14 ABSTRACTS OF CHEMICAL PAPERS.I n o r g a n i c C h e mi s t ry,Density of Liquid Oxygen. By S. WROBLEWSKI (Compt. rend.,97, 166-168).-A known volume of oxygen was liquefied at -130"in a calibrated tube, and the volume of t.he liquid was compared withthe volume of a known amount of carbonic anhydride liquefied in thesame tube at 0". The sp. gr. of the Iiquid oxygen was found to be0.89 and 0.90 ; mean, 0.895. C. H. B.Active Oxygen. By E. BAUMANN (Ber., 16, 2146--2152).-1t haslong been observed that substances which under ordinaiy conditionsare not affected by oxygen, are yet oxidised in presence of othersubstances, which combine directly with the oxygen of the air.Besides examples of these changes which have been brought for-ward by Hoppe-Seyler, the author, and also Leeds, have observed thatcarbonic oxide when passed over damp phosphorus is oxidised by theozone formed into carbonic anhydride. Remsen and Keiaer (Chem.SOC.J.. 1883, 454) have repeated these experiments, but with negativeresults, and they attribute the formation of carbonic anhydride totrhe oxidation by the ozone of the carbon of the corks used in theapparatus. The author has accordingly repeated his experimentswith an apparatus made entirely of glass, and has confirmed hiINORGANIC CHEMISTRY. 15former results. In one case 56.6 mgrm. of carbonic anhydride wereformed by the passage of 700 C.C. of carbonic oxide over phosphorus ;this represents an oxidation of 2.6 per cent,. of the carbonic oxide.In connection with the explanation of the phenomena of oxidationput forward on the one hand by Hoppe-Seyler and Leecls, and on theother by Traube, the author considers that the former is more satis-factory : for firstly it is simpler; secondly, it explains the greaternumber of phenomena ; and thirdly, Traube's hypothesis gives noaccount of the formation of ozone, hydrogen peroxide, and activeoxygen by the slow oxidation of dam.p phosphorus in the air.v. K. v,Conversion of Carbonic Oxide into Carbonic Anhydride byNascent Oxygen. By A. R. LEEDS (Clzeni. News, 48, 25-29).-The author refers to works which have been published in support ofand against the theory that ca,rbonic oxide is converted into carbonicanhydride during the oxidation of phosphorus in moist air ; he thenproceeds to describe the latest repetition of his original investigationin support of the theory.Sticks of phosphorus were put into a largeflat-bottomed bottle, fitted with a well ground stopper, filled withwater free from air and carbonic anhydride, and inverted in a,pneumatic trough. Well washed carbonic oxide and air, in equalvolumes, were now introduced, sufficient water being left in the bottleso as to partially cover the phosphorus. The stopper was put inwhile the neck of the bottle was still under water; the bottle wasthen reversed, and kept at a temperature of 24". After six days thegaseous mixture was withdrawn from the bottle ; and, the glass stopperbeing replaced by a cork saturated with paraffin, and fitted with thenecessary tubes, the .bottle was immediately inverted in a mercurytrough, and a little mercury allowed to enter in order to cover the cork.The tubes were respectively connected with an aspirator and an air-supply free from carbonic anhydride ; and the issuing gas was passedthrough baryta-water and potassium iodide solution.The baryta-watersoon became turbid, and was subsequently tested in a carbonic anhy-dride apparatus, when the potash bulbs in connection therewith hadincreased 0.0155 gram after the decomposition. The tube passing intothe barium hydrate became incrusted ; the incrustation when treatedwith acid evolved carbonic anhydride. There was no iodine set freein the potassium iodide solution, and therefore no ozone conld havebeen present.The author considers, therefore, that the above state-ment is now established by a rigid quantitative and qualitative analysis.D. A. L.Liquid Nitrous Anhydride. By R. H. GAINES (Chem. News, 48,97).-The author observes that this gas condenses at + 14.4" undera pressure of 755 mm., and when the liquid is heated, ebullition beginsat the same temperature. The pure liquid is dark green, but changesto blue on the addition of a very little water ; the blue and green liquidsare not readily miscible.Preparation of Caustic Potash and Soda. By LOWIG (Di?zyZ.polyt. J., 248, 260).-According to Lowig, an intimate mixture ofsodium or potassium carbonate and ferric oxide is exposed to a brightD. A. L16 ABSTRACTS OF CHEMICAL PAPERS.red heat until the whole of the carbonic anhydride has been expelled.The resulting melt of sodium or potassium ferride is treated with waterat 80-90", the ferride being decomposed into caustic soda or potashand ferric oxide.The latter on drying is used for a further opera-tion. D. B.History of the Preparation of Artificial Sodium Carbonatefrom Common Salt. By DUMAS (Compt. rend., 97, 209--214).-Ahistorical summary, eulogistic of Leblanc. C. H. B.Dimorphism of Silver Iodide. By MALLARD and LE CHATELIER(Co112pt. rend., 97, 102--105).-It has been shown (BUZZ. SOC. Min.,5, 214) that boracite belongs to the rhombic system at the ordinarytemperature, but changes to the cubic system at about 265" withabsorption of heat of 4-77 cal. for 1 gram, and retains this form up tothe melting point.Potassium sulphste behaves in a similar manner,passing from the rhomhic to the hexagonal system, and ammoniumnitrate, which belongs to the rhombic system at the ordinary tem-perature, suddenly changes to the cubic system at about 127". Silveriodide is dull red at a sufficiently high temperature, but is lightyellow at the ordinary temperature, and Wernicke has shown that at138-138.5" the colour suddenly changes from deep yellow to yellowish-white, or vice versc2, according as the temperature is rising or falling.By examination with polarised light, the authors find that when thesilver iodide is heated it suddenly passes from the hexagonal t o thecubic system, the change taking place at about 146", a temperaturenot very far removed from that determined by Wernicke. Measure-ments of the mean specific heat of the iodide between differentlimits of temperature, show that the passage from the hexagonal t othe cubic system is accompanied by an absorption of heat of 6.8 cal.for 1 gram.According t o Zepharovitch the ratio of the axes in thehexagonal form, h : a = 1.2294 : 1. In the cubic form, the ratio ofthe tertiary to the secondary axis will be h : a = 1.2247 : 1, fromwhich it appears that the hexagonal form closely approaches thecubic form, and that at the point of change from one form t o theother there must be contraction along the vertical axis, or expansionalong the horizontal axes, or both. This contraction and expansionhas been actually observed by Fizeau, and may be regarded as pre-paratory to the change of crystalline form.It follows from thesefacts that the cubicad expansion of silver iodide ought t o be normalabove 146", and Rodwell has observed that the iodide experiences asudden and considerable contraction between 142" and 145.5'. but . ..afterwards expands regularly up to the melting point,C. H. B.Chlorides of Lime and Lithia. By I(. KRAUT (Annalen, 221,108--124).-This communication contains no new experiments ; theauthor merely controverts Lunge's experiments and deductions there-from as t o the constitution of bleaching powder and the so-called'' chloride of lithia." V. H. VINORGANIC CHEMISTRY. 17Separation of Gallium. By L. DE BOISBAIJDRAN (Contpt. rend.,97, 66-67, 142-144) .-From TeZZurium.-The tellurium, whichmust be in the state of tellurous acid, is precipitated by hydrogen sul-phide in the cold, in presence of a considerable quantity of free hydro-chloric acid.The filtrate is concentrated if necessary, heated toboiling, and treated with a current of hydrogen sulphide, the pre-cipitate of tellurium sulphide being filtered off.From Sdicon.-The solution is made strongly acid with hydro-chloric acid (any free sulphuric acid having been previously neutm-lised), evaporated to dryness, and heated for some time a t 120-125".The residue is moistened with strong hydrochloric acid, and theevaporation and ignition repeated. The gallinm is then dissolwd outby boiling dilute hydrochloric acid, the solution filtered, and the silicawashed, first with dilute hydrochloric acid, and finally with water ;the last traces of gallium are removed from the silica by dissolving itin potassium hydroxide, acidifying with hydrochloric acid, and repeat-ing the above treatment.From JfoZybdenumf.-(l.) The molybdenum is precipitated in thecold by hydrogen sulphide in presence of a considerable quantitg ofhydrochloric acid, the precipitate being washed with dilute hydro-chloric acid saturated with hydrogen sulphide.The filtrate is heatedfor some time at sbost 70°, then boiled, and any precipitate whichforms is filtered off. The second filtrate is concentrated, and againtreated in the same way. (2.) The solution is supersaturated withammonia, mixed with ammonium sulphide in excess, gently heated,and then acidified with hydrochloric acid, heated t o expel hydrogensulphide, and filtered.The filtrate is concentrated, the ammoniumsalts destroyed by heating with aqua regia, and the solution againtreated with ammoninm sulphide. The molybdenum sulphide is dis-solved in aqua regia, and the precipitation repeated in order to removethe last traces of gallium. Before applying methods (1) and (2)chlorides and lower oxides of molybdenum must be oxidieed by meansof nitric acid, excess of the latter being expelled by boiling withhydrochloric acid. (3.) The solution is acidified with sulphuric acid,mixed with a slight excess of ammonium sulphate, and the galliumalum precipitated by concentrating the liquid and adding three orfour times its volume of alcohol of 85".The alum is purified byrecrystallisation. I n accurate analyses, the greater part of the molyb-denum should be precipitated as sulphide by (1) or (2), then thegreater part of the gallium removed as alum, the last traces Eeiiigseparated from the concentrated mother-liquor by (1) or (2).Salts of Aurous Oxide : Colorimetric Estimation of Gold. ByA. CARNOT (Coinpt. rend., 97,169-170).-When an aqueous solution ofhydrogen phosphide is added gradually to a very dilute solution of goldchloride, either alone or mixed with phosphoric or arsenic acid, a rosecoloration is produced. J t would appear that this rose coloration(this vol., p. 115) is not due simply to a complex compound of aurousoxide and ferric oxide, but is produced also by simple salts of aurousoxide, such as the phosphate or arsenate.The presence of ferricoxide apparently increases the stability of these salts, for if anyC. H. B.VOL. XLVI. 18 ABSTRACTS OF CHEMICAL PAPERS.foreign salt is added to a solution of the rose-coloured compound freefrom iron, the rose-coloured compound becomes blue and yields aslight bluish precipitate, whereas in presence of iron a purple pre-cipitate is formed, which can be dried a t 100" without alteration.This reaction furnishes a colorimetric method for the rapid approxi-mate estimation of gold. The intensity of the tint given by thesolution under examination is compared with that of solutions of goldcontaining different amounts of gold in a constant volume (100 c.c.)of liquid.To obtain comparable solutions, the neutral solution ofgold chloride is mixed with a drop of hydrochloric acid, one or twodrops of ferric chloride, and a, few drops of arsenic acid, andthen dilutedto 100 C.C. A small quantity of zinc powder is added, the liquidagitated, allowed to clarify by standing, and the clear liquid decantedoff. The mineral to be examined is finely powdered, dissolved inaqua regia, the solution diluted, filtered, evaporated twice to drynessafter addition of a little nitric acid, and heated to dull redness. Theresidue is treated with chlorine-water, which leaves the ferric oxideundissolved, the solution is filtered, the chlorine expelled by boiling,and the solution treated as above.C. H. B.Action of Hydrochloric Acid on Stannous Sulphide. ByA. DITTE (Compt. rend., 97, 42-45).-Dry hydrochloric acid gas hasno action on crystallised stannous sulphide at the ordinary tempera-ture, but on gently heating it, decomposition commences, and stannouschloride and hydrogen sulphide are formed.Aqueous hydrochloric acid, even in dilute solutions, attacks crystal-lised stannous sulphide a t the ordinary temperature, but the degree ofconcentration of the acid has great influence on the decomposition,the phenomena being similar to those observed by Berthelot in thecase of galena (Mkcan. Chim., 2, 562). Decomposition takes placewith a solution containing 8.3 per cent. of acid; stannous chlorideand hydrogen sulphide are formed, and after some time equilibrium isestablished under somewhat complex conditions.When hydrogen sulphide is passed into a solution of stannouschloride, deep brown stannous sulphide may be produced, but inmany cases the precipitate consists of brilliant reddish-brown mica-ceous plates of a stannous chlorosulphide, which is decomposed byhydrogen siilphide and by water.The heat of formation of this com-pound is probably intermediate between that of stannous chloride andthat, of stannous sulphide, and its more or less complete decomposi-tion by water plays a part in t'he establishment of equilibrium. Whenhydrochloric acid is brought in contact w i t h an excess of stannoussulphide, or when hydrogen sulphide is passed into a solution ofstannous chloride, equilibrium is eventually established between thehydrogen sulphide and the hydrochloric acid, the relative proportionsof the two bodies depending on the temperature.The conditions ofequilibrium are complicated by the fact that the solubility of hydrogensulphide in solutions of stannous chloride decreases as the concentra-tion of the latter increases. If, therefore, a dilute solution of hydro-chloric acid acts on stannous sulphide, a small quantity of hydrogensulphide is liberated, and dissolves in the liquid without saturating itINORGANIC CHEMISTRY. 19whilst at the same time a correspondingly small quantity of stannouschloride is formed. If, on the other hand, a current of hydrogensnlphide is passed into a strong solution of stannous chloride, in whichthe gas is almost insoluble, a small quantity of stannous sulphide isformed, and the hydrochloric acid which is liberated is sufficient t oestablish equilibrium with the small quantity of hydrogen sulphide bywhich the liquid is saturated.The greater part of the stannouschloride therefore remains unchanged. If, however, the experimentis arranged so that the proportion of free hydrogen sulphide is thesame in both solutions, the proportion of hydrochloric acid will alsobe the same in both, and equilibrium will be established, although thetwo solutions contain very different amounts of stannous chloride. Itwould appear, therefore, that the stannous chloride plays a consider-able, although mainly a mechanical, part in the establishment ofequilibrium, by virtue of the fact that its presence diminishes thesolubility of hydrogen sulphide.No hydrochloride of stannous chloride appears to exist.Whenhydrochloric acid gas is passed over hydrated stannous chloride,SnC1,,2H20, the latter melts and yields an acid solution of stannouschloride, and a new hydrate, SnC12,H,0. The same hydrate is formedby the action of concentrated aqueous hydrochloric acid ou thehydrate, SnCI2,2H,O. The decomposition of stannous sulphide byhydrochloric acid is also materially affected by the temperaturewhich not only accelerates the action of the acid, but at the sametime diminishes the solubility of the hydrogen sulphide.Hydrated stannous sulphide is attacked by hydrochloric acid at theordinary temperature, decomposition taking place in more dilutesolutions and with greater rapidity, than in the case of the anhydroussulphide, although the products of the reaction and the conditions ofequilibrium are the same in both cases.Stannoustelluride is decomposed when heated in a current of dry hydrochloricacid gas, but is not attacked by the aqueous acid in concentratedsolutions.C. H. B.Similar phenomena are observed with stannous selenide.Antimony Pentiodide. By J. H. PENDLETON (Chern. News, 48,97).-Evidence of the existence of pentiodides of phosphorus andarsenic have been obtained (Abstr., 1881, 507 ; and Chem. News, 44,194), and the author has now investigated a pentiodide of antimony.Pure antimony is fused with excess of iodine in an atmosphere ofan inert gas in a sealed tube, and the mixture is kept well fused for anhour or two ; the tube is then heated at 130" in a sulphuric acid-bath,one end which is cooled with wat,er serving to collect the surplusiodine which distils off.The residue is a dark-brown crystallinemass, decomposible by water or long exposure to moist air. It meltsat 78-79", but is very unstable, readily undergoing decompositionwhen exposed t o even a moderate increase of temperature. Analyticalresults correspond " pretty closely to the formula Sb15."Several experiments were tried in which the temperature of theacid-bath and time of exposure were varied. D. A. L.e 2 0 ABSTRACTS OF CHEMICAL PAPERS.Chromic Acid and Hydrogen Peroxide. By H. MOISSAN(Compt.rend., 97, 96-99). The dark-blue compound formed bythe action of hydrogen peroxide on chromic acid is rapidly decomposedby water, but is much more stable in ethereal solution, the stabilitybeing, however, much diminished by the presence of a small quantityof alcohol. By using carefully purified ether and a solution of50 grams of potassium dichromate per litre, it is possible to obtain a t0" an ethereal solution of the dark-blue compound containing 0.5 percent. of chromium. This solution may be digested over calciumchloride, and can be kept for six o r eight hours at 0" without under-going sensible decomposition, but in time it decomposes, chromic acidbeing deposited on the sides of the tube. Dilute solutions are morestable than concentrated solutions.The ethereal solution when eva-porated a t - Z O O in a vacuum leaves a deep indigo-blue oily liquid,which readily redissolves in ether. I n contact with sodium, it evolvesh-j-drogen, and when only very gently heated it rapidly decomposesinto oxygen and chromic acid. The volumes of oxygen and hydrogengiven off correspond with the formula CrO3,H2O2 ; it would thereforeappear that the blue substance is a compound of chromic acid andhydrogen peroxide. The ethereal solution, in contact with phosphoricanhydride or other dehydrating agents, is rapidly decomposed withevolution of oxygen. It is also decomposed by acids, bases, leaddioxide, carbon, manganese dioxide, red lead, mercuric oxide, andsodium : in the last case with evolution of a mixture of hydrogen andoxygen.The ethereal solution also bleaches the skin like hydrogenperoxide.The blue compound cannot be obtained by the action of ozone onchromic acid, and it is only formed by electrolysis when hydrogenperoxide is also produced. C. H. B.Reduction of Iron Oxide with Carbonic Oxide. By R. AKER-MANN and SARNSTROM (DiiLgZ. polyt. J., 248,291-239) .-Experimentson the reduction of ferric oxide prove that it is readily convertedinto ferrosoferric oxide, and that magnetite parts with its oxygen toform ferrous oxide more easily than does ferrous oxide to yield metalliciron. One volume of carbonic oxide diluted with 20 volumes of car-bonic anhydride reduces ferric oxide to ferrosoferric oxide a t 450'.One part of carbonic oxide with 2.1 parts carbonic anhydride effectsa similar reduction at 300-350". By raising the temperature to8.'JO-900° 1 volume of carbonic oxide to 2.6 volumes carbonic anhy-dride still reduces magnetite, but with the proportions of one tothree this is no longer the case even a t a temperature of 800".Ferricoxide is easily reduced to magnetite in layers 2 mm. thick withoutany further reduction taking place SO long as the gaseous mixturecoiitains a t least three volumes carbonic anhydride to one volume ofcarbonic oxide. Ferrous oxide is easily reduced to the metallic stateat 850" when the ratio of carbonic oxide to anhydride is 1 t o 0.4.Experiments were made to oxidise iron by heating it in mixtures ofcarbonic oxide and carbonic anhydride, the results were unsuccessful,however, as regards the production of ferrous oxide. The reducingpower of an equivalent of carbonic oxide on ferrous oxide does noMINERALOWCAL CHEMISTRY. 21equal the oxidising power of an equivalent of carbonic anhydride oniron. In both cases, however, an increase of temperature gives riseto increased action. To reduce magnetite at a temperature of 350" toferrous oxide, the quantity of carbonic anhydride must not exceedtwo volumes to one of carbonic oxide. The authors find that for theproduction of 100 kilos. iron at least 64.3 kilos. of carbon in the formof carbonic oxide are required. Much less than this, however, isrepe8tedly employed in blast furnaces ; it follows therefore that a re-duction has taken place by the carbon as such, a reaction requiringa, higher initial temperature than the reduction with carbonic oxide.In regard to the consumption of fuel, therefore, very little saving is tobe anticipated. D. B