THE ACTION OF MERCURIC CYANIDE ON METALLIC SALTS. 67 X.-The Action of Mercwic Cyanide on Metallic Salts. By LILANANDA GUPTA. THE action of mercuric cyanide on different metallic salts has been the subject of investigation by previous workers for example, Poggiale (Compt. rend. 1846 23 762) Varet (BUZZ. SOC. chim., 1891 [iii] 5 S) Bloxam (Ber. 1883 16 2669) etc. In this paper the action of mercurio cyanide on the chlorides of copper, cobalt nickel manganese cadmium chromium vanadium and aluminium the nitrates of silver and lead and ammonium molybdate has been studied. The method of procedure was how-ever somewhat different from that adopted by Poggiale and others. A concentrated solution of mercuric cyanide was added to an excess of the salt solutions and the mixture evaporated slowly on the water-bath.The precipitate which was obtained in each case was collected and washed as a rule with the minimum quantity of cold water and next -with alcohol. I n the case of the aluminium compound the concentrated solution was evaporated in a vacuum, and the crystals obtained were washed with alcohol alone. It is t o be noted that the general method of preparing a double cyanide of two metals is to bring together their respective cyanide 68 GUPTA THE ACTION OF MERCURIC or to treat an alkali cyanide with some other double cyanide of the metals in question. The action of mercuric cyanide on the chlorides of copper and cobalt and silver nitrate producing respectively, the compounds CuCN,2Hg(CN),,4H20 Co(CN),,Hg(CN),,5H20, and AgCN,Hg(CN),,4H2O described in this paper are thus excep-tions to this general rule.On the other hand Poggiale and Bloxam mention the compounds 2CoC12,Hg( CN),,4H20 and 7AgCN,ZHgO,Hg(CN), as the products of reaction. Hofmann and Wagner (Ber. 1908, 41 319) however obtained the salt AgN0,,Hg(CN)2,2H20 in the last case. Other compounds described in this paper are 4A1C1,,HgC1,,28H20 and MnC1,,Hg(CN),,2R2O. Cryoscopic determinations of the molecular weights in aqueous solutions of these salts have been carried out ; also measurements of the conductivity of the latter salt were made a t different dilutions. EXPERIMENTAL. Compound CuCN 2Hg( CN),,4H20. About 24 grams of cupric chloride were dissolved in 100 C.C. of water and a solution of mercuric cyanide containing 10 grams of the salt was added the final volume of the mixture being 300 C.C.This was evaporated slowly on the water-bath almost to one-third its original bulk and was set aside. Next morning a very fine precipitate had formed which was collected washed first with cold water and finally with alcohol and dried on a porous plate at 25O. The product had a pale violet colour shining like fine sand when held in the sunlight and under the microscope was seen to consist of cubes. It is not appreciably soluble in water but decomposes and passes into solution when treated with sulphuric acid : I. 0.2011 gave 0.0258 CuO. Cu=10*25. 0.2786 , 0-1680 Hg. Rg=60.32. 11. 0.2125 , 0.0262 CuO. Cu=9.97. 111. 0.2618 , 0.0318 CuO. Cu=9.70. 0.3242 , 0'1960 Hg. Hg=60*46. 0.2838 , 26.2 C.C.N a t 32O and 760 mm CN=18.39. 0.2161 , 18-6 C.C. N a t 24" and 760 mm. CN=18-21. CuCN,2Hg(CN),,4H20 requires Cu = 9.47 ; Hg = 60.15 ; CN = 19.54 per cent. Compound CO(CN),,H~(CN)~,~H,O. This was prepared by the action of mercuric cyanide on cobalt The proportions of the salts were as 1 :3 and the method chloride CYANIDE ON METALLIC SALTS. 69 of procedure was exactly the same as mentioned above. A visible reaction took place within ten minutes from the sta.rt; the colour of the solution deepened and slow precipitation commenced. The washings and drying were performed as before. The compound is sparingly soluble in water but decomposes and passes into solution by the action of sulphuric acid as is the case with all sparingly soluble cyanides : 0.3118 gave 0.043 Co.Co=12'79. 0.2440 ,? 0.0977 Hg. Hg=40*04. 0.1875 ,? 25-46 C.C. N a t 2 6 O and 760 mm. CN=27*78. Co(CN),,Hg(CN),,5H20 requires Co = 12-31 ; Hg = 41.40 ; CN = 27.13 per cent. Nicke I Cyanide Ni( CN),,3€€,0. About 20 grams of nickel chloride NiCl,,GH,O were dissolved in 100 C.O. of water and to the cold solution was added a filtered solution of mercuric cyanide ( 8 l O O ) . The details of the procedure were exactly the same as in the foregoing cases. Williams (" Cyanogen Compounds," p. 65) describes the com-pound Ni(CN),,7R2O which he obtained by treating an alkali cyanide with excess of a nickel salt. The compound obtained by the above method is not at all gelatinous it has a fine granular structure and is insoluble in water or dilute acids but is decom-posed by concentrated sulphuric acid.Two- preparations gave the same product in each case: I. 0.3852 gave 0.1751 NiO. 11. 0-2605 gave 0.1186 NiO. Ni=35*71. Ni=35-78. 0.2348 , 35.4 C.C. N a t Oo and 760 mm. CN=31.61. 0.4348 ,? 0.1408 H,O. H20=32.38.* Ni(CN),,3H20 requires Ni=35.75; CN=31.51; H,O=32.74 per cent. The estimation of water in this compound and in subsequent determinations was carried out in the following manner. I n a combustion furnace was placed a hard-glass combustion tube containing a known amount of the substance t o be dehydrated in a weighed porcelain boat. One end of this tube was joined successively to a sulphuric acid wash-bottle a calcium chloride U-tube a soda-lime tower a second sulphuric acid wash-bottle and finally to an air-supply.To the other end a U-tube containing pumice stone soaked in sulphuric acid was attached. A guard tube containing calcium chloride was also joined t o this U-tube. A thermometer was inserted through an opening a t the top of the * The dehydration mas completed at 170" 70 QUPTA TEE ACTION OF MERCURIC furnace just above the glass tube. Air-tight joints were secured, and blank experiments were previously conducted. The boat con-taining the salt was placed a t the end furthest from the U-tube containing pumice stone. The heating was gradual and began from the opposite end. A regular stream of dry air was passed through the combustion tube and the water set free from the sub-stance was condensed and collected in the weighed U-tube contain-ing pumice stone soaked in sulphuric acid.It was finally weighed again and the difference between the two weights gave the amount of water in the compound. Compotbnd AgCN,Hg(CN),,4H20. To 100 C.C. of a solution of silver nitrate containing 12 grams of the salt was added a filtered solution of 6 grams of mercuric cyanide in 100 C.C. of water. This mixture was allowed t o evaporate to half its original bulk. A slight turbidity was noticed almost from the beginning. From the ice-cold concentrated solu-tion milk-white needles were deposited within ten minutes. These were collected and washed first with a little water then with alcohol and dried on a porous plate a t 25O. The compound decomposes slightly in the presence of water and dissolves in dilute sulphuric acid : I.0.8172 gave 0.2403 AgCl and 0.4166 HgS. Ag=22*12; Hg = 43.94. 0.2735 gave 22-2 C.C. N a t 30° and 760 mm. CN=17*64. 11. 0.5797 gave 0.1751 AgCl and 0.2998 HgS. Ag=22.73; Hg = 44.59. 0.7988 gave 0.1196 H,O.* H,O = 14-97. AgCN,Hg(CN),,4H20 requires Ag = 23.5 ; Hg = 43-66 ; CN = 17.07 ; R20 = 15-77 per cent. Action of Mercuric Cyanide on Chromium Trichloride and Vanadium Trichloride. The action of mercuric cyanide on the above-mentioned salt solu-tions was tried in a similar manner but the attempt to bring about a combination failed and the product obtained in each case con-sisted of an oxide of the respective metal. The cyanides were no doubt formed a t first but these being unstable decomposed into * The sdt began to give off water from 85" onwards and the dehydration ww completed at 180' CYANIDE ON METALLIC SALTS.71 the oxide and free hydrocyanic acid the odour of the latter being distinctly perceptible. Thus : A . 2CrC1 + 3Hg(CN) + 6H,O = 3HgC1 + 2Cr(CN) + 6H20. U. 2VC1 + 3Hg(CN) + 3H,O + 0 = 3HgC1 + V,O + 6HCN. 2Cr(CN)3 + 6H20 = 2Cr(OH)3 + GHCN. Action of Mercuric Cyanide om Ammonium Molybdate and Lend Nitrate . These reactions also failed t o yield the expected cyanides; with ammonium molybdate was obtained metallic mercury in the shape of fine globules and the oxide Mo205 as a violet-black powder, whilst in the latter case lead oxide was produced. Compound MnCl,,Hg(CN),,2H20. About 25 grams of manganese chloride were dissolved in 150 C.C. of water and a solution of mercuric cyanide (10 100) was added.The mixture was evaporated on the water-bath for one and a-half hours and the bulk reduced to 100 C.C. This concentrated solution was well stirred and allowed to cool overnight. Next morning a crystalline mass of pinkish-white scales had formed which was powdered washed with the mother liquor and finally with alcohol. The compound was dried on a porous plate for two days a t 24-25O. This salt was found t o be dihydrated whereas Poggiale described i t as being trihydrated. It is very readily soluble in water and iinder the microscope appears to consist of prismatic crystals : Mn=13.41 ; I. 0.1617 gave 0.056 Mn,P,07 and 0.0918 HgS. Hg=48*97. 0.1617 gave 0.1197 AgC1. C1=18-32. 0.3274 , 18.8 C.C. N a t 24O and 760 mm. CN= 12.14.11. 0.2084 gave 0.0724 Mn,P,O and 0.1180 HgS. Mn=13.44; Hg=48.81. 0.2084 gave 0.1485 AgCl. C1=17*63. 0.8353 , 0*0780 HzO. HzO=9*34.* MnC1,,Hg(CN),,2H2O requires Mn = 13.28 ; Hg = 48.30 ; Cl= 17.15 ; CN = 12-56 ; B20 = 8.71 per cent. The conductivity of the salt a t different dilutions was determined. * The salt began t o give off water from 90' onwards. The dehydration was completed at 170' 72 QUPTA TIIE ACTION OF MERCURIC T = 23O. Molecular Dilution. tivity. p, forizinc V. k- chloride. 50 204 207 100 212 216 500 237 231 1000 245 235 i& conduc-The molecular conductivity of zinc chloride a t 23O for different dilutions is quoted from Kohlrausch's table and proves to be almost identical with that of the salt. Assuming the molecular Con-ductivity of zinc chloride and manganese chloride to be the same, as has been shown to be true by Kohlrausch then the above com-pound evidently undergoes dissociation in solution into the ions Mn" and C1' and the undissociated molecule Hg(CN),.This is confirmed by the fact that when treated with silver nitrate silver chloride but no silver cyanide is formed. The determination of the molecular weight of the compound by the cryoscopic method supports the above view. Solvent 19-86 grams of water. Solute. At. M.W. 1 ............ 0.4593 -0*40" 107.3 2 ............ 0-9031 - 0.75 112.0 M.W. calculated for no dissociation = 414. M.W. calculated for dissociation to Mn" Cl," and Hg(CN) = 103.5. Compound 4A1C1,,HgC1,,28H20. s About 25 grams of hydrated aluminium chloride were dissolved in 100 C.C.of water and a cold solution of 10 grams of mercuric cyanide in 100 C.C. of water was added. The mixture was evapor-ated to about 50 c.c. and as no deposit was obtained on allowing the solution t o remain overnight it was placed in a vacuum desiccator over sulphuric acid. After a day a crystalline crust was found to have been formed on the surface of the solution. This was broken and the crystallisation in a vacuum allowed to proceed for a further couple of days. During this interval the crust on the surface was broken from time t o time and through-out the whole process of evaporation a strong odour of hydrocyanic acid was noticed. The crystals were collected washed first with the mother liquor and then repeatedly with alcohol.The com-pound was finally dried in a vacuum for two days CYANIDE ON BIETALLIC SALTS. 73 I. 0.4973 gave 0.0877 HgS and 0.0804 A1,0,. Hg=15*19; A1 = 9.51. 0.4348 gave 0.6548 AgC1. C1= 36.43. 11. 0.4407 gave 0.0777 HgS and 0.0790 Al,O,. Hg=15*20; A1 = 8.57. 0.5995 gave 0.8831 AgCl. C1= 37.25. 4A1C1,,HgC1,,28H20 requires Hg = 15.27 ; A1 = 8.25 ; C1= 37.96 ; H,O = 38.52 per cent. Cryoscopic Determination of t h e Molecular Weight. Solvent 15.7251 grams of water. Solute. At. M.W. 1 ............ 0.0963 - 0.087O 130.84 2 ............ 0.3077 - 0.277 130.94 3 ............ 0.6056 -0.515 137.10 The value found in this way approximates to one-tenth of the molecular weight namely 1309. I n aqueous solution the double salt is decomposed as the solution is found t o be distinctly acid. Since the compound 4A1C1,,HgC~,28H20 is completely hydrolysed to hydrochloric acid aluminium hydroxide and mercuric chloride, the observed molecular weight ought to be one-twelfth of 1309 as the lowering of the freezing point due to the aluminium hydroxide and mercuric chloride is quite negligible. This anomaly is difficult to explain and the molecular weight determination does not there-fore furnish much evidence as to the true molecular weight of the double salt in solution. I avail myself of this opportunity to express my sincere thanks to Sir Prafulla Chandra Riiy for his helpful interest and encourage-ment in this work. COLLEGE OF SOIENCE, 92 UPPER CIR~ULAR ROAD CALCUrnl [Received June 6th 1919.) VOL. OXVII.