摘要:
J. CHEM. SOC. DALTON TRANS. 1982 1499Hexanuclear Rhodium Hydrido-carbonyl ClustersBy Brian T. Heaton and Luisella Strona, Chemical Laboratory, University of Kent, Canterbury CT2 7NHSecondo Martinengo and Donatella Strumolo, lstituto di Chimica Generale e Centro del C.N.R. sui bassiRobin J. Goodfellow, Department of Inorganic Chemistry, The University, Bristol BS8 1 TSIan H. Sadler, Department of Chemistry, University of Edinburgh, Edinburgh EH9 3JJstati di ossidazione, Via G. Venezian 21, 20133, Milano, ItalyMultinuclear n.m.r. studies (1 H,' H-{103Rh}, l3C, 13C-{103Rh), and lo3Rh) show that protonation of [Rh6(C0)15]2-and [Rh6(C0)15C]2- at low temperatures gives [Rh,H(CO),,]- and [Rh6H(CO)15C]- which contain terminaland triangular face-bridging hydrides respectively; warming these solutions to room temperature results in loss ofhydrogen and formation of [ Rhl 2(CO)30]2- and [ Rhl 2( CO),,( C),], - respectively.THE hydride ligand in transition-metal hydrido-carbonylclusters can be found (a) in a terminal site bonded to onemetal, ( b ) in a bridging site bonded to either two metalsor a metal and a ligand, (c) bridging a M3 triangular face,or ( d ) in an interstitial site which is usually octahedral orpseudo-octahedral., The difference in energy betweendifferent sites is often not very great and this results infacile H migration.Because neutron diffraction studieson [Co,H(CO),,]- showed the hydride to be in the centreof the Co, octahedron,2 it was of interest to establish thestructure of the analogous rhodium derivative and wenow report n.m.r.studies at low temperature whichshow that [Rh,H(CO),,]- is isostructural with [Rh,X-(CO),,]- (X = I,3 COEtJ4 or 0,CMe 4, with the hydrideligand occupying a terminal site whereas the relatedcluster, [Rh,H(CO),,C]-, which adopts a trigonal-prismatic metal skeleton, contains a triangular face-bridging hydride.RESULTS AND DISCUSSIONN.M.R. Studies on [Rh6H(CO),,]-.-It has previouslybeen reported that [Rh12(C0)30]2- reacts with hydrogento give [Rh,3H3(C0)24]2- through the intermediate form-ation of the unstable [Rh,H(CO),J-, which can bebetter prepared by addition of acid to [Rh,(CO),,]2- atlow ternperat~res.~~, In the latter reaction a high-fieldlH n.m.r. resonance was observed but, because the pro-duct is unstable and difficult to obtain pure by crystal-lisation, we have now made multinuclear n.m.r.measurements in order to make a complete structuralcharacterisation of this compound, which is shown to beisostructural with [Rh,X(CO),,]- (X = I? COEt,4 or0,CMe 4, (Figure 1) and thus different to the structureadopted by the analogous complex, [CO,H(CO),,]-.~There was no evidence for the formation of [Rh6H2-(CO),,] on addition of excess of CF,CO,H to either anacetone or dichloromethane solution of [Rh6(co),,]2-.Consistent with the structure shown in Figure 1, thelo3Rh-(lH) n.m.r.spectrum shows four resonances at-374, -475, -478, and -538 p.p.m. in the intensityratio 2 : 2 : 1 : 1 with the last resonance (RhA) appearingas a doublet without lH decoupling; the other resonanceof intensity 1 can be assigned to RhD (Table).Althoughthe lH resonance (6 -12.2 p.p.m.) is basically a doublet,[lJ(lH-lo3Rh) 22.9 Hz] , with further long-range coupling,lH-(lmRh) INDOR measurements could only establish8(lWRhA) -540 p.p.m., which is in close agreement withthat obtained by direct lo3Rh n.m.r. measurements.49p"FIGURE 1[Rh,H(CO),,]-; large spheres = Rh, small spheres = COSchematic representation of the structure ofThe structure in Figure 1 contains four face-bridgingcarbonyls and 11 terminal carbonyls. Consideration ofthe symmetry shows there to be three types of face-bridging carbonyls in the ratio 2 : 1 : 1 (CIO, C20, andC30 respectively) and six types of terminal carbonyls inthe ratio 1 : 2 : 2 : 2 : 2 : 2 (C40-C90 respectively).The13C n.m.r. spectrum is thus quite complicated [Figure2(e)] but, through 13C-{ lo3Rh) measurements, can beassigned.The l3C n.m.r. spectrum resulting from broad-bandlo3Rh decoupling is shown in Figure 2 ( d ) . I t clearly showsthe correct number of f ace-bridging carbonyl resonanceswith the correct intensities but there are fewer peaks thanmight have been expected in the terminal region. Thus,there are only four resonances at 191.8, 187.1, 184.6, and183.6 p.p.m. with relative intensities 1 : 4 : 2 : 4 respect-ively. Decoupling RhA collapses the doublet of intensity1 at 191.8 p.p.m. which can be thus assigned to C40,while leaving the remaining terminal carbonyl resonancesessentially unchanged [Figure 2(b)].Decoupling at3.158 818 MHz [8(lo3Rh) -374 p.p.m.1, which could b1500 J. CHEM. SOC. DALTON TRANS. 1982due to RhB or Rhc, collapses the doublet of intensity 4 at187.1 p.p.m. [Figure 2(c)] ; this resonance must be due toeither C50 and C60 or C 7 0 and CsO being coincident.Figure 2(c) also clearly shows the doublet of intensity 2SAssignment 1 2 3 4 5936 8FIGURE 2 Carbon-13 n.m.r. spectra (25.15 MHz) of [NBu,][Rh,H(CO),,] in (CH,),CO at -70 "C. ( a ) Decoupling at3.158 495 MHz [8(lo3Fth) = -476 p.p.m.1, (b) decoupling at3.158 300 MHz [6(lo3Rh) = -538 p.p.m.1, (c) decoupling at3.158 818 MHz [8(lo3Rh) = -374 p.p.m.1, ( d ) broad-bandRh decoupling at 3.158 547 MHz, ( e ) non-decoupled spectrum.S = solventat 184.6 p.p.m.due to C90 and a doublet of intensity 4 at183.6 p.p.m., which is again due to coincidences of eitherC50 and C60 or C70 and C80; both these doublets col-lapse on irradiation at 3.158 495 MHz [8(Io3Rh) -476p.p.m.1 which simultaneously decouples RhD and RhRor Rhc [Figure 2(a)]. Unfortunately, since it is not pos-sible to differentiate RhB from Rhc, it is not possible toassign unambiguously which of the resonances of inten-sity 4 is due to C50/C60 and C70/C80. Nevertheless, thestructure of [Rh,H(CO),,]- is clearly consistent withthat shown in Figure 1 and all the n.m.r. data on thiscompound are summarised in the Table.Since there are probably small energetic differencesbetween different sites occupied by hydrogen, it is un-wise to speculate too much on the different structuresadopted by [M,H(CO),,]- (M = Co or Rh).Neverthe-less, it could be associated with the increasing M-H bondstrength found with the heavier metals. Thus therhodium cluster is stabilised by a single terminal Rh-Hbond, whereas a multicentre C0,-H bond is required tostabilise the cobalt cluster.N.M.R. Studies on [Rh6H(CO)15C]-'.-The I3C n.m.r.spectrum of [Rh6(CO)l,C]2- has been previously shownto be entirely consistent with the solid-state structure.8Addition of excess CF3C02H to a dichloromethane solu-tion of [Rh6(CO),5C]2- at low temperature results in theappearance of (a) two new rhodium resonances [8(lo3RhA)- -144.5 and 8(lo3Rh~) = -412.5 p.p.m.1 with thelatter occurring as a doublet [1J(1H-103RhB) 15 Hz] in thenon-lH-decoupled Io3Rh spectrum, * and (b) a high-fieldquartet of quartets in the lH n.m.r. spectrum [8(lH) =-15.6 p.p.m., 1J(1H-103Rh~) 15, 2J(1H-103RhAI 1.5 Hz].These results are consistent with protonation of aRh, triangular face (Figure 3), which is confirmed by13C-{103Rh} measurements.The data resulting fromthese experiments, together with the corresponding datafor [Rh,(CO),,CI2- are summarised in Figure 3. I tshould be noted that RhB occurs at lower frequency thanRhA and this is in keeping with similar shifts to lowerfrequency observed for the interstitial rhodium atom in[Rhl,H5-n(CO)2,]n- (n = 1 4 ) on increasing protonation;thus, the interstitial rhodium chemical shifts in [Rhl,-H6-n(C0)24]n- are +6 370, +4 554, +3 547, and +2 917p.p.m.for n = 4, 3, 2, and 1 respe~tively.~ I J -(l03Rh-lo3Rh) has yet to be observed in direct lo3Rhn.m.r. measurements and appears to be immeasurablysmall in rhodium carbonyl clusters. The values ofJ(lo3Rh-13CO) remain essentially the same on protonationwhereas there are significant variations in 13C chemicalshifts with the average value of 8(13CO) in [Rh,H(CO),,-C]- (209.5 p.p.m.) being, as expected, lower than in[Rh6(CO),,Cl2- (215.4 p.p.m.) Without the benefit ofspin-decoupling experiments, it is necessary to be ex-tremely careful when assigning 8(13CO). Thus, although* The 103Rh chemical shifts were initially obtained by INDORmeasurements. Thus RhB was readily located by lH-{lo3Rh}INDOR and a triple-resonance lH-Po3Rh, lo3Rh: experimentlocated RhA. The rhodium chemical shifts of RhA and RhBobtained in this way are -143 and -412 p.p.m.respectivelyJ. CHEM. SOC. DALTON TRANS. 1982 1501N.m.r. data for [NBu,][Rh,H(CO),,] in (CD,),CO at -70 "C8(103Rh) q13co)LIL \ r tR h A R h B Rho RhD C'O C20 c30 c40 c50 c60 c70 cao c90 -- -638 -374" -475e -478 244.6 237.6 e 232.9 6 191.8 187.1 6 183.6 a 184.6(77.1) (70.3) (72.3) (68.4)In CD2C12 a t -75 "C, 6(1H) = -12.2 P.P.m. (SiMe, as standard), lJ(H - RhA = 22.9 Hz. For numbering see Figure 1.11.376 MHz = 0 p.p.m. a t such a magnetic field that the protons in SiMe, resonate a t exactly 360 MHz; negative chemical shiftsd Values in p.p.m. with SiMe, as standard; figures in parentheses indicate 1J(103Rh-13CO) values are to low frequency of 11.376 MHz.in Hz ( f 2).* Assignments could be reversed.one terminal carbonyl and one inter-triangular bridgingcarbonyl resonance in [Rh6H(C0),,C] - have similarchemical shifts (193.3 and 223.5 p.p.m.) to those found in[Rh6(CO),5C]2- (196.9 and 224.0 p.p.m.), specific Rh-decoupling measurements show them to be on differentplanes, see Figure 3. It should also be noted that theoriginal, somewhat contentious, assignment of 2J(Rh-CO)for the terminal carbonyls in [Rh6(CO)15C]2- to inter-triangdar coupling 7 has been substantiated by thepresent measurements. Thus, for [Rh,H(CO),,C]- thereare two resonances due to the terminal carbonyls whichplane is highly susceptible to electrophilic attack. Wehave also found a similar addition with Ag+ to give[Ag{Rh,(CO)l,C)2]3-,12 but we have, as yet, no evidencefor exposing the carbide by opening the metallic skele-ton through oxidation as was recently observed byBradley etFinally, it should be noted that warming either[Rh,H( CO),,] - or [Rh,H( cO),,c] - to room temperatureresults in loss of hydrogen with concomitant dimeris-ation to [Rhl2(C0),I2- and [Rh12(CO)2,(C)2]2- respect-ively.G(lo3Rh) 6(13CO) /p .p.m. lJ (Rh-CO) /Hz aJ(R.h-CO)/H~___--- 216.0 (224.0) 30.3 (30.6) e----- 163.1 (-285.1) --_ ;;* -Cod--- A 193.3 (196.9) 71.3 (79.3) 3.9 (3.9) a223.6 (224.0) 25.4 (30.6)CO c--- 189.6 (196.9) 78.1 (79.3) 3.9 (3.9) *-415.7 (-285.1)---* = p-co 8(lH) = -16.6 p.p.m., lJ(Rh-H) = 16 HzFIGURE 3 Schematic representation of the structure of [Rh,H(CO),,C]- together with lH, 13C, and lo3Rh n.m.r.data a t -25 "C.Values of J are f l Hz; superscript a denotes 2J(RhB-CO), and The values shown in parentheses are data for [Rh,(CO)l,C]2-.b, aJ(Rh*-CO)both appear as doublets of doublets. Decoupling- RhAcauses the resonances at 193.3 p.p.m. to become a doubletc2J(RhB-c0) 3.9 Hz] while the resonance at 189.6 p.p.m.also becomes a doublet [lJ(RhB-c0) 78.1 Hz]; relatedresults are obtained on decoupling RhB (Figure 3).However, similar long-range coupling is not found in[Rh,H(CO),,] -. The carbide resonance in [Rh,H(CO),,-C]- is complicated (quartet of quartets) but becomes asimple quartet on decoupling either RhA or RhB. Inkeeping with previous findings,lo the reduced negativecharge on [Rh,H(CO),,C]- results in the carbide re-sonance occurring at higher frequency than in [Rh,-I t has previously been shown that successive additionof up to two [Cu(NCMe)]+ groups occurs on the triangularfaces of [Rh6(CO)15C]2-,11 whereas we find no evidencefor formation of [Rh,H,(CO),,C], even with excessCF3S03H.However, the similar addition of both H+and [Cu(NCMe)]+ to the triangular face suggests that this(co) 15Cl2--EXPERIMENTAL1H-(103Rh}, 13C, and 13C-{ 103Rh} measurements weremade as described previously 14-16 and 103Rh n.m.r. spectrawere recorded on a Bruker WH 360 MHz spectrometer using15 mm n.m.r. tubes containing solutions (5 cm3, 0.5 mmol)doped with [Cr(acac),] (acac = acetylacetonate) at a levelsimilar to that used for 13C measurements.'& Optimumsignal to noise was obtained using a pulse angle of 13" and apulse repetition rate of 0.348 s.Hydrogen-1 n.m.r. spectrawere obtained on a JEOL PS-100 spectrometer using 5 mmn.m.r. tubes. It has previously been suggested l7 that[Rh,(CO)l,C]2- should be used as a reference (0 p.p.m.) forlo3Rh chemical shifts. This seems to us unwise because ofits general unavailability and because 4l03Rh) is significantlyaffected by both solvent and temperature effects. As aresult, we prefer to relate 8(lO3Rh) = 0 p.p.m. to a standardfrequency (3.16 or 11.376 MHz) at such a magnetic field thatthe protons in SiMe, resonate a t exactly 100 or 360 MHzrespectively. Converting the data of Gansow et a1.l' on[Rh,(co)l,c]2- to our scale gives 8(lo3Rh) = -277.2 p.p.m1502 J.CHEM. SOC. DALTON TRANS. 1982which, although it is not stated at what temperature thismeasurement was made, is in close agreement with ourdata, 8(lo3Rh) = -285 p.p.m., at -25 "C.13CO-Enrichments were carried out using standardvacuum line techniques and all preparations were carriedout under a nitrogen atmosphere using Schlenk tubetechniques.[NBu,],[Rh,(CO) 15] was prepared as described pre-viously la and converted to [NBu,][RhsH(CO),,] by additionof excess (ca. four times the stoicheiometric equivalent) ofCF,CO,H to an acetone or dichloromethane solution of[NBu,],[Rh,(CO),,] a t low temperature (- 76 "C). Thisresulted in an immediate colour change from deep green tobrown.Warming the brown solution to room temperatureresults in a further colour change to purple due to the form-ation of [Rhl,(C0),,]2-, which shows three 103Rh resonancesat + 158, -332, and -551 p.p.m. of relative intensities1 : 4 : 1. This supports our earlier 13C-(lo3Rh} measure-ments l 9 and contradicts recent claims for fluxional be-haviour of the metal skeleton.20[N(PPh,),] [Rh,(CO) 15C] was prepared from K,[Rh,-(CO),,C] a by adding an excess of [N(PPh,),]Cl in MeOH.The resulting precipitate was filtered off, dried, and re-crystallised from acetone-PriOH to give the desired pro-duct.[N(PPh,) ,] [Rh,(CO) 1513C] was prepared by addition ofl3Cc1, (6 p1, 92% 13C) to a methanol solution of [N(PPh,),!-[Rh(CO),] (0.25 g).There was an immediate reaction togive the crude product which was filtered off and recrystal-lised from acetone-PriOH, yield 80%.[N(PPh,),][Rh,H(CO),,C] was prepared by addition ofexcess (ca. four times the stoicheiometric equivalent) ofCF,CO,H to either an acetone or dichloromethane solution of[N(PPh,),][Rh,(CO),,C] at -30 "C.We thank the S.E.R.C. and C.N.R. for financial supportand the S.E.R.C. for a research fellowship (to L. S.) andfor high-field n.m.r. facilities.[I/ 1957 Received, 18th December, 198 13REFERENCESR. G. Teller and R. Bau, Struct. Bonding, 1981, 44, 1.D. W. Hart, R. G. Teller, C. Y . Wei, R. Bau, G. Longoni,S. Campanella, P. Chini, and T. F. Koetzle, Angew. Chem., Int.Ed. EngE., 1979, 18, 80.V. G.Albano, P. L. Bellon, andM. Sansoni, J . Chem. SOC. A ,1971, 678.G. Ciani, A. Sironi, P. Chini, and S. Martinengo, J . Organo-met. Chem., 1981, 213, C37.P. Chini, G. Longoni, and S. Martinengo, Chim. Ind. (Milan),1978, 60, 989.P. Chini, Gazz. Chim. Ital., 1979, 109, 225.V. G. Albano, P. Chini, S. Martinengo, D. J. A. McCaffrey,D. Strumolo, and B. T. Heaton, J . Am. Chem. SOC., 1974, 96,8106.V. G. Albano, M. Sansoni, P. Chini, and S. Martinengo, J .Chem. SOC., Dalton Trans., 1973, 651.B. T. Heaton, L. Strona, P. Chini, S. Martinengo, and R. J.Goodfellow, unpublished results.lo B. T. Heaton. L. Strona, and S. Martinengo, J . Organomet.Chem., 1981, 215, 415.l1 V. G. Albano, D. Braga, S. Martinengo, P. Chini, M. Sansoni,and D. Strumolo, J . Chem. SOC., Dalton Trans., 1980, 52.l2 B. T. Heaton, L. Strona, and S. Martinengo, unpublishedresults.1s J. S. Bradley, G. B. Ansell, and E. W. Hill, J . Am. Chem.Soc., 1979, 101, 7417.l* P. L. Goggin, R. J. Goodfellow, J . R. Knight, M. G. Norton,and B. F. Taylor, J . Chem. SOC., Dalton Trans., 1973, 2220.l6 G. Longoni, B. T. Heaton, and P. Chini, J . Chem. SOC.,Dalton Trans., 1980, 1537.l6 C. Brown, B. T. Heaton, L. Longhetti, W. T. Povey, andD. 0. Smith, J . Organomet. Chem., 1980, 192, 93.0. A. Gansow, D. S. Gill, F. J. Bennis, J. R. Hutchinson.J. L. Vidal, and R. C. Schoening, J . Am. Chem. SOC., 1980, 102,2449.18 S. Martinengo and P. Chini, Gazz. Chim. Ital., 1972, 102,344.19 B. T. Heaton, L. Strona, P. Chini, and S. Martinengo, J .Organomet. Chem., 1980, 194, C29.20 J. L. Vidal, R. C. Schoening, and W. E. Walker, A.C.S.Symp. Ser., 1981, 155, 61
ISSN:1477-9226
DOI:10.1039/DT9820001499
出版商:RSC
年代:1982
数据来源: RSC