SMWWttMffiK UNIVERSITY OF CALIFORNIA. FROM THE LIBRARY OF DR. JOSEPH LECONTE. GIFT OF MRS. LECONTE. No. n ( pyi-&~ ELEMENTS OF CHEMICAL ANALYSIS, INORGANIC AND ORGANIC. BY EDWARD ANDREW PARNELL, CHEMICAL ASSISTANT IN UNIVERSITY COLLEGE, LONDON. LONDON: PRINTED FOR TAYLOR AND WALTON, BOOKSELLERS AND PUBLISHERS TO UNIVERSITY COLLEGE, 28. UPPER GOWER STREET. 1842. LONDON : Printed by A. SPOTTISWOODE, New- Street- Square. TO THOMAS GRAHAM, ESQ., F.E.S. L. & ED., PROFESSOR OF CHEMISTRY IN UNIVERSITY COLLEGE, LONDON, PRESIDENT OF THE CHEMICAL SOCIETY, ETC. HAD I no other inducement, my dear Sir, in offering you the present Work, than your high and well- deserved scientific reputation, ample, indeed, would that have been; but the personal attention and kindness you have always manifested towards me constitute not a slight additional incentive to the very small tribute of esteem, the paying of which is now the privilege of Your late pupil, and attached friend, THE AUTHOR. A 2 j * '-* PREFACE. IN the first chapter and division of this work, in- structions are offered for performing the ordinary manipulations of analytical processes ; to which are added a description of the most important reagents employed, and the means of detecting the impurities with which these are occasionally contaminated. The second chapter is devoted to the statement of the appearances produced by the principal reagents when applied to various chemical substances, the results being thrown into a series of tables, and is followed by a chapter containing full instructions for the performance of the qualitative analysis of every variety of substance. This last chapter includes the use in analysis of the mouth blowpipe. In the fourth and last chapter and division, quan- titative analysis is treated of. The processes are there described by which all bodies which enter into the composition of familiar substances are separated and their weights determined j with an account, in conclusion, of the operations of organic analysis, including the recent improvements on the mode of determining the amount of nitrogen. A large proportion of the processes described are those recommended by our two greatest authorities A 3 VI PREFACE. in analytical chemistry Berzelius and Rose, to whose works, especially to the last German edition (1838) of the Handbuch der Analytischen Chemie of the latter, the reader is referred for much inform- ation which could not be included in the more limited plan of the present work. But several of the processes described are of very recent invention, and new to treatises on the subject. Some of these will be found of considerable value. An Appendix is added, in which the mode of calculating the atomic constitution of a substance from its per-centage composition, when known, is described, with the methods of taking the specific gravity of matter in different conditions. Numerical tables are also appended, which have been formed to expedite the calculations of analysis. University College, July, 1842. CONTENTS. CHAPTER I. MANIPULATIONS AND REAGENTS. Page SECTION I. Manipulations in qualitative Analysis - 1 SECTION II. Manipulations in quantitative Analysis - 4 SECTION III. Reagents - 12 Abbreviations and Symbols - - - 16 CHAPTER II. BEHAVIOUR OP SUBSTANCES WITH REAGENTS. Division of metallic Oxides and Acids into Three Classes according to their Behaviour with Sulphuretted Hydrogen and Hydrosulphate of Ammonia _ 17 Behaviour of Salts of First Class of Oxides with Reagents. No. 1. - 18, 19 Behaviour of Salts of First Class of Oxides with Reagents. No. 2. - 20, 21 Behaviour of Salts of Second Class of Oxides with Re- agents. A. (Alumina to Oxide of Manganese.) No. 1. 22, 23 Behaviour of Salts of Second Class of Oxides with Re- agents. A. No. 2. - - 24, 25 Behaviour of Salts of Second Class of Oxides with Re- agents. B. (Oxide of Manganese to Oxide of Zinc.) No. 1. - - 26, 27 Behaviour of Salts of Second Class of Oxides with Re- agents. B. No. 2. - _ 28, 29 Behaviour of Salts of Third Class of Oxides with Re- agents. A. (Oxide of Antimony to Iridic Oxide.) No. 1. . - - 30, 31 Behaviour of Salts of Third Class of Oxides with Re- agents. A. No. 2. - - 32, 33 Behaviour of Salts of Third Class of Oxides with Re- agents. B. (Iridic Oxide to Palladious Oxide.) No. 1. 34, 35 Behaviour of Salts of Third Class of Oxides with Re- agents. B. No. 2. - - 36 ; 37 Vlll CONTENTS. Page Behaviour of Salts of Third Class of Oxides with Re- agents. C. (Palladious Oxide to Peroxide of Tin.) No. 1. _ - 38, 39 Behaviour of Salts of Third Class of Oxides with Re- agents. C. No. 2. - - 40, 41 Behaviour of Third Class of Oxides with Reagents. Acids. D. (Antimonic Acid to Vanadic Acid.) No. 1. 42, 43 Behaviour of Third Class of Oxides with Reagents. Acids. D. No. 2. - 44, 45 Behaviour of Acids with Reagents. A. (Acetic to Formic.) 46, 47 Behaviour of Acids with Reagents. B. (Formic to Hy- drosulphuric.) - - 48, 49 Behaviour of Acids with Reagents. C. (Hydrosulphuric to Perchloric.) - 50, 51 Behaviour of Acids with Reagents. D. (Perchloric to Tar- taric.) - - 52, 53 CHAPTER III. QUALITATIVE ANALYSIS. Preliminary Observations. List of Bases and Acids to be sought - 54 SECTION I. Qualitative Analysis of a Salt consisting of a single Acid and a single Base contained in the List, pages 57* and 58. - 58 Examination for Base - - 60 Examination for Acid - - 63 SECTION II. Qualitative Analysis of a Compound which can con- tain all the Bases and Acids in the List, pages 57. and 5. 65 Determination of Bases - - 66 I. Examination of the Precipitate by Sulphuretted Hydrogen 66 II. Treatment of the Solution filtered from the Precipitate by Sulphuretted Hygrogen - 67 Treatment of the Solution in Hydrochloric Acid of the Pre- cipitate by Hydrosulphate of Ammonia - 68 III. Examination of the Solution filtered from the Precipitate by Hydrosulphate of Ammonia - - 69 Determination of Acids - - - 70 SECTION III. Qualitative Analysis of a Substance which may con- tain all inorganic Bodies whose Properties are well known 71 I. Examination for Bases and some Acids - 71 Treatment of the Precipitate formed by Sulphuretted Hy- drogen - - _ - 72 Treatment of the Solution filtered from the Precipitate by Sulphuretted Hydrogen - - 73 II. Examination for Acids - - - J4< CONTENTS. ix Page SECTION IV. Examples of the qualitative Analysis of Metallic Alloys - - - - - -75 Analysis of Berlin Silver - - - 76 Analysis of Newton's fusible Metal - - - 77 SECTION V. Qualitative Analysis of Silicates - - 78 Means of effecting Solution - 78 Constituents of Silicates - 80 Analysis of a Silicate insoluble in Hydrochloric Acid,, sus- pected to contain an Alkali - - 83 Examples of the qualitative Analysis of Silicates, Stilbite, Hornblende, Lepidolite - 84 SECTION VI. Qualitative Analysis of Mineral Waters: 1. For the Gases evolved. 2. For the Bodies rendered insoluble by Evaporation. 3. For the Bases in Solution filtered from the Precipitate produced by Ebullition. 4. For the Acids - 86 SECTION VII. Use of the Blowpipe in qualitative Analysis - 93 Construction of the Blowpipe, Blowpipe Flame, Blowpipe Supports - Q4> Behaviour of Alkalies and Earths before the Blowpipe 98, 99 Behaviour of Oxides of Metals proper before the Blowpipe (Oxide of Antimony to Oxide of Manganese) - 100, 101 Behaviour of Oxides of Metals proper before the Blowpipe (Oxide of Manganese to Oxide of Zinc) - 102, 103 Blowpipe Tests for Acids - 104 Blowpipe Operations - 106 First Operation Heating in a glass Tube closed at one end 106 Second Operation Heating alone in the open Air - 107 Third Operation Heating in a glass Tube open at both ends 108 Fourth Operation Heating with Reagents: 1. With Car- bonate of Soda. 2. With Borax. 3. With Microcosmic Salt - - 109 SECTION VIII. Analysis of Gases - - - 111 Properties of Gases - 1 1 1 Qualitative Analysis of a mixed Gas - 111 Analysis of Coal Gas (qualitative and quantitative) - 114 SECTION IX. Detection of Poisons in organic Mixtures - - 115 Arsenic - 115 Mercury - 120 Antimony - - - 120 Copper - 120 Lead - - 121 Nitric Acid - 122 Sulphuric Acid - - - 122 Hydrochloric Acid - - 123 Oxalic Acid - 123 Hydrocyanic Acid - 124 X CONTENTS. Page SECTION X. Qualitative Analysis of Urine and Discrimination of Urinary Calculi -124 Analysis of Urine - -124 Discrimination (by chemical tests) of Urinary Calculi, consist- ing of a single Deposit or of alternating Calculi - - 126 CHAPTER IV. QUANTITATIVE ANALYSIS. Preliminary Observations - - -128 SECTION I. Alkalies and their metallic Bases - - - 130 1. Potassium - -130 2. Sodium - - 131 Alkalimetry - - 132 3. Ammonia - - - 135 4. Lithium - 135 SECTION II. Alkaline Earths and Earths proper, and their metallic Bases - 136 1. Barium - 136 2. Strontium - - 137 3. Calcium - - - 138 4. Magnesium - - - - - 139 5. Aluminum - - 144 6. Yttrium - 147 7. Thorinum - - - - - - 147 8. Glucinum - - 147 9. Zirconium - - - 149 SECTION III. Metals proper, not precipitated by Sulphuretted Hydrogen from their Acid Solutions - - - 149 1. Cerium _ 149 2. Manganese - - - 150 Valuation of Peroxide of Manganese - - - 152 3. Iron . - 153 Analysis of Clay Ironstone - - - - 155 4. Nickel - 156 5. Cobalt - - 159 6. Zinc - - - 165 7. Uranium - - - - - -169 8. Chromium - . . _ . -170 9. Titanium - - - _ - 171 SECTION IV. Metals proper, precipitated by Sulphuretted Hydrogen from their Acid Solutions - 172 1. Lead - - 172 2. Cadmium - - 174 3. Bismuth - - - - - - 174 CONTENTS. XI Page 4. Copper - - - 176 5. Silver - 179 6. Gold - 180 7. Mercury - - 182 8. Platinum, Palladium, Osmium, Iridium, and Rhodium - 186 9. Tin - 195 10. Antimony - - - 196' 1 1 . Molybdenum - 200 12. Tungsten - - - 201 13. Arsenic r . - - - 202 SECTION V. Non-metallic Bodies - - 207 1. Silicon - 207 Analysis of natural Silicates - - 207 2. Sulphur - - 212 Analysis of Sulphurets - - - 214 3. Chlorine - - - 217 Chlorimetry - - 218 4. Iodine - - 220 5. Bromine - - 222 6. Fluorine - - - 223 7. Phosphorus - - 224 8. Boron - 227 9. Nitrogen - - - - - 228 Eudiometry - - 230 10. Carbon - - 233 Analysis of Gunpowder - 235 11. Water - - 237 Analysis of Mineral Waters : Saline, Chalybeate, and Car- bonated -_---_ 239 Alkaline Waters - - 247 Sulphureous Waters - - 248 Determination of the Degree of Hardness of Waters - 249 SECTION VI. Organic Analysis - - 254 Determination of Nitrogen ... 264 APPENDIX. Method of calculating the Atomic Constitution of a Body from its per-centage Composition - 269 Methods of taking the specific Gravity of Bodies - - 270 Addition on Chlorimetry - 279 Tables for Calculations in Analysis - - 282 Tables of Atomic Weights, with their Logarithms - . 297 General Index - - - - - 301 ELEMENTS OF CHEMICAL ANALYSIS. /^\SRA/ Or TH UNIVERSITY CHAPTER I. > Phosphate of Soda. Bar of Metallic Zinc Iodide of Potassium. Remarks. White, sol. in potash. Reduced lead. Yellow, sol. in ex. SO 3 . a white ppt. HC1. in a strong sol. a white ppt. Chromate of potash, a yellow ppt, sol. in pot. White. Mercury re- duced. Greenish yel- low, becom- ing black, sol. in ex. HC1. a white ppt. blackened by am. Chromate of pot. a red ppt. White, in ni- trate or sul- phate. A faint trou- bling in chloride. Mercury re- duced. First yellow, then scar- let, sol. in ex., sol. in HC1. Protochloride of tin in excess gives metal- lic mercury. In a galvanic circle, formed by a drop of sol. of mercury on gold, touched with a piece of iron, mer- cury is deposited on the gold. Brown black. Brownish white. Black, formed slowly. Black powder of reduced osmium. Black after standing. Free chloride of os- mium is decom- posed by water into osmic acid and os- mium, which is pptd. Brown. Black powder of reduced palladium. Black. Cyanide of mercury gives a yellowish white ppt. of cyan- ide of palladium, sol. in HC1. D 3 38 BEHAVIOUR OF SUBSTANCES Behaviour of Salts of THIKD CLASS Potash. Carbonate of Potash. Bicarbonate of Potash. Platinous oxide. Brown, slowly formed. Platinic oxide. Yellow, sol. in ex. if hot ; in- sol. in HC1. Yellow, insol. in ex. Yellow, insol. in ex. Rhodic oxide. On ebullition, gelat. brownish yellow. Yellowish, formed slowly. Silver, oxide of. Brown, insol. in ex., sol. in am. White, sol. in am. White, sol. in am. Telluric oxide. White, sol. in ex. White, sol. in ex. White, sol. in ex. Tin, prot- oxide of. White, sol. in ex., soln. gives metallic tin on heating. White, insol. in ex. White, insol. in ex. Tin, peroxide of. White, sol. in ex. White, efferves- cence ; sol. in ex. White, efferves- cence ; insol. in ex. WITH REAGENTS. 39 OF OXIDES with Reagents. C. No. 1. % Ammonia. Carbonate of Ammonia. Yellow Prus- siate of Potash. Red Prussiate of Potash. Green, cryst. Yellow, sol. in ex. if hot ; in- sol. in alka- lies. Yellow, insol. in ex. Yellow, insol. in acids, sol. in hot alkalies. Yellow, insol. in acids, sol. in hot alkalies. Yellowish, sol. in HC1. Yellowish, formed slowly. Brown, very sol. in ex. White, sol. in ex. White. Reddish brown. White, sol. in ex. White, sol. in ex. White, insol. in ex. White, insol. in ex. White, gelat. White. White, sol. in ex. White, efferves- cence ; insol. in ex. Yellowish, ge- lat. on stand- ing. D 4 40 BEHAVIOUR OP SUBSTANCES Behaviour of Salts of THIKD CLASS Hydrosulphate of Ammonia. Sulphuretted Hydrogen. Oxalic Acid. Platinous oxide. Dark brown, sol. in large ex. Black, formed slowly, slight- ly sol. in ex. Platinic oxide. Brownish black, sol. in ex. Brown, formed slowly. Rhodic oxide. Brown, insol. in ex. Brown, formed slowly. Silver, oxide of. Black, insol. in ex. Black. White, sol. in am. Telluric oxide. Brown, sol. in ex. Brown. Tin, prot- oxide of. Brown, sol. in ex. Brown. White. Tin, per- oxide of. Yellow, sol. in ex. Yellow, in acid sol. WITH KEAGENTS. OF OXIDES with Reagents. C. No. 2. 41 Phosphate of Soda. Bar of Metallic Zinc. Iodide of Potas- sium. Remarks. Black powder Black, metallic Behaviour with am. is of reduced appearance, characteristic. platinum. formed slowly. Ditto. Brown. Behaviour with potash and am. is charac- teristic. Chloride of potassium and mur. am. give yel- low ppts. Metallic rho- Dark ppt. formed Salts have mostly a dium. slowly. red colour. Yellow, sol. Metallic silver. Yellowish white, HC1. gives a white in am. sol. in ex., curdy ppt. of chlo- insol. in am., ride of silver, insol. insol. in NO. in acids, very sol. in am. (character- istic.) White. Tellurium, (See Blowpipe bulky black. Tables.) White. Metallic tin., Yellowish white, Chloride of gold gives grey. sol. in ex., sol. a blue ppt. (purple in HC1. of Cassius.) White. White, gelat., Behaviour with sul- perox. of phuretted hydrogen tin, hydro- and hydrosulph. gen evolved. am. is character- istic. 42 BEHAVIOUR OF SUBSTANCES Behaviour of THIRD CLASS OF OXIDES Acids (in combination). Nitrate of Silver. Chloride of Barium. Antimonic. White. White. Antimonious. White. White, slightly sol. in water. Arsenic. Chocolate red, sol. in ni- tric acid, sol. in am. White, sol. in acids, sol. in mur. am. Arsenious. Pale yellow, sol. in acetic acid, sol. in am. ; the ammonio-nitrate of silver gives the same ppt. in free acid. Chromic. Reddish brown. Canary yellow, sol. in N0 5 . Molybdic. White, sol. in water, ni- tric acid, and am. White, sol. in water ; sol. in NO 5 . Hyper-Manganic. A yellowish brown pre- cipitate. Manganic. Black, of perox. of man- ganese. Osmic. Brown. Brown. Selenious. White, slightly sol. in N0 5 . White, sol. in HC1. Tungstic. White. White. Yanadic. Bulky yellow, sol. in NO 5 , and in am. Bleaches on exposure. Bulky orange yellow. Bleaches on ex- posure. WITH REAGENTS. with Reagents. Acids. D. No. 1. 43 Acetate of Lead. Protonitrate of Mercury. Nitrate of Lime. White. White. White. White. White. White, slightly sol. in water. White, sol. in NO 5 , insol. in acetic acid. Yellowish white, sol. in NO 5 . White, sol. in HC1., in mur. am. and in acetic acid. White, sol. in NO 5 and in acetic acid. Yellowish white, sol. in NO 5 . Yellow, sol. in alka- lies. Yellowish red. White, sol. in NO 3 . Yellowish, sol. in N0 5 . White, sol. in water, in HC1. and in NO 5 . A ppt. of ox. of lead (carbonate?), and of perox. manganese. Ppt. of perox. man- ganese. Ditto. Ditto. Perox. manganese. Brown, with basic acetate of lead. Brown, sol. in NO 5 . Brown. White, slightly sol. in N0 5 . White, sol. in NO 5 . White, sol. in NO 5 . White. White. White. Yellow, sol. in water. Yellow. 44 BEHAVIOUR OF SUBSTANCES Behaviour of THIRD CLASS or OXIDES Acids (in combina- . tion). Sulphuretted Hydro- gen (in acid soln). Hydrosulphate of Ammonia. Antimonic. Orange yellow, .sol. in alkalies. Orange yellow, sol. in ex. Antimonious. Ditto. Ditto. Arsenic. Yellow, sol. in al- kalies. Yellow, on adding an acid. Arsenious. Ditto ; slightly sol. in pure water. Ditto. Chromic. Chromic acid re- duced to oxide of chromium. Greenish (oxide of chro- mium). Molybdic. Brown. Brown on adding an acid. Hyper-Manganic. In neutral soln. flesh colour. Flesh colour, insol. in ex. Manganic. Ditto. Ditto. Osmic. Brown black. Black, insol. in ex. Selenious. Citron yellow, sol. in hydrosulph. am. Yellow, sol. in ex. Tungstic. Slight turbidity. Brown with very dilute HC1. Vanadic. Grey. Sol. becomes brown. WITH REAGENTS. 45 with Reagents. Acids. D. No. 2. Rod of Metallic Zinc. Hydrochloric Acid. Remarks. With HC1. a black powder (antimony). White (antimonic acid). Ditto. White (antimonious) acid). With HC1. gives metallic arsenic. (See Section on Detection of Poisons.) Ditto. Sulphate of copper gives a green precipitate. Am monio -sulphate of cop- per gives a green ppt. in free acid. With sulphuric acid, chromic acid be- comes oxide. Chlorine evolved. Organic bodies reduce chro- mic acid to oxide of chromium. ~ Dark brown with HC1. (or sol. in ex.) Perox. Manganese. Chlorine evolved. Sol. decolorized by sul- phurous acid. Ditto. Ditto. Ditto. With HC1., reduced osmium. Free osmic acid, known by its odour. Sulphurous acid and or- ganic bodies slowly re- duce osmic acid. With HC1., reduced selenium. Sulphurous acid gives a floccyred ppt. of reduced selenium. With HC1. or S0 3 a fine blue colour. White, insol. in ex. Chlorine evolved. Sulphurous acid and or- ganic bodies reduce va- nadic acid to oxide of vanadium. 46 BEHAVIOUR OF SUBSTANCES Behaviour o/*AciDS with Reagents Acids (neutralized). Chloride of Barium. Nitrate of Lime. Acetate of Lead. Nitrate of Silver. Acetic. A cry st. ppt. if solns. be conctd. Boracic. White, sol. in ex., in acetic acid, and in mur. am. White, sol. in ex., sol. in mur. am. White, sol. in acetic acid. White, sol. in acetic acid. Bromic. White, sol. in water. White, sol. water. White, sol. in N0 5 . Carbonic. White. White. White. Brownish white, sol. in am. Chloric. Citric. White in in strong solns. White, slightly sol. in water. White. White. Formic. Except in very strong so- lutions. Heated slowly, a black pow- der of me- tallic silver. WITH REAGENTS. (Excess neutralized by an Alkali). A. 47 Protonitrate of Mercury. Sulphuric Acid (concentrated). Remarks. A cryst. ppt. if solns. be conctd. Yellowish white, sol. in ex., sol. in NO 6 . Yellow,, sol. in N0 5 . Light brown or black. White. When heated, mer- cury is reduced to the metallic state. Acetates moistened with SO 3 evolve acetic acid, known by its pungent odour. Borates moistened with SO 3 and then with al- cohol, cause the latter to burn with a green flame. Bromine is evolved, sol. becoming yellow. Effervescence of carb. acid, in- odorous. Bromates mixed with com- bustibles detonate when heated, or when mois- tened with oil of vitriol. Carbonates effervesce with acids. Peroxide of chlorine evolved ; liquid bleaches vegetable colours. Evolves formic acid, known by its pungent odour. The ppt. by a citrate in a salt of lead, when wash- ed, dissolves in ammonia. Characteristic test is heat- ing with nitrate of silver. 48 BEHAVIOUR OF SUBSTANCES Behaviour of AciDS with Reagenti Acids (neutralized). Chloride of Barium. Nitrate of Lime. Acetate of Lead. Hydrochloric. White, slightly sol. in water. Hydrocyanic. If moderately di- White. lute. Hydrobromic. White, sol. in N0 5 . Hydrofluoric. White, sol. in Bulky white. White, sol. in HCL HCL Hydriodic. Yellow, sol. in N0 5 . Hydroselenic. Black. Hydrosulpho- cyanic. if moderately di- lute. Hydrosul- phuric. Black. WITH REAGENTS. (Excess neutralized by an Alkali). B. 49 > Nitrate of Silver. Protonitrate of Mercury. Sulphuric Acid (concentrated). Remarks. Curdy white, insol. in NO 5 , sol. in am. White, insol. in cold NO 6 blackened by am. Evolves HC1. Nitrate of silver with nitric acid is a cha- racteristic test. White, insol. in dilute N0 5 ; sol. in am. and in cyanide of potm. Grey, reduced mercury. Evolves hy- drocyanic acid. Pppte a mixture of proto and per-salt of iron by potash, add a sol. cyanide, and treat with HC1. Pruss. blue remains. (Scheele's test.) Yellow, insol. in NO 5 . ; sol. in am. Yellow, insol. in cold NO 5 . Bromine is evolved. Chlorine water sets bromine free, soln. becoming brown ; colour removed by agitation with ether. Hydrofluoric acid evolved, ascertained by its corroding glass. Yellow, insol. in NO 5 , nearly insol. in am. Yellowish green, in- sol. in NO 5 . Iodine evolved. Chlorine water added gradually renders iodine free, ascer- tained by starch, which gives a blue colour. Black. Black. Selenietted hydrogen evolved ; odour of decayed cabbage. Odour produced be- fore the blowpipe is the most character- istic test. Curdy white, sol. in strong but not in dilute am. Grey, reduced mercury. Free hydro- sulphocyanic acid ; depo- sits a yellow powder. Gives a red colour with persalts of iron, destroyed by chlo- ride of mercury. Black, insol. in am. Black, insol. in NO 5 . Sulphuretted hydrogen disengaged. Foetid odour of free gas destroyed by Cl. with deposition of sulphur. 50 BEHAVIOUR OP SUBSTANCES Behaviour of ACIDS with Reagents Acids (neu- tralized). Chloride of Barium. Nitrate of Lime. Acetate of Lead. Nitrate of silver. Hyposul- phuric. Hyposul- phurous. unless con- centrated. White, slightly sol. in N0 5 . White, be- comes black. lodic. White, sol. in NO 5 . White, sol. in NO 5 . White, sol. in NO 5 . White, sol. in N0 3 and in am. Nitric. Nitrous. White, slightly sol. in water. Oxalic. White, nearly in- sol. in wa- ter, sol. in acids. White, in- sol. in water and dilute acids. White. White. Perchlo- ric. WITH REAGENTS. 51 (^Excess neutralized by an Alkali). C. ; Protonitrate of Mercury. Sulphuric Acid (concentrated). Remarks. Sulphurous acid e- volved ; no ppt. of sulphur. HC1. evolves sulphurous acid, and sulphuric acid remains, ascertained by a salt of ba- rytes. Black, insol. in NO 5 . Sulphurous acid dis- engaged, and free sulphur deposited. Soluble hyposulphites dissolve chloride of silver, forming a sweet solution. White, sol. in NO 5 . Sulphurous acid developes free iodine, indicated by starch, which gives a blue colour. With copper filings nitric oxide is e- volved. Free nitric acid with HC1. dissolves gold. Bleaches indigo, if hot. A nitrate, heated with protosulph. iron and oil of vitriol, gives a dark brown colour. Nitric oxide and ni- tric acid formed. With HC1. does not dissolve gold. White. Heated with sulphuric acid, a dry oxalate gives car- bonic oxide and carbonic acid gases, the first is in- flammable, the last is detected by causing lime-water to become milky. Sol. of oxalic acid ppts. a soln. of sulphate of lime. The free acid pptes. salts of potash in strong solu- tions. When heated to redness, perchlorates evolve oxygen gas, and leave chlorides. E 2 52 BEHAVIOUR OF SUBSTANCES Behaviour of ACIDS with Reagents Acids (neutralized). Chloride of Barium. Nitrate of Lime. Acetate of Lead. Phosphoric (tribasic). White, sol. in HC1. or mur. am. in ex. White, sol. in N0 3 . White, sol. in NO 5 , insol. in acetic acid. Phosphorous. White, sol. in HC1. sol. in mur. am. White, sol. in N0 3 . White, sol. in NO 3 , insol. in acetic acid. Selenic. White, insol. in N0 5 . White, sol. in N0 5 . White, insol. in N0 3 . Silicic. White, sol. in dilute HC1. White, sol. in dilute NOs. White, sol. in NO 5 . sol. in potash. Sulphuric. White, insol. in acids. White, sol. in water. White, slightly sol. in NO 5 . Sulphurous. White, sol. in HC1. White, sol. in water. White, sol. in dilute NO 5 . Tartaric. White, slightly sol. in ex. ; more sol. in dilute acids. White, almost insol. in water. White. WITH REAGENTS. 53 (Excess neutralized by an Alkali). D. Nitrate of Silver. Protronitrate of Mercury. Sulphuric Acid. Remarks. Yellow, sol. White, sol. Sol. of magnesia with in N0 5 , in NO 5 . am. gives a white sol. in am. cryst. ppt.* Brown, Grey, Characteristic pro- reduced reduced perty is power of silver. mercury. reducing salts of silver and mercury. White, sol. Seleniates mixed in the solid state with in NO 5 . mur. am. and heated, give selenium. Seleniates boiled with HC1. evolve Cl.j and sulphuretted hydrogen then causes a ppt. White, sol. YeUowish Produces a jelly with hydrochloric or in dilute white, sol. sulphuric acid, or leaves, on evapor- N0 5 . in dilute ating the mixture to dryness, a resi- N0 3 . due insol. in water. White, in a White, sol. in NO 5 . A salt of barytes is a characteristic test. very con- ctd. soln. White, sol. Grey, Sulphurous acid Decomposes iodic in NO 5 , reduced is evolved. acid, giving free sol. in am. mercury. iodine, ascertained by starch. White, sol. White, sol. The sol. tartrates in am. in NO 5 . mixed with bisul- phate of pot. cause a ppt. of bitartrate of pot., sol. in HC1. and alkalies. * In a salt soluble in water, phosphoric acid is easily recognized by the yellow precipitate by nitrate of silver in neutral solutions ; if the substance is insoluble in water, it is better to fuse it with an alkaline carbonate, when a phosphate of the alkali is formed, which together with the excess of carbonate, dissolves when the mass is treated with water. When made neutral by an acid, this solution may be tested with nitrate of silver. (See also Schulze's method of estimating phosphoric acid in quantitative analysis.) E 3 54 CHAPTER III. QUALITATIVE ANALYSIS. THE performance of a satisfactory qualitative analysis of a complex substance by the aid of the preceding tables, would be a matter of some difficulty in their present arrangement. In- stead of applying each reagent successively, the method adopted in examining the composition of single or mixed bodies, con- sists in determining the presence or absence of certain classes of substances by the application of a single test for each class ; these classes can be divided and subdivided by other tests, and the subdivisions examined for their individual members. A particular class of metals is, for instance, pre- cipitated by sulphuretted hydrogen ; by the behaviour of this class with hydrosulphate of ammonia, it admits of a farther division, some of the metals precipitated by sulphuretted hydrogen being soluble, the remainder insoluble, in that liquid ; thus affording two divisions, each of which, if both are found to exist, is then examined for its individual members. The advantage of this method of procedure arises from its being in general easier to discover classes than particular bodies ; and the absence of a class being demonstrated, it is of course unnecessary to examine by other tests for any particular member. The principal feature in this system is, therefore, that of exclusion ; on the application of a test, hydrosulphate of ammonia, for example, when we perceive none of the appearances which that test should produce with certain metals, we reject these as absent from the combina- tion. We thus have a negative result ; but to prove the absence of particular substances is in itself just as important in qualitative analysis as to demonstrate their presence. A substance may occasionally be presented for examina- QUALITATIVE ANALYSIS. 55 tion, whose composition can be very well ascertained by the application of two or three reagents ; but it is advisable, and will mostly, in the end, be found more satisfactory, especially when we are not assured of the purity of the substance under examination, to follow some general rule, or particular system, rather than to try the effect of a reagent at random, from some vague idea we had before entertained of the composition of the substance. Such systems are developed in the following tables, where, it will be seen, classes are formed, divided, subdivided, and, if necessary, divided fur- ther, characters being described by which the particular members of each subdivision are distinguished from one another. An examination of the physical characters of a substance must always precede its analysis. It is of importance to know its density, volatility, combustibility, and solubility. Before proceeding to determine the inorganic constituents of a substance, it is also necessary to ascertain whether or not any organic matter is present, this materially affecting many metals in their behaviour with reagents, as may be seen by the preceding tables. Almost all fixed organic bodies possess the property of becoming blackened or car- bonized by heating out of contact with the air. To discover organic matter, therefore, a small portion of the substance should be heated in a glass tube, free from lead, when, in general, blackening will occur, and an empyreumatic odour be developed, if organic matter is present. The tube should be held slightly inclined, and be heated by a spirit lamp. There are, however, a few inorganic bodies which become black on heating ; and, on the contrary, some organic bodies which do not blacken. But there is another character of organic bodies, which, taken with the blackening by heat, may be considered decisive. This is the property they all possess of deflagrating when thrown into fused nitre ; a pro- perty possessed, too, by a few inorganic bodies, but by none which also blacken on heating. The two tests conjoined are therefore conclusive. E 4 56 QUALITATIVE ANALYSIS. By the same operation of heating in a tube to discover organic matter, we ascertain (when heated gradually) whether the substance contains water. If so, this condenses on the sides of the glass tube before the heat is sufficient to carbonize the substance. If the condensed water is alkaline, this shows the presence of ammonia. If organic matter is present, and the object of the analysis is merely to discover the fixed bases, and not the acids of the substance, it should be destroyed by combustion in the open air, either in a platinum or a porcelain * crucible, uncovered, heated very strongly by a spirit lamp with circular wick, or by the inflammable mixture of gas and air described page 10. The proper solvent of the substance must next be sought. Pure water is first to be tried ; if neither wholly nor partially soluble in this, hot hydrochloric or nitric acid or aqua regia must be tried successively, until it dissolves entirely, or leaving only a slight residue, which must be collected and examined by other means. Acids frequently leave insoluble residues, which have been formed by their action on the substance to be dissolved, and did not therefore pre-exist in the substance under examination : these may be easily distin- guished in appearance from the original substance; thus silica remains as a jelly when some silicates are treated with acids ; sulphur when sulphurets, and selenium when sele- niurets are acted on by nitric acid. In the last two cases, sulphur and selenium are slowly dissolved by the further action of the acid, sulphuric and selenic acids being formed. The course of qualitative analysis to be followed differs somewhat for substances soluble and insoluble in water ; but the following tables have been so arranged as to include the necessary modifications for both. When insoluble, both in water and acids (as is the case with most silicates), the sub- stance must be rendered soluble in acids by fusion with an * When the substance can contain metals reducible by charcoal, and which would alloy with platinum at a red heat, a porcelain crucible must be used ; in other cases platinum is preferable. : QUALITATIVE ANALYSIS. 57 alkaline carbonate. This operation consists, 1. in reducing the substance to a state of extremely minute division in an agate mortar ; 2. mixing with twice or thrice its weight of pure carbonate of soda or potash ; and 3. fusing in a covered platinum crucible. The resulting mass is then treated with pure hydrochloric or nitric acid, and is in almost every case dissolved, leaving, in the case of silicates, a gelatinous resi- due of silicic acid. (See the Analysis of Silicates for de- tails of this operation.) When fusion with carbonate of soda is inadequate, caustic potash (potassa fusa) is to be substituted, and the fusion performed with care in a silver instead of a platinum crucible. The first section of this chapter contains directions for the qualitative analysis of a salt consisting only of a single acid and a single base, comprising all those which are of ordinary occurrence: with such, rather than with complex substances, the beginner will do well to exercise himself. Respecting the quantity of matter to be operated on, it will be found more convenient to work with small than with large amounts ; from ten to twenty grains will be sufficient in almost every case. A small quantity must always be re- served, in case of accidents, or to examine the action of any special test. LIST OF BASES TO BE SOUGHT FOB. Antimony, oxide of Bismuth, oxide of Cadmium, oxide of Copper, protoxide of Gold, oxide of Class !.-< Iron, peroxide of Lead, oxide of Mercury, protoxide of Mercury, suboxide of Silver, oxide of Tin, peroxide of Tin, protoxide of. 58 QUALITATIVE ANALYSIS. 'Alumina Chromium, oxide of Cobalt, oxide of Class 2.^ Iron, protoxide of Manganese, protoxide of Nickel, oxide of .Zinc, oxide of. fBarytes . . I Magnesia (_Strontian. "Ammonia Class 4 I Potash Soda. LIST OP ACIDS TO BE SOUGHT. Arsenic Hydrosulphuric Boracic lodic Carbonic Nitric Chloric Phosphoric Hydrochloric Sulphuric Hydrofluoric Sulphurous. Hydriodic SECTION I. QUALITATIVE ANALYSIS OF A SALT CONSISTING OF A SINGLE ACID AND A SINGLE BASE, CONTAINED IN THE PRECEDING LIST. Solution made in water, hydrochloric or nitric acid, aqua regia, or by fusion with carbonate of soda and solution in hydrochloric acid. I. EXAMINATION FOR BASE. According to the following table, in an examination for the base of a single salt, the first operations have for their object the assignment of the base to one of three classes, named A, B, and C; A being class 1., B class 2., and C classes 3. and 4., of the preceding list. In the table, it is directed EXAMINATION FOR BASE. 59 that (1.) sulphuretted hydrogen gas be passed through the solution of the substance made slightly acid. If a precipitate appears, the base is then among the list forming class 1., and may be determined by tests given under A page 61. If the base is not precipitated by sulphuretted hydrogen (2.), hydro- sulphate of ammonia is applied to the solution previously made alkaline. The production of a precipitate by this re- agent shows the base to be among the list forming class 2., and may be determined by tests described under B page 61. If neither sulphuretted hydrogen nor hydrosulphate of am- monia produces a precipitate (3.), the base is one contained in classes 3. and 4., and may be determined by operations described under C page 62. Frequent advantage will be taken of a tabular method of exhibiting at one view the entire course of an analysis by means of brackets, under the opposite arms of which will be mentioned the two results which can be afforded by the application of a reagent, namely, the precipitate and the remaining solution. The precipitate is mentioned at the left hand, and the remaining solution at the right hand arm of the bracket. In the analysis of mixed compounds the pre- cipitate and the solution are generally separated by filtra- tion, and each examined apart by other tests, the results which can be produced by these being, as before, mentioned under the opposite arms of brackets. This plan may be readily conceived by reference to the table, page 66. 60 QUALITATIVE ANALYSIS. 4 <= .5 II I II II "3 3 P e - I, I I LI EXAMINATION FOR BASE. 61 1 -a i. Lead. 1 ll 2. Bismuth. || 3. Copper. 4. Silver. ^ S - 5. Subox. of 5 03 t| 2 - S . 3 8.5 o 'o -S Mia si S3 a) -< i! 1 1/3 ' s s H |iil 11 ill o> oT ^ . - ta^'g 2 . * J j M 8 1 II 05 .S QUALITATIVE ANALYSIS. 67 II. TREATMENT OF THE SOLUTION FILTERED FROM THE PRECIPITATE BY SULPHURETTED HYDROGEN. Render it slightly alkaline by ammonia, and then add a very slight excess of hydrosulphate of ammonia. The pre- cipitate may consist of alumina, oxide of chromium, sulphu- rets of nickel, cobalt, iron, manganese, and zinc, and if the substance for analysis be insoluble in water, magnesia, lime, barytes, and strontian united with phosphoric, arsenic, and boracic acids, or their metallic bases with fluorine. Filter ; wash the precipitate with water to which a little ammonia has been added, and digest it in pure dilute hydro- chloric acid, until the odour of sulphuretted hydrogen is no longer perceptible. Sulphurets of nickel and cobalt remain undissolved ; filter and test for these by microcosmic salt or borax before the blowpipe. (See Blowpipe Tables.) F 2 68 w QUALITATIVE ANALYSIS. "T3 lilil** f "Hir I i iPfiili ilill!!! L g F&iSJjta ^AH .a .4* g 2 1*1 1 11 j 4^3 &1 * ^ 111 l"3-l1l &l -^ 1^ l-l & -I 1 ! | W g ^ | I *. "ftll- 8 sla * .~s c- rJllj 8* " * s^^lllljlil bvllr ills^il p i rvi t-g Q O -g cS 'i S &-' ,3 cS 13 ^^ rt !5 %S 1*8 ?-B*&l 'O g.SJ S S a p h- I *M QJ I I *> -S ^ g o s ^ *S 3 2 ^3 *" 5 1 tl^is, QUALITATIVE ANALYSIS. 69 F 3 70 QUALITATIVE ANALYSIS. si: BJ : 5 * !S ' ' is c g lie *U? S.S Ill s 3 I I 1 I I 8 S !! ^-f * -sJ^ I 1 1 I s.i QUALITATIVE ANALYSIS. 71 SECTION III. QUALITATIVE ANALYSIS OF A SUBSTANCE WHICH MAY CONTAIN ALL INORGANIC BODIES WHOSE PROPERTIES ARE WELL ASCERTAINED. The solution effected either in water, acids, or by fusion with .an alkali, and subsequent solution in an acid. I. EXAMINATION FOR BASES AND SOME ACIDS. Acidify the solution, if neutral, by hydrochloric or nitric acid. ( See Table, p. 60., for cases in which hydrochloric acid must not be used.) Substances precipitated by hydrochloric or nitric acid aided by heat, insoluble in an excess : silicic acid ; tungstic acid ; tantalic acid ; titanic acid : by strong nitric acid only ; peroxide of tin. These are distinguished from each other by blowpipe tests. (See Tables on the Blowpipe.) If a precipitate occur, filter; transmit sulphuretted hy- drogen gas through the solution to saturation, set aside for twelve hours, then boil to expel excess of sulphuretted hy- drogen, and filter. F 4 72 QUALITATIVE ANALYSIS. * 11 QUALITATIVE ANALYSIS. 73 74 QUALITATIVE ANALYSIS. II. EXTENDED EXAMINATION FOR ACIDS. Acids to be sought besides the list, page 58. Bromic Hyposulphurous Phosphorous Chromic Hyposulphuric Selenic Hydrobromic (bromides) Manganic Selenious Hypophosphorous Nitrous Silicic. Hyper manganic Oxalic The acids already discovered in the determination of the bases are arsenic, chromic, manganic, and selenious or selenic. The evolution of chlorine, when the substance is digested in hydrochloric acid, may proceed from chloric, iodic, bromic, selenic, manganic, and hypermanganic acids. The first operation consists in mixing the substance, either in the dry state or in a strong solution with concen- trated sulphuric acid, and applying heat in a test tube. Observe whether any volatile acid is evolved by holding a rod moistened with ammonia over the tube. S. I < - f Leaves a clear solution ; soluble sul- j_ phites dissolve sulphur. ("Leaves a milky solution from preci- tpitation of sulphur: hyposulphites dissolve moist chloride of silver. {Leaves sulphuric acid ; if pure hydro- chloric acid be used instead of sul- phuric, the liberated sulphuric acid may be observed. Oxygen gas is evolved from bromates by the action of sulphuric acid. Bromates part with oxygen when heated to redness. From bromides, chlorine water liberates bro- mine, giving a yellow solution, from which the bromine is removed by agitation with ether. Nitric oxide is evolved and forms ruddy fumes with the air. Nitric acid remains. Peroxide of chlorine liberated ; the liquid bleaches vegetable colours. Fumes of iodine. Apply the test of chlorine water and starch to the solution of the substance. 11 . Sulphurous. 'o fa ^ 111 Hyposulphurous. 53 .r- o _< 1*1 I|8 ill C8 Hyposulphuric. 1 Bromates and bro- J mides give the [_ yellowish red vapour of bro- mine. I { lydriodic iodides). Nitrous. Chloric. EXAMINATION FOR ACIDS. 75 || J's! Oxalic. j I Carbonic. Effervescence in cold with dilute acid. No effervescence in cold. Chloride of calcium gives a white precipitate, which effervesces with hydrochloric acid after, but not before ignition. Note If carbonic acid, a jar of the gas collected over water, agitated with a little lime-water, redissolves the precipitate it produces at first. If it be mixed with sulphurous acid gas, the last may be separated by its greater solubility in water, or by agitation with peroxide of lead. The other acids volatilized when the substance is heated with sulphuric acid, are hydrosulphuric, hydrofluoric, hydrochloric, and nitric. (Examine for these according to the Tables (A.), (B.), and (C.), pp. 64, 65.) Acids not volatilized when their combinations are heated with sulphuric acid, are arsenic, iodic, phosphoric, boracic, sulphuric, silicic, selenious, selenic, hypophosphorous, and phosphorous acids. (For the first five apply the special tests mentioned in Table (B.), p. 64.) r Acidulate the solution with hydrochloric acid, evaporate to Silicic. -! dryness, arid redissolve in dilute acid. Silicic acid remains undissolved. Confirm by blowpipe with carbonate of soda. Detected by the blowpipe. These acids are distin- guished from each other by sulphuretted hydrogen, which precipitates sulphuret of selenium from sele- nious, but not from selenic acid. By boiling with hydrochloric acid, selenic is reduced to selenious acid, and is then precipitated by sulph. hydrogen. These acids reduce salts of mercury and gold to the metallic state in acid solutions. Most phosphites and hypophosphites evolve phosphuretted hydrogen when heated to redness, the gas sometimes being sponta- neously inflammable. I Selenic and selenious acids. Phosphorous and hypo- phospho- rous acids. SECTION IV. EXAMPLES OP THE QUALITATIVE ANALYSIS OF METALLIC ALLOYS. The two following tables are illustrations of the quali- tative analysis of Berlin silver and Newton's fusible metal (conducted according to the Tables, pp. 66. 68.). In the first example, positive indications are given of the existence of copper, nickel, and zinc, in the alloy ; but the tests for all other metals afford negative results. In the second example, nothing is detected except tin, bismuth, and lead. 76 QUALITATIVE ANALYSIS. !. _r C .3 ph. of G ^ *-C ^ o **" *Q c lllllgllsli, EXAMPLES. 77 1 r "a E 0) 3 o r2 3 's a ^ A "8 I X y= o ,fi s T| I | - 1 . S nj 8| 1 .2 ^S '^ ' rt 1 "o 2 - ~ w * SB 3 11 e .S !l ^"T3 ts i .S g c N &i rt O . *^-3 ^3 5 >> a> 2 2 w ^^ > -g a, -^ 8| ^ - ^ 2 s ^^ S IFH 5^1 "-g^ 1 1 .ET.-< T3 ,1 1 o **- . w HH Z d O M *C *& m**( ^fc D OQ *H f> c3 w 2 o ^ ^ ^ PR CO *Jzj * Qi J- If J a 0) g,3 T3 -&'S :3 5 S ^ f ?| ^ -s SJ Ss S^- 1| tn J3 _, 45 a, R c I i ^P ?s s ^^ -5 a ^ ' -d 2 J2? il - 8 o ! 1 1 i ^ 1 * Mllill' 'ss-jasss i . HI -G 1 ^ B HH M & is 1 OS 3 11 S $* *SoT B| 11 il~ 8 8 pat 1 ! s fi^-5 ^ si e-|g So o g B 0) fl ^ 8 l!sli.n =3 8 1 S-S^ g 11 l C8 ^-/ o n3 * "S 2 g 1 i iltlitf! 3 -rt *3 *J J; 5J f * b 1 te 1 J |H c^j o *C ^^ * IM H 1 1 1 axj S g,S^' < .r W) - 5 -3 -B i* jjiwJ -a 1 '31 * a 111 'O a> '4l Wa a, ^ * c o > 1 I Jl'Kjjll OF- o.-g^S oj^ ^'S^* fl ^s< ill ^ *<* s'S^ 5.9 es ao Cj c/) S >i 9 &l O CONTAIN hydrochloric aci phate of ammoni <& 1 jj fl ution contains al dilute sulphuric _ A^, p^i i C J- p-j w x H fi W 5*3 g | f'B >-, 1. H 8^ r 1 |W 3 S ,-c . : i 2 i CO tJO g 03 ] *5 * S Q *O -o * "P< o" 5 1 g c/: ^3 'I 14 + ^ ** c rt &. <1 C o 1 *1 *5 O P3 ll 'a IfS 60 S w II" 1 w o eS 8 c 4-H -^ Qi PH 0, '| a rH 1-2 1 2 1| >-> 11 w ^O *T3 c e 'o. w *J >> O C8 -5 .f C = 9 *o 4 1 1 8 M w LJ I *" 3 11 wl! W X "^ ill H 'f H 1 1 o o HH =0 .s g % 3 O s I 1-3 0) li n * III! J* 1 * G 2 84 QUALITATIVE ANALYSIS. i EXAMPLES OF THE QUALITATIVE ANALYSIS OF SILICATES. 1. STILBITE. Dissolve the powdered mineral in hydrochloric acid; evaporate to dryness, and treat the residue with dilute hydrochloric acid. Insoluble residue is silica. (Red prussiate of potash gives no blue precipitate.) Add excess of caustic ammonia. (As free hydrochloric acid is present, the addition of muriate of ammonia is unnecessary.) ji ^^^^^^^^^^^^^^ t > > A white precipitate falls, which is Add oxalate of ammonia to the entirely soluble in potash ; hence solution, this is alumina. JL. ( \ Oxalate of lime is pre- A drop of the solution evaporated to dryness cipitated. on platinum foil leaves no fixed residue after ignition. The constituents of stilbite are therefore silica, alumina, and lime. 2. HORNBLENDE. After fusion with carbonate of soda, dissolve in hydrochloric acid, and separate silica by evaporation and re-solution of the residue. After a little nitric acid has been added to peroxidize iron, add, first muriate of am- monia (unless the solution is very acid), and then caustic ammonia. _A^ t \ The precipitate is filtered and Add oxalate of ammonia to the dissolved in hydrochloric acid. filtered solution. To the acid solution add excess of potash, and boil. t \ ( \ Peroxide Muriate of ammo- Oxalate of Phosphate of soda of iron nia added to the lime is gives a white remains alkaline solu- precipi- precipitate of undis- tion produces a tated. phosphate of solved. precipitate of magnesia and alumina. ammonia. Hornblende therefore contains silica, alumina, oxide of iron, lime, and magnesia. 3. LEPIDOLITE. (1.) Dissolve in hydrochloric acid, after fusion with carbonate of soda. Silica is separated in the usual manner. Add muriate of ammonia and caustic ammonia to the solution. ji A white precipitate, almost wholly Oxalate of ammonia produces no soluble in potash j hence alu- precipitate in the filtered solu- mina. (Sometimes a trace of tion. Hydrosulphate of am- manganese remains.) monia gives a flesh -colour pre- cipitate of sulphuret of man- ganese. MINERAL WATERS. 85 (2.) For alkalies by fusion with carbonate of barytes. To the solution in hydrochloric acid separated from silica (page 83.) add carbonate of ammonia and hydrosulphate of ammonia. Precipitate contains carbonate of ba- rytes, carbonate of lime, alumina, and sulphuret of manganese: fil- ter. Add a solution of oxalate of ammonia to the filtered solution. Traces of ba- The solution leaves a residue on rytes and evaporation and ignition. Tar- lime are taric acid and chloride of plati- precipi- num indicate the presence of tated. potash. Phosphate of soda and ammonia gives a white precipi- tate, on long standing, of phos- phate of lithia and soda. Lithia is also discoverable by the blowpipe. Lepidolite therefore contains silica, alumina, manganese, potash, and lithia. SECTION VI. QUALITATIVE ANALYSIS OF MINERAL WATERS.* The first point to be determined in the examination of a mineral water is, to which of the five great classes of waters does the one in question belong ? 1. Carbonated, containing free carbonic acid. 2. Sulphureous, containing sulphuretted hydrogen. 3. Chalybeate, containing carbonate of iron. 4. Alkaline, containing carbonate of soda. 5. Saline, containing much salts. 1. If the water reddens blue litmus paper before boiling, but not afterwards, and the blue colour of the reddened paper is restored on warming, it is carbonated. 2. If it possesses a nauseous odour, and gives a black pre- cipitate with acetate of lead, it is sulphureous. 3. If after the addition of a few drops of hydrochloric acid it gives a blue precipitate with yellow or red prussiate of potash, the water is a chalybeate. * If practicable, the water should always be collected near the mouth of the spring, and secured in well closed vessels. G 3 86 QUALITATIVE ANALYSIS. 4. If it restores the blue colour to reddened litmus paper, after boiling, it is alkaline. 5. If it possesses neither of the above properties in a marked degree, and leaves a large residue on evaporation, it is a saline water. The substances which commonly enter into the compo- sition of a mineral water are, Acids. Sulphuric, carbonic, phosphoric, silicic, hydro- chloric (chlorides). Bases. Potash, soda, lime, oxide of iron, magnesia, alu- mina. Besides these, other constituents are sometimes found, but they are comparatively of much rarer occurrence : these are, Acids. Nitric, sulphurous, boracic, and some organic acids produced by the decomposition of vegetable matter (crenic, apocrenic, and puteanic). Bases. Lithia, strontian, oxide of manganese, oxide of zinc, oxide of copper. To determine with accuracy on all the ingredients which may exist, the water should be concentrated by evaporation nearly to dryness, although a cursory examination, sufficient for many purposes, may be performed on the water in its ordinary state. On boiling, most waters evolve a gas, which in carbonated, alkaline, and chalybeate waters is chiefly car- bonic acid ; and in sulphureous, sulphuretted hydrogen : all waters which are exposed to the atmosphere at ordinary tem- peratures contain also oxygen and nitrogen, though not in the proportions in which these gases exist in the atmosphere. In an extended examination the liberated gas must be col- lected in the mercurial trough.* While the gas is evolved by ebullition, a precipitate is formed in chalybeate, carbonated, and most saline waters, which may contain silica, carbonates of lime, magnesia, and strontian, peroxide of iron, fluoride of calcium, and phosphate of alumina. As this precipitate is to be collected and ana- lyzed, particular attention must be paid to the perfect filtra- tion of the water before evaporation, that no solid matter * For means of collecting the gas, see the " Quantitative Analysis of Mineral V/aters." MINERAL WATERS. 87 diffused through the water may contaminate the precipitate actually deposited from a state of solution. When the water is evaporated to dryness, and the residue treated with distilled water, the precipitate formed during ebullition remains undissolved, while all the soluble matters dissolve, forming a solution which may contain potassium, sodium, calcium, and magnesium, combined with chlorine, iodine, and bromine, or as nitrates or sulphates of the oxides of these metals. By evaporating to dryness, and treating the residue with water, three classes of constituents therefore become sepa- rated, each of which is to be examined independently. This is the first step in the analysis : 1st, the gases held in solu- tion in the water in its natural state ; 2d, the bodies held in solution by the gas, and rendered insoluble by evapora- tion ; and 3d, bodies which are permanently soluble, and therefore redissolved when the dry residue is treated with pure water. The qualitative analysis of each class is con- ducted in the following manner : I. FOR THE GASES EVOLVED. Into the gas contained in a jar over mercury introduce a little solu- tion of potash free from carbonate. Absorbed carbonic acid and sul- Not absorbed by potash oxygen phuretted hydrogen. Test the and nitrogen : introduce a stick original water with lime-water of phosphorus. for carbonic acid, and with f ^ . acetate of lead for sulphuretted Oxygen is ab- Nitrogen re- hydrogen, (pp. 89, 90.) sorbed. mains. II. FOR THE BODIES RENDERED INSOLUBLE BY EVAPORATION. Treat the matter with pure hydrochloric acid, and eva- porate to dryness. Reserve a small portion of the residue to test for fluorine, by heating with sulphuric acid in a glass tube, and treat the remainder with dilute hydrochloric acid. The insoluble residue is silica. G 4 88 QUALITATIVE ANALYSIS. S g g w 52; M O c^ w w H I w .2 J3 O tjj r s| II l|! a .s s l d | S* 11 II ' -2 2 u O 2t si i il il*H -1 3 I* 8 4J 03 .v. a> f^I a. 3 ^ cs to a> 'g|5 ^ 8-g,'^ H Ill <3 .2 o a ^ 's S 9 .S 9 ^ fl) *> ^ B*ia M il 9 a f. cc 03 ^ - ^ o -II *H S 8 ra c -P 1|*a O C ,-j .222 5 " g-3 1^1 "1 bl ^ ^ ^ cf 'S S S'^ jl ll > The precipitate The filtered solution contains lime, barytes, and alka- is magnesia lies : add carbonate of ammonia. and lime ^ _____________ -^ (sulphate of Precipitate is The solution contains ammoniacal salts, barytes). carbonate potash, soda (lithia). Evaporate, of lime and and calcine to expel ammoniacal carbonate salts ; the residue consists of alka- of barytes. line chlorides. Dissolve in water, and examine by chloride of plati- num, tartaric acid, and blowpipe. To discover lithia, a large quantity of the water should be evaporated to dryness, the residue dissolved in a small quantity of water, and filtered to separate the insoluble portion. To the solution, phosphate and carbonate of soda are added, the mixture is evaporated to dryness, and the residue treated with water, when, if lithia is present, this alkali remains undissolved, in the state of phosphate of lithia and soda. IV. EXAMINATION FOR ACIDS. First Portion. Carbonic acid and sulphuretted hydrogen. If the water, fresh from its source, feebly reddens blue litmus paper, and the colour of the paper is restored on gently heating it, this indicates the presence of carbonic acid. Lime-water then occasions a pre- cipitation of carbonate of lime. Carbonic acid may exist in three different states in mineral waters. 1st, as the free uncombined acid dissolved in the 90 QUALITATIVE ANALYSIS. water; 2d, holding in solution earthy carbonates; and 3d, combined with potash or soda as bicarbonates. These different states may be distinguished by the following cha- racters: 1st. If the precipitate of carbonate of lime pro- duced in the water by the addition of a small quantity of lime-water is redissolved by an excess of the mineral water, it contains the uncombined acid. 2d. If, on the contrary, the precipitated carbonate of lime does not dissolve in any excess of the water, and during ebullition a precipitate is formed which effervesces with acids, the carbonic acid is combined with earthy carbonates. 3d. If no precipitate is formed while the gas is evolved by ebullition, and the solution becomes strongly alkaline, effervescing with acids, alkaline bicarbonates existed in the original water. Sulphuretted hydrogen is detected by the black precipitate produced by salts of lead or copper, and by the nauseous odour possessed by the water. If after boiling for a considerable period, a black precipitate is still produced by the metallic solutions, alkaline sulphurets are contained in the water. Second Portion. For sulphuric and hydrochloric acids. Hydrochloric acid is shown by the white curdy preci- pitate produced by nitrate of silver, insoluble in nitric acid, but soluble in ammonia. To the solution filtered from chloride of silver add a solution of chloride of barium and a little nitric acid ; a white precipitate indicates sulphuric acid. Third Portion. Iodine and bromine. Evaporate a large quantity of the water to a very small bulk, and divide into two portions. With one mix a little starch, and allow chlorine gas to fall on the surface of the liquid from a bottle of chlorine water. If a trace of iodine exists, it will be immediately recognised by the blue colour produced. To discover bromine, add chlorine water to the other por- tion : if this element is present, a brown colour is developed, which is removed by agitation with ether. On standing for some time, the ether rises to the top, carrying with it all the bromine : add potash to the ethereal solution, evaporate to dry- ness, redissolve in water, and apply other tests for bromine. MINERAL WATERS. 91 Fourth Portion. For nitric and boracic acids. Eva- porate to dryness. If nitric acid is present, the dry salt deflagrates when heated strongly and touched with a glowing match. When the dry residue is moistened with alcohol and a little sulphuric acid, if the alcohol burns with a green flame, boracic acid is present. Fifth Portion. For sulphurous acid. This acid is recog- nised by the action of iodic acid, (page 53.) Strong sulphuric acid causes an effervescence of sulphurous acid gas from a concentrated solution. Sixth Portion. For phosphoric acid. Sulphate of mag- nesia mixed with a little free ammonia causes a white crystalline precipitate of phosphate of magnesia and ammonia. Seventh Portion. For organic acids (crenic, apocrenic, and puteanic). The dry residue of evaporation becomes carbonized when heated to redness, if it contains organic bodies. To the preceding extended examination I subjoin a simpler and more cursory one, which will, in many cases, afford suf- ficient information on the nature of the water. Different quantities of water are to be used for each test. 1. Oxalate of ammonia or binoxalate of potash produces a white precipitate of oxalate of lime. 2. Subphosphate of ammonia, a white precipitate of phos- phate of magnesia and ammonia. If oxalate of ammonia and subphosphate of ammonia give precipitates before but not after ebullition, the lime and magnesia exist entirely as carbonates. 3. Yellow or red prussiate of potash gives a blue pre- cipitate if iron is present. If the water is carbonated, and has been boiled, a few drops of hydrochloric acid should be added to redissolve the precipitate formed on ebullition. 4. Tincture of nut-galls, with a few drops of lime-water, gives a blackish brown colour, when a very minute quantity of iron exists. If this tincture and prussiate of potash give 92 QUALITATIVE ANALYSIS. indications of iron after boiling for half an hour, when no acid has been added, the iron does not exist solely in the state of carbonate. For carbonic acid apply the tests of litmus and of lime- water ; for sulphuretted hydrogen, of acetate of lead ; for sulphuric acid, of chloride of barium, and for hydrochloric acid, of nitrate of silver. These directions for the qualitative analysis of mineral waters do not, however, indicate the actual salts which are present as they exist, but only their respective acids and bases. The arrangement of these in the water cannot be determined by any process of qualitative analysis, except by the application of a rule, which is not without exceptions, that the strongest acids are in combination with the strongest bases. But when the quantitative analysis of the water has been performed, the arrangement of the acids and bases can generally be ascertained with tolerable accuracy, from the difference which exists in the combining numbers of bodies. Thus, the quantities of sulphuric acid required for neutra- lization by the two bases, magnesia and potash, differ con- siderably ; so that an amount of sulphuric acid having been found which is equivalent to the magnesia, and not to the potash, the former, and not the latter, is concluded to be in the state of sulphate. Another acid will probably be found whose amount will correspond to that of the potash, with which therefore the alkali was probably united. SECTION VII. ON THE USE OP THE BLOWPIPE IN QUALITATIVE ANALYSIS. From its ready application and manageability, from its portability, convenience, and cheapness, and from the highly characteristic effects produced by it with chemical sub- stances, the blowpipe is invaluable to the chemist as a sub- ordinate or a confirmatory agent to the results previously USE OF THE BLOWPIPE. 93 Fig. 10. obtained or afterwards to be sought in humid analysis. By its means we can raise the temperature of a small quantity of a substance almost in- stantaneously to a white heat ; when this is done in different ways, as in a closed tube, in an open tube, alone or mixed with other bodies, or in different parts of the flame, we arrive at positive conclusions respecting the nature of the bodies operated on. This little instrument is simply a tube open at both extremities, the aperture in one being con- tracted so as to become capillary, through which a small stream of air is forced upon a flame. It would be out of place here to enter into details respecting the construction of the various kinds of blowpipes. A very convenient one is Voigt's, which is repre- sented in the annexed figure. It consists of a brass tube, with an ivory mouth-piece a s at one extremity, and at the other a circular box b, from which issues the small tube c, moveable round the centre d, by which any degree of obliquity may be given to the flame : e is a brass or platinum jet, which fits on the tube c. The only inconvenience attending this form is, that the tube c becomes loose by long usage. Griffin's (Jig. 11.), which is of the same con- struction as Black's, is a cheaper and equally efficient instrument. It is made of japanned tin, with a brass nozzle. As the current of air which is supplied ought to be con- tinuous, its production requires some attention and address. The air is not blown directly from the lungs, but is forced from the mouth by means of the cheeks. The difficulty con- sists in inspiring and expiring through the nose, while a con- tinued stream escapes from the mouth. This may be attained by attention to the following directions : Inflate the mouth fully, and then, with the lips firmly closed, and the back of fig. I \ 94 QUALITATIVE ANALYSIS. the mouth closed by the tongue, breathe freely through the nostrils. While the respiration proceeds, and the mouth is inflated, allow a little air to escape through a very minute opening between the lips, renewing the supply in the mouth by occasionally admitting air from the lungs without in- terfering with the process of respiration through the nose. A little practice renders the operation easy. The effect of this current of air through a flame is to invert it, or rather to. cause a double combustion : the in- terior, which before could not burn for want of air, now burns by means of the air forced through the blowpipe ; the external atmosphere supplying air for the combustion of the exterior of the flame as before. This will be more apparent by reference to the annexed figures. Fig. 12. represents a Fi s . 12. Fig. is. section of the ordinary flame of a candle : the interior cone (a Z) is composed of combustible va- pours or gases formed from the tallow, prevented from burning by having no access to the air, combustion taking place only on the exterior, b, where the vapour is in contact with the atmosphere. Let air be admitted into the interior, as is done by the blowpipe, and these gases and vapours are burned, as they are on the exterior; the flame has then the appearance of Jig. 13., a representing the interior portion of the flame where com- bustion takes place. Between the points b and c inflam- mable gases still unburned exist ; now this portion of the flame possesses properties very different to those possessed by the portions either beyond or within it. If oxide of copper be heated in this portion, oxygen is abstracted from the oxide by the combustible gases (which at this high tem- perature exhibit an intense affinity for oxygen), and metallic copper is produced : on the other hand, if metallic copper be heated beyond the point c, it is again oxidated, the oxygen being now supplied from the external atmosphere. From possessing such different properties, these portions of the USE OF THE BLOWPIPE. 95 flame are called the oxidating and the deoxidating flames. The latter is also called the reducing flame. The chief oxi- dating effect is, however, produced beyond the flame, and this action is greater the farther the substance is from it, provided the temperature be maintained sufficiently high. The apparatus to be used with the blowpipe can be but briefly described here. For a flame, where coal gas is avail- able, that is to be preferred : the aperture of the burner is made of an oblong shape, instead of round, the current of air being blown lengthways. A candle having a large wick may also be employed; but it is inconvenient from the melting of one side by the bent flame and by the radiant heat of the ignited substance under examination. The oil lamp recommended by Berzelius consists of a tin cylinder about one inch in diameter, and four inches long, having a flat wick about three quarters of an inch broad. It is supported at one end by a retort-stand. Various supports are used to hold the substance while exposed to the blowpipe flame, depending on the nature of the operation to which the matter is submitted. When heated alone, a pair of tongs or forceps, having small platinum points, a piece of platinum foil held by forceps, a platinum capsule, or a piece of dense charcoal, may be used. When heated with fluxes, as borax or microcosmic salt, a piece of small platinum wire is employed, bent twice at right angles at one extremity to retain the fused globule : a piece of charcoal or a bone-earth cupel is also employed for this purpose. Glass tubes, open or closed at one end, are used when it is sus- pected something volatile may be given off. None but hard German glass should be used for blowpipe experiments. The principal operations to be performed are, 1st. To heat the substance in a glass tube sealed at one end ; 2d, in a tube open at both ends ; 3d, alone ; and, 4th, with fluxes or other reagents. The chief objects of the first operation are, to ascertain whether a sublimate occurs, or whether the substance contains 96 QUALITATIVE ANALYSIS. any volatile matters; whether decrepitation takes place; whether the volatile products possess an acid or an alkaline reaction or are neutral ; whether the substance becomes black on heating, showing the presence of organic matter, and the degree of fusibility of the substance. The tubes employed should be about two or three inches in length, and from one eighth to one fourth of an inch in diameter. The body under examination may occasionally require to be mixed with some reagent, as bisulphate of potash or charcoal. The second operation is done with a view of ascertain- ing what effect a current of atmospheric air has on the sub- stance at a high temperature: to know if a sublimate is produced ; and if so, whether similar to that in the closed tube ; and whether any peculiar odour is developed. By inclining the tube more or less, we have the means of regu- lating the current of air which passes : if heated when held nearly horizontal, very little air passes through, but the air increases in proportion as the tube is held more and more vertical. In the third operation many objects are to be held in view. The degree of fusibility of the substance is observed ; for which purpose a comparative scale has been established, consisting of the following minerals, in the order of their fusibility, beginning with the least fusible: sulphuret of antimony, natrolite, garnet, hornblende, felspar, and diallage. The changes exhibited by the substance must be noticed, as change of colour, intumescence, or deflagration. It must be ascertained whether the substance appears the same in the reducing as in the oxidating flame, and whether any volatile matters are evolved. An important object is to ascertain whether the substance produces any change on the colour of the blowpipe flame. The change, if any, is best seen when the matter just touches the flame at the top, about the middle of its length. For this operation the substance is usually held in a pair of forceps having small platinum points. The fourth operation is heating with reagents. Except in a few particular cases, the only reagents employed are borax, USE OF THE BLOWPIPE. 97 microcosmic salt, and carbonate of soda. The object of heating with the two first, is to see what colour glass results when the substance is dissolved in the flux, both in the in- terior or reducing, and in the exterior or oxidating flame. Fi s- 14 - r- 1 A platinum wire, C-J bent as in the an- nexed figure, is employed in this operation for borax ; but as microcosmic salt gives a very fluid bead when heated, a few additional bends must be made, when this salt is used, to prevent the melted drop from falling. Instead of the pla- tinum wire, charcoal may be used to support microcosmic salt. In fusing with carbonate of soda, the objects to be ascer- tained are, whether the substance gives a reduced metal, and whether it is soluble in the carbonate, giving a glass, or insoluble, giving a slag. A platinum wire is sometimes em- ployed, but charcoal is the proper support for carbonate of soda ; a dense piece should be selected, and a cavity made in it, from one-third to one-half of an inch in diameter, bored to about one-eighth of an inch in depth. The carbon- ate of soda, in fine powder, is kneaded into a lump with a little water, mixed with the substance for examination, and placed in the charcoal cavity. To ascertain whether a re- duced metal has been formed by heating, the bead, when cold, is removed from the charcoal, and powdered with a little water, in an agate mortar. The particles of charcoal and soda are washed away by affusion, while any metallic particles present, subside to the bottom, and may be further examined as to colour, malleability, &c. In the first three of the following tables are exhibited the most remarkable of the phenomena which metallic oxides present before the blowpipe ; then the characters of some acids (in combination) ; after which tabular views are given of the successive operations (with their subdivisions) which are to be performed, and appearances to be looked for in the examination of an unknown substance. When, by following the processes of the latter tables, a clue to the nature of the substance is obtained, the result should be confirmed by examining whether it possesses all the characters which belong to that substance, as contained in the former tables. H 98 QUALITATIVE ANALYSIS. I. Behaviour of ALKALIES Alone. Colour of Flame jroduced by moist soluble Salts. With Borax. Potash. Fusible. Faint reddish or violet. The red glass of borax and nick- el is changed to blue. Soda. Fusible. Intense greenish yellow (cha- racteristic). The red glass of borax and nickel is not changed to blue. Lithia. Fusible. Crimson (cha- racteristic). Barytes. Infusible. The hydrate and carbonate are fusible. Pale apple green. Colourless bead, opaque if satu- rated with ba- rytes. Strontian. Ditto. Intense crimson (characteristic). Ditto. Lime. Infusible, gives a strong light. Reddish purple. Ditto. Magnesia. No change. Ditto. Alumina. Ditto. Colourless glass, remains so when cold. Opaque with excess. Glucina anc Yttria. Ditto. Transparent glass; becomes milky by an intermittent flame. Zirconia. Infusible, gives a very powerfu light. Ditto. BLOWPIPE TABLES. 99 and EARTHS. With Microcosm ic Salt. With Carbonate of Soda on Charcoal. Moistened with Solution of Nitrate of Cobalt. A colourless bead., becomes opaque when cold, if sa- turated. Is absorbed by the charcoal. Brick or yellowish red while hot ; colourless when cold. Ditto. Not dissolved. Black or grey. Ditto. Not dissolved ; is ab- sorbed by the char- coal. Ditto. Ditto. No action. Flesh colour after strong heating (characteristic). Transparent glass. Undissolved. A fine azure (character- istic). As with borax. No action. Greyish black. Scarcely transparent. Ditto. H 2 100 QUALITATIVE ANALYSIS. II. Behaviour of Alone. With Borax. Oxidating Reducing Flame. Flame. Oxidating Flame. Reducing Flame. Antimony, oxide of. Sublimes ; gives a green- ish blue flame ; is re- duced on charcoal. ^ Yellow glass while hot ; colourless when cold. Grey from re- duced metal. Bismuth, oxide of. Fusible ; dark brown while hot, yellow when cold on wire. Reduced on charcoal. Colourless bead. Ditto. Cadmium, oxide of. On platinum wire, no change. Is reduced on charcoal and volati- lized, giving an orange red coating of oxide. Yellowish glass, colourless when cold ; becomes opaque with an intermitting flame. Cerium, oxide of. Protoxide becomes per- oxide. Red or orange when hot, pale when cold. Colourless ; milk white if saturated. Chromium, oxide of. No change. Red while hot, green when cold. Green. Cobalt, oxide of. Ditto. Deep blue (characteristic). Copper, oxide of. Bromide and chloride give a blue flame; most other salts give green. Green while hot. Brown red ; increased by the addition of tin. Fuses. Reduced. Iron, per- oxide of. Blackens anc No change. becomes magnetic. Red while hot; pale when cold. Bottle-green. Lead, oxide of. Gives a clear blue flame. Fuses to a fine orange glass; is reduced on charcoal. Yellow ; nearly colourless when cold. Manganese, oxide of. Infusible. Protoxide becomes brown. Amethyst, black if in excess ; as- sisted by nitre. Colourless if quickly cooled. BLOWPIPE TABLES. 101 METALLIC OXIDES. (.) With Microcosmic Salt. With Carbonate of Soda. Oxidating Flame. Reducing Flame. Faint yellow glass : colourless when cold. Grey from reduced metal. Colourless glass on wire. Re- duced on charcoal. The reduced metal is very fusible and brittle, disengages a white smoke of oxide when heated in the air. Brownish yellow while hot ; co- lourless when cold. Reduced. Reduced on charcoal. The reduced metal is brittle ; heated on charcoal it gives a yellow coating of oxide. A transparent glass, opaque when cold, if saturated. Does not melt on the wire. On charcoal it is reduced, and cadmium is volatilized. The charcoal acquires a red- dish brown coating (cha- racteristic). As with borax. Colourless. Not dissolved. Red while hot, green when cold, in both flames (characteristic). In oxidating flame, dark orange when hot, opaque when cold. In reducing flame, opaque, becoming green on cooling. Deep blue (characteristic). Reduced on charcoal, giving a grey magnetic powder. As with borax. Green when hot ; colourless and opaque when cold. Is reduced on charcoal, giving a red malleable metal. As with borax. Green when hot, almost colourless when cold. Reduced on charcoal, giving a magnetic powder. Yellowish when saturated ; opaque when cold. A transparent glass on wire, yellow and opaque when cold. Instantly reduced on charcoal. As with borax (characteristic). II 3 102 QUALITATIVE ANALYSIS. III. Behaviour of Alone. With Borax. Oxidating Flame. Reducing Flame. Molybdic acid. Fuses, giving a smoke in the open inclined tube and on platinum foil. Is reduced with difficulty on charcoal. Colourless on pla- tinum wire. Brownish red on charcoal. Nickel, oxide of. No change. Orange red j al- most colourless when cold. Grey from re- duced metal. Silver, oxide of. Instantly reduced. Partly reduced and partly dissolved. Milk white or opaline when cold. Grey. Tellurium, oxide of. Gives the flame a green colour. Fuses and sublimes ; is ea- sily reduced on charcoal. Colourless. Grey. Tantalum, oxide of. No change. Colourless ; becomes opaque when cold by an intermittent flame. Tin, oxide of. In oxidating flame, protoxide becomes peroxide. In re- ducing flame, is reduced by long heating. Colourless. Titanium, oxide of. No change. Colourless ; be- comes opaque on cooling. First yellow, then amethyst, be- coming darker on cooling. Tungstic acid. Blackens, does not fuse. Colourless. Orange, becom- ing darker as it cools. Tin produces an enamel. Uranium, oxide of. Ditto. A Yellow. Green ; black- ened by an in- termittingflame Vanadium, oxide of. On charcoal, fuses, and is par- tially reduced. Yellow. B rown when hot; ; green when cold Zinc, oxide of. Gives a strong whitish green flame. Slightly yellow while hot (by daylight) ; white when cold. Becomes green with solution of nitrate of cobalt. Transparent glass, becoming milky by an intermittent flame. Reduced ; the metal volatilizes BLOWPIPE TABLES. 103 METALLIC OXIDES, (b.) With Microcosmic Salt. With Carbonate of Soda. Oxidating Flame. Reducing Flame. Colourless on plati- num wire. Dark blue or black : green when cold. Is reduced on charcoal. Red ; colourless when cold. As in oxidating flame. Tin de- stroys the colour. Gives the reduced jnetal on charcoal. A magnetic powder. Yellowish by day- light ; red by candle-light. Opa- lescent in ex. Grey. Reduced. Colourless. Grey. On the wire a colourless glass, which becomes opaque on cooling. Colourless glass. Combination with effervescence ; no solution nor reduction. Colourless. Easily reduced on charcoal, giving a soft and very fusible metal. Colourless. Yellow while hot ; violet when cold. Dissolves with effervescence, giving a yellowish glass, which becomes greyish white and opaque on cooling. Is not reduced on charcoal. Very slightly yel- low. Very fine blue ; ox. of iron makes it blood-red ; this is prevented by tin. Dark yellow glass ; becomes opaque on cooling. Is reduced on charcoal. Yellowish while hot ; greenish when cold. Green. Insoluble. As with borax in both flames. Dissolves. As with borax in both flames. Insoluble. Is reduced on charcoal, which becomes covered with a subli- mate of oxide. ii 4 104 QUALITATIVE ANALYSIS. BLOWPIPE TESTS TO DISTINGUISH ACIDS, OR THE ELECTRO-NEGATIVE CONSTITUENTS. Sulphates. Mixed with the colourless and transparent glass of silica and soda, sulphates give, in the interior flame, a deep yellow or red colour, either immediately or on cooling. Heated with soda on charcoal, in the interior flame, the resulting mass blackens metallic silver when moistened. Nitrates. Those which are fusible deflagrate on charcoal. Infusible nitrates, heated to redness in a glass tube give off nitrous vapours. All nitrates give deep red vapours when mixed with bisulphate of potash and heated. Borates. Fused with carbonate of potash on charcoal, and moistened with sulphuric acid and alcohol, a green flame is produced. Mixed with a flux of one part of fluor spar and four and a half of bisulphate of potash, borates give a dark green flame for an instant, when held at the point of the blue flame on a platinum wire. Phosphates. To detect phosphoric acid, fuse the substance with boracic acid on charcoal ; when fused, introduce the extremity of a steel wire and heat strongly in the interior flame. Borate and phosphuret of iron are formed ; on breaking the fused mass (wrapped in paper) in a mortar, the phosphuret of iron is perceived, having the appear- ance of a metallic button, which affects the magnet and is brittle. Moistened with sulphuric acid, and held in the platinum tongs at the point of the blue flame, phosphates give a pale green colour to the outer flame. Silicic acid is characterized by dissolving easily in carbonate of soda with effervescence, giving a transparent and colour- less bead. It is slowly dissolved in borax, giving a clear glass. Seleniates and selenites give the characteristic odour of sele- nium, with soda on charcoal, in the interior flame. They behave like sulphates with the glass of silica and soda. Arseniates and arsenites. On charcoal, in the interior USE OF THE BLOWPIPE. 105 . flame, they give the characteristic alliaceous odour of arsenic. Mixed with black flux and heated in a glass tube closed at one end, a brilliant sublimate of metallic arsenic is formed, crystalline in its interior surface. It may be chased up the tube, becoming slowly oxidized. If the bottom of the tube be cut off, and the sublimed metal heated, it is immediately converted into arsenious acid, which condenses as a white crystalline sublimate. Sutyhurets. Heated in an open inclined tube, or on char- coal, they evolve the odour of sulphurous acid, which acid bleaches moist Brazil-wood paper. Sulphurets behave as sulphates with the glass of silica and soda. After fusion with soda on charcoal, the moistened globule blackens metallic silver. Seleniurets, in the exterior flame, evolve a strong odour re- sembling putrid horse-radish. Heated in the open tube, metallic selenium sometimes condenses. With the glass of silica and soda, seleniurets behave as seleniates and sulphates. Antimoniurets give a sublimate of oxide of antimony when roasted in the open tube. Tellurets give, when roasted in the inclined open tube, a sublimate of a white powder, which, when heated, fuses into colourless drops before it sublimes. Chlorides, added to a glass of microcosmic salt, saturated with oxide of copper (or held by a very small brass instead of a platinum wire), and suddenly heated, communicate a bright colour to the flame. Iodides, with the glass of microcosmic salt and oxide of copper, give a fine emerald green flame. Heated with bisulphate of potash in a glass tube closed at one end, iodine vapour is evolved. Bromides, with the glass of microcosmic salt and oxide of copper, give a greenish blue flame. Heated with bisul- phate of potash, bromine vapour is expelled. Fluorides. Those which contain water, give off hydrofluoric acid when heated in a tube closed at one end. All fluorides, when mixed with bisulphate of potash, or with 106 QUALITATIVE ANALYSIS. microcosmic salt previously fused, and heated at the lower extremity of an open glass tube, evolve hydrofluoric acid, known by its corrosive action on glass, and its property of bleaching Brazil-wood test paper. OPERATIONS TO BE PERFORMED IN THE EXAMINATION OF AN UNKNOWN SUBSTANCE. r - "They sublime without > Ammoniacal salts.' decomposition, unless they contain a fixed g (a) Volatile salts. -< Most salts of mer- acid. "Heated with dry car- bonate of soda, give cury metallic mercury. 03 g ..Many chlorides of fixed oxides. "Oxide of anti- ' Sublimes in shining nee- c3 mony. dles. i (&) Oxides and Arsenious acid. White octahedral crys- tals. O acids sub-^ "Becomes converted into O ,2 limed. Arsenic acid. oxygen gas and arse- _p nious acid. Oxide of tellurium. Sublimate not crystalline. ^Osmic acid. Recognised by its odour. C3 (c) Volatile acids j" Ascertained by their action on red and bl ^ and am- -{ ,. A * litmus paper, monia. d 1 Tellurium. ( Vd *f!f l ? at a bri S ht \ red heat. w C Gives a yellowish-red l a ^ , . coatinar to charcoal Cadmium. -{ , ? , when heated on it in 1 (d) Volatile me-^ (^ the open air. i tals. Mercury. (Condenses in liquid (^ drops. W f Interior surface of the O sublimate is crystal- Arsenic. S T /> 1- ] line. Odour of garlic evolved. (e) Water. Condenses on the sides of the tube. (/") Sulphur. A yellow sublimate. (#) Selenium. Characterized by its odour. bodies } Become black > leavin S a residue of charcoal. BLOWPIPE OPERATIONS. 107 a i glucina ; lime ; magnesia ; strontian ; tantalic acid ; thorina ; oxides of ura- nium ; y ttria ; zirconia. Examine for sulphur and selenium by the glass of silica and soda. 2. Colours of beads with borax (while hot). (c) Metals whose compounds are re- ducible with soda- on charcoal in the reducing flame. (d) Neither fused nor reduced with soda on charcoal sed f ith-l 1. Colourless In oxidating flame. Alumina Antimony Barytes Bismuth Cadmium Glucina Lime Magnesia Molybdic acid Silicic acid Strontian Silver Tantalic acid Tellurium Thorina Tin Titanic acid Tungstic acid Yttria Zinc Zirconia. Colourless Reddish brown Yellow In reducing flame. Alumina Barytes Cadmium Cerium Glucina Lime Magnesia Manganese Silica Strontian Tantalic acid Thorina Tin Yttria Zinc LZirconia. f Copper L Molybdic acid. C Tungstic acid 1 Titanic acid. 110 QUALITATIVE ANALYSIS. Red Violet Blue I v/u. 1 Iro [Ni< In oxidating flame. f Cerium I Chromium Iron Nickel. {Lead Uranium Vanadium. f Chromium Green -J (when cold) I Copper. In reducing flame. r Chromium Green J Iron I Uranium ^ Vanadium. Blue Cobalt. Grey Antimony Bismuth NickeL Silver > Tellurium. Manganese. Cobalt. 3. Colours of beads with microcosmic salt (while hot). In oxidating flame. In reducing flame. "Alumina Colourless Red Yellow Green Antimonious acid Barytes Cadmium Glucina Lead Lime Magnesia Molybdic acid Strontian Tantalic acid Tellurium Thorina Tin Titanic acid Tungstic acid Yttria Zinc ^Zirconia. f Cerium J Iron \ Nickel I Chromium. Colourless- Red Yellow {Bismuth Silver Vanadium. {Copper Molydic acid Uranium (Chromium when cold). Green "Alumina Barytes Cadmium Cerium Glucina Lime Magnesia Manganese Strontian Tantalic acid Thornia Tin Yttria Zinc LZirconia. f Antimonious acid with iron. Nickel Titanic acid with iron Tungstic acid with iron. f Titanic acid (violet 1 when cold). " Chromium Iron Molybdic acid Uranium Vanadium (when cold). BLOWPIPE OPERATIONS. Ill Violet Blue In oxidating flame. Manganese. Cobalt. Blue Brown Grey I In reducing flame. Cobalt Molybdic acid Tungstic acid. Copper. Antimony Bismuth Lead Silver - Tellurium. SECTION VIII. ANALYSIS OP GASES. I. PROPERTIES OF GASES. Oxygen - 1 Re- in flame a glow- Nitrous oxide J ing taper. Ammonia Hydrochloric acid Chlorine - - V Soluble in Sulphurous acid water. Carbonic acid Sulphuretted hydrogen Cyanogen Olefiant gas Phosphuretted hydrogen Arsenietted hydrogen Carbonic oxide Hydrogen Light carburetted hydrogen II. QUALITATIVE ANALYSIS OF A MIXED GAS ; ITS CON- STITUENTS BEING CONTAINED IN THE FOLLOWING LlST. _ Absorbed by potash. ^Combustible. 1. Carbonic acid 2. Hydrochloric acid 3. Chlorine 4. Sulphurous acid 5. Sulphuretted hydrogen 6. Cyanogen 7. Phosphuretted hydrogen 8. Arsenietted hydrogen 9. Ammonia (A.) Into the dry gas contained over mercury, introduce moistened sticks of potash ; withdraw the potash, and dis- solve it in water. The first six of the list are absorbed. 10. Oxygen 1 1 . Nitrous oxide 12. Nitric oxide 13. Light carburetted hydrogen 14. Olefiant gas 15. Carbonic oxide 16. Hydrogen 17. Nitrogen. 112 QUALITATIVE ANALYSIS. f Remove the more soluble gases from a portion 1. Carbonic acid. -I of the original gas by water, and apply the [_ test of lime-water to the residue. f Neutralize the solution of the potash by pure 2. Hydrochloric acid, nitric acid, and test with nitrate of silver. and -{ Chlorine and hydrochloric acid are distin- 3. Chlorine. guished from each other by the power of L the first to bleach vegetable colours. 4. Sulphurous acid. J Per xid " of }*f a S itated in the ori S inal S as L absorbs sulphurous acid. 5. Sulphuretted J~ The solution of the potash produces a black hydrogen. \ precipitate in salts of lead. f To the solution of the potash apply the test 6. Cyanogen. -j of peroxide and protoxide of iron (Scheele's L test), page 49. (B.) After the absorption of the potash, remove the re- maining gas to another jar, and introduce a solution of ni- trate of silver. Nos. 7, 8, and 9. are absorbed. Combustible ; producing white fumes of phos- phoric acid ; is sometimes spontaneously inflammable. Gives a volatile deposit of metallic arsenic when passed through a red-hot tube. Kown by its great solubility in water, and its 1. Phosphuretted hydrogen. 2. Arsenietted hydrogen. 3. Ammonia. alkaline reaction. (C.) The gases which remain after the action of potash, and of nitrate of silver, are, Nos. 10, 11, 12, 13, 14, 15, 16, 17. 1. Introduce a stick of phosphorus, and allow it to remain so long as any contraction occurs. Oxygen is absorbed.* , 2. If the gas supports combustion before, but not after, agitation with water, it contains nitrous oxide. (D.) Into a portion of the remaining gas contained in a tube through which the electric spark can be passed f, pass up an equal volume of pure oxygen. 1. The production of ruddy fumes shows the presence of nitric oxide. 2. Pass the electric spark ; if explosion takes place, Nos. 13, 14, 15, and 16. may be present. * Phosphorus does not act upon oxygen in the presence of olefiant gas, un- less the pressure be greatly reduced. Phosphorus is likewise without action on perfectly pure oxygen. f Such as Dr. Ure's eudiometer, of which a figure is given under the subject of Eudiometry. ANALYSIS OF MIXED GASES. 113 (E.) To the other portion of the gas (not mixed with oxygen), add its own bulk of chlorine, and keep in a dark place. Contraction and form- ation of oily drops, show the presence olefiant gas. Withdraw the excess of chlorine by agi- tation with potash, and heat potassium in the residue so long as contraction occurs.* Carbonic oxide is ab- sorbed. To the residue add an equal bulk of oxy- gen, and ignite by the electric spark. If explosion takes place, both light car- buretted hydrogen and hydrogen may have been present. Carburetted hydro- gen produces its own volume of carbonic acid. By ascertaining, (l.) the volume of the gas before adding oxygen ; (2.) the amount of oxygen added ; (3.) of car- bonic acid produced ; and ( 4.) of Oxygen in excess (by removing it afterwards by phosphorus), the presence and quantity of hydrogen may be calculated. (See following table.) As nitrous oxide interferes with the above process, it must be previously removed by agitation with water. If any gas re- mains after detonation and absorption of carbonic acid by potash, and of oxygen by phosphorus, this must be nitrogen. * This operation is best performed in a glass tube closed at one end, about eight inches in length, with one inch and a half from its top recurved to hold the potassium. 114 QUALITATIVE ANALYSIS. I .3.8 S3 ECTION OF POISONS. 115 SECTION IX. DETECTION OF POISONS IN ORGANIC MIXTURES. When an inorganic poison is presented for examination unmixed with organic matters, its composition may be ascer- tained by processes already described ; but if the poison is to be looked for in the contents of the stomach, or any such organic mixture, its properties may have become so masked, its composition so altered, and the usual action of reagents so restrained, as to require a modification of the processes employed where organic matters are absent. In general, it is not convenient, in these examinations, to follow the course recommended in mineral qualitative analysis, of de- termining the presence or absence of certain classes of bodies, and afterwards proceeding to individual substances, but better to seek at once for a particular substance, by the ap- plication of its appropriate test. This plan is adopted for two reasons: first, because the act of poisoning is almost always committed with a single substance ; and second, be- cause there usually exist evidences, of a physiological or circumstantial character, which have excited a suspicion of the true nature of the poison ; the appropriate tests for which are then applied. The substances for the discovery of which, in organic mixtures, processes will be described, are the following : 1. Arsenic 6. Nitric Acid 2. Mercury 7. Sulphuric acid 3. Antimony 8. Hydrochloric acid 4. Copper 9. Oxalic acid 5. Lead 10. Hydrocyanic acid. 1. Arsenic. The ordinary liquid tests for arsenic can seldom be suf- ficiently depended on in organic solutions. The only un- equivocal test of the presence of arsenic is, then, the pro- I 2 116 QUALITATIVE ANALYSIS. duction of the metal itself, by processes of reduction. This reduction is performed in two different ways: first, by the formation and subsequent decomposition of arsenietted hydrogen gas ; and second, by the formation and reduction of the sulphuret of arsenic by means of black flux, a mixture of charcoal and carbonate of soda or potash. The principle of the first method (Marsh's test), which is the simplest and easiest, is, that whenever hydrogen gas is evolved in contact with an arsenical compound, soluble or insoluble, the arsenic unites with hydrogen, forming arsenietted hydrogen gas, in which the existence of arsenic is easily demonstrated. The arrangement represented in F , is the figure was contrived by Mr. Marsh, for the operation adverted to. The stopcock, which can be fitted to the tube by means of a cork, being removed, a fragment of pure zinc is introduced into the lower bulb, and dilute sulphuric acid poured upon it, up to the mark a. Hydrogen gas is evolved, with which the whole limb b is soon filled on closing the stopcock, the acid liquid being then forced into the upper bulb. As the zinc of commerce almost always contains some arsenic, it is necessary to examine whether the hydrogen is free from it, or not. For this purpose the gas is kindled at the jet, and allowed to burn with a small flame. On holding the bottom of a porcelain bason or a stoneware plate in the flame, a black spot, of a bright metallic lustre, will immediately be formed on it, in the centre of the flame, if the gas contains any, even a very small amount of arsenic. Another method of discovering the presence of arsenic in the gas, is to hold a piece of coloured glass over the flame, when the glass becomes covered with a white sublimate of arsenious acid, if any arsenic is present. If the hydrogen gas presents these appearances, some arsenic must have existed either in the acid or the zinc; but if it does not, pour off some of the dilute sulphuric acid, and mix with the remainder a portion of the liquid to be tested. Proceed as before to collect the gas, and burn it DETECTION OF POISONS. 117 against a stoneware plate, and under a piece of coloured glass, as before directed. Instead of burning the gas in the way described, it may be passed through a hard glass tube, a portion of which is heated red hot by a spirit-lamp, by which all the arsenic is reduced and condensed in the tube, near the flame of the lamp. For this purpose, a bottle si- milar to that represented in the an- nexed figure may be employed. The funnel is for adding more acid or sus- pected liquid. This modification of the previous process is more manage- able, and equally satisfactory. It may be left in action without requiring the constant presence of the oper- ator ; and the whole amount of arsenic originally contained in the liquid operated on, may be obtained without loss. It affords, moreover, the means of distinguishing between arsenic and antimony, these metals being easily confounded by the test of burning the gas at the jet. Antimony forms with hydrogen a gaseous compound corresponding to arsenietted hydrogen, which also gives, on burning, a white sublimate, and a metallic deposit in the interior of the flame on a stoneware plate. But when anti- monietted hydrogen is passed through a red-hot tube, the reduced antimony is deposited much nearer the part where heat is applied than arsenic is, and, unlike the latter metal, on both sides of the heated part. By these characters, anti- monietted hydrogen and arsenietted hydrogen may be dis- tinguished. When the ordinary apparatus with the jet is employed, a good method of discriminating between antimony and arsenic is by collecting and examining the white sublimate formed by the combustion of the gas. For this purpose, the subli- mate is most conveniently collected by condensing it within a glass tube open at both ends, held slightly oblique. This tube may be about six inches in length, and half an inch in diameter. When the sublimate produced by the combustion I 3 118 QUALITATIVE ANALYSIS. of twice the contents of a bulb has been obtained, it may be treated with a little water, to which a single drop of solution of ammonia has been added, when it dissolves, if it is arse- nious acid, giving a solution in which arsenic is easily recog- nised by the two principal fluid tests for arsenious acid, namely, ammonio-nitrate of silver and ammonio-sulphate of copper. If the metal is antimony, the proper indications which these reagents give with arsenious acid, mentioned in the tables, pp. 42. and 45., are not perceived. The test of ammonio-nitrate of silver is prepared by adding dilute am- monia very gradually to a solution of nitrate of silver, until the oxide at first precipitated is redissolved. As an excess of ammonia would be injurious, it is advisable to leave un- dissolved a trace of the oxide of silver, which may be sepa- rated afterwards by filtration. This reagent produces a precipitate of the yellow arsenite of silver, as well in free arsenious acid as in a solution of an arsenite in water. The test of ammonio-sulphate of copper is prepared in a similar manner, with ammonia and a solution of sulphate of copper. A sublimate of metallic arsenic may also be distinguished, and even separated from metallic antimony, by a solution of chloride of soda, or of chloride of lime, which dissolves the arsenic, but leaves the antimony untouched. (BischofF.) In the reduction of the sulphuret of arsenic to the metallic state, the other process of reduction, the following operations are necessary : " I. Preparation of the fluid : 1. Boil the matters with water and a few drops of nitric acid. 2. Strain through calico. 3. Precipitate animal matter by an excess of nitrate of silver, and subsequent addition of common salt. 4. Filter through paper. "II. Precipitation of the sulphuret of arsenic : 1. Transmit a stream of sulphuretted hydrogen through the liquid for half an hour. 2. Heat the liquid in an open vessel for a few minutes, to cause the precipitate to separate. DETECTION OF POISONS. 119 3. Wash the precipitate by affusion of water acidulated with hydrochloric acid, and subsidence. 4. Dry the precipitate at a temperature not exceed- ing 300. III. Reduction of the sulphuret of arsenic : 1. Mix the dried precipitate intimately with twice its weight of dry black flux (carbonate of potash and charcoal), and heat to redness in a glass tube, of the form and size of a or b exhibited below. 2. Heat slowly a particle of the metallic crust in a glass tube, c, and observe the formation of a white crystalline sublimate of arsenious acid." (Graham's Elements of Chemistry, pp. 633, 634.) Fig- 17. c A solution of arsenic acid is precipitated by sulphuretted hydrogen with less facility than a solution of arsenious acid ; a trace of arsenic acid may even effectually resist decomposi- tion by sulphuretted hydrogen. When the arsenic, there- fore, can be contained in a suspected liquid in the state of arsenic acid, it should be reduced to the state of arsenious acid before transmitting sulphuretted hydrogen. This may be readily effected by sulphurous acid at a boiling heat. (WohleiO An aqueous solution of sulphurous acid is mixed with the sus pected liquid, and the mixture boiled ; by this, the arsenic acid is reduced, while the excess of sulphurous acid is ex- pelled in the ebullition. Sulphuretted hydrogen may then be passed through the solution, and the precipitated sulphu- ret treated in the usual manner. I 4 120 QUALITATIVE ANALYSIS. 2. Mercury. The mixture suspected to contain mercury is to be eva- porated almost to dryness, and the organic matter destroyed by boiling the residue with aqua regia so long as red fumes are evolved. Evaporate nearly to dryness, to expel most of the excess of acid, dilute with water, and filter. The solu- tion thus obtained may be treated either with sulphuretted hydrogen, and the precipitated sulphuret subsequently re- duced by heating in a small glass tube with carbonate of soda, or the solution may be mixed with protochloride of tin, and the precipitated metallic mercury collected and sublimed in a glass tube. (p. 37.) 3. Antimony. When introduced as tartar emetic, antimony may exist in organic fluids either in solution, or it may be rendered insoluble by combining with organic compounds. A mix- ture of tartaric and muriatic acids is to be first added to the suspected mixture, the former to dissolve out oxide of anti- mony from its insoluble compounds, the latter to coagulate various animal principles which may be present. Filter and transmit sulphuretted hydrogen gas through the solution ; an orange-coloured precipitate of sulphuret of antimony will be formed, if that metal be present. Collect the precipitate, dry and heat it in a small capsule, with strong hydrochloric acid ; it is then dissolved, with evolution of sulphuretted hy- drogen : evaporate off most of the excess of hydrochloric acid, and pour the liquid residue into pure water, on which a white precipitate of oxichloride of antimony will be pro- duced, (p. 33.) 4. Copper. This metal, like the preceding, may be present in an organic mixture in two states ; in solution, and in a state of combination with organic principles, forming insoluble com- pounds. As acetic acid dissolves out the oxide of copper DETECTION OF POISONS. 121 from most of these organic compounds, the mixture, after being boiled, should be treated with that acid, diluted and filtered. Sulphuretted hydrogen gas is now passed through the solution, when, if copper is present, a black or brown precipitate of the sulphuret is formed, which, after boiling the liquid, is to be washed by affusion and subsidence. The precipitate is then collected in a porcelain capsule, heated to redness with access of air, to destroy any organic matter with which it may be mixed, and treated with nitric acid to dis- solve the copper. The solution then obtained may be tested with ammonia, or any other characteristic test of copper, (p. 33.) If necessary to examine the portions of the sus- pected mixture insoluble in acetic acid, the matter should be well dried, and heated to redness in a porcelain cru- cible, by which most of the copper will be reduced to the metallic state. The residue, treated with nitric acid, affords a solution to which the usual tests of copper may be applied. 5. Lead. Add a little nitric acid to the suspected mixture before filtration, to dissolve the compounds of oxide of lead and organic matters, and transmit sulphuretted hydrogen gas through the filtered liquid. Boil and collect the precipitate on a filter. To prove the existence of lead in the precipitate, digest it in moderately strong nitric acid, with the assistance of a very gentle heat, by which the lead will be dissolved, while almost all the sulphur remains unacted on. If strong nitric acid is employed, the sulphur will be converted into sulphuric acid, and the insoluble sulphate of lead formed. The acid solution of nitrate of lead must be poured off from the sulphur, evaporated to dryness, and redissolved in water. This solution can then be examined by chromate of potash, or any other test of lead. (p. 37.) Instead of digesting the sulphuret of lead in nitric acid, it may be examined before the blowpipe, by heating with soda on charcoal, when it gives a globule of metal easily recognised 122 QUALITATIVE ANALYSIS. as lead, by its softness, ready fusibility, and the peculiar lustre of a freshly cut surface. 6. Nitric acid. Neutralize the liquid for examination with potash, filter, and evaporate to obtain crystals of nitre. To the salt thus obtained, the test of protosulphate of iron and sulphuric acid may be applied (page 51.) ; or the crystals may be mixed with sulphuric acid and copper filings, when ruddy fumes of peroxide of nitrogen are perceived arising from the nitric oxide evolved. If much organic matter is present, the nitrate of potash will not crystallize ; in which case the dry residue of evaporation is redissolved, and a solution of acetate of silver added, which not only removes organic matters, but also decomposes the chloride of sodium present in almost all organic liquids, and which would interfere when the crys- tallized salt is treated with sulphuric acid, by producing free chlorine. After precipitation by acetate of silver, filter, and evaporate to obtain crystals. (Christison.) 7. Sulphuric acid. Although introduced in the state of oil of vitriol, sulphuric acid may soon exist in the form of a neutral sulphate, from coming in contact with bases or carbonates, or from the gradual formation of ammonia through decomposition of the animal matters. This acid, therefore, when suspected, must be looked for as well in the free as in the combined state. The existence of a free acid in considerable quantity having been ascertained by blue litmus paper, if the liquid does not give the usual appearance presented by sulphuric acid with a salt of barytes (page 52.), in consequence of the presence of organic matters, it must be distilled to dryness in a proper retort and receiver, the heat being increased to dull redness towards the close. The receiver should contain a dilute solution of ammonia, to arrest all the sulphurous acid which is formed by the action of the organic substances on sulphuric acid. On treating the distilled product with DETECTION OF POISONS. 123 aqua regia, the sulphurous acid is again converted into sulphuric acid, the presence of which is then easily ascer- tained by chloride of barium. If the sulphuric acid is in a state of combination with bases, the liquid suspected to contain it may be filtered, and the ordinary tests applied. In examining organic mixtures for sulphuric acid, it is always necessary to discriminate between the small quantity naturally contained in these mixtures, and the large excess which should be found when sulphuric acid is really administered with a criminal inten- tion. To do this satisfactorily, a quantitative analysis must generally be had recourse to. 8. Hydrochloric acid. Nitrate of silver, the ordinary test for hydrochloric acid, cannot be applied to the contents of the stomach, from the constant presence of chlorides, as well as of organic matters which precipitate salts of silver. The mixture suspected to contain hydrochloric acid must be cautiously distilled to dryness in a chloride of calcium bath, composed of about equal parts of that salt and water. The test of nitrate of silver with nitric acid (page 49.) is then applied to the pro- duct of the distillation. If the subject of examination is the contents of the stomach, and a mere trace only is detected, no conclusion as to poisoning can be drawn, as this may have actually existed as an ingredient of the gastric juice. 9. Oxalic acid. The mixture suspected to contain this acid is first filtered and mixed with carbonate of potash. The oxalic acid is then precipitated by a solution of acetate of lead ; the pre- cipitate (partly oxalate of lead) is collected on a filter, washed, diffused in water, and decomposed by sulphuretted hydrogen : sulphuret of lead and free oxalic acid are formed. The mixture is then filtered, heated to expel sulphuretted hydrogen, and a salt of lime and nitrate of silver are ap- plied as tests (page 50.) 124 QUALITATIVE ANALYSIS. If lime or magnesia has been administered as an antidote, this process must be somewhat modified. The suspected mixture is allowed to stand some time to permit the inso- luble oxalate to subside, and the supernatant liquid de- canted. The insoluble residue is now boiled with a solution of carbonate of potash, and filtered. By this operation there are formed a carbonate of the earth and the oxalate of potash. Add to the solution a slight excess of nitric acid, filter, render the liquid alkaline by carbonate of potash, and filter again. The object of these operations is to get rid of animal matter, some portion being removed at each step. The alkaline solution of oxalate of potash is now precipitated by acetate of lead, and the precipitated oxalate treated as above by sulphuretted hydrogen. 10. Hydrocyanic acid. If the organic mixture is of such a nature as to prevent the detection of this acid by its usual tests, recourse must be had to distillation. The mixture is first filtered, mixed with sulphuric acid, to combine with all the ammonia which may have been produced by the decomposition of organic matters, and then distilled in a water-bath until one eighth has been collected. To this product the tests of nitrate of silver and of potash and protosulphate of iron (page 49.) are applied. ' SECTION X. QUALITATIVE ANALYSIS OF URINE, AND DISCRIMINATION OF URINARY CALCULI. ANALYSIS or URINE. (Abridged from the " Traite de Chimie" of JBerzelius. ) The substances to be looked for are uric acid, albu- men, colouring matter of the bile, urea, phosphate of lime, phosphates of the alkalies, lime, sulphuric acid, phosphoric acid, fibrin, caseous matter, hydrochloric acid, mucus, free acid and alkali, and sugar. The following are the reagents to be used, with their applications : ANALYSIS OF URINE. 125 1. Nitric acid. This is employed to detect uric acid, urea, albumen, and the colouring matter of the bile. It is known whether uric acid is present by adding a few drops of nitric acid to half a pint of urine, and allowing it to stand for twelve hours, when uric acid will be deposited on the sides of the vessel. To detect urea, evaporate the urine to one eighth of its original bulk, and add half its reduced volume of nitric acid, when immediately, or on standing a short time, crystals of nitrate of urea are deposited. Nitric acid gives a floccy white precipitate with albumen, which is soluble in caustic potash after being washed, and is not reprecipitated by acetic acid. 2. Ammonia precipitates the phosphate of lime held in solution by the free acid of the urine. 3. Lime-water shows the presence of alkaline phosphates, by the precipitate of phosphate of lime which it produces. 4. Oxalate of ammonia is employed to precipitate the lime contained in urine. If ammonia is afterwards added, the ammoniacal phosphate of magnesia may be precipitated. Should no precipitation take place, add a solution of phos- phate of soda, to ascertain whether this is owing to the absence of magnesia or that of phosphoric acid. 5. Acetate of barytes is used to indicate sulphuric acid. The urine must be slightly acidified by acetic acid. 6. Neutral acetate of lead may precipitate the chloride and the phosphate of lead : distinguish these by the blowpipe. 7. Solution of alum causes a troubling in urine that con- tains albumen or fibrin in solution. 8. Chloride of mercury (corrosive sublimate) gives no pre- cipitate in acidified urine, unless albumen or caseous matter is present. 9. Infusion of gall-nuts or tannin precipitates at least two constituent principles, namely, mucus and the extractive matter, which last is also precipitated by acetate of lead. 10. Red and blue litimus papers are used to detect free alkali and acid. 11. Yeast is employed to discover the presence of sugar in urine, by exciting the vinous fermentation. 126 QUALITATIVE ANALYSIS. DISCRIMINATION (BY CHEMICAL TESTS) OF URINARY CALCULI CONSISTING OF A SINGLE DEPOSIT, OR OF ALTERNATING CALCULI. 1. Bone-earth calculus. Insoluble in potash and in acetic acid, soluble in dilute nitric and hydrochloric acids. Before the blowpipe it first becomes black, and afterwards white : it is fused with difficulty. 2. Ammonio-magnesian phosphate. It does not dissolve in potash, but evolves ammonia. Soluble in cold acetic and dilute nitric acids, and reprecipitated by ammonia. It gives off ammonia at 212, and melts into a white pearl before the blowpipe. 3. Fusible calculus. A. portion is dissolved by acetic acid, and the remainder by hydrochloric acid. It readily fuses into a pearly bead before the blowpipe. 4. Uric acid calculus. Readily soluble in potash, and is reprecipitated by acids. In strong nitric acid it dissolves with effervescence, the solution leaving, when evaporated to dryness, a residue, which on treating with an excess of ammonia becomes of a purple-red colour. Uric acid is nearly insoluble in hydrochloric acid. Before the blowpipe it evolves an ammoniacal odour and blackens, leaving a minute portion of a white ash, which possesses an alkaline reaction. 5. Urate of ammonia calculus. It is soluble in potash with evolution of ammonia. Is readily soluble in alkaline carbonates, while uric acid is not. With nitric acid it behaves as uric acid does. It usually decrepitates before the blowpipe. 6. Cystic oxide calculus. It is soluble in alkalies and in the carbonates of the fixed alkalies, giving a solution which is decomposed by heat, ammonia being first evolved, and after some time a combustible gas, smelling like bisulphuret of carbon. It is soluble in phosphoric, hydrochloric, sul- phuric, nitric, and oxalic acids ; and insoluble in water, alcohol, bicarbonate of ammonia, and tartaric, citric, and acetic acids. Before the blowpipe it exhales a peculiar foetid odour. ANALYSIS OF URINE. 127 7. Xanthic oxide calculus. It is completely dissolved by potash, and is reprecipitated by carbonic acid, white, becoming on drying, a pale yellow agglutinated mass, which possesses a waxy appearance. It is soluble in nitric acid with effer- vescence. When that solution is evaporated to dryness, and the residue treated with ammonia, no red colour is developed, as with uric acid. This calculus is very slightly soluble in hot water, and in hydrochloric and oxalic acids. Concen- trated sulphuric acid dissolves it, forming a yellow solution. 8. Oxalate of lime calculus. Insoluble in potash : it is decomposed by digestion in carbonate of potash, with form- ation of carbonate of lime and oxalate of potash. Insoluble in acetic, but soluble in hydrochloric and nitric acids. When heated to dull redness, it is converted into carbonate of lime, and then dissolves in acids with effervescence. Before the blowpipe, pure lime remains, which, when moistened, produces an alkaline reaction on test paper. 9. Carbonate of lime calculus. It dissolves with effer- vescence in dilute acids, affording a solution which is pre- cipitated by oxalate of ammonia. 10. Calculi containing silica leave, after calcination before the blowpipe, an infusible ash (principally silica), which dis- solves in a melted bead of carbonate of soda with effer- vescence, producing a vitreous pearl, more or less limpid. 128 CHAPTEE IY. QUANTITATIVE ANALYSIS. PRELIMINARY OBSERVATIONS. THE portion of material employed in a quantitative ana- lysis ought not to exceed thirty grains : from twenty to twenty-five grains of an inorganic body is the most con- venient quantity, except in those cases in which a sub- stance existing in minute proportion is to be determined, when more may be used. When a larger amount is ope- rated on, the chances of error in the analysis are seldom reduced, while, from the greater quantities of precipitates to be filtered and washed, and of liquids to be evaporated, a much longer time is necessary for the execution of of the analysis. If the substance to be analyzed is solid and pulverizable, it must be reduced to a fine powder, and cau- tiously dried, to get rid of all hygrometric moisture without expelling any of its chemically combined water. A temper- ature of from 100 to 150 Fahr. is quite sufficient to effect this ; but if the body parts with its combined water at that degree of heat, it is then necessary to dry it at common temperatures, by exposure to a surface of oil of vitriol in a clo*:e vessel. The oil of vitriol is most conveniently con- tained in a shallow bason, across the top of which are placed wires to support a piece of paper, on which the substance to be dried is strewed. The bason is placed on a level surface, and covered with a bell jar, having a ground bottom ; or, if necessary, under the receiver of an air-pump, in which a vacuum is maintained. Six or eight hours exposure is gene- rally sufficient. To prevent the absorption of any water during weighing, the substance should be weighed in a tube about two inches in length and a quarter of an inch in diameter, transferred directly to the vessel in which it is to be dissolved, and the weight of the empty tube with its adhering particles ascertained and deducted from the entire QUANTITATIVE ANALYSIS. 129 weight of the tube and substance. If the substance for analysis is a liquid, it may be weighed in a specific-gravity bottle, whose weight when empty has been previously ascer- tained; the liquid is poured out into a convenient vessel, and the bottle washed out several times, with distilled water, the washings being added to the original liquid. It has already been stated that a substance is generally obtained in a fit state for weighing by precipitation from a state of solution. Particular processes will now be described which are practised in the quantitative estimation of bodies, supposing those bodies or their compounds to exist in a state of purity, with the various methods of separating substances from one another in a state of mixture, for the purpose of their quantitative determination, commencing with the alkalies and earths, and proceeding to the metals proper and non-metallic bodies, those mixtures only being considered which occur in nature or are likely to be presented for examination. I have pursued as far as possible the plan adopted by Rose, in his valuable (S Handbuch der Anatytischen Chemie" of treating, under each substance, of the mode of separating it from those bodies which have been previously considered ; so that a method of separating any two substances being required, the process, if given, will be found under the last of the two bodies in question. In quantitative as in qualitative analysis, advantage is sometimes taken. of certain similarities in the properties of bodies, which permit of their classification into groups, sepa- rable from each other by a single agent : thus sulphuretted hydrogen removes one class of metals proper from a solution, and hydrosulphate of ammonia another class, leaving earths and alkalies, the former of which may be precipitated by an alkaline carbonate. 130 QUANTITATIVE ANALYSIS. SECTION I. ALKALIES AND THEIR METALLIC BASES. I. POTASSIUM. If the potash to be weighed exists either as sulphate or nitrate, or as chloride of potassium, the salts are weighed as such, if pure, and the quantity of base contained in them is calculated.* If in a state of solution as neutral sulphate, evaporate the liquid to dryness in a platinum capsule, or in one of porce- lain if the solution contains anything which can act on pla- tinum ; transfer the residue carefully to a weighed platinum crucible, ignite with the cover on, to prevent loss by de- crepitation, and weigh what remains. If an excess of sul- phuric acid existed in the solution, the bisulphate of potash will be formed, which must be converted into the neutral sulphate before weighing. This is done by placing a piece of dry carbonate of ammonia in the crucible, when heated nearly to redness, and afterwards heating more strongly. The excess of sulphuric acid volatilizes as sulphate of am- monia, neutral sulphate of potash remaining. When the potash exists as nitrate, the solution is evaporated to dryness in a weighed platinum capsule, and kept at a temperature of 212 until the weight becomes constant. If as chloride of potassium, the only particular precaution to be observed is, that this salt must not be heated above a dull red heat, as it is volatile at somewhat higher tempe- ratures. If united to the weaker volatile acids, the salt of potash must be converted into the sulphate, or into chloride of potassium, by the addition of sulphuric or hydrochloric acid, evaporating to dryness and igniting with the precautions above mentioned. When in combination with an acid which when free is soluble in alcohol, potash is sometimes estimated as the double chloride of platinum and potassium. The solution is eva- * Tables for calculations in analysis are given in the Appendix. SODIUM. 131 porated to a small bulk, mixed with solution of chloride of platinum, and evaporated nearly to dryness. When treated with alcohol, the residue dissolves, with the exception of the double chloride of platinum and potassium, which is collected on a weighed filter (page 11.), washed with alcohol, dried and weighed on the filter. From the weight of the double salt that of the potassium it contains may be calculated. II. SODIUM. Soda may be weighed either in the state of sulphate of soda, or chloride of sodium ; the same precautions being ob- served in the ignition of these salts as with the corresponding compounds of potassium. Separation of potash from soda. One process for the separation of these alkalies is founded on the solubility of the chloride of platinum and sodium, and the insolubility of the corresponding compound of chloride of potassium. Both alka- lies are converted into chlorides, ignited and weighed. They are then redissolved in water and mixed with 3-7 times their weight of crystallized chloride of platinum and sodium, which precipitates the chloride of platinum and potassium. The precipitate is collected and weighed in the manner described under potassium, and from its amount, that of the chloride of potassium is calculated. The weight of the latter deducted from the weight of the mixed chlorides, gives that of the chloride of sodium. If hyperchloric (perchloric) acid is added to a mixture of potash and soda salts, it precipitates potash, but not soda. The mixture should be evaporated nearly to dryness, and the residue treated with alcohol, in which hyperchlorate of potash is quite insoluble. This salt is collected on a filter, dried and heated carefully to redness, when it leaves chloride of potassium, which may be weighed, and the corresponding amount of potassium or of potash calculated. The weight of the mixed salt of soda and potash being known, the loss expresses the quantity of soda. When existing as sulphates (into which most salts of the alkalies may be converted by the addition of sulphuric acid), K 2 132 QUANTITATIVE ANALYSIS. add an excess of a solution of hyperchlorate of barytes: collect the precipitate, which consists of hyperchlorate of potash and sulphate of barytes, on a filter and wash it with hot water to dissolve out the former salt. If the sulphates did not contain excess of acid, the weight of the sulphate of barytes may be ascertained, and from it that of the sul- phuric acid calculated. Evaporate the filtered liquid to dryness, and treat the residue with alcohol, by which hyper- chlorate of soda and the excess of hyperchlorate of barytes are dissolved, and an insoluble residue of hyperchlorate of potash remains, which is filtered, washed with alcohol, dried, ignited, and weighed as chloride of potassium. The filtered alcoholic solution, containing hyperchlorates of soda and barytes, is evaporated to dryness, the salts are redissolved in water, and sulphuric acid is added to precipitate barytes. Filter and evaporate the solution, which now contains sul- phate of soda, to dryness : ignite the residue with a little carbonate of ammonia, and weigh as neutral sulphate of soda. If the alkalies exist as chlorides, a similar process to the last is followed, hyperchlorate of silver being substi- tuted for hyperchlorate of barytes, and hydrochloric acid for sulphuric acid. Chloride of silver is then formed in the steps in which sulphate of barytes would otherwise be obtained. % ALKALIMETRY. The determination of the proportion of pure soda existing in a specimen of impure carbonate, as the soda-salts of com- fig- 18 - merce, is a process of considerable importance in connection with the arts. An exceedingly simple and easy method of ascertaining this has been devised, which consists in finding what quantity of an acid is required to destroy the alkaline reaction of a known weight of the specimen to be analyzed, which quantity of acid is of course equivalent to the amount of base present. In this operation a tube measure, called an alka- limeteris employed (Jig. 18.), about fourteen inches in height, and five-eighths of an inch in diameter, mounted ALKALIMETKY. 133 on a foot, and capable of holding, when full, a little more than 1000 grains of water. It is graduated up to this amount into one hundred divisions, numbered from the top downwards ; each division, therefore, contains ten grains of water. The acid employed is sulphuric, made of such a strength, that a single measure of the alkalimeter (ten grains measure of distilled water) is sufficient to neutralize exactly one grain of pure soda. Acid of this strength is obtained in the following manner: 170*6 grains of pure carbonate of soda (obtained by heating the bicarbonate to dull redness in a platinum or porcelain crucible) are dissolved in four or five ounces of hot water. This quantity contains exactly 100 grains of pure soda. A quantity, say a pint, of a dilute acid is prepared by mixing one part of oil of vitriol with ten of water by measure, with which the alkalimeter is filled up to (1000 grains measure of water), and poured from this very gradually into the solution of carbonate of soda above mentioned, until it has destroyed the alkaline reaction of the latter on test paper, and the mixture has become very slightly acid. The quantity of acid which has been required to pro- duce this effect is accurately observed. Supposing it to be 90 measures, this quantity is of course equivalent to 100 grains of soda ; but an acid is wanted, of which 100 mea- sures, instead of 90, are required to neutralize the soda, to procure which we have simply to add to 90 volumes of the acid, 10 volumes of water, so that the same amount of real acid which was before contained in 90 shall now be con- tained in 100 parts. For this, any vessel of sufficient capa- city, divided accurately into 100 equal parts, will suffice. Such is the method of preparing the test acid, of which 1 alkalimeter measure (10 grains of water) is equivalent to 1 grain of pure soda: 22 measures should neutralize 100 grains of crystallized carbonate of soda, and 58 J measures, 100 grains of anhydrous carbonate of soda. In performing an actual analysis, 100 grains of the soda salt to be tested are weighed, dissolved in about three ounces of hot water, filtered if necessary, and neutralized with the K 3 134 QUANTITATIVE ANALYSIS. test acid poured from the alkalimeter. The number of measures used represents the amount of soda per cent, in the specimen. If, instead of 100 grains of the soda salt, 50 grains are operated on, the number of measures must of course be doubled to obtain the per-centage of soda. The ordinary impurities in soda salts are chloride of sodium, sulphate of soda, and insoluble matters ; but some- times other bodies are present, which interfere with the above process, by neutralizing a portion of the acid ; such as sulphuret of sodium, sulphite and hyposulphite of soda, which are liable to occasion a considerable error, by taking up the sulphuric acid. To avoid these sources of fallacy, it is recommended to mix the filtered solution of the salt with chlorate of potash, to evaporate to dryness, and to heat the residue to redness; by this, the sulphuret, sulphite and hyposulphite become converted into the sulphate of soda, which does not interfere in the process. The analysis of a mixture of caustic soda and carbonate of soda is effected by first making an alkalimetrical experiment to ascertain the whole amount of soda, and afterwards esti- mating the carbonic acid by a process which will be de- scribed under the analysis of carbonates. From the data thus obtained, the proportions of caustic soda and carbonate of soda are easily calculated. This may likewise be deter- mined by means of a solution containing muriate of am- monia and chloride of calcium, from which carbonate of lime is precipitated proportional in quantity to that of the car- bonate of soda in the alkaline mixture. The same test acid used for soda may also be employed in analyzing the carbonates of potash ; but as the equivalent of potash is higher than that of soda, the number of measures poured from the alkalimeter cannot represent the per-centage of this alkali, as it does of soda. A simple calculation, how- ever, reduces the numbers for soda to those for potash ; their respective equivalents being very nearly as 4 to 6, the number of measures necessary to saturate 100 grains of the specimen, multiplied by 6, and divided by 4, gives the per-centage of potash. AMMONIA. LITHIUM. 135 III. AMMONIA. If the ammonia whose amount is to be estimated exists in aqueous solution, a slight excess of hydrochloric acid is added, and the solution is evaporated to dryness in a weighed platinum capsule, by means of a water-bath. When dry, the residue is weighed, again heated in the water-bath, and weighed. This is repeated so long as it suffers any loss. From the weight of the muriate of ammonia obtained, that of the ammonia is calculated. After the last weighing, the capsule is heated strongly, to sublime the muriate of ammonia. The loss now experienced, which should be pure muriate of ammonia, affords another datum in the calculation. The estimation of ammonia in ammoniacal salts is per- formed in the following manner. A known weight of the salt is introduced into a small tubulated retort, having a tube funnel passing through a cork in the tubulure. The extremity of the funnel must reach nearly to the bottom of the retort, and be drawn out to a capillary just large enough to allow a liquid to flow through. The substance being in- troduced, the retort is connected with a receiver containing dilute hydrochloric acid, the beak of the retort just touching, but not dipping under, the surface of the liquid. A strong solution of potash is then introduced into the retort through the funnel, and heat is applied to distil the ammonia, which becomes free by the action of potash, into the receiver. If necessary, more solution of potash may be added, and after- wards water, to drive all the ammonia into the dilute acid. The solution of muriate of ammonia thus obtained in the receiver is evaporated in the manner above described. IV. LITHIUM. The best mode of estimating lithia, every base except pot- ash and soda having been previously separated from the solu- tion, is by means of the double phosphate of soda and lithia. Pure phosphate of soda and a little carbonate of soda are K 4 136 QUANTITATIVE ANALYSIS. added to the solution, and the whole is evaporated to dryness. On dissolving the dry mass in water, a white salt remains, insoluble in a solution of phosphate of soda, but slightly soluble in hot water. This is the double phosphate of soda and lithia. The quantity precipitated generally increases a little on standing. It is collected on a filter and washed with cold water. In consequence of the slight solubility of the double phosphate, the washing should not be pro- longed beyond what is absolutely necessary. It should be heated to redness before being weighed. If lithia is in combination with a volatile acid, and no other alkali is present, it may be estimated as sulphate, by simply adding sulphuric acid, evaporating to dryness, and heating to redness before weighing. Separation of lithia from potash and soda. When the three alkalies exist together in solution, first determine the entire weight of the mixed salts by evaporation to dryness ; after weighing, redissolve the residue, and estimate the potash as chloride of platinum and potassium. Through the solu- tion filtered from the double chloride, transmit a stream of sulphuretted hydrogen gas, to remove the platinum it con- tains from the excess of chloride employed, evaporate, and proceed to estimate lithia as the phosphate of lithia and soda. The sum of the weights of the potash and lithia salts (ascer- tained by calculation) deducted from the weight of the mixed salts first determined, leaves the weight of the soda salt. SECTION II. ALKALINE EARTHS AND EARTHS PROPER, AND THEIR METALLIC BASES. I. BARIUM. Barytes is weighed in the form of sulphate. On adding sulphuric acid to the solution containing barytes, the sul- phate is immediately precipitated, but occasionally does not subside readily, and then passes through the pores of the filter. On heating the liquor, or on the addition of a little STRONTIUM. 137 nitric or hydrochloric acid, the precipitate aggregates, and may then be collected on a filter. The sulphate of barytes is washed, dried, and ignited in a platinum or porcelain crucible. From its weight that of barium or of barytes is calculated. Barytes and alkalies are separated from each other by sulphuric acid; the liquid filtered from the precipitate of sulphate of barytes is evaporated to dryness, and the alkaline sulphates may be determined by one of the processes before described. II. STRONTIUM. Strontian is precipitated from its solutions for the pur- pose of weighing either as sulphate or as carbonate; but both of these salts possess a slight degree of solubility in water, and hence cannot be washed by water with impunity. If the strontian is united to an acid, which when free is soluble in alcohol, and nothing can be formed on the addition of sulphuric acid which is insoluble in weak alcohol besides sulphate of strontian, it is preferable to estimate the strontian as sulphate. The solution to be precipitated should be con- centrated, mixed with alcohol and sulphuric acid, and the precipitate washed with weak alcohol, dried, ignited, and weighed. To estimate strontian as carbonate, add to the solution an excess of carbonate of ammonia mixed with a little free am- monia, apply heat to the mixture, filter the precipitate, wash, dry, and heat it to redness. Carbonate of strontian does not part with carbonic acid at a red heat. Strontian and barytes are separated from each other when in solution by hydrofluosilicic acid, which precipitates ba- rytes (although incompletely), but not strontian. A more common, but less perfect method is, to convert the earths into dry chlorides of the metals, and then treat with alcohol, which dissolves chloride of strontium, but leaves chloride of barium undissolved. Strontian is separated from the alkalies either as sulphate, 138 QUANTITATIVE ANALYSIS. by sulphuric acid, or as carbonate,, by carbonate of am- monia. III. CALCIUM. Lime is best precipitated from its solutions as oxalate, by means of oxalate of ammonia. The precipitate, after being washed and dried, is ignited at a red heat, when it becomes converted into carbonate of lime, which is weighed. In performing this process several precautions are necessary. 1. That the solution to be precipitated is either neutral or slightly ammoniacal, as oxalate of lime is soluble in acids. If the solution is acid, a slight excess of ammonia must be added; but this alkali may occasion the precipitation of a salt of lime, soluble in acids, but insoluble in water, and therefore precipitated on neutralization by an alkali ; such, for instance, as the phosphate or arseniate. For such salts of 'lime the process by oxalate of ammonia cannot be em- ployed. 2. As the oxalate of lime subsides very slowly, the mixture must be boiled for some time, and allowed to stand until all the oxalate has subsided, otherwise it is impossible to filter the solution perfectly. 3. In the con- version of oxalate into carbonate of lime by ignition, if a strong heat is employed, a portion of the carbonic acid of the carbonate may be expelled, thus subjecting the estimation of the lime to a considerable deficiency; to avoid this, after weighing the ignited carbonate, add to it a few drops of a strong solution of carbonate of ammonia, again ignite at a dull red heat, and weigh. Repeat this as long as any in- crease in weight is apparent. In those salts which are soluble only in acids, and therefore precipitated from their solutions on the addition of ammonia, the lime is determined as sulphate in a manner precisely similar to that described under strontian. Separation of lime from strontian. When these two earths exist together in solution, their separation may be effected in the following manner. They are first converted into carbonates by the addition of carbonate of soda or potash ; the precipitated carbonates are filtered, washed, and MAGNESIUM. 139 dissolved in dilute nitric acid, to be converted into nitrates. The solution of the nitrates thus obtained is evaporated to dryness to expel the last traces of nitric acid, and the residue is treated with alcohol, which dissolves nitrate of lime, but leaves nitrate of strontian undissolved. Collect the nitrate of strontian on a filter, and wash it with alcohol. When washed, the salt may be dissolved in water, and the strontian estimated as sulphate ; the lime of the nitrate in the filtered alcoholic liquid is estimated as oxalate. Separation of lime from barytes. On adding sulphuric acid to the solution containing these two earths, both sul- phate of barytes and sulphate of lime are precipitated, but from the slight solubility of sulphate of lime in water, it may be completely separated from sulphate of barytes by a long continued washing on a filter. A solution of oxalate of ammonia is added to the filtered liquid, which should be con- centrated by evaporation before the precipitated oxalate of lime is collected on a filter. Lime is separated from the alkalies by oxalate of ammonia : on evaporating the liquid filtered from the precipitated oxalate of lime to dryness, and calcining the dry salt, the excess of oxalate of ammonia is expelled, together with any other ammoniacal salt which may be present, leaving a residue which may be treated in the usual way to obtain the alkali. The method of analyzing limestones will be described under the Analysis of Carbonates. IV. MAGNESIUM. A solution of phosphate of soda, to which pure ammonia or its carbonate has been previously added, is employed to pre- cipitate magnesia from its solutions for the purpose of weighing. The magnesia then falls in the state of phosphate of magnesia and ammonia. Several precautions are necessary in this operation. The solution should be concentrated to a small bulk by evaporation ; it should be quite cold, and either neutral or slightly ammoniacal, as the precipitated phosphate of magnesia and ammonia is soluble in free acids. 140 QUANTITATIVE ANALYSIS. Even carbonic acid, and, according to Stromeyer, alkaline carbonates, dissolve it to a small extent. If free carbonic acid exists in the solution, it must therefore be expelled by boiling ; and if alkaline carbonates are present, they should be decomposed by the addition of hydrochloric acid : the solution is afterwards boiled, and the excess of acid neutra- lized by pure ammonia. Phosphate of magnesia and am- monia is also slightly soluble in pure water ; hence, if the proportion of magnesia in the solution is small, no precipitate is at first produced on the addition of phosphate of soda with ammonia. The precipitate appears, however, on stirring for some time, which should never l3e omitted before filtration. Although soluble in pure water, phosphate of magnesia and ammonia is not in water which contains phosphate or muriate of ammonia in solution. A dilute solution of the latter salt is hence advantageously employed to wash the phos- phate when collected on a filter. When dried and ignited, the muriate of ammonia is expelled, and the phosphate of magnesia and ammonia becomes reduced to phosphate of magnesia, 2 Mg O +PO 5 , containing 36'67 per cent, of mag- nesia, in which state this earth is weighed. Magnesia may also be precipitated in the form of carbonate, by the fixed alkaline carbonates, but owing to the tendency of the latter to form soluble double salts with the carbonate of magnesia, this method of estimating magnesia is not more certain than the preceding. A solution of carbonate of soda (this salt being preferable to carbonate of potash) is added to the hot magnesian solution, and the liquid boiled until the precipitate is no longer bulky and light, but has become granular and heavy. As the presence of ammoniacal salts interferes with the precipitation of magnesia in the state of carbonate, these should be expelled from the solution by boiling with a sufficient quantity of carbonate of soda, until the odour of ammonia is no longer perceptible. The precipi- tate, if granular, and comparatively dense, is then collected on a filter, and the filtered liquid is evaporated to dryness. On treating the residue of the evaporation with boiling water, a minute quantity of carbonate of magnesia generally re- MAGNESIUM. 141 mains undissolved, which should be added to that already on the filter. The precipitate, when washed, is dried, ig- nited strongly to expel carbonic acid, and weighed as pure magnesia. If magnesia exists in solution unmixed with other fixed bases, and not combined with a fixed acid, it may be esti- mated in the state of sulphate. The solution is evaporated to dryness, and if any ainmoniacal salts are present, heated to redness in a platinum crucible. The residue is treated with dilute sulphuric acid, again heated to redness, and weighed when cold. If the magnesia already exists as sulphate, the addition of sulphuric acid is of course un- necessary. The separation of magnesia from lime can be effected by oxalic acid. Ammonia in very slight excess is added to the solution of the two earths in hydrochloric acid, a previous addition having been made of muriate of ammonia, to pre- prevent the precipitation of magnesia by ammonia. If the solution is very acid, a sufficient quantity of ammoniacal salt will be formed for the purpose, and the addition of muriate of ammonia is then unnecessary. If ammonia causes a pre- cipitate of magnesia, this must be redissolved in hydrochloric acid, and a slight excess of the alkali again added, on which no precipitation will occur. The lime is then precipitated by oxalic acid, or by oxalate of ammonia, the oxalate of lime is collected on a filter, and magnesia is precipitated from the filtered solution as carbonate or ammonio-phosphate of mag- nesia. After the precipitation of lime as oxalate, the solu- tion might deposit oxalate of magnesia on evaporation, if a large excess of oxalate of ammonia had been added : this is prevented by adding hydrochloric acid in very slight excess before the evaporation. If the magnesia in solution is to be precipitated as phosphate, the hydrochloric acid must be neu- tralized by ammonia before precipitation. Dobereiner has lately proposed an excellent process for the separation of magnesia and lime, by the use of chlorate of potash. The solution of the two earths in hydrochloric acid is evaporated to dryness, and the residue is calcined in a platinum 142 QUANTITATIVE ANALYSIS. crucible, to expel the excess of hydrochloric acid. Powdered chlorate of potash is then added so long as the evolution of chlorine is perceptible. On treating the dry mass with water, chloride of calcium and chloride of potassium are dissolved, arid a residue of pure magnesia remains, which is filtered, washed, ignited, and weighed. The lime in the filtered so- lution may be precipitated by oxalate of ammonia. When heated to redness alone, hydrated chloride of mag- nesium decomposes into magnesia and hydrochloric acid ; hydrated chloride of calcium, however, resists decomposition in this manner, or, at least, at the temperature necessary for the decomposition of chloride of magnesium. Hence when a mixture of the two hydrated chlorides is ignited, on treating the calcined mass with water, chloride of calcium alone redissolves, if the ignition has been continued a suf- ficient length of time for the chloride of magnesium to be fully decomposed. Magnesia then remains as an insoluble residue. It is, however, with difficulty that this decom- position of the hydrated chloride of magnesium can be com- pletely effected alone ; but by the addition of chlorate of potash, as in Dobereiner's process for the separation of lime from magnesia, it is remarkably facilitated. This arises from the decomposition of the hydrochloric acid, as it is evolved from the hydrated chloride, by the oxygen of the chlorate of potash ; water and chlorine being the products of the decomposition. Six equivalents of hydrated chloride of magnesium, and one of chlorate of potash, produce six equi- valents of magnesia, six of water, one of chloride of potas- sium, and six of chlorine : or expressed in symbols : KO, Cl 5 }= 6MgO. 6 HO. KC1. 6 Cl. Separation of magnesia from the fixed alkalies. One of the best means of effecting the separation of magnesia from the fixed alkalies is the following : The magnesia is first precipitated by subphosphate of ammonia, and the precipi- MAGNESIUM. 143 tated phosphate is filtered, washed, ignited, and weighed: acetate of lead is added to the filtered solution, to separate phosphoric acid (from the excess of phosphate of ammonia employed), and the precipitated phosphate of lead is sepa- rated by filtration. The lead in excess now existing in the solution is precipitated by supersaturation with sulphuretted hydrogen gas ; the solution is filtered and evaporated to dry- ness. The residue of the evaporation consists of acetates of the alkalies, which may be converted into sulphates or chlo- rides by the addition of sulphuric or hydrochloric acid, and estimated by methods already described. Liebig recommends a process for the separation of mag- nesia from the fixed alkalies, founded on the higher basic properties of barytes than magnesia. The solution con- taining magnesia and alkalies is mixed with muriate of ammonia, evaporated to dryness, calcined, and the residue is dissolved in a small quantity of water. To this solution, saturated barytes-water is added, and the magnesia hereby precipitated is collected on a filter : barytes is separated from the filtered liquid by sulphuric acid ; the solution is filtered, evaporated to dryness, and the residue calcined with a little carbonate of ammonia (page 130.). Neutral alkaline sulphates remain, which may then be weighed. Berzelius has recently recommended the following process for the separation of magnesia from the fixed alkalies, when existing in the state of chlorides, as is generally the case in mineral waters. The concentrated solution of the chlo- rides is mixed with protoxide of mercury (red oxide), and then evaporated to dryness. The chloride of magnesium and oxide of mercury undergo mutual decomposition, be- coming magnesia and chloride of mercury ; the latter com- bines with the alkaline chloride to form a double salt. The dry mass is treated with water, the solution is filtered, evaporated to dryness, and the residue calcined to expel all the chloride of mercury. What remains after the chloride of mercury is expelled, is the alkaline chloride, which may be weighed. The portion insoluble in water, consisting of a mixture of magnesia with the excess of oxide of mercury, 144 QUANTITATIVE ANALYSIS. is ignited strongly to expel the latter. Pure magnesia then remains, in a fit state for weighing. V. ALUMINUM. For the purpose of weighing, alumina is precipitated from its solutions by ammonia or carbonate of ammonia. As pure ammonia exercises a slight solvent action on alumina, the carbonate should be used in all cases where it is admissible. When pure ammonia must be employed, however, the alumina is completely precipitated on boiling the solution for a short time. The precipitate is very bulky, and must be washed with hot water. If alumina is to be precipitated from its solution in the caustic fixed alkalies, the alkali is first saturated with hydro- chloric acid, of which a sufficient quantity is added to re- dissolve the alumina at first precipitated, and a solution of carbonate of ammonia, or ammonia, is then added to preci- pitate alumina. Separation of alumina from magnesia, lime, and the fixed alkalies. Supposing these bases to exist together in solution, their separation can be effected in the following manner. Muriate of ammonia is first added to the solution, to prevent the precipitation of magnesia by ammonia, which alkali is next added to precipitate alumina. If the solution was previously very acid, the addition of muriate of ammonia is unnecessary, as a sufficient quantity of an ammoniacal salt will be formed for the purpose, by the neutralization with ammonia. The ammonia should be perfectly free from car- bonate, otherwise carbonate of lime will be formed. The pre- cipitate of alumina should be filtered as rapidly as possible, to prevent absorption of carbonic acid from the atmosphere by ammonia, and washed with distilled water. To the liquid filtered from the precipitate of alumina, oxalate of ammonia is added to precipitate oxalate of lime, which is collected on a filter, washed, and converted into carbonate by ignition in a platinum crucible (page 138.). The precipitate of alumina produced by ammonia, carries down a little magnesia with ALUMINUM. 145 it, whatever amount of muriate of ammonia may exist in the liquid : this magnesia is separated by dissolving the alumina in caustic potash, magnesia remaining undissolved by the alkali. To perform this, the precipitate of alumina, while still moist, should be first dissolved in hydrochloric acid, an excess of caustic potash is then added to the solution, and the mixture heated in a porcelain or platinum capsule. The alumina, at first precipitated by the potash, redissolves in the excess of alkali, leaving magnesia undissolved, which is collected on a filter and washed. To the alkaline solution of alumina, hydrochloric acid is added until the precipitate of alumina at first produced is entirely redissolved ; a so- lution of carbonate of ammonia in excess is then added to throw down the alumina. The small quantity of magnesia on the filter should be dissolved in hydrochloric acid, and this so- lution added to that filtered from the oxalate of lime, which contains the greater part of the magnesia, together with the alkalies. Processes for the separation of magnesia from alkalies have already been described. When it* is required to separate alumina from magnesia only, this may be accomplished by the same means as in the process just described, where lime and alkalies also are sup- posed to be present. Muriate of ammonia is first added to the solution, and then caustic ammonia. The small quantity of magnesia precipitated with the alumina is separated by hydrochloric acid and potash, in the manner described. The magnesia undissolved by potash is dissolved in hydrochloric acid, and added to the solution filtered from the precipitate by ammonia. From this solution magnesia may be preci- pitated, as the ammonio -phosphate (page 139.). To separate alumina from lime only, the former is preci- pitated from the solution containing these earths by ammonia, which must be quite free from carbonate of ammonia, other- wise carbonate of lime will also be precipitated. The alumina should be collected on a filter as rapidly as possible, to prevent absorption of carbonic acid from the atmosphere by ammonia. After the filtered ammoniacal solution has been boiled to expel most of the excess of ammonia, the lime it L 146 QUANTITATIVE ANALYSIS. contains may be precipitated by oxalate of ammonia or oxa- lic acid. From barytes, alumina is separated by precipitating the former earth as sulphate, by sulphuric acid or a soluble sul- phate. To separate alumina and strontian, the former may be precipitated by ammonia, the access of air being avoided as much as possible, to prevent the formation of carbonate of strontian. Alumina and the fixed alkalies may be separated by adding carbonate of ammonia, or free ammonia, to the neutral or acid solution. On evaporating the solution filtered from the precipitate of alumina to dryness, and calcining the residue, the ammoniacal salts which are present volatilize, leaving the salts of the fixed alkalies. Solution of aluminous minerals. The solution of a few minerals, which consist entirely or chiefly of alumina, is very difficult to effect by ordinary means. M. Rose has shown that when corundum and aluminates * are reduced to a fine powder, they may be completely dissolved by first fusing them with an excess of bisulphate of potash, and then digesting the fused mass in water. The mineral should be pounded in a steel crushing-mortar, and not in an agate mortar, that no silica may be acquired by abrasion of the latter : it should be reduced to a further state of division by sifting through very fine muslin. A platinum crucible may be employed for the fusion with bisulphate of potash. From the solution of the fused salt in water, alumina may be pre- cipitated by ammonia ; but as a subsulphate of alumina is precipitated, instead of pure alumina, the precipitate should be separated by filtration, dissolved in hydrochloric acid, and again precipitated by ammonia. It is then pure alumina, which may be washed, ignited, and weighed. * The minerals in which alumina acts the part of an acid are called aluminates. YTTRIUM THORINUM GLUCINUM. 147 VI. YTTRIUM. Potash and ammonia are both employed to precipitate yttria from its solutions, in the quantitative determination of that earth. Hydrate of yttria falls, which may be collected on a filter, washed, ignited, and weighed as pure yttria. Except when yttria is in the state of chloride, potash is a better precipitant to employ than ammonia, as the latter alkali precipitates subsalts of yttria from the sulphate and nitrate. The insolubility of yttria in potash affords the means of separating it from alumina) and its being precipitated by ammonia, although in the presence of an ammoniacal salt, from magnesia, lime, barytes, strontian and the alkalies. VII. THORINUM. Thorina is separated from most other bodies by precipi- tation as double sulphate of thorina and potash. A boiling saturated solution of sulphate of potash is added to the solution containing thorina, the mixture is allowed to cool, and the precipitate is filtered and washed with a cold satu- rated solution of sulphate of potash, in which the double salt is quite insoluble. After washing with solution of sulphate of potash, boiling water is poured on the filter, to dissolve the salt, the thorina of which is then precipitated by pure potash. Thorina is separated from alumina by its insolubility in potash ; and from magnesia and lime by being precipitated by ammonia in the presence of an ammoniacal salt. It is separated from alkalies by ammonia, which completely pre- cipitates thorina. VIII. GLUCINUM. Ammonia is the reagent employed to precipitate glucina from its solutions. A voluminous precipitate of glucina falls, which contracts considerably on drying. Glucina is separated from thorina and yttria by its solu- bility in potash, and from alumina by its solubility in car- L 2 148 QUANTITATIVE ANALYSIS. bonate of ammonia. To separate glucina from alumina, add a large excess of a saturated solution of carbonate of ammonia to the mixed solutions of the two earths, and allow the whole to stand for some time, with occasional agitation. When the precipitate of alumina no longer appears to decrease in bulk, collect it on a filter, wash, ignite, and weigh it, and evaporate the filtered solution to dryness. As the carbonate of ammonia volatilizes, the glucina is deposited : when the dry mass is heated to redness, to expel ammoniacal salts, it leaves pure glucina, the weight of which may then be ascertained. A process has been recently recommended for the sepa- ration of alumina and glucina, which is said to be preferable to the preceding. Both glucina and alumina are soluble in a concentrated solution of potash ; but if the alkaline solution is diluted considerably with water, and boiled, all the glucina is precipitated, while the alumina still remains in solution. The process is the following. Both earths are dissolved, with the assistance of heat, in a concentrated solution of potash ; the solution is cooled and diluted with water. The precipitate which falls (glucina) is filtered and washed, and the filtered solution is boiled to precipitate the remaining glucina. The alumina which remains dissolved may be estimated in the usual manner (page 144.). Berzelius * supposes it probable that glucina precipitated by ebullition from a dilute solution of potash, contains potash in chemical combination, for it is insoluble in cold alkaline ley after this operation ; but by washing, all the potash it may contain is removed, and it again becomes soluble in the cold ley. Glucina may be separated from magnesia by a process similar to that described for the separation of alumina from magnesia. The only difference is, that while caustic am- monia or carbonate of ammonia, may be used indiscriminately to precipitate alumina, caustic ammonia only should be em- ployed for glucina. Glucina is separated from lime, strontian, barytes, and the * Rapport Annuel, 1841, 2 s Annee, p. 79. (Paris edition). ZIRCONIUM CERIUM. 149 fixed alkalies, by methods the same as those by which alumina is separated from the same bodies. IX. ZIRCONIUM. Zirconia is best precipitated from its solutions by potash. Ammonia may also be employed, but this alkali sometimes precipitates subsalts instead of pure zirconia. The preci- pitate is washed, dried, ignited with care, and weighed. Zir- conia may be precipitated by sulphate of potash in the same manner as thorina. For this purpose, it is best to add sul- phate of potash in crystals to the solution containing zirconia. The precipitate which falls is a double sulphate of zirconia and potash. If all excess of acid present in the solution is neutralized with potash, zirconia may be completely precipi- tated by sulphate of potash. As the precipitate is partly soluble in pure water, it should be washed with water ren- dered alkaline by the addition of a little caustic ammonia. On boiling the double sulphate of zirconia and potash with a solution of pure potash, hydrate of zirconia remains, which may be washed, ignited, and weighed. SECTION III. METALS PROPER, NOT PRECIPITATED BY SULPHURETTED HYDROGEN FROM THEIR ACID SOLUTIONS. I. CERIUM. Potash is the best precipitant for the oxides of cerium, ammonia having a tendency to throw down subsalts instead of the pure oxides. If the solution previously contained protoxide of cerium, the precipitate of protoxide produced by potash becomes converted by calcination into the peroxide, from the weight of which the amount of protoxide originally present must be calculated. The oxides of cerium are separated from glucina and alu- mina by the insolubility of the former in caustic potash, and from magnesia and lime by ammonia. L 3 150 QUANTITATIVE ANALYSIS. II. MANGANESE. Carbonate of potash is the reagent usually employed to precipitate protoxide of manganese from its solutions. The solution should be boiling at the time of the precipitation. Carbonate of protoxide of manganese falls, which may be collected on a filter. When the solution filtered from the precipitated carbonate is evaporated to dryness, a little more carbonate generally remains undissolved when the dry mass is treated with hot water ; this portion is to be added to that already on the filter. As carbonate of manganese is slightly soluble in ammoni- acal salts, the whole oxide of manganese is not precipi- tated, if these are present in the solution, until a sufficient quantity of carbonate of potash has been added to decompose all ammoniacal salts. To effect their complete decomposition, the liquid should be mixed with carbonate of potash and evaporated nearly to dryness, fresh additions of the alkaline carbonate being made so long as any odour of ammonia is perceived. By this means ammoniacal salts may be com- pletely expelled from the solution. After washing and drying, the carbonate of manganese is heated to redness in an open platinum crucible ; it then parts with carbonic acid, and acquires oxygen from the air, being converted into the red oxide of manganese (manganoso- manganic oxide, Mn O + Mn 2 O 3 ), from the weight of which that of protoxide or deutoxide, originally present in the sub- stance under examination, may be calculated. Separation of protoxide of manganese from alumina. Pro- toxide of manganese may be separated from alumina by boiling the solution with an excess of caustic potash : of the mixture of oxide of manganese and alumina at first preci- pitated, the latter alone dissolves in the excess of alkali. The alumina in solution may then be estimated in the usual manner (page 144.), and the oxide of manganese, insoluble in potash, may be dissolved in hydrochloric acid, and precipitated as carbonate. MANGANESE. 151 Protoxide of manganese and alumina may also be separated by ammonia, muriate of ammonia having been first added to the solution, to prevent the precipitation of protoxide of manganese. Separation of protoxide of manganese from magnesia and lime. When the amount of manganese present is rela- tively small, the following process is recommended. A sufficient quantity of muriate of ammonia is first added to the solution (unless it contains a large excess of acid) to prevent precipitation by ammonia, which is next added, and then an excess of hydrosulphate of ammonia to precipitate sulphuret of manganese. The precipitated sulphur et is col- lected on a filter, washed with water containing a little hydrosulphate of ammonia, and dissolved in pure hydrochloric acid : sulphuretted hydrogen is thus developed, to expel which, the solution is heated until it becomes inodorous. The solution is now filtered, and the manganese precipitated by carbonate of potash. The solution filtered from the sulphuret of manganese, containing lime and magnesia, is slightly acidified by hydrochloric acid, boiled to expel sul- phuretted hydrogen, and filtered. The filtered solution is then supersaturated with ammonia ; first lime is precipitated as oxalate, by oxalate of ammonia or oxalic acid, and after- wards magnesia, as the ammonio-phosphate. When protoxide of manganese constitutes the chief in- gredient, the following process is preferable. Carbonates of manganese and of the two earths are precipitated by a fixed alkaline carbonate, the solution being boiling at the time of precipitation. The precipitate is ignited at a dull red heat, to convert the carbonate of manganese into the manganoso-manganic oxide, and is then treated with very dilute nitric acid. This acid slowly dissolves the lime and magnesia with effervescence, leaving the oxide of manganese unacted on, which may be collected on a filter, washed, again ignited, and weighed. Any trace likewise of manganese dissolved by the nitric acid can be precipitated by hydrosul- phate of ammonia, after neutralizing the acid by caustic L 4 152 QUANTITATIVE ANALYSIS. ammonia. The lime and magnesia may be separated from each other by a process already described. VALUATION OF PEROXIDE OF MANGANESE. A simple and expeditious process for the determination of the commercial value of the higher oxides of manganese, is of some importance from the extensive applications of these oxides in the arts. The following details of a process of this kind are given in Prof. Graham's " Elements of Chemistry " page 536. " The value of these oxides is exactly proportional to the quantity of chlorine which they produce when dissolved in hydrochloric acid, and the chlorine can be estimated by the quantity of protosulphate of iron which it oxidizes. " Fifty grains of the powdered oxide of manganese to be examined are weighed out, and also any known quantity, not less than 317 grains, of the sulphate of iron (copperas) employed in chlorimetry.* The oxide of manganese is thrown into a flask containing an ounce and a half of strong hydrochloric acid, diluted with half an ounce of water, and a gentle heat applied. The sulphate of iron is gradually added in small quantities to the acid, so as to absorb the chlorine as it is evolved, and the addition of that salt continued till the liquid, after being heated, gives a blue precipitate with the red prussiate of potash, and has no smell of chlorine, which are indications that the protosulphate of iron is present in excess. By weighing what remains of the sulphate of iron, the quantity added is ascertained ; say m grains. If the whole manganese were peroxide, it would require 317 grains of sulphate of iron, and that quantity would, therefore, indi- cate 100 per cent, of peroxide in the specimen ; but if a portion of the manganese only is peroxide, it will consume a proportionally smaller quantity of the sulphate, which quantity will give the proportion of the peroxide, by the proportion, as 317 : 100::m : per-centage required. The per-centage of peroxide of manganese is thus obtained by * Clean crystals of the salt, bruised in a mortar, and dried by strong pressure between folds of cloth. IKON. 153 multiplying the number of grains of sulphate of iron perox- idized, by 0. 317." III. IRON. This metal is best weighed in the state of peroxide, as obtained by precipitation from its solution by ammonia, potash, or by alkaline carbonates. If not already existing in solution as peroxide, but as protoxide or black oxide, it must be converted into the peroxide by heating with a little nitric acid, until ruddy fumes of nitric oxide are no longer evolved and the black colour sometimes acquired by the liquid on the addition of nitric acid has disappeared. The amount of metal, protoxide, or black oxide, which was formerly contained in the substance analyzed, is calculated in such a case from the quantity of peroxide obtained. Ammonia is the best precipitant to employ, as the peroxide of iron pre- cipitated by potash always carries down with it some portion of the alkali, which is not removed by washing without great difficulty ; and alkaline carbonates retain a little peroxide in solution at common temperatures. As much of the peroxide as possible is transferred, when washed and dried, to a crucible for ignition, the filter with the remaining peroxide being carefully burned outside the crucible (as described at page 8.), to prevent the reduction of any portions of oxide to the state of protoxide or- black oxide by the carbonaceous matter of the filter. Iron may be separated from several other bodies by pre- cipitation as sulphuret, by hydrosulphate of ammonia. If the solution contains any free acid, it must first be neutralized by ammonia, which alkali may be added in slight excess, as the precipitation of a small quantity of peroxide of iron does not interfere with the result. A sufficient quantity of hydrosul- phate of ammonia is then added, the precipitate is filtered, washed with water containing a little hydrosulphate of am- monia, and dissolved in pure hydrochloric acid. The solution thus obtained is heated to expel sulphuretted hydrogen, a little nitric acid is added to peroxidize the iron, the solution 154 QUANTITATIVE ANALYSIS. is filtered if necessary, and the peroxide of iron is precipitated by ammonia. Separation of peroxide of iron from protoxide of manganese. The perfect separation of these oxides is not accomplished without some difficulty. In the process generally preferred it is effected by means of a neutral succinate or benzoate of an alkali, which precipitates peroxide of iron as benzoate, in a neutral solution, but not protoxide of manganese. If any iron exists in solution as protoxide, it must first be con- verted into peroxide by gently heating with a little nitric acid or aqua regia. If the solution is acid, it must be rendered neutral by ammonia, muriate of ammonia being previously added to prevent the precipitation of protoxide of manganese. A very slight excess of ammonia may be added, so that a trace of peroxide of iron remains undissolved when the solution is gently heated. Solution of crystallized succinate of soda, of neutral succinate of ammonia, or of benzoate of am- monia may be used as the precipitant. A very bulky pre- cipitate of benzoate or succinate of peroxide of iron falls, which, after being filtered and washed, is decomposed by pouring caustic ammonia on it, while still on the filter. By this, a large portion of the benzoic or succinic acid is removed, in the state of a soluble benzoate or succinate of ammonia, peroxide of iron, or rather a subsalt of that oxide, remaining on the filter, which last is dried, ignited with exposure to air, and weighed as peroxide. The object in removing the organic acid by ammonia is to avoid the reduction of the oxide during ignition. From the solution filtered from the precipitated succinate or benzoate, protoxide of manganese may be precipitated by carbonate of soda.* Separation of peroxide of iron from magnesia. These bodies are separated by a process similar to that just described for the separation of peroxide of iron from protoxide of man- ganese. Instead of neutralizing any excess of acid in the * For another method of separating peroxide of iron from protoxide of manganese, see a process for the separation of oxide of nickel from peroxide of iron. IRON. 155 solution carefully, as when manganese is present, an excess of ammonia may be added at once, having made a previous addition of muriate of ammonia, unless the quantity of free acid is considerable. The precipitate, which consists of peroxide of iron with merely a trace of magnesia, is filtered, redissolved in hydrochloric acid, and the solution thus obtained is rendered exactly neutral by ammonia. Peroxide of iron is then to be precipitated as succinate or benzoate, and the liquid filtered from this precipitate is added to that filtered from the precipitate by ammonia, which contains the greater part of the magnesia. From the mixed solutions magnesia may be precipitated as the ammonio-phosphate. Peroxide of iron may be separated from alumina by adding an excess of caustic potash to the solution. The alumina con- tained in the precipitate redissolves, and may be separated, after filtration, by neutralizing the alkaline solution with hydrochloric acid and precipitating by carbonate of ammonia. Peroxide of iron may be separated from lime, strontian, and alkalies by ammonia, and from barytes by precipitating the latter as sulphate. For the means of separating peroxide and protoxide of iron from each other, see a process for the sepa- ration of peroxide of iron and oxide of nickel. * ANALYSIS OF CLAY IRONSTONE. The ordinary constituents of this ore are, carbonic acid, silica, protoxide- and peroxide of iron, alumina, magnesia, lime, and protoxide of manganese. Its analysis may be effected in the following manner : The powdered mineral is dissolved by boiling in aqua regia, with effervescence of carbonic acid and separation of silica, with, perhaps, a little alumina, as an insoluble powder. These are estimated, if necessary, by processes described under the analysis of Carbonates and of Silicates. The solution, which contains all the bases, with the exception of a little alumina, is evaporated to dryness ; the residue is redissolved in dilute hydrochloric acid, and the solution filtered. Unless * See also Note to the mode of estimating phosphoric Ttcid in soils. 156 QUANTITATIVE ANALYSIS. the solution is very acid, muriate of ammonia is now added, and afterwards an excess of caustic ammonia, which preci- pitates peroxide of iron and alumina, with small quantities of protoxide of manganese and magnesia. From the solution filtered from this precipitate, lime is to be precipitated, as oxalate, by oxalate of ammonia or oxalic acid. The pre- cipitate by ammonia being filtered and washed, is dis- solved in a small quantity of hydrochloric acid. This solution is then boiled with excess of caustic potash, to re- dissolve the alumina at first precipitated by potash, and which is estimated by the usual method (page 144.). The portion insoluble in excess of potash is dissolved in pure hydrochloric acid, the solution is carefully neutralized with ammonia, and peroxide of iron precipitated as succinate or benzoate, with the precautions already described. The filtered liquid, containing small quantities of magnesia and protoxide of manganese, is mixed with that filtered from the oxalate of lime, and, from the mixture, manganese is precipitated as sul- phuret by hydrosulphate of ammonia, and magnesia from the filtered solution, as ammonio-phosphate, by phosphate of soda with ammonia (pages 139. and 151.). If the amount of iron only is required, the other consti- tuents of the ore being neglected, add ammonia in excess to the solution in aqua regia ; after washing, boil the precipitate in solution of caustic potash to dissolve alumina, redissolve the peroxide of iron in hydrochloric acid, and precipitate it again by ammonia : dry, ignite, and weigh. IV. NICKEL. Nickel is weighed best in the state of oxide. It is com- pletely precipitated as oxide by potash from a hot solution, whether ammoniacal salts are present or not. The pre- cipitate, after washing with hot water, may be ignited and weighed. Nickel is separated from some substances by precipitation as sulphuret, by hydrosulphate of ammonia in a slightly ammo- niacal solution ; but sulphuret of nickel being slightly soluble NICKEL. 157 in that reagent, its complete precipitation is not easily effected. As small an excess as possible of hydrosulphate of am- monia should be employed; the mixture with that reagent should be diluted and exposed in an open vessel to a very gentle heat, that the excess of the hydrosulphate may be volatilized, or decomposed by the oxygen and carbonic acid of the atmosphere. When the liquid loses its brown colour, the whole of the sulphuret of nickel is precipitated, and may be collected on a filter and washed with water, to which a little hydrosulphate of ammonia has been added. The filter and sulphuret are transferred together to a porcelain capsule, and digested in nitric acid or aqua regia, until the separated sulphur has a pure yellow colour, or is dissolved by the acid ; the liquid is then diluted, thrown on a filter, and oxide of nickel is precipitated by potash. The digestion of the sulphuret in nitric acid should not be prolonged, as or- ganic matter from the filter may be dissolved, which will impede the subsequent precipitation of oxide of nickel by an alkali. Separation of oxide of nickel from peroxide of iron. These oxides may be separated by boiling the carbonate of barytes or carbonate of lime with a solution of the two oxides (chlorides) in hydrochloric acid. The perchloride of iron in solution is decomposed by the earthy carbonate, with precipitation of peroxide of iron, and the formation of a corre- sponding quantity of chloride of barium or chloride of calcium ; while chloride of nickel, on the contrary, is not affected by the earthy carbonate. The latter is added to the solution of the two oxides in hydrochloric acid so long as effervescence of carbonic acid occurs. When a sufficient quantity has been added, the mixture is boiled for a short time, and filtered. The precipitate of peroxide of iron with the excess of earthy carbonate, is then dissolved in hydrochloric acid, and if carbonate of lime has been used ammonia is added to re-precipitate peroxide of iron, avoiding exposure to the air as much as possible, to prevent the absorption of carbonic acid and formation of carbonate of lime. If carbonate of barytes has been used instead of carbonate of lime, barytes 158 QUANTITATIVE ANALYSIS. may first be separated from the solution in hydrochloric acid by sulphuric acid, and peroxide of iron afterwards from the solution filtered from the sulphate of barytes, by ammonia. From the solution filtered from the precipitate of peroxide of iron and excess of earthy carbonate, oxide of nickel may be precipitated by potash, then washed, dried, ignited, and weighed. A similar process is sometimes adopted to separate per- oxide of iron from protoxide of manganese, the two oxides being dissolved in hydrochloric acid. Peroxide of iron is precipitated through decomposition of the perchloride of iron, but protochloride of manganese is not affected by the earthy carbonate. The particular precautions necessary to be adopted in this process are two: 1. That neither sulphuric, phosphoric, arsenic, nor boracic acid is present in the solution; and 2. that all the iron exists in the state of peroxide, as salts of the protoxide of iron are not decomposed by carbonate of lime. On the property possessed by salts of the protoxide of iron to resist decomposition by an earthy carbonate, is founded a process for the separation of peroxide from protoxide of iron, when existing together in solution. When boiled with carbonate of barytes, the salt of the peroxide only is decom- posed; the solution filtered from the precipitated peroxide and the excess of carbonate of barytes is boiled with a little nitric acid, to convert the protoxide of iron into peroxide, which last is precipitated by ammonia after the barytes has been separated by sulphuric acid. Peroxide of iron can also be separated from oxide of nickel by means of an alkaline benzoate or succinate, as protoxide of manganese and peroxide of iron are separated from each other (page 154.). Separation of nickel from manganese. These metals are separated by a difficult process, founded on the reducibility of chloride of nickel to the metallic state, when heated to redness, by hydrogen gas, while chloride of manganese resists decom- position under similar circumstances. The reduction may be COBALT. 159 performed in the apparatus, of which a figure is given at page 160. Chloride of manganese may be separated from the metallic nickel by water. Oxide of nickel is separated from magnesia and the alkalies by hydrosulphate of ammonia, observing the precautions before mentioned in the precipitation of sulphuret of nickel, and from alumina by dissolving the latter in excess of caustic potash. V. COBALT. Cobalt is estimated by precipitating as oxide by caustic potash or soda, and weighing either as oxide or in the me- tallic state. The solution containing cobalt should be heated during precipitation, or before it is filtered, and the precipi- tated oxide washed with hot water and ignited gently. If ammoiiiacal salts are present in the solution, oxide of cobalt is not precipitated by potash or soda until the salts of am- monia are decomposed by the alkali, and all the ammonia is completely expelled by ebullition. But in such a case, it is better to precipitate cobalt as sulphuret by hydrosulphate of ammonia. To perform this, a slight excess of ammonia is first added to the solution, and then a sufficient quantity of hydrosulphate of ammonia to precipitate the cobalt completely. The precipitated sulphuret is collected on a filter and washed with water to which a little of the hydrosulphate has been added. When washed, the sulphuret is decomposed by di- gestion in nitric acid or aqua regia ; the solution thus obtained is diluted, and oxide of cobalt precipitated by potash. The ignition of protoxide of cobalt, previous to weighing, should be performed at a very gentle heat, as it acquires a little oxygen from the air when heated strongly in an open vessel, becoming converted into the peroxide. To avoid this source of error, the protoxide of cobalt may be collected in a weighed filter, placed when washed, together with the filter, in a platinum crucible, and heated in a water-bath so long as it suffers any loss in weight. (See description of the method of collecting, drying, and weighing a precipitate in a weighed filter, page 11.) 160 QUANTITATIVE ANALYSIS. In cases where great accuracy is requisite, instead of weighing cobalt in the state of oxide, it should be weighed in the metallic state, the metal being obtained by reduction of the oxide by heating it strongly in an atmosphere of dry hydrogen gas. This reduction is performed in the appa- ratus depicted in fhe annexed figure: a is a bottle in which hydrogen gas is generated by the action of dilute sulphuric acid on iron or zinc ; as the gas issues from the bottle, it is dried by causing it to pass through the tube Ib, which contains fragments of dry chloride of calcium. Con- nected with this chloride of calcium tube is the tube cc, having a bulb, d, blown in the middle, into which the oxide of cobalt for reduction is introduced. This tube should be of hard German glass. Having ascertained the weight of the tube cc when empty, the whole of the oxide of cobalt may be introduced into the bulb. As it is difficult, however, to introduce the oxide without loss, it is better to take a part only, and to ascertain the proportion which the amount introduced bears to the entire amount by the three following weighings : 1. The entire amount of oxide of cobalt after ignition. 2. The weight of the tube c when empty ; and, 3. The weight of the tube c when the portion of oxide is introduced, from which that of the oxide can be estimated, by deducting the weight of the empty tube. It may be calculated by the simple rule of proportion what quantity of metal is contained in the whole amount COBALT. 161 of oxide, when it has been determined, by experiment, what is contained in the portion operated on. The oxide being introduced, and the apparatus arranged as in the figure, hydrogen gas may be evolved by pouring con- centrated sulphuric acid down the funnel of the bottle a on zinc and water already in the bottle. After the apparatus is filled with hydrogen, the bulb is heated to redness * by a spirit lamp. The cobalt is presently reduced from the state of oxide to that of metal, with evolution of heat and light, and production of water. When the decomposition is complete, and the water is expelled by the application of heat along the tube, a current of the gas being still maintained, the bulb is allowed to cool, the tube is disconnected from the apparatus, gently inclined to allow the hydrogen to escape, and then weighed. The weight of the tube when empty being known, the increase is the quantity of metallic cobalt, from which the corresponding quantity of pure protoxide may be obtained by calculation. Separation of oxide of cobalt from oxide of nickel. These oxides are not separated from each other without some diffi- culty. Mr. Phillips has described a process for the purpose, founded on the precipitability of oxide of nickel by potash from an ammoniacal solution of that oxide, while oxide of cobalt remains dissolved. To the solution of these oxides in an acid, a sufficient quantity of muriate of ammonia is added to prevent precipitation by caustic ammonia, which is then applied in excess. The mixture is largely diluted with water, which has been recently boiled to expel atmo- spheric air, and oxide of nickel is precipitated by potash, which alkali is added in sufficient quantity to decompose the ammoniacal salt. The precipitation should be performed in a vessel which can afterwards be closed air-tight. When the oxide of nickel has subsided, it may be collected on a filter, washed with hot water, and weighed ; and the oxide of cobalt contained in the filtered solution, may be precipitated * Oxide of cobalt may be reduced at a temperature much below redness, but the metallic cobalt thus obtained acts as a pyrophorus when it comes in con- tact with the air. It does not, however, if heated more strongly. M 162 QUANTITATIVE ANALYSIS. by hydrosulphate of ammonia. The necessity of avoiding exposure to the air, and of diluting the solution with water deprived of air, arises from the property which the ammo- niacal solution of oxide of cobalt possesses, especially if strong, of absorbing oxygen, the protoxide of cobalt being converted into peroxide, which is insoluble in ammonia. Professor Liebig has described very recently a method of separating cobalt and nickel by means of cyanide of potas- sium. When cyanide of potassium is added to a solution of a salt of nickel, a precipitate of cyanide of nickel is formed, which re-dissolves on adding an excess of cyanide of potas- sium, a double compound of cyanide of nickel and cyanide of potassium being formed, which is soluble in water. This double cyanide is decomposed by dilute sulphuric acid, with precipitation of cyanide of nickel. When cyanide of potassium and an excess of hydrocyanic acid are added to any salt of cobalt, and the mixture warmed, there is formed the soluble double compound of percyanide of cobalt and cyanide of potassium (cobalti-cyanide of po- tassium), whose aqueous solution is not at all affected by boiling with hydrochloric, sulphuric, or nitric acid (Gmelin), unlike the analogous compound of cyanide of nickel and cya- nide of potassium. On adding hydrochloric acid in excess to a solution con- taining nickel and cobalt, with a little free acid, and gently heating, there are formed the double cyanide of nickel and potassium, and the double cyanide of cobalt and potassium. If to this mixture dilute sulphuric acid be added in the cold, three cases present themselves, according to the relative pro- portions which exist of nickel and cobalt. (1.) " If the cobalt and nickel in solution are in the pro- portion of two equivalents of cobalt to three equivalents of nickel (quantities which correspond to their atomic pro- portions in the cobalti-cyanide of potassium), the precipitate produced is cobalti-cyanide of nickel, and is of a bluish-white colour. (2.) " If the solution contains less nickel than cor- responds to the above proportions " (as would be the case in COBALT. 163 the analysis of an ore of cobalt), " there remains in solu- tion a certain quantity of cobalti-cyanide of potassium, and the precipitate is still cobalti-cyanide of nickel. (3.) " If there is more nickel present in the solution " (as in the analysis of an ore of nickel), " the precipitate is a mixture of cyanide of nickel and cobalti-cyanide of nickel."* In the first and second cases, the precipitate produced by dilute sulphuric acid is boiled with the acid fluid until not a trace of hydrocyanic acid is observed to escape (or it may be evaporated nearly to dry ness in a water-bath), and then slightly wanned with an excess of carbonate of potash or caustic potash ; the cobalti-cyanide of nickel is thereby de- composed into pure oxide of nickel, or the carbonate, on the one hand, which is collected on a filter, washed, dried, and weighed; and an alkaline liquid, on the other hand, which contains the whole of the cobalt. The latter is eva- porated to dryness, some nitrate of potash being added to it, and the residue ignited. On being afterwards treated with water, the oxide of cobalt remains behind, which may be weighed in the usual manner. In the third case, which is that of the analysis of nickel ores, having the quantity of cobalt relatively small, it is necessary to add a considerable excess of muriatic acid to precipitate the cyanides dissolved in the cyanide of potas- sium, and to keep the mixture boiling at least one hour. The precipitate, as first produced, is a mixture of cyanide of nickel and cobalti-cyanide of nickel, but by boiling in hydrochloric acid, the cyanide of nickel is decomposed into chloride of nickel and hydrocyanic acid, the latter being ex- pelled by the ebullition. The cobalti-cyanide of nickel is not affected by the acid. After the acid liquid has been boiled a considerable time, potash is added to the solution, as in the other cases. Chloride of nickel is then decomposed with precipitation of oxide, which falls completely if all the hydro- cyanic acid has been expelled by ebullition previous to the addition of the alkali, but not otherwise, as cyanide of potassium would be formed, which is capable of retaining * Philosophical Magazine, third series, vol. xx. p. 269. M 2 164 QUANTITATIVE ANALYSIS. nickel in solution. The cobalti-cyanide of nickel, insoluble in hydrochloric acid, is also decomposed by the potash, with formation of nickel and an alkaline solution of cobalt.. The oxide of nickel is collected on a filter, washed, dried, and weighed ; and oxide of cobalt is obtained from the alkaline solution, in the manner above described. Peroxide of iron may be separated from oxide of cobalt by processes similar to those by which the same oxide is separated from oxide of nickel. The best method is by boiling carbonate of barytes in the solution of the two oxides in hydrochloric acid, when peroxide of iron is precipi- tated through decomposition of the perchloride of iron, while the chloride of cobalt is not affected by the earthy carbonate. The peroxide of iron may be dissolved out from the excess of carbonate of barytes by sulphuric acid, and again pre- cipitated by ammonia. From the solution filtered from the peroxide of iron and excess of carbonate of barytes, oxide of cobalt may be precipitated as sulphuret by hydrosulphate of ammonia. Oxide of cobalt is separated from alumina by potash, and from magnesia, lime, and the alkalies by hydrosulphate of ammonia. VI. ZINC. Carbonate of potash and hydrosulphate of ammonia are the reagents generally employed to precipitate zinc from its solutions, for the purpose of weighing ; the zinc, in both cases, being afterwards weighed in the state of oxide. When carbonate of potash is used, a sufficient quantity should be added to decompose all ammoniacal salts which may be present, as these prevent the precipitation of subcarbonate of zinc. The mixture should be evaporated to dryness, the residue treated with hot water, boiled and filtered. If no ammoniacal salts are present, the solution may simply be boiled with carbonate of potash and filtered. On ignition, the subcarbonate of zinc loses carbonic acid, and becomes converted into the oxide. In the form of sulphuret, zinc may be separated from ZINC. 165 many other bodies by means of hydrosulphate of ammonia, in a neutral or ammoniacal solution. If the solution to be pre- cipitated is acid, caustic ammonia should be added in sufficient quantity to re-dissolve the precipitate of oxide at first oc- casioned by the alkali. When hydrosulphate of ammonia has been applied, the sulphuret of zinc which is formed is allowed to subside before filtration, as it would pass through the pores of the paper if the solution were filtered imme- diately on being precipitated. When the sulphuret has sub- sided, it may be transferred to the filter (the supernatant liquid being previously decanted), and washed with water to which a little hydrosulphate of ammonia has been added. When washed, it is dissolved in hydrochloric acid without removal from the filter, the solution is boiled to expel sul- phuretted hydrogen, filtered, and precipitated by carbonate of potash. A precaution which should never be neglected in esti- mating zinc when precipitated as subcarbonate, is to examine whether the liquid filtered from the precipitate contains any more zinc in solution. This is best ascertained by hydrosulphate of ammonia ; and the precipitated sulphuret, if any is formed, is converted into subcarbonate, and added to the quantity previously obtained. Separation of oxide of zinc from oxides of cobalt, nickel, and manganese. When the quantity of material operated on is very small (a few grains only), the oxides of cobalt, nickel, and manganese may be separated from oxide of zinc by means of caustic potash. On adding this alkali in excess to a solution of the oxides mentioned in hydrochloric acid, free from ammoniacal salts, oxide of zinc alone is re-dissolved by the excess, and it may be obtained from the alkaline liquid by neutralizing the latter with an acid, and precipitating by carbonate of potash. Although this method may be prac- tised when the weight of the material does not exceed a few grains, it is impossible to separate the oxide of zinc from the other oxides completely by any excess of potash. When a large amount of the oxides is operated on, another method is therefore had recourse to. In that generally followed, M 3 166 QUANTITATIVE ANALYSIS. the chlorides of the metals are formed, and chloride of zinc, which is comparatively volatile, is separated from the other chlorides by distillation. The first part of the process consists in ascertaining the united weights of the oxides by precipitation from their solu- tion by carbonate of potash, and conversion into oxides by ignition. This being done, the whole or a known proportion of the oxides is then introduced into a bulb blown on a tube of hard German glass, d of the annexed figure, and weighed Fig. 20. with the tube. The apparatus being arranged as shown in the figure, the mixed oxides are exposed to the action of a stream of dry hydrochloric acid gas, at the highest tem- perature procurable by the Rose spirit lamp. The hydro- chloric acid is evolved in the retort #, from a mixture of chloride of sodium and sulphuric acid ; the gas is passed first into the bottle b, which contains a little concentrated sulphuric acid, next through the chloride of calcium tube c, and then over the oxides in the bulb. By being heated in hydrochloric acid gas, the oxides become converted into chlorides, and the volatile chloride of zinc distils. The exit limb of the bulb tube is bent at right angles about its middle, to enable it to dip into the receiver/, which contains water rendered slightly alkaline by ammonia, the surface of the liquid being im- mediately below the extremity of the tube. When no more chloride of zinc distils, the transmission of hydrochloric acid may be discontinued, the whole of the chloride of zinc which has condensed in the tube near the bulb being previously driven farther on by the application of heat. The part of the tube which contains this portion of chloride of zinc is divided from the bulb by means of a triangular file, and the ZINC. 167 chloride is washed into the receiver, the solution in which should now contain all the zinc. From this solution, zinc may be precipitated by carbonate of potash, and weighed as oxide after being ignited. The bulb containing the fixed chlorides is weighed, together with the divided portion of the tube, made clean and dry. The difference between the weight of the tube when empty and its present weight is the weight of the fixed chlorides in the bulb. After weighing, the bulb may be washed out with water, and the cobalt, nickel, and manganese estimated by the ordinary methods. Although preferable to the process of separating zinc from cobalt, nickel, and manganese in which potash is used, the preceding method still does not afford accurate results. The last portions of chloride of zinc which distil over contain traces of chloride of nickel, and notwithstanding this, the chlorides remaining in the bulb are not entirely free from zinc. Berzelius has recently made known the following process for separating zinc from cobalt and nickel, which he considers to be very exact. Having first separated the greater part of the oxide of zinc by boiling the mixed oxicjes in caustic potash, the residue is washed, first with cold and afterwards with boiling water, to withdraw the last traces of potash, ignited, and weighed. After weighing, the oxides are inti- mately mixed with finely pulverized sugar, which does not leave an incombustible ash on calcination, the mixture is introduced into a porcelain crucible, and the sugar is car- bonized by careful heating. The porcelain crucible is covered, and placed in caustic magnesia contained in an earthen cruci- ble, also covered, and exposed for one hour to the highest tem- perature of a wind furnace. By this operation the oxides are reduced, zinc is volatilized and completely expelled, and nickel and cobalt remain in the crucible in the state of carburets. The residue in the crucible is treated with nitric acid, to dissolve nickel and cobalt, and the solution is evaporated to dryness in a weighed platinum crucible : on heating the residue to redness, metallic oxides remain, from whose weight that of M 4 168 QUANTITATIVE ANALYSIS. the oxide of zinc is calculated, being the loss upon the for- mer weight.* Another method has been proposed by M. Ullgren. The mixed oxides of zinc, cobalt, and nickel are weighed, and heated to dull redness in an atmosphere of hydrogen, in the apparatus of which a figure is given, page 160. The oxides of cobalt and nickel are reduced to the metallic state by hydrogen with formation of water, but oxide of zinc remains unaltered. When no more water is produced, the bulb is allowed to cool, the current of hydrogen being still main- tained. One of the extremities of the tube c is then sealed before the blowpipe, and the tube is filled with a solution of carbonate of ammonia : the open end is corked, and the tube is kept for twenty-four hours at a temperature near 100 Fahr. The oxide of zinc not reduced by the hydrogen dissolves completely in the carbonate of ammonia, from which it may be obtained by evaporating carefully to dryness and igniting the residue ; the oxide of zinc remain- ing is then weighed. The cobalt and nickel are dissolved in nitric acid, and determined by a process already de- scribed, page 161. Separation of oxide of zinc from peroxide of iron. On adding an excess of ammonia to the solution containing these ox-ides, the oxide of zinc at first precipitated is re-dissolved, and a residue of peroxide of iron left. When the proportion of zinc is small compared with that of iron, this process may succeed ; but if the zinc is in considerable quantity, no ex- cess of ammonia will separate it completely from the per- oxide of iron. In such a case, these oxides are best separated by means of an alkaline benzoate or succinate. (page 154.) Oxide of zinc may be separated from lime by means of oxalate of ammonia in a strongly ammoniacal solution, and from magnesia and the fixed alkalies by hydrosulphate of ammonia, (page 164.) * Rapport Annuel of Berzelius, 1841. (2 Ann4e, p. 81. Paris edition.) URANIUM. 169 VII. URANIUM. Both the peroxide and protoxide of uranium are precipi- tated from their solutions by ammonia ; but as the peroxide is precipitated more completely than the protoxide, it is con- venient, when the latter is to be estimated, to convert it into peroxide (uranic acid) by heating with a little nitric acid, before precipitation by ammonia. If the precipitated per- oxide, when collected~on a filter, is washed with pure water, it passes through the pores of the paper, making the filtered liquid milky : this inconvenience may be avoided, to a great extent, by employ ing, a dilute solution of muriate of ammonia, instead of water, in the washing. The precipitate, besides peroxide of uranium, contains ammonia and w^ater, both of which, together with oxygen, it loses on calcination, being converted into the protoxide of uranium, from the weight of which, that of the peroxide, if the latter was originally present, may be calculated. If the solution, besides peroxide of uranium, contains a considerable quantity of an earth, the latter is also precipi- tated by ammonia, in a state of combination with the peroxide, although the earth is one which is not precipitated by am- monia under ordinary circumstances, as lime, magnesia, and barytes. In such a case, when the precipitate is washed, it should be dissolved in hydrochloric acid, and the peroxide re- precipitated by H ammonia ; the precipitate which then falls is free from the earth, and may be washed with a solution of muriate of ammonia, dried, and calcined. By calcination it becomes the protoxide of uranium, which may be weighed. Separation of uranium from iron. On adding a large excess of carbonate of ammonia to a solution of the peroxides of these metals, the peroxide of uranium at first precipitated is re-dissolved in the excess, while peroxide of iron remains precipitated. If uranium and iron exist in solution as prot- oxides, the latter should be converted into peroxides by heating the solution with nitric acid before carbonate of am- monia is added. 170 QUANTITATIVE ANALYSIS. VIII. CHKOMIUM. Estimation of chromic acid and oxide of chromium. Chro- mic acid may be estimated by precipitation from a neutral solution by protonitrate of mercury ; chromate of the sub- oxide of mercury is precipitated, which may be transferred to a filter and washed. When heated to redness, the pre- cipitate leaves green oxide of chromium, from the weight of which that of the chromic acid may be calculated. Accord- ing to Rose, chromic acid may be determined more accu- rately by reducing it to the state of oxide or chloride, by boiling with hydrochloric acid, by which chlorine is liberated. The reduction of chromic acid may be accelerated by the cautious addition of alcohol, any excess of which should afterwards be expelled by evaporation. Oxide of chromium is best precipitated by ammonia: the liquid is heated gently, filtered, and the precipitate ignited in a covered platinum crucible. By ignition, oxide of chromium is rendered insoluble in acids. Separation of oxide of chromium from oxide of iron. Analy- sis of Chrome-iron. The mineral being reduced to a fine powder, is mixed with about three times its weight of fused potash, with the addition of a small quantity of nitre, and the mixture is exposed to a dull red heat in a silver cruci- ble. In this operation the oxide of chromium absorbs oxygen, and becomes converted into chromic acid, which unites with the alkali to form chromate of potash. The fused mass is digested in water until every thing soluble in that liquid is taken up, the alkaline solution is filtered, neu- tralized with acetic acid, and carbonate of ammonia is added to precipitate alumina. After the alumina has been gently ignited and weighed, it should be dissolved in hydrochloric acid to separate a little adhering silica, which remains undis- solved, and may be collected on a filter, ignited, and weighed. The weight of the silica should be deducted from that of the alumina and silica previously weighed together. The chromic acid existing in the liquid filtered from the precipitate of TITANIUM. 171 alumina is reduced to the state of oxide of chromium by boiling with hydrochloric acid, the oxide is precipitated by ammonia, filtered, dried, ignited, and weighed. The portion of the fused mass insoluble in water, consisting principally of peroxide of iron, is dissolved in hydrochloric acid, the solution is neutralized with ammonia, and peroxide of iron is precipitated by an alkaline succinate or benzoate in the manner described page 154. If magnesia exists in the ore, as sometimes occurs, it is contained in the liquid filtered from the succinate or benzoate of iron, from which it may be precipitated either as carbonate or ammonio-phosphate. (page 139.) IX. TITANIUM. Estimation of titanic acid. This acid is best precipitated from its solution by ammonia, as small an excess as possible of that alkali being employed. It is recommended to expel the excess of ammonia, after precipitation, by exposing the liquid to a gentle heat in an open vessel. The precipitate is then transferred to a filter, dried, and ignited. Titanic acid should be weighed soon after ignition, as it rapidly absorbs moisture from the atmosphere. Titanic acid may be separated from oxides of iron, co- balt, zinc, and manganese, in the following manner. A considerable quantity of tartaric acid is first added to the solution, which is then supersaturated with caustic ammonia ; if sufficient tartaric acid has been previously added, no titanic acid is precipitated by the ammonia. By adding a slight excess of hydrosulphate of ammonia, all the other metals are completely precipitated as sulphurets, leaving in solution titanic acid, which may be obtained by filtering, evaporating the liquid to dryness, and igniting the residue in a platinum crucible until all combustible matter is destroyed. The residue is pure titanic acid, which may be weighed. From lime and the fixed alkalies titanic acid may be separated by ammonia. 172 QUANTITATIVE ANALYSIS. SECTION IV. METALS PROPER, PRECIPITATED BY SULPHURETTED HYDROGEN FROM THEIR ACID SOLUTIONS. I. LEAD. Estimation of lead, and modes of separating it from other metals. Lead may be completely separated from all the metals which have been previously considered, by precipitation as sulphuret, by sulphuretted hydrogen, from a slightly acid solution. If the solution to be precipitated is not already acid, it is best rendered so by the addition of nitric acid, as hydrochloric, which is the acid generally used for that purpose with other metals, might cause a precipitate of chloride of lead, which is not decomposed by sulphuretted hydrogen with the same facility as a salt of lead in solution. When the proportion of lead is very small, solution of sulphuretted hydrogen in water may be employed ; but in other cases it is better to pass the gas through the solution contained in a bottle, having a conical glass stopper. In all cases of pre- cipitation by sulphuretted hydrogen, the gas must be trans- mitted, or the solution of the gas in water added, until the odour of the reagent becomes very perceptible, and con- traction no longer occurs from absorption of the gas above the liquid, when the bottle containing it is agitated. Lead cannot be weighed in the form of the precipitated sulphuret, as it is almost impossible to obtain the latter without an admixture of free sulphur, and it cannot be ignited without decomposition. To obtain the lead in a fit state for weighing, it is necessary to convert the sulphuret into the sulphate of oxide of lead, which is done in the following manner. After being collected on a small filter * and washed, the sulphuret * It sometimes happens that the sulphuret adheres so strongly to the sides of the vessel in which the precipitation was performed, and also to the tube through which the gas was passed, that it cannot be removed by washing out with water. When this is the case, having first washed out all the adhering liquid, the sul- phuret of lead may be readily detached by pouring a few drops of fuming nitric acid on it. Sulphate of lead is then formed, which may be washed out .nd added to the other quantity of sulphate. LEAD. 173 is dried and digested, while still in the filter, in a little fuming nitric acid, contained in a weighed platinum capsule, the acid being added very gradually to avoid loss by pro- jection. By this, the sulphuret is entirely converted into sulphate, if the nitric acid is sufficiently strong ; if such does not appear to be the case, from some sulphur having been set free without conversion into sulphuric acid, all the nitrate of lead then formed may be converted into sulphate by the addition of a few drops of sulphuric acid. The excess of acid is driven off by evaporation, and the capsule heated to bright redness to destroy the organic matter of the filter. In an open capsule the organic matter burns completely, without any reduction of the sulphate, which is then in a proper state for weighing. From the weight of the sulphate, that of metallic lead or oxide of lead originally present may be cal- culated. Lead is also precipitated from its solutions, for estimation, as oxalate, by oxalate of ammonia, as sulphate by sulphuric acid, and as subcarbonate by carbonate of ammonia. In general, Rose gives the preference to oxalate of ammonia. As in the precipitation of oxalate of lime, it is necessary that the solution be neutral, or slightly ammoniacal. After wash- ing and drying the precipitated oxalate, it is removed from the filter, which is burned separately, to avoid any reduc- tion of the oxide by the combustible matter. The ash of the filter is added to the oxalate of lead, which is converted into oxide by ignition, and then weighed. As caustic potash dissolves oxide of lead, it is sometimes used as the means of separating this oxide from those which are insoluble in potash ; after being filtered, the alkaline so- lution should be saturated with acetic acid, and the lead pre- cipitated by oxalate of ammonia. From the facility with which compounds of lead are re- duced to the metallic state, and an alloy with platinum formed, it is necessary to use porcelain crucibles in the igni- tion of all lead compounds, except the sulphate. Analysis of chromate of lead. The chromate being re- duced to a fine powder, it is boiled in hydrochloric acid, with 174 QUANTITATIVE ANALYSIS. the addition of a little alcohol, when it becomes converted into chloride of chromium, soluble, and chloride of lead, in- soluble, in alcohol. The chloride of lead is collected on a weighed filter, washed with dilute alcohol, dried and weighed with the filter, as described at page 11. The alcohol being expelled from the filtered liquid by evaporation, the oxide of chromium is precipitated by ammonia. (See p. 170.) II. CADMIUM. Cadmium may be separated from all those metals which have been previously considered, with the exception of lead, by passing sulphuretted hydrogen gas into the solution pre- viously rendered slightly acid. If the solution is neutral, hydrochloric acid may be employed to acidify it. The pre- cipitated sulphuret is collected on a filter, washed, and trans- ferred, while still moist and in the filter, into a capsule, and digested with the assistance of a gentle heat in concentrated hydrochloric acid. The sulphuret is thereby decomposed, being converted into chloride of cadmium, with evolution of sulphuretted hydrogen. When the conversion of sulphu- ret of cadmium into chloride is completed, and the odour of sulphuretted hydrogen no longer perceptible, the solution may be filtered, and cadmium precipitated from it as car- bonate, by carbonate of potash. The carbonate is collected on a filter, washed, dried, and heated to redness before being weighed. Ignition reduces carbonate of cadmium to the state of oxide. The filter must be burned separately, as described at page 8., when compounds of lead and some other metals are ignited, oxide of cadmium being easily reduced to the me- tallic state, and the metal lost from its volatility. The insolubility of oxide of cadmium in caustic potash affords the means of separating it from oxide of lead. (See p. 173.) III. BISMUTH. By means of sulphuretted hydrogen, bismuth may be BISMUTH. 175 separated from all bodies which are not precipitated by that reagent from their solutions in acids. The liquid through which sulphuretted hydrogen is passed should be diluted ; but as the dilution of bismuth solutions by pure water causes the precipitation of subsalts of bismuth, a quantity of acetic acid should be added before dilution. The acetic acid completely prevents precipitation when the liquid is diluted. The solution may now be saturated with sulphuretted hydrogen, and the precipitated sulphuret be filtered, washed, and decomposed by digestion in strong nitric acid : the nitrate of bismuth is thus formed, and sulphur is set free. When the separated sulphur has a pure yellow colour, the sulphuret of bismuth is completely decomposed; the solution is then diluted (a sufficient quantity of free acid being present to prevent pre- cipitation), filtered, the sulphur is washed with water slightly acidified by nitric acid, and oxide of bismuth is precipitated from the solution by carbonate of ammonia. Bismuth is completely precipitated from its solutions by carbonate of ammonia after standing for a few hours in an open vessel, although on first adding an excess of the carbonate, a con- siderable quantity of the precipitate may be re-dissolved. The precipitated subcarbonate is ignited in a porcelain crucible, and weighed as oxide. The filter should be burned separately to avoid reduction of the oxide. When bismuth is to be separated from substances which are not precipitated by carbonate of ammonia, it is unneces- sary to convert that metal into the sulphuret, but it may be precipitated at once by the alkaline carbonate. It is abso- lutely necessary, however, that the solution be free from hydrochloric acid, otherwise the precipitate of subcarbonate will be mixed with an oxichloride of bismuth, which resists decomposition by any excess whatever of the alkaline car- bonate. In such a case, the sulphuret of bismuth should first be precipitated by sulphuretted hydrogen, and that decomposed by nitric acid in the manner above described. Separation of bismuth from lead. According to Liebig, bismuth may be separated from lead by carbonate of lime, which throws down bismuth, but not lead, from a cold solu- 176 QUANTITATIVE ANALYSIS. tion of the nitrates of these metals. They may also be sepa- rated by adding to the solution in nitric acid, which must be free from hydrochloric acid, a sufficient excess of caustic potash to re-dissolve the oxide of lead at first precipitated. (See p. 173.) Oxide of bismuth remains undissolved. Another method of separating bismuth from lead has been recently described by M. Ullgren. Carbonate of ammonia is added to the solution of the two metals to precipitate the subcarbonates, which are filtered and dissolved in acetic acid. A piece of perfectly pure lead is immersed in the solution, and the containing vessel is carefully closed and set aside for several hours. Metallic bismuth is then precipitated on the lead, an equivalent quantity of the latter being dis- solved. When the precipitation is complete, the plate of lead may be separated from the bismuth, washed, dried, and weighed. The precipitated bismuth is collected on a filter, washed with distilled water which has been previously boiled to expel air, and dissolved in nitric acid. The solution of nitrate of bismuth is evaporated to dryness and calcined ; pure oxide then remains, which may be weighed. To the solution of lead filtered from the precipitated bismuth, carbonate of ammonia is added, and the precipitate is washed, calcined, and weighed as oxide of lead. By the loss in weight sus- tained by the metallic lead employed in the process, it is known how much oxide of lead must be deducted from the preceding quantity.* IY. COPPER. Copper is precipitated from its solutions, for the purpose of estimation, in the state of oxide by caustic potash : the liquid should be rather dilute when precipitated, and always boiled before filtration. In the cold, potash in excess produces with salts of copper a bulky blue precipitate of hydrate of the protoxide, which, on the application of heat, becomes dark brown or black, very dense, and anhydrous. The precipitate is washed on a filter with hot water, dried, * Berzelius, Rapport Annuel, 1841. (2 ? Annee, p. 83. Paris edition.) COPPER. 177 and ignited with the usual precautions to prevent reduction by the combustible matter of the filter. Should a portion of oxide, however, be reduced to the metallic state, it is gene- rally re-converted into oxide by the current of air produced by the combustion of the filter. A portion of the precipi- tated oxide of copper adheres so tenaciously to the bottom of the vessel in which it was boiled, as not to be removed to the filter without some pains by mechanical means : this, how- ever, is easily effected by dissolving it in a few drops of hydrochloric acid, and re-precipitating the oxide by potash. Copper is also precipitated for estimation in the metallic state by means of a piece of polished iron ; but this method is not admissible in a delicate analysis. As sulphuret, copper may be separated from many other metals by passing sulphuretted hydrogen gas through the acid solution. The precipitate should be filtered * rapidly, and washed with water impregnated with sulphuretted hydrogen; for when washed with pure water, or water holding air in solution, it oxidizes and dissolves to a small extent, and therefore passes through the filter. When clean, the sulphuret is transferred to a capsule, while still moist and on the filter, and digested in nitric acid or aqua regia, until entirely decomposed ; the separ- ated sulphur, if any remains -undissolved by the acid, having a pure yellow colour. The solution is diluted, filtered, and decomposed by potash. By the prolonged action of nitric acid or aqua regia on the filter, a soluble organic compound is produced, whose presence impedes the precipitation of oxide of copper by potash, some of the oxide still remaining in solution after an excess of potash has been added. When this is the case (it is perceived by the blue colour of the solution), the portion of oxide must be again precipitated by sulphuretted hydrogen (as the action of this reagent is not affected by the organic matter), and the precipitated sulphuret treated as before. To avoid the inconvenience from the presence of organic matter, the conversion of the sulphuret * The sulphuret which adheres tenaciously to the bottle in which the preci- pitation was effected, and to the tube which conducted the gas, may be detached by means of nitric acid. (See note, page 172.) m 178 QUANTITATIVE ANALYSIS. into nitrate should be effected as expeditiously as possible ; or when the greater part of the sulphuret has been disengaged from the filter by the acid, the filter may be removed into another capsule, and the sulphuret remaining attached to it decomposed by a somewhat weaker acid. Separation of copper from lead. M. Rose recommends the following method for the separation of these metals. Sul- phuric acid is added to their solution in nitric acid, and the mixture is evaporated to dryness, so that the excess of acid may be expelled. On treating the dry mass with water, sulphate of lead remains undissolved, which is collected on a filter, dried, ignited, and weighed. From the filtered solution oxide of copper is precipitated by potash. The alkaline so- lution filtered from the oxide of copper contains a trace of oxide of lead, dissolved at first by the water as sulphate, and retained in solution afterwards by the excess of potash. In a delicate analysis, this small portion of oxide is worth separating, which is done by first neutralizing nearly the whole excess of alkali by an acid, and then precipitating oxide of lead as oxalate, by oxalate of ammonia. Separation of copper from zinc: analysis of brass. Brass may be analyzed in the following manner. The alloy is dis- solved in a small quantity of nitric acid, the solution is di- luted and treated with sulphuretted hydrogen to precipi- tate sulphuret of copper, which is converted into oxide and weighed. From the liquid filtered from the sulphuret of copper, oxide of zinc may be precipitated, after boiling, by carbonate of potash, (page 164.) Estimation of the suboxide of copper. When it is required to estimate the suboxide of copper, the substance to be ana- lyzed should be dissolved in nitric acid. That acid yields oxygen to the suboxide, converting it into the protoxide, which may be precipitated by potash in the usual manner. The amount of suboxide is then calculated from the weight obtained of protoxide. SILVER. 179 V. SILVER. Silver may be precipitated from its solutions, and weighed with great exactness, in the state of chloride, in which form it may be separated from all metals whose chlorides are so- luble in water, or in dilute nitric or hydrochloric acids. The only metals from which silver cannot easily be separated in this form, are, mercury, when existing as suboxide, and lead, when the latter is in a very large proportion compared with the silver. Either hydrochloric acid, or the chlorides of po- tassium, sodium, or ammonium, may be employed to preci- pitate the chloride of silver, but preference is given to the first, as alkaline chlorides exert a slight solvent action on the precipitate. Before precipitation, the solution should be slightly acidified by nitric acid, and boiled for a few minutes previous to filtering, as the liquid would pass through the filter in a milky state without these precautions. Since chloride of silver is partially reduced by the agency of light, it is proper to keep it in the dark while being filtered : it answers very well to cover the funnel with a piece of blackened paper. When washed and dried, the chloride of silver is removed from the filter into a weighed porcelain crucible, and heated until fusion takes place. The filter is burned separately, its ash being received and calcined in a weighed platinum crucible cover, which is used, after calcining the ash, as the cover for* the crucible in which the chloride is fused. From the weight of the chloride of silver, that of the metal or oxide may be calculated. After weighing, the fused chloride can be easily detached from the crucible, unless the inside of the latter is rough, by digestion in water for a day or two. Estimation of silver in alloys. An alloy of silver with other metals, may be analyzed by dissolving it in nitric acid, and precipitating chloride of silver from the solution by hydrochloric acid ; or, if the alloy is insoluble in nitric acid, it may be treated with aqua regia, when chloride of silver is formed, while the other metals dissolve. A trace of chloride of silver may be dissolved by the acid, but it is precipitated N 2 180 QUANTITATIVE ANALYSIS. on dilution with water. The chloride of silver may be collected and weighed in the usual manner. From the solution filtered from the chloride of silver, the other metals may be obtained by processes described elsewhere. Alloys of silver with metals which are oxidable by the air at high temperatures, such as lead and copper, are analyzed or assayed by the process of cupellation, which is briefly the follow- ing. It consists in fusing the alloy, mixed with a considerable quantity of pure metallic lead, in a small crucible formed of bone earth, which is called the cupel. By placing the cupel near an opening in the side of a furnace, air is caused to pass over the surface of the fused metals, which, with the excep- tion of silver, become oxidized ; the oxides formed, although by themselves infusible, dissolve in the fused oxide of lead, and are absorbed with it by the porous cupel, leaving a button of pure metallic silver, which may afterwards be de- tached and weighed. This operation is practised to a great extent where several such assays are required at the same time ; but for a single analysis it is more convenient and ac- curate to dissolve the alloy in nitric acid, and precipitate the silver as chloride by dilute hydrochloric acid. When lead is present, the solution to be precipitated should be dilute, and the precipitate washed with hot water, for chloride of lead, though soluble in hot, is very slightly so in cold water, and might therefore be precipitated with the chloride of silver on adding hydrochloric acid. By long washing, however, with boiling water, the chloride of lead may be entirely removed. VI. GOLD. Gold may be separated from most other metals by pre- cipitation in the metallic state from its solution by deoxidizing agents, such as protosulphate or protochloride of iron, proto- nitrate of mercury, oxalic or sulphurous acids. Protosulphate and protochloride of iron are the precipitants generally em- ployed. On adding protochloride of iron to a solution of the chloride of gold, the latter yields all its chlorine to the former, which becomes perchloride of iron, metallic gold GOLD. 181 being precipitated. A corresponding change occurs when the protosulphate of iron is employed. If a large excess of nitric acid is present in the liquid to be precipitated, it should be expelled by evaporation to dryness before adding the salt of iron ; the residue is moistened with a little hydrochloric acid, to render it soluble in water, and also to prevent the subsequent deposition of a subsalt of the peroxide of iron on exposure to the air. Protosulphate or protochloride of iron being added, the mixture should be digested for a few hours at a gentle heat, and the precipitated gold filtered, ignited, and weighed. When gold is to be separated from metals, the admixture of iron with which would render their determination diffi- cult, oxalic acid, or an oxalate, is employed, instead of the proto-salt of iron. The excess of nitric acid in the liquid is first expelled by evaporation, and hydrochloric acid is added to redissolve the residue, and also to prevent the formation of any insoluble oxalates of the metals from which the gold is to be separated. The addition of oxalic acid causes an effer- vescence of carbonic acid and precipitation of metallic gold ; the gold is not, however, completely precipitated without digestion at a moderate heat for a day or two. The analysis of an alloy of gold and silver is one of frequent occurrence. If the proportion of silver is small compared with that of gold, a mass of the alloy is laminated, weighed, and heated with aqua regia, which dissolves the gold completely, leaving a residue of chloride of silver of the form of the alloy operated on. The chloride of silver should be disinte- grated with a glass rod, the liquid considerably diluted with water (on which a little more chloride of silver is generally deposited), boiled, filtered, and the chloride of silver weighed. The excess of acid in the filtered liquid being expelled by evaporating to dryness, the residue is diluted, and gold is precipitated from it by oxalic acid. The solution filtered from the gold may then be examined for copper and iron. If, however, the proportion of silver is large compared with that of gold, the coating of chloride of silver formed by the action of the aqua regia may effectually protect the N 3 182 QUANTITATIVE ANALYSIS. interior of the alloy from further action, although the latter is reduced to a very thin plate. In such a case, the alloy should be decomposed by boiling in strong nitric acid, the silver being then dissolved, and the gold separated in the form of a powder, which may be collected and weighed. If the alloy resists the action of hot nitric acid, it should be fused with thrice its weight of pure lead, such as that obtained by heating the acetate of lead to redness in a porcelain crucible over a spirit lamp. The lead alloy is readily acted on by nitric acid, all the metals being dissolved except gold, which remains in a pure state, and may be collected on a filter. From the filtered liquid, silver may be precipitated as chloride by a hot solution of chloride of lead. VII. MERCURY. The most convenient form in which mercury can be obtained fit for weighing, is in that of the metal itself, in which state it may also be separated from many other metals. The re- ducing agent generally employed to obtain metallic mercury from its compounds is protochloride of tin ; the reduction of all mercurial salts, whether soluble or insoluble in water and acids (with the single exception of the sulphuret), being easily effected by means of that reagent. When, however, the admixture of tin with the bodies from which mercury is to be separated would render the subsequent determination of the latter difficult, instead of the pro- tochloride of tin, phosphorous acid is employed, which be- comes phosphoric acid, when brought into contact with a solution of either of the oxides of mercury, with precipi- tation of mercury in the metallic state. If the mercury exists in solution, the routine of the operation is as follows. The liquid is first strongly acidified by hydrochloric acid, an excess of protochloride of tin which has been previously rendered clear by the addition of a few drops of hydrochloric acid is then added, and the mixture is boiled for a few minutes. The boiling should not be con- tinued longer than a few minutes, to avoid the risk of losing MERCURY. 183 mercury by evaporation. When the precipitated mercury is completely deposited, the supernatant liquid should be decanted, and the precipitate boiled with concentrated hydro- chloric acid, on which it generally loses its pulverulent appearance, and becomes converted into running globules. If the quantity of mercury is very small, it may be collected in a weighed filter, dried at a temperature not exceeding 120 or 130 Fahr., and weighed with the filter in a covered platinum crucible. (See page 11.) If the quan- tity of mercury, however, is not small, a better method of collecting it is the following. The precipitate having been boiled in hydrochloric acid (one object of which is to dissolve any peroxide of tin which may have been precipi- tated), the supernatant liquid is decanted, and the precipitate is washed by affusion of water acidulated with hydrochloric acid, and subsidence. When all the foreign bodies are re- moved by washing, the humid mercury is allowed to fall into a weighed porcelain capsule, and as much as possible of the liquid floating over the mercury is taken up by a piece of bibulous paper. The remaining moisture is expelled by evaporation at a temperature not exceeding 150 Fahr. When the mercury is dry, it may be weighed in the por- celain capsule. If nitric acid exists in the solution from which mercury is to be precipitated, it must be previously expelled by boiling, with the occasional addition of hydrochloric acid, so long as the formation of any chlorine or nitrous acid is perceived from the simultaneous decomposition of nitric and hydro- chloric acids. When the quantity of nitric acid is very con- siderable, it is better first to precipitate the mercury as sul- phuret, by means of sulphuretted hydrogen, and then to convert the sulphuret into metallic mercury in the following manner : The precipitate is collected in a small filter, and introduced while still moist and on the filter, into a flask containing about one ounce of pure and slightly diluted hydrochloric acid. Through this mixture, chlorine gas is passed, by which the sulphuret is completely decomposed, with formation of chloride of mercury (corrosive sublimate) N 4 184 QUANTITATIVE ANALYSIS, and free sulphur, some of the latter being slowly converted into sulphuric acid. When the undissolved sulphur is of a pure yellow colour, the transmission of chlorine may be dis- continued ; the liquid is then boiled to expel free chlorine, filtered, and metallic mercury is precipitated by protochloride of tin. If the mercurial compound to be analyzed is insoluble, it is introduced, excepting in the case of the sulphurets, into a flask, treated first with strong hydrochloric acid, and then with an excess of a saturated solution of pure protochloride of tin. The mixture is boiled for a few minutes, and allowed to cool, the flask being well corked. When the solution is quite cold, and the decomposition complete, the supernatant liquid may be decanted, and the mercury washed and col- lected in a porcelain capsule, as described above. Mercury in solution may be separated from many other bodies by sulphuretted hydrogen : if this reagent precipitates the pure protosulphuret (unmixed with free sulphur), the pre- cipitate may be collected on a weighed filter, dried carefully, and weighed with the filter in a covered platinum crucible, the quantity of metal, protoxide or suboxide, being calculated from the weight of the protosulphuret. If sulphuretted hydrogen precipitates the subsulphuret, this compound, as well as the protosulphuret when mixed with free sulphur -as would occur, if nitric acid, a salt of the peroxide of iron, chromic acid, &c. existed in the solution is decomposed by chlorine, in the manner before described, and the resulting chloride of mercury is reduced to metallic mercury, which is weighed. The reason why mercury cannot be well weighed in the state of subsulphuret, is, that a slight elevation of temperature causes the decomposition of the subsulphuret into protosulphuret and metallic mercury, and a portion of the latter might easily be lost, from its volatility. Separation of the oxides of mercury from the oxide of silver. When mercury exists in the state of protoxide (red oxide), it may be separated from silver by first precipitating the latter from the solution, as chloride, by hydrochloric acid ; and from the solution filtered from the chloride of silver, me- MERCURY. 185 tallic mercury is precipitated by protochloride of tin. This method cannot be practised when the mercury is in the state of suboxide (black oxide), as subchloride of mercury would then be precipitated with chloride of silver. In this case, it is necessary to convert the suboxide of mercury into protoxide, by gently heating with a little nitric acid ; then to precipitate the silver as chloride ; and, afterwards, if much nitric acid is present in the solution, mercury should be precipitated as sulphuret, which is converted into chloride by means of chlorine, in the manner before described. Separation of protoxide of mercury from protoxide of copper. If these oxides exist in the dry state, and unmixed with other bodies, their proportions may be determined by ex- pelling the oxide of mercury from a known weight of the mixture by heating to redness in a platinum crucible, and weighing the residuary oxide of copper. If in a state of solution, one method of effecting the separation is the fol- lowing. The protosulphurets of both metals are preci- pitated by passing sulphuretted hydrogen gas through the solution, which should not contain free nitric acid, nor any substance capable of precipitating free sulphur. The sul- phurets are collected in a weighed filter, washed rapidly, and dried in vacuo over sulphuric acid (page 128.). When dry and weighed, either the whole or a known proportion of the sulphurets is then introduced into a weighed tube retort with an elongated neck, and heated to redness : sulphuret of mercury is thereby volatilized, together with one half of the sulphur contained in the protosulphuret of copper : the latter, therefore, becomes reduced to the state of subsulphuret, Avhich is weighed. Having calculated the quantity of pro- tosulphuret of copper which corresponds to that of the sub- sulphuret weighed, the amount of sulphuret of mercury may be estimated as loss. Estimation of mercury in amalgams. The amount of mercury contained in an amalgam, that is, in an alloy of mercury with other metals, may be determined by ascertain- ing what loss in weight occurs on volatilizing the mercury by heating the amalgam to redness. If the other metals are 186 QUANTITATIVE ANALYSIS. not oxidable by the air at high temperatures, a small porcelain crucible may be employed in this operation ; but if the metals are oxidable, a tube retort should be used, the neck of which is elongated after the material is introduced, and sealed after the tube has been heated to redness. By means of sulphuretted hydrogen, mercury in solution may be separated from the metals proper not precipitated by this reagent in their acid solutions, from earths and from alkalies. The precipitated sulphuret of mercury is converted into the chloride by means of chlorine, as described page 183., from which mercury is precipitated in the metallic state by protochloride of tin. VIII. PLATINUM, PALLADIUM, OSMIUM, IRIDIUM AND RHODIUM. Analysis of platinum ore. The modes by which the metals accompanying native platinum are estimated and se- parated from each other, are well exhibited in the ordinary process for analyzing platinum ore. Besides the metals peculiar to the ore, namely, platinum, palladium, osmium," iridium and rhodium, it contains small quantities of iron and copper. For the method described we are indebted to Berzelius : although an exceedingly com- plicated process, it is, I believe, the best we possess. When every thing which seems to be foreign to the ore has been separated by mechanical means, it is digested in dilute hydrochloric acid, in order to remove the metallic iron and peroxide of iron with which the ore is generally mixed ; it is then washed, carefully dried, and about thirty grains are weighed out for analysis. After being weighed, the ore is ignited, and dissolved in aqua regia with the assistance of heat, in a glass retort, furnished with a proper receiver which should be kept constantly cold. When the acid has been distilled so far that the liquid remaining in the retort possesses the consistence of a thick syrup, the mass is dis- solved in a very small quantity of water, and the solution carefully poured off from the undissolved residue. The dis- tilled acid, which is yellow, and contains some of the ore PLATINUM, ETC. 187 projected by the effervescence, is returned to the residue in the retort and redistilled. If the liquid in the receiver is not then obtained in a colourless state, it must again be re- turned to the retort and redistilled. (A.) When procured colourless, the distilled liquid is diluted with water, and ammonia is added to it, so that a slight excess only of free acid remains. To this solution, sulphuretted hydrogen water is added, and the mixture is set aside, out of the contact of air, until the precipitate produced by sulphuretted hydrogen, which is sulphuret of osmium, has completely subsided. The clear supernatant liquid is drawn off with a syphon, and the sulphuret collected on a weighed filter, washed, dried and weighed. According to Berzelius, this sulphuret contains between 50 and 52 per cent, of os- mium. (B.) If the saline mass in the retort gives a solution smell- ing of free chlorine when dissolved in water, that gas proceeds from the decomposition of chloride of palladium, and must be expelled by boiling. Should a precipitate appear, it is oxide of palladium ; this is redissolved by hydrochloric acid, and the liquid is passed through a weighed filter, to separate some grains of a combination of osmium and iridium, to- gether with some sand which could not be separated before the analysis. When dried, these matters are weighed on the filter without being ignited. If aqua regia containing too much nitric acid is employed to dissolve the ore, peroxide of iridium is occasionally precipitated, which passes through the filter, having the appearance of charcoal. The filtered liquid is mixed, first, with twice its bulk of alcohol of sp. gr. 0-833, and afterwards with a concentrated aqueous solution of chloride of potassium, which must be added so long as a precipitate appears on agitation. The precipitate is essentially a mixture of chloride of pla- tinum and potassium, and chloride of iridium and potassium ; but it contains, besides these compounds, small quantities of the corresponding double salts of rhodium and palladium. It is collected on a filter, and washed with alcohol of sp. gr. 0-896, to which a little chloride of potassium has been added, 188 QUANTITATIVE ANALYSIS. till the filtered liquid gives no precipitate with sulphuretted hydrogen water. The subsequent operations are now divided into two series : first, the treatment of the precipitate; and, secondly, the treatment of the alcoholic solution. (C.) As much as possible of the double chlorides on the filter, when washed and dried, is removed, and carefully mixed with an equal weight of carbonate of soda : the filter, with what remains of the precipitate, being burnt, the ash is mixed with carbonate of soda, and added to the other mixture, and the whole is gently heated in a porcelain crucible until it becomes black throughout. The double chlorides are thus decomposed ; the chloride of platinum is reduced to the me- tallic state, while the iridium and rhodium are oxidized. The saline mass is treated with water, by which the greater part of it is dissolved, and hydrochloric acid is poured on the residue to extract the alkali combined with oxides of iridium and rhodium. The residue of the action of hydrochloric acid, consisting of metallic platinum and peroxides of iridium and rhodium, is washed, dried, ignited (the filter being burned separately), and weighed. This being done, it is fused in a covered platinum crucible with five or six times its weight of bisul- phate of potash, in which the rhodium dissolves, and com- municates to it a red colour. This operation is repeated with fresh bisulphate, until the salt no longer acquires a red or pink colour. (D.) The quantity of rhodium may be determined in two ways : if the platinum and peroxide of iridium which remain undissolved after fusion with bisulphate of potash are washed, ignited, and weighed, the loss on the weight of the mixture of platinum and peroxides of iridium and rhodium before the fusion indicates peroxide of rhodium, which con- tains 71 per cent, of metal. Or, instead of this method, the saline solution of rhodium may be mixed with an excess of carbonate of soda, evaporated to dryness, and ignited in a platinum crucible: on re-solution in water, peroxide of rhodium remains, which, when washed and dried, is reduced PLATINUM, ETC. 189 to the metallic state by heating in a tube in a stream of dry hydrogen gas. For the reduction, the apparatus of which a figure is given at page 160. may be employed. The rhodium thus obtained, sometimes contains palladium, which must be extracted by digestion in aqua regia ; the acid in the solution is afterwards neutralized by an alkali ; and pal- ladium is precipitated as cyanide by cyanide of mercury. On being heated to redness, cyanide of palladium is decom- posed into cyanogen gas and metallic palladium, which may be weighed. The weight of this portion of palladium is, of course, deducted from the rhodium. (E.) The next operation is to obtain the amount of iri- dium. The metallic mass which remains after the separation of rhodium by bisulphate of potash is digested with very dilute aqua regia, to remove the greater part of the platinum it contains. The solution thus obtained appears very dark, from the presence of some suspended peroxide of iridium ; but when the solid matter has subsided, it is of a pure yellow colour, and should then be decanted. The insoluble residue is treated with a mixture of concentrated aqua regia and chloride of sodium, and evaporated to dryness. The use of chloride of sodium is to prevent the formation of protochloride of platinum. When the dry mass is digested in water, per- oxide of iridium remains as an insoluble residue, which is filtered and washed, first with a solution of chloride of sodium, and afterwards with muriate of ammonia to remove the common salt. If washed with pure water, the peroxide of iridium would pass through the pores of the filter. It is now burned with the filter, and to be obtained in a fit form for weighing, should be reduced to the metallic state by hy- drogen gas, which may be done in the apparatus figured at page 160. A small quantity of iridium exists in the solution of chloride of sodium, to obtain which it is mixed with car- bonate of soda, evaporated to dryness, and ignited. The product, when freed from soda salts by water, and from pla- tinum by aqua regia, leaves peroxide of iridium, which is reduced to the metallic state, and added to that already obtained. 190 QUANTITATIVE ANALYSIS. (F.) Having by these means obtained the weights of the rhodium and indium, that of the platinum may be calculated. From the united weights of platinum, and peroxides of rho- dium and iridium (C.); subtract the peroxide of rhodium (D.), and the iridium (E.), (the latter being increased 12 per cent, to convert it into peroxide), the remainder is the weight of the metallic platinum. Such is one series of operations in this complicated analysis; there still remains for examination the alcoholic liquid fil- tered from the double chlorides. (Gr.) This liquid is introduced into a bottle or flask, which will permit of being closed air-tight, and a current of sulphuretted hydrogen gas is passed through it to saturation. The bottle or flask is then closed, allowed to stand in a warm place for twelve hours, and if the precipitate has sub- sided, the liquid is filtered, and the alcohol evaporated off. During evaporation, an additional precipitate falls, which is to be added to the former. The precipitate by sulphuretted hydrogen consists of the sulphurets of palladium, rhodium, iridium and copper ; and the liquid filtered from the preci- pitated sulphurets contains iron, rhodium and iridium, to- gether with a trace of manganese. During the evaporation of the alcohol, a greasy sulphuret is deposited, which cannot be detached from the sides of the vessel by simple washing, but may be easily removed after all the adhering liquid has been washed away, by dissolving it in ammonia. The am- moniacal solution is evaporated to dryness in a platinum crucible, the residue is mixed with the other sulphurets, and the whole is roasted by heating, with exposure to the air, so long as sulphurous acid is produced. When com- pletely oxidated, the mass is treated with muriatic acid, by which the subsulphates of copper and palladium are dissolved, while oxides of iridium and rhodium, with a trace of pla- tinum, remain undissolved. (H.) To the solution in muriatic acid of the roasted sulphurets of the last operation, a mixture of nitric acid and chloride of potassium is added. On evaporating the mixture to dryness, a dark-coloured saline mass is obtained, which PLATINUM, ETC. 191 consists of chloride of potassium, chloride of copper and potassium, and chloride of palladium and potassium,, of which the two first are separated from the third by treatment with alcohol of sp. gr. 0-833, in which the former dissolve, leaving the double salt of palladium as an insoluble residue, which is collected on a filter and washed with alcohol. When dried and ignited, this salt may be weighed: it contains 28-84 per cent, of palladium. The alcoholic solution filtered from the palladium salt is concentrated by evaporation, and oxide of copper precipitated from it by potash, (page 176.) (I.) The portion of the roasted sulphurets (G.) insoluble in hydrochloric acid, consisting of oxides of iridium and rhodium, with a trace of platinum, is fused with bisulphate of potash in a platinum crucible to dissolve out the rhodium ; an operation which is repeated until the bisulphate no longer acquires a red or pink colour. The rhodium is determined by one of the methods before described. (D.) The residue undissolved by bisulphate of potash, which is peroxide of iridium with a little platinum, is treated with aqua regia, and the peroxide reduced by hydrogen gas, in the same manner as described in a previous part of the analysis. (E.) (K.) The concentrated solution from which the sulphu- rets were precipitated contains a small quantity of iridium and rhodium, protochloride of iron, and a trace of manganese. The quantity of the last is commonly too minute to be de- termined. The iron is peroxidized by heating with nitric acid, and weighed as peroxide, obtained by precipitation by ammonia. But this peroxide of iron contains a small quan- tity of a combination of iridium and rhodium, to separate which, after weighing the peroxide, it must be reduced by hydrogen gas : the reduced metal is treated with hydrochloric acid to dissolve iron, and the black undissolved portion is collected on a filter, ignited with exposure to the air, and weighed ; its weight, deducted from that of the peroxide of iron previously obtained, leaves the quantity of the latter in a pure state. The solution filtered from the precipitate by ammonia is mixed with carbonate of soda in sufficient quantity to de- 192 QUANTITATIVE ANALYSIS. compose the ammoniacal salts, and evaporated to dryness. On treating the residue with water after a gentle ignition, peroxides of iridium and rhodium remain undissolved, but the amounts of these are in general too small for separa- tion. To facilitate the conception of this intricate analysis, I have annexed the following tabular view of the several steps in the process. The weighed ore is dissolved in aqua regia, distilling the acid: (.) The distilled liquid contains osmium ; dilute with water, neutralize most of the excess of acid by ammonia, and precipitate osmium, as sulphuret, by sulphuretted hydrogen ; weigh the sulphuret of osmium. (&.) Dilute the concentrated solution in the retort, and collect the undissolved grains of osmium and iridium (with sand) in a weighed filter; weigh this insoluble matter. Mix the filtered solution with twice its bulk of alcohol of sp. gr. 0*833, add chloride of potassium, filter, and wash with alcohol. The operation is now divided into, I. the examination of the precipitate; and, II. the examination of the alcoholic solution. 193 194 QUANTITATIVE ANALYSIS. V fc Hi. S - 9 i J s g i *-s -B a.-s TIN. 195 IX. TIN. Tin is best weighed in the state of peroxide. If the metal is contained in solution in the state of protoxide, or proto- chloride, strong nitric acid should be added to the solution, and the mixture evaporated nearly to dryness at a boiling heat; the protoxide of tin is thereby converted into the insoluble peroxide. Although peroxide of tin is insoluble in nitric acid after being boiled in the liquid, it is still slightly soluble in hydrochloric acid. If the latter acid is present, therefore, it should be expelled from the liquid by boiling, with the occasional addition of some nitric acid ; the hydrochloric and nitric acids suffer mutual decomposition, with the formation of nitrous acid and free chlorine, both of which are expelled by the ebullition. When the conversion of protoxide into peroxide of tin is complete, which is per- ceived by the cessation of nitrous vapours, the liquid may be diluted with water, and the peroxide collected on a filter. Peroxide of tin must be ignited previous to being weighed. If tin exists in solution in the state of peroxide, or per chloride, the solution is boiled with nitric acid in the same manner; peroxide of tin is thus precipitated, which may be filtered, ignited, and weighed. ~ If the substance of which the tin is an element is a me- tallic alloy, it should be boiled in moderately strong nitric acid, when the oxides of all metals which are soluble in that acid are formed and dissolved, while the tin is converted into insoluble peroxide. The liquid is evaporated until nearly all the excess of acid is expelled, diluted with water, and filtered to collect the peroxide of tin, which is ignited and weighed. As tin is precipitable by sulphuretted hydrogen, whether existing in the state of protoxide or peroxide, that reagent affords the means of separating tin from the alkalies, earths, and the metals not precipitated by sulphuretted hydrogen from their acid solutions. To estimate tin in such a case, the precipitated sulphuret, when washed and dried, is introduced o 2 196 QUANTITATIVE ANALYSIS. into a flask, and treated with fuming nitric acid, proper means being adopted to prevent loss of the solution by projection, from the intensity of the action. The tin is thus converted into peroxide, and the sulphur into sul- phuric acid. The acid liquid is poured into a platinum capsule or crucible, and evaporated carefully to dryness. When all the nitric and sulphuric acids are expelled, the residue should be heated to redness, after which, pure peroxide of tin remains. In case a little sulphuric acid should be retained by the peroxide, it is recommended, after the first weighing, to place a fragment of carbonate of ammonia in the crucible, and to heat again to redness and weigh. The analysis of the native oxide of tin is performed in the following manner : Mix the finely powdered ore with five or six times its weight of caustic potash or soda, heat the mixture to dull redness in a closely covered silver crucible for half an hour, and dissolve the resulting mass, which is usually green, in hydrochloric acid. Evaporate off most of the excess of acid, dilute the remaining liquid with water, and precipitate tin as sulphuret by sulphuretted hydrogen. Treat the precipitated sulphuret of tin with fuming nitric acid in the manner before described. The liquid filtered from the sulphuret contains protochloride of iron ; peroxidize the iron by boiling the solution with nitric acid, neutralize with ammonia, and precipitate peroxide of iron by a neutral alkaline benzoate or succinate. (Page 154.) From the liquid filtered from the precipitate of benzoate or succinate of iron, manganese may generally be precipitated by hydrosulphate of ammonia. X. ANTIMONY. In the estimation of antimony that metal is best precipitated as sulphuret, from its slightly acidified solutions, by means of sulphuretted hydrogen. It is indifferent, in this operation, whether the antimony exists in the state of oxide or the corre- sponding chloride, or as antimonious or antimonic acid. Before applying sulphuretted hydrogen gas, the liquid ANTIMONY. 197 should be diluted with water ; but if the solution is neutral, or nearly so, dilution with water may cause the precipi- tation of a subsalt of antimony, an inconvenience that is avoided by first adding a considerable quantity of tar- taric acid to the solution, after which, it may be diluted with water, without the production of any precipitate. The liquid, when saturated with sulphuretted hydrogen, should be left in a warm place exposed to the air, until the odour of the gas entirely disappears. The necessity of getting rid of the excess of sulphuretted hydrogen arises from the circumstance that water saturated with that gas possesses the property of retaining considerable traces of sulphuret of antimony in solution, which are deposited, however, when the sulphuretted hydrogen is dissipated. The precipitated sulphuret is collected on a weighed filter, dried and weighed with the filter in the usual manner. If the precipitate which is weighed is the pure sulphuret of antimony (Sb S 3 ) corresponding to the oxide of antimony (Sb O 3 ), we can, from its weight, calculate at once the amount of metal or oxide it represents. But that sulphuret is rarely precipitated in a state of sufficient purity for the ^calculation, it being generally contaminated with free sul- phur, and sometimes with the higher sulphurets of anti- mony, namely, those corresponding to antimonious and anti- monic acids, when it is obviously impossible to estimate, from the weight of the sulphuret, the amount of metal it contains. To ascertain, therefore, the purity of the precipi- tated sulphuret, after having weighed it, digest a small por- tion in concentrated hydrochloric acid : if it dissolves com- pletely, with evolution of sulphuretted hydrogen, it is the pure sulphuret corresponding to the oxide ; but if it leaves sulphur undissolved, it contains either free sulphur or a higher sulphuret of antimony. In the latter case, it is "necessary that another operation should be performed to ascertain the real amount of antimony contained in the sul- phuret. A known weight is introduced into a flask, and fuming nitric acid is added, drop by drop, proper means being taken to prevent loss of the liquid by projection. O 3 198 QUANTITATIVE ANALYSIS. When a considerable quantity of nitric acid has been thus gradually added, hydrochloric acid is applied, and the mixture is digested at a gentle heat until the whole or the principal portion of the sulphur is dissolved. The liquid is diluted, and the undissolved sulphur, if any remains, is collected in a weighed filter, dried carefully, and weighed on the filter in the ordinary manner. The sulphuric acid into which the remainder of the sulphur has been converted is precipitated by chloride of barium, after an addition of tartaric acid, to prevent the precipitation ofasubsalt of antimony, on dilution with water. From the weight of the precipitated sulphate of barytes, that of the sulphur it contains is calculated, which, added to that already weighed, and the sum deducted from the weight of the sulphuret of antimony employed, leaves the amount of metallic antimony in the sulphuret. Another calculation is then necessary to ascertain how much antimony was contained in the entire amount of the sul- phuret precipitated by sulphuretted hydrogen, as a portion of it only was operated on. Separation of antimony from other metals. From those metals whose sulphurets are insoluble in hydrosulphate of ammonia, (manganese., iron, zinc, cadmium, cobalt, lead, bis- muth, silver, mercury, and copper,) antimony may be separated by the ready solubility of its sulphuret in that liquid. The metallic alloy, or mixture of oxides, is dissolved in concen- trated hydrochloric acid or aqua regia, and the solution is supersaturated, first with ammonia, and afterwards with an excess of hydrosulphate of ammonia, which should contain its full dose of sulphur. After a gentle heating, the liquid is filtered, and the precipitate is washed with water to which some hydrosulphate of ammonia has been added. The filtered liquid is a solution of sulphuret of antimony in hydrosulphate of ammonia ; and the former may be precipitated by the addition of acetic acid, of which a slight excess should be used. The quantity of antimony in the precipitate is then determined by the process before described. In an alloy, antimony may be separated from many other metals by a process founded on the volatility of its chloride , ANTIMONY. 199 The alloy, reduced to a state of powder, is introduced into the bulb of the tube d, of the apparatus figured at page 166., and chlorine gas is caused to pass through the tube. The wash-bottle b may contain water to arrest hydrochloric acid. The bottle / is one half or two thirds filled with a weak solution of tartaric acid, to which a little hydrochloric acid is added. When the apparatus is full of chlorine, heat should be applied to the bulb; chloride of antimony then distils into the receiver, affording a limpid solution, if sufficient tartaric and hydrochloric acids are pre- sent. When all the chloride of antimony is distilled, the tube is divided near the bulb, and the adhering chloride of antimony is washed into the receiver. Through the solution thus obtained, sulphuretted hydrogen gas is passed to preci- pitate antimony as sulphuret, and the purity of the precipitate is tested in the manner before described. The quantity of the fixed chlorides in the bulb may be estimated by first weigh- ing the tube in its present state, then dissolving out the chlorides by water or an acid; and, lastly, weighing the empty tube when cleaned and dried. The difference in the two weighings is, of course, the weight of the fixed chlo- rides. From tin, the sulphurets of which metal are soluble in hydrosulphate of ammonia, antimony may be separated by precipitating it in the metallic state from the solution of the two chlorides by a bar of tin with the assistance of heat. The precipitated antimony may be collected, dried, and weighed ; and the united weights of the two metals being previously ascertained, the amount of tin originally present is known from the loss. If the metals exist in solution, and their united weight is not known, they may both be pre- precipitated together in the metallic state by a bar of zinc : after being collected and weighed, the mixed metals are dis- solved in hydrochloric acid with a few drops of nitric acid. Metallic antimony may then be precipitated by tin, collected on a weighed filter, dried, and weighed. Estimation of antimonic acid. If the antimonic acid is dissolved in an alkaline liquid, the excess of alkali is o 4 200 QUANTITATIVE ANALYSIS. first neutralized by adding nitric acid until a precipitate begins to appear. Antimonic acid is then precipitated in the form of the antimoniate of the suboxide of mercury, by adding a solution of the protonitrate of mercury. A few drops of ammonia may be added to the mixture to insure the complete precipitation of the antimoniate. The precipitate is collected, dried, and heated to redness. By ignition, it is decomposed, .oxide of mercury being given off, and pure antimonic acid remaining, in a fit state to be weighed. XI. MOLYBDENUM. To estimate molybdenum, or molybdic acid, the concen- trated solution of the substance is treated with an excess of hydrosulphate of ammonia ; the sulphuret of molybdenum is thus formed, and is dissolved by the excess of hydrosulphate. From the solution, when diluted, the tersulphuret of molyb- denum is precipitated by adding hydrochloric or acetic acid, and the mixture is digested at a gentle heat, until the odour of sulphuretted hydrogen is no longer perceptible. The pre- cipitated tersulphuret, when calcined in a small retort, parts with sulphur, being converted into the grey bisulphuret, in which state it is weighed. From the weight of the sul- phuret, the amount of molybdenum it contains, or molybdic acid it represents, may be estimated. Analysis of molybdate of lead. Dissolve the powdered mineral in dilute hydrochloric acid, with the assistance of heat, and collect the chloride of lead which is formed in a weighed filter ; after washing with water, and drying, weigh the chloride in the filter with the usual precautions. Neutralize the filtered liquid by ammonia, add an excess of hydrosulphate of ammonia, and digest the mixture for an hour in a closed flask. The lead not precipitated as chloride now falls as sulphuret, which may be collected and estimated by conversion into sulphate of lead (page 172.). To the liquid filtered from the sulphuret of lead, which is a solution of sulphuret of molybdenum in hydrosulphate of ammonia, add hydrochloric or acetic acid, and allow the mixture to TUNGSTEN. 201 digest at a gentle heat until the odour of sulphuretted hy- drogen is no longer perceived. Calcine the precipitated ter- sulphuret of molybdenum, when dry, in a small retort, and weigh it as the bisulphuret. XII. TUNGSTEN. Estimation of tungstic acid. The ordinary process for estimating tungstic acid in a solution resembles that by which molybdic acid is estimated. An excess of hydrosul- phate of ammonia is added to a concentrated neutral or alkaline solution containing the tungstic acid ; sulphuret of tungsten is thus formed, and is dissolved in the excess of hy- drosulphate of ammonia. The sulphuret in solution is now precipitated by dilute nitric acid, and the mixture is exposed to a gentle heat to expel the excess of sulphuretted hydrogen, on which all the tungsten is precipitated as tersulphuret, of a yellow colour. The precipitate is collected on a filter, washed with water slightly acidified by hydrochloric acid, dried, and ignited strongly in a platinum crucible with ex- posure to the air. The sulphuret thereby becomes oxidized and converted into tungstic acid, whose weight may be de- termined. When tungstic acid exists in a solution, with no other base but ammonia present, its weight may be determined by evaporating the solution to dryness, and calcining the resi- due, after which pure tungstic acid remains. Tungstic acid may be precipitated from its solution, for the purpose of weighing, by means of protonitrate of mercury, in precisely the same manner as antimonic acid. (See Esti- mation of antimonic acid, p. 199.) Analysis of tungstate of lime. The analysis of tungstate of lime may be performed in the following manner : The mineral is reduced to a fine powder, and digested at a gentle heat in three times its weight of concentrated nitric acid until nearly all the acid is evaporated. If the digestion and evaporation are repeated, the tungstate of lime is en- tirely converted into nitrate of lime, and the tungstic acid is set at liberty. When the excess of nitric acid is eva- 202 QUANTITATIVE ANALYSIS. porated off, alcohol is added to the dry mass to dissolve nitrate of lime, and the residue, which is tungstic acid, is dissolved in ammonia. If any thing remains undissolved after the successive action of alcohol and ammonia, it is probably silica or some stony matter foreign to the mineral. To estimate the lime, sulphate of ammonia is added to the alcoholic solution to precipitate sulphate of lime, which is dried, ignited, and weighed. The solution of tungstic acid in ammonia, when evaporated to dryness and calcined, leaves pure tungstic acid. Analysis of Wolfram (tungstate of iron and manganese). The finely powdered mineral is fused in a platinum crucible with twice its weight of carbonate of soda, and the fused mass is digested in water until nothing more is dissolved by that liquid. On treating the residue insoluble in water with hydrochloric acid, oxides of iron and manganese dissolve; these oxides are separated from each other by an alkaline suc- cinate or benzoate (p. 154.). From the alkaline solution of the fused mass in water, tungstic acid is precipitated by hydrochloric acid, washed, and redissolved in solution of ammonia. On evaporating this solution to dryness, and calcining the residue, pure tungstic acid remains, which may be weighed. Or instead of that, the tungstic acid may be precipitated by protonitrate of mercury, when the tungstate of the suboxide of mercury falls, which leaves a residue of pure tungstic acid when heated to redness in a fit state for being weighed. XIII. ARSENIC. Estimation of arsenic in alloys, arseniurets, arsenites, and arseniates. To estimate arsenic, the solution in an acid of the substance to be analyzed should be diluted with water, and the arsenic precipitated from it as sulphuret by sulphuretted hydrogen. The precipitated sulphuret is that which corre- sponds to the degree of oxidation of arsenic in the solution : if the arsenic in solution is in the state of arsenious acid, then sulpharsenious acid is precipitated; but if as arsenic ARSENIC. 203 acid, it is sulpharsenic acid which is formed. Frequently the precipitate consists of a mixture of the two sulphurets. When the subject of analysis is an arseniuret, that is, a com- bination of arsenic with another metal, it should be dis- solved by aqua regia : the whole of the arsenic is then con- verted into arsenic acid. If the compound for analysis is a soluble arsenite or arseniate, it should be dissolved in water, and the solution acidified strongly by hydrochloric acid be- fore applying the gas. As a solution of arsenious acid is pre- cipitated by sulphuretted hydrogen with greater facility than a solution of arsenic acid, it is advisable to reduce arsenic acid to the state of arsenious acid, before exposing it to sulphuretted hydrogen. This may be effected by means of sulphurous acid at a boiling heat. (See p. 119.) When saturated with sulphuretted hydrogen, the liquid should be gently heated in an open vessel until the excess of the gas is dissipated, as the aqueous solution of sulphuretted hydrogen is capable of dissolving a small quantity of the sul- phurets of arsenic. If no doubts of the purity of the precipitate are entertained, that is, of its containing no free sulphur, and of its being either entirely sulpharsenious or sulpharsenic acid, and not a mixture of these sulphurets, then it may be collected in a weighed filter, dried, and weighed with the filter, and from the weight of the sulphuret, the quantity of metallic arsenic it contains, or arsenious or arsenic acid it represents, may be deduced by calculation. But it rarely happens that the sul- phuret is obtained sufficiently pure for the calculation : for as the mere exposure of solution of sulphuretted hydrogen to the air causes a precipitate of sulphur, it is scarcely possible to obtain either of the sulphurets quite free from that element. Another operation must then be performed to ascertain the true amount of arsenic in the precipitate, which is precisely analogous to that already described for a similar purpose under antimony (page 197.). After the sulphuret is collected on a weighed filter, washed and dried, it is weighed together with the filter, and as much as can easily be removed is in- troduced into a flask, and digested with aqua regia. The 204 QUANTITATIVE ANALYSIS. remaining sulphuret with the filter is weighed, after another drying, to know the amount introduced into the flask. By digestion with aqua regia, the arsenic becomes converted into arsenic acid, and a portion or the whole of the sulphur into sulphuric acid. After a long digestion, to insure perfect de- composition of the sulphuret, the liquid is diluted, and the undissolved sulphur (if any) is collected in a weighed filter, dried and weighed. The sulphuric acid in the solution is estimated as sulphate of bary tes, the liquid being made pretty strongly acid to prevent the precipitation of arseniate of barytes. The sulphate of barytes being weighed, the quan- tity of sulphuret it contains is calculated, and added to that already collected: the sum, deducted from the weight of the sulphuret employed, leaves the amount of metallic arsenic. After having precipitated arsenic in the state of sulphuret from a solution of arsenious acid, by means of sulphuretted hydrogen, and determined the weight of the precipitate, the small quantity of free sulphur which is mixed with it may often be perfectly separated by treating the precipitate with ammonia, which dissolves the sulphuret of arsenic without affecting the sulphur. The quantity of sulphur can then be readily determined. (Rose.) In this manner the acids of arsenic may be separated from the alkalies, earths, and all metals not precipitated by sulphuretted hydrogen from their acid solutions.* From the greater number of those metals the solutions of which are precipitated by sulphuretted hydrogen, arsenic, like anti- mony, may be separated by the ready solubility of its sul- phurets in hydrosulphate of ammonia. To the solution of the substance, add first an excess of pure ammonia, and after- wards an excess of hydrosulphate of ammonia ; sulphurets of * It has been observed by M. Wohler, that when sulphuretted hydrogen gas is passed through a solution in a mineral acid of arsenic acid and oxide of zinc, the precipitate which is produced is a combination of sulpharsenic acid and sul- phuret of zinc (Zn S+ AsS 5 ), whatever excess of free acid may exist in the solu- tion. But if the arsenic acid is reduced to the state of arsenious acid before transmitting sulphuretted hydrogen (which may be effected by means of sul- phurous acid, page 1 19. ), sulpharsenious acid is precipitated in a pure state, all the zinc remaining in solution. AKSENIC. 205 the various metals are thereby formed, sulphuret of arsenic is redissolved, while the other sulphurets remain insoluble. When cold, the precipitate is collected on a filter, and washed with water containing a little hydrosulphate of ammonia. On acidifying the filtered liquid with acetic acid, sulphuret of arsenic and free sulphur are precipitated, and sulphuretted hydrogen is evolved. After the liquid has been digested at a gentle heat until the odour of sulphuretted hydrogen is no longer perceptible, the precipitate may be collected on a weighed filter, and its proportion of arsenic determined by one of the processes above described. Alloys of arsenic with metals whose chlorides are not volatile at a red heat, may be analyzed by exposing the alloy to chlorine gas, and distilling the volatile chloride of arsenic in precisely the same manner as the chloride of antimony in the analogous method of analyzing alloys of antimony. The method of separating arsenic from antimony is not very exact. If the substance for analysis is an alloy con- sisting of no other metal than arsenic and antimony, the former may be separated from the latter by distillation in an atmosphere of hydrogen gas. The alloy is powdered and introduced into the bulb of the tube (fig. 19. page 160.); dry hydrogen gas is passed through the apparatus, and the heat of a spirit lamp is applied to the bulb until all the arsenic is distilled and driven out of the tube. The weight of the residue of metallic antimony deducted from the original weight of the alloy gives that of the arsenic expelled. When arsenic and antimony occur mixed with other metals, or in a state of solution, the above process cannot be practised. In such a case, the solution of the substance (previ- ously mixed with tartaric acid to prevent the precipitation of a subsalt of antimony) is diluted with water, and the sul- phurets of both metals are precipitated by sulphuretted hydrogen. The two sulphurets should be well mixed by agitation, and then collected in a weighed filter, washed, dried and weighed. Two operations are now to be per- formed on the precipitate ; the first to ascertain its amount of sulphur, and the second to ascertain the relative propor- 206 QUANTITATIVE ANALYSIS. tions of antimony and arsenic present. 1. The weight of the whole precipitate being ascertained, about one half of it is removed into a flask, and the filter with the remainder is warmed and weighed, to learn the quantity introduced into the flask. That portion is very cautiously treated with aqua- regia, beginning by adding fuming nitric acid drop by drop, and afterwards adding hydrochloric acid with the same caution. Tartaric acid is added to the solution thus formed, which is diluted with water, and any sulphur which may re- main undissolved is collected and weighed. The sulphuric acid in solution is then estimated as sulphate of barytes, the proportion of sulphur in which added to that previously weighed, and the sum deducted from the weight of the sul- phurets employed, gives the proportion of mixed antimony and arsenic. 2. To obtain the relative proportions of the two metals, a known weight of the other portion of the pre- cipitated sulphurets is heated in a current of hydrogen gas in the apparatus (Jig. 19. p. 160.), when sulphur from the sulphuret of antimony is first disengaged, and afterwards sulphuret of arsenic sublimes, leaving a residue of metallic antimony, which may be weighed. Having previously ob- tained the united weight of both metals, the weight of the tin can now be calculated. When arsenic and antimony are mixed with metals which are precipitated by sulphuretted hydrogen from their acid solutions, but whose sulphurets are insoluble in hydrosulphate of ammonia, the two sulphurets may be obtained together, separate from the other sul- phurets, in the same manner as arsenic or antimony alone is separated from such metals. SILICON. 207 SECTION V. NON-METALLIC BODIES. I. SILICON. Estimation of silicic acid (silica). For the purpose of estimation, silicic acid may be separated from the solution in which it is contained, by first rendering the latter strongly acid, if it is not already so, and then evaporating the liquid to dryness. Hydrochloric acid is generally employed to acidify the solution of silicic acid. During evaporation, the silicic acid is deposited, most frequently in the form of an in- soluble jelly, but occasionally in a pulverulent state : it is not, however, completely separated until the liquid is brought to perfect dryness, and the residue treated with a dilute acid to remove all soluble matters. Silicic acid then remains undissolved, generally in a state of purity, and may be col- lected on a filter, washed, dried, and ignited in a platinum crucible. As thus obtained, silicic acid is highly hygro- metric ; hence the crucible in which it is contained should be kept well covered while being cooled and weighed. ANALYSIS OF NATURAL SILICATES. The first ingredient of silicates whose weight should be determined in the analysis of these bodies is water, to esti- mate which a known weight of the silicate previously de- prived of all its hygrometric moisture by gentle drying (page 128.) is heated to redness in a weighed platinum crucible for about half an hour. The loss of weight sus- tained by the ignition indicates the amount of water con- tained in the silicate. For the subsequent operations, it is better to employ another portion of the mineral, as silicates which have undergone ignition do not dissolve so readily as before being heated. For the purpose of analysis, silicates may be conveniently divided into two classes ; those decomposable by acids, and those undecomposable by acids. 208 QUANTITATIVE ANALYSIS. I. SILICATES DECOMPOSABLE BY HYDROCHLORIC ACID. Allophane Gadolinite Analcime Gehlenite Anorthite Haiiyne Apophyllite Helvine Botryolite Heulandite Calamine Ilvaite Cerine Laumonite Cerite Lazulite Chabasie Leucite Cronstedtite Meerschaum Datholite Mesole Dioptase Mesolite Epistilbite Natrolite Eudialite Nepheline Nosian Okenite Orthite Pectolite Potash-harmotome Pyrosmalite Scapolite Scolezite Siliceous oxide of copper Sodalite Stilbite Titariite (Sphene) Wollastonite. IL SILICATES NOT DECOMPOSABLE BY HYDROCHLORIC ACID. Acmite Emerald Petalite Albite Epidote Finite Andalusite Euclase Prehnite Autophyllite Felspar Pyroxene Axinite Garnet Serpentine Barytes-harmotome Hornblende Soapstone Carbonated manganese Idocrase Soda-spodumene Carpholite Labradorite Spodumene Chlorite Lepidolite Staurolite Chondrodite Mica Talc Diallage Obsidian Topaz Dichroite Olivine Tourmaline. Except, however, in estimating the amount of alkali con- tained in the mineral, the course to be pursued in the analysis of a silicate of one class differs from that of a silicate of the other class only at the commencement of the process. (1.) The following is the method commonly practised in the analysis of a silicate decomposable by hydrochloric acid. The mineral being reduced to a very fine powder, is dried at a gentle heat in a platinum crucible to expel hygrometric moisture, weighed and digested in hydrochloric acid in a porcelain basin. The bases formerly united to silicic acid thereby become chlorides, and silica, having the appearance of a jelly, is set free. If the decomposition is complete, no SILICON. 209 gritty particles of the mineral are perceived on drawing the point of a glass rod across the bottom and sides of the porcelain basin. The solution thus obtained is evaporated to perfect dry- ness, the residue moistened with hydrochloric acid, and then digested in hot water. The chlorides dissolve, leaving a residue of pure silicic acid, which may be filtered, washed, ignited, and weighed. (2.) Silicates not decomposable by hydrochloric acid must be fused with an alkaline carbonate, a caustic alkali, carbonate of barytes, or quick-lime, and afterwards dissolved in an acid. As this operation has already been fully de- scribed when treating of the " qualitative analysis of sili- cates," page 79., any further notice of it is unnecessary. After separation of the silica, the remaining steps in the analysis of silicates of both classes differ according to the nature of the constituents. Supposing the mineral to con- tain all the bases which ordinarily enter into the composition of silicates (page 80.), the usual course is, first to add a slight excess of ammonia to the solution filtered from the silicic acid, on which alumina and peroxide of iron, with traces of oxide of manganese and magnesia, are precipitated, methods of separating which have already been described (pages 154. and 155.). From the liquid filtered from those oxides, lime is precipitated by oxalate of ammonia (page 138.), and the oxalate of lime being separated, manganese is then thrown down as sulphuret by hydrosulphate of ammonia (page 151.); magnesia may be precipitated afterwards by phosphate of soda and ammonia. For details of the methods of estimating and separating these bases, I must refer to what has been previously said on the subject, under each particular base. When magnesia and manganese occur together, a better method than the above may be adopted, as an example of which the following analysis of the mineral colophonite is subjoined (Liebig. Handwb'rterbuch, p. 343.). The constituents of that mineral are: silica, alumina, magnesia, lime, peroxide of iron, and protoxide of manganese. After fusion with an alkaline carbonate and separation of P 210 QUANTITATIVE ANALYSIS. silica in the ordinary method, the solution is mixed, first, with chloride of soda, and then with bicarbonate of potash. .. /, t "~*\ The solution contains lime The precipitate contains alumina, peroxide and magnesia. of iron, and peroxide of magnesia. Treat it with solution of potash to dissolve out alumina. Dissolve the residue, which consists of peroxide of manganese and peroxide of iron, in hydrochloric acid, and saturate the solution with carbo- nate of barytes in the cold. ji ,___^ f~~ ~~~\ The solution contains manga- The precipitate is peroxide of nese and barytes : precipi- iron mixed with carbonate tate manganese by hydro- of barytes. Dissolve in hy- sulphate of ammonia. drochloric acid, and pre- cipitate peroxide of iron by ammonia. Estimation of alkalies in silicates. The amount of alkali contained in silicates decomposable by acids, and not con- taining magnesia, may be easily determined in the following manner: The mineral is dissolved in hydrochloric acid, and silica is separated in the usual manner. To the solution, which contains the chlorides of the bases formerly united to silica, carbonate of ammonia is added, and the liquid filtered from the precipitate. In the absence of magnesia, all the earthy bases are thereby precipitated, and the filtered solu- tion, when evaporated to dryness and the residue calcined, yields the alkali in the state of chloride, fit for weighing. But it is obvious that when fusion with an alkali is neces- sary to procure solution and the ordinary method is followed, the estimation of alkali will be extremely difficult, if not alto- gether impracticable. Other means of effecting decomposition are then adopted in the determination of the alkali, for which another portion of the mineral must be employed, all the other constituents having been previously estimated by the usual process. Several modes are proposed of procuring the decomposition of silicates by means of substances, of which the presence would offer no impediment to the esti- mation of the alkali ; as by means of hydrofluoric acid, by fusion with carbonate of barytes or caustic barytes, and by SILICON. 211 oxide of lead, each of which may be removed from the solution of the ignited mass in an acid, without interfering with the estimation of the alkali. A method which gives a very satisfactory result con- sists in heating to whiteness in a platinum crucible for an hour an intimate mixture of one part of the silicate with about six parts of finely powdered quick-lime. The silicic acid before in combination with alkali then unites with lime, and the alkali is therefore set free. The mass, which coheres considerably from the heating, is removed from the crucible after the ignition, and carefully powdered in a glazed mortar, placed on a sheet of paper to preserve any particles which may be projected. The powder is then transferred to a basin, and digested in water for ten or twelve hours at a moderate heat ; the solution thus obtained is filtered, and the undissolved lime washed. The filtered liquid is merely a solution of the alkali in lime-water, from which the former may be obtained in the following manner: Carbonic acid gas is passed through the liquid, until the car- bonate of lime at first precipitated is re-dissolved ; the liquid is boiled to re-precipitate carbonate of lime (by dissipating the excess of carbonic acid gas), and filtered. The filtered solution is then slightly acidified by hydrochloric acid, eva- porated to dryness, and the alkali weighed as chloride (page 130.). Instead of performing the latter part of the oper- ation as indicated, the calcined mass may be treated with hydrochloric acid, by which both earths and the alkali are dissolved, and the solution precipitated by an excess of car- bonate of ammonia ; nothing then remains in solution but the alkali which is to be estimated and ammoniacal salts. The solution is evaporated to dryness, the residue is calcined, and the alkali weighed as chloride. The large quantity of ammoniacal salts present renders it difficult to effect the calcination of the residue without some loss of the alkali by projection. The silica obtained in the analysis of silicates should always be tested as to purity after being weighed. This is best done by fusing a small quantity in ja bead of carbonate p 2 212 QUANTITATIVE ANALYSIS. of soda on a piece of charcoal before the blowpipe. If a colourless and transparent glass is obtained, the silica is pure, or nearly so ; but if an opaque bead is produced, the silica is not free from foreign earthy matter. II. SULPHUK. Estimation of sulphuric acid. Few bodies can be deter- mined with greater accuracy than sulphuric acid. The solution of the substance in water is acidified by nitric or hydrochloric acid, and the sulphuric acid precipitated in the form of sulphate of barytes, by adding nitrate of barytes or chloride of barium to the solution. Before filtration, the mixture should be heated to cause the aggregation of the precipitate, otherwise the latter generally passes through the pores of the filter. The precipitate must be ignited before being weighed. From the weight of the precipitated sul- phate of barytes, that of the sulphuric acid it contains is cal- culated. If the substance to be analyzed is insoluble in water, but soluble in dilute acids, pure hydrochloric or nitric acid should be used as the solvent. The sulphates insoluble, or nearly so, in water and dilute acids are, sulphates of barytes, lead, lime, and strontian. Sulphate of lead may be analyzed by fusing it with about three times its weight of carbonate of soda in a platinum crucible, by which sulphate of soda and oxide of lead are produced. All contact with organic matter must be carefully avoided, to prevent reduction of oxide of lead to the metallic state, which would inevitably destroy the crucible, by the formation of a fusible alloy of platinum with the reduced lead. When the fused mass is treated with water, the greater portion of it dissolves, leaving a residue of oxide of lead, which may be collected on a filter, and if pure, washed, dried, ignited, and weighed. If the oxide of lead is mixed with other oxides, these are separated by processes before described. To the filtered liquid, which contains sulphate and carbonate of soda, add a slight excess of nitric acid, and estimate the sulphuric acid by precipitation as sulphate of SULPHUR. 213 barytes. The aqueous solution of the fused mass generally contains a little lead, which may be precipitated as sulphuret by sulphuretted hydrogen, and weighed as the sulphate (page 172.). Sulphate of barytes and sulphate of strontian are analyzed by mixing them in a minute state of division with three times their weight of carbonate of soda, and fusing the mixture in a platinum crucible. Sulphate of soda and carbonates of the earths are formed, the latter remaining undissolved when the fused mass is treated with water. When washed, the earthy carbonates are dissolved in dilute hydrochloric acid, and estimated in the usual manner (pages 136. and 137.). The solution filtered from the earthy carbonates is acidified by hydrochloric acid, and sulphuric acid is precipitated as sul- phate of barytes by chloride of barium. Sulphate of lime may be completely decomposed by boiling it with a solution of carbonate of potash or soda, with the formation of carbonate of lime and an alkaline sulphate. The carbonate of lime is collected on a filter, dried, and weighed after a gentle calcination, and the sulphuric acid in the filtered solution is precipitated as sulphate of barytes, the excess of alkaline carbonate being previously neutralized by hydrochloric acid. After the carbonate of lime is weighed, it should be tested for any undecomposed sulphate of lime, by dissolving it in pure dilute hydrochloric or nitric acid. If an insoluble, residue remains, or if chloride of barium pro- duces a precipitate of sulphate of barytes in the acid solu- tion, the carbonate of lime has not been free from sulphate of lime. From the insolubility of sulphate of barytes both in water and dilute acids, sulphuric acid in solution may be separated from almost every other substance ; all the salts of barytes which are insoluble in water being soluble in hydro- chloric and nitric acids, with the single exception of the sulphate. Sulphurous acid, existing in a state of solution, may be esti- mated by converting it into sulphuric acid by means of strong nitric acid, by aqua-regia, or by chlorine gas, with the appli- p 3 214 QUANTITATIVE ANALYSIS. cation of a gentle heat ; the sulphuric acid formed being afterwards weighed as sulphate of barytes. ANALYSIS or SULPHURETS. The ordinary method of analyzing a metallic sulphuret is to digest the substance in nitric acid or aqua-regia; the nietal thereby oxidates, and commonly dissolves, and the sulphur is converted into sulphuric acid, which may be precipitated and weighed as sulphate of barytes. In general, the metal is oxidated and dissolved long before the sulphur ; and as a continuance of the digestion is tedious, from the slowness of the action, if the undissolved sulphur has a pure yellow colour, and if it does not appear to be mixed with any undecomposed sulphuret, it may be collected on a weighed filter (having first diluted the solution), washed carefully, dried, and weighed on the filter. From the filtered liquid, sulphuric acid is precipitated by chloride of barium. If the sulphuret is in a highly divided state, and the acid employed highly concentrated, the whole sulphur is some- times immediately converted into sulphuric acid. The metals existing in the solution filtered from the sul- phate of barytes are then separated and estimated in various ways by processes already treated of. If the presence of barytes arising from the excess of chloride of barium em- ployed interferes with the determination of these metals, the barytes may first be removed by adding sulphuric acid to the solution. Weak nitric acid should not be employed to dis- solve metallic sulphurets in analysis, as an evolution of sul- phuretted hydrogen gas might then be occasioned, with a consequent loss of sulphur. In all cases in which sulphur is separated in the elementary state, especially in the analysis of metallic sulphurets, the purity of the sulphur should be tested, after being weighed, by volatilizing it in a weighed porcelain crucible. If any fixed residue remains in the analysis of a metallic sul- phuret, it is commonly the oxide of the metal, and was, of course, weighed in the state of sulphuret with the sulphur. SULPIIUR. 215 After a strong calcination of the residue in the open air to insure perfect oxidation, and also to expel any sulphuric acid which may have been formed, it is weighed ; and the nature of the oxide being known, the weight of sulphuret which it represents is calculated and deducted from the weight obtained of sulphur. When this oxide is dissolved in hydrochloric acid, there usually remains a trace of stony matter or silicic acid undissolved, if the substance analyzed is a natural sulphuret. In this manner, most metallic sulphurets may be ana- lyzed; but a small number require slight modifications of the above process. The sulphuret of silver must be de- composed by pure nitric acid, and not by aqua-regia : when the decomposition of the sulphuret is complete, and the liberated sulphur is separated from the solution, silver may be precipitated by hydrochloric acid ; and after filtration from the chloride of silver, sulphuric acid may be precipitated by chloride of barium. Sulphuret of bismuth must also be treated with nitric acid, and not with aqua-regia. The liberated sulphur being collected and washed with dilute nitric acid, bismuth is precipitated from the solution by car- bonate of ammonia, and sulphuric acid from the filtered liquid by chloride of barium." Sulphiiret of lead may be analyzed by digesting it in fuming nitric acid, by which it is converted into the insoluble sulphate of lead. The greater part of the excess of acid being expelled by evaporation, a few drops of sulphuric acid are added, the mixture is diluted, and the sulphate of lead collected on a filter. It should be ignited previous to being weighed (page 173.). As the sulphurets of mercury cannot be decomposed by nitric acid, it is necessary to use aqua- regia to dissolve these compounds. The sulphuric acid pro- duced is precipitated by chloride of barium, and mercury is thrown down from the filtered liquid by sulphuretted hydro- gen (page 184.). The combinations of sulphur with most metals may be analyzed by heating the substance in an atmosphere of chlorine gas. In this operation the apparatus,^. 20. page p 4 216 QUANTITATIVE ANALYSIS. 166., may be employed. Chloride of sulphur and chlorides of the metals are thereby formed; the former, together with the volatile metallic chlorides, distils into the receiver, which should be kept one fourth filled with water; the fixed chlorides remain in the bulb. Before applying heat to the sulphuret in the bulb, all the air in the receiver should be displaced by chlorine. The chloride of sulphur which distils is decomposed, on coming in contact with the water in the receiver, with formation of free sulphur and hydrochloric and sulphurous acids ; the latter is immediately converted into sulphuric acid by chlorine in the presence of water. The sulphur is collected in a weighed filter, and solution of chloride of barium is added to the filtered liquid to precipitate sulphuric acid. The metallic chlorides remaining in the bulb may be dissolved out by water, and estimated Iby the ordinary processes. A method has been recently recommended for the analysis of metallic sulphurets, which consists in passing pure oxygen gas over the substance, heated to redness in a porcelain tube placed across a furnace : sulphurous acid is thus formed from combustion of the sulphur, and conducted into a vessel con- taining a strong aqueous solution of chlorine *, by which it is converted into sulphuric acid (Maugham). The analysis of alkaline and soluble earthy sulphurets may be performed in the following manner : A solution of chloride of copper is added to the aqueous solution of the substance to be analyzed, and the precipitate, which is sul- phuret of copper, is rapidly collected on a filter, and oxidized by fuming nitric acid. When the sulphur is completely dissolved, a solution of nitrate of barytes is added to preci- pitate the sulphuric acid into which the sulphur has been converted, and the amount of sulphur is calculated from that of the sulphate of barytes obtained. If the alkaline sulphuret is treated at once with nitric acid, some sulphur would escape oxidation and be lost, being evolved in the state of sulphu- retted hydrogen gas; hence the necessity of having the * A solution of chloride of soda would probably be found more convenient than chlorine water in this operation. CHLORINE. 217 sulphur in the state of sulphuret of copper, which does not yield sulphuretted hydrogen with strong nitric acid. The liquid, filtered from the precipitated sulphuret of copper may be treated with sulphuretted hydrogen, to separate copper ; and the alkali or earth remaining in solution may be deter- mined in the usual manner. Instead of mixing the solution of chloride of copper with the solution of the alkaline or earthy sulphuret, sulphuretted hydrogen gas may be produced by decomposing the sulphuret by hydrochloric acid, and the gas transmitted through so- lutions of the copper salt, contained in a series of Woulf's bottles, the last solution being supersaturated with am- monia. When the decomposition is complete, a saturated solution of carbonate of ammonia should be introduced into the generating bottle to cause a stream of carbonic acid gas to traverse the whole apparatus and thus carry forward the remaining sulphuretted hydrogen. The precipitated sul- phuret of copper is treated by nitric acid as above. III. CHLORINE. Analysis of chlorides. Hydrochloric acid and chlorine may be determined with great accuracy. The solution of the substance to be analyzed is acidified by nitric acid, warmed, and the chlorine is precipitated in the form of chloride of silver by adding a solution of nitrate of silver to the liquid. The precautions to be observed in the precipi- tation and ignition of chloride of silver have already been mentioned under the subject of silver (page 179.). From the liquid filtered from the chloride, the excess of silver may be removed by adding hydrochloric acid, and the bases present may then be determined by the ordinary processes. If the substance is insoluble in water, but soluble in acids, such as many subchlorides, nitric acid should be employed as the solvent. Chloride of silver and chloride of lead may be analyzed by reducing these compounds to the metallic state by hydrogen gas, in the apparatus jft/. 19. (page 160.) with the assistance 218 QUANTITATIVE ANALYSIS. of heat; hydrochloric acid gas is then disengaged. The degree of heat required for the decomposition of metallic chlorides by hydrogen gas is somewhat higher than that necessary to decompose the oxides of the same metals by hydrogen. When the decomposition of the chloride is com- plete, which is known by the absence of white fumes on holding a rod moistened with ammonia at the extremity of the apparatus, the tube c containing the reduced metals may be disconnected from the apparatus and weighed. The weight of the tube and substance having been previously determined, the loss of weight on reduction represents the amount of chlorine contained in the chloride. Subchloride of mercury (calomeT) may be analyzed by digestion in a solution of caustic potash ; the black suboxide of mercury and chloride of potassium are then formed. The filtered alkaline solution is supersaturated with nitric acid, and the chlorine contained in it is precipitated by nitrate of silver. It is advantageous to take another portion of calomel to determine the mercury by means of protochloride of tin. CHLORIMETRY. Chlorimetry, or the process of the valuation of chloride of lime (bleaching powder), is an important subject, in con- nection with the chemical arts. Of many methods which have been proposed, that which appears entitled to preference is one in which the chlorine is determined by finding what quantity of the chloride of lime to be analyzed is required to convert a known weight of the protoxide of iron into peroxide. For this change, protoxide of iron requires half an equivalent of oxygen, which is supplied in the present process by the decomposition of half an equivalent of water by chlorine of the chloride of lime. Half an equi- valent of chlorine, therefore, or 221-8 parts, effects this change upon a whole equivalent, or 439 parts of protoxide of iron, which quantity of protoxide is contained in one equivalent, or in 1728 parts of crystallized protosulphate of iron. Therefore 221-8 parts of chlorine can peroxidize 1728 CHLORINE. 219 parts of crystallized protosulphate of iron, or 10 grains of chlorine can peroxidize 78'1 grains of the crystallized proto- sulphate. Now the process consists in discovering how much of the specimen of chloride of lime is requisite to per- oxidize 78-1 grains of the protosulphate, which quantity of the chloride must contain 10 grains of chlorine. In the first place, 78 grains of the pure protosulphate (clean crystals of the salt dried by strong pressure between folds of cloth) are dissolved in about two ounces of water, and the solution is acidulated by a few drops of hydrochloric or sulphuric acid. Fifty grains of the chloride of lime are weighed out, mixed well in a mortar with about two ounces of tepid water, and the mixture poured into an alka- limeter (page 132.). The measure is filled up to with the washings of the mortar, and the liquids are well mixed by agitation, the mouth of the alkalimeter being closed by the palm of the hand. This solution of chloride of lime is then gradually added to the solution of 78 grains of proto- sulphate of iron until the latter is completely peroxidized, which is discovered by means of the red prussiate of potash. This salt gives a precipitate of Prussian blue with a salt of the protoxide of iron only, and not with a salt of the per- oxide ; so long, then, as it produces a blue precipitate, pro- toxide of iron still exists in solution, and more chloride of lime must be added. In applying red prussiate as the test, it is convenient to have a plate spotted over with small drops of its solution, which are touched with the drop of the liquid to be tested, taken out on the point of a glass stirrer. Stopping short exactly at that point at which the red prus- siate of potash ceases to afford Prussian blue, the number of measures poured from the alkalimeter is carefully observed ; the richer in chlorine the specimen of chloride of lime is, the less of course is required. As this amount, whatever it may be, but which, for example, we will suppose to be 70 mea- sures, contains always 10 grains of chlorine, it becomes a simple matter to calculate the per centage of chlorine in the specimen. For as 100 of the alkalimeter divisions con- tain 50 grains of the chloride, each single measure must 220 QUANTITATIVE ANALYSIS. contain half a grain ; the 70 measures poured out, contained therefore 35 grains of chloride of lime. Then, if 35 grains of the chloride contain 10 grains of chlorine, by simple pro- portion 100 grains of the chloride must contain 2 8 -5 7 grains of chlorine. The only calculation to obtain the per centage is, to divide 1000 by one half the number of measures poured from the alkalimeter : ^= 28-57. Or, what is the same thing, to divide 2000 by the actual number of measures poured out.* IV. IODINE. Analysis of iodides. The amount of iodine contained in iodides which are soluble in water may be estimated by pre- cipitating the iodine in the form of iodide of silver by the nitrate of silver. The precipitated iodide is collected and ignited in the same manner and with the same precautions as the chloride of silver. It is, however, more usual to deter- mine the amount of iodine as loss, that is, as the deficiency on the weight of the original substance, having ascertained the quantity of all the other ingredients of which the sub- stance is composed. To obtain, then, the amount of base combined with iodine, the iodide may be decomposed by sulphuric acid in a porcelain capsule with the assistance of heat ; hydriodic acid, sulphurous acid and iodine are evolved, and a residue remains, consisting of the sulphate of the oxide of the metal formerly united to iodine. The composition of the iodide being known, the weight of metal in the portion weighed is calculated and deducted from the original weight of the substance, to obtain the amount of iodine. In the case of iodides, whose metallic bases are precipitated from solution in an acid by sulphuretted hydrogen or hydrosulphate of ammonia, those reagents may be used to separate the metals. * In the preceding process, Mr. W. Crum uses a stoppered phial to contain the solution of sulphate of iron, which is shaken, instead of an open jar or tumbler, and thus prevents any escape of chlorine. He can then add the chlo- ride of lime in a dry state, from a weighed quantity, in the absence of an alkalimeter. ( Trans. Glasgow Phil. Society}. IODINE. 22 1 If the iodide for analysis is insoluble both in water and acids, it should be ignited with carbonate of soda, and the heated mass afterwards digested in water. Iodide of sodium and the excess of carbonate of soda dissolve, and leave the metallic oxide as a residue, if it is insoluble in a solution of the alka- line carbonate. The oxide may be weighed, the correspond- ing amount of metal calculated, and deducted from the original weight of the iodide to obtain the amount of iodine. Separation of iodine from chlorine. As iodides rarely exist in nature unmixed with chlorides, it is important to have the means of completely separating chlorine from iodine. Many processes have been proposed for this purpose, one of the latest and best of which is the following, by M. Rose : Both chlorine and iodine are precipitated from the mixed salts by nitrate of silver, and the mixed iodide and chloride of silver is collected, ignited, and weighed. The mixed iodide and chloride is then introduced into a tube of hard glass, having a bulb blown on its middle (such as the tube c of the apparatus (Jig. 19. page 160.), and after being weighed with the tube, chlorine gas is passed over the sub- stance. By the action of chlorine, the iodide of silver is decomposed, chloride of silver being formed, and iodine liberated, which may be driven out of the tube by the appli- cation of heat. On again weighing the tube, a loss will be experienced, which, multiplied by 1-389, gives the quan- tity of iodine contained in the mixed chloride and iodide of silver operated on. The number 1*389 bears the same proportion to 1 which the equivalent of iodine bears to the difference between the equivalents of chlorine and iodine. For every equivalent of iodine (1579*5) expelled by chlorine one equivalent of chlorine (442*65) is absorbed; the loss actually sustained in the experiment, therefore, is the differ- ence between the equivalents of chlorine and iodine, or 1136*85 for every 1579*5 parts of iodine, which numbers are as 1 to 1*389. Having thus obtained the iodine, the amount of chlorine may be calculated by deducting the weight of the iodide of silver (found by calculation) from the weight of the mixed chloride and iodide of silver ; the 222 QUANTITATIVE ANALYSIS. remaining chloride of silver contains 24-67 per cent, of chlorine. The proportion of iodine existing in iodine ley may be conveniently estimated by means of a solution composed of 1 part of crystallized sulphate of -copper, and 2J parts of crystallized protosulphate of iron. On adding such a solution, all the iodine is thrown down from the ley as subiodide of copper, the chloride of copper formed at the same time from the chlorides present in the ley remaining in solution. The precipitate must be collected on a weighed filter, washed, dried, and weighed. It contains 6 6 '6 per cent, of iodine. V. BROMINE. The amount of bromine contained in bromides may be estimated in the same manner as iodine is estimated in iodides ; that is, either by precipitation, as bromide of silver, which is collected and fused with the same precautions as the chloride of silver, or by determining the amounts of all the other ingredients, and estimating bromine as the loss on the original weight of the substance. The amount of me- tallic base may be ascertained either by conversion into sul- phate by heating the bromide with sulphuric acid in a porce- lain capsule, or by separating the metal from the solution of the substance by means of sulphuretted hydrogen or hydrosul- phate of ammonia. Bromides insoluble in acids may be analyzed by fusion with carbonate of soda, in a manner pre- cisely similar to that described for insoluble iodides. Separation of bromine from iodine. In a mixed bromide and iodide the respective proportions of bromine and iodine may be ascertained by a method similar to that used for the analysis of mixed chlorides and iodides. The mixture of iodide and bromide of silver is obtained, its weight de- termined, and the mixture is heated in an atmosphere pf bromine. A loss is found to occur after heating, which, multiplied by 2-627, gives the proportion of iodine. FLUORINE. 223 VI. FLUORINE. Analysis of fluorides. The ordinary method of deter- mining the amount of fluorine contained in fluorides is, to separate the fluorine in the state of hydrofluoric acid, and to calculate its quantity from the loss. To perform this, the dry weighed fluoride is moistened with oil of vitriol in a platinum capsule, and heated so long as acid vapours are given off. The residue, which consists of the sulphate of the oxide of the metal formerly in combination with fluorine, is weighed, and the proportion of metal it contains calculated; the weight of the metal deducted from the original weight of the substance leaves the amount of fluorine. If more than one base is combined with fluorine in the substance for analysis, it is necessary to undertake a full determination of the bases, and to calculate the amount of each metal. To obtain the amount of fluorine, the united weight of the metals is deducted from the weight of the fluorides operated on. When the substance to be analyzed contains water, that, it is obvious, will be included in the loss, and must, therefore, be estimated by another operation before the fluorine can be determined. The amount of water contained in many fluo- rides may be estimated by heating the substance alone to redness in a platinum crucible. The whole of the water may then be given off in a pure state, and its amount is learned by the loss in weight undergone by heating. But it frequently happens, that besides the water a little hydrofluoric acid is disengaged, which vitiates the determination of the former. In such a case, the weighed fluoride, in fine powder, is intimately mixed with freshly calcined and powdered prot- oxide of lead, the mixture is introduced into a small hard glass retort, and covered with a little more pure oxide of lead. The retort, with its contents, is weighed, and heated to redness to expel water, which is given off without a trace of acid, the latter being retained by the oxide of lead. The vapour of water in the retort at the close of the operation is 224 QUANTITATIVE ANALYSIS. removed by carefully sucking it out with the mouth by means of a narrow tube introduced into the retort through the neck. The loss in the weight of the retort indicates the amount of water expelled. By deducting the water from the united loss of fluorine and water previously arrived at by the determination of the bases, the amount of fluorine is ascertained. VII. PHOSPHORUS. Estimation of phosphoric acid. In the analysis of com- binations of phosphoric acid with substances, the weights of which are easily determined, it is usual, having obtained the amount of every other ingredient, to consider the deficiency on the weight of the matter employed as representing the phosphoric acid. To estimate phosphoric acid, directly, it may be precipi- tated from its neutral solutions either as phosphate of lead or as phosphate of barytes ; but as this acid forms compounds of different composition with the same base, the precipitate must, in general, be subjected to analysis after its weight is determined, to ascertain the actual amount of phosphoric acid it contains. The phosphate is analyzed by ascertaining directly the proportion of base it contains, and considering the loss as the phosphoric acid. The two phosphates mentioned are analyzed by the fol- lowing processes: Phosphate of lead is dissolved in dilute nitric acid, and oxide of lead is precipitated from the solution as sulphate, by sulphuric acid ; some alcohol is added to the mixture to prevent the solution of any trace of the sulphate, the latter being wholly insoluble in dilute alcohol. The pre- cipitated sulphate of lead is filtered, washed with weak alco- hol, ignited, and weighed. From the quantity of sulphate, that of oxide of lead is calculated, and the phosphoric acid estimated as loss. Phosphate of barytes is analyzed by dis- solving it in hydrochloric acid, and precipitating barytes from the solution by sulphuric acid. From the weight of the precipitated sulphate of barytes, that of the barytes is cal- PHOSPHORUS. 225 culated, which, again, deducted from the original weight of the phosphate, gives the amount required of phosphoric acid. If phosphoric acid exists alone in solution, its amount may be determined directly, by adding, of pure and freshly calcined protoxide of lead, a known weight, which is in excess, to the solution, and evaporating to dryness. The residue, which is a mixture of phosphate of lead with the excess employed of oxide of lead, is calcined in a pla- tinum crucible, and weighed. The increase on the quantity of protoxide of lead employed is the amount of phosphoric acid formerly contained in solution. Phosphoric acid may be separated from many metallic oxides by fusion with an excess of carbonate of soda. The mass, after being heated, contains the metallic oxide, phos- phate of soda, and the excess of carbonate of soda. If the oxide is quite insoluble in a solution of the alkaline carbonate, nothing but phosphate of soda and carbonate of soda dissolves when the dry mass is treated with water. The aqueous solution is exactly neutralized by nitric acid, and acetate of lead or chloride of barium is added to precipitate phosphoric acid. The precipitated phosphate of lead or phosphate of barytes is analyzed by means of its conversion into sulphate, as before described. Separation of phosphoric acid from alumina. Phosphoric acid cannot be separated from alumina by the above method, as phosphate of alumina is soluble in solution of carbonate of soda. To analyze phosphate of alumina, Fuchs recommends that a known weight of the combination should be dissolved in a solution of caustic potash, and the alumina be precipi- tated by a solution of silicate of potash, in the form of silicate of alumina and potash. The bulky and mucilaginous pre- cipitate which falls is collected on a filter, washed, and di- gested in hydrochloric acid ; the acid mixture is evaporated to dryness to separate silica. When the dry mass which re- mains after the evaporation is moistened with hydrochloric acid, and then treated with water, it affords an insoluble re- sidue of silicic acid, while alumina dissolves, and may be precipitated from the solution by carbonate of ammonia. Q 226 QUANTITATIVE ANALYSIS. The alkaline liquid filtered from the precipitate of silicate of alumina and potash contains all the phosphoric acid. Separation of phosphoric acid from sulphuric acid. If a substance is to be analyzed which contains both phosphoric and sulphuric acids, the latter may be easily separated from the former by the solubility of phosphate of barytes in acids. Either hydrochloric or nitric acid being first mixed with the solution, on adding an excess of chloride of barium, sulphate of barytes only is precipitated, which may be weighed. If the solution filtered from the precipitate of sulphate of ba- rytes is carefully neutralized with ammonia, phosphate of barytes is precipitated, the phosphoric acid contained in which may be determined by weighing and analyzing the pre- cipitate, in the manner before described. M. Schulze has recently proposed a method of determining the amount of phosphoric acid contained in soils, rocks, wood- ashes, &c., founded on the insolubility of phosphate of the peroxide of iron, and phosphate of alumina in acetic acid. In soils, the phosphoric acid generally exists in the state of phos- phate of lime and phosphate of magnesia. If to a solution of these phosphates in acetic acid a sufficient quantity of a solu- tion of acetate of peroxide of iron be added, all the phos- phoric acid may be precipitated, in the form of phosphate of the peroxide of iron, which may be collected and weighed. Instead of acetate of iron, acetate of alumina may be em- ployed with a similar result. As soils usually contain more peroxide of iron and alumina than is sufficient to combine with the whole of the phosphoric acid they contain, all that is required to be done, in estimating the amount of this acid contained in a soil, is to add excess of ammonia to the solu- tion of the earthy matter in hydrochloric acid, and to treat the precipitate with acetic acid. Nothing but phosphate of the peroxide of iron and phosphate of alumina remains un- dissolved, and after being weighed, these phosphates may be separated from each other by dissolving the latter in solution of caustic potash.* * The insolubility of the phosphate of the peroxide of iron in acetic acid also affords a means of separating the peroxide from the protoxide of iron ; the phosphate of the latter base being soluble in acetic acid. BORON. 227 VIII. BORON. Estimation of boracic acid. Like phosphoric acid, boracic acid is best determined by the loss experienced on the ori- ginal weight of the substance containing it, when the weight of every other constituent is ascertained. The estimation of boracic acid comes thus to consist in the perfect separation and determination of the weights of all the substances with which it is associated. The separation of the metals proper from boracic acid may be easily effected by sulphuretted hydrogen or hydrosulphate of ammonia, but the separation of the earths requires particular processes. When the com- pounds which contain this acid are decomposable by concen- trated sulphuric acid, they may be analyzed by the following method, proposed by Arfvedson. A known weight of the borate to be analyzed is mixed in a platinum crucible with four times its weight of pure fluor spar, both substances being in a state of fine powder, and sufficient oil of vitriol to make a thick paste. The mixture is heated gradually to redness, and maintained at that temperature so long as any acid va- pours are given off. All the boracic acid is thus con- verted into the gaseous fluoride of boron, which is entirely disengaged, together with the excess of sulphuric acid. The residue consists of sulphate of lime produced from the de- composition of fluor spar, and sulphates of the bases with which boracic acid was previously in combination. This residue is then subjected to a full analysis, by processes described elsewhere ; and the nature and proportions of the bases being ascertained, their united weights are deducted from the original weight of the borate operated on to obtain the amount of boracic acid. In the manner just described may be analyzed all borates which are decomposable by sulphuric acid, and which do not contain water. The same process may be followed if the borate contains lime ; the quantity of fluor spar must then be accurately weighed, and the lime contained in the sub- stance analyzed be deduced from the excess of sulphate of lime formed over the calculated quantity produced from the quantity of fluor spar employed. (Rose.) Q 2 228 QUANTITATIVE ANALYSIS. Instead of fluor spar, hydrofluoric acid may be used to decompose borates. The borate, in a state of fine division, is mixed with hydrofluoric and sulphuric acids in a weighed platinum crucible : the mixture evaporated to dryness and calcined to expel all the excess of sulphuric acid, leaves a residue, consisting of sulphates of the bases formerly united with boracic acid. Borate of barytes may be analyzed by dissolving it in hydrochloric or nitric acid, and precipitating barytes from the solution, as sulphate, by sulphuric acid. The weight of the barytes contained in the sulphate being calculated, the boracic acid may be estimated as loss. C. G. Gmelin has proposed the following process for the direct determination of boracic acid. The powdered sub- stance to be analyzed is heated to redness with carbonate of soda, and the resulting mass is digested in water, to dissolve the borate formed during the fusion. The small quantities of silica and alumina which the water may also dissolve are precipitated by digestion with carbonate of ammonia. The filtered liquid is evaporated to dryness, and the residue treated with sulphuric acid, to decompose the borate of soda ; the boracic acid thus liberated is dissolved out by alcohol. The alcoholic solution, when saturated with ammonia, evapo- rated to dryness and the residue calcined, leaves pure boracic acid, whose weight may be determined. IX. NITROGEN. Estimation of nitric acid. If nitric acid exists in solution in a free state, and unmixed with any other acid, its amount may be determined by adding a sufficient quantity of barytes water to render the liquid feebly alkaline, and evaporat- ing the solution of nitrate of barytes thus obtained to dry- ness. The excess of barytes combines with carbonic acid from the atmosphere during evaporation, to form the insoluble carbonate of barytes, which therefore remains undissolved when the dry residue is treated with water. The barytes in solution is then precipitated by sulphuric acid, and weighed as sulphate. The weight of the sulphate of barytes is, of NITROGEN. 229 course, proportional to that of the nitric acid, in the ratio of single equivalents of each. When the nitric acid exists in a state of combination, it is" generally estimated as loss, the amount of the base in com- bination with the acid being ascertained, either by conversion into sulphate, or by other means proper for the particular base. If the base is one which may be precipitated from its acid solutions by sulphuretted hydrogen, that re-agent is advan- tageously employed to separate it from nitric acid. The pre- cipitated sulphuret being separated from the solution, the nitric acid set at liberty may be determined by adding a slight excess of barytes-water to the solution, and evaporating to dryness. The small quantity of sulphuret of barium formed from the sulphuretted hydrogen in solution is converted, during evaporation, into hyposulphite and sulphate of ba- rytes, both insoluble salts. When the dry mass is treated with water, nothing but nitrate of barytes dissolves, which may then be converted into sulphate and weighed. When the analysis is performed in this manner, it is necessary to cease passing sulphuretted hydrogen so soon as the nitrate is entirely decomposed, as an excess of the reagent would act on the liberated nitric acid, occasioning its decomposition and consequent loss. Such subnitrates as are insoluble in water may be decom- posed as well as if soluble, by sulphuretted hydrogen ; they should be reduced to a state of minute division, and diffused through an ounce and a half or two ounces of water. These bodies may likewise be decomposed by digestion in a solution of sulphuret of barium ; by which an insoluble metallic sul- phuret is produced, on the one hand, and the soluble nitrate of barytes on the other. The excess of sulphuret of barium is removed from the solution by a current of carbonic acid gas, which precipitates carbonate of barytes, and causes an evolution of sulphuretted hydrogen gas. The solution is filtered and evaporated to dryness, the residue redissolved in water, and barytes precipitated from the solution by sul- phuric acid. From the amount of sulphate of barytes, that of the nitric acid may be calculated. Q 3 230 QUANTITATIVE ANALYSIS. EUDIOMETRY. To separate the oxygen and nitrogen gases, and to determine the amounts of watery vapour and carbonic acid contained in atmospheric air, several processes have been recommended, possessing various claims to accuracy as well as simplicity. Of those for estimating the oxygen and nitrogen the principal are the following : (1.) Into a measure of air observed at a known temper- ature and pressure, and contained in a narrow graduated tube standing over water, a stick of phosphorus is introduced and allowed to remain for twenty-four hours, or so long as any absorption of gas is seen to occur. When the white vapours have entirely disappeared, the phosphorus is with- drawn, and the amount of contraction noticed, together with the temperature and pressure, for which, if they do not agree with the observations at the time of the first measurement, corrections must be made * : the gas remaining is nitrogen. This is the simplest method, but probably the least exact. (2.) Gay-Lussac recommends a method in which oxygen is separated from nitrogen by means of slips of copper moistened with hydrochloric acid, which absorb oxygen with rapidity. (3.) Another simple method is that in which the oxygen is estimated by observing the contraction which occurs on the ignition of a mixture of a known measure of air with one half its bulk of pure hydrogen. The ignition may be effected either by the electric spark or by means of a pellet of clay and spongy platinum. In the latter case, combination of the hydrogen and oxygen takes place without explosion ; hence the experiment may be performed in a common graduated tube standing over mercury. One third of the contraction represents the volume of oxygen contained in the mea- sure of air employed. In igniting by the electric spark, Dr. lire's eudiometer is a convenient instrument to be used. It is formed of a straight tube {fig> 21.), about twenty inches in length, sealed at one end, , and doubly bent at the middle. * The means of making these corrections are given in an Appendix. EUDIOMETRY. 231 The limb with the sealed end is graduated. Two platinum wires are sealed into the glass near the closed end, with their extremities within the tube about one tenth of an inch apart. On placing one of these wires in communication with the knob of a charged Ley den jar, and the other with the ground, the spark passes within the tube. Having prepared a mixture of air and hydro- gen in known proportions, enough of this mixture is transferred into the tube at the water or mercurial trough, to occupy about three inches of the sealed limb; the amount of the mixture introduced is accurately observed, the liquid standing in both limbs at the same level. The mouth of the open limb is now closed by a cork which is secured by a wire, and the electric spark is taken through it. After the explosion, a contraction is found to occur, to observe the amount of which, the liquid must again be brought to the same height in both limbs. One third of the contraction is the bulk of oxygen contained in the original mixture. (4.) The most accurate mode of determining the compo- sition of the air is that recently employed by M. Dumas, which is a modification of a process previously proposed by M. Brunner. After the air is passed through potash and sulphuric acid successively, to absorb carbonic acid, water, and carburetted hydrogen, it is conducted over metallic copper heated to redness in a tube of hard Bohemian glass. The copper employed is that procured from reduction of the oxide by means of hydrogen gas. Oxygen is eagerly absorbed by the copper, while nitrogen passes onwards, and is received into a weighed exhausted globe, which is again weighed at the close of the experiment. The increase in the weight of the tube containing the copper gives at once the oxygen. In this process, therefore, both the oxygen and nitrogen are determined by direct weighing, and no corrections are re- Q 4 232 QUANTITATIVE ANALYSIS. quired for changes of temperature and pressure : hence its superiority to all eudiometrical processes in which the gases are estimated by measurement. The proportion of carbonic acid gas contained in the atmo- sphere may be determined by adding gradually barytes-water or lime-water of a known strength to the air to be examined, contained in a large globe of known capacity. The liquid is dropped into the globe by means of a graduated pipette, so long as the alkaline re-action of the barytes or lime is neu- tralised, which is best ascertained by reddened litmus paper. When the paper is rendered permanently blue, more barytes or lime has been added than is necessary to form carbonate with all the carbonic acid of the air. The carbonic acid is proportional to the quantity of lime or barytes neutralized. The most ready method of determining the amount of watery vapour in the air is by the hygrometer ; but the sub- ject of hygrometry belongs rather to physics than to chemistry. The same object may be attained by causing a known quan- tity of air to pass through a tube which contains some highly hygroscopic body ; a method followed by Brunner. A cy- lindrical bottle, a, (Jig. 22.), furnished with a stop-cock b, near Fig. 22. its bottom, is connected (air-tight) at top with a small tube CARBON. 233 c, containing fragments of dry chloride of calcium ; and that is connected, by means of a caoutchouc joint, with the tube d, which contains asbestos moistened with concentrated sul- phuric acid, an extremely useful hygrometric agent. The tube d being weighed, and the apparatus mounted as in the figure, the stopcock is opened, and the water which flows out is received in the graduated vessel e. That vessel serves as the measure of the air which enters the bottle a. If passed with a moderate degree of rapidity, the air, in entering, is entirely deprived of its water by the sulphuric acid in the tube rf; the increase in weight of which at the end of the experiment is the amount of water contained in a volume of air equal to the volume of water collected in the receiver e. X. CAKBON. Analysis of carbonates. The quantity of carbonic acid contained in all carbonates which are soluble in hydrochloric or sulphuric acid in the cold may be determined by the fol- lowing simple method : A known weight of the powdered substance is introduced into a flask similar to that represented v&fig. 23. The flask should be thin at the bottom to allow of Fi 23 the application of heat, and have the capacity of three or four ounces of water. It is fitted with a cork, which has a perforation to ad- mit a small bent tube, and the latter is con- nected by means of another cork with a somewhat larger tube, , containing frag- ments of dry chloride of calcium. The extremity b of this tube is drawn out so as to be capillary. The small tube c within the flask, sealed at one end, is intended to hold hydrochloric or sulphuric acid to decompose the carbonate, and is of such length that it will not fall flat on the bottom of the flask, but rests against the side at an angle of about 45 with the bottom, so that on inclining the flask, all the acid contained in this tube can be made to flow out. The apparatus being arranged, the weighed carbonate is 234 QUANTITATIVE ANALYSIS. introduced into the empty flask, with about half an ounce of water ; the small tube c, containing sufficient hydrochloric or sulphuric acid to decompose the carbonate, is then introduced, taking care that no acid comes in contact with the car- bonate, and the flask is closed by the cork attached to the chloride of calcium tube a. The whole apparatus is now weighed, after which the flask is inclined, in order that a little of the acid in the tube c may flow out and come in contact with the carbonate, which is repeated until the latter is completely decomposed. As the evolved carbonic acid gas is dried in passing through the chloride of calcium tube , nothing else than this gas escapes, and the loss in weight of the apparatus at the close of the experiment is the weight of the carbonic acid required. But as the flask is then full of carbonic acid gas, which is considerably heavier than air, it should not be weighed in its present state. To get rid of the remaining carbonic acid, the flask is very gently heated, so as to fill it with aqueous vapour, and thus drive out the gas, the steam itself not proceeding further than the chloride of calcium tube. On the condensation of the steam, air enters the flask, which, when cold, is in the same condition as it was when weighed before the decomposition of the car- bonate, excepting only in the loss of carbonic acid. By such an operation as this, in conjunction with an alka- limetrical experiment, (page 132.) the proportions of alkaline carbonate and caustic alkali contained in a mixture of these substances may be determined. Suppose, for example, an alkalimetrical experiment to indicate 41 per cent, of soda, and the process just described 20 per cent, of carbonic acid ; then, twenty parts of carbonic acid are equivalent to 28-2 of soda, and that deducted from 41 gives 12-8 of caustic soda, and consequently 48-2 per cent, of carbonate. In the analysis of alkaline carbonates by this method, it is better to employ dilute sulphuric than hydrochloric acid, as a little hydrochloric acid gas might easily be carried away by the carbonic acid, or evaporated on heating the flask at the close of the operation ; but for earthy carbonates whose sul- phates are insoluble, the employment of sulphuric acid is ANALYSIS OF GUNPOWDER. 235 not admissible. Lime-stones and marls are analyzed very conveniently by this process. Carbonates of oxides of the metals proper, which do not contain water, or any other volatile constituent, excepting carbonic acid, may be analyzed by simply heating a known weight of the substance to redness in a weighed platinum crucible. The loss in weight experienced by heating is the amount of carbonic acid. ANALYSIS OF GUNPOWDER. Separation of carbon from sulphur and nitrate of potash. The process of analyzing gunpowder consists, first, in the separation of nitrate of potash from carbon and sulphur, and, second, in the separation of carbon from sulphur. The gunpowder, when reduced to a fine powder and weighed, is boiled in about seven times its weight of water, to dissolve the nitre. The residue, which consists of carbon and sulphur, is collected on a weighed filter, washed with tepid water so long as any nitrate of potash is dissolved out, dried and weighed. The filtered solution of nitrate of potash is evapo- rated to dryness in a weighed platinum capsule in a water-bath, and the dry saline residue weighed. This residue generally contains a small quantity of chloride of sodium, to determine the quantity of which, the dry salt, after being weighed, is dissolved in water, and the amount of chlorine it contains is estimated by precipitating and weighing as the chloride of silver. The amount of chloride of sodium which corresponds to the weight of the chloride of silver is calculated, and deducted from the weight of the nitrate of potash previously obtained. Various methods are practised to ascertain the relative proportions of carbon and sulphur. One consists in convert- ing all the sulphur into sulphuric acid, which is weighed as sulphate of barytes. For this purpose, 1 part of the dry mixture of carbon and sulphur is mixed, first with 4 parts of carbonate of potash (free from sulphate), and afterwards with 8 parts of nitre, and 6 parts of chloride of sodium. When 236 QUANTITATIVE ANALYSIS. that mixture is heated strongly in a platinum crucible, the sulphur becomes sulphuric acid, which unites with potash. After being heated, the mass is dissolved in water, and the excess of alkaline carbonate is neutralized with hydrochloric acid : sulphuric acid is then precipitated in the form of sulphate of barytes, which is washed, dried, and weighed. The quantity of sulphur which corresponds to the weight of the sulphate of barytes is calculated, and the carbon is estimated as loss. Another method of separating the sulphur and carbon, advantageous from its allowing the weights of both these elements to be taken directly, consists in dissolving out the sulphur from the mixture by means of sulphite of potash ; hyposulphite of potash is thus formed, and a residue of pure carbon remains, which is collected on a weighed filter, dried, and weighed. From the solution of hyposulphite of potash all the sulphur may be precipitated by the addition of hydro- chloric acid : the precipitate is collected on a weighed filter, washed, dried, and weighed. (Pelouze.) Two other methods are also practised ; in the first of these carbon and sulphur are separated from each other by sub- liming the latter in an atmosphere of hydrogen, and weigh- ing the residue of carbon; in the second, by heating the mixture in an atmosphere of chlorine, when the volatile chloride of sulphur is formed. But neither of these methods possesses any advantage over those described, while their execution is more difficult. Separation of carbon from iron : analysis of cast-iron. The estimation of carbon contained in cast-iron may be effected by an operation similar to that of an organic analysis. (Regnault.) From 70 to 80 grains of the specimen is reduced to powder (either by pulverizing in a mortar or by filing), and mixed with from 12 to 14 times its weight of chromate of lead, previously fused and reduced to powder. About one third or one fourth of the mixture is set aside; the remainder is mixed with as much powdered chlorate of WATER. 237 potash as the weight of cast-iron operated on, and this mixture is introduced into a tube of hard glass similar to that used in organic analysis, but much shorter. The re- maining mixture of cast-iron and chromate of lead is now added, being placed above the mixture previously intro- duced, and the tube is connected with the potash apparatus employed in organic analysis. Heat is first applied to that part of the tube, the mixture in which contains no chlorate of potash, and is advanced gradually to the other end of the tube, the heat being increased very considerably at the close of the operation, so as to fuse the chromate of lead. By this means; the whole carbon of the cast-iron is converted into carbonic acid, which is condensed by the solution of potash. It is unnecessary to give the details of this process here, as they are all included in the account of the operation of or- ganic analysis. (For the methods of estimating carbonic acid when con- tained in aqueous solution see " Analysis of mineral Waters.") XI. WATEK. Estimation of water as contained in hydrates. The ordi- nary method of determining the amount of water contained in a substance consists in expelling the water by the appli- cation of heat, and estimating it by the loss in weight which the substance then experiences. The degree of heat requi- site varies according to the facility with which the water is expelled, and also with the nature of the hydrate under ex- amination. Either a platinum or porcelain crucible may be employed in this operation. In general, this simple process gives a satisfactory result, but a large number of bodies containing water exist, which can experience a change in weight on being heated, from cir- cumstances entirely unconnected with the presence of water. For such compounds, the process must be modified. If the hydrate to be analyzed possesses the property of uniting with oxygen at a high temperature, to obtain the water, the sub- 238 QUANTITATIVE ANALYSIS. stance is heated in an atmosphere of some gas which exerts no action on it, such as carbonic acid or hydrogen ; a current of which, previously dried by passing through a chloride of calcium tube, is sustained during the whole operation. Such an apparatus as that figured at page 160. may be used; and the water given off by the substance may be collected and weighed in a small chloride of calcium tube attached to the posterior extremity of the tube c. At the end of the process, and when the substance is cold, air is admitted into the tube, to bring it to the same state as when weighed be- fore the experiment. The loss in weight indicates the amount of water. When the substance to be analyzed contains an amrno- niacal salt, that is expelled, together with the water, by ignition. The entire loss in weight being determined, the amount of ammonia contained in the substance must then be ascertained by a separate experiment ; the corresponding quantity of ammoniacal salt is calculated and deducted from the whole loss, to obtain the amount of water. In determining the proportion of water in those salts which lose the whole or a portion of their acid at a high temperature, the substance should be intimately mixed with a known weight of a strong base which can retain the acid, forming a salt which is not decomposed at the tem- perature at which water is given off. Protoxide of lead is the base generally employed for this purpose, of which oxide, from four to six times the weight of the substance may be taken. On heating, water alone is expelled, unless the compound contains an ammoniacal salt, when free ammonia is also given off. In a case where the hydrate to be analyzed is that of an organic body, if the whole of the water is not removed by heating at a temperature lower than that at which the organic matter suffers decomposition, recourse must be had to an ultimate organic analysis, provided the substance possesses a definite constitution. The whole of the water is then obtained, condensed in the chloride of calcium tube (see the account of the operation), but with it an additional WATER. 239 quantity proceeding from the combustion of the hydrogen of the organic substance; the proportion of which in rela- tion to the carbon being known, the quantity of water it should produce is calculated and deducted from the whole amount obtained. The remainder is the quantity which ex- isted in the substance as water. In the determination of combined water, care should be taken that the body is en- tirely free from hygrometric moisture, and if in a crystallized state, that it contains no water lodged mechanically between the plates of the crystals. If such is suspected, the crystals must be powdered and carefully dried by heating a few de- grees above the ordinary temperature. As almost all hydrates of definite constitution evolve dif- ferent quantities of water at different elevations of tem- perature, it is advisable to commence by heating the sub- stance gently, and to increase the temperature progressively, weighing at each increase. This is especially necessary in all accurate investigations of hydrates of salts or of organic bodies : after drying as much as possible in the open air, the substance may be kept in a confined portion of air, supported over a surface of oil of vitriol ; then in vacuo, over oil of vitriol; afterwards it may be heated to 212 by a water- bath, or higher, if necessary, by means of a boiling saturated solution of some salt, such as chloride of sodium or nitre. The proportion of water contained in inorganic hydrates which do not part with their water by being heated is esti- mated as the loss, the weights of all the other constituents of the substance being first determined. ANALYSIS OF MINERAL WATERS. (1.) Saline, Chalybeate, and Carbonated. Although the qualitative analysis of a mineral water may exhibit the presence of some bodies of rare occurrence, such as iodine, bromine, strontian, lithia, ammonia, &c., the quan- tities of these are, in general, so small as to render it worth- less, except in special investigations, to attempt to estimate 240 QUANTITATIVE ANALYSIS. their weight. The following directions for the quantitative analysis of mineral waters, will not, therefore, be compli- cated with details of the modes of estimating such bodies (for which I refer elsewhere), but comprise the means of estimating those only which are of ordinary occurrence. In saline, chalybeate, and carbonated waters, the common consti- tuents are : lime, oxide of iron, magnesia, soda, manganese, silica, sulphuric acid, chlorine, and carbonic acid. The operations to be performed in the estimation of the fixed constituents are : 1. Evaporation to dry ness to ascertain the whole amount of saline matter. 2. Estimation of the magnesia, lime, oxide of iron, oxide of manganese, and silica, contained in the portion of the residue of evaporation which is insoluble in water. 3. Estimation of the lime, magnesia, and oxide of iron, contained in the portion of the residue of evaporation which is soluble in water. 4. Estimation of the whole amount of lime, magnesia, oxide of iron, and silica. 5. Estimation of the whole amount of sulphuric acid. 6. Estimation of the whole amount of hydrochloric acid (chlorine). 7. Estimation of the whole amount of chlorides. 8. Estimation of the whole amount of chloride of sodium. 1. To obtain the entire amount of saline matter contained in the water, a known quantity is evaporated to dryness in a porcelain bason, or what is better, in a weighed platinum capsule. The evaporation should be conducted below 212 to prevent ebullition, and consequent loss from spirting ; but when the residue is quite dry, the heat should be gradually increased to 300 or 400 Fahr. The weight of the dry residue affords a check on the subsequent determinations. During evaporation, carbonic acid gas may probably be dis- engaged, and a precipitate formed at the same time; the latter may consist of carbonate of lime, carbonate of mag- nesia, and oxide of iron. These bodies were formerly held in solution by the carbonic acid, and are consequently pre- WATER. 241 cipitatcd when the latter is expelled. On treating the dry residue of the evaporation with water, this precipitate, together with some silica, and perhaps oxide of manganese, remains undissolved. The solution in water may also con- tain portions of lime, magnesia and oxide of iron, but these portions did not originally exist in the water as carbonates. 2. The second operation consists in the estimation of the lime, magnesia, oxide of iron, and silica contained in the precipitate formed on ebullition. The dry residue of evapo- ration is treated with water, and as much of the precipitate as can be easily removed is thrown on a filter and washed. What remains of the precipitate attached to the sides of the capsule, after being washed, is dissolved in hydrochloric acid, and the acid liquid passed through the filter to form a solution of the whole precipitate. If an insoluble residue exists, it is silica, which should be washed, ignited, and weighed. The solution of the precipitate in hydrochloric acid may contain lime, magnesia, oxide of iron, and manganese. Its analysis is performed according to the following table : 242 QUANTITATIVE ANALYSIS. s s II K pq I* B I W H H H ^ I I s Ij 11 ll 1 * * 's 3 ^ I'll aPf 8^|| I ! ill B-S i*i|a* '^ "i/i *' ^'i ' 'o g t -S g S tjQ |llts. 6d. a dozen. They are placed two and two, so that the bottle containing the liquid to be examined may be set by the side of any one in the series, and the colour compared by looking through the liquid upon a broad piece of white paper stretched upon a board behind the instru- ment. " To explain the table it is necessary to state that the steeping vessels we employ contain, at the proper height for receiving goods, 14-40 gallons, or 288 measures of 5 gallons each, a mea- sure being the quantity easily carried at a time. In the following table, represents water, and the numbers 1, 2, 3, &c. are the strength of the liquor already in the vessel in 12ths of a degree of Twaddell, as ascertained by the chlorimeter. If the vessel has to be set anew, we see by the first table that 32 measures of liquor at 6 must be added to (256 measures of) water to pro- duce 288 measures of liquor at T \ths of a degree. But if the liquor already in the vessel is found by the chlorimeter to produce a colour equal to the second phial, then 24? measures only are necessary, and so on." To stand #> To stand ^ requires 32 measures. 1 28 requires 24 measures. 1 _ 20 2 _ 24 2 16 3 _ 20 3 12 4 16 4 8 5 12 5 4 6 8 7 4 To stand ^ To stand & requires 16 measures. 1 12 requires 12 measures. 1 8 2 8 2 4 3 4 282 APPENDIX. TABLES FOR CALCULATIONS IN ANALYSIS. In quantitative analysis the substance whose weight is to be de- termined, instead of being weighed in its free or uncombined state, is more frequently weighed in a state of union with another body forming a definite compound and one of known composition, and from the weight of which the weight of the particular con- stituent whose amount is the object of research is calculated. To facilitate calculations of this kind the following tables have been constructed to express the quantities of one substance contained in 1-00000, 2-00000, 3-00000, 4-00000, 5'00000, 6-00000, 7*00000, 8*00000, and 9*00000 parts respectively of another substance ; and by adding together the proportions corresponding to all the figures in the number which expresses the weight of a substance, the proportion in which one body is contained in or indicated by any amount of another body may be ascertained. The first column, headed " Found," contains the names of the substances weighed ; the second, headed " Required," contains the names of the substances whose amounts are to be determined ; and the re- maining columns, headed by the first nine whole numbers, express the proportions of the "required" substance contained in or in- dicated by as many parts of the "found" substance as the number at the head of the column. When the number which expresses the quantity of the substance found contains more than one figure as its whole number, the decimal point is advanced in proportion to the right ; but if less than one whole number, the decimal point is advanced to the left. The application of these tables will be better comprehended by an example. To find the amount of chlorine contained in 21-34? grains of chloride of silver proceed as follows : 20*00 The number in the column 2, with the decimal point advanced one figure to the right - 4*9340 1- The number in column 1, without changing the * decimal point - - - -2467 3 The number in column 3, with the decimal point advanced one figure to the left - - -0740 -04 The number in column 4, with the decimal point advanced two figures to the left - -0098 21*34 grains of chloride of silver contain grains of chlorine - - - - - 52645 APPENDIX. 283 3 8 1 3 8 1 r, SGQ nious As0 oO H S ' J* "S r - 284 APPENDIX. {t m PQ Id B B "8 I So |B PP a So fo* r o Carbonic CO 2 . o APPENDIX. 285 Requi ? lorine Cl. J, o ?! . 'IS U W lorin Cl. Carbonate of lim Ca 0, C 2 . Carbonate of lime. Ca O, C O 2 . arbonate of barytes. Ba O, C O 2 . Chloride o potassium. KCI. oride of silver. AgCl. Chloride of AgCl Chloride of s NaC um, oxide of. 2 3 . 286 APPENDIX. o 00 o 05 CO i oo (N GO g >o 3 CO CO 00 o o 00 rH rH rH CO (N *^ ^ 00 i-H rH "* 00 ^ t rH g CD -tf 05 2 1 CD 00 s 00 (M CO (N o CO CO J>. O CO (N CO CO ^ i-H rH CO 1 1 il | g 1 O5 O 1 1 CO CO ^ 00 *o ^| CO C^ O5 (^ CN >0 >0 CD ** l 1 CO | 8 oo 1 GO 00 o (N 00 . s CO rH GO O5 ^>. G O5 ,t^ !> CO rH O^ ^ ^- o rH i i 5 S o >o y 1 1 g 1 GO i CO rH CO 1 8 GO O5 rH 00 O i CO CO CO CM rH CO CO * '-' o CN 00 rH I 1 o *o ^ *o . % ^ ^ O5 s 00 rH V3 ^ (M rH rH ^o i>* 00 GO >0 rH CO CO CO 6 rH CO i O5 CO O5 1 S GO CD rH O oo "M Oi CO CO CO CO CO CO H 6 (N * 6 6 rH (^ CO >0 CO o g 22 ^ K^ CO CO 00 C^ CD d 1 oo CD 5 05 CO i 5 (N O rH rH 6 6 6 ,_, 00 CO 00 CO CO 00 O5 >o J> c^ rH CO . 1 3 00 rH CO oo 00 05 O5 05 CD rH CO 6 6 6 00 6 6 6 I I o oO F o Jo I* 8 if 1,3 f o * o o O ,n o x_x * B "ydrocyanic acid. H, C 2 N. Q W a 1 1 IJ IChromate of barytes. Ba O, Cr O 3 . Ichromate of lead. Pb 0, Cr 3 . ti |o ^6 o O i I I | 1 1 Cyanide of silver. Ag, C 2 N. 1 1^ O JS APPENDIX. 287 ?' s s lO . 1" s 2 1" m g'S * CJO 288 APPENDIX. 2 S s 8 CO co 8 o o .1 Jo I 'o cc .3 T3 . _: M %3 o ^aw S * Hi IS i-^L* TJ ^^ * enic ac Se0 3 . ot of i* cc u 3 294 APPENDIX. 8 c? 6 o be 02 00 Id o* cc s| cc PP VI APPENDIX. 295 f 02 O2 T3 3 V I S nti Sr. S Stro Sr ontian SrO. I A p ( | ^^ O Q2 J~ fs 02 |~ i 02 ^*H ^^ O ^ o ver S. 02 02 Sulph sil A u 4 Stronti SrO. e. * o 01 296 APPENDIX. g s -co s 3 H lls * s^ is- ff 00" ff L. sSo" 2-'2 h " 8 Id APPENDIX. 297 TABLE or ATOMIC WEIGHTS, WITH THEIR LOGARITHMS FOR CALCU- LATIONS IN ANALYSIS. Name. Symbol. Atomic Weight. Logarithm. Aluminum Al 171-17 2-233427 Ammonia NHL 213-13 2-32864 Antimony Sb 1612-90 3-20761 oxide of Sb0 3 1912-90 3-28169 sulphuret of SbS, 2216-40 3-34565 Arsenic - 'As 940-10 2-973174 sulphuret of AsS 3 1543-6 3-18854 sulphuret of AsS s 1945-9 3-28912 Arsenic acid As0 5 1440-1 3-15839 Arsenious acid As0 3 1240-1 3-09346 Barium - Ba 856-88 2-93292 chloride of BaCl 1299-5 3-11378 Barytes - BaO 956-88 2-98086 carbonate of Ba 0, C0 2 1231-9 3-09058 phosphate of 2 BaO, PO 5 2806-1 3-44811 sulphate of Ba 0, SO, 1458-1 3-16379 Bismuth - Bi 886-92 2-947884 Boron - B 136-25 2-134336 Bromine - Br 978-31 2-990476 Cadmium Cd 696-77 2-843089 Calcium - Ca 256-02 2-408273 chloride of CaCl 698-67 2-84427 Carbon - C 75-00 1-87506 Carbonic acid C0 2 275-00 2-43933 Carbonic oxide CO 175-00 2-24304 Cerium - Ce 574-70 2-759441 Chlorine - Cl 442-65 2-646060 Chromium Cr 351-82 2-546320 Cobalt - Co 368-99 2-567015 Columbium (tantalum) Ta 2307-43 3-363128 Copper - Cu 395-70 2-59737 protoxide of CuO 495-70 2-69522 suboxide of Cu 2 891-39 2-95007 Fluorine - P 233-8 2-368845 Glucinum G 331-26 2-520168 Gold Au 2486-03 3-395506 Hydrogen H 12-48 1-09621 Iodine - I 1579-50 3-198520 Iridium - Ir 1233-50 3-091138 Iron - Fe 339-21 2-53047 peroxide of - protoxide of - Fe 2 3 FeO 978-43 439-21 2-99053 2-64267 Lead Pb 1294-5 3-11210 carbonate of Pb 0, C0 2 1669-5 3-22259 chloride of - PbCl 1737-1 3-23983 phosphate of 2Pb O, PO 5 3681-3 3-56600 protoxide of PbO 1394-5 3-14442 sulphate of - Pb 0, S0 3 1895-7 3-27777 298 APPENDIX. TABLE or ATOMIC WEIGHTS, WITH THEIR LOGARITHMS FOR CALCU- LATIONS IN ANALYSIS continued. Name. Symbol. Atomic Weight. Logarithm. Lime - CaO 356-02 2-55147 carbonate of Ca O, CO 2 631-03 2-80004 sulphate of - Ca O, SO 3 857-19 2-93308 Lithium - L 80-33 1-904878 Magnesia - MgO 258-35 2-41220 sulphate of MgO,S0 3 759-52 2-88054 Magnesium Mg 158-35 2-199618 Manganese Mn 345-9 2-53895 Srotoxide of - MnO 445-9 2-64924 eutoxide of - Mn 2 3 991-8 2-99642 Mercury - Hg 1265-8 3-10236 subchloride of - oxide of HHg 2 Cl) HgO 1487-1 1365-8 3-17234 3-13539 suboxide of J.1 *** {j 2631-6 3.42022 Molybdenum fifo 598-52 2-777078 Nickel - Ni 369-68 2-567826 Nitrogen - N 175-6 2-24452 Osmium - Os 1244-49 3-094989 Oxygen - Palladium O Pd 100-00 665-9 2-000000 2-823409 Phosphoric acid - Po s 892-31 2-95052 Phosphorous acid Po 3 692-21 2-84030 Phosphorus p 392-31 2-59363 Platinum - Pt 1233-3 3-09107 ammonio-chloride of f(Pt C1 2 + 1 NH 4 C1) 2786-5 3-445059 potassio-chloride of f(PtCl 2 + 1 KC1) 3051-2 3-48447 Potash - KO 589-92 2-77079 carbonate of KO, C0 2 864-93 2-93698 sulphate of KO, S0 3 1091-1 3-03786 Potassium K 489-92 2-69013 chloride of KC1 932-57 2-96968 Rhodium - R 651-39 2-813841 Selenium - Se 494-58 2-694236 Silicon - '?- Si 277-31 2-442965 Silver Ag 1351-6 3-13085 chloride of - AgCl 1794-3 3-25389 oxide of - AgO 1451-6 3-16185 ditto 2 2903-2 3-46288 ditto 3 4354-8 3-63897 ditto 4 5806-4 3-76391 ditto 5 7258-0 3-86082 Soda NaO 390-9 2-59207 carbonate of Na O, CO 2 665-91 2-82341 sulphate of - Na 0, S0 3 892-07 2-95040 Sodium Na 290-9 2-46374 chloride of NaCl 733-55 2-86543 APPENDIX. 299 TABLE or ATOMIC WEIGHTS, WITH THEIR LOGARITHMS FOR CALCU- LATIONS IN ANALYSIS continued. Name. Symbol. AtomicWeight. Logarithm. Strontian SrO 647-29 2-81110 carbonate of Sr O, CO 2 922-3 2-96487 sulphate of Sr 0, S0 3 1148-5 3-060131 Strontium Sr 547-29 2-738217 chloride of SrCl 989-94 2-99561 Sulphur - S 20M7 2-30356 Sulphuric acid S0 3 501-17 2-69998 Sulphurous acid - S0 2 401-17 2-60333 Tellurium Te 801-76 2-904044 Thorium - Th 744-9 2-872098 Tin Sn 735-29 2-86645 peroxide of - protoxide of - SnO 2 SnO 935-29 835-29 2-97095 2-92184 Titanium - Ti 303-66 2-482387 Tungsten - W 1183-00 3-072985 Uranium - U 2711-36 3-433187 Vanadium V 856-89 2-932924 Water - HO, 112-48 2-05107 Ditto 2 224-96 2-35210 Ditto ... 3 337-44 2-52820 Ditto 4 449-92 2-65314 Ditto 5 562-40 2-75005 Ditto 6 674-88 2-82923 Ditto ... 7 787-36 2-89618 Ditto 8 899-84 2-95417 Ditto - 9 1012-3 3-00531 Ditto ... 10 1124-8 3-05107 Ditto 11 1237-3 3-09248 Ditto ... 12 1349-8 3-13027 Ditto 13 1462-2 3-16501 Yttrium - - - Y 402-51 2-604776 Zinc Zn 403-23 2-60555 oxide of - ZnO 503-23 2-70177 2 1006-5 3-00281 3 1509-7 3-17889 4 2012-9 3-30382 Zirconium Zr 420-20 2-623456 INDEX. ABBREVIATIONS, list of, 16. Acid, acetic, behaviour of, with re- agents, 46, 47. , antimonic, behaviour of, with re- agents, 42 45. , antimonic, estimation of, 199. , antimonious, behaviour of, with reagents, 42 45. , arsenic, behaviour of, with re- agents, 4245. , arsenic, estimation of, 202. , arsenious, behaviour of, with re- agents, 42 45. , boracic, behaviour of, with re- agents, 46, 47. , bromic, behaviour of, with re- agents, 46, 47. -, carbonic, behaviour of, with re- agents, 46, 47. , chloric, behaviour of, with re- agents, 46, 47. , chromic, behaviour of, with re- agents, 42 45. , chromic, estimation of, 170. , citric, behaviour of, ^vith re- agents, 46, 47. , formic, behaviour of, with re- agents, 46, 47. , hydrobromic, behaviour of, with reagents, 48, 49. , hydrochloric, 12. , hydrochloric, behaviour of, with reagents, 48, 49. , hydrochloric, detection of, in or- ganic mixtures, 123. r, hydrochloric, estimation of, 217. , hydrocyanic, behaviour of, with reagents, 48, 49. , hydrocyanic, detection of, in or- ganic mixtures, 124. , hydrofluoric, behaviour of, with reagents, 48, 49. , hydriodic, behaviour of, with re- agents, 48, 49. > , hydroselenic, behaviour of, with reagents, 48, 49. Acid, hydrosulphocyanic, behaviour of, with reagents, 48, 49. , hydrosulphuric, behaviour of, with reagents, 48, 49. , hypermanganic, behaviour of, with reagents, 42 45. , hyposulphuric, behaviour of, with reagents, 50, 51. , hyposulphurous, behaviour of, with reagents, 50, 51. , iodic, behaviour of, with reagents, 50, 51. , manganic, behaviour of, with re- agents, 42 45. , molybdic, behaviour of, before blowpipe, 102, 103. , molybdic, behaviour of, with re- agents, 4245. , nitric, 13. , nitric, behaviour of, with reagents, 50,51. , nitric, detection of, in organic mixtures, 122. , nitric, estimation of, 228. , nitrous, behaviour of, with re- agents, 50, 51. ,osmic, behaviour of, with reagents, 4245. , oxalic, 13. , oxalic, behaviour of, with re- agents, 50, 51. , oxalic, detection of, in organic mixtures, 123. , perchloric, behaviour of, with re- agents, 50, 51. , phosphoric, behaviour of, with re- agents, 52, 53. , phosphorous, behaviour of, with reagents, 52, 53. , selenic, behaviour of, with re- gents, 52, 53. , selenious, behaviour of, with re- agents, 42 45. , silicic, behaviour of, with reagents, 52, 53. , silicic, blowpipe test for, 104. , silicic, estimation of, 207. 302 INDEX. Acid, sulphuric, behaviour of, with re- agents, 52, 53. , sulphuric, detection of, in organic mixtures, 122. , sulphuric, estimation of, 212. , sulphurous, behaviour of, with reagents, 52, 53. , sulphurous, estimation of, 213. ., tartaric, 13. , tartaric, behaviour of, with re- agents, 52, 53. , titanic, estimation of, 171. , tungstic, behaviour of, with re- agents, 42 45. , tungstic, behaviour of, before blowpipe, 102, 103. . . , vanadic, behaviour of, with re- agents, 42 45. , uranic, estimation of, 169. Alkalies, behaviour of, before blowpipe, 98, 99. , estimation of, in silicates, 210. , fixed, separation of, from alu- mina, 146. , separation of, from lime, 139. , separation of, from magnesia, 142. Alkaline mineral waters, analysis of, 247. sulphurets, analysis of, 216. Alkalimeter, 132. Alkalimetry, 132. Alumina, behaviour of, before blowpipe, 98, 99. , behaviour of, with reagents, 22 25. , estimation of, 144. , separation of, from barytes, 146. , separation of, from fixed alkalies, 146. , separation of, from glucina, 148. , separation of, from lime, 145. , separation of, from phosphoric acid, 225. , separation of, from protoxide of manganese, 150. , separation of, from magnesia, 1 45. , separation of, from magnesia, lime and alkalies, 144. , separation of, from peroxide of iron, 155. , separation of, from strontian, 146. Aluminous minerals, solution of, 146. Amalgams, estimation of, mercury in, 185. Ammonia, 14. , behaviour of, with reagents, 18 21. , carbonate of, 14. , estimation of, 135. Ammonia, hydrosulphate of, 15. , muriate of, 15. , oxalate of, 13. Ammonio-nitrate of silver, 118. Ammonio-sulphate of copper, 118. Alloys, estimation of silver in, 179. Analysis of alkaline mineral waters, 247. of alkaline sulphurets, 216. of borate of barytes, 228. of borates, 227. of brass, 178. of bromides, 222. of calomel, 218. of carbonated waters, 239. of carbonates, 233. of cast-iron, 236. of chalybeate waters, 239. of chloride of lead, 217. of chloride of silver, 217. of chlorides, 217. of chromate of lead, 173. of chrome-iron, 1 70. of clay ironstone, 1 55. of coal gas, 114. of colophonite, 209. of earthy sulphurets, 216. of fluorides, 222. of gunpowder, 235. of iodides, 220. of metallic sulphurets, 214. of molybdate of lead, 2OO. of native oxide of tin, 1 96. of natural silicates, 207. of phosphate of barytes, 224. of phosphate of lead, 224. of platinum ore, 186. of platinum ore, table of, 193, 194. of saline mineral waters, 239. of sulphate of barytes, 213. of sulphate of lead, 21 2. of sulphate of lime, 213. of sulphate of strontian, 21 3. - of sulphureous mineral waters, 248. of sulphuret of bismuth, 215. of sulphuret of lead, 215. of sulphuret of mercury, 215. of sulphuret of silver, 215. of tungstate of lime, 201. of volatile organic substances, 261. of wolfram, 2O2. , organic, 254. , organic, by chromate of lead, 263. , organic, by mixture of oxide of copper arid chlorate of potash, 262. , organic, determination of nitro- gen in, 264. INDEX. 303 Analysis, qualitative, 54. , qualitative, of Berlin silver, 76. , qualitative, of hornblende, 84. , qualitative, of lepidolite, 84. , qualitative, of mineral waters, 85. , qualitative, of Newton's fusible metal, 77. , qualitative, of silicates, 78. , qualitative, of stilbite, 84. , qualitative, of urine, 124. , quantitative, of mineral waters, 239. , tables for calculations in, 281 296. Antimonic acid, behaviour of, with re- agents, 42 45. , estimation of, 199. Antimonious acid, behaviour of, with reagents, 42 45. Antimoniurets, blowpipe test for, 105. Antimony, detection of, in organic mixtures, 120. , estimation of, 1 96. , oxide of, behaviour of, before blowpipe, 100, 101. , oxide of, behaviour of, with re- agents, 30 33. , separation of, from arsenic, 205. , separation of, from other metals, 198. , separation of, from tin, 199. Arseniates, blowpipe tests for, 105. Arsenic, detection of, in organic mix- tures, 115. , estimation of, 202. , reduction test for, 118. , separation of, from antimony, 205. , separation of, from other metals, 204. Arsenic acid, behaviour of, with re- agents, 42 45. , separation of, from oxide of zinc, note, 204. , separation of, from other bodies, 204. Arsenious acid, behaviour of, with re- agents, 42 45. , separation of, from other bodies, 204. Arsenites, blowpipe test for, 105. Atomic constitution, how to calculate, 269. Atomic weights, with their logarithms, 297299. Barium, chloride of, 14. , estimation of, 136. Barytes, acetate of, 15. Barytes, analysis of borate of, 228. , analysis of sulphate of, 213. Barytes, behaviour of, before blowpipe, 98, 99. , behaviour of, with reagents, 18 21. , estimation of, 136. , nitrate of, 14. , separation of, from alumina, 146. , separation of, from lime, 139. , separation of, from strontian, 137. Berlin silver, qualitative analysis of, 76. Bismuth, analysis of sulphuret of, 215. , estimation of, 174. , oxide of, behaviour of, before blowpipe, 100, 101. , oxide of, behaviour of, with re- agents, 30 33. ' , separation of, from lead, 175. Black's blowpipe, 93. Blowpipe, use of, in qualitative analysis, 92. Boracic acid, behaviour of, with re- agents, 46, 47. , estimation of, 227. Borate of barytes, analysis of, 228. Borates, analysis of, 227. , blowpipe tests for, 1 04. Borax, colours of beads of, 109, 110. Brass, analysis of, 178. Bromic acid, behaviour of, with reagents, 46, 47. Bromides, analysis of, 222. , blowpipe test for, 105. Bromine, estimation of, 222. , separation of, from iodine, 222. Cadmium, estimation of, 174. , oxide of, behaviour of, before blowpipe, 100, 101. , oxide of, behaviour of, with re- agents, 30 33. , estimation of, 138. Calculations, tables for, in analysis, 281 296. Calculi containing silica, 127. Calculus, ammonio-magnesian phos- phate, 126. , bone-earth, 126. , carbonate of lime, 127. , cystic oxide, 126. , fusible, 126. , oxalate of lime, 1 27. , uric acid, 126. , xanthic oxide, 127. Calomel, analysis of, 218. Carbonated mineral waters, analysis of, 239. Carbonates, analysis of, 233. Carbon, estimation of, in organic bodies, 254. , separation of, from iron, 236. 304 INDEX. Carbon, separation of. from sulphur, 235. Carbonic acid, behaviour of, with re- agents, 46, 47. Carbonic acid gas, how to determine quantity of, in atmosphere, 232. Cast-iron, analysis of, 236. Cerium, estimation of, 149. , oxide of, behaviour of, before blowpipe, 100, 101. , oxide of, behaviour of, with re- agents, 22 25. Chalybeate mineral waters, analysis of, 239. Chloric acid, behaviour of, with re- agents, 46, 47. Chlorides, analysis of, 217. , blowpipe test for, 105. Chlorimetry, 218. ; Appendix, 279. Chlorine, estimation of, 217. , separation of, from iodine, 221. Chromate of lead, analysis of, 173. , use of, in organic analysis, 263. Chromic acid, behaviour of, with re- agents, 42 45. Chrome-iron, analysis of, 170. Chromic acid, estimation of, 170. Chromium, estimation of, 170. , oxide of, behaviour of, before blowpipe, 100, 101. , oxide of, behaviour of, with re- agents, 22 25. , separation of oxide of, from oxide of iron, i70. Citric acid, behaviour of, with reagents, 46, 47. Clay ironstone, analysis of, 155. Coal gas, analysis of, 114. Cobalt, estimation of, 159. , oxide of, behaviour of, before blowpipe, 100, 101. , oxide of, behaviour of, with re- agents, 22 25. , separation of oxide of, from oxide of nickel, 161. , separation of oxide of, from oxide of zinc, 165. , separation of oxide of, from per- oxide of iron, 164. Colophonite, analysis of, 209. Constitution, how to calculate atomic, 269. Copper, detection of, in organic mix- tures, 120. , estimation of, 176. , oxide of, behaviour of, before blowpipe, 100, 101. , protoxide of, behaviour of, with reagents, 30 33. Copper, estimation of suboxide of, 178. , preparation of oxide of, for or- ganic analysis, 256. , separation of, from lead, 178. , separation of oxide of, from prot- oxide of mercury, 185. , separation of, from zinc, 178. , suboxide of, behaviour of, with reagents, 30 33. Cupellation, 180. Density, 270. of gases, how to determine, 274. of liquids, how to determine, 271. of solids, how to determine, 272. of vapours, how to determine, 277. Earths, behaviour of, before blowpipe, 98, 99. Earthy sulphurets, analysis of, 216. Endiometer, Dr. Ure's, 231. Endiometry, 230. Filtration, 6. Filter, ignition of, 8. Flame, construction of, 94. Flame, oxidating, 94. , reducing, 95. Fluorides, analysis of, 222. , blowpipe test for, 105. Fluorine, estimation of, 222. Formic acid, behaviour of, with re- agents, 46, 47. Fusible metal, qualitative analysis of Newton's, 77. Gas, coal, analysis of, 114. Gases, change in bulk of, by aqueous vapour, 276. Gases, corrections for changes in bulk of, note, 275. Gases, how to determine density of, 274. Gases, mixed, analysis of, 111, et seq. Glucina, behaviour of, before blowpipe, 98, 99. , behaviour of, with reagents, 22 25. , estimation of, 147. , separation of, from alumina, 148. , separation of, from magnesia, 148. Gold, estimation of, 180. , oxide of, behaviour of, with re- agents, 3033. , separation of, from silver, 181. INDEX. 305 Griffin's blowpipe, 93. Gunpowder, analysis of, 235. Hardness of waters, determination of degree of, 249. Hornblende, qualitative analysis of, 48. Hydriodic acid, behaviour of, with re- agents, 48, 49. Hydrobromic acid, behaviour of, with reagents, 48, 49. Hydrochloric acid, behaviour of, with reagents, 48, 49. , detection of, in organic mixtures, 123. , estimation of, 217. Hydrocyanic acid, behaviour of, with reagents, 48, 49. detection of, in organic mixtures, 124. Hydrofluoric acid, behaviour of, with reagents, 48, 49. Hydrogen, estimation of, in organic bodies, 254. Hydroselenic acid, behaviour of, with reagents, 48, 49. Hydrosulphocyanic acid, behaviour of, with reagents, 48, 49. Hydrosulphuric acid, behaviour of, with reagents, 48, 49. Hypermanganic acid, behaviour of, with reagents, 42 45. Hyposulphuric acid, behaviour of, with reagents, 50, 51. Hyposulphurous acid, behaviour of, with reagents, 50, 51. lodic acid, behaviour of, with reagents, 50, 51. Iodides, analysis of, 220. , blowpipe test for, 105. , estimation of, 220. Iodine ley, estimation of iodine in, 221. Iodine, separation of, from bromine, 222. , separation of, from chlorine, 221. liridic oxide, behaviour of, with re- agents, 30 33. Iridium, estimation of, 186. Iron, analysis of cast, 236. , estimation of, 153. , peroxide of, behaviour of, before blowpipe, 100, 101. , peroxide of, behaviour of, with reagents, 22 25. , protoxide of, behaviour of, with reagents, 2225. , separation of, from carbon, 236. , separation of, from other bodies, 153. Iron, separation of oxide of, from oxide of chromium, 170. , separation of peroxide of, from alumina, 1 55. , separation of peroxide of, from earths and alkalies, 155. , separation of peroxide of, from magnesia, 154. , separation of peroxide of, from oxide of cobalt, 164. , separation of peroxide of, from oxide of nickel, 157. , separation of peroxide of, from oxide of zinc, 168. , separation of peroxide of, from protoxide of iron, 155., note. , separation of peroxide of, from protoxide of manganese, 154. , separation of uranium from, 169. Ironstone, clay, analysis of, 155. Lamp, Berzelius's, for blowpipe ex- periments, 95. , Rose's, 9. Lantanum, oxide of, behaviour of, with reagents, 22 25. Lead, acetate of, 15. , analysis of chloride of, 217. , analysis of chromate of, 173. , analysis of molybdate of, 200. , analysis of phosphate of, 224. , analysis of sulphate of, 212. , analysis of sulphuret of, 215. , detection of, in organic mixtures, 121. , estimation of, 172. , oxide of, behaviour of, before blowpipe, 100, 101. , oxide of, behaviour of, with re- agents, 34 37. , separation of, from bismuth, 1 75. , separation of, from copper, 178. Lepidolite, qualitative analysis of, 84. Lime, analysis of sulphate of, 213. , analysis of tungstate of, 201. , behaviour of, before blowpipe, 98, 99. , behaviour of, with reagents, 1 8 21. , estimation of, 138. , separation of, from alkalies, 139. , separation of, from alumina, 145. , separation of, from barytes, 139. , separation of, from magnesia, 141. >, separation of, from protoxide of manganese, 156. , separation of, from strontian, 138. 306 INDEX. Liquids, how to determine density of, 271. Lithia, behaviour of, before blowpipe, 98, 99. , behaviour of, with reagents, 18 21. , estimation of, 135. , separation of, from potash and soda, 136. Logarithms of atomic weights, 297 299. Magnesia, behaviour of, before blow- pipe, 98, 99. , behaviour of, with reagents, 1 8 21. , Magnesia, estimation of, 139. , separation of, from alumina, 145. , separation of, from fixed alkalies, 142. , separation of, from glucina, 148. , separation of, from lime, 141. , separation of, from peroxide of iron, 154. , separation of, from protoxide of manganese, 151. Manganese, dentoxide of, behaviour of, with reagents, 22 25. , estimation of, 150. , protoxide of, behaviour of, with reagents, 22 25. -, oxide of, behaviour of, before blowpipe, 100, 101. , separation of protoxide of, from alumina, 150. , separation of, from nickel, 158. , separation of protoxide of, from magnesia and lime, 151. , separation of protoxide of, from oxide of zinc, 165. , separation of protoxide of, from peroxide of iron, 154. , valuation of peroxide of, 152. Marsh's test for arsenic, 116. Mercury, analysis of subchloride of, 218. , analysis of sulphurets of, 215. , detection of, in organic mixtures, 120. , estimation of, 182. , estimation of, in amalgams, 185. , protoxide of, behaviour of, with reagents, 34 37. , separation of protoxide of, from oxide of copper, 185. , separation of oxides of, from oxide of silver, 184. , suboxide of, behaviour of, with reagents, 34 37. Metallic oxides, classification of, 17. Microcosmic salt, colours of beads of, 110, 111. Mineral waters, analysis of alkaline, 247. , qualitative analysis of 85. , quantitative analysis of, 239. , analysis of sulphureous, 248. Molybdate of lead, analysis of, 2OO. Molybdeum, estimation of, 20O. Molybdic acid, behaviour of, before blowpipe, 102, 103. acid, behaviour of, with reagents, 4245. acid, estimation of, 200. oxide, behaviour of, with reagents, 3437. Molybdous oxide, behaviour of, with reagents, 34 37. Nickel, oxide of, behaviour of, before blowpipe, 102, 103. , oxide of, behaviour of, with re- agents, 26 29. , estimation of, 156. , separation of alumina from oxide of, 159. , separation of, from manganese, 158. , separation of magnesia from oxide of, 159. , separation of oxide of, from oxide of cobalt, 161. , separation of oxide of, from oxide of zinc, 165. , separation of oxide of, from per- oxide of iron, 157. Nitrates, blowpipe test for, 104. Nitric acid, behaviour of, with reagents, 50, 51. , detection of, in organic mixtures, 122. , estimation of, 228. Nitrogen, determination of, in organic substances, 264. Nitrous acid, behaviour of, with re- agents, 50, 51. Organic analysis, 254. Organic substances, analysis of volatile, 261. Osmic oxide, behaviour of, with re- agents, 34 37. Osmic acid, behaviour of, with reagents, 42 45. Osmium, estimation of, 186. Oxalic acid, behaviour of, with reagents, 50, 51. INDEX. 307 Oxalic acid, detection of, in organic mixtures, 123. Oxides, metallic, behaviour of, before blowpipe, 100 103. Palladious oxide, behaviour of, with reagents, 34 37. Palladium, estimation of, 186. Perchloric acid, behaviour of, with re- agents, 50, 51. Platinic oxide, 38 41. Platinous oxide, behaviour of, with re- agents, 38 41. Platinum, estimation of, 186. Platinum ore, analysis of, 186. , tables of the analysis of, 193, 194. Phosphate of barytes, analysis of, 224. Phosphate of lead, analysis of, 224. Phosphates, blowpipe test for, 104. Phosphoric acid, behaviour of, with re- agents, 52, 53. , estimation of, 224. , separation of, from alumina, 225. , separation of, from metallic oxides, 225. , separation of, from sulphuric acid, 225. Phosphorous acid, behaviour of, with reagents. 52, 53. Poisons, detection of, in organic mix- tures, 115. Potash, behaviour of, before blowpipe, 98, 99. , behaviour of, with reagents, 18 21. , bicarbonate of, 14. , binoxalate of, 13. , carbonate of, 14. , caustic, 13. , chromate of, 15. , estimation of, 130. , separation of, from soda, 131. Potassium, estimation of, 130. , iodide of, 16. Precipitation, 4 6. Qualitative analysis, manipulations in, 1. Quantitative analysis, manipulations in, 4. , observations on, 128. Reagent, 2. Reagents, list of, with impurities, 12. Reduction, test for arsenic, 118. Rhodic oxide, behaviour of, with re- agents, 38 41. Rhodium, estimation of, 186. Saline mineral waters, analysis of, 239. Seleniates, blowpipe test for, 104. Selenic acid, behaviour of, with re- agents, 52, 53. Selenious acid, behaviour of, with re- agents, 42 45. Selenites, blowpipe test for, 104. Seleniurets, blowpipe tests for, 105. Silica, estimation of, 206. Silicates, analysis of natural, 78, 207. , constituents of, 80. , estimation of alkalies in, 210. , list of, decomposable by hydro- chloric acid, 2O8. , list of, net decomposable by hydrochloric aci. 1 , 208. , qualitative analysis of, 78. , solution of, 78. Silicic acid, behaviour of, with re- agents, 52, 53. , blowpipe test for, 104. , estimation of, 207. Silver, analysis of chloride of, 217. , analysis of sulphuret of, 215. , estimation of, 179. , nitrate of, 1 4. , oxide of, behaviour of, before blowpipe, 102, 103. , oxide of, behaviour of, with re- agents, 3841. , separation of, from gold, 181. , separation of oxide of, from oxides of mercury, 1 84. Soda, behaviour of, before blowpipe, 98, 99. , behaviour of, with reagents, 1 8 21. , carbonate of, 14. , estimation of, 131. , separation of, from potash, 131. , sulphate of, 12. Sodium, estimation of, 131. Solids, how to determine density of, 272. Solution of aluminous minerals, 146. Solution of silicates, 78. Specific gravity, 270. gravity of gases, how to deter- mine, 274. gravity of liquids, how to deter- mine, 271. gravity of solids, how to deter- mine, 272. gravity of vapours, how to de- termine, 277. Stilbite, qualitative analysis of, 84. Strontian, analysis of sulphate of, 213. , behaviour of, before blowpipe, 98, 99. 308 INDEX. Strontian, behaviour of, with reagents, 1821. , estimation of, 137. , separation of, from alumina, 146. , separation of, from barytes, 137. , separation of, from lime, 138. Strontium, estimation of, 137. Sulphate of barytes, analysis of, 213. of lead, analysis of, 212. of lime, analysis of, 213. of strontian, analysis of, 203. Sulphates, blowpipe tests for, 1O4. Sulphureous mineral waters, analysis of, 248. Sulphuret of bismuth, analysis of, 215. of lead, analysis of, 215. of mercury, analysis of, 215. of silver, analysis of, 215. Sulphuretted hydrogen, 14. Sulphurets, analysis of alkaline, 216. , analysis of earthy, 216. , analysis of metallic, 214. , blowpipe tests for, 105. Sulphuric acid, 12. , behaviour of, with reagents, 52, 53. , detection of, in organic mixtures, 122. , estimation of, 212. Sulphurous acid, behaviour of, with re- agents, 52, 53. , estimation of, 213. Sulphur, separation of, from carbon, 235. Sulphuric acid, separation of, from phosphoric acid, 225. Supports in blowpipe experiments, 95. Symbols, list of, 16. Tantalic acid, behaviour of, with re- agents, 26 29. Tantalum, oxide of, behaviour of, before blowpipe, 102, 103. Tartaric acid, behaviour of, with re- agents, 52, 53. Telluric oxide, behaviour of, with re- agents, 3841. Tellurets, blowpipe test for, 105. Tellurium, oxide of, behaviour of, be- fore blowpipe, 102, 103. Test glass, 2. : tube, 3. Thorina, behaviour of, with reagents, 2629. , estimation of, 147. Tin, analysis of, native oxide of, 1 96. , estimation of, 195. , means of separating other bodies from, 195. Tin, oxide of, behaviour of, before blowpipe, 102, 103. , peroxide of, behaviour of, with reagents, 38 41. , protochloride of, 15. , protoxide of, behaviour of, with reagents, 38, 39, 40, 41. , separation from antimony, 1 99. Titanic acid, behaviour of, with re- agents, 26, 27, 28, 29. , estimation of, 171. Titanium, oxide of, behaviour of, before blowpipe, 102, 103. Tungstate of lime, analysis of, 201 . Tungstic acid, behaviour of, before blowpipe, 102, 103. , behaviour of, with reagents, 42 45. , estimation of, 201. Uranium, estimation of, 169. , oxide of, behaviour of, before blowpipe, 102, 103. , oxide of, behaviour of, with re- agents, 26 29. , peroxide of, 26 29. , separation of iron from, 169. Urine, qualitative analysis of, 124. Vapours, how to determine density of, 277. Valuation of peroxide of manganese, 152. Vanadic acid, behaviour of, with re- agents, 42 45. Vanadium, oxide of, behaviour of, be- fore blowpipe, 102, 103. , oxide of, behaviour of, with re- agents, 2629. Voigt's blowpipe, 93. Volatile precipitates, drying and igni- tion of, 11. Water, estimation of, 237. Waters, how to determine degree of hardness of, 249. , qualitative analysis of mineral, 85. , quantitative analysis of mineral, 239. Weights, atomic, with their logarithms, 295 297. Wolfram, analysis of, 202. Yttria, behaviour of, before blowpipe, 98, 99. , behaviour of, with reagents, 26 29. , estimation of, 147. INDEX. 309 Yttria, separation of, from earths and alkalies, 147. Zinc, oxide of, behaviour of, before blowpipe, 102, 103. , oxide of, behaviour of, with re- agents, 26 29. , estimation of, 164. , separation of arsenic acid from oxide of, 204. note. , separation of, from copper, 178. , separation of oxide of, from pe- roxide of iron, 168. Zinc, separation of oxide of, from oxide of cobalt, 165. , separation of oxide of, from oxide of manganese, 165. , separation of oxide of, from oxide of nickel, 165. Zirconia, behaviour of, before blow- pipe, 98, 99. , behaviour of, with reagents, 26 29. , estimation of, 149. THE END. LONDON : Printed by A. SPOTTISWOOOE, New Street- Square. WORKS OX CHEMISTRY PRINTED I OR TAYLOR AND WALTON, UPPER GOWER STREET. PROFESSOR LIEBIG'S NEW WORK. In 8vo., price 9-y. 6d. cloth, ANXMAX. CHE1VIISTRY, OR ORGANIC CHEMISTRY, IN ITS APPLICATIONS TO PHYSIOLOGY AND PATHOLOGY. By JUSTUS LIEBIG, M.D., F.R.S., Professor of Chemistry in the Univer- sity of Giessen. Edited, from the MSS. of the Author, by WILLIAM GREGORY, M.D., F.R.S.E., Professor of Chemistry, King's College, Aberdeen. " While we have given but a very imperfect sketch of this original and pro- found work, we have endeavoured to convey to the reader some notion of the rich store of interesting matter which it contains. The chemist, the physiolo- gist, the medical man, and the agriculturist, will all find in this volume many new ideas, and many useful practical remarks. It is the first specimen of what modern Organic Chemistry is capable of doing for physiology ; and we have no doubt that from its appearance physiology will date a new era in her advance." Quarterly Review, No. 139. LIEBIG'S AGRICULTURAL CHEMISTRY. Small 8vo., 9*. 6d. CHEMISTRY, ITS APPLICATION TO AGRICULTURE AND PHYSIOLOGY. By JUSTUS LIEBIG, M.D., PH.D., Professor of Chemistry in the Univer- sity of Giessen. Edited from the MSS. of the Author by LYON PLAY- FAIR, PH.D. Second and cheap Edition, with very numerous additions, in- chiding a New Chapter on Soih. " It is not too much to say, that the publication of Professor Liebig's Organic Chemistry of Agriculture constitutes an era of great importance in the history of Agricultural Science. Its acceptance as a standard is unavoidable, for following closely in the straight path of inductive Philosophy, the conclusions which are drawn from its data are incontrovertible." "We can truly say that we have never risen from the perusal of a book with a more thorough conviction of the pro- found knowledge, extensive reading, and practical research of its author, and of the invincible power and importance of its reasonings and conclusions, than we have gained from the present volume." Sillimari's Journal, January, 1841. " Every page contains a mass of information. I would earnestly advise all practical men, and all interested in cultivation, to have recourse to the book itself. The subject is vastly important, and we cannot estimate how much may be added to the produce of our fields by proceeding on correct principles." 's Gardener K Magazine for March, 1841. TURNER'S CHEMISTRY. NEW EDITION. * Just published, in one thick Volume 8vo., including the Oily Acids, Price I/. 5s. , ELEMENTS OF CHEMISTRY, Including the actual state and prevalent doctrines of the Science. By the late EDWARD TURNER, M.D. Seventh Edition. Edited by JUSTUS LIEBIG, M.D., PH.D., F.R.S., &c., Professor of Chemistry in the Uni- versity of Giessen ; and WILLIAM GREGORY, M.D., Professor of Chemis- try in King's College, Aberdeen. *#* A second Supplement, completing the Work, will be ready very shortly. " To the student in Chemistry it is decidedly the best adapted of any manual in the English language. It is, indeed, a condensed epitome of the Science. This is one of the greatest recommendations to those who wish to possess a work giving, in detail and consecutive connection, the facts and doctrines of the Science." The Chemist, January, 1841. Just imported, EIGHT FOLIO PLATES, REPRESENTING THE CHEMICAL LABORATORY OF THE UNIVERSITY OF GIESSEN. Accompanied by a Pamphlet, in German, descriptive of the same. By Dr. JUSTUS LIEBIG. Price 12s. SCHOOLS OF CHEMISTRY. A LETTER TO THE EARL OF ABERDEEN ON THE STATE OF THE SCHOOLS OF CHEMISTRY IN THE UNITED KINGDOM. By WILLIAM GREGORY, M.D., F.R.S.E., Professor of Medicine and Chemistry in the University and King's College, Aberdeen. 8vo., Is. sewed. " We strongly recommend an attentive perusal of the pamphlet to all who are interested in the advancement of chemical knowledge in this country. The system here advocated has already been productive of important discoveries, for to the school of Giessen alone ' we are indebted for a very large proportion of those researches which have advanced organic chemistry to its present very flourishing condition.'" Medical Gazette, March 18. 1842. PRINTED FOR TAYLOR AND WALTON, 28. UPPER GOWER STREET. UPPER GOWER STREET, FEBRUARY, 1844. WORKS PRINTED FOR TAYLOR AND WALTON, BOOKSELLERS AND PUBLISHERS TO UNIVERSITY COLLEGE. NEW EDITION BY DR. SHARPEY AND MR. QUAIN. One Volume Octavo, illustrated bj, very numerous Engravings on Wood, many for the first time contained in this Edition, DR. QUAIN'S ELEMENTS OF ANATOMY. JFtftfj emtton. EDITED BY RICHARD QUAIN, PROFESSOR OP ANATOMY, AND WILLIAM SHARPEY, M.D., PROFESSOR OF ANATOMY AND PHYSIOLOGY IN UNIVERSITY COLLEGE, LONDON. *** Part I., 500 Pages and 150 Woodcuts, is now ready, price 13*. NEW DISSECTOR. In One Volume, crown 8vo, 776 pages, 12*. DEMONSTRATIONS OF ANATOMY; A GUIDE TO THE DISSECTION OF THE HUMAN BODY. By GEORGE VINER ELLIS, One of the Demon- strators of Anatomy in University College. " We think Mr. Ellis's ' Demonstrations' are in every way fitted for the purposes for which they are intended, and we therefore strongly recommend the work to the notice of every member of the profession. We are con- vinced that it will quickly become the general Text Book of every working student in anatomy." British and Foreign Medical Review, January 1841. COMPLETION OF MULLER'S PHYSIOLOGY. In Two Volumes 8vo, each Volume 20s. ELEMENTS OF PHYSIOLOGY. By J. MULLER, M.D. Translated from the German, with Notes, by WILLIAM BALY, M.D. With Steel Plates and very numerous Wood Engravings. Second Edition. " We do, however, recommend it very strongly, and we have no hesitation in saying that it will supersede all other works as a Text Book for Lectures, and one of refer- ence for students. It steers a middle course between the superficial brevity of many of our modern works, and the pleonastic and metaphysical jargon of Burdach." Med. Chir. Review, April 1838. * + * Subscribers are requested immediately to complete their sets. DISEASES OF THE LUNGS- Fcp 8vo, 6s. 6d. cloth, PHYSICAL DIAGNOSIS OF DISEASES OF THE LUNGS. By W.H. WALSHE, MD., Professor of Pathological Anatomy in University College, Physician to the Hos- pital for Consumption and Diseases of the Chest. " The treatise is one of extraordinary merit. Indeed, we do not hesitate to say that there exists in no language any work on the physical diagnosis of diseases of the lungs, suited for students, so clear and precise, and at the same time so comprehensive and practical, as this." British and Foreign Medical Review. ANATOMY AND OPERATIVE SURGERY OF THE ARTERIES. Parts 1 to 16, each containing 5 Plates, Imperial Folio, and Letterpress, Octavo, 12*. per Part. THE ANATOMY OF THE ARTERIES, WITH ITS APPLICATIONS TO PATHOLOGY AND OPERATIVE SURGERY. In Lithographic Drawings, the size of and drawn from Nature, with Practical Commentaries. By RICHARD QUAIN, Professor of Anatomy in University College, and Surgeon to University College Hospital. Parts I. to VII., forming Volume I. (if the Work be bound in 2 volumes,) may be had, half bound morocco, with the Plates folded into half their full size, price 51. 5*. *** Part 17 will complete the Work, and will be shortly published. WORKS PRINTED FOR TAYLOR AND WALTON. ! THE SURGICAL ANATOMY PRINCIPAL REGIONS OF THE HUMAN BODY. By THOMAS MORTON, Assistant-Surgeon at Univer- sity College Hospital, one of the Demonstrators of Anatomy in the same College. Royal 8vo. With Lithographic Plates and Wood Engravings. I. THE PERINEUM. 6*. plain, fs. 6d. coloured, II. THE GROIN, THE FEMORAL AND POPLITEAL REGIONS. 13*. coloured ; 9*. plain. III. INGUINAL HERNIA, THE TESTIS, AND ITS COVERINGS. 12s. coloured ; 9*. plain. IV. THE HEAD AND NECK, THE AXILLA, AND BEND OF THE ELBOW. (Preparing.) " We most cordially recommend Mr. Morton's treatise, as a satisfactory guide in the dissection of the perinseum and pelvis." British and Foreign Medical Review. " The production (volume on the Groin) is altogether one which we can conscientiously recommend to the working student. * * * The work will constitute a complete and elaborate treatise, that cannot fail to be highly useful to surgeons iu general." British and Foreign Medical Review. " The present work is a worthy successor to those (on the groin and perinseum), and will prove, we do not doubt, as great a favourite with students." Med. Chir. Review, April 1841. A SERIES OF ANATOMICAL PLATES, IN LITHOGRAPHY, With References and Physiological Comments. Illus- trating the Structure of the different Parts of the Human Body. Edited by JONES QUAIN, M.D., and ERASMUS WILSON, Lecturer on Anatomy and Physiology at the Middlesex Hospital. Complete in 2 vols. royal folio, half bound morocco, gilt tops, 12?. plain, 201. coloured. The Work consists of the following Divisions : THE MUSCLES. 51 Plates. 21. 16s. plain; 51. 5s, full coloured. THE VESSELS. 50 Plates. 21. Us. plain ; 31. 18s. with the vessels coloured. 38 Plates. 21. 4s. plain ; 4Z. 2s. full THE NERVES. coloured. THE VISCERA. Including the Organs of Digestion, Re- spiration, Secretion, and Excretion. 32 Plates. Price 11. 18s. plain ; 31. 10s. coloured. THE BONES AND LIGAMENTS. 30 Plates. 21. plain ; 21. 15s. coloured. *** Any Volume may be purchased separately. The' entire Work consists of 106 Numbers, contain- ing 201 Plates, bes ides Indexes, $c. Subscribers are requested immediately to complete their sets. ILLUSTRATED AND CHEAP EDITION OF DR. DAVIS'S MIDWIFERY. Now ready, in One Volume, 8vo, with a Mo Volume > f Plates, together 11. 7s. 6d. cloth, or 11. 5s. in Parts, ELEMENTS OF OBSTETRIC MEDICINE, DESCRIPTION AND TREATMENT OF SOME OF THE PRINCIPAL DISEASES OF CHILDREN. By DAVID D. DAVIS, M.D., late Professor of Obstetric Medicine in University College. Second Edition. Accompanied by all the Plates belonging to the original 4to Edition. " Having, during along series of years, accumulated a mass of important facts, Dr. Davis is entitled to the thanks of established practitioners, and to the gratitude of all the junior members of the Profession, for thus pre- senting to them, in an agreeable form, the results of an experience which it can be the lot of few, even eminent physicians, to enjoy." Notice of the First Edition. Lancet, Sept. 22, 1832. LINDLEY'S ILLUSTRATED BOTANY. In One Volume 8vo, with many Hundred Figures on Wood, 10s. 6d. cloth, lettered, ELEMENTS OF BOTANY, STRUCTURAL, PHYSIOLOGICAL, SYSTEMATI- CAL, AND MEDICAL. Being a Fourth Edition of the First Principles of Botany. By JOHN LINDLEY, Ph. D., F.R.S., Professor of Botany in University College, London. DRESSING AND BANDAGING- Fcap 8vo, with 100 Engravings on Wood, cloth 6*. 6d. THE SURGEON'S PRACTICAL GUIDE IN DRESSING, AND IN THE METHODIC APPLICATION OF BANDAGES. By THOMAS CUTLER, M.D., late Staff-Surgeon in the Belgian Army. Second Edition. ALLEN, (DR. M., of HIGH BEECH,) ON THE CLASSIFICATION OF THE INSANE. 8vo, 6*. DAVIS, (DR. DAVID D.,) ACUTE HYDRO- CEPHALUS, OR WATER IN THE HEAD, an Inflammatory Disease, and Curable by the same means with other Diseases of Inflammation. 8vo, 9*. 6d. HIND (G. W.), FRACTURES OF THE EXTREMITIES, exhibited in 20 Plates and 40 Woodcuts. Folio, 11. 4*. cloth. A PORTRAIT OF R. LISTON, ESQ.,Surgeou to University College Hospital. Drawn on Stone by GAUCI, from the original by Eddis. Price 2*. 6d. CHRISTISON'S (PROFESSOR) MEMOIR OF THE LATE DR. TURNER, Professor of Che- mistry in University College. 8vo, Is. (Now Publishing.) February, 1844. APPLIED CHEMISTRY; MANUFACTURES, ARTS, AND DOMESTIC ECONOMY. EDITED BY E. A. PARNELL, AUTHOR OF "ELEMENTS OF CHEMICAL ANALYSIS;" WITH NUMEROUS WOOD ENGRAVINGS, AND OTHER ILLUSTRATIONS. In Monthly Parts, 800., each 2s. Just completed, VOLUME I., price 12s., cloth lettered, containing, INTRODUCTION. I PRESERVATION OF WOOD. GAS ILLUMINATION. DYEING AND CALICO PRINTING. Illustrated by Wood- cuts, and specimens of Dyed and Printed Cottons. *** Part VI., price 2s., will be published on the 1st of March. Among the subjects which will be introduced in early Parts are MANUFACTURE OF SULPHURIC ACID. INDTGO. COFFEE, TEA, AND CHOCOLATE. CAOUTCHOUC. BORAX. ZINC. BICHROMATE OF POTASH. NEW WORK BY PROFESSOR LIEBIG. Second Edition, fcap 8vo, is. 6d. cloth, FAMILIAR LETTERS ON CHEMISTRY ; AND ITS RELATIONS TO COMMERCE, PHY- SIOLOGY, AND AGRICULTURE. By JUSTUS LIEBIG, M.D., Professor of Chemistry in the University of Giessen. Edited by JOHN GARDNER, M.D., Member of the Chemical Society. NEW EDITION OF LIEBIG'S AGRICULTURAL CHEMISTRY. Third Edition, with large additions and very numerous alterations, One Volume 8vo, 10*. 6d. cloth, CHEMISTRY, IN ITS APPLICATIONS TO AGRICULTURE AND PHYSIOLOGY. By JUSTUS LIEBIG, M.D., Ph.D.,F.R.S., Professor of Chemistry in the University of Giessen. Edited from the Manuscript of the Author, by LYON PLAYFAIR.Ph.D. " It is not too much to say, that the Publication of Professor Liebig's Organic Chemistry of Agriculture con- stitutes an era of great importance in the history of Agri- cultural Science. Its acceptance as a standard is un- avoidable, for following closely in the straight path of inductive Philosophy, the conclusions which are drawn from its data are incontrovertible." ft We can truly say, that we have never risen from the perusal of a book with a more thorough conviction of the profound know- ledge, extensive reading, and practical research of its author, and of the invincible power and importance of its reasonings and conclusions, than we have gained from the present volume." Silliman's Journal. NEW WORK ON ALKALIMETRY, &c. In One Volume, I2mo, 4*. cloth, NEW METHODS OF ALKALIMETRY, AND OF DETERMINING THE COMMERCIAL VALUE OF ACIDS, AND MANGANESE. By Drs. FRESENIUS and WILL, Chemical Assistants in the University of Giessen. Edited by J. LLOYD BULLOCK, Member of the Chemical Society, and late of the Giessen and Paris Laboratories. This little Work will prove of the highest importance to Calico Printers, Bleachers, Dyers, Manufacturers of Soap, Paper, and Prussiate of Potash ; also to Chemists and to Dealers in Alkalies, Acids, &c. NEW EDITION OF LIEBIG'S ANIMAL CHEMISTRY. In One Volume, 8vo, price 9s. 6d., the Second Edition of ANIMAL CHEMISTRY; Or, ORGANIC CHEMISTRY IN ITS APPLICATIONS TO PHYSIOLOGY AND PATHOLOGY. By JUSTUS LIEBIG, M.D., Ph. D., Professor of Chemistry in the University of Giessen. Edited, from the Author's Manuscript, by WILLIAM GREGORY, M.D., Professor of Chemistry, King's College, Aberdeen. " While we have given but a very imperfect sketch of this original and profound work, we have endeavoured to convey to the reader some notion of the rich store of in- teresting matter which it contains. The chemist, the physiologist, the medical man, and the agriculturist, will all find in this volume many new ideas and many useful practical remarks. It is the first specimen of what modern Organic Chemistry is capable of doing for Physiology ; and we have no doubt that from its appearance phy- siology will date a new era in her advance." Quarterly Review, No. 139. TAYLOR AND WALTON, 28, UPPER GOWER STREET. CHEMICAL WORKS, PRINTED FOR TAYLOR AND WALTON. TURNER'S CHEMISTRY, BY PROFESSORS LIEBIG AND GREGORY. Seventh Edition, in One Volume 8vo, containing the whole of Organic Chemistry, 11. 8s., TURNER'S ELEMENTS OF CHEMISTRY. Edited by JUSTUS LIEBIG, M.D., Ph.D., Professor of Chemistry in the University of Giessen ; and WILLIAM GREGORY, M.D., Professor of Chemistry in King's College, Aberdeen. *** Part ITI., No. 4, completing the Sixth Edition, and the Second Supplement, completing the Seventh Edition, are Published. " There is no English work on Chemistry, which has been in so many hands, and has met with such universal approbation, as ' Turner's Elements ; ' and there is scarcely any work which has received so many additions and improvements in passing through its numerous edi- tions. The present one appears to fulfil all that can be desired of a work of this kind. In the former editions, which were conducted by the late lamented Dr. Turner, the inorganic division of the subject was treated with that clearness, perspicuity, and beauty of arrangement so peculiarly his own ; but the organic part of the work, although giving a very good general outline of this part of chemistry, was not so full as could be wished, when the almost miraculous advances of this interesting branch of science were taken into consideration. The edition now before us, by W. Gregory and J. Liebig, leaves nothing to be desired in this respect ; they have rendered it exceedingly complete, carrying out at the same time the original idea of Turner. As a compendium of the present state of chemistry, and a text -book for all beginners, we consider it as unequalled by any in the English language ; and we even doubt whether there are any of the foreign manuals of an equal size, which can venture to compete with it." Chemical Gazette, Dec. 1, 1842. CHEMICAL ANALYSIS In One Volume, 8vo, with Diagrams on Wood, price 10s. 6d. cloth, ELEMENTS OF CHEMICAL ANALYSIS, INORGANIC AND ORGANIC. By EDWARD ANDREW PARNELL, late Assistant Lecturer on Chemistry in the Medical School of St. Thomas's Hospital. " As an adjunct to the exertions of the teacher, but by no means superseding the necessity of oral and practical instruction, the work of Mr. Parnell appears likely to be in the highest degree useful ; it will serve to convey to the attentive student a correct notion of the general routine of operations proper to each of the different classes of analytical research, and the principles upon which such modes of proceeding have been devised, besides affording him a fund of practical detail, an easy reference to which will certainly lighten his labours and accelerate his pro- gress." Chemical Gazette, Nov. 1842. PORTRAIT OF PROFESSOR LIEBIG, OP GIESSEN. A Line Engraving by RAUCH, after a Drawing by Trautschold. Price 7s. 6d. DR. BENCE JONES'S APPLICATION OF LIEBIG'!i PHYSIOLOGY. In One Volume, 8vo, price 6s. AN APPLICATION .OF PROFESSOR LIE BIG'S PHYSIOLOGY TO THE PREVENTION AND CURE OF GRAVEL, CALCULUS, AND GOUT. By H. BENCE JONES, M.A., Cantab. ; Licentiate of the Royal College of Physicians, London ; and Member of the Chemical Society. " In thus expressing our opinion of the practical valu< of this publication , it is satisfactory to find ourselves sup ported by Professor Liebig himself, under whose imme diate superintendence and sanction a translation into German is now preparing." Pharmaceut. Journal. DR. GREGORY ON SCHOOLS OF CHEMISTRY- / In 8vo, price Is. sewed, SCHOOLS OF CHEMISTRY. A Letter to the Earl of Aberdeen on the State of tho Schools of Chemistry in the United Kingdom. By WILLIAM GREGORY, M.D., F.R.S.E., Professor of Medicine and Chemistry in the University and King's College, Aberdeen. GIESSEN LABORATORY. Lately imported, price 12s. EIGHT FOLIO PLATES, REPRESENTING THE CHEMICAL LABORATORY OF THE UNIVERSITY OF GIESSEN. Accompanied by a Pamphlet (in German) descriptive of the same. By DR. JUSTUS LIEBIG. / ^ NEW CHEMICAL TABLES. In Folio, PART I., price 4. TABLES OF THE ELEMENTARY AND COMPOUND BODIES. Systematically arranged and adapted as Tables of Equivalents, or as Chemical Labels. By CHARLES BUTTON & WARREN DE-LA-RUE. CHEMICAL APPARATUS. For the Use of Analytical Chemists, Lecturers, Students, and Amateurs. TAYLOR & WALTON have on Sale a Collection of IMPROVED CHEMICAL APPARATUS, Selected from the Stock of GRIFFIN & Co., Glasgow. Descriptive Catalogues of Messrs. Griffin <$ Co.'s entire Stock may be had, price Is. Gd. BRADBURY AND SVANS, PRINTERS, WHITEFRIARS. LIBRARY USE RETURN TO DESK FROM WHICH BORROWED LOAN DEPT. THIS BOOK IS DUE BEFORE CLOSING TIME ON LAST DATE STAMPED BELOW i * i - Itl'ULD DEC ... . :f LD62A-30m-2,'71 (P2003slO)9412A-A-32 General Library University of California Berkeley 30m-l,'l5 m