摘要:
J. CHEM. SOC. DALTON TRANS. 1992 15092-Galla-arachno-tetraborane(lO), H2Ga B,H,: Synthesis,ProDerties and Structure of the Gaseous Molecule asdetermined by Electron Diffraction tColin R. Pulham," Anthony J. Downs,**" David W. H. Rankin*Sb andHeather E. Robertsonba Inorganic Chemistry Laboratory, University of Oxford, South Parks Road, Oxford OX 7 3QR, UKDepartment of Chemistry, University of Edinburgh, West Mains Road, Edinburgh EH9 3JJ, UKThe novel mixed hydride 2-galla-arachno-tetraborane( lo), H,GaB,H,, has been synthesised bymetathesis between neat monochlorogallane and tetra-n-butylammonium octahydrotriborate,[Bu",N] +[B,H,] -, at temperatures near 243 K. The compound has been characterised by chemicalanalysis, by its vibrational, 'H and "B NMR, and mass spectra, and by its reaction with ammonia.Electron-diffraction measurements, carried out on the vapour at ca. 273 K, are wholly consistent with thetetraborane(l0)-like structure of the molecule with gallium replacing boron at the 2 position. The r,structure features the following parameters: r[Ga-B(1,3)] 231.2(0.1), r[B(1,3)-B(4)] 185.2(1.3),r(Ga-H,) 144.2(1.1), r(Ga-H,) 176.0(2.8) and r(B-H,) 126.4(0.7) pm; and 11 4.4(0.6)O for the dihedralangle between the planes GaB(l)B(3) and B(l)B(3)B(4) (H, = terminal H atom; H, = bridging Hatom). The NMR spectra of solutions imply non-fluxional behaviour of the molecule in the temperaturerange 193-283 K. Gallatetraborane(l0) decomposes at ambient temperatures t o give either metallicgallium, dihydrogen and tetraborane(l0) or a solid with the composition GaB,, dihydrogen and diboraneaccording to whether it is in the condensed or gaseous phase, respectively.The recent synthesis of the simple base-free gallium hydridesHGa(BH,)2,'.2 Ga2H6 and H,GaBH, ' v 6 raises the questionof whether GaH or GaH, moieties can also be successfullyincorporated into polyborane or carborane clusters. Precedentsare sparse but the compounds Me2GaB,H8,7*8 Me,GaC2-B4H7,' MeGaC,B,H, l o and Tl[commo-3,3'-Ga( 1,2-C2B9-HI 1 ) 2 J '' do provide examples of gallium-substituted uruchno,nido or closo frameworks embodying 4-12 vertices. However,the inherent frailty of Ga-H bonds militates against high-energymethods of synthesis, requiring instead the sort of facile reactionthat can be engineered at low temperatures so as to outstripdecomposition.Foremost among the precursors to gallane 3,4and its derivatives is monochlorogallane, [H,GaCl],, acompound conveniently prepared by the reaction of gallium(II1)chloride with an excess of trimethylsilane and which we have nowcharacterised in some detai1.6*'2,'3 This has been shown toundergo metathesis with a salt of the nucleophile GaH,- 3*4 orBH, - , 5 * 6 typically under solvent-free conditions and attemperatures of 240-250 K, to give the elusive binary hydride,[GaH,],, or the mixed hydride gallaborane, [GaBH,],,respectively, in accordance with equation (1) (X = GaH, or$[H,GaCl], + Li+X- --++ [H,GaX], + Li'C1- (1)BH,). Accordingly we were prompted to investigate whether theoctahydrotriborate anion, B3H8 -, also displaces chloride frommonochlorogallane. The most likely product of such a reactionis the urachno-gallatetraborane( lo), H2Ga(p-H),B3H6, featur-ing bidentate ligation of the B3H8 group and with a structureanalogous to that of Me,GaB,H, l .7 7 8 However, there is alsothe possibility of forming a four-vertex gallatetraboraneH,Ga(p-H),B,H, featuring five-fold co-ordination of thet Non-SI unit employed: mmHg z 133 Pa.H H102gallium and tridentate ligation of the B3H8 group, and with astructure analogous to that of [Mn(B,H,)(CO),] 2.14 Yetanother option involves rearrangement to give a tetraborane( 10)derivative with gallium replacing boron at a hinge (1) ratherthan an apical (2) site.Indeed, we find that monochlorogallane reacts smoothly withfreshly recrystallised tetra-n-butylammonium octahydrotri-borate in uucuo at cu.243 K to give a volatile product shown tobe 2-galla-uruchno-tetraborane(l0), H,GaB,H,. As reportedbriefly in a preliminary communication," the gallane has beenauthenticated by chemical analysis and by its vibrational, 'Hand "B NMR, and mass spectra; the structure of the gaseousmolecule has also been determined on the basis of the electron-diffraction pattern of the vapour. The present paper isconcerned primarily with the synthesis, physical properties anddecomposition of the compound, as well as its reaction withammonia; a fuller account of its chemistry, now being studied,will be presented separately.' Our findings have alreadystimulated theoretical enquiries in which ub initio molecularorbital methods have sought to explore the equilibriumgeometries open to the GaB3H10 molecule.'7 In a practicalcontext, too, the gallane is noteworthy as a means of vapourtransport of gallium at low temperatures ( < 253 K) and also asa possible precursor to solid gallium-bearing films.1510 J.CHEM. SOC. DALTON TRANS. 1992Results and Discussion(i) Synthesis of 2-Galla-arachno-tetraborane( 10).-2-Galla-arachno-tetraborane( 10) has been synthesised by the reaction invacuo between monochlorogallane, [H,GaCl],, and powdered,freshly recrystallised tetra-n-butylammonium octahydrotri-borate, [Bu",N] +[B3H8] -, under solvent-free conditions.After co-condensation of the monochlorogallane with theoctahydrotriborate, the reaction is initiated by allowing themixture to warm to ca.243 K. On the evidence of chemical andspectroscopic analysis, the main volatile product is 2-gallatetra-borane(lO), H,GaB3H,. The reaction can be made to proceedwith ca. 90% conversion of monochlorogallane into gallatetra-borane in accordance with equation (2).243-268 K5[H2GaC112 + [Bu"4NI+[B3H81- -H2GaB,H8 + [Bu",N]+Cl- (2)As with the synthesis of gallane,3.4 scrupulous attention topractical technique is necessary for the successful synthesis andmanipulation of gallatetraborane. It is imperative rigorously toexclude moisture, grease, or other contaminants. Operationsmust be carried out at pressures mmHg in all-glassapparatus designed to give short distillation paths and whichhas been preconditioned by heating under continuousp ~ m p i n g .~ Moreover, the thermal frailty of the compoundrequires that all glassware to which it has access be maintainedat temperatures < 283 K to avoid decomposition (signalledtypically by the appearance of grey deposits on the walls of theapparatus and the formation of diborane). Unlike gallane,4however, gaseous gallatetraborane will survive the passagethrough a greaseless (Teflon) valve provided that steps aretaken to precondition the surfaces by preliminary exposure to aportion of the vapour which is subsequently pumped to waste.Further practical details are given in the Experimental sectionand ref. 4.Gallatetraborane melts at ca.178 K to give a colourlessmobile liquid which, on the evidence of its rate of vaporisation,has a vapour pressure of ca. 1 mmHg at 210 K. Although thecompound is thermally fragile, it is considerably more robustthan either gallane 394 or gallaborane, [GaBH6],.5*6 Samples ofthe material in the condensed phase (liquid or solution)decompose rapidly at temperatures in excess of 283 K; at roomtemperature and a pressure of ca. 100 mmHg the vapour has ahalf-life of the order of 30 min.(ii) Decomposition and Chemical Analysis.-Gallatetra-borane( 10) vapour decomposes at ambient temperatures inaccordance with equation (3), as attested by quantitative assayH2GaB3H, --+ GaB, + f B,H6 + 5 H2 (3)of the products, and this provides a satisfactory means ofanalysis of the gallane.The involatile solid residue forms abrown film on the walls of the container; while it answers to thecomposition GaB,, we have no definite evidence that it is asingle homogeneous product. Although the aluminium borideAlB, is known, the literature does not appear to contain anyreference to a gallium anal~gue,'~" and, beyond the reportedpreparation of GaB, , , l g b gallium borides of any description arevirtually uncharted.A potential complication arises from the existence of morethan one decomposition channel. As the local concentration ofthe gallane increases, for example in the condensed phases,bimolecular processes compete with the unimolecular onewhich is presumed to be the dominant primary step at lowpressures affording the products detailed above.Thus, NMRmeasurements reveal that the decomposition of gallatetra-borane( 10) in C2H8]toluene solution gives metallic gallium,dihydrogen and tetraborane( 10) as the principal products, as inequation (4), although these are also accompanied by smallamounts of unidentified boranes generated by secondaryreactions.(iii) Mass Spectrum.-The mass spectrum of the gallanevapour includes a weak peak at m/z = 114 corresponding to themolecular ion 71Ga'1B3H10+. That there is no sign of anyfeature at higher mass number suggests that the compoundvaporises as the monomeric molecule GaB,Hl,. There is afamily of peaks at m/z = 107-1 14 but the intensities show thatmost of the components originate only partially in the mole-cular ions XGaYB3H10+ (x = 69 or 71, y = 10 or ll), thecontributions of which are augmented by the products ofhydrogen loss, GaB,H,+ ( z = 6-9).Otherwise the fragment-ation pattern is dominated by successive loss of BH, givingGaB,H,+, GaBH,+, and, ultimately, GaH' and Ga+ as themost prominent ions together with an array of borane fragmentscontaining one, two or three boron atoms.(iu) Vibrational Spectra.-Infrared spectra have beenmeasured for gallatetraborane with reference (a) to the vapourat a pressure of ca. 50 mmHg and ambient temperatures, (b) tothe vapour species trapped in a solid nitrogen matrix at ca.20 K, and (c) to the annealed solid condensate formed on a CsIwindow at 77 K. The Raman spectrum of the solid condensateformed on a copper block at 77 K has also been measured.Thespectra are illustrated in Fig. 1 and the details itemised, withproposed assignments, in Table 1.That the vibrational spectra are essentially independent ofphase and temperature implies the presence of a molecular unitcommon to the vapour, matrix-isolated, and solid states.Reference to the vibrational spectra of compounds known tocontain terminal GaH, groups (e.g. Ga2H6,3'4 H2GaBH, 5 3 6and [H,GaCl], 1 2 v 1 3, and dihydrogen-bridged octahydro-triborate groups (e.g. Me,GaB3H, and B4Hlo2') leaves littledoubt that the source of our spectra is a derivative oftetraborane( 10) with gallium replacing boron at the 2 position(see Fig. 2). Thus, the strong IR doublet near 2000 cm-' and theintense absorption near 700 cm-' are both characteristic of theGaH, moiety and, with due allowance for the modes localisedmainly in this or in the GaMe, moiety, the spectra are virtuallysuperimposable on those of Me2GaB3H8 under comparableconditions.' Hence there is strong circumstantial evidence thatthe new molecule has the structure shown in Fig. 2 akin to 1.Such a molecule, H2GaB3H8, has C, symmetry with 36vibrational fundamentals spanning the a' and a" irreduciblerepresentations and each of which is active in both IR absorp-tion and Raman scattering.The low symmetry of the molecule,combined with the near-degeneracy of many of the funda-mentals, means that there will inevitably be considerable mixingof some local, group modes.In the absence of any informationabout the Raman spectrum (including polarisation properties)of the compound in a fluid phase, and of experiments involvingselective isotopic enrichment (cf: B4Hlo ' O ) , it is not yet feasibleto carry out meaningful normal-coordinate-analysis calcul-ations to arrive at a full vibrational assignment for the newmolecule. However, analogies with the vibrational propertiesof related molecules, notably Ga2H6,3*4 H,GaBH4,5*6[HzGaC1],,'2*13 [Me,NGaH,],,2' B4H10 2o and Me,GaB,-H8,7 do provide a reasonable basis for the assignmentssuggested in Table 1. Although the proposals are necessarilysimplified and tentative, they are consistent with the morerigorous assignments which have been advanced for moleculeslike Ga,H,, H,GaBH4, [H,GaCl], and B4H10, and areunlikely therefore to be too wide of the mark.( u ) NMR Spectra-The 'H and "B NMR spectra of thegallane, which are illustrated in Fig.3 and itemised in Table 2,provide unequivocal evidence of its identity. The normal 'Hspectrum of a C2H,]toluene solution at 233 K is broad anJ. CHEM. SOC. DALTON TRANS. 19921.4 Iy y - v - w - w v vI . I , I , I , I I I , A2800 2400 2000 1600 1200 800 400I 13000 2000 1400 800 200u/cm-'J151 1L11 800 1200 1600 2000 2400C/cm-'Fig. 1 Infrared spectra of 2-gallatetraborane( 10) vapour, (a) at a pressure of ca. 50 mmHg and ambient temperatures (sample contained in a cellfitted with CsI windows and having a pathlength of ca.10 cm), (6) isolated in a solid dinitrogen matrix at ca. 20 K (N2 : H2GaB,H, = ca. 200: 1) andof (c) an annealed solid film of 2-galIatetraborane(lO) condensed on a CsI window at 77 K. (d) Raman spectrum of an annealed solid film of 2-gallatetraborane( 10) condensed on a copper block at 77 KQFig. 2optimum refinement of the electron-diffraction analysisPerspective view of the 2-gallatetraborane( 10) molecule in theuninformative, partly as a result of the quadrupolar natures ofthe Ga and B nuclei, partly through the effects of correlation-time decoupling of the protons attached to boron.23 The "Bspectrum is much more informative [See Fig. 3(a)] in revealingthe presence of two types of boron in the proportions 2: 1. Thetriplet resonance at 6("B) -12.9 can be identified with theunique apical boron atom, B(4) in Fig.2. The two terminalhydrogen atoms attached to B(4) are inequivalent (em and endo)and so the l 1 B resonance should take the form of a doublet ofdoublets; in the event the doublets overlap to give the appearanceof a triplet. The resonance centred at 6("B) -44.0 is due to theboron atoms B( 1) and B(3). Its splitting pattern is unusual butcharacteristic of the resonances emanating from hinge boronatoms in related molecules, e.g. [Mn(B3Hs)(CO),].22The response of the corresponding 'H NMR spectrum tobroad-band and selective "B decoupling is shown in Fig. 3(6).The high-frequency signal, which is unaffected by "Birradiation, can be linked only to the terminal protons of theGaH, unit, the AB-type structure of the signal arising from thedistinct exo and endo sites of the protons with respect to thefolded GaB, ring and from the condition that J(H-H) iscomparable in magnitude with the difference in chemical shift.The corresponding doublet which appears at 6 , 2.78/2.09 onirradiation of the sample at the resonant frequency of B(4)must originate in the apical BH, unit.In this case thedifference in chemical shifts between the exo- and endo-protons appears to dwarf J(H-H), and no discernible H-Hcoupling is revealed under the conditions of our experiments.Irradiation at the resonant frequency of B(1) and B(3)sharpens the 'H resonances at ijH -0.80 and - 1.49 whichmust therefore be correlated with the two pairs of bridgingprotons, H, and H,' (Fig.2). It is impossible to deduce withcertainty which of these two signals corresponds to H, andwhich to H,'. However, the observation that the resonance athigher frequency is slightly broader when the sample isirradiated at the B(1,3) frequency than when it is subject tobroad-band irradiation suggests that this originates in theprotons H,' which come under the influence of B(4) as well asB(1,3). Irradiation at the resonant frequency of B(1,3) alsoresults in decoupling of the 'H resonance at tiH 1.16 which canbe attributed to the terminal protons of B(l) and B(3), Ht'.Support for this assignment comes from the relative intensity(2 H) and from the general observation that terminal protonsattached to boron resonate at higher frequency than dobridging ones.24Cooling the sample from 233 to 193 K causes the AB systemof the GaH, unit in the 'H NMR spectrum to sharpen butotherwise has little effect on the 'H-{ "B} or ''B NMR spectra1512 J.CHEM. SOC. DALTON TRANS. 1992Table 1 Vibrational spectral data (cm-') for 2-gallatetraborane( 10)"Solid condensate at 77 K IR spectrum of vapour at ambient temperaturesIR2541vs2467ms21 17s2008s1986s13 14w1231w1140w1061w1007w976w892w834w701s676s616m536mw492vw395m270msRaman2547m2534m2472m2137m2 124m2 107m2010 (sh)1984s1063w91 lw743s704m678m61 8w590w523vs502m492 (sh)396w268sTrapped in a nitrogenmatrix at 20 K Gaseous sample Assignment2558m2550 (sh)2488mw21 13m2063w2036w201 8m2004m1998m1987m1981m1320w1240vw1144w1012w980w888w832w7 12vs z:} 664m619mw2563s2486m2107s2007mw1996vs1989s1983 (sh)1327vw1235vw1153w101 lw984w879w834w789vw706678mw620wVasyrn(Ga-Ht)vs,m(Ga-H,)v(Ga-H,) + out-of-phaseW H )WH,)in-phase BH wagBH, wagBH, twisting or rockingBridge deformationiiGaH, waggingGaH, rockingSkeletal deformation?v(Ga-B)Skeletal deformationmodes i 535mw 529w405mw375w28Omw" Abbreviations: s = strong, m = medium, w = weak, v = very, sh = shoulder, H, = terminal H atom, Hb = bridging H atom.Pressure ca. 30mmHg in a cell with a pathlength of ca. 10 cm. Partially resolved rotational branches.Table 2in [ZH,]toluene solution at 233 KProton and IIB NMR parameters for 2-gallatetraborane(lO)'HAtom"H, of GaH, unitHb (Or Hb')H,' (or Hb)H,'HI of B(4)H2unit"BCouplingChemical shift, constant,4.02,4.56 486, ' J ( H-H)/Hz- 1.49- 0.801.162.09, 2.78 < 40CouplingChemical shift, constant,Atom" 6, 'J(B-H)/HzB(1,3) - 44.0 69B(4) - 12.9* ca. 125" Labelling of atoms as in Fig.2. AB quartet. Triplet of unusual form(cf: ref. 22). Triplet due to overlapping of two doublets (see text).Warming the sample to 273 K evokes little response from the'H-("B) spectrum, apart from a broadening of the GaH2resonance. The ' 'B resonances broaden and lose their structure,a change suggesting that some proton exchange is nowoccurring. With the sample at temperatures above 283 K, the"B NMR spectrum discloses the appearance and growth of atriplet centred at 6( ' ' B) - 7.4 and a doublet centred at 6(' 'B)-42.3 attributable, respectively, to B(2,4) and B(1,3) oftetraborane( In addition, there are some weaker featureswhich develop at this stage and signal the formation of otherboranes as yet unidentified. The 'H NMR spectrum showsthe simultaneous growth of a sharp resonance at 6H 4.6associated with dihydrogen2' at the expense of the GaH,resonance.Inspection of the NMR cell at this stage shows thewalls to be coated with a grey metallic film presumed to beelemental gallium.That the NMR spectra are essentially independent of temper-ature in the range 193-283 K argues that fluxional processesopen to the H2GaB,H8 molecule are slow on the NMR time-scale. The results naturally invite comparisons with thosereported for other metallotetraboranes and related sys-t e m ~ .~ * ~ ~ * ~ ~ . ~ ~ The ' 'B chemical shifts for the static metallo-tetraboranes vary only slightly with the nature of the metallofragment, and in this respect 2-gallatetraborane is unremarkable.Rather is it the fluxional behaviour of these systems which ismore remarkable. The behaviour appears to run the wholegamut from the essentially static behaviour of [Mn(B,H,)-(CO),] 2 2 to the rapid exchange of boron and hydrogen atomsin [Be(B3H8)(C5H5)] 27a and M +[B,H,]- even at ver1513EJ. CHEM. SOC. DALTON TRANS. 1992(a )-5 -15 -25 -35 -45 -55BB5.0 4.0 3.0 2.0 1.0 0.0 -1.0 -2.0SHFig.3 (a) Boron-11 NMR spectrum of a C2H,]toluene solution of 2-gallatetraborane(lO) at 233 K (measured at 96.25 MHz). (6) Proton NMRspectrum of the same solution at 233 K (measured at 300 MHz) showing the effects of broad-band and selective "B decoupling. The sharp resonancein the 'H spectra near 6,2.0 is due to the solventlow temperatures. The factors which determine whether suchcompounds are fluxional or not are but poorly understood, andit seems likely in any case that traces of impurity in the samplemay have a profound effect on the exchange rates andmechanisms.( v i ) Structure of Gaseous 2-Gallatetraborane( 10) as Deter-mined by Electron Diffraction.-Using the all-glass inlet systemspecially constructed for studies of gallane and related speciesand maintained at a temperature near 273 K, we have succeededin measuring the electron-diffraction pattern of the gallatetra-borane vapour.The measurements have been made at twocamera distances, viz. ca. 200 and 260 mm, at an electronwavelength of 5.67 pm to give results spanning the range 20-208nm-' in the scattering variable s. As with similar studies of othervolatile gallane 1+6*12,21*28 or tetrahydroborate 29 derivatives,the main obstacle to success is the propensity of the stronglyreducing vapour to react with the emulsion of the photographicplates.In the light of the evidence adduced by the vibrational andNMR spectra it seemed appropriate to adopt a structural modelakin to that of Me,GaB,H, with the GaH, unit replacing BH,in the 2 position of B4H103' to give a molecule conformingoverall to C, symmetry (see Fig.2). A full specification of such amodel requires the 14 independent geometrical parameterslisted in Table 3. Local CZ0 symmetry has been assumed for boththe apical (Ht)2Ga(Hb)2 and (H,),B(H,), units (H, = terminalhydrogen atom, H, = bridging hydrogen atom).The experimental molecular scattering curves for the twocamera distances are shown in Fig. 4. Combination of the scaledexperimental data sets yields the radial-distribution curvereproduced in Fig. 5 which exhibits no more than four welldeveloped peaks. The broad feature centred near 130 pm plainlycorrelates with the scattering from all the B-H, and B-H, atompairs.The feature at ca. 180 pm is also due to a blend ofscattering, this time from the B(l)-B(3), Ga-H, and B(l)-B(4)atom pairs. Scattering from the Ga-B(1,3) atom pairs is respon-sible for the peak near 230 pm, whereas the broad feature near320 pm reflects mainly the scattering of Ga B(4) augmentedby contributions from other non-bonded atom pairs, uiz.Ga H, B H and H H. The comparative simplicity ofthe radial-distribution curve, its obvious kinship to those ofB4H1030 and Me2GaB3H8,, and the absence of any majorcontribution from long-range atom pairs all tend to endorseH,GaB,H, as the predominant, if not the sole, vapour species,and to rule out the possibility that larger molecules are presentin appreciable concentrations. Tridentate ligation of the octa-hydrotriborate ligand, as found in [Mn(B,H,)(CO),] 2,14would be expected to give rather different results, with sig-nificantly more scattering arising from pairs of atoms separatedby ca.180 (corresponding to Ga-H, bonds) and 230-250 pm(corresponding to Ga-B contacts) and less from pairs separatedby ca. 320 pm.Our relatively meagre knowledge of the vibrationalproperties characterising gallatetraborane and related moleculesmakes it impossible to gauge the magnitudes of 'shrinkage'corrections or to estimate on this basis any of the amplitudes ofvibration. As may be expected, therefore, the analysis has beenhampered by the marked correlation between several of thestructural parameters, this being implicit in the compositenature of most of the peaks in the radial-distribution curve; forexample, the B(l)-B(3), B(l)-B(4) and Ga-H, distances are allstrongly subject to such correlation.The problems are exacer-bated, moreover, by the degree to which the molecular scatter-ing is dominated by the heavier atoms, making it particularlydifficult to locate precisely the positions of the hydrogen atoms.It has been necessary therefore to assign fixed values to most ofthe parameters defining the structure of the B3H8 group, thesebeing based on the corresponding dimensions of B4H10 30 andcertified, where possible, by R-factor plots. The structure predi-cated for the molecule has been constrained by the follow-ing assumptions. (a) The B( 1)-B(3) distance has been fixed at174 pm on the basis of an R-factor plot, although values in therange 170-1 78 pm could be accommodated almost equally well.(b) The H,-Ga-H, and H,-B-H, bond angles have both beenfixed at 11 5" on the basis of R-factor plots, but any value in therange 110-120" is found to be acceptable.(c) The mean B-H,distance has been fixed at 121 pm on the basis of an R-factorplot. (d) The difference between r[B(4)-H,] and r[B(1,3)-H,],A l , has been fixed at 2.8 pm in line with the dimensions ofB4H10.30 (e) The difference between the two r[B(1,3)-Hb]distances, A3, and between r[B( 1,3)-H,] and r[B(4)-Hb], A,1514 J. CHEM. SOC. DALTON TRANS. 1992Table 3 Molecular parameters deduced from the electron-diffraction pattern of 2-gallatetraborane( 10) and compared with the results of ab initiocalculationsDistance/pm or anglerAb initio VibrationalParameter Experimental" calculations amplitude "/pm(a) Independent parameters'r c ~ ( l w ( 3 ) i 174.0 179.2 4.3(2.6)r[Ga-B( 1,3)] 231.2(0.1) 234.2 6.8(0.5)rCBW)-B(4)1 185.2(1.3) 190.3 4.3Dihedral angle, a 114.q0.6) 117.1 -121.Od 118.8 - @-Ht)meanA, = O(~FHJ - ~cB(~,~)-H,I 2.8 + 0.4 -126.4(0.7) 131.1 - r(B-Hb)meanA2 = r[B(4)-&'] - i(r[B(1,3)-Hb'] + 0.0 15.7 -A3 = r[B(1,3)-Hb'] - r[B(1,3)-HJ 0.0 0.2r(Ga-Hb) 176.q2.8) 185.5 10.0Angle HI-Ga-H, 115.0d 130.2 -Angle Ht-B(4)-H, 1 15.0d 119.9 -Angle B(3)-B(l)-Ht' 1 lo.od 107.4 --rCB( ,3>-Hb1 1r(Ga-H,) 144.2( 1.1) 155.7 9.0(b) Interatomic distances and vibrational amplitudes 'ar(Ga-H,) 144.2( 1.1) 155.7 9.0r(Ga-Hb) 176.0(2.8) 185.5 10.Odr[Ga-B( 1,3)] 231.2(0.1) 234.2 6.8(0.5)r[Ga B(4)] 318.7(0.5) - 7.1(0.9)r[Ga H;] 320.0(0.3) - 13.5( 1.9)r[Ga HI'] 3 10.9( 1.2) - 13.5'~cB(~)-B(~)I 174.0 179.2 4.3(2.6)rCB(1,3)-B(4)1 185.2( 1.3) 190.3 4.3O(L3)-H1'l 122.4 118.6 11.4'125.8 11.4'126.4(0.7)' 126.0 11.4'141.6 11.4'1 19.6d 119.0 1 1.4(0.8) 1 r[B(l ,3)-Hb1r[B( ,3kHb'IrCB(4)-Hb'I~cB(~)-H,I" This work. Figures in parentheses are the estimated standard deviations of the last digits. Although allowance has been made for systematic errors inthe electron wavelength, nozzle-to-plate distances, etc., the values do not budget for systematic errors associated with the assumptions made about themodel; to this extent they may underestimate the true margins of error.Calculations use a DZP basis set (ref. 17). ' Atoms are numbered as in Fig. 2.Dihedral angle between the planes GaB(l)B(3) and B(l)B(3)B(4). Other distances involving non-bonded atompairs, viz. Ga H, B - H and H H, were included in the refinement calculations, but are not listed here. ' Tied to u(Ga - H,,'). Tied tou[B(4)-Ht]. j B-H, and B-H,' distances assumed to be equal.Fixed. Tied to u[B(l)-B(3)].Fig. 4(b) 201.08 mmExperimental and final difference molecular-scattering intensities for 2-gallatetraborane( 10) vapour; nozzle-to-plate distances (a) 259.48 andhave both been set at zero following checks by R-factor plots.(f) The B(3)-B(l)-H(l) angle has been assigned a value of 1 lo",again on the basis of an R-factor plot.In addition, the bridginghydrogen atoms have been taken t o be coplanar with theheavy-atom planes GaB( l)B(3) and B( l)B(4)B(3); in fact thiscondition runs counter to the experience with B,H,0,30 butsuch is the dearth of information about the finer structuraldetails of gallatetraborane(l0) that we have little alternative.With these provisions and simplifications, molecular-scatter-ing intensities have been calculated and the molecular structurehas been successfully refined on the basis of the model describedabove by full-matrix least-squares analysis.31 It has beeJ. CHEM. SOC. DALTON TRANS. 1992 1515Table 4elements ( x 100) with absolute values 2 50%Analysis of the electron-diffraction pattern of the H,GaB,H, molecule: portion of the least-squares correlation matrix listing off-diagonala- - 90* Scale factor.r(Ga-Hb) u[B(l)-B(3)] u[B(4)-Ht] u(Ga - - Hb') k , * k2 *90 -9172 7457 5195-6455 69 5880 6675- 69Fig.5 Observed and difference radial-distribution curves, P(r)/ragainst r, for 2-gallatetraborane( 10) vapour; before Fourier inversionthe data were multiplied by s . exp[( -0.oOO 035 s')/(zGa-fC.)(z,-fB)]possible in the process to refine simultaneously six out of the 14geometrical parameters used to specify the model. Thecalculations have admitted, in addition, the independentrefinement of no more than five amplitudes of vibration[relating to the B(l)-B(3), Ga-B(1,3), Ga B(4), B(4)-H, andGa Hb vectors]; the remaining amplitudes have beenassigned values in line either with these results or withcorresponding parameters of the related molecules B,H 0,30Me,GaB,H, and [H2GaC1],.6~'2 The most noteworthyaspect of the final least-squares correlation matrix, reproducedin part in Table 4, is the substantial correlation involvingr[B(1,3)-B(4)], the dihedral angle a, r(Ga-Hb) and theamplitude u[B( l)-B(3)].The success of the calculations may beassessed by the differences (a) between the experimental andsimulated radial-distribution curves (Fig. 5) and (b) between theexperimental and simulated intensities of molecular scattering(Fig. 4). Table 3 lists the values of the geometric and vibrationalparameters associated with the optimum refinement whichcorresponds to a value of R, = 0.089 (R, = 0.103). Fig.2 givesa perspective view of the molecule in its ultimate form.The detailed analysis of the electron-diffraction pattern thussupports the inferences drawn from the vibrational and NMRspectra, namely that the molecule is indeed 2-gallatetra-borane( 10) with the octahydrotriborate ligand bound to theGaH, fragment via two of its boron atoms and two singlehydrogen bridges. A comparison with the dimensions oftetraborane( 10) and some other octahydrotriborate derivativeswhose structures have been determined by diffraction tech-niques is presented in Table 5. There are three aspects whichmerit some comment.(a) r(Ga-B). At 231.2 pm, the Ga-B(1,3) distance is onlymarginally shorter than the corresponding distance (234.4 pm)in Me2GaB,H,,' as well as being comparable with theM-B( 1,3) distances in related molecules containing a bidentateB,H, moiety bound to a medium-sized metal atom M {e.g.[CU( B,H,)( PPh3)2],3 " [NMe,] [Cr(B,H 8)( CO),] 2b and[Mn,(B,H,)(CI-Br)(CO),] 32c}.On the other hand, it is con-siderably longer than the Ga-B distances of 216-218 pm foundin H2GaBH,,5*6 HGa(BH,), ' v 2 and Me,GaBH, where thegallium atom is bound to boron through a double hydrogenbridge. It is also appreciably longer than that in the closo-gallacarborane MeGaB,C,H, lo wherein direct gallium-boronbonding must be invoked with Ga-B distances (211 and 222pm) close to the sum of the tetrahedral covalent radii (ca. 214(6) r(Ga-Hb). The Ga-H, distance of 176.0 pm is consider-ably shorter than the corresponding distance (198.9 pm) inMe2GaB,H8,, and more nearly comparable with those in thegallium tetrahydroborates H,GaBH, (1 82.6 ~ m ) , ' , ~ HGa-(BH4)2 (177.0 pm), and Me,GaBH, (179.1 pm),,'O evenallowing for the relatively large uncertainty to which thisparameter is usually subject.Such a finding is particularlynoteworthy in relation to the dynamics of the H,GaB,H, andMe2GaB,H, molecules. The boron environments of Me,-GaB,H, undergo interchange which is rapid on the NMR time-scale at temperatures in excess of 250K and the moleculeassumes a wholly rigid structure only at 203 K; by contrast,H,GaB,H, presents an essentially rigid structure throughoutthis temperature regime. The fluxional behaviour may wellhinge on an initial bidentate - monodentate dissociativeprocess at the Ga(p-H),B, link;7 the shorter Ga-Hb distancereflects presumably a stronger bond in H2GaB3H8 than inMe,GaB,H,, implying a higher activation barrier to exchangein the former molecule.The difference arises in all probabilityfrom the charge distributions in the two molecules, with thegallium atom assuming a more positive charge in Me,GaB,H,than in H2GaB,H,.(c) r(Ga-H,). The Ga-H, distance is, at 144.2 pm, nominallyone of the shortest to be measured so far for a gallium hydride(cJ Ga2H6 151.9; H,GaBH, 158.6,5*6 [H,GaCl], 155.9,,*12Me,N-GaH, 149.7 28b and [Me2NGaHJ2 148.7 pm ,l). How-ever, the vibrational spectra reveal that the v(Ga-H,) modes alloccur at much the same energy (ca.1980 cm-') in unco-ordinated gallanes and 130-160 cm-' lower in energy in gallanecomplexes (e.g. H,POG~H,,~ Me,N-GaH, and Me,P-GaH,33b). The two sets of findings are difficult to reconcile,particularly as force constants, and hence vibrational wave-numbers, tend to be more sensitive to changes in bonding thando bond distances. There is no obvious reason for anybreakdown of the correlation between vibrational wavenumberand bond strength in the gallium hydrides, and the eccentricvariation of the measured Ga-H, distances may reflect under-estimated errors in the determination of these parameters. Sucherrors are likely to arise from the paucity of reliable independentinformation about the geometric and vibrational properties ofthe molecules.Research now under way is aimed at reducingthe limits of uncertainty affecting the results derived exclusivelyfrom electron-diffraction measurements. Preliminary inves-tigations suggest that the Ga-H, distance given here forPm)Table 5 Comparison of the molecular parameters of 2-gallatetraborane( 10) with those of tetraborane( 10) and other octahydrotriborate derivativesDistance/pm or angle/" *CompoundH2GaB3H8Phase/methodVapour/EDVapour/EDVapour/EDVapour/EDSolid/XRDSolid/XRDSolid/XRDSolid/XRDSolid/XRD170.5'170.5'170.5( 1.2)176.6(0.3)176(1)178(1)171(2)172.7(0.8)185.6'185.6'185.6(0.4)183.4(0.4)182(2)182(1)185(2){ ;:!:0.9)}B( 1,3)-Hb'126.4(0.7)l27.7( 1.7)127.0(2.4)13 1 S(0.9)1 15(9)115(7)108.8(5.8)1 05- 1 2 1 (2)112-1 17B(4)-Hb' B-H,126.4(0.7) 119.6-122.4'144.7( 1.7) 118-121(1)1 44.0( 2.4) 1 22- 1 25( 2)119.4(0.7),148.4(0.9) { 122.1 (1.4) }13&138(2) 105-1 14(2)152(9) 100-130I43( 7) 107- 122141-150 126-150143.8(2.7) -M-Hb176.0(2.8)198.9(4.8)190.6(4.1)148.4(0.9)150(2)185(5)178(6)182.0(4.6)150-1 76(I ED = Electron diffraction; XRD = X-ray diffraction.Figures in parentheses are the estimated standard deviations of the last digits. The labellingatom; H, = terminal H atom; Hb = bridging H atom; CI = angle between the planes MB( l)B(3) and B( 1)B(3)B(4). ' FixedJ. CHEM. SOC. DALTON TRANS. 1992 1517to the vibrational properties and microwave spectrum of thegaseous molecule.(vii) Reaction with Ammonia-With an excess of ammonia at195 K, 2-gallatetraborane reacts to give a white solid.Evapor-ation of the excess of ammonia leaves an involatile white residuewhich is long-lived at room temperature. At no stage isdihydrogen evolved. A mass balance of the reaction mixturereveals that the reacting proportions H,GaB3H8 : NH3 are 1 : 2.Details of the IR spectrum of the solid, including suggestedassignments, are itemised in Table 6. Clearly in evidence are theabsorptions characteristic of the anion B3H, -, as signalled bythe close resemblance which the spectrum bears to that of thesalt C S B ~ H ~ . ~ ' In addition, there are features diagnostic of co-ordinated leaving five bands (at 1966, 1941, 746,633 and 433 cm-') which are most plausibly identified withvibrations of the H,GaN2 skeleton of the cation[H,Ga(NH,),] + .The spectroscopic recognition of this cationis strongly supported by obvious parallels with the spectra ofthe compounds [H2Ga(NH3),]+C1-,13 [H,Ga(NH,),]+[GaH,] -,, [H,Ga(NH,),] +[BH,] -,6 and Me,GeH,.36 Henceit seems that the reaction proceeds in accordance with equation(9, with unsymmetrical cleavage of the Ga(p-H),B2 fragment of2-gallatetraborane( 10).Table 6 The IR spectral data (cm-') of the solid (as a CsI disc) derivedfrom the reaction of gallatetraborane with ammoniaSpeciesSolid a CsB,H, * Assignment2472s 24652423s 241 52366s 23552329m 23202225vw2125m 21202085m 20801966s1941s1606w (br)1399w (br)d13OOm1282m1179m1007m772s746s729s633w45om433m255170005775455v(N-H)In-phase vasym(BH2) of B(2),B(3)In-phase vSym(BH,) of B(2),B(3) +Vsym(BH2) of B(1)vasym(BH2) ofB(1)Out-of-phase vSyrn(BH2) of B(2),B(3)Out-of-phase vasym(BHZ) of B(2),B(3)vsym(B-Hb)vasyrn(B-Hb)Vasym(GaHJVsym(GaH2)6asyrn(NHJ6asymWH4 + )6syrn(NH3)WBH,)P(NH3)WaH2)p(GaH2)Sym.ring stretching, B3H8 -BH, wagging or rockingRing breathing, B3H8 -Ring deformation, B3H8-v(Ga-N)a Abbreviations: s = strong, m = medium, w = weak, v = very, sh =shoulder, br = broad. See ref. 36. For motions involving terminalhydrogen atoms of the B3H,- unit, the numbering scheme is such thatB(2) and B(3) are directly bonded boron atoms and B(l) is linked toB(2) and B(3) via single hydrogen bridges.dNH4+ present as animpurity arising from hydrolysis of the product.H2GaB3H, is closer to the mark than, say, the Ga-H, distanceof 158.6 pm deduced from similar, but incomplete, electron-diffraction experiments with H2GaBH,.'y6Table 3 also includes the results of some self-consistent field(SCF) ab initio molecular orbital calculations using a doublezeta with polarisation (DZP) basis set.17 There is reasonableagreement between the experimental and theoretical estimatesof several parameters (e.g. r[Ga-B( 1,3)], r(B-H) and thedihedral angle a } , but two notable discrepancies are apparent.First, the Ga-H, and Ga-Hb distances are calculated to beabout 10 pm longer than the experimental values we havedetermined.Secondly, the B(4)-Hb bond, with a length of141.6 pm, is calculated to be significantly longer than theother two B-Hb bonds {r(B(1,3)-Hb]}. In view of therelatively intractable correlation problems affecting theexperimentally determined B-H distances, it is quite possiblethat the theoretical value comes closer to the truth than theexperimental one, and the precedents set by related molecules,e.g. B4H10,30 tend to bear this out. In addition, the theoreticalstudies imply an H,-Ga-H, angle (1 30.2') substantially largerthan the value (115') to which we are led, not by independentrefinement, but by R-factor analysis. Irrespective of the precisebasis set adopted, all the ab initio calculations reported sofar 17*34 seem to point to a value near 130' for the H,-Ga-H,angle of the (H,),Ga(Xb), moiety (X = H or Cl).Althoughthis parameter cannot be determined very meaningfully fromthe electron-diffraction patterns of such molecules, preliminaryanalysis of the rovibrational spectrum of gaseous digallane 35argues for a value close to, if not greater than, that predictedby quantum-mechanical methods. There may be scope forinvestigating how the incorporation of some of the theoreticalparameters affects the refinement calculations for 2-galla-tetraborane( 10): better still would be the incorporation ofindependent experimental information relating, for example,(viii) Further Research.-Experiments are now under way, orare planned, to explore more fully the physical and chemicalproperties of 2-gallatetraborane( 10).Thus, it is hoped todetermine the rotational constants of the gaseous molecule in itsvibrational ground state by measuring the microwave spectrum,and then to carry out a combined analysis of the electron-diffraction, microwave, and vibrational properties (augmentedpossibly by the results of quantum-mechanical calculations)with the aim of gaining a superior structural definition of themolecule. The solid may also prove amenable to useful X-ray orneutron diffraction measurements. The thermal decompositionof the compound is also likely to repay more detailedinvestigation: thus, rapid, controlled thermolysis could providean expedient route to other gallium-substituted boranes, andthe nature of the solid with the composition GaB, also calls forcloser scrutiny.Moreover, the known reactivity of tetra-borane(lO),, holds out the promise of a rich and interestingchemistry for its 2-gallium-substituted derivative.ExperimentalSynthesis of 2-Gallatetraborane( lo).-All-glass apparatus ofthe sort described previously was rigorously preconditionedby heating under continuous pumping. In a representativeexperiment, monochlorogallane (ca. 180 mg, 1.7 mmol ofH,GaCl), itself prepared by the metathesis of gallium(rrr)chloride with an excess of trimethylsilane at ca. 250 K,12,13 wasco-condensed in uucuo with freshly recrystallised, powderedtetra-n-butylammonium octahydrotriborate (ca. 500 mg, 1.8mmol) at 77 K; the octahydrotriborate was prepared fromNaBH, (ex BDH) by the procedure developed by Gaines et aL3'Careful warming of the mixture to a temperature of ca.243 Kunder continuous pumping resulted in effervescence and theevolution of a volatile compound. With all parts of the glass-ware to which the vapours had access cooled to temperatures(283 K in a stream of cold nitrogen gas, the material wasfractionated between traps held initially at 210, 147 and 77 K.The main component collected in the trap at 147 K. After ca.1 h, while the reaction mixture was allowed gradually to warmto 268 K, the reaction appeared to be complete, and the variousfractions were isolated by sealing the traps at the relevantconstriction^.^ The fraction condensing at 147 K, shown to beessentially pure 2-gallatetraborane( lo), was kept at 77 K untilrequired, access to it being regained by way of a suitable break1518 J.CHEM. SOC. DALTON TRANS. 1992Table 7 Chemical analysis of 2-gallatetraborane( 10)Ga in B in Total B Total HFound Solid residue solid residue solid residue H2 evolved B,H6 evolved (solid + B2H6) (H, + B,H,)mmol - 1.069 2.1 19 3.28 0.542 3.205 9.82Proportion - 1 .oo 1.98 3.07 0.5 1 3 .oo 9.19Masslmg 97.4 74.5 22.9 1 6.61 15.0 34.65 9.90seal. The authenticity of such a sample was checked by referenceto its melting point (ca. 178 K), vapour pressure (ca. 1 mmHg at210 K), the IR spectrum of the vapour or of the solid condensateit forms at 77K, or the 'H or "B NMR spectrum of a['H8] toluene solution of the compound at low temperatures.The gallatetraborane(l0) amounted to ca.170 mg (1.5 mmol ofGaB,H,,), representing a yield of 88% based on equation (2)and the amount of monochlorogallane taken.Chemical Analysis.-A sample of the gallane was allowed todecompose in a preconditioned Pyrex ampoule of known massand volume equipped with a break-seal. It was left at roomtemperature for 24 h, and the temperature then raised to 333 Kfor 12 h to ensure complete decomposition. The volatilecontents of the ampoule were withdrawn and assayed, withfurther gentle heating of the solid residue to secure the release ofall the volatile material; this caused the appearance of the solidto change from a heterogeneous (metallic brown flecked withyellow inclusions) to a uniform brown one. The non-condens-able fraction consisted only of dihydrogen which was collectedand estimated by means of a Toepler pump.The only othervolatile product, which condensed in a spiral trap at 77 K, wasidentified by its IR spectrum4' as diborane, and this wasmeasured tensimetrically. The solid residue was dissolved in ARconcentrated nitric acid; its boron content was determined asboric acid (by titration in the presence of mannitol withstandard NaOH solution) and its gallium content by atomicabsorption spectroscopy. The results are presented in Table 7(Found: H, 8.3; B, 29.1; Ga, 62.6. GaB,H,, requires H, 9.0B, 28.9; Ga, 62.1%). The rather low figure for hydrogen wasalmost certainly caused by the retention of some elementalhydrogen by the solid residue.Reaction with Ammonia.-An excess of ammonia was co-condensed with a sample of the gallane at 77 K.Warming themixture to 195 K caused the ammonia to melt and a white solidto be precipitated. After 2 h at 195 K the mixture was warmed to238 K over a period of 1 h to ensure that the reaction proceededto completion. The excess of ammonia was recovered anddetermined tensimetrically. On the basis of the quantity ofammonia consumed (1.42 mmol) and the mass of the solidproduct (103 mg), the reacting proportions GaB,H,,: NH3were found to be 1 :2.02. A sample of the solid was powderedwith CsI (under dry nitrogen) and a disc pressed for IR measure-ments.Spectroscopic Measurements.-Infrared spectra were re-corded using one of two spectrometers, oiz. a Perkin-Elmermodel 580A dispersive instrument for the region 4000-200 cm-'or a Mattson Galaxy FT-IR instrument for the region 4000-400 cm-'.Solid nitrogen matrices, typically with compositionsN,:GaB,H,, estimated to be ca. 200:1, were prepared bycontinuous deposition of the gallane vapour (delivered via anappropriately cooled, preconditioned, glass inlet tube) with anexcess of the matrix gas on a CsI window cooled to ca. 20 K bymeans of a Displex closed-cycle refrigerate-r (Air Productsmodel CS 202); fuller details of the relevant equipment andprocedures are given el~ewhere.~' Raman spectra of solidsamples were excited at h = 514.5 nm with the output of aSpectra-Physics model 165 argon-ion laser and measured with aSpex Ramalog 5 spectrophotometer operating in conjunctionwith a Glen-Creston SCADAS data-handling system; theresolution was normally ca.5 cm-'. Solid films of the gallanewere presented for spectroscopic analysis by allowing thevapour to condense on a CsI window (for IR measurements) ora copper block (for Raman measurements) contained in anevacuated glass shroud and maintained at 77 K. Vapoursamples of the gallane for IR measurements were contained in aPyrex-bodied cell equipped with CsI windows and having apathlength of ca. 10 cm.The NMR measurements on C2H8]toluene solutions at tem-peratures in the range 193-300 K were made at 300 (for 'H) or96.25 MHz (for "B) with a Bruker model AM 300 spectro-meter. The mass spectrum of the gallane vapour was measuredusing a VG SPX800 Spectromass quadrupole mass spectrometerat the University of Reading; the ionising potential was 70 eV(ca.1.12 x J) and the system was controlled by aSpectrolab version 4 computer program. For these measure-ments the vapour over the sample at 238 K was admitted to aglass vacuum manifold via a ground-glass joint lubricated withhalocarbon grease. The manifold was preconditioned byexposure for 5 min to each of two samples of the vapour; afterre-evacuation, a fresh sample was admitted and from themanifold it was caused to bleed into the ionisation chamber ofthe spectrometer uia a greaseless valve.Electron-diffraction Measurements.-The electron-diffractionpatterns of 2-gallatetraborane( 10) vapour were recorded photo-graphically on Kodak Electron Image plates using theEdinburgh gas diffraction apparatus.42 To accommodate thereactivity and thermal frailty of the gallane, we used the all-glassinlet assembly described previ~usly,~ but the more robustnature of the compound (compared, say, with Ga2H6 andGaBH,) allowed some modification to be made. Thus, theassembly was fitted with a greaseless (Teflon) valve and ground-glass socket to enable samples to be changed so that thescattering of more than one sample could be recorded in a singleexperiment.The experience gained in manipulating gallatetra-borane suggested that decomposition was slow under theconditions of the experiments provided that the glass surfaceshad been adequately conditioned by prior exposure to a sampleof the vapour.The ampoule containing the sample of 2-galla-tetraborane(l0) was kept at 253 K and the inlet system wascooled to 273 K. Before any measurements were made the inletsystem was conditioned by exposing it to a slow stream of thegallane vapour for 2 min. The scattering pattern was thenrecorded at the two nozzle-to-plate distances listed in Table 8;also included in this table are details of the electronwavelengths, the weighting functions used to set up the off-diagonal weight matrix, the correlation parameters and therefined scale factors. The rotating sector was kept in motionuntil all of the sample in the diffraction chamber had beencondensed in the cold trap. This prevented a 'shadow' of thestationary sector from being superimposed on the photographicplate, leaving the plate fogged (by the vapour) but with ahomogeneous background. Each exposed plate was left underpumping for 4 h before removal, washed, and left in the air for24 h before being developed.The precise nozzle-to-platedistances and electron wavelengths were determined fromscattering patterns for benzene vapour recorded immediatelyafter the sample pattern.Details of the electron-scattering patterns were collected iJ. CHEM. SOC. DALTON TRANS. 1992 1519Table 8 Nozzle-to-plate distances, weighting functions, correlation parameters, scale factors and electron wavelengthsNozzle-to- Electronplate Correlation, Scale wavelength b/distance/mm Aslnm-' s,,,/nm-' sw,/nm-' sw,/nm-' s*,,/nm-' Plh factor, k" pm259.48 2 20 40 140 160 - 0.0679 0.71 l(25) 5.67120 1.08 4 52 72 176 208 0.4494 0.549(26) 5.670' Figures in parentheses are the estimated standard deviations of the last digits.Determined by reference to the scattering pattern of benzene vapour.digital form using a computer-controlled Joyce-Loebl MDM6microdensitometer with the scanning program describedp r e v i ~ u s l y . ~ ~ Calculations made use of well established pro-grams for data reduction 43 and least-squares refinement,,' thecomplex scattering factors being those listed by Fink andRow4,Chemicals.-Nitrogen gas (research grade) for matrix studieswas used as received from the British Oxygen Co. (B.O.C.). Thefollowing reagents, from the commercial sources indicated, werealso used as received: gallium metal (BDH or Aldrich), C1,(BDH), NaBH, (BDH), LiA1H4 (BDH), I, (BDH), Me,SiCl(Aldrich) and Bu",NI (Aldrich); NH, (B.O.C.) was dried overNa and purified before use by fractionation in UCLCUO.Mono-chlorogallane was prepared by the interaction of gallium(Ii1)chloride (derived from the elements and purified by vacuumsublimation) with trimethylsilane [derived from the reaction ofMe,SiCl with LiAlH4 in tetrahydrofuran (thf)],45 at250 K.6*'23'3 The solvents C2H,]toluene (Aldrich), Et,O(Fisons), thf (Fisons) and diglyme (2,5,8-trioxanonane)(Aldrich) were dried and purified by standard procedures.AcknowledgementsWe thank the SERC for the funding of a research studentshipand assistantship (for C.R.P.) and for financial support of theresearch, including the Edinburgh Electron Diffraction Serviceand the provision of the microdensitometer facilities atDaresbury.We are grateful also to Dr. P. T. Brain for hispractical assistance with the NMR measurements and to Dr.M. J. Almond for access to the mass-spectrometric facilitiesat Reading.ReferencesI A. J. Downs and P. D. P. Thomas, J. Chem. SOC., Chem. Commun.,1976, 825; M. T. Barlow, C. J. Dain, A. J. Downs, G. S. Laurensonand D. W. H. Rankin, J. Chem. SOC., Dalton Trans., 1982,597.2 P. D. P. Thomas, D.Phi1. Thesis, University of Oxford, 1977; A. J.Downs, L. A. Harman, C. R. Pulham, D. W. H. Rankin, H. E.Robertson and P. D. P. Thomas, unpublished work.3 A. J. Downs, M. J. Goode and C. R. Pulham, J.Am. Chem. SOC., 1989,111, 1936.4 C. R. Pulham, A. J. Downs, M. J. Goode, D. W. H. Rankin and H. E.Robertson, J. Am. Chem. SOC., 1991,113,5149.5 C. R. Pulham, P. T. Brain, A. J. Downs, D. W. H. Rankin and H. E.Robertson, J. Chem. Soc., Chem. Commun., 1990,177.6 C. R. Pulham, D.Phi1. Thesis, University of Oxford, 1991.7 J. Borlin and D. F. Gaines, J. Am. Chem. SOC., 1972,94, 1367.8 C. J. Dain, A. J. Downs and D. W. H. Rankin, J. Chem. SOC., DaltonTrans., 198 1, 2465.9 C. P. Magee, L. G. Sneddon, D. C. Beer and R. N. Grimes, J.Organornet. Chem., 1975,86, 159.10 R. N. Grimes and W. J. Rademaker, J. Am. Chem. SOC., 1969, 91,6498; R. N. Grimes, W. J. Rademaker, M. L. Denniston, R. F. Bryanand P. T. Greene, J. Am. Chem. Soc., 1972,94, 1865.1 1 M.A. Bandman, C. B. Knobler and M. F. Hawthorne, Znorg. Chem.,1989, 28, 1204; D. M. Schubert, M. A. Bandman, W. S. Rees, jun.,C. B. Knobler, P. Lu, W. Namand M. F. Hawthorne, Organometallics,1990, 9, 2046.12 M. J, Goode, A. J. Downs, C. R. Pulham, D. W. H. Rankin and H. E.Robertson, J. Chem. SOC., Chem. Commun., 1988,768.13 M . J. Goode, D.Phi1. Thesis, University of Oxford, 1987.14 S. J. Hildebrandt, D. F. Gaines and J. C. Calabrese, Znorg. Chem.,15 C. R. Pulham, A. J. Downs, D. W. H. Rankin and H. E. Robertson,16 C. R. Pulham, unpublished work.17 B. J. Duke and H. F. Schaefer 111, J. Chem. Soc., Chem. Commun.,1991,123.18 R. H. Moss, Chem. Br., 1983,19,733; D. J. Cole-Hamilton, Chem. Br.,1990,26,852; R. D. Dupuis, Science, 1984,226,623; R.J. M. Griffiths,Chem. Znd. (London), 1985, 247; G. J. Davies and D. A. Andrews,Chemtronics, 1988, 3, 3; The Chemistry of the SemiconductorIndustry, eds. S. J. Moss and A. Ledwith, Blackie, Glasgow andLondon, 1987.19 (a) B. Post, in Boron, Metallo-Boron Compounds and Boranes, ed.R. M. Adams, Interscience, New York, 1964, p. 301; N. N.Greenwood, R. V. Parish and P. Thornton, Q. Rev. Chem. SOC., 1966,20, 441; N. N. Greenwood and B. S. Thomas, The Chemistry ofBoron, Pergamon, Oxford, 1975, p. 697; (b) P. Laveant, Rev. Chim.Miner., 1965, 2, 175.1978, 17,790.J. Chem. SOC., Chem. Commun., 1990, 1520.20 A. J. Dahl and R. C. Taylor, Znorg. Chem., 1971,10,2508.21222324252627282930313233P. L. Baxter, A. J. Downs, D.W. H. Rankin and H. E. Robertson,J. Chem. SOC., Dalton Trans., 1985,807.D. F. Gaines and S. J. Hildebrandt, Znorg. Chem., 1978, 17, 794.T. J. Marks and J. R. Kolb, Chem. Rev., 1977,77,263.N. N. Greenwood and B. S. Thomas, The Chemistry of Boron,Pergamon, Oxford, 1975; Boron Hydride Chemistry, ed. E. L.Muetterties, Academic Press, New York, 1975; Supplement toMellor 's Comprehensive Treatise on Inorganic and TheoreticalChemistry, Longman, London, 1981, vol. 5, Part BI; Aduances inBoron and the Boranes, eds. J. F. Liebman, A. Greenberg and R. E.Williams, VCH, Weinheim, 1988; Gmelin Handbook of InorganicChemistry, 8th edn., 1977-1987, Syst. No. 13, Parts 14, 18 and 20,2ndand 3rd Supplements, vol. 1.E. Dayan, G. Widenlocher and M. Chaigneau, C.R.Hebd. SeancesAcad. Sci., 1963,257, 2455.J. B. Leach, T. Onak, J. Spielman, R. R. Rietz, R. Schaeffer and L. G.Sneddon, Znorg. Chem., 1970,9,2170.(a) D. F. Gaines, J. L. Walsh, J. H. Morris and D. F. Hillenbrand,Znorg. Chem., 1978, 17, 1516; (b) A. D. Norman and R. Schaeffer, J.Phys. Chem., 1966,70,1662; D. Marynick and T. Onak, J. Chem. SOC.A, 1970, 1160; H. Beall, C. H. Bushweller, W. J. Dewkett and M.Grace, J . Am. Chem. SOC., 1970,92,3484.(a) P. L. Baxter, A. J. Downs, M. J. Goode, D. W. H. Rankin andH. E. Robertson, J. Chem. Soc., Chem. Commun., 1986,805; J . Chem.SOC., Dalton Trans., 1990, 2873; (b) P. L. Baxter, A. J. Downs andD. W. H. Rankin, J. Chem. SOC., Dalton Trans., 1984, 1755.(a) M. T. Barlow, A. J. Downs, P. D. P. Thomas and D. W. H.Rankin, J. Chem. SOC., Dalton Trans., 1979, 1793; (b) C. J. Dain, A. J.Downs, M. J. Goode, D. G. Evans, K. T. Nicholls, D. W. H. Rankinand H. E. Robertson, J. Chem. Soc., Dalton Trans., 1991,967.C. J. Dain, A. J. Downs, G. S. Laurenson and D. W. H. Rankin,J. Chem. SOC., Dalton Trans., 1981,472.A. S. F. Boyd, G. S. Laurenson and D. W. H. Rankin, J. Mol. Struct.,1981,71,217.(a) S. J. Lippard and K. M. Melmed, Znorg. Chem., 1969,8, 2755; (6)L. J. Guggenberger, Inorg. Chem., 1970,9,367; (c) M. W. Chen, D. F.Gaines and L. G. Hoard, Znorg. Chem., 1980, 19, 2989; (d) J. C.Calabrese, D. F. Gaines, S. J. Hildebrandt and J. H. Morris, J. Am.Chem. Soc., 1976, 98, 5489; (e) P. D. Grebenik, J. B. Leach, J. M.Pounds, M. L. H. Green and P. Mountford, J. Organomet. Chem.,1990,382, C1; J. M. Pounds and P. Mountford, personal communica-tion.(a) J. R. Durig, K. K. Chatterjee, Y. S. Li, M. Jalilian, A. J. Zozuli1520and J. D. Odom, J. Chem. Phys., 1980,73,21; (6) J. D. Odom, K. K.Chatterjee and J. R. Durig, J. Phys. Chem., 1980,84, 1843.34 C. Liang, R. D. Davy and H. F. Schaefer 111, Chem. Phys. Lett., 1989,159,393; K. Lammertsma and J. Leszczynski, J. Phys. Chem., 1990,94, 2806; B. J. Duke, J. Mol. Struct., 1990, 208, 197; V. Barone,C. Minichino, F. Lelj and N. Russo, J. Comput. Chem., 1988,9, 518;K. Lammertsma and J. Leszczynski, J. Chem. SOC., Chem. Commun.,1989,1005.35 C. R. Pulham, I. M. Mills and A. J. Downs, unpublished work.36 D. F. Van de Vondel and G. P. Van der Kelen, Bull. SOC. Chim. Belg.,1965,74,467.37 J. Tomkinson, C. J. Ludman and T. C. Waddington, Spectrochim.Acta, Part A, 1979,35, 117.38 K. H. Schmidt and A. Miiller, Coord. Chem. Rev., 1976, 19, 41;K. Nakamoto, Infrared and Raman Spectra of Inorganic andCoordination Compounds, 4th edn., Wiley-Interscience, New York,1986.39 D. F. Gaines, R. Schaeffer and F. Tebbe, Inorg. Chem., 1963, 2, 526;H. C. Miller and E. L. Muetterties, Inorg. Synth., 1967,10,82.J. CHEM. SOC. DALTON TRANS. 199240 W. L. Smith and I. M. Mills, J. Chem. Phys., 1964, 41, 1479; J. L.Duncan, D. C. McKean, I. Torto and G. D. Nivellini, J. Mol.Spectrosc., 198 1,85, 16.41 See, for example, M. Hawkins and A. J. Downs, J. Phys. Chem., 1984,88, 1527,3042.42 C. M. Huntley, G. S. Laurenson and D. W. H. Rankin, J. Chem. Soc.,Dalton Trans., 1980,954.43 S. Cradock, J. Koprowski and D. W. H. Rankin, J. Mol. Strucf., 1981,77, 113.44 M. Fink and A. Ross, International Tables for Crystallography,International Union of Crystallography, Reidel, Dordrecht, in thepress.45 A. E. Finholt, A. C. Bond, jun., K. E. Wilzbach and H. I. Schlesinger,J. Am. Chem. Soc., 1947,69,2692.Received 6th January 1992; Paper 2/00047
ISSN:1477-9226
DOI:10.1039/DT9920001509
出版商:RSC
年代:1992
数据来源: RSC