AGRIC. DEPT, QUALITATIV CHEMICAL ANALYSIS A. A. NOYES A COURSE OF INSTRUCTION IN THE f QUALITATIV CHEMICAL ANALYSIS OF INORGANIC SUBSTANCES BY ARTHUR A. NOYES PROFESSOR OF THEORETICAL CHEMISTRY IN THE MASSACHUSETTS INSTITUTE OF TECHNOLOGY SIXTH EDITION NEW YORK: THE MACMILLAN COMPANY. LONDON: MACMILLAN & Co., LTD. 1914 CJ COPTBIGHT, 1914 BY ARTHTJK A. NOTES PREFACE. This text-book is an attempt, on the experimental side, to train the student of qualitativ analysis in careful manipulation and exact methods of procedure, such as are commonly employed in quantitativ analysis. It is an attempt, on the theoretical side, to make clear to the student the reason for each operation and result, and to accustom him to apply to them the laws of chemical equilib- rium and the principles relating to the ionization and complex-formation of substances in solution. It is believed that in both these ways the educational value of the subject is greatly increased. The book is divided into two main Parts, entitled The Course of Instruction and The System of Analysis. In presenting the System of Analysis the plan adopted in the earlier editions of this book has been followed, namely, that of separating sharply the description of the operations from the discussion and explanation of them. The operations are described with as great definitness as possible in short paragraphs entitled "Procedures"; and each of these is followed by "Notes" in which are given the reasons for the operations, the precautions necessary and difficulties encountered in special cases, the chemical behavior of the different elements, the indications afforded of their presence, and the application of the theoretical principles to the reactions involved. The system of procedure has been thoroly revised as a result of the extended investigations made in the laboratory of this Institute during the past six years and described in volumes 29, 30, 31, and 34 of the Journal of the American Chemical Society. As a result of these investigations, in which the author has had the able cooperation of Professor W. C. Bray and Professor E. B. Spear, the process of analysis has been made much more reliable, so that now it is possible with careful manipulation to detect one milligram of any constituent in the presence of 500 milligrams of any other (except in a few combinations where the limit of detectability is two milligrams) . At the same time the process has on the whole been considerably simplified. The larger size of the present edition is due, not to greater complexity of the process, but to the inclusion in the Procedures of the more explicit directions necessary to secure accuracy in the separations and reactions, to the insertion of confirmatory tests for most of the elements, to the development of a more systematic process for the detec- tion of the acidic constituents, and to the elaboration of the notes and especially the inclusion in them of the theoretical explanations made possible by the recent development of our knowledge of solutions. Vl PREFACE The Course of Instruction includes two sections one entitled Laboratory Experiments, giving the directions for the laboratory work; and the other entitled Questions on the Experiments, consisting of a series of questions to be studied in connection with the class-room exercises. The laboratory work described in the section on Laboratory Experiments is from beginning to end closely correlated with the systematic scheme of analysis. For experience has convinced the author that the plan followed in many text- books of requiring the student to study the separate reactions characteristic of the various elements before undertaking their systematic separation is highly unsat- 1 sfactory. However valuable the knowledge of the additional reactions may be, it is found in practis that the performance of such a large number of independent* disconnected experiments makes little impression on the student's mind and fails to awaken his interest in the subject. Qualitativ analysis affords an effectiv means of teaching a part of inorganic chemistry chiefly because it unites into a connected whole a great variety of isolated facts, and because the student sees a practical use of the information presented to him; but these advantages evidently do not apply to facts not directly related to the process of analysis. The Questions on the Experiments are intended to aid the student in under- standing the work that he is doing in the laboratory and to make sure that he derives from the subject the mental traning which it ought to afford. The questions are in large part of such a character that, in order to answer them properly, the student not only must study the Notes on the Procedures, but also must giv to the questions some independent thought. It is assumed in these questions, as well as in the Notes on the Procedures, that the student has acquired, in his previous course on Inorganic Chemistry, a general knowledge of the mass-action law and of the chemical aspects of the ionic theory. The best plan of conducting the course, when circumstances permit, is in the author's opinion as follows: The class, if large, is divided into sections of from 15 to 25 students. At the beginning of each laboratory exercise, which should, if possible, be three hours long, or at the beginning of every second lab- oratory exercise, the instructor holds a class-room conference with each section, at which the experiments to be next made are discussed in outline, and those made at the previous exercises are reviewed in detail. The conferences are carried on mainly by questioning the individual students and by encouraging them to ask questions as to matters which they do not understand. As prepa- ration for each conference the student is expected to study the Table in which is presented the outline of the analysis of the group to be next taken up, to study the Notes on the Procedures which he has worked through just pre- viously hi the laboratory, and (unless the course is an elementary one) to answer the questions on the corresponding Experiments. To ensure that the student givs proper study to the subject, a short written exercise may well be held at the beginning of each conference. In the laboratory work the class may be kept nearly together by giving to the faster working students additional unknown solutions on each group, and by allowing those who are falling behind PREFACE vii to omit some of the less important Experiments. In the laboratory great stress is laid on careful work, such as will enable the proportions of the various con- stituents present in unknown solutions to be estimated and small quantities of them to be detected. An effectiv means of teaching the details of manipula- tion is for the instructor to carry through in the lecture-room, after the students have had a little experience of their own in the laboratory, the com- plete process for the analysis of the copper-group. Even when the time available for the subject of qualitativ analysis does not permit of so complete a course as that here presented, the student gets, in the author's opinion, a better training by working through selected parts of an exact scheme of analysis carefully and thoroly than he does by covering the whole of an elementary scheme superficially. Experiments that may be well omitted in briefer courses are indicated by asterisks prefixed to the description of them in the section entitled Laboratory Experiments. PREFACE TO THE FIFTH EDITION. In this edition important changes and additions have been made as follows. The Procedures for the detection of the basic constituents and the Notes upon them have been improved in many matters of detail; and there have been intro- duced a simpler process for the detection of nickel and cobalt and a more satis- factory one for the analysis of the alkali-group. The part of the book relating to the detection of the acidic constituents has been entirely rewritten; and a more complete and more instructiv system of analysis for those constituents has been presented. Many additional tabular outlines have been included to assist the beginner in grasping the general plan of the separations; and the Questions on the Experiments have been revised. In making this revision the author has had the benefit of many valuable sug- gestions from Professors Henry Fay, W. T. Hall, and A. A. Blanchard of this Institute, and from Prof. G. S. Forbes of Harvard University. ARTHUR A. NOTES. MASSACHUSETTS INSTITUTE OF TECHNOLOGY, BOSTON, August, 1914. TABLE OF CONTENTS. PART I. THE COURSE OF INSTRUCTION. PAGE LABORATORY EXPERIMENTS 1 QUESTIONS ON THE EXPERIMENTS 1$ PART H. THE SYSTEM OF ANALYSIS. PREPARATION OF THE SOLUTION 27 DETECTION OF THE BASIC CONSTITUENTS 41 DETECTION OF THE ACIDIC CONSTITUENTS 92 APPENDICES. PREPARATION OF THE REAGENTS 117 PREPARATION OF THE TEST-SOLUTIONS 120 APPARATUS REQUIRED 122 IONIZATION-VALUES 123 ATOMIC WEIGHTS OF THE COMMON ELEMENTS 124 SOLUBILITIES OF SLIGHTLY SOLUBLE SUBSTANCES 124 PART I. THE COURSE OF INSTRUCTION, LABORATORY EXPERIMENTS. DETECTION OF THE BASIC CONSTITUENTS. Preliminary Work. Check off on an apparatus slip (corresponding to that printed on page 122) the apparatus found in the desk, and sign and hand in the slip. Make a 750 cc. wash-bottle, taking pains to bend the tubes and to cut them off so as to correspond closely with the model exhibited in the laboratory. Make also a 250 cc. wash-bottle (for washing with hot water and special solutions). Make a dropper about 10 cm. (4 inches) long by drawing out one end of a glass tube to a fairly wide capillary and slightly expanding the other end with the aid of a file while it is heated in a flame. Cap the expanded end with a rubber nipple. Make 3 stirring-rods about 15 cm. long by cutting a piece of glass rod into sections and then rounding the ends in a flame. Experiment i. Separation of the Basic Constituents into Groups. In connection with this experiment read the General Discussion on page 41 and refer to Table II (page 42). Measure out with the aid of a 10 cc. graduate 5 cc. portions of the test-solutions (see Note 1) of AgN0 3 , Cu(N0 3 ) 2 , Zn(N0 3 ) 2 , Ca(N0 3 ) 2 , and KN0 3 . Mix the por- tions in a conical flask, add 5 cc. 6-normal HN0 3 and 10 cc. NH 4 C1 solution, shake for a minute or two, and filter. Dilute the filtrate with water to a volume of 100 cc. Place it in a 200 cc. conical flask; insert a two-hole rubber stopper through which passes a tube leading to the bottom; and pass in a slow current of H 2 S, until, upon shutting off the gas and shaking thoroly, the liquid smells strongly of it. Fil- ter. To the filtrate add 10 cc. NH 4 OH and 3-5 cc. (NH 4 ) 2 S. Shake the mixture and filter. Evaporate the filtrate to a volume of about 10 cc., filter, and to the cold solution add 15 cc. (NH 4 ) 2 CO 3 reagent and 15 cc. alcohol. In this experiment and all subsequent ones observe carefully every- thing that happens, and record it clearly and neatly in the note-book in the way described in Note 2, together with the equations expressing all the chemical changes that take place. In connection with each Ex- periment study the Table and the Notes referred to in the directions. 1 ' LABORATORY EXPERIMENTS. Notes. 1. The solutions of constituents to be tested for, here called the test-solutions, are all so made up as to contain 10 mg. (10 milligrams) of the constituent per cubic centimeter of solution. The mixture used hi this ex- periment therefore contains 50 mg. of each of the basic constituents silver, copper, zinc, calcium, and potassium. The student should acquire the habit of working with definit quantities of the constituents and of noting the size of the precipitates which they yield. For a good qualitativ analysis should not only show the presence or absence of the various constituents, but should also furnish an estimate of the proportions in which they are present. Test-solutions should not be used in place of reagents, nor reagents in place of test-solutions, since the concentrations are as a rule quite different. In regard to the concentrations of reagents, see Note 3, page 44. 2. In the note-book the operations should be indicated very briefly; but ererything that happens should be recorded fully, tho concisely. Thus the report of the first experiment may be made in the following form: Expt i. Added HN0 3 : no change observed. Added NH 4 C1: white curdy ppt. Ag+NO 3 -+NH 4 + Cl- = AgCH-NH 4 + NO 3 -. Passed in H$: large black flocculent ppt. Cu ++ (NO 3 -) 2 +H 2 S = CuS+2H+N0 3 -. Solid substances involved in chemical reactions should be indicated by under- lining their formulas. Largely ionized dissolved substances should be written with + and signs attached to the formulas in such a way as to show the ions into which they dissociate. Slightly ionized dissolved substances should be distinguished by not attaching these signs to the formulas. In regard to the ionization of substances, see the Table on page 123. The neatness, accuracy, and completeness of the notebook record will be an important factor in determining the grade of the student. Experiment 2. Analysis of the Silver-Group. Mix in a conical flask 20 cc. of the test-solution of Pb(NOa)2 with 5 cc. portions of the test- solutions of Bi(NO 3 )3, AgNO 3 , and Hg 2 (N0 3 ) 2 , and treat the mixture by P. 11-15 (i.e., by Procedures 11-15 of the System of Analysis de- scribed in Part II, on pages 43-46); omit the confirmatory tests mentioned at the end of P. 12 and 13. Study Table III (page 43) before carrying out this experiment; and in connection with it read the Notes on P. 11-15. Note. In all these "Experiments" only the Procedures actually named in the directions should be worked through, omitting any others that may be inci- dentally referred to in the System of Analysis. Experiment 3. Precipitation by Hydrogen Sulfide. To 10 cc. of the test-solution of Bi(N0 3 ) 3 in a 200 cc. conical flask add 5 cc. HNO 3 , 10 cc. NH 4 C1 solution, and 75 cc. water; and, without filtering the mixture, pass H 2 S into it till it becomes saturated, in the way LABORATORY EXPERIMENTS. 3 described in the first paragraph of p. 21, omitting the filtration at the end. (The HNO 3 and NH 4 C1 are added and the solution is diluted to 100 cc. so as to have the same conditions as in an actual analysis.) Read Notes 1 and 2, P. 21. *To 10 cc. of the test-solution of H 3 As0 4 add 5 cc. HNO 3 , 10 cc. NH 4 C1 solution, and 75 cc. water. Treat this solution by the whole of P. 21. Read Note 3, P. 21. * Note. Experiments or parts of experiments preceded by an asterisk may be omitted in brief courses on the subject when the instructor so directs. Experiment 4. Effect of Acid on the Precipitation by Hydrogen Sulfide. Introduce into each of three test-tubes by means of a dropper (see Note 2, P. 11) 3 drops of the test-solution of Cd(N0 3 ) 2 . Add to the first tube 1 cc. HC1, to the second 3 cc. HC1, and to the third 9 cc. HC1. Then add to each solution enough water to make the volume about 20 cc., and pass a slow current of H 2 S into it for about a minute. Repeat the last test (with 9 cc. HC1), substituting Cu(NO 3 )2 for the Cd(N0 3 )2. Calculate the normal concentration of the HC1 in each tube and the number of milligrams of cadmium or copper per 100 cc. solution; and record the values in the note-book. Read Notes 4 and 5, P. 21. Experiment 5. Effect of Oxidizing Substances on Hydrogen Sulfide. To 20 cc. of the test-solution of Fe(N0 3 ) 3 add 10 cc. NH 4 C1 solution ; 5 cc. HN0 3 , and 65 cc. water, and pass in H 2 S till the solution is satu- rated. Repeat this experiment, substituting 20 cc. of the test-solution of K 2 Cr0 4 for that of the Fe(N0 3 ) 3 . Read Notes 6 and 7, P. 21. Experiment 6. Analysis of the Copper-Group. Mix 10 cc. portions of the test-solutions of Hg(NO 3 ) 2 , Pb(NO 3 ) 2 , Bi(NO 3 ) 3 , Cu(NO 3 ) 2 , and Cd(N0 3 ) 2 , add 5 cc. HN0 3 , 10 cc. NH 4 C1 solution, and 35 cc. water, treat the mixture by P. 21, and treat the precipitate so obtained by P. 31-38. Refer to Table V (page 54), and read the Notes on P. 31-38. Experiment 7. Analysis of an Unknown Solution for Elements of the Copper-Group. Ask the instructor for an unknown solution con- taining elements of the copper-group ("unknown A"), and analyze 10 cc. of it by P. 21 and P. 31-38, first adding 5 cc. HN0 3 and 10 cc. NH 4 C1 solution. Record and report the results as described in the following Note. Keep all final tests in properly labelled test-tubes or flasks until the report on the analysis has been returned by the in- structor. LABORATORY EXPERIMENTS. Note. Record the results of the analyses of unknown solutions in the note- book in three columns headed respectivly, "Operations," "Observations," "Conclusions." The operations and observations are to be recorded in the same brief form employed in the experiments with known solutions. In the column headed Conclusions are to be inserted the conclusions that may be drawn from each observation as to the presence or absence of any of the constituents that may be present in the unknown solution. The chemical equations involved need not be written. Sum up at the end the constituents that have been found to be present, giving also a rough estimate of the quantity of each of them per 10 cc. of solution. Quantities less than 5 mg. may be reported as "small"; those from 5 to 50 mg. as "medium"; and those greater than 50 mg. as "large." (It is to be noted, since one gram of a non-metallic solid substance is ordinarily taken for analysis, that 5 mg. corresponds to the presence of 0.5% and 50 mg. to the presence of 5% of the constituent in such a substance.) The quantity of any constituent present is to be estimated from the size of the precipitate obtained in the confirmatory test or in the Procedure preceding it. The student should make it a habit to compare this precipitate in any doubtful case with that obtained by subjecting a known quantity of the test-solution to the same final Procedure. For this purpose the test-solution may be measured out with the aid of a dropper, noting that three medium-size drops correspond to about 0.1 cc. of the solution or to 1 mg. of the constituent contained in it. The instructor will return the report of the student with an entry upon it showing the quantities of the various constituents which the unknown actually contained. The correctness of the results obtained in the analysis of these unknown solutions is an important factor in determining the grade of the student. Experiment 8. Behavior of Elements of the Tin-Group towards Hydrogen Sulfide and Ammonium Sulfide. To 5 cc. water in each of three test-tubes add from a dropper 6 drops respectivly of the test- solutions of AsCls, of SbCls, and of SnCU. (Note that 6 drops of a test-solution contains 2 mg. of the constituent to be tested for.) Pass H^S into each tube for half a minute. Then add from a graduate 2 cc. ammonium polysulfide solution. Finally add 3 cc. HC1 slowly to each tube, and shake the mixture. Compare these precipitates with that produced by mixing 5 cc. water, 2 cc. ammonium polysulfide solution, and 3 cc. HC1, and shaking. (In an actual analysis the ana- lyst decides from the appearance of the HC1 precipitate whether it contains an appreciable quantity of the tin-group.) Refer to Table IV (page 47). Experiment 9. Separation of the Tin-Group from the Copper-Group. To a mixture of 5 cc. portions of the test-solutions of Bi(NOa)s, AsCl 3 , SbCl 3 , and SnCU add 5 cc. HNO 3 , 10 cc. NH 4 C1 solution, and enough water to make the volume 100 cc. Treat the mixture by the first paragraph of P. 21, filter with the aid of suction (see Note 1, P. 23), treat the precipitate by P. 22, using 10 cc. ammonium polysulfide, reject the residue of Bi 2 S 3 , and treat the solution by P. 23. Treat LABORATORY EXPERIMENTS. 5 at once the precipitate obtained in P. 23 as described in Expt. 10. Refer to Table IV (page 47); and read the Notes on P. 22 and P. 23. Experiment 10. Analysis of the Tin-Group. Treat the precipi- tated sulfides obtained in Expt. 9 by P. 41-46. Refer to Table VI (page 60), and read the Notes on P. 41-46. Experiment n. Analysis of Unknown Solutions for Elements of the Copper and Tin Groups. Ask for two unknown solutions contain- ing elements of these groups ("unknowns B and C"), and analyze 10 cc. of each of them. First, in order to secure the proper acid con- centration for the H 2 S precipitation, make the solution exactly neutral by adding to it NH 4 OH drop by drop till it no longer reddens blue litmus paper, and add just 5 cc. HN0 3 and enough water to make the volume 100 cc. Then treat the mixture by P. 21-46. (Unknown B will contain only elements of the tin-group.) Experiment 12. Precipitation of the Aluminum and Iron Groups and Solution of the Group-Precipitate. Treat a mixture of 10 cc. portions of the test-solutions of Co(N03)2 and of Fe(N0 3 ) by P. 51 and by the first five lines of P. 52. Refer to Table VII (page 65), and read Note 1, P. 51. and Notes 1-2, P. 52. Experiment 13. Behavior of Elements of the Aluminum and Iron Groups towards Ammonium Hydroxide and Sulfide. To 5 cc. portions of the test-solutions of A1(NO 3 ) 3 , CrCl 8 , Fe(N0 3 ) 3 , FeS0 4 , Zn(N0 3 ), Mn(N0 3 )2, Ni(NO 3 )2, and Co(NO 3 )2, in separate test-tubes add 3 cc. NH 4 C1 solution and 8-10 drops of NH 4 OH, and note the result. Then add 2-3 cc. more NH 4 OH. Finally add 1-2 cc. (NH 4 ) 2 S to each tube. Filter out the NiS precipitate, and boil the filtrate for 2 or 3 minutes. Record the results of all these tests in a single table, so as to show what effect is observed and what compound is formed in the case of each element upon the addition of each reagent. Study the results, refer to Table VII, and read Notes 2-5 and 8-10, P. 51. ' Experiment 14. Behavior of Elements of the Aluminum and Iron Groups towards Sodium Hydroxide and Peroxide. To separate 5 cc. portions of the test-solutions named in Expt. 13, add 8-10 drops of NaOH, and note the result. Then add 2-3 cc. more, and again note the result. Finally to each of the mixtures add gradually from a dry 7-cm. test-tube 0.2-0.3 cc. Na 2 O 2 powder, and heat it to boiling. Re- cord all the results in a single table as in Expt. 13. Study the results, refer to Table VII, and read Notes 3-7, P. 52. *Experiment 15. Precipitation of Alkaline- Earth Elements by Ammonium Hydroxide in the Presence of Phosphate. Heat about 6 LABORATORY EXPERIMENTS. 0.3 g. solid Ca 3 (P0 4 ) 8 with 10 cc. water; then add 5 cc. HN0 3 . To the solution add NEUOH till the mixture after shaking smells of it; filter out the precipitate; and add 1-2 cc. (NH 4 ) 2 C0 3 reagent to the filtrate. Read Notes 6-7, P. 51, and Note 8, P. 52. Experiment 16. Analysis of the Aluminum-Group. Treat a mix- ture of 10 cc. portions of the test-solutions of A1(N0 3 ) 3 , CrCla, and Zn(N0 3 ) 2 by the second paragraph of P. 52 and by P. 53-57. Refer to Table VIII (page 71), and read the Notes on P. 53-57. Experiment 17. Analysis of the Iron-Group: Separation of Man- ganese and Iron. Treat a mixture of 10 cc. portions of the test- solutions of Mn(N0 3 ) 2 , Fe(NO 3 ) 3 , Zn(N0 3 ) 2 , Co(N0 3 ) 2 , and Ni(N0 3 ) 2 by the second paragraph of P. 52; and treat the precipitate there- by obtained by P. 61, 62, 64, and 66. Treat the precipitate containing the zinc, cobalt, and nickel as described in Expt. 18. Refer to Table IX (page 74), considering only the case where " phosphate is absent," and read the Notes on P. 61, 62, 64, and 66. *Experiment 18. Analysis of the Iron-Group: Separation of Zinc, Nickel, and Cobalt. Treat the residue and precipitate obtained in Expt. 17 by P. 67-70. Refer to Table X (page 78), and read the Notes on P. 67-70. Note. In brief courses this experiment may be omitted; and in the subse- quent analyses of unknowns cobalt and nickel may be detected (without dis- tinguishing them) by the formation of a black precipitate in P. 66, and zinc may be tested for only in the aluminum-group. *Experiment 19. Modification of the Analysis of the Iron-Group in the Presence of Phosphate for the Purpose of Detecting Alkaline- Earth Elements. Mix together 10 cc. portions of the test-solutions of Fe(N0 3 ) 3 , of Co(N0 3 ) 2 , and of Caa(PO 4 )s in HN0 3 . Treat one-tenth of this solution by P. 63, and treat the remainder of it by P. 65 and 66. To the filtrate obtained in P. 66 add 2-3 cc. (NH 4 ) 2 C0 3 reagent. Read the Notes on P. 65. Experiment 20. Analysis of Unknown Solutions for Elements of the Aluminum and Iron Groups. Ask the instructor for two unknown solutions for this purpose ("unknowns D and E"), and treat 10 cc. of each by P. 51-57 and 61-70. (Unknown E will contain phosphate.) Experiment 21. Precipitation of the Alkaline- Earth Group. To 2 cc. of the test-solution of Mg(N0 3 ) 2 add 10 cc. water and 1-2 cc. (NH4) 2 C0 3 reagent, and shake the mixture for about a minute. Then add (in accordance with P. 81) 15 cc. (NH 4 ) 2 C0 3 reagent and 15 cc. 95 per cent alcohol, and shake for a minute more. LABORATORY EXPERIMENTS. 7 To 2 cc. of the test-solution of Ca(N0 3 ) 2 add 10 cc. water and 2 cc. (NH 4 ) 2 C0 3 reagent, shake, and after 2-3 minutes filter out the pre- cipitate. To the filtrate add 15 cc. (NH^COs reagent and 15 cc. 95 per cent alcohol. Read the Notes on P. 81. Experiment 22. Analysis of the Alkaline- Earth Group. Mix to- gether in a small flask 3 cc. portions of the test-solutions of BaCl 2 , Sr(N0 3 ) 2 , Ca(N0 3 ) 2 , and Mg(N0 3 ) 2 . Add to the mixture 30 cc. of the (NH 4 ) 2 C0 3 reagent and 30 cc. 95 per cent alcohol, and shake it for about 5 minutes. Filter out the precipitate and treat it by P. 82-89. Refer to Table XI (page 81;, and read the Notes on P. 82-89. Experiment 23. Analysis of the Alkali-Group. Mix together 10 cc. portions of the test-solutions of KNOs and NaN0 3 , add 10 cc. NH 4 C1 solution and 10 cc. (NH 4 ) 2 C0 3 reagent, and treat the mixture by P. 91- 95. Refer to Table XII (page 86), and read the Notes on P. 91-95. Experiment 24. Analysis of an Unknown Solution for Elements of the Alkaline -Earth and Alkali Groups. Ask the instructor for an un- known solution for this purpose ("unknown F"), and analyze 10 cc. of it by P. 81-89 and P. 91-95. *Experhnent 25. Detection of Ammonium. Treat 0.2-0.3 g. solid NH 4 C1 by P. 96. Read the Notes on P. 96. *Experiment 26. Determination of the State of Oxidation of Iron. Treat 0.3 g. finely powdered Fe 3 O 4 by P. 97, omitting tests a, 6, and c. Read the Notes on P. 97. *Experiment 27. Determination of the State of Oxidation of Arsenic. Treat 1 cc. of the test-solution of H 3 As0 4 by P. 98. Read the Notes on P. 98. *Expenment 28. Detection of a Small Quantity of Arsenic. Add 6 drops of the test-solution of H 3 As0 4 to 10 cc. H 2 S0 4 , and treat the mixture by P. 99. Read the Notes on P. 99. Experiment 29. Analysis of Unknown Solutions for All the Basic Constituents. Ask the instructor for two unknown solutions for this purpose ("unknowns G and H") } and analyze 10 cc. of each of them by P. 11-95. Before precipitating with H 2 S, exactly neutralize the solution with NH 4 OH and add 5 cc. HN0 3 . Note. In complete analyses of this kind where a number of different precipi- tates and filtrates are sue essivly obtained, any of these that are set aside, even temporarily, should be distinctly labelled, in order to avoid mistakes. A convenient method of doing this is to mark on the label simply the Procedure by which the precipitate or filtrate is next to be treated; thus the H 2 S precipi- tate would be marked P. 22, and the filtrate from it P. 51. The final tests for any element may be marked Test for Pb, Test for Al, etc. 2 8 LABORATORY EXPERIMENTS. DETECTION OF THE ACIDIC CONSTITUENTS. Experiment 30. Detection of Readily Volatil Constituents. Test a mixture of 5 drops of the test-solution of Na 2 CO 3 and 5 drops of the test-solution of Na 2 S (in place of "0.3 g. of the finely powdered sub- stance") for carbonate and sulfide by P. 101. Test 5 drops of the test-solution of NaN02 for nitrite by P. 101. Test 5 drops of the test-solution of KCN for cyanide by P. 101. Read the General Discussion on pages 93-94, refer to Table XIII page 96), and read the Notes on P. 101. Experiment 31. Detection of Sulfate, Sulfite, and Fluoride. Mix together 2 cc. portions of the test-solutions of NajSCX, Na^Os, and KF, add 2 cc. HN0 3 , and treat the mixture by P. 102. (Reject the CaCl 2 precipitate.) Refer to Table XIV (page 97), and read the Notes on P. 102. Experiment 32. Detection of Halides and of Chlorate. Mix to- gether 2 cc. portions of the test-solutions of KC1 and KC10 3 , add 2 cc. HN0 3 , and treat the mixture by P. 103. Refer to Table XIV (page 97), and read the Notes on P. 103. Experiment 33. Detection of Phosphate. To 2 cc. of the test-solu- tion of Na 2 HPO 4 add 2 cc. HN0 3 , and treat the mixture by P. 104. Read the Notes on P. 104. Experiment 34. Detection of Thiocyanate. To 2 cc. of the test- solution of KSCN add 10 drops of HNOs, and treat the mixture by P. 105. Read the Notes on P. 105. Experiment 35. Detection of the Separate Halides. Mix together 2 cc. portions of the test -solutions of KC1, KBr, and KI, and treat the mixture by P. 106. Refer to Table XV (page 100), and read the Notes on P. 106. Experiment 36. Analysis of an Unknown Solution for the Acidic Constituents tested for in Nitric-Acid Solution. Ask the instructor for an unknown solution for this purpose ("unknown J"). To 10 cc. of it add 15 cc. water and 5 cc. HNOs, and treat portions of the mixture by P. 102-106. Refer to Tables XIV and XV (pages 97 and 100). *Experiment 37. Distillation with Phosphoric Acid and Detection of Carbonate, Halides, and Sulfate. Mix 3 cc. portions of the test-solu- tions of Na2C0 3 , KC1, and K 2 S0 4 , and treat this mixture (in place of "2 g. of the finely powdered substance") by P. 111. Treat a portion of the second distillate by the first paragraph of P. 116. Treat the LABORATORY EXPERIMENTS. 9 third distillate by P. 119. Refer to Table XVI (page 102), and read the Notes on P. Ill, 116, and 119. *Experiment 38. Detection of Carbonate and Sulfite in the Presence of Each Other. Mix 3 cc. portions of the test-solutions of Na 2 C03 and Na 2 SOs, add to the mixture 10 cc. Ba(OH) 2 solution, acidify with HAc, and treat the mixture (in place of "one-half of the first dis- tillate") by P. 112. Refer to the first part of Table XVII (page 105), and read the Notes on P. 112. *Experiment 39. Detection of lodin-Liberating Constituents. Treat 10 cc. of the test-solution of NaOCl (which contains also an equiva- lent quantity of NaCl and some Na2C0 3 ) by the first paragraph of P. 111. Treat the whole distillate so obtained by the first two paragraphs of P. 113. Treat 3 cc. of the test-solution of KNC>2 by the first two paragraphs of P. 113. Refer to the middle part of Table XVII (page 105), and read the Notes on P. 113. *Experiment 40. Detection of Cyanide. Treat 3 cc. of the test- solution of KCN by P. 115. Refer to the last column of Table XVII (page 105), and read Notes 1-3, P. 115. *Experiment 41. Analysis of an Unknown Solution for Acidic Con- stituents passing into the Phosphoric Acid Distillates. Ask the in- structor for an unknown solution for this purpose ("unknown K"), and treat 10 cc. of it by P. 111. Treat the distillates by P. 112-119. -Refer to Tables XVII and XVIII (pages 105 and 109). Experiment 42. Detection of Borate. Treat 0.2-0.3 g. of solid borax (Na 2 B 4 O 7 ) by P. 121. In connection with each one of Expts. 42-47 refer to Table XIX (page 111), and read the Notes on the Pro- cedure involved in the Experiment. Experiment 43. Detection of Fluoride. Treat 0.2-0.3 g. of solid CaF 2 by P. 122. Experiment 44. Detection of Nitrate. Treat 0.2-0.3 g. of solid KN0 3 by P. 124. *Experiment 45. Detection of Nitrite. Treat 1 cc. of the test- solution of KNO 2 by P. 125. *Experiment 46. Detection of Hypochlorite. Treat 3 cc. of the test-solution of -NaOCl by P. 126. *Experiment 47. Detection of Chlorate in Presence of Hypochlorite. Mix 3 cc. portions of the test-solutions of NaOCl and KC10 3 , and treat the mixture by P. 127. 10 LABORATORY EXPERIMENTS. PREPARATION OF THE SOLUTION AND COMPLETE ANALYSES OF UNKNOWN SOLID SUBSTANCES. Experiment 48. Substances Soluble in Water or Dilute Acid. Ask the instructor for two such unknown substances ("unknowns I and II"), and treat portions of each of them by P. 1, by P. 2 followed by P. 11-95, by P. 96-98, and by P. 100, 101-106, 121, and 124-126.- Read the Notes on P. 1 and 2. Record and report the results in the note-book as directed in the Note on Expt. 7. In the case of a solid substance not only report the constituents and the proportions of them present, but state the compound or compounds of which the substance seems to be mainly composed. Note. In analyzing unknown solids the quantity taken for the analysis should be weighed (within 0.1 g.) on a rough balance, not guessed at nor esti- mated by volume. Experiment 49. Non-Metallic Substances requiring Treatment with Concentrated Acids. Ask the instructor for two such substances ("unknowns III and IV"), and treat portions of each of them by P. 1, by P. 2-3 (or 2-4, if necessary) followed by P. 11-95 (or 21-95), by P. 96-98, and by P. 100, 111-119, and 121-127. If there is any undissolved residue (of silica or silicate) at the end of P. 4, disregard it in these analyses. Refer to Table I (page 29), and read Notes 1-4, P. 3, and Notes 1-2, P. 4. Experiment 50. Alloys Dissolved by Concentrated Acids. Ask the instructor for two such alloys ("unknowns V and VI") , and treat 0.5 g. of each by P. 3-4 and P. 11-70 (or P. 21-70). Read Notes 5-9, P. 3, and Notes 3-4, P. 4. *Experiment 51. Mineral Substances Not Completely Dissolved by Concentrated Nitric and Hydrochloric Acids. Ask the instructor for two such substances ("unknowns VII and VIII"). Treat 1 g. of each of them by P. 2-6, followed by P. 11-95. Refer to Table I (page 29), and read the Notes on P. 5 and 6. Treat fresh portions of each of the substances by P. 131. Read the Notes on P. 131. Ask the instructor for another such substance ("unknown IX"). Treat 1 g. of it by P. 2-4. Treat the solution obtained in P. 4 by P. 21-95, and the residue obtained in P. 4 by P. 7, followed by P. 21-89. Reserve, as directed, one-half of the aqueous extract of the fused mass obtained in P. 7, and test it for acidic constituents as de- scribed in the third and fourth paragraphs of P. 131. Read the Notes on P. 7. Treat fresh portions of the substance by P. 101 and P. 102-104. LABORATORY EXPERIMENTS. 11 *Experiment 52. Substances Containing Organic Matter. Ask the instructor for such a substance ("unknown X")> and treat portions of it by P. 1, by P. 8 followed by P. 11-95, by P. 96-98, and by P. 101-106. Read the Notes on P. 8. QUESTIONS ON THE EXPERIMENTS. DETECTION OF THE BASIC CONSTITUENTS. Experiment i. 1. In precipitating the silver-group in an actual analysis could the NH 4 C1 be replaced by NaCl? by HC1? (In the case of all questions which can be answered by "yes" or "no," giv the reasons for the answer.) 2. If the NH 4 C1 were not added, what would happen to the silver in the subse- quent parts of the experiment? 3. Of the five basic constituents present in the mixture why is silver the only one that is precipitated by NH 4 C1? 4. If enough H 2 S were not used to precipitate all the copper, how would it behave on the subsequent addition of NH 4 OH and (NH 4 ) 2 S? 5. What is the first reaction that takes place when NH^OH is added to the filtrate from the H 2 S precipitate? 6. What would happen to the (NH^S if it were added directly to the filtrate from the H 2 S precipitate, without first adding NH 4 OH? 7. What happens to the (NH 4 ) 2 S when the filtrate from the (NHOaS precipitate is evaporated? 8. If all the basic constituents had been present in the original mixture used for this experiment, what ones would have been precipitated by (a) NH 4 C1, (6) H 2 S, (c) NH 4 OH and (NH^S, (d) (NH 4 ) 2 CO 3 ? (e) What ones would have been left with the potassium hi the filtrate from the (NH 4 ) 2 COa precipitate? Experiment 2. 1. What would be meant by the statement that a certain quan- tity of lead nitrate is equivalent to a certain other quantity of ammonium chloride? 2. In making up one liter of a 1-normal solution of NH 4 C1, how many grams of the salt should be weighed out, and how much water should be added to it? (For the atomic-weight values needed in answering this and other questions see the table on page 124.) 3. In making up a liter of a 6-normal solution of H 2 S0 4 , how many cubic cen- timeters of 96% sulfuric acid (s. g., 1.84) should be used, and how much water should be added to it? 4. Approximately how many cubic centimeters of the NH 4 C1 solution would be required to precipitate 500 mg. of silver? (Calculate first the number of equivalents corresponding to 500 mg. Ag. Since 1 g. of the unknown substance is ordinarily taken for the analysis for basic constituents, 500 mg. is as large a quantity of any element as is likely to be present.) 5. Why is a considerable excess of NH 4 C1 added? (The word excess signifies the quantity added beyond the equivalent quantity theoretically required to produce the reaction in question.) 6. The solubility of Pbd2 at 20 in water is 0.070 equivalents per liter and in 0.2-normal NH 4 C1 solution is 0.018 equivalents per liter. Explain by the solubility- product principle why NH 4 C1 diminishes the solubility. (This question may be answered by shortening the complete explanation given in Note 6, P. 11, as follows: "In any dilute sol'n satur. with PbCl 2 , (Pb ++ ) X (Cl~) 2 = satur. value. NH 4 C1 added to such a sol'n causes, owing to its ionization into NH 4 + and Cl~, an increase in (Cl~), and therefore raises (Pb + +) X (Cl~) 2 above the saturation value, BO that 13 14 QUESTIONS ON THE EXPERIMENTS. Pbdz ppts." (All other questions as to the effect of one substance on the solubility of another substance should be answered in a similar way. Always consider the effect which the added substance may have on the concentration of each of the ions of the salt with which the solution is saturated, and state the reason for any such effect.) 7. Calculate by the solubility-product principle, from the fact that the solubility of PbCk in water at 20 is 0.070 normal, what its solubility would be in a solution 0.16 normal in chloride-ion (which is approximately the chloride-ion concentration in a 0.2 normal NHjCl solution). Assume the PbCl2 to be completely ionized. 8. From the data given in Question 6 calculate how many milligrams of lead would have to be present in 40 cc. water at 20, in order that any precipitation of PbCl2 may result on adding to it 10 cc. 1 normal NH^Cl solution. 9. In the precipitation of bismuth by the NKiCl solution, what ion -concentration product comes into consideration? What must be true of its value in order that bismuth may be precipitated? Explain why increasing the HNOa-concentration increases the quantity of bismuth that remains in solution. 10. The solubility of AgCl at 100 is 0.022 g. per liter. Calculate how many milligrams of silver might be lost if the chloride precipitate were washed with 100 cc. boiling water. 11. In testing for lead in P. 13, explain why the addition of an excess of H2SO4 to the solution diminishes the solubility of the precipitate, and thus increases the deli- cacy of the test. 12. What other elements besides lead might be precipitated by adding H2S04 to a solution in which they were present? Since these other elements are precipitable by H2SO4, why does the formation of a precipitate in P. 13 show the presence of lead? 13. Explain by the solubility-product principle why the formation of the complex salt Ag(NHa)2^Cl~ causes AgCl to be much more soluble in NBUOH solution than hi water. (Answer in accordance with the Note on Question 6.) 14. Formulate the mass-action expression for the equilibrium between the com- plex cation Ag(NH3)2 + and its constituents. Show by reference to this expression and the solubility-product principle why the addition of HNOs causes AgCl to be precipitated out of its solution in NH^OH. Experiments 3 ancl 4. 1. In precipitating with E^S in P. 21 what is the reason for adding 5 cc. HNOs and diluting the solution to 100 cc.? Why not use less acid and thus avoid all risk of failing to precipitate the elements of the copper and tin groups? 2. In passing H2S into a Cu(NOa)2 solution, at what stage in the process does the solution after shaking begin to smell of the gas? 3. What principle determine how the quantity of [28 dissolved by a given quantity of water varies with its partial pressure? What would its partial pressure be in a mixture made by mixing 1 volume of H^S with 4 volumes of air, each at a pressure of one atmosphere? 4. Why would a larger quantity of an element have to be present in order to giv a precipitate if in P. 21 the solution were treated with H^S in an open beaker, in- stead of in the closed flask? 5. Giv two reasons why a larger quantity of an element would have to be present to giv a precipitate if the solution were saturated with H2S at 80, instead of at 20. 6. The solubility (in equivalents per liter) of freshly precipitated ZnS in water is about 100 times as great as that of CdS. Calculate by the principles discussed in Note 4, P. 21, the ratio of the hydrogen-ion concentrations at which the precip- QUESTIONS ON THE EXPERIMENTS. 15 itation of cadmium and zinc will barely take place when the concentration of each of them has any definit value (for example, 0.0001 equivalents per liter). 7. The pressure* volume relations of perfect gases are expressed 03^ the equation p v I T = 82 N, when the pressure p is in atmospheres, the volume v in cubic centi- meters, and the temperature T in centigrade-degrees on the absolute scale, and when the quantity of the gas is N gram-molecular-weights. Calculate the number of cubic centimeters of H^S at 25 required to precipitate 500 mg. of copper. 8. By what reaction is the HNOs destroyed when the arsenic solution to which HC1 has been added is evaporated to dryness? Could HC1 be destroyed in the same way by evaporating a solution of chloride with HNOs? 9. If the HNOs were not so destroyed, what would happen when the H^S is passed into the hot, strongly acid solution? Experiment 5. 1. What substances besides ferric salts might be present which would cause precipitation of sulfur in P. 21? 2. Write the equation expressing the reaction between each of these substances and H2S, balancing the equations by the method described in Note 7, P. 21. 3. Write by the same method the equations expressing the oxidation of H^S, in one case to sulfur and in another to H2S04, by hot fairly concentrated HNOs, assuming that the HNOs is reduced to NO. Experiment 6. 1. Make a table showing briefly in the first column the chemical operations involved in analyzing a solution for mercury and lead (by P. 21 and P. 31-34), and showing in a second and in a third column the behavior of these two elements in each operation. "Behavior" in this and later questions means both the effect observed and the chemical compound produced. Thus, the first two operations and the results of them should be entered as follows: Operation Behavior of Mercury Behavior of Lead Satur. with H 2 S. Black ppt. of HgS. Black ppt. of PbS. Boil with 2-n. HNOs. No change. Ppt. diss., forming colorless sol'nofPb ++ (N0 3 -) 2 . 2. Make a similar table showing the operations involved in analyzing a solution for bismuth, copper, and cadmium (by P. 21, 31, 33, and 35-38), and showing the behavior of these elements. 3. Explain by the solubility-product principle the fact that CuS, which is only slightly soluble in hot dilute HC1, dissolves readily in hot dilute HNOa of the same concentration. 4. Write the equation expressing the dissolving of CuS in the HNOs. 5. Why is a black residue left undissolved by HNOs not sufficient evidence of the presence of mercury? 6. Suggest a reason why hi the confirmatory test for mercury the addition of HC1 with the SnCl2 tends to prevent the immediate reduction of the Hg&Ck to Hg. 7. Why does the evaporation with H2SO4 convert the salts present into sulfates? Could sulfates be converted into nitrates by evaporating with a large excess of HN0 3 ? 8. Explain with reference to the solubility-product principle why PbS04 is much more soluble in dilute HNOa than in water. ([2804 in dilute solution is dissociated almost completely into H + and HSO4~; but the latter ion is only to a moderate ex- tent dissociated into H + and SC>4~). 9. What effect, as compared with that of HNOa, would HC1 have on the solubility of PbS0 4 ? What effect would KNO 3 have? Giv reasons. (K 2 S0 4 in dilute solu- 16 QUESTIONS ON THE EXPERIMENTS. tion, like other unibivalent salts, but unlike [2804, is almost completely dissociated into the simple ions, K + and SO4 = , with formation of only a small proportion of the intermediate ion, KSCV.) 10. Explain by the solubility-product principle why the fact that PbAc 2 is a slightly ionized substance should cause PbSO4 to dissolve much more readily in NEUAc solu- tion than hi water. 11. Would you expect PbCrO4 also to be more soluble in NELiAc solution than in water? Why or why not? if so, why does PbCrO4 precipitate from the same NH 4 Ac solution that dissolves PbSO 4 ? 12. Arrange all the compounds of lead thus far met with in the order in which their solubility in water decreases. 13. If in an actual analysis a precipitate of BiOCl formed on diluting the solution before passing in H 2 S in P. 21 and that precipitate were filtered off, would a bis- muth precipitate be obtained with NEUOH in P. 35? if the BiOCl were not filtered off, would any change occur in it on passing in H 2 S? 14. Explain with the aid of the mass-action expressions involved why Cu(OH)2, a substance very slightly soluble in water, is not precipitated by the NH 4 OH. Show that the presence of the (NH4)2SC>4 in the solution must diminish the tendency of it to precipitate. 15. If the lead were not removed by the addition of [2804, would it be precipitated as Pb(OH) 2 on the addition of NEUOH? What knowledge hi regard to lead com- pounds would enable one to predict whether or not this precipitation would take place? 16. Write the equations expressing the formation of Na 2 SnO2 from SnCl2 and NaOH; also that expressing the spontaneous decomposition of NagSnC^ into tin and Na 2 Sn03; also that expressing its action on BiOgHs. 17. Lead hydroxide, like Sn(OH) 2 , is an amphoteric substance. What is meant by this statement? What experiments might be made to determin whether it is true? 18. If K4Fe(CN)e be added in P. 37 to the ammoniacal solution (without neutraliz- ing it with HAc), no precipitate is produced unless a fairly large quantity of copper is present. Explain this fact. 19. Show from the solubility-product expressions that, if Cd2Fe(CN)6 is much more soluble than Cu2Fe(CN) 6 , the former can not be precipitated till enough KiFe(CN)6 has been added to precipitate practically all the copper. 20. What change is observed which shows that the complex copper-ammonia ion is completely decomposed when KCN is added to the solution? 21. CdS is much more soluble in water than Cu 2 S. Why is CdS precipitated by H 2 S from the KCN solution, while Cu 2 S is not? 22. In dissolving the sulfides in P. 31, if a HNOs solution stronger than 2-normal were used, or if the solution were boiled too long, and in consequence some mercury dissolved, how would it behave in the subsequent Procedures? 23. If in evaporating the HNOs solution with H 2 SC>4 in P. 33 the evaporation were stopped before all the HNOs was expelled, what effect would this error have on the tests for lead, bismuth, and cadmium? Experiments 8-9. 1. Write chemical equations expressing the two stages of the hydrolysis of (NB^S. Explain by the ionic theory and the mass-action law why this hydrolysis takes place, taking into account the fact that water is ionized to a slight extent into H + and OH~. QUESTIONS ON THE EXPERIMENTS. 17 2. Write the equations expressing the action of HC1 on a solution of ammonium monosulfide, and on one of ammonium polysulfide (regarding the latter as (NH^Sj). 3. Write equations expressing the behavior of a solution of AsCla when treated in succession with H2S in P. 21, with ammonium polysulfide in P. 22, and with HC1 in P. 23. 4. Explain by the solubility-product principle why the addition of HC1 to a so- lution of (NH^SnSa causes the precipitation of 81182. 5. Make a table showing the solubility of each of the sulfides of the copper and tin groups in both ammonium monosulfide and ammonium polysulfide. Indicate in each case whether the sulfide is readily soluble, slightly soluble, or practically insoluble. 6. Why does not Bi 2 S 3 dissolve in (NH^S, just as Sb 2 S 3 does? Why does not SnS dissolve in ammonium monosulfide as well as in ammonium polysulfide? 7. Why not use in all cases ammonium polysulfide, since this readily dissolves all the tin-group sulfides? Why is the precipitate first treated with a small quantity of it, even tho this makes necessary a second treatment? 8. Why does the tendency of copper to form the complex copper-ammonia ion not cause CuS to dissolve in NEUOH? What must be the explanation of the fact that some CuS dissolves in ammonium polysulfide? Experiment 10. 1. Describe specifically the differences in the behavior of the sulfides of arsenic, antimony, and tin on which the separation of these three ele- ments is based. 2. In treating the sulfides with 12-normal HC1 why does much more As2S5 dis- solve if the solution be allowed to boil? Why does it boil at so low a temperature as 50-60? 3. Write by the method described in Note 7, P. 21, the equations expressing the action of HC1 on KClOs by which C\z is produced and that by which C1C>2 is pro- duced. 4. Explain why the Ck set free by the addition of the KClOs causes the As2Ss to dissolve even in the dilute HC1. 5. What is the expression for the solubility-product in the case of MgNH4As04? Why does it dissolve readily in HC1? 6. Why does the hydrolysis of this salt increase its solubility? Why is that hydrolysis decreased by an excess of NH^OH? How is the hydrolysis affected by the presence of NH 4 C1? Would NH 4 C1 affect the solubility in any other way? 7. What is a saturated solution? a supersaturated one? By what treatments can a precipitate be made to separate from a supersaturated solution? 8. What difference in the ionization causes the behavior of arsenic acid towards E^S to be different from that of other elements of the copper and tin groups? What causes its own behavior to be different in dilute and concentrated HC1 solutions? 9. Write the series of reactions which take place when H^S is passed into a dilute HC1 solution of HgAsOi. State what is known about the rate or equilibrium of each of these reactions. 10. Explain why antimony precipitates on the platinum rather than on the tin, and state how would the result be different if the tin did not touch the platinum? 11. Why does the tin dissolve in HC1 more readily when it is in contact with the platinum? 18 QUESTIONS ON THE EXPERIMENTS. 12. In the confirmatory test for tin, why is lead used instead of zinc? 13. If zinc were used, how would the procedure have to be modified? 14. In the confirmatory test for tin how does the precipitation of a mercury com- pound show the presence of tin? 15. If the antimony were not all precipitated in P. 44 on passing H 2 S into the hot solution, how would it behave in the subsequent procedure (P. 46) in which the tin is precipitated and its presence confirmed? Experiment 12. 1. In an actual analysis how many cubic centimeters of NH 4 OH would be required to neutralize the 5 cc. HNOs that are added before precipitating with H 2 S? 2. How much more NH 4 OH would be needed to neutralize the solution if 500 mg. Cu had been present in the form of Cu(NOs) 2 in the solution precipitated by H 2 S? (In all such calculations of the volume of the reagent needed, first reduce the weight of the constituent from grams to equivalents.) 3. How does testing the vapors above the solution with PbAc2-paper show that an excess of (NH^S, a non-volatil salt, has been added? 4. If in an actual analysis the mixture containing NH4OH and (NH 4 ) 2 S were allowed to absorb CC>2 from the air before filtering, what difference would it make? 5. Why is the (NHi) 2 S precipitate treated first with cold HC1? Why is HNO 3 subsequently added? Experiments 13 and 14. 1. Which elements are soluble, a, in excess of NH4OH (in the presence of NH4C1), but not in excess of NaOH; 6, in excess of NaOH, but not of NH 4 OH (in presence of NH 4 C1); c, in excess both of NBUOH and of NaOH; d, neither in excess of NaOH nor of NH 4 OH (hi presence of NHjCl)? 2. What are the explanations of the four typical cases a, 6, c, d, referred to in the preceding question? 3. Could the hydroxide of an element which does not form a complex ammonia cation be soluble in NHiOH and not in NaOH? Could an amphoteric hydroxide be readily soluble in NaOH and entirely insoluble in NH 4 OH? 4. Show by formulating and combining the two mass-action equations involved that the quantity of aluminum dissolved (as A1O 2 ~) in the presence of a base is pro- portional to the OH~ concentration in the solution. 5. Name all the elements that form ammonia complexes in all the groups thus far considered. What can be said as to the position of these elements in the periodic system? (Refer to a text-book of Inorganic Chemistry.) 6. If in an actual analysis no precipitate is obtained on the addition of NEUOH, what conclusion may be drawn? 7. Which of the hydroxides precipitated by NaOH undergo change on the addi- tion of NaaO 2 , and into what compound is each of these hydroxides converted? 8. What substances are produced by the action of Na^Oa on water? Experiment 15. 1. What does Expt. 15 show in regard to the behavior of the alkaline-earth elements in an actual analysis? 2. What must be the explanation of the fact that the phosphate combines with the iron rather than the calcium when both these elements are present? 3. If phosphate is known to be present, is it necessary to test for alkaline-earth elements in the filtrate from the (NH4) 2 S precipitate? 4. If the original substance were soluble in water, would it be necessary to test for these elements in the (NH 4 ) 2 S precipitate? QUESTIONS ON THE EXPERIMENTS. 19 5. If CaCO 3 were substituted for Ca^(POi) 2 in the first part of Expt. 15, what would the result have been? 6. If a solution of CaaCPO^ in HC1 were treated by the second paragraph of P. 52, what would happen on the addition of each reagent? What would happen, if a large proportion of Feds were also present in the phosphate solution? Experiment 16. 1. Make a table (like that described in Question 1 on Expt. 6) showing the operations involved and the behavior of the chromium and zinc in analyzing a dilute HNOs solution of ZnCrO^ beginning with the K^S precipitation (P. 21) and continuing through the analysis of the aluminum-group (P. 51-57). At the foot of the table write all the chemical equations involved. 2. In separating the aluminum from the chromium and zinc with NRiOH in P. 53, what would be the harm of adding too small an excess? what of adding too large an excess? 3. By what reagent other than BaC^ could the chromate be precipitated? What disadvantage would there be in the use of this reagent? 4. How can sulfate be present in the solution to which BaCl 2 is added? 5. How could a precipitate of ZnS be distinguished from one of sulfur by adding a suitable reagent to the liquid containing it? 6. What happens to nitrates, such as Zn(NC>3) 2 or Co(NOs)2, when they are ignited, as in the confirmatory test for zinc? Experiment 17. 1. What are the oxides of manganese corresponding to its three stages of oxidation occurring in P. 61 and 62? What is the valence of manganese in each of these oxides? How do they differ with respect to the formation of salts with acids and with bases? 2. Make a table (like that described in Question 1 on Expt. 6) showing the operations involved and the behavior of manganese in analyzing a dilute HNOs solution of CaMnO* beginning with the H 2 S precipitation and continuing through the final test for manganese (thus involving P. 21, 51, 52, 61, and 62). Write also all the chemical equations involved. 3. Why is a considerable excess of NH4OH added in precipitating the iron in P. 64? 4. If FeS were treated by P. 31-38, how would it behave with each of the reagents? 5. Why is it necessary to test for zinc in the analysis of the iron-group? 6. Why may zinc be precipitated by NaOH and Na2O 2 in the first treatment (in P. 52), and yet not be precipitated by them in the second treatment (in P. 67)? 7. When the original Na 2 O2 precipitate is so small that it need not be tested for zinc, how may P. 67-68 be simplified? 8. When the H2 precipitate obtained in P. 66 is small, how may P. 67-68 be simplified? Experiment 18. 1. If the H 2 S and NaA precipitates were not washed free from ammonium and sodium salts before dissolving them in P. 68, what might happen upon the addition of the HCl-ether reagent? 2. Suggest an explanation of the fact that, tho NiCl 2 is precipitated as a yellow compound by the HCl-ether reagent, it has a green color even in very concentrated aqueous solution. 3. Suggest any possible differences in the molecular state of CoCk, which might account for the facts that it has a pink color, like that of Co(NO3) 2 , in a dilute aqueous solution, and a blue color in an ether solution or a concentrated aqueous solution saturated with HC1. 20 QUESTIONS ON THE EXPERIMENTS. 4. Name all the elements thus far considered which in any state of oxidation form colored compounds in solution. What can be said as to the position of these elements in the periodic system? 5. Write equations expressing the steps in the chemical process by which the pre- cipitate of potassium cobaltinitrite may be considered to be formed. 6. Does the N(V coming from the excess of KNO 2 diminish the solubility of the precipitate in the same way as the K + does? What else, from a mass-action stand- point, might it be expected to do? Experiment 19. 1. Why must the HNOs be removed before testing for iron with KSCN? 2. Make a table showing the operations involved and the behavior of Ca 3 (PO4)2 in analyzing a HNO 3 solution of it by P. 51-52, P. 61, and P. 65. 3. Why does the NH4Ac solution become red only when the quantity of iron is more than equivalent to the quantity of phosphate present? 4. What is meant by a basic salt? What are two possible simple formulas for basic ferric acetate? Experiment 21. 1. What does this experiment show as to the precipitation by (NH^COs of magnesium and of the other alkaline-earth elements (which all behave nearly as calcium does)? 2. Why would a reagent which consisted of NH^COs not be suitable for the precipitation? 3. Why is there any advantage in adding more NHs than corresponds to the neu- tral salt (NH 4 ) 2 CO 3 ? 4. If it were desired to work out a procedure for separating calcium, barium, and strontium from magnesium by means of (NH ^COs, what experiments would one naturally make? Experiment 22. 1. In order to make a separation of 1 mg. barium from 500 mg. strontium, what must be the concentration of CrO4~, stated with reference to the saturation-values of the ion-concentration products of BaCrO4 and SrCrO4? 2. What must be true of the relativ values of these two saturation-values in order that this separation may be possible? What is the actual ratio of these two values? (See the Table of Solubilities on page 124.) 3. Write the chemical equations for the conversion of chromate-iron into hydro- chromate-ion, and for the conversion of the latter into bichromate-ion. Write also the mass-action expressions for the equilibrium of these reactions. Show by them what determins the proportion of CrC>4 = and of HCrCU" in any solution, and what determins the concentration of C^O?" in any solution. 4. In practis what three substances must be added in proper proportions in order to secure the right CrC>4~ concentration in the solution? 5. Why is the second K2CrO4 precipitate obtained in the confirmatory test for barium more conclusiv evidence of its presence than the first K2CrO4 precipitate? 6. On addition of NEUOH in P. 84 what chemical change causes the change in color from orange to yellow? Why from a mass-action standpoint does the addition of NEUOH cause this change to take place? Why does this change cause strontium to precipitate? 7. Explain fully with reference to the saturation-values of the ion-concentration products why the carbonate-chromate mixture used in P. 85 does not affect BaCrC>4, and why it converts SrCrO4 into SrCO 3 . QUESTIONS ON THE EXPERIMENTS. 21 8. Why does CaQjCU dissolve in dilute H 2 SO 4 , but not in HAc? 9. How does the confirmatory test for calcium distinguish it from barium and strontium, which form much less soluble sulfates? How does it distinguish calcium from magnesium? 10. Could magnesium be precipitated by any other reagent in the form of a com- pound closely analogous to magnesium ammonium phosphate? 11. Why is the production of a precipitate with NaaHPOi in the confirmatory test for magnesium (in P. 89) more conclusiv evidence of its presence than the production of the first Na^HPC^ precipitate (in P. 88)? Experiment 23. 1. If the ammonium salt were not completely removed by the ignition, how would it behave in the subsequent tests for potassium? 2. Why is the separation of potassium and sodium by the HC1O4 method satis- factory when these elements are present as chlorides or nitrates, but not when they are present as sulfates? Why is it satisfactory when they are present as phos- phates? 3. What is implied by the statement that Na3Co(NO 2 )e is a complex salt in solution? 4. How might a solution of NaaCotNQOe be prepared, judging from previous expe- rience with an analogous compound? Experiments 26 and 27. 1. Describe in detail the method by which it can be shown that the magnetic oxide of iron (FesOJ contains both ferrous and ferric iron. 2. Describe a method by which it can be shown that a solution contains both mercurous and mercuric salts. 3. What indications of the state of oxidation of tin would be afforded by the treatment with H 2 S in P. 21, and with (NH 4 ) 2 S in P. 22, of a solution of SnCl 2 ? of solution of SnCl 4 ? 4. What other constituents besides iron, mercury, and tin exist in two stages of oxidation in the form of salts? Giv the symbols of compounds illustrating these two states of oxidation. 5. Mention any phenomena that might be observed during the course of analysis which would distinguish from each other the two stages of oxidation of each of the constituents named in the answer to Question 4. Experiments 28 and 29. 1. What indication of the state of oxidation of arsenic may be obtained in the course of the analysis for the basic constituents? 2. What do the facts that concentrated HC1 converts much H 3 AsO3, but very little HsAsO 4 , into the corresponding chloride show as to the ionization of these two substances? 3. What reaction occurs when the CuSO 4 solution is poured into the hydrogen generator in P. 99? How does the addition of the CuSO 4 accelerate the evolution of hydrogen? 22 QUESTIONS ON THE EXPERIMENTS. DETECTION OF THE ACIDIC CONSTITUENTS. Experiment 31. 1. With the aid of the laboratory experience and the statement as to solubilities on page 31 arrange the barium salts of the acidic constituents listed on page 93 in four groups comprising respectivly: those readily soluble in water; those only slightly soluble in water, but readily soluble in HAc; those only slightly soluble in HAc, but readily soluble in dilute HC1; and those only slightly soluble in dilute HC1. 2. Explain the following facts with reference to the ionization-values given in the Table on page 123: a, a precipitate of BasfPO^ is converted into one of BaHPO4 by a nearly equivalent quantity of HAc; 6, BaHPO4 is readily soluble in an excess of HAc with formation of Ba(H2PC>4)2 and BaAc2; c, BaHPCU is readily soluble in dilute HC1 with formation of H 3 PC>4 and BaCl2,' d, BaSO 3 is not much more soluble in HAc than in water, but is readily soluble in dilute HC1; e, BaCrO4 is not readily soluble in HAc, even tho HCrO4~ is very slightly ionized. 3. State how fluoride when present in large quantity behaves in each step of P. 102; and explain this behavior with reference to the ionization of HF and the solu- bility-values given in the table on page 124. 4. What other acidic constituent besides fluoride is shown by the solubility-table on page 124 to form a calcium salt less soluble than the barium salt? Show by reference to the solubility and ioniaation values involved how this constituent (in comparison with fluoride) might be expected to behave in each step of P. 102. 5. Suggest a reason why chromate, tho included in P. 102, is not added to the mix- ture (of Na2S04, Na2SO 3 , and KF) used in Expt. 31 in illustrating that Procedure. Experiment 32. 1. What disadvantage would there be in substituting Pb(NO 3 )2 for Cd(NO 3 ) 2 in the test for sulfide in P. 103? 2. Arrange the silver salts of the acidic constituents listed on page 93 in three groups comprising respectivly: those readily soluble in water; those only slightly soluble in water, but readily soluble in dilute HNO 3 ; and those only slightly soluble in dilute HNO 3 . 3. Explain by reference to the solubilities of the salts (as given in the Table on page 124) and the ionizations of the corresponding acids why each of the salts in the second group referred to in the previous question dissolves in dilute HNO 3 . 4. Explain why the silver halides do not dissolve in dilute HNO 3 . 5. Explain why Ag2S is only slightly soluble in dilute HNO 3 , even tho the ionization of HS~ is extremely small. 6. Explain why Ag2(CN)2 is only slightly soluble in dilute HNO 3 , even tho HCN is only very slightly ionized. 7. Why would it not be satisfactory to precipitate (and thus partially identify by their color) the silver salts which are soluble in dilute HNO 3 , but not in water, by adding to the filtrate from the silver halides an excess of NaOH? Why not by adding an excess of NH 4 OH? Experiment 35. 1. State the principles involved in the process used for the detec- tion of the three halides in the presence of each other. 2. Does the fact that only a very small quantity of bromin is liberated in the first step of the process mean that the reaction between the bromide and KMnC>4 is in equilibrium under the conditions prevailing in the solution? 3. What is meant by the statement that a reaction is in equilibrium? How would one proceed to determin whether a given reaction-mixture is in equilibrium? QUESTIONS ON THE EXPERIMENTS. 23 4. If the mixture of KBr, NaAc, HAc, and KMn04 were allowed to stand a day or a week, what would happen? 5. Why must the iodin set free in the first part of the process be completely extracted by the chloroform? 6. Why is H^SOs added in the last part of the process? Write the equation for the reaction which it causes. 7. Why is HNO 3 added with the AgNO 3 at the end of the process? Experiment 37. 1. What two things detennin whether or not an acid passes over into the first distillate? 2. Show by reference to the mass-action expressions for the ionization of the two acids what deterrnins the extent to which an acid is displaced from its salt by another acid, taking K + CN~ and H + H 2 PO4~ as an example. 3. Explain by reference to the ionization- values involved why BaCOs dissolves on adding HAc, and why BaSOs does not. 4. Show how phosphoric, pyrophosphoric, and metaphosphoric acids are related to one another in composition. 5. If a compound of an element forming an insoluble phosphate (for example, CaCOs) were distilled, with HsPO^ would the insoluble phosphate separate in the distilling flask? Explain why or why not. 6. If E^SOs is found in the first distillate, which of the other substances that may be in that distillate is it unnecessary to test for? Write the equation for the reac- tion which would take place between H^SOs and each of these substances. 7. If any of the HsPC^ were thrown over mechanically into the first distillate, what erroneous conclusion might be drawn in connection with P. 111? Experiment 38. 1. What different constituents of the original substance may giv rise to sulfur in the distillate? what ones to sulfurous acid? 2. Show that HsPC^ which had been thrown over mechanically into the first dis- tillate would not interfere with the tests for sulfite and carbonate in P. 112. Experiment 39. 1. What different constituents in the original substance may giv rise to chlorin in the first distillate? Write chemical equations illustrating its production from each of these constituents. 2. What different substances that might be present in the distillate would cause iodin to be set free on the addition of KI? 3. What conclusions could be drawn from the tests of the first two paragraphs of P. 113 as to the presence or absence of each of the three halogens in a distillate containing the following halogens: a, 12, Br2, and Cfe; b, Br2j c, Ck; d, no halogen? 4. Explain the fact that iodin is extracted by chloroform more slowly from an aqueous solution when it contains iodide. State the law that determins the quan- tity extracted. 5. Write chemical equations illustrating the process by which a small quantity of HNO2 liberates a large quantity of iodin from KI. Experiment 40. 1. Write the chemical equations involved in the test for cyanide. 2. Into what compounds is K4Fe(CN)e decomposed when it is distilled with HaPC^? 3. Referring to the results of Expt. 26 with ferrous and ferric salts, suggest how a ferrocyanide could be distinguished from a ferricyanide by tests applied to a solu- tion of the original substance. Experiment 41. 1. To what extent is the analysis of the second distillate sim- plified when AgNOa givs no precipitate? 24 QUESTIONS ON THE EXPERIMENTS. 2. Some H 3 PC>4 may pass over into the second distillate. Would Ag 3 PO4, which is only slightly soluble in water, precipitate in the AgNO 3 test, and thus obscure the test for the other constituents? 3. When is it necessary to test the second distillate for sulfide? 4. Explain by the mass-action principles why a sulfide may giv off an appreciable quantity of H 2 S only in the second part of the distillation. 5. What constituents in the substance are likely to giv rise to chlorin in the second distillate? to bromin? to iodin? Experiment 42. 1. Why not test for boric acid in the distillates obtained in the H 3 PO 4 distillation? 2. What is the advantage in P. 121 of distilling the borate with H 2 SC>4 and CHaOH, rather than with H 2 S0 4 with water? Experiment 43. 1. Why must all the substances used in the test for fluoride be thoroly dry? 2. What happens to KHS0 4 when it is heated? 3. What are the main constituents of glass? What is the action of HF on it? 4. Since H 2 Si0 3 is non-volatil, how can the deposit be driven up the tube by heating? 5. Of what does the white deposit left after washing the tube with water consist? 6. Show why it is appropriate to call the compound H 2 SiFe "fluo-silicic " acid. 7. Why is it not satisfactory to test for fluoride in the distillates obtained in the HsP0 4 distillation? Experiments 44 and 45. 1. If chlorate were present in the substance, what would happen to it when treated by P. 124? 2. Iodin is not completely reduced to HI by FeS0 4 , the four substances I 2 , HI, ferrous salt, and ferric salt all being present in considerable quantity at equilibrium. If a solution containing ferric sulfate and KI were submitted to P. 124, what would be the result? 3. How might the distillation-process of P. 124 be modified so as to detect in two steps, first, the presence of nitrite, and second, when it is absent, the presence of nitrate? Experiment 46. 1. What different substances are present in the solution pro- duced by acidifying the NaOCl solution with HAc? 2. What advantage would there be in making the hypochlorite test in alkaline solution, rather than in HAc solution? What disadvantage? 3. Assuming that the oxidation of the PbAc2 to PbO 2 is caused by HOC1, but not by Ck, explain why the oxidation does not take place in a solution strongly acidified with HNO 3 . Experiment 47. 1. How may a chlorate giv off chlorin in the first part of the HsPO 4 distillation? How in the second part? 2. Write the equation expressing the reaction that takes place between NaOCl and NaAsO 2 . 3. If the hypochlorite were not reduced, AgCl and AgClO 3 would be formed in the HNO 3 solution by the action of the C1 2 on the AgN0 3 . Write the equation ex- pressing this reaction. QUESTIONS ON THE EXPERIMENTS. 25 PREPARATION OF THE SOLUTION. Experiment 48. 1. Of what elements is organic matter mainly composed, and what causes it to blacken on heating? 2. Describe a method by which a substance might be tested for organic matter by converting the carbon into carbon dioxide. 3. Could a sirup be tested for aluminum by diluting it with water and analyzing the solution in the usual way? Could it be so tested for iron? 4. Name the different forms in which water may be present in a substance. 5. If a substance were ignited at a red heat (for example, to destroy organic matter) before submitting it to analysis, what basic constituents would be lost? 6. State how each of the following substances would behave in the closed-tube test: a, Na 3 AsO 4 ; b, MgNH 4 PO 4 ; c, FeS 2 ; d, Pb(NO 3 ) 2 ; e, NaC 2 H 3 O2. 7. What acidic constituents will be detected, when present in considerable quan- tity, upon treating the substance with dilute HNO 3 ? 8. What is the behavior towards litmus of an aqueous solution of each of the following substances: Na 2 SO 4 ; Na^COs; NaHCO 3 ; NaHSO 4 ; Ca(NO 3 ) 2 ; Fe(NO 3 ) 3 ? 9. Name the basic constituents whose salts are nearly all readily soluble in water. 10. Name the basic constituents whose salts are decomposed by water, but dis- solved by dilute acids. 11. Name the acidic constituents whose salts are all or nearly all readily soluble in water. 12. Name the acidic constituents whose salts (except those of the alkali elements) are nearly all very slightly soluble in water. 13. Of the groups of salts named in the answer to Question 12, which are readily soluble in cold dilute HNO 3 or HC1? Why? 14. In a substance soluble in water which has been found to contain barium, which of the following constituents would it be unnecessary to test for: nitrate, phosphate, sulfide, sulfite, chloride, sulfate, carbonate? Experiments 49 and 50. 1. What group of salts not dissolved by cold dilute HNO 3 are decomposed by hot concentrated HNO 3 , because of its oxidizing action? what groups because of its action as a strong acid? 2. State what happens (that is, what chemical changes occur and what phenomena are observed) at each step of the process when PbSiO 3 (which is decomposed by strong acids) is treated by P. 3. 3. State what happens at each step, namely, on heating with strong HNO 3 , on evaporating to dryness and adding cold dilute HNO 3 , and on heating the residue with strong HC1, when each of the following substances is treated by P. 3 and 4: a, Sb 2 S 3 ; 6, MnO 2 ; c, PbSO 4 . 4. What metallic elements are scarcely ever found in alloys? How may the process of analysis therefore be shortened? 5. If there is no residue after treating an alloy by P. 3, what does it show as to the absence of certain elements? Experiment 51. 1. State what happens to each of the substances whose sym- bols are given under b in Table I on fusing it with Na 2 CO 3 , on treating the fused mass with water, and on treating the residue with dilute HNO 3 , as described in P. 7. 2. What are the only acidic constituents that it is ordinarily necessary to test for in minerals? 26 QUESTIONS ON THE EXPERIMENTS. 3. State what happens to each of the substances whose symbols are given under 6 in Table I when it is treated, as in P. 5: (1) with concentrated H^SO^* (2) then with HF; (3) then, after evaporating to fuming, with water. 4. State what happens to feldspar (potassium aluminum silicate, an example of a silicate not much decomposed by acids other than HF) when it is treated by P. 5. 5. Explain with reference to the solubilities of the substances involved why PbS(>4 is completely converted into PbCOs by boiling with Na^COa solution in P. 6; and why BaSC>4, when much of it is present, is only partly converted into BaCOs by the same treatment. PART II. THE SYSTEM OF ANALYSIS. PREPARATION OF THE SOLUTION. PRELIMINARY EXAMINATION. Procedure i. Preliminary Examination. If the substance is a non-metallic solid, note its color, odor, and texture; examin it with a lens to determin whether it is heterogeneous, and, if so, note the appear- ance of its constituents. To determin whether organic matter or water is present and to get other indications, heat gently at first, then strongly, about 0.1 g. (0.1 gram) of the finely powdered substance in a hard glass tube (of about 0.6 cm. bore and 10 cm. length) closed at one end. Note whether the substance blackens, whether a tarry, aqueous, or other deposit forms on the cold part of the tube, and whether any odor is emitted. If organic matter is thus proved to be absent, pass to P. 2 (Procedure 2) ; if proved to be present, to P. 8. If the substance is an alloy, treat it by P. 3. If the substance is a solution, evaporate a measured volume of it to dryness in a small weighed dish, dry the residue thoroly at 120- 130 in a hot closet or by keeping the dish in motion over a small flame, and weigh the dish again. Heat a portion of this residue in a closed tube as described above. Treat another portion by P. 2 if organic matter is absent, or by P. 8 if organic matter is present. Notes. 1. When a complete analysis in the wet way is to be made, it is usually not worth while to make a more extended preliminary examination in the dry way. The closed-tube test is, however, essential, in order to show whether organic matter is present; for certain kinds of organic matter, especially sugars and hydroxy-acids, such as tartaric, citric, and lactic acids, prevent the precipitation of the hydroxides of most of the elements by alkalies. Such organic matter must therefore be detected and removed in order to ensure the precipitation of aluminum and chromium by NEUOH. Moreover, a large quantity of organic matter of any kind interferes with the execution of the analysis; for example, with the operations of solution, filtration, and evapora- tion. Alloys do not contain organic matter or water; and therefore the closed- tube test need not be applied to them. 27 28 PREPARATION OF THE SOLUTION. P.I 2. Blackening accompanied by a burnt odor or by the formation of a tarry deposit shows organic matter. Blackening alone does not show it; for copper, cobalt, and nickel salts may turn black on heating, owing to the formation of the black oxides. 3. It is usually desirable to determin whether water is a constituent of the substance, and, if so, whether it is present in large or small proportion. This can be done with a fair degree of delicacy by the closed-tube test, provided care be taken to keep the upper part of the tube cool during the first of the heating. Water may be present as so-called water of constitution, as in FeOsHs or NagHPO^ as water of crystallization, as in MgSOi.THsO; as enclosed water, as in some hydrated silicates like the zeolites or as mother-liquor within crystals; and as hygroscopic moisture on the surface. Water of constitution may be expelled only at a fairly high temperature, while in the other forms it is seldom retained above 200. 4. The closed-tube test may also furnish evidence of the presence of certain basic and acidic constituents when they are present in considerable quantity. Thus all ammonium salts and mercury compounds are volatilized much below a red heat. Ammonium salts and the chlorides of mercury giv a white subli- mate. Most other mercury compounds giv a gray one, consisting of minute globules of mercury, made visible by a lens or by rubbing with a wire. Metallic As, As2O3, and As2Ss are also readily volatilized, forming black, white, and yellow sublimates, respectivly. Of the acid-forming elements or groups, free sulfur or a persulfide is shown by a sublimate of reddish-brown drops, changing to a yellow solid on cooling, and accompanied by odor of SO2J a moist sulfide, by the odor of [28; a nitrate or nitrite, by brown vapors of NO2J free iodin or a decom- posable iodide, by a black sublimate of Iz and by its violet vapor; a sulfite, by the odor of SOz', a peroxide, chlorate, or nitrate, by evolution of oxygen, recognized by its inflaming a glowing wood-splinter held in the tube; and a carbonate or oxalate, by the evolution of CO2, recognized by its causing tur- bidity in a drop of Ba(OH)2 solution. 5. If the substance to be analyzed is a liquid, it is desirable to determin by evaporation how much, if any, solid substance is present in it; for enough must be taken for analysis to enable small quantities of the basic constituents to be detected. Moreover, if it is dissolved in a volatil organic solvent the latter must be removed by evaporation. P.S PREPARATION OF THE SOLUTION. 29 PREPARATION OF THE SOLUTION. TABLE I. PREPARATION OF THE SOLUTION IN THE CASE OP NON-METALLIC SUBSTANCES. Heat the substance with water and dilute HNOz (P. 2). If it all dissolves, treat the If it does not all dissolve, add more HNOz, evaporate, dry completely, add dilute HN0 3 (P. 5). solution by P. 11. Substances decom- posed: many sul- fides and silicates. Treat the solution by P. 11. Residue:* a. Sb 2 O 6 , H 2 SnO 3 , MnQj, PbO 2 , HgS, PbCr0 4 , BaCrO 4 . b. C, A1 2 O 3 , Cr 2 O 3 , AgCl, CaF 2 , PbS0 4 , BaSO 4 , SrSO 4 , SiO 2 , and many silicates and fluosilicates. Heat with HCl or HCl and HNOs, evaporate, add dilute HCl (P. 4). Solution: substances under a. Treat by P. 21. Residue: substances under b. Heat with # 2 4 and HF, evaporate off the HF, add water, boil (P. 5). Gas: SiF 4 . Residue: Pb, Ba, Sr,(Cr), as sulfates. Treat by P. 6. Solution : other elements as sulfates. Treat by P. 11. * Only the more common substances that are likely to be present in the residue are here mentioned. Procedure 2. Treatment of Non-Metallic Substances Free from Organic Matter. Weigh out on a rough balance 1 g. of the finely powdered substance (see Note 1), add to it in a casserole 20 cc. water, heat the mixture to boiling if there is a residue, and test the solution with litmus paper. Add to the mixture 6-normal HN0 3 , a few drops at a time, till after shaking it becomes distinctly acid. Note whether there is an odor or effervescence. Then add, without filtering out any residue and without further heating, just 5 cc. 6-normal HN0 3 . If the substance has dissolved completely, treat the solution by P. 11. If the substance has not dissolved completely, treat the mixture, without filtering out the residue, by P. 3. (See Note 8.) Notes. 1. In order that difficultly soluble substances may be dissolved, the substance must be reduced to a very fine powder. This is usually best accomplished by grinding the substance, a small quantity at a time, in a por- celain or agate mortar. With hard substances, and in general with minerals, an 30 PREPARATION OF THE SOLUTION. P. 2 agate mortar should be used. As such a mortar is likely to be broken by a blow, the substance should be ground, not pounded, in it. 2. The quantity of the substance taken for analysis should always be approximately known; for a good qualitativ analysis should not only show the presence or absence of the various elements in the substance, but should enable their relativ quantities to be estimated. Since 1 or 2 mg. of almost any element can be detected by this system of analysis, the presence of 0.1-0.2% of an element will be detected when one gram of substance is taken, and this degree of delicacy is ordinarily sufficient. If much more than this quantity is taken, the precipitates may be so large that much time is consumed in fil- tering and washing them. Moreover, the directions given for some of the separations are based on the assumption that not more than 500 mg. of any one constituent is present. 3. When the substance dissolves only partly in water, it is not worth while to filter off the residue and analyze it and the solution separately, unless special information in regard to the soluble constituents is desired. It is, therefore, directed to treat at once with HNOs. The mixture is not heated after addition of the acid, so as to avoid oxidizing mercurous, arsenous, and ferrous salts. Only 20 cc. of water are used so that the acid may be strong enough to prevent the hydrolysis of salts of bismuth and tin, and thus ensure their solution. 4. Just 5 cc. 6-normal HNOs must be present in order that the acid con- centration may be properly adjusted in the subsequent H 2 S precipitation. For this reason, when the solution is alkaline or when a substance (like an undissolved oxide or carbonate) which neutralizes the acid is present, the solution is made distinctly acid before adding the 5 cc. of HNOs. 5. If the aqueous solution has an alkaline reaction, the addition of an acid may cause precipitation of any substance held in solution by an alkaline solvent; for example, sulfur or sulfides of the tin-group from an alkaline sulfide solution; silver chloride or cyanide from a potassium cyanide solution; silicic acid from sodium silicate solution; or basic hydroxides from solutions in alkalies. These last substances redissolve when the excess of HNOs is added. 6. An acid reaction of the aqueous solution towards litmus is due to hy- drogen-ion, which may arise from free acid, from an acid salt of a strong acid, or (by hydrolysis) from a neutral salt of a strong acid and a weak base. An alkaline reaction is due to hydroxide-ion, which may arise from a soluble hy- droxide, or (by hydrolysis) from a carbonate, sulfide, phosphate, borate, cyanide, or a salt of some other weak acid. 7. When the acid is added to the aqueous solution, the evolution of any gas and its odor should be noted, since this indicates the nature of the acidic constituents present. Thus carbonates effervesce with evolution of CO2,* sulfides produce the odor of H 2 S; sulfites and thiosulfates, that of S0 2 ; and cyanides, that of HCN. 8. There are certain compounds (especially those of antimony and tin, the oxides of iron, MnO 2 , and BaCrO 4 ) which dissolve more rapidly or in larger quantity in HC1 than in HNO 3 . When a substance fails to dissolve completely in the dilute HNO 8 and seems likely from its appearance or behavior to contain any of these compounds, it is therefore well to attempt to prepare an HC1 solu- tion of it by proceeding as follows: To a fresh 1 g. sample of the finely pow- dered substance in a small flask add just 5 cc. 6-normal HC1; heat the mixture P. t PREPARATION OF THE SOLUTION. 31 nearly to boiling for 3-4 minutes, covering the flask with a watch-glass to pre- vent evaporation; and then add 10 cc. water. If the substance has dissolved completely, reject the HNO 3 solution and residue, and treat the HC1 solution by P. 21. Otherwise, reject the HC1 solution and residue, and treat the HNO 8 solution and residue by P. 3. As to the reasons for recommending the use of HNO 3 , rather than of HC1, as the usual procedure, see Note 4, P. 3. 9. The following general statements may be made in regard to the solubility of substances in water and dilute acids: All the ordinary salts of sodium, potassium, and ammonium are readily soluble in water. The salts of mercurous and mercuric mercury, bismuth, antimony, and tin are hydrolyzed by water with precipitation of basic salts, which dissolve readily in dilute HNO 3 or HC1. The nitrates, nitrites, chlorates, and acetates of all the elements are readily soluble in water (except certain basic nitrates and acetates). The hydroxides, carbonates, phosphates, borates, arsenates, and arsenites of all the elements except the alkalies are only slightly soluble in water, but dis- solve readily in dilute HNO 3 or HC1. (Ba(OH) 2 , Sr(OH) 2 , and Ca(OH) 2 are, however, fairly soluble in water.) The chlorides, bromides, iodides, and thiocyanates of all the elements except lead, silver, and mercury, and the sulfates of all the elements except calcium, strontium, barium, lead, mercury, and silver are readily soluble in water. (In regard to the solubility of these and other salts of the alkaline earths and of silver and lead, see the Table of Solubilities of Slightly Soluble Substances on page 124.) Procedure 3. Treatment of Non-Metallic Substances not dissolved by Dilute Nitric Add and of Alloys. If the substance is non-metallic and has not dissolved in dilute HN0 3 , to the mixture obtained in P. 2 add 5 cc. 16-normal HN0 3 , and evaporate just to dryness. If the substance is an alloy, convert it into a form offering a large surface and treat 0.5 g. of it in a casserole with 10 cc. 6-normal HN0 3 . Cover the dish with a watch-glass, heat the mixture nearly to boiling as long as any action continues, adding a little 16-normal HNOi if action is renewed thereby, or a little water if crystalline salts have separated, and then evaporate just to dryness. Heat the residue obtained in either case at 100-130 until it is perfectly dry, by keeping the casserole in motion over a small flame. Loosen the residue from the dish and rub it to a fine powder with a pestle; add to it just 5 cc. 6-normal HN0 3 , cover the dish, and warm the mixture, taking care that none of the acid evaporates. Dilute with 20 cc. water, heat to boiling, and note whether there is any residue (see Note 7). If there is a residue, filter it out and wash it. (Residue, P. 4; solution, P. 11.) Notes. 1. On heating the HNOs solution, the presence of sulfides is indi- cated by the separation of sulfur as a spongy or pasty mass, which floats on 32 PREPARATION OF THE SOLUTION. P. 3 the surface and may be removed by means of a spatula or rod; and the presence of iodides is shown by the liberation of free iodin, which may separate as a black precipitate, which imparts a brown color to the solution, and which gives rise to violet vapors above it. 2. When a silicate is decomposed by acid, silicic acid may separate as a gelatinous precipitate, but even then a part of it always remains in solution, mainly as a colloid. When thoroly dried at 100-130, it is partially dehydrated and becomes entirely insoluble. The HNOs acid solution is therefore evapo- rated to dryness and the residue is heated at 100-130, in order to remove the silica at this point; for, if it were not removed, it would appear as a gelatinous precipitate at some later stage of the analysis; thus, if it did not separate earlier, it would be precipitated by NH 4 OH together with the iron group and might then be mistaken for aluminum hydroxide. In the case of nonmetallic substances which cannot contain silica, the heating may be omitted. 3. If the substance is nonmetallic, the residue insoluble in HNOs probably consists of one or more of the substances whose formulas are given in Table I under a and 6. Other less common insoluble substances are anhydrous chro- mium salts, phosphate of tin, ferrocyanide of iron, and silicon carbide. 4. In dissolving nonmetallic substances HC1 may be used in place of HNOs. Each of these acids has advantages and disadvantages of its own, as follows: HNO 3 dissolves, owing to its oxidizing power, many sulfides not attacked by HC1, but fails to dissolve certain substances, especially MnO2, Sb2O5, ^SnOs, and BaCrO 4 , which dissolve in HC1. HC1 may cause the precipitation of chlorides of the silver group; while strong HNOs on heating oxidizes sulfides partially to sulfates, and may cause the precipitation of lead, barium, strontium, and calcium sulfates; thus in either case making it sometimes impossible to deter- min whether complete decomposition has resulted. HNOs oxidizes mercurous, arsenous, antimonous, stannous, and ferrous compounds to the higher state of oxidation; consequently almost all the antimony and tin will usually be found in the residue insoluble in dilute HNOs after evaporation, all the mercury will be in the H 2 S precipitate, and sulfur will always be precipitated by H2S when iron is present. When HC1 is used as a solvent, mercury and arsenic in the arsenous form would be wholly or partly lost, owing to the volatility of their chlorides, in the subsequent evaporation, which is necessary in order to remove silica. For this last reason, and for the reason that the procedure is a more genera 1 one in that it provides for the solution of alloys and of a larger proportion of nonmetallic substances and for the isolation of the silver group, the use of HNOs is here recommended. 5. Alloys can not ordinarily be powdered by grinding in a porcelain or agate mortar. They may usually be converted into a form that offers a large surface by hammering in a steel mortar, filing with fine steel file, shaving with a knife, or converting into turnings with a lathe. Only 0.5 g. of an alloy is taken for analysis; for, owing to the absence of acidic constituents, the same quantity of basic elements is contained in a smaller amount of substance. 6. By the treatment of alloys with strong HNOs, all the more common elements are dissolved by strong HNOs except antimony, tin, and silicon. These are oxidized to antimonic acid (Sb2O5.nH 2 0),metastannic acid (n^SnOg), and silicic acid (HzSiOs), which separate at once as white amorphous pre- cipitates when considerable amounts of these elements are present. Certain P. 3 PREPARATION OF THE SOLUTION. 33 nitrates, especially that of lead, may separate in crystalline form from the strong HNOs, but these dissolve upon adding water and heating to boiling. 7. In the case of an alloy the evaporation to dryness and heating at 100- 130 serve to partially dehydrate the hydroxides of silicon, tin, and antimony, whereby they are rendered nearly insoluble in HNOs. This makes possible a conclusion in regard to their presence or absence. Thus, if after having thoroly dried the mixture at this temperature there is no residue insoluble in the HNOs, it shows the absence of silicon and tin in quantity as large as 1 mg., and that of antimony in quantity as large as 2 or 3 mg. The fact must not be overlooked, however, that in the dehydrated form even a very small residue or slight turbidity may correspond to an appreciable quantity of one of these elements. Therefore, if no residue can be seen, rub the sides of the dish gently with the rubber-covered end of a glass rod, pour into a small flask, allow the liquid to stand 2 or 3 minutes, and note whether there is any residue whatever. The knowledge that tin is absent enables the subsequent procedures for the detection of this element to be omitted. The subsequent procedures for antimony may, in the absence of a residue, also be omitted, provided quantities as small as 3 mg. are not to be tested for. In addition to the hy- droxides named above, the residue may also contain a considerable quantity of stannic phosphate or arsenate when tin and pho phorus or arsenic are simul- taneously present, or of bismuth hydroxide when both antimony and bismuth are present; also small quantities of various other elements enclosed in a residue consisting of the substances already mentioned. 8. The hydroxides of antimony, tin, and silicon usually separate also in the treatment of nonmetallic substances with HNOs when the corresponding elements are present; but the nonexistence of a residue must not, except in the case of silicon, be regarded as conclusiv evidence of their absence in such substances. For the presence of certain acidic constituents, such as chloride or sulfate, may cause a considerable quantity of tin or antimony to dissolve. 9. A black or metallic residue insoluble in HNOs, obtained in the case of an alloy, may contain carbon or carbides, certain alloys of iron, such as ferro- chrome or ferrosilicon, gold, or any of the platinum metals. If there is no such residue, it shows the absence of gold and platinum. Procedure 4. Treatment of the Residue Insoluble in Nitric Acid. To the residue insoluble in HNOs (P. 3) in a casserole add gradually 5 cc. 12-normal HC1, and heat as long as action continues, adding more acid if necessary. If the substance does not dissolve completely in HC1, add to the mixture without filtering a few drops of 16-normal HN0 3 , and heat gently as long as action continues, adding more of the acids if necessary. Evaporate this solution in HC1 alone, or in HC1 and HN0 3 , without filtering off any residue, just to dryness, taking care not to overheat the dry residue. Add to the residue 5 cc. 6-normal HC1, measured in a small graduate, and 10 cc. water; boil gently for a few minutes if there is a residue: filter, wash the residue thoroly with boiling water, and treat it by P. 5 (or P. 7). Unite the filtrate with the 34 PREPARATION OF THE SOLUTION. P. 4 HNOs solution obtained in P. 3 after treating the latter by P. 11; neutralize half the acid in the mixture by adding just 5 cc. 6-normal NH40H; and, without filtering off any precipitate, treat the mixture by P. 21. Notes. 1. Of the substances that may be present in the residue undis- solved by HNOs (see Table I), the peroxides of manganese and lead and the chromates of barium and lead are reduced and dissolved by concentrated HC1. Antimonic acid (HSbOs) and stannic acid (H^SnOs) are also dissolved by it. Upon the addition of HNOs, whereby the strongly oxidizing mixture known as aqua regia is produced, gold, platinum, and mercuric sulfide are entirely dissolved; and silver compounds, such as AgBr, Agl, and AgCN, are converted into AgCl. The chloride of silver and the sulfates of strontium and lead dissolve in large quantity in the concentrated acids, but only in much smaller quantity in the small amount of dilute HC1 added after the evaporation. Some of the other substances that may be in the residue, especially the oxides and certain sili- cates, are slowly attacked by the strong acids, but the solvent action is not rapid enough to make this a practicable method of getting them into solution. 2. The solution is evaporated to remove the large quantity of acid which would otherwise interfere with the [28 precipitation. Care is taken not to over- heat the dry residue, so as to avoid loss of mercury, antimony, and tin by volatilization of their chlorides. A measured quantity of HC1 is then added; and, after mixing this HC1 solution with the HNOs solution, an equivalent quantity of NB^OH is added, in order to produce the acid concentration required for the H2S precipitation. Only 10 cc. of water are added to the HC1 solution at first, so as to prevent the precipitation of SbOCl and thus make it possible to determin whether the substance has been completely decomposed. 3. If the original substance was an alloy, a residue after the treatment with HC1 and HNOs is likely to consist of metastannic or silicic acid or of carbon, a platinum metal, or an alloy of iron with chromium, silicon, etc. It is best treated with H2S04 and HF by P. 5, in order to test for and remove silica and to dissolve metastannic acid and iron-alloys. If a black or metallic residue still remains, it may be tested for graphite by rubbing a dried portion on the fingers or on paper; and to bring it into solution the remainder may then be fused with Na2O2 in a nickel crucible, the mass treated with water and HC1, and the solution analyzed as usual, except that nickel cannot be tested for. 4. If the original substance was an alloy and a large, nonmetallic residue remains after treatment with HNOs (P. 3), it is sometimes advantageous, instead of treating it by P. 4, to analyze the residue separately by the following procedure, by which a large quantity of metastannic acid is more readily dis- solved: Add to the residue in a casserole 3-4 cc. of 96 % H 2 SO 4 and heat under the hood until the acid has evaporated to a volume of about 2 cc. Cool, add an equal volume of water, cool again, add 5 cc. HC1 to dissolve antimonic oxide, and heat to boiling. Cool completely, filter if there is a residue (which may consist of silicic acid), and add the acid solution drop by drop, with con- stant shaking, to a mixture of 10 cc. ammonium monosulfide, 1 cc. ammonium polysulfide, and 10 cc. 15-normal NH 4 OH in a flask. Cover the flask and digest for a few minutes on a steam bath. Filter out the precipitate, which P. 5 PREPARATION OF THE SOLUTION. 35 may consist of small quantities of sulfides of the copper and iron groups. Dilute the filtrate, and make it slightly acid with HC1. Shake to coagulate the pre- cipitate, filter, and wash with hot water. Analyze the precipitate for the tin- group by P. 41; reject the filtrate or test it for phosphate by P. 104. *Procedure 5. Fluoride Treatment of the Residue Insoluble in the Common Acids. Transfer to a platinum crucible (see Notes 1 and 2) the residue after treatment with acids (P. 4), add 2 cc. of 96% H 2 S04 from a graduate, heat with a moving flame until white fumes are given off, and cool completely. To test for silicate, add carefully from a lead or hard-rubber tube capped with a rubber nipple pure 48% HF drop by drop until 5-6 drops have been added, and warm the mixture over a steam bath. (Formation of gas bubbles, presence of SILICA or SILICATE.) Then add 2-5 cc. more pure 48% HF, cover the crucible with a platinum cover, digest on a steam bath for about 15 minutes unless the residue dissolves more quickly; remove the cover, and evaporate under a hood until dense white fumes of H 2 S04 are given off, taking care to avoid spattering. (See Note. 3.) [Unless it is known from the presence of solid substance at this point or from other indications that the residue treated with H 2 S04 and HF contained other constituents than silica, determin this by evaporating off the H 2 S04 under a hood, taking care not to ignite the dry residue. If a significant residue remains, add from a graduate 1.5 cc. of 96% H 2 S0 4 , and heat until the residue is redissolved, not allowing the acid to evaporate.] Cool, pour the contents of the crucible into 10 cc. water, and rinse out the contents with a little water. Boil to dissolve slowly dissolving sulfates; cool, shake, filter, and wash the residue, first with 6-normal H 2 S04 and then with a little water. (Residue, P. 6; filtrate, P. 11.) Notes. 1. A student using this procedure for the first time should work under the direct supervision of an instructor. Great care must be taken not to breathe the fumes of HF nor to get it on the hands; for it is extremely irritating and pro- duces dangerous burns. 2. Whenever a residue or precipitate has to be transferred from a filter to a cru- cible in which it is to be ignited or fused, it is best to separate it as far as possible from the filter-paper, and then to incinerate the part of the paper to which much residue adheres by rolling it up, winding a platinum wire around it in the form of a spiral, and heating it in a gas-flame till the carbon is all burnt off. 3. When a liquid is to be evaporated in a crucible, it is well to heat it within a larger iron crucible, which serves as an air-bath. The smaller crucible may be supported upon a nicrome triangle set into holes bored in the side of the iron crucible, or upon a circular disk of asbestos-board with a round hole cut out in the middle and slots cut out along the sides. * If the use of a platinum crucible or of hydrofluoric acid is impracticable, the lees satisfactory, alternativ method described in P. 7 may be employed (see Note 8, P. 5). 36 PREPARATION OF THE SOLUTION. P. 6 4. The test for silica or silicate depends on the formation of SiF4 gas, which is insoluble in strong H2S04, but dissolves in water in the presence of HF with formation of fluosilicic acid, H 2 SiFe. With free silica the evolution of gas takes place in the cold; but with slowly decomposing silicates, such as feldspar, the test is obtained only upon warming. A few silicates are not acted upon by HF and H 2 SO 4 , and, of course, do not show the test for silica at this point. The test is delicate enough to enable 1 mg. of silica, whether free or in a de- composable silicate, to be detected. Moreover, after the substance has been treated with acids as in P. 4 and warmed with H 2 SO<, an evolution of gas with HF is not produced with the compounds of any element other than silicon. It should be borne in mind that a small quantity of silica will be introduced if ordinary niters (which have not been washed with HF; have been employed and have been destroyed by acids or by ignition, or if a strongly alkaline solu- tion has been boiled in glass vessels. 5. Since glass and porcelain consist of silicates which are readily attacked by HF, this acid must not be allowed to come into contact with these materials. In handling cold HF solutions, vessels and funnels of celluloid or paraffin or of glass coated with paraffin may be used; but platinum vessels must be em- ployed when the solutions are to be heated. Care must be taken not to intro- duce into a platinum vessel any solution containing chlorin or bromin or any acid mixture containing nitrates and chlorides by which chlorin would be evolved. Platinum is so slowly attacked by hot concentrated H 2 SO4 that even when 2-3 cc. of the acid are rapidly evaporated in a crucible less than 0.5 mg. passes into solution. 6. The digestion with HF decomposes most silicates and dissolves silica. The subsequent evaporation with H 2 S(>4 expels the excess of HF and decom- poses the fluorides produced, as well as some other substances that may have been left undissolved by the HNO 3 and HC1. The H 2 SO 4 solution is diluted with a small quantity of water so as to cause the complete precipitation of BaSO4, SrSC>4, and PbSO4. These sulfates are moderately soluble in strong H 2 SC>4 and may not appear till after dilution. The addition of much water is avoided, since SrSC>4 and PbSC>4 are somewhat soluble in water; and the residue is washed with dilute ^864 for the same reason. The solution is boiled so as to dissolve anhydrous sulfates, such as those of aluminum and iron. 7. The residue insoluble in dilute H 2 SC>4 contains as sulfates all the barium, strontium, and lead, and all of the calcium in excess of 5-10 mg., left undis- solved by HNOa and HC1; more or less of the chromium (according as the H 2 SO4 has been more or less strongly heated) as a pink anhydrous sulfate; and part of the bismuth as basic sulfate and antimony as antimonic hydroxide, when much of these elements was left undissolved by the previous treatments with acids. The residue may also contain still undecomposed substances, especially the following: silver chloride; corundum, Al 2 Os; chromite, FeCr 2 O4j cassiterite, SnO 2 ; some anhydrous silicates and fluosilicates, such as cyanite or andalusite (Al 2 SiOs) and tourmalin; graphite and carbides; and certain compounds of the rarer elements. For a method of bringing some of these substances into solution, see Note 3, P. 7. 8. If the use of a platinum crucible or of hydrofluoric acid is impracticable, the residue insoluble in HC1 and HNOs may be fused in a nickel crucible with P. 6 PREPARATION OF THE SOLUTION. 37 , as described in P. 7, instead of being treated by P. 5-6. This is, however, a far less satisfactory method of analysis for the following reasons. Compounds of the alkali elements are used as a flux; nickel is introduced from the crucible; and mercury compounds are volatilized; so that these elements can not be tested for in the subsequent analysis. Moreover, the treatment with HF and H2SO4 is almost always a shorter process, since when the residue consists only of silica, as is often the case with minerals, no further treatment is necessary, and since in other cases there is often no residue to be boiled with Na2COs solution (P. 6). A fusion in a platinum crucible with Na2COa would be less objectionable; but this is not possible, unless elements reducible to the metallic state are known to be absent in the residue (see Note 5, P. 7). Procedure 6. Treatment of the Residue from the Fluoride Treat- ment. Transfer the residue insoluble in dilute H 2 S0 4 (P. 5) to a casserole, add about 25 cc. 3-normal Na2COs solution, cover the casserole, and boil gently for 10 minutes. Filter and wash the resi- due thoroly. (Filtrate, reject.) Heat the residue with just 5 cc. HNOs and 10-20 cc. water. Filter out any undissolved residue, and treat the solution by P. 11, subsequently testing it only for lead, bismuth, chromium, barium, strontium, and calcium. Notes. 1. The boiling with NasCOs converts into carbonates the sulfates of lead, calcium, strontium, and bismuth completely, and at least 80% of the sulfate of barium, even when large quantities of them are present. A second treatment, which should be applied to the residue if there are indica- tions that barium is present, completely decomposes BaSO4. The carbonates dissolve readily in HNOa. Anhydrous chromic sulfate, which is left undis- solved by dilute H2S04 (P. 5) as a fine pink or gray powder, is slowly changed by boiling with Na2COs to a greenish blue hydroxide which dissolves in the HNOa, leaving behind the still undecomposed sulfate. Antimonic oxide dis- solves only to a small extent (2-4 mg.) in the N^COs solution or in the dilute HN0 3 . 2. Any residue insoluble in HNOs can therefore consist only of barium or chromic sulfate, of antimonic oxide, or of some of the original substance still undecomposed, which is likely to consist of one of the nativ oxides or silicates mentioned in P. 5, Note 5. If such a residue is obtained, it can ordi- narily be rendered soluble by fusion with Na2COs, K^COs, and KNOs, as de- scribed in P. 7; but in this case a platinum crucible may be used for the fusion, provided the residue be first heated with HC1 to extract any Sb20s that may be present and provided silver is not found present in the H2S04 solution obtained in P. 5. Procedure 7. Alternativ Treatment of the Residue Insoluble in the Common Acids. If the use of HF (P. 5) is impracticable, transfer the residue insoluble in acids (P. 4) to a nickel crucible (see Note 2, P. 5), heat until the residue is dry, mix the residue with ten to twenty times its weight of anhydrous Na2COs, cover the crucible, heat strongly 38 PREPARATION OF THE SOLUTION. P. 7 over a powerful burner, preferably within a cylinder of asbestos-paper, so that complete fusion takes place, and continue the heating for 10-20 minutes. If dark particles of undecomposed substance can still be seen, add gradually in small portions 0.1-0.5 g. of solid KN0 3 , and heat strongly for several minutes. Cool, boil the crucible and its contents with water until the fused mass is disintegrated, filter, and wash the residue thoroly. Warm the residue with HNO 3 until action ceases, and filter out any still undecomposed substance. Test one half of the carbonate solution for acidic constituents as described in the third and fourth paragraphs of P. 131. Mix a tenth of the HN0 3 solution with a tenth of the carbonate solution, making the mixture acid with HN0 3 , if it is not already so. If no precipitate forms, mix the remainder of the acid solution with the remainder of the carbonate solution. Add 3-5 cc. HC1 (or more if the solution is still alkaline), and filter. Test the precipitate for lead and silver by P. 13-14. Evaporate the solution, and heat the residue until it is thoroly dry at 120-130 by keeping it in motion over a small flame. Moisten the residue with 16-normal HN0 3 , and heat again till the residue becomes perfectly dry. Add from a graduate just 5 cc. 6-normal HNOs and about 20 cc. water, and heat to boiling. Filter out any residue, dilute the filtrate to 100 cc., and treat it by P. 21. If a precipitate forms on mixing the small portions of the HNOs solution and the carbonate solution, treat these solutions separately as described in the preceding paragraph, uniting the precipitates formed by the same group-reagent in the subsequent analysis. Notes. 1. Upon fusion with sodium carbonate most compounds undergo metathesis, the acidic constituent of the compound combining with the sodium, and the basic element with the carbonate. The carbonate formed is, however, sometimes decomposed by heat with production of the oxide or of the metal itself. Acid-forming oxides, such as SiC>2, As2O5, and less rapidly A^Oa, expel CQj from the carbonate and form sodium salts. Such reactions are illustrated by the following equations: BaSO 4 +Na2CO 3 =Na2SO 4 +BaCO 3 . 4AgCl Si0 2 After the treatment with water, the acidic constituent of the substance is therefore found with the excess of carbonate in the aqueous extract, while the basic element remains undissolved by the water and passes into the acid solu ion. The first and third reactions are examples of cases where the aqueous and acid solutions must not be mixed, for upon mixing BaSC>4 or AgCl would again be formed. P. 7 PREPARATION OF THE SOLUTION. 39 2. Of the basic elements that may be present, all or a part of the arsenic, antimony, tin, aluminum, chromium and manganese are contained in the carbonate solution; and this solution must therefore be analyzed for basic elements. Since this solution may also contain NagSiOs, the solution after the addition of acid is evaporated to dryness, and the residue is heated at 100-130, in order to dehydrate the silicic acid and render it insoluble. The resi- due insoluble in HNOs after this treatment usually consists only of silicic acid. To prove whether it consists wholly of this acid, it may be treated with H 2 SO4 and HF as described in P. 5. 3. Some substances which are not much acted upon by alkali carbonates alone are readily attacked when an oxidizing substance like KNOs is present. Thus, sulfides are converted into sulfates and chromium compounds (such as chromite, FeOC^Os) into chromates. A few substances, however, such as the nativ or ignited oxides of tin and aluminum, may be only partially de- composed even by long-continued fusion with the mixed fluxes. Such an un- decomposed residue may be fused with KOH in a nickel or silver crucible and the fusion treated first with water and then with HC1. The oxides of aluminum and stannic tin, if finely powdered, dissolve rapidly in fused KOH. The aqueous extract contains the aluminum as aluminate and most of the tin as stannate. The residue undissolved by water may consist of black nickel oxide from the crucible, stannic hydroxide, and of other hydroxides accompanying the aluminum or tin oxides; all of which dissolve in HC1. 4. A few milligrams of nickel are taken up from the crucible by the flux, so that this element, as well as the alkali elements, cannot be tested for later in the analysis. The crucible is, however, so little attacked by the flux that it can be used repeatedly. 5. Whenever it is permissible, it is somewhat better to make the fusion in a platinum crucible, since then no foreign substances are introduced from the crucible. It is not permissible, however, to ignite in platinum vessels compounds of the silver, copper, and tin groups; for these may be reduced to the metal by heating with an alkaline flux. The same is true of sulfur, sulfides, and in the presence of organic matter of phosphates; for all these elements form easily fusible alloys with the platinum, and thus spoil the crucible. More- over, alkaline hydroxides and strongly oxidizing fluxes (such as peroxides and nitrates) must not be fused in platinum, since they attack it fairly rapidly. Therefore, if the fusion is made in platinum, no more KNOs should be added than is necessary. Procedure 8. Destruction of Organic Matter. If the closed-tube test (P. 1) has shown the presence of organic matter, powder or cut into small pieces 1-5 g. of the substance (according to the quan- tity of organic matter present). Add to it in a casserole about 5 cc. of 96% H 2 S0 4 ; warm gently until the substance is well charred; cool; add slowly, with constant stirring, under a hood, 16-normal HN0 3 , until violent reaction ceases; warm gently for a few minutes, and then heat more strongly, keeping the dish moving, until the substance is thoroly charred. Cool, again add 16-normal HNOs as before, and 4 40 PREPARATION OF THE SOLUTION. P. 8 heat until thick fumes of H 2 S0 4 are evolved. Repeat this process till the mixture becomes light-colored and remains so when heated strongly. If the substance has dissolved completely (or even if it has not, if the use of HF is impracticable), evaporate off the H 2 S04 under a hood till only 1.5 cc. remains, cool completely, add very carefully 10-20 cc. water, and boil. (If a precipitate separates, filter it off and treat it by P. 6.) Treat the solution by P. 11. If the substance has not dissolved completely, transfer the mixture to a platinum crucible, evaporate off the [2864 till only 1.5 cc. re- main, cool completely, and treat the mixture by the second and third paragraphs of P. 5. Notes. 1. This method of destroying organic matter is of very general application, being effective even when such stable substances as paraffin and cellulose are present. Organic matter can also be destroyed by ignition, but this has the disadvantages of volatilizing certain elements, especially mercury and arsenic, and of making some substances very difficultly soluble. When the organic matter consists only of oil, as is the case with an oil paint, it may be better to extract it with ether, especially when it is desired to determin the proximate constituents of the substance. 2. The residue contains: any substances originally present that have not been attacked by HNOs or H2SO4, especially silicates; all the lead, strontium, and barium that may have been present in any form, since the sulfates of these elements are insoluble in dilute H2SO4J all the silica, since silicic acid is dehydrated and made insoluble by heating with E^SC^; some of the calcium, bismuth, antimony, and tin, when these elements are present in considerable quantity, since their sulfates (or oxides) are not readily soluble in dilute H2SO4; and substantially all of the chromium, since its sulfate is converted into the insoluble anhydrous form. 3. After the organic matter is destroyed, the solution is evaporated to 1.5 cc., in order that the concentration of the acid may be properly adjusted in the subsequent H 2 S precipitation. DETECTION OF THE BASIC CONSTITUENTS. GENERAL DISCUSSION. The science of qualitativ chemical analysis treats of the methods of determining the nature of the elements and of the chemical com- pounds which are present in any given substance. When the presence or absence of the various elements is alone determined, the process is called ultimate analysis; when the chemical compounds of which the substance is composed are identified, it is called proximate analysis. In the analysis of inorganic substances, to which this book is devoted, the object in view is ordinarily an intermediate one namely, that of de- tecting the base-forming and acid-forming constituents (called in this book for short the basic and acidic constituents) that are present in the substance. Thus the analysis of a substance consisting of calcium sul- fate, zinc chromate, and ferric oxide would show not only that the ele- ments calcium, sulfur, zinc, chromium, and iron were present, but also that the sulfur was in the form of sulfate (not sulfide or sulfite), the chromium in the form of chromate (not of a chromic salt), and the iron in the ferric (not the ferrous) state. The reason for this is that the analysis is carried out by dissolving the substance in water (with aid of acids, if necessary), and by treating the solution so obtained suc- cessively with a number of different chemical substances. Now, since the chemical reactions of substances in aqueous solutions are deter- mined by the nature of the ions which they yield, and since the ions correspond to the basic and acidic constituents, it is these constitu- ents whose presence or absence is established. For detecting the basic constituents a systematic method is employed which consists in adding to an acid solution of the substance in suc- cession ammonium chloride, hydrogen sulfide, ammonium hydroxide and sulfide, and ammonium carbonate. By each of these reagents (which is the name applied to substances added to produce any desired chemical reaction) a group of basic constituents is precipitated. Thus, ammonium chloride precipitates those constituents whose chlorides are only slightly soluble in water; hydrogen sulfide, those whose sulfides are only slightly soluble in dilute acid; ammonium hydroxide and sulfide, those whose sulfides or hydroxides are only slightly soluble in ammoniacal solutions; and ammonium carbonate, those whose car- bonates are only slightly soluble in water containing ammonium car- bonate. The way in which the basic constituents are thus separated into groups is shown in more detail in Table II on the following page. 41 SEPARATION INTO GROUPS. H Z S te. NHJ 2 S 1 i ni Filtrate: INUM-GRO CR ZN) as sodium salts See Table VIII 9 3 o HI -I* ^ 2 '^ *3 0! 3 3 O ' '& S I SH^i ...B|| a oQ S 1^1 ^a^:! i g a - ||| rt g ft O cT .. fe'o l^ 19^ p. 11 ANALYSIS OF THE SILVER-GROUP. 43 PRECIPITATION AND ANALYSIS OF THE SILVER-GROUP. TABLE III. ANALYSIS OF THE SILVER-GROUP. Precipitate: BiOCl*, PbCl 2 *, AgCl, Hg 2 Cl 2 . Treat with HCl (P. 12.) Solution: BiCl 3 . Evaporate, pour into water (P. IS). Residue: PbCl 2 , AgCl, HgaCla. Treat with hot water (P. IS). Solution: PbCl 2 . Add HzSOt (P. 18). Residue: AgCl, Hg2Cl 2 . Pour NH 4 OH through the filter (P. 14). Precipitate : BiOCl. Black residue: HgandHg^ Solution: Ag(NH 3 ) 2 Cl. AddHNO z (P.14). Precipitate: PbS0 4 . White precipitate: AgCl. * These precipitates form only when large quantities of bismuth and lead are present. Procedure n. Precipitation of the Silver-Group. Place the cold solution of the substance (prepared by P. 2, 3, 5, 6, or 8, and containing 5 cc. 6-normal acid in about 25 cc. of solution) in a conical flask, and add to it 10 cc. NH 4 C1 solution. (White precipitate, presence of SILVER- GROUP.) Let the mixture stand for 3 or 4 minutes; then filter it. (Precipitate, P. 12; filtrate, P. 21.) Notes. 1. It is recommended that in general hard-glass conical flasks (the to-called Erlenmeyer flasks of Jena or Bohemian glass), rather than beakers or test-tubes, be employed for holding solutions that are being subjected to the opera- tions of precipitation and heating. 2. Even in cases where it is not essential to add a perfectly deflnit volume of a reagent, the analyst should make it a practis to measure out the quantity to be added, rather than to pour in an indeflnit quantity from the reagent bottle. For this purpose a 10 cc. graduate should be constantly at hand. For adding smaller quantities than 1 cc. a dropper should be used. This may be made by drawing out one end of a short glass tube to a wide capillary and capping the other end with a rubber nipple. When more of a reagent than is needed has been poured into a graduate or other vessel, it should never be poured back into the reagent bottle, owing to the danger of contaminating the reagent. 5. Unless the concentration is specified, it is understood that all salt solutions used as reagents are 1-normal, that is, that they contain one equivalent of salt per liter of solution', also that the acid and base solutions used as reagents (those of HCl, HN0 3 , HzSOi, HAc, NHOH, and NaOH) are 6-normal. 4. By one equivalent of any substance is meant that weight of it which reacts with one atomic weight (1.008 grams) of hydrogen in any of its com- pounds or with the weight of any other substance which itself reacts with one 44 ANALYSIS OF THE SILVER-GROUP. P. 11 atomic weight of hydrogen. Thus, one equivalent is the quantity in grams corresponding to the following formulas: INaOH, Ba(OH) 2 , 1HC1, H 2 SO 4 , SH 3 PO 4 , 1NH 4 C1, Na 2 SO 4 , |CaS0 4 , JFeCl 3 . When a substance may take part either in a reaction of metathesis or in one of oxidation and reduction, its meta- thetical equivalent has to be distinguished from its oxidation equivalent. Thus the metathetical equivalent of nitric acid is 1HNO 3 ; but its oxidation equivalent (when it is reduced to NO) is |HNO 3 . In this book the term equivalent will always be used to denote the metathetical equivalent. Note that the number of equivalents of a substance is a certain quantity of it; but that the term normal denotes its concentration, that is, the quantity of it per unit-volume (more specifically, the number of equivalents of it per liter). 5. If NEUC1 produces no precipitate, it proves the absence of silver and mercurous mercury, but not of lead or bismuth, since PbCl 2 is fairly soluble in water and BiOCl is fairly soluble in dilute acid. The solubility of PbCl2 is, however, much less in a solution of NH 4 C1 or of any other chloride than it is in water owing to the common-ion effect explained in detail in the following note. 6. The fact that the solubility of PbCl2 is greatly decreased by the addition of NH 4 C1 or HC1 is explained as follows: The mass-action law requires that at a given temperature in all dilute solutions containing lead chloride the ratio of the product of the ion-concentrations* (Pb ++ ) X (Cl~) 2 to the concentration (PbCl 2 ) of the un-ionized salt have the same value; that is, (Pb ++ ) X (Cl~) 2 4- (PbCl2) = some definit value. Now in all solutions which have been saturated with lead chloride as a result of sufficiently long contact with the solid substance, the concentration of the lead chloride present as such (that is, as un-ionized PbCl 2 ) must evidently have the same value, and therefore in all such saturated solutions the ion-concentration product (Pb ++ )X(Cl~) 2 must also have the same value; that is, in all solutions saturated at a given temperature with lead chloride, (Pb + +) X (Cl~) 2 = some definit value. This particular value which the ion-concentration product has when the solution is saturated is commonly called the solubility-product', but.the principles involved are less likely to be mis- understood if it be called the saturation-value of the ion-concentration product. The saturation- value varies, of course, with the nature of the salt, and with the temperature in the case of a given salt. In the case of lead chloride at 20, whose solubility in water at 20 will be seen by reference to the Table on page 122 to be 70 milli-equivalents per liter, the saturation-value of the ion-concentration prod- uct in these units is evidently (70) X (70) 2 = 343, 000, provided the ionization be considered to be complete, as may be assumed to be true in these qualitativ considerations in the case of nearly all neutral salts. Any solution containing lead-ion and chloride-ion in which the ion-concentration product exceeds this saturation-value is evidently supersaturated and tends to deposit the solid sub- stance; and any solution in which the ion-concentration product is less than the saturation-value is evidently undersaturated and tends to dissolve more of the solid substance. Now, when NH 4 C1 or HC1 is added to a saturated solu- tion of PbCl2 in water, the immediate effect is to increase the value of (Cl~), and therefore of the product (Pb ++ )X(Cl~) 2 ; but the solution becomes thereby supersaturated, and PbCl 2 will precipitate out of it until the saturation- value of the product (Pb++) X(CJ-) 2 is restored. * In mass-action expressions of this kind, chemical formulas within parentheses denote the concentrations of the respectiv substances, that is, the quantities of them per liter of solution. P. 12 ANALYSIS OF THE SILVER-GROUP. 45 Procedure 12. Extraction and Detection of Bismuth. Pour re- peatedly through the filter containing the NH 4 C1 precipitate (P. 11) a cold 10 cc. portion of 2-normal HC1 (see Note 1). Treat the residue by P. 13. Evaporate the HC1 solution in a casserole almost to dryness (see Note 2), add a few drops of water, and pour the solution into a flask containing 100 cc. water previously heated to 50-70. (Fine white precipitate, presence of BISMUTH.) Confirm the presence of bismuth by filtering off the precipitate and treating it by P. 36. Notes. 1. When it is directed to dissolve a precipitate by pouring the solvent repeatedly through the filter, this is best done by pouring a single portion of the solvent from one test-tube through the filter into another test-tube back and forth three or four times. When the solvent is to be used hot (as in P. 13), it should be heated to boiling between each pouring. 2. When it is directed to evaporate a solution almost to dryness or just to dryness, the last part of the evaporation should be carried out by keeping the dish moving over a small flame in such a way as not to overheat the residue. 3. The precipitate is extracted with 2-normal rather than with 6-normal HCl, because PbCl 2 , AgCl, and Hg 2 Cl2 are much more soluble in the more con- centrated acid, owing to the formation of acids with complex anions, such as H+ 2 PbCl 4 =, H+ 2 AgCl 3 -, H + 2 HgCl 4 - 4. The white precipitate of BiOCl formed on the addition of water in the test for bismuth is produced by the hydrolysis of Bids. If HCl, the other product of the hydrolysis, is present in the solution, the reaction will not be complete, and a greater or less quantity of bismuth will remain in solution. This quantity increases very rapidly with the acid concentration. For this reason, nearly all the HCl must be removed by evaporation and the solution must be added to a large volume of water. Warm water is used, because the precipitation of BiOCl takes place more rapidly at the higher temperature. Antimony under these conditions givs a similar precipitate; but it can not be present in the HNOs solution of the substance in quantity sufficient to giv a precipitate with NH 4 C1, unless the substance contained also much chloride or sulfate. If it were so precipitated, it would behave in P. 12 like bismuth; but would be distinguished from it by the confirmatory test described in P. 36. 5. The expression for the solubility-product of BiOCl is (Bi +++ )X(O)- X(C1~)= const. Acids greatly increase its solubility because their H + ion combines with the O = ion to form water, which is a very slightly ionized sub- stance. Chlorides also increase its solubility; for the increase of (Cl~) which they directly produce is more than compensated by the decrease of (Bi+ ++ ) which arises from the formation of un-ionized Bids (and probably also of com- plex anions, such as BiCU"). Procedure 13. Extraction and Detection of Lead. Pour repeatedly through the filter containing the residue undissolved by HCl (P. 12) a 10 cc. portion of boiling water. Wash the residue thoroly with hot water, and treat it by P. 14. Cool the 10 cc. aqueous extract and add to it 10 cc. H 2 SO 4 . (Fine white precipitate, presence of LEAD.) Confirm the presence of lead by filtering out the precipitate, washing it with a little water, and treating it by P. 34. 46 ANALYSIS OF THE SILVER-GROUP. P. 14 Procedure 14. Detection of Silver and Mercury. Pour repeat- edly through the filter containing the residue insoluble in hot water (P. 13) a 5-10 cc. portion of NH 4 OH. (Black residue on the filter, presence of MERCUROUS MERCURY.) Acidify the filtrate with HN0 3 . (White precipitate, presence of SILVER.) When there is much black residue and little or no white precipitate, treat the residue by P. 15. Notes. 1 . When two quite different limiting quantities of the reagent are specified (for example, 510 cc. as in this procedure), the quantity added should be adjusted to the size of the precipitate. 2. The black residue that is produced by the action of NH 4 OH on H&Cl* is a mixture of finely divided mercury with the white mercuric compound HgClNH 2 . The reaction is expressed by the equation: Hg2Cl 2 +2NH4OH =HgClNH 2 +Hg+NH4Cl+2H 2 O. The compound HgClNH 2 may be considered to be a derivativ of HgCl 2 , formed by replacing an atom of chlorin by the univalent radical NH 2 . 3. An NH^OH solution contains a considerable proportion of (unhydrated) NHs; and AgCl dissolves readily in it, owing to the formation of a soluble com- plex salt, Ag(NH3) 2 Cl, which in solution is largely ionized into Ag(NH3) 2 " 1 " and Cl~ ions. This complex cation has so slight a tendency to dissociate into Ag + and NHs that the ratio of its concentration to that of the simple Ag + ion is about 10 7 in a normal solution of NB^OH. 4. If the PbCl 2 was not completely extracted from the NH 4 C1 precipitate by boiling water (in P. 13), it is converted into a basic salt (Pb(OH)Cl) by the NH 4 OH, and may pass through the filter, yielding a turbid filtrate. This basic salt will, however, dissolve on the addition of HNOa. Procedure 15. Detection of Silver in the Presence of Much Mer- cury. Wash the black residue undissolved by NH 4 OH (P. 14), and pour repeatedly through the filter containing it a mixture of 3 cc. HC1 and 10 cc. saturated Br 2 solution, at the same time rubbing the residue with a glass rod. Wash the filter, and pour repeatedly through it a 10 cc. portion of NH 4 OH. Acidify the solution with HN0 3 . (Yel- lowish-white precipitate, presence of SILVER.) Notes. 1. When much mercury is present a considerable quantity of silver (5 mg. or more) may be so completely retained in the black residue that scarcely any test for silver is obtained in P. 14. This is probably due to the fact that the AgCl is reduced to metallic silver by the metallic mercury. When much mer- cury is present it is therefore necessary to test the residue for silver, as described in this Procedure. 2. The bromin converts the mercury in the residue into soluble HgBr 2 and the silver into insoluble AgBr. The HC1 dissolves the HgClNH 2 present in the residue with formation of HgCl 2 . P. SI PRECIPITATION OF COPPER AND TIN GROUPS. 47 PRECIPITATION AND SEPARATION OF THE COPPER AND TIN GROUPS. TABLE IV. SEPARATION OF THE COPPER AND TIN GROUPS. HYDROGEN SULPHIDE PRECIPITATE: HgS, PbS, 61283, CuS, CdS; As 2 S 3 , As2S 5 , Sb 2 S 3 , Sb 2 S 6 SnS, SnS 2 . Treat with ammonium polysulfide (P. 22). Residue: HgS, PbS, BijjSs, CuS, CdS. See Table V. Solution: (NH 4 ) 3 AsS 4 , (NH 4 ) 3 SbS 4 , (NH 4 ) 2 SnS 3 . Add HCl (P. 23). Precipitate: As 2 S 5 , 80285, SnS 2 . See Table VI. Filtrate: NH 4 C1. Reject. Procedure 21. Precipitation of the Copper and Tin Groups. Dilute to 100 cc. the filtrate from the NH 4 C1 precipitate (P. 11) or the solution of the substance in HCl (P. 4), which should contain just 5 cc. of 6-normal HN0 3 , H 2 S0 4 , or HCl. Place this solution in a conical flask provided with a two-hole rubber stopper in which is a tube leading to the bottom of the flask. Pass into it in the cold a slow current of H 2 S, until, upon shutting off the gas and shaking thoroly, the liquid smells strongly of H 2 S. Filter at once, wash the precipitate with hot water (see Note 1), and treat it by P. 22. Heat the filtrate nearly to boiling (to. 70-90), and pass H 2 S into it at that temperature for 5-10 minutes. If there is no further precipitate, treat the solution by P. 51. If there is a further precipitate, evaporate the mixture almost to dryness. Add 3-5 cc. 12-normal HCl and evaporate just to dryness, to destroy the HN0 3 . Then add 10 cc. 6-normal HCl, saturate the cold solution with H 2 S, heat it to 70-90, and pass H 2 S into it for 5- 10 minutes. Cool the mixture, dilute it to 100 cc., and saturate it with H 2 S. Filter out the precipitate, wash it, treat it by P. 22, and unite the residue and the solution with those coming from the first H 2 S precipitate. Treat the filtrate by P. 51. Notes. 1. The washing of precipitates should in general be continued until the wash-water will no longer giv a test for any substance known to be present in the filtrate (for example, in this case for acid with blue litmus-paper or for chloride with AgNO' S ). Precipitates which are practically insoluble in water (like all the fulfides and hydroxides that are met with in this System of Analysis) are best washed with nearly boiling water, as this runs through the filter more rapidly and extracts soluble substances more readily. Precipitates which are appreciably 48 PRECIPITATION OF COPPER AND TIN GROUPS. P.gl soluble should be washed with cold water and with only a small quantity of it. The proper method of washing a precipitate is to cause a fine stream of water from a wash-bottle to play upon the upper edge of the filter. The wash-water should in general not be allowed to run into the filtrate, so as not to dilute it unnecessarily. When, however, a considerable proportion of the solution is likely to be retained in the filter and precipitate, it is well to add the first washings to the filtrate. 2. The formation of a white precipitate on diluting the solution to 100 cc. shows the presence of much bismuth or antimony. The precipitate, which con- sists of BiOCl or SbOCl, need not be filtered off. The formation, on passing in H 2 S, of a white or yellowish precipitate which immediately turns black with more H 2 S indicates mercury. (The white compound is HgCl2.2HgS, and this is converted into black HgS by the excess of H 2 S.) The formation of an orange precipitate shows antimony; of a yellow one, cadmium, arsenic, or stannic tin. All the other sulfides are black. 3. The solution is afterwards heated nearly to boiling and again saturated with H2S, in order to ensure the detection of arsenic; for this element, when present in the higher state of oxidation (as arsenic acid) is only very slowly precipitated by H 2 S in the cold. At 70-90 the precipitation is much more rapid, especially if the solution has been previously saturated with H 2 S in the cold. Under these conditions even 1 mg. As givs a distinct precipitate in less than 5 minutes. Continuous treatment with H 2 S at 70-90 in an open vessel does not, however, completely precipitate a large quantity of arsenic from such a weakly acid solution even within an hour. For this reason, when a considerable precipitate forms in the hot solution, it is directed to evaporate the filtrate, to add HC1 to destroy the HNOs (which in the concentrated state would decompose the H 2 S), to dissolve the residue in HC1, and to pass H 2 S through the hot solution. From this concentrated acid solution the arsenic precipitates completely in 5-10 minutes. The reasons for this peculiar behavior of arsenic in the higher state of oxidation are presented in Note 2, P. 43. The solution is finally diluted and saturated in the cold with H 2 S, since the other elements are not completely precipitated until the arsenic has been removed. 4. The effect of acid on the precipitation of the sulfides is explained by the mass-action law and ionic theory as follows: When a dilute solution, whether aqueous or acid, is saturated at a definit temperature with H 2 S gas under the atmospheric (or any definit) pressure the H 2 S present as such always hag the same concentration. This ionizes, however, to a slight extent into H + and HS", and to a still less extent into 2H + and S". It is only the latter form of ionization that needs to be considered here. Now between the H 2 S and its ions must be maintained the equilibrium expressed by the equation (H + ) 2 X(S = ) = const. X (H 2 S); or, since in this case (H 2 S)= const., as just stated, it follows that also (H + ) 2 X(S=)= const. From this it is evident that when (H + ) is increased by the addition of acid to the solution, (S*) must be decreased in the proportion in which the square of (H + ) is increased; thus, if (H + ) is doubled, (S = ) will be reduced to one-fourth. But in order that a sulfide for example, of the formula M+^S" may precipitate, the concentra- tion-product (M ++ ) X (S~) must attain its saturation-value. This value varies, however, with the nature of the sulfide and with the temperature; and there- fore the acid concentration that will barely permit of precipitation when (M + +) P.tl PRECIPITATION OF COPPER AND TIN GROUPS. 49 has a definit value (for example, 1 mg. in 100 cc.) will be different for different sulfides and for the same sulfide at different temperatures. Thus if the ele- ments are arranged in the order in which they are precipitated from cold HC1 solutions as the acid-concentration is progressivly decreased, the series is ap- proximately as follows: arsenic, mercury and copper, antimony, bismuth and stannic tin, cadmium, lead and stannous tin, zinc, iron, nickel and cobalt, manganese. The acid concentration which permits precipitation also varies with the ionization of the acid; thus zinc is precipitated from a fairly concen- trated solution of acetic acid, since, owing to the slight ionization of this acid, the H+ concentration is less than in a far more dilute solution of HC1. The three acids, HC1, HNOs, and H 2 SC>4, do not, however, differ greatly in this respect, since they are all largely ionized in dilute solution. 5. The acid concentration is made 0.3 normal (5 cc. of 6-normal acid being present in 100 cc.) and the solution is saturated with H2S gas in the cold, since under these conditions even 1 mg. of cadmium, lead, or tin precipitates, and even 300 mg. of zinc remain in solution. (This statement in regard to zinc is true, however, only when the solution contains also a considerable quantity of chloride-ion, such as was added in P. 11, and when it is not allowed to stand.) Moreover, even when a small quantity of any of the iron-group elements is present with a large quantity of a copper-group element, the former is not carried down in the H 2 S precipitate under these conditions to such an extent as to prevent its detection in the filtrate, provided as much of it as 1 mg., or in some combinations as much as 2 mg.. is present. 6. A white, finely divided precipitate of free sulfur will be formed if the solution contains substances capable of oxidizing H 2 S. The most important of these likely to be present are ferric salts, chromates, permanganates, and chlorates. The reduction by H2S of ferric salts to ferrous is attended by a change in color from yellow to colorless; of chromates to chromic salts, from orange to green; and of permanganates to manganous salts, from purple to colorless. Nitric acid, if it were fairly concentrated, would also destroy the H 2 S; but at the concentration in question (0.3 normal) it has scarcely any oxidizing action even in boiling solution. 7. In balancing equations expressing reactions of oxidation and reduction, like those referred to in the preceding note, the main thing is to determin the number of molecules of the oxidizing and reducing substances which react with one another. This can be done most simply by considering, in the way illustrated by the following examples, the changes which take place in the valences of the atoms of these substances. Thus, in the reduction of a ferric to a ferrous salt by hydrogen sulfide, the iron atom changes in valence from +3 to +2, and the sulfur atom changes in valence from 2 (in H 2 S) to zero (in ordinary sulfur). Since the total change in the number of valences must be equal and opposit in the two substances, it is evident that two molecules of ferric salt react with one of hydrogen sulfide, and therefore that the equation is : 2FeCl 3 +H 2 S = FeCl 2 +S +2HC1. In the reduction of HClOs to HC1 by H 2 S, the chlorin atom decreases in val- ence from +5 to 1 (thus by six positiv valences) ; hence there must be a decrease of six negativ valences in the reducing substance, and that this may be the case three molecules of H 2 S are evidently required. The equation is therefore: HC10 3 +3H 2 S =HC1+3S+3H 2 0. 50 SEPARATION OF COPPER AND TIN GROUPS. P. 22 In cases where the valence of an atom in a substance is in doubt, it can be found at once from the valences of the other atoms with the aid of the principle that in any compound the sum of all the positiv valences is equal to the sum of all the negativ valences; thus in chloric acid HClOs, since the three oxygen atoms have 6 negativ valences and the hydrogen atom has one positiv valence, the chlorin atom must, in order to make the compound neutral, have 5 positiv valences. Consider as another example the reduction of potassium permanganate (KMnO*) to manganous chloride (MnCl 2 ) by H 2 S in the presence of HC1. The number of valence of the manganese atom is seen to be +7 in KMn(>4 (since that of four oxygen atoms is 8 and that of the potassium atom is +1) and to be +2 in MnCl 2 . The proportion is therefore 2KMnC>4: 5H 2 S, and the reaction is: 2KMnO 4 +5H 2 S+6HCl=2MnCl 2 +2KCl+5S+8H 2 O. The amount of acid required in such cases can be seen by inspection, most readily by noting how many hydrogen atoms are needed to combine with the oxygen atoms of the substance undergoing reduction. Thus in this case 16 hydrogen atoms are evidently needed for this purpose; and, since ten are fur- nished by the H2S, six more must be supplied by adding 6 molecules of HC1 (or an equivalent quantity of some other acid). Procedure 22. Separation of the Copper and Tin Groups by Am- monium Sulfide. Transfer the H 2 S precipitate (P. 21) to a small casserole, add to it 10-25 cc. ammonium monosulfide (if the original substance was treated with strong HNOs in P. 3), or 5-10 cc. ammo- nium polysulfide (if it was dissolved in water or in dilute HNO* in P. 2), cover the dish, and warm the mixture slightly (to 40-60) for about 10 minutes with frequent stirring. Add 10 cc. water, filter, and wash once with hot water. [If the residue is large and much has been extracted from it by this treatment, as indicated by its appear- ance or as determined in P. 23, warm it again with ammonium mono- sulfide or polysulfide, and filter, collecting the filtrate separate from the first one.] Wash the residue thoroly with hot water. (Residue, P. 31; solutions, P. 23.) Notes. 1. When a small precipitate is to be treated with a solvent, this may be done by pouring a portion of the solvent repeatedly through the filter. When a considerable precipitate is to be so treated, the filter is opened, the portions to which no precipitate adheres are torn off, and the remainder is laid along the side of a casserole; the solvent is then poured over it and is swashed to and fro, the precipitate being rubbed at the same time with a glass rod, so as to remove it from the filter. If this succeeds, the filter is drawn out and thrown away; otherwise, it is allowed to disintegrate and filtered out together with any residue. 2. The ammonium monosulfide reagent is a solution of (NBU^S and of the products of its hydrolysis, NH^SH, NH4OH, and a little H 2 S. The polysulfide reagent contains in addition some of the disulfide (NH4) 2 S 2 and of its hydrolysis product (NH 4 )HS 2 . The monosulfide is prepared by saturating concentrated P.8S SEPARATION OF COPPER AND TIN GROUPS. 51 NH 4 OH with H 2 S gas (whereby NH 4 SH is formed), adding an equal volume of NH 4 OH, and diluting the solution. The polysulfide is prepared from this solution by dissolving sulfur in it. These reagents, especially the monosul- fide, should be kept as far as possible out of contact with the air, which is con- veniently done by storing them in small, completely filled, glass-stoppered bottles; for the oxygen of the ah* destroys the sulfide with liberation of sulfur, which at first combines with the still unchanged sulfide forming the polysulfide, but later precipitates when the oxidation becomes more complete. 3. The chemical action of ammonium sulfide in dissolving the sulfides of the tin-group depends on the formation of soluble salts of sulfoacids with complex anions. When the monosulfide is used, the reactions are as follows : _ j (NH 4 +) 3 AsS 3 ( AsoSs ) +3(NH4)2S _ H + (NH 4 ) 2 S = (NH 4 +) 2 SnSr. The disulfide present in the polysulfide oxidizes the lower sulfides (As2S 3 , SbjjSt, SnS) to the same sulfosalts as are obtained by dissolving the higher sulfides (A82S5, 80285, SnS2) in ammonium monosulfide. It will be seen that these sul- fo-salts are analogous to the salts of the familiar oxygen acids of these ele- ments, the difference being that sulfur has replaced oxygen; and they are so named as to indicate this relationship. Thus the five sulfo-salts whose formulas are given above are called ammonium sulfarsenite, sulfantimonite, sulfarsenate, sulfantimonate, and sulfostannate. 4. In separating the copper-group from the tin-group (colorless) ammonium monosulfide is used rather than (yellow) polysulfide whenever the H2S pre- cipitate must contain in the state of the higher sulfide (81182 or Sb2S5) any tin or any antimony that is present. This is the cas3 when hot concentrated nitric acid was used originally in dissolving the substance, but may not be so when water or dilute HNO 3 was used; hence the directions as to the choice between the two solvents. The polysulfide has the disadvantages that it yields a large precipitate of sulfur on acidification and that it dissolves a not inconsiderable quantity of CuS and HgS, thus making it more difficult to determin from the color of the HC1 precipitate obtained in P. 23 whether or not elements of the tin-group are present. The fact that it dissolves some CuS and HgS also diminishes the delicacy of the tests for copper and mercury. The polysulfide must, nevertheless, be used if tin may be present as SnS, or much antimony as Sb2Ss; for in the monosulfide SnS is only slightly soluble, and Sb2S 3 only moderately soluble. 5. More specifically, the behavior of the various sulfides, when warmed with 10 cc. of these reagents, is as follows: Of the sulfides of the copper-group none dissolves to a significant extent in ammonium monosulfide. 5-10 mg. CuS and 0.5-1.0 mg. HgS may, however, dissolve in the polysulfide when the substance contains a large quantity of these elements. Yet when only 2 mg. are present, either of these elements can be detected in the analysis of the copper-group, even when the polysulfide is used, provided only one treatment with it has been made. Of the sulfides of the tin-group, 500 mg. of As as A^Sa or As2Ss, of Sb as Sb2S5, or of Sn as SnS2 dissolves in either the monosulfide or polysul- 52 SEPARATION OF COPPER AND TIN GROUPS. P. S fide, and 500 mg. of Sn as SnS or of Sb as Sb2Sa dissolve in the polysulfide. Scarcely any SnS and only 50-100 mg. Sb as 80283 dissolve in the monosulfide. 6. Even when a quantity of only 1 or 2 mg. of arsenic or antimony is present with a large quantity (even 500 mg.) of an element of the copper-group, enough is extracted by either the monosulfide or polysulfide to be detected in the sub- sequent tests. With tin, however, the separation is imperfect; for, when a large quantity of elements of the copper-group and only 3-5 mg. of tin are present, the whole of this may remain undissolved; indeed, when much cad- mium is present and the tin is in the stannous state, as much as 15 mg. of the latter may be wholly left in the residue, even when the polysulfide is used. On this account it is necessary to test for tin in the course of the analysis of the copper-group. Procedure 23. Reprecipitation of the Tin-Group. Dilute in a small flask the first portion of the ammonium sulfide solution (P. 22) with about 20 cc. water, make it slightly acid with HC1 (see Note 2), and shake the mixture for 2 or 3 minutes to coagulate the precipitate. (White or pale yellow precipitate, absence of TIN-GROUP; deep yellow or orange precipitate, presence of TIN-GROUP. See Notes 4 and 5.) [Treat the second portion of the ammonium sulfide solution (P. 22) in the same way, and unite the precipitate, if considerable in amount, with the first one.] Filter out and wash the precipitate, using suction, finally sucking it as dry as possible. (Precipitate, P. 41; filtrates, reject.) Notes. 1. In cases where the filtration is slow, where the precipitate must be washed with very little water, or where (as in this case} it must be free ' as far as possible from water, it is advisable to filter with the aid of suction. This op- eration is carried out by reinforcing the ordinary filter with a small hardened filter placed below it in the funnel, by inserting the funnel in a rubber stopper in the neck of a filter-bottle, and connecting the side arm of the filter-bottle to a suction-pump by means of a rubber tube carrying a screw-clamp. The suction should be applied very gradually so as to avoid breaking the filter. The filtrate should be poured out of the filter-bottle before beginning to wash the precipitate. 2. Whenever it is directed to make a solution slightly acid or alkaline, this should be done carefully as follows: Add from a graduate somewhat less acid or alkali than will neutralize the alkali or acid known to be present in the solution. Then add from a dropper more of the acid or alkali, a drop or two at a time, till a glass rod dipped in the solution and touched to a piece of blue or red litmus paper placed on a watch-glass changes the color of the litmus. 3. When the HC1 is added to the solution of the sulfosalts, the correspond- ing sulfoacids which are liberated decompose immediately into H2S and the solid sulfides. These are now necessarily in the higher state of oxidation, since the lower sulfides, if originally present, have been oxidized by the polysulfide. The fact that the sulfide precipitates when the solution of the sulfosalt is acidified is a consequence of the mass-action-law. Thus, since the complex anions dissociate according to the equations, 80285+ 3S~, P. SS SEPARATION OF COPPER AND TIN GROUPS. 53 this law evidently requires that, in any solution saturated with the solid sulfide, the concentration of the complex anion, and therefore of the tin, arsenic, or antimony in the solution, increase with increasing concentration of the S" ion. Now in the solution of the largely ionized (NEU^S there is a large concentration of S" ion; but when the solution is made acid with HC1, the S" ion is almost completely converted by the relativly large concentration of the H + ion into the slightly ionized substances HS~ and H2S. 4. Much time is ordinarily saved by determining at this point whether or not any element of the tin-group is present. When ammonium monosulfide has been used, there is never any difficulty in drawing a definit conclusion in regard to this from the size and appearance of the HC1 precipitate; for in the absence of the tin-group only a very small, nearly white precipitate of finely divided sulfur separates. Even when ammonium poly sulfide has been used, it is usually possible to decide as to the presence or absence of a small quantity of arsenic, antimony, or tin; for, altho considerable sulfur is then liberated, yet the precipitate is distinctly lighter colored and smaller in appearance when it consists only of sulfur than it is when even 1 mg. of arsenic, antimony, or tin is present with the sulfur. The difference is most marked when only a slight excess of HC1 has been used in the acidification and when the acidified mixture has been shaken, but not heated. In any doubtful case, the precipitate should be compared with that produced by acidifying 5-10 cc. of am- monium polysulfide to which 1 mg. of arsenic, antimony, or tin has been added. 5. When, however, the HC1 precipitate from a polysulfide solution is fairly small and is dark brown (indicating copper) or dark gray or black (indicating mercury) or of unpronounced yellow or orange color, so as to make any con- clusion as to the tin-group doubtful, the precipitate is best treated as follows: Heat it with 15-20 cc. NEUOH almost to boiling for 5 minutes and filter; test the precipitate for copper by P. 31, 35, and 37, if it has not already been found present; add to the filtrate a few drops of ammonium monosulfide, filter out any precipitate, heat the filtrate to boiling, make it acid with HC1, shake, filter out the precipitate, and treat it by P. 41 as usual. The character of the HC1 precipitate now obtained will clearly indicate the presence or ab- sence of the tin-group; for by the treatment with NELjOH the excess of sulfur originally present and any CuS is left undissolved, and by the (NH^S added to the solution any mercury present is precipitated, so that the HC1 precipi- tate can contain only sulfides of the tin-group and a very little sulfur. A^Ss, Sb 2 S5, and SnS 2 all dissolve in NEUOH (tho in the cases of Sl^Ss and SnSj less abundantly than in ammonium sulfide), owing to the formation of a mix- ture of salts of partially sulfurated acids, such as HsAsOgS and HsAsC^. The addition of (NIL^S to the NEUOH solution and the heating serve to convert these into the fully sulfurated acids, such as HgAsS4 ; from which HC1 will then precipitate the simple sulfides much more completely. The inci- dental removal of the small amounts of CuS and HgS by the NKUOH treat- ment is not necessary so far as the analysis of the tin-group is concerned, since their presence does not interfere with the detection of even 1 mg. of arsenic, antimony, or tin; but it does enable 1 or 2 mg. of copper to be detected which might otherwise be lost. In this connection it may be mentioned that a mix- ture of SnS2 and SbzS$ does not always have a color intermediate between the yellow and orange colors of the separate sulfides, but that it may be brown or dark gray. 54 ANALYSIS OF THE COPPER-GROUP. ANALYSIS OF THE COPPER-GROUP. P. 31 TABLE V. ANALYSIS OF THE COPPER-GROUP. RESIDUE FROM AMMONIUM SULFIDE TREATMENT! HgS, PbS, 61283, CuS, CdS. Boil vrilh HN0 3 (P. 81). Residue: HgS. Add Br 2 solution (P. 82} . Solution: Pb, Bi, Cu, Cd as nitrates. Add HySOt, evaporate, add water (P. 55). Residue: Sulfur. Solution: HgBr 2 . Add SnCk- Precipitate: PbS0 4 . Dissolve in NHiAc, add K 2 Cr0 4 (P. 84). Filtrate. Add NH^OH (P. 85} . Precipitate: Bi(OH) 3 . Add NazSn0 2 . (P. 36). Filtrate: Cu(NH3)4SO 4 , Cd(NH 3 )4S0 4 . White or gray precipitate: Hg2Cl 2 or Hg. To a small part add H Ac and KtFe(CN)t (P. 57). To the remainder addKCN and H 2 S (P. 88). Yellow precipitate: PbCr0 4 . Black residue: Bi. Red precipitate: Cu 2 Fe(CN)* White precipitate : Cd 2 Fe(CN) 6 . Yellow precipitate: CdS. Solution: K3Cu(CN) 4 . Procedure 31. Treatment of the Sulfides with Nitric Add. To the residue from the ammonium sulfide treatment (P. 22) in a casse- role add 10-20 cc. 2-normal HNO 3 solution, heat to boiling, and boil gently for a minute or two. Filter, and wash the residue. (Residue, P. 32; solution, P. 33.) Notes. 1. Boiling HNOs of this concentration dissolves the sulfides of lead, bismuth, copper, and cadmium almost immediately, and is therefore preferable to a more dilute acid, with which the reaction would require for its completion several minutes' boiling. Only a little HgS is dissolved by this treatment, unless the boiling is long continued. 2. Moderately concentrated H&Os dissolves sulfides much more rapidly than HC1 or H2SO 4 of the same concentration; for with the latter acids the sulfide-ion is removed from the solution only by combination with the hydrogen- ion and by the volatilization of the H 2 S formed thereby, while with HNOs the sulfide-ion (or the H2S in equilibrium with it) may also be destroyed by oxidation to ordinary sulfur. The oxidizing action of HNOa is, however, slow, unless it is hot and as concentrated as 2-normal. P. SI ANALYSIS OF THE COPPER-GROUP. 55 3. That HgS, unlike the other sulfides, does not dissolve in the dilute HNOs is doubtless due to the much smaller concentration of its ions in its saturated solution and to the fact that at this small concentration sulfide-ion (or the [28 in equilibrium with it at a correspondingly small concentration) is oxidized only very slowly by the dilute HNOa. HgS is, however, readily dissolved by more vigorous oxidizing agents, such as aqua regia or bromin solution, since they react rapidly with sulfide-ion (or with H2S) even when its concentration is very small. 4. If more concentrated HNOa be used, or if the acid become concentrated by long boiling, the black HgS is dissolved in part, and the remainder is con- verted into a heavy, white, difficultly soluble compound (Hg(NC>3)2.2HgS). As a small quantity of HgS may be so converted even by the hot 2-normal HNOa, the residue should be tested for mercury by P. 32, even if it is light-colored. A light-colored residue must often be tested also for tin (see Notes 2 and 3, P. 32). 5. When much lead, copper, or bismuth is present the sulfur formed generally encloses enough of the undissolved sulfide to give it a black color. A black residue is therefore not necessarily HgS, but must be further tested for mercury as described in P. 32. 6. Some sulfur is always oxidized to H2SO4 by the boiling HNOs; but, even in the presence of much lead, PbSO4 is not precipitated, owing to its moderate solubility in HNOs. 7. Even when only 1 mg. of lead, bismuth, or copper is present with 500 mg. of mercury, enough of any of these elements is extracted from the residue by the HNOa to giv a satisfactory test in the subsequent procedures; but in the case that 15 mg. of cadmium are associated with 300-500 mg. of mercury, about four-fifths of the cadmium is retained in the residue, so that a corre- spondingly small precipitate of CdS is produced in the final test for cadmium, and 1-2 mg. may escape detection. Procedure 32. Confirmatory Test for Mercury. Transfer the residue undissolved by HNOs (P. 31), with the filter if necessary, to a casserole, add 10-40 cc. saturated Br 2 solution, cover the dish, and warm slightly for 5-10 minutes, with frequent stirring. Boil the mixture until the bromin is expelled, and filter. (Residue, see Note 3.) Cool the solution, and add to it 15-20 drops of HC1 and one drop of SnCU solution; then add 3-5 cc. SnCl2 solution. (White pre- cipitate turning gray, or gray precipitate, presence of MERCURY.) Notes. 1. In the final test for mercury HC1 is added to prevent the pre- cipitation of a basic tin salt when the SnCl2 reagent is diluted, to diminish the tendency of CuCl to precipitate, "and to cause the formation at first of white Hg2Cl2. For the last reason, also, a single drop of SnCk solution is first added to the cold solution. By the excess of SnCfe the white precipitate is re- duced to gray, finely divided mercury. This darkening of the precipitate dis- tinguishes Hg2Cl2 from CuCl, which may separate as a white precipitate if a large quantity of CuS was left undissolved by the HNOs in P. 31. 2. If elements of the copper-group are present in large quantity (50- 56 ANALYSIS OF THE COPPER-GROUP. P. 32 500 mg.), the residue from the Br 2 treatment should be tested for tin, in order to guard against overlooking the presence of a small quantity of this element in the substance. For, as stated in P. 22, Note 6, a quantity of tin as large as 5 mg. (or even larger when stannous tin and cadmium are both present) may remain entirely in the residue undissolved by ammonium sulfide when this residue is large. Any SnS or SnS 2 present in it is converted by the HNOs in P. 31 into metastannic acid (H 2 SnOs), very little of which is dissolved either by the acid or by the Br2 solution used in P. 32. 3. To recover the tin from the residue, proceed as follows : Digest the residue from the Br2 treatment, if it is still dark-colored, with another portion of Br2 solution (to extract the rest of the mercury), filter, reject the filtrate, and warm the residue slightly with 2-5 cc. ammonium monosulfide. Filter, and unite the solution with the main ammonium sulfide solution obtained in P. 22. By this procedure 2 mg. Sn can be detected. 4. In case tin need not be tested for at this point, the residue may be more quickly dissolved by warming it with HC1 and adding gradually a little solid KClOs. Metastannic acid, being soluble in HC1, then passes into solution with the mercuric salt, but it does not interfere with the test for mercury. Procedure 33. Precipitation of Lead with Sulfuric Acid. To the HN0 3 solution (P. 31) add 10 cc. H 2 S0 4 and evaporate in a casserole to a volume of 1-2 cc. until dense white fumes of H 2 S04 begin to come off. Cool the mixture and pour it into 10-15 cc. cold water, rinsing out the casserole with the same solution. Cool again, shake, and allow the mixture to stand 5 minutes, but not much longer. (Fine white precipitate, presence of LEAD.) Filter, and wash the pre- cipitate first with H 2 S04, and finally with a little water. (Precipitate, P. 34; filtrate, P. 35.) Notes. 1. PbSC>4 is somewhat soluble both in water and in concentrated H2SO4, but much less so in dilute H 2 SO4, its solubility being scarcely appreciable in the cold 6-normal acid. That the solubility in dilute H 2 SO4 is less than that in water is due mainly to the common-ion effect. Concentrated H 2 SC>4 is, of course, an entirely different solvent. PbSO4 dissolves fairly readily in dilute HNOs, owing to the tendency to form the intermediate HSO4~ ion; hence, to ensure complete precipitation of PbS04, the HNOs must be removed by evaporation. To ensure complete precipitation it is also necessary to let the mixture stand a few minutes; for the solutions of substances which, like PbSO4, are crystalline and appreciably soluble tend to remain supersaturated. 2. When much bismuth is present it ordinarily dissolves at first when the water is added to the concentrated H 2 SO4, provided the mixture is kept cold; but from this solution a coarsely crystalline precipitate of an oxysulfate, such as (BiO) 2 S04, separates slowly upon standing in the cold but almost imme- 1 diately upon heating, and to such an extent that there may remain in solution not more than 50 mg. of bismuth. If the precipitate is of this character, free it from bismuth before applying the confirmatory test for lead by pouring repeatedly through the filter a 5-10 cc. portion of HC1 and treating the solution so obtained by P. 33. The evaporation with H 2 S04 is necessary in order to ensure reprecipitation of the PbSO4 that has been dissolved by the HC1. P. 34 ANALYSIS OF THE COPPER-GROUP. 57 Procedure 34. Confirmatory Test for Lead. Pour repeatedly through the filter containing the H 2 S04 precipitate (P. 33) a 10-20 cc. portion of NH 4 Ac solution. (See Note 1, P. 12.) To the filtrate add a few drops of K 2 CrO 4 solution and 2-5 cc. HAc. (Yellow precipi- tate, presence of LEAD.) Notes. 1. This confirmatory test for lead should not be omitted; for the HjS04 precipitate may consist not only of PbSO* but of (BiO)2SO4 or of BaSC>4, which last closely resembles PbS(>4 in appearance. (BiO)2SO4 dissolves in NH^Ac solution and gives a yellow precipitate on adding K2OO4 ; but this precipitate, unlike PbCrO*, dissolves readily in acetic acid. 2. The solubility of PbSO 4 in NH 4 Ac solution depends on the formation by metathesis of undissociated PbAcs, this salt being much less ionized than most other salts of the same valence-type. On the addition of a chromate to this solution the much more difficultly soluble PbO0 4 is precipitated. BaSC>4 is not dissolved by NEUAc solution, owing to its very slight solubility in water and the fact that barium acetate, unlike lead acetate, is a largely ionized salt. Procedure 35. Precipitation of Bismuth with Ammonium Hy- droxide -To the H 2 S0 4 solution (P. 33) add NH 4 OH slowly until a strong odor of it persists after shaking. (White precipitate, possible presence of BISMUTH; blue solution, presence of COPPER.) Shake to cause coagulation, filter, and wash the precipitate thoroly. (Precipi- tate, P. 36; filtrate, P. 37 and 38.) Notes. 1. The precipitate produced by NH4OH may consist also of Fe(OH)a, or of other hydroxides of the iron-group, if these elements were carried down in the H^S precipitate or were not completely removed from it by washing. The formation of a small precipitate is, therefore, not a sufficient proof of the presence of bismuth, and the confirmatory test of P. 36 must be applied. 2. Cd(OH) 2 or Cu(OH)2, tho only very slightly soluble in water, dissolves in NH^OH owing to the combination of the Cd ++ or Cu ++ ion present in the saturated solutions with NHs, forming the complex cation Cd(NHa)4 ++ or Cu(NHa)4 ++ . These complex cations have an extremely small ionization tendency; thus in a normal NH^OH solution the ratio of the concentration of the complex cadmium ion to the simple cadmium ion is about 10 7 . The solubility of these hydroxides in NH^OH is greatly increased by the presence of ammonium salts, since these salts, owing to the common-ion effect, greatly reduce the ionization of the NH4OH, and therefore the OH~ concentration in the solution, thus enabling the Cd ++ or Cu ++ concentration, and there- fore also the corresponding complex ion concentration, to attain a much larger value than in the saturated solutions of Cd(OH)2 or Cu(OH)2 in NBUOH alone. Procedure 36. Confirmatory Test for Bismuth. Pour through the filter containing well washed NH 4 OH precipitate (P. 35) a cold freshly prepared solution of sodium stannite (see Note 1). (Black residue, presence of BISMUTH.) 58 ANALYSIS OF THE COPPER-GROUP. P. 86 Notes. 1. The solution of sodium stannite (NaaSnC^) is prepared when needed by adding NaOH solution, a few drops at a time, to 1 cc. SnCk reagent diluted with 5 cc. water, until the Sn(OH)2 first formed is dissolved and a clear or slightly turbid liquid results. The solution must be freshly prepared, because it decomposes spontaneously into sodium stannate (Na^SnOs) and metallic tin, and because it oxidizes in contact with air to sodium stannate. SnC^Bfe is an example of a so-called amphoteric substance one which acts either as a base or an acid, as is shown by its solubility in both acids and alkalies. 2. The final test with sodium stannite depends on the reduction of Bi(OH)s to black metallic bismuth. The test is an extremely delicate one. The other reducible substances, like HSb0 3 , Fe(OH) 3 , Pb(OH) 2 , or Cu(OH) 2 , that might possibly be present in the NH^OH precipitate are not reduced by short con- tact with stannite solution in the cold. Mercury, if present, would cause blackening; but it is not precipitated by NH^OH in the presence of ammo- nium salt. Procedure 37. Confirmatory Test for Copper. Acidify one- fourth of the NH 4 OH solution (P. 35) with HAc, add one drop K4Fe(CN)e solution, and allow the mixture to stand for 2 or 3 minutes. (Red precipitate, presence of COPPER.) Then add 4-5 cc. more K4Fe(CN) 6 solution. If it is uncertain whether there is a pre- cipitate, pour the solution through a filter and wash with a little water. (Pink color on the filter, presence of COPPER.) Notes. 1. The confirmatory test for copper is more delicate than the formation of a blue color with NHjOH (P. 35). It should, therefore, be tried even when the NH^OH solution is colorless. Cadmium is also precipitated by K4Fe(CN)e; but the precipitate is white, and does not prevent the pink color of the copper compound from being detected, provided only a small quantity of K4Fe(CN)e is added ; for the copper salt, owing to its smaller solubility, is first precipitated. 2. Nickel, like copper, givs a blue solution with excess of NH40H; but even if present it would giv a green, not a red, precipitate with K4Fe(CN)e. Procedure 38. Detection of Cadmium. Treat the remainder of the NH 4 OH solution (P. 35) with KCN solution (see Note 1), adding only a few drops if the solution is colorless, but enough to decolorize it if it is blue, and pass in H^S gas for about half a minute. (Immediate yellow precipitate, presence of CADMIUM.) Notes. 1. In working with KCN bear in mind that it is extremely poisonous. Take care not to get it on the hands; also not to breathe its fumes, especially when a solution containing it is acidified. 2. By the addition of KCN the copper salt is reduced from the cupric to the cuprous state and then combines with the excess of KCN to form the com- plex salt K + 3Cu(CN)3)2 depends upon the formation of a blue substance, whose formula is not definitely known; but it is doubtless a compound of the two oxides CoO and Al 2 Oa, and is probably cobalt alumi- nate,^Co(AlO2)2- It enables 0.5 mg. Al to be detected, or even 0.2 mg. after a little practis. No other element givs a blue color to the ash. It is essential to have the aluminum present in excess; for otherwise the blue color is ob- scured by the black oxide of cobalt. Moreover, when sodium, or potassium salts are present, the ash fuses together and the test is unsatisfactory. For this reason the sodium salts present should be completely washed out of the NH40H precipitate before dissolving it in HNOj. Procedure 55. Detection of Chromium. To the NH 4 OH solution (P. 53) add HAc, 1 cc. at a time, till the solution reddens blue litmus paper. If the solution is colorless, treat it by P. 57. If it is at all yellow, add about 10 'cc. BaCl 2 solution, and heat the mixture to boiling. (Yellow precipitate, presence of CHROMIUM.) Filter, and wash the precipitate. (Precipitate, P. 56; filtrate, P. 57.) Notes. 1. The presence of less than 0.5 mg. chromium as chromate in a volume of 50 cc. makes the solution distinctly yellow; and the addition of BaCl2 is therefore unnecessary when the solution is perfectly colorless. It is to be avoided, since BaSOi may be precipitated and has then to be removed by filtration. In doubtful cases the color of the solution should be compared with that of water. The color-test is, of course, not delicate by artificial light. 2. Since some sulfate may be present, the formation of a white precipitate with BaCl2 does not prove the presence of chromium. Whether the precipitate is pure white or yellow should therefore be carefully noted. The yellow color of a small BaCrO4 precipitate is most apparent when the precipitate has settled or when it has been collected on the filter. If there be sufficient sulfate present to obscure the yellow color of a little BaOC>4, the confirmatory test described in the next procedure will enable the chromium to be detected. P. 66 ANALYSIS OF THE ALUMINUM-GROUP. 73 Procedure 56. Confirmatory Test for Chromium. Pour repeatedly through the filter (P. 55) a warm mixture of 3 cc. HC1 and 10 cc. sat- urated SO2 solution. Evaporate the filtrate in a casserole just to dryness, taking care not to over-heat the residue ; and add a few drops of water. (Green coloration of the solution, presence of CHKOMIUM.) Note. The green color of CrCla is intense enough to enable less than 0.5 mg. of chromium in 1 cc. of solution to be detected. A yellow color may result from the action of the acids on the filter paper; but this color usually disappears on evaporating and dissolving the residue in water. Procedure 57. Detection of Zinc. Warm the HAc solution (P. 55) to 50 or 60, saturate it in a small flask with H 2 S, cork the flask, and allow it to stand for 5 or 10 minutes if no precipitate separates at once. (White flocculent precipitate, presence of ZINC.) Confirmatory Test for Zinc. Filter the H 2 S precipitate through a double filter (made by folding two niters together), wash it with a little water, and pour a 5-10 cc. portion of HN0 3 repeatedly through the filter. To the filtrate add from a dropper one drop of a Co(N0 3 ) 2 solution containing 1% Co and as many more drops of the same solution as there were estimated to be centigrams of zinc in the H 2 S precipitate. Evaporate the mixture in a casserole just to dryness to expel the acid, add 1 cc. Na 2 COs solution and 0.5 cc. more for each centigram of zinc estimated to be present. Evaporate to dryness, ignite gently by keeping the dish moving to and fro in a small flame until the purple color due to the cobalt disappears, and allow the casserole to cool. (Green color, presence of ZINC.) [If the ignited mass becomes black (owing to too strong heating), add a few drops HN0 3 , evaporate just to dryness, add the same quantity of Na 2 COs solution as was added before, evaporate and ignite gently as before.] Notes. 1. ZnS precipitates more rapidly, and in a somewhat more flocculent form, from a warm solution. Very small quantities of zinc (less than 1 mg.) may be missed unless a short time be allowed for the precipitate to coagulate; but since sulfur may then separate, the appearance of a white turbidity is not sufficient proof of the presence of zinc. The precipitate may be allowed to settle in order that the amount of zinc present may be better estimated. A double filter is used, since the ZnS is apt to pass through the filter. 2. The immediate formation of a white flocculent precipitate with H 2 S in acetic acid solution is so characteristic as to be a sufficient test for zinc. Man- ganese is the only other element of this group that forms a light-colored sulfide; and this, owing to its greater solubility in water, does not precipitate from an acetic acid solution. The confirmatory test described in the last paragraph of the procedure is, however, useful when only a small noncoagulating precipitate which may be sulfur results, or when, owing to the presence of a small quantity of other elements, the precipitate is dark-colored. In this test the use of an excess of Co(NOa) 2 is avoided, since otherwise the black color of the CoO ob- scures the green color of the cobalt zincate (CoZnO 2 ). 74 ANALYSIS OF THE IRON-GROUP. P. 61 ANALYSIS OF THE IRON-GKOUP. TABLE IX. ANALYSIS OP THE IRON-GROUP. PRECIPITATE PRODUCED BY SODIUM HYDROXIDE AND PEROXIDE: A. Phosphate absent: MnO(OH) 2 , Fe(OH) 3 , Co(OH) 3 , Ni(OH) 2 , B. Phosphate present: Also BaCO 3 , SrCO 3 , CaCO 3 , MgC0 3 , FePO 4 , Ca 3 (PO 4 ) 2 , etc. Dissolve in HNO S and H 2 2 , evaporate, heat with HN0 3 and KClOs (P. 61). Precipitate: MnO-j. Add HNO Z and bismuth peroxide (P. 62). Solution: Test a portion for phosphate with (NH^MoOi (P. 68). A. Phosphate absent: add NH*OH (P. 64). B. Phosphate present: add NH^Ac and FeCk, dilute, boil (P. 65). Precipitate: A. Fe(OH) 3 . B. Basic ferric acetate and FeP0 4 . Filtrate: add NH 4 OH, pass in HyS (P. 66). Precipitate: ZnS, CoS, NiS. See Table X. Filtrate: A. NHi salts. Reject. B. Ba, Ca, Sr, Mg. Treat by P. 81. Violet color: HMn0 4 . * This precipitate may contain all the zinc when elements of the iron-group are present in large quantity. Procedure 61. Precipitation of Manganese. Transfer the Na 2 O 2 precipitate (P. 52) to a casserole, together with the filter if neces- sary, and add 5-15 cc. HNO 3 . If there is still a residue, heat the mix- ture nearly to boiling and add 3% H 2 2 solution, a few drops at a time, - - stirring after each addition, till the residue is dissolved. Filter to remove the paper, and evaporate the filtrate almost to dryness. Add 10 cc. 16-normal HNO 3 and about 1 g. of powdered KC1O 3 , and heat to boiling. (Black precipitate, presence of MANGANESE.) If there is no precipitate, evaporate the solution almost to dryness, add 25 cc. water, and treat a part of the solution by P. 63 and the remainder by P. 64 or P. 65. If there is a precipitate, add to the mixture 10 cc. 16-normal HN0 3 , pour it into a 200 cc. conical flask, heat it to boiling, and add grad- ually (0.5 g. at a time) 3 g. powdered KC10 3 . Grasp the neck of the flask with a strip of stiff paper, and keep the flask in motion over a small flame so that the mixture boils gently for one or two minutes. Filter with the aid of suction through an asbestos filter, made by plac- ing in a funnel enough glass-wool to form a wad 1 cm. high, tamping it down gently with the finger, and pouring through it a suspension of asbestos in water, enough to form an asbestos layer about | cm. thick. Add to the filtrate 1 g. KC10 3 . If more precipitate separates, boil the P. 61 ANALYSIS OF THE IRON-GROUP. 75 mixture gently for a minute or two, and then filter it through the same asbestos filter. Wash the precipitate two or three times with 16-nor- mal HNOs which has previously been freed from oxides of nitrogen by warming it with a little KClOs, and treat the precipitate by P. 62. Evaporate the filtrate to about 5 cc., but not further, dilute it with 25 cc. water, and treat a part of it by P. 63 and the remainder by P. 64 or P. 65. Notes. 1. Pure concentrated HNOs does not dissolve hyd rated MnOj; but it may do so in the presence of filter-paper, whereby the HNOs is reduced to lower oxides. The action is rapid in the presence of E^CV, for the MnOj is thereby quickly reduced to Mn(NOs)2 with evolution of Og. 2. By HClOs in HNOs solution (but not by HNOs alone) manganous salts are rapidly oxidized to hydrated MnC>2 with formation of chlorin dioxide (C102), which escapes as a yellow gas. 3. The separation of manganese in this way from the other metals of this group is entirely satisfactory with the exception that a small quantity of iron (up to 1 mg.) may be completely carried down with a large quantity (500 mg.) of manganese. Procedure 62. Confirmatory Test for Manganese. Pour through the filter containing the KClOs precipitate (P. 61) 5 cc. hot HNOs to which 4 or 5 drops of 3% H 2 2 solution have been added. Collect the filtrate in a test-tube; cool it; add to it solid bismuth dioxide, 0.1 g. at a time, till some of the brown solid remains undissolved; and let the solid settle. (Purple solution, presence of MANGANESE.) Notes. 1. This confirmatory test for manganese is usually superfluous, since the precipitation of MnO2 by HClOs is highly characteristic. 2. In the presence of HNOs, MnO 2 is reduced by H2O2 with evolution of oxygen and formation of Mn(NOs)2. 3. Commercial bismuth dioxide, also often called sodium bismuthate, is a mixture of bismuth compounds which probably owes its oxidizing power to the presence of the dioxide Bi02. When it is not available, PbC>2 may be substituted for it; but in that case the mixture must be boiled for 2 or 3 minutes. Procedure 63. Test for Phosphate. Pour about one-tenth of the HNOs solution (P. 61) into two or three times its volume of ammonium molybdate reagent, and heat the mixture to 60-70. (Fine yellow precipitate, presence of PHOSPHATE.) If there is no precipitate, or only a very small one, treat the remainder of the HNOs solution by P. 64; otherwise by P. 65. Notes. 1. Phosphate is tested for at this point because a different treatment is necessary when it is present in significant amount, in order to separate from it alkaline-earth elements and to provide for their detection. When phosphate is not present, iron can be separated from nickel and cobalt by NH44 as a yellowish white precipitate. Upon boiling, the excess of iron separates completely as a dark red gelatinous precipitate of basic ferric acetate, leaving the supernatant liquid colorless, except when nickel or cobalt is present. 4. The solution is diluted to at least 100 cc., owing to the large volume of the precipitate; and it is heated in a capacious flask, owing to its tendency to boil over. Procedure 66. Precipitation of Zinc, Nickel, and Cobalt. Into the ammoniacal solution (P. 64 or P. 65) pass H^S gas until the mixture after shaking blackens PbAc 2 paper held above it. (Black precipi- tate, presence of NICKEL or COBALT.) Filter, wash the precipitate with water containing a very little (NH^S, and treat it by P. 67. Treat the filtrate by P. 81-89 if phosphate or much chromium was found present (in P. 63 or P. 55); otherwise, reject it. Notes. 1. In precipitating NiS, the use of H 2 S has the advantage that the nickel is all thrown down at once, while with (NKi^S some of it usually remains in the solution, giving it a dark brown color. If (NKi^S be used, the filtrate must be boiled to throw down the unprecipitated nickel, as described in P. 51. 2. The filtrate must be treated by the procedures for detecting alkaline- earth elements when phosphate has been found to be present, since the whole quantity of these elements present in the substance may then be contained in this filtrate, for the reasons stated in Note 6, P. 51, and Note 2, P. 65. The filtrate should also be so treated when much chromium is present, since all the magnesium present may have been carried down with it in the original precipitate. 78 ANALYSIS OF THE IRON-GROUP. P. 67 TABLE X. SEPARATION OF ZINC, NICKEL, AND COBALT. HYDROGEN SULPHIDE PRECIPITATE I ZnS, NiS, CoS. Treat with cold dilute HCl (P. 67}. Solution: ZnCl 2 , NiCl 2 *, CoCl 2 * Add NaOH and Na 2 2 (P. 67). Filtrate: Na 2 ZnO 2 . Add HAc and White precipitate: ZnS. Precipitate: Ni(OH) 2 , Co(OH) 3 , Add HCl, evaporate (P. 68). Residue: NiS, CoS. Dissolve in HCl and HNOs, evaporate (P. 68). Residue : NiCl 2 , CoCl 2 . Add HCl and ether (P. 68) . Yellow residue: NiCl 2 . Dissolve in water, add tartaric acid, NaOH, and HzS (P. 69). Brown coloration: presence of nickel. Blue solution: CoCl 2 . Evaporate, add HAc and KN0 2 (P. 70). Yellow precipitate: K 3 Co(NO 2 ) 6 . * A small proportion of the nickel and cobalt present always dissolves in the dilute HCL Procedure 67. Separation of Zinc from the Nickel and Cobalt. Transfer the H 2 S precipitate (P. 66) with the filter to a casserole, and add 10-30 cc. 1-normal HCl. Stir the cold mixture frequently for 5 minutes, and filter. Wash the residue and treat it by P. 68. Boil the HCl solution until the H 2 S is completely expelled, add NaOH solution until the mixture is slightly alkaline, transfer to a casserole, cool, and add 0.5-1 cc. Na 2 C>2 powder, a small portion at a time. Boil for several minutes to decompose the excess of Na 2 02, cool the mixture, and filter. Wash the precipitate thoroly, and treat it by P. 68, uniting it with the sulfide residue undissolved by the dilute HCl. Acidify the filtrate with HAc, and test it for ZINC by P. 57. Notes. 1. This treatment with dilute HCl serves to extract almost com- pletely the zinc which may be present in this precipitate owing to its having been carried down in the Na20 2 precipitate, as described in P. 52, Note 6. A small proportion of the nickel and cobalt present (5-20%) always dissolves in the dilute HCl, and the subsequent treatment with Na^O 2 serves to separate these elements from the zinc. This separation is satisfactory when, as in this case, the nickel and cobalt are present in small quantity; for then only an insignificant amount of zinc is carried down with them. When, therefore, the H 2 S precipitate is small, it may, instead of being treated with dilute HCl, be dissolved at once in aqua regia and the solution treated directly as described in the last paragraph of this Procedure. 2. This Procedure must always be followed in order to determin whether or not zinc is present in the substance, unless a satisfactory test for it has already been obtained in P. 57, or unless the original Na-jC^ precipitate (P. 52) P. 68 ANALYSIS OF THE IRON-GROUP. 79 was small. In either of these two cases this Procedure may be omitted and the H 2 S precipitate (P. 66) treated directly by P. 68. 3. The fact that NiS and CoS do not dissolve readily in 1-normal HC1 seems inconsistent with the non-precipitation of nickel and cobalt by B^S with the copper and tin groups from a solution which is only 0.3 normal in acid. This behavior probably arises from the fact that these sulfides exist in at least two allotropic forms of different solubilities. The form that tends to be produced by direct union of sulfide-ion with nickel-ion or cobalt-ion is soluble in dilute acid. The precipitate produced by (NH^S consists, however, only in part of this soluble form: it contains also a large proportion (increasing with the time that the precipitate has stood) of a form that is nearly insoluble in dilute acids. Procedure 68. Separation of Nickel and Cobalt. Transfer the sulfide residue undissolved by dilute HC1 and the thoroly washed Na 2 O2 precipitate (P. 67) with the filters to a casserole, add 5-15 cc. HC1 and a few drops HNOs, warm until the black precipitates are dissolved, and filter off the paper. Evaporate the solution just to dryness moisten the residue with 12-normal HC1, and evaporate again just to dryness. To the dry residue add 0.5-2.0 cc. 12-normal HC1; and, if it does not all dissolve, warm it gently with constant stirring. Cool, add 10 cc. ether previously saturated with HC1 gas, and stir for 1-3 minutes. Filter the mixture through a dry filter, and wash out the filter with 5-10 cc. of the HCl-ether reagent. (Yellow residue, presence of NICKEL; blue solution, presence of COBALT.) (Residue, P. 69; solution, P. 70.) Notes. 1. The nickel and cobalt must be present as chlorides in order to be separated by this process. The two evaporations with an excess of HC1 serve to destroy any nitrate present. 2. Even 500 mg. of either nickel or cobalt dissolve in 2 cc. 12-normal HC1 on gentle warming; and the residue should be so dissolved before adding the HCl-ether reagent. 500 mg. of cobalt will then remain in solution on the addi- tion of 10 cc. of that reagent. 3. Even 0.5 mg. of nickel yields a distinct yellow residue; and less than 0.5 mg. of cobalt imparts a pronounced blue coloration to the ethereal solution. It is therefore unnecessary to try the confirmatory test for cobalt when the ethereal solution is colorless. 4. CoCb in aqueous solution is pink, like Co(NC>3)2 or CoSO4, when its concentration is small; but it is blue, even in the cold, when its concentration is very large. It is blue, even when its concentration is only moderately large, in hot solutions, or in cold solutions which contain HC1 or other chlorides at a high concentration. It is also blue in organic solvents. The pink color is probably due to the cobalt-ion, which is doubtless hydrated (existing per- haps as Co(H 2 O)4 ++ in analogy with Co(NH 3 )4 ++ ), and to the hydrated un- ionized cobalt chloride which results directly from the union of this ion with chloride-ion. The blue color is probably due to anhydrous CoCk, or to com- plex anions, like Cods", which result from the union of it with chloride-ion. 80 ANALYSIS OF THE IRON-GROUP. P. 69 Procedure 69. Confirmatory Test for Nickel. Pour through the filter containing the residue undissolved by ether (P. 68) 10 cc. water, add 3-5 cc. 10% tartaric acid solution, neutralize with NaOH solu- tion, and add 2 cc. in excess. Pass in H 2 S gas for about 1 minute, filter out any precipitate that may form, and saturate the filtrate with H 2 S. Filter again if there is a precipitate. (Brown coloration, presence of NICKEL.) Notes. 1. When an alkaline tartrate solution containing a very little nickel (even 0.1-0.2 mg. in 20 cc.) is saturated with H 2 S, a clear brown solution is obtained. With somewhat larger amounts of nickel (10-20 mg.) the liquid is opaque, but runs through a filter very readily. The condition of the nickel in this solution is not known. The presence of the tartrate serves merely to prevent the precipitation of Ni(OH)2 by the NaOH solution, owing to the formation of a complex salt containing the nickel in the anion. The brown color does not appear until the alkaline solution is nearly saturated with H 2 S, so that care must be taken to use an excess of H 2 S. 2. This confirmatory test for nickel is not interfered with by moderate amounts of other elements of this group, such as cobalt and iron; for on leading H 2 S into an alkaline tartrate solution containing these elements, they are completely precipitated as sulfides and may be filtered off, yielding a filtrate which, when saturated with H 2 S, remains colorless if nickel is absent, but be- comes dark brown if it is present in even small amount. Procedure 70. Confirmatory Test for Cobalt. Heat the ethereal solution (P. 68) in a casserole on a steam-bath, or by floating the casserole in a vessel of nearly boiling water, till the ether is expelled; then evaporate it just to dryness over a small flame. To the residue add about 5 cc. water, and a few drops of NaOH solution. To the mixture add 5 cc. HAc and then 30 cc. 3-normal KNO 2 solution; and allow the mixture to stand at least ten minutes if no precipitate forms sooner. (Fine yellow precipitate, presence of COBALT.) Notes. 1. The precipitate is potassium cobaltic nitrite, 3KN0 2 .Co(NO 2 )s, or more properly, potassium cobaltinitrite, K3Co(NO 2 )e, since in solution it dissociates into K+ and the complex anion CotNO^e^. In the formation of this substance the cobaltous salt is oxidized to the cobaltic state by the nitrous acid displaced from its salt by the acetic acid, the cobaltic salt combining as fast as formed with the potassium nitrite. 2. The precipitate is somewhat soluble in water, but very difficultly soluble in a concentrated KNO 2 solution, owing to the common-ion effect of the potas- sium-ion. The formation of the KsCo(NO 2 )6 precipitate takes place slowly; but, even when only 0.1-0.2 mg. of cobalt is present, a distinct precipitate results within 10 minutes. 3. Nickelous salts are not oxidized by nitrous acid, and they are not precip- itated by KNO 2 , unless a very large quantity of nickel is present, in which case a dark-yellow or dark-red precipitate of potassium nickelous nitrite, K4Ni(NO 2 )e, may separate. P. 81 ANALYSIS OF ALKALINE-EARTH GROUP. 81 PRECIPITATION AND ANALYSIS OF THE ALKALINE-EARTH GROUP. TABLE XI. ANALYSIS OF THE ALKALINE-EARTH GROUP. AMMONIUM CARBONATE PRECIPITATE: BaCO 3 , SrCO 3 , CaCO 3 , Dissolve in HAc, add NH+Ac and KzCrO* (P. 82}. Precipitate: BaCrO 4 . Dissolve in HCl, evaporate (P. 88). Filtrate. Add NHtOH and alcohol (P. 84). Precipitate: SrCr0 4 . Treat with (NH^CO Z (P. 85). Filtrate: Ca and Mg salts. Add (ATtf 4 ) 2 C 2 4 (P. 86). Test in flame Add HAc, NHtAc, and KzCrO*. Precipitate: CaQA. Dissolve in dilute H 2 SOt, add alcohol (P. 87). Filtrate. Add NH&H and Na 2 HPOi (P. 88). Residue: SrCO 3 . Dissolve in HAc, add CaSO*. Green color : Ba. Precipi- tate: BaCrO 4 . Precipitate: MgNH 4 PO 4 . Precipitate : SrSO 4 . Precipitate: CaS0 4 . Procedure 81. Precipitation of the Alkaline-Earth Group. Evaporate the filtrate from the NH 4 OH and (NH 4 ) 2 S precipitate (P. 51) to a volume of about 10 cc., and filter off the sulfur. To the cold solution add 15 cc. (NH 4 ) 2 C0 3 reagent and 15 cc. 95% alcohol; and, if a large precipitate results, add 15 cc. more of each of these liquids. Shake the mixture continuously for 10 min- utes; or, let it stand, with occasional shaking, for at least half an hour. (Precipitate, presence of ALKALINE-EARTH ELEMENTS.) Filter, and wash the precipitate with a little (NH 4 ) 2 CO3 reagent, using suction if the precipitate is large. (Precipitate, P. 82; filtrate, P. 91.) Notes. 1. The nitrate from the (NH 4 ) 2 S precipitate is evaporated in order that the elements of the alkaline-earth group may be precipitated more quickly and more completely. The evaporation also serves to destroy (NHU) 2 S and to coagulate any sulfur that may separate. The volume to which the (NH^COs reagent is added should not exceed 10 cc. 2. The (NH4) 2 CO 3 reagent is prepared by dissolving 250 g. freshly powdered ammonium carbonate in 1 liter 6-normal NHiOH. 3. If the ammonium carbonate and hydroxide were added in only small excess, the precipitation of CaC0 3 , SrCO 3 , and BaCOs would not be complete, and additional tests for small quantities of these elements would have to be made in the nitrate. But, by the use of a concentrated solution of (NH 4 ) 2 CO 3 containing a large excess of NBUOH (so as to diminish the hydrolysis of the carbonate into (NH^+HCOr and NHiOH), the pre- 82 ANALYSIS OF ALKALINE-EARTH GROUP. P. 81 cipitation may be made practically complete, owing to the greatly increased concentration of carbonate-ion (COa"). 4. When the concentrations of (NH^COs and NH 4 OH are sufficiently great, magnesium is in the cold also completely precipitated. The precipitate, which is in this case a double carbonate, MgCOs^NH^COa^EkO, is, however, fairly soluble in cold water and readily soluble in hot water. 5. From a cold aqueous solution the precipitation of these elements takes place slowly, especially in the case of magnesium and calcium; but it is greatly accelerated by the addition of alcohol and by shaking. Under the conditions recommended in the procedure 0.5 mg. of any of the four elements is easily detected. Procedure 82. Precipitation of Barium Dissolve the (NH 4 ) 2 CO S precipitate (P. 81) by pouring repeatedly through the filter a 5-15 cc. portion of hot HAc, and evaporate the solution just to dryness, taking care not to ignite the residue. Add to the residue 3 cc. HAc, 20 cc. 3-normal NH^c, and 15 cc. water; and heat the solution to boiling in a flask. Measure out 8 cc. 3-normal K2Cr0 4 solution, and add it a few drops at a time, heat- ing and shaking after each addition. Finally, heat the mixture at 90-100 for 1 or 2 minutes, shaking at the same time. Filter, even tho the solution appear clear; remove the filtrate, and wash the precipitate thoroly with cold water. (Pale yellow precipitate, presence of BARIUM.) (Precipitate, P. 83; filtrate, P. 84.) Notes. 1. The solubility in water of the chromates of the alkaline-earth elements increases rapidly in the order, Ba, Sr, Ca, Mg, as shown in the Table on page 124. The difference in solubility of BaCrO4 and SrCrO4 is so great that under the conditions of the procedure 0.5 mg. Ba can be detected, while even 400 mg. Sr giv no precipitate. The amount of K 2 CrO4 added is sufficient to precipitate completely more than 500 mg. of barium. 2. Acetic acid is added to increase the solubility of SrCrC^. By its action the concentration of the chromate-ion is decreased, owing to its conversion partly into hydrochromate-ion and partly into bichromate-ion, according to the reactions: CrOr + H+ = HCrO 4 -; and 2HCrO 4 - = H 2 O + Cr 2 O 7 ". It is evident that the ratio of the CrO 4 ~ to the HCrO 4 ~ concentration must decrease as the H + concentration increases. For this reason the presence of an excess of a largely ionized acid (such as HC1 or HNOa) would prevent the pre- cipitation of BaCrO 4 ; but since acetic acid is a slightly ionized acid, and since a large amount of acetate is present, the addition of a considerable excess of acetic acid has but little effect. 3. The K 2 Cr0 4 is added slowly to the hot solution and the mixture is shaken and heated in the neighborhood of 100 before filtering, since otherwise the precipitate is liable to pass through the filter. By this method of precipitation almost all the barium is precipitated before an excess of K 2 CrO 4 is added. This is of importance, since, when much barium is present, as much as 3 mg. P. 83 ANALYSIS OF ALKALINE-EARTH GROUP. 83 Sr may be carried down completely if the K2(M)4 reagent be added quickly. If for any reason the filtrate is turbid after two or three filtrations, the pre- cipitate may be coagulated by boiling gently for 1 or 2 minutes. Vigorous or long- continued boiling is to be avoided, since, owing to loss of acetic acid, SrCrCU may then separate if much strontium is present. When less than 1 mg. Ba is present it is very difficult to distinguish the faint turbidity in the colored solution; but the pale yellow precipitate can be seen after filtering and washing the K2CrO4 out of the filter. The precipitate must be washed thoroly, in order to remove strontium as completely as possible, which otherwise would obscure the flame coloration in the confirmatory test for barium. Procedure 83. Confirmatory Test for Barium. Pour repeatedly through the filter containing the K2Cr04 precipitate (P. 82) a 5-10 cc. portion of hot HC1, and evaporate the solution just to dryness in a small casserole. Heat a platinum wire having a small loop at one end in a gas flame till it no longer colors the flame, dip it in HC1, touch it to the residue in the casserole, and introduce it again into the flame. (Green color, presence of BARIUM.) If this flame test is in- conclusiv, treat the residue by the second paragraph of P. 82. (Yel- low precipitate, presence of BARIUM.) Notes. 1. When the amount of barium present is very small, only a mo- mentary green color is seen as the yellow sodium color which first appears fades away. The only other elements that color the flame green are copper, boron, and thallium. Strontium givs a crimson and calcium an orange-red color. 2. If a small quantity of SrCrC>4 was precipitated in P. 82, owing to the pres- ence of a very large amount of strontium in the substance or owing to failure to follow the directions, it will not again precipitate in this second treatment; for the quantity of strontium now present in the solution is much less than before. A yellow precipitate obtained in this second treatment is therefore conclusiv evidence of the presence of barium. Procedure 84. Precipitation of Strontium. To the filtrate (P. 82), after cooling it, add NEUOH slowly until the color of the solution changes from orange to yellow, and then 10 cc. more. Dilute the solution to 65 cc., and add slowly, with constant shaking, 50 cc. 95% alcohol, and cool the solution. If a large precipitate forms or if a large precipitate of BaCr0 4 separated in P. 82, add 4 cc. 3-normal K 2 Cr04 solution and 10 cc. 95% alcohol, and shake. (Light yellow precipitate, presence of STRONTIUM.) Filter after 5-10 minutes with the aid of suction; suck the precipitate as dry as possible, but do not wash it. (Precipitate, P. 85; filtrate, P. 86.) Notes. 1. Under these conditions 1 mg. of strontium yields a precipitate, while even 400-500 mg. of calcium or magnesium do not do so. A moderate change in the conditions will not affect this result; but if the con- centration of alcohol or K2CrO4 is much less than is recommended, the pre- cipitation of strontium may be incomplete; while the addition of larger amounts 84 ANALYSIS OF ALKALINE-EARTH GROUP. P. 84 of alcohol and K2CrO4 may cause the precipitation of chromate of calcium or magnesium if much of these elements is present, or of K2CrO4 itself, since the latter is not very soluble in alcohol. The confirmatory test should therefore always be tried. 2. When a large quantity of strontium or barium is present, the K2OO4 added in P. 82 may have been so far removed from the solution that the stron- tium is not at first completely precipitated. In such cases, in order to ensure its complete precipitation, more K 2 CrO4 is added in this Procedure. 3. The precipitate is not washed, since SrCrO4 is fairly soluble in water. Procedure 85. Confirmatory Test for Strontium. If the K 2 Cr0 4 precipitate (P. 84) is small, pour three or four times through the filter containing it a hot mixture consisting of 10 cc. (NH 4 ) 2 C03 reagent and 5 cc. 3-normal K 2 Cr04 solution. If the K2Cr04 precipitate is large, transfer it to a casserole, heat it for 4-5 minutes with 15 cc. of the same mixture, and filter. Wash the residue thoroly with cold water. Pour repeatedly through the filter containing it a cold 5-10 cc. portion of 1-normal HAc. Evaporate the solution just to dryness, add 2-3 cc. water, transfer the solution to a test-tube (pouring it through a very small filter, if it is not perfectly clear), add 3 cc. satu- rated CaS04.2H 2 solution, and heat in a vessel of boiling water for at least ten minutes. (Fine white precipitate, presence of STRONTIUM.) Note. The mixture of (NH 4 ) 2 CO 3 and K 2 CrO 4 converts SrCrO 4 into the less soluble SrCO^; but it has no action on any BaCrC>4 that may be present, and only very little of any such BaCrO4 dissolves in the subsequent treatment with HAc. Any CaCrO4 present is converted by the mixture into CaCOs, and this dissolves in the HAc; but a solution of a calcium salt givs no pre- cipitate with saturated CaSO4 solution (tho it might, of course, do so with a concentrated solution of another sulfate, like K 2 SO4). Procedure 86. Precipitation of Calcium. To the filtrate from the rOi precipitate (P. 84) add 60 cc. water, heat to it boiling, add to it slowly 60 cc. boiling-hot (NH 4 ) 2 C 2 04 solution, and keep it nearly boiling for 5 minutes. (White precipitate, presence of CALCIUM.) Filter the mixture while it is still hot; and wash the precipitate once or twice with water. (Precipitate, P. 87; filtrate, P. 88.) Note. The solution is heated to boiling, and the ammonium oxalate solution is added slowly, in order to cause the CaC2O4 to precipitate in the form of coarser particles which can be more readily filtered. Moreover, the solution must be heated and must be kept hot, in order to prevent the precipitation of some MgC2C>4, which would precipitate when much magnesium is present if the mixture were allowed to cool. Procedure 87. Confirmatory Test for Calcium. Pour repeatedly through the filter containing the (NH 4 ) 2 C 2 04 precipitate (P. 86) a cold 5 cc. portion of H 2 S04; add 10-15 cc. 95% alcohol, and let the mixture stand for several minutes. (White precipitate, presence of CALCIUM.) P. 87 ANALYSIS OF ALKALINE-EARTH GROUP. 85 Notes. 1. CaC2O4.H 2 O is very difficultly soluble in water, but dissolves in dilute solutions of largely ionized acids, owing to the formation by metathesis of unionized HC 2 O4~. CaSO4 is somewhat soluble in dilute H2SO4, but is com- pletely thrown out as a flocculent precipitate by the addition of the alcohol. 2. One milligram of calcium produces a turbidity at once, 0.5 mg. in 1-3 minutes, and 0.2 mg. within 10 minutes. Even a large amount of magnesium does not interfere with the test. If strontium were present, a small amount of it would dissolve in the H 2 SO4, but only enough to giv a slight turbidity on the addition of alcohol, corresponding to that given by 0.2-0.3 mg. of calcium after standing a few minutes. Therefore anything more than a slight turbidity ia a conclusiv proof of the presence of calcium. Procedure 88. Detection of Magnesium. To the filtrate from the (NH 4 ) 2 C 2 4 precipitate (P. 86) add 10 cc. 15-normal NH 4 OH and 25 cc. Na 2 HP04 solution; cool, and shake the mixture; if no precipi- tate forms, let the mixture stand for at least half an hour, shaking it frequently. (White precipitate, presence of MAGNESIUM.) Filter out the precipitate, wash it once with alcohol, and treat it by P. 89. Note. This test for magnesium depends upon the precipitation of magne- sium ammonium phosphate, Mg(NEL|)PO4. This salt is fairly soluble even in cold water, owing chiefly to hydrolysis into NH 4 OH and Mg ++ HPO4~; and the test is therefore made in a strongly ammoniacal solution. Since the solubility increases rapidly with the temperature, the solution is cooled to the room temperature, or below. In an aqueous solution this substance shows a great tendency to form a supersaturated solution, and it is therefore usually directed to make the test in as small a volume as possible. In the presence of alcohol, however, precipitation takes place rapidly, and even 0.5 mg. of mag- nesium produces a distinct turbidity in a solution containing 190 cc. water and 50 cc. alcohol within half an hour. A small precipitate of this kind settles out on further standing and may then be detected by rotating the solution so as to cause the precipitate to collect in the center. Procedure 89. Confirmatory Test for Magnesium. Pour repeatedly through the filter containing the Na 2 HP0 4 precipitate (P. 88) a 5 cc. portion of 2-normal H 2 S0 4 ; add to the solution 20 cc. 95% alcohol, and shake it continuously for two or three minutes. Filter, if there is a precipitate; add to the filtrate 10 cc. water, 20 cc. NH 4 OH, and 5 cc. Na 2 HPO4 solution; and let the mixture stand at least half an hour. (White crystalline precipitate, presence of MAGNESIUM.) Notes. 1. This confirmatory test should be tried whenever Na 2 HPC>4 pro- duces a small precipitate that is not distinctly crystalline. For, if even a small quantity of strontium or calcium failed to be precipitated in P. 84 or 86, a flocculent precipitate of Sra(P04) 2 or Ca3(PO4)2 would come down on the addition of Na^HPCU. 2. The addition of H 2 SO 4 and alcohol precipitates strontium and calcium so completely that any precipitate (more than a very slight turbidity) produced on adding Na2HPO 4 to the HgSOi solution can not be due to these elements. 86 ANALYSIS OF THE ALKALI-GROUP. ANALYSIS OF THE ALKALI-GROUP. P. 91 TABLE XII. ANALYSIS OP THE ALKALI-GROUP. FILTRATE FROM AMMONIUM CARBONATE PRECIPITATE : NIL;, K, Na Salts. Evaporate and ignite the residue (P. 91). Vapor: NH4 salts. Residue: KC1, NaCl. Add HCIO*, evaporate, add alcohol (P. 92). Residue: KClOi. Dissolve in hot water, addNa z Co(NO z )s(P.9S). Solution: NaClO 4 . Saturate with HCl gas (P. 94). Precipitate: NaCl. Dissolve in water, add K 2 H 2 Sbz07 (P. 95). Yellow precipitate: K 2 NaCo(N0 2 )6. Crystalline precipitate: Na2H 2 Sb2O7. Procedure 91. Removal of Sulfate and of Ammonium Salts. Evaporate the filtrate from the (NH 4 ) 2 CO 3 precipitate (P. 81) to a volume of 10-15 cc. till it no longer smells of ammonia. To 10 drops of the filtrate add 4-5 drops HCl and 1-2 cc. BaCl 2 solution. (White precipitate, presence of SULFATE.) If BaCl 2 produces no precipitate, treat the rest of the filtrate by the last paragraph of this Procedure. If BaCl 2 produces a precipitate, add to the rest of the filtrate 5-10 cc. BaCl 2 solution, heat the mixture to boiling, and filter out the pre- cipitate. To the filtrate add 5-15 cc. (NH 4 ) 2 CO3 reagent, heat the mixture to boiling, filter out the precipitate, and and treat the fil- trate by the last paragraph of this Procedure. Evaporate the filtrate to dryness in a small casserole, and ignite the residue, at first gently, then just below redness, until no more white fumes come off, keeping the dish in motion over a flame and taking care to heat the sides as well as the bottom of the dish. Cool completely, add 5 cc. water, filter the mixture through a 5-cm. filter, evaporate the filtrate to dryness in a small casserole, and heat the Pish at a temperature just below redness. (White residue, presence of POTASSIUM or SODIUM.) Treat the residue by P. 92. Notes. 1. If sulfate is present, either because it was a constituent of the substance or because H 2 SO 4 was used in the preparation of the solution (in P. 5 or P. 8), it must be removed before attempting to separate potassium and sodium by P. 92. The process of removing it involves its precipitation as BaSO 4 by the addition of BaCl 2 , the removal of the excess of barium by the subsequent P. 91 ANALYSIS OF THE ALKALI-GROUP. 87 addition of (NH^COs, and finally the volatilization of the ammonium salts by ignition. If phosphate is present, it is also removed by the addition of BaCk to the neutral solution; but, as it is not necessary to remove it, the preliminary test with BaCl2 is made in HC1 solution, in which case sulfate alone produces a precipitate. 2. Great care must be taken to volatilize the ammonium salts completely, since even 1 mg. of ammonium would giv a precipitate in the subsequent test for potassium in P. 92. To ensure their removal the residue is ignited twice. The dish must not, however, be allowed to become red-hot during the ignition, since at that temperature potassium and sodium chlorides are somewhat volatil. 3. The formation of a precipitate with BaClz does not necessarily show that sulfate is present in the original substance, even when H 2 SO 4 has not been used in preparing the solution; for it may have been produced by the oxidation of sulfide, when the substance contains it, in preparing the solution in P. 3 and 4; or it may even be formed in small quantity by the oxidation of the H 2 S and (NH^S used as reagents. 4. A brown or black residue of organic matter, coming from impurity in the ammonium salts added in the course of analysis and from the alcohol and filter paper, may remain upon treating the ignited residue with water. There may also be a white residue of silica, coming from the action of the reagents on the glass and porcelain vessels throughout the course of the analysis. 5. Since in the dry form even a residue that seems very small may correspond to an appreciable quantity of potassium or sodium, the subsequent tests for these elements should be made if there is any residue whatever after the final ignition. Procedure 92. Separation of Potassium and Sodium. To the ig- nited residue (P. 91) add 5-15 cc. 2-normal HC1O 4 ; and evaporate, by keeping the dish in motion over a small flame, till thick white fumes of HC1O 4 come off copiously (which occurs when the liquid is reduced to about one-fourth of its original volume). Cool completely, add 10-20 cc. 95% alcohol, and stir the mixture for 2-5 minutes if there is much residue. If there is still a res 'due, add 3 cc. 2-normal HC104. (White residue, presence of POTASSIUM.) Filter through a dry filter-paper, and wash the residue with 95% alcohol. (Precipi- tate, P. 93; filtrate, P. 94.) Notes. 1. Enough HC1O4 must be added to convert the potassium and sodium chlorides completely into perchlorates, and the evaporation must be con- tinued till all the HC1 is expelled; for otherwise NaCl, being insoluble in alcohol, may be left as a residue with the KClOi. An unnecessary excess of HC1O4 is, however, to be avoided, since it makes the subsequent test for sodium somewhat less delicate. The quantity added is therefore varied (from 5 to 15 cc.) in ac- cordance with the size of the ignited residue obtained in P. 91. 2. The evaporation is continued till the HC1O4 fumes strongly, also for the purpose of removing most of the water; for this test for potassium and the subsequent test for sodium (in P. 94) are more delicate, the less the quantity of water present. When the directions given in the Procedure are followed, 1-1 J mg. of potassium produces a distinct precipitate. 3. If sulfate were present in the (NH 4 ) 2 CO8 filtrate and it were not removed 88 ANALYSIS OF THE ALKALI-GROUP. P. 92 in P. 91 by the addition of BaCk, this separation of potassium and sodium would be unsatisfactory; for, when sodium is present, Na2SC>4 would remain in the residue undissolved by the alcohol. This arises from the fact that H2SO4 is less volatil than HC1O4 and is therefore not expelled by it in the evaporation, and from the fact that Na2SC>4 is only slightly soluble in alcohol even in the presence of HC1O4. The presence of phosphate or borate does not, however, interfere with the separation; for, tho phosphoric and boric acids are not volatilized in the evaporation with HC1O4 and tho their sodium salts are very slightly soluble in pure alcohol, yet these salts are metathesized by the excess of per- chloric acid, since this acid is much more largely ionized than phosphoric or boric acid. The sodium therefore remains dissolved in the alcoholic solution. 4. A small quantity of sulfate is usually produced when the (NH 4 ) 2 COj filtrate is evaporated to dryness and the residue is ignited, owing to the action of the nitrate present on sulfur-compounds coming from the decomposition of the (NH^S reagent. To redissolve any precipitate of NazSC^ arising from this small quantity of sulfate, 3 cc. more of 2-normal HCICX are added at the end of the Procedure when there is a residue undissolved by the alcohol. Procedure 93. Confirmatory Test for Potassium. Pour repeatedly through the filter containing the HC104 precipitate (P. 92) a 5-10 cc. portion of boiling water, add to the solution 3-5 cc. NaNOa solution and 2-3 cc. HAc, and boil the mixture gently in a casserole for 5-6 min- utes. Cool the mixture, add to it 5 cc. Na 3 Co(NO 2 )6 reagent, and let it stand for 10 minutes. (Yellow precipitate, presence of POTASSIUM.) Notes. 1. This confirmatory test should not be omitted, owing to the pos- sibility that, "in consequence of imperfect manipulation, the residue left undis- solved by the alcohol in P. 92 consists of NH 4 C1O 4 , NaCl, or Na2SO 4 . The NH 4 C1O4 may result from incomplete removal of the ammonium salts in the ignition, the NaCl from incomplete conversion of the chlorides into percMorates in the evaporation with HCIO^, and the Na2SC>4 from the presence of sulfate which was not removed by the use of BaCl2. 2. The boiling with NaNO 2 and HAc serves to destroy any small quantity of ammonium (up to 20-30 mg.) which might possibly have escaped volatiliza- tion in the ignition. It is destroyed in virtue of the reaction NH4 + C1~+HNO2= N 2 +2H 2 O+H + C1~. If ammonium is present in the solution tested with NasCo(NO2)6, it yields a precipitate closely resembling that produced by potas- sium. 3. By this test the presence of 0.3 mg. of potassium may be detected within 510 minutes. The yellow color of the precipitate can be best seen by collecting it on a filter, and washing out the solution thoroly with cold water. Procedure 94. Detection of Sodium. Pour the alcoholic filtrate (P. 92) into a dry conical flask placed in a vessel of cold water, and pass into it a fairly rapid current of dry HC1 gas until the gas is no longer absorbed. (White precipitate, presence of SODIUM.) Treat the precipitate by P. 95. Notes. 1. This test for sodium is most delicate when the alcohol is com- pletely saturated with HC1 gas, in which case 1 mg. of sodium can be detected. P. 94 ANALYSIS OF THE ALKALI-GROUP. 89 2. The dry HC1 gas may be prepared by dropping 96% H 2 SO 4 from a separating funnel into a flask containing solid NaCl covered with 12-normal HC1, and passing the gas through a gas- wash-bottle containing 96% HzSOi. Such a gas-generator may be conveniently kept ready for use in a hood in the laboratory, as the evolution of gas soon ceases when no more H 2 SO 4 is added. 3. The alcoholic filtrate containing the excess of HC1O4 must not be evapo- rated, since a dangerous explosion is likely to result. Procedure 95. Confirmatory Test for Sodium. Filter out the HC1 precipitate (P. 94), wash it with a little 95% alcohol, pour a 5-10 cc. portion of water repeatedly through the filter, evaporate the solution just to dryness, add 1 cc. water and 2 cc. K 2 H 2 Sb2O7 reagent, pour the mixture into a test-tube, and let it stand for at least half an hour, pref- erably over night. (White crystalline precipitate, presence of SODIUM.) Notes. -1. The sodium pyroantimonate (Na2H2Sb2O7) separates as a heavy crystalline precipitate, which usually adheres in part to the glass in the form of distinct crystals, which can be best seen by inverting the test-tube. The test is not extremely delicate; but 1 mg. of sodium is easily detected. 2. With the antimonate reagent many other elements, even if present in small quantity, giv precipitates; thus a distinct turbidity is produced by even 0.1-0.2 mg. of calcium, barium, or magnesium, and by 1-2 mg. of aluminum. These elements produce, however, light, flocculent precipitates which look very different from the heavy crystalline precipitate obtained with sodium, especially if the mixture has been allowed to stand a few hours. The crystals of the so- dium salt may be separated from a flocculent precipitate by shaking the mixture, waiting long enough for the heavy crystals to settle, and decanting off the solution containing the flocculent precipitate in suspension. 90 SUPPLEMENTARY PROCEDURES FOR BASES. P. 97 SUPPLEMENTARY PROCEDURES FOR BASIC CONSTITUENTS. Procedure 96. Detection of Ammonium. Place 0.2-0.3 g. of the finely powdered original substance and 2 cc. NaOH solution in a 50 cc. round-bottom flask. Insert a stopper carrying a glass rod around whose end is wound a piece of moist red litmus paper; and heat the mixture nearly to boiling. (Blue coloration of the litmus paper and odor of ammonia, presence of AMMONIUM.) Confirmatory Test for Ammonium. If the litmus turns blue or the vapors smell of ammonia, pour into the flask 10 cc. water, insert a stopper fitted with a long, wide delivery tube leading to the bottom of a test-tube placed in a vessel of cold water, and distil slowly till about half the water has passed over. To the distillate add a 10% solution of K 2 HgI 4 in 3-normal NaOH drop by drop so long as the precipitate increases. (Orange precipitate, presence of AMMONIUM.) Notes. 1. Less than 0.2 mg. of ammonium can be detected with litmus paper and by the odor when the test is carried out as described in the first paragraph of the Procedure. The test described in the last paragraph is use- ful with very small quantities of ammonium as a confirmation, and with larger quantities as a means of better estimating the proportion of it present. 2. The orange precipitate produced by the action of NH 3 on alkaline K2Hgl4 is a complex compound of the composition HgO.HgNH 2 I. The test is extremely delicate, a distinct precipitate resulting even with 0.2 mg. NH4 in 5 cc. solution, and a pronounced yellow color with a much smaller quan- tity. This fact must be taken into account in estimating the quantity of ammonium present. Procedure 97. Determination of the State of Oxidation of Mer- cury, Tin, and Iron. If mercury, tin, or iron has been found to be present, heat 20-30 cc. 6-normal H 2 S04 to boiling in a small flask (to expel the air from the solution and flask); drop in 0.3 g. of the finely powdered original substance; and, if solution does not take place at once, boil the mixture for 2-3 minutes, covering the flask loosely with a small watch-glass. Pour 5 cc. portions of the solution (through a filter if it is not clear) into test-tubes containing: a, 5 cc. 1-normal HC1 (white precipitate, presence of MERCUROUS MERCURY or silver) ; 6, 5 cc. SnCU solution (white precipitate, presence of MER- CURIC MERCURY) ; c, 5 cc. HgCl 2 solution (white precipitate, presence of STANNOUS TIN); d, 5 cc. K 3 Fe(CN) 6 solution (blue precipitate, presence of FERROUS IRON) ; e } 5 cc. KSCN solution (red coloration, presence of FERRIC IRON). If a precipitate is produced by HC1 in a, filter and add the filtrate (instead of fresh portions of the H 2 S0 4 solu- tion) to the solutions named under b and c; and if silver is present in the substance, treat the HC1 precipitate with NH 4 OH by P. 14. P. 97. SUPPLEMENTARY PROCEDURES FOR BASES. 91 Notes. 1. If in preparing the solution for the analysis for basic constituents the substance was dissolved in water or cold dilute HNO 3 , the state of oxidation of mercury will have been determined by its presence or absence in the HC1 and H 2 S precipitate. But, if the substance was treated with hot or concen- trated HNOa, any mercurous compound present will have been partly or com- pletely oxidized to the mercuric state. 2. Stannous and ferrous salts oxidize rapidly in the air. Hence, if the tests for them are to be delicate, the contact with the air must be made as short as possible. 3. Ferrous and ferric salts show a different behavior also with K4Fe(CN)e; the former giving a white precipitate (of K 2 FeFe(CN) 6 ) which rapidly turns blue in the air and the latter a dark-blue precipitate (of ferric ferrocyanide, Fe^FeCeNe^). Since this reagent produces a precipitate with both kinds of salts, it is less suitable than K3Fe(CN)e for distinguishing between them. The precipitate produced by K3Fe(CN)e with ferrous salts is also ferric ferro- cyanide. Procedure 98. Determination of the State of Oxidation of Arsenic. If arsenic has been found to be present, place 0.3 g. of the solid substance in a 100 cc. round-bottom flask arranged for distillation as described in P. 101. Place in the receiving flask 50 cc. of water. Pour into the distilling flask 10 cc. 12-normal HC1 and distil till about 5 cc. have passed over. (See Note 1.) Pass H 2 S into the distillate. (Yellow precipitate, presence of ARSENITE.) If H 2 S produces no precipitate, pour into the distilling flask 5 cc. 12-normal HC1 in which 0.5 g. powdered FeS04 has been dissolved, and distil into a fresh 50 cc. portion of water till 5 cc. have passed over. Pass H 2 S into the distillate. (Yellow precipitate, presence of ARSENATE.) Notes. 1. As the arsenic vapors are very poisonous, care must be taken that they do not escape into the air. 2. This method of distinguishing arsenite and arsenate depends upon the facts that arsenite is largely converted by strong HC1 into AsCls, which is volatil with steam, while arsenate is not converted into the corresponding chloride, and is therefore not volatil. By the addition of FeSC>4, however, arsenate is reduced in the presence of HC1 to AsCls, which then volatilizes with the steam. Less than 1 mg. of arsenic in either state of oxidation is readily detected by this procedure. 3. When arsenite is present, it can be completely driven over into the distillate by distilling the substance with 10 cc. 12-normal HC1 till only 2-3 cc. remain, replacing the HC1 which has distilled off, distilling again, and repeating these operations till the distillate givs no precipitate with B^S. Then, to test for arsenate, a solution of FeSC>4 in HC1 may be added, the mixture again distilled, and the distillate saturated with B^S. The removal of the arsenite, is, however, so slow that six or eight repetitions of the distilla- tion may be necessary. 92 SUPPLEMENTARY PROCEDURES FOR BASES. P. 99 Procedure 99. Detection of Very Small Quantities of Arsenic and Antimony. Prepare a H 2 SO 4 solution of the substance by heating 0.2-0.3 g. of it with 10 cc. H 2 SO4 if it does not contain organic matter, or by treating a larger quantity of it by the first paragraph of P. 8 if it contains organic matter. Place 2-3 cc. of pure, finely granulated zinc in a 75-cc. flask fitted with a stopper through which pass a thistle-tube and a delivery-tube bent at a right angle. Connect the delivery-tube through a tube filled with small lumps of CaCl2 with a hard-glass tube drawn out to a point at one end and constricted to a fairly wide capillary tube in the middle. Pour into the flask 10-15 cc. water and 10-12 drops of CuSO 4 solution; then add enough H 2 S0 4 to produce a fairly rapid evolution of hydrogen. After the air has been expelled from the apparatus (as shown by the fact that, on filling an inverted 7-cm. test-tube with the gas and touching the mouth of it to a gas-flame, the gas burns quietly without exploding) light the gas at the end of the hard-glass tube and heat that tube just back of the cap- illary with a small gas flame. (See Note 4.) If after 2-3 minutes no black deposit appears in the capillary (showing the purity of the re- agents), pour into the flask the H 2 S04 solution of the substance, a little at a time, and let the gas continue to pass through the heated tube for 4-5 minutes. (Black deposit in the capillary, presence of ARSENIC or ANTIMONY.) Cool the hard-glass tube, and dip the capillary part of it in a test-tube containing NaOCl solution. (Partial or complete solution of the deposit, presence of ARSENIC; incomplete solution of the deposit, presence of ANTIMONY.) Notes. 1. This method of detecting arsenic and antimony depends upon the facts that, when hydrogen is produced in a solution containing these elements all of the arsenic is converted into hydrogen arsenide gas (AsHs) and a part of the antimony is converted into hydrogen antimonide gas (SbHs) ; and that these gases decompose into their elements at a moderately high temperature. 2. The treatment of the deposit with NaOCl solution serves to distinguish the two elements; for metallic arsenic is readily converted into HsAsO 4 by that reagent, while metallic antimony is not acted upon by it. Even when they are present together, it is usually possible to detect both of them; for arsenic, being more volatil, deposits in the part of the capillary further from the flame, and this part of the deposit may be seen to dissolve, and the other part to remain undis- solved, in the treatment with NaOCl solution. 3. The addition of the CuSO 4 solution serves to produce a deposit of copper on the zinc granules and thus to accelerate through the voltaic action the evo- lution of hydrogen and the formation of the AsHs and SbHs. 4. Hydrogen arsenide is an extremely poisonous gas. Therefore it must not be allowed to escape into the air through leaks in the apparatus or by discon- tinuing the heating of the hard-glass tube while much AsH 3 is being evolved. 5. This process is commonly employed for the detection of arsenic in papers, fabrics, and other organic materials. DETECTION OF THE ACIDIC CONSTITUENTS. GENERAL DISCUSSION. The acidic constituents whose detection is here provided for are : Carbonate. Cyanide. Bromate. Sulfite. Chloride. Nitrate. Thiosulfate. Bromide. Phosphate. Nitrite. Iodide. Sulfate. Hypochlorite. Thiocyanate. Borate. Sulfide. Chlorate. Fluoride. Somewhat different processes are described in this book for the de- tection of these constituents, according as the substance is: (1) completely dissolved by cold dilute acids; (2) decomposed only by hot concentrated acids; (3) not decomposed even by concentrated acids. In the process employed when the substance is dissolved by cold dilute acids, the constituents that yield with acids readily volatil prod- ucts (hereafter called for short the readily volatil constituents), namely, carbonate, sulfite, thiosulfate, sulfide, nitrite, hypochlorite, and cyan- ide, are first tested for by treating the substance with hot dilute sulf uric acid and bringing suitable solutions in the form of drops or test- papers into contact with the vapors. This process is outlined in Table XIII on page 96. A nitric acid solution of the substance is then pre- pared; and, by adding to portions of it various reagents, most of the other acidic constituents are detected through the formation of char- acteristic precipitates or colorations. This process is outlined in Tables XIV and XV on pages 97 and 100. The remaining constituents are finally tested for with samples of the original substance, as outlined in Table XIX on page 111. In the process employed when the substance is decomposed only by hot concentrated acids the substance is first boiled with a solution of phosphoric acid. The vapors which come off while the phosphoric acid is dilute contain the readily volatil constituents; these are con- densed in barium hydroxide solution, constituting the " first distillate." The vapors passing over when the phosphoric acid becomes more concentrated contain the acids corresponding to the less volatil con- stituents, namely, to chloride, bromide, iodide, thiocyanate, chlorate, 93 94 DETECTION OF ACIDIC CONSTITUENTS. and nitrate; these are condensed in water, constituting the "second distillate." Finally when the phosphoric acid has become very con- centrated, copper is added to it, causing sulfate to be reduced to sulfite and to pass over as sulfurous acid into the "third distillate." These distillates are then tested for the separate constituents by appropriate reagents. This system of procedure is summarized in Tables XVII and XVIII on pages 105 and 109. Portions of the solid substance are then tested for the remaining constituents by such of the procedures outlined in Table XIX on page 111 as the results of the previous tests make necessary. The second of these processes can, of course, be employed, in place of the first one, also with substances which are dissolved by cold dilute acids; and the analyst may prefer to use it as the general procedure for all substances decomposable by acids. In the process employed with substances not completely decomposed even by hot concentrated acids, samples of the solid substance are tested for the readily volatil constituents and portions of a nitric acid extract of the substance are tested for the other constituents, just as in the case of substances dissolved completely by dilute nitric acid. The residue undissolved by nitric acid is then fused with sodium car- bonate, the mass is extracted with water, and the solution is tested for the constituents that are likely to be present in insoluble substances, namely, for sulfate, sulfide, chloride, phosphate, borate, and fluoride. It is to be noted that the system of procedure for detecting the acidic constituents can often be much shortened by omitting the tests for certain constituents which are excluded by the known character or source of the substance, or by its solubility considered in connection with the basic constituents present. Thus, it is useless to test a mineral for nitrite, sulfite, hypochlorite, chlorate, or cyanide; or, in a neutral water-soluble substance containing barium or silver it is unnecessary to test for any of the acidic constituents which form insoluble com- pounds with these elements. A general statement as to the solubilities of substances in water or dilute acids will be found in Note 9 on page 31; and numerical values of the solubilities of some analytically im- portant substances are given in a table on page 124. In addition to the acidic constituents listed on page 93, the follow- ing ones, which are detected in the course of the analysis for basic con- stituents, are frequently met with in minerals or industrial products : Silicate. Arsenate. Chromate. Stannate. Areenite. Permanganate. P. 100 DETECTION OF ACIDIC CONSTITUENTS. 95 GENERAL DIRECTIONS. Procedure 100. General Directions. If the substance is completely dissolved by cold dilute HNOs (as used in P. 2), test samples of the substance for readily volatil constituents by P. 101. Prepare a solu- tion of the substance by dissolving 2 g. of it in 30 cc. 1-normal HN0 3 ; test portions of this solution by P. 102-104; and test the remainder of the solution by P. 105 and P. 106, if in P. 103 halides are found to be present. Test fresh samples of the substance for borate by P. 121 and for nitrate by P. 124; also for nitrite by P. 125 and for hypochlorite by P. 126, if the previous tests show that they may be present. If the substance is not completely dissolved by cold dilute HN0 3 , but is decomposed by hot concentrated acids (as used in P. 3 and 4), treat a 2 g. portion of it by P. Ill, and treat the distillates thus ob- tained as directed in P. 111. Test also fresh samples of the sub- stance for borate, fluoride, and phosphate by P. 121, 122, and 123; also for nitrite, hypochlorite, and chlorate by P. 124-127, if the previous tests show that these constituents may be present. If the substance is not completely decomposed even by concentrated acids, treat samples of the substance by P. 131. Notes. 1. When the substance is completely decomposed by dilute acids the analysis for acidic constituents can be made by testing for the various constituents in the manner described in the first paragraph of the foregoing Procedure. When, however, the substance is not decomposed by dilute acids, some of the acidic constituents might escape detection if tested for in these ways. With such a substance it is therefore necessary to use a more powerful decomposing agent. Phosphoric acid is suitable for this purpose, since it is a fairly strong acid whose solution can be made highly concentrated without caus- ing much volatilization or decomposition of the acid. When the substance does not dissolve completely in dilute HNOs it is therefore directed to boil it with dilute HsPCU (as described in P. Ill) and to condense the vapors which contain all the volatil acids resulting from the decomposition of the substance. This process serves at the same time to separate the acidic constituents of the substance from its basic constituents, thereby facilitating the detection of the acidic constituents. 2. Even with a substance which dissolves completely in dilute acids it is often advantageous to employ the distillation process of P. Ill; for tho the tests for the readily volatil acidic constituents described in P. 101 are very delicate, they do not enable a satisfactory estimate to be made of the propor- tions in which the constituents are present. Moreover, these tests fail to detect certain of the constituents when they are present together (for example, car- bonate in the presence of sulfite). 3. If the substance is not decomposed even by hot concentrated acids, it has to be decomposed by fusion with Na^OOa, as described in P. 131. 96 DETECTION OF ACIDIC CONSTITUENTS. P. 101 SUBSTANCES DISSOLVED BY DILUTE NITRIC ACID. TABLE XIII. DETECTION OF THE READILY VOLATIL ACIDIC CONSTITUENTS. Heat the substance with dilute # 2 4 (P. 101). Vapors: CO 2 , SO 2 , H 2 S, NO 2 , C1 2 , Br 2 , I 2 , HCN. Expose to the vapors: Ba(OH}z solution. PbAc% paper. Starch and KI paper. Fe(OH) 2 , Fe(OH^, and NaOH on paper. White turbidity: BaCO 3 or BaSO 3 . (ShoWS CARBONATE, STJLFITE, 01 THIO- SULFATE.) Black color: PbS. (ShoWS STTLFIDE.) Blue color: I 2 . (Shows NITRITE, HYPOCHLORITE, CHLORATE, BRO- MATE, Or IODIDE.) Formation of Na 4 Fe(CN) 6 . Dip in HCl. Blue color: Fe 4 (Fe(CN) 6 ) 3 . (ShoWS CYANIDE.) Procedure 101. Detection of Constituents Yielding Readily Volatil Products. Place 0.3 g. of the finely powdered substance in a 30 cc. conical flask, add 2 cc. of water, heat the mixture nearly to boiling, and to the hot mixture add 5 or 6 drops H 2 S0 4 . Note whether there is an odor or formation of gas bubbles. Insert in the flask a two-hole rubber stopper through which passes a glass rod from which is sus- pended a drop of Ba(OH) 2 solution. (White precipitate, presence of CARBONATE, SULFITE, or THiosuLFATE.) Make B, conical paper roll out of half a filter-paper; insert the narrow end of it, in place of the glass rod, in one hole of the rubber stopper; dip the other end of the roll in PbAc 2 solution contained in a 7-cm. test-tube; insert the stopper in the flask; and heat the liquid nearly to boiling. (Blackening of the paper, presence of SULFIDE.) Replace the paper roll by a fresh one which has been dipped in a solution of KI and starch; and heat the liquid again nearly to boiling. (Blue coloration of the paper, presence of NITRITE, HYPOCHLORITE, CHLORATE, BRO- MATE, or IODIDE.) Place in the flask a fresh 0.3 g. sample of the sub- stance, and add to it 2 cc. of water and 5-6 drops of H 2 SO4; insert a rubber stopper carrying a paper-roll which has been dipped first in a 2-normal FeS0 4 solution and then in a 1-normal NaOH solution; and heat the mixture in the flask nearly to boiling. Remove the roll, and dip it first in HCl and then in water. (Blue coloration of the paper, presence of CYANIDE.) P. 101 DETECTION OF ACIDIC CONSTITUENTS. 97 Notes. 1. These tests are all delicate enough to show 0.1-0.2 mg. of the respectiv constituents. When any of the tests yields a positiv result, the nature and quantity of the constituent in the substance which givs rise to the test is more definitly determind by later procedures, as described in the general directions given in P. 100. 2. A blue coloration of the starch-KI paper shows that there is present in the vapors either free iodin or one of the volatil substances which liberates iodin from KI; namely, chlorin, bromin, or nitrogen dioxide. In the presence of starch, which forms with water a colloidal solution, the iodin is dissolved by the minute starch globules or is adsorbed on their surface; and in this finely divided state it possesses a deep blue color. Chlorin or bromin usually arises from the presence in the substance of hypochlorite, chlorate, or bromate, together with a chloride or bromide. Iodin may be liberated from an iodide, when an oxidizing substance, such as a ferric salt, is also present. TABLE XIV. DETECTION OF THE ACIDIC CONSTITUENTS PRECIPITATED FROM ACID SOLUTIONS BY BARIUM AND SILVER SALTS. To a HNOz solution of the substance add BaCk (P. 102). To a HNOs solution of the substance add Cd(N0 3 ) 2 (P. 10S). Precipi- tate: BaSO 4 . (Shows SUL- FATE.) Filtrate. Add Br 2 . Yellow precipitate: CdS. (Shows SULFIDE.) Filtrate: add AgN0 3 . Precipi- tate: BaSO 4 . (Shows SULFITE.) Filtrate. Add NH* Ac. Precipitate: AgCl, AgBr, Agl, Ag2(CN) 2 , AgSCN. (Shows HALIDES, CYANIDE, or THIO- CYANATE.) Filtrate: AgC10 3 , AgBr0 3 . Add H 2 S0 3 . Yellow precipitate: BaCr0 4 . (Shows CHROMATE.) Filtrate. Add CaCk. Precipitate: CaF 2 . (Shows FLUORIDE.) Precipitate: AgCl, AgBr. (Shows CHLORATE Or BROMATE.) Procedure 102. Detection of Sulfate, Sulfite, Chr ornate, and Flu- oride.To 10 cc. of the HNO 3 solution (P. 100) add 10 cc. BaCl 2 solu- tion, and let the mixture stand for 5 minutes. (White precipitate, presence of SULFATE.) Filter the mixture, repeatedly if necessary; add to the filtrate (unless it smells of H 2 S) saturated Br 2 solution, 1 cc. at a time, till it is present in excess, and let the mixture stand for 5 minutes. (White precipitate, presence of SULFITE.) Filter the mixture, repeatedly if necessary, add to the filtrate 5 cc. 3-normal NH 4 Ac solution, and let the mixture stand 5 minutes. (Fine yellow precipitate, presence of CHROMATE.) 98 DETECTION OF ACIDIC CONSTITUENTS. P. 102 Filter the mixture, add to the filtrate 10 cc. CaCl2 solution, and let it stand 15 minutes. (White turbidity, presence of FLUORIDE.) Confirm the presence of fluoride by filtering off the precipitate, igniting the filter containing it in a spiral of platinum wire till it is incinerated, and treating the residue by P. 122. Notes. 1. Of the barium salts the sulfate is the only one that is precipitated by a moderate excess of BaC^ from a HNOs solution as strong as 0.5 normal. From the HAc solution produced by adding to such a HNOs solution 1? times as many equivalents of NH4Ac the chromate is precipitated completely, and the fluoride is precipitated when more than about 10 mg. of fluorin is present. Sulfite, if it were not previously removed by converting it into sulfate by the addition of Br2, would also precipitate on the addition of the NH4Ac. From a neutral solution phosphate, carbonate, and borate are also precipitated; but none of these separates from the HAc solution. 2. Of all the inorganic acidic constituents the fluoride is the only one whose calcium salt is much less soluble in water than its barium salts. BaF2 is soluble in water to such an extent that under the conditions of the procedure about 10 mg. of fluorin remain in the HAc solution. A precipitate produced by CaCfe therefore shows fluoride. The precipitate of CaF2 has, moreover, a characteristic appearance, separating first as a milky turbidity which slowly settles out in flocculent form. 3. The addition of NH*Ac to the HNOs solution may cause the precipitation of other substances than BaCrO4 and BaF2; namely, of any substances, such as bismuth salts, ferric or aluminum phosphate, silicic acid, which are dissolved by HNOs hut not by HAc. In such a case the test for chromate is obscured; but fluoride may still be detected in the nitrate with CaCfe. 4. Sulfide and sulfite can not be present together in an acid solution; for they destroy each other with the separation of sulfur. It is therefore useless to test a solution containing H2S for sulfite; moreover, the sulfur which would be precip- itated on adding bromin to such a solution might be mistaken for BaSO4. Procedure 103. Detection of Sulfide, of Halides, and of Chlorate or Bromate. To a 5 cc. portion of the HNOs solution (P. 100) add 5 cc. Cd(NO 3 )2 solution. (Yellow precipitate, presence of SULFIDE.) Filter out the precipitate; and to the filtrate add 20 cc. of water and 5 cc. AgNOs solution. (White precipitate, presence of CHLORIDE, CYANIDE, or THIOCYANATE; yellow precipitate, presence of BROMIDE or IODIDE.) Filter out the precipitate; to the filtrate add a few drops of AgNOs solution; then add 20 cc. HNOs and 5 cc. saturated S02 solution, and let the mixture stand for 5 minutes. If there is a precipitate, heat the mixture nearly to boiling. (White precipitate, presence of CHLORATE; yellow precipitate, presence of BROMATE.) P. 108 DETECTION OF ACIDIC CONSTITUENTS. 99 Notes. 1. The presence of sulfide is detected by the test-paper test in P. 101; but its precipitation as CdS enables the amount to be better estimated. 2. All the common silver salts, except the three halides, the cyanide, thio- cyanate, and sulfide, are either soluble in water (as are the nitrate, sulfate, chlorate, and fluoride), or dissolve readily in HNOs owing to displacement of the weaker acid (as do the phosphate, carbonate, borate, and sulfite). There are, however, some exceptions to the principle that salts of weak acids are readily soluble in a strong acid. Thus A&S does not dissolve in dilute HNO 3 because its solubility in pure water is so extremely small that there is only a very minute concentration of S~ ion in the saturated solution, and this can yield, in ac- cordance with the mass-action law, only a relativly small concentration of SH~ and unionized H 2 S with the H + ion of the HNOs. Silver cyanide has for another reason a very slight concentration of its anion in its saturated solution; namely, because of the fact that this salt exists in the solution mainly as Ag+ and Ag(CN) 2 ~, and scarcely at all as Ag+ and CN~. 3. The reduction of chlorate to chloride by the H 2 SOs is not instantaneous; but it is so rapid that 0.5 mg. ClOs in a volume of 60 cc. produces a precipitate in less than 5 minutes. 4. Before the addition of the H 2 SOs a few drops of AgNOs are added to make sure that the halides have been completely precipitated. A large quantity of HNOs is also added to prevent the precipitation of A^SOs; and, if a precipitate separates on adding the SO 2 solution, the mixture is heated nearly to boiling to make sure that the precipitate is not Ag^Os. 5. If much bromate is present, some of it is precipitated upon the first addition of AgNOs, along with the silver halides; but some of it also remains in the solution and shows the same behavior as chlorate. In order to distinguish between them, the final precipitate with AgNOs may be treated as follows: Suspend it in 25 cc. water, pass in H 2 S until the mixture is saturated with it, heat to boiling, filter off the precipitated Ag2S, boil the filtrate till the H 2 S is expelled, and test it for bromide and chloride by P. 106. 100 DETECTION OF ACIDIC CONSTITUENTS. P. 104 TABLE XV. DETECTION OF PHOSPHATE AND THE SEPARATE HALIDES. To portions of the HNO S solution of the substance (P. 100} : Add (NHAzMoOi (P. 104). Add FeCk (P. 105)* Add NaAc, HAc, KMnO*, and CHCk (P. 106)* Chloroform layer, purple: I 2 . (ShoWS IODIDE.) Water layer: add HzSO*, more KMnOi and CHCk. Yellow precipitate: (NH4) 3 P0 4 . 12 MoO 3 . (ShoWS PHOS- PHATE.) Red color: Fe(SCN) 3 . (Shows THIO- CYANATE.) Chloroform layer, orange: Br 2 . (ShoWS BRO- MIDE.) Water layer: boil out the Brz, add HNO& and AgNO z . Precipitate: AgCl (ShoWS CHLO- RIDE.) These procedures are followed only in case AgNOj produces a precipitate in P. 103. Procedure 104. Detection of Phosphate. Add 3 cc. of the HN0 3 solution (P. 100) to 6-8 cc. (NH^MoC^ solution, and let the mixture stand 5-10 minutes. (Yellow precipitate, presence of PHOSPHATE.) Notes. 1. The yellow precipitate produced is a complex compound, am- monium phosphomolybdate, of the composition (NH4) 3 PO4.12MoOa 2. In order that the test may be delicate, a large proportion of the (NEU) 2 MoO4 must be present to reduce the solubility of the precipitate; and a short time must be allowed for the formation of the complex phosphomolyb- date. This is promoted by gentle warming; but in a hot solution arsenate or silicate may giv rise to a similar yellow precipitate, while in the cold the re- action is given only by phosphate. By this test 0.1 mg. P(>4 may be easily detected. The great delicacy of this test should be borne in mind in estimating the quantity of phosphate present. Procedure 105. Detection of Thiocyanate. If AgN0 3 produced a precipitate (in P. 103), add to 2 cc. of the HNO 3 solution (P. 100) 4-5 drops of FeCl 3 solution. (Red color, presence of THIOCYANATE.) Note. The red color arises from the formation by metathesis of Fe(SCN) 3 , a substance whose degree of ionization is relativly small. A distinct reddish- yellow color is produced by 0.1 mg. SCN, a deep-red color by 1 mg. or more. Procedure 106. Detection of the Separate Halides. If AgNOa pro- duced a precipitate (in P. 103), add to the remaining 10 cc. of the HN0 3 solution (P. 100) in a conical flask 3-normal Na 2 CO 3 solu- tion, a few drops at a time, till the liquid no longer givs a decided red color to blue litmus-paper. (If too much has been accidentally added, add HN0 3 drop by drop till the solution again reddens blue P. 106 DETECTION OF ACIDIC! pQ%J[TfSpjJXyfai' 101 litmus-paper.) Then add 8 cc. NaAc solution, 2 cc. HAc, and (after filtering out any precipitate) 3 cc. chloroform (CHC1 3 ). Finally add 1% KMn0 4 solution, 1 cc. at a time, shaking vigorously after each addition, till the aqueous layer becomes pink. (Purple coloration of the chloroform, presence of IODIDE.) Pour the mixture through a moistened filter to remove the chloroform and precipitated Mn0 2 , and shake the filtrate once or twice with a fresh 10 cc. portion of chloroform to extract all the iodin. Place the aqueous solution and 3 cc. chloroform in a separating funnel, add 5 cc. H 2 S0 4 , and 1 cc. 1% KMn0 4 solution, unless such an excess is already present, and shake. (Yellow or orange coloration of the chloroform, presence of BROMIDE.) Transfer the aqueous layer to a casserole, add 5-20 cc. 1% KMn0 4 solution, and boil the mixture 3-5 minutes, or until the volume has been reduced to 10 cc. Filter off the Mn0 2 , and, if the solution is still pink, add H 2 SOs solution drop by drop until it is colorless. Dilute the solution to 100 cc., filter if necessary, and add 20 cc. HN0 3 and 5 cc. AgNOs solution. (White precipitate, presence of CHLORIDE.) Notes. 1. This separation is based upon the different rates at which KMnO 4 sets free by oxidation the three halogens from their salts in a solution of definit hydrogen-ion (H+) concentration. A dilute solution of acetic acid containing considerable sodium acetate has such a hydrogen-ion concentration that an iodide is immediately oxidized by KMnO 4 with liberation of iodin, while bromide and chloride are not oxidized to an appreciable extent in the time required for the operations. When the H + concentration is increased by the addition of the prescribed quantity of H 2 S0 4 , the bromide is oxidized very rapidly, while the rate of the corresponding reaction for the chloride is still so small at room temperature that scarcely any chlorin is set free. Even when the solution is boiled to expel the bromin, only a small fraction of the chloride present is oxidized to chlorin. 2. To secure satisfactory results, the directions as to the quantities of the acids added must be followed carefully. The proper quantity of H 2 SO 4 is that required to react with all the sodium acetate and to giv in addition an excess equal to about 1 cc. H 2 SO 4 per 20 cc. of solution. 3. The yellow color of bromin in 3-5 cc. chloroform enables about 0.5 mg. Br to be detected in this procedure. 4. A very small precipitate of AgCl obtained at the end of the procedure does not necessarily indicate the presence of chloride in the substance, unless the reagents used have been proved to be entirely free from chloride. Even then a very slight precipitate (corresponding to less than 0.1 mg. Cl) may result from a reaction between the permanganate and chloroform. For these reasons a blank test should be made in any doubtful case. 5. Before adding the AgNOs in the test for chloride the solution is diluted and HN0 3 is added to it, so as to prevent the precipitation of A&SC^ and Ag 2 SOs. 6. If HCN, H 2 S, or HSCN is present in the solution, it will be expelled or destroyed by the boiling with KMnO 4 before the test for chloride is made. 102 '.ti #F ACIDIC CONSTITUENTS. P. Ill SUBSTANCES DECOMPOSED ONLY BY CONCENTRATED ACIDS. TABLE XVI. BEHAVIOR OF THE ACIDIC CONSTITUENTS ON DISTILLATION WITH PHOSPHORIC Aero. Distil the substance with dilute H^PO* (P. 111). Collect the first half of the distillate in Ba(OH)z solution and the second half in water. To the residue add Cu and distil again, collecting this third distillate in water. FIRST DISTILLATE SECOND DISTILLATE THIRD DIST. NONVOLATIL RESIDUE COo from carbonate. HC1 from chloride. SO 2 from HPOa from phosphate. 862 from sulfite or HBr from bromide. sulfate. HBO2 from borate. thiosulfate. HI from iodide. H2SiOs from silicate. Cl2 from hypochlorite, HSCN from thio- chlorate, or chloride.* cyanate. Bis from bromate or HCN from f erro- or bromide.* ferri-cyanide. Ij from iodide.* H2S from insoluble HNO2 from nitrite. sulfides. H2S from sulfide. HNOs from nitrate. HCN from cyanide. Ck from chlorate or chloride.* Br2 from bromide.* 12 from iodide. * When the substance contains also an oxidizing compound. Procedure in. Distillation with Phosphoric Add. Place 2 g. of the finely powdered substance and a few glass beads in a distilla- tion-apparatus, arranged as shown in the figure, consisting of a 100 cc. round-bottom hard-glass flask fitted with a rubber stopper, through which pass a delivery tube and a safety-tube, 20-30 cm. long, leading to the bottom of the flask. Support the flask in an inclined posi- tion. Lead the end of the delivery tube through a two-hole stopper into 40 cc. of nearly saturated Ba(OH) 2 solution contained in a 100 cc. flask supported in a large beaker of cold water. Boil in a small flask for about a minute a mixture of 25 cc. water and 10 cc. 85% H 3 PO 4 (to expel any C0 2 present in it). Pour this mixture into the distilling flask with the aid of a small funnel connected with the safety-tube. Heat the mixture to boiling, distil till about 10 cc. have passed over, and then remove the distillate. (White precipitate, presence of CARBONATE, SULFITE, THIOSULFATE, SULFIDE, or SULFUR.) p. Ill DETECTION OF ACIDIC CONSTITUENTS. 103 Cool the distillate and make it slightly acid with HAc. (Com- plete or partial solution of the precipitate, presence of CARBONATE; residue (of S or BaSOs), presence in the substance of free SULFUR, SULFIDE, SULFITE, or THiosuLFATE.) If there is a residue, treat one- half of the mixture immediately by P. 112, and separate portions of the remainder by P. 113, 114, and 115. If there is no residue, treat separate portions of the whole distillate by P. 113, 114, and 115. Introduce the end of the delivery tube of the distilling flask into another receiving flask containing 35 cc. water. Continue the distil- lation until the liquid becomes sirupy, boils more quietly, and begins to giv off fine white fumes. Treat this distillate as directed in P. 116. To the contents of the distilling flask, while still warm, add 5-10 g. of copper filings or turnings. Distil for 3-5 minutes longer, col- lecting the distillate in 15 cc. of water. Note the odor of the distillate, and treat it by P. 119. Notes. 1. It is necessary to use a hard- glass flask, since one of ordinary glass is quickly destroyed by the action of hot, concentrated HsPO^ The boil- ing is sometimes violent, especially when much insoluble material is present. The addition of the glass beads serves to reduce the bumping; and placing the flask in an inclined position prevents material from being thrown over into the distillate, which would lead to error in the subsequent tests. In any case in which it seems possible that some of the boiling liquid has been thrown over into the distillate, a small portion of the latter should be tested for phos- phate by acidifying it with HNOa and adding an equal volume of (NEU^MoOi solution (see P. 104). 2. Phosphoric acid, which is ionized into H + and H2PO,r to a moderate extent (about 40% in 0.1 normal solution), displaces almost completely from their salts (unless these are very difficultly soluble) the much less ionized acids, 104 DETECTION OF ACIDIC CONSTITUENTS. P. Ill H 2 C0 3 , HN0 2 , H 2 S, HC10, HCN, HF, and HsBO 3 , and also to a large extent the moderately ionized H 2 SO3. Since all these acids, except HF and HaBOs, volatilize readily out of aqueous solution, they pass over almost or quite com- pletely into the first distillate, HC1O in the presence of chloride giving C1 2 . The largely ionized acids, HC1, HBr, HI, HSCN, HN0 3 , HC1O 3 , H 3 Fe(CN), and H4Fe(CN)e, are not found in any considerable proportion in the first dis- tillate, since the unionized acid is formed in much smaller proportion, and since in addition it is much less volatil. Of these the first five pass over unchanged and almost completely into the second distillate; for after the HaPOi has become fairly concentrated, the acids are displaced to a greater extent and volatilize more readily in consequence of the higher temperature at which the mixture boils and the smaller proportion of water it contains. From the stronger HgPC^ solution HF also passes over in large quantity; but this is not true of HaBOs and H 2 SO4, which volatilize only in insignificant amounts even when the acid has become nearly anhydrous. The three acids, HClOa, H4Fe(CN)e, and HsFe(CN)6, are not volatil as such, but are decomposed by the H 3 PO4 after it becomes fairly concentrated HC1O 3 with formation of Clj and OE, H 3 Fe(CN) 6 and H 4 Fe(CN) 6 with formation of free HCN. In regard to the acids that may be present in the two distillates, see also Table XVI. 3. The barium salts of all the acids passing into the first distillate, except the carbonate and sulfite, remain in solution. Phosphoric acid, if thrown over mechanically, would, however, also giv a precipitate. Sulfur, when present in the free state or when liberated from a polysulfide or thiosulfate, volatilizes with the steam, and givs a turbid appearance to the water condensed in the delivery tube and to the barium hydroxide solution, by which it is little acted on in the cold. Chlorin is converted by the barium hydroxide into barium chloride and hypochlorite; bromin, into bromide and hypobromite, and into bromide and bromate; and iodin, mainly into iodate and iodide. 4. On acidifying the first distillate slightly with HAc, BaCOs dissolves, but BaS0 3 does not. This difference in behavior is due to the fact that hydrocar- bonate-ion (HCO 3 ~) is much less ionized than hydrosulfite-ion (HSOa~). Sulfur, if present, also remains undissolved. The addition of HAc causes the liberation almost at once of chlorin, bromin, or iodin from a mixture of hypo- chlorite and chloride, hypobromite and bromide, or iodate and iodide; but bromin is set free somewhat more slowly from a mixture of bromate and bromide. 5. A small precipitate obtained in this procedure (or in the following one) does not prove the presence of carbonate in the mixture unless the prescribed precautions are carefully observed namely, the boiling of the original HsPO4 solution, and avoiding the exposure to the air of the various solutions, especially that of the BaO 2 H 2 . Even with these precautions, however, it is seldom pos- sible to prevent the absorption of enough CO 2 to produce a slight turbidity. 6. Upon boiling the H 3 PO4 with the copper, H 2 SO4, if present, is reduced to H 2 SO?; and this passes over into the distillate in the form of SO 2 gas. Less than 1 mg. SO4 can be detected by this process of distillation. The copper should be finely divided and should be added while the liquid is still warm, since on cooling it solidifies to a glassy mass, which consists of pyrophosphoric acid (H^O?). The heating should be continued for 5-10 minutes; but, if much more prolonged, the contents of the flask change to a solid mass, owing to conversion of the pyro to metaphosphoric acid (HP0 3 ), which can after- wards be removed only with much difficulty. P. 112 DETECTION OF ACIDIC CONSTITUENTS. 105 TABLE XVII. ANALYSIS OP THE FIRST DISTILLATE. FIRST DISTILLATE. Precipitate: BaCO 3 , BaSO 3 , S. Solution : Ba(ClO) 2 , Ba(BrO) 2 , Ba(IO 3 ) 2 , (with halides) ; Ba(NO2) 2 , BaS, Ba(CN) 2 . To the whole mixture add HAc (P. 111). Precipitate: BaSO 3 , S. Solution: H 2 CO 3 , C1 2 , Br 2 , 1 2 , HNCfe, H 2 S, HCN. Treat portions of the unfiltered mixture as follows: Add HCl and filter (P. 112}. Add CHCk Filter, add Heat with (P 118) Cd(NO ) NaOH and Residue: S. Solution. Add Br 2 . (P. 114). FeSOt (Shows Purple color : I 2 . f (P. 115). SULFIDE Or THIO- SULFATE.) Precipitate: BaSO 4 .* Boil the mia the dist Ba(C Solution: H 2 CO 3 . iture, collect 'Hate in >#) 2 . Orange color: Br 2 .J Precipitate: CdS. (Shows SULFIDE.) Solution: Na4Fe(CN) 6 . Precipitate: Fe(OH) 2 - 3 . Add HCL // the CHCk is colorless, add KI. Purple color: I 2 . (Shows NITRITE, HYPO CHLORITE, Or CHLORATE.) Blue pre- Precipitate: BaC0 3 . cipitate: (ShoWS CARBONATE.) Fe4(Fe- (CN),),. (Shows CYANIDE.) * ShoWS BULFITE Or THI08UI.PATB. fShowS IODIDE. J Shows BKOMATE or BROMIDE. Procedure 112. Detection of Carbonate and Sulfur-containing Con- stituents. To one-half of the first distillate (P. Ill), if there was a residue on adding HAc, add 1-2 cc. HCl. (Residue, presence of free SULFUR, SULFIDE, or THIOSULFATE.) Filter, and add to the filtrate saturated bromin solution till the liquid becomes slightly yellow. (White precipitate, presence of SULFITE or THIOSULFATE.) Transfer the mixture to a distilling apparatus such as is used in P. Ill, first filtering out the precipitate if it is large, distil for a minute or two, collecting the vapors in 20 cc. saturated Ba(OH) 2 solution. (White precipitate, presence of CARBONATE.) Acidify slightly with HAc. (Solution of the precipitate, presence of CARBONATE.) Notes. 1. See P. Ill, Notes 3-5. Since H 2 SOs slowly oxidizes to H 2 SC>4 in the air, the solution should be treated with HCl at once. If any H 2 SO4 has been formed in this way, it will be precipitated as BaSO 4 before the addi- tion of Br 2 . Care must be taken to add enough Br 2 to complete the oxidation, since otherwise in the subsequent distillation SO 2 will distil over and might be mistaken for carbonate. 106 DETECTION OF ACIDIC CONSTITUENTS. P. 112 2. If there is a large precipitate of BaSO 4 , it is filtered out, since otherwise it is difficult to avoid violent bumping during the distillation. Exposure to the air, and especially to the breath, should, however, be avoided so far as possible, so that CC>2 may not be absorbed from it. 3. A residue of sulfur may arise from the presence in the substance of free sulfur, of a persulfide, of an ordinary sulfide together with some oxidizing substance, or of a thiosulfate. Procedure 113. Detection of Nitrite and Halogen-liberating Constit- uents. To one-fourth of what remains of the first distillate (P. Ill), add 1-2 cc. HAc and 2-3 cc. of chloroform, and shake vigorously. (Purple coloration of the chloroform, presence in the distillate of free IODIN; yellow or orange coloration, of free BROMIN.) If there is no coloration, add 8-10 drops of 0.1-normal KI solu- tion, and shake the mixture. (Purple color, presence in the distillate of CHLORIN, or of NITROUS ACID; no color, absence of NITRITE and HYPOCHLORITE in the substance.) If there is a coloration either before or after the addition of KI, test fresh samples of the original substance for NITRITE, HYPOCHLO- RITE, CHLORATE, and BROMATE by P. 125, 126, and 127. Notes. 1. For the reactions between the halogens and barium hydroxide and their re-formation on acidifying with HAc, see P. Ill, Notes 3 and 4. 2. The characteristic purple color given to chloroform is so delicate a test that even 0.05 mg. of iodin in the solution tested can be detected by this pro- cedure. Bromin may be detected, but only in the absence of iodin, by the orange or yellow color of the chloroform layer when not less than 0.5 mg. of bromin is present in the solution tested. Chlorin gives no decided color to the chloroform, but causes liberation of iodin on the addition of KI. 3. The free halogens distribute themselves between the chloroform and water layers. In the case of pure bromin or iodin the ratio of the concentration in the chloroform to that in the water layer is very large; and this ratio is almost independent of the concentration, in accordance with the so-called dis- tribution law, which requires that the ratio of the concentrations of a given molecular species, such as Br2 or 1%, in any two non-miscible solvents be con- stant. When an iodide, like KI, is also present, as it is in the test for free chlorin and nitrous acid, the proportion of iodin extracted by the chloroform is greatly reduced, since the iodin in the aqueous layer is largely combined with the iodide in the form of the triiodide (KIs); but it is still sufficient to make the color-test a very delicate one, provided only a few drops of the KI solution have been added. 4. For extracting the halogens from aqueous solutions carbon tetrachloride or carbon bisulfide may be used instead of chloroform; but carbon bisulfide has the disadvantage of being highly inflammable. 5. If the tests described in both paragraphs of this procedure yield negativ results, it shows the absence in the substance of nitrite and hypochlorite, but not of the halides nor of chlorate and bromate, since these constituents may not P. 118 DETECTION OF ACIDIC CONSTITUENTS. 107 be decomposed or volatilized till the HaPC^ becomes concentrated in the later part of the distillation. 6. If the chloroform assumes a purple color when it is first added to the dis- tillate, it shows the presence in the substance of free iodin, of iodate, or of iodide (from which iodin has been liberated in the distillation by the action of the air or by some oxidizing compound present in the substance). If the chloroform assumes an orange color, it shows the presence in the substance of a bromate, or of a bromide together with some oxidizing compound. In these cases the presence or absence of other halogen-containing constituents and of nitrite has to be determined by the special tests described in P. 125-127. 7. If the chloroform becomes colored only after the KI is added, it shows the presence in the substance of nitrite, hypochlorite, or chlorate (or possibly only of chloride in rare cases where a powerful oxidizing substance, such as MnO2, KMnO4, or K 2 Cr 2 07, is also present). Which one of these constituents givs rise to the color has to be determined by the special tests. 8. Nitrous acid liberates iodin from KI owing to its reduction to nitric oxide. A peculiarity of this reaction is that the nitric oxide which is formed by it is rapidly reoxidized by the oxygen of the air to nitrous acid, which then reacts with the iodide, so that a continuous liberation of iodin results. Thus the nitrous acid acts as a catalyzer of the reaction between oxygen and HI. This progressiv liberation of iodin is highly characteristic of nitrous acid, but renders it difficult to estimate the amount of it present. Procedure 114. Detection of Sulfide. To one-half of what still remains of the first distillate (P. Ill), add 2-3 cc. Cd(N0 3 )2 solution. (Yellow precipitate, presence of SULFIDE.) Note. A negativ test in the first distillate does not prove the absence of sulfide in the original substance, unless the latter has dissolved completely in the dilute HsPC^; for some difficultly soluble sulfides, such as CuS, are decomposed only when the HjPOi becomes concentrated, as it does in the latter part of the distillation. It is therefore directed in P. 116 to test also the second distillate for sulfide. Procedure 115. Detection of Cyanide. Place what remains of the first distillate in a casserole; add 1 cc. NaOH solution and 0.5 cc. FeSOi solution; and boil for one minute. To the hot mixture add HC1, a few drops at a time, until, on shaking, the dark colored pre- cipitate of ferrous and ferric hydroxides is dissolved. Cool the mix- ture. If a precipitate is not plainly visible, filter, and v. ,sh out the filter-paper once with water. (Blue precipitate, presence of CYANIDE.) Notes. 1. This test is based upon the formation of sodium ferrocyanide by the action of the sodium cyanide on the ferrous hydroxide and upon the reaction between this ferrocyanide and the ferric salt which has been produced by the oxygen of the air. As a result of these two reactions, ferric ferrocyanide (Prussian blue) is formed, which is difficultly soluble in dilute hydrochloric acid. 2. A small precipitate is not readily detected in the hot reddish-yellow solution, but is more easily seen in the cold light-colored solution, especially 108 DETECTION OF ACIDIC CONSTITUENTS. P. 115 after standing, or when collected on a filter. If the precipitate on the filter is not dark blue, it should be washed with a little hot, dilute hydrochloric acid. With these precautions, 0.2 mg CN in the solution tested can be detected. 3. Cyanides may be present in the original substance in the form either of simple or of complex cyanides. The latter are characterized by complex anions (such as Ag(CN)?~ and Fe(CN)6~"). These differ very greatly in their stability towards decomposing agents, the difference depending on the extent to which they are dissociated into the simple ions (Ag + and CN~ or Fe+ + and CN~). Ferrocyanides, ferri cyanides, and cobalticyanides are so slightly dissociated in this way that scarcely any HCN is produced when dilute HC1, HNOs, or H2SO4 is added to their cold solutions; but almost all the other complex cyanides (such as KAg(CN)2 or K 2 Ni(CN)4) are readily decomposed by these acids. In the distillation with HsPO^ not only the simple cyanides, but also nearly all the complex cyanides are decomposed during the first part of the distillation; but a few very stable substances (such as Prussian blue) are completely decomposed only in the second part of the distillation. 4. The following procedure enables 2 mg. cyanide to be detected in the presence of ferro or f erricyanide : Place in a 20 cc. distilling flask provided with a thistle-tube 0.5-1 g. of the original substance, 2 g. powdered CaCOs, and 10 cc. water. Add very gradually through the thistle-tube 2 cc. HC1 (enough to decompose some, but not all, of the CaCOa). Allow the gas which is evolved to pass into a small test-tube containing 1 cc. NaOH and 5 cc. water. Finally heat the contents of the flask almost to boiling. Test the NaOH solu- tion for cyanide by P. 115. This separation depends upon the fact that HCN is displaced by ^COs from simple cyanides and from the relativly unstable complex cyanides, such as Ag(CN)2~ or Ni(CN)4 = , but not from ferro or ferricyanides. 5. Ferrocyanide and ferricyanide may be detected and distinguished from each other when only one of them is present, by adding a ferric salt to one portion of an aqueous or dilute acid solution, and by adding ferrous salt to another portion of the solution. A ferric salt givs a blue precipitate of ferric ferrocyanide with ferrocyanide, but no precipitate with a ferricyanide. A ferrous salt givs the same blue precipitate (of ferric ferrocyanide) with a ferri- cyanide; but it also givs with a ferrocyanide a precipitate (of ferrous ferro- cyanide), which is white if no ferric salt is present, but which rapidly turns blue in contact with the air. 6. Ferrocyanide and ferricyanide may be detected in the presence of each other by proceeding as follows: Add to an aqueous or dilute HNOs solution of the substance AgNOs and then a moderate excess of NEUOH. (White precipitate insoluble in NELiOH, presence of FERROCYANIDE. Orange to red precipitate readily soluble in NKLiOH, presence of FERRICYANIDE.) Filter out and wash the precipitate, and pour over it a little FeCls solution. (Blue coloration, presence of FERROCYANIDE.) Acidify the ammoniacal filtrate with HAc, filter out and wash the precipitate, and pour through the filter containing it a little FeSO4 solution. (Orange-red precipitate, which is turned blue by the FeSO4, presence of FERRICYANIDE). This procelure enables 0.2 mg. Fe(CN)e as either ferro or ferricyanide to be detected when present alone; but the test for ferri- cyanide is much less delicate in the presence of much ferrocyanide. P. 116 DETECTION OF ACIDIC CONSTITUENTS. 109 TABLE XVIII. ANALYSIS or THE SECOND AND THIRD DISTILLATES. SECOND DISTILLATE: H 2 S, HCN, HSCN, HC1, HBr, HI, C1 2 , Br 2 , I 2 . THIRD DIS- TILLATE: H 2 SOs. Add HCl, BaCk, and Br z (P.119). To a portion add AgN0 3 (P. 116). // AgNOs givs a precipitate, treat portions as follows: Add Cd(NO s ) 2 (P. W). Add FeCk and HCl (P. 117). Add CHCk (P. 118). Precipitate: AgCl, AgBr, Agl, AgSCN, Ag2S, Ag 2 (CN) 2 . CHCla layer: I 2 , Br 2 , C1 2 .* // colorless, add KL Water layer: HCl, HBr, HI. Test for sepa- rate HALIDES by P. 106. Precipitate: BaSO 4 . (Shows SULFATE.) Precipitate: CdS. (Shows SUL- FIDE.) Red color: Fe(SCN) 3 . (Shows THIO- CYANATE.) Purple color: I 2 . (Shows CHLORATE Or CHLORIDE.) * Purple coloration shows IODIDE, orange coloration, BROMIDE or BROMATE. Procedure 116. Detection of Constituents Precipitable by Silver Nitrate. To one-sixth of the second distillate (P. Ill) add 1 cc. HN0 3 and 1 cc. AgNOs solution. (White precipitate, presence of CHLORIDE, CYANIDE, or THiocYANATE; yellowish precipitate, presence of BROMIDE or IODIDE; black precipitate, presence of SULFIDE.) If there is a precipitate, test one-sixth of the second distillate for sulfide by P. 114, another sixth for thiocyanate by P. 117, and the remainder for free halogen and halides by P. 118 followed by P. 106. If there is no precipitate, reject the whole second distillate. Notes. 1. As to the solubilities of the various silver salts, see Note 2, P. 103. 2. It is not necessary to test for cyanide in this distillate; for even the insoluble ferro and ferricyanides are decomposed partly, tho not necessarily completely, in the first part of the distillation. It is, however, advisable to test for sulfide unless the substance dissolved completely in the hot dilute HaPO,!, or unless the AgNOa precipitate is pure white; for some insoluble sulfides begin to decompose only when the HsPOi becomes concentrated. Procedure 117. Detection of Thiocyanate. Dilute a sixth of the second distillate (P. Ill) to 5-10 cc., add 2-3 drops of FeCl 3 solu- tion and 2-3 drops of HCl. (Red color, presence of THIOCYANATE.) Notes. 1. The red coloration arises from the formation by metathesis of Fe(SCN)s, a substance whose degree of ionization is relativly small. The HCl is added to reduce the hydrolysis of the FeCls and diminish the color imparted by it to the solution. A distinct reddish-yellow coloration is produced by 0.1 mg. SCN. A deep red color is obtained when 1 mg. or more is present. 2. Since in the distillation with HsPC^, thiocyanates are destroyed by certain oxidizing agents, such as nitrates, it is sometimes advisable to apply this test also to a solution of the original substance. 110 DETECTION OF ACIDIC CONSTITUENTS. P. 118 Procedure 118. Detection of Chlorate, Bromate, and the Halides. If AgNO 3 gave a precipitate in P. 116, to the remainder of the second distillate in a separating funnel add 2-3 cc. chloroform and shake. (Purple color, presence of IODIDE; orange or yellow color, presence of BROMIDE Or BROMATE.) If the chloroform is colorless, separate it from the aqueous layer; and add to the chloroform layer a few drops of KI solution. (Purple color, probable presence of CHLORATE.) If there is no color, treat the aqueous layer left in the separating funnel by P. 106. If the chloroform becomes colored either before or after the addi- tion of KI, add to the mixture in the separating funnel enough H 2 S0 3 solution to reduce the halogen, draw off the chloroform layer if it is still in the funnel, and treat the aqueous layer by P. 106. Notes. 1. As to these tests see the notes to P. 113. 2. In the second part of the distillation chlorate or bromate, whether present alone or with a halide, is rapidly decomposed with evolution of Ch or Br2. Therefore, if free halogen is found present neither in the first nor second distillate, it shows the absence of chlorate or bromate. The presence of Cl2 or Br in the second distillate does not, however, necessarily indicate chlorate or bromate; for these halogens may be produced from the corresponding halides by the action of some oxidizing substance. The special test for chlorate or bromate described in P. 127 should therefore be tried when, and only when, free halogen is found in either the first or second distillate. Procedure 119. Detection of Sulfate. To the third distillate ob- tained upon heating with copper (P. Ill), add 1-2 cc. HC1, 3-5 cc. BaCl 2 solution, and saturated Br 2 solution till the liquid becomes yellow. (White precipitate, presence of SULFATE.) Notes. 1. By the action of copper in the presence of concentrated H^PC^ on sulfates (even on the very difficultly soluble BaSO4) SO2 is formed. This is oxidized by the Br2 to H 2 S04, which then precipitates as BaS(>4. In this way 1 nig. SC>4 may be detected. Even when this small amount is present in the substance, only a small proportion of it passes into the second distillate. 2. Much HaPOi also passes over into the distillate; and the HCi is added to prevent its precipitation as BaHPC>4. Too much HCI must not be added since BaS(>4 is appreciably soluble in it. 3. When a sulfide is present which has not already been decomposed, sulfur and H 2 S, or sulfur and SC>2, may pass into the third distillate, after the acid has become concentrated. The H 2 S may be tested for in a portion of the distillate by P. 114. Owing to the possible formation of SOz by the oxidation of sulfur at the high temperature attained at the end of the distillation, the test for sulfate is unreliable if sulfur, undecomposed sulfide, or S0 2 gas is present in the distilling flask at the end of the second part of the distillation. In that case a fresh 0.3 g. sample of the substance should be heated with 10 cc. 2-nonnal HCI, the mixture filtered, and the filtrate tested for sulfate by adding an equal volume of BaCl2 solution. P. 121 DETECTION OF ACIDIC CONSTITUENTS. Ill SUBSTANCES DECOMPOSED BY COLD DILUTE ACIDS OR BY HOT CONCEN- TRATED ACIDS! SUPPLEMENTARY PROCEDURES. TABLE XIX. SUPPLEMENTARY PROCEDURES FOR DETECTING THE ACIDIC CONSTITUENTS. Treat samples of the original substance as follows: Distil with CHzOH and #2/S0 4 (P. 121). Heat with SiO z and KHSO* (P. 122). Boil with HN0 3 , add (NHJtMoOt (P. 123). Distil with HzSOi and FeSO* (P. 124). Dissolve in water, intro- duce into an inverted tube filled with HCl solution of urea (P. 126). Heat with water, HAc, and PbAcz (P. 126). Dissolve in HN0 3 , add AgNOs (P. 127). Distillate: B(OCH)3. Collect in CH S OH and HCl, and add turmeric. Gases evolved: SiF 4 and H 2 O. Deposit on cold part of tube: Si0 3 H 2 . (Shows FLUORIDE.) Yellow precipitate : (NH4) 3 P0 4 . 12MoO 3 . (Shows PHOSPHATE.) Distillate: HNO 2 . Add KI and CPICk. Dark brown precipitate : Pb0 2 . (Shows HYPOCHLO- RITE.) Precipitate: AgCl, etc. Reject. Gas: N 2 . (Shows NITRITE.) Filtrate: AgC10 3 . Add H 2 SO Z Purple color: I 2 . (Shows NITRATE Or NITRITE.) Precipitate: AgCl. (Shows CHLORATE.) Orange color. (Shows BORATE.) Procedure 121. Detection of Borate. Place 1-2 g. of the finely powdered substance in the distilling apparatus used in P. Ill, and add 10 cc. of methyl alcohol (CH 3 OH) and two or three glass beads. Pour in carefully 3 cc. 96% H 2 SO4, and distil off most of the alcohol, collecting it in a mixture of 5 cc. CH 3 OH and 3 cc. 12-normal HCl. Cool the distillate, and add to it five drops of a saturated solution of turmeric in ethyl alcohol. (Red or orange color, presence of BORATE.) Note. Methyl alcohol reacts with boric acid to form methyl borate B(OCH3)a which is a readily volatil liquid. The color given by turmeric to a solution of boric acid in methyl alcohol and strong hydrochloric acid is so intense that the test is very delicate if the proportions given are reproduced. The presence of 1 mg. BO 2 in the substance distilled may readily be detected. To estimate roughly the quantity present, the color may be compared with that given by adding the turmeric solution to known quantities of borate dissolved in a mixture of 3 cc. 12-normal HCl and 15 cc. CHsOH. Procedure 122. Detection of Fluoride. Mix 0.2 g. of the dry, finely powdered substance with twice its weight of powdered KHSOi and with 10-20 mg. dry, finely powdered or precipitated Si0 2 . Blow 112 DETECTION OF ACIDIC CONSTITUENTS. P. 122 a thick-walled bulb 1J-2 cm. in diameter at the end of a glass tube of 5-8 mm. bore. Place the mixture in the bulb (not using more of it than will one-third fill the bulb). Heat the bulb carefully until the KHSO4 is melted, taking care that the mixture does not froth up into the tube. Continue to heat the bulb and the lower part of the tube until there is a deposit of a solid substance or of condensed acid 3 or 4 cm. above the bulb. After it has cooled, cut off the tube close to the bulb. Dip the tube several times in water, dry it in a flame, and heat it strongly. (White deposit in the middle part of the tube and etched surface at the lower end, presence of FLUORIDE.) Notes. 1. This test depends on the following reactions: 4HF +SiO 2 =SiF 4 +2H 2 O. 3SiF 4 +3H 2 O =H 2 SiO s +2H 2 SiF 6 . Some of the HF liberated by the molten KHSO4 volatilizes and takes the silica required for the first reaction from the glass, thus producing the characteristic etched surface in the lower part of the tube. The SiF4 gas and the water- vapor liberated react in the cooler part of the tube according to the second equation (forming a white ring of solid silicic acid and fluosilicic acid, H 2 SiF). The reaction is reversed at higher temperatures, so that the deposit may be driven up the tube by heating. This white deposit is the most characteristic part of the test for fluoride. A deposit of SOs and H^SO* may also form in the upper part of the tube, and might be mistaken for, or interfere with, the test for small amounts of fluoride, if the final washing with water is omitted. This procedure enables 0.5 mg. F to be easily detected. 2. The test fails with certain minerals which are not decomposed by fusion with KHSOi. Such cases are provided for by the treatment described in P. 131. 3. Fluoride is often tested for by heating the solid substance in a platinum crucible with H 2 S(>4 alone and detecting any HF evolved by its etching action on a watch glass coated with wax through which markings have been made. This test has the disadvantage that when silica or silicate is present, which is very often the case in minerals, it is unreliable owing to the conversion of the HF to SiF4 by the reaction given in Note 1. Procedure 123. Detection of Phosphate. To 0.1-0.2 g. of the finely powdered substance add about 5 cc. HNOs. If the substance does not dissolve, boil the mixture for 2 or 3 minutes, and filter. Add to the filtrate an equal volume of (NH 4 )2Mo0 4 solution, and allow it to stand 5 to 10 minutes. (Yellow precipitate, presence of PHOSPHATE.) Note. See the notes on P. 104. Procedure 124. Detection of Nitrate or Nitrite. Arrange a distil- lation-apparatus in the way shown in the figure under P. 111. Place in the distilling flask 0.3 g. of the solid substance, 5 cc. 2-normal FeSC>4 solution, and 15 cc. H 2 S0 4 ; and place in the receiv- ing flask a mixture of 20 cc. water and 1 cc. NaOH solution. Distil P. 124 DETECTION OF ACIDIC CONSTITUENTS. 113 till only about 5 cc. remain in the distilling flask. Acidify the distillate with H 2 S0 4 , add 2-3 cc. chloroform, and shake (to make sure that the chloroform remains colorless). Then add 8-10 drops of KI solution, and shake again. (Purple coloration of the chloroform, presence of NITRATE or NITRITE.) Notes. 1. In this procedure the nitrate is reduced by the FeSC>4 to nitric oxide (NO), which passes over as a gas into the receiver, where it is oxidized by the oxygen of the air to HNO 2 , which is then absorbed by the NaOH. When the solution is acidified and KI added, I 2 is liberated by the HNC>2. By this procedure 0.2 mg. NOs can be detected. 2. The reaction is highly characteristic for nitrates and nitrites, since other oxidizing substances (for example, chlorin or bromin) which might liberate iodin from potassium iodide are reduced by the FeSCU to compounds which, even if they pass over into the distillate, have no action on KI. The only substances that may interfere are iodides and thiocyanates ; if these are present, they should be removed before the distillation by treating the substance with 10 cc. H 2 SO4, adding solid Ag 2 SO4, shaking, and filtering. Procedure 125. Detection of Nitrite. To 0.1 g. of the substance add 1 cc. of water and 4-5 drops of HC1. Fill a 7-cm. test-tube with a 20% solution of urea in HC1, and invert it over a small dish containing more of the same solution. Introduce the solution of the substance into the test-tube by means of a small tube which has one end closed with a rubber nipple and the other end drawn out and bent so as to form a small U. Take care not to introduce an air- bubble at the same time. (Formation of gas, presence of NITRITE.) Notes. 1. The reaction between urea and nitrous acid is CO (NH 2 ) 2 -I- 2HNO 2 = CO 2 +2N 2 +3H 2 O. The N 2 is liberated hi the form of minute bubbles which collect at the top of the tube. When much CO 2 is produced, it also separates as a gas; but a small quantity remains dissolved hi the liquid. 2. This procedure enables 0.1 mg. NO 2 to be detected. The amount of nitrite present may be estimated by making a comparativ test with a known quantity of nitrite. 3. The mixture is acidified before it is introduced into the tube so that any carbonate present may be expelled from it. 4. The halogens, chlorin, bromin, and iodin, when dissolved in alkali, de- compose urea with evolution of nitrogen, but they do not do so when dissolved in concentrated HC1. They do not therefore interfere with the test when carried out as above described. Procedure 126. Detection of Hypochlorite. To 0.5 g. of the pow- dered substance add 5 cc. water, and then HAc, a few drops at a time, until the solution is acid. Filter if there is much residue, add 2-3 cc. PbAca solution, heat the mixture to boiling, and let it stand for ten minutes. (Brown precipitate, presence of HTPOCHLORITE.) 114 DETECTION OF ACIDIC CONSTITUENTS. P. 126 Notes. 1. Hypochlorites are commonly met with either in alkaline solution or in the form of a powder (for example, in bleaching powder). Since they are prepared by the action of chlorin on alkali, chloride is ordinarily present in nearly equivalent amount. When the solid powder is treated with water, the hypochlorite passes into solution; and from it the unionized HC1O is liberated upon the addition of the more largely ionized acetic acid. Chlorin is also formed in such quantity as will satisfy the equilibrium-conditions of the reaction HC1O+C1-+H+=C1 2 +H 2 O. When in neutralizing with HAc litmus paper is used, the paper will soon be bleached if hypochlorite is present; but the color at the first instant or on the edges of the bleached portion can usually be observed. 2. This test depends upon the oxidation of the lead salt to lead dioxide (PbC^) by the hypochlorite. The reaction takes place so slowly in the cold that not less than 10 mg. CIO in 5 cc. solution can be detected at room tempera- ture, even if the mixture be allowed to stand a few minutes. But when the mixture is heated, the limit of detectability is about 0.5 mg. in 5 cc. The solu- tion is acidified with HAc, since oxidation does not take place in the presence of a strong acid, such as HNOs. 3. Peroxides in alkaline solution react instantaneously with lead salts, forming PbO2; but this reaction does not take place in the presence of HAc, even on boiling. Therefore in the above procedure a peroxide will not be mis- taken for a hypochlorite. Peroxide and hypochlorite, moreover, cannot exist together, since they react very rapidly with formation of oxygen. 4. This test for hypochlorite may be made even more delicately in alkaline solution, provided peroxides are known to be absent. If the solution is only slightly alkaline, a small white precipitate of Pb(OH) 2 or PbCOs is first formed; but this turns brown if hypochlorite is present when the mixture is heated and allowed to stand. The delicacy is of course diminished by the presence of a large amount of Pb(OH) 2 or PbCOa; but 1 mg. CIO can be detected in the presence of even 2 or 3 g. of these substances, provided an excess of the lead salt is still present in the solution and the mixture is boiled vigorously, prefer- ably in a casserole. Procedure 127. Detection of Chlorate and Br ornate. Treat 0.3 g. of the powdered substance in the cold with 30 cc. water. (If hypo- chlorite is present as shown by P. 126, reduce it to chloride by adding NaAs0 2 solution till a drop of the mixture placed on filter-paper wet with starch and KI solution no longer produces a blue coloration.) Add 20 cc. HN0 3 and 10 cc. AgN0 3 solution. Shake the mixture, and filter off the precipitate. To the filtrate add 5 cc. of saturated 862 solution, and allow it to stand five minutes. If there is a pre- cipitate, heat the mixture nearly to boiling. (White precipitate, presence of CHLOKATE; yellowish precipitate, presence of BROMATE.) Note. See Notes 3 and 4, P. 103. P. 131 DETECTION OF ACIDIC CONSTITUENTS. 115 SUBSTANCES NOT DECOMPOSED BY CONCENTRATED ACIDS. Procedure 131. Detection of All the Acidic Constituents. Treat two 0.3 g. samples of the powdered substance by P. 101; or treat a 2 g. sample of it by the first two paragraphs of P. Ill, testing portions of the distillate thus obtained by P. 112-115. Treat 2 g. of the very finely powdered substance with 30 cc. of cold 1-normal HNOs, filter, and wash the residue. Treat portions of the solution by P. 102-106. Separate the residue from the filter, if possible; incinerating the paper, if much residue adheres to it, in a spiral of platinum wire. Transfer the residue (with the ash) to a nickel crucible, and dry it by igniting it gently. Mix it in the crucible with 10-15 g. dry Na2C03. Cover the crucible, and heat it, preferably within a cylindrical jacket of asbestos-paper, over a power- ful burner for 15-20 minutes. If a perfectly clear fusion does not result, add more Na 2 C0 3 , and heat again. Cool, place the crucible in a casserole, boil it with water till the fused mass is disintegrated, and filter, rejecting the residue. To three-fourths of the filtrate add HNOs till it is distinctly acid, then add 5 cc. more, filter if there is a precipitate (of silicic acid), and test portions of the solution for sulf ate, fluoride, sulfide, halides, and phosphate, by P. 102, 103, and 104. Test the remainder of the filtrate for borate by evaporating it to complete dryness, adding 96% H 2 S04 drop by drop as long as there is effervescence, and treating the mixture by P. 121. If silicate is present, test a fresh sample of the substance for fluoride by P. 122. Notes. 1. Fusion with NaaCOa metathesizes nearly all insoluble com- pounds in the way described in the Notes to P. 7. 2. In minerals or metallurgical products undecomposed by acids, it is usually necessary to test only for silicate, chloride, sulfate, phosphate, borate, and fluoride, since other acidic constituents are scarcely ever present. 3. In the process of fusion changes in the state of oxidation may take place. Thus sulfate may be wholly or partially reduced to sulfide, and sulfide wholly or partially oxidized to sulfate. These two constituents must therefore be distinguished by tests made with the unfused substance, as described in the first part of this Procedure. APPENDICES PREPARATION OF THE REAGENTS ACIDS. Acetic, 6-normal: Mix 350 cc. 99.5% acid with 650 cc. water. Hydrochloric, 12-normal: Use the C. P. acid of commerce of s. g. 1.19. Hydrochloric, 6-normal: Mix 12-normal HC1 with an equal volume of water. Hydrofluoric, 48 percent. : Use the pure acid sold in ceresin bottles. Nitric, 16-normal: Use the C. P. acid of commerce of s. g. 1.42. Nitric, 6-normal: Mix 380 cc. HNO 3 (s. g., 1.42) with 620 cc. water. Perchloric, 2-normal: Use the purest acid of commerce of s. g. 1.12. Phosphoric, 85 percent. : Use the C.P. acid of commerce. Sulfuric, 96 percent.: Use the C. P. acid of commerce of s. g. 1.84. Sulfuric, 6-normal: Pour 96% H 2 S0 4 into five volumes of water. Sulfurous, saturated: Saturate water at 20-25 with S0 2 gas made by dropping 96% H 2 S0 4 into hot NaHS0 3 solution. Tartaric, 10 percent.: To 100 g. of the solid add enough water to make 1000 cc. of solution. BASES. Ammonium hydroxide, 15-normal: Use the C.P. product of s. g. 0.90. Ammonium hydroxide, 6-normal: Mix 400 cc. 15-normal NH 4 OH with 600 cc. water. Barium hydroxide, saturated: Heat 60 g. Ba(OH) 2 .8H 2 with 1000 cc. water, cool to 15, decant or filter. Sodium hydroxide, 6-normal: Add to 250 g. NaOH "purified by alcohol" enough water to make the volume 1000 cc. AMMONIUM SALTS. Acetate, 3-normal : Mix equal volumes of 6-normal HAcand 6-normal NH 4 OH; or dissolve 250 g. of the solid salt in enough water to make the volume 1000 cc. Carbonate: Dissolve 250 g. freshly powdered ammonium carbonate in 1000 cc. 6-normal NH 4 OH, and filter if there is a residue. Chloride, 1-normal: Add to 54 g. NH 4 C1 enough water to make the volume 1000 cc. Molybdate: Dissolve 75 g. of the pure ammonium molybdate of commerce in 500 cc. water, pour the solution into 500 cc. 6-normal HN0 3 , and shake the mixture occasionally till the precipitate is dissolved. Monosulfide, 6-normal: Pass H 2 S gas into 200 cc. 15-normal NH 4 OH in a bottle immersed in running water or in iced water until the gas is no longer absorbed; then add 200 cc. 15-normal NH 4 OH and enough water to make the volume 1000 cc. 117 118 APPENDICES Oxalate, 0.5 normal: Dissolve 35 g. (NH 4 ) 2 C 2 O 4 .H 2 in 1000 cc. water. Polysulfide, 6-normal: Digest one liter of ammonium monosulfide with 25 g. flowers of sulfur for some hours and filter. Name of SaU Bariuin chloride Cadmium nitrate Calcium chloride Calciu^m sulfate Cobalt nitrate Copper Sulphate Ferric chloride Lead acetate Mercuric chloride Potassium chromate Potassium cyanide Potassium ferricyanide Potassium ferrocyanide Potassium iodide Potassium nitrite Potassium permanganate Potassium thiocyanate Silver nitrate Sodium acetate Sodium arsenite Sodium carbonate Sodium nitrite Sodium phosphate OTHER SALTS. Formula and formula-weight BaCl 2 .2H 2 0(244) Cd(N0 3 ) 2 .4H 2 O(308) CaCl 2 .6H 2 O(219) CaS0 4 .2H 2 0(172) Co(N0 3 ) 2 .6H 2 0(291) CuSO 4 .5H 2 O(250) FeCl 3 .6H 2 O(270) Pb(C 2 H 3 O 2 ) 2 .3H 2 0(379) HgCl 2 (271) K 2 Cr0 4 (194) KCN (65) K 3 Fe(CN) 6 (329) K 4 Fe(CN) 6 .3H 2 0(422) KI (166) KN0 2 (85) KMnO 4 (158) KSCN (97) AgNOs (170) NaC 2 H 3 2 .3H 2 0(136) NaAs0 2 (130) Na 2 C0 3 (106) NaNO 2 (69) Na2HPO 4 .12H 2 O(358) Concen- Grams tration 1 -normal 1-normal 1-normal saturated 1% cobalt 1-normal 1-normal 1-normal 0.2-normal 3-normal 1-normal 1-normal 1-normal 0.1 -normal 3-normal 1-per cent. 1-normal 1-normal 1-normal 1-normal 3-normal 3-normal 1-normal per liter 120 150 110 2 50 125 90 190 25 290 65 110 105 17 250 10 100 170 135 130 160 210 120 SPECIAL REAGENTS. Bromin, saturated solution: Shake liquid bromin with water, leaving a small excess of it in contact with the solution. Ether saturated with HC1: pass dry HC1 gas, as long as it continues to be absorbed, into a bottle of ether immersed in ice-water. Ferrous sulfate, 2-normal: Dissolve 280 g. FeS0 4 .7H 2 O in 0.6-normal H 2 S0 4 , and keep in contact with iron nails. Hydrogen peroxide, 3 per cent. Magnesium ammonium chloride, 1-normal in MgCl 2 : Dissolve 100 g. MgCl 2 .6H 2 O and 100 g. NH 4 C1 in water, add 50 cc. 15- normal NH 4 OH, and dilute to 1000 cc. Potassium mercuric iodide, 0.5 normal in K 2 HgIi: Dissolve 115 g. HgI 2 and 80 g. KI in enough water to make the volume 500 cc. ; add 500 cc. 6-normal NaOH ; and decant the solution from any precipitate that may form on standing. Keep this stock solution in the dark. APPENDICES. 119 Potassium pyroantimonate : Add 20 g. of the best commercial salt to 1000 cc. boiling water, boil for a minute or two till nearly all the salt is dissolved, quickly cool the solution, add about 30 cc. 10% KOH solution, and filter. Sodium cobaltinitrite : Dissolve 250 g. NaN0 2 in 500 cc. water, add 150 cc. 6-normal HAc and 25 g. Co(N0 3 ) 2 .6H 2 0, let the mixture stand over night, filter or decant the solution, and dilute it to one liter. Stannous chloride, 1-normal: Dissolve 115 g. SnCl2.2H2O in 100 cc. 12-normal HC1, dilute to 1000 cc., and keep in bottles containing granulated tin. Starch and potassium iodide: Rub 20 g. starch to a thin paste with a little water in a mortar, and pour the paste into 1000 cc. boiling water. Boil for five minutes, and pour the liquid through a funnel plugged loosely with cotton wool. Add to the filtrate 10 g. KI and 5 cc. chloroform. Turmeric: Shake turmeric powder with 95% alcohol and filter. Urea: Dissolve 200 g. urea in 1000 cc. 6-normal HC1. SOLID REAGENTS. Beads (glass). Bismuth dioxide (sold also as sodium bismuthate). Borax (anhydrous). Calcium chloride (dry lumps). Copper (turnings). Ferrous sulfate (powder). Lead (finely granulated). Potassium chlorate (powder). Potassium hydrogen sulfate. Potassium nitrate. Silica (precipitated). Silver sulfate. Sodium carbonate (anhydrous). Sodium peroxide (in 4 oz. cans). Tin (finely granulated). Zinc (finely granulated). SOLVENTS. Chloroform. Ethyl alcohol (95%). Methyl alcohol (acetone-free). PREPARATION OF THE TEST-SOLUTIONS. Of the powdered salt whose formula is given in the middle column of the fol- lowing table weigh out the number of grams given in the last column, and add enough water (or acid when so stated in the foot-note) to make the volume one liter. To prepare the test-solutions, dilute these stock solutions, which contain 100 mg. of the constituent per cubic centimeter, with nine times the volume of distilled water. In a few cases (indicated by the letter H) where the substance is not sufficiently soluble, the stock solution is made up so as to contain 50 mg. of the constituent per cubic centimeter and must be diluted with four times its volume of water to yield the test-solution. Since these solutions serve also for the preparation of the "unknown solutions," the purest salts that can be pur- chased should be employed. Constit- Formula Grams Constit- uent of salt per liter uent Ag AgNO 3 160 Cr Pb Pb(N0 3 ) 2 160 Zn Hg(ous) HgNO 3 .H 2 O 140(a) Fe(ous) Hg(ic) Hg(N0 3 ) 2 160(a) Fe(ic) Bi Bi(N0 3 ) 3 .5H 2 O 230(6) Mn Cu Cu(N0 3 ) 2 .3H 2 O 380 Ni Cd Cd(N0 3 ) 2 .4H 2 275 Co As(ous) As 2 O 3 13(c) Ba As(ic) As 2 O 5 150 Sr Sb SbCl 3 190(d) Ca Sn(ous) SnCl 2 .2H 2 O 190(e) Mg Sn(ic) SnCl 4 .3H 2 O 270(e) Na Al A1(NO 3 ) 3 .9H 2 700H K C0 3 Na 2 C0 3 180 Cl 80s Na 2 SO 3 160 Br SO 4 K 2 S0 4 90H I CIO NaOCl ...to) SCN s Na 2 S.9H 2 750 N0 3 CN KCN 250 C1O 3 N0 2 KNO 2 185 P0 4 Cr0 4 K 2 Cr0 4 170 P0 4 Formula of salt CrCls(50%sol'n) Zn(N0 3 ) 2 FeSO 4 .7H 2 Fe(N0 3 ) 3 .9H 2 O Mn(NO 3 ) 2 .6H 2 Ni(N0 3 ).6H 2 Co(N0 3 ) 2 .6H 2 BaCl 2 .2H 2 Sr(N0 3 ) 2 Ca(N0 3 ) 2 .4H 2 Mg(NO 3 ) 2 .6H 2 O NaNO 3 KN0 3 KC1 KBr KI KSCN KN0 3 KClOa Grams per liter 610 290 250H(/) 715 530 500 500 180 240 590 530H 370 260 210 150 130 170 165 75H Na 2 HP0 4 .12H 2 380 Ca 3 (P0 4 ) 2 160(6) (a) Dissolve in 0.6-normal HNO. (ft) Dissolve in 3-normal HNOi. (c) Digest with 500 oc. 12-normal HC1; then add 500 cc. water, yielding the test-solution of AsCls containing 10 mg. As per cubic centimeter. (d) Dissolve in 6-normal HC1; and, in making the test-solution, dilute with 2-normal HC1. () Dissolve in 6-normal HC1. (/) Dissolve in 1-normal HSO4, and keep in contact with iron nails. (0) To 600 cc. of the 31% solution of commerce add 450 cc. water, yielding the test-solution containing 10 mg. CIO per cubic centimeter. 120 APPENDICES. 121 UNKNOWN SOLUTIONS. The "unknown solutions" given to the student should contain the constituents to be tested for in quantities which are definitly known by the instructor. As a rule they may well contain in 10 cc. 300 mg. of one of the constituents, 30 mg. of another of the constituents, and 3 mg. of each of two or three of the remaining constituents of the group in question. Such solutions may be conveniently prepared in advance by mixing in a 250 cc. bottle 60 cc. of the stock solution of the first constituent (or 120 cc. if it is half-strength as shown by an H in the table), 6 cc. of the stock solution of the second constituent (or 12 cc. if half- strength), and 6 cc. of the test-solutions of the other constituents, and diluting with enough water to make the volume 200 cc. Of these "unknown solutions" just 10 cc. should be given out to each student for analysis. When time permits two unknowns being done on any group, the second may well contain only 2 mg. of some of the constituents. APPARATUS REQUIRED. Returnable. 4 Lipped Beakers, 120 cc. to 500 cc. 2 Casseroles, 50 cc. 2 Casseroles, 100 cc. 2 Casseroles, 200 cc. 1 Conical Flask, 30 cc. 4 Conical Flasks, 75 cc. 4 Conical Flasks, 200 cc., with 1 two- hole rubber stopper. 1 Conical Flask, 500 cc. 2 Jena Round-Bottom Flasks, 100 cc. with 1 two-hole rubber stopper. 1 Filter Flask, 500 cc., with a one-hole rubber stopper. 1 Flat-Bottom Flask, 750 cc., with a two-hole rubber stopper. 1 Flat-Bottom Flask, 250 cc., with a two-hole rubber stopper. 3 Funnels, 2.5 in. 1 Funnel, 2 in. 4 Watch-Glasses, 3 cm. 2 Watch-Glasses, 7 cm. 2 Watch-Glasses, 10 cm. 12 Test-Tubes, 15 cm. 6 Test-Tubes, 7 cm. 1 Separating Funnel, 100 cc. 1 Graduate, 10 cc. 1 Graduate, 50 cc. 1 Nickel Crucible, 20 cc. 1 Porcelain Mortar and Pestle. 1 Desk Key. 1 Gas Burner (Tin-ill). 1 Lamp-Stand and Rings. 1 Filter Stand. 1 Test-Tube Rack. Not Returnable. 12 Hardened Filters, 5.5 cm. 1 Pkg. Filters, 7 cm. 1 Pkg. Filters, 9 cm. 1 Monel-Metal Wire-Gauze, 12x12 cm. 1 Box Labels. 1 Note-Book. 1 Piece Platinum Foil, 2 cm. x 1 cm. 1 Piece Platinum Wire, 10 cm. 1 Horn Spoon (bowl 1 cm. long). 1 Piece Glass Rod, 50 cm. long. 1 Sponge. 2 Towels. 1 Nichrome Triangle. 1 Piece Rubber Tubing, 60 cm. long, 6 mm. bore. 1 Piece Rubber Tubing, 60 cm. long, 4.5 mm. bore. 1 Rubber Nipple. 2 Pieces Glass Tubing, each 75 cm. long, 4 mm. bore. 1 Piece Hard-Glass Tubing, 50 cm. long, 6 mm. bore. 1 Tube Blue Litmus Paper. 1 Tube Red Litmus Paper. 1 Box Matches. 1 Triangular File. 1 Screw Clamp. 1 Test-Tube Brush. 122 IONIZATION VALUES. The following table shows approximately the percentage of the substance which is dissociated into its ions in 0.1 normal solution at 25. In the case of the dibasic acids the value opposit the formula of the acid shows the percentage of the first hydrogen that is dissociated, and that opposit the acid ion (HA~) shows the percentage of it dissociated (into H> and A- for the case that these two ions are present in equal quantities). Salts of type B+A~ (e. g., KN0 3 ) 84% Salts of type B+ 2 A= or B++A- 2 (e. g., K 2 S0 4 or BaCl 2 ) 73 Salts of type B+ 3 A a or B+++A~ 3 (e. g., K3Fe(CN) 6 or AlCls) 65 Salts of type B++A- (e. g., MgSO 4 ) 40 KOH, NaOH 90 Ba(OH) 2 80 NH 4 OH 1 HC1, HBr, HI, HSCN, HN0 3 , HC1O 3 , HC10 4 , H 2 S0 4 , H 2 Cr0 4 .. .90 HsP0 4 , H 3 AsO 4 , H 2 S0 3 , H 2 C 2 O 4 , HS0 4 ~ $0-45 HN0 2 , HF 7-9 HAc, HCaO,-, HSO 3 ~ 1-2 H 2 S, H 2 C0 3 , H 2 PO 4 ~, HOOr 0.1-0.2 HB0 2 , HAs0 2 , HCN, HC0 3 - 0.002-0.008 HS-, HP0 4 - 0.0001-0.0002 HOH 0.00,000,02 123 ATOMIC WEIGHTS OF THE COMMON ELEMENTS. Aluminum Al Antimony Sb Arsenic As Barium Ba Bismuth Bi Boron B Bromin Br Cadmium Cd Calcium Ca Carbon C Chlorin Cl Chromium Cr Cobalt Co Copper Cu Fluorin F Gold Au Hydrogen H lodin. I 27.1 120.2 74.96 137.37 208.0 11.0 79.92 112.40 40.07 12.00 35.46 52.0 58.97 63.57 19.0 197.2 1.008 126.92 Iron Fe 55.84 Lead Pb 207.10 Magnesium Mg 24 . 32 Manganese Mn 54 . 93 Mercury Hg 200.6 Molybdenum Mo 96 . Nickel Ni 58.68 Nitrogen N 14.01 Oxygen O 16.00 Phosphorus P 31.04 Potassium K 39.10 Silicon Si 28.3 Silver Ag 107.88 Sodium Na 23.00 Strontium Sr 87 . 63 Sulfur S 32.07 Tin Sn 119.0 Zinc . . Zn 65 . 37 SOLUBILITIES OF SLIGHTLY SOLUBLE SUBSTANCES. The numbers in the table show the solubility in milli-equivalents per liter at 20. The letters v.s. (very soluble) denote a greater solubility than 1-normal. In the case of the carbonates the values have been corrected for hydrolysis so as to correspond to the ion-concentration product in the saturated solution. For a general statement in regard to the solubilities of other substances, see Note 9, page 31. Mg Ca Sr Chloride Bromide Iodide Thiocyanate Sulfate Chromate Carbonate Hydroxide Fluoride Oxalate Phosphate v.s. v.s. v.s. v.s. v.s. v.s. 10. 0.3 2.8 5. Ca v.s. v.s. v.s. v.s. 30. 60. 0.1 45. 0.4 0.09 0.7 v.s. v.s. v.s. v.s. 1.5 12. 0.1 130. 1.9 0.5 Ba v.s. v.s. v.s. v.s. 0.02 0.03 0.1 450. 18. 0.8 Pb 70. 45. 2.6 28. 0.28 0.0003 0.0004 0.2 5. 0.012 0.001 Ag 0.01 0.0005 0.00002 0.0008 50. 0.16 *0.2 0.18 v.s. 0.24 0.05 124 UNIVERSITY OF CALIFORNIA LIBRARY BERKELEY Return to desk from which borrowed. This book is DUE on the last date stamped below. 24Feb'58HK REC'D LD 1 1958 MAY 22 <5953 , nlun'SSRHf REC'D LD MAY 2 0-1933 LD 21-100m-ll,'49(B7146sl6)476 REG D MAR 2 1962 SEP racut, RPR 17 302162 UNIVERSITY OF CALIFORNIA LIBRARY