February 23, 1893] 



NATURE 



401 



with nitric acid or aqua regia and absorption of the liberated 

 tetroxide vapours in a solution of caustic potash. The dark 

 brown solution of potassium perosmate thus formed was largely 

 diluted with water, and reduced to osmate by the addition of 

 alcohol. After the expiration of about twenty-four hours almost 

 the whole of the osmium had separated in the form of beautiful 

 little crimson octahedrons of the salt ICOsOj.aHp.which, after 

 washing with dilute alcohol, proved to be quite free from 

 impurity, showing no trace of iridium. 



Previous observers have noticed that an aqueous solution of 

 potassium osmate, K._;Os04, is most remarkably affected by sun- 

 light, a rapid decomposition being brought about with deposi- 

 tion of a black precipitate to which the composition OsO^ zH^O 

 has been ascribed. The specimens experimented with, however, 

 undoubtedly contained iridium, and it was therefore of interest 

 to investigate the action of sunlight upon solutions of the pure 

 salt just described. When the crimson octahedrons of pure 

 K2OSO4.2H2O were dissolved in cold water, and the clear red- 

 dish violet-coloured solution was exposed to direct sunshine, no 

 evidence of change was apparent for several days, but the 

 moment the vessel containing the solution was immersed in a 

 bath of boiling water, while in bright sunshine, decomposition 

 commenced, and a black precipitate rapidly accumulated, until 

 after the expiration of two or three hours the whole of the 

 osmium present was deposited. As there is a marked tendency 

 for the production of the noxious fumes of osmium tetroxide during 

 this decomposition of the hot osmate solution by the waves of 

 light it is best to take the precaution of reducing their amount to 

 a minimum by the addition of a little alcohol, which acts as a 

 strong reducing agent under these circumstances, and by passing 

 a stream of hydrogen through the solution during the whole 

 operation. The precipitate is usually so finely divided that con- 

 siderable difficulty is experienced in separating it from the solu- 

 tion. The filtration succeeds best when the filter is previously 

 moistened with dilute acetic acid, when a clear colourless filtrate 

 is usually at once obtained. The precipitate cannot be dried in 

 a warm air bath, as it is largely converted thereby into the 

 volatile osmium tetroxide. It may safely, however, be dried 

 over phosphoric anhydride in the vacuum of an air-pump. 



The accurate analysis of an insoluble substance of the nature 

 of this precipitate, and containing a metal such as osmium, 

 which so readily oxidises to the volatile tetroxide, is a task of 

 exceptional difficulty. The usual method of reduction to metal 

 in a stream of hydrogen is insufTicient, for more or less of the 

 tetroxide is always formed during the process, necessitating the 

 use of an absorption apparatus containing a solution of caustic 

 potash, placed in front of the tube containing calcium 

 chloride to absorb the water formed. The difficulty is, then, 

 how to estimate the small quantity of osmium thus dissolved in 

 the large excess of alkali. It was eventually found that the 

 weak electric current from three Daniell's cells precipitates the 

 whole of the osmium from such a solution, contained in a nickel 

 dish which forms the negative electrode, in the form of pure 

 osmium dioxide, OsOj, which may conveniently be dried in vacuo 

 over phosphoric anhydride and weighed as such. 



By this mode of analysis the interesting fact was eventually 

 elicited, that the black insoluble substance formed by the action 

 of light upon a hot solution of potassium osmate is not, as was 

 previously supposed, a hydrate of osmium dioxide of the com- 

 position OsO.j.aH.iO, but is no other than free osmic acid itself, 

 the hydrate of osmium trioxide, OSO3. H._,0 or H.^OsOj. Osmic 

 acid is thus formed by the direct action of water, under the in- 

 fluence of sunlight and slight rise of temperature, upon the 

 potassium salt. This remarkable change is expressed by the 

 simple equation : 



K2OSO4 -f 2Yi.fi = H0OSO4 -f 2KOH. 



The liquid, as soon as the change commences, is observed to 

 exhibit a strong alkaline reaction, becoming, as indicated in the 

 equation, a solution of caustic potash. It is singular that the 

 presence of alcohol and the passage of a current of hydrogen 

 during the reaction do not cause any reduction, serving only to 

 hinder the further oxidation to the state of tetroxide. Indeed, 

 if the crimson octahedral crystals of potassium osmate are covered 

 in sunshine with warm alcohol and a current of hydrogen is 

 allowed to bubble through the liquid, no trace of blackening is 

 observed upon the faces of the crystals. The moment water is 

 added, however, decomposition is immediately brought about. 



Osmic acid, H.JOSO4, is a soot-black powder, which fumes 

 strongly in moist air, owing to its rapid conversion into the 



NO. 1217, VOL. 47] 



volatile osmium tetroxide, OSO4, but which is quite permanent 

 at the ordinary temperature when preserved under water con- 

 taining alcohol. It dissolves readily in nitric acid with forma- 

 tion of the hydrate of osmium tetroxide, the so-called per-osmic 

 acid. Cold hydrochloric acid attacks it but very slightly. Upon 

 warming, however, it is entirely soluble, forming an olive-green 

 liquid, which will be subsequently considered, with liberation of 

 a small quantity of chlorine. Sulphuric acid does not attack it. 

 Osmic acid reacts in a most energetic and interesting manner 

 with sulphuretted hydrogen gas. Even in the dry state at the 

 ordinary temperature the reaction proceeds with considerable 

 violence. If the experiment is conducted in a piece of combus- 

 tion tubing, upon which a bulb has been blown for the reception 

 of the osmic acid, the moment that the gas enters the tube the 

 whole of the black powder immediately becomes incandescent, 

 and drops of water and a large quantity of free sulphur are de- 

 posited in the portion of the tube not heated by the reacting 

 substances. The residual product of the reaction is a brown 

 powder, which has been found to be a hydrated oxysulphide of 

 osmium of the composition 20sS0. H^O. 



The reaction occurs in accordance with the equation — 

 2H-.OSO4 -t- 4H,S = 2OSSO.H2O + 5H2O 4- 2S. 



This oxysulphide of osmium is soluble in acids with decom- 

 position, even sulphuric acid decomposing it with evolution of sul- 

 phuretted hydrogen. It possesses acid properties, for it liberates 

 carbon dioxide from carbonate of soda and sulphuretted hy- 

 drogen when fused with sulphide of potassium. It would, 

 moreover, appear to contain SH groups, for it yields mercaptan 

 upon treatment with soda and ethyl iodide, the osmium being 

 reduced to the dioxide OsOo. Its probable constitution is there- 

 fore represented by the graphic formula : — 



OS = Q 



Osf o 



\SH 

 When this oxysulphide is warmed in dry sulphuretted hydrogen 

 another violent reaction occurs, the whole mass again becomes 

 incandescent, and the whole of the oxygen is eiimmated in the 

 form of water. The product of this second reaction with 

 sulphuretted hydrogen is pure osmium disulphide OsSo. 



OsaOaCSH), 4- 2H0S = 2OSS2 -f sHjO. 



Qi \!a^ halogen compounds of osmium only the chlorides have 

 been at all investigated, chiefly by Claus, whose observations 

 may be summarised in a few words. 



When finely-powdered metallic osmium is heated in a stream 

 of dry chlorine sublimates are formed. The fir^t chlorine com- 

 pound formed is chromous-green in colour, but is only produced 

 to a very slight extent. There is next deposited a dense black 

 sublimate, and finally a smaller quantity of a sublimate of the 

 colour of red lead. None of these three chlorine compounds 

 are crystalline. Claus subsequently stated that the lowest 

 chloride OsCl, is a bluish- black solid when isolated, and forms a 

 dark bluish-violet solution ; the sesquichloride OsjClg is reddish- 

 brown in the solid state, and gives with water a rose-red coloured 

 solution, and the dichloride O3CI2 is the compound which ex- 

 hibits the colour of red-lead, and yields a lemon-yellow solution 



These observations of Claus are completely confirmed by the 

 experiments of Prof. Moraht and Dr. Wischin, who, howevei, 

 have extended them, and have been able to isolate other and 

 higher chlorides of osmium. 



They commenced by warming a large quantity of the free 

 osmic acid above described for two days upon a water-bath with 

 concentrated hydrochloric acid, the flask in which the reaction 

 was conducted being connected with an upright condenser. A 

 little alcohol was added in order to prevent the formation of 

 osmium tetroxide. The osmic acid eventually entirely dissolved 

 with formation of the dark olive-green coloured solution pre- 

 viously incidentally mentioned, a little chlorine being evolved at 

 the commencement of the operation. It was found impossible 

 to evaporate the solution upon the water-bath without decom- 

 position, but evaporation in vacuo over sulphuric acid and solid 

 caustic potash, the latter to absorb the hydrochloric acid, 

 succeeded admirably. The solid left after complete evaporation 

 consisted of well formed crystals which assumed the habit of 

 six-sided pyramids. Thesecrystals were dark olive-green in colour 

 when moist, but when the last traces of superfluous water were 

 removed, exhibited a bright veimilion colour. They werereadi'y 



