results are not so good when considerable 
 quantities of iron or aluminium are present. In the presence of 
 large amounts of these metals, it is advisable to redissolve and 
 reprecipitate the phosphomolybdate. As sulphuric acid may be 
 carried down with the precipitate, it should be absent from the 
 solution. If present it may be removed by means of barium 
 chloride. 
 
 524. Standard Solutions. A standard solution of nitric acid 
 and one of caustic soda or potash free from carbon dioxide are 
 required. The solutions are made of such a strength that 1 c.c. is 
 equal to .001 gram of P 2 5 . Since 46 molecules of caustic soda or 
 potash are equal to 1 molecule of phosphorus pentoxide, a liter of 
 
 * Jour. Am. Chem. Soc., 1894, 278. 
 
 t Blair, The Chemical Analysis of Iron, 1901, p. 104. 
 
DETERMINATION OF PHOSPHORIC ACID. 361 
 
 normal solution will be equal to l/46th of the molecular weight of 
 P 2 5 , or 3.087 grains. A solution of convenient strength will be 
 .3239 normal, 1 liter of such a solution being equal to 1 gram of 
 P 2 5 . The nitric acid may be made by diluting 21 c.c. strong 
 nitric acid (sp. gr. 1.40) to 1 liter. The caustic soda solution may 
 be made by adding to 100 grams of pure caustic soda an amount of 
 water just insufficient to dissolve it. After allowing the solution 
 to settle in a tall covered vessel, 17 c.c. of the clear liquid is with- 
 drawn with a pipette and diluted to 1 liter. Most of the sodium 
 carbonate remains undissolved. If caustic potash is used, about 
 20 grams are weighed out and diluted to a liter. The carbon 
 dioxide present must be precipitated by means of barium hydroxide 
 or chloride. Both the caustic soda and caustic potash solutions 
 must be protected from the carbon dioxide of the air. 
 
 525. Standardization. The solutions may be standardized by 
 any of the methods given in Chapters XXI and XXII. The acid 
 must be compared with the alkali, using phenolphtha'ein as the 
 indicator. A much better method of standardization consists in 
 precipitating a known amount of phosphoric acid by means of 
 the molybdate solution and titrating the precipitate in exactly 
 the same manner as in the analysis of an unknown phosphate. 
 
 526. Precipitation of the Phosphoric Acid. The molybdate 
 solution is made according to the directions given on p. 507. For 
 washing the precipitate, a dilute solution of nitric acid is used, a 
 solution made by adding about 15 c.c. concentrated acid to 1 liter 
 of water being suitable. The acid is washed out by means of a 
 potassium nitrate solution made by dissolving 1 gram of potassium 
 nitrate in 1 liter of water. 
 
 The phosphate must be free from organic matter. If present, 
 it is destroyed by ignition of the material in a platinum dish or 
 crucible. If the percentage of phosphoric acid is large, a solution 
 should be made by dissolving a weighed amount in nitric acid and 
 diluting to a known volume. Portions are then measured out 
 containing not more than .050 gram P 2 5 . The solution is nearly 
 neutralized with ammonia, warmed to about 40, and 40 c.c. of 
 the molybdate solution added. The precipitation is best made in 
 an Erlenmeyer flask of about 250 c.c. The solution is allowed to 
 digest for ten to fifteen minutes, the temperature of about 40 
 
362 VOLUMETRIC METHODS. 
 
 being maintained by warming on the water-bath. By closing the 
 flask with a rubber stopper and shaking about five minutes, com- 
 plete precipitation will be insured < Allow the precipitate to settle 
 for a few minutes and decant the solution through a small filter- 
 paper. Transfer the precipitate and wash with dilute nitric acid 
 until free from molybdenum as indicated by the absence of a pre- 
 cipitate with ammonium sulphide. The precipitate is then washed 
 with the potassium nitrate solution until free from acid. The 
 washing must be thorough, but as little wash-water as possible 
 must be used, because of the slight solubility of the precipitate. 
 If properly done, 200 c.c. will be found sufficient. 
 
 527. Titration. The paper and the precipitate are placed in a 
 beaker and a measured amount of the standard caustic soda or 
 potash solution sufficient to dissolve the precipitate is added. 10- 
 c.c. portions may be added, and if on agitation the precipitate does 
 not dissolve, another 10-c.c. portion is added. Continue the addi- 
 tion of 10-c.c. portions until the precipitate is completely dissolved. 
 Add a few drops of phenolphthalein and titrate the excess of alkali 
 with the standard nitric acid. The amount of phosphoric acid 
 present is calculated from the number of cubic centimeters of 
 caustic soda used. 
 
 REDUCTION OF THE MOLYBDATE PRECIPITATE WITH 
 ZINC AND TITRATION WITH PERMANGANATE SO- 
 LUTION. 
 
 528. Standard Solution. A standard solution of potassium 
 permanganate will be required for this method. As 1 molecule of 
 P 2 5 is combined with 24 molecules of Mo0 3 , which will require 36 
 atoms of oxygen to reoxidize after being reduced to Mo 2 3 , we have 
 the relation Mol. wt. P 2 5 (142) =Mol. wt. 36 0(576). A liter of a 
 normal solution of potassium permanganate will therefore be equal 
 to 1.9722 gram of P 2 5 . From this figure the value of a fifth- 
 or tenth-normal solution may be computed. If standardized by 
 means of iron, the value of the solution in* phosphorus pentoxide 
 may be calculated directly from the factor .03528, which is the 
 equivalent of 1 gram of iron in phosphorus pentoxide. The best 
 method consists in reducing and titrating the molybdate precipitate 
 from a known amount of phosphoric acid. 
 
DETERMINATION OF PHOSPHORIC ACID. 363 
 
 529. Titration. The precipitation of the phosphate with the 
 molybdate solution is carried out exactly as given for the precedirg 
 method. Instead of washing with dilute nitric acid and potassium 
 nitrate, a solution of acid ammonium sulphate is used. This is 
 made by adding to 1 liter of water 15 c.c. of strong ammonia (sp. gr. 
 0.90) and 25 c.c. of concentrated sulphuric acid (sp. gr. 1.84). After 
 washing the precipitate until it is free from the molybdenum solu- 
 tion, as indicated by the test with ammonium sulphide, it is dis- 
 solved in dilute ammonia. During the washing, the phospho- 
 molybdate need not be entirely removed from the flask in which 
 it was precipitated. The ammonia solution of the precipitate is 
 allowed to flow into the flask together with the water used to wash 
 the paper. 5 grams of pulverized zinc and 15 c.c. of concentrated 
 sulphuric acid are now added. The flask is closed with a stopper 
 through which passes a glass tube bent at a slightly acute angle 
 and dipping into a saturated solution of sodium bicarbonate. 
 When all of the zinc is dissolved the solution is titrated with the 
 standard potassium permanganate solution. The solution before 
 titration should be green in color. If it is brown the determi- 
 nation should be rejected. A blank should be made in which the 
 same amount of ammonia, zinc, and sulphuric acid is added to 
 enough distilled water to make the volume of the solution the 
 same as that in the titration of the molybdate precipitate. After 
 the zinc is dissolved, the solution is titrated with the standard 
 permanganate and the amount added is subtracted from the vol- 
 ume used in the determination of phosphorus. 
 
TECHNICAL ANALYSIS. 
 
 CHAPTER XXVIII. 
 
 ANALYSIS OF IRON AND STEEL. 
 
 IRON and steel as manufactured to-day are quite complex 
 substances, containing almost invariably, besides IRON, the ele- 
 ments CARBON, SILICON, SULPHUR, PHOSPHORUS, and MANGANESE, 
 
 while special varieties may contain NICKEL, CHROMIUM, MANGANESE, 
 etc. 
 
 530. The Metals are undoubtedly present in the conditions in 
 which they usually exist in alloys. The analysis for the determina- 
 tion of the amount of the metals present other than iron is carried 
 out by the methods already given for the analysis of alloys. 
 
 531. The Acid Elements are for the most part combined directly 
 with the iron as carbide, silicide, sulphide, and phosphide. As 
 these elements are usually met with in combination with oxygen, 
 the ordinary methods of analysis are not always suitable for their 
 determination in iron and steel. The modification usually consists 
 in adopting a method of dissolving the metal by which the acid 
 element is certain to be thoroughly oxidized. Solution of iron or 
 steel in a non-oxidizing acid like hydrochloric results in the evo- 
 lution of more or less of the acid elements present as hydrogen 
 compounds, such as hydrogen sulphide, phosphine, and various hy- 
 drocarbons. Silicon even is liable to be volatilized as silicon tetra- 
 chloride. When solution is effected with nitric acid, silicon, phos- 
 phorus, and sulphur are converted into non-volatile oxygen com- 
 pounds, while more or less of the carbon is lost as carbon dioxide. 
 For the determination of carbon, the iron is dissolved in solutions 
 of cupric salts or the carbon dioxide evolved on dissolving the metal 
 in an oxidizing solution is collected and determined. 
 
 The determination of these elements must be carried out 
 with unusual care, since a very slight variation in the percentage 
 
 364 
 
SILICON. 365 
 
 produces a very considerable variation in the properties of the 
 metal. 
 
 532. To Obtain a Sample for Analysis a perfectly clean drill is 
 taken, and by means of a drill-press or an ordinary brace, holes are 
 bored in various parts of the casting or bar of iron or steel. Loose 
 dirt, sand, etc., must be removed from the surface so as not to fall 
 into the drillings. If the surface cannot be cleaned perfectly the 
 first drillings should be discarded. The remainder of the drillings 
 may then be collected on a piece of clean glazed paper. The iron 
 may also be separated from sand and dirt by means of a magnet. 
 Oil may be removed by washing a few times with ether. The drill- 
 ings are usually fine enough for most purposes. If not, they may 
 be crushed in an agate or hardened-steel mortar. In no case may 
 the fine portions be taken and the large pieces discarded. A fair 
 sample of both portions must be taken and further pulverized if 
 necessary. 
 
 DETERMINATION OF SILICON. 
 
 533. Solution of the Iron in Nitric Acid. One to five grams 
 of the drillings are placed in a beaker of 200- to 300-c.c. ca- 
 pacity and 40 c.c. of a mixture of 20 c.c. concentrated nitric 
 acid and 20 c.c. water are added. The action is generally quite 
 violent accompanied by frothing, so that to prevent loss it should 
 be moderated by placing the beaker in cold water. If action 
 does not begin as soon as the acid is added, the beaker may be 
 gently warmed and cooled when the action becomes violent. It 
 may also be moderated by adding the acid in several portions. 
 When the action has ceased, the beaker is warmed on the hot-plate 
 to dissolve the few remaining portions of metal. Solution may be 
 hastened by the addition of a little hydrochloric acid. The solu- 
 tion is evaporated to dryness and the residue heated in the air-bath 
 until nitrous fumes are no longer evolved. The residue is then 
 dissolved in 30 c.c. hydrochloric acid by warming gently. The 
 solution is diluted somewhat, warmed, and the silica filtered off 
 on an ashless paper and washed. Dilute hydrochloric acid is first 
 used and then water. 
 
 534. Volatilization of Silica. The moist paper is transferred 
 to a weighed platinum crucible, the paper burned, and the precipj" 
 
366 ANALYSIS OF IRON AND STEEL. 
 
 tate ignited, finally over the blast-lamp, and weighed. Even if 
 the silica is perfectly white, it is necessary to test its purity by add- 
 ing a few drops of concentrated sulphuric acid and then enough 
 pure hydrofluoric acid to dissolve the silica. The solution is then 
 evaporated to dryness and the residue ignited and weighed. The 
 weight of the residue is subtracted from the weight of the impure 
 silica and the amount of silicon calculated. 
 
 535. Drown's Method. A more rapid method which is very 
 much used for pig iron and which gives excellent results is based on 
 the fact that the silicic acid may be dehydrated by heating with 
 concentrated sulphuric acid. The drillings are dissolved in nitric 
 acid as already described. 20 c.c. of sulphuric acid consisting of 
 equal parts of concentrated acid and water are then added for each 
 gram of drillings taken. The solution is heated until copious 
 fumes of sulphuric acid are given off. The lumps of ferric sulphate 
 must be broken up with a rod and the solution stirred to prevent 
 bumping and spattering of the material. The concentrated acid 
 is allowed to cool somewhat, cautiously diluted with water, and 
 the solution warmed gently until the ferric sulphate is completely 
 dissolved. The hot solution is filtered immediately, as the silica 
 dehydrated in this manner gradually goes into solution again. 
 The silica is washed first with water and then with dilute hydro- 
 chloric acid and finally with water. It is ignited and weighed, 
 and then volatilized as directed in the first method. This method is 
 used when silica alone is to be determined. 
 
 536. Solution of the Iron in Hydrochloric Acid. The silica may 
 also be determined after solution of the iron in hydrochloric acid 
 and potassium chlorate. The evaporation of the solution to dry- 
 ness may then be accomplished more rapidly, but a slight loss of 
 silicon as SiCl 4 is almost inevitable. 
 
 537. Fusion of Silica Precipitate. Instead of ascertaining the 
 amount of impurity in the silica by treatment with hydrofluoric 
 and sulphuric acids, it may be fused with four or five times its 
 weight of sodium carbonate. The fused mass is dissolved in water, 
 acidified with hydrochloric acid, evaporated to dryness, redissolved 
 in hydrochloric acid and water, and the silica filtered off, washed, 
 ignited, and weighed. The purification of the silica by volatiliza- 
 
SULPHUR. 367 
 
 tion with hydrofluoric acid is more convenient if a pure acid is at 
 hand. 
 
 DETERMINATION OF SULPHUR. 
 
 538. Solution of the Iron in Nitric Acid. Sulphur may be deter- 
 mined in the filtrate from the silica by precipitation with barium 
 chloride and weighing as barium sulphate. The iron must be dis- 
 solved by the first method given, but concentrated nitric acid must 
 be used in place of the acid of sp. gr. 1.2 in order to insure com- 
 plete oxidation of the sulphur. If the metal dissolves slowly, a 
 few drops of hydrochloric acid may be added. As the percentage 
 of sulphur is usually small, the large amount (5 grams) should b<i 
 used. After solution has been effected a little sodium carbonate 
 is added to prevent loss of sulphur trioxide, and the solution is 
 evaporated to dryness. The residue is dissolved in hydrochloric 
 acid, and again evaporated to dryness to completely remove the 
 nitric acid and dehydrate the silicic acid. The residue is again 
 dissolved in hydrochloric acid, and the excess expelled by evapora- 
 tion. The solution must not be allowed to go to dryness, as a crust 
 of insoluble basic ferric chloride would be produced. The solution 
 is diluted and the silica and graphite filtered off and washed. 
 
 The filtrate is heated to boiling, and 10 c.c. of a 10% solution 
 of barium chloride added slowly and with constant stirring. The 
 solution is allowed to stand overnight in the cold. The precipitate 
 is then filtered off and washed with very dilute hydrochloric acid 
 until free from iron, and then with water. The barium sulphate 
 is ignited and weighed in the usual manner. 
 
 539. Errors. This method tends to give a low result on account 
 of the solubility of barium sulphate in hydrochloric acid and ferric 
 chloride. The result is apt to be high on account of the great diffi- 
 culty of obtaining a precipitate of barium sulphate which is free 
 from iron. The iron carried down seems to be present as a basic 
 ferric sulphate which loses S0 3 on ignition and leaves the barium 
 sulphate colored red by the ferric oxide. Fusing the precipitate 
 with sodium carbonate, dissolving in water, and filtering effects a 
 separation from the iron, but the barium sulphate obtained from 
 the filtrate will not represent the sulphur present in the iron because 
 
368 ANALYSIS OF IRON AND STEEL. 
 
 of loss of S0 3 during the first ignition. The presence of considerable 
 amounts of hydrochloric acid in the iron solution tends to prevent 
 the precipitation of the iron with the barium sulphate, but another 
 error is introduced because of the marked solubility of the latter 
 in the dilute hydrochloric acid and the ferric chloride. The large 
 excess of barium chloride used tends to produce complete precipi- 
 tation of the sulphuric acid. The success of the method therefore 
 depends on regulating the amount of hydrochloric acid so as to 
 prevent the precipitation of iron, but not that of the barium sul- 
 phate. The so-called evolution methods are therefore generally 
 preferred when applicable. In steel the sulphur is generally so 
 combined that the evolution method may be used, but the evolution 
 of the sulphur is not complete in all pig and cast irons. 
 
 540. For Pi Irons the following method is used: After solu- 
 tion of the pig iron in strong nitric acid, 10 grams of sodium car- 
 bonate are added, and the solution transferred to a platinum 
 dish, evaporated to dryness, and the residue ignited. It is then 
 digested with water, a little sodium carbonate added, and the 
 solution filtered. The insoluble material is washed with water 
 containing sodium carbonate. The filtrate is acidified with hydro- 
 chloric acid and evaporated to dryness to dehydrate the silica, 
 whJch is filtered off and washed. The sulphuric acid in the filtrate 
 is precipitated with barium chloride and weighed as barium sul- 
 phate. 
 
 541. Determination of Sulphur in the Reagents. All of the 
 reagents used in sulphur determinations must be free from sulphur. 
 Their purity is most easily assured by making a blank determina- 
 tion as follows: The total amount of nitric and hydrochloric acid 
 used is measured out into a beaker, a little sodium carbonate added, 
 and the solution evaporated to dryness. The residue is dissolved 
 in a little water, a few drops of hydrochloric acid added, and after 
 heating to boiling a few cubic centimeters of barium chloride 
 solution are added. Any precipitate formed is filtered off, washed, 
 ignited, and weighed, and the amount of barium sulphate obtained 
 deducted from the amount obtained from the solution of iron. 
 
SULPHUR. 369 
 
 EVOLUTION METHODS FOR SULPHUR. 
 
 The evolution methods are now more largely used than the 
 oxidation methods. In the evolution methods advantage is taken 
 of the fact that when iron or steel is dissolved in hydrochloric acid 
 the sulphur is evolved as hydrogen sulphide. Air as well as all 
 oxidizing agents must be excluded from the evolution-flask. 
 
 542. For the Absorption of the Hydrogen Sulphide a large 
 number of solutions have been suggested. Formerly a hydrochloric- 
 acicl solution of BROMINE was largely used. The sulphur was oxi- 
 dized by this solution to sulphuric acid, which was precipitated 
 and weighed as barium sulphate. The evolution of the disagreeable 
 bromine fumes may be obviated by the substitution of an ammo- 
 niacal solution of hydrogen peroxide or a 3% solution of sodium 
 peroxide. These reagents are, however, almost never free from 
 sulphur, necessitating a separate determination of the amount 
 present. 
 
 Another class of absorption liquids consists of solutions of 
 metallic salts. The hydrogen sulphide is precipitated as a metallic 
 sulphide, which in some cases is filtered off, washed, and weighed, 
 and in other cases is oxidized by means of an appropriate solution 
 to sulphuric acid which is precipitated as barium sulphate. The 
 use of solutions of silver and copper for this purpose seems to be 
 objectionable because of the reduction of the former by the hydro- 
 carbons evolved, and the formation of an insoluble phosphorus 
 compound in the latter. Alkaline solutions of lead, zinc, and 
 cadmium seem to be very efficient absorbents. Very good results 
 are obtained by washing the cadmium sulphide, drying at 100, 
 and weighing. 
 
 543. lodometric Determination of the Sulphur. The sulphur 
 in the cadmium and zinc sulphides may also be determined by 
 treatment with a measured excess of standard iodine solution, 
 acidulating, and titrating the excess with standard sodium thio- 
 sulphate. The following reactions then take place: 
 
 CdS+2HCl=CdCl 2 -fH 2 S, 
 H 2 S+I 2 =2HI + S. 
 
370 
 
 ANALYSIS OF IRON AND STEEL. 
 
 The hydrogen sulphide may also be absorbed by passing the 
 gas directly through a measured amount of standard iodine solu- 
 tion and titrating the excess. The volatility of the iodine intro- 
 duces an error in this determination. This may be obviated by 
 absorbing the hydrogen sulphide in a solution of caustic soda 
 and determining the sulphur with iodine solution in the same 
 manner as with cadmium sulphide. 
 
 544. All of the Sulphur is not Evolved as Hydrogen Sulphide 
 when a metal of high carbon content is analyzed. Organic sul- 
 phides are formed among which methyl sulphide (CH 3 ) 2 S has 
 been identified.* These sulphur compounds are expelled from 
 the evolution-flask with difficulty, and are not taken up so readily 
 by the absorption solution. By passing the gases through a red- 
 hot glass tube these compounds may be decomposed so that the 
 sulphur is absorbed. 
 
 545. A Convenient Form of Apparatus for carrying out the deter- 
 mination is shown in Fig. 59. The Kipp generator is charged 
 
 FIG. 59. 
 
 with zinc and acid for the generation of hydrogen. The wash- 
 bottle A contains a solution suitable for the absorption of hydro- 
 gen sulphide. 
 
 A solution of LEAD NITRATE in CAUSTIC POTASH is very efficient. 
 It is made by dissolving one part of caustic potash in two parts of 
 water and pouring in a solution of lead nitrate with constant 
 stirring until a permanent precipitate is produced. After being 
 
 * Philips, Stahl und Eisen, 16, 633. 
 
SULPHUR. 371 
 
 allowed to settle, the clear liquid is siphoned off for use. It should 
 be kept in a glass-stoppered bottle. The stopper should be coated 
 with a little paraffine to prevent its sticking. This solution may 
 also be used in the absorption-flasks C and E. 20 to 30 c.c. in 
 each flask will be sufficient, water being added until the flasks are 
 about half-full. The flask B should have a capacity of about \ 
 liter. 
 
 546. Solution of the Iron. Ten grams of filings are placed in the 
 flask B. If a portion weighing 13.732 grams is used, the number of 
 milligrams of barium sulphate obtained will be equal to the number 
 of one-thousandths per cent of sulphur. If, for instance, .0156 
 gram of barium sulphate is obtained, the percentage of sulphur 
 present will be .0156. The air in the apparatus is displaced with a 
 stream of hydrogen. The glass tube D is then heated to redness, 
 the dropping-funncl filled with dilute hydrochloric acid, and 
 connection made immediately with the Kipp generator. About 
 20 c.c. of dilute acid should be used for each gram of iron. Some 
 workers use for each gram of iron 10 c.c. of a dilute hydrochloric 
 acid composed of equal volumes of concentrated acid and water, 
 while others recommend 18 c.c. of a mixture of two-thirds dilute 
 hydrochloric acid and one-third dilute sulphuric acid. 
 
 When the glass tube has been heated to dull redness the acid 
 is allowed to flow into the flask B by turning the stop-cock of the 
 dropping-funnel. The flask is heated with a Bunsen burner until 
 the solution boils, a rapid stream of gas being allowed to pass 
 through the apparatus. After boiling for about fifteen minutes the 
 iron is usually dissolved. 
 
 547. Absorption of the Sulphur by Lead Solution. The Bun- 
 sen burner is removed and the stream of hydrogen allowed to flow 
 for about fifteen minutes more to sweep all of the. hydrogen sul- 
 phide out of the apparatus. The precipitate in the flask C is filtered 
 off on a small paper. If a precipitate has formed in the flask D, 
 it is filtered off on the same paper and washed once or twice with 
 hot water. The moist paper with the precipitate is thrown into a 
 beaker, a little potassium chlorate added, and 5 to 20 c.c. of con- 
 centrated hydrochloric acid added, the amount used depending on 
 the amount of lead sulphide present. Allow the beaker to stand in 
 a warm place until chlorine is no longer evolved, dilute the strong 
 
372 ANALYSIS OF IRON AND STEEL. 
 
 acid and filter. After washing the paper with hot water the fil- 
 trate is neutralized with ammonia and then acidified with 2 or 3 
 drops of hydrochloric acid. The sulphuric acid is then precipi- 
 tated with barium chloride and weighed as barium sulphate. 
 
 548. Absorption of the Sulphur by a Cadmium Solution. 
 Instead of using an alkaline solution of lead nitrate in the absorp- 
 tion-flasks C and D, an alkaline solution of cadmium may be 
 used as proposed by T. T. MorreL* 
 
 The cadmium sulphide is filtered off on a Gooch crucible and 
 washed with water containing a little ammonia. The precipitate 
 is dried at 100 and weighed. 
 
 The cadmium sulphide together with the paper may also be 
 placed in a beaker, 200 or 300 c.c. of cold water added, and then 
 dilute hydrochloric acid until the precipitate is dissolved. A few 
 cubic centimeters of freshly prepared starch solution are then intro- 
 duced, and standard iodine solution added from a burette until a 
 permanent blue color is produced. 
 
 DETERMINATION OF PHOSPHORUS. 
 
 The determination of phosphorus in iron or steel is invariably 
 preceded by solution of the iron and oxidation of the phosphorus 
 to phosphoric acid, which is then separated from the iron by pre- 
 cipitation as ammonium phospho-molybdate. The estimation of 
 the amount of phosphoric acid may then be carried out gravi- 
 metrically or volumetrically by any of the methods given in 
 Chapters IX and XXVII. 
 
 549. Oxidation with Nitric Acid and Separation of Silica. 
 From 1 to 5 grams of the iron or steel, according to the amount of 
 phosphorus present, are dissolved in nitric acid, sp. gr. 1.2, about 
 15 c.c. being used for each gram of iron. Evaporate to dryness, 
 and heat in the air-bath at 200 for one hour. At this tempera- 
 ture the oxidation of the phosphorus is completed. The residue 
 is treated with 30 c.c. concentrated hydrochloric acid, and warmed 
 until the oxide of iron is dissolved. The solution is evaporated to 
 dryness again to dehydrate the silica. Dissolve once more in 
 
 * Chem. News. XXVIII. 229. 
 
PHOSPHORUS. 373 
 
 30 c.c. hydrochloric acid. Evaporate the solution to a sirupy 
 consistency, dilute, and filter. The insoluble residue may be 
 used for the determination of silicon, if that is desired. 
 
 550. Separation of Titanium. If titanium is present it will form 
 an insoluble phosphate which will remain with the silica and 
 cannot be dissolved by means of hydrochloric acid. The paper 
 containing the washed silica and graphite is transferred to a plati- 
 num crucible and burned. The silica is volatilized by treatment 
 with hydrofluoric and a few drops of concentrated sulphuric acid 
 and heating until the sulphuric acid is volatilized. The residue, 
 consisting of Ti0 2 and P 2 5 , is fused for one half hour with 2 or 3 
 grams of sodium carbonate, the melt dissolved in water, and the 
 solution containing the P 2 5 as sodium phosphate is filtered. The 
 filtrate from the insoluble sodium titanate is acidified with nitric 
 acid and added to the first filtrate from the impure silica. The 
 hydrochloric acid is then expelled by addition of 30 c.c. of concen- 
 trated nitric acid and evaporating nearly to dryness. 
 
 551. The Phosphoric Acid may be Precipitated, after nearly 
 neutralizing the nitric acid with ammonia, by means of molybdate 
 solution and determined gravimetrically as directed in Chapter IX, 
 or volumetrically as directed in Chapter XXVII. 
 
 552. Oxidation with Potassium Permanganate. The destruc- 
 tion of the organic matter as well as the complete oxidation of the 
 phosphorus in the solution of the iron in nitric acid may be advan- 
 tageously accomplished by means of potassium permanganate as 
 well as by heating the dry nitrate at 200. After dissolving the 
 iron in nitric acid, 10 c.c. of a 1.5% solution of potassium perman- 
 ganate is added and the solution boiled until the pink color is 
 destroyed and the manganese is reduced to the dioxide. This 
 precipitate is dissolved by the addition of small amounts of sul- 
 phurous acid, ferrous sulphate, or sodium thiosulphate. When 
 the solution has cooled to about 40, it is neutralized with 
 ammonia, the precipitate dissolved with a little nitric acid, and 
 the phosphoric acid precipitated by the addition of molybdate 
 solution. 
 
 553. The Molybdate Precipitate may be directly weighed as 
 directed in Chapter IX. It may also be titrated with standard 
 caustic potash or soda as directed in Chapter XXVII, or with 
 
374 ANALYSIS OF IRON AND STEEL. 
 
 standard potassium permanganate solution after reduction with 
 zinc as directed in the same chapter. 
 
 DETERMINATION OF CARBON. 
 
 Carbon exists in iron and steel in several conditions, and the 
 qualities of the metal are considerably influenced by the condition 
 in which the carbon exists. Analytical methods are in use for 
 determining the amount of carbon existing in two conditions. The 
 so-called FREE or GRAPHITIC CARBON as well as the total amount of 
 carbon are determined directly. The COMBINED CARBON is usually 
 determined by difference. 
 
 TOTAL CARBON. 
 
 a. SOLUTION OF IRON AND OXIDATION OF CARBON BY 
 CHROMIC ACID. 
 
 This is a very accurate as well as a rapid and largely used 
 method of determining carbon in all grades of iron and steel, with 
 the exception of ferric chrome and irons high in silicon. The 
 carbon is oxidized by the chromic acid to carbon dioxide, which 
 may be absorbed in an appropriate apparatus and weighed or the 
 volume of the gas may be measured and the weight calculated. 
 
 554. The Following Solutions are prepared for the combustion 
 of the carbon- 
 
 1. Two hundred grams of copper sulphate are dissolved in a 
 liter of water. 
 
 2. A saturated solution of chromic acid. 
 
 3. A solution made by taking 35 c.c. of solution 2, 190 c.c. of 
 water, 750 c.c. of concentrated sulphuric acid, and 340 c.c. of 
 phosphoric acid, sp. gr. 1.4. 
 
 After adding a few cubic centimeters of sulphuric acid, solution 2 
 should be heated to boiling to destroy any organic matter that 
 might be present. Solution 3 should also be heated to boiling before 
 use. 
 
 555. The Apparatus is set up as shown in Fig. 60. The round- 
 bottomed flask A should have a capacity of about 250 c.c. The 
 short piece of combustion tubing B is filled with lumps of copper 
 oxide which are kept in place by short coils of copper gauze. The 
 
TOTAL CARBON. 
 
 375 
 
 tube is placed in a short tube furnace or is heated by several 
 Bunsen burners. This tube is heated to redness during the com- 
 bustion of the carbon in order to decompose the hydrocarbons 
 evolved. If it is omitted, the result must be increased 2% for 
 malleable iron and steel and 3J% for pig iron. The presence of 
 copper sulphate in the combustion-flask reduces the amount of 
 the hydrocarbons evolved. The U-tube C is filled with calcium 
 chloride for drying the carbon dioxide which is absorbed in the 
 caustic potash bulb D. If this bulb is not fitted with a small 
 calcium chloride tube, the U-tube E must be weighed as well as 
 
 FIG. 60. 
 
 the caustic potash bulb D. Another calcium chloride tube must 
 then be inserted between the U-tube E and the aspirator. The 
 U-tube attached to the dropping-funnel is filled with soda-lime 
 and serves to remove carbon dioxide from the air drawn through 
 the apparatus. The apparatus is set up and tested for leaks as 
 already directed for the determination of carbon dioxide in Chap- 
 ters IX and XV. 
 
 556. Process. From 1 to 5 grams of the iron, depending on the 
 amount of carbon present, is weighed out and placed in the flask A. 
 15 c.c. of copper sulphate solution is introduced through the drop- 
 ping^funnel. After one or two minutes introduce 15 c.c. of the 
 chromic acid solution and 135 c.c. of solution 3. Allow a slow 
 stream of air to pass through the apparatus, and heat the solution 
 gently with the Bunsen burner, bringing it to a boil in fifteen to 
 twenty minutes. Continue the boiling for about two hours. 
 
376 ANALYSIS OF IRON AND STEEL. 
 
 Aspirate about 2 liters of air through the apparatus, disconnect 
 the absorption-tubes, and weigh after allowing them to remain 
 in the balance-case for about fifteen minutes. 
 
 b. WEIGHING OF CARBON AFTER SEPARATION OF THE IRON BY 
 MEANS OF POTASSIUM CUPRIC CHLORIDE SOLUTION. 
 
 557. Action of Copper Solutions. On treating iron or steel 
 with a solution of copper, the copper is displaced by the iron which 
 unites with the acid, forming a ferrous salt. The combined carbon 
 as well as the graphitic carbon remains undissolved, no hydrogen or 
 hydrocarbons being evolved. 
 
 Formerly a neutral solution of COPPER SULPHATE was employed. 
 This has largely been replaced by a solution of CUPRIC AMMONIUM 
 CHLORIDE, by which the iron is dissolved much more rapidly, the 
 precipitated copper also going into solution. On account of the 
 difficulty of obtaining the double ammonium salt free from organic 
 material, the use of the double chloride of potassium and copper 
 has arisen. It has also been recently shown by the American 
 members of the International Steel Standards Committee that 
 more accurate results are obtained by the use of a copper solution 
 containing about 10 per cent of hydrochloric acid. The reactions 
 taking place may be represented as follows: 
 
 2 = FeCl 2 +Cu; 
 Cu+CuCl 2 = 2CuCl. 
 
 The cuprous chloride produced in this manner may be recon- 
 verted into cupric chloride by passing chlorine through the solu- 
 tion until it smells strongly of chlorine. Blair* states that a solu- 
 tion regenerated in this manner is more active than a fresh solution. 
 The same solution may be used repeatedly until the concentration 
 of the iron salts is too great. Blair has used the same solution 
 eleven times with good results. 
 
 558. The Solution of the Iron is effected by treating 1 gram of 
 pig iron, Spiegel, or ferromanganese with 100 c.c. of a saturated 
 solution of the potassium cupric chloride and 7.5 c.c. of concen- 
 trated hydrochloric acid. 3 grams of steel or puddled iron are 
 
 * Chemical Analysis of Iron, 1901, p. 165. 
 
TOTAL CARBON. 377 
 
 taken and treated with 200 c.c. of the copper solution and 15 c.c. 
 strong hydrochloric acid. The iron is dissolved very much more 
 rapidly if the solution is stirred. After digestion for several 
 minutes at the ordinary temperature the solution may be warmed 
 gently, but not higher than 60 or 70. If many determinations 
 are to be made, a mechanical device for stirring should be used. 
 
 559. Direct Weighing of Carbon. When the iron and the 
 precipitated copper are completely dissolved* the carbon is fil- 
 tered off and washed with hot water and a little hydrochloric acid 
 to dissolve the copper and iron salts. If the iron or steel contains 
 only combined carbon and is free from graphite the carbon may 
 be weighed directly, after being filtered off on a Gooch crucible 
 and dried at 100. The carbon is then burned off and the residue 
 weighed. The difference is carbonaceous matter which contains 
 about 70% of carbon. 
 
 c. COMBUSTION OF SEPARATED CARBON IN A STREAM OF OXYGEN. 
 
 560. Filtration of Carbon on a Platinum Boat. A more common 
 and generally applicable method consists in burning the separated 
 carbon in oxygen or chromic acid and collecting and weighing or 
 measuring the carbon dioxide produced. If the carbon is to be 
 burned in a stream of oxygen, the most suitable apparatus on which 
 to filter it off and wash it is a platinum boat with perforated bottom 
 which is of such a size that it may be placed in the glass or porcelain 
 tube used for the combustion. A suitable holder is provided for the 
 boat so that it may be used in the same manner as a Gooch crucible. 
 Asbestos which has been ignited so as to free it from organic matter 
 is made into a pulp by mixing with water, floated on the bottom 
 of the boat and sucked into a smooth compact layer with the 
 filter-pump. After filtering off the carbon on the boat and wash- 
 ing it, the whole is dried at 100 and transferred to the combus 
 tion-tube. Carbonaceous matter adhering to the sides of the beaker 
 in which the iron was dissolved is wiped off with a wad of ignited 
 asbestos held by a pair of forceps. The asbestos is then trans- 
 ferred to the boat. 
 
 * The metallic copper may easily be seen because of its bright-red color. : .* 
 
378 ANALYSIS OF IRON AND STEEL. 
 
 561. Filtration of Carbon on a Carbon Filter. Instead of the 
 perforated boat and holder a small platinum tube fitting into the 
 combustion-tube may be used for filtering off the asbestos. The 
 asbestos is held on a perforated platinum disk resting on a shoulder 
 hi the tube. After the carbon has been filtered off and washed, 
 the disk is loosened by pushing it further into the tube by means 
 of a glass or platinum rod. This is done to provide a passageway 
 for the oxygen used in the combustion. After drying, the tube con- 
 taming the asbestos and the carbon is inserted into the combustion- 
 tube. An ordinary glass carbon filter may also be used in which 
 the asbestos rests on a perforated platinum or procelain disk or a 
 Spiral of platinum wire. After filtering off the carbon, the disk 
 with the asbestos and the carbon is pushed out of the tube and 
 transferred to a platinum boat. The particles of asbestos and 
 carbon still adhering to the glass tube are wiped out by wads of 
 ignited asbestos held in a pair of forceps. The boat and contents 
 are then dried at 100 and transferred to the combustion-tube. 
 
 562. A Platinum, Glass, or Porcelain Combustion-tube may be 
 used. It should be long enough to extend several centimeters 
 beyond the furnace, so that the rubber stoppers are not softened 
 by the heat. The forward half of the portion of the tube which is 
 in the furnace is filled with granulated copper oxide which is held 
 in place by short rolls of copper gauze. Another roll of copper 
 gauze, to which a hook made of copper wire is fastened, is pro- 
 vided for insertion into the tube after the boat has been placed 
 in the empty rear end. 
 
 563. A Stream of Oxygen is obtained from a steel tank which 
 has been filled under pressure or from an ordinary gasometer which 
 has been filled with oxygen by heating gently in a retort an intimate 
 mixture of 1 part of powdered manganese dioxide and 20 parts of 
 potassium chlorate. When the oxygen begins to be given off from 
 this mixture the heating with the Bunsen burner must be discon- 
 tinued or the evolution of gas will be too violent. The oxygen is 
 purified by passing through a large U-tube or tower filled with 
 soda-lime and pieces of fused caustic potash. A similar apparatus 
 should be provided for purifying air. 
 
 564. Absorption Apparatus. The carbon obtained from iron 
 and steel is almost never free from sulphur. Some chlorides are 
 
TOTAL CARBON. 379 
 
 also generally present which cannot be washed out. Sulphur 
 dioxide, hydrochloric acid, and chlorine are therefore given off 
 on burning the carbon, and these gases must be absorbed before 
 the carbon dioxide is allowed to pass into the caustic potash. A 
 spiral of silver foil should be placed in the end of the combustion- 
 tube for the absorption of the chlorine. A U-tube filled with 
 pieces of pumice-stone which have been saturated with copper- 
 sulphate solution and thoroughly dried at a little over 200 should 
 be connected with the combustion-tube for the absorption of 
 hydrochloric acid. For the absorption of sulphur dioxide a layer 
 of lead peroxide is placed in the bottom of this tube and separated 
 from the copper sulphate by means of wads of asbestos or cotton 
 wool. 
 
 A U-tube filled with calcium chloride, a caustic potash bulb 
 for the absorption of carbon dioxide, and a guard tube containing 
 soda-lime and calcium chloride are attached in the order mentioned. 
 After the apparatus has been tested for leaks in the usual manner, 
 the combustion-tube is heated to redness and oxygen and air 
 passed through to burn out any organic matter which may be 
 present. 
 
 565. The Combustion. The caustic potash bulb is weighed 
 and replaced in the series of U-tubes. The boat containing the 
 carbon is inserted in the combustion-tube and pushed in until it 
 is close to the layer of copper oxide. The roll of copper gauze is 
 inserted in the end of the combustion-tube, which is closed with 
 the rubber stopper connecting with the supply of oxygen. The 
 apparatus is tested for leaks by sucking out a little air so that the 
 liquid rises in one limb of the caustic potash bulb. If this column 
 of liquid remains at a constant height for at least five minutes the 
 apparatus is tight. A slow stream of oxygen is now admitted, 
 and two burners are lit under the forward end and one under the 
 copper gauze in the rear end of the combustion-tube. When 
 these portions of the tube are red-hot, the third burner from the 
 forward end is lighted and then the other burners in order with a 
 few minutes ' interval. If a glass tube is used, the combustion may 
 be observed by the glowing of the carbon. The combustion is 
 usually complete fifteen minutes after the tube has become red- 
 hot. The asbestos in the boat then appears nearly pure white. 
 
380 ANALYSIS OF IRON AND STEEL. 
 
 The burners are then turned down and, after a few minutes, extin- 
 guished so as to cool the tube gradually. The stream of oxygen 
 is replaced by a stream of air of which from 1 to 2 liters should be 
 passed through the apparatus to sweep out the carbon dioxide. 
 Neither the oxygen nor the air should pass faster than three bub- 
 bles per second. The absorption apparatus is then disconnected, 
 wiped perfectly clean with a linen handkerchief, and after stand- 
 ing in the balance-case for not more than one-half hour is weighed. 
 Until experience has been attained in conducting the combustion, 
 it is advisable to insert the absorption apparatus and repeat the 
 combustion to ascertain if any of the carbon remains unburned. 
 
 566. If Several Determinations are to be made a duplicate 
 absorption apparatus should be prepared and weighed while the 
 first combustion is in progress. When the first absorption appara- 
 tus is disconnected, the second is inserted, and a boat contain- 
 ing the carbon from another sample of iron is inserted in place of 
 the first boat. During the progress of the second combustion the 
 absorption apparatus of the first is weighed and may then be used 
 for the third determination. When work is discontinued the 
 entrance of air or other gases into the apparatus is prevented by 
 closing all openings with glass plugs. 
 
 The carbon from which the iron has been separated may also 
 be burned with chromic and sulphuric acids. 
 
 d. COMBUSTION OF THE SEPARATED CARBON WITH CHROMIC ACID. 
 
 567. -An Apparatus similar to that shown in Fig. 61 is used. 
 The flask A has a capacity of about 250 c.c. The U-tube B con- 
 tains a sufficient amount of a solution of silver sulphate in con- 
 centrated sulphuric acid to fill the curved part of the tube. An 
 empty TJ-tube C is placed next and then a U-tube D containing 
 pumice-stone which has been saturated with copper sulphate solu- 
 tion and dried. The usual U-tube filled with calcium chloride for 
 drying the carbon dioxide, the caustic potash bulbs, and the guard 
 tube are then connected in the order mentioned. 
 
 568. The Combustion. The asbestos containing the carbon 
 is transferred to the flask and, after inserting the stopper, 10 c.c. 
 
TOTAL CARBON 
 
 381 
 
 of a saturated solution of chromic acid is introduced through the 
 funnel, followed by 100 c.c. of concentrated sulphuric acid which 
 has been brought to the boiling-point after the addition of a little 
 chromic acid. A slow current of air is passed through the apparatus 
 and the flask is gently heated with the Bunsen burner and gradu- 
 ally brought nearly to the boiling-point. The heating is con- 
 
 FIG. 61. 
 
 tinued until no more gas is evolved and the oxidation is complete. 
 The current of air is continued until from 1 to 2 liters have passed 
 through the apparatus, when the absorption-bulbs are discon- 
 nected and weighed. 
 
 e. DETERMINATION OF CARBON AFTER VOLATILIZATION OF THE 
 IRON IN A STREAM OF CHLORINE. 
 
 The carbon in irons containing chromium and a large percentage 
 of silicon can be determined only after volatilization of the iron 
 in a stream of chlorine. 
 
 569. Separations Effected. The iron is converted into ferric 
 chloride, which can be completely expelled by heating the tube. 
 The silica is also volatilized as silicon tetrachloride, SiQ 4 , and 
 may be absorbed in water which will also contain any titanium 
 present in the iron, being volatilized as titanium tetrachloride, 
 TiCl 4 . These elements may be separated by adding to the solu- 
 tion concentrated sulphuric acid and evaporating to fumes. The 
 silicic acid is dehydrated and may be filtered off and weighed 
 while the titanium remains in solution. Any metals present 
 
382 ANALYSIS OF IRON AND STEEL. 
 
 which form volatile chlorides will also be present in this solution. 
 In the boat hi which the iron was placed will remain all of the 
 carbon, most of the manganese, chromic chloride, and the slag 
 which was present in the iron. 
 
 570. Process. The chlorine is generated in a large flask of 
 about l.V liters capacity, charged with 190 grams of manganese 
 dioxide intimately mixed with 280 grams of sodium chloride. 
 The chlorine is evolved on adding from a dropping-funnel 350 c.c. 
 of a mixture of equal parts of concentrated sulphuric acid and 
 water. The chlorine is washed by passing through water and 
 dried by passing through concentrated sulphuric acid. The iron 
 is spread over the bottom of a porcelain boat which is placed in a 
 glass combustion-tube which has been drawn out and bent at right 
 angles where it emerges from the furnace so that it can dip under 
 water. All connections must be made of glass as far as possible, 
 the inner surface of rubber stoppers being coated with paraffine. 
 The air is displaced from the apparatus by the stream of chlorine 
 and the combustion-tube heated just sufficiently to expel the ferric 
 chloride from the boat. This is usually accomplished by heating 
 the tube to dull redness. The chlorides are dissolved out with 
 water and the carbon filtered off on ignited asbestos, which is re- 
 turned to the boat and dried. The carbon may then be burned 
 in a combustion-tube in a stream of oxygen, or it may be burned 
 with chromic and sulphuric acids as already directed. 
 
 DETERMINATION OF GRAPHITIC CARBON. 
 
 571. On dissolving iron and steel in acid, part of the carbon 
 remains undissolved while part goes into solution or is evolved in 
 combination with hydrogen. As graphite is insoluble in dilute 
 acid, all of the carbon present in this form remains undis- 
 solved. If dilute hydrochloric acid is used a part also of the 
 combined carbon remains undissolved. If nitric acid of specific 
 gravity 1.2 is used, only the graphitic carbon remains undis- 
 solved. For this determination 1 gram of pig iron is taken, 
 while as much as 10 grams of steel containing a very small per 
 cent of graphitic carbon is taken. 15 c.c. of the nitric acid is 
 taken for each gram of iron. The first violent action of the acid 
 
COMBINED CARBON. 383 
 
 is moderated by cooling the beaker with water. When all of the 
 iron is dissolved, dilute the acid, filter off the carbon on ignited 
 asbestos, and wash with water until free from iron. Treat the 
 residue on the filter with a hot solution of potassium hydroxide 
 of sp. gr. 1.1 to remove silica, and wash with water, then with 
 dilute hydrochloric acid, and finally with water. Determine the 
 amount of carbon present by burning in oxygen or in chromic 
 acid as already directed. 
 
 DETERMINATION OF COMBINED CARBON. 
 
 The amount of combined carbon may be found indirectly by 
 subtracting the amount of graphitic carbon from the total amount 
 of carbon found. 
 
 572. Colorometric Method of Eggertz. This method is founded 
 on the fact that when steel is dissolved in nitric acid of 1.2 
 specific gravity a brown color is given to the solution which is 
 proportional to the amount of combined carbon present. Two 
 specimens of steel which give the same depth of color contain the 
 same percentage of combined carbon. If the colors differ, the 
 darker solution may be diluted until the depth of color is identi- 
 cal. The ratio of the percentage of combined carbon in the two 
 samples of steel will then be the same as the ratio of the volumes 
 of the two solutions. If the percentage of combined carbon in 
 one of the steels is known, the amount present in the other may 
 be calculated. A so-called standard steel must therefore be se- 
 lected and the percentage of combined carbon determined by one 
 of the methods already given. 
 
 573. Standards. It has been found, however, that two samples 
 of steel containing exactly the same percentage of combined carbon 
 give different colors if one sample has been treated differently, so 
 far as tempering, hammering, etc., or if it has been made by a 
 different process of manufacture. This peculiarity is believed to 
 be due to the fact that the carbon exists combined in two different 
 conditions, and when existing in one condition does not impart the 
 same depth of color to the acid as when it exists in the other condi- 
 tion. Standards should therefore be at hand for each kind of 
 steel to be tested, such as Bessemer, crucible, open-hearth, etc,, 
 
384 ANALYSIS OF IRON AND STEEL. 
 
 and the rolling, hammering, tempering, etc., of the standard should 
 as far as possible be the same as that of the sample to be analyzed. 
 The percentage of carbon should also be as nearly as possible the 
 same in the standard as in the steel to be analyzed. 
 
 574- Solution of the Steel. A sufficient number of test-tubes 
 6 inches long and f inch in diameter are taken and 2 grams of each 
 sample, including the standard, are weighed out and placed in a 
 test-tube. The test-tubes are carefully numbered to correspond 
 to the numbers of the samples, and are placed in beakers or bottles 
 containing cold water. Nitric acid of 1.2 specific gravity is then 
 added to the test-tubes. Acid of this strength is made by diluting 
 concentrated nitric acid (sp. gr. 1.4) with an equal volume of 
 water. The acid is added from a burette, 3 c.c. being used for steels 
 containing .3% of carbon; 4 c.c. for .3 to .5% of carbon; 5 c.c. 
 for .5 to .8%; 6 c.c. for .8 to 1%; and 7 c.c. for over 1%. If the 
 amount of carbon present is unknown, 3 c.c. of the acid should be 
 added at first, and then successive portions of 1 c.c. as the increase 
 in the depth of color or the presence of undissolved carbonaceous 
 matter indicates the necessity for more acid. The nitric acid 
 used must be absolutely free from chlorine. A small funnel or a 
 glass bulb of suitable size should be placed in the mouth of each 
 test-tube. The test-tubes are now immersed in the water cf a 
 water-bath while held upright by some suitable device, such as a 
 metallic disk in which holes of the proper size to admit the tubes 
 have been cut. The water is heated to boiling and the tubes 
 shaken at intervals until the carbonaceous matter is completely 
 dissolved, which will require from fifteen to forty-five minutes. As 
 soon as solution is complete each tube is removed from the water- 
 bath and placed in cold water, the light being excluded as much 
 as possible. Direct sunlight bleaches the color very quickly. 
 
 575. For Comparing the Color cf the Solutions Eggertz used 
 graduated tubes. These tubes are of uniform bore, made of clear, 
 colorless glass, and are carefully graduated from the bottom up- 
 wards, having a capacity of 30, or more, cubic centimeters. Trans- 
 fer the solution having the lightest color to one of the tubes, using 
 the same amount of water to rinse the test-tube as the volume of 
 the solution. Mix the resulting solution thoroughly. In every 
 case the acid solution of the steel must be diluted with at least as 
 
MANGANESE. 385 
 
 much water as the amount of acid used to dissolve the steel in 
 order to destroy the color due to the ferric nitrate. 
 
 The solution of the steel to be compared with the one first 
 diluted is transferred to another tube in the same manner, and 
 the color of the two solutions is compared by holding the tubes 
 vertically and side by side in front of a piece of white paper between 
 the observer and the light. The darker solution is diluted by the 
 addition of water until the depth of color in the two tubes is iden- 
 tical. The percentage of combined carbon known to be present 
 in the standard steel is divided by the volume of the solution of this 
 steel. The quotient gives the amount of carbon represented by 
 each cubic centimeter of the solution. This number multiplied 
 by the number of cubic centimeters to which the solution of the 
 steel to be analyzed was diluted gives the percentage of combined 
 carbon present. 
 
 DETERMINATION OF MANGANESE. 
 
 The difficulties attending the determination of manganese in 
 iron and steel arise mainly from the presence of the large amounts 
 of iron. The iron may be precipitated as basic acetate, but the 
 precipitation must be repeated several times, and if the percentage 
 of manganese is small, so that a large amount of iron must be 
 taken for the analysis, the basic acetate precipitate becomes very 
 large and unwieldy. The separation of the iron by solution of 
 the ferric chloride in ether is a more rapid and convenient method, 
 which has already been described in Chapter XIII, p. 160. 
 
 a. FORD'S METHOD. 
 
 Equally as efficient is a method known in this country as 
 Ford's method.* Advantage is taken of the fact that potassium 
 chlorate converts the managnese present in strong nitric acid solu- 
 tion into manganese dioxide, which can be filtered off from the 
 iron which remains in solution. The separated manganese may 
 
 * Trans. Inst. Min. Engineers, IX, 397, X, 100. 
 
386 ANALYSIS OF IRON AND STEEL. 
 
 then be determined gravimetrically or volumetrically. The reac- 
 tions taking place may be represented as follows: 
 
 KC10 3 + HN0 3 = KN0 3 + HC10 3 ; 
 
 2HC10 3 + Mn(N0 3 ) 2 = Mn0 2 + 2C10 2 + 2HN0 3 . 
 
 576. Solution of the Steel. 5 to 10 grams of steel or pig iron 
 are taken, while only 1 gram of spiegel or ferro-manganese contain- 
 ing 20 to 40% of manganese, and \ gram if the per cent of manga- 
 nese is still higher. If less than 0.2% of silicon is present, the 
 weighed portion is dissolved in 60 c.c. of nitric acid of 1.2 specific 
 gravity and the solution evaporated to small bulk. A large beaker 
 should be used and the acid added cautiously. If the percentage 
 of silicon is high, the steel should be dissolved in hydrochloric 
 acid and the solution evaporated to dryness to dehydrate the 
 silica, which otherwise forms a gelatinous mass and interferes with 
 the subsequent filtration. Pig iron is also dissolved in hydro- 
 chloric acid, the solution filtered, and the filtrate evaporated to 
 dryness. The hydrochloric acid is removed in either case by 
 evaporation with 50 c.c. of strong nitric acid to a small bulk. 
 The silica may also be removed by dissolving the iron in nitric 
 acid to which a few drops of hydrofluoric acid have been added 
 and evaporating to a small bulk. 
 
 577. Precipitation of the Manganese. This solution, as well as 
 that obtained from the steel low in silica, is treated with 100 c.c. 
 of concentrated nitric acid and 5 grams of potassium chlorate 
 added. The solution is boiled for fifteen minutes, 50 c.c. strong 
 nitric acid are added, and 5 grams of potassium chlorate, and the 
 solution again boiled for fifteen minutes. When no more chlorine 
 fumes are evolved the solution is rapidly cooled by standing the 
 beaker in cold water and the precipitate filtered off on asbestos. 
 It is washed with strong nitric acid and then with water. It is 
 advisable to keep the strong nitric acid filtrate separate from the 
 wash-water and to test it for manganese by adding a little more 
 potassium chlorate and again boiling. 
 
 578. If the Manganese is to be Determined Gravimetrically 
 the manganese dioxide is dissolved in sulphurous acid or hydrogen 
 peroxide and hydrochloric acid. The excess of the sulphur dioxide 
 
MANGANESE. 387 
 
 or hydrogen peroxide is expelled by boiling the solution. The 
 small amount of iron present is oxidized by the addition of a little 
 bromine water the excess of which must be boiled out. It is precipi- 
 tated by adding an excess of ammonia to the hot solution. The 
 precipitate is filtered off, washed once or twice, dissolved in hydro- 
 chloric acid, and the solution boiled to expel a trace of chlorine 
 from the manganese dioxide and reprecipitated with ammonia, fil- 
 tered off, and washed thoroughly. The combined filtrates con- 
 tain all of the manganese and a little cobalt if that element was 
 present in the iron. The cobalt may be precipitated by passing 
 hydrogen sulphide through the solution after acidifying with acetic 
 acid and heating to boiling. After filtering off the precipitate 
 and boiling the solution to expel the excess of hydrogen sulphide, 
 the manganese may be determined either as sulphide or as phos- 
 phate as directed in Chapters VI and VII. 
 
 579. The Manganese may also be Determined Volumetrically 
 by dissolving the manganese dioxide in excess of oxalic acid or a 
 pure ferrous salt and titrating the excess with a standard solution 
 of potassium permanganate (see pp. 322, 325). As the per cent 
 of oxygen in the precipitate may not be exactly that correspond- 
 ing to the formula Mn0 2 , it is advisable to standardize the potas- 
 sium permanganate solution by using the precipitate obtained 
 from a sample of steel or iron in which the manganese has been 
 determined gravimetrically. The manganese dioxide obtained 
 from a weighed amount of this iron is dissolved in excess of a 
 reducing agent and the excess determined by means of the per- 
 manganate solution. 
 
 6. VOLHARD'S METHOD. 
 
 580. Solution of the Metal. This very accurate method of 
 determining manganese may be used for determining this element 
 in all samples of iron and steel except those containing very 
 minute amounts of manganese. Weigh out 5 grams of ferro- 
 manganese and from 1 to 2.5 grams of spiegeleisen, white pig iron, 
 steel, etc. Dissolve * in a mixture of 275 c.c. of water 125 c.c. of 
 
 * Reis, Zeit. f. angew. Ch., 1892, pp. 604, 672. 
 
388 ANALYSIS OF IRON AND STEEL. 
 
 concentrated nitric acid (sp. gr. 1.4) and 100 c.c. concentrated 
 sulphuric acid. Use from 25 to 50 c.c. of this acid mixture, accord- 
 ing to the amount of iron taken. After the iron is dissolved, evapo- 
 rate the solution until fumes of sulphuric acid are evolved. After 
 cooling, add 100 c.c. of water and 10 c.c. of the acid mixture and 
 warm until solution is complete. Transfer to a flask of 750- to 
 1000-c.c. capacity and add 3 grams of barium peroxide and 5 c.c. 
 concentrated nitric acid to oxidize the hydrocarbons present. 
 Boil several minutes to decompose the excess of barium peroxide. 
 
 581. Titration. Dilute the solution until the flask is about one- 
 half full, add zinc oxide suspended in water, shaking the solution 
 after each addition until the solution is neutral and the iron is pre- 
 cipitated, leaving it colorless or slightly milky with the excess of 
 zinc oxide. Warm the solution to about 90, but not to boiling, 
 and titrate with standard potassium permanganate solution as 
 directed in Chapter XXIV, page 323. 
 
 582. If the Zinc Oxide contains manganese it may be purified 
 as follows: Dissolve the impure oxide in hydrochloric acid, finally 
 adding an excess of the zinc oxide so that part of it remains undis- 
 solved. Add a little bromine and heat the solution to boiling and 
 filter. Precipitate the zinc oxide by means of ammonia, carefully 
 avoiding an excess. Wash the oxide thoroughly and then transfer 
 to a bottle with water. Shake the bottle well before withdrawing 
 any of the oxide for use. 
 
 c. DESHAY'S METHOD.* 
 
 This method is based on the fact that manganese nitrate 
 when boiled with nitric acid and lead peroxide is converted into 
 permanganic acid, which may then be titrated with a standard 
 solution of sodium arsenite. The method is especially applicable to 
 iron and steel containing less than 2% of manganese. It cannot 
 be used when chromium is present, as chromic acid is produced. 
 
 583. Process. 0.5 gram of steel or pig iron is dissolved in a 
 small beaker or test-tube in 30 c.c. nitric acid of 1.2 specific gravity. 
 The solution is boiled until the nitrous fumes are completely 
 
 * Bull. Soc Chim de Paris. June 20, 1878. 
 
MANGANESE. 389 
 
 expelled. From 1 to 3 grams of lead peroxide or red lead free 
 from manganese are then added and the solution diluted with 
 hot water to about 60 c.c. It is boiled for three or four minutes 
 and after allowing the peroxide of lead to settle, the solution is 
 decanted and the residue again boiled after the addition of 50 c.c. 
 of dilute nitric acid. If the solution again becomes colored with 
 permanganic acid, it must be decanted and the boiling repeated 
 after the addition of dilute nitric acid. The solution is filtered 
 through asbestos which is free from organic matter. The filtrate is 
 titrated with a standard solution of sodium arsenite. 
 
 584. The Sodium Arsenite Solution may be made by dissolving 
 4.5 grams of pure arsenious oxide in water to which 25 grams of 
 sodium carbonate has been added and diluting the solution to 2 
 liters. This solution should then be standardized by comparison 
 with a standard potassium permanganate solution, or a still better 
 plan is to standardize it with a sample of steel in which the man- 
 ganese has been determined gravimetrically. This is done by 
 weighing out \ gram of the steel, dissolving in nitric acid, and 
 oxidizing with peroxide of lead. The permanganic acid produced 
 in this manner is titrated with the arsenite solution to be standard- 
 ized. The amount of manganese corresponding to 1 c.c. of the 
 arsenite solution may then be calculated from the known amount 
 of manganese in the steel used. 
 
 585. Separation of Lead Peroxide by a Centrifugal Machine. 
 Instead of filtering the permanganic acid from the lead peroxide, 
 the separation may be effected by means of a centrifugal machine. 
 In order to secure complete oxidation of the manganese by boiling 
 once, a smaller amount of the steel is taken, from .05 to .10 gram 
 being used. 10 c.c. of the nitric acid of sp. gr. 1.2 are added to 
 dissolve the iron and after the nitrous fumes have been expelled 
 by boiling, 10 c.c. of hot water and \ gram of lead peroxide are 
 added and the solution boiled for three minutes. The solution 
 is washed into a wide-mouthed two-ounce bottle, enough water 
 being used to make 50 c.c. The bottle is then placed in the cen- 
 trifugal machine and rotated for two minutes. The liquid will 
 then be perfectly clear and may be poured out without disturbing 
 the lead peroxide, which forms a compact layer on the bottom of 
 the bottle. The permanganic acid is then titrated with a solution 
 
390 ANALYSIS OF IRON AND STEEL. 
 
 of sodium arsenite which is equal to about .0001 gram of man- 
 ganese per cubic centimeter. 
 
 PROXIMATE ANALYSIS OF COAL. 
 
 The methods used in the analysis of coal do not give the 
 amount of any definite chemical compound which is present in the 
 coal. The determination of MOISTURE gives very nearly the amount 
 of water present, but as this constituent is obtained by loss in 
 weight after drying the coal in the air, an error is introduced 
 because SULPHUR and other constituents present are oxidized at 
 the temperature used for drying the coal, and therefore the coal 
 loses weight for a time, and then begins to increase in weight. 
 The conditions under which this as well as other determinations 
 are made are somewhat arbitrarily fixed so that the results obtained 
 by various chemists may be comparable with each other and form 
 a basis for the comparison of the coals analyzed. The methods 
 given in this chapter are those adopted by a committee of the 
 American Chemical Society, consisting of Wm. A. Noyes, W. F. 
 Hillebrand, and C. B. Dudley.* 
 
 586. The Determination of Moisture is made by heating 1 gram 
 for one hour at 104 to 107. This temperature may most readily 
 be maintained by using a double-walled air-bath in which pure 
 toluene is boiled. By attaching a reflux condenser to the usual 
 steam vent, loss of toluene may be obviated. As considerable 
 moisture is lost in powdering the coal, the committee recommends 
 the making of two determinations of moisture, one on the coal as 
 powdered for the determination of sulphur, ash, etc., and another 
 on the material broken up only sufficiently fine for sampling. As 
 it is necessary that the percentages found shall apply to the unpow- 
 dered coal as it is actually used, a correction must be made on the 
 percentages found by using the powdered material. The differ- 
 ence in the percentages of moisture found is divided by the sum 
 of the percentages of the other constituents in the powdered 
 coal. The quotient multiplied by 100 is the percentage by which 
 
 * Jour. Am. Chem. Soc., XXI, 1116. 
 
VOLATILE MATTER. 391 
 
 the percentages of the other constituents are reduced. The illus- 
 tration given in the report mentioned is as follows: 
 
 Coarsely Ground Powdered 
 
 Coal. Coal. 
 
 Moisture 12.07 10.39 
 
 Volatile combustible matter. . 34.25 
 
 The correction factor will be 
 
 12.07-10.39 1.68 
 
 = .0187. 
 
 100-10.39 89.61 
 The true per cent of volatile combustible matter will be 
 
 34.25- (34.25 X .0187) =33.61. 
 
 587. The Volatile Combustible Matter is determined by heating 
 a fresh portion of the powdered coal in a tightly closed platinum 
 crucible with the full flame of a Bunsen burner for seven minutes. 
 The loss in weight less the moisture is the volatile combustible 
 matter, while the residue in the crucible is the coke. 
 
 This determination serves to distinguish the different kinds of 
 coal which are mined for various industrial purposes. The ANTHRA- 
 CITE COALS give only a small per cent of volatile combustible 
 matter, leaving a residue in the crucible which can hardly be 
 distinguished from the powdered coal. The BITUMINOUS and 
 CANNEL COALS melt on being heated in the crucible, and the vola- 
 tile matter in escaping produces a porous mass which is much 
 more bulky than the original coal. This material is what is known 
 in commerce as "COKE." A good bituminous coal gives about 
 60% of coke. Cannel coals, on the other hand, give 60% or 70% 
 of volatile combustible matter with a corresponding smaller per 
 cent of coke. Between these extremes coals may be found, 
 giving nearly all percentages of coke and volatile combustible 
 matter. 
 
 588. The Percentage of Ash is found by burning a weighed 
 portion of the coal in a platinum crucible. The percentage of ash 
 subtracted from the percentage of coke gives the percentage of 
 FIXED CARBON, the volatile combustible matter being composed of 
 carbon which has united with the hydrogen present, forming vola- 
 tile combustible hydrocarbons. 
 
392 ANALYSIS OF COAL. 
 
 It is evident that a low percentage of ash in coal is very desir- 
 able, as this constituent not only has no calorific value, but forms 
 a waste product which must be disposed of. 
 
 589. For Determining Sulphur various methods have been pro- 
 posed and used, the difference being found in the methods of 
 oxidizing the carbon and the sulphur, which are ultimately precipi- 
 tated by means of barium chloride and weighed as barium sul- 
 phate. ESCHKA'S METHOD, by which the coal is burned by the 
 oxygen of the air after mixing with magnesium oxide and sodium 
 carbonate, and the sulphur oxidized by means of bromine water, 
 is one of the most rapid as well as a very accurate method. As 
 the sulphuric acid is extracted from the alkaline flux with hot 
 water, some of the silica in the coal is also dissolved, and this is 
 subsequently carried down with the barium sulphate. More 
 accurate results may be obtained by evaporating the solution to 
 dryness after acidifying with hydrochloric acid in order to dehydrate 
 and remove this silica. 
 
 590. The Sulphur Exists in Coal in at least three conditions: 
 as a METALLIC SULPHIDE, especially PYRITES, as CALCIUM or BARIUM 
 SULPHATE, and as a SULPHURETTED HYDROCARBON. In a proximate 
 analysis of coal about one-half of the sulphur existing as pyrites 
 and all of the organic sulphur is probably volatilized with the 
 volatile combustible matter. The remainder of the sulphur exist- 
 ing as pyrites is expelled during the combustion of the fixed carbon, 
 only that existing as barium or calcium sulphate remaining with 
 the ash unless calcium or barium carbonate was present in the coal. 
 
 EXERCISE 70. 
 Proximate Analysis of Coal. 
 
 The carefully selected sample is coarsely ground in a porcelain mortar 
 and well mixed. A few grams are taken and powdered and placed in a 
 well-stoppered bottle. The main portion of the sample is preserved in the 
 same manner. 
 
 591. Determination of Moisture. The temperature of an air-bath is 
 brought to 104-107 by adjusting the flame of the Bunsen burner. Two 
 platinum or porcelain crucibles are cleaned and, after drying for a few 
 minutes in the air-bath, are cooled in desiccators and weighed. One gram 
 
VOLATILE MATTER. 393 
 
 of the coarsely ground coal is weighed and placed in one of the crucibles, 
 while one gram of the powdered material is weighed out and placed in the 
 other. The uncovered crucibles are then placed in the air-bath and heated 
 for one hour. They are then cooled in the desiccators after replacing the 
 covers, and weighed. If porcelain crucibles have been used, the coal is 
 carefully brushed out and the crucibles weighed, as the latter may have lost 
 weight during the heating. The loss in weight in each case in centigrams 
 gives the percentage of moisture. The percentage found in the coarsely 
 ground material is reported, while the difference between this percentage 
 and that found in the powdered material is used to calculate the correction 
 on the percentage of the other constituents. 
 
 592. Determination of Volatile Combustible Matter- A 20- to 30-gram 
 platinum crucible is cleaned, placed on a platinum or pipe-stem triangle, 
 and ignited with the Bunsen burner. It is cooled in a desiccator and weighed. 
 One gram of the powdered coal is weighed out and transferred to the 
 crucible, which is placed on a large ring of the iron ring-stand, and the 
 height of the ring so adjusted that the bottom of the crucible is 6 to 8 cm. 
 above the top of the Bunsen burner, which must have a clear blue flame 
 fully 20 cm. high when burning free. The cover of the crucible, which must 
 fit tight, is adjusted, and the lighted burner placed under the crucible, which 
 is heated seven minutes. The crucible must be protected from draughts 
 during the heating, and the upper surface of the cover must burn clear. The 
 hot crucible is transferred to the desiccator, and after cooling for fifteen to 
 twenty minutes is weighed. The loss in weight less the moisture in the 
 powdered coal is the volatile combustible matter. Compute the percentage, 
 making the correction for difference in percentage of moisture in the coarsely 
 ground and powdered coal. 
 
 COKE. The residue in the crucible is COKE. The percentage is cor- 
 rected as already directed. 
 
 593- Ash is determined by burning one gram of the coal placed in a 
 platinum crucible. The portion used for the determination of moisture 
 may be taken, or a fresh portion of the powdered material may be weighed 
 out. The coke left after expelling the volatile matter burns with difficulty. 
 The crucible is placed on its side on a pipe-stem or platinum triangle, and 
 is heated at first with a very low flame. The burner must not be placed 
 so that the flame passes in front of the crucible, thus preventing access of 
 air. The combustion is accelerated by bringing fresh portions of coal to 
 the surface by turning the crucible when the carbon has burned out, leaving 
 the nearly white ash on the surface. When no more black particles are visi- 
 ble which burn on exposure to the air, the crucible is cooled in the desiccator 
 and weighed, and the percent of ash calculated. The per cent of coke minus 
 the per cent of ash gives the per cent of FIXED CARBON. 
 
 594. Determination of Sulphur by Eschka's Method. An intimate mix- 
 ture of 1 part of dry sodium carbonate and 2 parts of magnesium oxide is 
 made. The magnesium oxide must be light and porous, not compact and 
 
394 ANALYSIS OF COAL. 
 
 heavy. One and one-half grams of this mixture are weighed and placed in 
 a large platinum crucible, or, still better, in a platinum dish having a capac- 
 ity of 75 to 100 c.c. One gram of the finely powdered coal is weighed out 
 and placed on the magnesia mixture, and the two thoroughly mixed by 
 stirring with a platinum or glass rod. The dish is heated very cautiously 
 with the Bunsen burner, which is at first held in the hand. The heat 13 
 raised very slowly, especially with soft coals. When the strong glowing has 
 ceased the heat is gradually increased, until in fifteen minutes the bottom 
 of the dish is at a low red heat. 
 
 When the carbon is completely burned the residue is transferred to a 
 beaker, the dish is rinsed with about 50 c.c. of water, 45 c.c. of bromine- 
 water is added, and the solution is boiled for five minutes. After allowing 
 the insoluble matter to settle the solution is decanted through a filter. 
 The residue is treated twice with 30 c.c. of water, which is brought to a boil, 
 and decanted through the filter, which is then washed until the filtrate 
 gives only a slight opalescence with silver nitrate and nitric acid. To the 
 filtrate, which should have a volume of about 200 c.c., li c.c. of concen- 
 trated hydrochloric acid or 3 c.c. of dilute acid are added. The solution is 
 boiled until the bromine is expelled, and 10 c.c. of a 10 per cent solution of 
 barium chloride is added drop by drop with constant stirring, especially at 
 first. Digest on the hot plate until the solution is clear, filter oft the barium 
 sulphate, and wash free from chlorides with hot water. Transfer the moist 
 precipitate to a weighed platinum crucible which is heated with a small 
 flame until the paper is burned. Finally heat to redness, cool in the desicca- 
 tor, and weigh. 
 
 A BLANK DETERMINATION should be made to ascertain if the fusion mix- 
 ture is free from sulphur. This determination is carried out exactly as 
 directed for the determination of sulphur in the coal. The amount of 
 barium sulphate obtained is subtracted from that obtained from the coal. 
 The percentage of sulphur is then calculated. 
 
 If the coal contains much pyrite or calcium sulphate, the residue of 
 magnesium oxide must be dissolved in hydrochloric acid and barium chloride 
 added to the hot solution. Any barium sulphate precipitated is filtered 
 and weighed as already directed. 
 
 595. Determination of Sulphur with the Parr Calorimeter. If a determina- 
 tion of the calorific value of the coal is made by means of the Parr calorim- 
 eter, the sulphur may be conveniently determined in the same sample. 
 The sodium peroxide oxidizes the sulphur to sodium sulphate. The cartridge 
 is placed in a porcelain dish and the fused mass dissolved out with hot 
 water. After rinsing the cartridge thoroughly, the solution is acidified 
 with hydrochloric acid and evaporated to dryness on the water-bath to 
 dehydrate the silica. The residue is taken up with water and a little 
 hydrochloric acid, and filtered. After washing the paper free from chlorides, 
 the solution is heated to boiling and the sulphuric acid precipitated by 
 means of barium chloride, which is added slowly and with constant stirring. 
 The precipitate is filtered off on a quantitative paper, washed free from 
 chlorides, and ignited and weighed in the usual manner. 
 
THERMAL UNITS. 395 
 
 DETERMINATION OF THE CALORIFIC VALUE 
 OF FUELS. 
 
 596. Effect of Composition on Calorific Value. The results of a 
 proximate analysis of a fuel give only a general idea of its calorific 
 or heating value by indicating the class in which the fuel belongs. 
 Anthracite coals are characterized by a small percentage of vola- 
 tile combustible matter (usually less than 10%), while bituminous 
 coals contain a large amount of this constituent (usually 30-40%), 
 and in canriel coals this is the main constituent. The proportion 
 of this constituent is dependent on the amount of hydrogen pres- 
 ent, as this element forms volatile compounds with the carbon. 
 When hydrogen burns, it evolves a much larger amount of heat 
 than is produced by the combustion of the same amount of carbon. 
 Coals containing a large proportion of volatile combustible matter 
 have therefore generally a high calorific value. On the other hand 
 the calorific value is reduced by the presence of ash which has no 
 calorific value, and by water which absorbs heat when it is con- 
 verted into steam. It is evident that the calorific value of a 
 fuel could be calculated from the percentage of carbon and 
 hydrogen present. This method is subject to the error arising 
 from the fact that the heat of combustion of hydrogen and car- 
 bon varies with the nature of their chemical combination with 
 each other. Direct methods for determining calorific values are 
 therefore more reliable. The instruments devised for this pur- 
 pose are called calorimeters. 
 
 597. Thermal Units. Various units for measuring heat are in 
 use. They are all based upon the fact that a definite amount of 
 heat will raise the temperature of a definite amount of water a 
 definite number of degrees. Those based on the metric system are 
 the large and the small calorie. A small calorie is defined as the 
 amount of heat necessary to raise 1 gram of water from 4 to 5 
 Centigrade. The large calorie is the amount of heat necessary to 
 raise 1 kilogram of water from 4 to 5 C., and is equal to 1000 
 small calories, and is denoted by the capital C. A unit of heat equal 
 to 100 small calories has also been used, arid is designated by K. 
 
396 CALORIFIC VALUE OF FUELS. 
 
 As the specific heat * of water changes but slightly with ordinary 
 changes of temperature, only a very slight error will be intro- 
 duced by using other temperatures than 4 to 5 C. when meas- 
 uring the rise in temperature. For practical purposes, there- 
 fore, a calorie may be defined as the heat necessary to raise one 
 gram of water one degree Centigrade. Another unit of heat in 
 common use is the British Thermal Unit, designated by the initial 
 letters B. T. U. This unit is the amount of heat necessary to 
 raise the temperature of 1 pound of water from 39 to 40 Fahren- 
 heit, or, for ordinary purposes, one degree F. at ordinary tem- 
 peratures. 
 
 The heat of combustion of a fuel may therefore be expressed 
 in calories or in B. T. U., the relation between these units being 
 easily calculated from the known equivalents of grams and 
 pounds and degrees Centigrade and degrees Fahrenheit (large 
 calorie = 3.96832 B. T. U.). However, when the heat of com- 
 bustion of a substance is given in calories, 1 gram or 1 kilo- 
 gram is the amount of the substance for which the heat is given, 
 while if the heat is given in B. T. U., the number given refers 
 to the amount of heat obtained from 1 pound of combustible 
 substance. As 1 kilogram = 2,20462 pounds,! to convert the 
 heat of combustion in calories to the heat of combustion in B. T. U., 
 multiply the number of calories by 1.8 (3. 96732-:- 2. 20462). 
 
 598. Calorimeters embody three essential features : (a) A com- 
 bustion chamber; (b) a tank of water with a delicate thermometer 
 for indicating the amount of heat absorbed; and (c) an insulating 
 device to prevent the external heat from reaching the water and 
 the thermometer, and also to prevent the heat generated in the 
 combustion chamber from escaping from the apparatus. 
 
 599. The Bomb Calorimeter of Berthelot is the most accurate cal- 
 orimeter which has been devised. It is adapted for the combustion 
 of any solid or liquid combustible substance. The bomb is the 
 
 * SPECIFIC HEAT OF WATER ACCORDING TO ROWLAND AND PERUET. 
 
 5 1.0054 9 1.0026 13 1.0009 17 0.9993 21 ' 0.9977 
 
 6 1.0047 10 1.0019 14 1.0005 18 0.9988 22 0.9974 
 
 7 1.0040 11 1.0014 15 1.0000 19 0.9984 23 0.9974 
 
 8 1.0033 12 1.0012 16 0.9995 20 0.9979 24 0.9972 
 
 t For these and other equivalents see Van Nostrand's Chemical Annual. 
 
THE PARR CALORIMETER. 397 
 
 characteristic part of this instrument. It consists of a brass or 
 steel receptacle of 400 to 600 cubic centimeters capacity, which 
 can be tightly closed and can sustain a pressure of at least 25 
 atmospheres. A small platinum cup is suspended from the lid 
 of the bomb. The substance to be burned is placed in this cup. 
 Two insulated wires are led through the lid and extend to within 
 a short distance of the cup. These wires are connected with a 
 short piece of fine iron or platinum w r ire which can be fused by 
 means of an electric current and thus ignite the combustible 
 material at any desired moment. Pure oxygen under pressure 
 is led into the bomb by means of a tube passing through the lid 
 and containing a close-fitting valve. The oxygen is forced in 
 until a pressure of 25 atmospheres is produced. In this atmos- 
 phere of pure oxygen under pressure, all substances may te burned. 
 The bomb is immersed in a tank of water in which a delicate 
 thermometer is suspended to register the increase in temperature, 
 the water being agitated by means of an appropriate stirring 
 devise to insure uniform distribution of the heat. The tank of 
 water is enclosed in a vessel made of some good insulating material. 
 
 THE PARR CALORIMETER. 
 
 600. Chemical Reactions Involved. This instrument differs from 
 the bomb calorimeter in that sodium peroxide or other strong 
 oxidizing substances are mixed with the fuel to produce the com- 
 bustion. This results in two decided advantages. In the first 
 place the combustion chamber may be very materially reduced in 
 size (from 300 to 600 c.c. in the bomb calorimeter to 25 c.c. in the 
 Parr calorimeter). In the second place the combustion chamber is 
 not required to stand high pressure because the oxygen is not 
 present as a gas under pressure and the products of combustion 
 (C0 2 , S0 3 , H 2 0, etc.) are not given off in the gaseous form. The 
 alkali resulting from the decomposition of the sodium peroxide com- 
 bines with the carbon dioxide, sulphur trioxide, and water, forming 
 sodium carbonate, sodium sulphate, and sodium hydroxide. Veiy 
 little gas pressure is therefore produced in the combustion chamber. 
 An additional error is introduced, however, because heat is evolved 
 by the combination of the products of combustion with the alkali. 
 
398 
 
 CALORIFIC VALUE OF FUELS. 
 
 These are mainly carbon dioxide and water. The former reacts 
 with the sodium oxide to form sodium carbonate, while the latter 
 forms sodium hydroxide. The chemical reactions which take 
 place may therefore be represented in simple form by the follow- 
 ing equations: 
 
 2NaA + C = Na 2 CO, + Na 2 ; 
 Na 2 2 +2H=2NaOH. 
 
 FIG. 62. 
 
 It has been found by experience as well as by calculation from 
 the heats of formation of the compounds given in the equations, 
 that very nearly 73 per cent of the heat produced in the reac- 
 tion is due to the oxidation of the carbon to carbon dioxide and 
 of the hydrogen to water. The factor 0.73 is, therefore, used as a 
 correction constant in the final calculation. 
 
 601. The Sodium Peroxide. Another source of error is met with 
 in the fact that sodium peroxide readily absorbs water, forming the 
 compound Na 2 2 .2H 2 0. When heated this compound decomposes 
 
THE PARR CALORIMETER. 399 
 
 into sodium hydroxide, oxygen, and water, with evolution of heat. 
 In one experiment 10 grams of sodium peroxide absorbed 0.5 gram 
 of water during an exposure of 1 hour to the air. The presence 
 of this amount of water caused an increase of 0.194 degree dur- 
 ing the combustion. 
 
 A special grade of sodium peroxide is, therefore, manufactured for 
 the Parr calorimeter. It is carefully protected from exposure to the 
 air and is sold in hermetically sealed tin cans. When desired 
 for use, the contents of the can are transferred to a glass jar pro- 
 vided with a tightly fitting lid. Glass-stoppered bottles are not 
 satisfactory. The " Lightening" or "Putnam" jar, which closes 
 with a lever clamp, has been found well adapted for this purpose. 
 The sodium peroxide must also be finely divided. It is not advis- 
 able to attempt to grind ordinary samples of the peroxide because 
 of the unavoidable exposure to the air. 
 
 602. Construction of the Instrument. Fig. 63 shows the con- 
 struction of the cartridge which is made of brass. Both top and 
 bottom are held in place by means of the cylinders or caps F and 
 E which are screwed to the body of the cartridge. The joints are 
 made air-tight by means of rubber washers. An insulated wire / 
 passes down through the stem and at / is connected to the wire 
 H by means of a piece of soft iron wire which may be fused by 
 means of an electric current and thus ignite the combustible mix- 
 ture in the bottom of the cylinder. 
 
 Fig. 64 shows the cartridge D in place in the instrument. 
 It rests on a pivot at F and can be rotated by means of the pulley 
 P. The blades attached to the center of the cartridge tend to 
 force the water downwards within the cylinder E, and produce 
 an upward current outside of this cylinder, thus keeping the tem- 
 perature of the water uniform. Both the small cylinder E and 
 the metallic vessel A are nickel-plated and highly polished. Such 
 surfaces neither radiate nor absorb heat readily. Loss or gain 
 of heat is still further prevented by the two vessels C and B of 
 indurated fiber, as well as by the air spaces between them. A 
 double close-fitting lid of the same material is also provided. 
 Two liters of water for the absorption of the heat are placed in 
 the vessel A, the temperature being measured by the thermometer 
 T. Radiation of heat cannot be wholly prevented, however, so 
 
400 
 
 CALORIFIC VALUE OF FUELS. 
 
 that if the temperature of the water is below that of the sur- 
 rounding objects, it will gain heat, while if it is above that of the 
 surrounding objects, it will lose heat. Error from this source is 
 compensated for by starting a determination with the water as 
 much below atmospheric temperature as it will rise above this 
 
 j. FIG. 64. 
 
 temperature after the combustion. The rate at which the water 
 increases or decreases in temperature may also be determined, 
 so that the amount of heat gained or lost may be calculated. 
 
 603. Water Equivalent of the Instrument. All of the heat 
 produced by the combustion of the fuel does not enter the water. 
 The combustion cartridge, the containing vessel, as well as the 
 
THE PARR CALORIMETER. 401 
 
 thermometer, absorb and retain a portion of the heat. The 
 amount of heat lost in this manner may be easily calculated. 
 These parts of the instrument are weighed and the water equiva- 
 lent calculated by multiplying the total weight by the specific 
 heat of the material used.* The cartridge and containing vessel 
 in the Parr calorimeter are made of brass, specific heat 0.092. 
 The thermometer is made of glass and mercury and only a por- 
 tion is immersed in the water and is affected by the changes in 
 temperature of the water. In order to ascertain the water 
 equivalent of the thermometer, advantage is taken of the fact 
 that the water equivalent of equal volumes of mercury and glass 
 is the same. This follows from the fact that the product of the 
 specific heat and specific gravity of those substances is the same: 
 2.5X0.19-0.47 for glass and 13.6X0.034=0.46 for mercury. 
 That is, one cubic centimeter of either mercury or glass is raised 
 one degree by .46-.4T calories of heat. The volume of that por- 
 tion of the thermometer which is immersed in the water is ascer- 
 tained by immersing the thermometer in a graduated cylinder 
 partially filled with water to the same point to which it is immersed 
 when used in the calorimeter. The volume of the portion of 
 the thermometer which is immersed is then read directly on the 
 cylinder. 
 
 604. Accelerators. The use of sodium peroxide alone has not 
 been found entirely satisfactory. The addition of small amounts 
 of substances like potassium chlorate or a persulphate is advan- 
 tageous because the oxygen in these compounds is more easily 
 liberated, and this assists in starting the reaction which proceeds 
 easily when the temperature is raised sufficiently to decompose the 
 sodium peroxide. The same object is attained by the addition of 
 substances which are more readily oxidized than the coal or other 
 substance to be burned. Tartaric acid is such a substance. 
 Ammonium compounds are also very easily oxidized on account of 
 the large amount of hydrogen present. Ammonium persulphate 
 has been found to be admirably adapted for this purpose because 
 
 * SPECIFIC HEATS. 
 
 Brass, 0.092-0.094 Nickel, 0.110 
 
 Glass, 0.190 Silver, 0.057 
 
 Mercury, 0.034 Platinum, 0.032 
 
402 CALORIFIC VALUE OF FUELS. 
 
 it contains both the ammonium radical and also the easily liberated 
 oxygen. The term accelerator is used to designate this class of 
 substances. 
 
 605. Correction Factors. A correction factor must be intro- 
 duced into the calculation when accelerators are used. r ii,u heat 
 of formation of potassium chlorate is 11.9 calories for the gram 
 molecular weight when formed from KC1 and 1J 2 , while the 
 same constant for sodium peroxide is 119.8 when formed from 
 Na 2 and 0. When ammonium persulphate is used, heat is evolved 
 by the combustion of the hydrogen. The correction for J grain of 
 potassium chlorate is .040 C., and for J gram of a mixture of 
 one part of ammonium persulphate and two parts of potassium 
 persulphate, .197 C. This persulphate mixture has been found 
 especially advantageous for use with anthracite coal. 
 
 No appreciable oxidation of nitrogen takes place in the Parr 
 calorimeter. A correction must be introduced for sulphur and 
 for ash. When using J gram of coal, these corrections are 
 - .006 C. for each per cent of sulphur, and - .001 C. for each 
 per cent of ash. A correction of .008 C. must be made for 
 each length of 10 mm. of fine iron wire. 
 
 Another correction should also be introduced when bitu- 
 minous coals are burned, on account of the oxygen or hydroxyl 
 compounds present. No such correction is necessary for anthra- 
 cite coals. The corrections may be summarized as follows: 
 
 Corrections for average anthracite coals, using J gram of 
 coal, 10 grams of Na 2 2 , and i gram of persulphate mixture: 
 
 \ gram persulphate mixture . 0.197 
 
 1 per cent of sulphur 0-006 
 
 7 " " " ash 0.007 
 
 Fuse wire.. . 0.008 
 
 Total 0.218C., 
 
 or 0.392F. 
 
 Corrections for average semi-bituminous coal (not over 25 
 per cent volatile matter), using J gram of coal, 10 grams of 
 Na 2 2 , and J gram KC10 3 : 
 
THE PARR CALORIMETER. 403 
 
 | gram KC10 3 .040 
 
 1 per cent sulphur 0-006 
 
 7 " " ash 0.007 
 
 4 " " combined water. .. 0-013 
 Fuse wire.. . 0-008 
 
 Total -074 C., 
 
 or 0.133F. 
 
 Corrections for average bituminous coal (over 25 per cent 
 volatile matter) : 
 
 \ gram KC10 3 0.040 
 
 2 per cent sulphur .012 
 
 10 " " ash 0.010 
 
 10 " " combined water... 0.033 
 Fuse wire.. . 0.008 
 
 Total . 103 C., 
 
 or 0.185F. 
 
 EXERCISE 71. 
 
 Determination of the Calorific Value of Coal by Means of the 
 Parr Calorimeter. 
 
 606. Electric Ignition. Grind the coal fine enough to pass through a 
 sieve of .25 mm. mesh or 100 to the square inch. Weigh out exactly 
 \ grain and dry for 1 hour in an air-bath heated to 104-107 C. If the 
 coal contains less than 2 per cent of water, the drying is unnecessary. By 
 adding cold or hot water, bring the temperature of two liters of water to 
 about 2 F. below room temperature. Measure out exactly two liters of 
 this water and place in the calorimeter. Transfer one full measure (10 
 grams) of sodium peroxide to the cartridge, which must be absolutely dry. 
 Transfer the coal to the cartridge, brushing off the watch-crystal carefully. 
 If the coal is anthracite, add exactly \ gram of the persulphate mixture. If 
 a bituminous coal is being tested, add exactly | gram of potassium chlorate. 
 Mix the contents of the cartridge thoroughly by stirring with a rod which is 
 afterwards well brushed off. The more recent instruments are supplied with 
 a false cap with burnished face, which is fastened temporarily in place, and 
 the contents of the cartridge mixed by shaking. After thorough mixing in 
 this manner, the provisional cap is removed and the cap with stem and 
 terminals substituted. In this way no chemical can come in contact with 
 
404 CALORIFIC VALUE OF FUELS. 
 
 the base of the terminals and produce a short circuit. The cartridge is then 
 tapped lightly to shake all the material from the upper part of the cylinder 
 and to settle the contents. When the electric method of ignition is used, the 
 fine iron wire (34 American gauge) is wrapped firmly with good contact around 
 the ends of the terminals, so as to leave a U-shaped loop extending about 
 three-quarters of an inch below the terminals and well into the chemical mix- 
 ture. If the loop is too long, it is not readily fused. The proper length of 
 wire may be ascertained by a preliminary test without screwing the top on the 
 cartridge. The current required is from 2 to 4 amperes and is readily obtained 
 by placing in parallel four to eight 16-candle-power lamps in an ordinary 
 lighting circuit of 110 volts. The current may also be obtained from 6 
 to 8 dry cells connected in series. The adjustment of the wire must be made 
 before the sodium peroxide is placed in the cartridge, as it rapidly absorbs 
 moisture. When the top has been pressed firmly in place and screwed 
 down, the vanes are adjusted and the cartridge placed in the can of water. 
 The cover is adjusted and the thermometer inserted so that the bulb is 
 about half-way down in the water. The pulley on the stern is connected 
 with the motor and the cartridge rotated to the right, or as the hands of a 
 watch turn, so as to deflect the water downwards in the can. The ther- 
 mometer is read at three-minute intervals until the readings are constant 
 or the rise in temperature is constant. These readings must be made with 
 the greatest care and tenths of the smallest divisions estimated as care- 
 fully as possible. 
 
 After igniting the charge by bringing the spring in contact with the 
 revolving stem of the cartridge, the temperature rises rapidly at first, 
 and then more slowly, until the maximum temperature is reached. The 
 readings are made at three-minute intervals as before, until the tempera- 
 ture is constant or it begins to fall. The cartridge is removed and the 
 parts unscrewed and placed in a dish of hot water to dissolve the melt. 
 It is then rinsed thoroughly with distilled water and dried, when it is 
 ready for another determination. 
 
 607. Ignition by Hot Wire. If the ignition is made by means of a hot 
 wire, the method of filling the cartridge and rotating it until the tempera- 
 ture is constant is the same as when the electric ignition is used. The short 
 piece of soft iron wire is placed on a wire gauze and heated to redness in a 
 Bunsen-burner flame and dropped quickly into the opening at the upper end 
 of the valve. With the pincers the valve is now pressed completely down 
 and released with a quick motion, so as to prevent the escape of heated air 
 from the interior. During this operation the cartridge is kept revolving. 
 The remainder of the operation is the same as when the electric method of 
 ignition is used. 
 
 608. Calculation. The various correction factors, as given on page 402, 
 are subtracted from the total rise in temperature. The remainder is multi- 
 plied by 3117. This factor is obtained from the following formula: 
 
CALCULATION. 405 
 
 Total water eqv.X 0.73 X Corrected rise Calorific va i ue . 
 Weight of fuel taken 
 
 Substituting the known values, 
 
 2135 X 0.73 Xr 
 ~05" 
 
 which reduces to 3117Xr. The product will be the calorific power of the 
 coal in B. T. U. per pound of coal. 
 
CHAPTER XXIX. 
 WATER ANALYSIS. 
 
 A CHEMICAL analysis of water is carried on by very different 
 methods, depending on the use to which the water is to be put. 
 These methods may conveniently be divided into two classes. In 
 the first class are placed those methods by which the POTABILITY 
 of a water is determined, while in the second class are included 
 the methods by which the suitability of a water for INDUSTRIAL 
 PURPOSES is determined. As the requirements of the various 
 industries are somewhat diverse, the methods used in testing a 
 water for industrial purposes cannot be so definitely stated as those 
 employed when the water is to be used for domestic consumption. 
 
 In the latter case no poisonous substances may be present even 
 in minute traces. The absence of lead, copper, etc., being insured, 
 we may conveniently classify the methods commonly used into 
 those by which the amount of ORGANIC impurities and those by 
 which the amount of INORGANIC impurities are determined. As 
 the commonly occurring inorganic salts, such as sulphates, chlorides, 
 and carbonates of lime, magnesia, the alkalies, iron, etc., are not 
 objectionable unless present in excessive amounts, only the total 
 amount of the calcium and magnesium salts present is determined 
 and reported as hardness of the water. If the water is classed as a 
 MINERAL WATER, and is therefore to be used because of the inor- 
 ganic salts present, a complete analysis of these must be made, as 
 well as determinations of the organic impurities. The most 
 important determinations for indicating the presence of organic 
 impurities are the estimation of the amount of nitrogen in its 
 various forms and chlorine. 
 
 SANITARY ANALYSIS. 
 
 609. Bacteriological Examination. As has been said, the sani- 
 tary analysis consists in the determination of the organic and 
 the inorganic impurities present. The most serious ill effect of 
 
 406 
 
SANITARY ANALYSIS. 407 
 
 using impure water undoubtedly arises from taking into the 
 system DISEASE GERMS which exist in the water. It would seem, 
 therefore, that the examination of the water should primarily be 
 entrusted to the bacteriologist. Unfortunately, it is extremely 
 difficult, if not impossible, in the present condition of bacterio- 
 logical science, to detect and identify the specific disease germs 
 even if they undoubtedly exist in a given water. This is largely 
 due to the fact that the great majority of the germs that exist in 
 water are either harmless or even beneficial. The continual drink- 
 ing of a water containing relatively few disease germs by a person 
 in whose system the germ can live, ultimately results in the ac- 
 cumulation of the germs in sufficient quantity to produce 
 disease. 
 
 The routine bacteriological tests generally made on water 
 consist of a count of the total number of bacteria present and 
 a test for the presence or absence of B. coli. These bacteria are 
 almost invariably present in the intestines of human beings and 
 other warm-blooded animals. Their presence in water is regarded 
 as evidence of contamination with sewage, which almost always 
 contains disease germs excreted from the body, thus rendering 
 the water unwholesome and dangerous. The presence of an ex- 
 cessivr number of bacteria in water also renders it deleterious to 
 health. 
 
 610. Source of Impurities. The chemical examination gives 
 indirect evidence of the presence of disease germs by revealing 
 the presence and the amount of organic matter in the water, as 
 well as whether it is of animal or vegetable origin, and the extent 
 to which it has undergone decomposition. As SEWAGE is undoubt- 
 edly the most common source of contamination of water with 
 pathogenic germs, the evidence furnished by the chemist of its 
 presence is quite sufficient to condemn a water for household use, 
 even if the specific disease germ is not found. If the water is liable 
 to be contaminated with sewage, it is almost certain that sooner 
 or later it will be rendered poisonous by the presence of the germs 
 of disease. It is evident that after organic impurities have been 
 found in a water it is important to know their source before a deci- 
 sion can be reached as to the potability of a water. Whenever 
 possible, therefore, the chemical examination should be supplemented 
 
408 WATER ANALYSIS. 
 
 by an examination of the source of the water-supply; that is, whether 
 it is a spring, lake, river, well, etc. If surface or underground 
 drainage can mingle with the water, the entire region drained 
 should be examined, if possible, for sources of contamination. 
 
 611. The Samples must be taken with the greatest care. As the 
 impurities at best are present in very small amount, careless wash- 
 ing of the bottle, or collecting water contaminated after leaving 
 the source of supply, may easily introduce as much impurity as 
 was originally present. The bottles used should be thoroughly 
 cleaned by means of strong dichromate solution and then thor- 
 oughly rinsed out with distilled water and drained. They should 
 then be rinsed out with the water to be examined. 
 
 If water from a CITY SUPPLY is to be examined, it must be 
 allowed to flow from the pipes a considerable time before filling 
 the bottle, unless the water is to be examined for lead or other 
 contamination from the pipes. 
 
 If a WELL-WATER is to be examined, the stagnant water in the 
 pump must first be completely removed. 
 
 A sample of SPRING, RIVER, or LAKE WATER is best taken by allow- 
 ing the bottle to fill after immersion some distance under the sur- 
 face of the water, so as to avoid the floating surface contamination. 
 
 The TIME which may elapse between the collection of a sample 
 and the beginning of its analysis varies with the character of the 
 sample and other conditions. The following limits are generally 
 safe : for fairly pure surface-water, 24 to 48 hours ; and for normal 
 ground-water, 48 to 72 hours. Polluted water requires analysis 
 within 12 hours. A bacteriological examination should in all cases 
 be made within 6 hours. If for any reason the examination is 
 delayed, the sample must be preserved on ice. 
 
 PHYSICAL EXAMINATION. 
 
 The physical examination includes observations of the tem- 
 perature, general appearance, color, turbidity, and the odor in hot 
 and cold samples. 
 
 612. The Temperature should be taken at the time of collec- 
 tion to the nearest 0.5 Centigrade. 
 
 613. Turbidity. The general appearance of the water should 
 be determined by inspection in strong light after standing several 
 
PHYSICAL EXAMINATION. 409 
 
 hours. Substances remaining in suspension are then classified 
 as turbidity on standing, and substances settling to the bottom as 
 sediment. Instead of expressing the turbidity as none, slight, dis- 
 tinct, decided, etc., it is advisable to use the NUMERICAL STANDARDS 
 recently introduced by which the turbidity is expressed in parts 
 of silica per one million. The stock suspension of silica is prepared 
 as follows : Diatomaceous earth, as free from amorphous silica and 
 sponge spicules as may be obtained, is washed with water to free 
 it from any soluble salts, and ignited to remove any organic matter. 
 It is then treated with warm dilute hydrochloric acid, and washed 
 with successive portions of distilled water until free from acid. 
 The material, now composed of practically pure diatomaceous 
 frustules, is ground to an impalpable powder in an agate mortar, 
 sifted through a sieve that has 200 meshes to the inch, and dried 
 in a desiccator. One gram of this prepared diatomaceous earth 
 in 1 liter of distilled water gives the stock suspension which has 
 a turbidity of 1000, i.e., contains 1000 parts of silica por million. 
 
 STANDARDS for comparison should be prepared from this 
 stock suspension by dilution with distilled water. For turbidity 
 readings below 20, standards of 0, 5, 10, 15, and 20 are kept in 
 gallon bottles made of clear white glass; for readings above 20, 
 standards of 20, 30, 40, 50, 60, 70, 80, 90, and 100 are kept in 100- 
 c.c. Nessler jars, approximately 20 mm. in diameter. 
 
 COMPARISON OF THE WATER under examination with the stand- 
 ards is made by viewing them sidewise towards the light, looking 
 at some object and noting the distinctness with which the margins 
 of the object can be seen. The standards are kept stoppered, 
 and both sample and standards are thoroughly shaken before 
 making the comparison. In order to prevent any bacterial or 
 alga3 growths from appearing in the standards, a small amount of 
 bichloride of mercury may be added to them. 
 
 614. The Color of the water is measured by comparison with 
 the color of a solution of platinum and cobalt and is expressed in 
 parts of platinum per million. A standard solution which has a 
 color of 500 is made by dissolving 1.246 grams potassium platinic 
 chloride (PtCl 4 .2KCl), containing 0.5 gram platinum, and 1 gram 
 of crystallized cobaltous chloride (CoCl 2 .6H 2 0), containing 0.25 
 gram of cobalt in water, with 100 c.c. concentrated hydrochloric 
 
410 WATER ANALYSIS. 
 
 acid, and making up to 1 liter with distilled water. By diluting 
 this solution standards are prepared having colors of 0, 5, 10, 15, 
 20, 25, 30, 35, 40, 50, 60, and 70 parts per million. These are kept 
 in 100-c.c. Nessler jars of such dkmeter that the liquid has a depth 
 between 20 and 25 cm. and is protected from dust. 
 
 The color of a sample is observed by filling a similar tube with 
 the water and comparing it with the standards. The observation 
 is made by looking vertically downwards through the tubes upon 
 a white surface placed at such, an angle that light is reflected 
 upwards through the column of liquid. The reading is recorded 
 to the nearest unit. Waters that have a color darker than 70 are 
 diluted before making the comparison, in order that no difficulties 
 may be encountered in matching the hues. Water containing 
 matter in suspension is filtered until no visible turbidity remains. 
 
 615. The Odor of the water both hot and cold is observed. To 
 obtain the odor of the cold water the sample in the large collecting- 
 bottle, which should be about one-half or two-thirds full, is shaken 
 vigorously, the stopper removed, and the nose immediately placed 
 to the mouth of the bottle. To obtain the odor of the hot water 
 it is poured into a moderately large beaker until it is about one- 
 third full. The beaker is covered with a watch-crystal and the 
 water is rapidly brought to a boil. It is immediately removed 
 from the source of heat and allowed to cool for about five minutes. 
 The beaker is then shaken with a rotary motion, the watch-crystal 
 slipped to one side, and the odor observed. The odor is expressed 
 by such terms as the following: vegetable, aromatic, grassy, fishy, 
 earthy, mouldy, musty, disagreeable, peaty, and sweetish. The 
 intensity of the odor is usually indicated by the numbers 1, 2, 3, 4, 
 and 5, 1 indicating very faint; 2, faint; 3, distinct; 4, decided; 
 and 5, strong. 
 
 CHEMICAL EXAMINATION. 
 
 The first determination carried out should be that of the 
 amount of nitrogenous organic matter present, as this material is 
 continually undergoing changes, and the condition in which the 
 nitrogen exists is taken as an index of the source of the contamina- 
 tion, and the time which has elapsed since it has been introduced. 
 
NITROGEN. 411 
 
 616. The Nitrogen may exist as so-called ALBUMINOID NITRO- 
 GEN, AMMONIA, NITRITES, or NITRATES. The albuminoid nitrogen 
 is undoubtedly present as part of a class of complex organic com- 
 pounds, called proteins, which may have been derived from plant, 
 but more likely animal, tissues which are present in the water. 
 These tissues undergo a more or less rapid decomposition or 
 putrefaction by which the nitrogen assumes for the most part the 
 condition of FREE AMMONIA, which may then be oxidized to nitrites 
 or nitrates. The presence of the unstable nitrites generally indi- 
 cates the presence of sewage or other organic matter in a state of 
 decomposition. While nitrates may be produced by the oxidation 
 of organic nitrogen, this is by no means the only source, since the 
 atmosphere and the soil contain considerable amounts of nitric 
 acid or nitrates, which are dissolved by the ram. The amount of 
 nitrates in a surface-water increases with the density of popula- 
 tion in the region drained. 
 
 617. Determination of Free and Albuminoid Ammonia. 
 Ammonia is determined by means of the so-called NESSLER SOLU- 
 TION, to which a brown color is given by very small amounts of 
 ammonia or ammonium salts. The reagent consists of a strongly 
 alkaline solution of the double iodide of potassium and mercury, 
 HgI 2 .2KI. Ammonia produces in this solution a dark-brown col- 
 oration or precipitate of dimercuric ammonium iodide, NHg 2 I.H 2 0, 
 according to the following equation: 
 
 2(HgI 2 .2KI) + NH 3 + 3KOH = NHg 2 I.H 2 + 7KI + 2H 2 0. 
 
 The depth of the color produced is proportional to the amount 
 of ammonia present, so that by comparison with solutions con- 
 taining known amounts of ammonia, the amount of ammonia 
 present in a given solution may be determined, the solution con- 
 taining the known amount of ammonia being so adjusted that 
 its color exactly matches the solution containing the unknown 
 amount. 
 
 618. Preparation of Nessler Solution.* Dissolve 61.75 grams 
 of potassium iodide in 250 c.c. of ammonia-free water and add a 
 cold solution of mercuric chloride which has been saturated by 
 
 * D. D. Jackson, Tech. Quar., XIII, No. 4, Dec. 1900, p. 320. 
 
412 WATER ANALYSIS. 
 
 boiling with excess of the salt. Pour in the mercury solution 
 cautiously and add an amount just sufficient to make the color a 
 permanent bright red, which will require about 400 c.c. Dissolve 
 the red precipitate by adding exactly .75 gram potassium iodide. 
 Then add 150 grams of potassium hydrate dissolved in 250 c.c. 
 of water and dilute the solution to 1 liter. Mix thoroughly and 
 allow the solution to stand until any precipitate formed has settled, 
 leaving a pale straw-colored solution which is siphoned or decanted 
 off into another bottle for use. The solution improves with age 1 . 
 Its sensitiveness is increased by adding mercuric chloride and 
 decreased by adding potassium iodide. The 2-c.c. portions used 
 for each test must be measured quite carefully as the depth of 
 color produced with a given amount of ammonia depends to a 
 certain extent upon the amount of Nessler solution used. 
 The following solutions will also be required : 
 
 619. Alkaline Potassium Permanganate Solution. Dissolve 
 200 grams of solid caustic potash* and 8 grams of potassium 
 permanganate in 1250 c.c. of distilled water. Boil the solution 
 down to 1 liter and preserve for use. 
 
 620. Sodium Carbonate Solution. Dissolve 50 grams of the 
 pure salt in 300 c.c. of ammonia-free water. 
 
 621. Standard Ammonia Solution. Dissolve 1.5701 grams of 
 pure dry ammonium chloride in ammonia-free water and dilute to 
 500 c.c. 1 c.c. of this solution will contain 1 mg. of ammonia. 
 10 c.c. should be diluted to 1 liter with ammonia-free water. 1 c.c. 
 of this solution will contain .01 mg. of ammonia and is the stand- 
 ard solution used. 
 
 622. Ammonia-free Water. Ordinary distilled water gener- 
 ally contains considerable amounts of ammonia. It may be 
 tested by adding 2 c.c. of Nessler solution to 50 c.c. of the water 
 placed in a NESSLER TUBE. These tubes are made of colorless glass 
 and contain 50-c.c. when filled, to a mark near the top. A tube of 
 suitable proportions has a diameter of 1.7 cm. and a height of 
 21 cm. to the 50-c.c. mark. If after standing five minutes no trace 
 of a yellow tint is developed in the water it may be used without 
 purification. Generally, however, it must be freed from ammonia. 
 
 * If caustic potash free ir^m carbonates is used, the bumping so apt to occur 
 while boiling an alkaline solution will be very much reduced. 
 
NITROGEN. 
 
 413 
 
 This may be done by nearly filling a large glass retort or a suitable 
 copper or tin vessel with distilled water to which some of the alka- 
 line potassium permanganate solution has been added. A glass 
 or block-tin condenser is attached and the water is distilled until 
 50-c.c. portions tested with the Nessler solution are found free 
 from ammonia. The distillate is then collected in a clean bottle 
 for use. 
 
 The distilled water may also be freed from ammonia by adding 
 enough bromine water to color it distinctly and then boiling until 
 the excess of bromine is expelled. The boiling may be omitted 
 if a drop of caustic soda is added to the water tinted with bromine, 
 and after standing ten minutes the undecomposed hypobromite 
 removed by adding a little potassium iodide. 
 
 The preparation of the ammonia-free water and all ammonia 
 determinations must be carried out in a room whose atmosphere 
 is free from ammonia and ammonium salts. The ordinary labora- 
 tory is, therefore, not suited to the purpose. 
 
 DETERMINATION OF FREE AMMONIA. 
 
 623. Distillation. A liter fiask is connected by means of a 
 short piece of rubber tubing or a rubber stopper with a Liebig 
 condenser. A tubulated glass retort capable of holding a liter of 
 solution is also very convenient for this purpose, as the reagent 
 can be introduced through the tubulure. The stem of the retort 
 may be introduced into the end of the condenser-tube and a tight 
 joint made by means of a short piece of rubber tubing. The dis- 
 tillate is led into the Nessler tubes by means of an adapter. 
 
 Although the apparatus is thoroughly cleaned before being set 
 up, to completely remove ammonia about 200 c.c. of ammonia- 
 free water is introduced into the retort and 10 c.c. cf the sodium 
 carbonate solution added. The solution is boiled and the first 
 50 c.c. of the distillate discarded while the second 50-c.c. portion 
 is tested for ammonia by adding 2 c.c. of the Nessler solution. If 
 no color develops after standing five minutes the apparatus may 
 be considered free from ammonia and the water to be analyzed 
 may be introduced into the retort. Unless the water contains 
 an unusual amount of ammonia 500 c.c. will be found to be a 
 
414 
 
 WATER ANALYSIS. 
 
 convenient amount. A preliminary test may be made by adding 
 2 c.c. of Nessler solution to 50 c.c. of the water to be tested. The 
 color developed is compared with the color produced by adding 
 0.5 c.c. of the standard ammonia solution to 50 c.c. of ammonia- 
 free water. If the color produced in the tube containing the water 
 to be analyzed is the darker, less than 500 c.c. of the water should 
 be placed in the retort and the difference made up by ammonia- 
 free water. 
 
 624. Nesslerizing. The distillate is collected in the Nesslerizing 
 tubes, which are placed in a rack in order after being filled to the 
 50-c.c. mark. The flame of the Bunsen burner should be so 
 adjusted that 50 c.c. are distilled in about eight minutes. As soon 
 
 FIG. 65. 
 
 as the second tube is filled and has come to room temperature it 
 is Nesslerized. 2 c.c. of the Nessler solution is added. Several 
 clean Nessler tubes are taken and carefully measured amounts of 
 the standard ammonia solution added from a burette, such as 
 1 c.c., 2 c.c., 3 c.c., etc. The tubes are then filled with ammonia- 
 free water to the 50-c.c. mark, and 2 c.c. of the Nessler solution 
 added to each tube. After a little practice the solutions may 
 readily be mixed by shaking with the hand. Long clean glass rods 
 may also be used for stirring the contents of the tubes. 
 
 The colors are compared by holding two tubes in the hand 
 and looking through the length of the tubes, which are held over a 
 white paper or piece of porcelain in a good light. If the color of a 
 
FREE AMMONIA. 415 
 
 tube does not match that of any of the standards made, others must 
 be made with the amount of standard ammonia solution judged 
 necessary until the color of the unknown ammonia-tubes have 
 been matched. After adding the Nessler solution, five minutes 
 must elapse before comparing the color, as it does not develop 
 instantly. After being fully formed the color is quite permanent, 
 so that a set of standard tubes may be used during a day's work 
 without error. A permanent set of standard tubes has been 
 devised by D. D. Jackson.* The temperature of the water may 
 vary between 18 and 25 without material change in depth of color. 
 
 If the ammonia in the second tube corresponds to 1 c.c. or less 
 of the standard solution, further distillation may be dispensed with. 
 Otherwise 50-c.c. portions must be collected until this limit is 
 reached. Usually three 50-c.c. portions are collected. If the 
 ammonia in the second tube corresponds to 2 c.c. or less of the 
 standard solution, the first tube may be Nesslerized in the same 
 manner as the second. If the second tube contains a greater 
 amount of ammonia, a measured portion of the first tube must be 
 taken and diluted with ammonia-free water and Nesslerized. Or 
 the contents of the first and third tubes may be mixed and 50 c.c. 
 of the mixture Nesslerized. In no case should the color of a tube 
 be read which responds to more than 10 c.c. of the standard ammo- 
 nia solution. 
 
 625. Calculation. The various amounts of ammonia found in 
 the different tubes are added and the result stated in parts per 
 million. If, for instance, the first tube was found to match a tube 
 containing 9 c.c. of standard ammonia solution, the second 3 c.c., 
 the third 1 c.c., and the fourth free from ammonia, then, as 1 c.c. 
 is equal to .01 mg. of ammonia, the total amount of ammonia 
 would be .09+. 03+. 01 = .13 mg. As 500 c.c. of the water was 
 taken and as 1 mg. is the millionth part by weight of a liter 
 of water, the amount of free ammonia present in the water would 
 be .26 parts per million. 
 
 *Tech. Quar., Vol. XIII, No. 4, Dec. 1900, p. 320. 
 
416 WATER ANALYSIS. 
 
 DETERMINATION OF ALBUMINOID AMMONIA. 
 
 626. Distillation. When all the free ammonia has been dis- 
 tilled off, which is usually the case when 150 c.c. of distillate has 
 been collected, 50 c.c. of the alkaline permanganate solution is 
 added and the distillation continued, 50-c.c. portions being col- 
 lected in the Nessler tubes and the ammonia determined and cal- 
 culated exactly as given for free ammonia. 
 
 The albuminoid ammonia does not always come off as readily 
 as the free ammonia. This is undoubtedly due to the greater or 
 less difficulty of decomposing organic matter from different sources. 
 Organic matter derived from vegetables is generally decomposed 
 with more difficulty than the contamination derived from partially 
 decomposed animal refuse as found in sewage. Each 50-c.c. por- 
 tion of the distillate for albuminoid ammonia must therefore be 
 tested to be sure that all of the ammonia has been obtained, the 
 solution in the retort being boiled nearly dry. As it is almost 
 impossible to obtain all of the albuminoid ammonia from many 
 waters, especially highly colored surface-waters, the practice is 
 growing of collecting five 50-c.c. portions of the distillate for the 
 albuminoid ammonia. In this way comparative results are ob- 
 tained which are of almost as great value as if the total amount 
 of albuminoid ammonia were obtained in each case. Much time 
 and labor are also saved. About one-half of the total organic 
 nitrogen will generally be obtained in the 250 c.c. of distillate. 
 
 DETERMINATION OF NITRITES. 
 
 A colorimetric process similar to that used to determine 
 ammonia is in use for the estimation of the amount of nitrites 
 present. Of the several methods which have been suggested, 
 that of GRIESS will be given. It depends on the fact that a red 
 coloration is produced when a solution of SULPHANILIC ACID and 
 NAPHTHYLAMINE HYDROCHLORiDE are added to an acidified solu- 
 tion of a nitrite. This test is said to be capable of showing one 
 part of nitrogen as nitrite in one thousand million parts of water. 
 
NITRITES. 417 
 
 The following reagents must be prepared: 
 
 627. Sulphanilic Acid. Dissolve 0.8 gram of the acid in 100 c.c. 
 of hot water. 
 
 628. Naphthylamine Hydrochloride. Dissolve 0.8 gram of the 
 salt in 100 c.c. of water and 1 c.c. of strong hydrochloric acid. 
 Decolorize the solution by adding a little powdered charcoal. 
 Place the solution in a glass-stoppered bottle with the charcoal 
 and filter as required for use. 
 
 629. Standard Solution of Sodium Nitrite. Pure silver nitrite 
 is prepared by mixing a warm concentrated solution of 8 parts of 
 sodium nitrite with a warm concentrated solution of 16 parts of 
 silver nitrate. The precipitate, when cold, is washed with cold 
 water and quickly dried over a water-bath with as little exposure 
 to the light as possible. Weigh .1091 gram of the dry nitrite and 
 dissolve in warm water. Add a slight excess of pure sodium 
 chloride and dilute to 1 liter. After allowing the silver chloride 
 to settle, draw out 10 c.c. and dilute to 1 liter. This will give a 
 standard solution, 1 c.c. of which will be equal to .001 mg. of 
 nitrogen as nitrite 
 
 630. The Determination is carried out as follows: 100 c.c. of 
 the water to be examined are placed in a cylinder capable of hold- 
 ing 100 c.c., which is made of the same kind of colorless glass as 
 the Nessler tube, although a shorter and wider tube is found con- 
 venient, a suitable size being 3 cm. in diameter and 13.2 cm. from 
 the bottom to the 100-c.c. mark. One drop of concentrated hydro- 
 chloric acid is added and then 2 c.c. of the sulphanilic acid solution 
 and 2 c.c. of the solution of hydrochloride of naphthylamine. 
 After mixing well, the tube is tightly stoppered and allowed to 
 stand for ten minutes until the color is fully developed. In the 
 meantime tubes are prepared containing known amounts of 
 nitrite, 1 c.c., 2 c.c., 3 c.c., etc., of the standard nitrite solution 
 being added to tubes containing 100 c.c. of ordinary distilled 
 water, or, still better, water redistilled as directed for ammonia- 
 free water. After the addition of 2 c.c. of each reagent to each of 
 the color solutions and a drop of hydrochloric acid, and allowing 
 the tubes to stand for twenty minutes, the colors of the known and 
 the unknown solutions are compared and the amount of nitrogen 
 present per million parts is computed. If 100 c.c. of the water 
 
418 WATER ANALYSIS. 
 
 is taken and the color is matched by the tube containing 3 c.c. of 
 the standard nitrite solution, the water contains .03 part per 
 million of nitrogen as nitrite. As this test is extremely delicate, 
 it is advisable to make a blank determination to detect accidental 
 contamination which at times arises from nitrous fumes in the air. 
 
 DETERMINATION OF NITRATES. 
 
 This determination also is carried out by a colorimetric 
 process, the yellow AMMONIUM SALT OF NITROPHENOL-SULPHONJC 
 ACID * being produced when the water to be examined is evap- 
 orated to dryness and treated with PHENOL-SULPHONIC ACID and 
 ammonia. The following reactions take place: 
 
 C 6 H 4 .OH.S0 3 H+ HN0 3 = C 6 H 3 .OH.S0 3 H.N0 2 + H 2 ; 
 C 6 H,.OH.SO,H.N0 8 + NH 4 OH = C 6 H 3 .ONH 4 .S0 3 H.N0 2 + H 2 0. 
 
 The intensity of the color developed is compared with the color 
 produced by the same process in water containing a known amount 
 of nitrate. 
 
 The following solutions are required: 
 
 631. Phenol-sulphonic Acid. In 148 c.c. of pure concentrated 
 sulphuric acid 24 grams of pure phenol are dissolved by heating 
 on the water-bath and 12 c.c. of distilled water added. 
 
 632. Standard Potassium Nitrate Solution. 0.722 gram of pure 
 dry potassium nitrate is weighed out and dissolved in 1 liter of 
 distilled water. 10 c.c. of this solution are evaporated to dryness, 
 2 c.c. of the phenol-sulphonic acid added, and quickly and thor- 
 oughly mixed with the residue by means of a glass rod. The 
 mixture is dissolved in distilled water and the solution diluted to 
 1 liter. 
 
 The color-tubes for comparison are prepared by measuring out 
 portions of this solution into 100-c.c. tubes of the size used for 
 the nitrite determination, adding 5 c.c. of strong ammonia and 
 diluting to 100 c.c. These standards are permanent for several 
 months if the tubes are well corked with stoppers from which the 
 coloring matter has been extracted by boiling water. 1 c.c. of the 
 dilute nitrate solution is equal to .001 mg. of nitrogen as nitrate. 
 
 * Jour. Am. Chem. Soc., XXVI, 388. 
 
NITRATES. 419 
 
 633. The Determination is carried out as follows: 25 c.c. of the 
 water to be analyzed are taken, or, if the amount of nitrate present 
 is small, 50 or even 100 c.c. are taken and evaporated to dryness. 
 The residue is treated with 1 c.c. of the phenol-sulphonic acid and 
 well mixed by stirring with a glass rod. 10 c.c. of distilled water 
 are added and then 5 c.c. of strong ammonia. The solution is 
 washed into a 100-c.c. tube and diluted to the mark with distilled 
 water. The depth of color is now compared with that of the stand- 
 ard color-tubes containing various amounts of the standard nitrate 
 solution. 
 
 When the color has been exactly matched, the parts per mil- 
 lion of nitrogen as nitrate are calculated. If, for instance, 50 c.c. 
 of the water were taken and the color was matched by the standard 
 color-tube containing 3 c.c. of the dilute nitrate solution, then 50 
 c.c. of the water would contain .003 mg. of nitrogen as nitrate, 
 which would be .06 part per million. If 25 c.c. of a water had been 
 taken, the result would be .12 part per million if the color produced 
 were matched by a color-tube containing 3 c.c. of the standard 
 nitrate solution. 
 
 DETERMINATION OF THE OXYGEN REQUIRED TO 
 OXIDIZE ORGANIC MATTER. 
 
 The methods already given have shown the amount of nitro- 
 gen present in a given water in the four forms: ammonia, 
 nitrous and nitric acid, and albuminoid substances. A very 
 simple method is in use for the determination of the amount of 
 carbon present. It consists in ascertaining the amount of per- 
 manganate required to oxidize the organic matter present. The 
 water is boiled for ten minutes with sulphuric acid and an excess 
 of standard potassium permanganate solution. The excess of the 
 permanganate is then titrated back with standard oxalic acid. 
 
 634. Errors. It is evident that by this process any substance 
 other than carbonaceous material capable of being oxidized under 
 the conditions will also reduce the permanganate. On the other 
 hand, the oxidation of the organic matter present may not be 
 complete. It has indeed been shown that the extent to which 
 different kinds of organic matter are oxidized varies greatly. As 
 the method, therefore, is an empirical one, the conditions of the 
 oxidation must be rigidly adhered to. If this is done, very con- 
 
420 WATER ANALYSIS. 
 
 stant results are obtained and the method has shown itself of 
 value in forming a judgment of the amount of organic matter in 
 the water, as other oxidizable substances are rarely present. 
 Ferrous iron and sodium chloride are the only common excep- 
 tions. The iron is quickly precipitated as ferric oxide after a few 
 hours' exposure of the water to the air. 
 
 If a considerable amount of SODIUM CHLORIDE is present, a cor- 
 rection must be applied as found by a blank test. Sodium chloride 
 is added to distilled water until there is present the amount of 
 chloride which has been found in the water being examined. The 
 oxidation with the permanganate is then carried out in the same 
 manner as when the water was tested. The amount of perman- 
 ganate required is subtracted from that used in testing the water. 
 
 635. The Following Solutions are required and are made with 
 water free from organic matter. This is prepared by adding a 
 little potassium permanganate and caustic alkali to distilled water 
 and redistilling it. The first portions of the distillate are discarded. 
 The steam or distilled water should not come in contact with rubber 
 or other organic matter. A suitable distilling apparatus may be 
 made by connecting a retort with a Liebig condenser by inserting 
 the tube of the retort into the inner tube of the condenser as 
 shown in Fig. 65, page 414, and packing the joint with ignited 
 asbestos instead of the rubber connection. 
 
 636. Potassium Permanganate Solution is made by dissolving 
 .3953 gram of the pure salt in 1 liter of the redistilled water. 1 c.c. 
 of this solution will contain 0.1 mg. of available oxygen. 
 
 637. Oxalic Acid Solution is made by dissolving .7875 gram 
 of the pure recrystallized and carefully dried salt in 1 liter of water. 
 If this salt as well as the permanganate are pure, these solutions 
 will be exactly equal. They should be titrated against each other 
 by measuring out 10 c.c. of the oxalic acid, adding 10 c.c. of dilute 
 sulphuric acid and 200 c.c. of the redistilled water. The whole 
 is heated to boiling and the potassium permanganate solution 
 introduced until a faint, permanent pink coloration is produced. 
 Neither the oxalic acid nor the permanganate solutions are perma- 
 nent and must be made up fresh or standardized from time to time. 
 
 Dilute sulphuric acid is made by diluting 1 part of concentrated 
 acid with 3 parts of the redistilled water. 
 
ORGANIC MATTER. 421 
 
 638. The Determination is carried out by placing 100 c.c. of 
 the water to be tested in a flask and adding 10 c.c. of the dilute 
 sulphuric acid. The standard potassium permanganate solution 
 is introduced from a burette until the water is colored. It is 
 heated to boiling and more permanganate is introduced if the 
 color fades. If brown flakes of manganese dioxide are formed, 
 more dilute sulphuric acid must be added. The boiling is continued 
 for five minutes. An excess of oxalic acid is then introduced and 
 titrated back with the permanganate solution. The volume of 
 oxalic acid introduced is subtracted from the total amount of the 
 permanganate solution used, and the number of milligrams of oxy- 
 gen consumed is calculated. 
 
 The result is stated in parts of oxygen required per million 
 parts of water. 
 
 If the amount of organic matter in the water is considerable, 
 less than 100 c.c. must be taken for this determination. The 
 volume should in each case be made up to 100 c.c. with the redis- 
 tilled water. As the success of this determination is assured only 
 by working under absolutely fixed conditions, the determination 
 should be repeated several times until constant results are obtained. 
 
 DETERMINATION OF CHLORIDES. 
 
 The chlorine in natural waters exists for the most part in 
 the form of sodium chloride, and as such is harmless unless present 
 in very large amounts. As considerable amounts of salt are used 
 very generally in articles of food, it is present in large amounts in 
 urine, and therefore in sewage. For this reason the amount of 
 chlorides present is used as an index of the amount of contamina- 
 tion of the water. As almost all natural waters, including rain- 
 water, contain chlorides in greater or less amount, only that amount 
 of chlorides which is above the average of similar but pure waters 
 can be used as an argument for condemning a water as contami- 
 nated. 
 
 639. Silver Nitrate Solution. The chlorine is determined volu- 
 metrically, using a standard silver nitrate solution with potassium 
 chroma te as indicator, as described in Chapter XXVI. The silver 
 
422 WATER ANALYSIS. 
 
 nitrate solution is made of such a strength that 1 c.c. is equal to 
 1 mg. of chlorine. For this purpose 4.7938 grams of pure silver 
 nitrate are weighed out and dissolved in a liter of water. The 
 potassium chromate indicator is made by dissolving 5 grams of the 
 pure salt in 100 c.c. of distilled water. 
 
 640. The Determination is carried out by placing 100 c.c. of the 
 water to be examined in a porcelain dish and adding 1 c.c. of the 
 indicator. The silver nitrate solution is added until a faint per- 
 manent red color is produced. This red color should be removed 
 by adding a pinch of sodium chloride or a drop or two of a chloride 
 solution and the titration repeated, using the first solution for com- 
 parison. This solution will have a pure-yellow color with a white 
 silver chloride precipitate equal to that obtained in the second 
 titration. The end-point may be made still sharper by preparing 
 a solution of distilled water containing 1 c.c. of the indicator, 
 through which the solution being titrated may be observed. 
 
 As an appreciable amount of the silver nitrate solution may be 
 required to produce the color indicating the end-point, a blank 
 determination should be made by adding 1 c.c. of the^ indicator to 
 100 c.c. of distilled water and then adding silver nitrate solution 
 until the red color is observed. The volume of silver nitrate added 
 should be subtracted from the amount added to the water being 
 tested. 
 
 When the amount of chlorides present is very small, or great 
 accuracy is required, an amount of water greater than 100 c.c. 
 should be taken and evaporated down to a bulk of about 100 c.c. 
 after the addition of 0.1 c.c. of a saturated solution of sodium 
 carbonate. This solution is then titrated as already given. If 
 the water contains considerable coloring-matter, so that after 
 concentration the color of the indicator cannot be observed, the 
 coloring-matter may be removed before concentration by shaking 
 the water with some freshly precipitated aluminum hydrate and 
 filtering before measuring out the portion for evaporation. If 
 the water is acid, it must be neutralized with sodium carbonate. 
 
 641. The Determination of Hardness in water, both temporary 
 and permanent, has already been given in Chapter XXI, page 
 269. 
 
 642. Determination of Total Solids. This determination is 
 conducted by evaporating a measured volume of the water, 100 
 
TOTAL SOLIDS. 423 
 
 to 500 c.c. being usually taken, to dryness in a weighed platinum 
 dish. Various temperatures from 105 to 180 have been used 
 for drying the residue before weighing, 103 having now been 
 fixed as the standard temperature. After weighing, the organic 
 matter is volatilized by gently igniting the residue. When the 
 carbon has been burned off, the alkaline-earth metals are recon- 
 verted to carbonates by treatment with carbonic-acid water, and 
 after evaporating and gently igniting, the residue is again weighed. 
 The loss in weight was formerly reported as ORGANIC MATTER. 
 
 It is generally recognized that the results of this determination 
 are at best but roughly approximate. During the evaporation 
 considerable organic matter may be lost, still more being volatilized 
 if the residue is dried at 180; while if the residue is dried at 
 105, much water will be retained if hygroscopic salts, such as cal- 
 cium chloride, are present. During the ignition this water will be 
 expelled together with considerable amounts of sodium chloride, 
 if this commonly occurring salt is present. The LOSS ON IGNITION 
 is therefore considered of very little value, the chief reason for 
 igniting the residue being to observe the amount of carbonization 
 which takes place. The blackening of the residue by the forma- 
 tion of carbon, which subsequently burns off, is taken as an indica- 
 tion of the presence of organic matter, even if the loss in weight is 
 inconsiderable; while the absence of a charred residue indicates 
 the absence of organic matter, even if considerable loss in weight 
 occurs. The presence of nitrates is indicated by sparking of the 
 carbon on ignition. On account of the difference in practice, the 
 temperature at which the total solids were dried should be stated in 
 reporting the analysis. 
 
 BACTERIOLOGICAL EXAMINATION OF WATER. 
 
 643. Bacteria Present in Water. All natural waters contain 
 varying numbers of bacteria which are microscopic single-celled 
 organisms generally living on organic matter. The bacteria which 
 feed on living organic matter, and are therefore parasitic, con- 
 stitute the class of disease germs. This class is far less numer- 
 ous than the bacteria which live on dead matter and frequently 
 cannot exist for any length of time out of their living host. Only 
 a few of this class can multiply or even live in water, the most 
 
424 WATER ANALYSIS. 
 
 common being B. /typhi, which produces typhoid fever when 
 it secures a foothold in the human system. Dysentery and similar 
 intestinal diseases are produced by bacteria which are carried 
 by water, but which have not been identified and studied. Another 
 bacterium, B. coli, which is almost always present in the intestines 
 of healthy human beings and other warm-blooded animals, can 
 also live in water. The presence of this bacterium in water is 
 therefore taken as evidence of sewage contamination. 
 
 Such water is always regarded as unwholesome, since disease 
 germs from the human system may at any time be present in 
 the water. B. typhi will almost certainly be present at times in such 
 water if sewage is the source of contamination, because cases 
 of typhoid fever may always be found in any large community. 
 
 644. Test for B. Coli. Fortunately a very simple method of 
 testing for B. coli has been devised. During its life activity, this 
 bacterium decomposes carbohydrates with the production of a 
 gas which is mainly hydrogen, with 25 to 40 per cent of carbon 
 dioxide. Bile has the property of inhibiting the growth of other 
 gas-forming bacteria. The sample of water is therefore mixed 
 with bile and lactose and kept at a temperature of 37J C. for 48 
 hours. If gas is evolved during this incubation period, the water 
 almost certainly contains B. coli. An idea of the number of 
 these organisms present can be obtained by making three tests, 
 using respectively .1 c.c., 1 c.c., and 10 c.c. of the water. The test 
 is made in two test-tubes of such a size that the smaller one can 
 be inverted within the larger one. 
 
 645. Bacterial Count. In addition to the test for B. coli, the 
 total number of bacteria in a water is determined. An excessive 
 number of bacteria in a water indicates contamination with sew- 
 age or other decomposing animal or plant tissue, and renders the 
 water unfit for consumption. This test is made by adding a 
 measured amount of the water to a plate containing a sterile 
 nutrient medium which is composed of meat extract and pep- 
 tone to nourish the growing bacteria, and gelatin or agar-agar 
 to make a semi-solid media in which the bacteria will remain 
 fixed. Each bacterium develops a colony around it, which ulti- 
 mately becomes visible to the eye under only very slight magni- 
 fication. By counting these spots or colonies, the number of 
 
BACTERIOLOGICAL EXAMINATION OF WATER. 425 
 
 bacteria originally present in the water can be ascertained. This 
 determination is only approximate, as all of the bacteria do not 
 grow under the conditions of incubation sufficiently to be counted. 
 Ill material used must be thoroughly cleaned and sterilized by 
 heat, so as to kill all organisms present before the water to be tested 
 is added. 
 
 EXERCISE 72. 
 
 Bacterial Count. 
 
 646. Sterilization of Apparatus. Clean thoroughly 12 test-tubes and 
 insert firmly a plug of cotton. Clean 6 Petri dis"hes and 6 one-c.c pipettes, 
 also 2 small glass-stoppered bottles. Place the pipettes in a copper tube 
 having a close-fitting cap, or in a large test-tube closed with a plug of cot- 
 ton. Sterilize this apparatus by placing it in an air-oven heated to 150 a 
 for 1 hour. Prepare sterile water by nearly filling a 250-c.c. flask with 
 distilled water and boiling vigorously for 15 to 20 minutes. Insert a 
 plug of cotton and allow to cool. 
 
 647. Preparation of Nutrient Gelatin. Place 500 grams of lean beef in a 
 large beaker or flask, and add 1000 c.c. of distilled water. The meat must 
 be as free as possible from fat and chopped fine or run through a sausage 
 grinder. The beaker or flask is weighed and distilled water added to 
 make up the loss by evaporation after standing 24 hours. Place a piece 
 of new cotton flannel, with the wool side up, in a large funnel and cover 
 with a layer of clean cotton wadding. Filter the meat infusion through 
 the flannel, squeezing out the last portions. It is advisable to place a 
 second smaller funnel containing the same filtering medium below the 
 first funnel, so that the filtered solution from the first funnel passes through 
 the second, thus giving a very clear solution. Allow the solution to flow 
 into the inner vessel of an agate-ware double boiler of at least two quarts 
 capacity. This vessel should be weighed empty as well as after receiving 
 the meat infusion. Add an amount of Witte's peptone equal to 1 per 
 cent of the weight of the infusion, and 10 per cent of the best quality of 
 sheet gelatin (gold label). Dissolve the peptone and gelatin by stirring 
 with a thermometer and heating the solution, not allowing the tempera- 
 ture to risa above 60 C. For this purpose the outer vessel only of the 
 double boiler is heated with the Bunsen burner, the inner vessel being 
 surrounded by the hot water. After the gelatin is dissolved, the water 
 in the outer vessel is brought to a boil and kept boiling for 30 minutes, the 
 inner vessel being covered. 
 
 648. Adjusting the Acidity. After the boiling has continued for 20 min- 
 utes, withdraw 5 c.c. of the solution with a pipette and place in a porcelain 
 dish or casserole. Add 45 c.c. of distilled water and boil for 1 minute 
 over the Bunsen-burner flame. Add 1 c.c. of phenolphthalein solution 
 
426 WATER ANALYSIS. 
 
 and titrate while hot (preferably while boiling) with N/20 caustic soda. 
 Add the soda solution until within a drop or two of the end-point. Cool 
 by standing the dish in cold water, and if a distinct pink color does not 
 develop, add the soda solution drop by drop until the end-point is reached. 
 
 So much of the acid in the solution must be now neutralized that the 
 amount of acid remaining in 1000 grams will neutralize 10 c.c. of normal 
 caustic-soda solution. The calculation of the amount of soda to be added 
 is made as follows: If the 5-c.c. portion titrated required 4 c.c. of the N/20 
 soda solution, and the weight of the solution is 950 grams, the entire solution 
 will require 855 c.c. of the N/20 soda, or 42.7 c.c. of normal soda. As the 
 acid equivalent to 10 c.c. of normal soda solution must be left free, 32.7 c.c. 
 of normal soda solution is added. 
 
 To coagulate finely suspended matter so that the solution may be 
 filtered clear, the white of an egg is added at this point. The solution is 
 cooled to 60^-70 by immersing the containing vessel in cold water, and the 
 white of an egg added with stirring; it is then heated in the double boiler 
 and finally boiled for 2 minutes over the free flame, with constant stir- 
 ring. Weigh and add distilled water to make up for loss by evaporation. 
 Take out 5 c.c. and titrate with N/20 sodium hydroxide and calculate the 
 acidity as before. If the acidity per 1000 grams is less than S c.c. or more 
 than 12 c.c. of normal acid, acid or alkali should be added to bring the 
 acidity to the standard 1 per cent. Take out another 5-c.c. portion and 
 titrate a third time to ascertain if the adjustment of the acidity has been 
 correctly carried out. 
 
 649. Sterilization of the Culture Medium. Filter the solution again 
 through absorbent cotton and cotton flannel, passing the filtrate through the 
 filter until clear. The nutrient gelatin must now be measured off in 10-c.c. 
 portions into the sterilized test-tubes closed with cotton plugs. For this 
 purpose a glass tube graduated every 10 c.c. is convenient. The cotton 
 plug is removed from a sterilized test-tube and held between the first and 
 second finger and again inserted into the test-tubes as soon as the 10-c.c. 
 portion of gelatin has been added. The gelatin should not be allowed 
 to touch the upper portion of the test-tube, which is set aside in an upright 
 position to cool. When all of the gelatin has been measured out into test- 
 tubes, it is sterilized by heating for 5 minutes in an autoclave at 120, 
 which is the temperature of steam at 15 pounds pressure. The tubes of 
 gelatin must be kept in an ice-chest. If the gelatin has not been com- 
 pletely sterilized, colonies will make their appearance in a few days. 
 
 650. Count of Bacteria. To make the count of bacteria in water, the 
 sample must be taken in a sterilized glass-stoppered bottle. It is advisable 
 before sterilization to cover the stopper with tin-foil to prevent entrance of 
 bacteria with the dust from the air; 1 c.c. is withdrawn with a sterilized 
 pipette and transferred to a sterilized test-tube, and 9 c.c. of sterilized 
 water added. This gives a dilution of 1 to 10. By diluting this solution 
 
BACTERIOLOGICAL EXAMINATION OF WATER. 427 
 
 in the same manner, a dilution of 1-100 is obtained. A clean pipette 
 must be used for each dilution. To one of these dishes 1 c.c. of the 
 sample of water is transferred; to the second dish 1 c.c. of the 1-10 dilu- 
 tion, and to the third dish 1 c.c. of the 1-100 dilution. Tubes of the 
 nutrient gelatin are melted by placing in warm water, the plug of cotton 
 removed, the open end of the test-tube sterilized by passing it through the 
 flame of a Bunsen burner, and the gelatin poured into three sterilized Petri 
 dishes. The covers are raised only when pouring in water or gelatin, so 
 as to prevent entrance of bacteria from atmospheric dust. The water 
 and nutrient gelatin are mixed by slightly tilting the dish, so that the con- 
 tents flow from one side to the other. The dish is then placed in a hori- 
 zontal position in a thermostat or refrigerator kept at about 20 C. After 
 48 hours the number of colonies on the plate are counted. If the number 
 is not over 200, the total number may be counted. If the number is large, 
 some counting device must be employed. This generally consists of a 
 plate marked off in sections or other divisions, so that the total number 
 of divisions covered by the plate may be counted, and the average num- 
 ber of colonies per division is obtanied by counting the number of colonies 
 in several divisions, so as to obtain a fair average of the number per divi- 
 sion. Small specks of dust must not be mistaken for colonies which are 
 circular in shape. The count must generally be made with a small mag- 
 nifying lens. When the colonies are small it is sometimes advisable to 
 allow the sample to incubate for 72 hours. This fact should be stated, 
 however, in reporting the analysis. The best results are generally obtained 
 by counting the colonies on plates containing about 200. 
 
 651. Preparation of Nutrient Agar-agar. When this medium is prepared 
 the meat is soaked in one-half the quantity of water, that is, 500 c.c. Fifteen 
 grams of thread agar are dissolved in 500 c.c. of water by boiling for one- 
 half hour. The loss of water is then restored and the infusion allowed to 
 cool to about 60 C. The remaining operations are identical with those 
 used in preparing gelatin, except that 2 per cent of Witte's peptone is 
 dissolved in the filtered meat infusion, after which the agar is added to the 
 meat infusion, care being taken to keep the temperature below 60 C. 
 
 The samples of water are plated as described for gelatin. It is neces- 
 sary, however, to heat the agar to a higher temperature in order to melt 
 it. It should be cooled to about 40 before pouring on the plate. The 
 plates are incubated at body temperature (37-40 C.) for 48 hours. 
 The number of colonies obtained is usually higher than with gelatin. It 
 is advisable to use porous earthenware covers with agar to prevent the 
 spreading of colonies by the condensation of water. 
 
428 WATER ANALYSIS. 
 
 EXERCISE 73. 
 Test for B. Coli. 
 
 Obtain a quantity of fresh ox-bile. Place in a flask and sterilize by 
 heating in an autoclave at 15 pounds pressure for 20 minutes, or heat in 
 steam for one-half to three-quarters of an hour. Filter through cotton 
 and add 1 per cent of lactose. Add a sufficient amount of the bile 
 to a large test-tube containing an inverted smaller test-tube, so that the 
 bile will fill the small tube and act as a seal at the bottom. Fill other 
 test-tubes in the same manner until all the bile has been utilized, and 
 plug with cotton. Sterilize in an autoclave at 15 pounds pressure for 5 
 minutes. The air in the smaller tube will be expelled by the high tem- 
 perature, and the steam produced, so that on cooling, the bile will entirely 
 fill the smaller tube. Cool and store in a cool place. 
 
 Take three of these tubes and add, with sterilized pipettes, 0.1 c.c., 1 c.c., 
 and 10 c.c. of the water to be tested. Incubate at 37 C. for 48 hours. 
 The presence of gas in the small tube is evidence of the presence of B. coli 
 in the water. 
 
 INTERPRETATION OF THE RESULTS OF A SANITARY 
 WATER ANALYSIS. 
 
 No definite limits can be set for the amount of chlorides, nitro- 
 gen, or other elements which may be present in water without 
 rendering it unsafe for domestic use. None of these substances 
 are themselves poisonous or injurious in the amounts found in 
 water. The amount present can merely be used as evidence of more 
 or less recent contamination of the water with sewage. A knowl- 
 edge of the condition of the water before contamination is the best 
 possible basis for a judgment. On comparing the analysis of a 
 water which is suspected to have produced disease with the analysis 
 of the same water before the disease appeared, if a marked increase 
 in the amount of chlorides or nitrogen is shown, contamination of 
 the water may be suspected, and its source should be looked for by 
 examining the watercourse from which the supply is drawn. 
 
 In cities where the water is analyzed at regular intervals and 
 where a careful record of the cases of typhoid and other diseases 
 known to be due to impure water is kept the amount of nitrogen, 
 chlorides, etc., in the water which indicates a contaminated water- 
 supply is soon learned. 
 
INTERPRETATION OF RESULTS. 429 
 
 652. Normal Chlorine. These difficulties in interpretation 
 arise from the fact that the substances introduced into the water 
 with sewage also find their way into the water from sources which 
 cannot produce dangerous contamination. While, for instance, 
 the chlorides present in water may be due to sewage contamina- 
 tion, which always contains large amounts of chlorides, their pres- 
 ence is by no means prima facie evidence of such contamination. 
 As the earth contains many salt deposits with which the water 
 may come in contact, and the salt from the ocean undoubtedly 
 diffuses through the earth for considerable distances inland, 
 natural waters may contain large amounts of chlorides derived 
 from these sources. The amount of chlorides derived from these 
 sources by the water of a given region is known as the normal 
 chlorine for the region. Maps have been prepared for many states 
 giving isochlors, or lines passing through regions of equal chlorine. 
 When the amount of chlorine found is in excess of the normal 
 chlorine, contamination of the water with sewage is indicated. 
 
 653. Nitrogen. As has been already stated, the ALBUMINOID 
 AMMONIA may be derived from vegetable as well as from animal 
 sources. A surface-water which has passed through a swampy 
 region almost always contains a large amount of albuminoid 
 ammonia. The presence of dissolved vegetable matter is almost 
 always indicated by the high color of the water. On the other 
 hand, the presence of much albuminoid ammonia in a light-colored 
 water is almost always suspicious, the absence of color making 
 it probable that the organic matter is partly of animal origin. If 
 the free ammonia and nitrites are also high, the evidence of pol- 
 lution is strong. 
 
 654. The Characteristics of Polluted Surface-waters are given 
 by Geo. C. Whipple as follows: 
 
 "They contain an excess of chlorine above the normal of the 
 region. 
 
 "They contain considerable organic matter, shown by the 
 albuminoid ammonia and loss on ignition ; but it should be remem- 
 bered that the dissolved albuminoid ammonia may be explained 
 by a high color. 
 
 "They contain free ammonia and nitrites, if the pollution is 
 recent and the organic matter is in a decomposing state. 
 
 "They show coincident high nitrates and high hardness. 
 
430 WATER ANALYSIS. 
 
 11 They are characterized usually by a mouldy or musty odor 
 and, if much polluted, by turbidity and sediment." 
 
 655. Ground- waters are very different from surface-waters. 
 As a rule they are clear, almost colorless, and without odor. Hav- 
 ing been much in contact with the soil, they are rich in mineral 
 matter. The fixed solids are higher, and usually the hardness. 
 The amount of chlorine and nitrogen may be high, indicating pollu- 
 tion, but if the nitrogen is nearly all present as nitrate, the water 
 has been purified by filtration through the soil and oxidation of 
 the organic matter so that it may be safely used. 
 
 The determination of the amount of TOTAL SOLIDS as well as 
 HARDNESS is carried out not so much to determine the sanitary 
 character of the water as its suitability for laundry purposes, 
 though the continued drinking of hard water is undoubtedly 
 injurious to many constitutions. The presence of considerable 
 amounts of iron renders water unfit for laundry use, the presence 
 of more than 0.5 part per million being objectionable. If the 
 water contains much organic matter, more iron may be present 
 without being objectionable. A hardness of 25 to 50 is most de- 
 sirable for ordinary use. A water of hardness of 50 is noticeably 
 hard, while if above 100 it is classed as very hard. 
 
 The COLOR of water should not exceed 25. The amount of 
 color which this figure implies may be judged from the fact that 
 if the color of a water is 15 to 20 it will be noticed when the 
 water is allowed to flow into a bath-tub, while if the color is 30 it 
 may be noticed in a tumblerful. 
 
 Waters having a TURBIDITY of more than 3 or 4 are objection- 
 able. 
 
 ANALYSIS OF WATER FOR USE IN BOILERS. 
 
 Aside from the sanitary analysis of water, the most commonly 
 conducted analysis is that carried out for the purpose of deter- 
 mining the adaptability of the water in question to use in boilers. 
 The water may not be suitable for use in boilers for two reasons. 
 
 It may, in the first place, contain substances which cause the 
 CORROSION of the iron of which the boiler is made. FREE ACIDS, 
 
 MAGNESIUM CHLORIDE, AMMONIUM SALTS, MUCH DISSOLVED OXY- 
 GEN, HUMUS, AND FATTY SUBSTANCES are the commonly occurring 
 substances which have been found to cause corrosion in boilers. 
 
BOILER-SCALE. 431 
 
 A water may also be objectionable for boiler use from contain- 
 ing SCALE-FORMING INGREDIENTS, among which the most impor- 
 tant are CALCIUM SULPHATE and CARBONATE and MAGNESIUM 
 CARBONATE. Others occurring commonly, though in smaller 
 amounts, are SILICA and the OXIDES OF IRON AND ALUMINIUM. 
 Although rarely forming parts of the boiler-scale, the alkalies 
 should in a careful analysis be determined, as the distribution of 
 the hydrochloric and sulphuric acids present may then be more 
 correctly determined. 
 
 656. Hardness. The fitness of a water for boiler use may be 
 somewhat roughly determined by simply determining the tem- 
 porary and permanent hardness, as given in Chapter XXI, page 
 269. All of the calcium and magnesium carbonates constituting 
 temporary hardness are precipitated by boiling and become part 
 of the boiler-scale. The calcium and magnesium existing as per- 
 manent hardness may be present as chlorides or sulphates. The 
 determination of permanent hardness does not give the amount 
 of magnesium chloride present which causes corrosion, nor the 
 amount of calcium sulphate present which is precipitated at the 
 temperature existing in the boiler and constitutes part of the 
 boiler-scale. 
 
 657. For a Complete Analysis an amount of water should be 
 taken which will give from J to 1 gram of residue. It is evapo- 
 rated to dryness in a platinum dish and the TOTAL SOLIDS deter- 
 mined as already directed. The ignition of the residue to volatilize 
 organic matter and burn the carbon must be conducted at dull 
 redness and the ignition continued only for a few minutes, or a loss 
 of alkalies may occur. After weighing, the residue is digested 
 with water and a few cubic centimeters of hydrochloric acid. 
 The insoluble residue is filtered off, washed, ignited in the 
 platinum dish and weighed. It is mainly SILICA, though a little 
 iron, aluminium, and calcium sulphate may be present. The 
 silica may be volatilized by treatment with hydrofluoric acid 
 and a drop or two of sulphuric acid, the residue weighed after 
 ignition, and the weight deducted from that of the impure silica. 
 The .residue in the dish is dissolved by digestion with a little 
 hydrochloric acid and water and is added to the main solu- 
 tion. 
 
432 WATER ANALYSIS. 
 
 The IRON and ALUMINIUM may be precipitated with ammonia 
 and weighed as oxides in the usual manner. 
 
 The CALCIUM is precipitated as oxalate from the nitrate and 
 weighed as oxide. 
 
 The MAGNESIUM and the ALKALIES in the nitrate may be weighed 
 together as sulphates, the solution being evaporated to dryness 
 in a platinum dish and the ammonium salts volatilized after the 
 addition of a few drops of sulphuric acid. A small amount of 
 silica and other insoluble matter generally remains with the sul- 
 phates of magnesium and the alkalies. This is separated by treat- 
 ing the residue in the dish with water and filtering. The filtrate 
 is again evaporated to dryness in the dish, and after gentle igni- 
 tion is weighed. The weighed sulphates are dissolved in water, 
 the solution diluted to 100 c.c. and divided into two portions, in 
 one of which the magnesium is determined in the usual manner 
 as pyrophosphate, while in the other portion the potassium is 
 precipitated as platinochloride, washed according to the Lindo- 
 Gladding method and weighed. The amount of magnesium 
 present as sulphate is calculated from the weight of the pyro- 
 phosphate, and the weight of potassium as sulphate is calcu- 
 lated from the weight of the potassium platinochloride. The 
 difference between the sum of these two weights and the weight 
 of the combined magnesium and alkali sulphates gives the weight 
 of sodium sulphate present. 
 
 The amount of CHLORIDES present is determined volumetric- 
 ally according to the method given under the sanitary analysis 
 of water on p. 421. The amount of SULPHATES present is also 
 determined in a separate portion. Unless a considerable amount 
 is present, 500 c.c. are evaporated down to a bulk of about 100 c.c. 
 and, after acidifying with a drop or two of hydrochloric acid, the 
 sulphuric acid is precipitated by the addition of barium chloride 
 solution. After digestion on the water-bath the barium sul- 
 phate is filtered off, washed with hot water, ignited, and weighed 
 in the usual manner. 
 
 If FREE MINERAL ACIDS are present, the amount is determined 
 volume trically by titration with a standard alkali. 
 
 658. Method of Combining Acids and Bases. The elements 
 present in the water having been determined, it remains to com- 
 
CALCULATION OF RESULTS. 433 
 
 bine them in such a manner that the properties of the water 
 when used in a boiler may be easily shown, no attempt being made 
 to combine them as they are believed to exist when in solution. 
 The following method is generally followed : The chlorine is com- 
 bined with the bases in the following order: sodium, potassium, 
 magnesium, and calcium. The sulphuric acid is combined with 
 the alkalies if the amount of chlorine was insufficient, then with 
 the calcium and, if any remains, with the magnesium. The cal- 
 cium and magnesium remaining are calculated as carbonates. 
 
 That this method of combination does not exist in solution 
 is evident from the consideration that the strong base calcium 
 would undoubtedly be in combination with some of the hydro- 
 chloric acid, while the alkalies would also be combined partly 
 with the sulphuric and partly with the hydrochloric acid. When, 
 however, the temperature was reached at which the calcium 
 sulphate is almost absolutely insoluble in water this salt would 
 be precipitated, even if the sulphuric acid were combined with 
 the alkalies and the calcium with the hydrochloric acid. The 
 ultimate result would therefore be that the alkalies would be found 
 in combination with the chlorine, and the calcium with the sul- 
 phuric acid. Some authorities recommend that as much magne- 
 sium chloride as can be formed from the amounts of these elements 
 present be reported in order to bring out the fact that the presence 
 of the two elements together leads to the formation of hydrochloric 
 acid and magnesium oxide. This method of reporting the analy- 
 sis probably overstates the danger, as some of the magnesium is 
 undoubtedly precipitated as carbonate if that acid is present in 
 the solution. 
 
 659. Calculation of Results. The simplest way in which to 
 state the results is in grams per liter or in milligrams per liter, 
 which would be identical with parts per million. The results 
 must frequently be given in grains per United States gallon (58,318 
 grains) or per imperial gallon (70,000 grains), to suit the conve- 
 nience of engineers who still use the English instead of the metric 
 system. 
 
 The method of calculation may be more clearly understood 
 from the following example. The analysis gave the following 
 results : 
 
434 WATER ANALYSIS. 
 
 Si0 2 0.0063 gram per liter 
 
 S0 3 0.0245 " " " 
 
 Cl 0.0075 " " " 
 
 K 2 0.0021 " " " 
 
 Na,0 0.0065 " " " 
 
 MgO 0.0125 " " " 
 
 CaO 0.0223 " " " 
 
 Fe 2 3 +Al 2 3 0.0040 "' " " 
 
 The sodium is first calculated to sodium chloride as follows: 
 
 Na 2 0:2NaCl:: 0.0065 :x 
 62. : 117.9 ::0.0065:o; 
 x = 0.0123 gram Nad. 
 
 The amount of chlorine in this weight of sodium chloride is 
 then calculated: 
 
 NaCl: Cl :: 0.0123 :x 
 
 58.5: 35.45:: 0.0123 :z 
 
 x = 0.0075 gram Cl. 
 
 As this is the amount of chlorine found by analysis, no other 
 chlorides are present. 
 
 The potassium is next calculated to sulphate : 
 
 K 2 0:K 2 S0 4 :: 0.0021: x 
 94.3: 174.36:. 0.0021:* 
 x = 0.0039 gram K 2 S0 4 
 K 2 = 0.0021 " 
 
 S0 3 = 0.0018 " 
 
 The amount of potassium oxide being subtracted, it is found 
 that 0.0018 gram of sulphur trioxide has been taken to combine 
 with the potassium, leaving 0.0227 gram (0.0245-0.0018) which is 
 combined with calcium oxide to form sulphate. 
 
 S0 8 :CaS0 4 :: 0.0227: a; 
 80:136.2 ::0.0227:z 
 x =0.0386 gramCaSO, 
 S0, = 0.0227 " 
 
 CaO =0.0159 
 
CALCULATION OF RESULTS. 435 
 
 0.0064 gram (0.0223-0.0159) of calcium oxide remains after 
 all of the sulphur trioxide has been combined to form sulphate. 
 This calcium oxide will be present as carbonate. 
 
 CaO-CaC0 3 ::0.0064:z 
 56 :100 ::0. 0064:s 
 x = 0.01 14 gram of CaC0 3 . 
 
 The chlorine and sulphur trioxide being all combined, all of 
 the magnesium will be present as carbonate. 
 
 MgO:MgC0 3 ::0.0125:z 
 40.3 : 84.3:: 0.0125: x 
 j = 0.0261 gram MgC0 3 . 
 
 The result of the analysis is therefore reported as follows: 
 
 NaCl= 0.0123 gram per liter 
 
 K 2 S0 4 = 0.0039 " " " 
 
 CaS0 4 = 0.0386 " " " 
 
 CaC0 3 = 0.0114 " il li 
 
 MgC3 3 = 0.0261 " " " 
 
 Fe 2 3 +Al 2 3 = 0.0040 " " " 
 
 Si0 2 =0.0063 " " " 
 
 LITERATURE. 
 
 WATER SUPPLY. Mas^n (John Wiley & Sons) 
 
 AIR, WATER, AND FOOD. Richards & Woodman (John Wiley & Sons). 
 
 WATER SUPPLY. Nichols (John Wiley & Sons). 
 
 SEWAGE. Rideal (John Wiley & Sons) . 
 
 EXAMINATION OF WATER. Leffman (Blakiston). 
 
 WATER SUPPLIES. T. C. Thresh (Rebman Pub. Co., London). 
 
 SPECIAL REPORT ON EXAMINATION OF WATER SUPPLY, 1890, by the Mass. 
 State Board of Health. 
 
 SPECIAL REPORT ON PURIFICATION OF SEWAGE AND WATER, 1890, by the 
 Mass. State Board of Health. 
 
 TYPHOID FEVER, ITS CAUSATION, TRANSMISSION, AND PREVENTION. Geo. 
 C. Whipple (John Wiley & Sons). 
 
 ELEMENTS OF WATER BACTERIOLOGY, WITH SPECIAL REFERENCE TO SANI- 
 TARY WATER ANALYSIS. Prescott and Winslow (John Wiley & Sons), 
 
 THE BACTERIOLOGICAL EXAMINATION OF WATER-SUPPLIES. W. G. Savage 
 (Blakiston & Sons, Philadelphia). 
 
CHAPTER XXX. 
 ANALYSIS OF FATS AND OILS. 
 
 660. The Quantitative Analysis of a fat or oil is conducted 
 for the purpose of identifying the constituent fats or oils as well 
 as ascertaining the proportion in which they are present. The 
 difficulties of the subject arise from the fact that not only are 
 most pure fats and oils mixtures of a considerable number of chemi- 
 cal substances, but a given fat or oil, being produced as a part of 
 an animal or vegetable structure, is by no means uniform in com- 
 position, but varies with the conditions under which it was pro- 
 duced. Changes in climate, season, food, etc., produce consid- 
 erable variations in the fat or oil produced. The number of oils 
 and fats which are produced on a commercial scale is very large, 
 thus adding very much to the complexity of the subject. The 
 mineral oils, being obtained by distillation of crude products from 
 various parts of the earth, differ still more widely in composition. 
 The strictly chemical examination is confined almost exclusively 
 to vegetable fats and oils. The methods employed are essen- 
 tially inorganic, and therefore come within the scope of this work. 
 The interpretation of the results should be undertaken only after 
 the consultation of a more extensive work on the subject,* except 
 in the case of pure oils and comparatively simple mixtures. 
 
 66 1. Chemical Composition. All animal and vegetable oils 
 and fats are organic salts or esters in which a common base, 
 GLYCERINE, is always present. Besides glycerine, most fats and 
 oils and especially the waxes contain a larger or smaller quantity 
 of a base peculiar to themselves. Of these, Benedikt and Ulzer 
 
 * A very excellent work for this purpose is that by Lewkowitsch on "Chemical 
 Analysis of Oils, Fats, and Waxes." (Macmillan & Co.) A recent German 
 publication is also excellent, "Analyse der Fette und Waehsarten." Benedikt 
 und Ulzer. (Julius Springer, Berlin.) 
 
 436 
 
FREE ACID. 437 
 
 mention twelve. Three of these belonging to the aromatic series 
 furnish a means of distinguishing between fats and oils of vege- 
 table or animal origin. CHOLESTERIN and ISOCHOLESTERIN occur 
 exclusively in substances of animal origin, while vegetable products 
 contain another base called PHYTOSTERIN. These may be liber- 
 ated from the fatty acids by saponificaticn, and after extraction 
 with ether, may be crystallized from alcohol, cholesterin and 
 isocholesterin forming irregular plates or tables, phytosterin, 
 groups of needles. The first melts at 147, the second at 138, 
 and the third at 132-134. Failure to obtain these substances, 
 however, is no indication that the oil or fat was not derived from 
 its proper source, since they often exist in such small quantity as 
 to be isolated with difficulty. The organic acids present in com- 
 bination with the glycerine are very numerous and differ greatly 
 in their properties. On these differences are founded most of the 
 methods of identifying the oils and fats. 
 
 662. Acidity. In many oils and fats, especially those of vege- 
 table origin, part of the acid exists uncombined with the base. 
 The amount of this free acid increases with the age of the oil, 
 this being especially true of palm-oil. The amount of the free 
 acid in most fresh animal oils is generally so small as to be neg- 
 ligible. Old and rancid fats become strongly acid. 
 
 The method of determining this value, as well as preparing the 
 solutions required, is given in the following paragraphs. 
 
 663. Standard N/2 Hydrochloric Acid Solution. Prepare 2 
 liters of a half -normal solution of hydrochloric acid by measuring 
 out about 40 c.c. concentrated acid and diluting to a liter. Repeat 
 the operation and pour the two solutions together and mix thor- 
 oughly. Standardize the solution by one of the methods given 
 in Chapter XXI, page 258, and dilute to exact strength. 
 
 664. Standard N/2 Caustic Potash Solution. Prepare some 
 pure alcohol by adding to about 2 liters of the ordinary alcohol a 
 few pieces of caustic potash, shaking thoroughly, and after allowing 
 the solution to stand for a few hours, distilling off the alcohol. To 
 1 liter of the alcohol add 30 grams of caustic potash. The alcohol 
 should be poured into a liter bottle, the potash added, and the 
 bottle closed with a rubber stopper. Shake until the caustic pot- 
 ash is dissolved. The potassium carbonate will not dissolve, but 
 
438 ANALYSIS OF FATS AND OILS. 
 
 will adhere quite firmly to the sides and bottom of the bottle if 
 the solution is allowed to stand undisturbed for a few hours. 
 Decant or siphon off the clear liquid into another liter bottle, 
 exposing the solution to the air as little as possible. 
 
 The solution is standardized by titrating with the standard 
 hydrochloric acid, using phenolphthalein as the indicator. It 
 should be protected from the carbon dioxide of the air both 
 when being withdrawn from the bottle and during a titration. A 
 simple method of accomplishing this object is to arrange a siphon 
 for withdrawing the solution from the bottle, as shown in Fig. 50, 
 p. 290. The siphon should be passed through a rubber stopper 
 with two holes, a straight tube containing soda-lime being inserted 
 in the second hole of the stopper The end of the siphon exposed 
 to the air should be kept closed with a piece of rubber tubing 
 having one end closed with a glass plug. After being filled, the 
 burette should be immediately closed with a small rubber stopper 
 through which passes a soda-lime tube. 
 
 665. Determination of Free Acid. For the determination of the 
 acid value of oils, calculate the volume of the N/2 caustic potash 
 solution to measure out and dilute with alcohol to 250 c.c. to make 
 a fifth-normal solution. As the solution is not stable it must be 
 used very soon after being made. All solid fats must be melted at 
 as low a temperature as possible, and after standing until the 
 impurities have settled, the clear fat is carefully decanted through 
 a dry filter-paper. The vessel containing the oil or fat is weighed 
 together with a small glass rod or spoon. From 1 to 2 grams are 
 transferred to an Erlenmeyer flask, the exact amount being obtained 
 by difference. 50 c.c. of the redistilled alcohol are now added and 
 the flask shaken until the oil is dissolved. If the fat does not 
 readily dissolve in alcohol, 50 c.c. of a mixture of equal parts of 
 alcohol and ether may be used. One drop of phenolphthalein is now 
 added. The solution is titrated with the fifth-normal caustic potash 
 solution, which should be added drop by drop with constant and 
 vigorous shaking. The first appearance of pink color is taken as the 
 end-point. It may fade on standing a few minutes, but this is due 
 to saponification of the neutral glycerides by the excess of alkali. 
 
 When oils of high free acid value, such as palm-oil or old and 
 rancid *ts are being tested, less than 1 gram may be taken for a 
 
KOTTSTORFER VALUE. 439 
 
 titration. When the acid value is small, as in most solid fats, as 
 much as 5 or 10 grams may be taken to obtain a reliable determina- 
 tion. The number of milligrams of KOH per gram of fat or oil is 
 calculated, which is the acid value. The acid value is also fre- 
 quently expressed in terms of the amount of oleic acid corre- 
 sponding to the amount of alkali used. The molecular weight of 
 oleic acid being 282, 1 c.c. of N/5 alkali is equal to 0.0564 gram of 
 oleic acid. 
 
 666. Kottstorfer Value. As the acids present in fats and oils 
 differ in molecular weight and basicity, varying amounts of alkali 
 will be required to neutralize the acid in a definite amount of the 
 oil or fat. Caustic potash is the alkali invariably used, 1 gram 
 of the fat or oil being taken for the determination. The so-called 
 Kottstorfer value is the number of milligrams of KOH required to 
 neutralize the acids present in 1 gram of the fat or oil. When an 
 oil or fat is treated with an alcoholic solution of caustic potash 
 the union between the glycerine and the fatty acids is broken, 
 so that the acid is left in combination with the potassium. This 
 process is called SAPONIFICATION. An excess of standard caustic 
 potash dissolved in alcohol is added to a weighed amount of the 
 fat or oil. When the saponification is complete the excess of 
 alkali is titrated with standard acid. The number of milligrams 
 of KOH neutralized by one gram of the fat or oil is then calcu- 
 lated to find the Kottstorfer value. 
 
 667. Determination of Kottstorfer Value. To determine the 
 Kottstorfer or saponification value, weigh out as directed for the 
 determination of free acid from 1 to 2 grams of the fat or oil into 
 a small Erlenmeyer flask fitted with a cork to a return condenser. 
 Add 25 c.c. of the alcoholic caustic potash solution with a pipette 
 or a burette, and heat on the water-bath until the oil or fat is 
 dissolved, which will require from ten minutes to half an hour, 
 depending on the fat or oil being investigated. Measure out 25 c.c. 
 of the caustic potash solution into a similar flask connected with 
 a return condenser, and digest on the water-bath for the same time. 
 Cool for a few minutes and titrate the contents of each flask with 
 the N/5 hydrochloric acid, using rather more phenolphthalein than 
 usual as the indicator. The difference between the acid used in 
 the two titrations gives the acid necessary to neutralize the alkali 
 
440 ANALYSIS OF FATS AND OILS. 
 
 used in the saponification. Calculate the number of milligrams of 
 KOH necessary to saponify 1 gram of the fat or oil. 
 
 668. Ether Value. The ether value is the difference between 
 the saponification and the acid value. 
 
 669. Reichert and Reichert-Meissl Value. Some of the fatty 
 acids of small molecular weight are quite volatile, and also soluble 
 in water. It has been found that by liberating the acids con- 
 tained in a given fat or oil, adding water, and distilling off this 
 water again, a distillate is obtained which contains a considerable 
 percentage of the volatile acids. It is found in practice that the 
 total amount of volatile acid present is not separated in this 
 manner by one distillation. The amount which comes over depends 
 upon the dilution of the solution, the amount of the distillate, and the 
 rapidity of the distillation. The amount of acid in the distillate is 
 determined by titration with decinormal caustic potash. The 
 REICHERT VALUE indicates the number of cubic centimeters of N/10 
 potash required to neutralize the acid from 2.5 grams of fat 
 contained in 100 c.c. of distillate, 110 c.c. of which has passed 
 over in one-half hour. The REICHERT-MEISSL value is obtained 
 in the same manner, 5 grams of fat having been used instead of 
 2.5 grams. This number is not necessarily twice the Reichert 
 number. This determination has been used more especially to 
 distinguish butter from the artificial substitutes which have been 
 put on the market. 
 
 670. Determination of Reichert-Meissl Value. For the deter- 
 mination of the Reichert-Meissl number a sample of butter-fat 
 should be used which has been freed from casein, water, salt, 
 etc., by melting and filtering as already directed. A quantity of 
 this fat as nearly as possible 5 grams is weighed out into a 250-c.c. 
 Erlenmeyer flask, 2 c.c. water, 8 c.c. alcohol, and a stick of caustic 
 soda weighing about 2 grams are added. Attach the flask to an 
 inverted condenser by means of a cork stopper and heat on the 
 water-bath for one hour, shaking the contents of the flask occa- 
 sionally. Remove or incline the condenser and distil off the alco- 
 hol by immersing the flask in the hot water and heating for about 
 three-quarters of an hour. After the alcohol is removed add 
 100 c.c. recently boiled distilled water and dissolve the soap by 
 warming on the water-bath and shaking. Cool the solution to 
 
POLENSKE VALUE. 441 
 
 60 or 70 and add 50 c.c. of a solution of sulphuric acid containing 
 25 c.c. concentrated acid per liter. Again attach the flask to the 
 inverted condenser and heat on the water-bath until the insoluble 
 acids collect in a clear oily layer on top of the liquid. Allow the 
 contents of the flask to cool to the room temperature. Introduce 
 a few pieces of pumice-stone which have been heated to redness 
 and dropped into distilled water. Attach the flask to an inclined 
 condenser and distil off 110 c.c. into a graduated cylinder, regulating 
 the heat so that thirty minutes as nearly as possible shall be 
 required for the distillation. Mix the distillate thoroughly by 
 pouring back and forth into a dry beaker. Filter through a dry 
 paper into a 100-c.c. flask. Pour into a beaker, add a few drops 
 of phenolphthalein, and titrate with N/10 alcoholic KOH made by 
 diluting the half -normal solution. When the end-point is reached 
 the 100-c.c. flask is rinsed and the titration completed. The num- 
 ber of cubic centimeters of N/10 alkali used multiplied by 1.1 gives 
 the Reichert-Meissl number. 
 
 671. The Polenske Value. This method of examining butter- 
 fat has come into use for the detection of cocoanut-oil, which has 
 been largely used for the adulteration of butter. The Reichert- 
 Meissl value has been generally relied on to detect the admixture 
 of foreign fats in butter, on account of its high value in the case 
 of butter-fat and the low value given by other fats, except cocoa- 
 nut-oil, which gives a Reichert-Meissl value of 7 to 7.8, so that a 
 considerable amount of this oil can be mixed with a butter having 
 a high value, 30 for instance, producing a mixture which would 
 have a Reichert-Meissl value above the legal limit of 20 to 24. 
 Polenske found that the greater part of the volatile fatty acids 
 of butter obtained in the Reichert-Meissl determination are solu- 
 ble in water, while the reverse is true of cocoanut-oil. The 
 amount of the soluble and insoluble acids is obtained, as in the 
 Reichert-Meissl determination, by titration with decinormal caus- 
 tic potash, and the values are given in cubic centimeters of the 
 alkali. For butter-fat the insoluble volatile fatty acids required 
 1.5 to 3.0 c.c. of the decinormal alkalies, while 16.8 to 17.8 c.c. 
 was required for the insoluble volatile fatty acids from cocoanut- 
 oil. The Polenske value varies with the Reichert-Meissl value, 
 as is shown in the table on page 459. 
 
442 ANALYSIS OF FATS AND OILS. 
 
 672. The Determination of the Polenske Value is made in a 
 manner very similar to the Reichert-Meissl determination. It 
 has been found that the details of the method must be adhered 
 to very strictly in order to obtain concordant results. Five grams 
 of the butter-fat are placed in a 300-c.c. flask and saponified with 
 caustic soda and glycerine. For this purpose 2 c.c. of a caustic- 
 soda solution, prepared from equal parts of caustic soda and water, 
 and 20 grams of glycerine, are added, and the flask heated over 
 the free flame until saponification is complete. The solution is 
 allowed to cool below 100, and 90 c.c. of water added. The mass 
 is dissolved by warming on the water-bath to 50 C. The solu- 
 tion must be clear and colorless. If it is brown it has been over- 
 heated during the saponification and must be rejected and the 
 saponification repeated. 50 c.c. of dilute sulphuric-acid solution 
 containing 25 c.c. concentrated sulphuric acid per liter are now 
 added, and the volatile acids distilled off. For this purpose i 
 gram of pumice is added. This pumice must be broken and sifted 
 so as to include only pieces of 0.5 to 1 mm. diameter. The flask 
 is placed on copper or brass gauze of 0.5 mm. mesh. The flask 
 is connected to an upright condenser by means of a bulb tube. 
 The bulb is immediately above the stopper of the flask, and the 
 tube passing to the condenser is inclined toward the bulb, which 
 must be 70 mm. from the condenser. The heat is regulated so 
 that 110 c.c. are distilled off in 19 to 20 minutes. The cooling 
 water must flow rapidly enough through the condenser to cool 
 the distillate to 20-23 C. as it drops into a 110-c.c. flask, which 
 serves as a receiver. When 110 c.c. have distilled over, the flask 
 is replaced with a 20-c.c. measuring cylinder. 
 
 The 110-c.c. flask containing the distillate is immersed, with- 
 out shaking, in water at 15 C. After about 5 minutes the 
 neck of the flask is tapped lightly, so as to cause the drops of oil 
 floating on the surface of the liquid to adhere to the walls of the 
 flask. After another 10 minutes the consistency of the oily drops 
 is noted. If the butter is pure and has a high Reichert-Meissl 
 value, the drops will be semi-solid and milky. If the Reichert- 
 Meissl value is low, or 10 per cent or more of cocoanut-oil is pres- 
 ent, the insoluble fatty acids do not solidify, but remain clear 
 and oily. The contents of the flask are mixed by inverting it 
 
DETERMINATION OF THE POLENSKE VALVE. 443 
 
 several times without violent shaking, and 100 c.c. filtered through 
 an 8-cm. filter-paper and titrated with tenth-normal caustic potash. 
 The insoluble acids on the paper are washed free from soluble acids 
 by washing with three 15-c.c. portions of water which have been 
 passed through the condenser tube and the 20-c.c. cylinder and 
 the 110-c.c. flask. The insoluble acids are then dissolved in neu- 
 tral 90 per cent alcohol by passing three 15-c.c. portions through 
 the condenser tube, the 20-c.c. measuring cylinder, the 110-c.c. 
 flask, and the filter-paper. Each portion is allowed to drain 
 completely before adding a fresh portion. The combined alco- 
 holic solution is then titrated with tenth-normal alkali to obtain 
 the Polenske value. 
 
 673. The Hehner Value is the converse of the Reichert value, 
 the percentage of insoluble acids being determined by saponifying 
 the fat, adding an excess of mineral acid so as to liberate the fatty 
 acids, then filtering off, washing, and weighing the insoluble portion. 
 
 674. The Iodine Value of oils and fats is obtained by allowing 
 a measured amount of an iodine solution to act on the oil or fat. 
 It has been found that a considerable amount of the halogen is 
 absorbed by the fatty acids present. The percentage of the halo- 
 gen absorbed is taken as the iodine value. The chemical reactions 
 which take place are not fully understood. In general it is thought 
 that in the unsaturated acids the double bond is broken and the 
 iodine enters at this point into the molecule. There seems to be 
 no doubt, however, that substitution takes place to a limited 
 extent, an atom of iodine taking the place of an atom of hydrogen, 
 which unites with another atom of iodine, forming hydroiodic 
 acid. Solutions of bromine have been used in the same manner, 
 while recently solutions of iodine monochloride and also solutions 
 of iodine monobromide have come into use. The results are almost 
 universally computed as percentage of iodine absorbed. Slightly 
 different values are obtained by the use of these different solu- 
 tions. Strictly comparable results can be obtained only by 
 working with the same solutions and under the same conditions, 
 especially as to the time during which the halogen is allowed to 
 act on the oil or fat. 
 
 675. Wijs* Iodine Monochloride Solution. For the deter- 
 mination of the iodine, the Wijs solution of iodine is the most 
 
444 ANALYSIS OF FATS AND OILS. 
 
 advantageous for use. This is a solution of iodine monochloride 
 in glacial acetic acid. The acid must be the strongest and purest 
 obtainable, giving no reduction with sulphuric acid and potas- 
 sium dichromate. The percentage of acid must not be less 
 than 9-9 and can be most readily determined by taking the melt- 
 ing-point which varies with the amount of water present, as is 
 evident from the following table: 
 
 Per cent of H.C 2 H 3 2 . Melting-point. 
 
 100 16.75 
 
 99.5 15.65 
 
 99 14.80 
 
 97.09 11.95 
 
 95.24 9.40 
 
 The test is made by half filling an 8-inch test-tube with the 
 acetic acid. This test-tube is suspended within a larger test-tube 
 by means of a cork, through which the smaller tube is passed. A 
 delicate and accurate thermometer is placed in the acid, which 
 is cooled by immersing the whole in ice-water As soon as the 
 acetic acid begins to crystallize, the temperature rapidly rises to 
 the true melting-point of the acid. The advantage of using this 
 solution is that its titre remains unchanged for months, and the 
 time necessary for a determination is reduced from four or six 
 hours to a few minutes. 
 
 Thirteen grams of pure iodine is dissolved in a liter of the 
 acid. The exact strength is then determined by tit rat ion against 
 a standard thiosulphate solution. Chlorine free from hydro- 
 chloric acid is passed into the solution until the dark-brown color 
 changes to a light-brown or yellow color. The strength should 
 then be determined again by titration against the thiosulphate 
 solution, and should be double the original strength. It i| advis- 
 able, however, to have a slight excess of iodine present rather 
 than of chlorine. 
 
 676. Harms' Monobromide Solution. The Hanus solution 
 serves the same purpose as the Wijs solution, and is more easily 
 made. Instead of passing chlorine into the acetic-acid solution 
 of iodine, liquid bromine is added until the titre is doubled. About 
 3.0 c.c. will be found sufficient. 
 
IODINE VALVE. 445 
 
 677. Sodium-thiosulphate Solution. An N/10 sodium-thio- 
 sulphate solution is made up and standardized by means of re- 
 sublimed iodine. This solution is then titrated against the iodine 
 solution. For this purpose 20 c.c. of the iodine solution are meas- 
 ured out into a beaker, 100 c.c. distilled water added, and 1 gram 
 potassium iodide dissolved in 10 c.c. water. The thiosulphate 
 solution is then added until a faint yellow color is obtained, the 
 starch solution is added, and the titration completed. 
 
 678. Determination of Iodine Value. To determine the iodine 
 value of an oil, from 0.15 to 0.2 gram of a drying-oil, 0.3 to 0.5 
 gram of a non-drying oil, or 0.7 to 1 gram of a solid fat is weighed 
 out and transferred to a glass-stoppered bottle. The most con- 
 venient method of weighing out oils is to pour a suitable amount 
 into a small beaker, place in the beaker a short glass tube, and 
 weigh carefully. By means of the glass tube, drop some of the 
 oil into the bottle, replace the tube in the beaker, and weigh again. 
 The oil may also be weighed in a tared homeopathic shell vial. 
 The vial containing the oil is introduced into the glass-stoppered 
 bottle. The oil or fat is dissolved in 10 c.c. of chloroform and 25 to 
 50 c.c. of the iodine solution are added. The iodine must be pres- 
 ent in excess. If the iodine solution is very much decolorized by 
 the oil, more must be added. The iodine should be allowed to act 
 for fifteen minutes for non-drying oils, thirty minutes for semi- 
 drying oils, and one hour for drying oils. A blank determination 
 is carried on at the same time, 10 c.c. of the chloroform being 
 treated with iodine solution for the same length of time. 
 
 After the iodine has acted on the oil, 10 c.c. of a 10% solution 
 of potassium iodide free from iodate are added to the bottle, and 
 this is followed, after thorough agitation, by about 100 c.c. of 
 distilled water, care being taken that all of the iodine solution is 
 washed down from the stopper, neck, and walls of the bottle. 
 The excess of iodine is then titrated with thiosulphate. The num- 
 ber of cubic centimeters of thiosulphate used, subtracted from the 
 amount required for the blank, represents the amount of iodine 
 absorbed by the oil. The equivalent weight of iodine thus found 
 is divided by the weight of oil taken, and the result expressed in 
 per cent as the iodine number. 
 
446 ANALYSIS OF FATS AND OILS. 
 
 679. Determination of Rosin in Shellac by A. C. Langmuir's 
 Method.* The iodine value of unbleached shellac varies from 15 to 
 18 per cent, while that of rosin varies from 175 to about 260. It 
 is possible, therefore, to ascertain approximately the amount of 
 rosin present in a given sample of shellac by a determination of 
 the iodine number. For calculating the result, 18 is used as the 
 iodine number of the shellac, and 228 as that of the rosin. By 
 using these figures the percentage of the adulterant which is 
 found may be slightly lower than that actually present. This 
 is in accord with the accepted commercial principle that the 
 inaccuracies in a method of analysis shall favor the seller rather 
 than the buyer, by assigning the lowest possible value to the 
 amount of impurities present. The calculation is made accord- 
 ing to the formula 
 
 A-M 
 
 where ?/ = per cent of rosin; 
 
 M = iodine number of shellac; 
 N= " " " rosin; 
 A= " " " mixture. 
 
 Substituting the fixed values of M and N, the formula becomes 
 
 A-18 
 
 y-100 
 
 210 
 
 The iodine number is determined by means of the Wijs solu- 
 tion. Exactly 0.2 gram of the ground sample is introduced into 
 a dry 250-c.c. glass-stoppered bottle. 20 c.c. of glacial acetic 
 acid is introduced and warmed until the shellac is dissolved, with 
 the possible exception of a little wax. 10 c.c. of chloroform is 
 then added and the solution cooled to 21-24 C. This tem- 
 perature must be maintained during the remainder of the test. 
 20 c.c. of the Wijs solution is introduced by means of a pipette. 
 The bottle is allowed to stand in water having a temperature of 
 22 to 23 for exactly 1 hour. 
 
 *Ber., 31, 750. J. Soc. Chem. Ind., 17, 609. Jour. Am. Chem. Soc., 29, 
 1221. 
 
ACETYL VALUE. 447 
 
 Pure shellac will scarcely alter the color of the Wijs solution. 
 If rosin is present, a reddish-brown color will be produced, the 
 depth of which depends on the amount of rosin present. After 
 standing 1 hour, 10 c.c. of a 10 per cent potassium-iodide solution 
 is added. The excess of iodine is immediately determined by 
 means of a tenth-normal solution of sodium thiosulphate. When 
 the iodine color is nearly gone, a little starch is added and the 
 titration completed. Any color returning after a half -minute 
 is disregarded. 
 
 On account of the large coefficient of expansion of the Wijs 
 solution, a blank determination should be made with that of the 
 shellac. The determination is carried out in exactly the same 
 manner, with the exception of the addition of the shellac. 
 
 680. Acetyl Value. This permits the identification of castor- 
 oil, grapeseed-oil, and of oxidized fish oils in mixtures. Besides 
 this, many fish oils show a comparatively high acetyl value. 
 The test is founded on the action of acetic anhydride upon alco- 
 holic hydroxyls as they exist in the oxyacids, as is shown in the 
 following reaction: 
 
 C 17 H 32 (OH)COOH + (C 2 H 3 0) 2 = C 17 H 32 (OC 2 H 3 0)COOH + HC 2 H 3 2 . 
 
 Upon saponification of the acetylated fatty acid thus produced, 
 the hydroxyl is again replaced and the acetyl group liberated as 
 an acetate. The amount of acetic acid liberated may be deter- 
 mined, giving the acetyl value. Benedikt first proposed this 
 method, operating on the fatty acids, but the process was modified 
 by Lewkowitsch, who works on the oils or fats directly, giving 
 more exactly the true content of hydroxy acids. 
 
 68 1. Determination of Acetyl Value. The method adopted by 
 the Association of Official Agricultural Chemists is carried out as 
 follows: Boil a portion of the oil or fat with an equal volume of 
 acetic anhydride for two hours and pour the mixture into a large 
 beaker containing 500 c.c. of water, and boil for half an hour. To 
 prevent bumping, a slow current of carbon dioxide is passed into 
 the liquid through a finely drawn-out tube, reaching nearly to the 
 bottom. Allow the mixture to separate into two layers, siphon 
 off the water, and boil the oily layer with fresh water until it is 
 no longer acid to litmus-paper. The acetylated fat is then sepa- 
 rated from the water, filtered, and dried in a drying-oven. 
 
448 ANALYSIS OF FATS AND OILS. 
 
 Weigh from 2 to 4 grams of the acet-ylated fats into a flask and 
 saponify with alcoholic potash, as in the determination of the 
 saponification number. If the distillation process is to be adopted 
 it is not necessary to work with a standardized alcoholic-potash 
 solution. In case the filtration method is used, which will be 
 found much shorter, it is necessary that the alcoholic potash be 
 measured exactly. In either case evaporate the alcohol after 
 saponification and dissolve the soap in water. Now two pro- 
 cedures are possible, either distillation or filtration. 
 
 (a) DISTILLATION PROCESS. Acidify with dilute sulphuric acid 
 (1-10) and distil the liquid, as in the Reichert test. As several 
 hundred cubic centimeters must be distilled, either a current of 
 steam is run through or portions of water are added from time 
 to time. From 500 to 700 c.c. of distillate will be found suffi- 
 cient. Filter the distillates to remove any insoluble acids carried 
 over by the steam, and titrate the filtrate with decinormal potas- 
 sium hydroxide, using phenolphthalein as the indicator. Multiply 
 the number of cubic centimeters of alkali employed by 5.61, and 
 divide by the weight of substance taken. This gives the acetyl 
 value, which is the number of milligrams of KOH required to 
 neutralize the acetic acid liberated from one gram of the acetyl- 
 ated fatty acid. 
 
 (b) FILTRATION PROCESS. Add to the soap solution a quantity 
 of the standard sulphuric acid exactly corresponding to the 
 amount of alcoholic potash added, warm gently, and the free fatty 
 acids will collect on top. Filter off the liberated fatty acids, wash 
 with boiling water until the washings are no longer acid, and 
 titrate the filtrate with decinormal potassium hydroxide, using 
 phenolphthalein as the indicator. Calculate the acetyl value as 
 before. 
 
 PHYSICAL TESTS. 
 
 Besides the constants obtained by strictly chemical quantita- 
 tive means, there are a few physical tests that are extremely im- 
 portant and widely used, which it seems advisable should be 
 mentioned under the head of oil-analysis. 
 
 682. Specific Gravity. This may be very accurately obtained 
 by means of a Westphal balance as described on p. 275. 
 
PHYSICAL TESTS. 449 
 
 If the oil is viscous, however, a pycnometer must be used as 
 described on p. 274. 
 
 683. Index of Refraction. This is found by means of an 
 instrument called a refractometer, of which there are several on 
 the market. The Zeiss butyro-refractometer, one of the simplest 
 of these, is also one of the most efficient. The oil or fat is intro- 
 duced as a thin film between two halves of a Nicol's prism, which 
 are enclosed in a hollow metal jacket through which water is 
 allowed to flow, maintaining a constant temperature. The degree 
 of diffraction is indicated on an arbitrary scale by a shadow which 
 is observed through a telescope attached to the prism, and this 
 reading, by comparison with a table, made especially for the 
 instrument, is converted into the refractive index. 
 
 The Abbe refractometer is better suited for general use, being 
 constructed with prisms and water-jackets like the Zeiss butyro- 
 refractometer, but instead of an arbitrary scale fixed in the tele- 
 scope, the border of a shadow, visible through the latter, is 
 brought to the intersection of cross-hairs by moving an arm at the 
 side of the instrument, and the refractive index may then be read 
 directly on a scale along which the movable arm slides. 
 
 684. Maumene Number and Specific Temperature Reaction. 
 It is found that by mixing various oils with concentrated sul- 
 phuric acid a reaction ensues, causing a rise of temperature in the 
 mixture, which for the same oil under the same conditions and 
 with the same concentration of acid is a constant. In order to 
 bring more uniformity into the results obtained by different 
 operators working under different conditions and using different 
 strengths of acid, the old Maumene test has been modified with 
 considerable success. Maumene mixed 50 grams of oil with 10 c.c. 
 of concentrated acid in a small beaker sunk in a larger vessel 
 filled with cotton or some other insulating material. Before mix- 
 ing, both oil and acid were at the same temperature, as close 
 to 20 as possible. This temperature was noted, and, after com- 
 bining, the highest point registered by the thermometer used for 
 stirring the mass was observed, which, minus the initial tempera- 
 ture, gave the Maumene number. The errors introduced by 
 different degrees of insulation and strength of acid have been 
 eliminated by conducting another test exactly as the one just 
 
450 ANALYSIS OF FATS AND OILS. 
 
 described, using the same apparatus and the same acid, but sub- 
 stituting 50 c.c. of pure water for the oil. Then, by the following 
 formula : 
 
 100A 
 
 & B 
 
 in which S = specific temperature number, 
 ^4.= Maumene number, 
 
 B--= rise of temperature obtained with pure water under 
 the same conditions as those used in the Maumene 
 test, 
 
 results are obtained which compare admirably, provided the acid 
 used is fairly concentrated between 95 and 99 per cent.* 
 
 685. The determination is carried out as follows: f A beaker, 5 
 inches by H inches, is placed inside another, 6 inches by 3 inches, 
 and a wet mixture of asbestos and plaster of Paris tightly packed 
 around the inner one. This, when dried, makes a hard, solid 
 packing which radiates heat very slowly. 
 
 Remove the inner beaker, weigh into it 50 grams of fat or oil, 
 and note the temperature carefully. Then from a pipette which will 
 deliver it in approximately one minute add 10 c.c. of 96-98% 
 sulphuric acid, which is at the same temperature as the oil. While 
 the acid is being introduced, stir the oil and acid with an accurate 
 thermometer. Then hold the thermometer-bulb carefully in the 
 centre of the mixture, and when the mercury reaches the highest 
 point note the reading. It is easy to determine this point, as the 
 column of mercury remains stationary for some time. It is neces- 
 sary to take care not to read the temperature too soon, as some 
 oils take considerable time to reach their maximum point. The 
 difference between the initial reading and the final reading, ex- 
 pressed in degrees centigrade, gives the Maumene number. 
 
 In order to get the specific temperature number the operation 
 is repeated with the same apparatus and the same acid, using 
 50 c.c. of pure water in place of the oil. It is advisable in this 
 case to cover the beaker with a piece of cardboard or heavy paper 
 
 * References: Journal Soc. Chem. Ind., 1191, 10, 233; Jour. Amer. Chem. 
 Soc., 1902, 24, 266. 
 
 t Mien, Com. Org. Anal., 3d Ed., Vol. II, Pt. 1, p. 76. 
 
PHYSICAL TESTS. 451 
 
 twice the diameter of the latter, the thermometer being thrust 
 through a small hole in the centre. This interferes with the 
 evaporation of the water and the consequent fall of the tempera- 
 ture of the mixture, but if employed should be used with the oils 
 also. 
 
 686. Melting-point of Fatty Acids. This test is much used in 
 corroborating results obtained by other methods when the ques- 
 tion of identity or purity of an oil or fat arises. The sample is 
 saponified by boiling with an excess of alcoholic potash, and the 
 fatty acids liberated, as in the Hehner method, with dilute sul- 
 phuric acid and washed with warm water until the washings show 
 a neutral reaction. The melted fatty acids are then drawn into a 
 very thin-walled capillary tube 1 or 2 inches long according to the 
 length of the bulb of the thermometer used. The capillary is 
 sealed about one-fourth inch above the filled portion by holding 
 in the outer mantle of a Bunsen flame and drawing out when the 
 glass softens. The tube thus prepared, after cooling in a refrig- 
 erator, is fastened to the bulb of the thermometer by an elastic 
 band and immersed in a glycerine bath, which, to secure a slower 
 and more even rise, may in turn be surrounded by a second bath 
 of the same material which is directly heated by a Bunsen flame. 
 By regulating the flame the rise of temperature may be readily 
 controlled. A slow rotary motion is maintained by the ther- 
 mometer with its capillary until the fatty acids become trans-- 
 parent, when the temperature is noted. It is almost always 
 necessary to conduct three tests; the first to get the approximate 
 temperature, after which the heating may be done very slowly as 
 the melting-point approaches, and the average of the two last 
 determinations taken as the true point. 
 
 687. Detection of Phytosterol and Cholesterol. Fats and oils 
 of vegetable origin may be detected in those derived from animals 
 by the presence of phytosterol and, vice versa, animal products 
 in vegetable oils and fats by the presence of cholesterol. Boil 
 50 grams of the fat or oil in a flask having a reflux condenser, with 
 75 c.c. of 95% alcohol, for five minutes, and separate the alcoholic 
 solution. Repeat with another portion of alcohol and separate. 
 Mix the alcoholic solution with 15 c.c. of 30% sodium hydroxide, 
 
452 ANALYSIS OF FATS AND OILS. 
 
 and boil in a flask with an air condenser until one-fourth of the 
 alcohol is evaporated. Evaporate nearly to dryness in a porce- 
 lain dish and extract the residue with successive portions of ether. 
 Evaporate the ethereal solution to dryness, take up with a little 
 ether, filter, again evaporate, dissolve in hot 95% alcohol, and 
 allow to crystallize. Examine a few of the crystals under a micro- 
 scope. Phytosterol thus prepared appears as tufts of needles, 
 cholesterol as thin rhombic tables. The melting-point of phy- 
 tosterol is 130-137.5, that of cholesterol 146. 
 
 The melting-point is better determined, however, on the acetates 
 of these alcohols. The crystals obtained as described are heated 
 in a dish, with 2-3 c.c. of acetic anhydride, over a small flame until 
 it boils, the dish being covered with a watch-glass. The watch- 
 glass is then removed and the excess of acetic anhydride evapo- 
 rated off on the water-bath. The contents of the dish are next 
 heated with the smallest quantity of absolute alcohol, and in 
 order to prevent immediate solidification or crystallization, a few 
 c.c. of alcohol are added, and the mass allowed to crystallize by 
 spontaneous evaporation of the alcohol, of which one-half or a 
 third is allowed to volatilize. The crystals are filtered off through 
 a small filter and washed with a little 95% alcohol. They are 
 then dissolved in 5-10 c.c. of hot absolute alcohol and again allowed 
 to crystallize. These crystals are filtered off, pressed dry, and 
 their melting-point determined. Cholesteryl acetate melts at 
 114.3 to 114.8 (corr.), and phytosteryl acetate at 125.6 to 137 
 (corr.). In doubtful cases the crystals should be redissolved in 
 hot absolute alcohol and recrystallized at least three times before 
 obtaining the melting-point. 
 
 688. Flash Point. This is the temperature at which the vapors 
 given off by the oil may be ignited, producing a flash when brought 
 in contact with a small flame. This test is made in order to deter- 
 mine the temperature at which the use of the oil involves the 
 danger of fires or explosions. The flash point varies with the 
 form of apparatus and the method of making the test, so that 
 the conditions must be rigidly fixed in order to obtain constant 
 results. Various forms of apparatus have been devised, all of 
 which embody a few essential features, such as a cup to hold the 
 oil, some device for gradually raising the temperature of the oil, 
 
PHYSICAL TESTS. 453 
 
 a thermometer to indicate the temperature, and a small flame 
 to be used for igniting the vapor. A cover containing two open- 
 ings, one for the thermometer and another at which to apply 
 the test-flame, is sometimes provided. Such an instrument is 
 called a closed tester, while an instrument without a cover is 
 called an open tester. The open testers give higher readings 
 than the closed testers. More accurate determinations can be 
 made with the closed testers. 
 
 689. Oil Testers. A simple tester may be made from ordi- 
 nary laboratory apparatus. A shallow porcelain crucible of about 
 100 c.c. capacity will serve as the cup to contain the oil. It may 
 be heated on a sand-bath, or it may be placed in a circular opening 
 in a piece of asbestos board and heated over a Bunsen-burner 
 flame. A thermometer is suspended in the oil, which is then slowly 
 heated while a small test-flame is passed over the surface of the 
 oil. The test-flame may be produced by drawing out a piece 
 of glass tubirg to a fine tube and attaching it to the gas-supply 
 by means of rubber tubing. The temperature is noted at the 
 moment a flash is produced. After the oil has cooled some- 
 what, the test is repeated, the oil being heated more slowly as 
 the temperature approaches the point at which the oil flashes. 
 By placing a metallic or glass cover containing two openings 
 over the crucible, it can be used as a closed tester. 
 
 Tagliabue's tester is constructed entirely of copper and con- 
 tains two cups. The large one serves as a receptacle for a non- 
 volatile oil in which the smaller cup containing the oil to be 
 tested is immersed. The heat is applied to the larger cup. By 
 this device the temperature of the oil in the inner cup is raised 
 more uniformly. In a similar form of tester known as the " Say- 
 bolt," the test-flame is replaced by a spark of electricity produced 
 by an induction-coil. 
 
 690. Viscosity is that property of a liquid which determines 
 the relative rate of flow under fixed conditions. It is usually 
 determined by noting the time required by a definite volume of 
 oil at a definite temperature to flow through a small aperture. 
 Water is usually taken as the standard. The simplest form of 
 apparatus is that used in the Pennsylvania and other railroad 
 laboratories. A 100-c.c. pipette, having a bulb 1.5 to 1.75 inches 
 
454 ANALYSIS OF FATS AND OILS. 
 
 in diameter and about 4J inches long, is graduated so that the 
 100-c.c. mark is just at the bottom of the bulb. The opening of 
 the pipette is then made of such a size that 100 c.c. of water 
 at 100 F. will run out of the pipette in 34 seconds. After dry- 
 ing the pipette, the oil to be tested is warmed to 100 and 100 c.c. 
 is drawn up into the pipette. The time required for the oil to 
 flow out is noted. This gives the viscosity of the oil. It is also 
 becoming customary to take the ratio between the number of 
 seconds required by the water and the oil as the viscosity. 
 
 A considerable number of instruments have been devised for 
 this determination, Engler's, Tagliabue's, Redwood's, and Gibbs' 
 being standard instruments. All of these instruments contain 
 a cup having a small opening at the bottom, through which the 
 oil is allowed to flow. A jacket is generally provided which can 
 be filled with water or oil and heated, thus maintaining the oil 
 at a constant temperature. The oil is delivered into a flask of 
 known capacity. 
 
 691. Oil Analysis. In the examination of an unknown oil 
 or fat, the first test to be made is the determination as to whether 
 the oil is of mineral or organic origin, or a mixture of these two 
 classes of oils. For this purpose advantage is taken of the fact 
 that organic fats and oils can be saponified, while mineral oils 
 and waxes cannot be converted into a soap, but can be extracted 
 with ether, while the soap produced from the organic fats and 
 oils is insoluble in ether. On evaporating off the ether, the mineral 
 oil can be recovered in pure condition. On acidifying the water 
 solution of the soap, the fatty acids of the vegetable or animal 
 oils separate out as an oily layer. As the fatty acids represent 
 about 95% of most animal or vegetable oils, the weight of these 
 acids can be used to calculate the amount of these constituents. 
 A determination of the chemical and physical constants of the 
 fatty acids can be made for the identification of the oils present. 
 If mineral oils are absent, these determinations are made on the 
 original oil. 
 
 692. The Saponification is carried out in the same manner as 
 the determination of the Kottstorfer value (page 439). Stand- 
 ard solutions of the caustic potash and acid are not necessary. 
 
PHYSICAL TESTS. 455 
 
 60 grams of caustic potash are dissolved in one liter of 95% alco- 
 hol. 5 to 10 grams of the fat or oil to be tested are weighed out 
 into an Erlenmeyer flask and 50 to 75 c.c. of the alcoholic caustic 
 potash added. The flask is heated on the water-bath with a 
 return condenser until the oil is saponified. 75 c.c. of water is 
 then added and the heating continued without the condenser 
 until the alcohol has been expelled. The soap solution is trans- 
 ferred to a separatory funnel and 75 c.c. of ether added. The 
 funnel is corked and the contents thoroughly shaken and then 
 allowed to stand quietly until the upper ether solution of the 
 unsaponified or mineral fat is completely separated from the 
 lower water solution of the soap formed from the animal or vege- 
 table fat. The lower layer is drawn off into a beaker, the stop- 
 cock being closed when the last drop of the water solution has 
 passed out. A little ether may be allowed to pass out with the 
 water solution, but none of the latter must re allowed to remain 
 with the ether. The ethereal solution is drawn off into a small 
 weighed flask and the water solution returned to the funnel and 
 the extraction with ether repeated. Three or four extractions 
 with ether are generally sufficient to remove all the unsaponified 
 fat or oil. The last portions of ether should leave no residue 
 when allowed to evaporate on a watch-crystal. The flask contain- 
 ing the ethereal solution is connected with an inclined condenser 
 and the ether distilled off over a water-bath, or, still better, over 
 an electric-light bulb. As the ether vapor is both inflammable 
 and explosive when mixed with air, a flame should not be brought 
 near vessels containing ether. The residue remaining after dis- 
 tilling off the ether is dried on the water-bath, or at a lower tem- 
 perature if volatile oils are present and weighed. This gives the 
 weight of mineral oil present. 
 
 693. Determination of the Fatty Acids. The aqueous solution 
 of the soap is acidified with hydrochloric acid and heated on the 
 water-bath until the fatty acids form a clear layer on the water 
 solution. The oily layer is filtered off on a wet filter-paper which 
 has been dried for one hour in the steam-oven together with a 
 small beaker and weighed. The oil is washed with hot water 
 without allowing the funnel to become more than half empty. 
 When the chlorides have been completely washed out, the paper 
 
456 ANALYSIS OF FATS AND OILS. 
 
 containing the fatty acids is transferred to the beaker which was 
 weighed with the paper, and the whole placed in the steam-oven 
 to dry. After one hour it is weighed and again dried and weighed 
 until the weight is constant. In order to determine what organic 
 oils are present in the sample being analyzed, a determination 
 of the various constants of the fatty acids is made. 
 
 694. Identification of Oils. For the identification of an un- 
 known organic oil, the most useful constant is the iodine value, 
 on account of the great difference in this value for various oils. 
 When two or more oils have the same iodine value as given in the 
 table, page 457, other constants must be determined. An iodine 
 value of 103, for example, might indicate cottonseed, mustard, 
 rape, or sesame oil. A saponification value of 193 would show 
 that the oil must be cottonseed or sesame oil, while an acetyl value 
 of 16.6 would indicate that the oil must be cottonseed. This 
 conclusion could be confirmed by qualitative color tests, of which 
 some very excellent ones are known. 
 
CONSTANTS. 
 
 457 
 
 IODINE VALUE OF OILS. 
 
 
 By Hiibl's 
 Solution. 
 
 By Wijs's 
 Solution. 
 
 ByHanus 
 Solution 
 
 Authority. 
 
 Butter 
 
 35.3 
 
 36.2 
 
 35.3 
 
 Tolman and Munson 
 
 a 
 
 34.8 
 
 35.9 
 
 35.4 
 
 u n tt 
 
 u 
 
 30 7 
 
 
 30 6 
 
 Hanus 
 
 It 
 
 19 5 to 38 9 
 
 
 
 
 Cocoanut 
 
 8 93 
 
 9 05 
 
 8 6C 
 
 Tolman and Munson 
 
 
 
 9 03 
 
 
 9 0? 
 
 Hanus 
 
 (i 
 
 8 to 9 5 
 
 
 
 
 Corn 
 
 119 
 
 123 2 
 
 120 2 
 
 Tolman and Munson 
 
 (i 
 
 119 
 
 122 2 
 
 119 6 
 
 (t tt tt 
 
 ii 
 
 123 3 
 
 129 2 
 
 126 
 
 (i tt tt 
 
 ii 
 
 124.8 
 
 128.4 
 
 
 Wijs 
 
 u 
 
 111 2 to 122 9 
 
 
 
 
 Cottonseed-oil. . . . 
 ii 
 
 ii 
 u 
 
 Konut 
 
 103.8 
 106.2 
 104.8 
 100. 9 to 116. 9 
 6.09 
 
 105.3 
 107.3 
 106.2 
 
 6 43 
 
 105.2 
 107.8 
 106.7 
 
 6 40 
 
 Tolman and Munson 
 tt ti 
 
 tt tt tt 
 tt tt u 
 
 Lard-oil. 
 
 73 7 
 
 74 5 
 
 73 9 
 
 (t tt tt 
 
 i < 
 
 69 3 
 
 70 5 
 
 69 8 
 
 tt tt it 
 
 ii 
 
 56 4 
 
 
 56 9 
 
 TTami 5 ? 
 
 ii 
 
 50 4 to 77 28 
 
 
 
 
 Linseed-oil 
 
 (Decreases to 22 
 169.8 
 
 with age") 
 186 5 
 
 184 5 
 
 Tolman and Munson 
 
 ii 
 
 179 5 
 
 188 7 
 
 183 7 
 
 (t tt tt 
 
 ii 
 
 174 8 
 
 177 3 
 
 m* 
 
 Hunt 
 
 <i 
 
 180.9 
 
 182 1 
 
 
 Wijs 
 
 iC 
 
 170 2 
 
 
 mn 
 
 
 II 
 
 171 to 188 
 
 
 
 
 Magnolia-oil. . 
 
 81 7 
 
 79 4 
 
 78 Q 
 
 
 u 
 
 76 1 
 
 75 6 
 
 74 
 
 tt tt tt 
 
 Mustard-oil. . . 
 
 110 4 
 
 118 5 
 
 115^ 
 
 tt tt tt 
 
 (i 
 
 113 
 
 118 2 
 
 11fi 8 
 
 tt tt tt 
 
 a 
 
 98 4 
 
 104 3 
 
 103 S 
 
 tt tt tt 
 
 tt 
 
 103 5 
 
 112 5 
 
 m9 
 
 tf tt tt 
 
 (( 
 
 106 4 
 
 117 3 
 
 114 8 
 
 tt tt tt 
 
 a 
 
 92.1 to 106 57 
 
 
 
 
 Oleo-oil 
 
 42 6 
 
 43 ^ 
 
 43 3 
 
 tt tt tt 
 
 Oleomargarine . . . 
 u 
 
 53.6 
 
 52.8 
 
 53.5 
 53.7 
 
 52.3 
 52.2 
 
 tt tt tt 
 
 t t 
 
458 
 
 ANALYSIS OF FATS AND OILS. 
 
 IODINE VALUE OF OILS * Continued. 
 
 
 By Hiibl's 
 Solution. 
 
 By Wijs's 
 Solution 
 
 By Hanus' 
 Solution. 
 
 Authority. 
 
 Oleomargarine 
 
 52.5 
 
 52 9 
 
 52 
 
 Tolman and Munson 
 
 
 66.3 
 
 66 
 
 64 8 
 
 ii u u 
 
 Olive-oil 
 
 79.2 
 
 79 9 
 
 80 6 
 
 it ti n 
 
 
 80 5 
 
 80 9 
 
 81 6 
 
 it n it 
 
 it 
 
 84 8 
 
 86 7 
 
 86 5 
 
 ti tt tt 
 
 ti 
 
 89 8 
 
 91 4 
 
 90 
 
 ii tt tt 
 
 " av. 36 samples . 
 
 84.1 
 77.28 to 91 5 
 
 85.3 
 
 84.7 
 
 
 Peanut- or Arachis-oil 
 a it ii 
 
 96.3 
 94.5 
 
 88.3 
 
 99.0 
 95.2 
 
 97.4 
 94.1 
 
 88.4 
 
 Tolman and Munson 
 it u it 
 
 Hanus 
 
 ti ti ii 
 
 87.2 
 
 87.2 
 
 
 Wijs 
 
 ti u it 
 ii it n 
 
 Poppv. . 
 
 91.8 
 85. 6 to 105 
 133.4 
 
 93.4 
 135 2 
 
 91.6 
 132 9 
 
 Hanus 
 it tt tt 
 
 a 
 
 134.9 
 
 139 1 
 
 138 4 
 
 u tt tt 
 
 tt 
 
 122.4 
 
 
 122 6 
 
 Hanus 
 
 it 
 
 134.6 
 
 
 135 2 
 
 it 
 
 u 
 
 119 6 
 
 119 6 
 
 
 Wijs 
 
 it 
 
 132.6 to 143 3 
 
 
 
 
 Rape 
 
 101 3 
 
 105 7 
 
 105 2 
 
 Tolman and Munson 
 
 tt 
 
 100 2 
 
 104 1 
 
 102 8 
 
 it it it 
 
 u 
 
 99 3 
 
 
 98 8 
 
 Hanus 
 
 tt 
 
 102.9 
 
 103.3 
 
 
 Wijs 
 
 ii 
 
 103.0 
 
 102.1 
 
 101.9 
 
 Hunt 
 
 tt 
 
 97 to 106 
 
 
 
 
 Sesame 
 
 106.4 
 
 107.0 
 
 106 5 
 
 Tolman and Munson 
 
 it 
 
 110.3 
 
 111.7 
 
 
 Wijs 
 
 ii 
 
 107.1 
 
 
 107.5 
 
 Hanus 
 
 it 
 
 102.7 to 112 
 
 
 
 
 Sunflower 
 
 106.4 
 
 109 2 
 
 107.7 
 
 Tolman and MunsoD 
 
 
 117.8 
 
 119.0 
 
 
 Wijs 
 
 it 
 
 119.7 to 135 
 
 
 
 
 
 
 
 
 
 * For other iodine values and other constants, see Van Nostrand's Chemical 
 Annual. 
 
CONSTANTS. 
 
 459 
 
 REICHERT-MEISSL-VALUE. 
 
 c.c. N/10 KOH. 
 
 c.c. N/10 KOH. 
 
 Butter 19.8-33.5 
 
 Legal minimum in Great 
 Britain, Germany, and 
 
 France 24.0 
 
 Legal minimum in Sweden. . . 23.0 
 
 " Italy 20.0 
 
 Rabbit-fat.. 5.6 
 
 Cocoanut-oil * 7.0-7.8 
 
 Margarine 2.6 
 
 Oleomargarine 0.8.-0.9 
 
 Palm-oil 5.0 
 
 Wool-fat , 8.0 
 
 Beef-marrow 2.0 
 
 Turkey-fat 5.6 
 
 POLENSKE VALUE OF BUTTER-FAT. 
 
 VALUES FOUND BY E. POLENSKE. VALUES FOUND BY M. FRITZSCHE 
 
 Reichert- Normal Maximum ON DUTCH BUTTER. 
 
 Meissl Value. 
 c.c .C/loKOH. 
 
 20-21 
 
 Pol. Value. 
 C.c. N/10 KOH. 
 
 1.3-1.7 
 
 Limit, 
 c.c. > /1 KOH. 
 
 2.1 
 
 V^il 
 
 Reichert- 
 Meissl Value, 
 c.c. N/10 KOH. 
 
 1 .LHJ J. \_/-LJ. 13 \J J 
 
 Normal 
 Pol. Value. 
 c.c. N/1U KOH. 
 
 ..EJXV. 
 
 Maximum 
 Limit, 
 c.c. N/10 KOH. 
 
 21-22 
 
 1.4-1.8 
 
 2.2 
 
 30-31 
 
 2.1-3.2 
 
 3.2 
 
 22-23 
 
 1.5-1.9 
 
 2.3 
 
 31-32 
 
 1.9-3.2 
 
 3.2 
 
 24-45 
 
 1.7-1.8 
 
 2.3 
 
 32-33 
 
 2.1-3.4 
 
 3.4 
 
 25-26 
 
 1.8-1.9 
 
 2.4 
 
 33-34 
 
 2.5-2.9 
 
 3.4 
 
 26-27 
 
 1.9-2.0 
 
 2.5 
 
 
 
 
 27-28 
 
 2.0-2.2 
 
 2.7 
 
 
 
 
 28-29 
 
 2.2-2.5 
 
 3.0 
 
 
 
 
 29-30 
 
 2.5-3.0 
 
 3.5 
 
 
 
 
 KOTTSTORFER SAPONIFICATION VALUE. 
 
 Mg. KOH per Gram. Mg. KOH per Gram. 
 
 Butter 216-233 
 
 Oleomargarine 192-200 
 
 Cocoanut-oil 246-270 
 
 Lard 193-200 
 
 Olive-oil 187-195 
 
 Niger-oil 189-191 
 
 Linseed-oil 190-195 
 
 Cottonseed-oil 191-196 
 
 Castor 176-186 
 
 Cod-liver oil . . 171-189 
 
 Sperm 123-147 
 
 Rape 167-1 7f 
 
 Cocoa-butter 192-202 
 
 Sesame 187-193 
 
 Wool-fat 98-127 
 
 Almond 187.9-195.5 
 
 Peanut 185.6-197 
 
 Palm 202 
 
 Beeswax 90-100 
 
 Mustard-oil. . . 170-180 
 
 HEHNER VALUE. 
 
 AVERAGE PERCENTAGE OF INSOLUBLE ACIDS. 
 
 Butter 87.5 
 
 Cocoanut-oil 86.43 
 
 Palm-nut oil 91.1-95.6 
 
 Lard 96.15 
 
 Oleomargarine 95 . 56 
 
 Olive 95.43 
 
 Rape 95.10 
 
 Cod-liver oil 95.46-97.05 
 
 Almond-oil 96. 6 
 
 Cocoa-butter 94.59 
 
 Tallow 95.50 
 
 Cottonseed 95.75 
 
 Peanut-oil 95 .00 
 
 Sesame. . .95.48 
 
 * De Negri and Fabris have found as low a number as 3. 
 
460 
 
 ANALYSIS OF FATS AND OILS. 
 
 MAUMENE NUMBER OF OILS. 
 
 
 
 Authority. 
 
 Linseed , 
 
 103 
 
 Maumene 
 
 " 
 
 , 104-124 
 
 Baynes 
 
 tt 
 
 104-111 
 
 Allen 
 
 " 
 
 122-126 
 
 De Nigri and Fabris 
 
 Niger 
 
 82 
 
 Baynes 
 
 tt 
 
 81 
 
 Allen 
 
 Cottonseed (raw) 
 
 84 
 
 Baynes 
 
 
 
 61 
 
 Dobb 
 
 tt it 
 
 70 
 
 Archbutt 
 
 tt (t 
 
 67-69 
 
 De Nigri and Fabris 
 
 Cottonseed (refined) 
 
 77 
 
 Baynes 
 
 K ft 
 
 75-76 
 
 Archbutt 
 
 it tt 
 
 74-75 
 
 Allen 
 
 Sesame 
 
 68 
 
 Maumene 
 
 tt 
 
 65 
 
 Archbutt 
 
 u 
 
 63-64 
 
 De Nigri and Fabris 
 
 Colza 
 
 57-58 
 
 Maumene 
 
 u 
 
 54-56 
 
 Dobb 
 
 tt 
 
 55-64 
 
 Archbutt 
 
 tt 
 
 51-60 
 
 Allen 
 
 tt 
 
 49-51 
 
 De Nigri and Fabris 
 
 Almond 
 
 52-54 
 
 Maumene 
 
 " 
 
 51-53 
 
 De Nigri and Fabris 
 
 Peanut 
 
 67 
 
 Maumene 
 
 u 
 
 47-60 
 
 Archbutt 
 
 " 
 
 45.5-51 
 
 De Nigri and Fabris 
 
 Olive 
 
 42 
 
 Maumene 
 
 tt 
 
 40 
 
 Baynes 
 
 tt 
 
 39-43 
 
 Dobb 
 
 tt 
 
 41-45 
 
 Archbutt 
 
 it 
 
 41-43 
 
 Allen 
 
 tt 
 
 32-37 
 
 De Nigri and Fabris 
 
 Castor 
 
 47 
 
 Maumene 
 
 " 
 
 46 
 
 Archbutt 
 
 M 
 
 46-47 
 
 De Nigri and Fabris 
 
 
 36 
 
 Olsen and Benoit 
 
CONSTANTS. 
 
 461 
 
 SPECIFIC TEMPERATURE NUMBER OF OILS.* 
 
 Authority. 
 
 Linseed 320-349 Thomson and Ballantyne 
 
 " 313 Jenkins 
 
 Cottonseed 163-170 Thomson and Ballantyne 
 
 " 174.3 Tolman and Munson 
 
 " butter-oil 172.9 " " " 
 
 " summer white.... 191.1 " " " 
 
 " cooking-oil 192 . 4 " " " 
 
 blown 164 Jenkins 
 
 Colza 125-144 Thomson and Ballantyne 
 
 " 130 Jenkins 
 
 Rape 135.6-152.5 Tolman and Munson 
 
 Peanut 105-137 Thomson and Ballantyna 
 
 " 129.1-135.3 Tolman and Munson 
 
 Olive 89-94 Thomson and Ballantyne 
 
 " 94 Jenkins 
 
 " 94.5-109.7 Tolman and Munson 
 
 Boiled linseed 248 Jenkins 
 
 Castor 89 Thomson and Ballantyne 
 
 Sperm 100 " " " 
 
 Gocoanut 44 Tolman and Munson 
 
 Lard-oil 96.0 Olsen and Benoit 
 
 Mustard (black) 189.4 Tolman and Munson 
 
 (brown) 165.4 " << " 
 
 " (black) 169.3 " ?, " 
 
 (yellow) 130.9 " " " 
 
 Almond 117.6 " " " 
 
 Sunflower 166.7 " " " 
 
 Maize 190.2 " "< " 
 
 ACETYL VALUE OF OILS. 
 
 Linseed 8.5 Whale-oil (old) 23.1 
 
 Sesame 11.5 " (recent) 13.2 
 
 Cottonseed 16.6 (various sam- 
 
 Colza 6.3 pies) 11.6-17.2 
 
 Almond 5.8 Grape-seed 144 . 5 
 
 Peanut 3.4 Cod-liver oil 41 .1-50.6 
 
 Olive 10.6 Seal-oil 33.0-33.9 
 
 Castor 153 .4-156 
 
 * This modification of Maumene's test being comparatively new, there are 
 less data of results to choose from than is the case with the older methods. 
 
CHAPTER XXXI. 
 GAS ANALYSIS. 
 
 695. General Methods of Gas Analysis. Two general methods 
 are in use for the analysis of gases. The gas to be determined 
 may be absorbed in an appropriate liquid and the amount deter- 
 mined by VOLUMETRIC or GRAVIMETRIC METHODS. The determina- 
 tion of carbon dioxide illustrates these methods. This gas may 
 be absorbed in caustic potash solution in a Liebig bulb which is 
 weighed before and after absorption. It may also be absorbed 
 in a measured volume of standard barium hydroxide solution 
 and the excess of alkali titrated with standard acid. Such 
 methods should always be used when the gas to be determined 
 is mixed with a large amount of other gases, as is the case with 
 carbon dioxide in the air. 
 
 If the gas to be determined constitutes a relatively large pro- 
 portion of a gaseous mixture, its amount may be ascertained by 
 taking a MEASURED VOLUME of the gaseous mixture and measuring 
 the residue after the constituent to be determined has been 
 absorbed. Such determinations may be made very rapidly and 
 with a high degree of accuracy, the methods having been well 
 developed and excellent apparatus devised by Bunsen, Hempel, 
 and others. 
 
 ABSORBENTS FOR OXYGEN. 
 
 Two reagents are commonly in use for absorbing oxygen in 
 gas analysis an ALKALINE SOLUTION OF PYROGALLOL and YELLOW 
 
 PHOSPHORUS IN STICKS. 
 
 696. The Pyrogallol, being in solution, may be used in appa- 
 ratus of a great variety of form. The solution must not be used 
 at a temperature less than 15, since at lower temperatures the 
 
 462 
 
OXYGEN. 463 
 
 absorption of oxygen is very slow. The absorption is not com- 
 plete unless a large excess of the reagent is present. When the 
 solution has absorbed a few per cent of the amount of oxygen 
 which it will take up, it leaves, during subsequent use, a larger and 
 larger residue of unabsorbed oxygen in the gas which is in contact 
 with it. A solution which has been exposed to the air for even 
 a short time must therefore not be used. A record should be 
 kept of the amount of oxygen absorbed by a given solution, so 
 that it may be discarded when no longer efficient. The solution 
 is made according to Hempel * by dissolving 5 grams of pyrogallol 
 in 15 c.c. of water and adding 120 grams of caustic potash dis- 
 solved in 80 c.c. of water. The caustic potash purified by alcohol 
 should not be used. 1 c.c. of this solution will absorb completely 
 2 c.c. of oxygen. 
 
 697. Yellow Phosphorus has the advantage that its absorbing 
 power is practically unlimited. It should be protected from 
 the light, which converts it into the red inactive modification. 
 It is fully as efficient an absorbent of oxygen as the solution of 
 pyrogallol. It is kept in a gas pipette under water, which, besides 
 protecting it from the oxygen of the air, serves to dissolve the ox- 
 ides of phosphorus produced during the absorption of the oxygen. 
 As these oxides are solid and have practically no vapor tension 
 it is not necessary to wait for their absorption before reading the 
 volume of the residual gas. The temperature of the phosphorus 
 must not be allowed to fall below 15 or the absorption will be 
 very slow, 20 being a more favorable temperature. 
 
 698. Pure Oxygen at atmospheric pressure is not absorbed by 
 phosphorus. If the oxygen is diluted by admixture with another 
 gas or the pressure reduced by means of an air-pump, action 
 begins when the pressure of the oxygen is 75% of atmospheric 
 pressure. The action is then violent, being accompanied by the 
 evolution of light and heat, so that the phosphorus is melted. 
 A gas containing more than 50% of oxygen should not be brought 
 into contact with the phosphorus. If a gas containing more than 
 this amount of oxygen is to be analyzed, it may be diluted with 
 nitrogen which is prepared from air by absorption of the oxygen 
 
 *Gas Analysis, 1902, p. 149. 
 
404 GAS ANALYSIS. 
 
 by phosphorus. The absorption of oxygen is entirely or partly 
 prevented, according to Schonbein, by the presence of ethylene 
 and other hydrocarbons, ethereal oils, alcohol, and traces of 
 ammonia. 
 
 699. The Phosphorus is Formed into Sticks by melting a suffi- 
 cient amount in a test-tube under a layer of water. The test- 
 tube is most conveniently heated by placing it in warm water. 
 Glass tubes having an internal diameter of 2 to 3 mm. are selected 
 and inserted into the test-tube with the larger end down. It 
 will be found that the diameter of most glass tubing varies so 
 that tubes having a slight difference in diameter at the two ends 
 may easily be found. On placing the finger over the upper end 
 of the glass tube and taking it out it will remain filled with phos- 
 phorus, which may be solidified by holding the tube in cold water. 
 If, on cooling, the stick of phosphorus does not drop out on tap- 
 ping the tube, it may be shoved out with a stiff wire. The operation 
 is more easily performed if a short piece of rubber tubing with a 
 pinch-cock is placed on the upper end of each glass tube. While 
 one tube is cooling in the water others may be filled with phosphorus. 
 
 700. Chromous Chloride. Oxygen may also be quantitatively 
 absorbed by a solution of chromous chloride, which does not absorb 
 hydrogen sulphide nor carbon dioxide. 
 
 701. For the Absorption of Carbon Dioxide a solution of 
 caustic potash is used. It is made by dissolving 1 part of com- 
 mercial caustic potash in 2 parts of water. 1 c.c. of this solution 
 will absorb quantitatively 40 c.c. of carbon dioxide. 
 
 ABSORBENTS FOR CARBON MONOXIDE. 
 
 This gas is absorbed by CUPROUS CHLORIDP: dissolved in HYDRO- 
 CHLORIC ACID or in AMMONIA. As solutions of cuprous chloride 
 rapidly absorb oxygen from the air, the cuprous chloride being 
 converted into cupric chloride, the solution must be protected 
 from the atmosphere. Hydrochloric acid solutions may be kept 
 reduced by the presence of metallic copper, the most convenient 
 form being wire gauze. Any cupric chloride formed in such a 
 solution is slowly reduced to cuprous chloride by the metallic 
 
CARBON MONOXIDE. 465 
 
 copper, which dissolves, the brown solution becoming colorless. 
 If the hydrogen in a mixture of gases is to be absorbed by 
 palladium after the absorption of the carbon monoxide, the 
 ammoniacal solution must be used. 
 
 702. The Absorbing Power of neither solution is large, and 
 the carbon monoxide is held in such loose combination that after 
 a moderate amount has been absorbed the solution gives up a 
 small part of the carbon monoxide again to a gas in which it is 
 absent. As acetylene and ethylene are also absorbed by cuprous 
 chloride solutions, these gases must be removed before determin- 
 ing carbon monoxide. This is accomplished by the absorbents 
 used to determine the " heavy hydrocarbons" or "illuminants." 
 
 703. The Hydrochloric Acid Solution of Cuprous Chloride may 
 be made, according to Winkler,* by adding a mixture of 86 grams 
 of copper oxide and 17 grams of finely divided metallic copper to 
 1036 grams of hydrochloric acid (sp. gr. 1.124), the mixture being 
 slowly introduced and the acid frequently stirred. The copper 
 powder is best prepared by reducing copper oxide in a stream of 
 hydrogen. The solution should be placed in a flask which it very 
 nearly fills. A spiral of copper wire is introduced into the flask, 
 which is closed with a rubber stopper and allowed to stand in 
 the dark until the color has disappeared. 1 c.c. of this solution 
 will absorb 4 c.c. of carbon monoxide. 
 
 The solution may also be made by dissolving 15 grams of 
 commercial cuprous chloride in 120 c.c. of concentrated hydro- 
 chloric acid. Twenty grams of copper drillings are added, and the 
 solution is allowed to stand twenty-four hours. It is diluted with 
 water to 200 c.c. and decanted from the metallic copper for use. 
 
 704. The Ammoniacal Solution of Cuprous Chloride may be 
 prepared by placing 15 grams of commercial cuprous chloride in 
 a 250-c.c. flask and adding 200 c.c. of 7% ammonia. After in- 
 serting the stopper the solution is shaken until the cuprous chlo- 
 ride is dissolved. 
 
 705. The Cuprous Chloride may be prepared by dissolving 
 30 grams of copper sulphate and 15 grams of sodium chloride in 
 100 c.c. of water. Sulphur dioxide is passed into this solution 
 
 * Hempel, Gas Analysis, 1902, p. 202. 
 
466 GAS ANALYSIS. 
 
 until precipitation is complete. The solution is decanted and the 
 cuprous chloride is washed several times by decantation and dis- 
 solved in ammonia or hydrochloric acid as desired. 
 
 ABSORBENTS FOR THE " ILLUMINANTS." 
 
 Ethylene and its homologues, as well as benzene and its homo- 
 logues, which taken together are known as heavy hydrocarbons 
 or illuminants in the analysis of illuminating-gas, are absorbed by 
 
 FUMING SULPHURIC ACID Or BROMINE-WATER. 
 
 706. The Fuming Sulphuric Acid gives off vapors of sulphur 
 dioxide and trioxide, so that after absorption of the hydrocar- 
 bons the gas remaining must be passed over caustic potash to free 
 it from the acid vapors. The fuming sulphuric acid must con- 
 tain a large amount of sulphur trioxide and should on cooling 
 deposit crystals. The commercial acid is seldom strong enough. 
 It may be concentrated by distilling the sulphur trioxide from 
 one portion into a second portion. The distillation should be con- 
 ducted in a tubulated retort, using as a receiver a flask the neck 
 of which fits quite snugly over the stem of the retort. The flask 
 may be placed in a large funnel to the stem of which is attached 
 a rubber tube leading to the sink. A stream of water is allowed 
 tc flow over the flask. If crystals form in the receiver, they may 
 be dissolved at the end of the distillation by gently warming the 
 flask. The distillate should be exposed to the air as little as pos- 
 sible. One c.c. will absorb 8 c.c. of the illuminants. 
 
 707. If Bromine is used as the absorbent, a few cubic centi- 
 meters of liquid bromine are placed in the absorption pipette and 
 distilled water added. On shaking, the water becomes saturated 
 with bromine. When the gas containing heavy hydrocarbons is 
 introduced the bromine fumes above the water disappear. On 
 shaking the liquid more bromine vapors appear, only to be again 
 absorbed until all the heavy hydrocarbon vapors have been re- 
 moved, when the bromine fumes persist and may be removed by 
 passing the gas over caustic potash. According to Cl. Winkler,* 
 
 * Zeit. f. anal. Chem., 28, 269. 
 
HYDROGEN. 467 
 
 the absorption of ethylene by means of bromine is not as complete 
 as by fuming sulphuric acid. 
 
 708. Absolute Alcohol. According to Hempel and Dennis,* 
 benzene and its homologues may be absorbed by absolute alcohol, 
 1 c.c. being sufficient for 100 c.c. of illuminating-gas. The alcohol 
 is placed in a pipette over mercury. The alcohol vapor is after- 
 wards removed by washing the gas with 1 c.c. of water. As these 
 vapors are soluble in caustic potash, they should be absorbed 
 before passing the gas over caustic potash, otherwise the result 
 for carbon dioxide will be high and for heavy hydrocarbons low. 
 
 DETERMINATION OF HYDROGEN. 
 
 No liquid absorbent for hydrogen is known. It is determined 
 by adding a sufficient amount of oxygen to form water with the 
 hydrogen present and igniting the mixture. The volume of the 
 mixed gases being measured before and after combustion, the 
 diminution in volume is noted. If the gas was saturated with 
 water vapor before combustion, two-thirds of this diminution 
 will represent the volume of hydrogen present, since two volumes 
 of hydrogen unite with one volume of oxygen, and the water 
 formed occupies no volume except that necessary to saturate the 
 gases present with water vapor. 
 
 709. Explosion of Hydrogen and Oxygen. Various methods 
 are in use for causing the union of the hydrogen with the oxygen. 
 The method longest in use consists in producing an explosion in 
 the gas mixture by means of an electric spark. The electric cur- 
 rent is led in through platinum wires fused into the glass of which 
 the confining vessel is made. 
 
 The proportion of explosive gases present to non-explosive must 
 for various reasons be carefully regulated. By explosive gases is 
 meant the combined volume of oxygen and hydrogen in the 
 proportion to form water. Any hydrogen or oxygen present in 
 excess constitutes a non-explosive gas, as well as nitrogen, carbon 
 dioxide, etc. If all or nearly all of the gas present were explosive, 
 
 * Journ. fur Gasbel., 1891, p. 414. 
 
468 GAS ANALYSIS. 
 
 the action would be so violent as to shatter the confining vessel. 
 If not violent enough to shatter the vessel, another difficulty 
 would be encountered in that if any nitrogen were present some 
 of it would be oxidized, giving a greater diminution in volume than 
 corresponds to the amount of hydrogen present. On the other 
 hand if the explosion were too weak, the combustion of the hydro- 
 gen would be incomplete. According to Bunsen there should be 
 present for 100 volumes of non-explosive gases at atmospheric 
 pressure not less than 26, nor more than 60 volumes of explosive 
 gases, while the most favorable proportion is 26 to 40 parts o f 
 explosive gas to 100 parts of non-explosive. 
 
 If an insufficient amount of the explosive gas is present, oxy- 
 hydrogen gas produced by the electrolysis of water may be added. 
 As this gas is entirely converted into water on explosion, the 
 exact amount added need not be noted. If the proportion of 
 explosive gas is too great, air or pure oxygen may be added. 
 Whenever a gas poor in hydrogen is to be analyzed, the required 
 amount of oxygen should be added in the form of the pure gas. 
 This may be produced according to Bunsen by blowing a small 
 retort from glass tubing and half filling it with pure powdered 
 potassium chlorate. The oxygen is evolved by heating the bulb 
 with the Bunsen burner, the first portions being discarded because 
 contaminated with air. If the gas to be analyzed contains a 
 large percentage of hydrogen, air should be used to furnish the 
 oxygen. 
 
 710. Slow Combustion of Hydrogen. Because of the large 
 amount of inert gas which must be present to diminish the violence 
 of the explosion, only small amounts of the gas to be analyzed 
 can be operated upon when it is exploded. This difficulty 
 is obviated by the method proposed by Dennis and Hopkins,* by 
 which the combustion is effected by means of a red-hot platinum 
 wire. Pure hydrogen may be burned by this method. It is 
 introduced into the pipette containing the red-hot platinum wire, 
 and a mixture of equal parts of air and oxygen containing more 
 oxygen than sufficient to combine with the hydrogen is intro- 
 duced slowly into the pipette containing the hydrogen. The 
 
 * Jour. Am. Chem. Soc., 21, 398. 
 
HYDROGEN. 469 
 
 hydrogen is burned almost as rapidly as the necessary oxygen 
 is introduced. 
 
 711. Combustion of Hydrogen by Palladium Sponge. The 
 mixture of hydrogen and oxygen may also be burned by passing 
 the gas over palladium sponge or palladium asbestos,* which is 
 prepared as follows: One gram of palladium is converted into 
 chloride, and the solution treated with a few cubic centimeters of 
 a saturated solution of sodium formate, and enough caustic soda 
 to make the solution strongly alkaline. One gram of fine silky 
 asbestos is immersed in the solution, which should be entirely 
 taken up by the asbestos. The material is dried by gently heat- 
 ing on the water-bath. The palladium is deposited on the asbes- 
 tos, to which it adheres firmly. The asbestos is placed in a funnel 
 and thoroughly washed with hot water and again dried. This 
 asbestos may be twisted into threads by moistening with water 
 and inserted into capillary tubes and again dried on the water- 
 bath before being used. 
 
 712. Theory of Combustion of Hydrogen by Palladium Sponge. 
 The palladium is able to absorb large quantities of hydrogen, es- 
 pecially at high temperatures. When air is led over palladium 
 saturated with hydrogen, water is formed, and the combustion is 
 so energetic that the palladium becomes red-hot, and in this con- 
 dition it is superficially oxidized. If hydrogen is led over this 
 palladium, even at the ordinary temperature, it is burned at the 
 expense of the oxygen of the palladous oxide. If the hydrogen 
 has previously been mixed with pure oxygen or air, the palladium 
 which has been heated by the combustion of some of the hydrogen 
 with the oxygen of the palladous oxide is able to absorb more 
 hydrogen, which then can combine with the gaseous oxygen 
 present. When all of the hydrogen present has been burned a 
 film of palladous oxide will be formed on the hot palladium by 
 the excess of oxygen present, when the palladium will be in con- 
 dition for the next combustion, and the oxidation will begin at 
 the ordinary temperature. If the palladium has not been rendered 
 active in this manner, it must be heated to 100, or even to redness, 
 before action begins, the hydrogen being absorbed more readily 
 
 * Cl. Winkler, Technische Gas-Analyse, p. 86. 
 
470 GAS ANALYSIS. 
 
 at high temperatures, and the film of oxide being always formed 
 when the palladium is heated to redness in the presence of oxygen. 
 
 713. Temperature. As methane is also oxidized when a mix- 
 ture of oxygen and hydrogen containing this gas is passed over 
 palladium heated to 200 or higher, it is advisable to heat the 
 palladium only to 100 when methane is present, in order to effect 
 the combustion of the hydrogen alone. The tube containing the 
 palladium is placed in boiling water and the mixture of gases is led 
 through it very slowly, so that the temperature shall not rise to 200. 
 
 714. Determination of Methane by Combustion. As no ab- 
 sorbent for methane is known, this gas must also be determined 
 by combustion with oxygen. The explosion by means of the 
 electric spark and the combustion with the red-hot platinum wire 
 may be used exactly as given for hydrogen. The same precautions 
 must be taken in regard to the proportion of explosive to non- 
 explosive gas. It must be remembered that one volume of methane 
 requires for its combustion two volumes of oxygen. The volume 
 of methane will be equal to one-half of the contraction noted 
 after the explosion, provided all other combustible gases are 
 absent. The volume of methane will also be equal to the volume 
 of carbon dioxide present after the explosion, provided all carbon 
 dioxide and all gases which produce carbon dioxide on combustion 
 are removed before the explosion. 
 
 715. A Mixture of Hydrogen and Methane may readily be 
 analyzed by combustion. The contraction produced is noted, and 
 then the amount of carbon dioxide present is determined. The 
 volume of methane will be exactly equal to the volume of carbon 
 dioxide produced, as 1 molecule of each of these gases contains 
 1 atom of carbon. Twice the volume of methane, or of the carbon 
 dioxide found, is the contraction due to the combustion of the 
 methane. On subtracting this volume from the total contraction, 
 that due to the hydrogen present is obtained, and two-thirds of 
 this volume will give the amount of hydrogen present. By sub- 
 tracting the combined volumes of methane and hydrogen from 
 the original volume of the gas analyzed, the amount of NITROGEN 
 or other gas present may be ascertained. 
 
 716. Determination of Nitrogen. As no convenient method 
 of absorbing nitrogen is known, this gas is generally determined by 
 difference, all other gases having been absorbed or determined. 
 
APPARATUS FOR THE ANALYSIS. 471 
 
 APPARATUS FOR THE ANALYSIS. 
 
 The apparatus used in gas analysis, of which a great variety 
 has been devised, must provide for two essentially different opera- 
 tions, namely, the measuring of the gas and the absorption of 
 individual constituents. The measuring of the gas volumes must 
 be accompanied by observations of the temperature and pressure 
 to which the gas is subjected. 
 
 717. Regulation of Temperature. Almost invariably the gas is 
 allowed to assume the temperature of the atmosphere of the work- 
 room and the temperature is noted from a thermometer which 
 is suspended near the measuring-tube. All sources of artificial 
 heat are removed from the immediate proximity of the gas or 
 from the room entirely. The heat emanating from the body and 
 especially the hands of the operator is sufficient to appreciably 
 change the volume of the gas unless precautions are taken to 
 prevent it. For this reason the glass measuring-tubes are fre- 
 quently water-jacketed, the temperature of water changing much 
 more slowly than that of gases. 
 
 718. Regulation of Pressure. The temperature of the gas 
 being fixed as that of the atmosphere, the volume varies inversely 
 with the pressure. In some forms of apparatus the pressure is 
 kept constant, that of the atmosphere being generally used, the 
 volume only being read, while in other forms both volume and 
 pressure are carefully noted, the latter as well as the former being 
 allowed to vary. Both the temperature and pressure of the 
 atmosphere being quite constant for short intervals of time, gas 
 analyses are frequently carried out in which the volume only of 
 the gas is noted, the results being given in percentage by volume. 
 In carrying out an analysis in which the weight of a gaseous con- 
 situent must be obtained, both temperature and pressure must 
 be noted as well as the volume. 
 
 719. Confining Liquids. Both WATER and MERCURY are used 
 as the confining liquids in gas apparatus. The tension of mercury 
 vapor at ordinary temperatures being very small and all gases 
 being almost absolutely insoluble in this metal, it is almost an ideal 
 confining liquid, and for some operations is indispensable. Because 
 
472 GAS ANALYSIS. 
 
 of the slight solubility of most gases in water, the results obtained 
 when this liquid is used are slightly inaccurate. The errors are 
 considerably reduced, however, by using water which has been 
 saturated with the gas to be analyzed. This is accomplished by 
 shaking the water with a portion of the gas or by passing a stream 
 of it through the water. 
 
 HEMPEL'S GAS APPARATUS. 
 
 Very convenient forms of apparatus have been perfected by 
 Hempel. Water is used as the confining liquid and the measure- 
 ments of volume are taken at atmospheric temperature and 
 pressure. 
 
 720. The Simple Gas Burette consists of two glass tubes set 
 in iron or loaded wooden feet, as shown in Fig. 66. The tubes are 
 bent at right angles at the bottom and drawn out so as to afford 
 convenient attachments for the rubber tube about 120 cm. long 
 which joins them. The measuring-tube A ends in a capillary tube, 
 c, over which a short piece of rubber tubing may be slipped. A 
 Mohr pinch-cock, d, placed on this tube, makes a convenient and 
 perfectly tight stop-cock. The tube is graduated in cubic centi- 
 meters, and is capable of holding 100 c.c. of gas, the lower mark 
 being made a little above the foot. The numbers run both up 
 and down. The second tube B is called the level tube, and serves 
 to hold the confining liquid. 
 
 721. The Modified Winkler Gas Burette is similar in construc- 
 tion, except that a glass stop-cock is placed at both ends of the 
 measuring- tube. The upper stop-cock a is the ordinary two-way 
 stop-cock, while the one at the bottom, 6, is a three-way cock, the 
 third opening being through the key and the tube c. A rubber 
 tube may be attached to this tube and gas admitted to the meas- 
 uring-tube without coming in contact with the water in the level 
 tube. This burette is used when the gas to be analyzed contains 
 a constituent easily soluble in water. The tube is dried out by 
 rinsing it with a little absolute alcohol and ether, the vapor of 
 which is displaced by a stream of air. The three-way stop-cock 
 is turned so that its horizontal opening communicates with the 
 
HEMPEL'S APPARATUS. 
 
 473 
 
 inside of the burette, and by means of a tube attached to this 
 opening the gas to be analyzed is passed through the burette 
 until the air is entirely expelled. The three-way stop-cock is 
 closed and then the upper stop-cock. The reagent for absorbing 
 the soluble gas is then introduced into the burette through a 
 
 FIG. 66. 
 
 FIG. 67 
 
 funnel attached to the horizontal opening c in the three-way 
 stop-cock. After the absorption of this constituent the remainder 
 of the analysis is conducted in the usual manner. Whenever 
 possible the burette with the Mohr pinch-cock should be used in 
 place of the Winkler burette, as glass stop-cocks are not as reliable 
 as rubber tubing and a pinch-cock. 
 
474 
 
 GAS ANALYSIS. 
 
 722. The Absorption Pipette in its simplest form is shown in 
 Fig. 68. It consists of two bulbs, a and b, having capacities of 
 100 c.c. and 150 c.c. respectively. A capillary tube, c, bent into 
 a U-shape is sealed to the bulb 6, and the whole is fastened to 
 the iron stand. This pipette is used for liquids which attack rub- 
 ber and which do not need protection from the air, such as caustic 
 potash and bromine-water. The bulb b is filled completely, 
 while a is left empty, so that when the gas is introduced through 
 the tube c into the bulb b the reagent expelled may pass into 
 the bulb a. 
 
 FIG. 68. 
 
 FIG. 69. 
 
 A modified form of this pipette which is suitable for solid 
 reagents is shown in Fig. 69. After the solid has been intro- 
 duced through i the opening is closed with a solid-rubber stopper, 
 or, still better, with a closed glass tube of suitable size, over which 
 a short piece of rubber tubing has been slipped. Very little rub- 
 ber is then exposed to the action of the reagent. Whichever 
 stopper is used, it should be firmly wired on. This pipette is 
 suitable for the sticks of phosphorus used for the absorption of 
 oxygen. 
 
HEM PEL'S APPARATUS. 
 
 475 
 
 723. The Double Pipette is used for reagents, such as cuprous 
 chloride, which must be protected from the air. The absorbing 
 reagent is placed in the bulb a, and when this bulb is filled with 
 gas the liquid is forced into the bulb b. The bulb c is filled with 
 water, which serves to retain the atmosphere of nitrogen, carbon 
 dioxide, or other indifferent gas which is kept in the bulb b over 
 
 m 
 
 FIG. 70. 
 
 the reagent. To fill this pipette a stream of carbon dioxide or 
 nitrogen is passed into the empty pipette at m. The stream of 
 nitrogen may be produced by placing in a small flask a mixture of 
 equal parts of ammonium chloride and sodium nitrite. After the 
 flask has been attached to the pipette by means of a rubber stopper 
 and a delivery-tube the nitrogen is evolved by gently heating the 
 flask with the Bunsen burner. When the air has been displaced 
 the delivery-tube is disconnected and the bulb c filled with water. 
 A glass tube, about a meter long, is connected at I by means of a 
 short piece of rubber tubing, and the cuprous-chloride solution 
 transferred to the bulb a by pouring it into a funnel attached to 
 the upper end of the long glass tube. 
 
 If FUMING SULPHURIC ACID is placed in a double pipette, the 
 liquid seal in c and d should be concentrated sulphuric acid. Any 
 water in the bulbs may be rinsed out with concentrated sulphuric 
 
476 
 
 GAS ANALYSIS. 
 
 acid. On account of the action of the fuming sulphuric acid on 
 rubber the pipette must be filled by inverting it so that the end 
 of the capillary tube may be dipped into the acid. Suction is 
 applied by the lungs or a Bunsen filter-pump at m until the bulb 
 a is full. The pipette is then inverted and concentrated sulphuric 
 acid poured into the bulb d to prevent the entrance of moisture 
 from the atmosphere. 
 
 724. The Double Pipette for Solid and Liquid Reagents has 
 an opening in the bulb a for the insertion of solids. This bulb is 
 used for the acid cuprous chloride solution. The bulb a is filled with 
 glass tubes containing coils of copper wire. After pouring in the 
 cuprous chloride solution the opening is closed with a rubber stopper 
 or glass stopper as directed for the single pipette for solid reagents. 
 This bulb need not be filled with nitrogen, as any cupric chloride 
 formed by the oxygen present is soon reduced by the metallic 
 copper, the air in it being soon deprived of its oxygen. 
 
 725. The Explosion Pipette consists of a thick-walled glass bulb 
 closed with a glass stop-cock at d. The level bulb b is connected 
 
 FIG. 71. 
 
 with the bulb a by means of a piece of thick-walled rubber tubing. 
 Mercury is used as the confining liquid in the pipette, since under the 
 high pressure developed by the explosion a considerable amount 
 
HEM PEL'S APPARATUS. 
 
 477 
 
 of carbon dioxide would dissolve in water. Platinum wires are 
 sealed into the glass at c. The spark which ignites the gases is 
 produced between these wires. The current from several Bunsen 
 dichromate cells or other convenient battery is passed through a 
 Ruhmkorff induction-coil. A coil about 15 cm. long is of con-, 
 venient size. The secondary current produced by the induction- 
 coil should produce a bright spark between the platinum wires. A 
 box made of strong wire netting is very convenient for placing 
 over the pipette for protection in case it should be shattered by 
 the explosion. If this is not at hand, it should be placed in a hood 
 or covered with a towel during the explosion. 
 
 726. The Pipette Arranged for Combustion with a Hot Platinum 
 Wire is shown in Fig. 72. A stout iron wire inclosed in a glass 
 
 FIG. 72. 
 
 tube is pushed through the rubber stopper and the lower end 
 of the glass tube closed with a piece of rubber tubing which is 
 wired around the iron wire. A second stout iron wire is pushed 
 through the rubber stopper and the upper end connected with the 
 first iron wire by means of a coil of platinum wire J mm. in diameter. 
 The coil should be about 2 mm. in diameter and contain from 
 20 to 30 turns. After filling the bulb with mercury or water the air 
 is sucked out of the glass tube containing the iron wire by closing 
 the capillary tube with a rubber tube and pinch-cock and applying 
 suction with a pump to the tube of the level bulb. The gas to be 
 
478 GAS ANALYSIS. 
 
 burned is transferred to the pipette, and a measured volume of 
 oxygen is slowly passed into the pipette after heating the plati- 
 num wire to redness with a current of electricity. 
 
 727. The Method of Manipulation of the Hempel apparatus 
 will be more fully understood from the following description of 
 the analysis of illuminating-gas. 
 
 EXERCISE 7^. 
 Analysis of Illuminating-gas. 
 
 72-8. Collecting the Sample. 200 c.c. of water are placed in a convenient- 
 sized flask, and the gas to be analyzed allowed to bubble through for a few 
 minutes, the flask being occasionally shaken. The level tube of the gas 
 burette is filled with the water saturated with the gas, and the air displaced 
 from the measuring-tube by opening the pinch-cock and raising the level 
 tube. The measuring-tube is connected with the gas-supply by means of a 
 rubber tube, from which the air is expelled by allowing the gas to flow a few 
 minutes. If the gas has not been burned for some time, it should be allowed 
 to flow* for a few minutes in order to sweep out any air which has accumu- 
 lated in the supply-pipes, which are seldom gas-tight. More than 100 c.c. 
 are drawn into the measuring-tube by lowering the level tube. The pinch- 
 cock of the measuring-tube is closed, and the tube connecting with the gas- 
 supply is removed. Exactly 100 c.c. of the gas is obtained by holding the 
 level tube in the hand, so that the bottom of the meniscus of the water 
 is exactly opposite the 100-c.c. mark on the burette. The pinch-cock is 
 now opened for a moment, thus allowing the excess of gas to escape. If 
 the gas does not measure exactly 100 c.c. when the level tube is held so 
 that the water is on the same level in the two tubes, the operation is repeated. 
 After a little practice the exact amount of gas is readily obtained. The gas 
 measuring-tube should be handled as little as possible. By taking hold of 
 the foot heating the glass, and consequently the inclosed gas, is avoided. 
 
 729. Carbon Dioxide. The burette is now connected to the caustic potash 
 bulb, as shown in Fig. 73. The capillary tube F is inserted into the rubber 
 tube d to the pinch-cock. The rubber tube m is then pinched with the 
 fingers to expel the air, and the other limb of the capillary tube inserted. 
 After noting the height of the caustic potash solution in the long arm of the 
 capillary U-tube of the pipette, the gas is passed into the pipette by opening 
 the pinch-cock and raising the level tube B. The pinch-cock is closed, and 
 three minutes allowed for the absorption of the carbon dioxide, the pipette 
 being shaken occasionally. The gas is then passed back into the burette 
 by opening the pinch-cock and lowering the level tube, the caustic potash 
 solution being brought to the same point in the capillary tube to which it 
 came before passing in the gas. The pinch-cock is closed, and after three 
 
ILL UMINA TING-GAS. 
 
 479 
 
 minutes the volume of the gas is read by bringing the water in the two tubes 
 to the same level by raising or lowering the level tube. The three-minute 
 intervals are most readily gauged by means of a small three-minute sand-glass. 
 
 FIG. 73. 
 
 730. Uluminants. While waiting three minutes for the water to run 
 down in the burette and the gas to reach the atmospheric temperature, the 
 caustic potash pipette should be replaced by the pipette containing the 
 fuming sulphuric acid. After reading and recording the volume, the gas is 
 passed over the fuming sulphuric acid and shaken for three minutes. The 
 gas is then passed back into the measuring-burette, and once more passed 
 into the caustic potash bulb, then back into the burette and measured 
 after three minutes. If the white fumes of the oxides of phosphorus do 
 not appear oxygen is absent or the phosphorus is too cold. 
 
 73 1 . The Oxygen is absorbed by passing the gas into the phosphorus 
 pipette. After three minutes it is passed back into the burette and meas- 
 ured. 
 
 732. The Carbon Monoxide is absorbed by passing the gas into the pipette 
 containing the cuprous chloride solution. If this solution is fresh, the carbon 
 
480 GAS ANALYSIS. 
 
 monoxide will be completely absorbed after shaking for three minutes, and 
 the residue may be passed into the burette and measured. If a fresh solu- 
 tion of cuprous chloride, and also one which has been used considerably, 
 are at hand, the bulk of the carbon monoxide may be absorbed by shaking 
 for two minutes with the old solution, and the last traces absorbed by 
 shaking the gas with the fresh solution for three minutes. A careful record 
 should be kept on a piece of paper pasted on these pipettes of the amount 
 ot carbon monoxide absorbed. If a hydrochloric acid solution of cuprous 
 chloride is used, the gas should be passed into the caustic potash bulb before 
 measuring in order to remove the hydrochloric acid gas. 
 
 733. Explosion of Hydrogen and Methane- The gas now remaining con- 
 sists of methane, hydrogen, and nitrogen. These constituents may be deter- 
 mined by explosion, or by combustion with the red-hot platinum wire. If 
 the explosion pipette is to be used, the gas is passed back into the cuprous 
 chloride pipette, which is closed wi' h a rubber tube and a pinch-cock. The 
 water in the burette is replaced by distilled water which has been saturated 
 with air. From 12 to 15 c.c. of the gas residue is then drawn into the burette 
 and carefully measured. Air is then drawn into the burette until the total 
 volume of the gas is very nearly 100 c.c. The volume is again carefully 
 read and the gas is passed into the explosion pipette, the large bulb of 
 which, including the capillary tube, has been filled with mercury. When 
 the gas has been passed in, enough water to nearly fill the capillary tube 
 should also be forced over and the glass stop-cock of the pipette closed. 
 A pinch-cock is placed on the rubber tube, which is securely wired on the 
 capillary tube of the pipette. The end of th's rubber tube is also closed 
 with a piece of glass rod. After placing the pipette under the hood or 
 other safe place, the secondary coil of the Ruhmkorf coil is connected with 
 the platinum wires of the explosion pipette, arid the primary coil is connected 
 with several Bunsen or other cells connected in series. The explosion 
 should be vigorous enough to present a single flash of light, the course of 
 which cannot be followed by the eye across the bulb. The glass stop-cock 
 of the pipette is opened, and the gas passed into the burette, and after 
 three minutes the volume is read. The burette is then connected with 
 the caustic potash pipette, the carbon dioxide absorbed, and the volume 
 of the gas again determined with the burette. 
 
 If the explosion was not satisfactory the operation should be repeated, 
 using a little more of the gas residue. As this operation is attended with 
 considerable liability of error, it is well in any case to repeat the explosion. 
 
 734. Combustion of Hydrogen and Methane by Red-hot Platinum Wire. 
 If the pipette fitted with the platinum wire for burning the gas is at hand, 
 it should be used rather than the explosion pipette. The entire gas residue 
 may then be used. After being carefully measured, it is transferred to the 
 pipette containing the platinum wire heated to redness. 10% more oxygen 
 than the volume of the gas residue is carefully measured in the burette, 
 and then slowly passed into the pipette, the wire having been heated to 
 redness by a suitable current of electricity. The oxygen used must be freed 
 from carbon dioxide by pa^in<r ihr uo-h istrong caustic potash solution or 
 
ORSAT APPARATUS. 481 
 
 by being shaken in the caustic potash pipette. The presence of nitrogen 
 is not objectionable. The amount of contraction as well as the volume oi 
 carbon dioxide formed must be noted as when the explosion pipette is used. 
 735. Calculation of the Analysis. The following analysis indicates the 
 method of recording and calculating the results: 
 
 Volume of gas taken 100.0 c.c. 
 
 After absorption by caustic potash 97 . 4 c.c. 
 
 Per cent of carbon dioxide fc 26 
 
 After absorption by fuming sulphuric acid 84 . 2 c.c. 
 
 Per cent of illuminants 13 .2 
 
 After absorption by phosphorus 83 . 9 c.c. 
 
 Per cent of oxygen .3 
 
 After absorption by cuprous chloride 56.6 c.c. 
 
 Per cent of carbon monoxide 27.3 
 
 Taken for explosion 14 c.c. 
 
 After addition of air 98 . 5 c.c. 
 
 Volume after explosion 75 . 9 c.c. 
 
 Contraction 22 . 6 c.c. 
 
 Volume after absorption by caustic potash 69.9 c.c. 
 
 Volume of carbon dioxide 6.0 c.c. 
 
 Volume of methane in 14 c.c. residue 6.0 c.c. 
 
 The methane in 56.6 c.c. is found by the propor- 
 tion 14 : 56.6 : : 6 : x, z = 24.3 c.c. 
 
 Per cent of methane 24.3 
 
 The contraction due to methane, being twice its 
 
 volume, will be 12 c.c. 
 
 Contraction due to hydrogen (22.6 12) 10.6 c.c. 
 
 The volume of hydrogen, being two-thirds of this 
 
 contraction, will be 7.1 c.c. 
 
 The volume of hydrogen in 56.6 c.c. will be found 
 
 by the proportion 14 : 56.6 : : 7.1 : x, z = 28.7 c.c. 
 
 Per cent of hydrogen 28.7 
 
 Per cent of nitrogen 3.6 
 
 The combined volume of methane and hydrogen being 53.0 c.c., the 
 remainder, or 3.6 c.c., is considered to be nitrogen. 
 
 736. The Orsat Apparatus. A very convenient and compact 
 apparatus for gas analysis was devised by Orsat, and with a num- 
 ber of modifications is very largely in use in the form shown in 
 Fig. 74. The measuring-burette a has a capacity of 100 c.c., 
 although the graduations do not always extend over the enlarged 
 portion at the top, the absorbable constituents in many cases 
 amounting to less than 50% of the gas analyzed. This burette is 
 
482 
 
 GAS ANALYSIS. 
 
 water-jacketed, so that the gas may quickly come to a constant 
 temperature. The absorption-bulbs, b, c, and d, contain glass 
 tubes so as to give a larger absorbing surface to the reagent. 
 Each of these bulbs is connected with a bulb of the same capacity, 
 which is only partially seen at the rear in Fig. 74. The reagent 
 forced out of the bulb in front by the gas passes into the bulb in 
 the rear. The bulb 6 is filled with caustic potash for absorbing 
 
 FIG. 74. 
 
 carbon dioxide, c is filled with an alkaline solution of pyrogallol 
 for the absorption of oxygen, while d is filled with cuprous chloride 
 solution for the absorption of carbon monoxide. The palladium 
 tube /, with the alcohol-lamp g and the bulb h, to be used for the 
 absorption of hydrogen, are not always attached, since the appa- 
 ratus is very largely used for the analysis of flue gases in which 
 hydrogen is absent. The large U-tube at the side is filled with 
 
ORSAT APPARATUS. 483 
 
 cotton, which removes dust particles from the gas as it enters the 
 apparatus. 
 
 The bottle m, connected by means of a rubber tube with the 
 measuring-tube, serves as a level tube, by which the gas may be 
 forced into or out of the measuring-tube and absorption-bulbs. 
 Water is used as the confining liquid. The whole apparatus is 
 enclosed in a wooden case, the front and back of which may be 
 readily removed. 
 
 737. The Manipulation of this apparatus is very similar to 
 that of the Hempel apparatus. The stop-cocks and rubber joints 
 must first be tested for leaks by bringing the solution in the ab- 
 sorption-bulbs to the zero-marks on the capillary stems and clos- 
 ing the stop-cocks. If after a few minutes the liquid has fallen 
 in any of the bulbs, the stop-cock or rubber connection leaks, and 
 the key of the former must be removed, cleaned, and lubricated 
 with vaseline and the rubber joints made tight by wiring the rub- 
 ber on the glass tubes. The stop-cock k and the rubber connec- 
 tions with the measuring-tube are tested by bringing the water in 
 the measuring-tube to the zero-mark on the upper capillary by 
 holding the level bottle at the same height. On closing the stop- 
 cock k and lowering the bottle, no air should enter the apparatus 
 as indicated by the height of the water-column when the bottle 
 is again raised. 
 
 All of the joints having been made tight and the solutions in 
 the absorption-bulbs being at the marks on the capillary stems, 
 the measuring-tube is filled with the gas to be analyzed. In order 
 to displace the air from the connecting-tubes it is necessary to 
 fill the measuring-tube several times and then expel the gas. 
 When a fair sample is finally obtained the volume is made 
 exactly 100 c.c. by holding the bottle so that the level of the 
 water is the same as that in the measuring-tube. The stop- 
 cock k is then opened and immediately closed. The gas is then 
 forced into the absorption-bulbs in turn and drawn back into 
 the measuring-tube in order to note the diminution of volume. 
 The absorption of oxygen and carbon monoxide being some- 
 what slower than that of carbon dioxide, it is well to pass the 
 gas back into the same bulb after measurement to ascertain by a 
 second measurement if absorption is complete. As the reagents 
 
484 GAS ANALYSIS. 
 
 are not well protected from the air, they must frequently be 
 renewed or tested for their efficiency. 
 
 738. In the Analysis of Flue Gases for judging the efficiency 
 of the firing, the most difficult part of the operation is taking 
 the sample of the gas. When the fires are hand stoked the com- 
 position of the flue gas will vary with the time which has elapsed 
 since fresh coal was put on the grate, as the following analyses * 
 show: 
 
 One Minute Twelve Minutes 
 After Stoking. Later. 
 
 Carbon dioxide 13.5 4.0 
 
 " monoxide 0.0 0.0 
 
 Oxygen 5.5 16.5 
 
 Nitrogen 81.0 79.5 
 
 739- Taking an Average Sample. A very common practice 
 consists in collecting the sample in a large bottle or gasometer from 
 which the water is allowed to flow slowly, so that six, twelve, or 
 twenty-four hours are required to fill the gas receiver. This aver- 
 age sample is then analyzed. When this method is used every 
 precaution must be taken to make the connections absolutely tight 
 and to shellac all rubber connections. The method is subject 
 to the errors due to the solubility of the gases in the water used. 
 
 The tube extending into the flue may be of iron if the temper- 
 ature of the gases is low. Should the iron become heated to even 
 very dull redness alternate oxidation and reduction may take 
 place, which very materially changes the composition of the gases. 
 A glass or porcelain tube must then be used. Several openings 
 should be made in the tube so that the gas in the entire cross- 
 section of the flue may be sampled. The entrance to the flue 
 should be made as near the fire as practicable, as the diffusion of 
 both air and the products of combustion through the usual leaks 
 in the flues is considerable. 
 
 740. Interpretation of Results. The method of taking an 
 average sample of the flue gases as already given is much more 
 suitable for mechanically stoked fires than those hand fed. For 
 the latter it is advisable to take numerous samples at various 
 
 * Lunge, Chem.-tech. Untersuchungsmethoden, Vol. I, p. 183. 
 
PROBLEMS. 485 
 
 intervals of time after stoking. The ideal flue gas should contain 
 only carbon dioxide and nitrogen. The percentage of the former 
 will not be quite equal to that of the oxygen in the air, because 
 some of the oxygen combines with the hydrogen of the coal. 
 The remainder of the oxygen, however, will give an equal volume 
 of carbon dioxide, since a molecule of the latter contains a molecule 
 of oxygen. Generally, however, the flue gases from even well- 
 managed fires contain several per cent of unused oxygen. If 
 oxygen is entirely absent and carbon monoxide is present, the 
 draught of air is insufficient. Carbon monoxide may also occur 
 in the presence of oxygen if clinkers are allowed to remain on the 
 grate. 
 
 Problem 16. A coal contains 3% hydrogen, 82% carbon, and 15% 
 ash. What would be the composition by volume of the flue-gas (a) if per- 
 fect combustion took place, all the oxygen of the air being used up and no 
 combustible left in the ash? (6) If only half of the oxygen of the air were 
 used? 
 
 Problem 17. A flue-gas produced by the same coal contains 3.3% 
 carbon dioxide and 16.7% oxygen. Calculate the excess of air passing 
 through the grate. 
 
CHAPTER XXXII. 
 STOICHIOMETRY. 
 
 741. Atomic-weight Tables. In the calculation of the re- 
 sults of many chemical analyses the atomic weight of one or 
 more of the elements is used. The percentage found will vary 
 somewhat according to the atomic weights used. Several tables 
 of atomic weights compiled by various authorities are in use. 
 Besides differing in the values assigned to elements whose atomic 
 weights are in dispute, the tables differ as to the element whose 
 atomic weight is taken as the basis for computing the remaining 
 values. Until recently hydrogen has been taken as the basis, the 
 value 1.000 having been assigned to it. Tables computed on the 
 basis of 16.000 as the value for oxygen are now more commonly 
 used. On this basis hydrogen has the value 1.008. It is advis- 
 able to select a set of atomic weights and in all calculations to use 
 these values. The values used in this book are those given in 
 Table I, p. 509, which is the table published by the International 
 Atomic Weights Committee in 1908. The figures given by this 
 Committee are based on the value for oxygen of 16.000, the values 
 based on hydrogen as unity being no longer given. 
 
 742. Limit of Accuracy in Atomic Weights. It will be observed 
 that the number of figures given after the decimal point varies 
 from one to three. This is due to the fact that the values have 
 been determined with a varying degree of accuracy. In each case 
 the last figure given is somewhat doubtful. The atomic weight of 
 silver, for instance, is certainly between the values 107.9 and 
 108.0, the value 107.93 being the average of several determinations, 
 in which the variations were in the second place of decimals. To 
 write the atomic weight of silver with three figures after the deci- 
 mal point, as 107.930, would be introducing an absolutely mean- 
 
 486 
 
FACTORS. 487 
 
 ingless figure. As most quantitative work is far less accurate 
 than atomic-weight determinations, even the last figure given may 
 frequently be dropped. Generally the figure 107.9 for silver will 
 be near enough to its true value, the figure dropped representing 
 only -yinnr of its atomic weight, which is smaller than the ordi- 
 nary analytical errors. If in the same calculation the value of 
 another element enters whose atomic weight is known with still 
 less certainty, even one more figure may be dropped. If the atomic 
 weight of iron, for instance, which is uncertain within 1 part 
 in 560, enters into the calculation, the atomic weight 108 may be 
 used for silver, the error being 7 parts in 10,793, or about 1 
 pail in 1540. Numbers obtained by multiplying or dividing such 
 numbers need not be given to more than four significant figures. 
 
 743. " Factors." If iron has been precipitated as ferric hy- 
 droxide and weighed as ferric oxide the amount of iron in the 
 precipitate is found by calculation, using the atomic weights of 
 iron and oxygen. From the formula of ferric oxide, Fe 2 3 , we 
 know that 2 atoms of iron, or 111.8 parts, are combined with 3 
 atoms of oxygen, or 48 parts. In 159.8 parts of the precipitate 
 there are present 111.8 parts of iron. If the weight of the pre- 
 cipitate is represented by p, the amount of iron, x, is found by the 
 following proportion: 
 
 Fe 2 3 : 2Fe : : p : x 
 159.8 : 111.8 : : p : x 
 
 111.8 
 ~ 159.8 P ' 
 
 111 8 
 The fraction 1 ' g , which is equal to .6996, represents the percent- 
 
 age of iron in ferric oxide and enters into every calculation of the 
 amount of iron as found by weighing ferric oxide. For each 
 precipitate one or more factors are calculated in this manner, thus 
 obviating the use of the proportion and simplifying the calcula- 
 tion. The factors for the commonly used precipitates are given 
 in Table II, p. 510. 
 
 744. Logarithms. To obtain the weight of a given element in 
 a precipitate the weight of the precipitate is multiplied by the 
 proper factor. The labor of multiplication may be greatly 
 
488 STOICHIOMETRY. 
 
 lessened by using logarithms. For this reason the logarithms of 
 the factors are also given in Table II. In order to multiply two 
 numbers together their logarithms are added. The sum is the 
 logarithm of the product of the two numbers. Two numbers 
 may be divided by subtracting the logarithm of the divisor from 
 the logarithm of the dividend, the remainder being the logarithm 
 of the quotient. The abbreviation log. is frequently used for 
 logarithm. The logarithms of the numbers from 1 to 9999 may be 
 obtained from Table III, page 514. The logarithm of a number is 
 composed of two parts, a mantissa and a characteristic. The 
 former is a positive decimal fraction and is the same for any 
 number composed of the same figures in the same order irrespec- 
 tive of the position of the decimal point. The characteristic is a 
 positive or negative integral number and indicates the position of 
 the decimal point of the number. In the first column at the left 
 of Table III are found the natural numbers beginning w'th 10. 
 The mantissas of the logarithms of any of these numbers are 
 found in the next column headed 0. In the following columns 
 are found the mantissas of the numbers formed by adding to 
 those in the first column the significant figure which is found at 
 the head of one of the columns. For instance, the mantissa of 21 
 is found in the column headed opposite 21 and is .3222. In the 
 next column opposite 21 is found .3243, which is the mantissa of 
 211. Following this number is .3263, which is the mantissa of 212. 
 If the mantissa of a number of four places is desired the columns 
 of differences must be utilized. The mantissa of 211 being .3243 
 and that of 212 being .3263, the mantissa of a four-place number 
 between 2110 and 2120 will have a value between .3243 and .3263. 
 The difference between these values is .0020, while the difference 
 between 2110 and 2120 is 10. For each unit added to 2110, .0002 
 must be added to the mantissa of 2110, or .3243. The mantissa 
 of 2115 is .3253, 10 having been added. This number may be 
 found in the column of differences headed 5 opposite the natural 
 number 21. As ordinary quantitative results are not accurate 
 beyond four significant figures, there is no necessity of obtaining 
 the mantissa to the fifth significant place. 
 
 745. The Characteristic is the number placed before the 
 mantissa- to designate the position of the decimal point in the 
 
LOGARITHMS. 489 
 
 natural numbers. The characteristic of the log. of a number 
 with one significant figure to the left of the decimal point is 0, 
 if there are two figures it is 1, with three figures it is 2, etc. The 
 characteristic of -the log. of a decimal is a negative number and is 
 equal to the number of places by which its first significant figure 
 is removed from the place of units. In order to make this char- 
 acteristic positive it is sometimes added to 10, which is placed 
 with a minus sign after the mantissa. The logarithms may then 
 be added, the proper number of tens being subtracted from the 
 result. According to this rule the complete logs, of the numbers 
 containing the figures 2115 are given in the following table: 
 
 Numbers. Logarithms. 
 
 2115 3.3253 
 
 211.5 2.3253 
 
 21.15 1.3253 
 
 2.115 0.3253 
 
 .2115 1.3253 or 9.3253-10 
 
 .02115 3/3253 or 8.3253-10 
 
 .002115 3.3253 or 7.3253-10 
 
 As the position of the decimal point may generally be ascer- 
 tained by inspection of the numbers to be multiplied or divided, 
 the characteristic is frequently omitted from the logarithms. 
 Illustration. If a precipitate of ferric oxide weighing .3252 
 gram is obtained from .8276 gram of material, the percentage 
 of iron is calculated by logarithms as follows: 
 
 log. of .3252 1.5122 or 9.5122-10 
 
 log. of factor of Fe in Fe 2 3 1 . 8449 or 9 . 8449-10 
 
 by addition log. of weight of iron 1 . 3571 or 19 . 3571-20 
 
 log. of .8276 1.9178 or 9.9178-10 
 
 by subtraction and addition of 2 (log. of 100) 
 
 log. of percentage of iron . . 1 . 4393 or 1 . 4393 
 percentage of iron 27 . 50 
 
 The number corresponding to a given logarithm is obtained 
 from the table of antilogarithms in exactly the same manner as 
 the logarithm of a number is obtained from the table of logarithms, 
 the first column of figures and the figures at the head of the other 
 
490 STOICHIOMETRY. 
 
 columns being the logarithms, while the numbers corresponding 
 are found in the body of the table. 
 
 746. Factors from a Single Precipitate. A given precipitate 
 may be used for the calculation of a large number of different 
 substances, as is shown by the following indicated calculations 
 based on determinations in which the weighed substance is 
 barium sulphate. The weight of barium sulphate is indicated 
 by p. 
 
 Substance weighed BaS0 4 . 
 Desired, the amount of barium. 
 
 BaS0 4 : Ba : : p : x 
 Mol. wts. 233 137 
 
 137 137 
 
 fact r= 233- 
 
 Desired, the amount of barium oxide. 
 
 BaS0 4 : BaO : : p : x 
 Mol. wts. 233 153 
 
 153 153 
 
 factor= 233* 
 
 Desired, the amount of sulphur trioxide. 
 
 BaS0 4 : S0 3 : : p : x 
 Mol. wts. 233 80 
 
 80 , 80 
 
 X= 233 P > factor= 233' 
 
 Desired, the amount of sulphuric acid. 
 
 BaS0 4 : H 2 S0 4 : : p : x 
 Mol. wts. 233 98 
 
 98 , 98 
 
 factor= 233' 
 
 Desired, the amount of sodium sulphate. 
 BaS0 4 : Na 2 S0 4 : : p : x 
 Mol. wts. 233 142 
 
 142 , 142 
 
 a . = _ p; factor =233. 
 
INDIRECT GRAVIMETRIC ANALYSES. 491 
 
 Desired, the amount of crystallized sodium sulphate. 
 
 BaSO, : Na 2 S0 4 .10H 2 : : p : x. 
 Mol. wts. 233 322 
 
 322 322 
 
 Desired, the amount of sulphur. 
 BaS0 4 : S : : p 
 Mol. wts. 233 32 
 
 32 32 
 
 factor= 233' 
 
 747. The Calculation of the Theoretical Percentage of an Ele- 
 ment in a Compound is made in exactly the same manner as 
 the calculation of a " factor" where the element or radicle 
 sought is part of the substance weighed. The theoretical per- 
 centage is the ratio of the atomic weight of the element sought to 
 the molecular weight of the compound. For example, the theo- 
 retical percentage of silver in silver nitrate is obtained from the 
 
 rati WWch 6qUalS 169^94 r 6 
 
 The theoretical percentage of S0 4 in crystallized magnesium 
 
 SO 
 sulphate is obtained from the ratio i\/r QO 7TT (V wn ^ c ^ e( l ua l s 
 
 Hm or 38 - 96% - 
 
 748. Indirect Gravimetric Analyses are more difficult to cal- 
 culate. General rules or factors cannot readily be given. The 
 general method may be seen from the calculation of a determina- 
 tion of potassium and sodium in a mixture of the chlorides of 
 these metals. This determination is readily carried out by weigh- 
 ing the sodium and potassium chlorides together and then deter- 
 mining the amount of chlorine present. As the percentage of 
 chlorine in potassium chloride is 47.53, and in sodium chloride is 
 60.6, the percentage of chlorine in a mixture of these two salts will 
 be between these numbers. For example, a sample of the mixed 
 chlorides which weighed 1.5802 grams gave .9023 gram of chlo- 
 rine, which is equal to 57.1 per cent. If onlv potassium chloride 
 
492 STOICHIOMETRY. 
 
 had been present .7511 gram of chlorine would have been ob- 
 tained. The difference between .9023 and .7511 is the amount of 
 chlorine due to the presence of sodium chloride. This difference 
 (.1512) divided by the difference in the percentage of chlorine in 
 the two salts (.1307) gives the weight of sodium chloride present 
 
 = 1.1568). The weight of the mixed chlorides less the 
 
 \.1307 
 
 weight of sodium chloride gives the weight of potassium chloride 
 present (1.5802 -1.1568 = .4234). The problem may be solved 
 algebraically as follows: 
 
 Let x = the weight of sodium chloride ; 
 
 and y =ii " " potassium chloride ; 
 
 then x+ y = 1.5802 the weight of the mixed chlorides; 
 
 and .606:r + A753y = .9023 the weight of chlorine. 
 
 Solving these simultaneous equations we find that 
 
 .9023 -(1.5802X. 4753) 
 .606 -.4753 
 
 On carrying out the indicated operations we obtain a value for x, 
 or the amount of sodium chloride present, which is identical with 
 that already obtained. 
 
 The disadvantage of using indirect methods of this kind is found 
 in the fact that the difference upon which the weight of one of the 
 constituents is calculated is very small. In the illustration given 
 
 it must be multiplied by the factor 7.652 ( 1007)' ^ e ex P er ^~ 
 
 mental errors of the analysis are increased in this ratio in the per- 
 centage found. The method is, therefore, not applicable to the 
 determination of small amounts of one constituent in the presence 
 of large amounts of the other. 
 
 749. Factor Weights. When many determinations of a given 
 constituent must be made it is found convenient to so choose the 
 amount of substance to be weighed out that the weight of the 
 precipitate obtained shall have a simple ratio to the percent- 
 age to be calculated. For example, if 1.373 grams of a substance 
 containing .375 per cent of sulphur is weighed out, the sulphur 
 converted into sulphuric acid and weighed as barium sulphate, 
 
FACTOR WEIGHTS. 493 
 
 .0375 gram or the latter will be obtained. The number of milli- 
 grams of barium sulphate obtained is evidently equal to the hun- 
 dredths of per cent of sulphur present. The general method of 
 finding the amount to be weighed out in a given case may be 
 obtained from the following general case: 
 
 Let w =the amount of substance weighed out; 
 and i>=the weight of precipitate obtained; 
 and p =the per cent of the given constituent in the precipitate v. 
 The percentage is obtained by the following indicated calcula- 
 tions: 
 
 pXv 
 
 w 
 
 If the value of w bears a simple ratio to p, the percentage will bear a 
 simple ratio to v. In the illustration already given p, or the per- 
 centage of sulphur in barium sulphate, is 13.73. w was made one- 
 tenth of this number, or 1.373. The percentage is therefore ten 
 times v or the weight of the barium sulphate obtained, as may be 
 seen by substituting the values of v and w in the formula giving 
 
 vX 13.73 
 
 1.373 ' 
 
 If w is made 13.73 grams, the weight of v in grams will be equal to 
 the percentage of sulphur in the substance analyzed. In general, 
 then, the amount of substance taken should bear a simple ratio to 
 the percentage in the precipitate to be weighed of the constituent 
 to be determined. This ratio will be inversely equal to the ratio 
 of the percentage to the weight of the precipitate. For example, 
 the percentage of chlorine in silver chloride being 24.72, the per- 
 centage of chlorine in a given substance will be ten times the 
 weight of silver chloride obtained if 2.472 grams of the substance 
 to be analyzed are weighed out. 
 
 CALCULATION OF THE FORMULAE OF SALTS AND 
 ISOMORPHOUS MIXTURES. 
 
 The results of many analyses may be expressed by a chemical 
 formula. The composition of all pure chemical compounds as well 
 as a large number of minerals may be expressed in this manner. 
 All chemical formulas have been derived from quantitative anal- 
 
494 STOICHIOMETRY. 
 
 yses and express primarily the proportions by weight in which 
 the elements are present. 
 
 750. Salts. Instead of using the percentage notation, the system 
 of equivalent or atomic weights has been devised so that a single 
 number or its simple multiple expresses the proportion in which a 
 given element is present in its innumerable compounds. Sodium 
 chloride, for example, contains 39.40 per cent of sodium, and 
 60.60 per cent of chlorine as found by analysis, but this fact is 
 generally stated by giving the formula NaCl, which states that 
 23.05 parts of sodium are combined with 35.45 parts of chlorine. 
 The ratio of the percentages 39.40 and 60.60 must be the same as 
 the ratio of the numbers 23.05 and 35.45, and what is not quite so 
 self-evident, the ratio of 23.05 to 39.40 must be the same as the 
 ratio of 35.45 to 60.60. This ratio is 1.709; that is, 39.40 parts of 
 sodium represent 1.709 atoms of this element and 60.60 parts of 
 chlorine represent 1.709 atoms of chlorine. In this way it is 
 found from the percentage composition and atomic weights that 
 in sodium chloride an equal number of sodium and chlorine atoms 
 are present. If the percentages of copper and chlorine found in 
 crystallized cupric chloride are divided by the atomic weights of 
 these elements the numbers 
 
 537 30 41 58 
 
 -^V=.5864, ^-^ = 1.1728 
 o3.o 35.45 
 
 are obtained. The number of chlorine atoms is evidently twice 
 the number of copper atoms. The formula then must be CuCl 2 . 
 The number of acid radicles, molecules of water, etc., may be 
 found in the same manner, as may be seen from the calculation of 
 the formula of potash alum from its analysis. 
 
 Per Cent Found. Atomic Weights Atomic Ratios. 
 
 K 8.17 39.14 3JnS = -2087 
 
 Al 5.78 27.1 |^| - .2132 
 
 S0 4 40.26 96.06 g^ = 
 
 45.74 18.02 TO| ==: 
 
 Total.. . 99.95 
 
CALCULATION OF THE FORMULA OF SALTS. 495 
 
 The atomic ratios of potassium and aluminium being nearly equal, 
 one atom of each of these elements will be present with two sul- 
 phuric acid radicles and twelve molecules of water, since their 
 atomic ratios are respectively twice and twelve times as great as 
 that of potassium, giving the formula KA1(S0 4 ) 2 .12H 2 0. 
 
 751. Many Minerals are found in the form of sharply defined 
 homogeneous crystals, which contain varying amounts of several 
 elements. These minerals contain several salts which crystallize 
 in the same system, and are therefore called isomorphous. Dolo- 
 mite is an example of such a mineral, in which varying amounts 
 of magnesium and calcium carbonates are present. Small amounts 
 of iron, aluminium, and silica are also present. The silicic acid is 
 able to replace a portion of the carbonic acid forming calcium or 
 magnesium silicates, while the iron and aluminium are able to 
 replace a portion of the calcium and magnesium forming carbon- 
 ates or silicates. A general formula, M"R0 3 , will express the 
 composition of the mineral. M" represents the bivalent bases, 
 calcium, magnesium, ferrous iron, or even two valences of an 
 aluminium atom. R represents the carbon or silicon. 
 
 The method of obtaining the formula from an analysis is shown 
 by the following analysis of dolomite: 
 
 46 80 
 
 Per cent of C0 2 ....... 46.80 
 
 Si0 2 ....... 62 = .010 
 
 " * " 
 
 Total acid equivalents, R"0 2 = 1.074 
 " " " FeO ....... 21 71^9= - 003 
 
 " " " 
 
 " " 
 
 CaO ...... 32.35 
 
 MgO ...... 19.83 .491 
 
 Total ..... 99.81 
 
 Total basic equivalents, M"0 = 1.071 
 
 Ratio of acids to bases ' =1.003 
 
496 STOICHIOMETRY. 
 
 The formula of the dolomite will therefore be MgC0 3 ,CaC0 3 , small 
 amounts of the base being replaced by FeO, and a little of the acid 
 by Si0 2 . 
 
 For each atom of bivalent metal there is present one acid radi- 
 cle, the number 1.003 being so close to 1.000 that the difference 
 may be ascribed to the unavoidable experimental errors of the 
 analysis. Ratios as high as 1.02 or even 1.03 may be produced 
 by errors which occur in fairly good analytical work, although the 
 analytical worker should not be satisfied unless the ratio is usu- 
 ally less than 1.01. 
 
 BALANCING OF EQUATIONS. 
 
 The calculation of nearly all volumetric determinations is 
 based on the chemical equation which represents the reaction 
 taking place. While it is true that the chemical equation is the 
 expression of a quantitative analysis and should not be written 
 unless the quantitative analysis has first been made, this view of the 
 matter is taken only by the research chemist. In ordinary analyt- 
 ical work the correctness of the equation is assumed, and the 
 quantitative result is calculated from it. The equation must 
 therefore be written before the calculation is made. 
 
 752. The Simple Synthetical Equation, such as 
 
 2H 2 + 2 =2H 2 
 or 
 
 CH 4 +20 2 =C0 2 +2H 2 0, 
 
 is required mainly for gas analysis. 
 
 753. The Analytical Equations, such as 
 
 CuC 2 4 =Cu+2C0 2 
 or 
 
 2H 2 0=2H 2 +0 2 , 
 
 are most commonly required to express electrical decompositions. 
 When the elements or compounds produced by the reaction are 
 known, the equations are balanced by counting the number of 
 atoms of each element on each side of the equation. For exam- 
 ple, in the second equation given, the molecule of methane requires 
 for its oxidation two atoms of oxygen for the carbon atom and 
 two atoms for the four hydrogen atoms. The first member will 
 
BALANCING OF EQUATIONS. 497 
 
 therefore contain C . . . 1, H . . . 4, and ... 4, while 
 the second member will contain C . . . 1, H . . . 4, and 
 ... 4. These numbers being identical for each element, the 
 equation is balanced. 
 
 754. Metathetical Equations are far more commonly met with, 
 as they represent the two types of chemical reaction which are 
 most commonly carried out; namely (a) those involving the pro- 
 duction of a precipitate when two substances in solution are 
 brought together, such as 
 
 AgN0 3 + Nad = AgCl+ NaN0 3 
 or 
 
 BaCl 2 +H 2 S0 4 =BaS0 4 +2HCl; 
 
 (6) those involving the production of a substance which is hi the 
 gaseous form under the conditions of the experiment, as 
 
 CaC0 3 + 2HC1 =CaCl 2 + C0 2 + H 2 
 or 
 
 2NH 4 Cl+Ca(OH) 2 =2NH 3 +CaCl 2 +2H 2 0. 
 
 Equations such as these may be balanced by adding or sub- 
 tracting molecules of one or the other substance until the number 
 of atoms of each element is identical in each member of the equa- 
 tion. It is far better, however, to take into consideration the fact 
 that some of the elements are combined into groups or radicles 
 which do not separate into their constituent atoms. N0 3 is such a 
 radicle, and in the first equation given the number of N0 3 radicles 
 on each side of the equation must be identical. S0 4 acts as a 
 radicle in the same manner. In the third equation the carbon 
 remains combined with two atoms of oxygen as C0 2 , and two 
 molecules of hydrochloric acid must be introduced, because one 
 atom of calcium requires two atoms of chlorine to form calcium 
 chloride, while the two atoms of hydrogen unite with the remain- 
 ing atoms of oxygen in the calcium carbonate. In a similar man- 
 ner the ammonia radicle NH 3 is considered as a unit in the fourth 
 equation. 
 
 755. Oxidation and Reduction Equations are more difficult to 
 balance since two distinct reactions generally take place. The 
 
STOICHIOMETRY. 
 
 valence of two substances changes, involving the transference of 
 oxygen or its equivalent, and in the second place an acid or an 
 alkali reacts with some of the elements present to form salts. In 
 balancing these equations the oxidizing and reducing substances 
 must first be introduced in equivalent amounts. Then the amount 
 of acid or base necessary to combine with the bases or acid liber- 
 ated by the first reaction must be introduced. 
 
 In order to balance the oxidizing and reducing equations the 
 amount of oxygen absorbed or liberated must be known for each 
 substance present. For instance, a molecule of potassium dichro- 
 mate breaks up as follows: 
 
 K 2 Cr 2 7 =K 2 + Cr 2 3 + 30, 
 
 liberating three atoms of oxygen. Potassium permanganate breaks 
 up so as to give five atoms of oxygen for every two molecules of 
 the salt, as follows: 
 
 2KMn0 4 =K 2 + 2MnO + 50. 
 
 Amongst reducing substances two atoms of ferrous iron take up 
 one atom of oxygen to form ferric iron while one atom of stan- 
 nous tin takes up one atom of oxygen or its equivalent. Two 
 molecules of potassium permanganate giving five atoms of oxygen 
 are therefore able to oxidize ten atoms of ferrous iron. If the 
 iron is present as ferrous sulphate one molecule of sulphuric acid 
 will be required to convert two atoms of ferrous iron into ferric 
 iron as follows: 
 
 2FeS0 4 +H 2 S0 4 +0=Fe 2 (S0 4 ) 3 +H 2 0. . . . (1) 
 
 As five atoms of oxygen are given by the two molecules of potas- 
 sium permanganate we must multiply each member of the equa- 
 tion by five, giving: 
 
 4 + 5H 2 S0 4 +50=5Fe 2 (S0 4 ) 3 + 5H 2 0. , ; . (2) 
 
 The potassium and manganese of the potassium permanganate are 
 also converted into sulphates, two molecules of potassium per- 
 manganate requiring three molecules of sulphuric acid as follows: 
 
 2KMn0 4 +3H 2 S0 4 =K 2 S0 4 +2MnS0 4 + 3H 2 + 50. . . (3) 
 
CALCULATION OF VOLUMETRIC DETERMINATIONS. 499 
 Combining equation (2) and (3), we obtain the complete equation 
 10FeS0 4 + 2KMn0 4 + 8H 2 S0 4 =5Fe 2 (S0 4 ) 3 + K,S0 4 + 2MnS0 4 
 
 The free oxygen disappears from both equations since it is present 
 in the first member in the potassium permanganate and in the 
 second as water. 
 
 The oxidation of stannous chloride with potassium dichromate 
 can also be written in two equations which may then be combined 
 into one. The potassium dichromate liberates oxygen in the 
 presence of . reducing agent according to the following equation : 
 
 K 2 Cr 2 7 +8HCl=2KCl + 2CrCl 3 +30 + 4H 2 0. . . . (1) 
 
 The stannous chloride in the presence of acid and an oxidizing 
 agent reacts as follows: 
 
 + 2HCl+0=SnCl 4 +H 2 ...... (2) 
 
 As one molecule of potassium dichromate liberates three atoms of 
 oxygen equation (2) must be multiplied by three, giving: 
 
 =3SnCl 4 +3H 2 0. .-. . . .(3) 
 
 By combining equations (1) and (3) we obtain the complete 
 equation as follows : 
 
 K,Cr a 7 + 3SnCl 2 + 14HC1 =2KC1+ 2CrCl 3 + 3SnCl 4 + 7H 2 0. 
 
 CALCULATION OF VOLUMETRIC DETERMINATIONS. 
 
 As volumetric solutions are generally made on the normal 
 basis of strength the calculation of the results is quite simple. 
 Roundabout and very cumbersome methods of calculation are 
 nevertheless often used. 
 
 756. Standard Solutions. When solutions such as hydrochloric 
 or sulphuric acid have been standardized by weighing precipi- 
 tates of silver chloride or barium sulphate the weight of the 
 acid radicle present in the precipitate may be computed. From 
 this weight the corresponding weight of acid may be com- 
 puted, and finally the ratio of this weight to the theoretical 
 weight for a normal solution may be calculated. The same 
 
500 STOICHIOMETRY. 
 
 result may, however, be obtained by a single calculation. As 
 a liter of normal hydrochloric acid contains the gram molecular 
 weight of the acid it must give a gram molecular weight of sil- 
 ver chloride according to the equation, 
 
 HC1 + AgN0 3 = AgCl + HN0 3 . 
 
 If 10 c.c. of the acid has been taken for the determination 
 the silver chloride precipitate should weigh 1.4338 or I/ 100th of 
 the gram molecular weight. The ratio of this number to the 
 weight found will give the strength of the acid in terms of normal. 
 For example, if 1.6728 grams of AgCl were obtained, the acid is 
 
 (1 A7OQ\ 
 i' 4 oQQJ- To dilute the acid to exact strength there 
 
 must be added 166.7 c.c. of water to one liter of the acid. If the 
 
 -I 4QQC \ 
 
 ' - 00 ), the number .85712 is obtained 
 
 and the dilution must be made by diluting 857.12 c.c. of the 
 strong acid to one liter. 
 
 757. Acidimetry and Alkalimetry. Having obtained a normal 
 acid solution the calculation of the amount of any substance 
 titrated is equally simple. For all univalent bases titrated one 
 liter of the acid is equal to the gram molecular weight of the base. 
 For example, one liter of a normal acid, whether it is sulphuric 
 or hydrochloric, is equal to 17.034 grams of ammonia, while one 
 c.c. is equal to one-thousandth of this amount or .017034 gram. 
 If 10 c.c. of acid has been used to titrate a given amount of ammo- 
 nia, there is present .17034 gram of NH 3 . If this ammonia has 
 been formed from the nitrogen contained in a substance analyzed, 
 the amount of nitrogen may be computed directly, since one liter 
 of a normal acid will be equivalent to 14.01 grams of nitrogen in 
 the form of ammonia. If, for example, the ammonia has been 
 obtained from an ammonium salt, such as ammonium sulphate, 
 the value of the normal acid in terms of this salt is obtained from 
 its molecular weight, half of which is taken since each molecule 
 contains two molecules of ammonia. 
 
 The value of a standard acid in terms of a given salt will some- 
 times vary with the indicator used, as already explained in Chap- 
 ter XX, page 250. If disodic phosphate is titrated with a strong 
 
CALCULATION OF VOLUMETRIC DETERMINATIONS. 501 
 
 acid, using methyl orange as the indicator, the following reaction 
 will take place: 
 
 Na 2 HP0 4 +HCl =NaCl+NaH 2 P0 4 . 
 
 A liter of normal acid will therefore be equal to a gram molecular 
 weight of disodic phosphate, which will be 358.4 grams of the crys- 
 tallized salt. While the disodic phosphate is alkaline to methyl 
 orange it is neutral to phenolphthalein, to which the trisodium salt 
 is alkaline. If the latter salt is titrated with acid, using phenol- 
 phthalein as the indicator, the following reaction takes place: 
 
 Na 3 P0 4 + HC1 =Na 2 HP0 4 + NaCl. 
 
 One liter of a normal acid is therefore equal to the gram molecular 
 weight of this salt or 380.44 grams of the crystallized salt. A 
 mixture of trisodic and disodic phosphates may therefore be ana- 
 lyzed by titration with a standard acid, using phenolphthalein and 
 methyl orange as the indicators. The method of calculation is 
 seen from the following illustration: One gram of a sample of 
 disodic phosphate was titrated with fifth-normal acid. With 
 phenolphthalein 1.50 c.c. of acid were used, and when methyl 
 orange had been added 13.5 c.c. acid were required to again give 
 the acid reaction. The acid used with the phenolphthalein rep- 
 resents the amount of trisodic phosphate present, 1 c.c. of fifth- 
 normal acid being equal to .07608 gram of this salt. 
 
 IOTT^ 1. 50 X. 07608 
 Percentage of Na 3 P0 4 .12H 2 = - OOQ =11.41. 
 
 The 13.50 c.c. of acid used with the methyl orange neutralized the 
 disodic phosphate present in the original material and that pro- 
 duced by the titration with phenolphthalein. The acid used for 
 the latter purpose will be exactly equal to that used with the 
 phenolphthalein, in this case 1.50 c.c. The acid necessary to 
 neutralize the disodic phosphate originally present will be 12.00 
 c.c. (13.50-1.50). The percentage of Na 2 HP0 4 .12H 2 will be 
 
 12.00 X. 0717 
 
502 STOICHIOMETRY. 
 
 758. Percentage Given by Number of Cubic Centimeters of 
 Acid Used. Frequently it is convenient to use a solution of 
 such a strength that when a given weight of substance is 
 taken the number of cubic centimeters used in the titration 
 shall be equal to the percentage of a given constituent. For 
 this purpose the solution must be made of such a strength 
 that 100 c.c. are equal to the amount of substance to be 
 weighed out, provided the substance is absolutely pure. If sodium 
 carbonate is the substance to be titrated and one gram is the 
 amount to be weighed out, 100 c.c. of the acid must be equal to 
 1 gram of pure sodium carbonate. As 100 c.c. of a normal acid 
 are equal to 5.3 grams of sodium carbonate, the acid required must 
 be -|rd normal. If an acid double this strength is used the num- 
 ber of cubic centimeters used must be multiplied by 2 to give the 
 percentage. The amount of substance to be weighed out may 
 also be so chosen that the percentage shall be equal to the number 
 of cubic centimeters used in a titration. The acid may then be of 
 any desired strength. The amount of substance weighed out must 
 in every case be equal to the weight of the pure substance, which 
 will just neutralize 100 c.c. of the acid. In the case of a fifth-nor- 
 mal acid 100 c.c. are equal to 1.06 grams of pure sodium carbonate. 
 If this amount of the impure article be weighed out and titrated 
 with the acid the number of cubic centimeters used will be equal 
 to the percentage of Na 2 C0 3 present. This is evident from the 
 consideration that while 1.06 grams of the pure compound will 
 require 100 c.c., any decrease in the purity of the substance will 
 produce a proportionate decrease in the amount of acid used in 
 its titration. 
 
 759. Oxidation and Reduction Titrations. In calculating the 
 results obtained by means of oxidizing and reducing solutions 
 the amount of oxygen absorbed or . liberated by the substances 
 acted upon must always be kept in mind. Equivalent weights 
 of reducing or oxidizing substances absorb or give up the same 
 amount of oxygen. It has already been shown that 31.63 grams 
 of KMn0 4 contain the same amount (8 grams) of available oxygen 
 as 49.05 grams of K 2 Cr 2 7 . A liter of solution containing either 
 of these -amounts of oxidizing substances is a normal solution. 
 120.97 grams of iodine liberate the same amount of oxygen to 
 
CALCULATION OF VOLUMETRIC DETERMINATIONS. 503 
 
 reducing substances, and dissolved in a liter give a normal oxidiz- 
 ing solution. These different amounts of oxidizing substances 
 are therefore equal to each other, because the same amount of 
 free oxygen is furnished by each. 
 
 The method of shortening calculations by applying this princi- 
 ple may be seen from the following examples: A potassium per- 
 manganate solution which had been standardized by means of 
 pure iron wire was used to standardize a sodium thiosulphate 
 solution. The standardization of the potassium permanganate 
 solution with iron takes place according to the following equation: 
 
 2KMn0 4 + 10FeS0 4 + 8H 2 S0 4 =K 2 S0 4 + 5Fe 2 (S0 4 ) 3 + 2MnS0 4 
 
 + 8H 2 0. 
 
 Two molecules of potassium permanganate are therefore equal to 
 ten atoms of iron. A measured volume of the permanganate 
 solution was acidified with hydrochloric acid after the addition of 
 potassium iodide. Iodine was liberated according to the following 
 equation : 
 
 KMn0 4 + 8HC1 + 5KI = 51 + 6KC1 + MnCl 2 + 4H 2 0. 
 
 The iodine liberated is then titrated with sodium thiosulphate 
 solution. By this equation two molecules of potassium perman- 
 ganate are equal to ten atoms of iodine. As by the first equation 
 the same amount of potassium permanganate is equal to ten atoms 
 of iron, one atom of iron is equal to one atom of iodine; that is, 
 55.9 parts of iron are equal to 126.97 parts of iodine. As 1 c.c. of 
 the potassium permanganate solution was found equal to .01115 
 gram of iron the value of 1 c.c. in iodine is obtained from the 
 following proportion: 
 
 55.9 : 126.85 :: .01115:* 
 x = . 0253 gram. 
 
 As 25 c.c. of the potassium permanganate solution required 23 c.c. 
 of the thiosulphate solution to titrate the iodine liberated, the 
 
 25 X 0253 
 
 iodine value of 1 c.c. of the latter solution is equal to ^ , or 
 
 & 
 
 .0275 gram. 
 
504 STOICHIOMETRY 
 
 760. lodometric Titrations. In a similar manner the value of 
 the thiosulphate solution in terms of any substance it may be used 
 to titrate may be calculated. The intermediate reactions are used 
 simply to find the number of iodine atoms equivalent to one mole- 
 cule of the substance titrated. For example, manganese dioxide 
 may be boiled with hydrochloric acid, the chlorine evolved being 
 absorbed in potassium iodide, and the iodine liberated titrated 
 with the thiosulphate solution. The following reactions take 
 place : 
 
 Mn0 2 +4HCl =MnCl 2 +Cl 2 +2H 2 
 
 C1 2 +2KI=2KC1+I 2 
 I 2 +Na 2 S 2 3 =2NaI+Na 2 S 4 6 . 
 
 One molecule of manganese dioxide is therefore equal to two atoms 
 of iodine. The value of 1 c.c. of the sodium thiosulphate solution 
 in terms of manganese dioxide is obtained from the following pro- 
 portion: 
 
 I 2 : Mn0 2 : : .0275 : x 
 253.7: 87 : : .0275 : x 
 =.00943 
 
APPENDIX. 
 
 REAGENTS. 
 
 IT is important that the reagents used should be of known defi- 
 nite strength and that the strength should be indicated on the 
 bottles. So far as possible also the reagents should be of uniform 
 strength. This is not always possible on account of the solubilities 
 of the salts employed. The strength is most easily indicated by 
 the normal system of nomenclature. Chemically pure material 
 should always be used. The impurities which may be present 
 and the tests to be applied are given in the notes on pp. 32-35. 
 Many salts when in solution, especially those that react alkaline, 
 act on the glass becoming contaminated with silica and other 
 constituents of the glass. Only recently prepared solutions of 
 such reagents should be used. It is therefore advisable to keep 
 the dry salts at hand and make up the solutions only as required 
 for immediate use. 
 
 REAGENTS OF FIVE TIMES NORMAL STRENGTH. LABEL, 5 N. 
 
 Sulphuric Acid, H 2 S0 4 Equivalent =49 
 
 Ordinary concentrated acid, sp. gr. 1.84, has a strength thirty- 
 six times normal, and may be designated 36 N. One volume of 
 strong acid diluted with 6 volumes of water = 5 N. 
 
 Nitric Acid, HN0 3 Equivalent =63 
 
 Concentrated acid, sp. gr. 1.40 = 15 N. One volume of strong 
 acid diluted with 2 volumes of water =5 N. 
 
 Hydrochloric Acid, HC1 Equivalent =36.5 
 
 Concentrated acid, sp. gr. 1.20 = 13 N. Five volumes of strong 
 acid diluted with 8 volumes of water =5 N. 
 
 505 
 
506 APPENDIX. 
 
 Acetic Acid, H.C 2 H 3 2 Equivalent =60 
 
 Glacial acid, M.P. 10 C. ; = 17 N. One volume of glacial acid 
 diluted with 2J volumes of water =5 N. 
 
 Potassium Hydroxide, KOH Equivalent =56 
 
 280 grams dissolved in water and diluted to one liter =5 N. 
 
 Sodium Hydroxide, NaOH Equivalent =40 
 
 200 grams dissolved in water and diluted to one liter =5 N. 
 
 Ammonia (solution in water, NH 4 OH) Equivalent =35 
 
 The strong solution, sp. gr. 0.90 = 15 N. One volume of the 
 strong solution diluted with 2 volumes of water =5 N. 
 
 Ammonium Sulphide, (NH 4 ) 2 S Equivalent =34 
 
 600 c.c. of 5N ammonia are saturated with sulphuretted 
 hydrogen; this gives hydrogen ammonium sulphide, NH 4 HS. 
 This is made up to 1 liter by adding 5N ammonia. (This 
 reagent is slowly decomposed by atmospheric oxygen, ammo- 
 nia is evolved, and yellow ammonium sulphide, (NH 4 ) 2 S 2 , is 
 formed.) 
 
 Sodium Sulphide, Na 2 S Equivalent =39 
 
 "200 grams of sodium hydroxide are dissolved in 800 c.c. of 
 water. 400 c.c. of this solution are saturated with sulphuretted 
 hydrogen and the remaining half added, together with water 
 sufficient to make the volume up to 1 liter. When hydrogen 
 sodium sulphide, NaHS, is required, the sodium hydroxide is 
 simply saturated with sulphuretted hydrogen, without the further 
 addition of sodium hydroxide. 
 
 Ammonium Chloride, NH 4 C1 Equivalent =53.5 
 
 267.5 grams of the salt dissolved in water and diluted to one 
 liter- 5 N. 
 
 Ammonium Carbonate, (NH 4 ) 2 C0 3 Equivalent =48 
 
 200 grams of ammonium sesquicarbonate (commercial carbon- 
 ate) dissolved in 350 c.c. of 5 N ammonia, and diluted with 
 water to one liter = 5 N. 
 
 Ammonium Acetate, NH 4 C 2 H 3 2 Equivalent =77 
 
 To 300 c.c. of glacial acetic acid (17 N) an equal volume of 
 water is added, the acid neutralized with strong ammonia and 
 diluted to one liter =5 N. 
 
REAGENTS. 507 
 
 REAGENTS OF NORMAL STRENGTH. LABEL, N. 
 
 The following normal reagents are prepared by dissolving the 
 equivalent weight in grams of the various salts in water, and 
 diluting to one liter. In all cases the nearest whole number to 
 the equivalent weight may be taken as sufficiently exact. 
 
 Barium chloride, BaCl 2 .2H 2 Equivalent weight 122.0 
 
 Disodium phosphate, HNa 2 PO 4 .12H 2 0. " " 119.3 
 
 Lead acetate, Pb(C 2 H 3 2 ) 2 .3H 2 " " 189 . 5 
 
 " Magnesia mixture " (MgCl 2 . (NH 4 C1) 2 , and NH 4 OH). 
 68 grams of MgCl 2 .6H 2 0, together with 165 grams of ammo- 
 nium chloride, are dissolved in 300 c.c. of water; 300 c.c. of 5 N 
 ammonia are added, and the solution diluted with water to one 
 liter. 
 
 Stannous chloride, SnCl 2 .2H 2 Equivalent weight 112.5 
 
 112 grams of the salt are dissolved in 200 c.c. of 5 N hydro- 
 chloric acid and the solution diluted with water to one liter. 
 Fragments of granulated tin should be placed in the solution. 
 
 REAGENTS OF VARIOUS STRENGTHS. 
 
 Ammonium oxalate, (NH 4 ) 2 C 2 4 .2H 2 0. Equivalent weight 80.0 
 
 N 
 
 40 grams dissolved in one liter = solution. 
 
 A 
 
 Molybdate Solution. The molybdate solution is made as fol- 
 lows: 75 grams of ammonium molybdate are dissolved in 500 c.c. 
 of water with the addition of a little ammonia if necessary. If 
 still turbid, the solution is filtered and then poured with constant 
 stirring into 500 c.c. of a mixture of 250 c.c. of concentrated nitric 
 acid (sp. gr. 1.40) and 250 c.c. of water. The solution may also 
 be made from molybdic oxide by dissolving 60 grams in 440 c.c. of 
 water and 60 c.c. strong ammonia (sp. gr. 0.90), and pouring the 
 solution into 500 c.c. of nitric acid diluted as already directed. 
 The freshly made molybdate solution must be allowed to stand in 
 a warm place for several days. The clear solution is decanted or 
 filtered off for use. Large amounts of the solution should not be 
 
503 APPENDIX. 
 
 made up at one time, since it will not keep more than several 
 months, as the molybdic acid is slowly precipitated. 
 
 Mercuric chloride, HgCl 2 Equivalent weight 135.5 
 
 N 
 
 27 grams dissolved in one liter =- solution. 
 
 5 
 
 Silver nitrate, AgN0 3 Equivalent weight 170 . 
 
 N 
 
 3.4 grams dissolved in 100 c.c. =7- solution. 
 
 o 
 
 Potassium dichromate, K 2 Cr 2 7 Equivalent weight 294.5 
 
 100 grams are dissolved in one liter of water. 
 
TABLE I. INTERNATIONAL ATOMIC WEIGHTS 
 FOR 1909* 
 
 Name. 
 
 Symbol. 
 
 Atomic 
 Weight. 
 
 Name. 
 
 Smybol. 
 
 Atomic 
 Weight. 
 
 Aluminium 
 
 Al 
 Sb 
 A 
 As 
 Ba 
 Bi 
 B 
 Br 
 Cd 
 Cs 
 Ca 
 C 
 Ce 
 Cl 
 Cr 
 Co 
 Cb 
 Cu 
 Dy 
 Er 
 Eu 
 F 
 Gd 
 Ga 
 Ge 
 Gl 
 Au 
 He 
 H 
 In 
 I 
 Ir 
 Fe 
 Kr 
 La 
 Pb 
 Li 
 Lu 
 Mg 
 3 n 
 Hg 
 
 27.1 
 120.2 
 39.9 
 75.0 
 137.37 
 208.0 
 11.0 
 79.92 
 112.40 
 132.81 
 40.09 
 12.00 
 140.25 
 35.46 
 52.1 
 58.97 
 93.5 
 63.57 
 162.5 
 167.4 
 152.0 
 19.0 
 157.3 
 69.9 
 72.5 
 9.1 
 197.2 
 4.0 
 1.008 
 114.8 
 126.92 
 193.1 
 55.85 
 81.8 
 139.0 
 207.10 
 7.00 
 174.0 
 24.32 
 54.93 
 200 
 
 Molybdenum 
 Neodymium . 
 
 Mo 
 Nd 
 Ne 
 Ni 
 N 
 Os 
 
 Pd 
 P 
 Pt 
 K 
 Pr 
 Ra 
 Rh 
 Rb 
 Ru 
 Sm 
 Sc 
 Se 
 Si 
 
 A 
 
 Sr 
 S 
 Ta 
 Te 
 Tb 
 Tl 
 Th 
 Tm 
 Sn 
 Tl 
 W 
 U 
 V 
 Xe 
 Yb 
 
 Yt 
 Zn 
 Zr 
 
 96.0 
 144.3 
 20.0 
 58.68 
 14.01 
 190.9 
 16.00 
 106.71 
 31.0 
 195.0 
 39.10 
 140.6 
 226.4 
 102.9 
 85.45 
 101.7 
 150.4 
 44.1 
 79.2 
 28.3 
 107.88 
 23.00 
 87.62 
 32.07 
 181.0 
 127.5 
 159.2 
 204.0 
 232.42 
 168.5 
 119.0 
 48.1 
 184 
 238.5 
 51.2 
 128 
 172.0 
 
 89.0 
 65.37 
 90.6 
 
 Antimony . . .. . . * 
 A rtron 
 
 Neon 
 
 
 Nickel 
 
 Barium 
 
 Nitrogen 
 
 Bismuth. 
 
 
 Boron 
 
 
 Bromine 
 
 
 Cadmium . 
 
 
 Caesium . . 
 
 Platinum 
 
 Calcium . . 
 
 Potassium 
 
 Carbon 
 
 Praseodymium . . . 
 Radium 
 
 
 Chlorine 
 
 Rhodium 
 
 Chromium ... 
 
 
 Cobalt 
 
 
 
 Samarium 
 
 Copper 
 
 Scandium .... 
 
 Dysprosium 
 
 
 
 Silicon 
 
 Europium . 
 
 Silver 
 
 Fluorine 
 
 Sodium 
 
 Gadolin ium 
 
 Strontium 
 
 Gallium 
 
 Sulphur 
 
 Germanium 
 
 Tantalum 
 
 
 Tellurium 
 
 Gold 
 
 Terbium .... 
 
 HeliuTn . . . 
 
 Thallium 
 
 Hydrogen .... 
 
 
 Indium 
 
 
 Iodine 
 
 Tin 
 
 Iridium 
 
 
 Iron 
 
 Tungsten 
 
 Krypton 
 Lanthanum 
 
 Uranium 
 
 Vanadium . . 
 
 Lead 
 Lithium ... . 
 
 Xenon . . . 
 
 Ytterbium 
 
 Lutecium. 
 
 1 (Neoyt terbium). 
 Yttrium 
 
 Magnesium 
 
 Manganese 
 
 Zinc . 
 
 Mercury 
 
 Zirconium 
 
 
 * Compiled by the International Committee on Atomic Weights con- 
 sisting of F. W. Clarke, W. Ostwald, T. E. Thorpe, and G. Urbain. 
 
 509 
 
10 TABLES. 
 
 TABLE II. CHEMICAL FACTORS AND THEIR LOGARITHMS. 
 
 
 Atomic 
 Weight. 
 
 Weighed. 
 
 Required 
 
 Factor. 
 
 Per Cent. 
 
 Loga- 
 rithms.* 
 
 Aluminium .... 
 Antimony 
 
 27.1 
 
 120.2 
 75.0 
 
 137.37 
 
 208.0 
 
 79.92 
 112.40 
 
 40.09 
 
 12.00 
 35.46 
 
 52.1 
 
 58.97 
 C3.57 
 
 Al,0 3 
 
 A1P0 4 
 ii 
 
 Sb a O 4 
 
 As a S 3 
 
 Mg a As a O 7 
 tt 
 
 
 
 BaSO 4 
 (i 
 
 14 
 
 
 
 BaCOs 
 
 it 
 
 o 
 
 
 BaCrO 4 
 ii 
 
 Bi,0 3 
 BiOCl 
 
 AgBr 
 (i 
 
 CdO 
 
 CdS 
 
 u 
 
 CaO 
 
 
 CaCO 3 
 u 
 
 CaSO* 
 
 CO 2 
 AgCl 
 
 Ag 
 
 u 
 
 Cr 3 O, 
 
 it 
 
 PbCrO 4 
 ii 
 
 < 
 tt 
 
 BaCrO 4 
 11 
 ii 
 (t 
 Co 
 CoSO 4 
 
 14 
 
 2CoSO 4 .3K 2 SO 4 
 
 14 
 
 Cu 
 
 Al 
 
 Al 
 A1 3 O, 
 Sb 
 As 
 As,O 3 
 As 
 As,0 9 
 As a O, 
 Ba 
 BaCl a 
 BaCl a .2H 2 
 BaO 
 Ba 
 BaO 
 C 
 CO 2 
 Ba 
 BaO 
 Bi 
 Bi 
 Bi a 3 
 Br 
 HBr 
 Cd 
 Cd 
 CdO 
 Ca 
 CaCO, 
 Ca 
 CaO 
 Ca 
 CaO 
 C 
 Cl 
 HCI 
 Cl 
 HCI 
 Cr 
 CrO, 
 Cr0 4 
 Cr 
 CrO 3 
 CrO 4 
 Cr a O 3 
 Cr 
 CrO 3 
 CrO 4 
 Cr 2 O 3 
 CoO 
 Co 
 CoO 
 Co 
 CoO 
 CuO 
 CuSO 4 .5H 2 O 
 
 53.03 
 22.20 
 41.85 
 78.97 
 60.92 
 80.42 
 48.29 
 63.74 
 74.04 
 58.85 
 89.23 
 104.66 
 65.71 
 69.61 
 77.71 
 06.08 
 22.29 
 54.20 
 60.51 
 89.65 
 80.17 
 89.42 
 42.56 
 43.11 
 87.54 
 77.80 
 88.88 
 71.47 
 178.44 
 40.06 
 56.04 
 29.44 
 41.20 
 27.27 
 24.74 
 25.44 
 32.87 
 33.80 
 68.46 
 131.54 
 152.57 
 16.12 
 30.97 
 35.92 
 23.55 
 20.55 
 39.49 
 45.80 
 30.02 ' 
 127.14 
 38.04 
 48.36 
 14.16 
 18.00 
 125.17 
 392.83 
 
 9.72455 
 9.34625 
 9.62170 
 9.89749 
 9.78478 
 9.90536 
 9.68383 
 9.80441 
 9.86947 
 9.76975 
 9.95054 
 0.01982 
 9.81760 
 9.84261 
 9.89046 
 8.78390 
 9.34817 
 9.73396 
 9.78181 
 9.95257 
 9.90399 
 9.95142 
 9.62896 
 9.63461 
 9.94220 
 9.89099 
 9.94879 
 9.85415 
 0.25150 
 9.C0265 
 9 74850 
 9.46899 
 9.61484 
 9.43573 
 9.39337 
 9.40543 
 9.51680 
 9.52886 
 9.83546 
 0.11905 
 0.18345 
 9. 20737 
 9. 49096 
 9.55536 
 9.37191 
 9.31291 
 9.59650 
 9.66090 
 9.47745 
 0.10426 
 9. 58019 
 9.68445 
 9.15107 
 9.25534 
 0.09750 
 0.59420 
 
 
 
 
 Cadmium .... 
 
 
 Carbon 
 
 
 Chromium 
 Cobalt 
 
 Copper . 
 
 
 * The-10 after each of the logarithms has been omitted. 
 
TABLES. 511 
 
 CHEMICAL FACTORS AND THEIR LOGARITHMS- (Continued). 
 
 
 Atomic 
 Weight. 
 
 Weighed. 
 
 Required. 
 
 Factor. 
 
 Per Cent. 
 
 Loga- 
 rithms. 
 
 Fluorine 
 
 19.0 
 
 1.008 
 126-92 
 
 55.85 
 207.10 
 
 24.32 
 54.93 
 
 200.0 
 
 58.68 
 14.01 
 
 CuO 
 
 Cu 2 8 
 
 H 
 
 CuCNS 
 CaF 2 
 
 (4 
 
 H 2 
 Agl 
 
 PdI 2 
 Fe 2 3 
 FeP0 4 
 
 a 
 
 PbO 
 PbO 2 
 PbS 
 
 PbS0 4 
 PbCl 2 
 PbCr0 4 
 
 MgO 
 Mg 2 P 2 7 
 
 MgS0 4 
 
 Mn 3 O 4 
 
 i 
 
 MnS 
 
 ( t 
 
 Mn 2 P 2 O 7 
 
 1 1 
 
 MnSO 4 
 
 
 Hg 
 HgS 
 
 a 
 
 HgCl 
 
 (t 
 
 Ni 
 NiO 
 NiSO 4 
 
 (NH 4 ) 2 PtCl 8 
 
 i 
 C 20 H 16 N 4 .HNO 
 
 Cu 
 Cu 
 CuO 
 Cu 
 CuO 
 F 
 HF 
 H 
 I 
 HI 
 HI 
 I 
 Fe 
 FeO 
 Fe 
 FeO 
 Pb 
 Pb 
 Pb 
 PbO 
 Pb 
 PbO 
 Pb 
 PbO 
 Pb 
 PbO 
 Mg 
 Mg 
 MgO 
 Mg 
 MgO 
 Mn 
 MnO 
 MnO 2 
 Mn 
 MnO 
 Mn 
 MnO 
 Mn 
 MnO 
 HgO 
 Hg 
 HgCl 2 
 HgO 
 Hg 
 HgO 
 NiO 
 Ni 
 Ni 
 NiO 
 N 
 NH 3 
 NH 4 
 HNO 3 
 N 2 5 
 
 79.89 
 79.86 
 99.96 
 52.26 
 65.41 
 48.66 
 51.24 
 11.19 
 54.06 
 54.48 
 70.97 
 70.40 
 69.94 
 89.98 
 37.02 
 47.63 
 92.83 
 86.62 
 86.59 
 93.28 
 68.31 
 73.59 
 74.49 
 80.25 
 64.08 
 69.03 
 60.32 
 21.85 
 36.22 
 20.20 
 33.49 
 72.03 
 93.01 
 113.98 
 63.14 
 81.53 
 38.70 
 49.98 
 36.38 
 46.97 
 108.00 
 86.18 
 116.74 
 93.07 
 84.94 
 91.73 
 127.26 
 78.58 
 37.92 
 48.26 
 6.31 
 7.68 
 8.13 
 15.47 
 12.87 
 
 9.90250 
 9.90232 
 9.99981 
 9.71814 
 9.81564 
 9.68718 
 9.70964 
 9.04884 
 9.73283 
 9.73627 
 9.85704 
 9.84760 
 9-84475 
 9.95416 
 9.56848 
 9.67789 
 9.96768 
 9.93760 
 9.93747 
 9.96979 
 9.83449 
 9.86681 
 9.87211 
 9.90443 
 9.80671 
 9.83903 
 9.78044 
 9.33938 
 9.55894 
 9.30538 
 9-52493 
 9.85749 
 9.96851 
 0.05685 
 9.80029 
 9.91131 
 9.58774 
 9.69876 
 9.56083 
 9.67185 
 0.03342 
 9.93541 
 0.06722 
 9.96883 
 9.92911 
 9.96253 
 9.10471 
 9.89529 
 9.57886 
 9.68357 
 8.80024 
 8 88514 
 8.91004 
 9.18952 
 9.10958 
 
 Hydrogen. 
 Iodine 
 
 
 Lead .... 
 
 Magnesium . . . 
 Manganese . . - 
 
 Mercury. . . 
 
 Nickel . . . 
 
 Nitrogen . . . 
 
 
512 TABLES. 
 
 CHEMICAL FACTORS AND THEIR LOGARITHMS (Continued). 
 
 
 Atomic 
 Weight. 
 
 Weighed. 
 
 Required. 
 
 Factor. 
 
 Per ( ent. 
 
 Loga- 
 rithms.* 
 
 Platinum 
 
 195.0 
 
 Pt 
 
 N 
 
 14.37 
 
 9.15744 
 
 
 
 ti 
 
 NH, 
 
 17.47 
 
 9.24234 
 
 
 
 i < 
 
 NH 4 
 
 18.50 
 
 9.26724 
 
 Phosphorus .... 
 
 31.0 
 
 Mg 2 P 3 7 
 
 P 
 
 27.85 
 
 9.44478 
 
 
 
 
 P0 4 
 
 85.34 
 
 9.93114 
 
 
 
 It 
 
 P.0 5 
 
 63.78 
 
 9.80468 
 
 Molybdenum... 
 
 96.0 
 
 NH 4 ),PO 4 .12MoO, 
 
 P 
 
 1.65 
 
 8.21787 
 
 
 
 " 
 
 P0 4 
 
 5.06 
 
 8.70423 
 
 
 
 (i 
 
 F.O, 
 
 3.78 
 
 8.57777 
 
 Potassium 
 
 39.10 
 
 KC1 
 
 K 
 
 52.44 
 
 9.71967 
 
 
 
 (i 
 
 K 2 O 
 
 63.17 
 
 9.80051 
 
 
 
 K,PtCl 6 
 
 K 
 
 16.09 
 
 9.20661 
 
 
 
 " 
 
 K 2 
 
 19.38 
 
 9.28745 
 
 
 
 K 2 S0 4 
 
 K 
 
 44.87 
 
 9.65199 
 
 
 
 n 
 
 K 2 O 
 
 54.06 
 
 9.73283 
 
 Selenium 
 
 79.2 
 
 Se 
 
 SeO 2 
 
 140.40 
 
 0.14737 
 
 Silicon 
 
 28.3 
 107.88 
 
 Si0 2 
 AgCl 
 
 Si 
 Ag 
 
 46.93 
 75.26 
 
 9.67147 
 9.87657 
 
 Silver 
 
 
 
 AgBr 
 
 . & 
 
 Ag 
 
 57.44 
 
 9.75924 
 
 
 
 Agl 
 
 Ag 
 
 45.95 
 
 9.66224 
 
 
 23.00 
 
 Ag 2 S 
 Nad 
 
 Ag 
 
 Na 
 
 87.06 
 39.34 
 
 9.93982 
 9.59487 
 
 
 
 
 Na 2 O 
 
 53.03 
 
 9^72451 
 
 
 
 Na 2 SO 4 
 
 Na 
 
 32.38 
 
 9.51026 
 
 
 
 < < 
 
 Na 2 
 
 43.64 
 
 9. 63989 
 
 
 
 Na 2 C0 3 
 
 Na 
 
 43.40 
 
 9.63745 
 
 
 
 <. 
 
 Na 2 
 
 58.49 
 
 9.76708 
 
 Strontium 
 
 87.62 
 
 SrCO, 
 
 Sr 
 
 59.35 
 
 9.77345 
 
 
 
 " 
 
 SrO 
 
 70.19 
 
 9.84G29 
 
 
 
 SrS0 4 
 
 Sr 
 
 47.70 
 
 9.67852 
 
 
 
 " 
 
 SrO 
 
 56.41 
 
 9.75136 
 
 Sulphur 
 
 32.07 
 
 BaS0 4 
 
 S 
 
 13.74 
 
 9.13793 
 
 
 
 11 
 
 S0 2 
 
 27.45 
 
 9.43S48 
 
 
 
 (i 
 
 S0 3 
 
 34.30 
 
 9.53530 
 
 
 
 
 
 S0 4 
 
 41.16 
 
 9.61442 
 
 
 
 " 
 
 H 2 S0 4 
 
 42.02 
 
 9.C2345 
 
 Tellurium 
 
 127.5 
 
 Te 
 
 Te0 2 
 
 125.09 
 
 0.09725 
 
 
 
 . Sn 
 
 SnO 
 
 113.44 
 
 0.05478 
 
 
 
 M 
 
 SnO 2 
 
 126.89 
 
 0.10343 
 
 Tin 
 
 119.0 
 
 SnO, 
 
 Sn 
 
 78.81 
 
 9.89657 
 
 Titanium 
 
 48.1 
 
 k '" v^2 
 
 Ti0 2 
 
 Ti 
 
 60.05 
 
 9.77852 
 
 Zinc 
 
 65.37 
 
 ZnO 
 
 Zn 
 
 80.34 
 
 9.90492 
 
 
 
 Zn 2 P 2 O t 
 
 Zn 
 
 42.90 
 
 9. 63248 
 
 
 
 
 ZnO 
 
 53.40 
 
 9.72756 
 
 
 
 ZnS 
 
 Zn 
 
 67.00 
 
 9.82664 
 
 
 
 it 
 
 ZnO 
 
 83.51 
 
 9.92172 
 
TABLES OF LOGAKITHMS AND 
 ANTILOGARITHMS. 
 
514 
 
 TABLES. 
 
 TABLE III. LOGARITHMS. 
 
 Natural 
 Numbers. 
 
 
 
 1 
 
 o 
 
 3 
 
 4 
 
 5 
 
 C 
 
 
 
 r 
 
 8 
 
 9 
 
 PROPORTIONAL PARTS. 
 
 1 
 
 9 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 0000 
 
 0043 
 
 0086 
 
 0128 
 
 0170 
 
 0212 
 
 0253 
 
 0294 
 
 0334 
 
 0374 
 
 4 
 
 8 
 
 12 
 
 17 
 
 21 
 
 2529 
 
 33 
 
 37 
 
 11 
 
 0414 
 
 0453 
 
 0492 
 
 05310569 
 
 0607 
 
 0645 
 
 0682 
 
 0719 
 
 0755 
 
 4 
 
 8 
 
 11 
 
 15 
 
 19 
 
 2326 
 
 30 
 
 34 
 
 12 
 
 0792 
 
 0328 
 
 0864 
 
 08990934 
 
 0969 
 
 1004 
 
 1038 
 
 10721106 
 
 3 
 
 7 
 
 10 14 
 
 17 
 
 21 24 28 
 
 31 
 
 13 
 
 1139 
 
 1173 
 
 1206 
 
 1239 1271 
 
 1303 
 
 1335 
 
 1367 
 
 1399 1430 
 
 3 
 
 6 
 
 1013 
 
 16 
 
 19 23 26 
 
 29 
 
 14 
 
 1461 
 
 1492 
 
 1523 
 
 1553 1584 
 
 1614 
 
 1644 
 
 1673 
 
 1703 1732 
 
 3 
 
 G 
 
 9 
 
 12 
 
 15 
 
 1821 24 
 
 27 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 15 
 
 1761 
 
 1790 
 
 1818 
 
 1847 
 
 1875 
 
 1903 
 
 1931 
 
 1959 
 
 1987 
 
 2014 
 
 3 
 
 6 
 
 8 
 
 11 
 
 14 
 
 17 20 22 
 
 25 
 
 16 
 
 2041 
 
 2068 
 
 2095 
 
 2122 
 
 214S 
 
 2175 
 
 2201 
 
 2227 
 
 2253 
 
 2279 
 
 3 
 
 5 
 
 8 
 
 11 
 
 13 
 
 16 18 21 
 
 24 
 
 17 
 
 2304 
 
 2330 
 
 2355 
 
 2380 
 
 2405 
 
 2430 
 
 2455 
 
 2480 
 
 2504 
 
 2529 
 
 2 
 
 5 
 
 7 
 
 10 
 
 12 
 
 15 17 20 
 
 22 
 
 18 
 
 2553 
 
 2577 
 
 2601 
 
 2625 
 
 2648 
 
 2672 
 
 2695 
 
 2718 
 
 2742 
 
 2765 
 
 2 
 
 5 
 
 7 
 
 9 
 
 12 
 
 14 16 19 
 
 21 
 
 19 
 
 2788 
 
 2810 
 
 2833 
 
 2856 
 
 2878 
 
 2900 
 
 2923 
 
 2945 
 
 2967 
 
 2989 
 
 2 
 
 4 
 
 7 
 
 9 
 
 11 
 
 13 16 18 
 
 20 
 
 20 
 
 30103032 
 
 3054 
 
 3075 3096 
 
 3118 
 
 3139 
 
 3160 
 
 3181 
 
 3201 
 
 2 
 
 4 
 
 6 
 
 8 
 
 11 
 
 1315 
 
 17 19 
 
 21 
 
 32223243 
 
 3263 
 
 3284 3304 
 
 3324 
 
 3345 
 
 3365 
 
 3385 
 
 3404 
 
 2 
 
 4 
 
 6 
 
 8 
 
 10 
 
 12 14 16 18 
 
 22 
 
 3424 3444 
 
 3464 
 
 348313502 
 
 3522 
 
 3541 
 
 3560 
 
 3579 
 
 3598 
 
 2 
 
 4 
 
 6 
 
 8 
 
 10 
 
 12 14 15 17 
 
 23 
 
 36173636 
 
 3655 
 
 3674 ! 3692 
 
 3711 
 
 3729 
 
 3747 
 
 3766 
 
 3784 
 
 2 
 
 4 
 
 6 
 
 7 
 
 9 
 
 11 
 
 13 15 17 
 
 24 
 
 3802 
 
 3820 
 
 3838 
 
 38563874 
 
 3892 
 
 3909 
 
 3927 
 
 3945 
 
 3962 
 
 2 
 
 4 
 
 5 
 
 7 
 
 911 
 
 12 
 
 1416 
 
 25 
 
 3979 
 
 3997 
 
 4014 
 
 4031 
 
 4048 
 
 4065 
 
 4082 
 
 4099 
 
 4116 
 
 4133 
 
 2 
 
 8 
 
 5 
 
 7 
 
 
 
 10 
 
 12 
 
 14 
 
 15 
 
 26 
 
 41504166 
 
 4183 
 
 4200 4216 
 
 4232 
 
 4249 
 
 4265 
 
 4231 
 
 4298 
 
 2 
 
 3 
 
 5 
 
 7 
 
 8 
 
 10 
 
 11 
 
 1315 
 
 27 
 
 43144330 
 
 4346 
 
 43624378 
 
 4393 
 
 4409 
 
 4425 
 
 4440 
 
 4456 
 
 2 
 
 3 
 
 5 
 
 8 
 
 8 
 
 911 
 
 13 14 
 
 28 
 
 4472 
 
 4487 
 
 4502 
 
 4518 4533 
 
 4548 
 
 4564 
 
 4579 
 
 4594 
 
 4609 
 
 2 
 
 3 
 
 5 
 
 6 
 
 8 
 
 911 1214 
 
 29 
 
 4624 
 
 4639 
 
 4654 
 
 4669 4683 
 
 4698 
 
 4713 
 
 4728 
 
 4742 
 
 4757 
 
 1 
 
 3 
 
 4 
 
 6 
 
 7 
 
 9 
 
 10 12 13 
 
 30 
 
 4771 
 
 4786 
 
 4800 
 
 4814 
 
 4829 
 
 4843 
 
 4857 
 
 4871 
 
 4886 
 
 4900 
 
 
 3 
 
 4 
 
 6 
 
 7 
 
 9 
 
 10 
 
 11 
 
 13 
 
 31 
 
 4914 
 
 4928 
 
 4942 
 
 4955 4969 
 
 4983 
 
 4997 
 
 5011 
 
 5024 
 
 5038 
 
 
 3 
 
 4 
 
 G 
 
 7 
 
 8 
 
 10 
 
 11 12 
 
 32 
 
 5051 
 
 5065 
 
 5079 
 
 5092! 5105 
 
 5119 
 
 5132 
 
 5145 
 
 5159 
 
 5172 
 
 
 3 
 
 4 
 
 5 
 
 7 
 
 8 
 
 9 
 
 11 12 
 
 33 
 
 5185 
 
 5198 
 
 5211 
 
 5224 5237 
 
 5250 
 
 5263 
 
 5276 
 
 5289 
 
 5302 
 
 
 3 
 
 4 
 
 5 
 
 6 
 
 8 
 
 9 
 
 1012 
 
 34 
 
 5315 
 
 5328 
 
 5340 
 
 5353 
 
 5366 
 
 5378 
 
 5391 
 
 5403 
 
 5416 
 
 5428 
 
 
 3 
 
 4 
 
 5 
 
 6 
 
 8 
 
 9 
 
 1011 
 
 35 
 
 5441 
 
 5453 
 
 5465 
 
 5478 
 
 5490 
 
 5502 
 
 5514 
 
 5527 
 
 5539 
 
 5551 
 
 1 
 
 2 
 
 4 
 
 5 
 
 6 
 
 7 
 
 9 
 
 1011 
 
 36 
 
 5563 
 
 5575 
 
 5587 
 
 5599 
 
 5611 
 
 5623 
 
 5635 
 
 5647 
 
 5658 
 
 5670 
 
 1 
 
 2 
 
 4 5 
 
 6 
 
 7 
 
 8 
 
 1011 
 
 37 
 
 56 2 
 
 5694 
 
 5705 
 
 5717 
 
 5729 
 
 5740 
 
 5752 
 
 5763 
 
 5775 
 
 5786 
 
 1 
 
 2 
 
 3 5 
 
 6 
 
 7 
 
 8 
 
 910 
 
 38 
 
 5798 
 
 5809 
 
 5821 
 
 5832 
 
 5843 
 
 5855 
 
 5866 
 
 5877 
 
 5888 
 
 5899 
 
 1 
 
 2 
 
 3 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 10 
 
 39 
 
 5911 
 
 5922 
 
 5933 
 
 5944 
 
 5955 
 
 5966 
 
 5977 
 
 5988 
 
 5999 
 
 6010 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 7 
 
 8 
 
 9 10 
 
 40 
 
 6021 
 
 6031 
 
 6042 
 
 6053 
 
 6064 
 
 6075 
 
 6085 
 
 6096 
 
 6107 
 
 6117 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 8 
 
 9 10 
 
 41 
 
 6128 
 
 6138 
 
 6149 
 
 6160 
 
 6170 
 
 6180 
 
 6191 
 
 6201 
 
 6212 
 
 6222 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 9 
 
 42 
 
 6232 
 
 6243 
 
 6253 
 
 6263 
 
 6274 
 
 6284 
 
 6294 
 
 6304 
 
 6314 
 
 6325 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 9 
 
 43 
 
 6335 
 
 6345 6355 6365 
 
 6375 
 
 6385 
 
 6395 
 
 6405 
 
 6415 
 
 6425 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 G 
 
 7 
 
 8 9 
 
 44 
 
 6435 
 
 6444 
 
 6454 
 
 6464 
 
 6474 
 
 6484 
 
 6493 
 
 6503 
 
 6513 
 
 6522 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 9 
 
 45 
 
 6532 
 
 6542 
 
 6551 
 
 6561 
 
 6571 
 
 6580 
 
 6590 
 
 6599 
 
 6609 
 
 6618 
 
 1 
 
 2 
 
 3 
 
 4 
 
 8 
 
 6 
 
 7 
 
 8 9 
 
 46 
 
 6628 
 
 6637 
 
 6646 
 
 6656 
 
 6665 
 
 6675 
 
 6684 
 
 6693 
 
 6702 
 
 6712 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 7 8 
 
 47 
 
 6721 
 
 6730 
 
 6739 
 
 6749 
 
 6758 
 
 6767 
 
 6776 
 
 6785 
 
 6794 
 
 6803 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 5 
 
 6 
 
 7 8 
 
 48 
 
 6812 
 
 6821 
 
 6830 
 
 6839 
 
 6848 
 
 6857 
 
 6866 6875 
 
 6884 
 
 6893 
 
 1 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 
 
 u 
 
 7 8 
 
 49 
 
 6902 
 
 6911 
 
 6920 
 
 6928 
 
 6937 
 
 6946 
 
 6955 
 
 6964 
 
 6972 
 
 6981 
 
 1 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 
 
 6 
 
 7 8 
 
 50 
 
 6990 
 
 6998 
 
 7007 
 
 7016 
 
 7024 
 
 7033 
 
 7042 
 
 7050 
 
 7059 
 
 7067 
 
 1 
 
 2 
 
 3 
 
 3 4 
 
 5 
 
 6 
 
 7 
 
 
 
 51 
 
 7076 
 
 7084 
 
 7093 
 
 7101 
 
 7110 
 
 7118 
 
 7126 
 
 7135 
 
 7143 
 
 7152 
 
 1 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 6 7 
 
 8 
 
 52 
 
 7160 
 
 7168 
 
 7177 
 
 7185 
 
 7193 
 
 7202 
 
 7210 
 
 7218 
 
 7226 
 
 7235 
 
 1 
 
 2 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 7 
 
 7 
 
 53 
 
 7243 
 
 7251 
 
 7259 
 
 7267 
 
 7275 
 
 7284 
 
 7292 
 
 7300 
 
 7308 
 
 7316 
 
 1 
 
 2 
 
 2 
 
 3 
 
 4 
 
 
 6 6 
 
 - 
 
 54 
 
 7324 
 
 7332 
 
 7340 
 
 7348 
 
 7356 
 
 7364 
 
 7372 
 
 7380 
 
 7388 
 
 7396 
 
 1 
 
 2 
 
 2 
 
 3 
 
 4 
 
 
 
 6 6 
 
 7 
 
TABLES. 
 LOGARITHMS. 
 
 515 
 
 in 
 
 
 
 
 
 
 
 
 
 
 
 PROPORTIONAL PARTS. 
 
 ll 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 
 
 
 
 
 
 
 
 
 
 | 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 9 
 
 3 
 
 4 
 
 fi 
 
 6 
 
 7 
 
 8 
 
 9 
 
 55 
 
 7404 
 
 74127419 
 
 7427 
 
 7435 
 
 7443 
 
 7451 
 
 7459 
 
 7466 
 
 7474 
 
 1 
 
 2 
 
 2 
 
 3 
 
 4 
 
 5 
 
 5 
 
 6 
 
 7 
 
 56 
 
 7482 
 
 7490 7497 
 
 7505 
 
 7513 
 
 7520 
 
 7528 
 
 7536 
 
 7543 
 
 7551 
 
 1 
 
 2 
 
 2 
 
 3 
 
 4 
 
 5 
 
 5 
 
 6 
 
 7 
 
 57 
 
 7559 
 
 7566 7574 
 
 7582 
 
 7589 
 
 7597 
 
 7604 
 
 7612 
 
 7619 
 
 7627 
 
 1 
 
 2 
 
 2 
 
 
 4 
 
 5 
 
 5 
 
 6 
 
 7 
 
 58 
 
 7634 
 
 7642 7649 
 
 7657 
 
 7664 
 
 7672 
 
 7679 
 
 7686 
 
 7694 
 
 7701 
 
 1 
 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 
 
 6 
 
 7 
 
 59 
 
 7709 
 
 7716 7723 
 
 7731 
 
 7738 
 
 7745 
 
 7752 
 
 7760 
 
 7767 7774 
 
 1 
 
 
 2 
 
 3 
 
 4 
 
 4 
 
 6 
 
 6 
 
 7 
 
 60 
 
 7782 
 
 7789 7796 
 
 7803 
 
 7810 
 
 7818 
 
 7825 
 
 7832 
 
 7839 7846 
 
 1 
 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 
 
 6 
 
 6 
 
 61 
 
 7853 
 
 7860 7868 
 
 7875 
 
 7882 
 
 7889 
 
 789: 
 
 7903 
 
 7910 7917 
 
 1 
 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 
 
 6 
 
 6 
 
 62 
 
 7924 
 
 7931 
 
 7938 
 
 7945 
 
 7952 
 
 7959 
 
 7966 
 
 7973 
 
 7980 7987 
 
 1 
 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 6 
 
 6 
 
 63 
 
 7993 
 
 8000 8007 8014 
 
 8021 
 
 8028 
 
 8035 
 
 8041 
 
 80488055 
 
 1 
 
 1 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 5 
 
 6 
 
 64 
 
 8062 
 
 806980758082 
 
 8089 
 
 8096 
 
 8102 
 
 8109 
 
 81168122 
 
 1 
 
 1 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 5 
 
 6 
 
 65 
 
 8129 
 
 8136 
 
 8142 
 
 8149 
 
 8156 
 
 816281698176 
 
 8182 8189 
 
 1 
 
 1 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 5 
 
 6 
 
 66 
 
 8195 
 
 8202 8209 8215 
 
 8222 
 
 822882358241 
 
 82488254 
 
 1 
 
 1 
 
 2 
 
 a 
 
 3 
 
 4 
 
 5 
 
 5 
 
 6 
 
 67 
 68 
 
 8261 
 8325 
 
 8267 8274 8280 
 833183388344 
 
 8287 
 8351 
 
 82938299830683128319 
 8357 8363 8370 8376 8382 
 
 1 
 1 
 
 
 2 
 
 2 
 
 3 
 3 
 
 3 
 3 
 
 4 
 4 
 
 5 
 
 4 
 
 5 
 5 
 
 6 
 6 
 
 69 
 
 8388 
 
 839584018407 
 
 8414 
 
 8420 8426 8432 8439 8445 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 
 
 6 
 
 70 
 71 
 
 8451 
 8513 
 
 8457 8463 8470!8476 
 8519852585318537 
 
 8482 8488 8494 8500*8506 
 8543 8549 8555 8561 8567 
 
 1 
 
 
 2 
 2 
 
 2 
 2 
 
 3 
 3 
 
 4 
 4 
 
 4 
 4 
 
 5 
 
 5 
 
 6 
 5 
 
 - 72 
 
 8573 
 
 8579858585918597 
 
 8603 8609 8615 8621 8627 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 4 
 
 4 
 
 fi 
 
 5 
 
 73 
 
 8633 
 
 8639 8645 8651 
 
 8657 
 
 8663 8669 8675 8681 8686 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 4 4 
 
 5 
 
 5 
 
 74 
 
 8692 
 
 8698 8704 
 
 8710 
 
 8716 
 
 87228727873387398745 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 4|4 
 
 5 
 
 5 
 
 75 
 
 8751 
 
 8756 
 
 8762 
 
 8768 8774 
 
 8779878587918797 
 
 8802 
 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 5 
 
 76 
 
 8808 
 
 8814 
 
 8820 8825 8831 
 
 8837 8842 8848 8854 8859 
 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 5 
 
 '77 
 
 8865 
 
 8871 
 
 8876 8882 8887 
 
 8893 8899 8904 8910 8915 
 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 78 
 
 8921 
 
 8927 
 
 8932 
 
 89388943 
 
 8949 8954 8960 8965 8971 
 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 79 
 
 8976 
 
 8982 
 
 89S7 
 
 89938998 
 
 9004 9009 9015 9020 9025 
 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 80 
 
 9031 
 
 9036 
 
 9042 
 
 9047 9053 
 
 9058 9063 9069 9074 
 
 9079 
 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 4 
 
 4 
 
 5 
 
 . 81 
 
 9085 
 
 9090 
 
 9096 9101 9106 
 
 9112911791229128 
 
 9133 
 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 82 
 
 9138 
 
 9143 
 
 914991549159 
 
 91659170917591809186 
 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 83 
 
 9191 
 
 9196 
 
 920192069212 
 
 9217 9222 9227 9232 9238 
 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 84 
 
 9243 
 
 9248 
 
 9253 9258 9263 
 
 9269 9274 9279 
 
 9284 9289 
 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 85 
 
 9294 
 
 9299 
 
 9304 9309 9315 
 
 9320 9325 9330 
 
 9335 9340 
 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 86 
 
 9345 
 
 9350 9355 9360:9365 
 
 9370 9375 9380 
 
 9385 9390 
 
 
 I 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 87 
 
 9395 
 
 9400940594109415 
 
 9420 9425 9430 
 
 9435 9440 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 88 
 
 9445 
 
 9450 9455 9460 
 
 9465 
 
 9469 9474 9479 
 
 94849489 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 8 
 
 4 
 
 4 
 
 89 
 
 9494 
 
 9499 9504 
 
 9509 
 
 9513 
 
 9518 9523 9528 
 
 95339538 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 90 
 
 9542 
 
 9547 9552 
 
 9557 
 
 9562 
 
 9566 9571 9576 
 
 9581 
 
 9586 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 91 
 
 9590 
 
 9595960096059609 
 
 9614 9619 9624 
 
 9628|9633 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 92 
 
 9638 
 
 96439647 
 
 9652 9657 
 
 9661 9666 9671 
 
 9675 
 
 9680 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 93 
 
 9685 
 
 96899694 
 
 96999703 
 
 970S 9713 9717 
 
 9722 
 
 9727 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 94 
 
 9731 
 
 9736 9741 
 
 97459750 
 
 9754 9759 9763 
 
 9768 
 
 9773 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 95 
 
 9777 
 
 9782 9786 
 
 9791 9795 
 
 9800 9805 9809 
 
 9814 
 
 9818 
 
 
 
 1 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 96 
 
 9823 
 
 9827 9832 
 
 983619841 
 
 9845 9850 9854 9859 
 
 9863 
 
 
 
 1 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 97 
 
 9868 
 
 9872 9877 
 
 9881*9886 
 
 9890 9894 9899,9903 
 
 9908 
 
 
 
 1 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 93 
 
 9912 
 
 9917 9921 9926 9930 9934 9939 9943 9948 
 
 9952 
 
 
 
 1 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 99 
 
 
 
 9956 
 
 9961 9965 9969 9974 
 
 9978 9983 9987 
 
 9991 
 
 9996 
 
 
 
 1 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 3 
 
 4 
 
516 
 
 TABLES. 
 TABLE IV. ANTILOGARITHMS. 
 
 Logarithms. 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 PROPORTIONAL PARTS. 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 .00 
 
 1000 
 
 1002 
 
 1005 
 
 1007 
 
 1009 
 
 1012 
 
 1014 
 
 1016 
 
 1019 
 
 1021 
 
 
 
 
 
 1 
 
 
 1 
 
 1 
 
 Q 
 
 2 
 
 2 
 
 .01 
 
 1023 
 
 1026 
 
 1028 
 
 1030 
 
 1033 
 
 1035 
 
 1038 
 
 1040 
 
 1042 
 
 1045 
 
 
 
 
 
 1 
 
 
 1 
 
 1 
 
 o 
 
 4 
 
 2 
 
 2 
 
 .02 
 
 1047 
 
 1050 
 
 1052 
 
 1054 
 
 1057 
 
 1059 1062 
 
 1064 
 
 1067 
 
 1069 
 
 
 
 
 
 
 
 1 
 
 1 
 
 2 
 
 2 
 
 2 
 
 .03 
 
 1072 
 
 1074 
 
 1076 
 
 1079 1081 
 
 1084 
 
 1086 
 
 1089 
 
 1091 
 
 1094 
 
 
 
 
 
 
 
 1 
 
 1 
 
 2 
 
 2 
 
 2 
 
 .04 
 
 1096 
 
 1099 
 
 1102 
 
 1104 
 
 1107 
 
 1109 
 
 1112 
 
 1114 
 
 1117 
 
 1119 
 
 
 
 1 
 
 
 
 1 
 
 2 
 
 2 
 
 2 
 
 2 
 
 .05 
 
 1122 
 
 1125 
 
 1127 
 
 1130 
 
 1132 
 
 1135 
 
 1138 
 
 1140 
 
 1143 
 
 1146 
 
 
 
 1 
 
 
 
 1 
 
 2 
 
 2 
 
 2 
 
 2 
 
 .06 
 
 1148 
 
 1151 
 
 1153 
 
 1156 
 
 1159 
 
 1161 
 
 1164 
 
 1167 
 
 1169 
 
 1172 
 
 
 
 1 
 
 
 
 1 
 
 2 
 
 2 
 
 2 2 
 
 .07 
 
 1175 
 
 1178 
 
 1180 
 
 1183 
 
 1186 
 
 1189 
 
 1191 
 
 1194 
 
 1197 
 
 1199 
 
 
 
 1 
 
 
 
 1 
 
 2 
 
 2 
 
 2 2 
 
 .08 
 
 1202 
 
 1205 
 
 1208 
 
 1211 
 
 1213 
 
 1216 
 
 1219 
 
 1222 
 
 1225 
 
 1227 
 
 
 
 1 
 
 
 
 1 
 
 2 
 
 2 
 
 2 
 
 3 
 
 .09 
 
 1230 
 
 1233 
 
 1236 
 
 1239 
 
 1242 
 
 1245 
 
 1247 
 
 1250 
 
 1253 
 
 1256 
 
 
 
 1 
 
 
 
 1 
 
 2 
 
 2 
 
 2 
 
 3 
 
 .10 
 
 1259 
 
 1262 
 
 1265 
 
 1268 
 
 1271 
 
 1274 
 
 1276 
 
 1279 
 
 1282 
 
 1285 
 
 
 
 1 
 
 
 
 1 
 
 2 
 
 2 
 
 2 
 
 3 
 
 .11 
 
 1288 
 
 1291 
 
 1294 
 
 1297 
 
 1300 
 
 1303 
 
 1306 
 
 1309 
 
 1312 
 
 1315 
 
 
 
 1 
 
 
 
 2 
 
 2 
 
 2 
 
 2 
 
 3 
 
 .12 
 
 1318 
 
 1321 
 
 1324 
 
 1327 
 
 1330 
 
 1334 
 
 1337 
 
 1340 
 
 1343 
 
 1346 
 
 
 
 1 
 
 
 
 2 
 
 2 
 
 2 
 
 2 
 
 3 
 
 .13 
 
 1349 
 
 1352 
 
 1355 
 
 1358 
 
 1361 
 
 1365 
 
 1368 
 
 1371 
 
 1374 
 
 1377 
 
 
 
 1 
 
 
 
 2 
 
 2 
 
 2 
 
 3 
 
 3 
 
 .14 
 
 1380 
 
 1384 
 
 1387 
 
 1390 
 
 1393 
 
 1396 
 
 1400 
 
 1403 
 
 1406 
 
 1409 
 
 
 
 1 
 
 
 
 2 
 
 2 
 
 2 
 
 3 
 
 3 
 
 .15 
 
 1413 
 
 1416 
 
 1419 
 
 1422 
 
 1426 
 
 1429 
 
 1432 
 
 1435 
 
 1439 
 
 1442 
 
 
 
 1 
 
 
 
 2 
 
 2 
 
 2 
 
 3 
 
 3 
 
 .16 
 
 1445 
 
 1449 
 
 1452 
 
 1455 1459 
 
 1462 
 
 .1466 
 
 1469 
 
 1472 
 
 1476 
 
 
 
 1 
 
 
 
 2 
 
 2 
 
 2 
 
 3 
 
 3 
 
 .17 
 
 1479 
 
 1483 
 
 1486 
 
 1489 1493 
 
 1496 
 
 1500 
 
 1503 
 
 1507 
 
 1510 
 
 
 
 1 
 
 
 
 2 
 
 2 
 
 2 
 
 3 
 
 3 
 
 .18 
 
 1514 
 
 1517 
 
 1521 
 
 1524 
 
 152* 
 
 1531 
 
 1535 
 
 1538 
 
 1542 
 
 1545 
 
 
 
 1 
 
 
 
 2 
 
 2 
 
 2 
 
 3 
 
 3 
 
 .19 
 
 1549 
 
 1552 
 
 1556 
 
 1560 
 
 1563 
 
 1567 
 
 1570 
 
 1574 
 
 1578 
 
 1581 
 
 
 
 1 
 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 3 
 
 .20 
 
 1585 
 
 1589 
 
 1592 
 
 1596 
 
 1600 
 
 1603 
 
 1607 
 
 1611 
 
 1614 
 
 1618 
 
 
 
 1 
 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 3 
 
 .21 
 
 1622 
 
 1626 
 
 1629 
 
 1633 
 
 1637 
 
 1641 
 
 1644 
 
 1648 
 
 1652 
 
 1656 
 
 
 
 1 
 
 
 2 
 
 2 
 
 2 
 
 3 
 
 3 
 
 3 
 
 .22 
 
 1660 
 
 1663 
 
 1667 
 
 1671 
 
 1675 
 
 1679 
 
 1683 
 
 1687 
 
 1690 
 
 1694 
 
 
 
 1 
 
 
 2 
 
 2 
 
 2 
 
 3 
 
 3 
 
 3 
 
 .23 
 
 1698 
 
 1702 
 
 1706 
 
 1710 
 
 1714 
 
 1718 
 
 1722 
 
 1726 
 
 1730 
 
 1734 
 
 
 
 1 
 
 
 2 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 .24 
 
 OR 
 
 1738 
 
 770 
 
 1742 
 
 709 
 
 1746 
 
 17QC 
 
 1750 
 
 1 701 
 
 1754 
 
 I 7QC 
 
 1758 
 
 7QQ 
 
 1762 
 
 -IOAO 
 
 1766 
 
 1807 
 
 1770 
 QI i 
 
 1774 
 
 1Q1 fi 
 
 
 
 1 
 
 
 2 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 pZQ 
 
 .26 
 
 1 1 O 
 
 1820 
 
 tQft 
 
 1824 
 
 1 /OO 
 
 1828 
 
 i /yi 
 1832 
 
 i <yo 
 1837 
 
 . /yy 
 1841 
 
 loUo 
 
 1845 
 
 loUi 
 
 1849 
 
 .Oil 
 
 1854 
 
 J.O AU 
 
 1858 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 3 
 
 4 
 
 .27 
 
 1862 
 
 1866 
 
 1871 
 
 1875 
 
 1879 
 
 1884 
 
 1888 1892 
 
 1897 
 
 1901 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 3 
 
 4 
 
 .28 
 
 1905 
 
 1910 
 
 1914 
 
 1919 
 
 1923 
 
 1928 1932 1936 
 
 1941 
 
 1945 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 .29 
 
 1950 
 
 1954 
 
 1959 
 
 1963 
 
 1968 
 
 1972 
 
 1977 1982 
 
 1986 
 
 1991 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 .30 
 
 1995 
 
 2000 
 
 2004 
 
 2009 
 
 2014 
 
 2018 
 
 2023 2028 
 
 2032 
 
 2037 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 .31 
 
 2042 
 
 2046 
 
 2051 
 
 2056 
 
 2061 
 
 2065 2070 
 
 2075 
 
 2080 
 
 2084 
 
 
 
 ] 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 .32 
 
 2089 
 
 2094 
 
 2099 
 
 2104 
 
 2109 
 
 2113 
 
 2118 
 
 2123 
 
 2128 
 
 2133 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 .33 
 
 2138 
 
 2143 
 
 2148 
 
 2153 
 
 2158 
 
 2163 
 
 2168 
 
 2173 
 
 2178 
 
 2183 
 
 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 .34 
 
 2188 
 
 2193 
 
 2198 
 
 2203 
 
 2208 
 
 2213 
 
 2218 
 
 2223 
 
 2228 
 
 2234 
 
 1 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 .35 
 
 2239 
 
 2244 
 
 2249 
 
 2254 
 
 2259 
 
 22652270 
 
 2275 
 
 2280 
 
 2286 
 
 1 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 5 
 
 .36 
 
 2291 
 
 2296 
 
 2301 
 
 2307 
 
 2312 
 
 2317 
 
 2323 
 
 2328 
 
 2333 
 
 2339 
 
 1 
 
 1 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 5 
 
 .37 
 
 2344 
 
 2350 
 
 2355 
 
 2360 
 
 2366 
 
 2371 
 
 2377 
 
 2382 
 
 2388 
 
 2393 
 
 1 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 5 
 
 .38 
 
 2399 
 
 2404 
 
 2410 
 
 2415 
 
 2421 
 
 2427 
 
 2432 
 
 2438 
 
 2443 
 
 2449 
 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 5 
 
 .39 
 
 2455 
 
 2460 
 
 2466 
 
 2472 
 
 2477 
 
 2483 
 
 2489 
 
 2495 
 
 2500 
 
 2506 
 
 
 
 2 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 5 
 
 .40 
 
 2512 
 
 2518 
 
 2523 
 
 2529 
 
 2535 
 
 2541 
 
 2547 
 
 2553 
 
 2559 
 
 2564 
 
 
 
 2 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 5 
 
 .41 
 
 2570 
 
 2576 
 
 2582 
 
 2588 
 
 2594 
 
 2600 
 
 2606 
 
 2612 
 
 2618 
 
 2624 
 
 
 
 2 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 5 
 
 .42 
 
 2630 
 
 2636 
 
 2642 
 
 2649 
 
 2655 
 
 2661 
 
 2667 
 
 2673 
 
 2679 
 
 2685 
 
 
 
 2 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 
 
 .43 
 
 2692 
 
 2698 
 
 2704 
 
 2710 
 
 2716 
 
 2723 
 
 2729 
 
 2735 
 
 2742 
 
 2748 
 
 
 
 2 
 
 3 
 
 3 
 
 i 
 
 4 
 
 5 6 
 
 .44 
 
 2754 
 
 2761 
 
 2767 
 
 2773 
 
 2780 
 
 2786 
 
 2793 
 
 2799 
 
 2805 
 
 2812 
 
 
 
 2 
 
 3 
 
 3 
 
 4 
 
 4 
 
 j 6 
 
 .45 
 
 2818 
 
 2825 
 
 2831 
 
 2838 
 
 2844 
 
 2851 
 
 2858 
 
 2164 
 
 2871 
 
 2877 
 
 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 5 6 
 
 .46 
 
 2884 
 
 2891 
 
 2897 
 
 2904 
 
 2911 
 
 29172924 
 
 2931 
 
 2938 
 
 2944 
 
 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 5 6 
 
 .47 
 
 2951 
 
 2958 
 
 2965 2972 
 
 2979 
 
 2985|2992 
 
 2999 
 
 3006 
 
 3013 
 
 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 5 6 
 
 .48 
 
 3020 
 
 3027 
 
 30343041 
 
 3048 
 
 3055' 3062 
 
 3069 
 
 3076 
 
 3083 
 
 1 
 
 1 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 
 
 6 6 
 
 .49 
 
 3090 
 
 3097 
 
 31053112 
 
 3119 
 
 31263133 
 
 3141 
 
 3148 
 
 3155 
 
 1 
 
 1 
 
 2 
 
 3 
 
 4 
 
 4 
 
 5 
 
 6 
 
 6 
 
TABLES. 
 ANTILOGARITHMS. 
 
 517 
 
 1 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 PROPORTIONAL PARTS. 
 
 1 2 
 
 3 
 
 5 
 
 4 
 
 7 
 
 .|. 
 
 .50 
 
 3162 
 
 53170 
 
 3177 
 
 3184 
 
 t319 
 
 3199 
 
 320 
 
 3214 
 
 322] 
 
 3228 
 
 j j 
 
 
 
 
 f 
 
 e ! 7 
 
 .51 
 
 3236 
 
 ,3243 
 
 3251 
 
 325 
 
 3266 
 
 3273 
 
 3281 
 
 3289 
 
 329C 
 
 13304 
 
 2 
 
 2 
 
 4 
 
 
 I 
 
 6 7 
 
 .52 
 
 3311 
 
 3319 
 
 3327 
 
 3334 
 
 3342 
 
 3350 
 
 3357 
 
 3365 
 
 3373 3381 
 
 f 
 
 2 
 
 4 
 
 
 5 
 
 6 7 
 
 .53 
 
 3388 
 
 3396 
 
 3404 
 
 3412 
 
 3420 
 
 3428 
 
 3436 
 
 3443 
 
 3451 3459 
 
 m 
 4 
 
 2 
 
 4 
 
 
 6 
 
 6 7 
 
 .54 
 
 3467 
 
 3475 
 
 3483 
 
 3491 
 
 3499 
 
 3508 
 
 3516 
 
 3524 
 
 3532 
 
 3540 
 
 2 
 
 2 
 
 4 
 
 5 
 
 6 
 
 6 7 
 
 .55 
 
 3548 
 
 3556 
 
 3565 
 
 3573 
 
 3581 
 
 3589 
 
 3597 
 
 3606 
 
 3614 
 
 3622 
 
 2 
 
 
 
 4 
 
 4 
 
 
 6 
 
 7 7 
 
 .56 
 
 3631 
 
 3639 
 
 3648 
 
 3656 
 
 3664 
 
 3673 
 
 3681 
 
 3690 
 
 3698 
 
 3707 
 
 2 
 
 3 
 
 4 
 
 | 
 
 6 
 
 7 8 
 
 .57 
 
 3715 
 
 3724 
 
 3733 
 
 3741 
 
 3750 
 
 3758 
 
 3767 
 
 3776 
 
 3784 
 
 3793 
 
 2 
 
 3 ' 
 
 4 
 
 5 
 
 6 
 
 7 
 
 1 
 
 8 
 
 .58 
 
 3802 
 
 3811 
 
 3819 
 
 3828 
 
 3837 
 
 3846 
 
 3855 
 
 3864 
 
 3873 
 
 3882 
 
 2 
 
 3 ' 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 .59 
 
 3890 
 
 3899 
 
 3908 
 
 3917 
 
 3926 
 
 3936 
 
 3945 
 
 3954 
 
 3963 
 
 3972 
 
 2 
 
 3 < 
 
 5 
 
 5 
 
 6 
 
 7 
 
 8 
 
 .60 
 
 981 
 
 3990 
 
 3999 
 
 4009 
 
 4018 
 
 4027 
 
 4036 
 
 4046 
 
 4055 
 
 4064 
 
 2 
 
 3 
 
 5 
 
 6 
 
 6 
 
 7 
 
 8 
 
 .61 
 
 074 
 
 4083 
 
 4093 
 
 4102 
 
 4111 
 
 4121 
 
 4130 
 
 4140 
 
 4150 
 
 4159 
 
 2 
 
 3 ' 
 
 5 
 
 6 
 
 7 
 
 8 
 
 Q 
 
 .62 
 
 169 
 
 4178 
 
 4188 
 
 4198 
 
 4207 
 
 4217 
 
 4227 
 
 4236 
 
 4246 
 
 4256 
 
 2 
 
 3 * 
 
 5 
 
 6 
 
 7 
 
 8 
 
 Q 
 
 .63 
 
 266 
 
 4276 
 
 4285 
 
 4295 
 
 4305 
 
 4315 
 
 4325 
 
 4335 
 
 4345 
 
 4355 
 
 2 
 
 3 ^ 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 .64 
 
 365 
 
 4375 
 
 4385 
 
 4395 
 
 4406 
 
 4416 
 
 4426 
 
 4436 
 
 4446 
 
 4457 
 
 2 
 
 3 ' 
 
 5 
 
 6 
 
 7 
 
 8 
 
 Q 
 
 .65 
 
 467 
 
 4477 
 
 4487 
 
 4498 
 
 4508 
 
 4519 
 
 4529 
 
 4539 
 
 4550 
 
 4560 
 
 2 
 
 3 ' 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 .66 
 
 571 
 
 4581 
 
 4592 
 
 4603 
 
 4613 
 
 4624 
 
 4634 
 
 4645 
 
 4656 
 
 4667 
 
 2 
 
 3 4 
 
 5 
 
 6 
 
 7 
 
 9 
 
 10 
 
 .67 
 
 677 
 
 4688 
 
 4699 
 
 4710 
 
 4721 
 
 4732 
 
 4742 
 
 4753 
 
 4764 
 
 4775 
 
 2 
 
 3 4 
 
 5 
 
 7 
 
 8 
 
 9 
 
 10 
 
 .68 
 
 786 
 
 4797 
 
 808 
 
 4819 
 
 4831 
 
 4842 
 
 4853 
 
 4864 
 
 4875 
 
 4887 
 
 2 
 
 3 A 
 
 6 
 
 7 
 
 8 
 
 9'lO 
 
 .69 
 
 898 
 
 4909 
 
 920 
 
 4932 
 
 4943 
 
 4955 
 
 4966 
 
 4977 
 
 4989 5000 
 
 2 
 
 3 I 
 
 6 
 
 7 
 
 8 
 
 910 
 
 70 
 
 012 
 
 5023 
 
 035 
 
 5047 
 
 5058 
 
 5070 
 
 5082 
 
 5093 
 
 51055117 
 
 2 
 
 4 
 
 6 
 
 7 
 
 8 
 
 911 
 
 .71 
 
 129 
 
 5140 
 
 152 5164 
 
 5176 
 
 5188 
 
 5200 
 
 5212 
 
 52245236 
 
 2 
 
 4 
 
 6 
 
 7 
 
 8 
 
 11 
 
 .72 
 
 248 
 
 5260 
 
 272 5284 
 
 5297 
 
 5309 
 
 5321 
 
 5333 
 
 5346535 
 
 2 
 
 4 
 
 6 
 
 8 
 
 91011 
 
 .73 
 
 ^370 
 
 5383 
 
 5395 5408 
 
 5420 
 
 5433 
 
 5445 
 
 5453 
 
 5470 5483 
 
 3 
 
 4 
 
 6 
 
 8 
 
 91011 
 
 .74 
 
 495 
 
 5508 
 
 521 5534 
 
 5546 
 
 5559 
 
 572 
 
 5585 
 
 5598 
 
 5610 
 
 3 
 
 4 
 
 6 
 
 8 
 
 9 
 
 10 12 
 
 .75 
 
 623 
 
 5636 
 
 649 5662 
 
 5675 
 
 5689 
 
 5702 
 
 715 
 
 5728 
 
 5741 
 
 3 
 
 4 
 
 7 
 
 8 
 
 9 
 
 1012 
 
 .76 
 
 754 
 
 5768 
 
 781 5794 
 
 5808 
 
 5821 
 
 834 
 
 848 
 
 5861 
 
 5875 
 
 3 
 
 4 
 
 7 
 
 8 
 
 9 
 
 11 12 
 
 .77 
 
 888 
 
 5902 
 
 916 5929 
 
 5943 
 
 5957 
 
 970 
 
 984 
 
 5998 
 
 6012 
 
 3 
 
 4 
 
 7 
 
 8 
 
 
 
 11 
 
 12 
 
 .78 
 
 026 
 
 6039 
 
 05316067 
 
 6081 
 
 095 
 
 109 
 
 124 
 
 6138 
 
 6152 
 
 3 
 
 4 
 
 7 
 
 8 
 
 
 
 11 
 
 13 
 
 .79 
 
 166 
 
 6180 
 
 194 
 
 6209 
 
 6223 
 
 237 
 
 252 
 
 6266 
 
 6281 
 
 6295 
 
 3 
 
 4 
 
 7 
 
 9 
 
 
 
 11 13 
 
 .80 
 
 3310 
 
 6324 
 
 3339 
 
 6353 
 
 6368 
 
 6383 
 
 397 
 
 6412 
 
 6427 
 
 6442 
 
 3 
 
 4 6 
 
 7 
 
 9 
 
 
 
 12 
 
 13 
 
 .81 
 
 457 
 
 6471 
 
 3486 
 
 6501 
 
 6516 
 
 531 
 
 546 
 
 6561 
 
 6577 
 
 6592 
 
 3 
 
 5 : 
 
 8 
 
 9 
 
 1 
 
 12 
 
 14 
 
 .82 
 
 607 
 
 6622 
 
 637 
 
 6653 
 
 6668 
 
 683 
 
 699 
 
 6714 
 
 6730 
 
 6745 
 
 3 
 
 > 
 
 8 
 
 9 
 
 1 
 
 l 
 
 14 
 
 .83 
 
 761 
 
 6776 
 
 792 
 
 6808 
 
 6823 
 
 839 
 
 3855 
 
 6871 
 
 6887 
 
 6902 
 
 3 
 
 5 6 
 
 8 
 
 9 
 
 1 
 
 13 
 
 14 
 
 .84 
 
 918 
 
 6934 
 
 950 
 
 6966 
 
 6982 
 
 998 
 
 015 
 
 7031 
 
 7047 
 
 7063 
 
 
 5 6 
 
 8 
 
 10 
 
 1 
 
 13 
 
 15 
 
 .85 
 
 079 
 
 7096 
 
 112 
 
 7129 
 
 7145 
 
 161 
 
 178 
 
 7194 
 
 7211 
 
 7228 
 
 
 5 7 
 
 8 
 
 10 
 
 2 
 
 13 
 
 15 
 
 .86 
 
 244 
 
 7261 
 
 278 
 
 7295 
 
 7311 
 
 328 
 
 345 
 
 7362 
 
 7379 
 
 7396 
 
 
 5 7 
 
 8 
 
 10 
 
 2 
 
 13 
 
 15 
 
 .87 
 
 413 
 
 7430 
 
 447 
 
 7464 
 
 7482 
 
 499 
 
 516 
 
 7534 
 
 7551 
 
 7568 
 
 
 5 7 
 
 9 
 
 10 
 
 2 
 
 14 
 
 16 
 
 .88 
 
 586 
 
 7603 
 
 621 
 
 7638 
 
 7656 
 
 674 
 
 691 
 
 7709 
 
 7727 
 
 7745 
 
 . 
 
 5 7 
 
 9 
 
 11 
 
 2 
 
 14 
 
 16 
 
 .89 
 
 762 
 
 7780 
 
 798 
 
 7816 
 
 7834 
 
 852 
 
 870 
 
 7889 
 
 7907 
 
 7925 
 
 4 
 
 5 7 
 
 9 
 
 11 
 
 3 
 
 14 
 
 16 
 
 .90 
 
 943 
 
 7962 
 
 980 
 
 7998 
 
 8017 
 
 035 
 
 054 
 
 8072 
 
 3091 
 
 8110 
 
 
 6 7 
 
 9 
 
 11 
 
 3 
 
 15 
 
 17 
 
 .91 
 
 128 
 
 8147 
 
 166 
 
 8185 
 
 8204 
 
 222 
 
 241 
 
 8260 
 
 3279 
 
 8299 
 
 
 6 8 
 
 9 
 
 11 
 
 3 
 
 15 
 
 17 
 
 .92 
 
 318 
 
 8337 
 
 356 
 
 8375 
 
 8395 
 
 414 
 
 433 
 
 8453 
 
 3472 
 
 8492 
 
 
 6 8 
 
 10 
 
 12 
 
 4 
 
 15 
 
 17 
 
 .93 
 
 511 
 
 8531 
 
 551 
 
 8570 
 
 8590 
 
 610 
 
 630 
 
 8650 
 
 3670 
 
 8690 
 
 4 
 
 6 8 
 
 10 
 
 12 
 
 4 
 
 16 
 
 18 
 
 .94 
 
 710 
 
 8730 
 
 750 
 
 8770 
 
 8790 
 
 810 
 
 831 
 
 8851 
 
 3872 
 
 8892 
 
 4 
 
 6 8 
 
 10 
 
 12 
 
 4 
 
 16 
 
 18 
 
 .95 
 
 913 
 
 8933 
 
 954 
 
 8974 
 
 8995 
 
 016 
 
 036 
 
 9057 
 
 3078 
 
 9099 
 
 4 
 
 6 8 
 
 10 
 
 12 
 
 5 
 
 L7 
 
 19 
 
 .96 
 
 120 
 
 9141 
 
 162 
 
 9183 
 
 9204 
 
 226 
 
 247 
 
 9268 
 
 3290 
 
 9311 
 
 
 6 8 
 
 11 
 
 13 
 
 5 
 
 17 
 
 19 
 
 .97 
 
 333 
 
 9354 
 
 376 
 
 9397 
 
 9419 
 
 441 
 
 462 
 
 9484 
 
 3506 
 
 9528 
 
 
 7 9 
 
 11 
 
 13 
 
 5 
 
 17 
 
 20 
 
 .98 
 
 550 
 
 9572 
 
 594 
 
 9616 
 
 9638 
 
 661 
 
 683 
 
 9705 
 
 3727 
 
 9750 
 
 4 
 
 7 9 
 
 11 
 
 13 
 
 3 
 
 18 
 
 20 
 
 .99 
 
 772 
 
 9795 
 
 817 
 
 9840 
 
 9863 
 
 886 
 
 908 
 
 9931 
 
 3954 
 
 9977 
 
 
 7 9 
 
 11 
 
 14 
 
 6 
 
 18 
 
 20 
 
518 
 
 TABLES. 
 
 TABLE V. SPECIFIC GRAVITY OF AMMONIA SOLUTIONS AT 15 C. 
 LUNGE AND WIERNIK.* 
 
 Snecific 
 Gravity 
 15 
 *4T 
 
 in Vacuo. 
 
 Per Cent 
 NH 3 , 
 Grams. 
 
 1 Liter 
 Contains 
 NH 3 , 
 Grams. 
 
 Correction 
 of Specific 
 Gravity 
 for 1* 
 
 Specific 
 Gravity 
 
 4 
 in Vacuo. 
 
 Per Cent 
 NH 3 , 
 Grams. 
 
 1 Liter 
 Contains 
 NH 3 , 
 Grams. 
 
 Correction 
 of Specific 
 Gravity 
 for 1. 
 
 1.000 
 
 0.00 
 
 0.0 
 
 0.00018 
 
 0.940 
 
 15.63 
 
 146.9 
 
 0.00039 
 
 0.998 
 
 0.45 
 
 4.5 
 
 0.00018 
 
 0.938 
 
 16.22 
 
 152.1 
 
 0.00040 
 
 0.996 
 
 0.91 
 
 9.1 
 
 0.00019 
 
 0.936 
 
 16.82 
 
 157.4 
 
 0.00041 
 
 0.994 
 
 1.37 
 
 13.6 
 
 0.00019 
 
 0.934 
 
 17.42 
 
 162.7 
 
 0.00041 
 
 0.992 
 
 1.84 
 
 18.2 
 
 0.00020 
 
 0.932 
 
 18.03 
 
 168.1 
 
 0.00042 
 
 0.990 
 
 2.31 
 
 22.9 
 
 0.00020 
 
 0.930 
 
 18.64 
 
 173.4- 
 
 0.00042 
 
 0.988 
 
 2.80 
 
 27.7 
 
 0.00021 
 
 0.928 
 
 19.25 
 
 178.6 
 
 0.00043 
 
 0.986 
 
 3.30 
 
 32.5 
 
 0.00021 
 
 0.926 
 
 19.87 
 
 184.2 
 
 0.00044 
 
 0.984 
 
 3.80 
 
 37.4 
 
 0.00022 
 
 0.924 
 
 20.49 
 
 189.3 
 
 0.00045 
 
 0.982 
 
 4.30 
 
 42.2 
 
 0.00022 
 
 0.922 
 
 21.12 
 
 194.7 
 
 0.00046 
 
 0.980 
 
 4.80 
 
 47.0 
 
 0.00023 
 
 0.920 
 
 21.75 
 
 200.1 
 
 0.00047 
 
 0.978 
 
 5.30 
 
 51.8 
 
 0.00023 
 
 0.918 
 
 22.39 
 
 205.6 
 
 0.00048 
 
 0.976 
 
 5.80 
 
 56.6 
 
 0.00024 
 
 0.916 
 
 23.03 
 
 210.9 
 
 0.00049 
 
 0.974 
 
 6.30 
 
 61.4 
 
 0.00024 
 
 0.914 
 
 23.68 
 
 216.3 
 
 0.00050 
 
 0.972 
 
 6.80 
 
 66.1 
 
 0.00025 
 
 0.912 
 
 24.33 
 
 221.9 
 
 0.00051 
 
 0.970 
 
 7.31 
 
 70.9 
 
 0.00025 
 
 0.910 
 
 24.99 
 
 227.4 
 
 0.00052 
 
 0.968 
 
 7.82 
 
 75.7 
 
 0.00026 
 
 0.903 
 
 25.65 
 
 232.9 
 
 0.00053 
 
 0.966 
 
 8.33 
 
 80.5 
 
 0.00026 
 
 0.906 
 
 26.31 
 
 238.3 
 
 0.00054 
 
 0.964 
 
 8.84 
 
 85.2 
 
 0.00027 
 
 0.904 
 
 26. 93 
 
 243.9 
 
 0.00055 
 
 0.962 
 
 9.35 
 
 89.9 
 
 0.00028 
 
 0.902 
 
 27.65 
 
 249.4 
 
 0.00056 
 
 0.960 
 
 9.91 
 
 95.1 
 
 0.00029 
 
 0.900 
 
 23.33 
 
 255.0 
 
 0.00057 
 
 0.958 
 
 10.47 
 
 100.3 
 
 0.00030 
 
 0.898 
 
 29.01 
 
 260.5 
 
 0.00058 
 
 0.956 
 
 11.03 
 
 105.4 
 
 0.00031 
 
 0.896 
 
 29.69 
 
 266.0 
 
 0.00059 
 
 0.954 
 
 11.60 
 
 110.7 
 
 0.00032 
 
 0.894 
 
 30.37 
 
 271.5 
 
 0.00060 
 
 0.952 
 
 12.17 
 
 115.9 
 
 0.00033 
 
 0.892 
 
 31.05 
 
 277.0 
 
 0.00060 
 
 0.950 
 
 12.74 
 
 121.0 
 
 0.00034 
 
 0.890 
 
 31.75 
 
 282.6 
 
 0.00061 
 
 0.948 
 
 13.31 
 
 126.2 
 
 0.00035 
 
 0.888 
 
 32.50 
 
 288.6 
 
 0.00062 
 
 0.946 
 
 13.88 
 
 131.3 
 
 0.00036 
 
 0.886 
 
 33.25 
 
 294.6 
 
 0.00063 
 
 0.944 
 
 14.46 
 
 136.5 
 
 0.00037 
 
 0.884 
 
 34.10 
 
 301. 4* 
 
 0.00064 
 
 0.942 
 
 15.04 
 
 141.7 
 
 0.00038 
 
 0.882 
 
 34.95 
 
 308.3 
 
 0.00065 
 
 * Zeit. f. angew. Chem., 1889, 181. 
 
TABLES 
 
 519 
 
 TABLE VI.-SPECIFIC GRAVITY OF HYDROCHLORIC ACID AT 15 C. 
 LUNGE AND MARCHLEWSKI.* 
 
 Specific Grav- 
 
 ityatf 
 
 in Vacuo. 
 
 Degrees 
 Baume.t 
 
 Degrees 
 Twaddell. 
 
 100 Parts by 
 Weight Contain 
 Grams, 
 HC1. 
 
 1 Liter Contains 
 in Kilograms, 
 HC1. 
 
 1.000 
 
 0.0 
 
 
 
 0.16 
 
 0.0016 
 
 1.005 
 
 0.7 
 
 1 
 
 1.15 
 
 0.012 
 
 1.010 
 
 1.4 
 
 2 
 
 2.14 
 
 0.022 
 
 1.015 
 
 2.1 
 
 3 
 
 3.12 
 
 0.032 
 
 1.020 
 
 2.7 
 
 4 
 
 4.13 
 
 0.042 
 
 1.025 
 
 3.4 
 
 5 
 
 5.15 
 
 0.053 
 
 1.030 
 
 4.1 
 
 6 
 
 6.15 
 
 0.064 
 
 1.035 
 
 4.7 
 
 7 
 
 7.15 
 
 0.074 
 
 1.040 
 
 5.4 
 
 8 
 
 8.16 
 
 0.085 
 
 1.045 
 
 6.0 
 
 9 
 
 9.16 
 
 0.096 
 
 1.050 
 
 6.7 
 
 10 
 
 10.17 
 
 0.107 
 
 1.055 
 
 7.4 
 
 11 
 
 11.18 
 
 0.118 
 
 1.060 
 
 8.0 
 
 12 
 
 12.19 
 
 0.129 
 
 1.065 
 
 8.7 
 
 13 
 
 13.19 
 
 0.141 
 
 1.070 
 
 9.4 
 
 14 
 
 14.17 
 
 0.152 
 
 1.075 
 
 10.0 
 
 15 
 
 15.16 
 
 0.163 
 
 1.080 
 
 10.6 
 
 16 
 
 16.15 
 
 0.174 
 
 1.085 
 
 11.2 
 
 17 
 
 17.13 
 
 0.186 
 
 1.090 
 
 11.9 
 
 18 
 
 18.11 
 
 0.197 
 
 1.095 
 
 12.4 
 
 19 
 
 19.06 
 
 0.209 
 
 1.100 
 
 13.0 
 
 20 
 
 20.01 
 
 0.220 
 
 1.105 
 
 13.6 
 
 21 
 
 20.97 
 
 0.232 
 
 1.110 
 
 14.2 
 
 22 
 
 21.92 
 
 0.243 
 
 1.115 
 
 14.9 
 
 23 
 
 22.86 
 
 0.255 
 
 1.120 
 
 15.4 
 
 24 
 
 23.82 
 
 0.267 
 
 1.125 
 
 16.0 
 
 25 
 
 24.78 
 
 0.278 
 
 1.130 
 
 16.5 
 
 26 
 
 25.75 
 
 0.291 
 
 1.135 
 
 17.1 
 
 27 
 
 26.70 
 
 0.303 
 
 1.140 
 
 17.7 
 
 28 
 
 27.66 
 
 0.315 
 
 1 . 1425 
 
 18.0 
 
 
 28.14 
 
 0.322 
 
 1.145 
 
 18.3 
 
 29 
 
 28.61 
 
 0.32S 
 
 1.150 
 
 18.8 
 
 30 
 
 29.57 
 
 0.340 
 
 1.152 
 
 19.0 
 
 t 
 
 29.95 
 
 0.345 
 
 1.155 
 
 19.3 
 
 3i 
 
 30.55 
 
 0.353 
 
 1.160 
 
 19.8 
 
 32 
 
 31.52 
 
 0.366 
 
 1.163 
 
 20.0 
 
 
 32.10 
 
 0.373 
 
 1.165 
 
 20.3 
 
 33 
 
 32.49 
 
 0.379 
 
 1.170 
 
 20.9 
 
 34 
 
 33.46 
 
 0.392 
 
 1.171 
 
 21.0 
 
 
 33.65 
 
 0.394 
 
 1.175 
 
 21.4 
 
 35 
 
 34.42 
 
 0.404 
 
 1 180 
 
 22.0 
 
 36 
 
 35.39 
 
 0.418 
 
 1.185 
 
 22.5 
 
 37 
 
 36.31 
 
 0.430 
 
 1.190 
 
 23.0 
 
 38 
 
 37.23 
 
 0.443 
 
 1.195 
 
 23.5 
 
 39 
 
 38.16 
 
 0.456 
 
 1.200 
 
 24.0 
 
 40 
 
 39.11 
 
 0.469 
 
 *Zeit. angew. Chem., 1891, 135. 
 
 t Fur the American Standard Baumd scale and the specific-gravity table adopted by the 
 Manufacturing Chemists' Association of the United States, see Van Nostrand's Chemical 
 Annual. 
 
520 
 
 TABLES. 
 
 TABLE VII. SPECIFIC GRAVITY OF NITRIC ACID AT 15 C. 
 LUNGE AND REY.* 
 
 Specific 
 Gravity 
 
 -? 
 
 in Vacuo. 
 
 Degrees 
 Baume'.t 
 
 Degrees 
 Twaddell. 
 
 100 Parts by Weight 
 Contain, Grams, 
 
 1 Liter Contains in 
 Kilograms, 
 
 N 2 5 . 
 
 HN0 3 . 
 
 NA,. 
 
 HNOa. 
 
 1.000 
 
 
 
 
 
 o.os 
 
 0.10 
 
 0.001 
 
 0.001 
 
 1.005 
 
 0.7 
 
 1 
 
 0.85 
 
 1.00 
 
 0.003 
 
 0.010 
 
 1.010 
 
 1.4 
 
 2 
 
 1.62 
 
 1.90 
 
 0.016 
 
 0.019 
 
 1.015 
 
 2.1 
 
 3 
 
 2.39 
 
 2/0 
 
 0.024 
 
 0.028 
 
 1.020 
 
 2.7 
 
 4 
 
 3.17 
 
 3.70 
 
 0.033 
 
 0.038 
 
 1.025 
 
 3.4 
 
 5 
 
 3.94 
 
 4.60 
 
 0.040 
 
 0.047 
 
 .030 
 
 4.1 
 
 6 
 
 4.71 
 
 5.50 
 
 0.049 
 
 0.057 
 
 .035 
 
 4.7 
 
 7 
 
 5.47 
 
 6.3"> 
 
 0.057 
 
 0.066 
 
 .040 
 
 5.4 
 
 8 
 
 6.22 
 
 7.26 
 
 0.064 
 
 0.075 
 
 .045 
 
 6.0 
 
 9 
 
 6.97 
 
 8.13 
 
 073 
 
 0.035 
 
 .050 
 
 6.7 
 
 10 
 
 7.71 
 
 8.99 
 
 1 
 
 0.094 
 
 .055 
 
 7.4 
 
 11 
 
 8.43 
 
 .f4 
 
 0.0^9 
 
 0.104 
 
 .060 
 
 8.0 
 
 12 
 
 9.15 
 
 10.63 
 
 0.097 
 
 0.113 
 
 .065 
 
 8.7 
 
 13 
 
 9.87 
 
 11.51 
 
 0.105 
 
 0.123 
 
 .070 
 
 9.4 
 
 14 
 
 10.57 
 
 12.33 
 
 0.113 
 
 0.132 
 
 .075 
 
 10.0 
 
 15 
 
 11.27 
 
 13.15 
 
 0.121 
 
 0.141 
 
 .OSO 
 
 10.6 
 
 16 
 
 11.96 
 
 13.95 
 
 0.129 
 
 0.151 
 
 .OS5 
 
 11.2 
 
 17 
 
 12.64 
 
 14.74 
 
 0.137 
 
 0.160 
 
 .090 
 
 11.9 
 
 18 
 
 13.31 
 
 15.53 
 
 0.145 
 
 0.169 
 
 .095 
 
 12.4 
 
 19 
 
 13.99 
 
 16.32 
 
 0.153 
 
 0.179 
 
 .100 
 
 13.0 
 
 20 
 
 14.67 
 
 17.11 
 
 0.161 
 
 0.188 
 
 .105 
 
 13.6 
 
 21 
 
 15.34 
 
 17.^9 
 
 0.170 
 
 0.198 
 
 .110 
 
 14.2 
 
 22 
 
 16.00 
 
 18.67 
 
 0.177 
 
 0.207 
 
 .115 
 
 14.9 
 
 23 
 
 16.67 
 
 19.45 
 
 0.186 
 
 0.217 
 
 .120 
 
 15.4 
 
 24 
 
 17.34 
 
 20.23 
 
 0.195 
 
 0.227 
 
 .125 
 
 16.0 
 
 25 
 
 18.00 
 
 21.00 
 
 0.202 
 
 0.236 
 
 .130 
 
 16.5 
 
 26 
 
 18.66 
 
 21.77 
 
 0.211 
 
 0.246 
 
 .135 
 
 17.1 
 
 27 
 
 19.32 
 
 22.54 
 
 0.219 
 
 0.256 
 
 .140 
 
 17.7 
 
 23 
 
 19.93 
 
 23.31 
 
 0.223 
 
 0.266 
 
 .145 
 
 18.3 
 
 29 
 
 20.64 
 
 24. OS 
 
 0.237 
 
 0.276 
 
 .150 
 
 18.8 
 
 30 - 
 
 21.29 
 
 24.84 
 
 0.245 
 
 0.26 
 
 .155 
 
 19.3 
 
 31 
 
 21.94 
 
 25.60 
 
 0.254 
 
 0.296 
 
 .160 
 
 19.8 
 
 '32 
 
 22.60 
 
 26.36 
 
 0.262 
 
 0.306 
 
 1.165 
 
 20.3 
 
 33 
 
 23.25 
 
 27.12 
 
 0.271 
 
 0.316 
 
 1.170 
 
 20.9 
 
 34 
 
 23.90 
 
 27.88 
 
 0.279 
 
 0.326 
 
 1.175 
 
 21.4 
 
 35 
 
 24.54 
 
 28.63 
 
 0.238 
 
 0.336 
 
 1.180 
 
 22.0 
 
 36 
 
 25.18 
 
 29.33 
 
 0.297 
 
 0.317 
 
 1.185 
 
 22.5 
 
 37 
 
 25.83 
 
 30.13 
 
 0.306 
 
 0.357 
 
 1.190 
 
 23.0 
 
 38 
 
 26.47 
 
 30.88 
 
 0.315 
 
 0.367 
 
 1.195 
 
 23.5 
 
 39 
 
 27.10 
 
 31.62 
 
 0.324 
 
 0.378 
 
 1.200 
 
 24.0 
 
 40 
 
 27.74 
 
 32.36 
 
 0.333 
 
 0.388 
 
 1.205 
 
 24.5 
 
 41 
 
 23.36 
 
 33.09 
 
 0.342 
 
 0.399 
 
 1.210 
 
 25.0 
 
 42 
 
 23.99 
 
 33.82 
 
 0.351 
 
 0.409 
 
 1.215 
 
 25.5 
 
 43 
 
 29.61 
 
 34.55 
 
 0.360 
 
 0.420 
 
 1.220 
 
 26.0 
 
 44 
 
 30.24 
 
 35.23 
 
 0.369 
 
 0.430 
 
 1.225 
 
 26.4 
 
 45 
 
 30.88 
 
 36.03 
 
 0.378 
 
 0.441 
 
 1.230 
 
 26.9 
 
 46 
 
 31.53 
 
 36.78 
 
 0.387 
 
 0.452 
 
 1.235 
 
 27.4 
 
 47 
 
 32.17 
 
 37.53 
 
 0.397 
 
 0.463 
 
 1.240 
 
 27.9 
 
 48 
 
 32.82 
 
 38.29 
 
 0.407 
 
 0.475 
 
 1.245 
 
 28.4 
 
 49 
 
 33.47 
 
 39.05 
 
 0.417 
 
 0.486 
 
 1.250 
 
 28.8 
 
 50 
 
 34.13 
 
 39.82 
 
 0.427 
 
 0.498 
 
 * Zeit. angew. Chem., 1891, 165. 
 
 + For the American Standard Baume' scale and the specific-gravity table adopted by the 
 Manufacturing Chemists' Association of the United States, see Van Nostrand s Chemical 
 Annual. 
 
TABLES. 
 
 521 
 
 SPECIFIC GRAVITY OF NITRIC ACID AT 15 C (Continued). 
 
 Specific 
 Gravity 
 
 *v 
 
 *n Vacuo. 
 
 Degrees 
 Baume. 
 
 Degrees 
 Twaddell. 
 
 100 Parts by Weight 
 Contain, Grams, 
 
 I Liter Contains in 
 Kilograms, 
 
 N 2 5 . 
 
 HNO 3 . 
 
 N.OS. 
 
 HN0 3 . 
 
 1.255 
 
 29.3 
 
 51 
 
 34.78 
 
 40.53 
 
 0.437 
 
 0.509 
 
 1.260 
 
 29.7 
 
 52 
 
 35.44 
 
 41.34 
 
 0.447 
 
 0.521 
 
 1.265 
 
 30.2 
 
 53 
 
 36.09 
 
 42.10 
 
 0.457 
 
 5.3 
 
 1.270 
 
 30.6 
 
 54 
 
 36.75 
 
 42.87 
 
 0.467 
 
 0.5^4 
 
 1.275 
 
 31.1 
 
 55 
 
 37.41 
 
 43.64 
 
 0.477 
 
 0.556 
 
 1.280 
 
 31.5 
 
 56 
 
 38.07 
 
 44.41 
 
 0.487 
 
 0.56J 
 
 1.285 
 
 32.0 
 
 57 
 
 38.73 
 
 45.18 
 
 0.498 
 
 0.5 1 
 
 1.290 
 
 32.4 
 
 58 
 
 39.39 
 
 45.95 
 
 0.503 
 
 0.593 
 
 1.295 
 
 32.8 
 
 59 
 
 40.05 
 
 46.72 
 
 0.519 
 
 0.605 
 
 1.300 
 
 33.3 
 
 60 
 
 40.71 
 
 47.49 
 
 0.529 
 
 0.617 
 
 1.305 
 
 33.7 
 
 61 
 
 41.37 
 
 48.26 
 
 0.540 
 
 0.630 
 
 1.310 
 
 34.2 
 
 62 
 
 42.06 
 
 49.07 
 
 0.551 
 
 0.643 
 
 1.315 
 
 34.6 
 
 63 
 
 42.76 
 
 49.89 
 
 0.562 
 
 0.656 
 
 1.320 
 
 35.0 
 
 64 
 
 43.47 
 
 50.71 
 
 0.573 
 
 0.669 
 
 1.325 
 
 35.4 
 
 65 
 
 44.17 
 
 51.53 
 
 0.5S5 
 
 0.6S3 
 
 1.330 
 
 35.8 
 
 66 
 
 44.89 
 
 52.37 
 
 0.597 
 
 0.697 
 
 1.3325 
 
 36.0 
 
 665 
 
 45.26 
 
 52.80 
 
 0.603 
 
 0.704 
 
 1.335 
 
 36.2 
 
 67 
 
 45.62 
 
 53.22 
 
 0.609 
 
 0.710 
 
 1.340 
 
 36.6 
 
 68 
 
 46.35 
 
 54.07 
 
 0.621 
 
 0.725 
 
 1.345 
 
 37.0 
 
 69 
 
 47.03 
 
 54.93 
 
 0.633 
 
 0.739 
 
 1.350 
 
 37.4 
 
 70 
 
 47.82 
 
 55.79 
 
 0.645 
 
 0.753 
 
 1.355 
 
 37.8 
 
 71 
 
 48.57 
 
 56.66 
 
 0.653 
 
 0.768 
 
 1.360 
 
 38.2 
 
 72 
 
 49.35 
 
 57.57 
 
 0.671 
 
 0.783 
 
 1.365 
 
 38.6 
 
 73 
 
 50.13 
 
 58.48 
 
 0.6S4 
 
 0.798 
 
 1.370 
 
 39.0 
 
 74 
 
 50.91 
 
 59.39 
 
 0.698 
 
 0.814 
 
 1.375 
 
 39.4 
 
 75 
 
 51.69 
 
 60.30 
 
 0.711 
 
 0.829 
 
 1.380 
 
 39.8 
 
 76 
 
 52.52 
 
 61.27 
 
 0.725 
 
 0.846 
 
 1.3833 
 
 40.0 
 
 
 53.08 
 
 61.92 
 
 0.735 
 
 0.857 
 
 1.385 
 
 40.1 
 
 77 
 
 53.35 
 
 62.24 
 
 0.739 
 
 0.862 
 
 1.390 
 
 40.5 
 
 78 
 
 54.20 
 
 63.23 
 
 0.753 
 
 0.879 
 
 1.395 
 
 40.8 
 
 79 
 
 55.07 
 
 64.25 
 
 0.76S 
 
 0.896 
 
 1.400 
 
 41.2 
 
 80 
 
 55.97 
 
 65.30 
 
 0.783 
 
 0.914 
 
 1.405 
 
 41.6 
 
 81 
 
 56.92 
 
 66.40 
 
 0.800 
 
 0.933 
 
 1.410 
 
 42.0 
 
 82 
 
 57.86 
 
 67.50* 
 
 0.816 
 
 0.952 
 
 1.415 
 
 42.3 
 
 83 
 
 58.83 
 
 68.63 
 
 0.832 
 
 0.971 
 
 1.420 
 
 42.7 
 
 84 
 
 59.83 
 
 69.80 
 
 0.849 
 
 0.991 
 
 1.425 
 
 43.1 
 
 85 
 
 60.84 
 
 70.98 
 
 0.867 
 
 1.011 
 
 1.430 
 
 43.4 
 
 86 
 
 61.86 
 
 72.17 
 
 0.885 
 
 1.032 
 
 1.435 
 
 43.8 
 
 87 
 
 62.91 
 
 73.39 
 
 0.903 
 
 1.053 
 
 1.440 
 
 44.1 
 
 88 
 
 64.01 
 
 74.68 
 
 0.921 
 
 1.075 
 
 1.445 
 
 44.4 
 
 89 
 
 65.13 
 
 75.98 
 
 0.941 
 
 1.098 
 
 1.450 
 
 44.8 
 
 90 
 
 66.24 
 
 77.28 
 
 0.961 
 
 1.121 
 
 1.455 
 
 45.1 
 
 91 
 
 67.38 
 
 78.60 
 
 0.981 
 
 1.144 
 
 1.460 
 
 45.4 
 
 92 
 
 68.56 
 
 79.98 
 
 1.001 
 
 1.168 
 
 1.465 
 
 45.8 
 
 93 
 
 69.79 
 
 81.42 
 
 1.023 
 
 1.193 
 
 1.470 
 
 46.1 
 
 94 
 
 71.06 
 
 82.90 
 
 1.045 
 
 1.219 
 
 1.475 
 
 46.4 
 
 95 
 
 72.39 
 
 84.45 
 
 1.068 
 
 1.246 
 
 1.480 
 
 46.8 
 
 96 
 
 73.76 
 
 86.05 
 
 1.092 
 
 1.274 
 
 1.485 
 
 47.1 
 
 97 
 
 75.18 
 
 87.70 
 
 1.116 
 
 1.302 
 
 1.490 
 
 47.4 
 
 98 
 
 76.80 
 
 89.60 
 
 1.144 
 
 1.335 
 
 1.495 
 
 47.8 
 
 99 
 
 78.52 
 
 91.60 
 
 1.174 
 
 1.369 
 
522 
 
 TABLES. 
 
 SPECIFIC GRAVITY OF NITRIC ACID AT 15 C. (Continued). 
 LUNGE AND REY. 
 
 Specific 
 Gravity 
 
 Degrees 
 Baum. 
 
 Degrees 
 Twaddell. 
 
 100 Parts by Weight 
 Contain, Grams, 
 
 1 Liter Contains in 
 Kilograms, 
 
 at 4 o 
 in Vacuo 
 
 N 2 5 . 
 
 HNO 3 . 
 
 N 2 5 . 
 
 HN0 3 . 
 
 1.500 
 
 48.1 
 
 100 
 
 80.65 
 
 94.09 
 
 1.210 
 
 1.411 
 
 1.501 
 
 .... 
 
 . 
 
 81.09 
 
 94.60 
 
 1.217 
 
 1.420 
 
 1.502 
 
 .... 
 
 . . . 
 
 81.50 
 
 95.03 
 
 1.224 
 
 1.42S 
 
 1.503 
 
 .... 
 
 ... 
 
 81.91 
 
 95.55 
 
 1.231 
 
 1.436 
 
 1.504 
 
 
 
 82.29 
 
 96.00 
 
 1.238 
 
 1.444 
 
 .505 
 
 48*.4 
 
 ioi 
 
 82.63 
 
 96.39 
 
 1.244 
 
 1.451 
 
 .506 
 
 .... 
 
 
 82.94 
 
 96.76 
 
 1.249 
 
 1.457 
 
 .507 
 
 
 
 83.26 
 
 97.13 
 
 1.255 
 
 . 1.464 
 
 .508 
 
 48 '.5 
 
 
 83. 53 
 
 97.50 
 
 1.260 
 
 1.470 
 
 .509 
 
 .... 
 
 
 83.87 
 
 97.84 
 
 1.265 
 
 1.476 
 
 .510 
 
 48*.7 
 
 102 
 
 84.09 
 
 93.10 
 
 1.270 
 
 1.481 
 
 .511 
 
 .... 
 
 . 
 
 84. 23 
 
 93.32 
 
 1.274 
 
 1.486 
 
 .512 
 
 .... 
 
 
 84.46 
 
 98.53 
 
 .277 
 
 1.490 
 
 .513 
 
 .... 
 
 ... 
 
 84.63 
 
 98.73 
 
 .280 
 
 1.494 
 
 .514 
 
 
 ? 
 
 84.78 
 
 98.90 
 
 .233 
 
 1.497 
 
 .515 
 
 49.6 
 
 ios 
 
 84.92 
 
 99.07 
 
 .237 
 
 1.501 
 
 .516 
 
 .... 
 
 
 85.04 
 
 99.21 
 
 .29 
 
 1.504 
 
 .517 
 
 
 
 85.15 
 
 99.34 
 
 .292 , 
 
 1.507 
 
 .518 
 
 
 ... 
 
 85.26 
 
 99.46 
 
 .294 
 
 1.510 
 
 .519 
 
 
 
 85.35 
 
 99.57 
 
 .296 
 
 1.512 
 
 1.520 
 
 49*.4 
 
 104 
 
 85.44 
 
 99.67 
 
 .299 
 
 1.515 
 
TABLES 
 
 523 
 
 TABLE VIII. SPECIFIC GRAVITY OF SULPHURIC ACID. 
 LUNGE, ISLER, AND NAEF.* 
 
 Specific 
 Gravity 
 
 Degrees 
 Baume.f 
 
 Degrees 
 Twaddell. 
 
 100 Parts by Weight 
 Contain, Grams, 
 
 1 Liter Contains in 
 Kilograms, 
 
 at 40 
 in Vacuo. 
 
 SO 3 . 
 
 H2SO 4 . 
 
 SO 3 . 
 
 H2S0 4 . 
 
 1.000 
 
 
 
 
 
 0.07 
 
 0.09 
 
 0.001 
 
 0.001 
 
 1.005 
 
 0.7 
 
 1 
 
 0.68 
 
 0.83 
 
 0.007 
 
 0.008 
 
 1.010 
 
 1.4 
 
 2 
 
 1.28 
 
 1.57 
 
 0.013 
 
 0.016 
 
 1.015 
 
 2.1 
 
 3 
 
 1.88 
 
 2.30 
 
 0.019 
 
 0.023 
 
 1.020 
 
 2.7 
 
 4 
 
 2.47 
 
 3.03 
 
 0.025 
 
 0.031 
 
 1.025 
 
 3.4 
 
 5 
 
 3.07 
 
 3.76 
 
 0.032 
 
 0.039 
 
 1.030 
 
 4.1 
 
 6 
 
 3.67 
 
 4.49 
 
 0.038 
 
 0.046 
 
 .035 
 
 4.7 
 
 7 
 
 4.27 
 
 5.23 
 
 0.044 
 
 0.054 
 
 .040 
 
 5.4 
 
 8 
 
 4.87 
 
 5.96 
 
 0.051 
 
 0.062 
 
 .045 
 
 6.0 
 
 9 
 
 5.45 
 
 6.67 
 
 0.057 
 
 0.071 
 
 .050 
 
 6.7 
 
 10 
 
 6.02 
 
 7.37 
 
 0.063 
 
 0.077 
 
 .055 
 
 7.4 
 
 11 
 
 6.59 
 
 8.07 
 
 0.070 
 
 0.085 
 
 .060 
 
 8.0 
 
 12 
 
 7.16 
 
 8.77 
 
 0.076 
 
 0.093 
 
 .065 
 
 8.7 
 
 13 
 
 7.73 
 
 9.47 
 
 O.OS2 
 
 0.102 
 
 1.070 
 
 9.4 
 
 14 
 
 8.32 
 
 10.19 
 
 0.089 
 
 0.109 
 
 1.075 
 
 10.0 
 
 15 
 
 8.90 
 
 10.90 
 
 0.096 
 
 0.117 
 
 1.080 
 
 10.6 
 
 16 
 
 9.47 
 
 11.60 
 
 0.103 
 
 0.125 
 
 1.085 
 
 11.2 
 
 17 
 
 10.04 
 
 12.30 
 
 0.109 
 
 0.133 
 
 1.090 
 
 11.9 
 
 18 
 
 10.60 
 
 12.99 
 
 0.116 
 
 0.142 
 
 1.095 
 
 12.4 
 
 19 
 
 11.16 
 
 13.67 
 
 0.122 
 
 0.150 
 
 1.100 
 
 13.0 
 
 20 
 
 11.71 
 
 14.35 
 
 0.129 
 
 0.158 
 
 1.105 
 
 13.6 
 
 21 
 
 12.27 
 
 15.03 
 
 0.136 
 
 0.166 
 
 1.110 
 
 14.2 
 
 22 
 
 12.82 
 
 15.71 
 
 0.143 
 
 0.175 
 
 1.115 
 
 14.9 
 
 23 
 
 13.36 
 
 16.36 
 
 0.149 
 
 0.183 
 
 1.120 
 
 15.4 
 
 24 
 
 13.89 
 
 17.01 
 
 0.156 
 
 0.191 
 
 1.125 
 
 16.0 
 
 25 
 
 14.42 
 
 17.66 
 
 0.162 
 
 0.199 
 
 1.130 
 
 16.5 
 
 26 
 
 14.95 
 
 18.31 
 
 0.169 
 
 0.207 
 
 1.135 
 
 17.1 
 
 27 
 
 15.48 
 
 18.96 
 
 0.176 
 
 0.215 
 
 .140 
 
 17.7 
 
 28 
 
 16.01 
 
 19.61 
 
 0.183 
 
 0.223 
 
 .145 
 
 18.3 
 
 29 
 
 16.54 
 
 20.26 
 
 0.189 
 
 0.231 
 
 .150 
 
 18.8 
 
 30 
 
 17.07 
 
 20.91 
 
 0.196 
 
 0.239 
 
 .155 
 
 19.3 
 
 31 
 
 17.59 
 
 21.55 
 
 0.203 
 
 0.248 
 
 .160 
 
 19.8 
 
 32 
 
 18.11 
 
 22.19 
 
 0.210 
 
 0.257 
 
 .165 
 
 20.3 
 
 33 
 
 18.64 
 
 22.83 
 
 0.217 
 
 0.266 
 
 .170 
 
 20.9 
 
 34 
 
 19.16 
 
 23.47 
 
 0.224 
 
 0.275 
 
 .175 
 
 21.4 
 
 35 
 
 19.69 
 
 24.12 
 
 0.231 
 
 0.283 
 
 .180 
 
 22.0 
 
 36 
 
 20.21 
 
 24.76' 
 
 0.238 
 
 0.292 
 
 1.185 
 
 22.5 
 
 37 
 
 20.73 
 
 25.40 
 
 0.246 
 
 0.301 
 
 1.190 
 
 23.0 
 
 38 
 
 21.26 
 
 26.04 
 
 0,253 
 
 0.310 
 
 1.195 
 
 23.5 
 
 39 
 
 21.78 
 
 26.68 
 
 0.260 
 
 0.319 
 
 1.200 
 
 24.0 
 
 40 
 
 22.30 
 
 27.32 
 
 0.268 
 
 0.328 
 
 1.205 
 
 24.5 
 
 41 
 
 22.82 
 
 27.95 
 
 0.275 
 
 0.337 
 
 1.210 
 
 25.0 
 
 42 
 
 23.33 
 
 28.58 
 
 0.282 
 
 0.346 
 
 1.215 
 
 25.5 
 
 43 
 
 23.84 
 
 29.21 
 
 0.290 
 
 0.355 
 
 1.220 
 
 26.0 
 
 44 
 
 24.36 
 
 29.84 
 
 0.297 
 
 0.364 
 
 .225 
 
 26.4 
 
 45 
 
 24.88 
 
 30.48 
 
 0.305 
 
 0.373 
 
 .230 
 
 26.9 
 
 46 
 
 25.39 
 
 31.11 
 
 0.312 
 
 0.382 
 
 .235 
 
 27.4 
 
 47 
 
 25.88 
 
 31.70 
 
 0.320 
 
 0.391 
 
 .240 
 
 27.9 
 
 48 
 
 26.35 
 
 32.28 
 
 0.327 0.400 
 
 .245 
 
 28.4 
 
 49 
 
 26.83 
 
 32.86 
 
 0.334 0.409 
 
 .250 28.8 
 
 50 
 
 27.29 
 
 33.43 
 
 0.341 0.418 
 
 .255 
 
 29.3 
 
 51 
 
 27.76 
 
 34.00 
 
 0.348 
 
 0.426 
 
 .260 29.7 
 
 52 
 
 28.22 
 
 34.57 
 
 0.356 
 
 0.435 
 
 *Zeit. angew. Chem., 1890, 131; Chem. Ind., 1883, 39. 
 
 + For the American Standard Baumd scale and the specific-gravity table adopted by the 
 Manufacturing Chemists' Association of the United States, see Van Nostrand's Chemical 
 Annual. 
 
524 
 
 TABLES. 
 
 SPECIFIC GRAVITY OF SULPHURIC ACID (Continued). 
 
 Specific 
 Gravity 
 
 Degrees 
 Baume\ 
 
 Degrees 
 TwaddeU. 
 
 100 Parts by Weight 
 Contain, Grams, 
 
 1 Liter Contains in 
 Kilograms , 
 
 at 40 
 in Vacuo. 
 
 S0 3 . 
 
 H2SO-4. 
 
 SO 3 . 
 
 H 2 S0 4 . 
 
 1.265 
 
 30.2 
 
 53 
 
 28.69 
 
 35.14 
 
 0.363 
 
 0.444 
 
 1.270 
 
 30.6 
 
 54 
 
 29.15 
 
 35.71 
 
 0.370 
 
 0.454 
 
 1.275 
 
 31.1 
 
 55 
 
 29.62 
 
 36.29 
 
 0.377 
 
 0.462 
 
 1.280 
 
 31.5 
 
 56 
 
 30.10 
 
 36.87 
 
 0.385 
 
 0.472 
 
 1.285 
 
 32.0 
 
 57 
 
 30.57 
 
 37.45 
 
 0.393 
 
 0.4S1 
 
 1.290 
 
 32.4 
 
 58 
 
 31.04 
 
 38.03 
 
 0.400 
 
 0.490 
 
 1.295 
 
 32.8 
 
 59 
 
 31.52 
 
 38.61 
 
 0.40S 
 
 0.500 
 
 1.300 
 
 33.3 
 
 60 
 
 31.99 
 
 39.19 
 
 0.416 
 
 0.510 
 
 1.305 
 
 33.7 
 
 61 
 
 32.46 
 
 39.77 
 
 0.424 
 
 0.519 
 
 1.310 
 
 34.2 
 
 62 
 
 32.94 
 
 40.35 
 
 0.432 
 
 0.529 
 
 1.315 
 
 34.6 
 
 63 
 
 33.41 
 
 40.93 
 
 0.439 
 
 0.538 
 
 1.320 
 
 35.0 
 
 64 
 
 33.88 
 
 41.50 
 
 0.447 
 
 0.548 
 
 1.325 
 
 35.4 
 
 65 
 
 34.35 
 
 42.08 
 
 0.455 
 
 0.557 
 
 1.330 
 
 35.8 
 
 66 
 
 34.80 
 
 42.66 
 
 0.462 
 
 0.567 
 
 1.335 
 
 36.2 
 
 67 
 
 35.27 
 
 43.20 
 
 0.471 
 
 0.577 
 
 1.340 
 
 36.6 
 
 68 
 
 35.71 
 
 43.74 
 
 0.479 
 
 0.586 
 
 .345 
 
 37.0 
 
 69 
 
 36.14 
 
 44.28 
 
 0.486 
 
 0.596 
 
 .350 
 
 37.4 
 
 70 
 
 36.58 
 
 44.82 
 
 0.494 
 
 0.605 
 
 .355 
 
 37.8 
 
 71 
 
 37.02 
 
 45.35 
 
 0.502 
 
 0.614 
 
 .360 
 
 38.2 
 
 72 
 
 37.45 
 
 45.88 
 
 0.509 
 
 0.624 
 
 .365 
 
 38.6 
 
 73 
 
 37.89 
 
 46.41 
 
 0.517 
 
 0.633 
 
 .370 
 
 39.0 
 
 74 
 
 38.32 
 
 46.94 
 
 0.525 
 
 0.643 
 
 .375 
 
 39.4 
 
 75 
 
 38.75 
 
 47.47 
 
 0.533 
 
 0.653 
 
 .380 
 
 39.8 
 
 76 
 
 39.18 
 
 48.00 
 
 0.541 
 
 0.662 
 
 .385 
 
 40.1 
 
 77 
 
 39.62 
 
 48.53 
 
 0.549 
 
 0.672 
 
 .390 
 
 40.5 
 
 78 
 
 40.05 
 
 49.06 
 
 0.557 
 
 0.682 
 
 .395 
 
 40.8 
 
 79 
 
 40.48 
 
 49.59 
 
 0.564 
 
 0.692 
 
 .400 
 
 41.2 
 
 80 
 
 40.91 
 
 50.11 
 
 0.573 
 
 0.702 
 
 .405 
 
 41.6 
 
 81 
 
 41.33 
 
 50.63 
 
 0.581 
 
 0.711 
 
 .410 
 
 42.0 
 
 82 
 
 41.76 
 
 51.15 
 
 0.589 
 
 0.721 
 
 .415 
 
 42.3 
 
 83 
 
 42.17 
 
 51.66 
 
 0.597 
 
 0.730 
 
 .420 
 
 42.7 
 
 84 
 
 42.57 
 
 52.15 
 
 0.604 
 
 0.740 
 
 .425 
 
 43.1 
 
 85 
 
 42.96 
 
 52.63 
 
 0.612 
 
 0.750 
 
 1.430 
 
 43.4 
 
 86 
 
 43.36 
 
 53.11 
 
 0.620 
 
 0.759 
 
 1.435 
 
 43.8 
 
 87 
 
 43.75 
 
 53.59 
 
 0.628 
 
 0.769 
 
 1.440 
 
 44.1 
 
 88 
 
 44.14 
 
 54.07 
 
 0.636 
 
 0.779 
 
 1.445 
 
 44.4 
 
 89 
 
 44.53 
 
 54.55 
 
 0.643 
 
 0.789 
 
 1.450 
 
 44.8 
 
 90 
 
 44.92 
 
 55.03 
 
 0.651 
 
 0.798 
 
 1.455 
 
 45.1 
 
 91 
 
 45.31 
 
 55.50 
 
 0.659 
 
 0.808 
 
 1.460 
 
 45.4 
 
 92 
 
 45.69 
 
 55.97 
 
 0.667 
 
 0.817 
 
 1.465 
 
 45.8 
 
 93 
 
 46.07 
 
 56.43 
 
 0.675 
 
 0.827 
 
 1.470 
 
 46.1 
 
 94 
 
 46.45 
 
 56.90 
 
 0.683 
 
 0.837 
 
 1.475 
 
 46.4 
 
 95 
 
 46.83 
 
 57.37 
 
 0.691 
 
 0.846 
 
 1.480 
 
 46.8 
 
 96 
 
 47.21 
 
 57.83 
 
 0.699 
 
 0.856 
 
 1.485 
 
 47.1 
 
 97 
 
 47.57 
 
 58.28 
 
 0.707 
 
 0.865 
 
 1.490 
 
 47.4 
 
 98 
 
 47.95 
 
 58.74 
 
 0.715 
 
 0.876 
 
 1.495 
 
 47.8 
 
 99 
 
 48.34 
 
 59.22 
 
 0.723 
 
 0.885 
 
 1.500 
 
 48.1 
 
 100 
 
 48.73 
 
 59 . 70 
 
 0.731 
 
 0.896 
 
 1.505 
 
 48.4 
 
 101 
 
 49.12 
 
 60.18 
 
 0.739 
 
 0.906 
 
 1.510 
 
 48.7 
 
 102 
 
 49.51 
 
 60.65 
 
 0.748 
 
 0.916 
 
 1.515 
 
 49.0 
 
 103 
 
 49.89 
 
 61.12 
 
 0.756 
 
 0.926 
 
 1.520 
 
 49.4 
 
 104 
 
 50.28 
 
 61.59 
 
 0.764 
 
 0.936 
 
 1 525 
 
 49.7 
 
 105 
 
 50.66 
 
 62.06 
 
 0.773 
 
 0.946 
 
 1 530 
 
 50.0 
 
 106 
 
 51.04 
 
 62.53 
 
 0.781 
 
 0.957 
 
TABLES. 
 
 525 
 
 SPECIFIC GRAVITY OF SULPHURIC ACID (Continued). 
 
 Specific 
 Gravity 
 15 
 *lo- 
 in Vacuo. 
 
 Degrees 
 Baumd. 
 
 Degrees 
 Twaddell. 
 
 100 Parts by Weight 
 Contain, Grams, 
 
 1 Liter Contains in 
 Kilograms, 
 
 S0 3 . 
 
 H 2 S0 4 . 
 
 SO 3 . 
 
 H2SO,. 
 
 1.535 
 
 50.3 
 
 107 
 
 51.43 
 
 63.00 
 
 0.789 
 
 0.967 
 
 1.540 
 
 50.6 
 
 108 
 
 51.78 
 
 63.43 
 
 0.797 
 
 0.977 
 
 .545 
 
 50.9 
 
 109 
 
 52.12 
 
 63.85 
 
 0.805 
 
 0.987 
 
 .550 
 
 51.2 
 
 110 
 
 52.46 
 
 64.26 
 
 0.813 
 
 0.996 
 
 .555 
 
 51.5 
 
 111 
 
 52.79 
 
 64.67 
 
 0.821 
 
 1.006 
 
 .560 
 
 51.8 
 
 112 
 
 53.12 
 
 65.08 
 
 0.829 
 
 1.015 
 
 .565 
 
 52.1 
 
 113 
 
 53.46 
 
 65.49 
 
 0.837 
 
 1.025 
 
 .570 
 
 52.4 
 
 114 
 
 53.80 
 
 65.90 
 
 0.845 
 
 1.035 
 
 .575 
 
 52.7 
 
 115 
 
 54.13 
 
 66.30 
 
 0.853 
 
 1.044 
 
 .580 
 
 53.0 
 
 116 
 
 54.46 
 
 66.71 
 
 0.861 
 
 1.054 
 
 .585 
 
 53.3 
 
 117 
 
 54.80 
 
 67.13 
 
 0.869 
 
 1.064 
 
 .590 
 
 53.6 
 
 118 
 
 55.18 
 
 67.59 
 
 0.877 
 
 1.075 
 
 .595 
 
 53.9 
 
 119 
 
 55.55 
 
 68.05 
 
 0.886 
 
 1.085 
 
 .600 
 
 54.1 
 
 120 
 
 55.93 
 
 68.51 
 
 0.895 
 
 1.096 
 
 .605 
 
 54.4 
 
 121 
 
 56.30 
 
 6S.97 
 
 0.904 
 
 1.107 
 
 .610 
 
 54.7 
 
 122 
 
 56.63 
 
 69.43 
 
 0.913 
 
 1.118 
 
 .615 
 
 55.0 
 
 123 
 
 57.05 
 
 69.89 
 
 0.921 
 
 1.123 
 
 .620 
 
 55.2 
 
 124 
 
 57.40 
 
 70.32 
 
 0.930 
 
 1.139 
 
 .625 
 
 55.5 
 
 125 
 
 57.75 
 
 70.74 
 
 0.938 
 
 1.150 
 
 .630 
 
 55.8 
 
 126 
 
 58.09 
 
 71.16 
 
 0.947 
 
 1.160 
 
 .635 
 
 56.0 
 
 127 
 
 58.43 
 
 71.57 
 
 0.955 
 
 1.170 
 
 .640 
 
 56.3 
 
 128 
 
 58.77 
 
 71.99 
 
 0.964 
 
 1.181 
 
 .645 
 
 56.6 
 
 129 
 
 59.10 
 
 72.40 
 
 0.972 
 
 1.192 
 
 .650 
 
 56.9 
 
 130 
 
 59.45 
 
 72.82 
 
 0.981 
 
 1.202 
 
 .655 
 
 57.1 
 
 131 
 
 59.73 
 
 73.23 
 
 0.9S9 
 
 1.212 
 
 .660 
 
 57.4 
 
 132 
 
 60.11 
 
 73.64 
 
 0.993 
 
 1.222 
 
 .665 
 
 57.7 
 
 133 
 
 60.46 
 
 74.07 
 
 1.007 
 
 1.233 
 
 .670 
 
 57.9 
 
 134 
 
 60.82 
 
 74.51 
 
 1.016 
 
 1.244 
 
 .675 
 
 58.2 
 
 135 
 
 61.20 
 
 74.97 
 
 1.025 
 
 1.256 
 
 .680 
 
 58.4 
 
 136 
 
 61.57 
 
 75.42 
 
 1.034 
 
 1.267 
 
 .685 
 
 58.7 
 
 137 
 
 61.93 
 
 75.86 
 
 1.043 
 
 1.278 
 
 .690 
 
 58.9 
 
 138 
 
 62.29 
 
 76.30 
 
 1.053 
 
 1.289 
 
 .695 
 
 59.2 
 
 139 
 
 62.64 
 
 76.73 
 
 1.062 
 
 1.301 
 
 1.700 
 
 59.5 
 
 140 
 
 63.00 
 
 77.17 
 
 1.071 
 
 1.312 
 
 1.705 
 
 59.7 
 
 141 
 
 63.35 
 
 77.60 
 
 1.030 
 
 1.323 
 
 1.710 
 
 60.0 
 
 142 
 
 63.70 
 
 78.04 
 
 1.019 
 
 1.334 
 
 1.715 
 
 60.2 
 
 143 
 
 64.07 
 
 78.48 
 
 1.099 
 
 1.346 
 
 1.720 
 
 60.4 
 
 144 
 
 64.43 
 
 78.92 
 
 .103 
 
 1.357 
 
 1.725 
 
 60.6 
 
 145 
 
 64.78 
 
 79.36 
 
 .118 
 
 .369 
 
 1.730 
 
 60.9 
 
 146 
 
 65.14 
 
 79.80 
 
 .127 
 
 .381 
 
 1.735 
 
 61.1 
 
 147 
 
 65.50 
 
 80.24 
 
 .136 
 
 .392 
 
 1.740 
 
 61.4 
 
 148 
 
 65.86 
 
 80.68 
 
 .146 
 
 .404 
 
 1.745 
 
 61.6 
 
 149 
 
 66.22 
 
 81.12 
 
 .156 
 
 .416 
 
 .750 
 
 61.8 
 
 150 
 
 66.58 
 
 81.56 
 
 .165 
 
 .427 
 
 .755 
 
 62.1 
 
 151 
 
 66.94 
 
 82.00 
 
 .175 
 
 .439 
 
 .760 
 
 62.3 
 
 152 
 
 67.30 
 
 82.44 
 
 .185 
 
 .451 
 
 .765 
 
 62.5 
 
 153 
 
 67.65 
 
 82.88 
 
 .194 
 
 .463 
 
 .770 
 
 62.8 
 
 154 
 
 68.02 
 
 83.32 
 
 .204 
 
 .475 
 
 .775 
 
 63.0 
 
 155 
 
 68.49 
 
 83.90 
 
 .216 
 
 .489 
 
 .780 
 
 63.2 
 
 156 
 
 68.98 
 
 84.50 
 
 .228 
 
 .504 
 
 .785 
 
 63.5 
 
 157 
 
 69.47 
 
 85.10 
 
 .240 
 
 .519 
 
 .790 
 
 63.7 
 
 158 
 
 69.96 
 
 85.70 
 
 .252 
 
 .534 
 
 .795 
 
 64.0 
 
 159 
 
 70.45 
 
 86.30 
 
 .265 
 
 1.549 
 
 .800 
 
 64.2 
 
 160 
 
 70.94 
 
 86.90 
 
 1.277 
 
 1 . 564 
 
526 TABLES. 
 
 SPECIFIC GRAVITY OF SULPHURIC ACID (Continued). 
 
 Specific 
 Gravity 
 
 a* -40- 
 in Vacuo. 
 
 Degrees 
 Baume\ 
 
 Degrees 
 Twaddell. 
 
 100 Parts by Weight 
 Contain, Grams, 
 
 1 Liter Contains in 
 Kilograms, 
 
 SO 3 . 
 
 H 2 S0 4 . 
 
 SO 3 . 
 
 H2SO 4 . 
 
 1.805 
 
 64.4 
 
 161 
 
 71.50 
 
 87.60 
 
 1.291 
 
 .581 
 
 1.810 
 
 64.6 
 
 162 
 
 72.08 
 
 88.30 
 
 .305 
 
 .598 
 
 1.815 
 
 64.8 
 
 163 
 
 72.69 
 
 9.05 
 
 .319 
 
 .621 
 
 1.820 
 
 65.0 
 
 164 
 
 73.51 
 
 90.05 
 
 .338 
 
 .639 
 
 1.821 
 
 
 
 73.63 
 
 90.20 
 
 .341 
 
 .643 
 
 1.822 
 
 65.1 
 
 
 73.80 
 
 90.40 
 
 .345 
 
 .647 
 
 1.823 
 
 .... 
 
 
 73.96 
 
 90.60 
 
 .348 
 
 .651 
 
 1.824 
 
 65.2 
 
 
 74.12 
 
 90.80 
 
 .352 
 
 .656 
 
 1.825 
 
 .... 
 
 ies 
 
 74.29 
 
 91.00 
 
 .356 
 
 .661 
 
 1.826 
 
 65'.3 
 
 
 74.49 
 
 91.25 
 
 .360 
 
 .666 
 
 1.827 
 
 .... 
 
 
 74.69 
 
 91.50 
 
 .364 
 
 .671 
 
 1.828 
 
 Q5A 
 
 
 74.86 
 
 91.70 
 
 .368 
 
 .676 
 
 1.829 
 
 
 
 75.03 
 
 91.90 
 
 .372 
 
 .681 
 
 1.830 
 
 
 iee 
 
 75.19 
 
 92.10 
 
 .376 
 
 .685 
 
 1.831 
 
 65 '.5 
 
 
 75.35 
 
 92.30 
 
 1.380 
 
 .690 
 
 1.832 
 
 
 
 75.53 
 
 92.52 
 
 1.384 
 
 .695 
 
 1.833 
 
 65! 6 
 
 
 75.72 
 
 92.75 
 
 1.388 
 
 .700 
 
 1.834 
 
 .... 
 
 , 
 
 75.96 
 
 93.05 
 
 1.393 
 
 1.706 
 
 1.835 
 
 65!7 
 
 i67 
 
 76.27 
 
 93.43 
 
 1.400 
 
 1.713 
 
 1.836 
 
 .... 
 
 
 76.57 
 
 93.80 
 
 1.406 
 
 1.722. 
 
 1.837 
 
 .... 
 
 
 76.90 
 
 94.20 
 
 1.412 
 
 1.730 
 
 1.838 
 
 65^8 
 
 
 77.23 
 
 94.60 
 
 1.419 
 
 1.739 
 
 1.839 
 
 
 
 77.55 
 
 95.00 
 
 .426 
 
 .748 
 
 1.840 
 
 65'.9 
 
 ies 
 
 78.04 
 
 95.60 
 
 .436 
 
 .759 
 
 1.8405 
 
 .... 
 
 
 78.33 
 
 95.95 
 
 .451 
 
 .765 
 
 1.8410 
 
 
 
 79.19 
 
 97.00 
 
 .453 
 
 .7,6 
 
 1.8415 
 
 .... 
 
 
 79.76 
 
 97.70 
 
 .469 
 
 .799 
 
 1.8410 
 
 .... 
 
 
 80.16 
 
 98.20 
 
 .476 
 
 .808 
 
 1.8405 
 
 .... 
 
 . 
 
 80.57 
 
 98.70 
 
 .483 
 
 1.816 
 
 1.8400 
 
 .... 
 
 
 80.98 
 
 99.20 
 
 .490 
 
 1.825 
 
 1.8395 
 
 .... 
 
 
 81.18 
 
 99.45 
 
 .494 
 
 1.830 
 
 1.8390 
 
 .... 
 
 
 81.39 
 
 99.70 
 
 .497 
 
 1.834 
 
 1.8385 
 
 
 
 
 81.59 
 
 99.95 
 
 .500 
 
 1.838 
 
 TABLE IX. VAPOR TENSION OF WATER IN MILLIMETERS OF 
 
 MERCURY. 
 
 MAGNUS. 
 
 Temp. 
 
 Millimeters. 
 
 Temp. 
 
 Millimeters. 
 
 Temp. 
 
 Millimeters. 
 
 Temp. 
 
 Millimeters. 
 
 
 
 4.525 
 
 8 
 
 7.964 
 
 15 
 
 12.677 
 
 23 
 
 20.909 
 
 4-1 
 
 4.867 
 
 9 
 
 8.525 
 
 16 
 
 13.519 
 
 24 
 
 22.211 
 
 2 
 
 5.231 
 
 10 
 
 9.126 
 
 17 
 
 14.409 
 
 25 
 
 23.5-2 
 
 3 
 
 5.619 
 
 11 
 
 9.756 
 
 1$ 
 
 15.351 
 
 26 
 
 25.026 
 
 4 
 
 6.032 
 
 12 
 
 10.421 
 
 19 
 
 16.345 
 
 27 
 
 26.547 
 
 5 
 
 6.471 
 
 13 
 
 11.130 
 
 20 
 
 17.396 
 
 23 
 
 28.148 
 
 6 
 
 6.939 
 
 14 
 
 11.882 
 
 21 
 
 18.505 
 
 29 
 
 29>32 
 
 7 
 
 7 436 
 
 
 
 22 
 
 19 . 675 
 
 30 
 
 31.602 
 
 
 
 
 
 
 
 
 
TABLES. 
 
 527 
 
 TABLE X. DENSITY OF WATER. 
 
 ROSSETTI. 
 
 Tempera- 
 ture. 
 
 Density (Density 
 at = 1). 
 
 Volume (Volume 
 atO = l). 
 
 Density (Density 
 at4 = l). 
 
 Volume (Volume 
 at4 = l). 
 
 -10 
 
 0.998274 
 
 1.001729 
 
 0.998145 
 
 .001858 
 
 - 9 
 
 0.998556 
 
 1.001449 
 
 0.998427 
 
 .001575 
 
 - 8 
 
 0.998814 
 
 1.001191 
 
 0.998685 
 
 .001317 
 
 - 7 
 
 0.999040 
 
 1.000963 
 
 0.998911 
 
 .001089 
 
 - 6 
 
 0.999247 
 
 1.000756 
 
 0.999118 
 
 .000883 
 
 - 5 
 
 0.999428 
 
 1.000573 
 
 0.999298 
 
 .000702 
 
 - 4 
 
 0.999584 
 
 1.000416 
 
 0.999455 
 
 .000545 
 
 - 3 
 
 0.999719 
 
 1.000281 
 
 0.999590 
 
 .000410 
 
 - 2 
 
 0.999832 
 
 1.000168 
 
 0.999703 
 
 .000297 
 
 - 1 
 
 0.999926 
 
 1.000074 
 
 0.999797 
 
 .000203 
 
 
 
 1.000000 
 
 1.000000 
 
 0.999871 
 
 .000129 
 
 + 1 
 
 1.000057 
 
 0.999943 
 
 0.999928 
 
 .000072 
 
 2 
 
 1.000098 
 
 0.999902 
 
 0.999969 
 
 .000031 
 
 3 
 
 1.000120 
 
 0.999880 
 
 0.999991 
 
 .000009 
 
 4 
 
 1.000129 
 
 0.999871 
 
 1.000000 
 
 .000000 
 
 5 
 
 1.000119 
 
 0.999881 
 
 0.999990 
 
 .000010 
 
 6 
 
 1.000099 
 
 0.999901 
 
 0.999970 
 
 .000030 
 
 7 
 
 1.000062 
 
 0.999938 
 
 0.999933 
 
 .000067 
 
 8 
 
 1.000015 
 
 0.999985 
 
 0.999886 
 
 .000114 
 
 9 
 
 0.999953 
 
 1.000047 
 
 0.999824 
 
 .000176 
 
 10 
 
 0.999876 
 
 1.000124 
 
 0.999747 
 
 .000253 
 
 11 
 
 0.999784 
 
 1.000216 
 
 0.999655 
 
 .000345 
 
 12 
 
 0.999678 
 
 .000322 
 
 0.999549 
 
 .000451 
 
 13 
 
 0.999559 
 
 .000441 
 
 0.999430 
 
 .000570 
 
 14 
 
 0.999429 
 
 .000572 
 
 0.999299 
 
 .000701 
 
 15 
 
 0.999289 
 
 .000712 
 
 0.999160 
 
 .000841 
 
 16 
 
 0.999131 
 
 .000870 
 
 0.999002 
 
 .000999 
 
 17 
 
 0.998970 
 
 .001031 
 
 0.998841 
 
 .001160 
 
 18 
 
 0.998782 
 
 .001219 
 
 0.998654 
 
 .001348 i 
 
 19 
 
 0.998588 
 
 1.001413 
 
 0.998460 
 
 .001542 
 
 20 
 
 0.998388 
 
 1.001615 
 
 0.998259 
 
 .001744 - 
 
 21 
 
 0.998176 
 
 1.00182S 
 
 0.998047 
 
 .001957 
 
 22 
 
 0.997956 
 
 1.002048 
 
 0.997828 
 
 .002177 
 
 23 
 
 0.997730 
 
 1.002276 
 
 0.997601 
 
 .002405 
 
 24 
 
 0.997495 
 
 1.002511 
 
 0.997367 
 
 .002641 
 
 25 
 
 0.997249 
 
 1.002759 
 
 0.997120 
 
 .002888 
 
 26 
 
 0.996994 
 
 1.003014 
 
 0.996866 
 
 .003144 
 
 27 
 
 0.996732 
 
 1.003278 
 
 0.996603 
 
 .003408 
 
 28 
 
 0.996460 
 
 .003553 
 
 0.996331 
 
 .003682 
 
 29 
 
 0.996179 
 
 .003835 
 
 0.996051 
 
 .003965 
 
 30 
 
 0.99589 
 
 .00412 
 
 0.99577 
 
 .00425 
 
 31 
 
 0.99560 
 
 .00442 
 
 0.99547 
 
 .00455 
 
 32 
 
 0.99530 
 
 .00473 
 
 0.99517 
 
 .00486 
 
 33 
 
 0.99498 
 
 .00505 
 
 0.99485 
 
 .00518 
 
 34 
 
 0.99465 
 
 .00538 
 
 0.99452 
 
 .00551 
 
 35 
 
 0.99431 
 
 .00572 
 
 0.99418 
 
 .00586 
 
 36 
 
 0.99396 
 
 1.0060S 
 
 0.99383 
 
 .00621 
 
 37 
 
 0.99360 
 
 1.00645 
 
 0.99347 
 
 .00657 
 
 38 
 
 0.99323 
 
 1.00682 
 
 0.99310 
 
 .00694 
 
 39 
 
 0.99286 
 
 1.00719 
 
 0.99273 
 
 .00732 
 
 40 
 
 0.99248 
 
 1.00757 
 
 0.99235 
 
 .00770 
 
 41 
 
 0.99210 
 
 1.00796 
 
 0.99197 
 
 .00809 
 
 42 
 
 0.99171 
 
 1.00 36 
 
 0.99158 
 
 .00849 
 
528 
 
 TABLES. 
 DENSITY OF WATER (Continued). 
 
 Tempera- 
 ture. 
 
 Density (Density 
 atO = l). 
 
 Volume (Volume 
 at = 1). 
 
 Density (Density 
 at 4 = 1). 
 
 Volume (Volume 
 at4 = l). 
 
 43 
 
 44 
 
 0.99131 
 0.99091 
 
 1.00876 
 1.00917 
 
 0.99118 
 0.99078 
 
 1.00889 
 1.00929 
 
 45 
 
 0.99050 
 
 1.00953 
 
 0.99037 
 
 1.00971 
 
 46 
 
 0.99009 
 
 1.01001 
 
 0.98996 
 
 1.01014 
 
 47 
 
 0.98967 
 
 1.01044 
 
 0.98954 
 
 1.01057 
 
 48 
 
 0.98923 
 
 1.01088 
 
 0.98910 
 
 1.01101 
 
 49 
 
 0.98878 
 
 1.01134 
 
 0.98865 
 
 1.01148 
 
 50 
 
 0.98832 
 
 1.01182 
 
 0.98819 
 
 1.01195 
 
 51 
 
 0.98785 
 
 1.01230 
 
 0.98772 
 
 1.01243 
 
 52 
 
 0.98737 
 
 1.01279 
 
 0.98725 
 
 1.01292 
 
 53 
 
 0.98689 
 
 1.01328 
 
 .98677 
 
 1.01341 
 
 54 
 
 0.9S642 
 
 1.01377 
 
 98629 
 
 1.01390 
 
 55 
 
 0.98594 
 
 1.01426 
 
 98531 
 
 1.01439 
 
 56 
 
 0.98547 
 
 1.01475 
 
 0.98534 
 
 1.01488 
 
 57 
 
 0.93499 
 
 1.01524 
 
 0.98486 
 
 1.01537 
 
 58 
 
 0.98450 
 
 1.01574 
 
 0.08437 
 
 1.01587 
 
 59 
 
 0.9S401 
 
 1.01625 
 
 0.98388 
 
 1.01638 
 
 60 
 
 0.98350 
 
 1.01678 
 
 0.98338 
 
 1.01691 
 
 61 
 
 0.98299 
 
 1.01731 
 
 0.93286 
 
 1.01744 
 
 62 
 
 0.98247 
 
 1.01785 
 
 0.98234 
 
 1.01798 
 
 63 
 
 0.98194 
 
 1.01839 
 
 0.98182 
 
 1.01852 
 
 64 
 
 0.98140 
 
 1.01895 
 
 0.98128 
 
 1.01908 
 
 65 
 
 0.98036 
 
 1.01951 
 
 0.98074 
 
 1.01964 
 
 66 
 
 0.93032 
 
 1.02003 
 
 0.98019 
 
 1.02021 
 
 67 
 
 0.97977 
 
 1.02065 
 
 0.97954 
 
 1.02078 
 
 68 
 
 0.97921 
 
 1.02124 
 
 0.97903 
 
 1.02137 
 
 69 
 
 0.97864 
 
 1.02183 
 
 0.97851 
 
 1.02196 
 
 70 
 
 0.97807 
 
 1.02243 
 
 0.97794 
 
 1.02256 
 
 71 
 
 0.97749 
 
 1.02303 
 
 0.97736 
 
 1.02316 
 
 72 
 
 0.97690 
 
 1.02365 
 
 0.97677 
 
 1.02378 
 
 73 
 
 0.97631 
 
 1.02427 
 
 0.97618 
 
 1.02440 
 
 74 
 
 0.97571 
 
 1.02490 
 
 0.97553 
 
 1.02503 
 
 75 
 
 0.97511 
 
 1.02553 
 
 0.97498 
 
 1.02566 
 
 76 
 
 0.97450 
 
 1.02617 
 
 0.97438 
 
 1.02630 
 
 77 
 
 0.97389 
 
 1.02681 
 
 0.97377 
 
 1.02694 
 
 78 
 
 0.97328 
 
 1.02745 
 
 0.97316 
 
 1.02758 
 
 79 
 
 0.97267 
 
 1.02809 
 
 0.97255 
 
 1.02322 
 
 80 
 
 0.97206 
 
 1.02874 
 
 0.97194 
 
 1.02887 
 
 81 
 
 0.97145 
 
 1.02939 
 
 0.97132 
 
 1.02952 
 
 82 
 
 0.97033 
 
 1.03005 
 
 0.97070 
 
 1.03018 
 
 83 
 
 0.97020 
 
 1.03072 
 
 0.97007 
 
 1.03085 
 
 84 
 
 0.96956 
 
 1.03139 
 
 0.96943 
 
 1.03153 
 
 85 
 
 0.96892 
 
 1.03207 
 
 0.96879 
 
 1.03221 
 
 86 
 
 0.96323 
 
 1.03276 
 
 0.96315 
 
 1.03289 
 
 87 
 
 0.96764 
 
 1.03345 
 
 0.96751 
 
 1.03358 
 
 88 
 
 0.96699 
 
 1.03414 
 
 0.96687 
 
 1.03427 
 
 89 
 
 0.96634 
 
 1.03484 
 
 0.96622 
 
 1.03497 
 
 90 
 
 0.96563 
 
 1.03554 
 
 0.96556 
 
 1.03567 
 
 91 
 
 0.96502 
 
 1.03625 
 
 0.96490 
 
 1.03638 
 
 92 
 
 0.96435 
 
 1.03697 
 
 0.96423 
 
 1.03710 
 
 93 
 
 0.96363 
 
 1.03770 
 
 0.96356 
 
 1 03782 
 
TABLES. 
 DENSITY OF WATER (Continued). 
 
 529 
 
 Tempera- 
 ture. 
 
 Density (Density 
 atO = l). 
 
 Volume (Volume 
 at = 1). 
 
 Density (Density 
 at 4 = 1). 
 
 Volume (Volume 
 at 4 = 1) 
 
 94 
 
 0.96300 
 
 1.03844 
 
 0.9628& 
 
 1.03-56 
 
 95 
 
 0.96231 
 
 1.03918 
 
 0.96219 
 
 1.03931 
 
 96 
 
 0.96161 
 
 1.03993 
 
 0.96149 
 
 1.04006 
 
 97 
 
 0.96091 
 
 1.04069 
 
 0.96079 
 
 1.040*2 
 
 93 
 
 0.96020 
 
 1.04145 
 
 0.96008 
 
 1.04158 
 
 99 
 
 0.95949 
 
 1.04222 
 
 0.95937 
 
 1.04235 
 
 100 
 
 0.95879 
 
 1.04299 
 
 0.95866 
 
 1.04312 
 
INDEX. 
 
 Accelerators, use of, in Parr calorim- 
 eter, 401. 
 
 Accuracy, limit of, 4, 122. 
 Acetic acid, reagent, preparation of, 
 
 506. 
 
 Acetyl value of fats and oils, 447. 
 determination of, 447. 
 distillation process for, 448. 
 filtration process for, 448. 
 table of values of, 461. 
 Acid, 
 
 best for general use, 255. 
 boric, titration of, 293. 
 carbonic. See carbon dioxide, 
 chromic, titration of, 334. 
 free, in fats and oils, determina- 
 tion of, 438. 
 hydrochloric. See hydrochloric 
 
 acid. 
 
 oxalic. See oxalic acid, 
 phosphoric. See phosphoric acid, 
 standard, most desirable strength 
 
 for general use, 256. 
 sulphuric. See sulphuric acid, 
 sulphurous, oxidation of, 306. 
 Acid radicles, secondary, electrolytic 
 
 reactions of, 200. 
 Acids, 
 
 determination of, 94. 
 normal, 247. 
 standard, 258. 
 
 dilution of, to exact strength, 
 
 262, 265. 
 standardization of, by 
 
 anhydrous oxalic acid, 260. 
 crystallized oxalic acid, 259. 
 evaporation with ammonia, 
 
 261, 263. 
 gravimetric methods, 261 , 
 
 265, 272. 
 
 potassium dichromate, 260. 
 " tetroxalate,260. 
 sodium carbonate, 258, 264. 
 specific gravity, 261, 271. 
 titration of, by iodometric meth- 
 ods, 343. 
 
 Acids, weak, titration of, 255. 
 
 titration of base in salts of. 
 
 253. 
 Acidimetry, 247. 
 
 calculation of titrations in, 500. 
 Acidity of fats and oils, 437. 
 
 determination of, 438. 
 of nutrient gelatin, adjusting of, 
 
 Ex. 72, 425. 
 Air, 
 
 collection of samples of, 302, 305. 
 determination of carbon dioxide 
 in, 301, Ex. 55, 304. 
 
 calculation of results of, 305. 
 displacement in weighing, 
 correction for, 11, 234. 
 effect of, on weighing, 11. 
 drying of, by calcium chloride, 45. 
 exclusion of, from ferrous solu- 
 tions, 312. 
 
 leaks, testing for, 184. 
 Albuminoid ammonia, determination 
 
 of, 411, 416. 
 
 Alcohol, absolute, absorption of ben- 
 zine vapor by, 467. 
 
 methyl, distillation of boric 
 
 acid with, 116. 
 
 Alkali, caustic, action on glass by, 53, 60. 
 precipitation of copper, man- 
 ganese, nickel, and cobalt 
 by, 60. 
 
 cyanides, analysis of, by iodomet- 
 ric methods, 342. 
 Alkalies, 
 
 action on glass by, 60. 
 contamination of precipitates 
 
 with, 60. 
 
 determination of, in silicates, 190, 
 195. 
 
 in feldspar, 195. 
 in water, 431. 
 
 separation of, from each other, 187. 
 from magnesium, 186. 
 from zinc, 145. 
 standard, 279. 
 titration of, 265. 
 
 531 
 
532 
 
 INDEX. 
 
 Alkalimetric method o? titrating phos- 
 phoric acid, 360. 
 Alkalimetry, calculation of titrations 
 
 in, 500. 
 
 Alkaline earth, carbonates, purification 
 of, 30, 33. 
 
 metals, separation from zinc, 
 
 145. 
 
 Alkaline solutions, method of keeping, 
 289. 
 
 standard, not permanent, 279. 
 Alkalinity of water, 268. 
 
 determination of, 269. 
 Alloys, analysis of, 121. 
 brass or bronze, 146. 
 Britannia metal, 142. 
 determination of silver in, 355. 
 German silver, 148. 
 manganes e-phosphorus-bronze, 
 
 149. 
 
 Rose's metal, 129. 
 silver coin, 123. 
 soft solder, 128. 
 taking samples of, for analysis. 
 
 153. 
 
 type-metal, 140. 
 Aluminium, 
 
 determination of, in chromite, 164. 
 in cinnabar, 175. 
 in dolomite, 181. 
 in feldspar, 194. 
 in potash alum, Ex. 10, 54. 
 in salts of volatile and organic 
 
 acids, 71. 
 in silicates, 189. 
 in smaltite, 170. 
 in water, 432. 
 electrolytic separation from lead, 
 
 222. 
 
 hydroxide, precipitation of, 51. 
 effect of organic matter on, 54. 
 washing and ignition of, 52, 
 
 54. 
 oxide, determination of silica in, 
 
 53. 
 
 pure compounds of, 31. 
 separation, as basic acetate from 
 manganese, cobalt, nickel, and 
 zinc, 157. 
 
 as basic carbonate from man- 
 ganese, cobalt, nickel, and 
 zinc, 159. 
 as hydroxide, from cobalt, 
 
 nickel, and zinc, 156. 
 from chromium, 159, 164. 
 from iron, 160, 162. 
 from zinc, 145. 
 volumetric from iron, 162. 
 Ammeter, 205. 
 Ammonia, 
 
 albuminoid, determination of, in 
 water, 411, 416. 
 
 Ammonia, determination of, as am- 
 monium chloride, 105. 
 
 as platino-chloride, 106. 
 
 in ammonium chloride, Ex. 
 
 52, 282. 
 free, determination of, in water, 
 
 411, 413. 
 
 free water, preparation of, 412. 
 reagent, preparation of, 506. 
 solution, standard, 412. 
 specific gravity of, table of, 518. 
 standard solutions of, 282. 
 not permanent, 279. 
 titration of, 287. 
 Ammonium, 
 
 acetate, reagent, 506. 
 carbonr.te, precipitation of metals 
 by, 64. 
 
 reagent, 506. 
 
 solution of arsenic sulphide 
 
 in, 134. 
 
 chloride, determination of ammo- 
 nia as, 105. 
 
 pure, preparation of, 32. 
 reagent, 506. 
 
 hydroxide, precipitation of metals 
 by, 51. 
 
 reagent, 506. 
 
 oxalate, precipitation of calcium 
 by, 66, 67. 
 
 reagent, preparation of, 507. 
 testing for purity, 32. 
 platino-chloride, determination of 
 
 ammonia as, 103. 
 salts, removal of, 85. 
 sulphate, pure, preparation of, 33. 
 sulphocyanate, standard solution 
 
 of, 355. 
 
 Analysis, limit of accuracy in, 122. 
 taking of sample for, 152. 
 technical, 364. 
 
 Analytical equations, balancing of, 496. 
 Anions, definition of, 197. 
 Anode or cathode, rotation of, 209, 
 
 217, 221, 225. 
 Anodes, platinum, 205. 
 Anthracite coal, definition of, 391, 395. 
 Antilogarithms, table of, 516. 
 Antimony, determination of, as sul- 
 phide, 78. 
 
 by iodometric methods, 343. 
 in bronze or brass, 147. 
 in pyrites, 173 
 in smaltite, 169. 
 in type-metal, 141. 
 in Wood's metal, 131. 
 electrolytic separation, from ar- 
 senic and tin, 137. 
 
 from lead, 222. 
 oxide, analysis of, by iodometric 
 
 methods, 342. 
 pure compounds of, 31. 
 
INDEX. 
 
 533 
 
 Antimony, separation, from arsenic 
 and tin, 134. 
 
 by H. Rose's method, 136. 
 from copper, 214. 
 from lead, 127. 
 from tin by F. W. Clarke's 
 method, 134, 144. 
 
 by reduction with iron, 
 
 136, 141. 
 from zinc, 145. 
 Apparatus, 
 
 calibration of, 226. 
 volumetric, 227. 
 
 Arms of balance, determination of rela- 
 tive length of, Ex. 2, 15. 
 Arsenic, 
 
 compounds, solution of, 139. 
 determination, as magnesium py- 
 roarsenate, 88. 
 
 as sulphide, 78, 81. 
 in arsenious oxide, Ex. 21, 81. 
 in brass or bronze, 147. 
 in pyrites, 173. 
 in smaltite, 169. 
 in soft solder, 128. 
 in type-metal, 142. 
 in Wood's metal, 131. 
 electrolytic separation from anti- 
 mony, 137. 
 
 pure compounds of, 31. 
 separation, as arsenious chloride, 
 138. 
 
 from antimony by H. Rose's 
 
 method, 136. 
 
 from antimony and tin, 134. 
 from copper, 214. 
 from tin by F. W. Clarke's 
 
 method, 134, 144. 
 
 Arsenic acid, separation of, from phos- 
 phoric acid, 118. 
 
 Arsenious chloride, distillation of, 139. 
 oxide, analysis of, by iodometric 
 methods, 342. 
 
 determination of arsenic in, 
 
 Ex. 21, 81. 
 oxidation of, 306. 
 percentage of arsenic in, 81. 
 pure, testing of, 33. 
 standardization of iodine 
 
 solutions with, 339, 341. 
 Arsenites, oxidation of, 306. 
 Arsenopyrite, analysis of, Ex. 36, 172. 
 Ash, determination of, in coal, 391, 
 
 393. 
 
 Atomic weights, limit of accuracy in, 
 486. 
 
 international table of, 509. 
 limit of accuracy in, 486. 
 Available oxygen, 307. 
 
 determination, in manganese ores, 
 322, Ex. 58, 325. 
 
 in potassium die iromate, 307. 
 
 Available oxygen, determination, in 
 potassium permanganate, 307. 
 
 in pyrolusite, Ex. 58, 325. 
 
 B. coli, in water, test for, 424. 
 present in intestines, 407. 
 test for, in water, Ex. 73, 428. 
 Bacteria present in water, 423. 
 Bacterial count, in water, 424, Ex 72 
 425. 
 
 preparation of, nutrient agar- 
 agar for, Ex. 72, 127. 
 nutrient gelatin for, Ex. 
 , 72, 425. 
 Bacteriological examination of water, 
 
 423, 
 
 Balance, centre of gravity of, 9. 
 construction of, 7. 
 determination of relative length of 
 
 arms, Ex. 2, 15. 
 maximum load of, 15. 
 parts of, 7. 
 rules for using, 20, 
 sensibility of, 9. 
 
 determination of, Ex. 1, 15. 
 testing for equality of arms, 13. 
 Balancing of equations, 496. 
 Barium, carbonate, ignition of, 66, 71. 
 pure, preparation of, 33. 
 weighing of, 60, 71. 
 chloride, determination of barium 
 in, Ex. 18, 75. 
 
 of water in, Ex. 8, 44. 
 percentage of barium in, 76. 
 pure, testing of, 33. 
 reagent, preparation of, 507. 
 determination, as chromate, 89. 
 in barium chloride, Ex. 18, 75. 
 in salts of organic acids, 71. 
 in salts of volatile acids, 76. 
 hydroxide, standard solutions of, 
 281. 
 
 titration of, with oxalic acid, 
 
 303, 305. 
 
 use of, in determining carbon 
 dioxide in gases, 301, 303, 
 304. 
 precipitation, as carbonate, 64. 
 
 as sulphate, 73, 75. 
 pure compounds of, 31. 
 separation as chromate from cal- 
 cium and strontium, 178. 
 
 as sulphate from magnesium 
 
 and the alkalies, 177. 
 from calcium as nitrate, 180. 
 from calcium and strontium, 
 
 176. 
 
 from lead, 127. 
 from the alkali metals, 176. 
 sulphate, determination of sulphur 
 in, 103. 
 
 dissociation of, 21. 
 
534 
 
 INDEX. 
 
 Barium, sulphate, ignition of, 74. 
 
 properties of, 73. 
 
 thiosulphate, standardization of 
 iodine solutions with, 338, 340. 
 weighing as chromate, 179. 
 Bases present in fats and oils, 436. 
 Basicity of acids with indicators, 251. 
 Baume* scale, 276. 
 
 American Standard, 277. 
 Beam of balance, 7. 
 Benzene vapor, absorption by absolute 
 
 alcohol, 467. 
 
 Berthelot, bomb calorimeter of, 396. 
 Bismuth, 
 
 determination, as oxide, 130. 
 as oxychloride, 92, 127. 
 as sulphide, 78, 79. 
 hi Britannia metal, 143. 
 in Rose's metal, 130. 
 in salts of volatile and organic 
 
 acids, 71. 
 
 in Wood's metal, 131. 
 oxide, pure, preparation of, 33. 
 oxychloride, properties of, 92. 
 pure compounds of, 31. 
 separation, from copper, 214. 
 from lead as chloride, 126. 
 from zinc, 145. 
 
 Bituminous coal, definition of, 391, 395. 
 Bleaching powder, 
 
 active substance in, 346. 
 determination of available chlo- 
 rine in, Ex, 67, 347. 
 iodometric titration of, 347. 
 sampling of, 347. 
 titration with sodium arsenite, 347. 
 Bolting-cloth, use of , in sifting, 155. 
 Bomb calorimeter of Berthelot, 396. 
 Borates, solution of, 116. 
 Borax, anhydrous, fusion of carbonates 
 
 with, 111. 
 
 Boric acid, apparatus for determina- 
 tion of, 115. 
 
 decomposition of organic 
 
 matter containing, 295. 
 determination of, by the 
 
 method of Gooch, 115. 
 distillation of, with methyl 
 
 alcohol, 116. 
 titration of, 293. 
 
 by Jones' mannitol meth- 
 od, 294. 
 by Thompson's glycerine 
 
 method, 293. 
 weighing of, 116. 
 
 oxide, decomposition of silicates 
 by, 190. 
 
 purification of, 191. 
 volatilization of, 191. 
 Brass, analysis of, Ex. 31, 146. 
 Britannia metal, analysis of, Ex. 30, 
 142. 
 
 Britannia metal, decomposition of, by 
 
 means of chlorine, 142. 
 British Thermal Unit, definition of, 396. 
 Bromates, analysis of, by iodometric 
 
 methods, 343. 
 Bromine, 
 
 absorption of hydrogen sulphide 
 by, 369. 
 
 of " ilium inants" by, 466. 
 determination, as silver bromide, 
 95. 
 
 in metallic salts, 95. 
 distillation of, 99. 
 free, determination of, by iodo- 
 metric methods, 343. 
 separation from chlorine and io- 
 dine, 97. 
 
 from hydrocyanic acid, 100. 
 from iodine as palladous io- 
 dide, 96. 
 Bronze, analysis of, Ex. 31, 146. 
 
 manganese-phosphorus, analysis 
 
 of, Ex. 33, 149. 
 B. typhus in water, 424. 
 Bun sen, 
 
 conditions for explosion of hydro- 
 gen, 468. 
 
 distilling apparatus of, 344. 
 filter-pump, 24. 
 method of producing pure oxygen, 
 
 468. 
 
 valve, 311. 
 Burettes, 
 
 calibration of, Ex. 47, 244. 
 
 by Morse-Blalock bulbs, 238. 
 curve of corrections for, Fig. 41. 
 
 243. 
 
 errors in reading, 229. 
 Hempel gas, 472. 
 irregularities of bore, 241 
 reading of, 228. 
 total capacity of, 240. 
 use of floats with, 229. 
 Winkler modified gas, 472. 
 Butter fat, table of Polenske values of, 
 459. 
 
 Cadmium, 
 
 absorption of hydrogen sulphide 
 by alkaline solution of, 369, 372. 
 carbonate, properties of, 68. 
 
 ignition of, 69. 
 determination, as sulphide, 78. 
 
 in Wood's metal, 131. 
 electrolytic separation, from cop- 
 per, 214. 
 
 from lead, 222. 
 iodide, pure, testing of, 33. 
 oxide, ignition of, 69. 
 pure compounds of, 31 . 
 separation, from copper, 130, 131. 
 
 from zinc, 145. 
 
INDEX. 
 
 535 
 
 Cadmium, precipitation by sodium 
 
 carbonate, 68. 
 Calcium, carbonate, 64. 
 
 determination of calcium in, 
 
 Ex. 15, 67. 
 ignition of, 65. 
 percentage of calcium oxide 
 
 in, 68. 
 
 precipitation of carbon diox- 
 ide as, 299. 
 
 pure, preparation of, 33. 
 weighing of, 65. 
 
 chloride, fused, drying properties 
 of, 44, 45. 
 
 effect of temperature on dry- 
 ing properties of, 45. 
 determination, in calcium carbon- 
 ate, Ex. 15, 67. 
 in cinnabar, 175. 
 in dolomite, 181. 
 in salts of organic acids, 71. 
 in salts of volatile acids, 76. 
 in water, 432. 
 
 OZalate, conversion into carbonate 
 and oxide, 67, 68. 
 
 precipitation and digestion of, 
 
 66, 67. 
 
 titration of, 328. 
 oxide, weighing of boric acid with, 
 
 117. 
 phosphate, standardization with, 
 
 359. 
 
 precipitation, as carbonate, 64, 
 195. 
 
 by ammonium oxalate, 66, 67. 
 pure compounds of, 31. 
 separation, as nitrate from barium, 
 180. 
 
 as nitrate from strontium, 
 
 179. 
 
 as oxalate from the alkalies, 
 magnesium, barium, and 
 strontium, 176, 181. 
 as sulphate from the alkalies 
 
 and magnesium, 177. 
 from barium as chromate, 178. 
 from barium and strontium, 
 
 176. 
 
 from the alkali metals, 176. 
 volumetric determination of, 328. 
 weighing, as carbonate, 65, 68. 
 
 as oxide, 65, 68, 71. 
 Calculation, by logarithms, 487. 
 
 of acidimetry and alkalimetry, 500. 
 
 of calibration of weights, 17. 
 
 of determinations of iron as ferric 
 
 oxide, 59, 60. 
 
 of iodometric titrations, 504. 
 of oxidation and reduction titra- 
 tions, 502. 
 
 of results of analysis of illuminat- 
 ing-gas, 481. 
 
 Calculation, of titrations with acids, 
 248. 
 
 of volumetric determinations, 499. 
 of water analysis, 432. 
 Calibration, 233. 
 
 of burettes, Ex. 47, 244. 
 
 by means of Morse-Blalock 
 
 bulbs, 235. 
 
 by weighing water, 234. 
 of flasks, Ex. 46, 244. 
 
 by the Morse-Blalock bulbs, 
 
 238. 
 
 by weighing water, 233. 
 temperature of water used in, 
 
 237. 
 
 of Morse-Blalock bulbs, 236. 
 of volumetric apparatus, 226. 
 of weights, Ex. 3, 16. 
 
 calculation of results of, 17. 
 Calorie, definition of, 395. 
 Calorific value, of coal, determination 
 of, Ex. 71 , 403. 
 of fuels, determination of, 395. 
 
 effect of composition on, 395. 
 Calorimeter, Parr, 397. 
 Calorimeters, description of, 396. 
 Cannel-coal, definition of, 391, 395. 
 Capsule, Hofmeister, use of, 290. 
 Carbon, 
 
 combined, determination of, 383. 
 combustion of, in a stream of oxy- 
 gen, 377. 
 
 with chromic acid, 380. 
 condition in which present in iron 
 
 and steel, 3G4. 
 
 determination by the colorimetric 
 method of Eggertz, 383. 
 
 in chrome steel or iron high in 
 
 silica, 381. 
 
 in iron and steel, 374. 
 direct weighing of, 377. 
 filtration of, on a platinum boat, 
 377. 
 
 on a carbon filter, 378. 
 fixed, determination of, in coal, 
 
 393. 
 
 graphitic, determination of, 382. 
 oxidation of, by chromic acid, 374. 
 
 apparatus for, 374. 
 separation from iron by coppej 
 
 solution, 376. 
 
 total, determination of, 374. 
 Carbon dioxide, 
 
 absorption, by caustic potash solu- 
 tion, 464. 
 
 by soda-lime, 114. 
 in caustic potash, 114, 183. 
 determination of, 109. 
 
 by fusion of carbonates with 
 
 borax, 111, 182. 
 by fusion of carbonates with 
 microcosmic salt. 111. 
 
536 
 
 INDEX. 
 
 Carbon dioxide, determination of, by 
 loss, by laboratory apparatus, 110. 
 by the Schrotter appa- 
 ratus, 109. 
 by the apparatus of Fre- 
 
 senius. 112, 183. 
 in air, 301, Ex. 55, 304. 
 
 calculation of results of, 
 
 304. 
 
 in bottled waters, 300. 
 in dolomite, 182. 
 in gases, 300. 
 
 by Pettenkofer's method, 
 
 301. 
 
 in illuminating-gas, 478. 
 in water, 296. 
 
 by Pettenkofer's method, 
 298. 
 
 free, 296. 
 
 semi-combined, 297. 
 total, 298. 
 direct weighing of, 112, 183. 
 
 in dolomite, 183. 
 drying and absorption of, 114. 
 effect on indicators, 250. 
 precipitation of, 266, 268. 
 
 as calcium carbonate, 299. 
 purification and absorption of , 378. 
 titration of, 265, 267, 293. 
 Carbon di-sulphide, extraction of sul- 
 phur by, 79. 
 
 Carbon monoxide, absorbents for, 464. 
 Carbonates, determination of carbon 
 dioxide in, by fusion with borax or 
 microcosmic salt, 111. 
 Cathions, definition of, 197. 
 Cathode, rotation of. 209, 217, 221, 225. 
 Cathodes, platinum dishes as, 205. 
 
 roughening of, 205. 
 
 Caustic potash, bulbs, Geissler, use of, 
 114, 183. 
 
 standard alcoholic solution of. 281, 
 437. 
 
 See potassium hydroxide. 
 Caustic soda. See sodium hydroxide. 
 Cells, primary and storage, 203. 
 Centre of gravity of balance, 9. 
 Centrifugal machine, separation of lead 
 
 peroxide by, 389. 
 
 Chalcopyrite, analysis of, Ex. 26, 172. 
 Characteristic of logarithms, method of 
 
 finding, 363. 
 
 Chemical examination of water, 410. 
 Chlorates, analysis of, by iodometric 
 
 methods, 343. 
 
 Chloric acid, determination of, 100, 343. 
 Chlorides, 
 
 determination of, in water, 421 . 
 titration of, with silver nitrate, 
 
 350, 351. 
 Chlorine, 
 
 absorption of, by silver-foil, 379. 
 
 Chlorine, available, determination of, 
 in bleaching powder, 346, Ex. 67. 
 347. 
 
 decomposition of Britannia metal 
 
 by, 142. 
 
 determination, as silver chloride, 
 94, 265. 
 
 in metallic salts, 95. 
 free, determination of, by iodc- 
 
 metric methods, 343. 
 normal in water, 429. 
 oxidation of sulphur compounds, 
 
 with, 104. 
 
 separation from bromine and io- 
 dine, 97. 
 
 from hydrocyanic acid, 100. 
 from iodine as palladous io- 
 dide, 96. 
 from iodine as thallous iodide, 
 
 97. 
 
 Cholesterol and phytosterol, detection 
 of in fats and oils, 151. 
 
 melting-points of, 452. 
 Chromate of silver as indicator, 350. 
 Chromates, analysis of, by iodometric 
 
 methods, 343. 
 Chrome-iron ore, decomposition of, 
 
 with sodium peroxide, 333. 
 Chromic acid, oxidation of carbon by, 
 374, 380. 
 
 titration of, 334. 
 
 Chromite, analysis of, Ex. 34, 163. 
 Chromium, chromic, precipitation of, 
 as hydroxide, 51. 
 
 determination, as barium or lead 
 chromate, 89. 
 
 in iron ores, 333, Ex. 63, 333. 
 
 in chromite, 163. 
 
 in salts of volatile and organic 
 
 acids, 71 
 in smaltite, 170. 
 
 hydroxide, effect of organic matter 
 on precipitation of, 54. 
 
 washing and ignition of, 52. 
 oxidation with chlorine or bro- 
 mine, 160. 
 
 with potassium nitrate, 159. 
 with sodium peroxide, 160, 
 
 163. 
 oxide, determination of silica in, 
 
 53. 
 
 pure compounds of, 31. 
 separation, as hydroxide from 
 cobalt, nickel, and zinc, 156. 
 from aluminium, 160, 164. 
 from iron, 160, 163. 
 from iron, aluminium, man- 
 ganese, cobalt, nickel, and 
 zinc, 159. 
 from zinc, 145. 
 
 Chromous chloride, absorption of oxy- 
 gen by, 464. 
 
INDEX. 
 
 537 
 
 Cinnabar, analysis of, Ex. 37, 174. 
 Clarke's method of separation of ar- 
 senic and antimony from tin, 134. 
 Classen's method of determining, zinc, 
 224. 
 
 copper, 215. 
 
 Cleaning mixture for volumetric appa- 
 ratus, 240. 
 Coal, 
 
 anthracite, definition of, 391 . 
 bituminous, definition of, 391. 
 cannel-, definition of, 391. 
 determination of, coke, 391, 393. 
 ash, 391 , 393. 
 
 calorific value of, Ex. 71, 403. 
 fixed carbon, 393. 
 moisture, 390, 392. 
 sulphur, 392, 393. 
 
 by means of the Parr 
 
 calorimeter, 394. 
 volatile combustible matter, 
 
 391, 393. 
 forms in which sulphur is present 
 
 in, 392. 
 proximate analysis of, 390, Ex. 70, 
 
 392. 
 
 Cobalt, determination of, in salts of 
 volatile acids, 76. 
 
 electrolytic, determination of, 21 9. 
 by ammonia or Fresenius 
 and Bergmann's meth- 
 od, 220. 
 
 by ammonium oxalate or 
 Classen's method, 219. 
 separation, from copper, 214. 
 from lead, 222. 
 from manganese, 167. 
 oxide, ignited, composition of, 61. 
 precipitation with caustic alkali, 
 
 60. 
 
 pure compounds of, 31. 
 separation, as sulphide from man- 
 ganese, 166. 
 from iron, 160. 
 from iron and aluminium, 157. 
 from iron, aluminium, and 
 
 chromium, 156, 157. 
 from manganese and nickel, 
 
 166. 
 
 from nickel, as tri-potassium 
 cobaltic nitrite, 167, 171. 
 by means of nitroso-/?- 
 
 naphthol, 168. 
 from zinc, 145. 
 Cobaltous hydroxide, properties of, 61. 
 
 oxide, properties of, 61. 
 Cochineal, use as indicator, 251, 253. 
 Coke, definition of, 391. 
 
 determination of, 391, 393. 
 Color, 
 
 comparison of, in carbon determi- 
 nations. 384. 
 
 Color, determination of, in water, 
 
 409. 
 
 Combustible matter, volatile, determi- 
 nation of, in coal, 391, 393. 
 Composition of fats and oils, 436. 
 Conditions in which water is held, 39. 
 Congo red, as indicator, 251. 
 Constant weight, 41. 
 Constitution, water of, determination 
 
 of, 40. 
 
 Copper, compounds of, conversion into 
 sulphides, 78. 
 
 determination, as sulphide, 78, 80. 
 by iodometric methods, 343. 
 in brass or bronze, 147. 
 in Britannia metal, 143. 
 in cinnabar, 175. 
 in copper pyrites, 174. 
 in copper sulphate, Ex. 12, 62, 
 
 Ex. 20, 80. 
 in Rose's metal, 130. 
 in salts of volatile and organic 
 
 acids, 71. 
 
 in a silver coin, 124, 219. 
 in type-metal, 141. 
 in Wood's metal, 132. 
 electrolytic, deposition, from a 
 nitric or sulphuric acid solution, 
 213, Ex. 41, 217. 
 
 from an ammonium ox- 
 alate solution accord- 
 ing to Classen, 215. 
 determination of, 213. 
 
 in copper sulphate, Ex. 
 
 40, 215. 
 in a nickel coin, Ex. 44, 
 
 220. 
 
 in a silver coin, 219. 
 in ores, 214. 
 
 separation, from lead, 222. 
 from other metals, 214. 
 from silver, 217. 
 hydroxide, properties of, 61. 
 oxide, ignited, composition of, 61. 
 use of, in Kieldahl digestion, 
 
 283. 
 
 precipitation, as cuprous sulpho- 
 cyanate, 78. 
 
 with caustic alkali, 60, 62. 
 pure compounds of, 31, 33. 
 separation, from iron, 160. 
 
 from zinc, 145. 
 sulphate, 
 
 determination, of copper in, 
 40, Ex. 12, 62, Ex. 20, 80, 
 215. 
 
 of water in, Ex. 7, 41. 
 percentage of copper in, 62. 
 80. 
 
 water in, 42, 50. 
 pure, preparation of, 33, Ex. 
 4, 36. 
 
538 
 
 INDEX. 
 
 Correction factors for the Parr calorim- 
 eter, 402. 
 
 Counterpoise, use of, in weighing, 14. 
 Crucible, Gooch, filtration with, 25. 
 
 Rose, use of, 80. 
 Crystallization, determination of water 
 
 of, 40. 
 
 Cubic centimeter, definition of, 232. 
 Culture medium, for bacterial count, 
 preparation of, Ex. 72, 425. 
 
 sterilization of, Ex. 72, 426. 
 Cupric hydroxide, properties of, 61 . 
 Cuprous chloride, absorption of carbon 
 monoxide by. 464. 
 
 ammoniacal solution of, 465. 
 hydrochloric acid solution of, 465. 
 preparation of, 465. 
 Current, electrical, 
 
 method of obtaining, 202, 203. 
 reduction, of voltage of, 202. 
 
 of strength by incandescent 
 
 lamps, 202. 
 
 significance of density of, in elec- 
 trolytic methods, 200. 
 Cyanides, titration of, with silver ni- 
 trate, 351, 354. 
 
 Cyanogen, determination of, in potas- 
 sium cyanide, Ex. 68, 354. 
 Cylinders, platinum, as electrodes, 
 206. 
 
 Daniell gravity cell, 203. 
 Decantation, washing by, 26. 
 Density, current, significance of, in 
 electrolytic methods, 200. 
 of water, table of, 527. 
 Deshay's method of determining man- 
 ganese, 388. 
 Desiccator, use of, 19. 
 Digestion-flasks, Kjeldahl, 284. 
 Digestion of precipitates, 23. 
 Dihydric potassium phosphate, stand- 
 ardization by, 358. 
 
 Dimethyl amidobenzine, use of, as indi- 
 cator, 251. 
 Dishes, platinum, as cathodes, 205. 
 
 roughening of, 205. 
 
 Disodium phosphate, standardization 
 by, 358. 
 
 reagent, preparation of, 507. 
 Distillation, apparatus, of Bunsen, 344. 
 of Mohr, 345. 
 of Kjeldahl, 285. 
 of arsenious chloride, 139. 
 Dolomite, analysis of, Ex. 38, 180. 
 calculation of formula of, 495. 
 Double salts, preparation of, 29. 
 Drown's method of determining silicon 
 
 in iron, 366. 
 
 Drying-agents for gases, 44. 
 Dudley, C. B., on coal analysis, 390. 
 Duplicates, agreement of, 5. 
 
 Economizing time, 124. 
 Edison, Lalande cell, 204. 
 
 primary cell, 204. 
 
 Eggertz, colorimetric method for com- 
 bined carbon, 383. 
 Electric motors, use of, for rotation of 
 
 electrodes, 212. 
 
 Electrical charge on ions, amount of, 
 198. 
 
 potential of, 199. 
 
 Electrification, error in weighing by, 14. 
 Electrodes, for rotation, 209. 
 platinum cylinders as, 206. 
 platinum gauze cylinders as, 207. 
 rotation of, 207. 
 Electrolysis, 
 
 ionic theory of, 197. 
 of warm solutions, 208. 
 Electrolytic, 
 
 determination of metals, 213. 
 methods, 197. 
 
 significance of current density 
 
 in, 200. 
 
 secondary reactions, of acid radi- 
 cles, 200. 
 
 of metals, 201. 
 
 separation of metals, by current 
 density, 200. 
 
 by potential, 199, 205. 
 stands, 205, 207. 
 
 Elements, difficulty of complete sepa- 
 ration of, 121. 
 Equations, 
 
 analytical, 496. 
 balancing of, 496. 
 metathetical, 497. 
 oxidation and reduction, 497. 
 Errors, 
 
 determination of amount of, 323. 
 in reading, burettes, 229. 
 
 pipettes, 231. 
 in weighing, 14. 
 of graduation, 232, 233. 
 sources of, in water determina- 
 tions, 40. 
 unavoidable, 5. 
 
 Frythrosene, use of, as indicator, 270. 
 Eschka's method of determining sul- 
 phur in coal, 392, 393. 
 Ether value of fats and oils, 440. 
 
 Factors, from a single precipitate, 490. 
 
 method of calculating, 487. 
 
 table of, 510. 
 Factor weights, method of obtaining, 
 
 492. 
 
 Faraday's law, 198. 
 Fats and oils, 
 
 acetyl value of, 447, 461. 
 
 acidity of, 437. 
 
 analysis of, 436. 
 
 chemical composition of, 436. 
 
INDEX. 
 
 539 
 
 Fats and oils, detection of phytosterol 
 and cholesterol in, 451. 
 
 determination of fatty acids of, 
 
 455. 
 
 ether value of, 440. 
 free acid in, determination of, 438. 
 Hehner value of, 443, 459. 
 identification of, 456. 
 iodine value of, 443, 457. 
 Kottstorfer value of, 439, 459. 
 Maumene number of, 449, 460. 
 melting-point of fatty acids of, 451 . 
 physical tests on, 448. 
 Polenske value of, 441. 
 
 determination of, 442. 
 refraction, index of, 449. 
 Reichert and Reichert-Meissl 
 
 value, 440, 459. 
 specific gravity of, 448. 
 specific temperature reaction of, 
 
 449, 461. 
 Fatty acids, determination of, 455. 
 
 melting-point of, 451. 
 Feldspar, analysis of, Ex. 39, 192. 
 Ferric, chloride, analysis of, by iodo- 
 metric methods, 343. 
 
 sulphocyanate, as indicator, 350. 
 Ferrous, ammonium sulphate, prepa- 
 ration of, 29. 
 
 pure, testing of, 33. 
 standardization of dichro- 
 mate solutions by, 331, 332. 
 of permanganate solu- 
 tions by, 313, 316. 
 iron, determination of, in ores, Ex. 
 
 57, 320. 
 
 salts, determination of available 
 oxygen in pyrolusite by, 322, 
 326. 
 
 silicate, decomposition of, 319. 
 sulphate, purification of, 30. 
 titanate, decomposition of, 319. 
 Filter-paper, 
 ashless, 54. 
 ignition of, 52, 56. 
 
 on platinum wire, 58. 
 Filter-pump, Bunsen, use of, 24. 
 Filtration, 24. 
 
 by suction, 24. 
 with Gooch crucible, 25. 
 Flash point of oils, determination of, 
 
 452. 
 Flasks, standard, 227. 
 
 calibration of, Ex. 46, 244. 
 
 by Morse-Blalock bulbs, 238. 
 by weighing water, 233. 
 for delivering capacity, 234. 
 Floats for burettes, 229. 
 Flue gases, 
 
 interpretation of results of analy- 
 sis of, 484. 
 taking sample of, 484. 
 
 Ford's method for the determination of 
 manganese, 385. 
 
 Formulae of salts and isomorphous mix- 
 tures, calculation of, 493. 
 
 Forster's modification of the Kjeldahl 
 method, 288. 
 
 Fractional weights, 8. 
 
 Fresenius, apparatus for determining 
 carbon dioxide, according to, 112. 
 
 Fuels, determination of calorific value 
 of, 395. 
 
 Fuming nitric acid, oxidation of sul- 
 phur compounds with, 102. 
 
 Funnel, hot-water, 28. 
 
 Gas, analysis, apparatus for, 471. 
 
 double absorption pipette for, 
 475. 
 
 method of filling, 475. 
 explosion pipette for, 476. 
 general methods of, 462. 
 HempeFs apparatus for, 472. 
 Orsat apparatus for, 481. 
 manipulation of, 483. 
 pipette for explosion with hot 
 
 platinum wire, 477. 
 simple absorption pipette for, 
 
 474. 
 burette, simple, 472. 
 
 Winkler's modified, 472. 
 explosive, definition of, 467. 
 illuminating-, analysis of, Ex. 72, 
 478. 
 
 determination of, carbon di- 
 oxide in, 478. 
 
 carbon monoxide in, 479. 
 hydrogen in, 480. 
 "illuminants" in, 479. 
 methane in, 480. 
 oxygen in, 479. 
 
 non-explosive, definition of, 467. 
 pressure, regulation of, 471. 
 temperature, regulation of, 471. 
 Gases, 
 
 confining liquids for, 471 . 
 determination of carbon dioxide 
 
 in, 300. 
 
 drying-agents for, 44. 
 Geissler caustic-potash bulbs, use of, 
 
 114, 183. 
 Gelatinous precipitates, filtration of, by 
 
 means of paper pulp, 27, 54. 
 General operations, 21. 
 German silver, analysis of, Ex. 32, 148,. 
 Glass, expansion of, 232. 
 
 solution of, in ammonia, 53. 
 Glycerine, use of, in titrating boric 
 
 acid, 293. 
 
 Gold, determination of. in silver coin. 
 123. 
 
 electrolytic separation of, from 
 lead, 222. 
 
540 
 
 INDEX. 
 
 Gooch, crucible, use in filtering by suc- 
 tion, 25. 
 
 determination of boric acid by 
 
 method of, 115. 
 Graduation, errors of, 232. 
 Graphitic carbon, determination of, 382. 
 Gravimetric methods, advantage of, 2. 
 definition of, 1, 226. 
 standardization of acids by, 
 
 261, 263, 265, 272. 
 Greiss, method of, for determining 
 
 nitrites in water, 416. 
 Gunning's modification of the Kjeldahl 
 
 method, 287. 
 
 Gunning-Jodlbauer modification of the 
 Kjeldahl method, 288. 
 
 Halogens, determination, 94. 
 as silver salts, 94. 
 in metallic salts, 95. 
 separation of, 96. 
 
 from hydrocyanic acid, 100. 
 Hanus' monobromide solution, 444. 
 Hardness, 
 
 permanent, determination of, in 
 
 water, 269, Ex. 50, 269. 
 temporary, determination of, in 
 
 water, 268, Ex. 50, 269. 
 Heat of combustion, definition of, 396. 
 Hehner value of fats and oils, 443. 
 
 table of, 459. 
 
 Hempel, double absorption pipette for 
 gases, 475. 
 
 method of filling, 475. 
 explosion pipette, 476. 
 gas, apparatus, 472. 
 
 burettes, 472. 
 simple absorption pipette for 
 
 gases, 474. 
 
 Hildebrand,W. F., on coal analysis, 390. 
 Hofmeister capsule, use of, 290. 
 Hot-water funnel, 28. 
 Hydriodic acid, pure compounds of, 32. 
 Hydrobromic acid, pure compounds of, 
 
 32. 
 
 Hydrocarbons, formation of, in solu- 
 tion of iron and steel, 364. 
 
 heavy, absorbents for, 466. 
 Hydrochloric acid, 
 
 absorption of, by copper sulphate, 
 
 379. 
 fifth-normal, preparation of, Ex. 
 
 48,263. 
 
 gas, generation of, 37, 140. 
 normal, 247. 
 pure compounds of, 32. 
 reagent, preparation of, 505. 
 specific gravity of, table of, 519. 
 standard, advantages of, 255. 
 preparation of, 263, 437. 
 Hydrocyanic acid, determination of, 
 100. 
 
 Hydrocyanic acid, separation of, from 
 
 the halogens, 100. 
 
 Hydrogen, basis for computing atomic 
 weights, 486. 
 
 gaseous, combustion by palladium 
 sponge, 469. 
 
 determination of, 467. 
 
 in illuminating-gas, 480. 
 in presence of methane, 
 
 470, 480. 
 explosion of, with oxygen, 
 
 467, 480. 
 
 explosion pipette for, 476. 
 ignition of sulphides in, 80. 
 slow combustion of, 468, 480. 
 theory of combustion with 
 
 palladium sponge, 469. 
 peroxide, absorption of hydrogen 
 sulphide by, 369. 
 
 determination of, by iodo- 
 
 metric methods, 343. 
 pipette for combustion with hot 
 
 platinum wire, 477. 
 sulphide, formation of, in solution 
 of iron and steel, 364. 
 
 reduction of iron by, 320. 
 solutions for absorbing, 369. 
 Hydrometers, use of, 275. 
 Hygroscopic, substances, weighing of, 
 19. 
 
 water, determination of, 40. 
 Hypochlorites, analysis of, by iodo- 
 metric methods, 343. 
 
 " Illuminants," 
 
 absorbents for, 466. 
 determination of, in gas, 479. 
 Illuminating-gas, analysis of, Ex. 72, 
 
 478. 
 Incrustants, 269. 
 
 determination of, 270. 
 Index of refraction, determination of, 
 
 449. 
 Indicator, 
 
 iodine as, 308. 
 
 potassium, ferrocyanide as, 308. 
 
 permanganate as, 308. 
 Indicators, 250. 
 
 basicity of acids with, 251. 
 effect of carbon dioxide on, 250. . 
 for oxidation reactions, 307. 
 for precipitation methods, theory 
 
 of, 349. 
 
 properties of various, 251, 253. 
 Indirect determinations, by means of 
 silver-nitrate solution, 353. 
 
 calculation of, 491. 
 Industrial processes, controlled by 
 
 quantitative analysis, 6. 
 International atomic weights, table of, 
 
 509. 
 lod^te, potassium acid, purity of, 35. 
 
INDEX. 
 
 541 
 
 Iodine, 
 
 determination, as silver iodide, 95, 
 99. 
 
 in metallic salts, 95. 
 distillation of, 99. 
 monobromide, Hanus' solution of, 
 
 444. 
 monochloride, Wijs' solution of, 
 
 443. 
 
 oxidizing power of, 306, 307. 
 pure, preparation of, 34. 
 reducing solutions used with, 336. 
 resublimed, standardization of 
 thiosulphate solution with, 337, 
 341. 
 
 separauv n, from chlorine as thal- 
 lous iodide, 97. 
 
 from chlor^.c and bromine as 
 
 palladous iodide, 96. 
 from hydrocyanic acid, 100. 
 solution, preparation and stand- 
 ardization of, Ex. 64, 339. 
 titration of, with sodium thio- 
 sulphate solution, 340. 
 solutions, conditions of use, 335. 
 standard, preparation of, 33G. 
 standardization of, 336. 
 
 by arsenious oxide, 339, 
 
 341. 
 by barium thiosulphate, 
 
 338, 340. 
 by potassium dichromate, 
 
 337. 
 
 by potassium perman- 
 ganate, 337. 
 by resublimed iodine, 
 
 337, 341. 
 
 by the iodates of potas- 
 sium and sodium, 338, 
 341. 
 
 solvents of, 335. 
 value of fats and oils, 443. 
 determination of, 445. 
 table of, 457. 
 
 lodometric, determination of, 
 antimony, 343. 
 copper, 343. 
 sulphur, 369. 
 methods, Chap. XXV, 335. 
 
 analysis of, oxidizing sub- 
 stances by, 342. 
 
 reducing agents by, 342. 
 sulphides by, 342. 
 titration, of acids by, 294, 343. 
 of bleaching powder by, 
 
 336. 
 
 titrations, calculation of, 504. 
 lonization, 
 
 constant, 21. 
 
 law of, 21. 
 
 of barium sulphate, 21 . 
 
 of metallic salts, 197. 
 
 Ions, 
 
 amount of electrical charge on, 198. 
 definition of, 21, 197. 
 potential of electrical charge on, 
 199. 
 
 Iron, 
 
 analysis of, 364. 
 determination, as sulphide, 82. 
 in brass or bronze, 148. 
 in Britannia metal, 143. 
 in chromite, 163. 
 in cinnabar, 175. 
 in dolomite, 181. 
 in feldspar, 194. 
 in German silver, 148. 
 in ores, 318, Ex. 57, 320, 331, 
 
 Ex. 62, 333. 
 in pyrites, 174. 
 in salts of volatile acids, 71. 
 in silicates, 189. 
 in smaltite, 169. 
 in soft-iron wire, Ex. 11, 57. 
 in soft solder, 129. 
 in type-metal, 141. 
 in water, 432. 
 in Wood's metal, 132. 
 of manganese in, 
 
 by Deshay's method, 388. 
 by Ford's method, 385. 
 by Volhard's method, 
 
 387. 
 
 of phosphorus in, 372. 
 of silicon in, 365. 
 
 after solution in hydro- 
 chloric acid, 366. 
 by Brown's method, 366. 
 of sulphur in, 367. 
 
 by solution in nitric acid, 
 
 367. 
 
 of total carbon in, 374. 
 electrolytic separation of, from 
 
 copper, 214. 
 from lead, 222. 
 
 ferric, precipitation as hydroxide, 
 51 . 
 
 effect of organic matter on, 
 
 54. 
 washing and ignition of, 52, 
 
 58. 
 
 in zinc, titration of, 312. 
 method of obtaining sample of, for 
 
 analysis, 3^5. 
 oxide, determination of silica in, 
 
 53. 
 
 pig, method for determining sul- 
 phur in, 3C8. 
 pure compounds of, 31. 
 reduction of, with hydrogen sul- 
 phide, 194, 320. 
 
 with stannous chloride, 319. 
 with sulphurous acid, 320. 
 with zinc, 311, 316, 319. 
 
542 
 
 INDEX. 
 
 Iron, separation, as basic acetate from 
 manganese, cobalt, nickel, and zinc, 
 157. 
 
 as basic carbonate from man- 
 ganese, cobalt, nickel, and 
 , zinc, 159. 
 
 as hydroxide from cobalt, 
 
 nickel, and zinc, 156. 
 by extraction with ether from 
 chromium, aluminium, 
 manganese, cobalt, nickel, 
 and copper, 160, 169. 
 by J. W. Roth's method, 160, 
 
 169. 
 from aluminium as aluminate, 
 
 162. 
 
 from manganese, 385. 
 from zinc, 132, 145. 
 volumetric, from aluminium, 
 
 102. 
 
 solution in potassium cupric chlo- 
 ride solution, 376. 
 standardization of permanganate 
 solution by standard solution of, 
 310. 
 
 volatilization of chloride, 381. 
 volumetric determination of, 194, 
 318. 
 
 in manganese-phosphorUs- 
 
 bronze, 149. 
 
 wire, determination of iron in, Ex. 
 11, 57. 
 
 standardizations of dichro- 
 mate solutions by, 330, 332. 
 of permanganate solu- 
 tions by, 309, 315. 
 
 Jannasch and Aschoff, separation of 
 halogens according to, 97. 
 
 Jannasch and Heidenreich, decomposi- 
 tion of silicates according to, 191. 
 
 Jones' mannitol method of titrating 
 boric acid, 294. 
 
 Kjeldahl, 
 
 digestion, 284, 290. 
 digestion-flasks, 284. 
 distillation, 285, 291. 
 
 apparatus for, 286. 
 method of determining nitrogen, 
 283. 
 
 Forster's modification of, 288. 
 determination of nitro- 
 gen in potassium ni- 
 trate by, Ex. 54,291. 
 Gunning's modification of, 
 287. 
 
 determination of nitro- 
 gen in milk by, Ex. 53, 
 289. 
 
 Ounning-Jodlbauer modifica- 
 tion of, 288. 
 
 Kjeldahl, method of, substances decom- 
 posed with difficulty by, 289. 
 
 Wilfarth's modification of, 
 
 287. 
 
 Knife-edge, of balance, 7. 
 Kottstorfer value of fats and oils, 439. 
 determination of, 439. 
 table of, 459. 
 
 Lacmoid, use as indicator, 251. 
 Lamps, incandescent, reduction of cur- 
 rent strength by, 202. 
 Langmuir's method of determining 
 
 rosin in shellac, 448. 
 Lead, 
 
 absorption of hydrogen sulphide 
 by alkaline solution of, 369, 370 
 371. 
 acetate, reagent, preparation of, 
 
 507. 
 carbonr.te, G4. 
 
 ignition of, 04. 
 
 determination of, as chromate, 89. 
 as sulphate, 74. 
 as sulphide, 78. 
 in brass or bronze, 147. 
 in Britannia metal, 143. 
 in lead nitrate, Ex. 17, 72. 
 in ores, 223. 
 in Rose's metal, 129. 
 in salts of volatile and organic 
 
 acids, 71, 77. 
 in soft solder, 128. 
 in type-metal as chloride, 
 
 140. 
 
 in Wood's metal, 131. 
 electrolytic deposition of, 201, 
 213, 222. 
 
 conditions of, 222. 
 separation and determination 
 
 of, 213. 
 separation of, from silver, 
 
 217. 
 
 peroxide, dissolving off platinum, 
 223. 
 
 oxidation of manganese by, 
 
 388. 
 separation of, by centrifugal 
 
 machine, 389. 
 
 precipitation by ammonium car- 
 bonate, 64. 
 
 pure compounds of, 31. 
 separation, as chloride from bis- 
 muth, 126. 
 
 as sulphate, 126, 129. 
 
 from antimony and barium, 
 
 127. 
 
 of silver from, 123. 
 sulphate, determination of sulphur 
 in, 104. 
 
 ignition of, 75. 
 properties of, 74. 
 
INDEX. 
 
 543 
 
 Limit of accuracy, in analysis, 122. 
 
 in atomic weights, 486. 
 Lindo-Gladding method of determining 
 
 potassium, 89. 
 Linen square, use of, in filtration, 24. 
 
 use of, in sifting, 155. 
 Liquids, taking samples of, 152. 
 Liter, 
 
 definition of, 232. 
 Mohr, 232. 
 
 various standards, in use, 232. 
 Litmus, use of, as indicator, 251. 
 Logarithms, 
 
 division by, 488. 
 method of using, 487. 
 multiplication by, 488. 
 table of, 514. 
 
 Lunge, and Marchlewski, table of spe- 
 cific Tavity of hydrochloric acid, 519. 
 and Rey, table of specific gravity 
 
 of nitric acid, 520. 
 and Weirnik, table of specific grav- 
 ity of ammonia solutions, 518. 
 Isler and Naef, table of specific 
 
 gravity of sulphuric acid, 523. 
 Lunge's nitrometer, use of, in standard- 
 izing permanganate solutions, 314, 
 316. 
 
 Magnesia mixture, 117. 
 
 preparation of, 507. 
 Magnesium, ammonium arsenate, 
 
 determination of phosphoric 
 
 acid as, 117. 
 ignition of, So, 87. 
 precipitation and ignition of, 
 
 88. 
 
 ammonium phosphate, precipita- 
 tion of, 85, 87. 
 
 carbonate, determination of mag- 
 nesium in, Ex. 19, 77. 
 determination, as phosphate, 85, 
 87. 
 
 in chromite, 165. 
 in dolomite, 182. 
 in magnesium carbonate, Ex. 
 
 19, 77. 
 in magnesium sulphate, Ex. 
 
 23, 87. 
 
 in salts of organic acids, 71. 
 in salts of volatile acids, 71 , 76. 
 in water, 432. 
 
 interference of, in determination 
 of carbon dioxide in water, 298. 
 pure compounds of, 31. 
 pyroarsenate, determination of 
 
 arsenic as, 88, 135. 
 separation from the alkalies, 186. 
 by means of barium hydrox- 
 ide, 187. 
 
 by means of mercuric oxide, 
 186. 
 
 Magnesium, sulphate, determination of 
 magnesium in, Ex. 23, 87. 
 
 of water in, Ex. 9, 47. 
 percentage of magnesium in. 
 
 88. 
 
 pure, preparation of, 35. 
 Magnus, table of vapor tension of 
 
 water, 526. 
 
 Manganese, ammonium phosphate, pre- 
 cipitation of, 84. 
 
 ignition of, 86. 
 carbonate, properties of, 69. 
 
 ignition of, 69. 
 
 determination, as phosphate, 84. 
 as sulphide, 82. 
 in chromite, 164. 
 in cinnabar, 175. 
 in dolomite, 181. 
 in iron and steel, 385. 
 
 Deshay's method for, 
 
 388. 
 
 Ford's method for, 385. 
 Volhard's method for, 
 
 387. 
 in potassium permanganate, 
 
 Ex. 22, 83. 
 
 in pyrolusite, Ex. 59, 326. 
 in salts of organic acids, 71. 
 in salts of volatile acids, 71, 
 
 76. 
 
 in smaltite, 171. 
 volumetrically, 387, 388. 
 dioxide, analysis of, by iodometric 
 methods, 343. 
 
 decomposition of potassium 
 
 permanganate by, 308. 
 determination of available 
 oxygen in, Ex. 65, 322, 344, 
 Ex. 58, 325. 
 precipitation of, 385. 
 electrolytic, deposition of, 201. 
 separation, from cobalt and 
 nickel, 167, 171. 
 from copper, 214. 
 from lead, 222. 
 hydroxide, properties of, 61. 
 ores, analysis of, 322. 
 oxidation of, to permanganic acid, 
 
 388. 
 
 oxide, ignited, composition of, 61. 
 phosphorus-bronze, analysis of, 
 
 Ex. 33, 149. 
 
 precipitation, as dioxide, 385. 
 with caustic alkali, 60. 
 with sodium carbonate, 68. 
 protosesquioxide, composition of, 
 
 69. 
 
 pure compounds of, 31. 
 separation, from chromium, 325. 
 from copper, 214. 
 from iron, 160, 385. 
 from iron and aluminium, 159. 
 
544 
 
 INDEX. 
 
 Manganese, separation, from iron, 
 aluminium, and chromium, 157. 
 from lead, 222. 
 from nickel and cobalt, 166, 
 
 167, 171. 
 from zinc, 145. 
 
 sulphide, green, precipitation of, 
 82. 
 
 ignition of, 83. 
 the two sulphides of, 81. 
 volumetric determination in man- 
 ganese-phosphorus-bronze, 149. 
 weighing as oxide, 69. 
 Mannitol, use of, in titrating boric acid, 
 
 294. 
 
 Mantissa, definition of, 488. 
 method of finding, 488. 
 Marshand U-tube, 46. 
 Mass, measurement of, 9. 
 Maiimene* number, determination of, 
 
 449. 
 
 Maiimene" numbers, table of, 460. 
 Maximum load of balance, 15. 
 Medicus, solution of lead ores, accord- 
 ing to, 223. 
 
 Melting-point of fatty acids, determi- 
 nation of, 451 . 
 Mercuric chloride, pure, testing of, 34. 
 
 reagent, 508. 
 
 Mercurous chloride, properties of, 92. 
 Mercury, 
 
 as confining liquid for gases, 471 . 
 determination, as mercurous 
 chloride, 92. 
 
 as sulphide, 78, 174. 
 by distillation, 175. 
 in cinnabar, 174, 175. 
 electrolytic separation of, from 
 
 lead, 222. 
 
 pure compounds of, 31. 
 pure, testing of, 34. * 
 separation from copper, 214. 
 from silver, 123. 
 from zinc, 145. 
 
 use of, in Kjeldahl digestion, 283. 
 Metallic oxides, properties of, 50. 
 Metals, 
 
 determination, as oxide, 50. 
 
 as phosphate, chromate, and 
 
 chloride, 84. 
 as sulphate and as sulphide, 
 
 73. 
 
 as sulphide, 78. 
 by evaporation with sulphuric 
 
 acid, 76. 
 by ignition of salts of volatile 
 
 acids, 70. 
 
 electrolytic determination of, 213. 
 secondary electrolytic reactions of, 
 
 201. 
 
 specific heat of, 401. 
 taking samples of, for analysis, 153. 
 
 Metals, washing and drying of electro- 
 
 lytically deposited, 208. 
 Metathetical equations, balancing of, 
 
 497. 
 Methane, combustion of, 470. 
 
 determination, of hydrogen in 
 presence of, 470,480. 
 
 of, in illuminating-gas, 480. 
 Methods of weighing water, 39. 
 Methyl alcohol, distillation of boric acid 
 
 with, 116. 
 Methyl chloride, volatilization of boric 
 
 acid by, 191. 
 Methyl orr.nge, 
 
 as indicator, use of, 253. 
 effect of carbon dioxide on, 250. 
 theory of action as indicator, 349. 
 Microcosmic salt, determination of car- 
 bon dioxide in carbonates by means 
 of, 111. 
 
 standardization by, 358. 
 Milk, determination of nitrogen in, by 
 the Kjeldahl-Gunning method, Ex. 
 53, 289. 
 
 weighing sample of, 290. 
 Mineral oils, detection of, 454. 
 Minerals, analysis of, 152. 
 arsenopyrite, 172. 
 chalcopyrite, 172. 
 chromite, 163. 
 cinnabar, 174. 
 dolomite, 180. 
 feldspar, 192. 
 pyrite, 172. 
 smaltite, 169. 
 
 calculation of formulas of, 493. 
 taking samples of, for analysis, 
 
 153. 
 
 Mohr, distillation apparatus of, 345. 
 Mohr's salt, standardization of perman- 
 ganate solutions by, 316. 
 Moisture, determination of, in coal, 390, 
 
 392. 
 
 Molybdate solution, preparation of , 507. 
 Morse-Blalock bulbs, 
 
 calibration by means of, 235. 
 calibration of, 236, Ex. 45, 242. 
 burettes by, 238. 
 flasks by, 238. 
 description of, 235. 
 temperature of water used with, 
 
 237. 
 Mortars, agate, 155. 
 
 steel, 154. 
 
 Motors for rotation of electrodes, 210. 
 Mullers, steel, 153. 
 
 Naphthylamine hydrochloride, solution 
 
 of, for determining nitrites, 417. 
 Neatness, necessity of, 3. 
 Nessler, solution, preparation of, 411. 
 tubes, 412. 
 
INDEX. 
 
 545 
 
 Nesslerizing, 414. 
 Nickel, 
 
 ammonium sulphate, determina- 
 tion of nickel in, 62. 
 coin, electrolytic analysis of, Ex. 
 
 44, 220. 
 
 determination of, in German sil- 
 ver, 148. 
 
 in nickel ammonium sulphate, 
 
 Ex. 13, 62. 
 
 in salts of organic acids, 71. 
 in salts of volatile acids, 71, 
 
 76. 
 
 in smaltite, 171 
 
 electrolytic, determination of, 219. 
 by ammonia or Fresenius 
 and Bergmann's meth- 
 od, 220. 
 
 by ammonium oxalate or 
 Classen's method, 219. 
 in a nickel coin, 221. 
 separation from copper, 214. 
 from lead, 222. 
 from manganese, 167. 
 oxide, ignited, composition of, 61. 
 precipitation of, with caustic al- 
 kali, 60, 62. 
 
 pure compounds of, 31. 
 separation, as sulphide from man- 
 ganese, 166. 
 
 from cobalt, as tri-potassium 
 cobaltic nitrite, 167, 171. 
 by means of nitroso-/?- 
 
 naphthol, 168. 
 from iron, 160. 
 from iron and aluminium, 
 
 159. 
 from iron, aluminium, and 
 
 chromium, 156, 157. 
 from manganese and cobalt, 
 
 166. 
 
 from zinc, 145. 
 
 Nickelous, hydroxide, properties of, 61. 
 oxide, 61. 
 
 reduction to nickel, 63. 
 Nitrates, 
 
 'determination, by the Kjeldahl 
 method, 288. 
 in water, 418. 
 of nitrogen in, 283, 288, Ex. 
 
 54, 291. 
 
 of N 2 O 5 in, 107. 
 teduction of, by Kjeldahl method, 
 
 283. 
 Nitric acid, 
 
 determination of, by means of 
 nitron, 106. 
 
 in nitrates, 107. 
 
 fuming, oxidation of sulphur com- 
 pounds with, 102. 
 pure compounds of, 32. 
 reagent, preparation of, 505. 
 
 Nitric acid, removal of, in sulphur de- 
 terminations, 104. 
 
 specific gravity, table of, 520. 
 Nitrites, determination of, in water, 416. 
 Nitrogen, 
 
 determination, 105. 
 
 by the Kjeldahl method, 283. 
 in gases, 470. 
 in milk, Ex. 53, 289. 
 in potassium nitrate by the 
 Kjeldahl method, Ex. 54, 
 291. 
 
 in water, 411. 
 
 organic, conversion of, into ammo- 
 nium sulphate, 283. 
 Nitrometer, Lunge's, use of, in stand- 
 ardization of permanganate solu- 
 tions, 314, 316. 
 Nitron, determination of nitric acid by 
 
 means of, 106. 
 Nitrous acid, determination of, by iodo- 
 
 metric methods, 343. 
 Normal, acids, 247. 
 
 hydrochloric, 247. 
 percentage given by the num- 
 ber of cubic centimeters 
 used, 249. 
 sulphuric, 247. 
 chlorine in water, 429. 
 oxidizing solutions, 307. 
 reducing solutions, 307. 
 solutions, 247. 
 
 Note-book, method of keeping, 3. 
 Noyes, Wm. A., on coal analysis, 390. 
 Nutrient agar-agar for bacterial count, 
 
 preparation of, Ex. 72, 427. 
 Nutrient gelatin, adjusting acidity of, 
 Ex. 72, 425. 
 
 preparation of, 'Ex. 72, 425. 
 
 Odor, numerical standards for, 410. 
 
 determination of, in water, 410. 
 Ohm's law, 203. 
 Oil analysis, 454. 
 Oils, 
 
 determination of fatty acids of, 
 455. 
 
 of flash point of, 452. 
 of viscosity of, 453. 
 identification of, 456. 
 mineral, detection of, 454. 
 saponification of, 454. 
 See also fats and oils. 
 Oil testers, use of, 453. 
 Operations, general, 21. 
 Ore-crushers, 154. 
 Ores, determination of, lead in, 223. 
 
 zinc in, 225. 
 pulverizing of, 153. 
 taking samples of, for analysis,153. 
 Organic, acids, 
 
 ignition of salts of, 71. 
 
546 
 
 INDEX. 
 
 Organic, acids, titration of, 255. 
 
 of base in salts of, 253. 
 matter, containing boric acid, de- 
 composition of, 295. 
 
 in water, determination of, 
 
 419. 
 
 removal of, 54. 
 
 Organic sulphides, produced by solu- 
 tion of steel in acids, 370. 
 Orsat apparatus, for gas analysis, 481. 
 
 manipulation of, 483. 
 Oxalates, 
 
 electrolytic decomposition of, 201. 
 solubility of , 121. 
 
 Oxalic acid, anhydrous, preparation of, 
 260. 
 
 standardization of acids with, 
 
 260. 
 
 crystallized, standardization of 
 acids with, 259. 
 
 pure, preparation of, 259. 
 determination of available oxygen 
 
 in pyrolusite by, 322, 325. 
 pure compounds of, 32. 
 pure, testing and preparation of, 
 
 34, 259. 
 
 separation of arsenic and anti- 
 mony from tin by, 134. 
 solution, preparation of, 420. 
 standard solution of, 303. 
 
 use of, in determining carbon 
 
 dioxide in air, 303. 
 standardization of permanganate 
 
 solutions by, 313. 
 Oxidation, 
 
 and reduction equations, balanc- 
 ing of, 497. 
 and reduction, methods, 306. 
 
 titrations, calculation of, 502. 
 definition of, 306. 
 indicators for, 307. 
 Oxides, metallic, properties of, 50. 
 Oxidizing, 
 
 solutions, normal, 307. 
 substances, 306. 
 
 analysis of, by iodometric 
 
 methods, 342. 
 Oxygen, 
 
 available, determination of, in 
 manganese ores, 322, Ex. 58, 
 325. 
 
 in pyrolusite, Ex. 58, 325. 
 in potassium dichromate, 307. 
 in potassium permanganate, 
 
 307. 
 basis for computing atomic 
 
 weights, 496. 
 
 determination of, in illuminating- 
 gas, 479. 
 
 gaseous, absorbents for, 462. 
 pure, not absorbed by phospho- 
 rus, 463. 
 
 Oxygen, pure, method of obtaining 
 stream of, 378. 
 
 method of producing, 468. 
 required for water, determination 
 
 of, 419. 
 
 standardization of permanganate 
 by measurement of, 314, 316. 
 
 Palladium sponge, combustion of hy- 
 drogen by, 469. 
 
 Pslladous iodide, properties of. 9G. 
 Pans of balance, 7. 
 
 Paper pulp, use of, in filtration, 27, 54. 
 Parr calorimeter, 397. 
 
 chemical reactions involved in, 
 
 397. 
 construction of the instrument, 
 
 399. 
 
 correction factors for, 402. 
 determination, of calorific value of 
 coal by means of, Ex. 71, 403. 
 of sulphur in coal by means 
 
 of, 894. 
 
 sodium peroxide for, 398. 
 use of accelerators in. 401. 
 water equivalent of, 400. 
 Pemberton's alkalimetric method of tit- 
 rating phosphates, 360. 
 Penfield's method of determining 
 
 water, 43. 
 Percentage given by number of cubic 
 
 centimeters of acid used, 249, 502. 
 Persulphuric acid, analysis of, by iodo- 
 metric methods, 343. 
 Pettenkofer's method of determining 
 carbon dioxide in water, 298. 
 
 in gases, 301 . 
 
 Phenacetolin, use of, as indicator, 251. 
 Phenol, use of, in the Kjeldahl diges- 
 tion solution, 283. 
 Phenol-sulphuric acid, solution of, for 
 
 determination of nitrates, 418. 
 Phenolphthalein, effect of carbon diox- 
 ide on, 250. 
 
 use of, as indicator, 251, 254. 
 Phosphate precipitates, ignition of, 86. 
 Phosphates, 
 
 titration of, with permanganate 
 solution, 362 
 
 with uranium solution, 357. 
 Phosphine, formation of, in solution of 
 
 iron and steel, 364. 
 
 Phosphomolybdate, direct weighing of, 
 119. 
 
 precautions in precipitation of, 
 
 118. 
 
 precipitation of, 119. 
 separation of phosphoric acid as, 
 
 118. 
 titration of, 360. 
 
 by Pemberton's alkalimetric 
 method, 3CO. 
 
INDEX. 
 
 547 
 
 Phosphomolybdate, titration of, with 
 permanganate solution, 362. 
 
 washing of, 119. 
 Phosphoric acid, 
 
 determination of, 117, 357. 
 precipitation of, as magnesium 
 ammonium phosphate, 117, 120, 
 361. 
 
 pure compounds of, 32. 
 separation of silicic and arsenic 
 
 acids from, 118. 
 
 titration of, by Pemberton's meth- 
 od, 360. 
 
 by the uranium method, 357. 
 with permanganate solution, 
 
 363. 
 weighing as phosphomolybdate, 
 
 119. 
 Phosphorous acid, reduction of mercury 
 
 by, 92. 
 Phosphorus, 
 
 condition in which present in iron 
 
 and steel, 364. 
 
 determination, in iron and steel, 
 372. 
 
 in manganese - phosphorus - 
 
 bronze, 150. 
 
 solution of iron in nitric acid 
 and separation of silica for, 
 372. 
 yellow, absorption of oxygen by, 
 
 463. 
 
 moulding of, into sticks, 464. 
 oxidation of, with potassium per- 
 manganate, 373. 
 
 pentoxide, drying properties of, 44, 
 46. 
 
 method of filling U- tubes 
 
 with, 47. 
 use of, in Kjeldahl digestion, 
 
 283. 
 pure, oxygen not absorbed by, 
 
 463. 
 
 separation from titanium, 373. 
 Physical, examination of water, 408. 
 
 tests of fats and oils, 448. 
 Phytosterol and cholesterol, detection 
 of, in fats and oils, 451 . 
 
 melting-points of, 452. 
 Pipettes, 227. 
 
 double absorption for gases, 475. 
 
 method of filling, 475. 
 explosion, 476. 
 for combustion with hot platinum 
 
 wire, 477. 
 
 method of using, 231. 
 simple absorption, for gases, 474. 
 Platino-chloride, ammonium, proper- 
 ties of, 103. 
 
 potassium, properties of, 89. 
 Platinum, 
 
 anodes, 205. 
 
 Platinum, boat, for filtration of carbon 
 on, 377. 
 
 cone, use of, 24. 
 
 crucible, damaged by ignition of 
 
 phosphates, 87. 
 cylinders, as electrodes, 206. 
 dish, precipitating iron in, 57. 
 dishes, as cathodes, 205. 
 gauze cylinders as electrodes, 207. 
 lead peroxide on, dissolving of, 223. 
 tin deposits on, dissolving of, 221. 
 utensils, rules for using and clean- 
 ing, 38. 
 
 zinc deposits on, dissolving of, 224. 
 Polenske value, of fats and oils, 441. 
 determination of, 442. 
 of butter-fat, table of, 459. 
 Policeman, 55. 
 Post of balance, 7. 
 
 Potash alum, calculation of formula of, 
 494. 
 
 determination of aluminium in, 
 
 Ex. 10, 54. 
 
 percentage of A1 2 O 3 in, 57. 
 pure, preparation of, 34. 
 Potassium, 
 
 acid iodate, purity of, 35. 
 
 standardization of sodium- 
 thiosulphate solution with, 
 338, 341 . 
 bromide, pure, testing of purity of, 
 
 35. 
 
 chromate, fusion of nitrates with, 
 107. 
 
 use of, as indicator, 350, 353. 
 cyanide, determination of cyano- 
 gen in, with silver nitrate, 351, 
 Ex. 68, 354. 
 
 fusion of stannic phosphate 
 
 with, 150. 
 
 solution, electrolytic deposi- 
 tion of silver from, 217. 
 determination, as platino-chloride, 
 89. 
 
 by Lindo-Gladding method, 
 
 89. 
 
 in feldspar, 196. 
 in presence of sodium, 353,491 . 
 in salts of volatile acids, 76. 
 dichromate, indicator for, 308. 
 oxidizing power of, 307. 
 properties of, 330. 
 pure, preparation of, 35, 330. 
 reagent, preparation of, 508. 
 solution, standardization of, 
 by ferrous ammonium sul- 
 phate, 331, 332. 
 by iron wire, 330, 332. 
 standard solution of, 330. 
 conditions of use of, 330. 
 preparation of, Ex. 61, 
 332. 
 
548 
 
 INDEX. 
 
 Potassium, dichromate, standardiza- 
 tion, of acids with, 260. 
 
 of sodium thiosulphate 
 
 solution with, 337. 
 ferricyanide, as indicator, 308. 
 
 in uranium titration, 357. 
 hydroxide, absorption of carbon 
 dioxide by, 464. 
 
 reagent, preparation of, 506. 
 standard alcoholic solution of, 
 
 281, 437. 
 
 iodide, pure, testing of, 35. 
 magnesium sulphate, preparation 
 
 of, 29, Ex. 5, 36. 
 
 nitrate, determination of nitrogen 
 in, by the Kjeldahl method, Ex. 
 54, 291. 
 
 purity of, 35. 
 standard solution of, 418. 
 permanganate, decomposition of, 
 by manganese dioxide, 308. 
 determination of manganese 
 
 in, Ex. 22, 82. 
 impurities in, 309. 
 its own indicator, 308. 
 oxidizing power of, 306, 307. 
 pure, testing of, 35. 
 permanganate solution, 
 
 alkaline for water analysis, 
 
 412. 
 determining manganese by, 
 
 323. 
 preparation of, 309, Ex. 56, 
 
 315. 
 removal of manganese dioxide 
 
 from, 309, 315. 
 stability of, 309. 
 standardization of, 309, 315. 
 by direct measurement of 
 
 oxygen, 314, 316. 
 by ferrous ammonium 
 
 sulphate, 313, 316. 
 by oxalates, 313, 316. 
 by oxalic acid, 313. 
 by pure iron wire, 309, 
 
 315. 
 
 by standard iron solu- 
 tions, 310. 
 
 standardization of sodium 
 thiosulphate solution with, 
 337. 
 titration of phosphoric acid 
 
 with, 362. 
 
 platmo-chloride, properties of, 89. 
 pure compounds of, 31. 
 separation from sodium, 187, 196. 
 sulphate, pure, testing of, 35. 
 
 use of, in the Kjeldahl diges- 
 tion, 293. 
 
 tetroxalate, preparation and stand- 
 ardization of acids with, 260. 
 pure, preparation of, 35. 
 
 Potassium tetroxalate, standardization 
 of permanganate solutions by, 313, 
 316. 
 Potential, 
 
 electrolytic, separation of metals 
 
 by, 199. 
 
 of electrical charge on ions, 199. 
 Precipitates, contamination of, with 
 soluble salts, 22. 
 digestion of, 23. 
 
 finely divided, treatment of, 23. 
 ignition of, 56. 
 
 transferring to the filter-paper, 54. 
 washing of, 26. 
 
 by decantation, 26. 
 without filtration, 91. 
 weighable, properties of, 50. 
 Precipitation, 
 
 by change of solvent, 30. 
 by double decomposition, 30 
 complete, theory of, 21. 
 excess of reagent in, 22. 
 methods, theory of indicators for, 
 
 349. 
 
 of double salts, 29. 
 Preparation, 
 
 of potassium magnesium sulphate, 
 
 Ex. 5, 36. 
 
 of pure copper sulphate, Ex. 4, 36. 
 of pure sodium chloride, Ex. 6, 37. 
 Pressure, 
 
 regulation of, in gas analysis, 471. 
 Primary cells, 203. 
 Problems, 245, 246, 257, 277, 278, 334, 
 
 356, 485. 
 Proximate analysis of coal, 390, Ex. 70, 
 
 392. 
 
 Pure salts, preparation of, 26. 
 Pycnometers, determination of specific 
 gravity of oils by, 449. 
 
 use of, 274. 
 Pyrite, analysis of, Ex. 36, 172. 
 
 in coal, 392. 
 Pyrogallol, absorption of oxygen by, 
 
 462. 
 
 Pyrolusite, commercial uses of, 322. 
 determination, of available oxygen 
 in, 322, Ex. 58, 325, Ex. 65, 345. 
 of manganese in, by Volhard's 
 method, 323, Ex. 59, 326. 
 
 Quantitative analysis, 
 
 neatness, necessary for, 3. 
 
 object of, 1. 
 
 patience and honesty necessary 
 
 for, 3. 
 
 required by industries, 6. 
 skill and knowledge required for, 2. 
 utility and importance of, 6. 
 
 Reagents, preparation of, 505. 
 Records, care in keeping, 3. 
 
INDEX. 
 
 549 
 
 Recrystallization of salts, 27. 
 Reducing, agents, analysis of, by iodo- 
 metric methods, 342. 
 solutions, 307. 
 substances, 306. 
 
 Reduction and oxidation, equations, 
 balancing of, 497. 
 
 methods, 306. 
 
 titrations, calculation of, 502. 
 Reichert and Reichert-Meissl, 
 value of fats and oils, 440. 
 determination of, 440. 
 table of, 459. 
 Results, calculation of, 4. 
 Retainers in water determinations, 40. 
 Rider of balance, 8. 
 Rocks, 
 
 pulverizing of, 153. 
 
 taking samples of, for analysis, 
 
 153. 
 
 Rose crucible, use of, 80. 
 Rose's metal, analysis of, Ex. 27, 129. 
 determination, of bismuth in, 130. 
 of copper in, 130. 
 of lead in, 129. 
 Rosin in shellac, determination of, by 
 
 A. C. Langmuir's method, 446. 
 Rosolic acid, use of, as indicator, 251 . 
 Rossetti, table of density of water, 527. 
 Rotation of anode or cathode, 209, 217, 
 
 221 , 225. 
 
 Roth's ether separation of iron, 160. 
 Rules for using balance, 20. 
 
 Salicylic acid, use of, in the Kjeldahl di- 
 gestion, 283, 288. 
 Salt, common, purification of, 30, Ex. 
 
 6, 37. 
 Salts, 
 
 calculation of formulae of, 493. 
 
 drying of, 28. ^ 
 
 pure, preparation of, 26. 
 
 by recrystallization, 27. 
 reason for analysis of, 121. 
 weighing of, 41. 
 Sample, 
 
 drying of, 156. 
 
 for analysis, selection and prepara- 
 tion of, 152, 347. 
 of air, collection of, 301, 305. 
 of flue gases, taking of, 484. 
 average, taking of, 484. 
 of illuminating-gas, collection of, 
 
 478. 
 
 of iron and steel, method of ob- 
 taining, 365. 
 of water, taking of, 408. 
 Sand-bath, use of, 42. 
 Sanitary analysis of water, 406. 
 Saponification of oils, 454. 
 
 value of fats and oils, definition 
 of, 439. 
 
 Scale-forming ingredients of water, 431 . 
 Schrotter apparatus, determination of 
 carbon dioxide by, 109. 
 
 sources of error in use of, 110. 
 Selenium, determination of, 105. 
 Sensibility of balance, 9. 
 definition of, 10. 
 determination of, Ex. 1, 15. 
 Separation of elements, difficulty of 
 
 complete, 121. 
 Sewage, contamination of water by, 
 
 407. 
 Sewaschen, apparatus for determining 
 
 carbon dioxide in air, 301. 
 Shellac, determination of rosin by A. C. 
 
 Langmuir's method, 446. 
 Sieves, brass, 155. 
 Sifting of powdered samples, 155. 
 Silica, determination of, in chromite, 
 164. 
 
 in dolomite, 181. 
 in feldspar, 193. 
 in ferric oxide, 59. 
 in precipitates, 69. 
 in pyrite, 174. 
 in silicates, 193. 
 in smaltite, 169. 
 in water, 431 . 
 
 fusion of nitrates with, 107. 
 precipitates, fusion of, 366. 
 volatilization with hydrofluoric 
 
 acid, 59, 181, 193, 365. 
 Silicates, 
 
 decomposition of, 188. 
 
 by the method of J. Lawrence 
 
 Smith, 190. 
 
 determination, of alkalies in, 190. 
 of iron, aluminium, and titan- 
 ium in, 189. 
 of silica in, 189. 
 
 fusion of, with alkali carbonates, 
 188, 192. 
 
 with boric oxide, 190. 
 
 with calcium carbonate and 
 
 ammonium chloride, 190. 
 Silicic acid, dehydration of, 189. 
 
 separation of, from phosphoric 
 
 acid, 118. 
 
 Silicon, condition in which present in 
 iron and steel, 364. 
 
 determination of, in iron and steel, 
 365 
 
 after solution of the iron in 
 
 hydrochloric acid, 366. 
 Drown's method for, 366. 
 tetrachloride, formation of, in 
 
 solution of iron and steel, 364. 
 Silver, 
 
 bromide, properties of, 95. 
 chloride, as indicator, 351. 
 
 properties of, 90. 
 chromate, use of, as indicator, 350. 
 
550 
 
 INDEX. 
 
 Silver, coin, analysis of, Ex. 25, 123. 
 
 determination of silver in, Ex. 
 
 69, 355. 
 division of. volumetrically, 
 
 124. 
 electrolytic analysis of, Ex. 
 
 43, 218. 
 
 cyanide, determination of hydro- 
 cyanic acid as, 100. 
 determination, as chloride, 90. 
 as sulphide, 78. 
 by Volhard's method, 355. 
 in a coin, Ex. 69, 355, 
 in alloys, 355. 
 in a silver coin, Ex. 25, 123, 
 
 Ex. 43, 218. 
 in silver nitrate, Ex. 24, 93, 
 
 Ex. 42, 218. 
 
 electrolytic deposition of, 201, 
 217, 218. 
 
 determination of, in silver 
 
 nitrate, Ex. 42, 218. 
 iodide, properties of, 95. 
 metallic, purity of, 35. 
 nitrate, 
 
 determination, of potassium 
 and sodium by, 353. 
 
 of silver in, Ex. 24, 93, 
 
 Ex. 42, 218. 
 indirect determination by 
 
 means of, 353. 
 purity of, 35. 
 
 reagent, preparation of, 508. 
 
 solution, preparation of, 421. 
 
 . standard solution of, 352, 354. 
 
 titration of chlorides 
 
 with, 350, 351. 
 
 of cyanides with, 
 
 351, 354. 
 
 nitrite, preparation of, 417. 
 pure compounds of, 31. 
 separation, from copper, 123, 214. 
 from lead, 123. 
 from lead and copper, 217. 
 from mercury, 123. 
 from other metals, 123. 
 from zinc, 145. 
 volumetric determination of, 226, 
 
 349. 
 
 Smaltite, analysis of, Ex. 35, 169. 
 Smith and Riche, method of determin- 
 ing tin, 224. 
 
 Smith, J. Lawrence, method of, for de- 
 composing silicates, 190, 195. 
 Soda-lime, absorption of carbon diox- 
 ide by, 114. 
 Soda reagent, preparation and use of, 
 
 269, 270. 
 Sodium, 
 
 ammonium phosphate, pure, test- 
 ing of, 35. 
 arsenite, solution, 336. 
 
 Sodium, arsenite, standard solution of, 
 389. 
 
 titration, of bleaching powder 
 with, 347. 
 
 of permanganic acid by, 
 
 389. 
 
 bicarbonate and carbonate, analy- 
 sis of mixtures of, 265, 267. 
 bicarbonate, conversion to carbon- 
 ate, 258, 204. 
 
 carbonate, determination of, in 
 caustic soda, Ex. 49, 267. 
 
 and hydroxide, analysis of 
 mixtures of, 265, 267. 
 calculation of titration 
 of, with standard acid, 
 248. 
 
 ignition and weighing of, 71. 
 precipitation of metals by, 
 
 68. 
 
 pure, preparation of, 258. 
 purification of, 30, 259. 
 solution, determination of 
 carbon dioxide in water 
 by, 296. 
 
 for water analysis, 412. 
 standardization of acids with, 
 
 258, 264. 
 
 testing for impurities, 258. 
 chloride, purification of, 30. 
 
 preparation of pure, Ex. 6, 37. 
 determination, as carbonate, 71. 
 in feldspar, 196. 
 in presence of potassium, 353, 
 
 491. 
 in salts of organic acids, 71. 
 
 of volatile acids, 76. 
 hydroxide, and carbonate analysis 
 of mixtures of, 265, Ex. 49, 267. 
 calculation of titration of, 
 
 with standard acid, 248. 
 determination of, in caustic 
 
 soda, Ex. 49, 267. 
 fifth-normal solution, prepa- 
 ration of, Ex. 52, 282. 
 precipitation of metals by, 60. 
 reagent, preparation of, 506. 
 solution, determination of 
 carbon dioxide in water by, 
 296. 
 
 standard solutions of, 279. 
 removal of carbon diox- 
 ide from, 280. 
 
 iodate, standardization of sodium 
 thiosulphate solution with, 338. 
 nitrite, standard solution of, 417. 
 peroxide, for the Parr calorimeter, 
 398. 
 
 fusion of sulphur compounds 
 
 with, 102. 
 
 phosphate, pure, testing of, 35. 
 pure compounds of, 31. 
 
INDEX. 
 
 551 
 
 Sodium, separation of, from potassium, 
 187, 196. 
 
 sulphide, reagent, preparation of, 
 
 503. 
 sulphide solution, preparation of, 
 
 137. 
 
 sulphite, determination of sulphur 
 dioxide in, Ex. 66, 346. 
 
 extraction of sulphur by, 79. 
 thiosulphate solution, 336, 339. 
 preparation and standardiza- 
 tion of, Ex. 64, 339, 443. 
 standardization, by arsenious 
 oxide, 339, 341. 
 
 by barium thiosulphate, 
 
 333, 340. 
 by potassium dichro- 
 
 mate, 337. 
 
 by potassium perman- 
 ganate, 337. 
 by resublimed iodine, 
 
 337, 341. 
 
 by the iodates of sodium 
 and potassium, 338, 
 341. 
 
 Solder, soft, analysis of, Ex. 26, 128. 
 determination, of arsenic in, 129. 
 of iron in, 129. 
 of lead in, 128. 
 of tin in, 128. 
 of zinc in, 129. 
 Solids, total, in water, determination of, 
 
 422. 
 Solutions, 
 
 calculation of standardization of, 
 
 499. 
 
 standard, 226, 247. 
 
 Sources of error, in water determina- 
 tions, 40. 
 
 Specific gravity, definition of, 272. 
 determination of. 272. 
 
 by means of, hydrometers, 
 275. 
 
 pycnometers, 274. 
 
 the Westphal balance, 
 
 275. 
 
 of fats and oils, 448. 
 standardization of acids by, 
 
 261. 
 
 standard temperatures for, 273. 
 Specific heat, of metals, 401. 
 
 of water, 396. 
 
 Specific temperature, reaction of oils 
 and fats, 449. 
 
 table of, 461. 
 Standard, 
 
 acid, most desirable strength of, 
 256. 
 
 percentage given by the num- 
 ber of cubic centimeters 
 used, 249, 502. 
 acids, 258. 
 
 Standard alcoholic caustic-potash solu- 
 tions, 281, 437. 
 alkalies, 279. 
 alkaline solutions not permanent, 
 
 279. 
 
 ammonia solutions, 282. 
 barium-hydroxide solutions, 281. 
 for set of weights, 18. 
 hydrochloric acid, preparation of, 
 
 Ex. 48, 263. 
 liters in use, 232. 
 sodium-hydroxide solution, 279, 
 282. 
 
 preparation of, Ex. 52, 282. 
 removal of carbon dioxide 
 
 from, 280. 
 solutions, 226, 247. 
 
 calculation of standardization 
 
 of, 499. 
 steels for carbon determinations, 
 
 383. 
 
 temperature for specific gravity, 
 273. 
 
 for volumetric work, 231. 
 Stands, electrolytic, 205, 207. 
 Stannic oxide, impurities present in. 
 125. 
 
 purification of, 125, 128. 
 separation of tin as, 125. 
 Stannic phosphate, decomposition of, 
 
 150. 
 
 Stannous chloride, analysis of, by iodo- 
 metric methods, 342. 
 
 determination of tin in, 327, 
 
 Ex. 60, 328. 
 
 reagent, preparation of, 507. 
 reduction of iron by, 319. 
 Starch iodide paper, as indicator, 348. 
 
 preparation of, 348. 
 Starch solution, as indicator, 308, 339. 
 
 method of preparing, 339. 
 Steel, 
 
 analysis of, 364. 
 
 determination of manganese in, 
 385. 
 
 by Deshay's method, 
 
 388. 
 
 by Ford's method, 385. 
 by Volhard's method, 
 
 387. 
 
 of phosphorus in, 372. 
 of silicon in, 365. 
 of sulphur in, 367. 
 
 by evolution method. 
 
 369. 
 by solution in nitric acid, 
 
 367. 
 
 of total carbon in, 374. 
 method of obtaining sample of, for 
 
 analysis, 365. 
 
 standard, for carbon determina- 
 tions, 383. 
 
552 
 
 INDEX. 
 
 Steel, solution of, for carbon determi- 
 nations, 384. 
 
 in potassium cupric-chloride 
 
 solution, 376. 
 
 Sterilization of culture medium for bac- 
 terial count, Ex. 72, 426. 
 Stirrups of balance, 7. 
 Stoichiometry, 486. 
 Stoking, effect of, on flue gases, 484. 
 Storage, batteries, 204. 
 
 cells, 203. 
 Strontium, 
 
 carbonate, determination of stron- 
 tium in, Ex. 14, 66. 
 ignition of, 65, 71 . 
 percentage of strontium in, 
 
 66. 
 
 pure, preparation of, 33. 
 weighing of, 66. 
 determination, as sulphate, 74. 
 in salts of organic acid, 71. 
 in salts of volatile acids, 76. 
 in strontium carbonate, Ex. 
 
 14, 60. 
 
 precipitation, by ammonium car- 
 bonate, 64. 
 
 separation, as nitrate from cal- 
 cium, 179. 
 
 as sulphate from magnesium 
 
 and the alkalies, 177. 
 from barium and calcium, 176. 
 from barium as chromate, 178. 
 from the alkali metals, 176. 
 sulphate, properties of, 74. 
 pure compounds of, 31. 
 Suction, use of, in filtration, 24. 
 Sulphanilic acid, solution of, for deter- 
 mination of nitrites, 417. 
 Sulphates, determination of, in water, 
 432. 
 
 insoluble, decomposition of, 103. 
 solubility of, 121. 
 
 Sulphides, analysis of, by iodometric 
 methods, 342. 
 
 determination of sulphur in, 101, 
 
 104. 
 
 drying of, 79. 
 
 ignition of, in hydrogen, 80. 
 Sulphur, 
 
 absorption of, by cadmium solu- 
 tion, 372. 
 
 by lead solution, 371 . 
 available for sulphuric-acid manu- 
 facture, determination of, 172. 
 condition in which present, in iron 
 and steel, 364. 
 in coal, 392. 
 crude, determination of sulphur in, 
 
 104. 
 determination of, 101. 
 
 by fusion with alkali nitrates 
 and carbonates, 101, 173. 
 
 Sulphur, determination of, by fusion 
 with sodium peroxide, 102. 
 by oxidation, with aqua regia, 
 171, 172. 
 
 with chlorine, 104. 
 with fuming nitric acid, 
 
 102. 
 with liquid bromine, 103, 
 
 104. 
 with potassium chlorate, 
 
 103, 104. 
 in coal, 392, 393. 
 
 by means of the Parr 
 
 calorimeter, 394. 
 Eschka's method for, 
 
 392, 393. 
 in iron and steel, 367. 
 
 after solution in nitric 
 
 acid, 367. 
 in pig iron, 368. 
 in pyrite, 172. 
 in reagents, 368. 
 in smaltite, 171. 
 in steel by evolution method, 
 369. 
 
 apparatus for, 370. 
 solution of iron for, 371. 
 dioxide, absorption of, by lead per- 
 oxide, 379. 
 
 determination of, in sodium 
 
 sulphite, Ex. 66, 346. 
 evolution methods for, in steel, 
 
 369. 
 
 iodometric determination of, 369. 
 removal from sulphides, 79. 
 weighing as cadmium sulphide, 
 
 372. 
 
 Sulphuric acid, concentrated, drying 
 properties of, 44, 46. 
 
 method of filling U-tubes 
 
 with, 46. 
 
 fuming, absorption of illuminants 
 by, 466. 
 normal, 247. 
 pure compounds of, 32. 
 reagent, preparation of, 505. 
 specific gravity, table of, 523. 
 standard, fifth-normal, prepara- 
 tion of, Ex. 51, 271. 
 
 titration of barium hydroxide 
 
 with, 303, 305. 
 Sulphurous acid, oxidation of, 303. 
 
 reduction of iron by, 320. 
 Synthetic equations, balancing of, 496. 
 
 Tafole of Polenske values of butter-fat, 
 
 459. 
 
 Tagliabue's oil tester, 453. 
 Technical analysis, 364. 
 Tellurium, determination of, 105. 
 Temperature, 
 
 regulation of, in gas analysis, 471. 
 
INDEX. 
 
 553 
 
 Temperature, standard, for volumetric 
 
 work, 231. 
 Tetroxalate of potassium, 
 
 preparation of, 260. 
 standardization of acids with, 
 
 260. 
 
 Thallous iodide, properties of. 97. 
 Theoretical percentage, calculation of, 
 
 491. 
 
 Thermal units, definition of, 395. 
 Thompson's glycerine rae j hod 01 titrat- 
 ing boric acid, 29b. 
 
 Thomson, R. T., classification of indi- 
 cators, 250. 
 
 table of basicity of acids with indi- 
 cators, 252. 
 
 Time, economizing, 124. 
 Tin, 
 
 deposits on platinum, dissolving 
 
 off, 221. 
 
 determination of, in brass or 
 bronze, 146. 
 
 in Britannia metal, 144. 
 in salts of volatile acids, 71. 
 in smaltite, 169. 
 in soft solder, 128. 
 in type-metal, 142. 
 in Wood's metal, 131. 
 electrolytic, determination of, 221. 
 by ammonium-oxalate 
 method of Classen, 221 . 
 separation from antimony 
 
 137. 
 
 precipitation as sulphide, 78. 
 pure compounds of, 31. 
 separation, as stannic oxide, 
 125. 
 
 from antimony, by H. Rose's 
 method, 136. 
 
 by means of metallic 
 
 iron, 136, 141. 
 
 from arsenic and antimony, 
 134. 
 
 F. W. Clarke's method 
 
 for, 134, 144. 
 from copper, 215. 
 from zinc, 145. 
 
 volumetric determination of, 327. 
 weighing as oxide, 80. 
 Titanium, 
 
 determination in feldspar, 194. 
 
 in silicates, 189. 
 
 separation of, from silica and phos- 
 phorus, 373. 
 
 Turbidity, determination of, in water, 
 408. 
 
 numerical standards for, 409. 
 Turmeric, use as indicator, 251. 
 Twaddell scale, 276. 
 Type-metal, analysis of, Ex. 29, 140. 
 
 Units of volume, definition of, 232. 
 
 Uranium, 
 
 method of titrating phosphoric 
 acid, 357. 
 
 indicator for, 357. 
 solution, titration of phosphates 
 with, 359. 
 
 standardization of, with cal- 
 cium phosphate, 359. 
 standard solution, preparation of, 
 
 358. 
 U-tube, 
 
 making impervious connections 
 
 with, 47. 
 Marshand, 46. 
 
 method of filling, with calcium 
 chloride, 46. 
 
 with phosphorus pent- 
 oxide, 47. 
 with sulphuric acid, 46. 
 
 Vacuo, reducing weights to, 11, 234. 
 Valence, charge on ions proportional to, 
 
 198. 
 
 Valve, Bunsen, 311. 
 Vapor tension, of water, 526. 
 Viscosity of oils, determination of, 453. 
 Volatile combustible matter, determina- 
 tion of, in coal, 391, 393. 
 Volhard's method of determining man- 
 ganese, 323, 327, 387. 
 
 calculation of results of, 324. 
 determination of amount of 
 
 error in, 323. 
 interfering metals, 324. 
 method of adding perman- 
 ganate solution in, 324. 
 method of determining silver, 355. 
 Voltage of current, 
 reduction of, 203. 
 separation of metals by, 199. 
 Volt-meter, 205. 
 Volume, units of, 232. 
 Volumetric apparatus, 227 
 calibration of, 233. 
 cleaning mixture for, 240. 
 standard temperature for, 231. 
 determinations, calculation of . 499. 
 methods, 226 
 . advantage of, 2 
 definition of, 1, 226. 
 
 Wash-bottle, hot-water, 55 
 Washing precipitates, 26. 
 
 by decantation, 26. 
 
 without filtration, 91 
 Water, 
 
 albuminoid ammonia in, determi- 
 nation of, 411, 416. 
 
 alkalies in determination of, 432. 
 
 ammonia-free, preparation of, 412. 
 
 analysis of, 406. 
 
 calculation of results of, 433. 
 
554 
 
 INDEX. 
 
 Water, analysis of, for use in boilers, 
 
 430. 
 
 literature of, 435. 
 sanitary, interpretation of re- 
 sults of, 428. 
 bacteria present in, 423. 
 bacterial count in, 424, Ex. 72, 425. 
 bacteriological examination of, 
 
 423. 
 
 B. typhus in, 424. 
 calcium in, determination of, 432. 
 calibration, by weighing of, 233, 
 
 236. 
 
 chemical examination of, 410. 
 chlorides in, determination of, 421. 
 color, determination of, 409. 
 conditions in which it is held, 39. 
 confining liquid for gases, 471. 
 density of, table of, 527. 
 determination, 39. 
 
 in barium chloride, Ex. 8, 44. 
 in copper sulphate, Ex. 7, 41. 
 in magnesium sulphate, Ex. 
 
 9, 47. 
 of carbon dioxide in,, 296. 
 
 by Pettenkofer's method, 
 
 298. 
 in bottled carbonated 
 
 waters, 300. 
 
 of free mineral acid in, 297. 
 sources of error in, 40. 
 direct weighing of, 47. 
 expansion of, 232. 
 free ammonia in, determination of, 
 
 411, 413. 
 
 free mineral acids in, determina- 
 tion of, 432. 
 
 ground, characteristics of, 430. 
 hardness, determination of, 268, 
 
 431, Ex. 50, 269. 
 hygroscopic, determination of, 40. 
 iron and aluminium in, determina- 
 tion of, 432. 
 magnesium in, determination of, 
 
 432. 
 
 motor, use of, for rotation of elec- 
 trodes, 210. 
 
 nitrates in, determination of, 418. 
 nitrites in, determination of, 416. 
 nitrogen in, determination of, 411. 
 normal chlorine of, 429. 
 numerical standards for odor of, 
 
 410. 
 
 odor, determination of, 410. 
 of constitution, determination of, 
 
 40. 
 of crystallization, determination 
 
 of, 40. 
 organic matter in, determination 
 
 of, 419, 421. 
 
 oxygen, required, determination 
 of, 419. 
 
 Water, Penfield's method of determin- 
 ing, 43. 
 
 physical examination of, 408. 
 
 sanitary analysis of, 406. 
 
 scale-forming ingredients of, 431. 
 
 silica in, determination of, 431. 
 
 source of impurities in, 407. 
 
 specific heat of, 396. 
 
 sulphates in, determination c 1 , 43 9 
 
 surface, characteristics of, 429. 
 
 taking sample of, 408. 
 
 temperature' '6f, 408. 
 
 test for B. cbli in, 424, Ex. 73, 
 428. 
 
 total solids in, determination of, 
 422. 
 
 turbidity, determination of, 408. 
 
 vapor tension of, 526. 
 
 weighing substances containing, 
 
 19. 
 Weighable precipitates, properties of, 
 
 50. 
 Weighing, 
 
 a salt, 41. 
 
 bottle, 19. 
 
 errors in, 14. 
 
 hygroscopic substances, 19. 
 
 independent of force of 'gravity, 9. 
 
 limit of error in, 5. 
 
 method of substitution for, 12. 
 
 methods of, 7, 12. 
 
 .substances containing water, 19. 
 
 tube, 49. 
 Weights, 
 
 calibration of, Ex. 3, 16. 
 
 constant, 41. 
 
 factor, method of obtaining, 492. 
 
 fractional, 8. 
 
 reducing to vacuo, 11. 
 
 standard for, 18. 
 
 Westphal balance, use of, 271, 275. 
 Whipple, Geo. C., characteristics of pol- 
 luted surface-waters, 429. 
 Wijs' iodine-monochloride solution, 
 
 443. 
 Wilfarth's modification of the Kjeldahl 
 
 method, 287. 
 
 Winkler modified gas burette, 472. 
 Wood's metal, analysis of, Ex. 28, 
 131. 
 
 Zeiss butyro-refractometer, use of, 
 
 449. 
 
 Zero-point of balance, 10. 
 determination of, 15. 
 variations in, 11. 
 Zinc, 
 
 absorption of hydrogen sulphide 
 
 by alkaline solution of, 369. 
 ammonium phosphate, ignition of, 
 86. 
 
 precipitation of, 84. 
 
INDEX. 
 
 555 
 
 Zinc, ammonium sulphate, determina- 
 tion of zinc in, Ex. 16, 70. 
 
 percentage of zinc in, 70. 
 carbonate, properties and ignition 
 
 of, 69. 
 deposits on platinum, difficulty of 
 
 dissolving, 224. 
 
 determination, as phosphate, 84. 
 as sulphide, 81. 
 in brass or bronze, 148. 
 in German silver, 148. 
 in ores, 225. 
 in salts of volatile and organic 
 
 acids, 71. 
 in soft solder, 129. 
 in Wood's metal, 132. 
 in zinc ammonium sulphate, 
 
 Ex. 16, 70. 
 
 electrolytic, determination of, 
 224. 
 
 Zinc, electrolytic, separation, from 
 copper, 214. 
 
 from lead, 222. 
 
 oxide, neutralization of manganese 
 solution with, 388. 
 properties of, 70. 
 purification of, 388. 
 precipitation with sodium carbon- 
 ate, 68, 69. 
 
 pure compounds of, 31. 
 reduction of iron solutions by, 311, 
 
 319. 
 
 separation, as sulphide, 145. 
 as zincate, 145. 
 from iron, 132. 
 from iron and aluminium, 159. 
 from iron, aluminium, and 
 
 chromium, 156, 157. 
 sulphate, pure, preparation of, 35. 
 titration of iron in, 312. 
 
STANDARD TEXT BOOKS 
 
 PUBLISHED BY 
 
 D. VAN NOSTRAND COMPANY, 
 
 NEW YORK. 
 
 ABBOTT, A. V. The Electrical Transmission of Energy. A Manual 
 for the Design of Electrical Circuits. Fifth Edition, entirely rewritten 
 and enlarged. Fully illustrated. 8vo, cloth net, $5.00 
 
 ASHE, S. W., and KEILEY, J. D. Electric Railways, Theoretically 
 and Practically Treated. Vol. I, Rolling Stock. Second Edition, 
 Revised. 12mo r cloth. 290 pp., 172 illustrations net, $2.50 
 
 ASHE, S. W. Vol. II. Engineering Preliminaries and Direct Current 
 
 Sub-Stations. 12mo, cloth. Illustrated net, $2.50 
 
 -Electricity Experimentally and Practically Applied. Funda- 
 mentals, Questions, Problems. 12mo, cloth. Illustrated net, $2.00 
 
 ATKINSON, A. A., Prof. Ohio University. Electrical and Magnetic 
 
 Calculations, for the use of Electrical Engineers and Artisans, Teachers, 
 Students, and all others interested in the Theory and Application of Elec- 
 tricity and Magnetism. Third edition, revised. 12mo, cloth. Illus- 
 trated net, $1.50 
 
 - PHILIP. The Elements of Electric Lighting, including Electric 
 Generation, Measurement, Storage, and Distribution. Tenth edition. 
 Illustrated. 12mo, cloth $1.50 
 
 The Elements of Dynamic Electricity and Magnetism. 
 
 Fourth edition. 120 illustrations. 12mo, cloth $2.00- 
 
 Power Transmitted by Electricity, and its Application by the 
 
 Electric Motor, including Electric Railway Construction. Fourth edition, 
 fully revised, new matter added. Illustrated. 12mo, cloth $2.00 
 
 AUCHINCLOSS, W. S. Link and Valve Motions Simplified. Illus- 
 trated with 29 wood-cuts, 20 lithographic plates, together with a Travel 
 Scale, and numerous useful tables. Fifteenth edition, revised. 8vo, 
 cloth $1.50 
 
 BAKER, A. L., Prof. Quaternions as the Result of Algebraic 
 
 Operations. 12mo, cloth. Illustrated net, $1.25 
 
 BARRUS, G. H. Engine Tests: Embracing the Results of over 
 
 one hundred feed-water tests and other investigations of various kinds of 
 steam-engines, conducted by the author. With numerous figures, tables, 
 and diagrams. 8vo ; cloth. Illustrated $4.00 
 
2 STANDARD TEXT BOOKS. 
 
 BAB WISE, S. M. The Purification of Sewage. Being a brief accunt 
 of the Scientific Principles of Sewage Purification and their Practical Appli- 
 cation. 12mo, cloth. Illustrated. New edition net, $3.50 
 
 BATES, E. L., and CHARLESWORTH, F. Practical Mathematics 
 
 and Geometry. 12mo, cloth. Illustrated $1.50 
 
 BEAUMONT, ROBERT. Color in Woven Design. With 32 colored 
 plates and numerous original illustrations. Thick 12mo, cloth $7.50 
 
 BEDELL, F., and PIERCE, C. A. Direct and Alternating Current 
 
 Manual, with directions for testing, an a discussion of the theory of ele tri al 
 apparatus. Second edition, enlarged. 8vo, cloth, 360pp. .net, $2.00 
 
 BEEKMAN, J. M. Principles of Chemical Calculations. 
 
 In Press 
 
 BEGTRTJP, J., M.E. The Slide Valve and its Functions. With 
 Special Reference to Modern Practice in the United States. With numerous 
 diagrams and figures. 8vo, cloth $2.00 
 
 BERNTHSEN, A. A Text-Book of Organic Chemistry. Translated 
 by George McGowan, Ph.D. Fifth English Edition. Revised and 
 
 extended by the author and translator. Illustrated. 12mo, cloth. . .$2.50 
 
 BIGGS, C, H. W. First Principles of Electricity and Magnetism. 
 
 A book for beginners in practical work, containing a good deal of useful 
 information not usually to be found in similar books. With numerous 
 tables and 343 diagrams and figures. 12mo, cloth, illustrated $2.00 
 
 BINNS, C. F. The Potter's Craft: a Practical guide for the studio 
 
 and workshop. 12mo, cloth. Illustrated net, $2.00 
 
 BLAINE, R. G. The Calculus and its Applications. A practical 
 treatise for beginners, especially engineering students. 12mo, cloth. Illus- 
 trated net, $1.50 
 
 BLYTH, A. W. Foods: Their Composition and Analysis. A 
 
 Manual for the use of Analytical Chemists, with an Introductory Essay 
 on the History of Adulterations. With numerous tables and illustrations. 
 Fifth edition, thoroughly revised, enlarged and rewritten. 8vo, 
 cloth $7.50 
 
 BOOMER, G. R. Hydraulic Motors and Turbines. For the use of 
 
 Engineers, Manufacturers and Students. Third edition, revised and 
 enlarged. With 192 illustrations. 12mo, cloth $5.00 
 
 BOWIE, AUG. J., Jr., M.E. A Practical Treatise on Hydraulic 
 
 Mining in California. With Descriptions of the Use and Construction of 
 Ditches, Flumes, Wrought-iron Pipes and Dams; Flow of Water on Heavy 
 Grades, and its Applicability, under High Pressure, to Mining. Tenth 
 edition. Quarto, cloth. Illustrated $5.00 
 
 BOWSER, E. A., Prof. An Elementary Treatise on Analytic Geometry. 
 
 Embracing Plane Geometry, and an Introduction to Geometry of Three 
 Dimensions. Twenty-third edition, 12mo, cloth $1.75 
 
 An Elementary Treatise on the Differential and Integral Calculus. 
 
 With numerous examples. Twenty-third edition, enlarged by 640 
 additional examples. 12mo, cloth $2.25 
 
STANDARD TEXT BOOKS. 3 
 
 BOWSER,E. A., Prof. An Elementary Treatise on Analytic Mechanics. 
 
 With numerous examples. Twenty-first edition, 12mo, cloth. . .$3.00 
 
 An Elementary Treatise on Hydro-Mechanics. With numerous 
 examples. Sixth edition. 12mo, cloth $2.50 
 
 A Treatise on Roofs and Bridges. With numerous Exercises. Espe- 
 cially adapted for school use. Second edition. 12mo, cloth. Illustrated. 
 
 net, $3.00 
 
 BRAGG, E. M. Marine Engine Design; including the Design of 
 
 Turning and Reversing Engines. 12mo, cloth. Illustrated net, $2.00 
 
 BREW, WM. Three-phase Transmission: a practical Treatise 
 
 on the economic conditions governing the Transmission of Electrical 
 Energy by Underground and Overhead Conductors. 8vo, cloth. Illus- 
 trated net, $2.00 
 
 BROWN, SIR HANBURY, K.C.M.G. Irrigation: Its Principles and 
 
 Practice as a Branch of Engineering. 8vo, cloth, 301 pp. Illustrated. 
 
 net, $5.00 
 
 BRUCE, E. M., Prof. Pure Food Tests : the Detection of the Com- 
 mon Adulterants of Foods by simple Qualitative Tests. A ready manual 
 for Physicians, Health Officers, Food Inspectors, Chemistry Teachers, and 
 all especially interested in the Inspection of Food. 12mo, cloth, illustrated. 
 
 net, $1.25 
 
 BRUHNS, Dr. New Manual of Logarithms to Seven Places of 
 Decimals. Seventh Edition. 8vo, half morocco $2.50 
 
 BUSKETT, E. W. Fire Assaying. A Practical Treatise on the Fire 
 
 Assaying of Gold, Silver and Lead, including descriptions of the appliances 
 used. 12mo, cloth. Illustrated net, $1.25 
 
 CAIN, W., Prof. Brief Course in the Calculus. With figures and dia- 
 grams. 8vo, cloth, illustrated net, $1.75 
 
 CARPENTER, R. C., Prof., and DIEDERICHS, H., Prof. Internal 
 
 Combustion Engines. With figures and diagrams. Third edition, 
 revised. 8vo, cloth, illustrated net, $4.00 
 
 CATHCART, WM. L., Prof. Machine Design. Part I. Fastenings. 
 
 8vo, cloth. Illustrated net, $3.00 
 
 CATHCART, W. L., and CHAFFEE, J. I. Elements of Graphic 
 
 Statics, and of General Graphic Methods. 8vo, cloth. Illustrated . .net, $3.00 
 
 CHAMBERS' MATHEMATICAL TABLES, consisting of Logarithms 
 
 of Numbers 1 to 108,000. Trigonometrical, Nautical and other Tables. 
 New edition. 8vo, cloth $1.75 
 
 CHRISTIE, W T . WALLACE. Chimney Design and Theory. A book 
 for Engineers and Architects, with numerous half-tone illustrations and 
 plates of famous chimneys. Second edition, revised. 8vo, cloth. . .$3.00 
 
 - Boiler Waters. Scale, Corrosion, Foaming. 8vo, cloth. Illus- 
 trated net, $3.00 
 
 CORNWALL, H. B., Prof. Manual of Blow-pipe Analysis, Quali- 
 tative and Quantitative. With a Complete System of Determinative 
 Mineralogy. With many illustrations. 8vo, cloth $2.50 
 
4 STANDARD TEXT BOOKS. 
 
 CRAMP, W. Continuous-Current Machine Design.T 8vo, cloth. 
 
 Illustrated net, $2.50 
 
 CROCKER, F. B., Prof. Electric Lighting. A Practical Exposition of 
 the Art, for Use of Engineers, Students, and others interested in the In- 
 stallation or Operation of Electrical Plants. Eighth edition, thoroughly 
 revised and rewritten. Vol. I. The Generating Plant. 8vo, cloth. 
 Illustrated $3.00 
 
 - Vol. II. Distributing Systems and Lamps. Sixth edition. 8vo, 
 cloth. Illustrated $3.00 
 
 -and WHEELER, S. S. The Management of Electrical Ma- 
 chinery. A thoroughly revised and enlarged edition of "The Practical 
 Management of Dynamos and Motors." Eighth edition, twenty- 
 fourth thousand. 131 illustrations. 12mo, cloth. 223 pp. . .net, $1.00 
 
 CROCKER, f F7 B.,~ and ARENDT, M. Electric Motors; their 
 
 Action Control and Application. 8vo, cloth. Illustrated net, $2.00 
 
 DEL MAR, F W. A. Electric Power Conductors. 8vo, cloth. Illus- 
 y^trated V net, $2.00 
 
 DINGER, H. C., Lieut., U.S.N. Handbook for the Care and Opera- 
 tion of Naval Machinery. l(jmo, cloth, illustrated net, $2.00 
 
 DORR, B. F. The Surveyor's Guide and Pocket Table-Book. Seventh 
 edition, revised, and greatly extended. With a second appendix up 
 to date. 16mo, morocco, flaps $2.00 
 
 DRAPER, C. H. Heat, and the Principles of Thermo-Dynamics. 
 
 With many illustrations and numerical examples. 12mo, cloth $1.50 
 
 ELIOT, C. W., Prof., and STORER, F. H. A., Prof. Compendious 
 
 Manual of Qualitative Chemical Analysis. Revised with the co-operation 
 of the authors, by Prof. William R. Nichols. Illustrated. Twenty-sec- 
 ond edition, newly revised by Prof. W. B. Lindsay and F. H. Storer. 
 12mo, cloth $1.50 
 
 ENNIS, W. D., Prof. Applied Thermodynamics for Engineers. 
 Second edition. 8vo, cloth. Illustrated net, $4.50 
 
 EVERETT, J. D. Elementary Text-Book of Physics. Illustrated. 
 Seventh edition. 12mo, cloth $1.40 
 
 EWING, A. J., Prof. The Magnetic Induction in Iron and other 
 metals. Third edition, revised. 159 illustrations. 8vo, cloth $4.00 
 
 FANNING, J. T. A Practical Treatise on Hydraulic and Water- 
 supply Engineering. Relating to the Hydrology, Hydro-dynamics and 
 Practical Construction of Water-works in North America. 180 illustrations. 
 Se venteenth edition, revised, enlarged, and new tables and illustrations 
 added 650 pp. 8vo, cloth $5.00 
 
 FAY, I. W., Prof. The Chemistry of the Coal-tar Dyes. 8vo, cloth. 
 
 net, $4.00 
 
 FISH, J. C L. Lettering of Working Drawings. Thirteen plates, 
 with descriptive text. Oblong, 9X 12, boards $1.00 
 
STANDARD TEXT BOOKS, 5 
 
 FLEMING, J. A., Prof. The Alternate-current Transformer In 
 Theory and Practice. Vol. I. The Induction of Electric Currents. 
 611 pp. New .edition, illustrated. 8vo, cloth $5.00 
 
 Vol. II. The Utilization of Induced Currents. Illustrated. 8vo, 
 
 cloth $5.00 
 
 Electrical Laboratory Notes and Forms, Elementary and 
 
 Advanced. 4to, cloth, illustrated $5.00 
 
 A Handbook for the Electrical Laboratory and Testing Room. 
 Vol. I. Equipment, Resistance, Current, Electromotive Force and 
 Power Measurement. 538 pages. Illustrated. 8vo, cloth. . .net, $5.00 
 
 - Vol. II. Meter, Lamp, Cable, Dynamo, Motor and Transformer 
 Testing. 650 pages. Illustrated. 8vo, cloth net, $5.00 
 
 FOSTER, H. A. Electrical Engineers' Pocket-Book. With the 
 Collaboration of Eminent Specialists. A handbook of useful data for 
 Electricians and Electrical Engineers. With innumerable tables, diagrams, 
 and figures. Sixth edition, completely revised and enlarged. 
 
 Pocket size, flexible leather, elaborately illustrated with an extensive index 
 and patent thumb index tabs. 1636 pp $5.00 
 
 FOX, WM., and THOMAS, C. W., M.E. A Practical Course in Mechan- 
 ical Drawing. With plates. Third edition, revised. 12mo, cloth.$1.25 
 
 GANT, L. W. Elements of Electric Traction, for Motormen and 
 
 Others. Illustrated. 217 pp., 8vo, cloth net, $2.50 
 
 GEIKIE, J. Structural and Field Geology, for Students of Pure 
 
 and Applied Science. With figures, diagrams, and half-tone plates. 8vo/ 
 cloth net, $4.00 
 
 GIBBS, W. E. Textbook of Physics. Part I. Solids and Fluids. 
 
 (Carnegie Technical Schools Textbooks.) 8vo, boards. Illustrated. 
 
 net, $1.50 
 
 GILLMORE, Q. A., Gen. Practical Treatise on the Construction of 
 
 Roads, Streets, and Pavements. Tenth edition. With 70 illustrations. 
 12mo, cloth $2.00 
 
 GOODEVE, T. M. A Text-Book on the Steam-Engine. With a 
 Supplement on Gas-Engines. Twelfth edition, enlarged. 143 illustra- 
 tions. 12mo, cloth $2.00 
 
 GUNTHER, C. O., Prof. Integration by Trigonometric and Imag- 
 inary Substitution. With an Introduction by J. Burkitt Webb. Illus- 
 trated. 12mo, cloth net, $1.25 
 
 GUY, A. E. Experiments on the Flexure of Beams, resulting in the 
 
 Discovery of New Laws of Failure by Buckling. Reprinted from the 
 "American Machinist." With diagrams and folding plates. 8vo, cloth, 
 illustrated. 122 pp net, $1.25 
 
 HAEDER, HERMAN, C. E. A Handbook on the Steam-Engine. 
 
 With especial reference to small and medium sized engines. Third 
 English edition, re-edited by the author from the second German edi- 
 tion, and translated with considerable additions and alterations by H. H. 
 P. Powles. Nearly 1100 illustrations. 12mo, cloth $3.00 
 
6 STANDARD TEXT BOOKS. 
 
 HALE, W. J., Prof. (Univ. of Mich.) Calculations of General Chem- 
 istry, with Definitions, Explanations, and Problems. Second edition, 
 revised. 174 pp. 12mo, cloth net, $1.00 
 
 HALL, WM. S., Prof. Elements of the Differential and Integral 
 Calculus. Sixth edition, revised. 8vo, cloth. Illustrated net, $2.25 
 
 - Descriptive Geometry; With Numerous Problems and Practical 
 
 Applications. Comprising an 8vo volume of text and a 4to Atlas of illus- 
 trated problems. Second edition. Two vols., cloth net, $3.50 
 
 HALSEY, F. A. Slide-Valve Gears : an Explanation of the Action and 
 
 Construction of Plain and Cut-off Slide- Valves. Illustrated. Eleventh 
 edition, revised and enlarged. 12mo, cloth $1.50 
 
 HANCOCK, HERBERT. Text-Book of Mechanics and Hydrostatics. 
 
 With over 500 diagrams. 8vo, cloth $1.75 
 
 HATT, J. A. H. The Colorist; Designed to Correct the commonly 
 
 held theory that red, yellow, and blue are the primary colors, and to supply 
 the much needed easy method of determining color harmony. With a 
 chart of color nomenclature and other practical information for artists and 
 workers or designers in colors. Small quarto, cloth. Illustrated .. net, $1.50 
 
 HAWKESWORTH, J. Graphical Handbook for Reinforced Concrete 
 
 Design. A series of plates, showing graphically, by means of plotted 
 curves, the required design for slabs, beams, and columns under various 
 conditions of external loading, together with practical examples showing 
 the method of using each plate. 4to, cloth net, $2.50 
 
 HAY, A. Alternating Currents; Their Theory* Generation, and 
 
 Transformation. 8vo, cloth, illustrated net, $2.50 
 
 -Principles of Alternate-Current Working. 12mo, cloth, 
 illustrated $2.00 
 
 An Introductory Course of Continuous Current Engineering 
 
 With 183 figures and diagrams. Illustrated. 327 pp. 8vo, cloth. 
 
 net, $2.50 
 
 HECK, R. C. H., Prof. The Steam-Engine. Vol. I. The Thermo- 
 dynamics and the Mechanics of the Engine. 8vo, cloth, 391 pp. Illus- 
 trated net, $3.50 
 
 - Vol. II. Form, Construction, and Working of the Engine. The 
 
 Steam-Turbine. 8vo, cloth. Illustrated net, $5.00 
 
 -Notes on the Graphics of Machine Forces. 8vo, boards. 
 Illustrated net, $1.00 
 
 - Notes on Elementary Kinematics. 8vo, boards. Illustrated 
 
 net, $1.00 
 
 HERRMANN, GUSTAV. The Graphical Statics of Mechanism. A 
 
 Guide for the Use of Machinists, Architects, and Engineers; and also a 
 Text-Book for Technical Schools. Translated and annotated by A. P. 
 Smith, M.E. 7 folding plates. Sixth edition. 12mo, cloth $2.00 
 
 HIROI, I. Statically-Indeterminate Stresses in Frames Commonly 
 
 Used for Bridges. With figures, diagrams, and examples. 12mo, cloth, 
 illustrated net, $2.00 
 
STANDARD TEXT BOOKS. 7 
 
 HOPKINS, N. MONROE, Prof. Experimental Electrochemistry. 
 
 Theoretically and Experimentally Treated. 300 pp., 8vo. Illustrated. 
 
 net, $3.00 
 
 HOUGHTON, C. E. The Elements of Mechanics of Materials. A 
 
 text for students in engineering courses. Illustrated. 194 pp., 12mo, 
 cloth net, $2'00 
 
 HURST, H. E., and LATTEY, R. T. Textbook of Physics. 8vo; 
 cloth. Illustrated net, $3.00 
 
 HUTCHINSON, R. W., Jr. Long Distance Electric Power Trans- 
 mission: being a treatise on the Hydro-electric Generation of Energy; its 
 Transformation, Transmission, and Distribution. 12mo, cloth, illustrated, 
 345 pp net, $3.00 
 
 JAMIESON, ANDREW, C. E. A Text-Book on Steam and Steam- 
 
 Engines, including Turbines and Boilers. Specially arranged for the Use 
 of Science and Art, City and Guilds of London Institute, and other Engineer- 
 ing students. Fifteenth edition, revised. Illustrated. 12mo, cloth. 
 
 $3.00 
 
 - Elementary Manual on Steam, and the Steam-Engine. Spe- 
 cially arranged for the Use of First-Year Science and Art, City and Guilds 
 of London Institute, and other Elementary Engineering Students. Twelfth 
 edition. 12mo, cloth $1.50 
 
 JANNETTAZ, EDWARD. A Guide to the Determination of Rocks: 
 
 being an Introduction to Lithology. Translated from the French by G. 
 W. Plympton, Professor of Physical Science at Brooklyn Polytechnic 
 Institute. Second edition, revised. 12mo, cloth . . $1.50 
 
 JOHNSTON, J. F. W., Prof., and CAMERON, Sir CHARLES. Elements 
 
 of Agricultural Chemistry and Geology. Seventeenth edition, 12mo, 
 cloth $2.60 
 
 JONES, H. C. The Electrical Nature of Matter and Radioactivity. 
 Second edition, completely revised. 8vo, cloth net, $2.00 
 
 IvAPP, GISBERT, C. E. Electric Transmission of Energy, and its 
 
 Transformation, Subdivision, and Distribution. A practical handbook. 
 Fourth edition, revised. 12mo, cloth $3.50 
 
 KELLER, S. S., Prof. Mathematics for Engineering Students 
 
 (Carnegie Schools Textbook Series. 12mo, half leather, illustrated.) 
 
 Algebra and Trigonometry, with a chapter on Vectors. 282 pp.. .net, $1.75 
 
 Special Algebra Edition. 113 pp net, $1.00 
 
 - Plane and Solid Geometry, 212 pp net, $1.25 
 
 Analytical Geometry and Calculus, 359 pp net, $2.00 
 
 KEMP, JAMES FURMAN, A.B., E.M. A Handbook of Rocks; for 
 
 use without the microscope. With a glossary of the names of rocks and 
 other lithological terms. Fourth edition, revised. 8vo, cloth. Illus- 
 trated : $1.50 
 
 KERSHAW, J. B. C. Electrometallurgy. Illustrated. 303 pp. 8vo, 
 cloth net, $2.00 
 
8 STANDARD TEXT BOOKS. 
 
 KLEIN, J. F. Design of a High-Speed Steam-engine. With notes 
 IT'ldiagrams, formulas, and tables. Second edition, revised and en- 
 
 l larged. j^Svo, cloth. & Illustrated. 257 pp net, $5.00 
 
 -'Physical Significance of Entropy, or the Second Law.*? 8vo, 
 
 cloth. ^ Illustrated $1.50 
 
 KNIGHT, A. M., Lieut.-Com., U.S.N. Modern Seamanship. Illus- 
 trated with 136 full-page plates and diagrams. Third edition, revised. 
 
 8vo, cloth, illustrated net, $6.00 
 
 Half morrocco $7.50 
 
 KOESTEB, F. Steam-Electric Power Plants and their Construction. 
 
 A Practical Treatise on the Design of Central Light and Power Stations and 
 their Economical Construction and Operation. Second edition. 473 pp., 
 340 illustrations. 8vo, cloth net, $5.00 
 
 KBAUCH, C., Dr. Testing of Chemical Reagents for Purity. 
 
 '^Authorized translation of the Third Edition, by J. A. Williamson and L. W. 
 i v Dupre. With additions and emendations by the author. 8vo, cloth, 
 
 net, $3.00 
 
 LAMBORN, L. L. Cottonseed Products. A Manual of the Treatment 
 of Cottonseed for its Products and Their Utilization in the Arts. With 
 tables, figures, full-page plates, and a large folding map. 8vo, cloth, 
 illustrated net, $3.00 
 
 LANCHESTER, F. W. Aerodynamics: Constituting the First 
 
 Volume of a Complete Work on Aerial Flight. With Appendices on the 
 Velocity and Momemtum of Sound Waves, on the Theory of Soaring Flight, 
 etc. With numerous diagrams and half-tones. Illustrated. 442 pp , 8vo, 
 cloth net, $6.00 
 
 LASSAR-COHN, Dr. An Introduction to Modern Scientific Chem- 
 istry, in the form of popular lectures suited to University Extension Students 
 and general readers. Translated from the author's corrected proofs for 
 the second German edition, by M. M. Pattison Muir, M.A. 12mo, cloth. 
 Illustrated $2.00 
 
 LATTA, M. N. Handbook of American Gas-Engineering Practice. 
 
 ,With diagrams and tables. 8vo, cloth, illustrated, 460 pp.. net, $4.50 
 
 LEEDS, C. C. Mechanical Drawing for Trade Schools. High School 
 
 *\ edition. (Carnegie Technical Schools Textbooks.) Text and Plates. 
 
 'Second edition, revised. ^ 4to, oblong cloth net, $1.25 
 
 Mechanical Drawing for Trade Schools. Machinery Trades' 
 edition. 43 lessons in text and plates. Second edition, revised. 
 4to, oblong cloth, 150 pp net, $2.00 
 
 LIVERMORE, V. P., and WILLIAMS, J. How to Become a Com- 
 petent Motorman. Being a Practical Treatise on the Proper Method of 
 Operating a Street Railway Motor Car; also giving details how to over- 
 come certain defects. Revised edition, entirely rewritten and en- 
 larged. 16mo, cloth, illustrated $1.00 
 
 LODGE, OLIVER J. Elementary Mechanics, including Hydrostatics 
 and Pneumatics. Revised edition. 12mo, cloth $1.50 
 
STANDARD TEXT BOOKS. 9 
 
 LOEWENSTEIN, L. C., and CRISSEY, C. P. Centrifugal Pumps; 
 
 their Design and Construction. With 8 folding plates. 8vo, cloth. Illus- 
 trated net, $4.00 
 
 LUCRE, C. E. Gas Engine Design. With figures and diagrams 
 Second edition, revised. 8vo, cloth, illustrated net, $3.00 
 
 LU .< .E, G., Ph.D. Technical Chemists' Handbook. Tables and 
 methods of analysis for manufacturers of inorganic chemical products. 
 283 pp. 12mo, leather net, $3.50 
 
 LUQUER, LEA McILVAINE, Ph.D. Minerals in Rock Sections. 
 
 The Practical Method of Identifying Minerals in Rock Sections with the 
 Microscope. Especially arranged for Students in Technical and Scientific 
 Schools. Third edition, revised. 8vo, cloth. Illustrated net, $1.50 
 
 MASSIE, W. W., and UNDERBILL, C. R. Wireless Telegraphy 
 
 and Telephony Popularly Explained. With a special article by Nikola 
 Tesla. 76 pp. 28 illustrations. 12mo, cloth net, $1.00 
 
 MELICK, C. W., Prof. Dairy Laboratory Guide. 12mo, cloth, 
 illustrated net, $1.25 
 
 MERCK, E. Chemical Reagents: Their Purity and Tests. 250 
 
 pp net, $1 .50 
 
 MILLER, E. H. Quantitative Analysis for Mining Engineers. Second 
 edition, revised. 8vo, cloth net, $1.50 
 
 MINIFIE, WM. Mechanical Drawing. A Text-Book of Geometrical 
 
 Drawing for the use of Mechanics and Schools, in which the Definitions and 
 Rules of Geometry are familiarly explained ; the Practical Problems are ar- 
 ranged from the most simple to the more complex, and in their description 
 technicalities are avoided as much as possible. With illustrations for Draw- 
 ing Plans, Sections, and Elevations of Railways and Machinery; an Introduc- 
 tion to Isometrical Drawing, and an Essay on Linear Perspective and 
 Shadows. Illustrated with over 200 diagrams engraved on steel. Tenth 
 thousand. With an appendix on the Theory and Application of Colors. 
 8vo, cloth $4.00 
 
 MINIFIE, WM. Geometrical Drawing. Abridged from the Octavo 
 
 Edition, for the use of schools. Illustrated with 48 steel plates. Ninth 
 edition. 12mo, cloth $2.00 
 
 MOSES, ALFRED J., and PARSONS, C. L. Elements of Mineralogy, 
 
 Crystallography, and Blow-Pipe Analysis, from a Practical Standpoint. 
 336 illustrations. Fourth edition. 8vo, cloth $2.50 
 
 NASMITB, JOSEPB. The Student's Cotton Spinning. Thirteenth 
 thousand, revised and enlarged. 8vo, cloth. Illustrated $3.00 
 
 NUGENT, E. Treatise on Optics; or, Light and Sight theoretically 
 
 and practically treated, with the application to Fine Art and Industrial 
 Pursuits. With 103 illustrations. 12mo, cloth $1.50 
 
10 STANDARD TEXT BOOKS. 
 
 OLSEN, Prof. J. C. Text-Book of Quantitative Chemical Analysis 
 
 by Gravimetric, Electrolytic, Volumetric, and Gasometric Methods. With 
 seventy-two Laboratory Exercises giving the analysis of Pure Salts, Alloys, 
 Minerals, and Technical Products. Fourth edition, revised and en- 
 larged. 8vo, cloth. Illustrated. 513 pp net, $4.00 
 
 OLSSON, A. Motor Control as Used in Connection with Turret 
 
 Turning and Gun Elevating. (The Ward Leonard System.) Illustrated. 
 8vo, Pamphlet, 27 pp. (U. S. Navy Electrical Series, No. 1.) net, $.50 
 
 OUDIN, MAURICE A. Standard Polyphase Apparatus and Systems. 
 
 With many diagrams and figures. Sixth edition, thoroughly revised. 
 
 Fully illustrated. 8vo, cloth $3.00 
 
 PALAZ, A., Sc.D. A Treatise on Industrial Photometry, with special 
 application to Electric Lighting. Authorized translation from the French 
 by George W. Patterson, Jr. Second edition, revised. 8vo, cloth. 
 Illustrated... ..$4.00 
 
 PARSHALL, H. F., and HOBART, H. M. Armature Windings of 
 
 Electric Machines. With 140 full-page plates, 65 tables, and 165 pages 
 of descriptive letter-press. Second edition. 4to, cloth $7.50 
 
 - Electric Railway Engineering. With numerous tables, figures, 
 
 and folding plates. 4to, cloth, 463 pp., illustrated net, $10.00 
 
 PAULDING, CHAS, P. Practical Laws and Data on Condensation 
 
 of Steam in Covered and Bare Pipes. 12mo, cloth. Illustrated. 102 
 pages net, $2.00 
 
 The Transmission of Heat through Cold-Storage Insulation. 
 
 Formulas, Principles, and Data relating to Insulation of every kind. A 
 Manual for Refrigerating Engineers. 12mo, cloth. 41 pp. Illustrated. 
 
 net, $1.00 
 
 PERRINE, F. A. C., A.M., D.Sc. Conductors for Electrical Dis- 
 tribution; Their Manufacture and Materials, the Calculation of the Cir- 
 cuits, Pole Line Construction, Underground Working and other Uses. 
 With diagrams and engravings. Second edition, revised. 8vo, 
 cloth net, $3.50 
 
 PERRY, JOHN. Applied Mechanics. A Treatise for the Use of 
 
 Students who have time to work experimental, numerical, and graphical 
 exercises illustrating the subject. New edition, revised and enlarged. 
 
 650 pages. 8vo, cloth net, $2.50 
 
 PLATTNER. Manual of Qualitative and Quantitative Analysis with 
 
 the Blow-Pipe. From the last German edition, revised and enlarged. 
 by Prof. Th. Richter, of the Royal Saxon Mining Academy. Translated 
 by Prof. H. B. Cornwall, assisted by John H. Caswell. Illustrated with 
 78 woodcuts. Eighth edition, revised. 463 pages. 8 vo, cloth, .net, $4.00 
 
 POPE, F. L. Modern Practice of the Electric Telegraph. A Tech- 
 nical Handbook for Electricians, Managers, and Operators. Seventeenth 
 edition, rewritten and enlarged, and fully illustrated. 8vo, cloth .$1.50 
 
STANDARD TEXT BOOKS. 11 
 
 PRELINI, CHARLES. Tunneling. A Practical Treatise containing 
 149 Working Drawings and Figures. With additions by Charles S. Hill, 
 C.E., Associate Editor "Engineering News." Fourth edition, thor- 
 oughly revised. 8vo, .cloth. Illustrated $3.00 
 
 - Earth'aild Rock Excavation. A Manual for Engineers, Contractors, 
 and Engineering Students. Second edition, revised. 8vo, cloth. 
 Illustrated. 350 pp net, $3.00 
 
 Graphical Determination of Earth Slopes. Retaining Walls, 
 
 and Dams. 8vo, cloth, illustrated, 136 pp net, $2.00 
 
 PRESCOTT, A. B., Prof. Organic Analysis. A Manual of the 
 Descriptive and Analytical Chemistry of Certain Carbon Compounds in 
 Common Use; a Guide in the Qualitative and Quantitative Analysis of 
 Organic Materials in Commercial and Pharmaceutical Assays, in the Esti- 
 mation of Impurities under Authorized Standards, and in Forensic Exami- 
 nations for Poisons, with Directions for Elementary Organic Analysis. 
 Sixth edition.,8vo, cloth. $5.00 
 
 and SULLIVAN, E. C. First Book in Qualitative Chemistry. 
 
 Eleventh edition. 12mo, cloth net, $1.50 
 
 and OTIS COE JOHNSON. Qualitative Chemical Analysis. A 
 
 Guide iri the Practical Study of Chemistry and in the Work of Analysis. 
 Sixth revised and enlarged edition, entirely rewritten, with an 
 Appendix by H. H. Willard containing a few improved methods of Analysis. 
 8vo, cloth net, $3.50 
 
 RANKINE, W. J. MACQUORN, C.E., LL.D., F.R.S. Machinery and 
 
 Mill-work. Comprising the Geometry, Motions, Work, Strength, Con- 
 struction, and Objects of Machines, etc. Illustrated with nearly 300, 
 woodcuts. Seventh edition. Thoroughly revised by W. J. Millar. 8vo, 
 cloth . . $5.00 
 
 - The Steam-Engine and Other Prime Movers. With diagram of 
 the Mechanical Properties of Steam. With folding plates, numerous 
 tables and illustrations. Fifteenth edition. Thoroughly revised by W. J. 
 Millar. 8vo, cloth $5.00 
 
 -Useful Rules and Tables for Engineers and Others. With 
 appendix, tables, tests, and formulae for the use of Electrical Engineers. 
 Comprising Submarine Electrical Engineering, Electric Lighting, and 
 Transmission of Power. By Andrew Jamieson, C.E., F.R.S.E. Seventh 
 edition, thoroughly revised by W. J. Millar. 8vo, cloth $4.00 
 
 A Mechanical Text-Book. By Prof. Macquorn Rankine and E. E. 
 Bamber, C.E. With numerous illustrations. Fifth edition. 8vo, 
 cloth $3.50 
 
 RANKINE, W. J. MACQUORN, C.E., LL.D., F.R,S. Applied Me- 
 chanics. Comprising the Principles of Statics and Cinematics, and Theory 
 of Structures, Mechanics, and Machines. With numerous diagrams. 
 Eighteenth edition. Thoroughly revised by W. J. Millar. 8vo, cloth $5.00 
 
 Civil Engineering."^ Comprising^ Engineering, Surveys, Earthwork, 
 Foundations, Masonry, Carpentry, Metal-Work, Roads, Railways, Canals, 
 Rivers, Water-Works, Harbors, etc. With numerous tables and illus- 
 trations. Twenty-third edition. Thoroughly revised by W. J. Millar. 
 8vo, cloth ; $6. 50 
 
12 STANDARD TEXT BOOKS. 
 
 RATEAU, A. Experimental Researches on the Flow of Steam 
 
 Through Nozzles and Orifices, to which is added a note on The Flow of 
 Hot Water. Authorized translation by H. Boyd Brydon. 12mo, cloth. 
 Illustrated ' net, $1.50 
 
 RAUTENSTRAUCH, W., Prof., and WILLIAMS, J. T. Machine 
 
 Drafting and Empirical Design. A Textbook for Students in Engineering 
 Schools and Others Who are Beginning the Study of Drawing as Applied 
 to Machine Design. Part I. Machine Drafting. Illustrated, 70 pp., 
 
 8vo, cloth net, $1.25 
 
 Complete in Two Parts. Part II in preparation. 
 
 RAYMOND, E. B. Alternating Current Engineering Practically 
 Treated. Third edition, revised and enlarged, with an additional 
 chapter on "The Rotary Converter." 12mo, cloth. Illustrated. 232 pages. 
 
 net, $2.50 
 
 REINHARDT, CHAS. W. Lettering for Draughtsmen, Engineers and 
 
 Students. A Practical System of Free-hand Lettering for Working Draw- 
 ings. New and revised edition. Thirty-seventh thousand. Oblong boards. 
 
 $1.00 
 
 RICE, J. M., Prof., and JOHNSON, W. W., Prof. On a New Method 
 of Obtaining the Differential of Functions, with especial reference to the 
 Newtonian Conception of Rates of Velocities. 12mo, paper $0.50 
 
 RIPPER, WILLIAM. A Course of Instruction in Machine Drawing 
 
 and Design for Technical Schools and Engineer Students. With 52 plates 
 and numerous explanatory engravings. Second edition. 4to, cloth. $6 .00 
 
 ROBINSON, J. B. Architectural Composition. An attempt to order 
 and phrase ideas which hitherto had been only felt by the instinctive taste 
 of designers. 233 pp., 173 illustrations. 8vo, cloth net, $2.50 
 
 ROGERS, ALLEN. A Laboratory Guide of Industrial Chemistry. 
 
 Illustrated. 170 pp. 8vo, cloth net, $1.50 
 
 ROTH, W. A. Exercises in Physical Chemistry. Translated by 
 A. T. Cameron. 8vo, cloth. Illustrated net, $2.00 
 
 SCHMALL, C. N. First Course in Analytic Geometry, Plane and 
 Solid, with Numerous Examples. Containing figures and diagrams. 12mo, 
 half leather, illustrated net, $1.75 
 
 and SHACK, S. M. Elements of Plane Geometry. An Elemen- 
 tary Treatise. With many examples and diagrams. 12mo, half leather, 
 illustrated net, $1.25 
 
 SCOTT, W. W. Qualitative Chemical Analysis; a Laboratory 
 
 guide. 8vo, cloth. Illustrated net, $1.50 
 
 SEATON, A. E., and ROUNTHWAITE, H. M. A Pocket-book of 
 
 Marine Engineering Rules and Tables. For the Use of Marine Engineers 
 and Naval Architects, Designers, Draughtsmen, Superintendents and all 
 engaged in the design and construction of Marine Machinery, Naval and 
 Mercantile. Seventh edition, revised and enlarged. Pocket size. 
 Leather, with diagrams $3 00 
 
STANDARD TEXT BOOKS. 13 
 
 SEIDELL, A. (Bureau of Chemistry, Wash., D. C.). Solubilities of 
 
 Inorganic and Organic Substances. A handbook of the most reliable 
 Quantitative Solubility Determinations. Second printing, with cor- 
 rections. 8vo, cloth, 367 pp net, $3.00 
 
 SENTEB, G. Outlines of Physical Chemistry. Second edition, 
 revised. 12mo, cloth. Illustrated $1.75 
 
 SEVER, Prof. G. F. Electrical Engineering Experiments and Tests 
 
 on Direct-Current Machinery. With diagrams and figures. Second 
 edition, thoroughly revised and enlarged. 8vo, pamphlet. Illus- 
 trated net, $1.00 
 
 and TOWNSEND, F. Laboratory and Factory Tests in Elec- 
 trical Engineering. Second Edition, thoroughly revised and enlarged. 
 
 8vo, cloth. Illustrated. 236 pages net, $2.50 
 
 SEWALL, C. H. Lessons in Telegraphy. For use as a textbook in 
 schools and colleges or for individual students. 12mo, cloth. Illustrated. 
 
 $1.00 
 
 SHELDON, S., Prof., MASON, H., and HATJSMANN, E. Dynamo 
 
 Electric Machinery ; its Construction, Design, and Operation. Direct-Current 
 Machines. Eighth edition, completely revised. 12mo, cloth. Illustrated. 
 
 net, $2.50 
 
 SHELDON, S., MASON, H., and HAUSMANN, E. Alternating Current 
 
 Machines. Being the second volume of the authors' "Dynamo Electric 
 Machinery; its Construction, Design, and Operation." With many dia- 
 grams and figures. (Binding uniform with volume I.) Ninth edition, 
 completely rewritten. 12mo, cloth. Illustrated net, $2.50 
 
 SHIELDS, J. E. Notes on Engineering Construction. Embracing 
 Discussions of the Principles involved, and Descriptions of the Material 
 employed in Tunneling, Bridging, Canal and Road Building, etc. 12mo, 
 cloth $1.50 
 
 SHUNK, W. F. The Field Engineer. A Handy Book of Practice 
 
 the Survey, Location and Track-work of Railroads, containing a large 
 collection of Rules and Tables, original and selected, applicable to both the 
 standard and Narrow Gauge, and prepared with special reference to the 
 wants of the young engineer. Nineteenth edition, revised and enlarged. 
 With addenda. 12mo, morocco, tucks $2.50 
 
 SMITH, C. A M. Handbook of Testing: MATERIALS. 8 vo, cloth. 
 Illustrated $2.50 
 
 SMITH, F. E. Handbook of General Instruction for Mechanics. 
 
 Rules and formulae for practical men. 12mo, cloth, illustrated. 324 pp. 
 
 net, $1.50 
 
 SOTHERN, J. W. The Marine Steam Turbine. A practical descrip- 
 tion of the Parsons Marine Turbine as now constructed, fitted and run, 
 intended for the use of students, marine engineers, superintendent engineers 
 draughtsmen, works managers, foremen, engineers and others. Third 
 edition, rewritten up to date and greatly enlarged. 180 illustrations 
 and folding plates, 352 pp. 8vo, cloth net, $5.00 
 
14 STANDARD TEXT BOOKS. 
 
 SPEYERS, C. L. Textbook of Physical Chemistry. 8vo, cloth. 
 
 $2.25 
 
 STAHL, A. W., and WOODS, A. T. Elementary Mechanism. A Text- 
 Book for Students of Mechanical Engineering. Sixteenth edition, en- 
 larged. 12mo, cloth $2.00 
 
 STALEY, CADY, and PIERSON, GEO. S. The Separate System of 
 
 Sewerage; its Theory and Construction. With maps, plates, and illus- 
 trations. Third edition, revised and enlarged, with a chapter on 
 " Sewage Disposal." 8vo, cloth $3.00 
 
 STANSBIE, J. H. Iron and Steel. 8vo, cloth. Illustrated, net, $2.00 
 
 STODOLA, Dr. A. The Steam-Turbine, With an appendix on Gas 
 Turbines and the future of Heat Engines. Authorized Translation from 
 the Second Enlarged and Revised German edition by Dr. Louis C. Loew en- 
 stein. 8vo, cloth. Illustrated. 434 pages net, $4.50 
 
 SUDBOROUGH, J. J., and JAMES, T. C. Practical Organic Chem- 
 istry. 92 illustrations. 394 pp., 12mo, cloth net, $2.00 
 
 SWOOPE, C. WALTON. Practical Lessons in Electricity. Princi- 
 
 C, Experiments, and Arithmetical Problems. An Elementary Text- 
 k. With numerous tables, formulae, and two large instruction plates. 
 Twelfth edition. 12mo, cloth. Illustrated net, $2.00 
 
 TITHERLEY, A. W., Prof. Laboratory Course of Organic Chemistry, 
 
 including Qualitative Organic Analysis. With figures. 8vo, cloth. Illus- 
 trated net, $2.00 
 
 THURSO, JOHN W. Modern Turbine Practice and Water-Power 
 Plants. Second edition, revised. 8vo, 244 pages. Illustrated. 
 
 net, $4.00 
 
 TOWNSEND, F. Short Course in Alternating Current Testing. 8vo 
 
 pamphlet. 32 pages net, $0.75 
 
 TURRILL, S. M. Elementary Course in Perspective. 12mo, 
 cloth. . Illustrated net, $1.25 
 
 UNDERBILL, C. Solenoids, Electromagnets, and Electromagnetic 
 
 Windings. 8vo, cloth. Illustrated net, $2.00 
 
 URQUHART, J. W. Dynamo Construction. A practical handbook 
 for the use of Engineer-Constructors and Electricians in charge, embracing 
 Framework Building, Field Magnet and Armature Winding and Group- 
 ing, Compounding, etc., with examples of leading English, American, 
 and Continental Dynamos and Motors. With numerous illustrations. 
 12mo, cloth $3.00 
 
 VAN NOSTRAND'S Chemical Annual, based on Biederman's " Chiem- 
 
 ker Kalender." Edited by Prof. J. C. Olsen, with the co-operation of 
 Eminent Chemists. Revised and enlarged. Second issue 1909. 12mo, 
 cloth net, $2.50 
 
STANDARD TEXT BOOKS. 15 
 
 VEGA, Von (Baron). Logarithmic Tables of Numbers and Trig- 
 onometrical Functions. Translated from the 40th, or Dr. Bremiker's 
 thoroughly revised and enlarged edition, by W. L. F. Fischer, M.A., F.R.S. 
 Eighty-first edition. 8vo, half morocco $2.50 
 
 WEISBACH, JULIUS. ' A Manual of Theoretical Mechanics. Ninth 
 American edition. Translated from the fourth augmented and im- 
 proved German edition, with an Introduction to the Calculus by Eckley B. 
 Coxe, A.M., Mining Engineer. 1100 pages, and 902 woodcut illustrations. 
 
 8vo, cloth $6.00 
 
 Sheep $7.50 
 
 an d HERRMANN, G. Mechanics of Air Machinery. Author- 
 ized translation with an appendix on American practice by Prof. A. 
 Trowbridge. 8vo, cloth, 206 pages. Illustrated net, $3.75 
 
 WESTON, EDMUND B. Tables Showing Loss of Head Due to 
 
 Friction of Water in Pipes. Fourth edition. 12mo, full leather. . .$1.50 
 
 WILLSON, F. N. Theoretical and Practical Graphics. An Educational 
 Course on the Theory and Practical Applications of Descriptive Geometry 
 and Mechanical Drawing. Prepared for students in General Science, 
 Engraving, or Architecture. Third edition, revised. 4to, cloth, 
 illustrated net, $4.00 
 
 -Descriptive Geometry, Pure and Applied, with a chapter on 
 
 Higher Plane Curves, and the Helix. 4to, cloth, illustrated net, $3.00 
 
 WILSON, GEO. Inorganic Chemistry, with New Notation. Revised 
 and enlarged by H. G. Madan. New edition. 12mo, cloth $2.00 
 
 WIMPERIS, H. E. The Internal Combustion Engine. A textbook 
 on Gas, Oil, and Petrol Engines for the use of Students and Engineers. 8vo, 
 cloth. Illustrated net, $3.00 
 
 WINCHELL, N. H., and A. N. Elements of Optical Mineralogy. 
 
 An introduction to microscopic petrography, with descriptions of all 
 minerals whose optical elements are known and tables arranged for their 
 determination microscopically. 354 illustrations. 525 pages. 8vo, 
 cloth '. net, $3.50 
 
 WRIGHT, T. W., Prof. Elements of Mechanics, including Kinematics , 
 Kinetics, and Statics. Seventh edition, revised. 8vo, cloth $2.50 
 
 -and HAYFORD, J. F. Adjustment of Observations by the 
 
 Method of Least Squares, with applications to Geodetic Work. Second 
 edition, rewritten. 8vo, cloth, illustrated .net, $3.00 
 
 ZEUNER, A., Dr. Technical Thermodynamics. Translated from the 
 Fifth, completely revised German edition of Dr. Zeuner's original treatise 
 on Thermodynamics, by Prof. J. F. Klein, Lehigh University. 8vo, cloth, 
 two volumes, illustrated, 900 pages net, $8.00 
 
UNIVERSITY OF CALIFORNIA LIBRARY 
 This book is DUE on the last date stamped below. 
 
 Fine schedule: 25 cents on first day overdue 
 
 50 cents on fourth day overdue 
 One dollar on seventh day overdue. 
 
 LNGllNEERING LIBRARY 
 
 LD 21-100m-12,'46(A2012sl6)4120 
 
.-'': fSs 
 
 YC 32749 
 
 
 Engineering 
 Library 
 
 258752 
 
 '6 
 
 :-' : V^.;<' ill