DIVERS AND HhGA : IMIDO SULPHONATES. C1X.- Euzidosu@honates. Part 11. By EDWARD DIVERS, M.D., P.R.S., and TAMsMAsA HAGA, D.Sc. (Japan), Y.C.S. THIS communication to the Society is supplementary to that which appeared in 1892 (Trans., 61, 943). I t contains an account of some imidosulphonates not there described, and a collation of our results with those obtained by Berglund, called for by the existence of some radical differences between them. It was this chemist, now deceased, who first made known the existence of imidosulphonates, although several of them had been already obtained and described by others, under various names and formule His important memoir on these salts was published in Swedish, but summaries of it by C l h e and by himself, respectively, appeared in the journals of the French and German Chemical Societies.References to all are given in our first paper. In English, also, a good summary was first published by Watts in the 2nd Supplement of his Dictionary ; this had escaped our notice, and, up to the time of publishing our first paper, we had only seen the account in the Berichte, and an appreciative notice of the Swedish paper in Raschig’s paper on “ Fremy’s Sulphazotised Salts.’’ But, soon after that, Dr. Raschig spontaneously sent us, with the greatest kindness, his own copies of Berglund’s Swedish papers on ‘‘ Imidosulphonic acid ” and “ Amidosulphonic acid.” A perusal of these led pus to resume work on the subject, with the intention only of examining into the differences between his results and onrs, but we went further afield, and prepared a few new salts, because they promised to be of interest.F o r convenience of reference, we follow in this paper the order of description observed in the former paper, and reproduce, as brieflyDIYERS AND HAGA : IMIDOSULPHONATES. 1621 as possible, matters of interest in Berglutid’s paper in Swedish, not t o be found in chemical literature outside it. A full list of the salts prepared by him is given by Watts (op. cit.). Berglund’s first source of imidosulphonates was ammonia and chlorosulphonic acid, but afterwards he prepared them from potas- sium nitrite and mlphite, following Fremy, with modifications, which do not call for notice here. Alkali Imidosulphonates. Ammonium ImidosuZphonates.-Berglund failed to get normal am- monium imidosulphonate, the statement i n Watts’ Dictionary that he did being erroneous. He believed Rose’s “ vitreous sulphat- ammon ” to be this salt, but we find that, both before and after its crystallisation from water, i t is the two-thirds normal imidosulphon- ate, and that Rose’s “ flocculent sulphatammon,” not noticed by Berglund, is anhydrous normal ammonium imidosulphonate.TO Woronin is due the accurate distinction between the two salts. The mother liquor of the crystals of the two-thirds normal ammo- nium imidosulphonate (“ parasulphatammon ”) gave Rose, on evapo- ration, his so-called “ deliquescent salt ; ” this, Berglnnd, judging from his later paper on “ Amidosulphonic acid,” considered to have been a mixture of amidosulphonate and the two-thirds normal imidosulphonate with ammonium acid sulphate.S o d i u m 1midosulphonnte.s.-That, contrary to Frem y’s experience, sodium nitrite can be sulphonated as easily as potassium nitrite, was ascertained independently by Raschig and by us about the same time ; later, me prepared the normal and two-thirds normal sodium salts, as well as some compound salts. Berglund, however, had pre- pared the normal aodium salt (and also the mercury sodium salt) from ammoninm imidosulphonate by boiling it with sodium hydroxide till all ammonia had been expelled; he found it more satisfactory, however, to precipitate the two-thirds normal potassium salt by adding potassium chloride to the ammonium salt, and to boil the product with a solution of sodium hydroxide and chloride ; on cool- ing, the normal sodium salt was deposited.His description of this salt, so far as it goes, agrees with ours. The crystalline salt, which Le sometimes got, in place of the ordinary normal sodium salt, and which he believed to be that salt in the anhjdrous state, we take to have been a double salt of normal sodium imidosulphonate and potassium chloride. He did not obtain the two- thirds normal sodium irnidosulphonate, or its compounds wich ammonium nitrate and with potassium nitrate, or sodium ammonium imidosulphonate. VOL. LXIX. 5 R1622 DIVERS AND HAGA : INIDOSULPHOKATES. Barium I ~ ~ i d o ~ d p h onat es. Berglund's account of normal barium imidosulphonate agrees with ours ; he was really the first to prepare this salt, and did himself an injnsthe in crediting Woronin and Jacquelain with its previous pre- paration.Woronin never anxlysed his preparation, which was almost certainly a double salt, and, although Jacquelain did carefully analyse his, the result shows it t o have been a barium ammonium salt. Berglund found that the normal barium salt gave up its 5H,O almost entirely a t 100'; we maintain the accuracy of our statement that i t loses water only very slowly, even a t 115' (see further on this point, our account of the strontium salt). Barium imidosulphonate is soluble in a solution of ammonium chloride. T wo-thirds normal barium imidosulphonate was fully described by Berglund, his account agreeing with ours. Double salts of barium with the alkalis received peculiar treatment by Berglund. He did not formally recognise their existence, and relegated to footnotes observations which, he admitted, made their existence probable.He could not satisfactorily formulate the com- position he found them to have, and for him they remained as impure barium salt only. Our own work confirms his results, and m&es it possible to give formulE to his preparations. First, there is Ba12K8H(NS20,),,,llH20, which he constantly obtained when he added normal potassium imidosulphonate to barium chloride. Had he thought of introducing the atom of hydrogen there is in the abol--e formula, he would probably not hare considered his analytical results as incapable of interpretation. A sodium salt, described by us, comes very near to this salt, being R~,,Na,(NSzO,),,,l~HzO ; for, if we sub- tract a molecule of the tlwo-thirds normal barium salt, BaHNS,O,, from his formula and write Na for K, we get ours.Evidently, either salt is mainly BaK(or Na)NS206 with a little Ba,(NS,O,),. By adding two-thirds normal potassium salt to ammonia and barium chloride, he got Ba,,K7(NH,)2(NS,0,J,l,f9H,0. Lastly, the salts ~a6(NH,),H(N8,0,),,21H,C and Ba7(NH4) (NS20,),,22H,0 ; all three having an obvious relation to the first. We have described a still more ammoniated salt, Bas,( NH4 ),( N S,O,) (,8Hz0. A 11 these double salts of barium are granular and powdery, and quite unlike the peculiarly soft, clinging, pure barium salt. Stroii tium Irniclosz~~pl~ onates. According to Berglund, normal strontium and normal calcinm imidosulphonates are, in properties, water of crystallisation, arid con - ditions of formation, as like each other as two salts can be, but unlike the normal barium salt.This account of the normal salts puzzledDITTVRS AND HAGA : IRIIDOSULPHONATES. 1623 US, but when we found hih description of the properties and method of preparstion clf tho calcium salt applied perfectly, so far as it goes, to a salt described 1 q us in our first paper, which we had found to be calcium sodium imidosulphonate, we lnst all confidence as to the accuracy of his accomt. We therefore studied the strontium salts for ourselves, having omitted to do SO when preparing our first paper. Morrnal strontium imidosulphonate, according to Berglund, gradu- ally separates in acicnlar prisms when a solution of strontinm chloride, moderately concentrated, is mixed with one of either noi-ma] sodium imidosulphonate or of diammonium imidosulphonate to which ammonia has been added.Its composition is expressed by the formula Sr,(NSz06)2,6Hz0, ana it loses only two-thirds of its water, even a t 130-140°. I n preparing it, it is iinnecessaig, he said, to take any care to have the strontium chloride ii. excess, although it is very important to keep the barium chloride in excess when preparing the normal barium imidosulphonate. Normal strontium im idosu7,phonate, according to our experience, cannot be obtained by mixing together strontium chloride and normal sodium imidosulphonate, strontium sodium imidosulphonate being formed in this way ; the normal salt, however, can be obtained from this in the same way as the normal barium salt is prepared from a barium potassium or barium sodium salt, namely, by dissolr- ing it in dilute hydrochloric acid and pouring the solution at once into a slight excess of a warm, concentrated solution of strontium Iipdroxide, repeating the operation twice, or until a,ll the sodium has been removed.So long as the reprecipitated salt contains sodium, it is a hard, granular precipitate, but, when it is free irom sodium, it separates in glistening, thin, scaly crystals, which felt together into soft, roluminous flocks, and these, dried on the tile, form coherent flakes, retaining water in their interstices with great obstinacy, like the barium salt, to which i t has, indeed, much resemblance. It is more soluble in water than the barium salt, and is actually soluble in hot water to a considerable extent. For analysis, we comminuted its flaky masses, and pressed the particles between filter paper till it seemed quite dry ; but, on exposure to air for days, such a prepara- tion continuously loses water, and much of it before the crystalline lustrc sensibly dim in is hes.The freshly dried salt contains 12H20, m the following numbers show : Strontium. Sulphur. Water. Calculated ....... 31-75 15-51 26.13 Found. ...... .... 31-86 15.49 - Tested fifter 14 days’ exposure, ihe water amounted to only 7 mds., 5 R 21624 DITERS AND RAGA : IMIDOSULPHONATES. although efflorescence had only then just become apparent. The strontium was then 35.61 per cent., calculation for 7H20 giving 35.66 per cent. By decomposing the normal hydroxy-lead imidosulphonate with ammonium hydrogen carbonate, so as to obtain a weak solution of normal ammonium imidosulphonate, and then evaporating to a small volume, we obtained a concentrated solution of the two-thirds normal ammoniam salt, the strength of which we determined by analysis.To it we added enough ammonia to make i t a little more alkaline than the normal salt, and then mixed it with w slight excess of a con- centrated solution of strontium chloride and left it to stand. Only a very small quantity of precipitate formed, but, on adding excess of somewhat concentrated ammonia water, allowing the whole to evapo- rate nearly to dryness in a desiccator over anhydrous potassium carbonate, and then adding water, a white, opaque powder was left, which, on analysis, proved to be normal strontium imidosulpho- nate, with a slight quantity of ammonia in addition.Our calculation i 3 fcr a pure strontium salt with 53 mols. water. The salt may, however, be regarded as having only 5 mols. water. We determined the alkalinity of the salt, and this expressed as strontium, shows i t t o be more than a third of the total strontium. The excess of base may be regarded as consisting of ammonium, the presence of a little of which wa8 established. Calc. Found. Strontium.. ................ 37.00 37.02 Alkalinity, as strontium. ..... 12.33 12-66 Sulphur.. .................. 18-04 17-98 When a solution of the normal strontium salt, with 12R20, is boiled, a nearly insoluble sandy precipitate is formed; this is a slightly basic* strontium salt with about 5H20.The calculation given is for normal strontium imidosulphonate with 5H20. Calc. Found. Strontium.. ................ 37.47 39.02 Sulphur.. .................. 18.27 18.36 An opaque, powdery salt is also precipitated when the mixed solutions of strontium chloride and normal ammonium imido- sulphonate are boiled ; it contains a very little ammonia, and is, no doubt, a, slightly basic ammoniacal strontium salt. We have not quantitatively analysed it. We have not attempted to prepare the two-thirds normal strontium * We have got results indicating the existence of a heinihydroxy-salt, (HOSr)3NS206, corresponding with the lead salt, but hare not had time to establish t h e fact.DIVERS AND HAGA : IMIDOSULPHONATES.1625 salt, although it could, no doubt, be easily got like the calcium salt (p. 1626). Berglund prepared a solution of it, but found the salt so soluble and so difficult to crystallise that he did not examine it. Strontium sodium irnidosulphonate is obtained when solutions of normal sodium imidosulphonate and strontium chloride are mixed. Probably, the proportions matter little, but we have used 2 mols of the imidosulphonate to 3 mols. of the strontium salt in moderately concentrated solutions. When the salts are mixed, precipitation occurs, but the precipitate redissolves on shaking, and soon small prisms of the strontium sodium salt separate. The salt is sparingly soluble, and resembles the calcium sodium salt described in our first paper. Like that salt, too, it contains 3H20, its formula being SrNaN S2O6,3H2O.Found. Celc. rL- 7 Strontium ..... 25.84 25.96 25.97 Sodium.. ...... 6-81 - 6.69 Water 15.95 - - Berglund got granular precipitates of a strontium potassium salt soon after mixing normal potassium imidosulphonate solutions with strontium chloride solutions. Calculating from his results, we find the salt waa represented by the formula Sr,,fCBH(NS,0s),1,13H,0, in close agreement with the composition of his barium potassium pre- cipitates, as calculated by us. Sulphnr ....... 18-93 18.89 - ......... Calcium Imidosu ?phonates. Berglund states that, on mixing solutions of normal sodium imido- sulphonate and of calcium chloride, he got the normal calcium salt, whereas we always got the calcium sodium salt described in our first paper.Berglund’s description and ours agree, and, as Berglund estimated the calcium from the weight of the residue left on ignition, the difference between this and the weight of the residue left by the sodium calcium salt would be within the limits of error of an ordinary analysis. Berglund’s salt is qaite unlike the normal calcium salt which we now describe for the first time. Norm a1 Calcium I?~aidosulphonate.-The true normal salt can be prepared by acting on the insoluble normal silver salt with its eqni- valent of calcium chloride in solution, decanting from silver chloride, and evaporating the solution in a desiccator till the new salt crystal- lises out. It is only sparingly soluble in water when once separated from solution, and crptallises i n rectangular prisms and tables, stablo in the air.Its composition is expressed by Ca3(NS,0s),,8H,0. It is strongly alkaline to litmus.DIVERS AXD HAGX : I3IIDOSULPHONATES. Calc. Found. Calcium.. ................ 19.60 19.60 Sulphur.. ................ 20.92 21.06 Bei-glund states that the normal calcium salt is also got from a niixed solution of ammonium imidosulphonate, s.mmonia, and calcium chloride. Two-thirds Norvial Culcium Im.idosulphonate.--This salt is obtained by decomposing normal silrer imidmulphonate by two-thirds of its equivalent of calcium chloride and one-third of its equivalent, or, for safety, just a, very little less, of hydrochloric acid. The solution, when evaporated in a desiecator, becomes a mass of radiating prisms. Crushed and drained dry, the salt is permanent in the air, at least for some days.Its formula is CaHNSz06,~H~0. This mixture gave us no insoluble sz,lt. Its reaction is slightly aci.1 t o litmus. Cdc. Found. Calcium ................ 14.87 14-70 Sulphur ................ 23 '79 23-59 Calcium Sodium Inaidc,sdphonate, CaNa.XS2O6,3H2O, is described in our first paper, and has been referred to tibove as being what we get by following Berglund's directions for preparing his normal calcium salt. The mercury calcium salt. will be found demribed on page 1630. Lead Iwtidosulphonates. We have indicated the existence of 8-2 unstable salt, PbHNS,06, in solution, and Berglund has done the sami?. The crystalline, normal hjdroxy-lead irnidosulphonate, (HO),Pb4NS2O6, has also been de- scribed by both of us, ar-d our accounts agree ; his way of preparing it, however, was to add ammonia gradually to mixed solutions of cliammonium imidosnlphonate and lead acetate so long as the amorphous precipitate at 6rst produced gives place to a crystalline one ; when t h e last formed e.morphous precipitate no longer changes, acetic acid is added until this has just been converted into the crjs- talline precipitate.The amorphous precipitate he found t o be a basic lead imido- sulphonate of varying composition. We have shown in our first Faper, however, that a basic salt can be uniformly obtained of the composition, (HOPb)JVS,O,. Xi 1 c e: Imid osu lp Lona t es. We have described t h i s and two otlier crystalline, well characterised salts, Ag,NaNS,06 and AgNazlTS2O6, the latter being obtained when silver nitrate is added to excess of xiorma1 sodium imidosulphonate.Accord- ing to Berglund, however, the precipitate in that case is richer in Bergluud has described only the normal silver salt.DIVERS AND HhGA : INIDOSULPHOXATES. 1627 silver than the normal salt. This is explicable when we consider that he worked with very dilute solutions, for, as we have pointed out, the salt, AgNa2NS206, is partly decomposed by much water into silver oxide and t wo-thirds normal sodium imidosulphonate. Mercury Tmidosulphonatcs. Oxyrnercuric Hydrogen Imidosulphonate, HN( S03HgO)2Hg (Divers and Haga) ; Normal Oqmerczwic Imidosulphonate, Hg[N( SO3Hg),0I2 (Berglund).-These basic mercuric imidosulphonates differ i n that the proportion of sulphur to mercury in ours is S,: Hg6, whilst in Berglund's it is S,: Hg6.His salt was prepared from mercury potassium imidosulphonate and mercuric nitrate, and ours from normal sodium imidosulphouate and mercuric nitrate ; they ought, therefore, to have been the same. Berglund tried the use of normal potassiurii imido- sulphonate, and thus got a more basic product, but this he attributed to the presence of mercuric oxide or basic nitrate in the precipitate. As Berglund was strongly impressed by the tendency of mercury to displace the imidic hydrogen, whilst we were similarly struck with bhe fact that whenever sulphuryl occurs in combination with oxylic mercury in a Precipitated salt, that mercury functions as the bivalent radicle, -IlgOHgOHg-, it will be seen that Berglund's ratio of sulphur to mercury accords with his preconception, and not with oum, and that our ratio accords with our preconception, and not with his.To make sure that we had not been mistaken, we made further experiments ; but, before describing these, we have two adverse com- ments to make on, Berglund's experiments. One is as to the precipitation of mercuric oxide or basic nitrate inferred by him. However prepared, our product has always been one of the whitest of precipitates, only assuming a faint buff tint when kept for some time at 100" or higher, in the dry state. The presence of very little oxide of basic nitrate should have shown itself by a yellowish tinge. No precipitate that we have tested either of the present salt, of the oxymercuric sodium salt (first paper, p.9831, of the mercurous salt (this paper, p. 1630), or of the rnercurosic salts (this paper, p. 1632), has ever shown the presence of nitric acid in it. Moreover, Berglund's supposition that mercnri c oxide or basic nitrate might be precipitated is not probable, when it is considered that the mother liquor of the precipitating salt is much more strongly acid than the solution of mercuric nitrate used. The other matter is the unlikelihood of mercury taking o r retaining the imidic relation in a salt precipitating from such very acid mother liquors ; for, as we have shown in our first paper, dilute nitric acid replaces such mercury by hydrogen. Directly the attempt is made t o lessen the quantity of nitric acid sufficiently to permit of mercury1628 DIVERS AND HAGA : 13IlDOSULPHONhTES.taking the imidic relation, sodium or potassium also enters the salt, displacing half the oxylic mercury. For we then find that we pass abruptly from HN( SO,ElgO),Hg to Hg<N(S033Na)2 “SO HgO),Hg . ,Experiment 1.-To excess of very dilute mercuric nitrate (neces- sarily acid) was added a very dilute solution (1 in 50) of mercury sodium imidosulphonate. The crystalloidal precipitate, washed by decantation with much water, was dried on a tile ; both the salt and the mother liquor were free from sulphate. Composition : Mercury, 73.70 per cent. ; sulphur, 7.98 per cent. ; sodiua, 0.08 per cent. Experiment 11.-Mercuric nitrate solution, 250 c.c., prepared from 6 grams of mercuric oxide dissolved in l i t h equivalent of nitric acid, 250 C.C.of a solution containing 4 grams of mercury sodium imido- sulphomte ; the voluminous, crystalline precipitate formed on mixing these solutions, after being washed once with dilute nitric acid, and then repeatedly with water, was drained dry on a tile; it weighed 6+ grams, or four-fifths of the calculated quantity. It was free from sulphate, as was also the mother liquor. Composition: Mer- cury, 73.44 per cent. ; sulphur, 7.94 per cent. ; sodium, 0.08 per cent, Experiment 111.-Dissolred 2.8 grams of mercury potassium imido- sulphonate i n 230 C.C. water, and added it t o 3 granis of concentrated mercuric nitrate solution containing almost 1.5 grams of mercury as nitrate. A crystalline precipitate formed at once; after a few moments’ active stirring, the precipitate was allowed to settle, and the bright mother liquor was decanted into another vessel containing 4.5 grams more of the mercuric nitrate solution ; a second precipi- tate was thus obtained.Both were washed bg decantation, and drained on tiles. The first weighed 2 grams ; the second, 1.5 grams ; by calculation, each would have weighed 2.9 grams, bad none remained dissolved, according to the equations, 1. HgN,(SOs)*Kd + 2Hg(NOS), + 2H20 = HN(SO,),Rg302 + 211(N03 + ?HN03 + IIN(SO,),K,; + 2KN03 + 4HN0,. 2. HN(SOs)ZK, + 3Hg(NOa)2 + 2HZO = HN(SO,),Hg,O2 The first precipitate contained mercury, 72.85 per cent. ; sulphur, 8.03 per cent. ; potassium, 0.3 per cent. The second contained mer- cury, 72.61 per cent. ; sulphur, 8.11 per cent.; potassium, 0.38 per cent. It will be seen that in no case is alkali metal absent, the quantity present being markedly greater in the case of potassium than in that of sodium, and that the potassium is slightly higher if more nitric acid is present. This apparently strange result is due t o the fact that nitric acid dissolves mercuric imidosulphonate, but has no action on the two-thirds normal potassium or sodium salts.DIVERS AND HAQh : IMIDOSULPHONATES. 1629 The results of the above experiments, together with those described in our first paper, can, we think, leave no doubt that the basic mer- curic salt has the composition HN( SO,HgO)zHg, the percentage numbers for which are-mercury, 74.35 ; sulphur, 7.93 ; the potas- sium in the precipitates of Experiment 111 fully accounts for the slight deficiency of mercury in them.As alreadF stated, it is probable that Berglund got his analytical results as a consequence of the presence of unobserved potassium. Nercury Ammonium 1midosuZphonate.-W hen the mercury calcium salt, described later (p. 1630), was decomposed with just enough am- monium hydrogen carbonate, a secondary decomposition set in after a very few minutes; that is to say, at first calcium carbonate was precipitated with effervescence due t o escape of carbon dioxide ; but after stirring well, till the effervescence had subsided, the solu- tion began again to effervesce and deposit a basic mercuric calcium salt, possibly analogous to the sodium salt, and, therefore, The precipitate proved to be a mixture of calcium carbonate and basic mercuric calcium imidosulphonate, roughly separable by dilute nitric acid ; whilst the mother liquor was a solution of normal and two- thirds normal ammonium imidosulphonate, with a very little mercury calcium imidosulphonate, or the equivalent of these salts.No ami- dated mercury salt was produced. Berglund also tried to make the mercury ammonium salt from the mercury barium salt and ammonium sulphate, but failed. The salt, therefore, appears t o be incapable of continued existence. Mercury potassium Imidosu1phonate.-According to Berglund, dilute nitric acid has no action on this salt, but, as we have pointed out, i t is converted into the insoluble two-thirds normal potassium salt. and mercuric nitrate. Jfercury Sodium Imidosu lyhonate.-This salt was described by us fully, under the belief that Berglund had not prepared it, iii which, we were mistaken.We hare agaiu prepared and examined the salt, and now find it contains 5 atoms of water, and not 6, which agrees with Berglund’s results. The formula is therefore HgN,( 80,),Nal,5Hz0, We have also found now that the salt left in a vacuum over sulphurio acid for weeks loses all its water, and not merely two-thirds of it, as stated in our first paper. Berglund dried his salt. at looo. Oxymercuric Sodium Imidosulphonate.-Berglund had no know- ledge of such a salt as the basic mercury sodium salt, described i n our first paper, or of a corresponding potassium salt. I t is interest- ing, however, to find that on mixing normal mercury potassium, imidosulphonate with silver nitrate even in excess, he was unable to1630 DIVERS AND HAGh : IJIIDOSULPBOSATES.get the mercury silver salt, but only a mercury silver potassium salt. ' ( so"g)2,3H20, is strictly analo- N( SO&) 2 His formula for this salt, Hg< p u s to ours for the basic mercury sodium salt, (instead of 4H20, formerly adopted by us). Mercury Calcium Imidosu7phonate.-Although Berglund obtained and described mercury barium, mercury strontium, and mercury magnesium imidosulphonates, he failed t o get the corresponding calcium salt, on account of its being so soluble in water. We have prep:tred it by dissolving mercuric oxide in a warm solution of two- thirds normal calcium imidosulphonate, filtering, evaporating, crjs- tallising, and, finally, recryetallising from water.It forms small, brilliant prisms. Mercury. Calcium. Sulphur. Calculated . . . . . . 25.90 10.36 16-58 Found .. .... .... 25.92 10.20 16.61 A compound of this salt with mercuric chloride, which can be formulated as C1Hg2(NSz0,Ca)3,12Hz0, and, therefore, be compared with apatite, was got in good, although small, crystals by treating oxymercuric hydrogen imidosulphonate with calcium chloride solu- fiou in the proportion IINS206Hg~Oz : CaC1,. The solution, filtered from tho mercuric oxychloride, and evaporated in a desiccator, gave the salt i n question. By dissolving this in water and precipitating by absolute alcohol, the mercuric chloride can, for the most part, be removed, but only with great loss of the imidosulphonate, as the latter also is soluble in alcohol.Mercury. Calcium. Sulphur. Chlorine. Calculated. . . . 30.92 9.28 14.84 2-74! Found ... .... 31.00 9.35 14-55 2.63 ,Vemirous lmidosui$honate.--No mercurous imidosulphonate has yet been described. There seems to be only one, a basic salt, having Hg', : S2, formed when two-thirds normal sodium imidosulphonate is added to mercurous nitrate in powder which has been stirred up with h u t water until it has all passed into solution except a little of the soft, voluminous, basic nitrate, the latter being quite free from any yellow, granular particles. As the nitric acid liberated dissolves some of the salt, a third, at least, of the imidosulphonate remains in solution ; tile normal sodium imidosulphonate would, therefore, be preferable t o use, were its use not subject to a disturbing effect, t o be noticed presently.The mercuric nitrate should be i n someDIVERS AND HAGA : IMIDOSULPHONATES. 1631 excess, about one-fifth more than the calculated quantity. tion expressing the reaction is The equa- 4(HgNOs), + H,O + 2HN(SO,Wa), = O[Hg4N(S0J2], + 4NaN0, t- 4HN03. Of the sodium salt, 1.05 gram, dissolved in about 150 C.C. water, added, with stirring, gradually to 4.5 grams of mercurous nitrate in about 20 C.C. of water, gave a precipitate weighing 2.i5 grams, and nitric acid in solution weighing 0.33 gram, the ratio of these weights being in accordance with the above equation. The precipitate was free from nitrate and sulphate, and the mother liquor contained no sulpliuric acid. The new salt is flocculent and quite white, and, most probably, has the constitution expressed by the formula 0 [ Hg’2N(S0,)2Hg’,],,6H20.Calculated ........ 77.22 6-18 Found ........... 77.27 6.18 It loses very little i n weight at 100’ or 120°, and part of that loss will be due to volatilisation of mercury, for the salt becomes very grey; when more strongly heated, it becomes nearly black tem- porarily, and then white again. Then, or while still black, the altered salt gives some mercuric chloride as well as mercurous chloride when triturated with a solution of sodium chloride. A t an incipient red heat, the whitened, altered salt fuses and effer- vesces, evolving nitrogen, but no sulphur dioxide. The black-red liquid consists mainly of the mercury sulphates. Mercurous imidosulphonate dissolves in dilute nitric acid much more readily than mercuric; imidosul phonate does, and the addition of sodium chloride to the solution precipitates all the mercury. Triturated with sodium chloride solution, the salt becomes of a per- manent, dull, and somewhat greenish-orange colour, though quite free from the mercuric radicle ; and the sodium chloride solution becomes very alkaline to litmus, no doubt because of the formation of normal sodium imidosulphonate.Concentrated hydrochloric acid soon causes, even in the cold, the formation of some mercury and mercuric chloride and, cn heating, this change becomes complete.* Concentrated solu- tion of potassium iodide at once, in the cold, dissolves it, all but half its mercury left as metal. Normal sodium imidosulphonate converts it into mercury and the sparingly soluble mercuric sodium imidosulphonate, and this makes it undesirable to use the normal sodium salt for preparing the mer- curous salt.When it is used, a blue-grey cloud of niercurg is Mercury. Sulphur. * Ox-ing to the production of amidosulphonic acid (see our paper on this acid).1632 DIVERS AND HAGA : IMIDOSULPHONATES. formed, but by incessant and violent stirring for 10 minutes from the time of adding the normal sodium salt, not in excess, to the mer- curous nitrate (which may here be used in a very dilute nitric acid solution), the grey precipitate becomes almost completely white, and is then the mercurous imidosulphonate. Thus prepared, we have found i t to contain 78.20 and 78.55, instead of 77.22 per cent.of mercury, but the right quantity of sulphur, namely, 6.20 (twice), theory requiring 6.18 per cent. Mercurosic ImidosuZphonates.-Berglund found that mercuric potas- sium imidosulphonate when added to a solution of mercurous nitrate deposited some oxymercuric imidosulphonate only after standing 6ome time. He must ttierefore have used a solution of mercuroua nitrate containing an unnecessary excess of nitric acid, for precipita- tion is immediate if but little nitric acid is present. The precipitate varies in composition with the proportions of the salts used, but, siill within well-marked limits. The variation is very great in the quantities of bivalent and univalent mercury, but very small indeed in the total quantities of mercury and of sulphur.The composition of the precipitates is such that they may be regarded as the oxymercurous imidosulphonate just described, modified in having one-half to three-eighths of its mercurous radicles replaced by mercuric radicles, for it varies within the limits expressed by They differ from the purely mercurous salt in having only half as much water of hydration, and they cannot be represented as mixtures of the known oxymercurous salt with the known oxymercuric salt. O[Hg”N(S0,),Hg‘2~~,3H,0 (atomic ratios Hg”, : Hg‘, : S,) is obtained by adding mercuric sodium imidosulphonate, a salt neutral to litmus, to half its weight of normal mercurous nitrate, which makes 3Hg” be present for every 2Hg’,, and leaves a neutral, or even alkaline, mother liquor. The mercurous nitrate is used in the form described in the preparation of oxymercurous imidosulpho- nate.After adding the mercuric sodium imidosulphonate to. it, the mixture is well stirred t o insure the completion of the action, and until the mother liquor has just lost its acidity. The precipitate s white, settles quickly, and can be freely washed; when dry, its colour is dull. It contains nearly all (+;) of the mercury of t h e nitrate, but only $- of the mercuric radicle, and even only f of the imidosulphonic radicle, its moi,her liquor being very rich in imido- sulphonate, and this gives assurance that there is no mercurous The compoundDIVERS AND HAGA : IMIDOSULPHONATES. 1633 nitrate in the precipitate. contains any snlphate. Neither precipitate nor mother liquor Calc.Found. Univalent mercury. , . 49-45} 74,1, 5p.76} 73.99 Bivalent mercury .... 24.72 23-23 Sulphur ............ 7.91 Sodium - ............ '7.87 0-04 I n the analysis, the two mercury radicles were estimated by dis- solving the salt in dilute nitric acid, diluting the solution, precipita- ting mercurous chloride by dilute hydrochioric acid, and precipitating the mercury in the filtrate as sulpbide. After the salt has been hydrolysed in a sealed tube by hydrochloric acid for sulphur estima- tion, the mercuric radicle is found increased in quantity at the expense of the mercurous radicle; the oxjgen in the air sealed up in the tube having given rise to the increase of mercuric radicle, as shown by the lessened pressure (Hada, this vol., p. 1676). (atomic ratios Hg", : Hg',, T S,) is obtained if the mercurous nitrate is in excess, about five parts being taken for every two parts of mer- curic sodium imidosulphonate, which provides 7Hgf2 for every 2Hg". The precipitate has much the same appearance as that obtained when the imidosulphonate is taken i n excess, and contains nearly 3 of the mercuric and imidosulphonic radicles, but only a little more than + of the mercury of the nitrate used. The mother liquor is accord- ingly comparatively rich in mercurous salt ; i t is also acid. Both the mother liquor and precipitate are free from sulphate, and the pre- cipitate from nitrate. The dry precipitate is dull white. Calc. Found. UnivaleQt mercury.. 63.89) 73.37. 62.11 } 73.57. Bivalent mercury ... 10.48 11.46 Sulphur ............ 7.22 7.20 Sodium.. .......... - 0.01 It will now be seen how remarkably the percentages of sulphur and of total mercury approach each other in the two precipitates, widely as the ratio of the two mercury radicles varies. We append the results of analyses of two other preparations; they support the conclusions drawn from the above extremes.1634 DIVERS AND HhaA : AMIDOSULPHONIO ACID. A. B. Univalent mercury.. . 59.90) '72.78s Bivalent mercury.. . . 12.88 73.46t Sulphur. . . . . . , . . . . . 7.04 7.56 Imperial University , Tokyo, Japan.