I NO RG Ll N I C C H E M IS TRY. 31 Inorganic Chemistry. The Development of Inorganic Chemistry in the last Forty Years. HANS LANDOLT (Ber. 1907 40 4627-4637).-A lecture delivered before the German Chemical Society. Activity of the Halogens Chlorine Bromine and Iodine in Relation to,Mercury. M. C. SCIIUYTEN (Shem. Zeit. 1907 31 1135. Compare Abstr. 1896 ii 524).-Even in the presence of an excess of the salt the halogen in mercuric chloride bromide or iodide may be replaced by either of the other halogens tha only exception being the action of iodine on mercuric bromide; in this case no bromine is displaced by the iodine. New Polymorphous Form of Iodine. W. A. I~URBATOFF (Zeitsch. anorg. Chem. 1907 56 230-232).-When solutions of iodine in various solvents are evaporated to crystsllisation the element separates either as rhombic plates or as dendritic aggregates.The separation of a definite crystalline form does not depend on the colour of the solution but on the temperature the common rhombic crystals being obtained at the higher temperatures. The trimsition temperature could not be determined by the usual thermal method as there was no break i n the cooling curve of iodine from 100-O' but mas established approximately by condensation on a glass surface kept at different temperatures ; below 46-47' the dendritic crystals mere obtained a t Action of Iodine on some Elements in Vacuum. MARCEL GUICHARD (Conapt. rend. 1907 145 807-808).-When a tube con- taining iron wire and iodine separated by an asbestos plug is evacuated sealed and heated in such a manner that the iron is kept a t 500' and the iodine a t 180° ferrous iodide is formed and sublimes in the form of deep red crystals.Uranium and nickel also form iodides under similar conditions the former as dark grey lamellie. In the case of very volatile iodides such as those of aluminium and silicon it is necessary to keep the three portions of the tube (bent so that the middle portion is horizontal and the others pointing slightly down- wards) a t three different temperatures the icdine a t one end at 180° the metal in the middle at 500" and t8he otiicr eud (to receive the W. H. Q. W. H. G. higher temperatures the rhombic modification. (3 s.32 AUSl'ItACTS OF CHEMICAL PAPEILS. iodide) a t 15". the iodine vapour and retards the action.Otherwise the iodide formed remains mixed with E. H. Amorphous Sulphur. V. The System Sulphur-Iodine. ALEXANDER SMITH and CHARLES I\I. CARSON (Zeitsch. physikal. Chena. 1907 61 200-208. Compare Abstr. 1906 ii 157; 1907 ii 20)- The proportion of 8 in the equilibrium mixture of X h and Sy a t a given temperature is raised by the addition of iodine. Thus at 150' and in the absence of iodine the percentage of S in the equilibrium mixture is 6.7 ; when one part of iodine is present per one hundred parts of sulphur the percentage of S is 12.26; when five parts of iodine are present per one hundred parts of sulphur the percentage of 8 is 17-86. The effect of the presence of two parts of iodine per one hundred parts of sulphur at different temperatures is shown in the following table Temperature ..............................150' 165' 220' 310' 448' ...... 6.7 14.0 29.4 32.6 34.1 with iodine ............ 13.8 24.1 47.0 54.2 62.7 insoluble sulphur A study of the freezing points of mixtures of sulphur and iodine shows t h a t neither compounds nor solid solutions are formed (compare Boulouch Abstr. 1903 ii 538). The authors reply to Hoffmann and Rothe (Abstr. 1907 ii 539). J. C. P. Electro-Syntheses. SIMA M. LOSANITSCH (Ber. 1907 40 4656-4666).-An investigation of the behaviour of various substances i n a gaseous form both when alone and mixed together under the influence of a silent electric discharge. The apparatus and method employed have been described (Abstr. 1897 i 179). Sulphur dioxide is converted into sulphur trioxide with liberation of sulphur; in the presence of hydrogen or hydrogen sulphide water is formed and sulphur liberated.Nitric oxide yields nitrogen and nitrogen peroxide; when mixed with hydrogen i t is reduced to nitrogen the latter interacting with the water formed yielding ammonium nitrite. A mixture of nitric oxide and hydrogen sulphide yields an aqueous solution of ammonium polysulphide. Carbon disulphide vapour is converted into a polymeride (CS2)l an almost black substance which decomposes into its components when strongly heated. A mixture of carbon disulphide and hydrogen or hydrogen sulphide yields a brown substance C H2S previously inaccurately described as a polymeride of carbon monosulphide (CS) (Zoc. cit.) and by Berthelot (Abstr. 1899 i 657) as a substance having the composition C2H2S,. Carbon disulphide yields with carbon mon- oxide a brown insoluble substance 3CS2,2C0 ; with ethylene a brownish-yellow insoluble substance 5CS2,2G,H ; with acetylene an almost black insoluble substance 3CS,12C2H2.Acetylene alone is converted into a viscid or solid mass containing (1) a fragrant viscid swbstance soluble in alcohol and ether which rapidly absorbs oxygen from the air forming a s u b s t a ~ ~ c e (3C,H,,C,H,,O) ;INORGANIC CHEMISTRY. 33 (2) an insoluble substccnce C4?H, yellowish-brown by reflected and yellowish-red by transmitted light having an empyreumatic odour ; i t absorbs oxygen forming the substance C,,H,608. These acetylene condensation products decompose a t 100' ; they also evolve an emana- tion which liberates iodine from potassium iodide and produces a reducing action through aluminium or gold leaf on a photographic plate ; the emanation is not a radioactive emanation the action pro- duced being probably due to absorbed oxygen present in a labile state.The same substances are produced together with polymerised acet- aldehyde when water is present with the acetylene. Equal volumes of acetylene and hydrogen or methane give a clear yellow product contain- ing a viscid liquid soluble in ether and an insoluble solid of the formulzle C,Hl9 and C,H respectively. Equal volumes of acetylene and ethyl- ene yield (1) a substance C,H,,C,H obtrtined as a red viscid fragrant liquid ; (2) a solid substance 3C,H2,C,H insoluble in all solvents and decomposing when heated.Equal volumes of acetylene and hydrogen sulphide yield (1) a yellow substance C,H,,S soluble in ether ; (2) a yellow substance Cl,Hi4S insoluble in ether. Acetylene with carbon monoxide yields a yellowish-brown insoluble substance which absorbs oxygen rapidly forming a substance 4C2H,,co,0,. Equal volumes of acetylene and sulphur dioxide condense with the formation of a dark brown insoluble substance O,II,OS. Ethylene is converted into a yellowish-red oily substance soluble in alcohol ; it absorbs oxygen forming a viscid substance (Gl,H,?O)2. Equal volumes of ethylene and methane yield a clear yellow viscid liquid which absorbs oxygen forming a viscid substance C,,H,,O,. Ethylene and hydrogen sulphide yield a substance (C2H4S)6 obtained as a viscid yellowish-red liquid It is probable that ethyl mercaptan is first formed since this compound yields the same substance under the influence of a silent discharge.Ethylene and carbon monoxide yield (1) a substance (2C,H,,CO) obtained as a yellowish-red viscid liquid soluble in ether &c. ; (2) an insoluble szcbstnnce (2C,H,,CO),. W. H. G. Atomic Weight of Tellurium. WILLY MARCKWALD (Bey. 1907 40 4730-4738).-hlthough both Staudenmaier and Bfyliixs had drawn the conclusion from a series of fractional crystallisations that telluric acid is a uniform substance these authors had not sufficiently considered the possibility of their having dealt with an isomorphous mixture which could be separated only by a systematic method of crystallisation. The author accordingly examined about 1500 grams of telluric acid from the latter standpoint and after several hundred crystallisations obtained twenty fractions of about equal weights ; between the first and last fractions there was not however t h e slightest difference.There is no doubt as to the uniformity of tellurium For the determination of the atomic weight of tellurium the author has modified the method of Staudenmaier namely the conversion of telluric acid into tellurium dioxide (Abstr. 1896 ii 97). Crude tellurium obtained from the Selmeczbanya mines was converted into dioxide and separated from copper lead silver Src. VOL. XCIV. ii. 334 ABSTRACTS OF CHEMICAL PAPERS. The dioxide was then oxidised to telluric acid by means of chromic acid and submitted t o a prolonged series of crystallisations from water.The telluric acid used in the doterminations was dried over phosphoric oxide in a vacuum. It is worthy of note that when telluric acid is exposed over phosphoric oxide for two months in a vacuum it remains practically constant in weight. The method is based on the action as represented by the equation H,TsO = TeO + 0 + 3H,O. The heating was conducted in a platinum vessel by means of a n electrical oven the temperature at the bottom of the vessel being measured through a Le Chatelier clement by means of a millivolt meter. The temperature a t the top of tho platinum vessel was maintained lower than that at the bottom in order to guard against volatilimtion of the dioxide The temperature was gradually raised in the course of two to three hours from 100" to 160" and then gradually in the course of three to four hours more to 650° which temperature was maintained for another hour.The residue was always white and quite free from the trioxide. The weight did not alter on further heating for one hour. The various precautions taken are indicated. As R mean of six experiments the value 126*85( +,0*02) was obtained on the basis of 0 = 16 and H = 1.005. This valuo is lower tlian that of iodine (I= 136.97). LEON EEUNEL and PAUL Wooc (Conxpt. s*end. 1907 145 922-924. Compare Woltereck Abstr. 1904 ii 115).-Attempts to effect the synthesis of ammonia by passing a mixture of pure hydrogen (3 vols.) and nitrogen (1 vol.) over a mixture of thorium and cerium oxides pulladinised pumice quicklime soda lime calcium chloride cltlciurn molybdate anhydrous barium or strontium oxide manganese dioxide aluminium phosphate or magnesium phosphate a t 15' to 350° were unsuccessful. The gaseous mixture reacts with nickel sesqiii- oxide at 180" t o 200' to form water and a trace of ammonia but the reaction is fugitive.If however a mixture of air and excess of hydrogen is passed over heated nickel sesquioxide the latter is alternately reduced and reoxidised and the energy liberated by the reactions causes the formation of small quantities of ammonia ; the reaction is continuous provided that the nickel oxide is not allowed to become incandescent owing to the heat of the reaction. Thisis effected by suspending the tube containing the nickel oxide in a bath of petroleum b. p. 200-240".Boiling Point of Liquid Ammonia. EDWARD C. FIXANULIN (Ann. Ply& 1907 [iv] 24 367-369).-R critical review of the determinations made by de Porcrand (AWL Chiin. Phys. 1903 [vii] 28 537) Gibbs (Abstr. 1905 ii 670) Yerman and Davies (Abstr. 1906 ii 743) and Brill (Abstr. 1906 ii 847) leads the author to the conclusion that Gibbs' value ( - 33.46') is probably correct to within 0-1". J. C. 1'. Formation of Nitric Oxide in High Tension Arcs. FRITZ HABEE and ADOLF KOENIG (Zeitsch. EZekti-ocitem. 1907 13 725-743). -An alt ernsting current ~7as passed through mixtures of oxygen and A . McK. Catalytic Synthesis of Ammonia from its Elements. M. A. W.INORGANIC CHEMISTRY. 35 nitrogen under pressures varying from about 40 mm. to 200 mm. Electrodes of platinum oxidised iron and Nernst glowers were used The gas mas passed through water-cooled tubes of hard glass or quartz of 5 to 7 mm.bore along the axis of which the discharge passed. A t the lower pressures the whole mass of gas in the tube glowed with a pink light with the exception of a thin film next to the wall of the tube; at the higher pressures the glowing coIumn of gas tended to contract further from the walls. The best results are obtained when the gas does not flow over the electrodes and when the latter are very hot. Electrodes of iron or Nernst filaments give the best results. The percentage of nitric oxide in the gas increases with the current up to a maximum which is not affected by further increase of current. The greatest concentration of nitric oxide is obtained with gas a t 100 mm.pressure The maximum percentage of nitric oxide obtained with each mixture was (at 100 mm.) Percentage of oxygen in mixture 20.9 48.9 44.4 75.0 81.7 Pencentage of nitric oxide ...... 9.8 14.4 14.3 12.77 12.1 Assuming that the formation of nitric oxide is a purely thermal phenomenon and using Nernst's determinations of the equilibrium concentrations these results indicate that the temperature oE the gas in the path of the arc was between 4300' and 5000O absolute. From the thickness of the non-luminous film of gas between the luminous gas and the walls of the vessel the authors calculate that the temperature of the hot gas could not have exceeded 3000'; further the rate of cooling of the gas was smaII so that nitric oxide formed by very high temperature alone would probably have been decomposed.The authors incline rather to believe that the temperature was low and the high percentage of nitric oxide found was produced by collisions of electrons ; owing to the low temperature the nitric oxide thus formed escaped decomposition which is not the case in the hotter flames produced in gases under atmospheric pressure. Non-existence of a Common Solvent for White and Red Phosphorus. ALBERT COLSON (Compt. rend. 1907 145 11 67-1 168).-Schroetter stated that red phosphorus is soluble in oil of turpentine; this statement has been contested and the author shows that pure red phosphorus does not dissolve in oil of turpentine or in the polymerised products obtained by the action of heat on the solvent even when heated a t 270' in n sealed tube provided that air is excluded.Further red phosphorus is not altered when heated a t 275-285O in a sealed tube containing a solution of phosphorus in turpentine. There exists moreover no common solvent for white and red phos phorus for Lemoine has shown t h a t the two varieties have the same vapour density corresponding with the molecule P ; according to van't Hoff's hypothesis therefore the two varieties mould become identical in a common solvent. Action of Hydrogen Phosphide on Mercuric Chloride or Bromide PHg,CI and P,Hg,Br,. PAUL LEnrouLT (Compt. rend. T. E M. A. w. 1907 145 1175-1177) -H. R O F ~ (YOYY. A I Z ~ ~ . 1837 40 75 87h 3-. 236 ABSTRACTS OF CHEMICAL PAPERS obtained a yellow amorphous compound PHg,CI 1 &€120 by the action of hydrogen phosphide on mercuric chloride and an analogous compound of a bromn colour when mercuric bromide replaced the mercuric chloride.The author finds however that mercuric chloride gives with hydrogen phosphide t,he anhydrous compound PHg,C! whereas mercuric bromide forms the compound P2Hg5Br4 according t o the equations f . PII + SHgCl = 3HC1+ PHg3C13 ; 11. 2PH3 + 5HgBr2 = 6HBr + P,Hg,Br4. The presence of excess of hydrogen phosphide even locally as when the gas is bubbled through a solution of the mercuric halide leads to a reduction of the salt with liberation of mercury and the best results are obtained when an aqueous solution of hydrogen phosphide is carefully poured on to the surface of a concentrated solution of the mercuric halide and the cowesponding potassium salt and the mixture suddenly shaken; the precipitate thus obtained is uniform in colour and of constant composition.M. A. W. Action of Arsine on Solutions of some Metallic Salts. HANS RECKLEBEN GEORG LOCKEMANN and ALFRED ECKARDT (Zeitsch. anal. Chenz. 1907 46 671-709).-Arsine is absorbed rapidly and completely by silver nitrate solution but slowly by mercury copper lead tin and iron salt solutions ; the latter solutions are useless for the quantitative absorption of arsine. The reaction with silver nitrate solution does not take place exactly according to Lassaigne's equation ASH + GAgNO,+ 3H,O = H3As03 + 6Ag + 6HNO but in part ac- cording to the equation ASH + 3AgN0 = Ag,As + 3HN0,. The silver arsenide is not very stable in the presence of the nitric acid and further action takes place Ag3As + 3AgN0 + 3H20 = H,AsO + 6Ag + SHNO,.I n dilute ammoniacal silver nitrate solution three reactions take place consecutively or simultaneously (a) As f3 t- S(AgNK,)NO = Ag3As + 3NH4N03 ( b ) Ag,As + 3(AgNH,)NO + EH,OH + H,O = NH,AsO + 6Ag + SNHI,N03 and (c) NH,AsO + 3(AgNH3)N0 + 8NH40H = (NH,),AsO + 2Ag + 2NH4N0,. Metallic arsenic when warmed with ammoniacal silver nitrate solution is oxidiaed thus As + 5(AgNH,)NO + 3NH40H + II,O = (NH,),AsO + 5Ag + 5NH,NO,. I n ammoniacal solution and in the presence of air srsine is readily oxidised. T t is also mentioned in the paper that finely-divided silver is not attacked by 6.3% nitric acid within four hours at the ordinary temperaturo and N/10 nitric acid is only very slowly reduced a t the ordinary temperature by arsenious acid.w. P. x. Compounds of Arsenic Sulphates with Potassium Calcium and Lead Sulphates. HUGO KUHL (Arch. Pltccrm. 1907 245 377-379).-The crystalline compounds 2K20 As20,,4SO CaO As,0,,3S03 PbO,As,O,,%SO were obtained by dissolving arsenious oxide and potassium calcium or lead sulphate in concentrated sulphuric acid and driving off sulphuric acid by heating ; with strontium and barium sulpbrttes similar compounds could not be obtained. C. F. B.INORGANIC CHEMISTRY 37 Identity of Graphite and ‘ I Temper ” Graphitic Carbon in Cast Irons. GEORGES CHARPY (Compt. rend. 1907,145,1173-1 174). -It was stated by Forquignon and Ledebur that the carbon (graphite) contained in cast iron that has been cooled gradually differs from the carbon (“ temper ” carbon) that separates in rapidly-cooled cast iron in that the latter is volatile and the former non-volatile when a stream of hydrogen or nitrogen is passed over the red-hot cast iron.Wust and Geiger find however (Abstr. 1006 ii SS) that pure hydrogen o r nitrogen is without action on “temper ” carbon or on graphite. The author has therefore examined the chemical behaviour of the free carbon obtained from two portions of cast iron one of which had been gradually the other rapidly cooled and the results show that graphite and the so-called ‘ I temper ” carbon are identical both with regard to the velocity with which they are oxidised to graphiticacid and to the rate at which they are completely eliminated in the gaseous form when a stream of pure hydrogen is passed over the cast iron a t 1000”.151. A. W. Solubility of Potassium Iodide in Water and of Water in Potassium Iodide at Low Temperatures. ROBERT KREMANN and F. KERSCIIBAUM (Zeitsch. nnorg. Chem. 1907 56 218-222).- Meusser (Abstr. 1905 ii 317) has recently determined the complete solubility curves for water with potassium chloride and bromide respectively and found no evidence of the formation of hydrates but in the case of water and potassium iodide there was a break in the observations from 42.7% to 53.5% of the salt the eutectic point being determined by extrapolation. The authors have now determined the complete curve including observations within the limits indicaked and find no evidence of chemical combination ; the eutectic point lies at - 23*1° the mixture containing 52.2% by weight of potassium iodide.G. S. Waterglass. VI. JOHN 81. ORDWAY (Amer. J. Sci. 1907 [iv] 24 473-478).-1n a.n earlier paper (Amer. J. Sci. 1865 40 190) it has been shown that when alcohol is added to solutions of sodium and potassium silicates basic silicates are precipitated. On mixing strong solutions of sodium silicate and lithium chloride precipitates are produced containing varying quantities of lithium and sodium silicates. By re-dissolving these precipitates and adding lithium chloride to the solutions the proportion of lithium in the product can be gradually increased but pure lithium silicate cannot be obtained in this manner. When however freshly-prepared silica is digested with lithium hydroxide solution a t the ordinary temperature a solution of lithium silicate Li,SiO is produced.On heating this solution a precipitate is produced which re-dissolves on cooling. Soluble rubidium silicates can be obtained of composition varying from Rb,O,SiOz to 2Rbz0,9Si0,. I n the case of the lithium silicates the range is not so great. When ammonia is added to a solution of a silicate containing a large proportion of silicic acid a precipitate is usually produced. E. G.38 A13S'rRACTS OF CHEMICAL PAPERS. Ammonium Amalgam. G . MCPIJAIJ SMITH (Bey. 1907 40 4893. Compare Abstr. 1907 ii 615).-A reply to the criticisms of Tra,vers (Abstr. 1907 ii 865). It is argued that ammonium amslgam is completely analogous t o sodium and potassium amalgams which a,re regarded as solutions of NaHg or IiEig in mercury.E. F. A. Amalgams. The Hydrargyrides of the Alkali and Alkaline Earth Metals. G. McPIrArL SMITII (Anzer. Chena. J. 1907 38 671-683).-F~irther evidence in favour of the views expressed regarding the nature of amalgams in previous papers (Abstr. 1905 ii 164 450; 1906 ii 6 7 3 ; 1907 ii 462 463 615) is brought forward and it is shown that unlike the amalgams of the alkaline earth metals those of the common alkali metals do not increase in stability with increase in the atomic weight of the alkali metal used (the order of stability being Li K Na). The amalgams used were liquid and were prepared by electrolysing a solution of the chloride of the appropriate metal in presence of mercury. The relative stabilities of the amalgams were tested (I) by the rate of decomposition on exposure to air (2) by the capacity for decomposing water (3) by the rapidity with which they were decom- posed by ammonia solution and (4) by the composition of the equilibrium amalgam obtained when the amalgam under investigation was acted on by an aqueous solution of sodium chloride and the chloride of its own metal.By all four methods of investigation i t was found that the relative stabilities of the amalgams examined increased in the following order K Rb Cs. The mercury compound of lithium i s the least stable i n the alkali metal group ~vhilst that pf sodium is the most stable. Sodium amalgam also behaves in an exceptional manner ; thus on exposure t o air no sodium peroxide is formed whereas the corresponding peroxide is formed in each case when an amalgam of potassium rubidium or cesium is exposed to air.The metals of the pairs Cs,Na and Rb,Na are like most of the other metals of the alkali and alkaline earth groups reversibly displaceable in aqueous solution in presence of mercury. T. A. H. Artificial Reproduction of Barytes Celestine and Anglesite and Isomorphous Mixtures of these Substances. PAUL GAURERT (Cowipt. rend. 1907 145 877-879).-Behren's micro- chemical reaction for the detection of certain metals (Abstr. 1886 917; 1891 766) gives confused results in the cases of barium and strontium. By a modification of the method the author has obtained crystals of the sulphates of barium strontium lead or calcium similar to those of the natural minerals and of such dimensions that the optic axial angles which are characteristic for each substance could be measured.The method consists in evaporating to dryness 100 C.C. of a saturated sulphuric acid solution of the metallic sulphate a t the boiling point or a t n slightly lower temperature and in the case of barium or strontium sulphate the crystals thus obtained were sufficiently large to admit of gouiometrical measurement. MixedINORGANTC CIIEMISTRY. 39 crystals of barium and strontium sulphate or of either of theso sulphates with that of lead were prepared similarly. Calcium sulphate not being isomorphous with the preceding sulphates does not form with them mixed crystals although the crystal forms of these sulphates are modified by the presence of calcium sulphate in the solution.M. A. W. Colloidal and Gelatinous Calcium and Magnesium Com- pounds. A kad. miss. Bedin 1907 620-822. Compare Neuberg and Neimann Abstr. 1906 ii 753).-Calcium sulphate calcium phosphate and calcium oxalate are precipitated in a gelatinous form when sulphuric acid phosphoric acid and oxalic acid respectively are added t o a methyl- alcoholic solution of calcium oxide. The clear viscous colloidd solution of calcium carbonate obtained by passing a stream of carbon dioxide into a methyl-alcoholic solution of calcium oxide may be evaporated without undergoing decomposition to a thick liquid which sets t o a jelly. If a current of carbon d.ioxide is passed into a suspension of calcium oxide in methyl alcohol the liquid after about five to six hours is transformed into a solid jelly which dissolves slowly in methyl alcohol.The colloidal calcium carbonate is miscible with several organic solvents for example benzene chloroform aud ether. Analogous colloidal magnesium compounds may be similarly prepared from a methyl-alcoholic solution of magnesia. Analysis of the isolated gelatinms salts (sulphate phosphate and oxalate) ofAcalcinm and magnefiium showed that they did not contain combined methyl alcohol. CARL NEUBERG [and in part 13. REJvALD] (Sitxungs6ei.. W. H. G. The Autoreduction of some Metallic Oxides in the Vacuum of the Cathode Light and on the Volatility of the Corrs- sponding Sulphides. . F. DAMM and FRrEDRrCH KRAFFT (BOY. 1907 40 4775-4778).-The behaviour of oxides when heated is sometimes quite otherwise in a vacuum than under ordinary pressure.Thus cadmium oxide a t 1000' decomposes into oxygen and cadmium whereas in air it is completely stable. The experiments were carried out in quartz tubes heated in a Heraeus electric oven At 750° lead oxide gives a metallic mirror and bismuth oxide decomposes slowly at 650". Antimony oxide volatile a t about TOOo is not completely de- composed a t 1050O. It is conjectured that tho autoreduction of the common metals may proceed in stages like that of antimony oon- sisting (1) of simultaneous oxidation and reduction and (2) conversion of a higher oxide into a lower. The sulphides are more volatile than the corresponding oxides cinnabar sublimes at 400" and cadmium sulphide very quickly at 770-780' whereas lead sulphide volatilises at 600' ; the sulphides of bismuth antimony and arsenic sublime a t 740° 530° and 230" in the vacuum of the cathode light T t is possible to separate by distiIIation mixtures of arsenic and 4Sb,06 = 3Sb,08 + 4Sb,,40 ABSTRACTS OF CHEillICAL PAPERS.antimony sulphides or mercury and lead sulphidcs under those con- ditions as well as sulphur selenium and tellurium. The suphides of copper tin and silver are decomposed; the first two lose half their sulphur content and the last is converted into metal and sulphur. W. R. Solubility of Certain Lead Compounds in Water. MAX PLEISSNER (Clzem. Zenir. 1907 ii 1055-1056 ; from Arb. Kais. G'esundh.-Amt F1907 26 384-443).-This investigation was under- taken in connexion with the risk of contamination of water supplies by lead.Lead oxide and hydroxide and the normal and basic carbonates sulphates and chlorides have been studied. When hot solutions of lead salts are treated with alkali hydroxides lead oxide PbO is precipitated whilst from cold solutions hydrated oxides separate. The lead oxide forms greyish-yellow scales with a metallic lustre and gives a greenish-yellow powder. The same oxide is produced by the action of water containing much dissolved oxygen on lead whilst by thb action of water containing but little oxygen the hydrates are formed. A hydrate Pb,02(OH)2 has been identified but higher hydrates probably exist the solubility increasing with the degree of hydration. The oxide and its hydrates appear t o dissociate thus Pb(OH) Pb(0H)' +OH' the degree of dissociation in a solution saturated at 18" amounting to about 25%.The solubility of lead sulphate and chloride in water is diminished by the presence of sulphuric or hydrochloric acid whilst that of the carbonate is increased by the presence of carbonic acid. The following basic lead salts have been obtained and their individuality established by application of the phase rule. &-Basic lead carbonate Pb0,2PbC0,,H20 ; $-basic lead sulphate and chloride PbO,PbSO and P b0,PbC12,H,0 ; $-basic lead sulphate and chloride 3Pb0,PbS0,,H20 and SPbO,PbCI,,H,O. These salts are less soluble in water than the corresponding normal salts. The solubility (millimols. P b per litre) and the specific conductivity of the saturated solutions after deducting the value for the conductivity of the water are as follows a t 1 8 O PbO Pb,O,(OH) PbCO Pb,(CO,),( OH) PbSO Solubility ......0.31 0.45 0'0002 < 0.0002 0-126 Specific conductivity 19.5 27'3 - - 33 *9 Solubility ...... 0.050 0'106 33'6 0.38 0'10 conductivity 8.8 9.3 4512 68 19 Pb,O( SO4) Pb,O,(OH),(SO,) PbCI PbCI,(OH) Pb4C1,02(OH)2 Specific E. G. Aluminium Titanide. WILHELM MANCHOT and P. RICHTER (Artnalem 1907 357 140-144. Compare Manchot and Fischer this vol. ii 46).-Attempts to prepare double aluminium titanides with the object of investigating the capacity of titanium t o form chains have resulted only in the formation of aluminium titanide,INORGANIC CHEMISTllY. JL A1,Ti (Wiihler Annwleit 1860 113 248; 115 10s). The titanide is best prepared by fusing 24 parts of potassium titanofluoride with 45.5 parts of aluminium and gently heating the regulus with dilute sodium hydroxide.The product forms silver-white leaflets D 5.5 is brittle hardly scratches steel glass or fluorspar contains tervdent titanium since i t forms a blue solution in hot dilute sulphuric acid evolving twelve atomic proportions of hydrogen and evolves thirteen atomic proportions of hydrogen when dissolved in aqueous potassium hydroxide. The titanide may have the double formula AI,Ti*TiRl but this cannot be decided. HENRY REYNOLDS (Clmn. News 1907 96 260).-The author considers that manganese which shows only the slightest resemblance t o the halogens should be placed in the eighth group with iron nickel and cobalt. It resembles these metals in physical properties and in the formation of similarly con- stituted and isomorphous simple and double salts M”SO 7H,O and R,S0,1M”S0,,6H,0 alums metallo- and metalli-cyanides. Moreover potassium manganate and permanganete find a parallel in potassium osmite potassium ruthenate and per-ruthenate.I n many cases a regular gradation of properties can be traced from manganese to cobalt and a corresponding gradation from ruthenium G. Y. Manganese and the Periodic Law. to palladium and from osmium to platinum. c. s. [Determination of the] Melting Points of the Iron Group Elements by a New Radiation Method. GEORQE K. BURGESS (Bureau Stand. ITashington 1907 3 (3) 345-355).-The m. p. of minute quantities of difficulty fusible substances have been determined by a method based on the measurement of the intensity of a particular monochromatic radiation from platinum (compare Burec6u Stand.IVasliington 1905 1 189; 1907 3 1). Within a blackened brass cylinder filled with hydrogen is a strip of platinum 60 x 4 x 0.02 mm. which is heated electrically to any desired temperature. On this strip is placed about 0.001 mg. of a metal or its oxide which is matched through a mica window in the cylinder by one observer through a microscope whilst simultaneously a second observer reads the temperature of the platinum strip by means of a Hol born-Kurlbaum optical pyrometer. The indications of the pyrometer are subject to two corrections one for the reflection and absorption of the mica the other for the selective emission of the platinum for the light used red light A = 0 .6 6 ~ . The following metals in a state of maximum obtainable purity have m. p. iron 1505’ ; chromium 1489” ; cobalt 1464” ; nickel 1435’; manganese 1201’. The m. p. of cobalt and nickel are correct t o within 5O those of the other three to within less than 10”. C. €3. Constitution of Roussin’s Salts. LIVIO CAMBI (Atti R. Accad. Lincsi 1907 [v] 16 ii 658-660. Compere Abstr. 1907,ii 960).-The simultaneous action on potassium heptanitrosulphide of mercuric cyanide in amount equivalent t o the sulphur present and potassium cyanide in the quantity necessary for the complete formation of ferro- cyanide yields mercuric sulphide alkali nitrite and ferrocyanide,42 ABSTRACTS OF CHEMICAL PAPERS. as would be expected from the constitution attributed to Roussin's salts by the author (Zoc.cit.). The phenomena observed in tho precipitation of Roussin's salts by silver nitrate have led the author to assume that hyponitrous acid rednces ferric to ferrous salts. Preliminary experiments indicate that this reduction does not take place. The investigation is being continued. T. €1. P. 1 2Dichlorotetramminecobalt Salts ; Ammonia-violeo-Salts. ALFRED WERNER (Be?.. 1907,40 4S17-4825).-Nurnerous examples of stereoisomerism of cobalt salts containing organic amino-residues have been described (JGrgensen Abstr. 1898 ii 226 ; Werner and others 1901 i 510 512 ; 1907 i lSS 290 482 590). Hitherto the simp1 es t tetramm ineco b d t salts for example the dichloro t e t rammine- cobalt salts [Cl,Co(NH,),] X have been obtained in one form only the so-called pmseo-salts ; a stereoisomeric series of salts the uioteo-salts has now been prepared by the action of hydrochloric acid on octamminedioldicobalt salts (Abstr. 1907 ii 965).When concen- trated hydrochloric acid is used a t low temperatures the reaction proceeds according t o the equation [(NHJ,*CO(OH)~CO(NH,),IX + 2HC1= [(NH,),Co(OH),]Cl + [CI,Co(NH,),]Cl a mixture of 1 2-di- aquotetramminocobalt chloride and 1 2-dichlorotetrarnminecobalt chloride being formed; the former is readily soluble in water whereas the latter forms sparingly soluble intensely blue crystals. The violeo- salt contains small amounts of the isomeric praseo-salt from which it may be freed by conversion into the practically insoluble dithionate.The chloride can then be regenerated by rubbing the thionate with ammonium chloride. The following salts have been analysed chloi-ide YCl,iH,O [Y = Cl,Co(NH,),] bromide iodide and nitrate all anhydrous dithionate YzSz06 and sulphccts Y,SO,. The chloride becomes anhydrous a t 60° and all are characterised by an intense violet-blue colour. The aqueous solutions are somewhat less stable that those of the corre- sponding praseo-salts ; even a t 0" the solutions assume a reddish-violet colour owing t o the formation of the chloroaquotetramminecobalt salts. When suspended in concentrated hydrochloric acid the violeo- chloride is converted into the praseo-chloride. The violeo-salt is regarded as a &-compound with the 1 2-constitution. Complex Metal Ammonia Derivatives.VI. Octammine-p- *NH amino-01-dicobalt Salts (NH,),Co "Co(N H,),]X,. ALFRED WERNER (Bey. 1907,40,4605-4615. Compare Abstr. 1907 i 1012 ; ii 965).-When an ammoniacal cobalt nitrate solution is oxidised by a slow current of air and a n excess of sulphuric acid then added Vort- mann's insoluble sulphate separates. The author has shown previously (Abstr. 2898 ii 223) that this product is not homogeneous since two compounds a red and a green can be separated from it both of which belong t o the class of complex cobalt salts. I n the present paper the constitution of the red salt is elucidated. Salts of the types in question are decomposed by a mixture of hydro- chloric and sulphuric acids into pentammino and tetrammine salts from which it folIows that 4 molecules of ammonia are attached J.J. S. [ OH2INORGANIC CHEMISTRY. 43 t o each of the two cobalt atoms. The constitution of the red series of salts is discussed. The formula [(NH,),Co:NH*CooH2 X is not favoured because an imino-group could not be detected; no salt formation took place either with acetic acid or dilute minernl acids; no acetylation took place by the action of acetic anhydride and no nitroso-compound resulted from the action of nitrous acid. Further the compounds in question were neutral in aqueous solution and not acid as would be the case if the formulation just stated were correct. (NH3)4 1 On the other hand the formulation [(NH,),CO.~~,,,CO(NH,) *OH* 1 X accords with the experimental resul&. Since the gobalt atom; are saturated with respect to co-ordination numbers the fact that all the acid groups are inorganic in character is explained.The complex radicle contains no water and the salts do not consequently give an acid reaction when dissolved in water. The amino-group behaves like ammonia in metal ammonia compounds being indifferent towards chemical reagents in the cold. The hydroxyl group also does not have the property of adding on hydrogen ions. For those amino- groups which bind two metal atoms together the author proposes the nomenclature ‘‘ p-amino.” Octammine-p-sulphatoamino-dicobalt nitrate obtained from Vort- mann’s sulphate was shaken with concentrated hydrochloric acid until the reddish-violet tint of the salt had changed to brown. The product was washed with alcohol and hydrochloric acid added t o its aqueous solution when the brownish-violet chloro-chloride separates. The f 011 owing oct ammine-p-amino -01-d icobalt salts were prepared.The &Zooride YCl,,BH,O (Y = (NH,),Co. *NR og C O ( N H ~ ) ~ ~ ) forms [ glistening brownish-red prisms ; by the action of hydrochloric acid it is converted into pentammine and tetrammine cobalt salts. The bromide YBr,,4H20 forms glistening red scales. The nitrute S(NO,) forms dark red needles and scales. The mlpiiate Y(S04),,2R20 forms brownish-red silvery scales. The dithionate Y(S,0,),,2H20 forms silvery raspberry-coloured leaflets. The tlLiocyanate Y (CNS) forms red crystals A. McK. Complex Metal Ammonia Compounds. VII. Hexammine- trioldicobalt Salts. ALFRED WERNER [and in part EMIL RINDSCHEDLER and ADOLF GRSN] (Ber.1907 40 4834-4844. Com- pare Abstr. 1898 ii 223; 1899 ii 658; 1907 i 482 1012; ii 965 and preceding abstract).-~examm~netr~o~d~cobccZt saEts of the r *OH- 7 type (NH,),Co*OH*Co(NH,) X may be prepared by the follow- 1 OH- ing methods. 1. The chloride is obtained by the action of sodium hydroxide solution on dichloroaquotriarnmine cobalt chloride. 2. The sulphate by the action of sodium hydroxide solution on chlorodiaquo- triammine cobalt sulphate. 3. The bromide by warming chlorobromo- aquotriammine cobalt bromide with a little water a t 60’. 4. The44 SBSX'RACTS OF CHEMICAL PAPERS. sulphate by leaving chlorodiaquotriammine cobalt sulphate with potass- ium bromide solution for several days at the ordinary temperature. When decomposed with halogen hydracids the salts give quantita- tive yields of triamminecobal t salts indicating that three ammonia residues are attached to each cobalt atom.The acid residues are readily ionised as each salt can be transformed into the others by double decomposition. Their aqueous solutions are quite neutral t o litmus and hence the salts cannot be aquo-salts. They are isomeric with the black dodecamminehexoltetracobalt salts [Co(OH)6tco(RH3)41 1x6. The following salts have been prepared. C l d o d e [ ~ N H ~ ~ ~ C o ( 0 H ~ 3 c o ( N ~ 3 ~ ~ ~ c 1 ~ ~ H ~ 0 ~ best obtained by rubbing the sulphate with ammonium chloride acd water brownish-red needles and prisms from dilute alcohol decom- pose a t 100". Byonaide anhydrous red prismatic crystals soluble in about thirteen times its weight of water a t the ordinary temperature.Nitrate red prisms containing 2H20 becomes anhydrous when kept over sulphuric acid. Sulphate crystallises with 6H20 pale red prisms loses SH,O over phosphoric oxide and the last molecule a t 77". Bithionate [ (NH,),Co( OH)3Co(NH3j,] 2( S,O,!,,H,O obtained by the &ion of a saturated solution of sodium dithionate on a solution of the bromide at Oo dark red needles and prisms. Thiocpnate anhydrous dark red needles soluble in 3 parts of water at the ordinary temperature. J. J. S. Pentamminecobalt Salts with Several Nuclei. JULIUS SAND and G . B~KMAN (Be?. 1907 40 4497-4504).-Sand and Genssler have described two a black and a red series of pentamminenitroso- cobalt salts (Abstr. 1903 ii 549; 1904 ii 39).These salts especi- ally those of the black series are characterised by their great reac- tivity The present paper contains an account of the compounds obtained by the action of iodine on the black pentamminenitrosocobalt chloride and nitrate in alcoholic solution and of the complex salts derived from these. When boiled with alcoholic iodine the black chloride Co2(N2Oz)(NH,)lOC14~ yields the chloride [Co3C11,(NH3)i,(H,0)2]C16 which after being boiled with hydrochloric acid separates from its ice-cold aqueous solution on addition of hydrochloric acid potassium chloride or alcohol in green crystals ; with v ? ~ ~ has the molecular conductivity p = 731 and gradu- ally decomposes in aqueous solution at 25O the conductivity increasing with the time. The action of potassium iodide on the chloride in ice- cold aqueous solution leads to the formation of the green crystalline iodide [CO,C~I~(NH~),,(H~O)~]I~ whilst the action of nitric acid leads to the formation of the nitrate [Co3C1I,(NH,),,(H20),](NO3),.The solubility of these three salts decreases from the chloride through the nitrate to the iodide which is only very sparingly soluble. When treated with cold nlcoholic iodine the black nitrate Co2(N,O,)(NH,)1,( KO,),,PHYSIOLOGICAL CHEMISTRY. 45 forms a peen substaszce which when boiled with 20% nitric acid yields the green complex scclt [Co,I,(NO,)(NII,),,(H,O),I(NO,),. Equilibrium of the System Nickel-Bismuth. A. PORTEVIN (Compt. rend. 1907 145 1168-1 17O).-The freezing-point curve of mixtures of nickel and bismuth consists of three branches the points of intersection are given by the reactions Ni + Bi 2 NiBi(?) at 654" and Ni+NiBi(?) I n neither case however is the reaction complete.14. A. ?V. ALBERT COLSON (Ann. Cliinz. Phys.. 1907 [viii] 12 433-467).-A rksumt' of work already published (compare Abstr. 1905 ii 94 460 592 639; 1906 ii 74 233 ; 1907 ii 177 267 356 474 780). New Compound of Uranium the Tetraiodide. MARCEL GUICHARD (Compt. Tend. 1907 145 921-922).-When iodine vapour is passed over uranium at 500' in sealed vacuum tubes (this vol. ii 31) the tetraiodide UI is formed as a crystalline sublimate con- sisting of fine black needles in. p. about 500° DI5 5.6. It is reduced when heated in hydrogen decomposed by chlorine yielding the chloride and iodine trichloride readily oxidised by oxygen or air to form the oxide U,O and dissolves in water to form a green acid solution giving the charnc teris tic reactions of uranium salts.G. Y. Nisi a t 462". The Green Chromium Sulphates. AT. A. W. M. A. W. Colloidal Hydroxides of Thorium Zirconium and Uranium. B ~ L A SZILARD (J. China. Phys 1907 56 488-494).-The author considers that there are two distinct classes of colloids those contain- ing a little electrolyte which are very stable and those quite free from electrolytes which are also fairly stable. Colloidal thorium hydroxide free from electrolytes has been prepared by decomposing thorium nitrate with excess of dilute ammonia and mashing the residue until free from ammonia which takes three to six days The product is a fine milky suspension in which the particles cannot be separated either by filtration or decanta- tion ; it is precipitated by electrolytes and by the carbon dioxide of the air.Colloidal zirconium hydroxide was prepared by tho same method and shows similar properties. Colloidal thorium hydroxide containing a little electrolyte has been obtained by adding precipitated thorium hydroxide to thorium tetrachloride so long as the former is dissolved. This solution is much more stable than that free from electrolytes is not affected by light or by boiling and is not precipitated by small quantities of electrolytes or by weak acids (compare Miiller Abstr. 1906 ii 762). Colloidal uranyl hydroxide containing a little electrolyte has been prepared by the gradual addition of uranyl hydroxide to a dilute solu- tion of uranyl nitrate as long as the former is dissolved ; the solution thus obtained is orange-yellow in colour and very stable.The uranyl hydroxide for this purpose was obtained by exposing to light a mixture of uranyl acetate and ether and thoroughly washing the resulting precipitate. G. S.46 ABSTRACTS OF CHEMICAL PAPERS. Spitting of the Acid Vanadates of Univalent Metals WILHELM PRANDTL and HANS MUESCIIHAUSER (Zeitsch. ccrtorg. Clrmm. 1907 56 173-2OS).-In a previous paper (Abstr. 1905 ii 170) it was shown that when certain alkali acid vanadates xMz0,yV205 are allowed to cool from high temperatures a vigorous evolution of oxygen takes place on solidification the mass spitting as solidifying silver does. On again heating in air oxygen is absorbed and the acid vanadates are regenerated.This phenomenon is due t o the reversible change of the acid vanadates xM,O,yV,O to vanadylvanadates xRT20,(y-x) V205,xV204) on solidification. Only the oxides of the univalent elements of the first group in the periodic tlable give acid vanadstes which spit on solidification. The respective oxides were mixed with vanadium pentoxide in varying proportions and the composition of the mixtures which gave the highesl; proportion of oxygen determined ; the results varied with the nature of the oxide and were as follows Ns,0,6V20 ; Ag,O,6V2O5 ; K,0,5V20 ; Rb20,5V,O ; Cs,0,5V2O ; Li,0,2V2C!,. From mixtures of the same alkali oxido wlth varying proportions of vanadium pentoxide the same vanadylvanadate is obtained the composition of the four most important being as follows Na,0,5V2O,,V,O ; Ag20,5V,0,,V20 ; 2 h ' 2 0 9 ~ 2 0 5 ~ 2 ( ~ 1 ; 4Li,o,7~,?,,V20,.The com- position of these compounds was determined approximately by treating the mixtures containing them with boiling dilute ammonia which dissolves unaltered V,O,. The amounts of oxygen evolved from solidifying sodium and silver vanadates correspond fairly closely with those calculated from the composition of the vanadylvanadates as determined above so that the reaction is nearly complete but in the case of the potassium rubidium caesium and lithium vanadates the amounts of evolved oxygen are much less than the calculated values showing that the decomposition on cooling is incomplete. When mixtures of vansdium pentoxide and two alkali oxides are employed the amount of oxygen evolved is approximately the mean of that obtained with the separate oxides.The addition of boron trioxide in moderate amount to the mixtures does not affect the evolution of oxygen but when considerable amounts of phosphates are present the fused mass solidifies without the liberation of oxygen probably owing to the previous formation of vanadyl phosphate. G. 8. Silicon Chains. WILHELM MANCHOT and H. FISCIIER (Anirixden 1907 357 129-139. Compare Manchot and Kieser Abstr. 1905 ii 165 ; 1906 ii 83 ; Vigouroux Abstr. 1906 ii 30).-I. Vcc?2c~diu(tl2 AZuminium SiZicides.-W hen heated together potassium . silicofluoride aluminium and ammonium metavanadate form three vanadiurn aZuminiurn silicicles depending on the proportions of the mixture.I n presence of an excess of vanadium and small amounts of silicon the chief product is tt silicido crystallising in small dark octahedra together with small amounts of a silicide crystallising in large hexagonal prisms. As the proportion of silicon is increased and t h a t of vanadium diminished the octahedra disappear and a silicideINORGANIC CHEMISTRY. 47 crystallising in monoclinic leaflets is formed becoming the chief product when the mixture contains a large excess of silicon. Of these vanadium aluminium silicides only the hexagonal silicide V A12Si13 has been isolated. It crystallises in greyish-white prismatic needles J ) 4.3 hardness slightly above 5 and reacts with hydrofluoric acid with development of heat forming in absence of air a reddish-violet solution containing vanadium dijuoride VP which on exposure to air in presence of excess of hydrofluoric acid evolves hydrogen and forms the green vanadium trifluoride solution.The hexagonal silicide is stable towards boiling concentrated hydro- chloric nitric or sulphuric acid aqum regicc or fused potassium chlorate remains unchanged when heated in a current of oxygen over the bunsen flame and is not attacked by chlorine at the ordinary temperature but at a red heat is converted into volatile chlorides and is dissolved by fused alkalis or alkali carbonates. 11. Constitution of the SiEicides.-When treated with an excess of hydrogen fluoride as was described in the case of the chromium salts (Abstr. 1906 ii 63) the hexagonal vanadium aluminium silicide evolves seventy-four atomic proportions of hydrogen whereas if tlie ailicon mere completely transformed into the fluoride eighty-two atomic proportions including eight from the conversion of the difluoride into the trifluoride should be evolved.It is considered that this points t o linkings between silicon atoms which are not resolved by hydrofluoric acid. It is found impossible to construct a formula for the hexagonal silicide in which not more than eight of the thirteen silicon atoms are linked to form a chain. G. Y. Action of Sodium and Barium Peroxides on Gold. Aurates. FEENAND MEYEB (Compt. rend. 1907 145 805-807).-When pre- cipitated gold is added t o fused sodium peroxide a vivid reaction takes place and sodium aurate is formed.A similar but much less complete reaction occurs with barium peroxide. From the aurate auric hydroxide is prepared by means of sulphuric acid adopting the precautions recommended by Fremy (Ann. Chim. Plqs. 1851 31 [iii] 478). Analyses of the bydroxide dried in a vacuum in the dark agree with the formula Au20,,3H20 ordinarily written Au(OH) but from analyses of the aurates the author proposes the formula AU,O,(OH)~,ZH,~. Iiruss has described the hydroxide Au,O,,H,O which be writes AuO.OH. The aurates of sodium (bright green needles) and potassium are prepared by treating tt solution of the hydrate with the alkali in au atmosphere free from carbon dioxide and evaporating in a vacuum in the dark. They have formulx of the type M2(Au02)2. 'l'he salts of the alkali earth metals aiw obtained by treating solutions of their hydroxides with an alkali aurate solution. They form voluminour light green precipitates which retain much water are sensitive to light and have the forniulrt M(huO,),. The degree of hydration of ths aurates is diiiticult t o determine but it tends towards the limit of 611,O for those of potassium calcium strontium and barium and 2H20 for that of sodium. The aurates are stable towards lie& when dry but iu48 ABSTRACTS OF CHEMICAL PAPERS. solution are decomposed by light or heat giving a violet-black residue of aurous oxide Au,O. Sulphur dioxide and alcohol (slowly) precipitate metallic gold from their solutions which with dilute sulphuric and nitric acids form auric hydroxide and a sulphate or nitrate. Hydrochloric acid dissolves them forming gold chloride and an alkali or alkali earth chloride. With organic-matter they form powders wliich detonate when heated. E. H. Gold Chromate. N. A. ORLOFF (Clmz Zeit. 1907 31 1182).- A solution of auric clwoinate Au,(CrO,) is obtained by treating a large excess of freshly-precipitated silver chromate with a solution of auric chloride On evaporation the solution deposits at first gold but subsequently the chrornude Au,( CrO,),CrO crystallises from the red mother liquor. W. H. G.