72 ABSTRACTS OF CHEMICAL PAPERS. I n o r g a n i c C h e mi s t r y. A Source of Atmospheric Carbonic Anhydride. By M. S . MEUNIER (Bied. Centr., 1880, 633-634) .-According to Cloez, iron which contains carbon and manganese yields carbonic anhydride on treatment with hot water ; this reaction may be assumed to go on in the neighbourhood of hot springs in metnlliferous districts, and thus to be a source of atmospheric carbonic anhydride. J. K. C. Id Liquid Sulphur Phosphide. By H. SCHULZE (J.pr. Chem. [2 22, 113-130) .-After referring to the experiments of Berzelius &n Rumme (Ber., 12, 1350), the latter of whom observed that there was no rise of temperature when phosphorus and sulphur were combined under warm water, and that on distilling the liquid product with steam, phosphorus passes over, leaving sulphur with a smaller propor- tion of phosphorus behind.The aiithor says he has repented the experiments and can confirm these observations, for not only was there no rise, but there was a slight fall of temperature on bringing sulphur and phosphorus together under water at 20". On cooling the liquid of the composition PIS, it was found to deposit crystals of phosphorus which redissolved when the temperature rose to 15", but separated out on again cooling. Liquid P,S deposited sulphur crystals. Wicke (Ann. Chenz. Pharm., 86, 115) had found that sulphur dissolved in P,S is deposited again on long standing in a cool place. Contrary to what is stated by Berzelius, the author finds that phosphorus rolatilises, when heated at 100" in a stream of hydrogen, whilst the hydrogen is partly converted into phosphine.The author did not succeed in effecting complete separation in this manner, as sometimes below and sometimes above lOO", a reaction took place in the liquid, causing an explosion. After a less violent reaction he has, however, occasionally succeeded, by very gradually heating small quantities in the paraffin-bath, in removing phosphorus, far below its boiling point, in a current of carbonic anhydride, and obtaining a residue of solid sulphur phosphide. The author finds, contrary to Gmelin-Kraut's notice, that carbon bisulphide is incapable of dissolving out the phosphorus from the sulphur ; but he succeeded in effecting a partial separation by dissolv- ing in carbon bisulphide, and then agitating the concentrated solution with alcohol, ether, chloroform, light petroleum, or other liquids, miscible with carbon bisulphide, but of slight solvent power for sul- phur, when most of the sulphur is separat.ed, and a liquid produced containing more or less sulphur in solution, according t o the amount of carbon bisulphide present.Phosphorus is also precipitated from its solution in carbon bisulphide by these reagents, so that it is only at first that pure sulphur is deposited. R. Bottger (J. pr. Chem., 12, 358) found that by shaking trans- parent phosphorus with potassium pdysulphide, and nllo wing it to stand for four days in a dark place, transparent sulphurphosphide wasINORGANIC CHEMISTRY. 73 produced. This the author can confirm, as also the decomposition of hydrogen persulphide by transparent phosphorus.In reviewing the facts as thus presented, it would seem that these sulphur phosphides are not chemical compounds, as there is no evolu- tion of heat on their formation, moreover they are readily decomposed by lowering the temperature, by solvents, and by heating in a current of gas. The decomposition of the potassium polysulphides and of hydro- gen persulphide by phosphorus seems to point the other way, but simi- lar actions are seen in crystalline salts giving up their water of crys- tallisation to the air, and the fact observed by Schone (Gmelin-Kraut, 2, 38), that alcohol precipitates potassium t'etrasulphide from a con- centrated solution of the pentasulphide, dissolving sulphur (and poly- sulphide).Since neither phosphorus nor sulphur by itself can decompose water, it seems remarkable that, a simple mixture of the two should have the power. But there are several examples which show that two bodies acting in common on a third, need by no means be combined. The action of chlorine and carbon on alumina is a case in point. The lique- faction of solid salts soluble in water, by contact with ice, is perfectly analogous to the formation of the liquid sulphur phosphide from its elements. F. L. T. The Earths of Samarskite. By C. MARIGNAC (Ann. Chim. Phys. [ S ] , 20, 535--557).-1n this memoir, the author describes only those oxides which form nitrates of relative high stability, and have been separated from those oxides, such as erbia, ytterbia, &c., which form nitrates easily decomposed by heat.They have been separated one from another by taking advantage of the different solubilities uf their double sulphates with potassium in a saturated solution of potassium sdphate. Details of this and other methods of separation employed, are given. The earths present in samarskite (from North America,) were thus divided into four groups :- I . Double Sulphates Soluble in less than 100 vob. of K,S04 Solution. Molecdar weight rises to 119.-Ytt.ria and terbia, with traces of the oxides of decipium and didymium, and probably traces of the oxide of Y a found in Group 11. The terbia, even after strong ignition, had a faint chamois tint, which disappeared on heating in a current of hydrogen, and reappeared when the oxide was again heated in the presence of air.These changes of colour were accompanied by very alight rariations in weight. 11. Double Subhates soluble in 100-200 ~01.9. of K,S04 Solution. Molecula?. weight varies between 119 and 120.-This portion consisted mainly of the oxide of a metal which the author provisionally desig- nates Ya. I t s molecular weight is about 120.5, a maximum between the molecular weights of those earths which most nearly approach it in their behaviour with potassium.sulphate solution. It forms colour- less salts, the solutions of which show no absorption spectrum. It is distinguished from all tbe metals of this class except yttrium and ytterbium by the very faint orange-yellow colour of its oxide, and by the fact that its salts show no absorption spectrum: from yttrium by the sparing solubility of its formate, and of its double potassium74 ABSTRACTS OF CHEMICAL PAPERS. sulphate ; from ytterbium by the much greater stability of its nitrate, and the ease with which the ignited oxide dissolves in dilute acids.Yr, is probably identical with that base, the'existence of which has been indicated by Delafontaine (Compt. rend., 90). Mole- culur weight falls from 119 to 115 as the Solubility decreases.-This group contains the oxide of Ya, a small quantity of the oxide of didymium which cannot be completely separated, and the oxide of a metal of Yp, which forms a nitrate readily decomposed by heat, but is included in this group, because its double potassium sulphate is very sparingly soluble in K,S04 solution.The molecular weight of the oxide of Y/3 is probably somewhat lower than 115.6. Its sulphate forms small, short, crystals similar to those of the sulphates of yttrium and didymium, but of a sulphur-yellow colour ; they have the constitution- 111. Dozd2.e Xulphates very slightly soluble in K,S04 Solution. 3(YP0.S03) + 8H,O. Solutions of the salts of this base give a well-defined absorption spectrum which closely resembles that ascribed by Delafontaine to decipium, or, still more closely, that described by Boisbaudran as peculiar to the new element samarium. Since these three bodies were obtained from samarskite by the same methods, it is probable that they are essentially one and the same substance. The molecular weight of the oxide of decipium is, howeber, according to Delafontaine, 130, whilst that of the oxide of YB is below 115.6.Further, the salts of decipium are colourless, those of Y have a yellow colour, the intensity of which increases the further the purification is carried. Probably either the substance obtained by Delafontaine is mixed with a considerable pro- portion of some other base having a high molecnlar weight, or that, obtained by the author contains a base forming yellow salts, a d having a much lower molecular weight. Further investigations are necessary to decide this point. IV. Double Sulphates irmlu hle in K2S04 Solution.-This group con- tained only didymium, which, however, could not be completely puri- fied from traces of other metals. Some of the products, pa,rticularly those most soluble in R,SO, solu- tion, contained traces of a substance which formed a yellowish oxide and is probably identical with the philippium of Delafontaine.The author has attempted to determine the relative solubilities of the formates of the metals present in samarskite and similar minerals, but without success. The solubility of these salts varies very greatly with the mechanical treatment to which they are subjected, and their solutions readily assume a state of supersaturation which they retain for a considerable length of time, even when in contact with the solid salt. C. H. B. metallic Oxides of the Iron Group. By H. MOISSAN (AWL. Chim. Phya. [ 51, 21, 199-255).--lron.-The black pyrophoric powder obtained by heating ferrous oxalate, or ferric oxide in a current of hydrogen or carbonic oxide, consists of ferrous oxide, a d contains TLO nietallic iron, or only traces.INORGANIC CHEMISTRY.75 The reduction of ferric oxide in a current of hydrogen commences a little above 330" ; between this temperature and the boiling point of sulphur, 440", the product is triferric tetroxide, Fe304. The same oxide is obtained when hydrated or anhydrous ferric oxide is heated to the melting point of zinc, 420°, in a current of pure and dry car- bonic oxide. If the carbonic oxide contains carbonic anhydride, a small quantity of ferruginous carbon is deposited. Between 500" and 600" ferrous oxide FeO, pyrophoric a t ordinary temperatures, is ob- tained. The results are the same even if the ferric oxide be mixed with a considerable proportion of alumina. By heating ferrous oxide to the softening point of glass in a current of hydrogen, metallic iron is obtained, but it is more or less agglutinated and is not pyrophoric.It is evident that the reduction of ferric oxide to metallic iron takes place in well-defined stages. When ferric oxide is heated to dull red- ness in a current of carbonic oxide, a more or less agglutinated powder is obtained, which is not pyrdphoric, and dissolves in dilute sulphuric acid with evolution of hydrogen, leaving a residue of carbon. Pyro- phoric metallic iron may, however, be obtained by heating ferric oxide in a current of perfectly dry hydrogen at 440" for a very long time, i e . , several days ; or by distilling iron amalgam at the lowest possible temperature. In all these cases of reduction, it is necessary, in order to obtain good results, that the reducing gas should be as pure and as dry as possible, and the current should be somewhat rapid.In addition to the methods above dcscribed, ferrous oxide may be prepared by heating ferrous oxalate out of contact with the air. That the pyrophoric black powders obtained in these experiments and shown by analysis to correspond in composition with FeO were really ferrous oxide, and not mixtures of triferric tetroxide with metallic iron was proved by the fact that they gave no hydrogen when treated with dilute acids, and did not decolorise an aqueous solution of iodine. When ferrous carbonate is heated out of contact with air, triferric tetroxide, Fe,04, is formed. Now, when carbonic anhydride is passed over heated ferrous oxide, carbonic oxide and triferric tetroxide are produced. It is probable therefore that the ferrous carbonate first splits up into ferrous oxide and carbonic anhydride, which react on each other in accordance with the equation 3Fe0 + CO, = Fe304 + CO.If the ferrous carbonate be heated slowly in a current of nitrogen, ferric oxide is formed: if rapidly heated, a small quantity of pyro- pboric ferrous oxide is produced. Allotropic ModiJicution of Triferric Tetyoxide.-In addition to the methods described above, triferric tetroxide may be obtained by heat- ing ferric oxide, or metallic iron reduced by hydrogen, to redness in a current of hydrogen saturated with aqueous rapour a t 90" ; by heating ferrous carbonate to dull redness in a currertt of carbonic anhydride; and by heating reduced iron in a current of carbonic anhydride at 440".As thus prepared a t low temperatures, it is a black magnetic powder, density 4.86, is readily attacked by concen- trated nitric acid, and when heated becomes incandescent, forming Fea03. This is one modification of triferric tetroxide. When the ferric oxide produced by heating this modification is exposed to a high tempcratnre, about 1500c, it gives off oxygen, and is reconverted into76 ABSTRAOTS OF CHEMICAL PAPERS. triferric tetroxide, a black magnetic substance, sp. gr. 5 to 5.09, almost unattacked by boiling concentrated nitric acid, and not forming ferric oxide when heated ; it is, in fact, the most stable oxide of iron. When the first modification of the triferric tetroxide is heated to whiteness in a current of nitrogen, it agglomerates, increases in deiisity, no longer forms Fe,O, when heated, and has all the properties of the second modification.It is evident that there are two well-defined allotropic forms of triferric tetroxide, one formed a t low, the other at high temperatures. The author explains the fact that the modifica- tion ,8 formed at high temperatures does not give ferric oxide when heated, by assumifig that when the modification a, formed a t low tem- peratures, is converted into ferric oxide, the heat evolved is less than that evolved when the modification a is converted into the modifica- tion 6, and therefore the formation of ferric oxide from the latter would be an endothermic reaction.Allotropic Modijicatiow of -Ferrous Oxide.-When prepared a t tem- peratures below GOO", ferrous oxide is a pulverulent, ivory-black sub- stance, which readily burns in the air, forming Fe203, the temperature of the combustion being sufficiently high to convert a portion of the Fe,O, into Fe304. It is oxidised with incandescence by nitric acid, burns, when gently heated in nitrogen monoxide or dioxide, displaces ammonia from its combinations, and decomposes water slowly a t ordinary temperatures, more rapidly a t 100". If, however, it be ob- tained a t a high temperature, as for example, by heating reduced iron in a current of carbonic anhydride, it is not pyrophoric a t ordinary temperatures, and is converted into triferric tetroxide when burnt. When pyrophoric ferrous oxide is heated in an atmosphere of nitrogen to the melting point of silver, it loses its pyrophoric properties, is not attacked by dilute acetic acid, and when heated burns like tinder, forming Fe304.Further, when the modification of ferrous oxide formed a t low temperatures is heated in a current of carbonic anhy- dride, it is converted into the low temperature modificat'ion of triferric tetroxide, but the high temperature modification of ferrous oxide, when heated, gives the k3 or high temperature modification of triferric tetroxide. 1Cfmgal.~ese.-When manganese dioxide, obtained by calcining man- ganous nitrate, is heated at 280" in a current of hydrogen, it becomes incandescent, and is reduced to the monoxide MnO. The reduction of the dioxide commences at 230", and if the temperature is kept constant at this point, manganese sesquioxide, Mn203, having a deep maroon colour: is obtained.At a somewhat higher temperature trimanganic tetroxide, Mn,04, is formed, and this, when carefully heated in the presence of air, is converted into the sesquioxide which, when strongly heated, gives trimanganic tetroxide, unalterable by heating in contact with air. When the trimanganic tetroxide obtained at a low tempera- ture is heated a t 260" in a current of pure and dry hydrogen, it is rapidly reduced with formation of the monoxide, a green powder which oxidises rapidly, and if previously heated to 140" takes fire when thrown into the air. Out of contact with air, this oxide is but slightly acted on by water, but in presence of air it quickly oxidises.A non- pyrophoric variety of the monoxide has been obtained by Deville inINORGANIC CREMISTRY. 77 regular octohedra of a beautiful green tint, by reducing a higher oxide a t a red heat in a current of hydrogen containing hydrochloric acid. When the crystalline amalgam of manganese obtained by electro- lysis, is distilled a t 440" in a current of pure hydrogen, a light, porous, blackish grey mass of metallic manganese is left. It is oxidised with incandescence by nitric acid, and portions of it take fire when thrown into the air. Nickel.-Hydrated or anhydrous nickel sesquioxide, Ni203, heated a t 190" in a current of hydrogen, is reduced to the grey magnetic oxide, Ni,O,. At ZL somewhat higher temperature the yellowish-green monoxide is obtained.This undergoes no further change at 200", but at 230-240" is reduced to metallic nickel, in the form of a black powder, which is pyrophoric at ordinary temperatures, but does not burn so brilliantly a s metallic iron reduced at 440". That the tem- perature of combustion is not very high is shown by the fact that the product consists mainly of the sesquioxide, whereas the only oxide stable a t a high temperature is the monoxide. When nickel amalgam is distilled in a current of hydrogen a t the lowest possible temperature, the nickel obtained is not pyrophoric. Nickel monoxide obtained by reduction is green when cold, yellow when hot. It readily absorbs oxygen even a t ordinary temperatures, and is partially converted by nitric acid, with elevation of tempera- ture, into the sesquioxide, which dissolves in acids with evolution of oxygen.The monoxide when heated in air or oxygen to 350-440", forms a blackish powder, the composition of which depends on the temperature ; at 600" this black powder is reconverted into the mon- oxide, showing that the higher oxides of nickel can only exist below a certain temperature. Nickel monoxide dissolves in hydrochloric and sulphuric acids, with rise of temperature ; it forms a beautiful violet solution with ammonia, and displaces ammonia from its salts. Cob&.-The sesquioxide is reduced a t the same temperature as nickel sesquioxide, viz., 190-200". A t 250" metallic cobalt is ob- tained in the form of a black powder, which takes fire a t ordinary temperatures, forming mainly the oxide Co304.If the reduction takes place a t 700°, the metallic cobalt is never pyrophoric ; and the metal obtained by distilling the amalgam in hydrogen at the lowest possible temperature behaves in the same way Cobalt monoxide obtained by reduction is a dark colonred, readily oxidisable powder, which when slightly heated in presence of air, becomes incandescent, and is converted into the sesquioxide. The sesquioxide at a higher temperature forms the magnetic oxide, Cos04, and this when still further heated gives the monoxide, COO, the oxide stable a t high temperatures. Chromium.-The sesquioxide is not reduced by hydrogen at any known temperature. When strongly heated, it becomes incandescent', and is rendered almost insoluble in acids.The ignited oxide when heated a t 440" in a current of hydrogen sulphide, dry chlorine, oxygen, or bromine vapour, undergoes no change whatever. When the non-ignited oxide is heated at 440" in a current of hydro- gen sulphide, it forms chromium sesquisulphide, the properties of which have been previously described by the author (this Journal,78 ABSTRACTS OF CHEMICAL PAPERS. Abstr., 1880, 527). Heated in a current of oxygen, the sesquioxide is oxidised to the dioxide, CrOz, as already pointed out by Kriiger. This dioxide has a deep grey colour, and when calcined evolves oxygen and is reconverted into the sesquioxide. When heated with hydrochloric acid or a mixture of snlphuric acid and sodium chloride, chlorine is given off. Fused with potash, out of contact with air, it gives potas- sium chromate and chromium sesquioxide.When the hydrated sesquioxide is gradually heated in a current of chlorine, it first loses water, and at 440" is converted into chromyl dichloride. The dehydrated but not ignited sesquioxide is partially converted into chromic chloride, but no chromyl dichloride is formed ; if, however, the chlorine be saturated with aqueous vapour a t 8- lo", the latter compound is formed in considerable quantity. I f the chlorine be saturat,ed with aqueous vnpour at 20°, but little chromyl dichloride is produced. Chromyl dichloride is also formed when a current of moist chlorine is passed over chromic chloride heated a t 440°, but a moist inert gas, such as carbonic anhydride, does not pro- duce this change. Further, it is found that the hydrated sesquioxide when dried at 440" contains 5-10 per cent.of water. Probably in all these cases chromic chloride i s first formed, and then decomposed with production of chromyl dichloride, by the excess of chlorine, and the water present either in the gas or in the oxide. By arresting the action of moist chlorine on the chromium sesqui- oxide a t the moment when the red vapours of Cr02Clz are given off, a brown powder is obtained, which approaches in composition the oxy- chlorides described by Moberg. Bromine vapour, under the same conditions, also attacks the non- ignited chromium sesquioxide. Chromium amalgam may be obtained by the action of sodium amal- gam on the chloride, bromide, or iodide of chromium. Like the amalgams of all the metals of this gronp, it oxidises readily.When distilled in a current of hydrogen a t the lowest possible temperature, metallic chromium is left in the form of a black, amorphous, very slightly agglutinated substance, which takes fire when exposed to the air. If distilled at temperatures above 355", the residual chromium is not, pyrophoric. The chromium thus obtained is more readily oxidised, and is more soluble in acids than that obtained by Deville by reducing the sesquioxide with carbon. When heated to dull redness in a carrent of dry carbonic anhydride, it is converted into the sesquioxide. If this family of elements be arranged in the order, chromium, manganese, iron, cobalt, nickel, their affinity for oxygen and the heats of formation of the oxides, chlorides, bromides, iodides, and sulphides decrease as the atomic weight increases.Atomic Weight of Antimony. By R. SCHNEIDER (J. pr. Chem [: 21, 22,131-147).-A controversial paper in reply to strictures passed by Kessler on the author's results. The ant'hor, together with Cooke, finds 120 as the atomic weight; Ressler, in common with Dexter and Dumas, obtains 122. The author's method is by reduction of a pure specimen of antimony C. H. B.INORQANIC CHEMISTRY. 79 glance by hydrogen, and in 1856 (Pogg. AWL., 98,293-308) obtained 120.3 by this means. The author gives three recent determinations made on the same plan, and with an antimony glance containing- Insoluble residue Antimony sulphide. (quartz). Calcium carbonate. Ferrous sulphide. 99-81 1 0-108 0-048 0.033 As on heating it in a current of hydrogen or carbonic anhydride it decrepitated, giving out a slight empyreumatic odour, it was heated before reduction in hydrogen to 290-320” until decrepitation had ceased and thg issuing gas was odourless.Antimony sulphide itself undergoea no reduction at this temperature. The mean result from the three experiments is 120.182, which com- pares well with 120.3, the one previously obtained. Dexter, like Berzelius, determined the atomic weight by oxidation of the metal with nitric acid and ignition of the residue. The suppo- sition that the residue on ignition is a thoroughly stable body of the composition of antimony tetroxide is incorrect, as Bunsen has shown (Annalem, 192, 317). Kessler’s method (Pogg. Anw,., 95,204 (1855), and 113,134 (1861)) is to oxidise antimonious chloride in concentrated hydrochloric acid b-y a volumet,ric solution of potassium dichromate in excess, and to add, after dilution of the solution, an excess of ferrous chloride and then titrate back the excess of ferrobs chloride by chromium solution.I n 1855 he obtained the atomic weight, 123.7, and in 1861 one experi- ment with antimony, one with antimony trioxide, and a third with antimonious chloride, gave 122.29. The author criticises Kessler’s method, and infers that no reliance is to be placed on it. Products of Decomposition and Metamorphosis of Uranyl Sulphide. By C. Z~MMERMANN (AnnuZen, 204-2’24) .-Uranium does not exhibit a great affinity f o r sulphur, since uranyl sulphide is con- verted into uranyl hydroxide by washing with warm water.It has been noticed by Reme16 (Pogg. Ann., 124, 120) that when uranyl sul- phide is left in contact with ammonium sulphide the liquid becomes of a brown to deep black colour. The author shows that this is due to the solubility of the uranyl sulphide i n ammonium carbonate con- tained in the ammonium sulphide. When precipitated uranyl sulphide is left for a long time in contact with ammonium sulphide in the cold, two bodies are obtained, one red the other black. The author shows that when the ammonium sulphide contains a considerable amount of thiosulphate the red body is formed, but that in absence of thio- sulphate the black body is obtained. The thiosulphate is formed by the action of atmospheric oxygen on the ammonium snlphide.The formula assigned to uranium-black is Ur,O,< O>Ur + 2Ur,0,. Uranium-red seems to consist of an oxygen-compound which contains, besides sulphur, a second base (potassium, sodium, ammonium, 01’ barium), and some uranium, probably as uranyl sulphide. Anhydrous F. L. T. 080 ABSTRACTS OF CHEMICAL PAPERS. uranium-red, on the assumption that it is perfectly free from admixture, OK may be formulated thus : U6SK209 = 2Ur203 + UrzOz< SK. G. T, A.72 ABSTRACTS OF CHEMICAL PAPERS.I n o r g a n i c C h e mi s t r y.A Source of Atmospheric Carbonic Anhydride. By M. S .MEUNIER (Bied. Centr., 1880, 633-634) .-According to Cloez, ironwhich contains carbon and manganese yields carbonic anhydride ontreatment with hot water ; this reaction may be assumed to go on inthe neighbourhood of hot springs in metnlliferous districts, and thus tobe a source of atmospheric carbonic anhydride. J.K. C.Id Liquid Sulphur Phosphide. By H. SCHULZE (J.pr. Chem. [222, 113-130) .-After referring to the experiments of Berzelius &nRumme (Ber., 12, 1350), the latter of whom observed that there wasno rise of temperature when phosphorus and sulphur were combinedunder warm water, and that on distilling the liquid product withsteam, phosphorus passes over, leaving sulphur with a smaller propor-tion of phosphorus behind. The aiithor says he has repented theexperiments and can confirm these observations, for not only was thereno rise, but there was a slight fall of temperature on bringing sulphurand phosphorus together under water at 20".On cooling the liquidof the composition PIS, it was found to deposit crystals of phosphoruswhich redissolved when the temperature rose to 15", but separated outon again cooling. Liquid P,S deposited sulphur crystals. Wicke(Ann. Chenz. Pharm., 86, 115) had found that sulphur dissolved inP,S is deposited again on long standing in a cool place.Contrary to what is stated by Berzelius, the author finds thatphosphorus rolatilises, when heated at 100" in a stream of hydrogen,whilst the hydrogen is partly converted into phosphine. The authordid not succeed in effecting complete separation in this manner, assometimes below and sometimes above lOO", a reaction took place inthe liquid, causing an explosion. After a less violent reaction he has,however, occasionally succeeded, by very gradually heating smallquantities in the paraffin-bath, in removing phosphorus, far below itsboiling point, in a current of carbonic anhydride, and obtaining aresidue of solid sulphur phosphide.The author finds, contrary to Gmelin-Kraut's notice, that carbonbisulphide is incapable of dissolving out the phosphorus from thesulphur ; but he succeeded in effecting a partial separation by dissolv-ing in carbon bisulphide, and then agitating the concentrated solutionwith alcohol, ether, chloroform, light petroleum, or other liquids,miscible with carbon bisulphide, but of slight solvent power for sul-phur, when most of the sulphur is separat.ed, and a liquid producedcontaining more or less sulphur in solution, according t o the amountof carbon bisulphide present. Phosphorus is also precipitated from itssolution in carbon bisulphide by these reagents, so that it is only atfirst that pure sulphur is deposited.R. Bottger (J.pr. Chem., 12, 358) found that by shaking trans-parent phosphorus with potassium pdysulphide, and nllo wing it tostand for four days in a dark place, transparent sulphurphosphide waINORGANIC CHEMISTRY. 73produced. This the author can confirm, as also the decomposition ofhydrogen persulphide by transparent phosphorus.In reviewing the facts as thus presented, it would seem that thesesulphur phosphides are not chemical compounds, as there is no evolu-tion of heat on their formation, moreover they are readily decomposedby lowering the temperature, by solvents, and by heating in a currentof gas.The decomposition of the potassium polysulphides and of hydro-gen persulphide by phosphorus seems to point the other way, but simi-lar actions are seen in crystalline salts giving up their water of crys-tallisation to the air, and the fact observed by Schone (Gmelin-Kraut,2, 38), that alcohol precipitates potassium t'etrasulphide from a con-centrated solution of the pentasulphide, dissolving sulphur (and poly-sulphide).Since neither phosphorus nor sulphur by itself can decompose water,it seems remarkable that, a simple mixture of the two should have thepower. But there are several examples which show that two bodiesacting in common on a third, need by no means be combined.Theaction of chlorine and carbon on alumina is a case in point. The lique-faction of solid salts soluble in water, by contact with ice, is perfectlyanalogous to the formation of the liquid sulphur phosphide from itselements. F. L. T.The Earths of Samarskite. By C. MARIGNAC (Ann. Chim. Phys.[ S ] , 20, 535--557).-1n this memoir, the author describes only thoseoxides which form nitrates of relative high stability, and have beenseparated from those oxides, such as erbia, ytterbia, &c., which formnitrates easily decomposed by heat. They have been separated onefrom another by taking advantage of the different solubilities uf theirdouble sulphates with potassium in a saturated solution of potassiumsdphate.Details of this and other methods of separation employed,are given. The earths present in samarskite (from North America,)were thus divided into four groups :-I . Double Sulphates Soluble in less than 100 vob. of K,S04 Solution.Molecdar weight rises to 119.-Ytt.ria and terbia, with traces of theoxides of decipium and didymium, and probably traces of the oxide ofY a found in Group 11. The terbia, even after strong ignition, had afaint chamois tint, which disappeared on heating in a current ofhydrogen, and reappeared when the oxide was again heated in thepresence of air. These changes of colour were accompanied by veryalight rariations in weight.11. Double Subhates soluble in 100-200 ~01.9. of K,S04 Solution.Molecula?. weight varies between 119 and 120.-This portion consistedmainly of the oxide of a metal which the author provisionally desig-nates Ya. I t s molecular weight is about 120.5, a maximum betweenthe molecular weights of those earths which most nearly approach itin their behaviour with potassium.sulphate solution.It forms colour-less salts, the solutions of which show no absorption spectrum. It isdistinguished from all tbe metals of this class except yttrium andytterbium by the very faint orange-yellow colour of its oxide, and bythe fact that its salts show no absorption spectrum: from yttrium bythe sparing solubility of its formate, and of its double potassiu74 ABSTRACTS OF CHEMICAL PAPERS.sulphate ; from ytterbium by the much greater stability of its nitrate,and the ease with which the ignited oxide dissolves in dilute acids.Yr, is probably identical with that base, the'existence of which hasbeen indicated by Delafontaine (Compt.rend., 90).Mole-culur weight falls from 119 to 115 as the Solubility decreases.-Thisgroup contains the oxide of Ya, a small quantity of the oxide ofdidymium which cannot be completely separated, and the oxide of ametal of Yp, which forms a nitrate readily decomposed by heat, but isincluded in this group, because its double potassium sulphate is verysparingly soluble in K,S04 solution. The molecular weight of the oxideof Y/3 is probably somewhat lower than 115.6. Its sulphate formssmall, short, crystals similar to those of the sulphates of yttrium anddidymium, but of a sulphur-yellow colour ; they have the constitution-111. Dozd2.e Xulphates very slightly soluble in K,S04 Solution.3(YP0.S03) + 8H,O.Solutions of the salts of this base give a well-defined absorptionspectrum which closely resembles that ascribed by Delafontaine todecipium, or, still more closely, that described by Boisbaudran aspeculiar to the new element samarium.Since these three bodies wereobtained from samarskite by the same methods, it is probable that theyare essentially one and the same substance. The molecular weight ofthe oxide of decipium is, howeber, according to Delafontaine, 130, whilstthat of the oxide of YB is below 115.6. Further, the salts of decipiumare colourless, those of Y have a yellow colour, the intensity of whichincreases the further the purification is carried.Probably either thesubstance obtained by Delafontaine is mixed with a considerable pro-portion of some other base having a high molecnlar weight, or that,obtained by the author contains a base forming yellow salts, a dhaving a much lower molecular weight. Further investigations arenecessary to decide this point.IV. Double Sulphates irmlu hle in K2S04 Solution.-This group con-tained only didymium, which, however, could not be completely puri-fied from traces of other metals.Some of the products, pa,rticularly those most soluble in R,SO, solu-tion, contained traces of a substance which formed a yellowish oxideand is probably identical with the philippium of Delafontaine.The author has attempted to determine the relative solubilities ofthe formates of the metals present in samarskite and similar minerals,but without success.The solubility of these salts varies very greatlywith the mechanical treatment to which they are subjected, and theirsolutions readily assume a state of supersaturation which they retainfor a considerable length of time, even when in contact with the solidsalt. C. H. B.metallic Oxides of the Iron Group. By H. MOISSAN (AWL.Chim. Phya. [ 51, 21, 199-255).--lron.-The black pyrophoric powderobtained by heating ferrous oxalate, or ferric oxide in a current ofhydrogen or carbonic oxide, consists of ferrous oxide, a d contains TLOnietallic iron, or only tracesINORGANIC CHEMISTRY. 75The reduction of ferric oxide in a current of hydrogen commences alittle above 330" ; between this temperature and the boiling point ofsulphur, 440", the product is triferric tetroxide, Fe304.The sameoxide is obtained when hydrated or anhydrous ferric oxide is heatedto the melting point of zinc, 420°, in a current of pure and dry car-bonic oxide. If the carbonic oxide contains carbonic anhydride, asmall quantity of ferruginous carbon is deposited. Between 500" and600" ferrous oxide FeO, pyrophoric a t ordinary temperatures, is ob-tained. The results are the same even if the ferric oxide be mixedwith a considerable proportion of alumina. By heating ferrous oxideto the softening point of glass in a current of hydrogen, metallic ironis obtained, but it is more or less agglutinated and is not pyrophoric.It is evident that the reduction of ferric oxide to metallic iron takesplace in well-defined stages.When ferric oxide is heated to dull red-ness in a current of carbonic oxide, a more or less agglutinated powderis obtained, which is not pyrdphoric, and dissolves in dilute sulphuricacid with evolution of hydrogen, leaving a residue of carbon. Pyro-phoric metallic iron may, however, be obtained by heating ferric oxidein a current of perfectly dry hydrogen at 440" for a very long time,i e . , several days ; or by distilling iron amalgam at the lowest possibletemperature. In all these cases of reduction, it is necessary, in orderto obtain good results, that the reducing gas should be as pure and asdry as possible, and the current should be somewhat rapid.In addition to the methods above dcscribed, ferrous oxide may beprepared by heating ferrous oxalate out of contact with the air.Thatthe pyrophoric black powders obtained in these experiments andshown by analysis to correspond in composition with FeO were reallyferrous oxide, and not mixtures of triferric tetroxide with metallic ironwas proved by the fact that they gave no hydrogen when treated withdilute acids, and did not decolorise an aqueous solution of iodine.When ferrous carbonate is heated out of contact with air, triferrictetroxide, Fe,04, is formed. Now, when carbonic anhydride is passedover heated ferrous oxide, carbonic oxide and triferric tetroxide areproduced. It is probable therefore that the ferrous carbonate firstsplits up into ferrous oxide and carbonic anhydride, which react oneach other in accordance with the equation 3Fe0 + CO, = Fe304 + CO.If the ferrous carbonate be heated slowly in a current of nitrogen,ferric oxide is formed: if rapidly heated, a small quantity of pyro-pboric ferrous oxide is produced.Allotropic ModiJicution of Triferric Tetyoxide.-In addition to themethods described above, triferric tetroxide may be obtained by heat-ing ferric oxide, or metallic iron reduced by hydrogen, to rednessin a current of hydrogen saturated with aqueous rapour a t 90" ; byheating ferrous carbonate to dull redness in a currertt of carbonicanhydride; and by heating reduced iron in a current of carbonicanhydride at 440".As thus prepared a t low temperatures, it is ablack magnetic powder, density 4.86, is readily attacked by concen-trated nitric acid, and when heated becomes incandescent, formingFea03. This is one modification of triferric tetroxide.When the ferricoxide produced by heating this modification is exposed to a hightempcratnre, about 1500c, it gives off oxygen, and is reconverted int76 ABSTRAOTS OF CHEMICAL PAPERS.triferric tetroxide, a black magnetic substance, sp. gr. 5 to 5.09, almostunattacked by boiling concentrated nitric acid, and not forming ferricoxide when heated ; it is, in fact, the most stable oxide of iron. Whenthe first modification of the triferric tetroxide is heated to whitenessin a current of nitrogen, it agglomerates, increases in deiisity, nolonger forms Fe,O, when heated, and has all the properties of thesecond modification.It is evident that there are two well-definedallotropic forms of triferric tetroxide, one formed a t low, the other athigh temperatures. The author explains the fact that the modifica-tion ,8 formed at high temperatures does not give ferric oxide whenheated, by assumifig that when the modification a, formed a t low tem-peratures, is converted into ferric oxide, the heat evolved is less thanthat evolved when the modification a is converted into the modifica-tion 6, and therefore the formation of ferric oxide from the latterwould be an endothermic reaction.Allotropic Modijicatiow of -Ferrous Oxide.-When prepared a t tem-peratures below GOO", ferrous oxide is a pulverulent, ivory-black sub-stance, which readily burns in the air, forming Fe203, the temperatureof the combustion being sufficiently high to convert a portion of theFe,O, into Fe304.It is oxidised with incandescence by nitric acid,burns, when gently heated in nitrogen monoxide or dioxide, displacesammonia from its combinations, and decomposes water slowly a tordinary temperatures, more rapidly a t 100". If, however, it be ob-tained a t a high temperature, as for example, by heating reduced ironin a current of carbonic anhydride, it is not pyrophoric a t ordinarytemperatures, and is converted into triferric tetroxide when burnt.When pyrophoric ferrous oxide is heated in an atmosphere of nitrogento the melting point of silver, it loses its pyrophoric properties, is notattacked by dilute acetic acid, and when heated burns like tinder,forming Fe304.Further, when the modification of ferrous oxideformed a t low temperatures is heated in a current of carbonic anhy-dride, it is converted into the low temperature modificat'ion of triferrictetroxide, but the high temperature modification of ferrous oxide,when heated, gives the k3 or high temperature modification of triferrictetroxide.1Cfmgal.~ese.-When manganese dioxide, obtained by calcining man-ganous nitrate, is heated at 280" in a current of hydrogen, it becomesincandescent, and is reduced to the monoxide MnO. The reduction ofthe dioxide commences at 230", and if the temperature is kept constantat this point, manganese sesquioxide, Mn203, having a deep marooncolour: is obtained.At a somewhat higher temperature trimanganictetroxide, Mn,04, is formed, and this, when carefully heated in thepresence of air, is converted into the sesquioxide which, when stronglyheated, gives trimanganic tetroxide, unalterable by heating in contactwith air. When the trimanganic tetroxide obtained at a low tempera-ture is heated a t 260" in a current of pure and dry hydrogen, it israpidly reduced with formation of the monoxide, a green powder whichoxidises rapidly, and if previously heated to 140" takes fire whenthrown into the air. Out of contact with air, this oxide is but slightlyacted on by water, but in presence of air it quickly oxidises. A non-pyrophoric variety of the monoxide has been obtained by Deville iINORGANIC CREMISTRY.77regular octohedra of a beautiful green tint, by reducing a higher oxidea t a red heat in a current of hydrogen containing hydrochloric acid.When the crystalline amalgam of manganese obtained by electro-lysis, is distilled a t 440" in a current of pure hydrogen, a light, porous,blackish grey mass of metallic manganese is left. It is oxidised withincandescence by nitric acid, and portions of it take fire when throwninto the air.Nickel.-Hydrated or anhydrous nickel sesquioxide, Ni203, heateda t 190" in a current of hydrogen, is reduced to the grey magneticoxide, Ni,O,. At ZL somewhat higher temperature the yellowish-greenmonoxide is obtained. This undergoes no further change at 200", butat 230-240" is reduced to metallic nickel, in the form of a blackpowder, which is pyrophoric at ordinary temperatures, but does notburn so brilliantly a s metallic iron reduced at 440".That the tem-perature of combustion is not very high is shown by the fact that theproduct consists mainly of the sesquioxide, whereas the only oxidestable a t a high temperature is the monoxide.When nickel amalgam is distilled in a current of hydrogen a t thelowest possible temperature, the nickel obtained is not pyrophoric.Nickel monoxide obtained by reduction is green when cold, yellowwhen hot. It readily absorbs oxygen even a t ordinary temperatures,and is partially converted by nitric acid, with elevation of tempera-ture, into the sesquioxide, which dissolves in acids with evolution ofoxygen.The monoxide when heated in air or oxygen to 350-440",forms a blackish powder, the composition of which depends on thetemperature ; at 600" this black powder is reconverted into the mon-oxide, showing that the higher oxides of nickel can only exist below acertain temperature. Nickel monoxide dissolves in hydrochloric andsulphuric acids, with rise of temperature ; it forms a beautiful violetsolution with ammonia, and displaces ammonia from its salts.Cob&.-The sesquioxide is reduced a t the same temperature asnickel sesquioxide, viz., 190-200". A t 250" metallic cobalt is ob-tained in the form of a black powder, which takes fire a t ordinarytemperatures, forming mainly the oxide Co304. If the reduction takesplace a t 700°, the metallic cobalt is never pyrophoric ; and the metalobtained by distilling the amalgam in hydrogen at the lowestpossible temperature behaves in the same wayCobalt monoxide obtained by reduction is a dark colonred, readilyoxidisable powder, which when slightly heated in presence of air,becomes incandescent, and is converted into the sesquioxide. Thesesquioxide at a higher temperature forms the magnetic oxide, Cos04,and this when still further heated gives the monoxide, COO, the oxidestable a t high temperatures.Chromium.-The sesquioxide is not reduced by hydrogen at anyknown temperature.When strongly heated, it becomes incandescent',and is rendered almost insoluble in acids. The ignited oxide whenheated a t 440" in a current of hydrogen sulphide, dry chlorine, oxygen,or bromine vapour, undergoes no change whatever.When the non-ignited oxide is heated at 440" in a current of hydro-gen sulphide, it forms chromium sesquisulphide, the properties ofwhich have been previously described by the author (this Journal78 ABSTRACTS OF CHEMICAL PAPERS.Abstr., 1880, 527).Heated in a current of oxygen, the sesquioxide isoxidised to the dioxide, CrOz, as already pointed out by Kriiger. Thisdioxide has a deep grey colour, and when calcined evolves oxygen andis reconverted into the sesquioxide. When heated with hydrochloricacid or a mixture of snlphuric acid and sodium chloride, chlorine isgiven off. Fused with potash, out of contact with air, it gives potas-sium chromate and chromium sesquioxide.When the hydrated sesquioxide is gradually heated in a current ofchlorine, it first loses water, and at 440" is converted into chromyldichloride.The dehydrated but not ignited sesquioxide is partiallyconverted into chromic chloride, but no chromyl dichloride is formed ;if, however, the chlorine be saturated with aqueous vapour a t 8- lo",the latter compound is formed in considerable quantity. I f thechlorine be saturat,ed with aqueous vnpour at 20°, but little chromyldichloride is produced. Chromyl dichloride is also formed when acurrent of moist chlorine is passed over chromic chloride heated a t440°, but a moist inert gas, such as carbonic anhydride, does not pro-duce this change. Further, it is found that the hydrated sesquioxidewhen dried at 440" contains 5-10 per cent.of water. Probably inall these cases chromic chloride i s first formed, and then decomposedwith production of chromyl dichloride, by the excess of chlorine, andthe water present either in the gas or in the oxide.By arresting the action of moist chlorine on the chromium sesqui-oxide a t the moment when the red vapours of Cr02Clz are given off, abrown powder is obtained, which approaches in composition the oxy-chlorides described by Moberg.Bromine vapour, under the same conditions, also attacks the non-ignited chromium sesquioxide.Chromium amalgam may be obtained by the action of sodium amal-gam on the chloride, bromide, or iodide of chromium. Like theamalgams of all the metals of this gronp, it oxidises readily.Whendistilled in a current of hydrogen a t the lowest possible temperature,metallic chromium is left in the form of a black, amorphous, veryslightly agglutinated substance, which takes fire when exposed tothe air. If distilled at temperatures above 355", the residualchromium is not, pyrophoric. The chromium thus obtained is morereadily oxidised, and is more soluble in acids than that obtained byDeville by reducing the sesquioxide with carbon. When heated todull redness in a carrent of dry carbonic anhydride, it is convertedinto the sesquioxide.If this family of elements be arranged in the order, chromium,manganese, iron, cobalt, nickel, their affinity for oxygen and the heatsof formation of the oxides, chlorides, bromides, iodides, and sulphidesdecrease as the atomic weight increases.Atomic Weight of Antimony.By R. SCHNEIDER (J. pr. Chem[: 21, 22,131-147).-A controversial paper in reply to strictures passedby Kessler on the author's results.The ant'hor, together with Cooke, finds 120 as the atomic weight;Ressler, in common with Dexter and Dumas, obtains 122.The author's method is by reduction of a pure specimen of antimonyC. H. BINORQANIC CHEMISTRY. 79glance by hydrogen, and in 1856 (Pogg. AWL., 98,293-308) obtained120.3 by this means.The author gives three recent determinations made on the sameplan, and with an antimony glance containing-Insoluble residueAntimony sulphide. (quartz). Calcium carbonate. Ferrous sulphide.99-81 1 0-108 0-048 0.033As on heating it in a current of hydrogen or carbonic anhydride itdecrepitated, giving out a slight empyreumatic odour, it was heatedbefore reduction in hydrogen to 290-320” until decrepitation hadceased and thg issuing gas was odourless. Antimony sulphide itselfundergoea no reduction at this temperature.The mean result from the three experiments is 120.182, which com-pares well with 120.3, the one previously obtained.Dexter, like Berzelius, determined the atomic weight by oxidationof the metal with nitric acid and ignition of the residue.The suppo-sition that the residue on ignition is a thoroughly stable body of thecomposition of antimony tetroxide is incorrect, as Bunsen has shown(Annalem, 192, 317).Kessler’s method (Pogg. Anw,., 95,204 (1855), and 113,134 (1861))is to oxidise antimonious chloride in concentrated hydrochloric acidb-y a volumet,ric solution of potassium dichromate in excess, and toadd, after dilution of the solution, an excess of ferrous chloride andthen titrate back the excess of ferrobs chloride by chromium solution.I n 1855 he obtained the atomic weight, 123.7, and in 1861 one experi-ment with antimony, one with antimony trioxide, and a third withantimonious chloride, gave 122.29.The author criticises Kessler’s method, and infers that no relianceis to be placed on it.Products of Decomposition and Metamorphosis of UranylSulphide. By C. Z~MMERMANN (AnnuZen, 204-2’24) .-Uranium doesnot exhibit a great affinity f o r sulphur, since uranyl sulphide is con-verted into uranyl hydroxide by washing with warm water. It hasbeen noticed by Reme16 (Pogg. Ann., 124, 120) that when uranyl sul-phide is left in contact with ammonium sulphide the liquid becomes ofa brown to deep black colour. The author shows that this is due tothe solubility of the uranyl sulphide i n ammonium carbonate con-tained in the ammonium sulphide. When precipitated uranyl sulphideis left for a long time in contact with ammonium sulphide in the cold,two bodies are obtained, one red the other black. The author showsthat when the ammonium sulphide contains a considerable amount ofthiosulphate the red body is formed, but that in absence of thio-sulphate the black body is obtained. The thiosulphate is formed bythe action of atmospheric oxygen on the ammonium snlphide. Theformula assigned to uranium-black is Ur,O,< O>Ur + 2Ur,0,.Uranium-red seems to consist of an oxygen-compound which contains,besides sulphur, a second base (potassium, sodium, ammonium, 01’barium), and some uranium, probably as uranyl sulphide. AnhydrousF. L. T.80 ABSTRACTS OF CHEMICAL PAPERS.uranium-red, on the assumption that it is perfectly free from admixture,OK may be formulated thus : U6SK209 = 2Ur203 + UrzOz< SK.G. T, A