摘要:
J. CHEM. SOC. DALTON TRANS. 1992 769Origin of the Two Significantly Different W-CI Bond Lengthsfor Chemically Equivalent Bonds in mer-[WCI,( PMe,Ph),l tKeum Yoon.B Gerard Parkin."Se David L. Hughesb and G. Jeffery Leighba Department of Chemistry, Columbia University, New York, NY 70027, USABrighton BN19R0, UKAFRC Institute of Plant Science Research, Nitrogen Fixation Laboratory, University of Sussex,The observation of two significantly different W-CI bond lengths for chemically equivalent bonds,previously reported for mer- [WCl,( PMe,Ph),], has been reinterpreted as an artifact due to cocrystallizationwith an isostructural 0x0 impurity, cis-mer- [WOCI,( PMe,Ph),].The synthesis and molecular structure of the tungsten(rI1)complex mer-[WCI,( PMe,Ph),] have been recently reported.'Tungsten(rI1) complexes are relatively rare and mer-[WCI,-(PMe'Ph),] represents the first neutral mononuclear tung-sten(rI1) derivative to be structurally characterized by X-raydiffraction (Fig.1). Of particular interest, it was revealed thatalthough the three W-P bond lengths [2.514(1), 2.536(1) and2.555( 1) A] were similar, the three W-CI bond lengths [2.295(2),2.437( 1) and 2.441(1) A] were significantly different, with onebeing exceptionally short. The most surprising observation wasthat the short W-CI bond length did not correspond to theunique chloro substituent that was located trans to PMe,Ph,but rather corresponded to one of the chloro substituents thatare mutually trans. Since this complex has a molecular plane ofsymmetry, the unexpected result that two chemically equivalentW-CI bonds may have substantially different bond lengths[2.437(1) and 2.295(2) A] is obtained.Here we report areinvestigation of the molecular structure of mer-[WCl,-(PMe'Ph),] and offer an explanation for the observation ofsignificantly different W-Cl bond lengths for chemicallyequivalent bonds.Results and DiscussionThe subjects of the original reports of 'distortional'' or 'bond-stretch' , isomerism were molybdenum 0x0 complexes of theclass cis-mer-[MoOCI,( PR,),]. We have recently reinvestigatedthe structures of these complexes in more detail, and determinedthat the observation of apparent 'isomers' with unusually longMo=O bond lengths is an artifact due to cocrystallization withthe trichloride impurity mer-[MoCI,( PR3)3].4 Incorporation ofchloride into the 0x0 site results in an apparent increase of the'Mo=O' bond length, since the chloride ligand is located at amuch greater distance from the molybdenum centre (ca.2.45 Afor Mo-C1 uersus ca. 1.68 A for Mo=O). In order to demonstratefurther that cocrystallization may result in the determination ofincorrect bond lengths, we have also investigated the apparentstructural effects of doping a series of tris(3-tert-butylpyrazo1yl)-hydroboratozinc complexes, [ZnX(HB(3Bur-pz),)] (X = C1,Br, I or CH,) with isostructural impurities.' In each case, only asingle 'atom' is observed at the disordered site, corresponding toan upparetit Zn-X bond length that is a composite for the pair ofcomplexes concerned.In view of the above studies, and also the close similarity ofmer-[WCI,( PMe,Ph),] to the aforementioned molybdenumPMe2Phhe2PhFig. 1system, we considered that a likely explanation for theobservation of significantly different W-CI bond lengths forchemically equivalent bonds may be cocrystallization of mer-[WCl,(PMe,Ph),] with the isostructural 0x0 complex cis-mer-[WOCI,(PMe,Ph),].Thus, in an analogous fashion to the factthat the incorporation of chloride into the 0x0 site of cis-mer-[MoOCI,( PMe,Ph),] results in an apparent lengthening of theMo=O bond: the incorporation of 0x0 into one of the chloridesites of mer-[WCI,(PMe,Ph),] would be expected to result inan apparent shortening of the W-Cl bond.Evidence that the above rationale may be likely was providedby examination of the ,'P-{ 'H} NMR spectrum.The spectrumof the 0x0 complex cis-mer-[WOCl,(PMe2Ph),] matchedexactly, both in chemical shift and also W-P coupling constants,that reported for mer-[WCI3(PMe2Ph),].$.' Since mer-[WCI,-(PMe'Ph),] is paramagnetic, it is not unexpected that thisspecies would not be readily observed by 3'P-(1H) NMRspectroscopy, thus resulting in only the diamagnetic cis-mer-[WOCI,(PMe,Ph),] impurity being detected. Although wecould not observe ,' P-('H) NMR resonances assignable tomer-[WC13( PMe'Ph),], paramagnetically shifted 'H NMRresonances are readily observed, in agreement with theoriginal report. The most distinctive features of the ' H NMRspectrum are the two resonances at 6 -24.7 and - 16.4 (inC6D6) in the ratio 2: 1, assignable to the two sets of methylgroups of the trans- and cis-PMe,Ph ligands, respectively.However, contamination by the diamagnetic 0 x 0 complex cis-mer-[WOC1,(PMeZPh),1 in this sample is revealed by theobservation of two triplets at 6 1.92 and 1.67, and adoublet at 6 1.31, due to the methyl groups of the PMe,Phligands.The ' H and 3'P-{'H) NMR studies described above clearlyindicate that the bulk sample of nter-[WCl,( PMe,Ph),] iscontaminated with cis-mer-[ W0Cl2( PMe,Ph),], the origin ofwhich is unknown.As such, the possibility exists that co-crystallization may occur and thereby result in the observ-ation of an uppcirerrtlr short W-CI bond length.Molecular structure of mer-[WCl,(PMe,Ph),]-t Supplcnienfary iiaara uwifubii~: see Instructions for Authors, J.C'iti~r?i.Soc., Dularon Trans., 1992, Issue I , pp. xx-xxv.However, i t should be noted that the 'J,,couplingconstants reportedin ref. 1 are half the correct values770 J. CHEM. SOC. DALTON TRANS. 1992w)- IFig. 2 An ORTEP drawing of cis-nter-[WOCI,( PMe,Ph),]Table 1lo3) for c*is-nter-[ WOCI,( PMe, Ph),]Atom coordinates ( x lo4) andX2 131(1)2 442( I )3 535( 1)1012(1)3 290( I )1 869(2)1 332(3)1 428(6)3 123(6)2 973 5)3 657(6)4 032( 7)3 737(7)3 048(6)2 683(6)4 530(5)3 984(6)3 433(5)4 226(6)4 158(7)3 31 l(7)2 558(6)2 612(6)1 245(7)- 134(6)836(5)93(5)21(7)697( 7)1 440(7)1519(6)Y4 923( 1)6 603(2)3 612(2)6 229(2)6 037(2)3 374(2)4 303(5)7 366(9)7 972(8)6 083( 7)6 769(9)6 3 1 3( 1 1)5 183(10)4 499( 10)4 961(7)4 548(8)2 718(8)2 393(7)1 883(8)I 026(8)673(9)1 137(8)I 984(8)6 266(9)5 542( 10)7 928(7)8 397(9)9 679( 10)10 524( 10)10 088( 8)8 798(7)thermal parameters (A2 xZ1 708( 1)876( I )1 909( I )2 029( I )2 924( 1)2 636( I )782(3)1 20( 5 )I414(6)206( 4)104(5)-415(6)- 828(6)- 750( 5)- 228( 5)2 OOl(5)2 861(5)1 147(4)1 115(5)5 17(6)- 77( 6)- 37(6)568( 5 )3 104(5)1525(6)1 771(5)1 123(5)925(6)1 385(7)2 041(7)2 232(5)* Equivalent isotropic U defined as one third of the trace of theorthogonalized U j j tensor.The molecular structure of ci.s-nzc~r-[WOCl,( PMe,Ph),] wasdetermined by X-ray diffraction and found to be bothisomorphous and isostructural with tm~r.-[WCI,(PMe,Ph)J.An ORTEP drawing is presented in Fig.2, and atomiccoordinates are given in Table I , with selected bond lengths andangles in Tables 2 and 3. The W=O bond length is 1.752(4) A;however, in view of the problems associated with thecorresponding molybdenum complexe~,~ we hesitate to attachFig. 3 An ORTEP drawing of mer-[WC1,(PMe2Ph),]Table 2 Comparison of bond lengths (A) for mer-[WCI,(PMe,Ph),]and cis-mrr-[WOCI,(PMe,Ph),][wcl3( PMe2 Ph )3]LI h c' [WOCl,( PMePh),] 'W-P(1) 2.513(5) 2.514(1) 2.513(5) 2.483(2)W-P(2) 2.561(5) 2.555(1) 2.555( I ) 2.538(2)W-P(3) 2.542(5) 2.536( I ) 2.536( 1 ) 2.508(2)W-CI( I ) 2.423(4) 2.437( I ) 2.437( 1 ) 2.493(2)W-Cl(2) 2.428( 5) 2.44 1 ( I ) 2.442( 1 ) 2.478(3)W-Cl(3) 2.342(4) 2.295(2) 2.3 132)w-0 I .752(4)a This work.Ref. I . Labelling scheme is modified from the originalreport forcomparison purposes. Re-evaluated data from ref. 1 (see text).Table 3 Selected bond angles ( ) for ci.s-nier-[WOCI,( PMe,Ph),]P( I)-W-P(2)P( 2)- w- P( 3)P( 2)-W-CI( 1 )P( I )-w-CI( 2)P( 3)-W-C1( 2)P( I )-w-0P(3)-W-0Cl(Z)-W-O94.2( 1 )160.5( 1 )80.0( 1 )175.2( 1 )81.9( 1 )86.6( 2)95.9(2)97.8( 2)P( 1 )-w-P(3)P( 1 )-w-CI( 1 )P(3)-W-CI( I )P( 2)-W-C1( 2)CI( 1 2 ) )-w-CI(P( 2)-w-0CI( 1 )-w-0w-P( I )-C( 1 1 )95.7( 1 )86.2( 1 )84.0( 1 )86.8( 1 )89.4( 1 )101.4( 2)172.8(2)114.4(4)too much significance to this bond length, which is slightlylonger than the average (1.70 A) for monooxotungsten com-p l e x e ~ .~A crystal of nrur-[WCI,(PMe,Ph),], shown by ' H NMRspectroscopic analysis to come from a sample that was con-taminated with e - ~ . 20 molar O 0 c~is-nrc~r-[WOCl,(PMe,Ph),],was investigated by X-ray diffraction. Atomic coordinates forme)r-[WCI,( PMe,Ph),] are presented in Table 4, with selectedbond lengths and angles in Tables 2 and 5. An ORTEP drawingis illustrated in Fig. 3. Although many of the bond lengths forthe structure of IIIPI'-[WCI,( PMe2Ph),] reported here aresimilar to those in the previous report,' a significant difference isobserved for the two trtrns W-CI bonds, as illustrated in Table 2,which provides a comparison of the co-ordination environmentfor the two structures of n~cv--[WCl,(PMe~Ph),] and alsofor cis-nrcr.-[WOCI,( PMe,Ph),].Significantly, the very shortW-CI bond length [2.295(2) A] reported for the originaJ. CHEM. SOC. DALTON TRANS. 1992 77 ITable 4lo3) for mer-[WCI,(PMe,Ph),]Atom coordinates ( x lo4) and thermal parameters (A2 xY2 199(1)2 441(3)3 548(3)1 030(3)3 321(3)1 905(3)1050(3)1 430( 13)3 113(14)2 972( 10)2 641( 12)2 997( 15)3 684( 15)4004(15)3 662( 13)4 007( 12)4 539( 11)3 482( 1 1)4 242( 14)4 221(15)3 399( 18)2 614( 14)2 654( 13)1 282( 13)850( 11)92( 12)670( 17)1412(16)1510(13)- 99( 12)-3(15)1'4 979( I )6 6 12(4)3 601(4)6 272(4)6 067(4)3 413(4)4 092( 5 )7 359( 17)7 995( 15)6 036( 15)4 946( 17)4 475( 19)5 152(19)6 229(20)6 717(18)2 673( 18)4 568( 16)2 429( 14)1914(18)1 062(2 I )674(23)1 065( 18)I 933( 15)5 559( 18)6 376( 16)7 967( 15)8 377( 18)9 642( 19)10 51 l(19)10 121( 18)8 857( 16)1 825(1)1915(2)2 114(3)2 978(2)2 688(3)683(3)175(10)1401(22)243(9)904(2)- 209( 10)-746(11)- 8 18( 13)-400(13)149( 12)2 844( 10)2 041(12)1 151(9)1111(11)- 48( 13)20( 12)614( 10)1 690(13)3 182(9)1801(8)I 176(11)929( 13)1 350( 15)1 968( 14)2 214(12)5 17( 12)* Equivalent isotropic U defined as one third of the trace of theorthogonalized U i j tensor.Table 5 Selected bond angles (") for mer-[WCI,(PMe,Ph),]P( 1 )-w-P(2)P( 2 )- w- P( 3 )P( 2)-W-C1( 1 )P( 1 )-w-CI( 2)P(3)-W-C1(2)P( I )-W-C1(3)P( 3)-w-c1(3)CI(Z)-W-CI( 3)93.4(2) P( 1 )-W-P(3) 96.4(2)89.7( 1 )83.0( 1 ) P( 3)-W-CI( 1 ) 86.7( 1 )177.9( 1 ) P( 2)-W-CI( 2) 88.4( 2)82.0(2) C1( I )-W-C1(2) 9 I .7( 1 )I65.7( 1 ) P( 1 )-W-CI( 1 )87.7(2) P( 2)-W-CI( 3) 10 1.2( 2)89.6( 2) CI( l)-W-C1(3) 175.2(2)90.8( 2)Fig.4 Partial ORTEP drawing of n~(.r-[WCl,(PMe~Ph)~lstructure is not reproduced for the structure described here;instead, a substantially longer W-CI bond length of 2.342(4) Ais observed. Although the two truns W-CI bond lengths are stillnot identical, the difference between the two structures providesgood evidence that the uppurently short W-CI bond lengthis due to cocrystallization with cis-mer-[WOCI,( PMe,Ph),]impurity, the two structures differing slightly due to differentlevels of contamination.Indeed, the structure of pure mer-[WCI,(PMe,Ph),] remains to be determined.The observation of two significantly different W-CI bondlengths for chemically equivalent bonds in mer-[WCI3( PMe2-Ph),] is, therefore, an artifact due to cocrystallization withcis-mer-[WOCI,(PMe,Ph),]. The common belief that singlecrystals are invariably pure, coupled with the difficulty ofdetecting certain impurities by X-ray diffraction, accounts forthe fact that this explanation was not originally considered.The question arises as to why the presence of the 0x0 impurityin mer-[WCI,( PMe,Ph),] was not detected crystallographi-cally, especially since it may be expected that disorder betweenoxygen and chlorine would be readily revealed by the presenceof abnormal thermal parameters. Figs.4 and 5 illustrate partialORTEP drawings for mer-[WCI3(PMe2Ph),] and cis-mer-[WOCI2(PMe2Ph),] in a plane defined by the atoms shown.Clearly, the thermal parameters associated with Cl(3) of mer-[WCI,(PMe,Ph),] are larger than those for both Cl(1) andCI(2). Larger thermal parameters would be expected for refininga site incorrectly as chlorine, if the site was partially occupied byan atom with lower atomic number, e.g. oxygen. However, arethe thermal parameters of Cl(3) anomalously large? Alter-natively, could the thermal parameters associated with Cl( 1 )and Cl(2) be anomalously small? It is not easy to answer thisquestion definitively on the basis of the ORTEP drawing shown,although it may be ussumed that Cl(1) and Cl(2) are correctsince they are more similar.Electron-density maps for mer-[WCI,(PMe,Ph),] and cis-mer-[WOCI,(PMe,Ph),] are shown in Figs.6 and 7, re-spectively. The projections shown correspond to those of thepartial ORTEP drawings in Figs. 4 and 5, and the contours arespaced at the same interval for both plots. It is apparent fromFig. 6 that there is slightly less electron density at C1(3), withseven electron-density contours, compared with Cl( 1 ) andC1(2), each of which are characterized by nine contours. Thisobservation is also supported by an electron-density differencemap for mer-[WCI,( PMe,Ph),].Reduced electron density atthe site occupied by Cl(3) would naturally be expected if the sitewas partially occupied by an oxygen atom. However, it is notclear that this evidence alone would lead to a convincingsuggestion that the site was compositionally disordered.Supporting evidence for compositional disorder in mer-[WCI,(PMe,Ph),] has been obtained by refining the occu-pancies of a composite atom composed of both CI and 0located at the site 'Cl(3)'. This procedure resulted in the siteoccupancies CI [0.80(3)] and 0 [0.20(3)], identical to the valuesestimated by 'H NMR analysis of the bulk sample. I t should benoted that performing a similar operation for the orderecl siteCl(1) resulted in the occupancies CI [ l .O l ( l ) ] and 0 [O.Ol(l)],as expected for occupancy only by chlorine. Thus, for thissystem (although not necessarily for other systems), refining siteoccupancies provides a useful indication of compositionaldisorder.We have also re-evaluated the data of Hills e f cii.,' forwhich ' H NMR spectroscopy also revealed the presence of the0x0 impurity cis-nw-[WOCI,( PMe,Ph),]. Using a similarapproach, the occupancy (S) of the Cl(3) atom site was allowedto vary, and an additional oxygen atom O(3) with partialoccupancy ( 1 - S ) was placed along the W-Cl(3) vector, withd(W-0) = 1.7 A. After several cycles of refining the atomcoordinates and either S or the thermal parameters of the Cl(3)and O(3) atoms, followed by two cycles on the whole structure,we found that the R factors were reduced significantly toR = 0.041 and R' = 0.040.Importantly, the site occupancy forCl(3) refined to c u . 0.79, clearly supporting a model involvingcompositional disorder. The result of this refinement procedurewas a slight increase in the apparent W-Cl(3) bond length to2.31 3 2 ) A, but the location of the oxygen atom was not precise772 J. CHEM. SOC. DALTON TRANS. 1992CI(2)Fig. 5 Partial ORTEP drawing of cis-mer-[WOCl,(PMe,Ph),lFig. 6illustrated in Fig. 4Electron-density map for mer-[WCI,(PMe,Ph),] in the planeFig. 7plane illustrated in Fig. 5Electron-density map for c.i.~-ntcr-[WOCI,(PMe,Ph)~] in thewith d(W-0) = 1.61(2) A. Changes in the rest of the moleculewere not significant (see Table 2).This kind of analysis can onlybe taken so far, because in addition to the chlorine/oxygenreplacement, the cocrystallized molecules are not of identicaldimensions. Allowing only one set of structural parameters tovary during refinement is clearly an approximation, and itshould be noted that the apparent W-Cl(3) bond lengthobtained by this procedure is still significantly different to thatobserved for W-CI( 1) [2.437( 1 ) A].It is not entirely clear why this compositional disorder isconfined to the Cl(3) position and is negligible at Cl(1).However, it is evident that the Cl(1) and Cl(3) sites areinequivalent, due to the conformations of the substituents onthe phosphine ligands, and it appears that the Cl(3) site is moresterically demanding than is the C1( 1) site.Therefore, on thisbasis, we suggest that Cl(3) would be the site that would bepreferentially occupied by the smaller oxygen atom, and therebyallow disorder preferentially at only one site.The effect that an impurity may have upon a structuredetermined by X-ray diffraction can be very subtle. We haveshown that, depending upon the nature of the impurity, themost sensitive probe for the presence of an impurity at acompositionally disordered site may be the apparent bondlength itself, and not the observation of abnormal thermalparameter^.^.' Coupled with the common belief that singlecrystals are invariably pure, it is understandable how easy it is tomisinterpret X-ray diffraction data for such cases. For example,the problem described above is reminiscent of that observed forthe dimethyl derivative [Hf(qS-CSH,),Me2] in which twosignificantly distinct Hf-Me bond lengths C2.31 8(8) and 2.382(7)A] were originally reported.8 This result has subsequently beenreinterpreted as arising from cocrystallization with the chloridederivative [Hf(q '-CS H s)2 MeC1].9ConclusionThe determination of two apparently different W-CI bondlengths for the chemically equivalent bonds in mer-[WCI,-(PMe,Ph),] is an artifact.The origin of this artifact iscocrystallization with the 0x0 impurity cis-mer-[WOCl,-(PMe,Ph),]. Such contamination results in one of the chlorosites being partially occupied by an 0x0 ligand, therebyapparently shortening the W-CI bond length.ExperimentalGeneral Considerations.-All manipulations were performedusing a combination of glove-box, high-vacuum or Schlenktechniques.O Solvents (except alcohols) were purified anddegassed by standard procedures. The NMR spectra weremeasured on Varian VXR 300 and 400 spectrometers; ,'Pspectra are referenced relative to external H3PO4.Preparation of cis-mer-[WOCI,( PMe,Ph),].-The complexcis-mer-[ WOCI,( PMe,Ph),] was prepared by a modification ofa previously reported synthesis.' ' A suspension of [WC14-(PMe,Ph),] (2.0 g, 3.32 mmol) in tetrahydrofuran (thf) (100cm3) was treated with PMe,Ph (1.38 g, 10.5 mmol) giving adark red solution. Water (0.06 cm3, 3.32 mmol) was added andthe mixture stirred overnight at room temperature. A furtherequivalent of water (0.06 cm3, 3.32 mmol) was added and themixture heated at 45 "C to complete the reaction.The mixturewas filtered and the solvent was removed under reducedpressure. The sticky residue obtained was washed with pentanegiving a purple solid. The latter was extracted into hot methanoland filtered. The filtrate deposited deep blue-purple crystalsof pure cis-mer-[WOCI,( PMe,Ph),] upon cooling to roomtemperature (0.45 g, 20%). NMR data (C6D6): 'H 6 1.49 [d,JpH = 8.6, cis-P(CH,),Ph], 1.85 [t, JpH = 4.7, 2 truns-P[(CH,)(CH,)Ph], 2.1 1 [t, J,, = 4.3,2 trans-P(CH,)(CH,)Ph]and 7.0-7.6 [m, 3 PMe,(C,H,)]; 31P-f'H), 6 -23.5 (s, Jpw =431, cis-PMe,Ph) and - 14.2 (s, Jpw = 340 Hz, 2 truns-PMe, Ph).Prepration 01. mev-[WCI,( PMe, Ph),].-This complex wasprepared by a slight modification of the published method.' Asolution of [WC14(PMePh),] (3.0 g, 4.1 mmol) in thf (25 cm3)was treated with zinc granules (0.53 g, 8.1 mmol) at roomtemperature.The mixture was stirred overnight at roomtemperature over which period the red solution became yellow-brown. The mixture was filtered and the solvent removed fromthe filtrate under reduced pressure giving a sticky brownresidue. The residue was washed with pentane giving impurJ. CHEM. SOC. DALTON TRANS. 1992 773Table 6 Crystal and intensity collection dataFormulaFormula weightLatticeSpace group4 Ah/AcyAPIVIA,ZF ( o wCrystal size/mm28 range/'DJg cm13p(Mo-Kz)/cm 'Octants collectedNumber of independent reflectionsNumber of reflections with F > 6o(F)Number of parameters variedRR'ci.s-nter-[WOCI,( PMe,Ph),]685.2MonoclinicP2,/c (no.14)16.1 14(4)10.4 1 3(2)17.955( 3)I l4.50( 1 )2741.4(9)413520.36 x 0.60 x 0.723-45.,1.6648.4h,k, & I3568275928 10.03 190.042 1C,,H,,CI*OP,Wmer-[WCl,( PMe2Ph),]704.7MonoclinicP2,/c (no. 14)I6.228( 3)10.344(2)I8.lOO(3)1 13.79( 1 )2780.0(7)413880.12 x 0.48 x 0.603 45"I .6848.6h,k, k I3635228823 10.05970.0692C24H33C13P3Wmrr-[WCI,( PMe'Ph),] as a brown-green solid (1.5 g, 52%).Proton NMR analysis of this sample revealed that at this stagethe product was already contaminated with cis-mer-[WOCI,-(PMe,Ph),]. Brown crystals were obtained by extraction intohot ethanol followed by cooling to room temperature.ProtonNMR data (C,D,): 6 -24.7 [br, 2 rruns-P(CH,),Ph], - 16.4[br, cis-P(CH,),Ph] and 7.3-1 2.2 [several broad resonances, 3PMe2(C,H,)I.X- Raj. Structure Derernrinarion Procedures.-Crystal data,data collection and refinement parameters for cis-mer-[WOCI,-(PMe,Ph),] and impure mer-[WCI,(PMe,Ph),] are sum-marized in Table 6. A typical procedure for the structuredetermination is as follows. A single crystal was mounted in aglass capillary and placed on a Nicolet R3m diffractometer. Theunit cell was determined by the automatic indexing of 25 centredreflections and confirmed by examination of the axial photo-graphs. Intensity data were collected using graphite-mono-chromated Mo-KZ X-radiation ( h = 0.71073 A).Check re-flections were measured every 100 reflections, and the data werescaled accordingly and corrected for Lorentz, polarization andabsorption effects. The structure was solved using Pattersonand standard difference-map techniques on a Data GeneralNOVA 4 computer using SHELXTL." A weighting scheme ofthe form H-' = o'(F) + gF' was employed for both structures.Space-group assignments were determined uniquely bysystematic absences. Hydrogen atoms were included incalculated positions [d(C-H) = 0.96 A; Uiso(H) = l.2Uiso(C)].The composite 'atom' at the disordered site of impure mer-[WCI,( PMe'Ph),] was modelled by refining each atom of thecomposite pair with common positional and isotropic thermalparameters.The site occupancies were allowed to refine subjectto the constraint that their sum was 1.0.Additional material available from the Cambridge Crystal-lographic Data Centre comprises H-atom coordinates, thermalparameters and remaining bond lengths and angles for bothstructures.AcknowledgementsG. P. is the recipient of an A. P. Sloan Research Fellowship(1991-1993) and a Camille and Henry Dreyfus Teacher-ScholarAward (1991-1996).References1 A. Hills, D. L. Hughes, G. J. Leigh and R. Prieto-Alcon, J . Cheni.Soc., Dalton Trans., 199 I , I 5 15.2 J. Chatt, L. Manojlovic-Muir and K. W. Muir, Cheni. Contntun.,1971, 655; L. Manojlovic-Muir, J . Chmt. Soc. A . 1971, 2796; L.Manojlovic-Muir and K. W. Muir, J . Cheni. Soc.. Dalton Trans.,1972,686 B. L. Haymore, W. A. Goddard, 111 and J. N. Allison, Proc.h i . Conf: Coord. Cl~ent., 1984, 23, 535.3 Y. Jean, A. Lledos. J. K. Burdett and R. Hoffmann,J. Ant. Chent. Soc.,1988, 110,4506; J . Chent. Soc., Chent. Contntun., 1988, 140.4 K. Yoon, G. Parkin and A. L. Rheingold, J . Ant. Cheni. Soc., 1991.113, 1437.5 K. Yoon and G. Parkin, J. Ant. Cheni. SO(.., 1991, 113, 8414.6 A. V. Butcher, J. Chatt. G. J. Leigh and P. L. Richards, J. Cheni. Soc.,7 J. M. Mayer, Inorg. Chent.. 1988,27, 3899.8 F. R. Fronczek, E. C. Baker, P. R. Sharp, K. N. Raymond, H. G. Altand M. D. Rausch, Inorg. Cheni., 1976, 15, 2284.9 W. E. Hunter, D. C. Hrncir, R. V. Bynum, R. A. Penttila and J. L.Atwood. Organontetullit:v, 1983,2, 750.10 J. P. McNally, V. S. Leong and N. J. Cooper, ACS Swtp. Set-.,1987, 357, 6; B. J. Burger and J. E. Bercaw, ACS Simp. Ser., 1987,357, 79.I I E. Carmona, L. Sinchez, M. L. Poveda, R. A. Jones and J. G. Hefner,Po!,*/terlron, 1983, 2, 797.12 G. M. Sheldrick, SHELXTL. An Integrated System for Solving,Refining and Displaying Crystal Structures from Diffraction Data,University of Gottingen. Gottingen, 198 1 .Dalton Truns., 1972, 1064.Rewivtd 17th Sc~pttwihtv 199 1; Pciper 1104807
ISSN:1477-9226
DOI:10.1039/DT9920000769
出版商:RSC
年代:1992
数据来源: RSC