INORGANIC CHEMISTRY.IN the present Report the aim has been to continue to present a representativeaccount of new work in the many fields of inorganic chemistry which, asshown by the number of original papers published, continues to attract anincreasing amount of attention. Owing to the time-lag between the publisheddate of some foreign journals and the actual time of their appearance in thiscountry, it has not been possible to include all the work published near theend of 1950; for the same reason reference has been made to a few paperspublished near the end of 1949.Thepresent classification under groups of the Periodic Table is convenient thoughit has its limitations ; inevitably some investigations are concerned withelements in more than one group. In such cases the work is described underthe group heading associated either with the first-mentioned element or withthe most important feature.The chemistry of some elements has receivedmuch attention during 1950; of others, virtually none. The two featureswhich are perhaps the most outstanding are the continued rapid increase inthe interest in the chemistry of fluorine and the wide range of applications ofion-exchange resins, not only for the separation of chemically very similarelements but also for the study of their physicochemical properties and evenfor preparative purposes.Of the reviews which have been published during the year reference maybe made to a collection of papers by a number of authors on a variety ofaspects of complex-compound formation which has been published in honourof the 75th birthday of Paul Pfeiffer, the pioneer in this field,l and to reviewsby Lister 2 on the Chemistry of the Transuranic Elements and by Sharpeon Interhalogen Compounds and Polyhalides.A review, by C ~ a t e s , ~ hasalso been published on the Organometallic Compounds of the First ThreePeriodic Groups which includes a discussion of some uncommon structuraland valency problems provided by these compounds. The centre of interestin organometallic compounds, apart from their use as preparative reagents,has seemed in recent years to belong more and more to the realm of InorganicChemistry, and mention is made in the present report of a number of neworganometallic compounds.Group 0.-A most interesting series of clathrate or cage-like compoundsof the inert-gases argon,5 krypton,6 and xenon has been described byPowell.These are prepared by slowly cooling an aqueous solution of quinolunder a high pressure of inert gas (necessary because of its low solubility).No single system of classification of topics is likely to be ideal.1 Angew. Chem., 1950, 62, 201.2 Lister, Quart. Reviews, 1950, 4, 20.Ibid., p. 217.6 Powell and Guter, Nature, 1949, 164, 240; Powell, J . , 1950, 298.Ibid., p. 115.Powell,ibid., p. 300. Idem, ibid., p. 468FAIRBROTHER. 99The general series of clathrate compounds of ideal formula SC,H,(OH),,Mformed by quinol with a, variety of substances M, which in their ordinarystates at room temperature may be volatile liquids or gases, consist of cage-like structures formed by the quinol on crystallisation in the presence of thesecond component, the atoms or molecules of which are trapped inside thecage.There is no chemical bonding in the usual sense between the inert-gasmolecules and the quinol cage. Nevertheless, these " compounds " areremarkably stable at room temperature and pressure although the inertgas may be effectively under a high pressure [more than 70 atm. for3C6H4( OH),,O.88 ; the fully saturated 3C,H4( OH),,A would contain argonat an effective pressure of 91 atm.]. Since the inert gas may be liberated bywarming or by dissolution of the quinol framework in some solvent, these com-pounds constitute a highly concentrated and portable source of inert gas.Similar compounds may be formed with other organic molecules whichare capable by hydrogen- bond formation of providing the necessary cage-likestructure.The inability of helium to form a clathrate compound withquinol is believed to be due to its small diameter, which permits it to escapethrough holes in the cage walls.Traces of nitrogen and oxygen present in commercial argon or helium canbe effectively removed by passing the gas over titanium powder at 850'.8Group 1.-A simple method of extraction of metallic czesium from theSwedish mineral pollucite, due to Hackspill and tho ma^,^ consists in heatingthe dried ore directly with calcium in vacuo at 900' : the czesium distils offand can be purified by redistillation in vacuo at 350400'.Lithium hydride can also be used lo as a reducing agent for the prepar-ation of other alkali metals by heating it with the corresponding chlorideunder reduced pressure.Superoxide formation, which occurs easily only with the large alkalimetals, has been extended by Stephanou, Schechter, Argersinger, andKleinberg l1 to include sodium superoxide, NaO,.The crystal structure ofthis has now been measured by Templeton and Dauben l2 and differs fromthat of the potassium, rubidium, and caesium superoxides, which are iso-morphous with CaC, and in which each pair of 0-0 atoms is oriented withits axis parallel to the tetragonal c-axis, in that it possesses a rock-saltstructure with the 0,- ions occupying positions like the chlorine ions butwith a random orientation of the 0-0 axis.The effects of the nature of the anion, the basicity of the amine, and thenature of the solvent on the properties of complexes between cupric and cuproussalts and long-chain amines have been investigated by Burkin.13 Cuprouschloride, bromide, and iodide complexes with monoamines l4 polymeriseMallett, Ind. Eng.Chem., 1950, 42, 2095.Compt. rend., 1950, 230, 1119.lo Pearce, Burns, and Gantz, Proc. Indiana Acad. Sci., 1949, 58, 99.l1 J . Amer. Chem. Xoc., 1949, 71, 1819.l2 Ibid., 1950, 72, 2251.l3 J., 1950, 122.Wilkins and Burkin, J . , 1950, 127100 INORGANIC CHEMISTRY.to a tetrameric form, whilst cuprous bromide also forms at the temperatureof freezing benzene bisamine complexes with a dimeric bridge structurewhich dissociate reversibly on warminginto the tetrameric monoamine complexesand free amine.The cuprous halideRH,NRH,N~~ k ~ ~ / \NH,R complexes of empirical formula CuX,NH,Rare oxidised in the air and in presence ofamine salts HX,NH,R form the corresponding cupric halide complexesCuX,,2NH,R.16 I n both cupric and cuprous complexes the usual co-ordination number of 4 is maintained. On the other hand, polarographicstudies of copper complexes with dipyridyl, o-phenanthroline, and thioureaby Onstott and Laitinen l6 indicate that in the presence of considerableexcess (50- to 200-fold) of complexing agent (dipyridyl) the cupric ion canco-ordinate three molecules, i.e., exhibit a co-ordination number of 6, whilstunder the same conditions the cuprous ion only co-ordinates two molecules.Other work, by Pfeiffer and Werdelmann,17 also indicates that in phenan-throline complexes with univalent central atoms (Ag,TI) the central atom hasa co-ordination number of 4.Bivalent perchlorates of Cu, Sn, Fe, Co,Be, Mg, Zn, Cd, and Hg on the other hand are reported to co-ordinate 3phenanthroline molecules with a co-ordination number of 6.Some light has been thrown by X-ray investigations hy Rundle andGoring of the crystal structure of the salt AgC1O4,C,H6 on the mechanismby which the argentous ion can become attached to the benzene ring. Itis found that each silver ion lies between two obliquely inclined benzenerings, equally bonded to two carbons in each, a structure which suggeststhe formation of x bonds between the silver and the aromatic ring.The compounds 5Ag,0,1,07, 2Ag,0,1,07 ,3H,O, and Ag,O,I,O, ,4H,Ohave been isolated by Gyani l 9 from the ternary system Ag20-H10,-H,O.Group II.-Hartford, Lane, and Meyer 2o have prepared and examinedthe two crystalline hydrates of calcium dichromate, CaCr2O7,3H,O, and thepreviously unknown CaCr,07,4.5H,0. When heated, the hydrates undergopartial decomposition by hydrolysis to CaCrO, and CrO, : there is no evidenceof the existence of anhydrous CaCr,O, as a separate crystalline phase.Onfurther heating, oxygen is lost to give a mixture of‘ CaCrO, and Cr203 whichat 11OO--l20Oo loses oxygen to form CaCr,04.Quinoline complexes of zinc and cadmium sulphates have been studied byBattacharya and Sinha.2l Polarographic studies by Douglas, Laitinen, andBailar 2, have demonstrated the existence of complexes, [Cd(en)3] + + ,[Cd( pn),] + + , [ Cd( dien),] + + , [ Cd( trien)] + + , [ Cd( dipy ),I + + , and [Cd(o -phen),] + + ,in the presence of considerable (100-2000-fold) excess of the respectivecomplexing agent.Mehta and Kabadi 2, have isolated sodium zincate, 2NaOH,Zn(OH),,H,O1 5 Wilkins and Burkin, J ., 1950, 132.l6 J . Amer. Chem. SOC., 1950, 72, 4724.l8 J . Amer. Chem. SOC., 1950, 72, 5337.20 J . Amer. Chem. SOC., 1950,72, 3353.s2 J . Amer. Chem. SOC., 1950, 72, 2484.NH2R Br Yli cu’ li c ul7 2. anorg. Chem., 1950,261, 197.lo J . Indian Chem. SOC., 1950, 27, 5.a1 J . Indian Chem. SOC., 1950,27, 21.23 J . Univ. Bombay, Sect.A, 1949,18, 54FAIRBROTHER. 101by slow evaporation in a vacuum of solutions of Zn(OH), in concentratedaqueous sodium hydroxide.Jander 24 has extended his work on acid-base systems in non-aqueousmedia to a study of molten mercuric bromide (m. p. 236") as an electrolyticsolvent in which many reactions can be carried out. Mercuric oxide, titratedconductimetrically with thallous bromide or mercuric sulphate, behaves as aweak base in HgBr,, dissolving in it much as NH, dissolves in water. Assuch it can be displaced from basic salts, e.g., BH,O,HgSO,, by the strongbase TlBr. In a study by Lamure Z5 of the system HgC1,-Cu(OH),-H20between 17" and loo", two definite compounds were found : 2HgC12,3Cu( OH),and HgC1,,3Cu(OH),. The paramagnetism of CU++ is much diminished inthese compounds, which with other observations suggests that they aresalts of complex CU++ ions.Group III.-Mass-spectrometric examination by Norton ,* of residuesfrom B,Hll which had been stored for a long time at -78" showed, in additionto traces of B10H14, B6HI0, and B5H,, species around mass 105-108.It isbelieved that these represent a new hydride B,H,, belonging to the stableB,H, + 27 series, although a formula BgH15 belonging to the less stableBnHn+6 series is not excluded.Stone and Emelkus g8 have examined the reaction of diborane with somealkene oxides and vinyl compounds. Diborane reacts rapidly with ethyleneor propylene oxide at -80" to give respectively diethoxyborine or diiso-propoxyborine and polymers of the type H*[CHR*CH,*O],*BH,, whereR = H or CH,, respectively.Boron trifluoride and ethylene oxide similarlyreact vigorously to give dioxan and a liquid boron-containing polymer.It may be noted that this polymerisation, unlike the polymerisation ofisobutene by boron triflu~ride,~~ does not require the presence of a hydroxylicco-catalyst : it is suggested that possibly the ethylene oxide itself may actas co-catalyst.28The systems BF,-PH,, BF,-ASH,, and BF3-HBr have been investigatedby Martin and Dial.30 The absence of compound formation in the last twosystems has been ascribed on the one hand to the greater radius of arsenicthan of phosphorus, and on the other hand to the small ionic character ofHBr. ASH, co-ordinates with both BCl, and BBr, in which the arsenicatom can approach more closely to the boron than in BF, : also the moreionic HF alone among the hydrogen halides forms co-ordination compoundswith BF,.Brown and Johannsen 31 have investigated the 1 : 1-addition compounds24 Jander andBrodersen, 2.anorg. Chem., 1950, 262, 3 3 ; Jander, Angew. Chem.,25 Compt. rend., 1949, 228, 1731.26 J . Amer. Chem. SOC., 1950, 72, 1849.27 Wiberg, Ber., 1936,69, 2816.28 J . , 1950, 2755.29 Evans and Meadows, J . Polymer Sci., 1949, 4, 359; Symposium on Friedel-Crafts30 J . Amer. Chem. Soc., 1950, 72, 852.1950, 62, 264.Catalysts, Nature, 1949, 164, 655.31 Ibid., p. 2934102 INORGANIC CHEMISTRY.which boron trifluoride forms with benzonitrile, o-, m-, and p-tolunitrile, andmesitonitrile.The dissociation pressure of boron trifluoride over thebenzonitrile addition compound is sufficiently low at room temperature topermit of purification by sublimation, whereas at 80" it is about 55 mm. ofHg : it has been suggested as a method for the purification of boron tri-fluoride.The specific electrical conductivity, electrolysis, and kinematic viscosityof liquid boron trifluoride-ethyl ether BF,,(C,H5),0 in the temperature range-10" to +45" have been the subject of an investigation by Greenwood,Martin, and E m e l k ~ s . ~ ~ The results indicate that considerable ionic characteris introduced into ethyl ether when co-ordinated t o BF, and that the doublecompound is ethyl ethoxyfluoroborate C,H5*[BF,*O*C,H,]. Freezing-pointdata for the COC1,-BF3 system 33 indicate the existence of two compoundsCOCl,,BF, (f.p.- 134.3") and (COC1,),,BF3 (f.p. - 137.0").A phase-rule investigation 34 of the system Al,O,-SO,-H,O at 60" in thebasic region between sulphur trioxide concentrations of 5.10-22.02% hasgiven evidence that only two stable solid phases, Al,O3,SO,,6H,O andAl,O3,2SO,,l1H,O, exist in this range, and that many basic aluminiumsulphates which exist at lower temperatures 36 are not stable chemicalindividuals at 60".An examination of the system MgS04-Al,(S04)3-H,0 by Bassett andWatt 36 has indicated the existence of one double salt MgAl,(S04)4,22H,0,which is the essential constituent of pickeringite, although specimens of thismineral usually contain also excess of MgS0,,7H20.Bassett 37 has also prepared a lead alunite which contains 44% ofPb[Al,(OH),( H,0),][S04], mixed with H,0[A1,(OH)5,H20][S04]2 and alittle H,O[AI,(OH),( H,0),][Al(OH)4],, which supports the view that in thealunite lattice one-third of the aluminium positions can be vacant.The equilibrium diagram of the system As,05-A1,03-H,0, studied byGuerin and Martin,38 indicates the existence of two aluminium arsenates-an acid salt 2As,05,Al,0,,3H,0 which exists in equilibrium with solutions ofhigh (43.2-71 -5 yo) As,05 content, and an orthoarsenate As,05,Al,03,2H,0at lower As,O, concentrations.Wallace and Willard 39 have studied the rate of exchange of radio-chlorinebetween solid aluminium chloride and carbon tetrachloride.The resultssuggest that the exchange takes place by the surface-ionisation of the carbon-halogen bond on charged centres of the ionic AlCl, lattice rather than throughthe formation of soluble molecular or ionic complexes such as occur in ahomogeneous system.Davidson and Jirik40 have found that the anodic oxidation of thallium33 J., 1950, 3030.33 Martin and Faust, J .Phys. CoZZ. Chem., 1949, 53, 1255.34 Henry and King, J . Amer. Chem. SOC., 1950,72, 1282.35 Bassett and Goodwin, J . , 1949, 2239.86 J . , 1950, 1408. 5 7 J . , 1950, 1460.313 Compt. rend., 1950, 230, 2025.39 J . Amer. Chem. SOC., 1950,72, 5275. 'O Ibid., p. 1700FAIRBROTHER. 103in anhydrous acetic acid yields exclusively thallous ion Tl+. The anodicoxidation of aluminium, gallium, and indium leads to a loss of metal from theelectrode considerably greater than that corresponding to Paraday's law onthe assumption that a triply charged metal ion is formed in each case.Itis suggested that the primary anodic products may be mixtures of singly andtriply charged metal ions.It has been shown by Glazer, McRoberts, and Schulman 41 that aluminiumdodecanoate (trilaurate), prepared by reaction between lauric acid andtrimethylaluminium in dry benzene, has no gelling properties in hydrocarbonsolution. This result may be ascribed to the fact that this aluminiumtri-soap does not contain any hydroxy-bonds by which hydrogen- bondedcross-linking can take place. Anhydrous aluminium triacetate and tripro-pionate have also been prepared.42Some aluminium selenates have been prepared and examined byBassett 43 and shown to be very similar to the corresponding sulphates-aresult that may be helpful in elucidating the structures of the sulphates,since the selenium atoms are much more easily distinguished from thealuminium atoms by X-ray diffraction than are sulphur atoms.It is becoming increasingly evident that the frequently postulatedchloro-acids of the type HAlC1, or HGaCl, do not exist as such.I n thisconnection the observations of Brown, Pearsall, and Eddy are significant :no reaction could be observed between hydrogen chloride and aluminium orgallium chloride under a variety of conditions from -120" to 300". Itwould appear that the simultaneous presence of a third, slightly basicprotophilic molecule may be necessary to bring about the ionisation of thehydrogen-chlorine bond.Clusius and Hitzig 46 have studied the halogenation of gallium metal byheating it with a number of other metal halides, and the reduction of thetrichloride by a number of metals.The reaction between silver and GaC1,is reversible : Ag + GaCl, AgCl + GaC1,. GaBr,,(CH3)20 andGaCI,,(CH,),O have been prepared, and their vapour-pressure curvesmeasured, by Van Dyke and Crawford : 46 trimethylamine forms compoundswith gallium chloride or bromide of the type GaC1,,2(CHJ3N. Van Dyke4' hasalso measured the conductance of these two halides in nitrobenzene, whichis increased by the addition of dimethyl ether. The gallium halides and theirether complexes are very weak electrolytes in nitrobenzene; the bromide ismonomeric in nitrobenzene solution.A study48 of some chlorogallates and related co~pounds of the typeM'M"'Cl,, where M' = NH,+, Li+, K+, or Cs+, has yielded information, notonly in respect of the preparation and properties of these compounds, butalso of the relative acidities of the several Lewis acids (Ga+++, Fe+++, A,+++)involved.The same authors have also found 48* 49 that KH,GaCl,, NH,FeCl,,41 J., 1950, 2082.4a Hood and Ihde, J . Amer. Chem. SOC., 1950, 72, 2094.44 J . Amer. Chem. SOC., 1950,72, 5347.46 J . Amer. Chem. SOC., 1950, 72, 2829.48 Friedman and Taube, ibid., p. 2236.43 J., 1950, 1191.4 5 Helv. Chim. Acto, 1950, 33, 506.4 7 Ibid., p. 2823.4~4 I b i d . , p. 3362104 INORGANIC CHEMISTRY.and NH,AlCl, when treated with dimethyl ether distribute themselvesbet-ween the solid phases and the saturated solutions without change incomposition, but that some other compounds of this type including LiGaC1,undergo a change in composition, the saturated ether phase being enrichedin M”’Cl, relative to M‘C1.The conditions under which hydrous gallium oxide is precipitated havebeen reinvestigated electrometrically by Moeller and King.” In thepresence of SO,- ion precipitation begins at a mole ratio of [OH-] : [Ga+++]of approximately unity and is complete a t a ratio of 3 : ’1. In the presenceof C1-, Br-, and NO,-, however, precipitation does not begin until the ratioreaches 3 : 1, whereupon complete flocculation takes place.Measurements of the magnetic susceptibility of TICl, and TI,CI, made byMeir and Garner 51 show that both these compounds are diamagnetic like thegallium and indium dihalide~.~, These compounds therefore contain themetals in 1 and 3 oxidation states or metal-metal bonds.The preparation of milligram quantities of pure actinium by the trans-mutation of radium in a chain-reacting pile, 226Ra (n ; y ) + 227Ra &227Ac, has made possible for the first time a direct observation of the proper-ties of pure actinium since this isotope of mass 227 has theconveniently long half-life of 21.7 years,% in contrast to the short-livednaturally occurring isotopes.Nine pure actinium compounds, all isomor-phous with the corresponding lanthanum compounds, have been preparedon a microgram scale and examined by X-ray diffra~tion.~~ The crystaldata provide additional evidence that actinium can be regarded as the firstmember of the “ actinon ’’ series.Considerable attention has been paid to the study of thechemistry of the lanthanons (rare-earth elements).Methods of separationand purification which have been studied include synthetic-resin ion-exchangemethods 56 and fractional precipitation of complexes with ammoniumacetate,67 ethylenediamine-NNN’N’-tetra-acetic acid,5* and other amino-acids.59 Lutetium may be purified 6o by the fractionation of hexa-antipyrinelutetium iodide [ LU(C,,H,,N,O)~]I,.The unusual dark brown Pr6011 which results when any other oxide, orsalts such as the nitrate or oxalate, of praseodymium are ignited in air, hasbeen examined by X-ray and other methods.The constancy of the latticedimensions of Pr6011 under widely different methods of preparation foundLunthanom.50 J . Phys. Coll. Chek., 1950, 54, 999.51 J . Chem. Physics, 1950, 18, 237.52 Klemm and Tilk, 2. anorg. Chem., 1932,207, 175.53 Hagemann, J . Amer. Chem. SOC., 1950,72,768.64 Curie and Bouissihres, Cahiers Phys., 1944,26, 1.6 5 Fried, Hagemann, and Zacharisen, J . Amer. Chem. SOC., 1950,72, 771.56 Spedding, Fulmer, Butler, and Powell, ibid., pp. 2349, 2354 ; Spedding and Dye,57 Vickery, J . , 1950, 1101 ; Perey, J . Chim. physique, 1949,46, 485.6 8 Marsh, J., 1950, 1819.ibid., p. 5350; Huffman and Ostwalt, ibid., p. 3323.59 Vickery, J . , 1950, 2058.8o Marsh, J . , 1950, 577FAIRBROTHER. 105by McCullogh 61 indicates that Pr601, is a separate and distinct phase inthe Pr-0 system, and a fluorite type of structure is suggested similar to thatassumed by Pro,, but with the random omission of one-twelfth of the anionsfrom the structure. Tensimetric measurements in conjunction with X-rayexamination and thermoelectric measurements 62 confirm that the structureis of a fluorite-minus-oxygen type but indicate that Pr forms a series ofnon-stoicheiometric oxides between Pro,., and Pro,., and that the air-ignited oxide is properly represented by Pr01.83. Pure Pro, can be readilyprepared by heating any lower oxide in oxygen at 50 atm. and about 300",but there is no evidence of oxidation beyond the 4+ state.61The hydrated normal lanthanon carbonates can be very convenientlyprepared by hydrolysis of aqueous solutions of the soluble trichlor~acetates.~~The solubility of ytterbium oxalate in buffered sodium oxalate solutions hasbeen studied by Crouthamel and Martin by a radio-chemical method.The results indicate the formation of Yb(C,O,)+ and Yb(C,O,),- but notof Yb(C,O,):-.In a search for new refractories, the sulphides of cerium have beenstudied.66 In addition to the known red Ce,S,, two new compounds Ce3S,and CeS, which are respectively black and brass-coloured, have been pre-pared.All three are very refractory, especially CeS (m. p. 2450" & 100")which has a v.p. of only lo3 mm. of Hg at 1900". Magnetic-susceptibilitydata show all three compounds to have one free electron, the additionalelectrons available for compositions with less sulphur than Ce,S, presumablybeing used for additional bonding between czsium atoms.The availability of milligram quantities of pure promethium (At.no. 61)from uranium fission products has made it possible for Parker and Lantz 66to make a detailed examination of its absorption spectrum. The solutions,which are rose or pink, show bands in the visible spectrum similar to those ofits neighbour neodymium but separable from them by at least 8 mp.Group N.-It has been shown by MacNevin and Carson 67 that carbonmonoxide is oxidised by iodine in acid aqueous solution : CO + I, +H,O = CO, + ZHI, the reaction being strongly catalysed by palladouschloride, and Katz and Halpern 68 have showed that it can be removedfrom a stream of air at ordinary temperatures by silver permanganatedeposited on a variety of metallic oxides, although dry silver permanganatealone is virtually unreactive towards carbon monoxide.The density, viscosity, and surface tension of anhydrous liquid hydrogencyanide have been determined by Coates and Davies 69 in the temperaturerange - 13.3 to + 25".The variation of these properties has been dis-cussed in relation to the linear association of hydrogen cyanide moleculesdue to hydrogen bonding.61 J . Arner. Chem. SOC., 1950, 72, 1386.62 Martin, Nature, 1950, 165, 202.64 Ibid., p. 1382.6 6 Ibid., p. 2834.8 8 Ind. Eng. Chem., 1950, 42, 345.Salutsky and Quill, J .Amer. Chem. SOC., 1950, 72, 3306.6 5 Eastman, Brewer, Bromley, Gilles, and Lofgren, ibid., p. 2248.e7 Ibid., p. 42.68 J., 1950, 1194106 INORGANIC CHEMISTRY.An examination of the isotherms at 10" and 31" for the system Na,O-Si0,-H,O has shown 70 that the following 6 compounds may occur as solidphases : Na,SiO, with 9, 8, 6 and 5H20; Na,HSiO, with 5 and 2H,O.Strickland 71 has shown that heteropolysilicomolybic acid exists in twoforms with the same empirical formula but different structures and formedaccording to the proportions of acid and MO," ion.When silicon tetrachloride is hydrolysed, even in moist air, the hydrolysisto silica is usually complete, but by partial hydrolysis in ethereal solution a t-78" a series of chlorosiloxanes can be prepared.72The rearrangement reactions previously reported by Anderson betweenmixed inorganic halides have now 73 been extended to include some fluoro-silanes.The redistribution reaction is an efficient method of preparationof bromofluorosilanes, isocyanatofluorosilanes, and especially of iodo-fluorosilanes. Three new iodofluorosilanes have been prepared by thismethod, vix., SiFI,, SiF,I,, and SiF,I, from the redistribution of SiF, andSiI,.SiC1,Br and SiCI,Br, react with Grignard reagents '4 with replacementof both chlorine and bromine, the proportion of bromine replaced beinghigher the larger the alkyl group ; ethyl bromodichlorosilane and ethyl-dibromochlorosilane have been prepared by this means ; SiClBr, and SiBr,are unreactive towards Grignard reagents.Eaborn 75 has examined theinteractions of triethyl- and trimethyl-silicon halides and pseudohalideswith silver salts and has shown that in the series R,SiI+ (R,Si),S+R,SiBr + R,SiNC --+ R,SiCI + R,SiNCS + R,SiNCO a compoundmay be converted into any other ou its right by boiling with the appro-priate silver salt, and has discussed this and other possible similar series interms of the solubilities of the salts and of the energy necessary to breakthe several Si-X bonds into Si+ and X-.A number of new alkylsilicon isocyanates and isothiocyanates have beenprepared 76 by reaction of the appropriate alkylchlorosilanes with silverisocyanate or isothiocyanate.Burg and Kuljian 77 have prepared a number of new silylamines andsilylaminoboron compounds.Trisilylamine (SiH,),N, prepared by thereactions SiCl, -+ 78 SiH, -+ 79 SiH,C1 -+ 8o (SiH,),N, reacts a tLIAIH, HC1 NH,-78" with BCl, to give (SiH3),NBCl, : (SiH,),N + BCI, -+ SiH,Cl +(SiH,),NBCl,. This new silylaminoboron compound is similar in severalAlCl,70 Baker and Jue, J . Phys. Goll. Chem., 1950,54, 299; Wills, ibid., p. 304.7 1 Chem. and Ind., 1950, 393.72 Goubeau and Warncke, 2. anorg. Chem., 1949, 259, 109; Schumb and Stevens,J . Amer. Chem. Soc., 1950, 72, 3178.73 Anderson, ibid., p. 2091.74 Wilkins, Brown, and Stevens, J., 1950, 163.7 5 J . , 1950, 3077.76 Anderson, J . Amer. Chem. SOC., 1950, 72, 196.7 7 Ibid., p. 3103.7 8 Finholt, Bond, Wilzbach, and Schlesinger, ibid., 1947, 69, 2694.79 Stock and Somieski, Ber., 1919, 52, 695.*O Idem, ibid., 1921, 54, 740FAIRBROTEER. 107respects to its carbon analogue (CH,),NBCI,.81 In a similar mannerB,H,Br reacts a t -78" with (SiH,),N with the quantitative formation ofB,H,,SiH,Br and the monomer and dimer of (SiH,),$BH, : this monomerreadily adds diborane to give (SiH,),NB,H,. (SiH,),N also reacts at roomtemperature with (CH,),BBr to give a number of products, including thenew and difficultly volatile, (CH,),BN(SiH,Br),. CH3N( SiH& reacts withBCI, a t -78" to give (CH,NSiH,)BCI,, which in turn decomposes quanti-tatively into SiH,CI and (CH,NBCl),.Milligan and Kraus 82 have prepared tristriphenylgermanylsilane byreaction of sodium triphenylgermanide with SiHCl, in ethereal solution.The hydrogen atom in this silane can be replaced by lithium in ethylaminesolution and by bromine in ethyl bromide solution.Tristriphenylgermanyl-silyl chloride, amine, and the corresponding silo1 have also been prepared.On treatment of the lithium salt of the silane with ethyl bromide in liquidammonia solution, the metal is replaced by the ethyl radical with theformation of ethyltristriphenylgermanylsilane.Bistriethylgermanium sulphate and diethylgermanium sulphate may beprepared by the action of concentrated sulphuric acid on the correspondingoxide.8,Everest and Terrey84 have shown that germanous hydroxide can beobtained in a form which may be either white or coloured according to theparticle size and degree of hydration. Conductivity measurements giveno indication of any reaction between Ge( OH), and NaOH ; no dissolution ofthe germanous hydroxide occurs even when ten equivalents of alkali areadded, and the evidence of the potential of the germanate-germanite couplesuggests that the fact that bivalent germanium can exist at all in alkalinemedia is due to the insolubility of &(OH),.Germanylsodium (NaGeH,), prepared by the quantitative reactionbetween sodium and monogermane in liquid ammonia, reacts with methyliodide and ethyl and propyl bromides to give the corresponding monoalkyl-germanes.Teal and Kraus 85 also find that when a solution of germanyl-sodium in liquid ammonia is electrolysed with a platinum anode and mercurycatahode, the anode reaction is not, as might be expected, the formation ofdigermane but of monogermane and nitrogen : 6GeH,- + 2NH, = GGeH, +N, + 6e-.A new and simplified method for the synthesis of ethylgermaniumtrichloride and diethylgermanium dichloride has been devised by Rochow 86which involves the direct reaction between ethyl chloride and elementarygermanium in the presence of a copper catalyst.The nature and conditions of formation of germanyl ferrocyanide,[Ge(OH),]Fe(CN), or (GeO),Fe(CN),,2H20, formed as an insoluble whiteprecipitate when ferrocyanide ion is added to an acid solution of germanium81 Wiberg and Schuster, 2. anorg. Chem., 1933,213, 77.82 J . Amer. Chem. SOC., 1950, 72, 5297.83 Anderson, ibid., p. 194.04 J., 1950, 2282.a 5 J. Amer. Chem. Soc., 1950,72,4706.B 6 Ibid., p. 198108 INORGANIC CHEMISTRY.dioxide, have been examined by Peisach, Pugh, and Sebba.87 This is thefirst time such a compound has been described.Experiments on tetraethyltin 88 have shown that it can crystallise in atleast ten forms, the m.p.s of which all lie within the range 137" H. to 148" K.Tetraethyl-lead also crystallises in many forms. Tetramethyl-tin and -leadand tetraethylgermanium do not show this unusual kind of polymorphismwhich, it is tentatively suggested, results from the molecular dimensionsenabling the molecules to pack into different lattices in which they havedifferent configurations, that is, to exhibit a form of rotational isomerism inthe solid state.Although previous studies have indicated that the only sodium stannate isNa2H8Sn0,, Grillot 89 has now shown that there exists, in addition to thecorresponding potassium compound, also K2Sn( OH), , which is formed whenthe solution contains excess of potassium hydroxide.The difficult separation of zirconium and hafnium has been attacked bya number of authors and by several methods.The observation by Hansenand Gunnar that silica gel will adsorb hafnium in strong preference tozirconium from a methanol solution of the tetrachloride has now been used 91for the purification of zirconium and, by differential extraction of the usedsilica gel, to the preparation of a concentrated solution of hafnium. Huffmanand Lilly 92 have found that adsorption of zirconium and hafnium as negativetluoro-ions on a strongly basic anion-exchange resin followed by elution witha mixture of O-~M-HC~ and O-O~M-HF gives an excellent separation on amilligram scale, the progress of elution being studied by a tracer method.Kraus and using larger quantities of the two elements and undersomewhat different conditions, have also achieved a partial separation bythis method.A good separation has also been accomplished by Gruen andK a t ~ , 9 ~ by the fractionation of 3ZrC1,,2POC13 (b. p. 360") and 3HfC1,,2POC13(b. p. 355"), by Huffman and Beaufait 95 by solvent extraction of the per-chlorates with thenoyltrifluoroacetone, and by Schultz and Larsen t~ byextraction of a hydrochloric acid solution of the ions with trifluoroacetyl-acetone.Wardlaw and Bradleyg7 have prepared a number of alkyl ortho-estersof zirconium by reaction of zirconium tetrachloride with a primary alcoholand anhydrous ammonia. When secondary alcohols are used in the reactionthe products are hydrolysed esters, the pure esters in this case being obtainedby ester interchange between the ethyl ortho-ester and the secondary alcohol.The ortho-esters of zirconium are much less volatile than those of eithersilicon or titanium.A series of new chloride ethoxides of zirconium has alsobeen ~ r e p a r e d . ~ ' ~87 J . , 1950,949. Staveley, Paget., Goulby, and Warren, J . , 1950, 2290.Compt. rend., 1950, 230, 1179.J . Amer. Chem. SOC., 1949,71,4158.9 1 Hansen, Gunnar, Jacobs, and Simmons, ibid., 1950, 72, 5043.92 Ibid., 1949, 71, 4147.9 j Ibid., p .3179.974 Bradley, Abd-el Halim, and Wsrdlaw, J., 1950, 3450.93 Ibid., p . 3263.96 Ibid., 1950,72, 3610.Ibid., p . 3843.97 Nature, 1950, 165, 75FAIRBROTHER. 109Solvent extraction, with pentan- 1-01 or any of a number of higher alcoholsand ketones, has been usedQ8 as a method of extracting thorium fromaqueous solutions of its naturally occurring mixtures with the lanthanons.D'Eye Qg has made an examination of the crystal structure of thorium bro-mide, with results which indicate that, as in the chloride, the thorium-halogen bonds are partly covalent in character.A series of refractory sulphides of thorium and uranium has beenprepared, loo viz. :Compound. Colour. M. p. d ( g . / c . c . ) .ThS ........................... Silvery > 2200" 9.57Th,S, ........................Brown 1950" f 50" 7.88Th,S,, ........................ Black 1770" f 30" 7.78ThS, ........................ Purple 1905" f 30" 7.36US ........................... Grey metallic >2000" 10.9 u,s, - -USJ, ........................... BIack 1850" f 100" 7.908.8 1 ...........................The compound Th,S, is of particular interest since it is the only definitelyestablished thorium compound which contains the element in the +3 oxid-ation state. None of these sulphides is paramagnetic, which means that allthe electrons even in the lower valency states are paired, and it is suggestedthat the electrons in the first three not used in bonding with the sulphur areused in bonding between the thorium ions in the (rock-salt) lattice just asbonding occurs in a metal lattice.Group V.-The electrolysis of a solution of aluminium iodide in liquidammonia with a platinum cathode and aluminium anode gives an intenselyblue colour similar to those of the alkali metals : lol the colour first appearson the platinum cathode and spreads through the solution.The need for simple compounds lo2 of thorium and uranium that areappreciably soluble in liquid ammonia without reaction more extensive thansolvation has led to the study of the behaviour of a number of compounds ofthese elements in liquid ammonia.A number of ammonolyic reactions havebeen observed.Although many investigations have been carried out on the gradual transi-tions which occur in some pure solids, relatively little is known of thephenomenon in binary mixed crystals of which one or both componentsexhibit such a transition.Mandleber and Staveley lo3 have now investigatedthe volume-temperature relationships of mixed crystals of ammoniumchloride and ammonium bromide.A comprehensive series of investigations into the chemistry of thenitronium ion NO,+ and its compounds has been carried out by Ingold andAsselin, Audrieth, and Comings, J. Phys. Coll. Chem., 1950, 54, 640; Templetonand Hall, ibid., pp. 954, 958.O9 J., 1950, 2764.loo Eastman, Brewer, Bromley, Gilles, and Lofgren, J. Amer. Chem. SOC., 1950, 72,l01 Davidson, Kleinberg, Bennett, and McElroy, ibid., 1949, 71, 377 ; McElroy,lo* Watt, Jenkins, and McCuiston, ibid., p. 2260. lo3 J., 1950, 2736.4019.Kleinberg, and Davidson, ibid., 1950, 72, 5178110 INORCIANIU OHEMISTRY.his ~ o l l a b o r a t o r s .~ ~ ~ 1 ~ Cryoscopic measurements show that NO,+ ions areformed when HNO,, N205, or N204 is dissolved in concentrated sulphuricacid.lM The freezing-point diagram of concentrated nitric acid, over therange of N205-H20 compositions in the neighbourhood of " pure " HNO,,indicates that N20, in nitric acid solution is completely dissociated intosolvated NO,+ and NO3- ions, and that in analytically anhydrous nitric acidat -40" some 3.4% is dissociated to give 102% of NO,+, 1.7% of NO,-, and06% of water.lo6 A study of its Raman spectra 112 has shown that whenN204 is dissolved in dilute nitric acid it is almost completely dissociated,partly homolytically, but chiefly heterolytically : 2N0, N204NO+ 3- NO,-.The nitrosonium ion NOf forms a molecular compound with nitrogendioxide: &0*&02, which may also be produced from nitronium ion andnitric oxide &O2-I$0, and it is believed that this compound arises throughthe formation of a single-electron bond by resonance between the structureskO,*kO and & 0 2 * & 0 .1 1 2 p 113A number of pure crystalline nitronium salts have been prepared,ll* zfiz.,the perchlorate (N02+)(C104-), hydrogen disulphate (N02+)(HS20,-),disulphate (NO,+),(S,O,"), trisulphate (NO,+),( S3010-), and fluorosulphonate(NO,+)(F*SO,-), the structural lattice components of some of which have beenidentified by means of the Raman spectra of the crystalline compounds, whichhas also given experimental confirmation that solid dinitrogen pentoxideis ionic nitronium nitrate (NO,+)(NO,-).Schomaker and Chia-Si Lu 115 have shown by electron-diffraction measure-ments that the molecule of nitrogen trifluoride is a symmetrical pyramid,with a F-N-F bond angle of 102.5' and not a nearly planar molecule as hadbeen conjectured earlier from its very small dipole moment.A study has been made 116 of the direct oxidation of phosphorus by steam,to give phosphoric acid and hydrogen, P, + 16H20 = 4H3P0, + 10H2,by passage of the vapours over a suitable catalyst a t 650-800". Bothplatinum and palladium supported on aluminium metaphosphate orzirconium pyrophosphate are found to be active catalysts, as are also theless stable, but cheaper, supported copper catalysts.104 Gillespie, Graham, Hughes, Ingold, and Peeling, J., 1950, 2504.lo6 Gillespie and Graham, J ., 1950, 2532.lo6 Gillespie, Hughes, and Ingold, J . , 1950, 2552.lo' Millen, J., 1950, 2606.lo8 Ingold, Millen, and Poole, J . , 1950, 2576.log Millen, J . , 1950, 2589.110 Idem, J . , 1950,2600.111 Ingold and Millen, J . , 1950, 2612.112 Goulden and Millen, J . , 1950, 2620.lla Goulden, Ingold, and Millen, Nature, 1950,165, 565.114 Goddard, Hughes, and Ingold, J., 1950, 2559.115 J . Amer. Chem. SOC., 1950, 73, 1182.118 Shultz, Tarbutton, Jones, Deming, Smith, and Cantelou, Ind. Eng. Chem.,1950,42, 1608FAIRBROTHER. I11Much work has recently been published on the chemistry of the condensedp01yphosphates.l~~ The study of the appropriate phase-diagrams 118 hasindicated the existence of two new phosphates, an unstable zinc phosphateZn(H2P0,),,2H3P0,, and an anhydrous manganous phosphate Mn(H,PO,),.Lange and Livingstone 119 have prepared anhydrous difluorophosphonic acid,HP0,F2, by the action of gaseous POF, on anhydrous monofluorophosphonicacid, POF( OH),.A new method has been described by Geach, Jeffery, and Shelton 120 forthe preparation of pure oxide-free arsenic, which involves the reduction in asealed tube of arsenious oxide by zirconium metal powder, mixed with 15 yoby weight of zirconium oxide to mitigate the violence of the reaction.The absorption spectra of vanadium(n1) and vanadium(1v) ions inaqueous chloride and perchlorate solutions have been examined by Furmanand Garner.121 The blue colour of vanadium(1rr) perchlorate solution is dueto the hydrated V3+ ion : on warming to 70" the colour changes to green,and reversibly to blue on cooling, owing to increased hydrolysis a t the highertemperature by the endothermic reaction V+++ + H20 VOH++ + H+.The spectra of aqueous vanadium chloride solutions indicate the formationof complex ions.Jantsch, Bergmann, and Rupp 122 have studied the am-moniates of vanadium(n1) chloride, vix., VCl,(NH,),, where x = 2, 3, 5.6, 7, or 12.Both the English (columbium) and the German (niobium) name forelement41 continue to appear in the literature.* There is still some question 123as to whether the latter will find general acceptance.There is no doubt thatthe oxide of' this element was extracted and recognised as new, and a numberof its properties correctly described, by Charles Hatchett in London in 1801 124and named by him columbium, before Ekeberg's discovery of tantalum 125(1 802) and more than 40 years before Heinrich Rose 126 extracted it and namedit Niobe, under the then current erroneous impression that Hatchett'scolumbium and Ekeberg's tantalum were one and the same.by use of the respectiveradio-isotopes, that columbium (niobium) and tantalum can be efficientlyseparated by fractional adsorption on an anion-exchange resin from solutionin a mixture of ~M-HCI and O-OSM-HF.11' Thilo and Riitz, 2. anorg. Chem., 1949, 260, 255; Andress, Gehring, and Fischer,ibid., p.331 ; Thilo and Hauschild, ibid., 1950,261,324; Van Wazer and Holst, J . Amer.Chem. Soc., 1950,72,639; Van Wazer, ibid., pp. 644,647; Van Wazer and Campanella,ibid., p. 655; Van Wazer, ibid., p. 906; Mehrotra and Dhar, Proc. Nat. Inst. Sci. India,1950, 16, 59; Jones and Monk, J., 1950, 3475; For further references see also Topley,Quart. Reviews, 1949,3, 345.It has been demonstrated by Kraus and11* Salmon and Terrey, J., 1950, 2813.120 J., 1950, 1207.122 Z.anorg. Chem., 1950,262,223.lZ4 Phil. Trans., 1802, 49.126 Pogg. Ann., 1844, 63, 317.* The Commission of Nomenclature of Inorganic Chemistry at the AmsterdamMeeting of the International Union of Pure and Applied Chemistry in 1949 recom-mended the adoption of the name niobium for element 41.llg J .Amer. Chem. SOC., 1950,72, 1280.121 J . Amer. Chem. SOC., 1950,72, 1785.125 Nicholson's J., 1802, 3, 251.12' J . Amer. Chem. Soc., 1949, 71, 3855.las Wichers, J . Amer. Chem.Soc., 1950,72,1431112 INORGANIC CHEMISTRY.Tantalum pentaiodide has been prepared by Alexander and Fairbrother 128by heating the metal in iodine vapour by means of high-frequency inductioncurrents. It forms shiny black crystals which can be sublimed in a vacuumwithout decomposition and have m. p. 496", b. p. 543". Measurements havebeen made of its vapour pressure from about 300" to near the boiling point.The reaction between columbium metal and an excess of iodine undersimilar conditions leads to the formation of well-defined lustrous crystals,essentially CbI,, with a brass colour, which are instantly attacked on exposureto air and lose iodine on gentle heating.2-Tartrato-2-columbic (-niobic) acid, Cb205( C4H405)2, 10H20, has beenprepared by dissolution of the freshly precipitated hydrous oxide in tartaricacid and precipitation in a micro-crystalline form by addition of alcohol.A heteropoly-acid structure of the type H,[Cb '4 ],2-5H20 is sug-ge~ted.l,~ The alkali-metal, alkaline-earth, and other metal salts have alsobeen investigated.130Group VI.-A clarification of the state of knowledge regarding the saltsof some of the simpler peroxy-acids has been made by Partington andFathallah.131 They have shown in the first place that the alkali peroxy-borates are true peroxy-acid salts and not simply addition compounds ofhydrogen peroxide and metaborates.Peroxyborates have also beenprepared, containing a greater proportion of active oxygen than previouslyreported, and corresponding to empirical formule LiBO,,H,O, KBO,,H,O,RbB04,0.5H20, and CsB04,0-5H,0. The previous contradictory experimentalwork on the alkali peroxycarbonates has also been reinvestigated by the sameauthors and, in addition to the confirmation of a number of compoundspreviously reported, the new compounds CsOOH,H,O,, KHCO,, RbHCO,,Rb,C,O,, and Li,C04,H20 have been prepared.Sulphuryl fluoride has been prepared by Woyski 133 by passing sulphurylchloride vapour or an equimolar mixture of sulphur dioxide and chlorineover sodium or potassium fluoride a t 400".Trifluoromethylsulphur penta-fluoride CF,*SF, has been prepared by Silvey and cad^.',^ The fluorin-ation of methanol gives trifluoromethyl hypofluorite CF,*OF. The fluorin-ation of methanethiol, however, either by cobalt(II1) fluoride at 250" orby dilute fluorine and a copper-supported silver fluoride catalyst at ZOO",instead of yielding the sulphur analogue, gives CF3*SF, as one of the products.This compound (b. p. -20.4", m. p. --81"), which is also obtained byfluorination of carbon disulphide, greatly resembles sulphur hexafluoride inits inertness.A new investigation of the solid-liquid equilibria in the system SO,-H,Ohas been carried out by Gable, Betz, and Maron 135 over a range of 0-93.7%of SO,. Eight solid phases, vix., H,S04 with 6, 4, 3, 2, or 1H20 : H,SO,,H2S20,, and ice, are found within this range. All of these except( c ,H 4 0 6 1Is* J., 1949, 2472.1*0 Idem, ibid., pp. 300, 381.las J .Amer. Chem. SOC., 1950,72, 919.laS Srinivasan, PTOC. Indian Acad. Sci., 1950, 31, 194.131 J., 1949, 3420.la' Ibid., p. 3624.13a J . , 1950, 1934.lS6 Ibid., p. 1445FAIRBROTHER. 113H2S04,6H20 and H2S04,3H20 exhibit congruent melting points. Thehydrate H,S04,3H20 has not been reported previously.The behaviour of sulphuric acid as an ionising solvent has beenexamined by Gillespie, Hughes, and Ingold. 136 The experimental methodsof cryoscopy in this solvent have been developed to a high degreeof precision. The cryoscopic constant of sulphuric acid (kf = 5-98 deg.g.-mol.-l kg.) has been obtained by using three types of solute, (a) sulphurylchloride and chlorosulphonic acid, which are un-ionised, ( b ) potassiumsulphate and ammonium sulphate, which are ionised, e.g., K2S04 + H2S04 =2K+ + 2HS04-, and ( c ) acetone and acetic acid, which behave as strongbases, e.g., B + H2S04 = BH+ + HS04-.Results with the three types ofsolute are in good agreement.From a study of the freezing points in the H,O-SO, system at composi-tions near that of sulphuric acid 13’ it is concluded that the ionisation of wateras a base in solvent sulphuric acid is appreciably incomplete, and that onthe other hand, disulphuric acid (H,S207) present in solutions with a higherSO, content, is moderately ionised. This cryoscopic method has also beenused to investigate the nature of the species present in sulphuric acid and indilute 01eum.l~~ In addition to the presence of H2S207 in oleum, evidencehas been obtained of the presence of higher polysulphuric acids, H3S3010and H2S4013 being definitely recognised.Freezing-point depressions of nitronium perchlorate and ammoniumperchlorate in solvent sulphuric acid 139 show that the perchlorate ion inthese salts is largely solvolysed to free perchloric acid ; which means thereforethat perchloric acid is a very weak acid when dissolved in sulphuric acid.On the other hand nitro-compounds e.g., nitromethane, nitrobenzene, andderivatives of nitrobenzene, behave as fairly strong bases in anhydroussulphuric acid, 140 the percentage ionisation in 0.1 M-SOlUtiOn varying from21 yo for nitromethane to 73% for p-nitrotoluene.Cations of high positive valency in aqueous solution usually combinewith the oxygen ions present to form oxy-cations of lower valency beforesalt formation occurs, so that the salts obtained from such solutions areoxy-salts.A method has now been developed by Hayek and Engelbrecht 141for the preparation of sulphates of higher-valency cations, which consistsin treating the chloride of the ion in question, in the absence of water, withsulphur trioxide in sulphuryl chloride solution. In this manner Ti(SO,),,Sn(SO4)2, v2°(s04)& Sb,O(SO4)4, CrO(SO4)2, MoO(SO4)2, WO(SOJ2, andU(SO,), have been prepared.The green solution of chromic sulphate, familiar to anyone who hasheated a solution of chrome alum, and the corresponding heated solutions ofthe chloride, have received considerable attention in the past.On the otherhand, little work has been done on the nitrate, probably because there islittle colour difference between a fresh solution and one that has been boiled136 J., 1950, 2473.138 Idem, J . , 1950, 2516.140 Idem, J . , 1950, 2542.la’ Gillespie, J., 1950, 2493.13@ Idem, J., 1950, 2537.lol Momtsh., 1949, 80, 640114 INORGANIU CHEMISTRY.and cooled. Hall and Eyring 142 have now attacked the problem of chromiccomplexes by a new method, using the so-called ammonium paramolybdate,(NH,),Mo,O~~,~H,O. Conductimetric titration with solutions of this reagentpermits the determination of the average number of oxygen bridges betweenchromium atoms in boiled solutions of chromic salts and has made it possibleto show that a structural change also accompanies the heating or ageing of asolution of chromic nitrate.The formation of oxalate complexes of chromium has been studied bySh~tt1eworth.l~~ Pure anhydrous chromic iodide has been prepared forthe first time by Handy and Gregory by heating CrI, in iodine vapour at1 atm.pressure and 500". The equilibrium Cr13(s) Cr12(s) +has been studied between 309" and 373". A rather long extrapolation leadsto an approximate value of atm. for the dissociation pressure a t 25",the compound being very stable at room temperatures.Hartford and Lane lg5 have described the preparation of a number ofnew compounds belonging t o the class of double chromates of some of' thetransition elements.The constitution of the peroxychromates has beenexamined by G 1 a ~ n e r . l ~ ~ It is suggested that the familiar ether-solubleblue perchromate is formed by the addition of an HO, radical to chromiumtrioxide : CrO, + HO, HCrO,. At a pH above 4.5 or on dilution bythe addition of an excess of hydrogen peroxide, which raises the pH, anothermolecule of hydrogen peroxide is added forming the violet perchromate :HCrO, + H,O, _T H,CrO,. The quinquevalency of the chromium in theperchromates and the reduction of chromic acid by hydrogen peroxide in acidsolution are considered to be due to the odd electron in the HO, radical.Molybdenum blues have been prepared by Sacconi and Cini14' with aMO(VI) : Mo(v) ratio from 9.0 : 1 to 0.56 : 1 and a water content of from17.96% to 19.38%.On heating at 195", almost all the water is lost to givea typical semi-conductor which decomposes at 350" to give a mixture ofMOO, and a more reduced blue. These compounds have a small para-magnetism, which is interpreted by the authors as indicating the presenceof Mo(v)-Mo(v) covalent bonds. The formula of sodium paratungstate hasbeen discussed by Saddington and Cahn.148 As a result of some new chemicaland crystallographic analyses they conclude that the crystalline salt has thecomposition Na,oWl,O,, ,Z8H20.Mair 149 has examined the reaction between ferric salts and sodiumparatungstate in boiling aqueous solution and has isolated 11 -tungstoferric(III)acid, the first member of a new series of 11 -heteropoly-tungstic acids withmainly positive central ions.Mair and Waugh 150 have also prepared threefurther 11-heteropoly-tungstic acids with tervalent central ions, vix., 11-tungstoaluminic, 11 -tungstochromic(IIr), and 1 1 -tungstomanganic(r) acids.1 4 2 J . Amer. Chern. SOC., 1950, 72, 782.143 J . Amer. Leather Chem. ASSOC., 1950, 45, 41.144 J . Amer. Chem. SOC., 1950,72,5049.1Q6 J . , 1950, 2795.145 Ibid., p. 1286.J., 1950, 3526. J., 1950, 2364. 150 J . , 1950,2372.1 4 7 J . Chem. Phys., 1950,18,1124FAIRBROTHER. 115From all three, salts have been prepared. A new and convenient method ofpreparation of free heteropoly-acids, and thence of other salts, has beendevised by using cation-exchange re~ins.1~1 The method is applicable evento the preparation of the free acid from an almost insoluble salt.A series of lithium tungsten '' bronzes " similar in composition to thewell-known sodium t,ungsten " bronzes " has been prepared.ls2Uranium and the Transuranic Elements.This year has seen the public-ation of an account of that part of the immense amount of research workcarried out during the War under the Manhattan Project and the AtomicEnergy Commission of the United States which was known as the PlutoniumProject. The account, which consists of a collection of some 162 originalresearch papers by 114 authors, edited by Seaborg, Katz, and Manning, iscontained in two volumes under the title of " The Transuranic Elements ".ls3The papers deal principally.with the chemistry and physics of neptuniumand plutonium but include also many relating to special techniques and toother radioactive elements such as actinium, thorium, radium, americium,and curium and to the general problem of the transuranic elements.Evidence is steadily increasing, as new elements are made, that thisseries of elements does in fact constitute a new series, variously referred to inthe literature as " actinide ", " actinon ", or " uranide " elements, in whichthe 5f electron shell is in process of being filled, somewhat in a similar mannerto that in which the 4f shell is filled in the rare-earth or lanthanon series.There is, however, still some diversity of opinion 154 as to where this seriesactually starts, that is, whether actinium or uranium should be regarded as thefirst member of the series and which ions contain f electrons and how many.The evidence, of the presence o f f electrons in any particular ion, fromabsorption or magnetic measurements, however, may not only depend on theelectronic configuration of the neutral gaseous atom and the oxidation stateof the ion, but may also depend on its environment, in particular if themultiplicity is reduced by covalent-bond formation.The problem is madeless simple by the fact that the 5f electrons are apparently less firmly boundthan the 4f electrons in the lanthanons. A review of the general lines ofevidence regarding the electronic structure of the heaviest elements fromchemical and physical data has been given by S e a b ~ r g .l ~ ~The hydrolysis of the uranyl and other uranium ions in aqueous solution,and the various acidic and basic compounds that are precipitated on theaddition of bases, have received further attention. By determining thechanges in pH and conductivity of solutions of uranyl nitrate or acetate, onaddition of measured amounts of NH,*OH, NaOH, or KOH, Tridst 156 hasfound that a basic salt is first produced and eventually a hydrated trioxideU03,xH20 at 1.5 mols. of NH,*OH or 1.6 mols. of NaOH or KOH per mol.151 Baker, Loev, and McCutcheon, J . Amer. Chem. SOC., 1950,72,2374.152 Straumanis and Hsu, ibid., p. 4027.153 " The Transuranic Elements," Edited by Sesborg, Katz, and Manning, McGraw15* Cf.Ann. Reports, 1949, 46, 88.165 Ref. 153, p. 1492.Hill Co., New York, 1949.lS6 Ann. Chim., 1950,5,368116 INORGANIC CHEMISTRY.of UO,++. Uranyl nitrate yields a diuranate, M,0,2U03, with an excess ofbase, whereas uranyl acetate first gives an unstable monouranate which isconverted into diuranate in the presence of water or uranyl salts.The existence of the ion U4+ in acid aqueous solution has been confirmedby Kraus and Nelson,ls7 and the hydrolysis of the U4+ and the Pu4+ ionstudied in further detail. The striking similarity between the sharp- bandedabsorption spectra of unhydrolysed Pu4+ and U4+ suggests that both ele-ments are members of the actinon series, but that Pu4+ has only two 5felectrons instead of the expected four.A comparative potentiometric and photometric absorption investigation 15*of uranyl monochloroacetate indicates that three complexes are formed :(UO,)*A+, (UO,)A,, and (U02)A3-, where A = CH,Cl*CO*O-.The solubility of some inorganic nitrates in ether has been examined inrelation to the now familiar purification of uranyl nitrate by ether-extrac-t 3 0 n .l ~ ~ The solubility of bismuth nitrate and ferric nitrate in ethyl etheris considerably greater than that of 23 other common nitrates. The presenceof uranyl nitrate in the ether lowers the solubility of other nitrates, and thepartition coefficients are such that one or two aqueous extractions of anethereal solution will remove any likely impurities from uranyl nitrate. Thethermal instability of the uranyl nitrate-ether complex forme din this extractionhas been examined.160 The concentrated solution, from which the complexUO,(N0,),,2H,0,2(C,H5),0 can be isolated, deflagrates at 85-90" owingto oxidation of the ether by the high concentration of nitrate ion.Glueckauf and McKay 161 have pointed out that the chemical propertiesof uranium and other actinon elements suggest that f-electron orbitals maybe used in bond formation in covalent complexes.The magnetic susceptibilities of elements 92 to 95 (uranium to curium)in most of their oxidation states have been measured in aqueous solutionby Howland and Calvin : 162 the results could only be interpreted on the basisof electronic configurations involving 5f electrons even though the suscepti-bilities were in general lower than the theoretical and experimental values forlanthanon ions with the same number of 4f electrons.Dawson and Lister 163have measured the magnetic susceptibilities of UO, and intermediate oxidesup to U,O, over a range approximately 90-570"~. The susceptibilitiesare consistent with the view that these oxides contain quadri- and sexi-valentbut no quinquevalent uranium.Measurements have also been made 164 of the magnetic susceptibilities ofsolid solutions of uranium dioxide in thorium dioxide. The results indicatethat the magnetic moment of U4+ shows little dependence on the concen-tration of UO, in the solid, and at the lowest concentration (2% UO,) is inagreement with the " spin only " formula for two unpaired electrons.This is15' J . Amer. Chem. Soc., 1950,72, 3901.158 Ahrland, Acta Chem. Scand., 1949, 3, 783.159 Bachelet, Cheylan, and Bris, J. Chim. phys., 1950,47, 62.160 Cheylan, Bull. SOC. chim., 1949, 641.162 J. Chem. Phys., 1950, 18, 239.lS1 Nature, 1950,165, 594.lea J . , 1950, 2181.Trzebistowski and Selwood, J . Amer. Chem. SOC., 1950, 72, 4504FAIRBROTHER. 117interpreted as an indication that these two unpaired electrons should beassigned to the 6d rather than to the 5f shell, on the assumption that quench-ing of the orbital contribution only occurs for the d electrons.Sheft, Fried, and Davidson 165 have prepared pure uranium trioxideby heating U,O, at 600-700" in pure dry oxygen a t a pressure of' about 25atm. This procedure gives a lOOyo yield of UO, over a wide range oftemperature. Above 750°, however, a higher pressure of oxygen is requiredto prevent the formation of oxides intermediate in composition betweenU,O, and UO,.When uranyl oxalate (U0,C,04,3H,0) is heated in theair,166 it loses 2H,O at 120" and the third a t 210". The anhydrous salt beginssuddenly to decompose at 310" to form UO, and U,O,. Continued heatingin the air gives a mixture of U,O, and UO,.The precipitation of uranium as U0,,2H20 from uranyl nitrate andhydrogen peroxide is substantially quantitative at pH 2 . 5 - 3 ~ 5 . ~ ~ ~ Twodifferent crystalline forms of this peroxide are formed, according to whetherthe uranyl nitrate or the hydrogen peroxide is in excess : the U-U distanceis the same in both types of crystal.U04,2H,0 dissolves in aqueous solutionsof sulphites without gas evolution to give solutions which contain, in additionto sulphate, also uranyl sulphite, an intensely yellow compound notpreviously reported.The system uranyl sulphate-water, investigated by Secoy,168 shows atwo-liquid phase region around 300-375" at all concentrations up to 72% ofUO,SO,. The lower critical solution temperature is 295" a t a weightconcentration of about 28% of UO,SO,. The solubility curve of solidU0,S0,,H20, which has been followed up to 363", does not intersect thetwo-liquid phase region.It is reported 169 that intermediate fluorides of uranium, includingU4F1,,U2F9, and ct- and p-forms of UF,, are formed by treatment of finelydivided UP, with gaseous UF, a t suitable temperatures and pressures.Dawson, Ingram, and Bircumshaw 170 have studied the reduction ofuranium hexafluoride with hydrogen.A need for some simple compounds of thorium and uranium that areappreciably soluble in liquid ammonia without other reaction than solvationhas led to the examination of the behaviour of a number of these compoundsin liquid ammonia.171 Thorium(rv) nitrate and sulphate, uranyl nitrate, anduranium(r1r and IV) chlorides and bromides react with the solvent, butthorium(rv) iodate and oxalate and uranium peroxide dihydrate are unreactive.Polarographic measurements of the N~(III)-N~(Iv) couple in perchlorica.nd hydrochloric acid have been made by Hindman and K r i t ~ h e v s k y .~ ~ ~185 J .Amer. Chem. SOC., 1950, 72, 2172.Boulle, Jary, and Dominb-Berg&, Compt. rend., 1950, 230, 300.16' Watt, Achorn, and Marley, J . Amer. Chem. SOC., 1950,72, 3341.168 Ibid., p. 3343.16s Agron and Weller, U.S.P. 2,510,850; Chem. Abs., 1950,44, 71649.170 J . , 1950, 1421.171 Watt, Jenkins, and McCuiston, J . Arner. Chem, SOC,, 1950, 72, 2260,Ibid., p. 953118 INORGANIO CHEMISTRY.The couple is polarographically reversible in hydrochloric acid with a redoxpotential Np4+ + e- Np3+ = +0.142 & 0-005 volt, but is irreversiblein perchloric acid.Some interesting facts have emerged in connection with the separationof americium and curium from rare-earth fission p r 0 d ~ c t s . l ~ ~ As is the casefor the lanthanon elements, the only thermodynamically stable oxidationstate of americium and curium in aqueous solution is the 3+ state, and sincetheir crystal radii overlap those of the lighter lanthanons, being about thesame as that of neodymium, the solubilities of the salts of these elements arevery similar to those of the lanthanon salts.A satisfactory separation hasbeen achieved by the now familiar method of adsorption on a syntheticion-exchange resin and fractional elution with ammonium citrate solution.Whereas, however, the eluent usually removes the elements in reverse orderof their atomic number, Le., the heaviest first, and with 6~-hydrochloric acidamericium and curium are eluted in their normal order, yet elution with13.3~-hydrockloric acid not only separates the americium and curium fromall the rare earths but the expected order of their elution is reversed. This,it is suggested may be due to weak covalent co-ordination of additionalC1- ions in concentrated hydrochloric acid, using the 5f electrons of theseheavy elements, available as a consequence of the smaller separation instability of the 5f and 6d orbitals as compared with the 4f and 5d of thelanthanon elements, and would indicate that the covalent complex ofamericium is slightly more stable than that of curium.Although Am3+ is the stable oxidation state in aqueous solution, it isreported 174 that it can be completely oxidised by ammonium persulphate in0*2~-nitric or -chloric acid to a formal valency of 6+.The colour changesfrom the characteristic pink to a strong yellow, and on addition of sodiumacetate a compound, presumably sodium americyl acetate, was obtained,which was isomorphous with the corresponding sodium uranyl and plutonylacetates, Na(UO,)(OAc), and Na(PuO)( Oh),.The production by Thompson, Ghiorso, and Seaborg 175 of a radioactiveisotope of element 97, which has been named Berkelium (Bk), has madepossible the examination of some of the properties of this element by tracertechnique.176 This isotope (half-life 4.6 hours, decaying by electron capturewith about 0.1% branching decay by a-particle emission) was prepared bythe bombardment of 241Am with about 35 MeV.helium ions and is believed tobe 243Bk or possibly 244Bk. The sequence of elution by ammonium citrate ofthe group berkelium-curium-americium from a synthetic ion-exchange resinindicates the same kind of break in ionic radius a t the point of half filling ofthe 5f electron shell (curium in the actinon series) as is known to occur in thelanthanon series a t gadolinium.Consequently, just as terbium can beoxidised in the solid state to TbO, (though not in aqueous solution), it was173 Street and Seaborg, J . Amer. Chem. Soc., 1950,72, 2790.Asprey, Stephanou, and Penneman, ibid., p. 1425.Phys. Review, 1950, 77, 838.176 Thompson, Cunningham, and Seaborg, J . Amer. Chem. SOC., 1950,72,2798FAIRBROTHER. 119expected that Bk3+ should be capable of oxidation to Bk4+ and that this mightbe achieved even in aqueous solution. This has now been confirmed bytracer experiments, which consisted of the measurement of the activitycarried down by a zirconium phosphate precipitate from solutions of knownoxidation potentials.I n this way a formal oxidation potential for thecouple Bk4+ + e- = Bk3+ of about + 1.6 V. has been indicated.The latest element to be prepared,17' element 98, obtained by bombard-ment of M2Cm with 35-Mev. helium ions, and probably 24498, has been giventhe name Californium (Cf). This isotope has a half-life of about 45 minutesand decays at least partially by ct-particle emission. An examination ofsome of its chemical properties has been made by Street, Thompson, andSeaborg,17* by methods similar to those used for berkelium. Its propertiesindicate that it fits in well as the ninth member of the actinon series, especi-ally as shown by the relative rates of elution by ammonium citrate from acation exchange-resin in the series Cf-Bk-Cm-Am as compared with theirlanthanon analogues, Dy-Tb-Gd-Eu. Only the 3+ oxidation state has sofar been examined, and the existence of a higher oxidation state in the formof an oxygenated ion (CfO,)+, stabilised by the two oxygen atoms, is stilluncertain.Group VII.-An account has now been published 179 of the Symposiumon the Chemistry of Fluorine held at the Royal Institution in November 1949.A new estimation of the dissociation energy of the fluorine molecule, forwhich the values in the literature range between about 64 and 33 kcals.per g.-mol., has been made by Evans, Warhurst, and Whittle; 180 it isconcluded that all the experimental evidence a t present available stronglyindicates a value for D(F2) of about 37 kcals.g.-mol.-l and certainly notgreater than about 45 kcals. g.-mo1.-1. This in turn leads inescapably to avalue for the electron affinity of a gaseous fluorine atom which is less thanthat of chlorine although in solution the total electron affinity of fluorine(i.e., including the heat of solvation of the ion) is still found to be greaterthan that of chlorine.The chemistry of the inter-halogen compounds and especially of thehalogen fluorides has been extensively studied by Emelhus and his collabor-ators. From this work has emerged the striking versatility of brominetrifluoride as an electrolytic solvent and fluorinating agent.The electrical conductivities of several inter-halogen compounds havebeen measured.The specific conductivities of ClF,, BrF,, and IF, 181 arerespectively <1eS (at O"), 8.0 x lo3 (at 25"), and 2-3 x ohm-1cm.-l The conductivity of BrF,, which obeys Ohm's law, decreases withtemperature in the range 15-60", but for IF, there is a positive temperaturecoefficient. The conductivity of molten IClYfe2 in which the degree ofionisation is of the order of 1%, passes through a maximum a t 40". Theconductivity of solid IC1, increases rapidly with temperature and passesThompson, Street, Ghiorso, and Seaborg, Phys. Review, 1950, 78, 298.17* J. Amer. Chem. SOC., 1950, 72,4832.180 J., 1950, 1524. J., 1949, 2861.17g Emelkus, Nature, 1950, 165, 224.lS2 Greenwood and Emelbus, J., 1950, 987120 INORGANIC OHEMISTRY.without marked discontinuity into that of the molten compound, whichshows a maximum at 111".To explain the phenomena in bromine trifluoride, Banks, Emelkis, andWoolf 181 postulated the existence of BrF,+ and BrF,- ions, 'uz'x., ZBrF,BrF,+BrF,- , an assumption amply supported by succeeding investigationswhich have also demonstrated the existence of an " acid-base " system inthis s01vent.l~~ Compounds such as (BrF,)SbF, and (BrF,)2SnF, behaveas acids, and KBrF,, AgBrF,, and Ba(BrF,), as bases, and neutralisationreactions have been shown, by conductivity measurements, to occurbetween them to give salts of SbF,- and SnF,-.Bromine trifluoride has also been found to be a valuable fluorinating agent.Sharp has prepared a compound, AuBrF,, by dissolving gold in BrF,.This compound also behaves as an " acid " in BrF, and with bromotetra-fluorides (" bases ") forms fluoroaurates such as AgAuF,.Similarly, salts of a wide variety of complex fluoro-acids of metals and ofthe nitronium ion have been prepared by Woolf and Emelhus 385 either bytreating an oxy-salt, such as a borate or metaphosphate, directly withBrP,, to give a tetrafluoroborate or hexafluorophosphate, respectively, or by" neutralisation '' reactions in BrF,.Nitronium salts were prepared bydissolving N,O, and an " acid " forming substance, e.g., Sb,O, or SnF,, inBrF, ; in this manner (NO,)BF,, (NO,),SnF,, (NO,)PF,, (NO,)AsF,,(NO,)SbF,, and (NO,)AuF, have been prepared.A variety of nitrosyl complex fluorides capable of functioning as " acids "in BrF, solution, and nitrosyl fluorosulphonate which can undergo reactionswith both " acids " and " bases " to give nitrosyl salts and fluorosulphonates,respectively, have been prepared including the following new salts :AgSO,F, (NO)SO,F, (NO,)SO,F, (NO)AuF,, (NO)PF,, (NO),GeF,,(NO),SnF,.186 Bromine trifluoride has also been used to prepare PtF,,RhF,, and PdF, from the halides of these metals.Niobium (columbium), tantalum, or their pentoxides react with BrF,to give fluorobromonium hexafluorotantalate (BrF,)NbF, and (BrF,)TaF,which act as " acids " in bromine trifluoride and are neutralised by bromo-fluorides. Bismuth pentoxide [70% Bi(v)] reacts with BrF, to give BiOF,containing BiF,, and BiF, forms (BrF,)BiF, which behaves as an " acid "in BrF,.laaThe only simple anhydrous fluoride of molybdenum known hitherto is thehexafluoride. The compound MoF, has now been prepared lag by heatingthe tribromide to 600" in hydrogen fluoride.Reactions of BrF, with mixturesof VCl, and potassium, silver, or barium chlorides give the correspondinghexafluorovanadates. The reaction between BrF, and a number of oxidesand some oxy-acid salts has been studied by Emelbus and W00lf.~~OComplex fluorides prepared by " acid-base " reactions in BrF, are183 Woolf and Emelkus, J., 1949, 2865.185 J., 1950, 1050.188 Gutmann and EmelBus, J . , 1950, 1046.18Q Emel6us and Gutmann, J., 1950, 2979.18' J., 1949, 2901.18' Sharpe, J., 1950, 3444.loo J ., 1950, 164.ls6 Woolf, J., 1950, 1053FAIRBROTHER. 121often impure and contain bromine. It has now been shown by Sharpe lglthat this result is due to solvolysis by the BrF, or incomplete reaction betweenacid and base.Further types of neutralisation reactions in hydrogen fluoride and iodinepentafluoride have been studied by W00lf.l~~ These lead to the formationof complex fluorides by such reactions as :K+€€F,- + H,F+BF,- = KBF, + 3HFKfIF,- + IF,+SbF,- = KSbF, + ZIP,Potassium iodohexafluoride KIF,, has been prepared by Emelbus andSharpe by dissolving K F in excess of IF,, the surplus IF, being removedby evaporation in vacuo. The preparation of iodine heptafluoride from itselements and a study of some of its properties have been described bySchumb and Lynch.lg4 The Raman spectra of IF, and IF, indicate that theIF, molecule is a tetragonal bipyramid with four F atoms at the corners ofa square base, and the iodine and fifth fluorine atom a t the ends of the four-fold axis normal to the base.I n IF, the iodine appears to be located at thecentre of a pentagon of five F atoms, the remaining F atoms being situatedabove and below the iodine, forming a pentagonal bipyramid.lg5The formation of some fluoride complexes of bivalent and tervalentmetals in aqueous solution has also been examined, and the existence of MnF,3-and CrF2+ dem0n~trated.l~~ Woolf and Greenwood lg7 have studied theformation of complex fluorides by addition and neutralisation reactions, inrelation to the conductivities of the non-aqueous liquid fluorides used assolvents.Tantalum trifluoride has been prepared by Emelkus and Gutmann lg8 bythe action of hydrogen fluoride on “tantalum hydride” at 250-300”:some pentafluoride was formed simultaneously. Niobium (columbium), its“ hydride ”, or the impure trichloride, on the other hand, gave only thepentafluoride under similar conditions.Pure tungsten dibromide wasprepared by reducing the tribromide in hydrogen. A number of non-metallic oxyfluorides, wiz., SOF,, SO,F,, H*SO,F, POF,, and SeOF,, havebeen prepared by Wiechert lg9 by treating the corresponding acid chlorideswith hydrogen fluoride without the use of a catalyst.The system NH,F-KF-H,O a t 25” is a simple one; no double salts areformed, the solid phases in equilibrium with the solutions being NH,F, KF,and KF,2H,0.200 Banks and Rudge ,01 have measured the density of liquidchlorine trifluoride under its own vapour pressure from -5” to +as0.Dunitz and Hedberg ,02 have re-investigated the structures of C10 andC10, by electron-diffraction measurements.lg2 J., 1950, 3678.lS1 J., 1950, 2907.lQ4 Ind. Eng. Chem., 1950, 42, 1383.lg5 Lord, Lynch, Schumb, and Slowinski, J. Amer. Chem. Soc., 1950,72, 522.lg6 Schaffer and Hammaker, ibid., p. 2575.leg 2. anorg. Chem., 1950, 261, 310.201 J., 1950, 191.lo3 J., 1949, 2206.lQ7 J . , 1950, 2200. lS8 J., 1950, 2115.Haendler and Jacke, J . Amer. Chem. SOC., 1950, 72, 4137.202 Dunitz and Hedberg, J . Amer. Chem. SOC., 1950,72, 3108122 INORGANIC CHEMISTRY.Lead monoxide is reported to give good yields as a, reducing agent in thepreparation of sodium chlorite from chlorine di~xide.~O~GufmannZM has shown that iodine monochloride also behaves as anionising solvent.A saturated solution of antimony pentachloride (1 mol. yo)possesses a conductivity which is greater than that of the solvent and believedto be due to the ionisation : I S b C l , ~ I + + SbC1,-. Addition of potas-sium chloride reduces the conductivity to a minimum at [KCl] = [SbCl,],followed by a rise caused by the formation of KICI,. Typical neutralisationreactions are observed.The course of the reaction between iodine monochloride and phenolshas been studied by Bennett and Sharp205 under different experimentalconditions : in the absence of a solvent the main reaction is chlorination;in solution it is iodination.This is accord with the increase of dipole momenton dissolution and the prediction that in a solvent of sufficiently highdielectric constant, iodine monochloride would undergo electrolytic dissoci-ation into I+ and C1-.,06A new determination of the acid dissociation constant of hypoiodous acidby Josien and Sourisseau 207 leads to the value 2 x for [H+][IO-]/[IOH]which is greater than the previous estimate by Furth 208 (2-3 x 10-ll) andabout the same as the basic dissociation constant obtained by Murrayzo9(3.2 x 10-l').Further evidence of the existence of iodine, and also to some extentbromine, as unipositive cations has been obtained by several workers.Hildebrand, Benesi, and Mower 210 have shown that the increase of solubilityof iodine in p-xylene and mesitylene, owing to complex formation betweenthe solvent and the iodine, is in good agreement with equilibrium data fromearlier spectroscopic evidence.211 Cromwell and Scott 212 have made are-determination of the equilibrium constant of the formation of the benzenecomplex, C6H6 + I, C,H6,1,, and estimate the heat of formation( A H ) of the complex to vary from -1317 -J= 50 to 1452 80 cals.accordingto the concentration of the solutions. Complex formation also betweeniodine monochloride and benzene and certain of its derivatives has beendemonstrated by Keefer and Andrews by measurement of the absorptionspectra.Several new salts of positive iodine, stabilised by pyridine or a-picoline,with halogenobenzoic acids have been prepared by Zingaro, VanderWerf,and Kleinbe~-g,~l~ and a very labile needle-shaped compound of iodine withdioxan has been obtained by Kortum-Seiler and K ~ r t U r n .~ ~ ~ The heat offormation in cyclohexane solution from dioxan and solid iodine is estimatedto be -4.3 kcals./mole.203 Holst, Ind. Eng. Chem., 1950, 42, 2359.205 J . , 1950, 1383.20' Bull. SOC. chim., 1950, 255.209 J . , 1925, 127, 885.211 Benesi and Hildebrand, ibid., 1949, 71, 2703.212 Ibid., 1950, 72, 3825.914 Ibid., p. 5341.*04 Research, 1950, 3, 337.206 Fairbrother, J., 1936, 847.208 2. Elektrochem., 1922, 28, 57.210 J . Amer. Chem. SOC., 1950, 72, 1017.213 ]bid., p .5170.215 2. Elektrochem., 1950, 54, 70FAIRBROTHER. 123Gonda-Hunwald, Graf, and Korosy 216 have obtained further evidenceof the existence of Br+ from some electrodialysis experiments, and Keefer andAndrews217 have shown that bromine displays a major absorption band inthe neighbourhood of 300 mp. in aromatic solvents similar to that shown byiodine.Pairbrother 2 l 8 has shown that iodine cyanide undergoes a polarisationin an electron-donor solvent which can stabilise the structure I+CN- similarto that shown by molecular iodine.219 It has been shown that the behaviourof the cyanohalogens as halides or as cyanides in their reactions (e.g., ClCNusually behaves as a chloride of cyanogen, and ICN as a cyanide of unipositiveiodine) is due to solvent-solute interaction in a basic solvent and to theenergy relationships involved in the particular reactions.A crystallographicexamination 220 of a series of solid solutions of iodine and bromine over therange 0-38 rnol.% of bromine suggests that the crystals contain orientedIBr molecules.Potassium permanganate and the oxides of manganese, when gentlyheated in fluorine, yield MnF, contaminated with a little MnF, but no oxy-fluoride. On the other hand, as Aynsley, Peacock, and Robinson 221 haveshown, rhenium dioxide and potassium per-rhenate by the same treatmentgive the two new oxyfluorides ReOF, and Re02F,.A spectrophotometric study of Mn(m), which is present in the brown-redsolution obtained when &(II) in hydrochloric acid is oxidised or when MnO,or KMnO, are dissolved in hydrochloric acid has been made by Ibers andDavidson.222 A blue sodium pentamanganate has been prepared by Levi.223Accurately characterised specimens of MnO, &(OH),, Mn,O,, a-Mn203,y-Mi20,, MnO-OH, and MnO, have been prepared, and their chemical and.physical properties Compounds of manganous chloride withpyridine have been studied by F ~ f e .~ ~ ~The thermal explosion in air and the decomposition of ammoniumpermanganate have been studied by Bircumshaw and Tayler.2Z6The technique of ion-exchange separation has been extended to the separ-ation of a number of seventh-group anions.227 Halide ions may be separatedfrom one another as well as TcO,- from Re0,- : permanganate cannot beseparated, as it reacts with the resin.The absorption spectra of rhenium(m), rhenium(v), and variousrhenium(rv) species and their reactivities towards common oxidising andreducing agents have been investigated by Maun and Davidson.228 Theyellow- brown solution of Re@) in hydrochloric acid or reduced per-rhenateis much more easily oxidised than the light green hexachlororhenate.21u Nature, 1950, 166, 68.217 J .Amer. Chem. SOC., 1950, 72, 4677.21g Fairbrother, J., 1948, 1051.220 Heavens and Cheesman, Acta Cryst., 1950,3, 197.222 J . Amer. Chem. SOC., 1950,72,4744.124 Moore, Ellis, and Selwood, J. Amer. Chem. SOC., 1950, 72, 856.225 J., 1950, 790.227 Attenbury and Boyd, J. Amer. Chem. SOC., 1950,72,4804.Z * 8 J., 1950, 180.221 J., 1950, 1622.223 Gazzetta, 1949, 79, 630.aa6 J ., 1950, 3674.228 Ibid., p. 2254124 INORGANIC CHEMISTRY.Aynsley, Peacock, and Robinson 221 have confirmed the observations ofearlier workers that the highest fluoride of rhenium that can be prepared isthe hexafluoride.Group VIII.-The formation of a thin film of y-Fe203 on iron immersedin 0-1N-sodium hydroxide and its relation to the inhibition of corrosion insodium hydroxide solution have been studied by Mayne, Menter, andP r y ~ r . ~ ~ ~ The thermal dissociation of the anhydrous ferric halides has beeninvestigated by Gregory and T h a ~ k r e y , ~ ~ ~ and by Kangro and Petersen.231Feitknecht and Keller 232 have shown that the dark green oxidation productof Fe(OH), does not correspond to a single chemical compound.Crystals of nickel ferrite, NiFe204, of up to 2 mm.side, have been preparedby slowly cooling a mixture of Fe203 and NiO dissolved in borax glass from1330°.233Nyholm has prepared and studied complexes of ferrous and ferrichalides,234 bivalent and tervalentand tervalent rhodium 237 halides with the di(tertiary arsine) chelate groupo-phen ylene bisdime th ylarsine.Azidopentamminocobalt (1x1) complexes have been prepared by Linhardand Flygare 238 by the substitution of N3- for H20 in [Co(NH3),,H20j3+ orby the synthesis of [Co(NH3),N3I2+ from Co2+, N3-, NH,, and NH,+ byatmospheric oxidation.Measurements of the diamagnetic susceptibilities of the cobalt(II1)complex of 3 : 6-dithia-1 : 8-di(salicy1ideneamino)octane and of thecorresponding p-hydroxy- a-naphthylidene complex 239 have confirmed theview, deduced earlier2& from a resolution of the former compound intooptical antipodes, that these chelate molecules can attach themselves at sixpoints to the central metal atom.The corresponding ferrous complex of thefirst-named chelating agent is paramagnetic, with a moment correspondingto two unpaired electrons, which has been interpreted as meaning that theFe(I1) is bound to the sexadentate in a similar manner to that in which Fe(II1)is bound in ferriheme hydroxide2,1 by four bonds resonating among sixpositions.A comparison of the absorption spectra of geometric isomers of cobalt(111)complex compounds of known configuration indicates that absorptionspectra may be used to establish the configuration in other cases wheneverthe positions of the absorption maxima of the second and third bands aresignificantly shifted.242Bassett and Henshall 243 have examined the systems Na,SO,-CoSO,-H20 and have shown that the double salt Na2S04,CoS0,,H20 has a definite229 J ., 1950, 3229.231 2. anorg. Chem., 1950, 261, 157.233 Matthias and Remeika, Phys. Review, 1950, 79,391.234 J., 1950, 851.237 J . , 1950, 857.239 Dwyer, Lions, and Mellor, J. Amer. Chem. SOC., 1950, 72, 5037.240 Dwyer and Lions, ibid., 1947, 89, 2917.241 Pauling and Coryell, ibid., 1937, 59, 633.242 Basolo, ibid., 1950, 72, 4393.bivalent and tervalent230 J . Amer. Chem. SOC., 1950,72,3176.233 Ibid., 1950, 262, 61.238 J . , 1950, 2061; Nature, 1950, 165, 154.238 2. anorg. Chem., 1950,262, 328.235 J., 1950, 2071.24s J., 1950, 1970FAIRBROTHER. 125composition and that isomorphous replacement of Co(H,O), by Na,(H,O),does not occur.that theonly solid phases in equilibrium with their solutions are the hexa- and mono-hydrates.Hume and Kolthoff 245 have shown by polarographic studies of the re-duction of the Ni(CN,)" ion that the dissociation constant of this ion is ofthe order of and that the solid nickel cyanide in equilibrium with asaturated solution is actually Ni[Ni(CN),]. Nickel complex compoundscontaining triethylenetetramine have been studied by Jonassen and Douglas.246Spectrophotometric measurements indicate the presence of [Ni trien],+and [Ni2trien,]4+ in solution. The same complexing agent has beenused 247 to prepare the compounds [Pt trien][PtCl,], [Pd trien][PdCl,], and[trien H,(PtC14)2],2H,0.The lower oxidation states of ruthenium, RU(III) and Ru(Iv), have beenstudied in acid perchlorate solution by Wehner and hind ma^^.^,^ Drychlorine a t 700" converts metallic ruthenium completely into the trichloride,which sublimes in two different polymorphic ~ a r i e t i e s . ~ ~The complex compounds PtBr2,3AsR, and PdBr2,3AsR, (where AsR,is methyldiphenylarsine) have been isolated and studied by N y h ~ l r n . ~ ~ Conductivity measurements in acetone of the palladium compound supportthe view that they are triarsine salts, e.g., [Pd(AsR,),Br]+Br-.UbbelohdeZ5l has studied the effect of adsorption of hydrogen on thewetting of palladium by mercury and has discussed the results in terms ofpseudo-metallic bonding in hydrides.Burstall, Dwyer, and Gyarfas 252 have prepared [ OsC12,2dipy]C1,3H,0and [ Os,3dipy]Cl,,GH,O and the corresponding bromides by the reactionbetween potassium osmichloride K,[ OsCl,] or ammonium osmibromide(NH,J,[ OsBr,] and excess of 2 : 2'-dipyridyl. Optically active enantiomorphsof' the antimony1 tartrate and iodide of the bivalent complex have beenobtained.Several new platinum(I1) and platinum(1v) complex compounds havebeen prepared, vix., C,H4PtC1 and (C,H4PtC12),,253 trans-[ Pt en2C12]C1,,254and a series of octahedral complexes of quadrivalent platinum with thegeneral formula Pt( Hal),2AsR3, various halides and tertiary arsines beingChatt 256 has studied the oxidation of the non-ionic simple com-plexes c~~-[(C,H~)~P],P~C~, and trans-[ (C,H7),P],PtC12 and the bridgedcomplex [ (C,H, ),P,,PtCl,], to the corresponding platinic compounds, and hasdiscussed the role of the d orbitals of the platinum atom in stabilising theplatinous bridge. F. FALRBROTHER.A study of the system NiSe0,-H,Se04-H20 at 30" has shown844 Rohren and Froning, J . Amer. Chem. SOC., 1950,72,4656.246 Ibid., 1949, 71, 4094.248 Ibid., 1950, 72, 3911.250 J . , 1950, 848.253 Chatt and Wilkins, Nature, 1950,165, 859.p54 Basolo and Tam, J . Amer. Chem.. SOC., 1950,72, 2433.255 Nyholm, J . , 1950, 843.245 Ibid., p. 4423.247 Jonassen and Cull, ibid., p. 4097.2*9 Hill and Beamish, ibid., p. 4865.251 J . , 1950, 1143. J . , 1950, 953.256 J . , 1950, 2301