J. CHEM. SOC. DALTON TRANS. 1992 1761Chromium-(11) and -(HI) Complexes containing Hydrazines orHydrazinecarboxylates as Ligands tDennis A. Edwards,*na David Thompsetta and John M. Bellerby"a School of Chemistry, University of Bath, Bath BA2 7AY, UKFaculty of Military Science, Cranfield Institute of Technology, Shrivenham, Swindon SN6 BLA, UKThe complexes [{CrX,(R2R1NNH,),},] (X = CI, R' = H, Me or Ph, R2 = H; R1 = R2 = Me; X = Br, R1 = Hor Ph, R2 = H) have been prepared and spectroscopic data acquired which suggest that themonomethylhydrazine complex is analogous to the well known hydrazine complexes with bridgingNH,NHR ( R = H or Me) ligands and terminal halide ligands, whereas the N,N-dimethyl- and phenyl-hydrazine complexes involve unidentate hydrazine and bridging halide ligands.The quadruple metal-containing bridging NH,NR1R2 ligands and [Cr,(O,CMe),( PhNHNH,),] containing unidentateN H,N H Ph ligands have also been prepared and characterised. Chromium(i1) hydrazinecarboxylate,[{Cr(O,CNHNH,),( H,O)},] has been prepared by either cleavage of the metal-metal bonds of[Cr,(O,CMe),LJ ( L = H,O or jNzH4) or ligand-displacement reactions of mononuclear chromium(ii)complexes. Its infrared spectrum and that of the fully deuteriated analogue have been recorded andvibrational assignments proposed. Oxidation of [{Cr(O,CNHNH,),( H,O)},] or other chromium(1i)species in aqueous [N,H,] [O,CNHNH,] leads to the chromium(l1i) complex [Cr(O,CNHNH,),]-2H,O.The substituted hydrazinecarboxylates [Cr{O,CN (Me) N H,},]=H,O and [Cr,(O,CN H N H Ph),( MeOH),]have also been isolated, the latter probably containing carboxylate-O,O' groups bridging a metal-metalbonded Cr, unit in the manner well established for other carboxylate anions.metal bonded complexes [{Cr,(02CMe)4(p-R2R1NNH,)},] (R1 = H or Me, R2 = H; R' = R2 = M elAnhydrous hydrazine, monomethylhydrazine and N,N-di-methylhydrazine have been used as propellants and fuels in, forexample, launch vehicles for spacecraft and the in-orbit controlof satellites.A problem related to this application is theincreased rate of propellant decomposition in metal storagetanks brought about by carbon dioxide contamination.' Inthe case of hydrazine, this enhanced decomposition has beentraced in part to a reaction producing hydrazinecarboxylicacid,, NH,NHC02H, which exists in excess of hydrazineas [N2H,][02CNHNH2]. In contact with steels, for example,metal ions are leached from the surfaces into the bulkpropellant, the resulting metal complexes (most likely hydra-zinecarboxylates) acting as homogeneous catalysts for hydra-zine decomposition.Recent work3 has shown chromium to bean active metal in this regard and so we have carried out somestudies aimed at establishing the nature of the species formed inseveral hydrazines.The reported chemistry of hydrazine complexes of chromiumis not extensive. The only well characterised complexes are[{CrX,(p-N,H,),],] (X = C1, Br or I),,*' an unstable, impure,fluoro-analogue also having been reported.6 Severalchromium(II1) complexes e.g.[CrC13(N2H,), ]-2H20,' [Cr-2(C20,)3]~xN2H,-yH20 (x = 2 or 6, y = 0; x = y = 4; x = 7,y = 1)8 and [Cr(N,H,)2(C104)2]C10.+g are mentioned in theliterature, but the authenticity of these complexes may be indoubt having been prepared using hydrazine hydrate which actsas a strong base producing hydroxochromium(rI1) species.Indeed, the removal of chromate from waste waters as'chromium(rI1) hydroxide' is achieved commercially usingaqueous hydrazine. The complexes [CrF3(N,H4),I6 and[CrCl,(MeHNNHMe),]" are less open to doubt, having beenobtained from reactions employing anhydrous hydrazines.We report here some of our studies on the chemistry ofchromium-(II) and -(HI) with the hydrazines NH,NRIRZ (R' =t Non-SIunit employed: eV = 1.60 x 1O-l' J.H, Me or Ph, R2 = H; R' = R2 = Me) and with hydrazine-,monomethylhydrazine- and phenylhydrazine-carboxylic acids.A preliminary account of some chromium(1r)-hydrazinereactions has been published.''ExperimentalStarting Materials.-The chromium(r~) compounds CrC1,-4H20,' CrBr,06H~0,'~ [Cr,(O,CMe),(H,O),] ' and[{ CrC12(NCMe)2},]14 were prepared by literature methods.Anhydrous [Cr2(02CMe),] was prepared by thermal de-hydration of the hydrate in uucuo. Anhydrous hydrazine wasprovided by Royal Ordnance plc, Westcott and conformedto the US monopropellant specification (1% H 2 0 maximum,0.003% CO, maximum). If necessary, it was distilled fromcalcium hydride under nitrogen before use.Monomethyl- andN,N-dimethyl-hydrazines were distilled from potassiumhydroxide under nitrogen before use.[2H6}Hydrazine hydrate (minimum 98% D) was a commer-cial product (MSD isotopes).Hydrazinecarboxylic acid wasprepared by slowly passing CO, gas through an aqueoussolution (30 cm3) of hydrazine (30 cm3) at 0 "C for 12 h. Thecolourless solid was filtered off under N, and purified by stirringin anhydrous methanol, followed by filtration under nitrogen,washing with diethyl ether and drying in uacuo. Yield 92%(Found: C, 15.6; H, 5.40; N, 36.7. Calc. for CH,N202: C, 15.8;H, 5.30; N, 36.8%). [2H,]Hydrazinecarboxylic acid wasobtained by saturating a solution of N,D,*D,O (1.00 cm3) inD20 (1.00 cm3) with C 0 2 until a colourless precipitate beganto form. C2H4]Methanol (4.00 cm3) was layered on to thissolution and after 72 h the solid that had formed was filtered off,washed with diethyl ether and dried in ULICUO.Yield 1.31 g,9 1 .Ox. Methylhydrazinecarboxylic acid was prepared in asimilar manner,'' C 0 2 being passed through a solution ofMeNHNH, (2.30 g, 50 mmol) in ethanol (50 cm3) for 3 h. Thecolourless, highly hygroscopic, solid was filtered off undernitrogen and dried in vacuo (Found: C, 26.1; H, 6.70; N, 30.91762 J. CHEM. SOC. DALTON TRANS. 1992Calc. for C2H6N,02: C 26.7; H, 6.70; N, 31.1%). Phenyl-hydrazinecarboxylic acid was prepared as PhNHNHC0,H.PhNHNH, l6 (or [PhNHNH,][PhNHNHCO,]), by passingC02 through a solution of PhNHNH, (10 cm3, 102 mmol) inmethanol (150 cm’) for 2 h. The colourless precipitate wasfiltered off, washed with diethyl ether and dried in vacuo(Found: C, 60.5; H, 6.30; N, 21.9.Calc. for C13H16N402: C,60.0; H, 6.20 N, 21.5%).Solvents were thoroughly dried by standard methods(MeOH and EtOH, reflux over Mg under N,; MeCN, refluxover P205 under N,) and purged with N2 before use.Physical Measurements.-Infrared spectra of Nujol mullswere recorded over the range 4000-200 cm-’ using a Perkin-Elmer 597 spectrophotometer. Magnetic moment determin-ations were carried out at room temperature by the Gouymethod using [Ni(en),]S203 (en = ethane-1,2-diamine) ascalibran t.Electronic spectra of mulls were recorded using a Perkin-Elmer 330 spectrophotometer fitted with a diffuse reflectanceattachment. Differential thermal analysis (DTA) and thermalgravimetry (TG) experiments were carried out on a StantonRedcroft STA 780 thermal analyser, the powdered samplesbeing heated at 10°C m i d in air or in flowing argon, thereference being a-alumina.The X-ray photoelectron spectrawere recorded using a VG spectrometer, samples being loadedas powders dusted on to double-sided adhesive tape. The Al-Kal,, line was used as the excitation source and surface-charging effects reduced by use of an electron ‘floodgun’.Chromium (II)-Hydrazine Complexes.--(a) [ ( C rX , (N , -H,),),] (X = C1 or Br). These established complexes wereprepared by treating ethanol solutions of CrC12-4H20 orCrBr2-6H,0 with a solution of anhydrous hydrazine inethanol. The pale lilac precipitates were filtered off under N,,washed with ethanol and diethyl ether and dried in a stream ofN,.Their identity was confirmed by infrared ~pectroscopy.~The chloride was also isolated from a reaction between amethanol solution of [(CrCl,(NCMe),},] (1.50 g, 7.3 mmol)and hydrazinecarboxylic acid (1.11 g, 14.6 mmol) under N,.The bromide was also obtained when an ethanol solution ofCrBr,-6H20 (1.44 g, 4.5 mmol) was added to hydrazinecarb-oxylic acid (3.04 g, 40.0 mmol) under N2 at -10 “C. Themixture was stirred for 1 h and then warmed to roomtemperature.(6) [(CrCl,(MeNHNH,),),]. A solution of MeNHNH,(0.53 cm3, 10.01 mmol) in anhydrous MeCN (20 cm’) wasadded under N2 to a suspension of anhydrous CrCl, (0.59 g,4.79 mmol) in anhydrous MeCN (25 an3). The light blueproduct which formed rapidly was filtered off under N,, washedwith MeCN and diethyl ether and dried in a stream of N2(Found: C, 10.9; H, 5.20; N, 25.5.Calc. for C2Hl2CI2CrN4: C,11.2; H, 5.65; N, 26.0%).(c) [(CrC12(Me2NNH2)2}n]. This complex was prepared asa pale blue-green solid from acetonitrile solution using themethod outlined in (b). The reactants were CrCl, (0.71 g, 5.74mmol) and Me,NNH, (0.92 cm3, 12.03 mmol) (Found: C, 19.5;23.0%).(d) [(CrX,(PhNHNH,),),] (X = Cl or Br). A methanolsolution of CrC12*4H,0 (0.975 g, 5.0 mmol) or CrBr2.6H,O(1.60 g, 5.0 mmol) was added to a solution of PhNHNH, (1.02cm3, 10.4 mmol) in diethyl ether (30 cm3). Pale blue productsprecipitated immediately and were filtered off under N,,washed with diethyl ether and dried in a stream of Nz (Found4.75; N, 16.5%).(e) [ (Cr2(O2CMe),(p-N2H,)},J. To a degassed solution of[Cr,(O,CMe),(H,O),] (2.04 g, 5.42 mmol) in methanol (30cm3) was added a degassed solution of anhydrous hydrazine(5.0 cm3, 15.78 mmol) in methanol (10 cm3) under N,.AfterH, 6.85; N, 22.7. Cak. for C,H,6C12CrN,: C, 19.8; H, 6.65; N,C, 42.2; H, 4.85; N, 16.4. Calc. for C12H16C12CrN4: C, 42.5; H,stirring under N, at room temperature for 12 h, the orangeproduct was filtered off, washed with methanol and diethylether and dried in a stream of N2 (Found: C, 25.3; H, 4.25; Cr,27.9; N, 7.55%). Selected IR bands (cm-I): 1595vs (br),v,,,(CO,) + 6(NH,); 1453vs, v,,,(COz); 1349m, o(NH,);v(CC); 679s, 6(CO,); and 621m, x(C0,).(f) [(Cr,(O,CMe),(p-MeNHNH,)),,]. A suspension ofanhydrous [Cr,(O,CMe),] (1.15 g, 3.37 mmol) in ethanol (25cm3) under N, was treated with a solution of MeNHNH, (0.54cm3, 10.2 mmol) in ethanol (15 cm3).After stirring for 1 h atroom temperature the red-orange product was filtered offunder N,, washed with diethyl ether and dried in a stream of N,(Found: C, 28.0 H, 4.95; N, 6.95. Calc. for C9Hl,Cr2N,0,:C, 27.9; H, 4.70; N, 7.25%).(g) [(Cr,(02CMe)4(p-Me2NNH,)),J. A suspension ofanhydrous [Cr,(O,CMe),] (1.18 g, 3.45 mmol) in ethanol (25cm3) under N, was treated with a solution of Me,NNH, (0.84cm3, 11.0 mmol) in ethanol (10 cm’). The product was filteredoff under N,, washed with diethyl ether and dried in a stream ofN2 (Found: C, 30.0; H, 5.60; N, 7.35.Calc. for CloH&r~N208:C, 30.0; H, 5.05; N, 7.00%).(h) [Cr2(OzCMe),(PhNHNH,),]. To a suspension of[Cr,(O,CMe),(H,O),] (1.20 g, 3.20 mmol) in ethanol (20 cm3)was added a solution of PhNHNH, (0.94 cm’, 9.6 mmol) inethanol (10 cm’). The mixture was stirred under N, at roomtemperature for 3 h before filtration of the pale orange productwhich was then washed with ethanol and diethyl ether beforedrying in a stream of N, (Found: C, 43.1; H, 5.35; N, 10.3. Calc.for C20H2SCr2N408: C, 43.2; H, 5.05; N, 10.1%).27.6; N, 7.55. Cak. for C8H16Cr,N,Os: c , 25.8; H, 4.35; Cr,112% pt(NH2); 1049~, 1032~, p,(CH,); 9 6 1 ~ , v(NN); 9 2 6 ~ ,Chromium-Hydrazinecarboxylate Complexes.--(a) Bis-(hydrazinecarboxylato)chromium(n) monohydrate. This com-pound can be prepared by treating a variety of chromium(@sources e.g.[Cr2(02CMe)4(H20)2], [{ Cr2(02CMe)4(p-N2-H4)},,], CrCl,-4H20, CrBr,-6H,O and [(CrXZ(p-N2H4),},,](X = C1 or Br) with either hydrazinecarboxylic acid orhydrazinium hydrazinecarboxylate. Two typical preparationsare given below.(i) Passage of CO, through an aqueous solution (5.0 cm3) ofN,H,-H20 (0.78 cm3, 16.0 mmol) for 2 h generated a solutioncontaining [N2H,][O2CNHNH2] (8.0 mmol). Addition of anaqueous solution (1.50 cm3) of CrC1, (0.139 g, 1.13 mmol)under N, caused rapid precipitation of a lilac product whichwas stirred under N2 for 15 min before filtration, washingsuccessively with water, methanol and diethyl ether and dryingunder N,. Yield: 0.152 g (61.0%, based on CrCl, used) (Found:C, 11.2; H, 3.45; Cr, 23.8; N, 25.8.Calc. for C$&Cr&05: C,10.9; H, 3.65; Cr, 23.6; N, 25.5%).(ii) A mixture of [Cr2(02CMe),(H20),] (1.00 g, 2.7 mmol)and hydrazinecarboxylic acid (1.50 g, 19.8 mmol) was heatedin refluxing methanol (50 cm3) under N, for 2 h. After cooling,the lilac product was filtered off under N,, washed withmethanol and diethyl ether and dried in a stream of N,. Thisreaction must be carried out at reflux temperature as it wasfound to be extremely slow at lower temperatures. Thus, largelyunreacted hydrazinecarboxylic acid was recovered after 4 weeksat room temperature.Fully deuteriated [Cr(02CNHNH2),(H20)] was preparedby method (i) above using N2D4-D20 (0.78 cm3, 16.0 mmol) inD20 (3.5 cm’) to generate ~,D,][O,CNDND,] beforereaction with CrCl, in D20.After filtration, the pale lilacproduct was washed with D,O, CH,OD and diethyl ether.Yield: 0.13 g (55.2%, based on CrCl, used).(b) Tris( h ydrazinecarboxylat o)chromium(m) dihydra te. A1 -though this complex can be prepared by oxidation of[Cr(O,CNHNH,),(H,O)], CrC1,.4H2O, CrBr,m6Hz0 or[{CrX,(p-N,H,),),J (X = C1 or Br) in the presence of[N,H,][O,CNHNH,], it is most easily prepared from[CrC12(H,0)4]C1~2H,01. CHEM. SOC. DALTON TRANS. 1992 1763Table 1 Magnetic and spectroscopic data for [{CrC12(R2R 'NNH,),},] and related complexesIR/cm-'C o m p 1 ex pelf, r.t. Reflectance spectrum, h/nm" v(CrN) v(CrX) Ref.4.85 5704.80 5704.59 5944.85 6624.59 7044.61 d4.79 7524.84 6854.57 714427m, 346m b c420m, 338m b c426w, 359w b C441 (sh), 434s 302s c390w 323s C393w b c320vs 14219m 303s 1718Broad band assigned to overlapping 5B1.s - 5B2g, 5E,.Not observed above 200 cm-'. This work. * Spectrum could not be recordedreproducibly because of rapid surface oxidation.A solution of anhydrous hydrazine (17.0 an3, 531 mmol) inwater (75 cm3) was saturated with CO, for 1 h while beingheated to 80 "C. Then with the CO, flow being maintained, asolution of [CrC12(H,0)4]CI*2H20 (1.60 g, 6.0 mmol) in water(5 cm3) was added dropwise. The red solution was kept at 80 "Cfor 24 h and then cooled to produce a pink microcrystallineprecipitate. This was filtered off, washed with water, ethanoland diethyl ether and dried in uacuo.Yield: 1.02 g, 54% based onCr (Found: C, 11.7; H, 4.20; Cr, 16.3; N, 26.8. Calc. forC3H13CrN,0,: C, 11.5; H, 4.20; Cr, 16.6; N, 26.8%).(c) Tris( me t h y ih y dr azinecarboxy la to) ch rom ium( 11 I) hydrate.( i ) A solution of MeNHNH, (2.30 g, 50 mmol) in ethanol (90cm3) was treated with CO, for 2 h whilst kept at 60°C. Asolution of [CrCl2(H,O),]CI~2H20 (1.33 g, 5.0 mmol) inethanol (40 cm3) was then added dropwise. The resulting pinkmicrocrystalline product was filtered off, washed with ethanoland diethyl ether and dried in uacuo (Found: C, 21.0; H, 4.65;15.4; N, 24.9%).( i i ) This complex could also be prepared by oxidation ofchromium(I1). A solution of MeNHNH, (1.15 g, 25 mmol) inmethanol (30 cm3) was treated with CO, for 2 h and thencooled under N, to - 10 "C.A solution of CrCI,-4H20 (0.195 g,1.0 mmol) in methanol (15 cm3) was added dropwise initiallyto give a purple solution which on standing slowly produced apink precipitate of [Cr(O2CN(Me)NH,},]-H2O.chromium(I1). A solution of CrC1,-4H20 (0.975 g, 5.0 mmol) inmethanol (15 cm3) was added under N, to a suspension ofPhNHNHC02H-PhNHNH2'6 (3.0 g, 11.5 mmol) in methanol(30 cm3). A salmon-pink product precipitated rapidly, wasfiltered off under N,, washed with methanol and diethyl etherand dried in a stream of N, (Found: C, 46.5; H, 4.70; N, 14.8.Calc. for C30H36Cr2NS010: C, 46.6; H, 4.70; N, 14.5%).Cr, 15.3; N, 24.3. CdC. for C6H,,CrN60,: c , 21.4; H, 5.10; Cr,(d)Bis(methanol)tetrakis( pheny1hydrazinecarboxyiato)di-Results and Discussion(a) Compiexes of Chromium(I1) Halides with Hydrazines.-The insoluble, polymeric, complexes [{CrX,(R2R'NNH,),},](X = C1, R' = H, Me or Ph, R2 = H; R' = R2 = Me; X =Br, R' = H or Ph, R2 = H) have been prepared andcharacterised by solid-state methods (Table 1). The substitutedhydrazine complexes are especially air-sensitive.The resultsobtained suggest that two structural types are formeddependent on the nature of the hydrazine ligands. Thepreviously reported4 complexes [(CrX,(p-N,H,),},] (X = C1or Br) have been prepared by both established and new routes,e.g. displacement of the acetonitrile ligands of [{ Cr(pCI),(NCMe),},] l4 by hydrazine or reaction of anhydrousCrCI, with anhydrous hydrazine in methanol. Infrared andvisible spectroscopic data4 indicate the complexes to beoctahedral with bridging hydrazine ligands and terminal axial,halides as established crystallographically for [{ MC12(p-N2H4),},] (M = Mn or Zn).The spectroscopic resultsobtained in this work, which confirm earlier result^,^ supportthe formation of tetragonally distorted octahedral, magneticallydilute, high-spin chromium(I1) complexes. The infrared spectracontain two bands in the 430-330 cm-' region, assignable tov(Cr-N) modes (2e, assuming CrN4X, units of D,, symmetry),but no v(Cr-X) stretches have been found, the weakness of theaxial Cr-X bonds probably resulting in such modes appearingbelow 200 cm-'. These results are in line with infrared andRaman data reported" for [{MX,(p-N,H,),},] (M = Mn, Coor Zn; X = C1 or Br) which show that bands with high v(MN)character are found in the 400-340 cm-' region whereas bandswith high v(MX) character appear at exceptionally lowfrequencies for terminal metal-halide stretches e.g.v(ZnC1) at285 and 149 cm-'. Additional Jahn-Teller-induced distortioncan weaken the chromium-halide interactions still further.Other infrared bands e.g. v(NN) at ca. 965 cm-' are in the rangesuggested for bridging hydrazine l i g a n d ~ . ~ * ~ ~ * ~ 'No reactions of chromium-(II) or -(III) compounds withMeNHNH,, Me,NNH, or PhNHNH, appear to have beenreported previously. Reactions of hydrated chromium(I1)chloride or bromide with methanol solutions of MeNHNH,afforded only impure products containing hydroxo groups.However, reaction of [(CrCl,(NCMe),},] with MeNHNH,in anhydrous acetonitrile produced [(CrCl,(MeNHNH,),},].The reflectance spectrum, magnetic moment and infraredv(Cr-N) stretching frequencies (Table 1) are all very similar tothose of [(CrCl2(p-N2H4),},], so a structure involving bridgingMeNHNH, ligands and weak, axial, Cr-CI bonds is proposed.This structural type has been suggested 22 for analogues of Fe",Co" and Nil'.For Me,NNH, the inductive effect of the methyl groupswill enhance the basicity of the adjacent nitrogen atom, soelectronic considerations favour MtNMe2NH, co-ordination.However, steric interactions involving the methyl groupsmay result in the NH, nitrogen being the favoured donor site,as in [RUH(~~-CO~)(M~,NNH~)~][PFJ~~ (cod = cycloocta-1,5-diene) and the boroxin [(EtBO),(Me,NNH,)].24 Bridgingco-ordination has also been established2' as in [{ RuC1(H)(q4-As for MeNHNH,, the reactions of hydrated chromium(I1)chloride or bromide with Me,NNH, in ethanol resulted inhydroxide formation, however [{ CrCl,(Me,NNH,),},] couldbe obtained from [{CrCl,(NCMe),},] in anhydrous aceto-nitrile.The room-temperature magnetic moment is similar toothers listed in Table 1 but the maximum of the broad peak inthe visible reflectance spectrum is shifted to lower energy bynearly 100 nm relative to that of [{CrCI2(p-N2H4),},,], beingcloser to those of [{Cr(pCI),L,},,] species (L = NH,, C5H5Nor MeCN) which possess CrC14N2, rather than CrC12N4,tetragonally distorted octahedral arrangements. Again, unlikecod)} 2 (I.L-Me,"H,)l1764 J.CHEM. SOC. DALTON TRANS. 1992Table 2 Magnetic and spectroscopic data for Cr2(0,CMe), productsI R/cm-Reflectance spectrum, h/nm v(CrN) v(Cr0)478s (br), 360s (sh), 333vs 469m 402s, 384m (sh)513s (br), 361s (br), 350s (br),336s (br)476m (br), 349m (sh), 331s 475w 398s476m (br), 346m (sh), 3 2 4 v ~ , ~ 427w 401ms2 6 5 s (br)478m (br), 360s (sh), 333vs486s (br), 347s (sh), 326vs475408s407m, 384m[{CrCl,(p-RNHNH,),},] (R = H or Me), the infraredspectrum of [{ CrCl,(Me,NNH,),},] contains a strong bandassignable to a v(CrC1) stretch at a very similar frequency tothose of [{Cr(p-Cl),L,},] (L = CSH,N or MeCN). Therefore,it is proposed that in [{CrC12(Me2NNH2)2}n] the Me2NNH,ligands are unidentate and most likely bonded to the metal viathe NH,-nitrogen atoms.Octahedral [{ MX2(Me2NNH2)2)n](M = Fe, X = C1 or Br;,, M = Ni, X = C126) complexes areknown but the nature of the bonding of the halide andMe,NNH, ligands was not resolved.Phenylhydrazine also reacts with first-row transition-metalhalides producing [{ MX,(PhNHNH,),},] complexes (M =Mn, Fe, Co, Ni or Zn; X = C1 or Br),27 but the bonding modesof the ligands remain unestablished. The hydrazine ligand isunidentate in both [V(salen)(PhNHNH,)]I [H,salen = N,N'-ethylene bis(salicy1ideneimine)l and [ { M o(N, Ph),( OMe), (Ph-NHNH,)},], bonding via the more basic NH,-nitrogen atom."Reactions of [Cr(H,O),Cl,] and [Cr(H,O),]Br, withPhNHNH, in ethanol-diethyl ether afford the very air-sensitivecomplexes [{CrX,(PhNHNH,),),] (X =C1 or Br).The severeair-sensitivity of the bromide did not allow reproducibleanalyses or good reflectance spectra to be obtained. Themagnetic moment values, although somewhat reduced from thespin-only value, are indicative of high-spin mononuclearchromium(II), the actual values obtained perhaps resulting froma little oxidation on handling rather than additional electronicinteractions through bridging groups as in [{ Cr(p-Cl)2-(NH3),},].'* The similarity of the reflectance spectrum and theinfrared v(CrC1) frequency of [{ CrC12(PhNHNH2),},] tothose of [{Cr(p-Cl),L,},] (L = MeCN, CSHsN or NH,)suggests that the chloride ligands are bridging and thephenylhydrazine ligands are unidentate.The infrared spectra ofthe chloride and bromide complexes are virtually identicalbetween 3500 and 400 cm-' suggesting an analogous structurefor the bromide.In summary, it is suggested that, probably as a reflection ofcompetition between steric and electronic influences, reactionswith N2H4 or MeNHNH, give hydrazine-bridged complexeswhile use of PhNHNH, or Me2NNH, results in halide-bridgedcomplexes.(b) Complexes of Chromium(1r) Acetate with Hydrazines.-The co-ordination behaviour of N2H4, MeNHNH, andPhNHNH, in the complexes formed with dinuclear chromium-(11) acetate parallels that found in the dichloride complexes.However, Me2NNH2 displays an alternative co-ordinationmode.Reaction of [Cr,(O,CMe),(H,O),] with anhydrous hyd-razine in ethanol gives the sparingly soluble, air-sensitive[{ Cr,(O,CMe),(p-N,H,)},].The reflectance spectrum and thevery weak paramagnetism of the product (Table 2) indicateretention of a quadruple metal-metal bonded structure, theaxial aqua ligands being replaced by bridging hydrazine ligandsto generate a polymer of alternating Cr,(02CMe), and N2H,units, as in [(Cr,(O,CMe),(p-pyz)},] (pyz = pyra~ine).~' Theinfrared spectrum (see Experimental section) contains bandsassignable to both bridging acetate and bridging hydrazineligands and a band at 469 cm-' has been assigned to v(CrN) inagreement with that at 475 cm-' for [Cr,(02CMe),(NH3),].31Reaction of [Cr,(O,CMe),(H,O),] with MeNHNH, inethanol gave an impure product which nevertheless analysedclose to that required for [{Cr,(O,CMe),(p-MeNHNH,)},].However, replacing the hydrate by the bis(ethano1) adduct gavethe pure product.The spectroscopic and magnetic properties,(Table 2) are closely similar to those of the hydrazine product.The reaction between [Cr,(O,CMe),(H,O),] and Ph-NHNH2 in ethanol gave air-sensitive [Cr,(O,CMe),-(PhNHNH,),], the spectroscopic and magnetic properties ofwhich strongly suggest retention of the metal-metal bonds withaxial unidentate phenylhydrazine ligands. An additional bandfor this complex in the reflectance spectrum can be assigned to an - n* transition of the ligated phenylhydrazine.Finally, reaction of [Cr,(O,CMe),(EtOH),] with Me,NNH,in ethanol afforded [{Cr2(02CMe),(p-Me2NNH2)},], thestoichiometry implying the presence of bridging Me,NNH,ligands, a different co-ordination mode to that suggested for[{Cr(p-C1),(Me,NNH2),},]. There is some indication that theCr-NMe, bond is easily cleaved.Thus, the UV/VIS solid-stateand methanol solution spectra are rather different, the solutionspectrum being closer to that of [Cr,(O,CMe),(H,O),],suggesting replacement of an axial nitrogen donor by an oxygendonor. Also, addition of ethanol to [{Cr,(O,CMe),(p-Me,-NNH,)},] gives a product which has an infrared spectrumcontaining bands assignable to both Me,NNH, and ethanolligands. No further characterisation of this product wasattempted.(c) Chromium Hydrazinecarboxylate Complexes.-(i)Chromium(I1) and -(HI) hydrazinecarboxylates. We havepreviously noted l 1 that treatment of [{Cr,(O,CMe),(p-N2H4)},] with hydrazinecarboxylic acid under nitrogen yields[(Cr(O,CNHNH,),(H,O)},]. Subsequently, this complex wasalso isolated3, from an aqueous chromium(I1) chloride solution.Full details are now reported, including additional preparativeroutes.Apart from the reactions mentioned above, chromium(1r)hydrazinecarboxylate monohydrate also results from thereactions of [Cr,(O,CMe),(H,O),] with either hydrazine-carboxylic acid in aqueous hydrazine (essentially [N2H5]-[02CNHNH2]), or hydrazine and carbon dioxide in refluxingmethanol. These reactions are noteworthy in that the metal-metal bonds of the acetate are cleaved yielding a mononuclearchromium(I1) product, the bridging hydrazine or unidentateaqua ligands and the bridging acetate ligands being replaced byhydrazinecarboxylate groups.In view of this metal-metal bondcleavage, it is not surprising that [{Cr(O,CNHNH,),(H,O)),]may also be prepared from a variety of mononuclearchromium(I1) reactants. Thus, reactions of deoxygenateJ. CHEM. SOC. DALTON TRANS. 1992 1765aqueous solutions of chromium(i1) chloride or bromide, orsuspensions of [{CrX2(p-N2H4)2}n] (X = C1 or Br), withaqueous [N,H,][O2CNHNH2] (prepared by passage of C02into an aqueous solution of hydrazine) all result in theformation of deep blue solutions from which good yields(> 60%) of the lilac product are precipitated. The product isinsoluble in all common organic solvents and more stable inmoist air than many mononuclear chromium(I1) compounds,being only slowly subject to oxidation to chromium(rI1) asshown by reflectance spectrum changes or the generation ofpeaks in the X-ray photoelectron spectrum at similar bindingenergies to those of chromium(II1) hydrazinecarboxylatedihydrate (oxidised [{Cr(02CNHNH2),(H20)},,]: 589.9, 579.3and 46.1 eV, [Cr(O2CNHNH2),]*2H2O: 589.1, 579.4 and46.6 eV, relating to Cr 2p,, 2pt and 3s, respectively).Prolongedcontact of [{ Cr(0,CNHNH2)2(H20)},] with an excess ofaqueous [N,H5][02CNHNH2] results in dissolution andoxidation to a bright red solution from which [Cr(O,-CNHNH,),]-2H20 can be isolated (see below). The rate ofoxidation appears to increase with increasing concentration of[N , H 5] [ 0 , CNHNH,], the successful isolation of [ { Cr(0 , C-NHNH,),(H,O)),] depending on its rapid precipitation andisolation from solution.Analytical data would not distinguish between [{Cr-(02CNHNH2),(H20)},] and the hydroxo-bridged chromium-(111) species [{ Cr(p-OH)(0,CNHNH,),}2].However, the lattercan be rejected because decomposition of the product in HClunder nitrogen generated Cr2+(aq) only and a redox titrationshowed that in basic solution 1 mol of the product consumed9 mol of [Fe(CN),I3-, in agreement with a one-electron changefor chromium and two four-electron changes for the hydrazine-carboxylato-groups (O2CNHNH,- --+ N, + C 0 2 + 3H' + 4e-).Attempts to remove the water from [{Cr(02CNHNH,),-(H,O)},,] using refluxing 2,2-dimethoxypropane or triethylorthoformate were unsuccessful and DTA/TG results point tovery strongly held water in the structure.Whereas DTAdeterminations on other hydrated metal hydrazinecarboxylatese.g. [M(O,CNHNH,),].xH,O (M = Mg, Ca or Mn) indicatewater loss in the range 130-1 70 "C, the chromium(I1) compoundshows a major exothermic peak at 214 "C both in air and inargon relating to water loss as indicated by complementary TGdata. A second exothermic DTA peak at 304°C in argonrepresents carboxylate decomposition leading ultimately toCr20, formation. These results differ markedly from thosereported by Manoharan and Patil.,, The intermediate [Cr-(02CNHNH2),] was not isolable and we infer from this thatthe water ligand plays a crucial role in the stability of thehydrazinecarboxylate via extensive hydrogen bonding to thehydrazinecarboxylate groups in addition to co-ordination tothe metal centre in a polymeric structure.The magnetic moment of [{ Cr(O,CNHNH,),(H,O)},] at298 K (peff = 4.73) confirms the presence of high-spin mono-nuclear chromium(r1) centres but it is puzzling to note thatManoharan and Patil report3, a value of 2.79, this reducedvalue being attributed to weak metal-metal bonding or strongmagnetic exchange. We have found reproducible values inagreement with high-spin chromium(I1) irrespective of thepreparative method employed.The reflectance spectrumconsists of a broad band with a maximum at 550 nm (5B1, +5B2g, 5Eg), rather similar to that of [{CrCl,(p-N,H,),},] butquite different to that of the glycinate2' [{ Cr(02CCH2NH2)2-(H20)),J (614 nm) which would be expected to have thesame donor atoms as the hydrazinecarboxylate.Finally, we consider the infrared spectrum of the hydrazine-carboxylate.Although vibrational data of metal hydrazine-carboxylates have been reported, assignments have been basedon comparisons with data from metal carboxylates and metal-hydrazine complexes, no consideration being given to skeletalvibrations of the OCNN core. In an attempt to propose firmerassignments, the infrared spectra of [(Cr(O,CNHNH,),-(H,O)},] and its fully deuteriated analogue, prepared byreaction of anhydrous CrCl, with C0,-saturated N2D,-D,0in D,O, have been recorded. The data presented in Table 3show a division between bands with H:D ratios of 1.25-1.36: 1involving considerable NH character and those with H:Dratios of 1.00-1.03:l in which the vibrations are essentiallycarboxylate or skeletal motions.Vibrations of the NHNH,groups are assigned by comparison with those of MeNHNH,33and metal-hydrazine complexes2' and will not be discussedfurther except to note that v(0H) has not been assigned,perhaps being broad and weak as a result of hydrogen bonding.The separation between the vasym and vsym vibrations of the0,CN moieties is 209 cm-', a low value for unidentate carboxyl-ate, but acceptable for a structure involving five-memberedCrOC(O)NHNH, chelate rings particularly if bridging of aC=O oxygen to another chromium atom to generate anoctahedral polymer, as in [(Cd(02CNHNH2)2(H20)}n],34occurs. Hydrogen bonding between the C=O and the watermolecules present could also make the carboxylate bonds moreequivalent than expected for unidentate carboxylate.Assignment of bands at 604,778 and 803 cm-', to carboxylatemodes, leaves three bands that are only slightly deuteriation-sensitive. In agreement with data on metal-hydrazine com-plexes,20 the band at 985 cm-' is assigned to a skeletal vibrationthat is primarily v(NN).The bands at 1215 and 1460 cm-',shifted to 1187 and 1424 cm-' on deuteriation, must involveconsiderable v(C0) and v(CN) character. Deuteriation-insensitive bands at 417 and 369 cm-' are likely to be v(CrN) orv(Cr0) vibrations.Chromium(r1r) hydrazinecarboxylate has been reportedpreviously as a red, crystalline, dih~drate,,~ a green-violet orblue anhydrous or most recently as a red trih~drate.~,We have found that the only product isolated from reactionscarried out under conventional conditions is the dihydrate.Ourinterest in this complex arises from the observation that it canbe isolated from reactions of various chromium(I1) complexeswith an excess of [N,H5][O2CNHNH2] solution. Thus, when[{Cr(O,CNHNH,),(H,O)}fl] or precursors such as hydratedchromium(I1) chloride or bromide, [{CrX,(p-N,H,),},] (X =C1 or Br) or anhydrous CrC1, are allowed prolonged contactwith an excess of aqueous hydrazine and CO, is passed into themixture a bright red colour develops in the solution and air-stable [Cr(O2CNHNH,),]-2H20 can be isolated.Theseoxidations proceed both in the presence and absence of air,suggesting that N2H4, N2H5' or 02CNHNH2- may beinvolved. Although N,H, and N,H5+ are usually regarded asreducing species, thermodynamic considerations indicate thatthey can behave as oxidants, being reduced to NH, or NH4+. Akinetic study3' of the Cr2+-N2H4 reaction in a perchloratemedium has been interpreted as producing Cr3+ and NH,.A more convenient route to [Cr(O2CNHNH,),]-2H,O isthe reaction of [CrCl,(H,0)4]Cl~2H,0 with aqueous [N2H5]-[O2CNHNH,]. The complex normally precipitates as a pinkfinely divided solid but on increasing the Cr3 + : 02CNHNH, -ratio to 1 : 50 it is formed as red microcrystalline aggregates.Suitable crystals for a structure determination were notproduced.The product is insoluble in all common solvents, butis soluble without change, as indicated by spectroscopy, inanhydrous N2H4 or concentrated [N,H5][0,CNHNH,]solutions. The magnetic moment of 3.82 at 295 K ischaracteristic of octahedral chromium(rI1). The reflectancespectrum, (maxima at 698vw, 523s and 395m nm) and solutionspectrum in anhydrous N2H4 [maxima at 700vw, 522s and402s (br) nm] can be interpreted on the basis of three N,O-chelating hydrazinecarboxylate anions bonded to the metal (inOh symmetry: 4A2, --- ,Eg, - 4T2g and -- 4Tlg, respec-tively, with lODq = 19 100 cm-', B = 587 cm-' and p = 0.64).These results can be compared with those for the glycinate3*[Cr(02CCH,NH,),]-H,0 which has N,O-chelating aminoacidate ligands (689vw, 503s and 384m nm; 1ODq = 19900cm-', B = 578 cm-' and p = 0.63).The octahedral Cr03N,I 1766 J. CHEM. SOC. DALTON TRANS. 1992Table 3 Infrared bands (cm-') of [Cr(O2CNHNH,),]-2H,O, [(Cr(O,CNHNH,),(H,O)),,J and its D,-analogue[Cr(02CNHNH,),]~2H,03329s3205s3106ms (sh)306oms1668vs1648s (sh)1615s (br)1570 (sh)1454vs1367m (br)1342m (br)1308w1246m (sh)1226s (sh)1216s983m783m710vw619m488mw408m330m3155s1627s161 Is (sh)I579vs1460s1370s1319mw1215m (sh)1 194s1102m985w803m778mw693w (br)636m (br)604s417m369wC(Cr(O,CNDND,),(D,O)),I2456s2415ms (sh)2385s2327s1219ms1187m1572vs1556s (sh)1424vs1372vs (sh)997m1187m896m879m989m802s76omw538vw (br)483m (br)598ms41 5ms368mwHID1.341.361.341.361 .oo1.031 .oo1.321.021.331.251 .oo1 .oo1.021.291.321.011.011 .ooSkeletal stretchVs,m(CO,)NH, wag>Skeletal vib.NH, twistSkeletal vib.S(OC0)NH, rock,NH def.W O , )C O , )unit could have the donor oxygen and nitrogen atoms co-ordinated in a fac or a mer arrangement. Both [Cr(02CCH2-NH2),]-Hz03* and [N2H,][Ni(02CNHNH2)3]-H203g arefac isomers.A further similarity of [Cr(02CNHNH2)3]-2H,0with [Cr(0,CCH2NH2)3]-H20 is the strong involvement ofwater in the crystal lattices. The glycinate is similarly insolublein a wide range of solvents, the structure involving an extensivenetwork of hydrogen bonding between the water of crystal-lisation and the amino and carboxylato groups.We have foundthat the water of the hydrazinecarboxylate cannot be removedby treatment with refluxing 2,2-dimethoxypropane or triethylorthoformate and heating in air at 110°C was ineffective(weight loss found 0.9%; calc. for loss of 2H20, 11.5%).Endothermic peaks with maxima at 136 and 169 "C from DTAexperiments carried out in air correspond to the stepwise loss of2H20 (TG supporting evidence), but pure samples of theseintermediates could not be obtained, more extensive de-composition occurring. An exothermic DTA peak with amaximum at 222 "C relates to decomposition of the hydrazine-carboxylate groups with ultimate formation of Cr203. Theseresults, which are rather different than those reported for thesupposed trih~drate,~, support extensive hydrogen bondingbetween the waters of crystallisation and the hydrazinecarboxy-late ligands.Loss of the water molecules seems to inducecollapse of the structure with concomitant decomposition of thehydrazinecarboxylate ligands. The report36 of a proposedanhydrous chromium(rr1) hydrazinecarboxylate, prepared froman aqueous medium, therefore seems questionable. The infraredspectrum of the dihydrate is typical of an 0,N-chelatedhydrazinecarboxylate, bands being listed in Table 3. Overlyingthe v(NH) bands is a very broad envelope associated with thehydrogen-bonded water molecules.( ii ) Chromium ( 11 I) me thy lhydr az ine carboxy la te . Pass age ofCO, into a solution of MeNHNH, in ethanol gives a colourless,highly hygroscopic, precipitate of methylhydrazinecarboxylicacid.I5 The favoured site of CO, attack on MeNHNH, willprobably be the more basic MeNH nitrogen, so the product isformulated as H,NN(Me)CO,H rather than MeNHNHC0,H.Given the formation of [{Cr(02CNHNH2)2(H20))nJ,isolation of a chromium(r1) methylhydrazinecarboxylate wasexpected, for example, from a reaction of hydrated chromium(r1)chloride with methylhydrazinecarboxylic acid in ethanol at 0 "Cunder nitrogen and carbon dioxide.The formation of an initialdark blue solution is observed, as in the hydrazinecarboxylatesystem, but rapid oxidation then ensues and the chromium(r1x)complex [Cr{ O,CN(Me)NH,} 3]-H20 is precipitated as apink, microcrystalline, air-stable solid.A similar, but muchslower, oxidation of [{Cr(O2CNHNH,),(H20)),1 to [Cr-(02CNHNH,)3]-2H20 has already been discussed. It thereforeseems that in the methylhydrazinecarboxylate system oxidationto chromium(xr1) occurs before a chromium(I1) product can beprecipitated.The complex [Cr{O,CN(Me)NH,},]~H,O is more con-venien tl y prepared from [ CrCl , (H O),] Ch2H , 0 and methyl-hydrazinecarboxylic acid in ethanol at 70 "C. The magneticmoment (3.84 at 292 K) and reflectance spectrum (517s and384m nm, ,A2 -,Tzg and -4T1, respectively; lODq= 19 300 cm-', B = 644 cm-', p = 0.70) are consistentwith chromium(Ir1) octahedrally co-ordinated by three 0 , N -chelating anions. A report36 that passage of C02 into anethanol solution of MeNHNH, and chromium(I1r) chlorideproduced violet-blue anhydrous [Cr(O,CN(Me)NH,},] couldnot be substantiated by us, the blue precipitate obtained alwayscontaining hydroxochromium(rr1) species. Indeed, DTA/TGevidence suggests that removal of water of crystallisation fromthe hydrate cannot be achieved without decomposition of themethylhydrazinecarboxylate groups.The decomposition path-way could not be elucidated, the TG trace not showing a specificstep relating to water loss and the DTA showing five exothermicpeaks with maxima at 195, 230, 255, 290 and 410 "C, the finalproduct being Cr,03. We conclude, as for the hydrazine-carboxylate, that hydrogen-bonded water molecules play a keyrole in the stability of the solid.The infrared spectrum of the hydrate has been recorded butfew useful comparisons are possible [major peaks at 3330m(vbr), v(0H); 3182ms, 3041ms, v(NH); 2930mw, v(CH3);1635vs, 1580(sh), vasym(C02) + 6(NH,); 1432m, skeletalstretch; 1380ms, vsYm(C0,); 1266s, skeletal stretch, mainlyv(CN); 1008m, skeletal stretch, mainly v(NN); 792ms, 6(OCO);615m, x(C0,); 407s, 342w, 300w cm-', v(Cr0) + v(CrN)].(iii) Chromium(I1) phenylhydrazinecarboxylate.Shortly afterFischer first reported the isolation of PhNHNH, in 1875, hedescribed the preparation of PhNHNH,.PhNHNHCO,H byreaction with C02.40 We have prepared this compoundJ. CHEM. SOC. DALTON TRANS. 1992 1767probably better formulated as [PhNHNH3][02CNHNHPh],by precipitation from methanol. This formula implies C 0 2attack at the more basic NH2 nitrogen of PhNHNH2 and forelectronic reasons it would be anticipated that the nitrogen lonepair adjacent to the phenyl ring in the anion might be lessavailable for donation to metals than the 02CNH nitrogen lonepair.However, N,O chelation would then involve a strainedMNCO ring system, unlike the well established MNNCO ringsystems found for many metal hydrazinecarboxylates. Alter-native potential co-ordination modes for PhNHNHC02- inwhich it behaves as a functionally substituted carboxylate aretherefore worthy of consideration e.g. (i) unidentate 0 bonding,(ii) 0,O' chelation, as established crystallographically for[(Ba(02CNHNH2)2(N2H4)),1,41 or (iii) 0,O' bridging asfound for binuclear chromium(I1) carboxylates and the diethyl-carbamate [Cr2(p-02CNEt2)4(NHEt2)2]?2Reaction of hydrated chromium(I1) chloride with [PhNHN-H3][02CNHNHPh] in methanol under a C02 atmosphere at0 "C resulted in the precipitation of an extremely air-sensitivesalmon-pink solid.It was almost diamagnetic (peff = 0.56 at296 K) and the reflectawx spectrum was characteristic of abinuclear chromium(I1) compound [bands at 469m and 455m(br), 6 --+ x*; 332s (sh) and 320vs nm, p + x * ] . The natureof this complex is clearly different to that of mononuclearchromium(r~) hydrazinecarboxylate, analysis and the aboveresults suggesting that it is [Cr2(02CNHNHPh),(MeOH)2].The infrared spectrum indicates the presence of methanol andphenylhydrazinecarboxylate ligands although the carboxylatestretches appear in the same region of the spectrum as do strongphenyl vibrations (bands at 1599, 1563, 1483, 1460 and 1373cm-').Tentatively, we suggest that [Cr2(02CNHNHPh-)4(MeOH),] has a metal-metal bonded structure akin to that of[Cr2(02CNEt2),(NHEt2)2]:2 with a change of co-ordinationmode from N,O chelation for the hydrazinecarboxylate to 0,O'bridging for the phenylhydrazinecarboxylate, the methanolligands being axial. Attempts to prepare chromium(w)phenylhydrazinecarboxylate, a compound previously re-ported,36 have not been successful. Thus, reactions of[CrC12(H20)4]Cl-2H20 or [CrCl,(thf),] (thf = tetrahydro-furan) with [PhNHNH,][O,CNHNHPh] did not lead tostoichiometric products.- -References1 J. M. Bellerby, J. Hazard. Muter., 1983,7, 187 and refs therein.2 R.Stolle and K. Hofmann, Ber., 1904,37,4523.3 J. M. Bellerby, J. Hazard. Muter., 1986, 13, 57; J. M. Bellerby, D. A.Edwards and D. Thompsett, Proceedings of the 3rd Conference onthe Environmental Chemistry of Hydrazine Fuels, eds. D. A. Stoneand F. L. Wiseman, Air Force Engineering and Services Centre,Tyndall Air Force Base, FL, 1987, p. 86.4 A. Earnshaw, L. F. Larkworthy and K. S. Patel, Z. Anorg. Allg.Chem., 1964,334, 163 and refs. therein.5 G. B. Kaufmann and N. Sugisaka, 2. Anorg. Allg. Chem., 1966,344,92.6 P. Glavic, J. Slivnik and A. Bole, Vestn. Slov. Kem. Drus., 1982,29,227.7 R. Y. Aliev, J. Gen. Chem. USSR (Engl. Transl.), 1972,42,2365.8 V. A. Sharov, N. V. Povarova, V. A. Perelyaev, E. I. Krylov andG. M. Adamova, Koord.Khim., 1979, 5, 537; V. A. Sharov, N. V.Povarova and E. I. Krylov, Russ. J. Znorg. Chem. (Engl. Transl.),1980,25,1193.9 K. C. Patil, K. C. Nesamani and V. R. Pai Verneker, Chem. Ind.(London), 1979,901.10 P. Ban and H. P. Fritz, 2. Naturforsch., Teil B, 1972,27, 1131.11 J. M. Bellerby, D. A. Edwards and D. Thompsett, Znorg. Chim.12 A. Earnshaw, L. F. Larkworthy and K. S. Patel, J. Chem. SOC., 1965,13 J. C. Reeve, J. Chem. Educ., 1985,62,444.14 D. M. Halepoto, L. F. Larkworthy, D. C. Povey, B. J. Tucker andG. W. Smith, J. Chem. Soc., Dalton Trans., 1990,699.15 J. M. Bellerby and H. A. El-Magied, unpublished work.16 A. Stern, J. Prakt. Chem., 1899,60,235.17 D. G. Holah and J. P. Fackler, Znorg. Chem., 1965,4, 11 12; R. J. H.Clark and C.S. Williams, Znorg. Chem., 1965,4,350.18 L. F. Larkworthy and J. M. Tabatabai, J. Chem. Soc., Dalton Trans.,1976,814.19 A. Ferrari, A. Braibanti and G. Bigliardi, Acta Crystallogr., 1963,16,498; A. Ferrari, A. Braibanti, G. Bigliardi and F. Dallevalle, 2.Kristallogr., 1963, 119,284.20 M. Goldstein and W. D. Unsworth, Spectrochim. Acta, Part A, 1972.Acta, 1986,117, L31.3267.21222324252627282930313233343536373839404142. ,28, 1297; W. Personette, J. Szczepaniki, K. KuBulat and M. Vala,J. Raman Spectrosc., 1989,20, 147.L. Sacconi and A. Sabatini, J. Inorg. Nucl. Chem., 1963,251389.A. Anagnostopoulos, D. Nicholls and J. Reed, Znorg. Chim. Acta,1979,32, L17 and refs. therein.T. V. Ashworth, M. J. Nolte and E. Singleton, J. Chem. Soc., DaltonTrans., 1978, 1040.M. K. Das, J. F. Mariategui and K. Niedenzu, Inorg. Chem., 1987,26,3114.T. V. Ashworth, M. J. Nolte, R. H. Reimann and E. Singleton, J.Chem. Soc., Dalton Trans., 1978, 1043.D. Nicholls and R. Swindells, J. Inorg. Nucl. Chem., 1968, 30,2211.W. K. Glass and J. 0. McBreen, J. Inorg. Nucl. Chem., 1974,36,747;M. S . Novakovskii, V. A. Starodub and E. V. Golovinova, Zh. Neorg.Khim., 1974,19,3288.G. J. Leigh and J. R. Sanders, Polyhedron, 1989, 8, 1782; V. Chilou,P. Gouzerh, Y. Jeannin, G. Olivares, F. Robert, T. C. Hsieh andJ. Zubieta, Polyhedron, 1989,8,29.L. F. Larkworthy and J. M. Tabatabai, Znorg. Chim. Acta, 1977,21,265.F. A. Cotton and T. R. Felthouse, Znorg. Chem., 1980, 19,328.L. F. Larkworthy and J. M. Tabatabai, Znorg. Nucl. Chem. Lett.,1980,16,427.S. S . Manoharan and K. C. Patil, Proc. Indian Acad. Sci., (Chem.Sci.), 1989,101, 377.D. P. Dowling and W. K. Glass, Spectrochim. Acta, Part A, 1988,44,1351.A. Braibanti, A. M. M. Lanfredi, A. Tiripicchio and F. Bigoli, ActaCrystallogr., Sect. B, 1969,25, 100.H. Funk, A. Eichhoff and G. Giesder, Omagiu Raluca Ripan, Edit.Acad. Rep. SOC. Romania, 1966,247 (Chem. Abstr., 1967,67,39661~).A. K. Srivastava, Croat. Chem. Acta, 1979,52,293.C. F. Wells and M. A. Salam, J. Chem. SOC. A , 1968,1568.R. F. Bryan, P. T. Greene, P. F. Stokely and E. W. Wilson, Znorg.Chem., 1971,10,1468.A. Braibanti, A. M. M. Lanfredi and A. Tiripicchio, 2. Kristallogr.,1967,124,335.E. Fischer, Liebigs Ann. Chem., 1878,190,123.D. A. Edwards, J. F. Keily, M. F. Mahon, K. C. Molloy and D.Thompsett, unpublished work.M. H. Chisholm, F. A. Cotton, M. W. Extine and D. C. Rideout,Znorg. Chem., 1978, 17,3536.Received 6th January 1992; Paper 2/00055