J . CHEM. SOC. DALTON TRANS. 1994 1 1 12,2‘ : 6‘,2”-Terpyridine (terpy) acting as a Fluxional BidentateLigand. Part 3.’ Ruthenium Carbonyl Halide Complexes,[RuX,(CO),(terpy)] (X = CI, Br or I) and MetalTetracarbonyl Complexes [M(CO),(terpy)] ( M = Cr, Mo orW): Nuclear Magnetic Resonance Studies of their SolutionDynamics and Synthesis of trans-[RuX,(CO)(terpy)](X = CI, Br or I)Edward W. Abel, Keith G. Orrell, Anthony G. Osborne, Helen M. Pain and Vladimir SikDepartment of Chemistry, The University, Exeter EX4 400, UKUnder mild conditions 2,2‘:6’,2’’-terpyridine reacted with [{RuX,(CO),},,] (X = CI, Br or I) and with[M(CO),(nbd)] [M = Cr, Mo or W; nbd = norbornadiene (bicyclo[2.2.l]hepta-2,5-diene)] t o form theoctahedral complexes trans, cis- [RuX,(CO),(terpy)] and [M (CO),(terpy)] respectively, in which theterpyridine acts as a bidentate chelate ligand.In solution these complexes are fluxional with the terpyridineoscillating between equivalent bidentate modes. Heating trans,cis-[RuX,(CO),(terpy)] (X = CI, Br or I) ina high boiling solvent formed trans- [RuX,(CO) (terpy)] whereas cis- [RuX,(CO) (terpy)] was formed byheating the solid dicarbonyl complex.2,2‘ : 6’,2”-Terpyridine (terpy) has long been known to formstable complexes with transition metals, but in recent yearsthere has been increased interest shown in the co-ordinationcompounds formed by terpy and other oligopyridines. *Attention has been particularly focused on complexes ofruthenium(r1) because of the photochemical and electrochemicalproperties exhibited by some of these c ~ m p l e x e s .~ . ~ Althoughterpy primarily functions as a terdentate ligand, a number ofcomplexes have been reported in which it is acting as a bidentateligand.’-12We have reported l 3 our preliminary results of a detailedNMR study of some complexes of Pt’”, Re’ and Wo with terpyand our completed results for trimethylplatinum halides l4 andrhenium pentacarbonyl halides. In these studies it was shownthat in solution terpy was definitely acting in a bidentate chelatemode and was also undergoing an oscillatory fluxional motion.We now report further examples of terpy behaving as a fluxionalbidentate ligand in the areas of Ru”, Cr’, Moo and Wocomplexes.This paper describes the reactions of the polymeric rutheniumdicarbonyl halides [{RuX,(CO),},] (X = C1, Br or I) and thetetracarbonyl norbornadiene [n bd (bicyclo[2.2.1] hepta-2,5-diene)] complexes [M(CO),(nbd)] (M = Cr, Mo or W) withterpy, solution NMR studies of the dynamic stereochemistry ofthe resulting trans,cis-[RuX,(CO),(terpy)] 1 and [M(CO),-(terpy)] complexes respectively, and the conversions of 1 to cis-and trans-[RuX,(CO)(terpy)] 2 and 3.ExperimentalMaterials.-The compounds [M(CO),(nbd)] (M = Cr orMoI5 and W 1 6 ) , trans,cis-[RuX,(CO),(terpy)] (X = C1 orBr) were prepared by previous methods with the modificationfor Ru complexes that the solutions were allowed to cool toapproximately 40 “C before addition of the terpy.2,2‘:6’,2”-Terpyridine (terpy) was purchased from Aldrich.Synthesis of Complexes.-All preparations were carried outunder purified nitrogen using standard Schlenk techniquesusing freshly distilled, dried and degassed solvents.XOC;d,= N,OC A N ”> 1X2 3Tetra carbony l( 2,2‘ : 6’, 2”- terpyr idyl )chromium( 0).The com-plex [Cr(CO),(nbd)] (0.18 g, 0.70 mmol) was added to a stirredsolution of terpyridine (0.15 g, 0.64 mmol) in hexane (30 cm3).The pale yellow solution was protected from light and stirred for3 to 4 d. The supernatant liquid was then decanted and the redprecipitate was washed with hexane (3 x 10 cm3) and driedunder vacuum. Yield 0.03 g (12%). Because of its instability thiscompound was not obtained pure.complex [Mo(CO),(nbd)] (0.13 g, 0.43 mmol) and terpyridine(0.1 g, 0.43 mmol) were heated in hexane (40 cm3) under refluxfor 2 h.The initially pale-yellow solution darkened to red-purple and a dark purple solid was precipitated. The solid wasisolated by filtration, washed with hexane (2 x 10 cm3) andrecrystallised from dichloromethane-hexane to yield dark red,needle-like crystals. Yield 0.09 g (47%).Tetracarbonyl(2,2’ : 6‘,2”-terpyridyl) tungsten(0). The complex[W(CO),(nbd)] (0.17 g, 0.43 mmol) and terpyridine (0.1 g, 0.43mmol) were heated in hexane (30 cm3) under reflux for 10 h toproduce a mauve solution and a black precipitate. The solid wasisolated by filtration, washed with hexane (2 x 10 cm3) andrecrystallised from dichloromethane-hexane to yield a dark red-purple crystalline solid. Yield 0.05 g (23%).ruthenium(I1). The complex [ { RuCl,(CO),},] (0.04 g, 0.18Tetracarbonyl(2,2‘ : 6‘,2’’-terpyridyl)molybdenum(o). Thecis-Dicarbonyl-trans-diiodo(2,2’ : 6‘,2”-terpyridyl)112 J.CHEM. SOC. DALTON TRANS. 1994Table 1(X = C1, Br or I)Synthetic and analytical data for the complexes [M(CO),(terpy)] (M = Cr, Mo or W), [RuX,(CO),(terpy)] and trans-[RuX,(CO)(terpy)]Analysis (%)t rans,cis-[RuCl ,(CO), (terpy)]trans, cis- [ RuBr , (CO) , (t erp y )]trans,cis-[RuI,(CO),(terpy)]trans- [ RuCl ,(CO)( terp y)]trans-[Ru Br ,(CO)( terp y)]trans-[RuI,(CO)(terpy)]Appearance Yield a (%)Red 12Red-purple 47Darkpurple 23Green-yellow 49Red and yellowd 45Dark red 44Red 86Brown-red 83Dark red 81vco */cm-'2010 (s) 1832 (s)1903 (vs)1860 (sh)201 5 (m) 1832 (s)1905 (vs)1885 (sh)2009 (s) 1826 (s)1892 (vs)1880 (sh)2066 (s)2005 (s)2064 (s)2005 (s)2058 (s)2005 (s)1966 (s)1968 (s)1965 (s)C-50.8(51.7)42.4(43. I )45.0(44.3)36.8(37.1)31.0(31.7)45.0(44.4)36.2(36.8)32.8(32.5)N-9.3(9.5)7.4(7.9)9.2(9.1)7.4(7.6)6. I(6.5)9.8(9.7)7.5(8.0)6.3(6.5)=a Yield quoted relative to metal-containing reactant.Recorded in CH,Cl, solution; s = strong, v = very, m = medium, sh = shoulder.Calculated values in parentheses. Ref. 9. Allows for OSMe,CO.mmol) and LiI (0.2 g, 1.5 mmol) were heated in methanol (10cm3) under reflux for 3 h. The dark yellow solution was cooledto approximately 40 "C, terpyridine (0.05 g, 0.21 mmol) addedand the resulting solution heated under reflux for 4 min.Coolingthe solution produced a fine dark red solid which was washedwith methanol (2 x 10 cm3) and then recrystallised from hotmethanol. Yield 0.05 g (44%).tran~-Carbonyldihalogeno(2,2' : 6',2"-terpyridyl)ruthenium(11).The chloro, bromo and iodo monocarbonyl complexes wereprepared in a similar fashion by heating the dicarbonylcomplexes in boiling tetrachloroethane for 3 to 4 h. Theprogress of the reaction was monitored using infraredspectroscopy, to note the disappearance of the carbonyl bandsof the dicarbonyl species. The cooled solution was concentratedunder reduced pressure to a low volume and the solid thusproduced was isolated by decantation, washed with hexane(2 x 10 cm3) and recrystallised from acetone. The synthetic andanalytical data for the three complexes are given in Table 1.Physical Methods.-Hydrogen- 1 NMR spectra were record-ed on Bruker AM250 or AC300 spectrometers operating at250.13 and 300.13 MHz respectively. A standard B-VT1 000variable-temperature unit was used to control the probetemperature, the calibration of this unit being checkedperiodically against a Comark digital thermometer.Thetemperatures are considered accurate to k 1 "C. Rate data werebased on band-shape analysis of 'H spectra using the authors'version of the standard DNMR program,18 and activationparameters based on experimental rate data were calculatedusing the THERMO program." Infrared spectra were recordedon a Perkin Elmer 881 spectrometer calibrated from the 1602cm-' signal of polystyrene, and on a Nicolet Magna IR 550 FTspectrometer. UV/VIS spectra were recorded on a Philips PU8730 spectrophotometer. Elemental analyses were performed byButterworth Laboratories Ltd., Teddington, Middlesex.Results and DiscussionThe complexes trans,cis-[RuX,(CO),(terpy)] 1 (X = C1, Br orI) were readily obtained from [{RuX,(CO),),] and terpy bya modification of a previous method.' In solution the complexesshow two strong carbonyl stretching bands (Table 1) whichare indicative of a cis stereochemistry for the carbonyl groupsand in agreement with the reported values.' NMR spectroscopy(see below) provided conclusive evidence for terpy acting asa bidentate chelate ligand in these complexes.The stereo-chemistry of these complexes has been established by X-raycry~tallography.~Previous work has shown that decarbonylation of trans,cis-[RuX,(CO),(terpy)] with trimethylamine N-oxide producedcis-[RuX,(CO)(terpy)], the cis arrangement of halogens beingconfirmed by X-ray crystallography. We find that heatingtrans,cis-[RuX,(CO),(terpy)] in tetrachloroethane causes de-carbonylation with the formation of trans-[RuX,(CO)(terpy)].This identification was based upon analytical data, IR data (asingle carbonyl band) (Table l), a 'H NMR spectrum whichcomprised only six chemically shifted signals (Table 2)indicative of a terdentate terpy, and a visible absorption band atapproximately 450 nm (Table 3) in agreement with reportedvalues for the trans isomers.'*20In contrast to this, heating solid trans,cis-[RuX,(CO),-(terpy)] above 140°C resulted in the formation of cis-[RuX,(CO)(terpy)].The stereochemical identification of thesecompounds was based on the observation of a band in thevisible spectrum at approximately 415 nm which is in agreementwith a previous report.' The far infrared data obtained for thecis and trans monocarbonyl compounds were also consistentwith previous work (Table 3). The 'H NMR spectra of thesecompounds were not recorded as they would not aidstereochemical assignment.The complexes [M(CO),(terpy)] (M = Cr, Mo or W) wereprepared from [M(CO),(nbd)] and terpy. These complexes,which are all air and light sensitive in solution, show fourcarbonyl stretching vibrations (Table I), in agreement with aprevious report and consistent with a M(CO), moiety of C2ysymmetry.This implies a bidentate terpy and NMRspectroscopy (see below) provided confirmation of thisconclusion.Ambient and Above-ambient Temperature NMR Studies.-[RuX,(CO),(terpy)]. At ambient temperature the 'H NMRspectra of these complexes consisted of a complex pattern oJ. CHEM. SOC. DALTON TRANS. 1994 113Table 2 Proton NMR chemical shift data for terpy and its chromium(o), molybdenum(o), tungsten(o) and ruthenium(I1) complexesCompoundterpy[Cr(CO)dterpy )ICMo(CO),(terpy)lCW(CO),(terpy)Itrans,cis-[ R uCI,(CO),(terpy)] 'trans,cis-[R uBr,( CO),( terpy)]trans,cis-[ R uI ,( CO) , (terpy )] 'trans-[RuCI,(CO)(terpy)] 'trans- [ RuBr ,( CO)( terpy)]trans-[ Rul ,( CO)( t erp y)] 'T/"C3030- 70- 303030303030306AE &IF 6CG8.69 7.35 7.889.27 (A) 7.35 (B) 7.85 (C)8.81 (E) 7.35 (F) 7.70 (G)9.07 (A) 7.47 (B) 7.99 (C)8.73 (E) 7.42 (F) 7.91 (G)9.22 (A) 7.46 (B) 8.02 (C)8.75 (E) 7.43 (F) 7.91 (G)9.15 (A) 7.68 (B) 8.16 (C)8.84 (E) 7.56 (F) 7.95 (G)9.15 (A) 7.66 (B) 8.14 (C)8.84 (E) 7.56 (F) 7.94 (G)9.13 (A) 7.60 (B) 8.09 (C)8.82 (E) 7.53 (F) 7.91 (G)9.02 7.67 8.278.83 7.67 8.288.79 7.64 8.266DH8.648.12 (D)7.61 (H)8.20 (D)7.68 (H)8.26 (D)7.69 (H)8.28 (D)8.20 (H)8.27 (D)8.19 (H)8.26 (D)8.15 (H)8.618.628.65a Relative to SiMe, (internal) 6 = 0.See Fig. 2 for hydrogen labelling.Recorded at 250 MHz. Recorded at 300 MHz.6,7.968.00 (J)8.10 (J)8.12 (J)8.19 (J)8.18 (J)8.13 (J)8.558.608.576 K L8.478.10 (K)7.68 (L)8.23 (K)7.68 (L)8.23 (K)7.66 (L)8.28 (K)7.91 (L)8.28 (K)7.85 (L)8.27 (K)7.73 (L)8.958.758.72Table 3 Spectroscopic data for cis- and trans-[RuX,(CO)(terpy)] (X = CI, Br or I)Compoundcis-[RuCl,(CO)(terpy)]cis-[ RuCl ,(CO)( terpy)]cis- [R uC1 (CO)( terp y )]cis-[RuBr,(CO)(terpy)]cis-[RuBr,(CO)(terpy)]cis-[RuI,(CO)( terpy)]trans- [R uC1 , (CO)( t erp y )]trans- [RuCl , (CO)( terpy )]trans-[RuBr,(CO)(terpy)]trans-[RuI,(CO)( terpy)]UV/VJS a IR b3c~maxlnm (400-200)/cm-'41 741941542 141741 346 I46 145545 13 18m, 278m, 255m31 5m, 277m, 255m32Om, 261m, 24Om258m, 243w, 190111,164~250w3 15m, 302 (sh)322m, 305 (sh)248m--Recorded in CH,CI, solution.m = Medium, sh = shoulder, w = weak. Recorded as CsI discs.SourceRef. 9Ref. 20This workRef. 9This workThis workRef. 20This workThis workThis workH9.0 8.5 8.0 7.5 7.06Fig. 1in (CDCI,), at 30 "C. Signal labels refer to Fig. 2300 MHz 'H NMR spectrum of trans,cis-[RuC1,(CO),(terpy)]HE/ 'HFFig. 2 Interconverting structures of the terpy complexes showing thehydrogen labelling: (a) M = Ru, L' = L2 = CO, L3 = L4 = CI, Br orI; (6) M = Cr, Mo or W, L' = L2 = L3 = L4 = COoverlapping signals which was clearly associated with anunsymmetrically co-ordinated terpyridine. The results for[RuCl,(CO),( terpy)] will serve to demonstrate the analysis ofthe problem.The spectrum (Fig. 1) was fully assigned by a two-dimensional correlation spectroscopy (COSY) experiment tocleven non-equivalent protons labelled A to L in Figs. 1 and 2,and hence was clearly consistent with a bidentate terpyridineligand. All chemical shift and coupling constant data are givenin Tables 2 and 4. Hydrogens HA and HE, alpha to the N atoms,give rise to signals at the highest frequencies, with HAexperiencing an additional co-ordination-induced shift.On increasing the solution temperature to ca. 70°C exten-sive spectral changes took place as indicated in Fig. 3. Exchangebroadening occurred between analogous pairs of protonsnamely HA/E, HB,F, H,,,, HD/H and HK L. As observed pre-viously with compounds of Re' and PJv involving bidentateterpy, proton H, retained its triplet structure and didnot undergo exchange.These exchange processes can berationalised by the dynamic spin problems ABCD EFGH'and JKL JLK.At temperatures greater than 70°C a new set of signalsappeared amidst the broad coalescing signals of the bidentateterpy species. The intensity of these new signals increasedrapidly with temperature (Fig.3) whilst the exchange-broadenedsignals decreased in intensity. The triplet structure of the signaldue to H, also collapsed. At 110 "C six sharp well resolvedchemically shifted signals were observed, which are indicative ofterpy symmetrically bonded to Ru. On cooling the sample toambient temperature the spectrum remained consistent with asymmetrical terpyridine and it indicated that only a traceamount of the starting complex involving a bidentate terpy waspresent.These spectral changes can be rationalised as arisingfrom two processes, first the terpyridine being involved in afluxional process in which the co-ordination complex isoscillating between two forms both of which involve terp114 J. CHEM. SOC. DALTON TRANS. 1994scalar couplingC1, Br or I)ComplexCCr(CO),(terpy 11rMo(CO),(terPY)lCW(CO),(terPY)Itrans,cis-[RuCl,(CO),(terpy)]trans, cis-[ RuBr , (CO) , (terpy )]trans,cis-[RuI,(CO),(terpy)]trans- [RuCl , (CO)( terpy)]trans-[RuBr,(CO)(terpy)]trans- [RuI , (CO)( terpy)]constant data" for3JAB/3JEF5.715.55.314.55,014.35.514.95.515.25.414.98.28.08.1(M = Cr, Mo or W),JAC/4EGbbb1.611.61.411.71.511.71 .o1.31.1JBC/3fFGb7.517.27.117.06.317.86.917.85.817.86.36.05.94JBD/4 JFHbbbb1.1/1.11.211 .o1 .o1.2I .23JCD/3fGH6.7/b8.817.88.417.0bl7.98.117.88.017.98.18.08.0" In Hz.See Fig. 2 for labelling of hydrogen atoms. Not measured accurately. All five-bond couplings < 0.5 Hz.and trans-3 J , ~ I 3 J,,.7.717.77.917.87.817.88.lj8.l7.917.77.917.78.17.67.7Ii/c JbLUS-'A 500-110401 30I U O200 HzFig. 3 300 MHz 'H NMR spectra of [RuCI,(CO),(terpy)] in(CDCI,), in the temperature range 30-1 10 "C, illustrating the fluxionalprocess and the formation of trans [RuCI,(CO)(terpy)].The uppermostspectrum was obtained after cooling the sample to 30 "C. Computersimulated spectra are shown on the right with 'best-fit' rate constants kfor the fluxional processfunctioning as a bidentate chelate ligand (Fig. 2), and secondly,at temperatures greater than 70 "C the simultaneous formationof a complex in which terpy is functioning as a terdentate ligand.The fluxionality of the terpyridine is analogous to that which wehave reported for complexes involving Pt" and Re'.The energetics of the fluxional process were analysed bythe application of standard band-shape analysis methods.Because of the formation of the terdentate complex at highertemperatures, studies were necessarily restricted to the tem-perature range 30-100 "C.As reported for the Re complexesit was found to be sufficient to apply the method to theexchanging pairs of signals A and E and fitting the AEportion of the A B e E F dynamic spectrum. The experi-mental and computed spectra are compared in Fig. 3.Analogous spectral changes were observed for the other twocomplexes [RuX,(CO),(terpy)] (X = Br or I).[M(CO),(terpy)]. The ambient temperature 'H NMRspectra of [Mo(CO),(terpy)] and [W(CO),(terpy)] both showsix chemically shifted signals, three of which are stronglyoverlapping. Cooling the solutions to -70 "C and -30 "Crespectively produced spectra containing eleven chemicallyshifted signals which could be unambiguously assigned bydecoupling experiments and by comparison with the spectra ofthe analogous rhenium complexes Sac-[ReX(CO),(terpy)](X = Cl, Br or I).Hence for these complexes the terpyridylligand is fully fluxional at ambient temperature but the motioncan be arrested at low temperatures. The energetics of thisprocess were calculated in a manner similar to that described forthe ruthenium complexes.The ambient temperature NMR spectrum of [Cr(CO),-(terpy)], although it contained some signals attributed toimpurities because of the instability of the complex in solution,showed eleven chemically shifted signals all of which could beassigned to terpyridine in a bidentate non-fluxional bondingmode. Attempts to study any fluxional process by increasing thesolution temperature led to rapid decomposition of the sample.Hence no quantitative information could be obtained but it canbe inferred that the fluxional process in the Cr complex involvesa greater activation energy than for the Mo and W complexessince the process is slow on the NMR time-scale at ambienttemperature.trans-[RuX,(CO)(terpy)].These complexes contained ter-dentate terpy and were stereochemically rigid, their 'H NMRspectra consisting of six chemically shifted signals at alltemperatures. Comparison of the data in Table 2 for terpy and[RuX,(CO)(terpy)] shows that on terdentate complexationhydrogens A/E, B/F, C/G, J and K/L all experience highfrequency shifts (A6 = 0.64-4.1) but hydrogens D/H experiencea very small low frequency shift (A6 = 0.034.01). The direc-tions of these co-ordination-induced shifts are the same asobserved for the Re complexes but the magnitudes differ asmight be expected.The explanation for the observed high andlow frequency shifts on complex formation is as described forthe Re complexes, and involves a combination of high frequencymetal co-ordination shifts and the effects of the necessaryreorientation of the pyridine rings into a cis,cis arrangementon terdentate complexation.Energies and Mechanisms of the Fluxion in [RuX,(CO),-( t e r p y ) ] and [ M (CO), ( t erp y )] Complexes . -Th e activationparameters for the fluxional process were calculated from 'best-fit' rate data and are listed in Table 5. Values for the Rucomplexes range from 75.8 to 78.7 kJ mol-' and exhibitslight halogen dependence. The order C1 < Br < I is identica115 J.CHEM. SOC. DALTON TRANS. 1994Table 5 Activation parameters for M-N fluxion in trans,cis-[RuX,(CO),(terpy)] and [M(CO),(terpy)] complexesComplex Temperature range/"C AHt/kJ mol-' ASt/J K-' mol-' AG* */kJ molCMo(CO),(terPY)l -70 to 30 52.4 f 1.6 13.3 f 6.6 48.4 f 0.4CW(CO),(terPY)l -30 to 40 69.7 f 1.9 45.9 f 6.7 56.0 f 0.1trans,cis-[RuCl,(CO),(terpy)] 3&100 88.3 f 1.4 41.8 ? 4.1 75.8 f 0.2t runs, cis-[ Ru Br , (CO),( terp y )] 86.2 _+ 2.6 29.4 f 7.5 77.4 f 0.3trans,cis-[RuI,(CO),(terpy)] 30-100 83.1 f 0.9 14.8 2 2.6 78.7 f 0.130- 1 00* At 298.15 K .Mechanism (i)LMechanism (ii)Fig. 4(i) and the 'tick-tock' twist mechanism (ii)Two possible mechanisms for the M-N linkage fluxion for a bidentate terpy in an octahedral metal complex, namely the rotation mechanismto that observed for the terpy complexes of Re' but is in contrastto that for PtIV where no halogen dependence was observed.Theonly values available for direct comparison are those from ourwork with the complexes [PtXMe,(terpy)] and [ReX(CO),-(terpy)] (X = C1, Br or I) where the AG values range from 61.5to 62.5 and 70.3 to 73.0 kJ mol-' respectively. Thus the orderingof AG with respect to metal for these d6 octahedral complexesis Ru" > Re' > PtIV > Wo > Moo, for the metal-terpyridinecommutation.Although we were unable to obtain quantitative data for Cr',the qualitative observation of AGt values in the orderCr > W > Mo is in contrast to values for fluxional motions ofM(C0)5 moieties undergoing 1,2 or 1,3 shifts on N ligands 21q22or on S ligands23,24 where the orders are W z Cr > Mo andW > Cr > Mo respectively.Two possible mechanisms for the terpy fluxion are illustratedin Fig.4. Mechanism (i) involves a five-co-ordinate intermediatewhereas mechanism (ii) involves a seven-co-ordinate inter-mediate. In the [PtXMe,(terpy)] complexes, mechanism (ii)was shown to operate since exchange of the equatorial (truns-N)Pt-Me environments occurred. When we reported our resultson the [ReX(CO),(terpy)] complexes we were unable to deducewhich mechanism was operating because of problems ofsolubility of the complexes. We have now overcome theseproblems and have been able to show by 13C NMRspectroscopy that the equatorial (trans-N) carbonyl environ-ments do exchange and hence mechanism (ii) is also operatingin the rhenium complexes.In the present work we have beenunable to deduce the precise mechanism because of a com-bination of insolubility, instability and conversion to terdentateligand species. However in view of the structural similarity ofthe metal complexes involved, we believe that the 'tick-tock'mechanism (ii) is also operating in these Mo, W and Rucomplexes. This postulate receives further support from theobservation that for the Ru complexes the fluxion and thedecarbonylation occur simultaneously and the trans orientationof the halogens is retained in the resulting terdentate terpycomplex. If it is assumed that the fluxion and the decar-bonylation involve the same intermediate, then the seven-co-ordinate intermediate of mechanism (ii) is favoured since thefive-co-ordinate intermediate of mechanism (i) would be highlyfluxional and probably result in the formation of cis- and trans-monocarbonyl products.AcknowledgementsWe are most grateful to the SERC for aH.M. P.References1 Part 2, E. W. Abel, V. S. Dimitrov, NA. G. Osborne, H. M. Pain, V. Sik,maintenance grant forJ. Long, K. G. Orrell,M. B. Hursthouse andM. A. Mazid, J. Chem. Soc., Dalton Trans., 1993, 597.2 E. C. Constable, Adv. Inorg. Chem. Radiochem., 1987,30, 69.3 E. Seddon and K. Seddon, The Chemistry of Ruthenium, Elsevier,4 A. Juris, V. Balzani, F. Barigelleth, S.Campagna, P. Belser and5 M. C. Ganorkar and M. H. B. Stiddard, J. Chem. SOC., 1965, 5346.6 C. C. Addison, R. Davis and N. Logan, J. Chem. Suc., Dalton Trans.,7 R. D. Chapman, R. T. Loda, J. P. Riehl and R. W. Schwartz, Znorg.8 A. J. Canty, N. Chaichit, B. M. Gatehouse, E. E. George and9 G. B. Deacon, J. M. Patrick, B. W. Skelton, N. C. Thomas andAmsterdam, 1984, p. 1173.A. Von Zelewsky, Coord. Chem. Rev., 1988,84,85.1974,2070.Chem., 1984,23, 1652.G. Hayhurst, Znorg. Chem., 1981, 20, 2414.A. H. White, Aust. J. Chem., 1984,37, 929.10 N. C. Thomas and J. Fischer, J. Coord. Chem., 1990,21, 1 19.11 P. A. Anderson, F. R. Keene, E. Horn and E. R. T. Tiekink, Appl.Organomet Chem., 1990,4523.12 E. R. Civitello, P. S. Dragovich, T. B. Karpishin, S. G. Novick,G . Bierach, J. F. O'Connell and T. D. Westmoreland, Znorg. Chem.,1993,32, 237.13 E. W. Abel, N. J. Long, K. G. Orrell, A. G. Osborne, H. M. Pain andV. Sik, J. Chem. Soc., Chem. Commun., 1992, 303.14 E. W. Abel, V. S. Dimitrov, N. J. Long, K. G. Orrell. A. G. Osborne,V. Sik, M. B. Hursthouse and M. A. Mazid. J. Chem. Soc., DaltonTrans., 1993,291.15 M. A. Bennett, L. Pratt andG. Wilkinson, J. Chem. Soc., 1961,2037.16 R. B. King and A. Fronzaglia, Inorg. Chem., 1966,5, 1837.17 D. F. Shriver, Manipulation of Air-sensitive Compounds, McGraw-Hill, New York, 1969116 J. CHEM. SOC. DALTON TRANS. 199418 D. A. Kleier and G. Binsch, DNMR3, Program 165, Quantum19 V. Sik, Ph.D. Thesis, University of Exeter, 1979.20 B. P. Sullivan, J. M. Calvert and T. J. Meyer, Inorg. Chem., 1980,19,21 M. N. Ackerman, R. M. Willett, M. H. Englert, C. R. Barton and22 K. R. Dixon, D. T. Eadie and S. R. Stobart, Inorg. Chem., 1982,21,23 E. W. Abel, S. K. Bhargava, P. K. Mittal, K. G. Orrell and V. Sik,J. Chem. SOC., Dalton Trans., 1985, 1561.24 E. W. Abel, S. K. Bhargava, T. E. MacKenzie, P. K. Mittal,K. G. Orrell and V. Sik, J. Chem. Soc., Dalton Trans., 1987,757.Chemistry Program Exchange, Indiana University, IN, 1970.1404.D. B. Shewitz, J. Organomet. Chem., 1979, 175,205.4318. Received 20th September 1993: Paper 3105644