zyxwvu zyxwvuts zyxw zyxw zyxw zyxw Organometallics 1987, 6 , 2548-2550 2548 Cationic Intermediates in Oxidative Addition Reactions of Alkyl Halides to de Complexes: Evidence for the SN2 Mechanism Margarita Crespo and Richard J. Puddephatt" Department of Chemistty, University of Western Ontario, London, Ontario, Canada N6A 587 Received April 28, 1987 The oxidative addition reactions of RX to [PtMezLz](1) to give [PtXMe2RL2](3) (L, = 2,2'-bipyridine, RX = MeI, PhCH,Br; L = PMe2Ph,RX = MeI) are shown to occur by way of an ionic intermediate, [PtSMe2RL2]+X-(2), when the solvent (S) = CD3CN and, for RX = Me1 and Lz = 2,2'-bipyridine, when S = CD30D or acetone4 The intermediates 2 were detected and characterized by low-temperature *H and, when L = PMezPh,Fs;PNMR studies. A t temperatures above -20 "C the ligand substitution of Xfor S occurred rapidly and hence the intermediateswere detected only for the most reactive systems, which underwent oxidative addition at temperatures below -20 "C. Less reactive systems, for example, Et1 with [PtMez(bpy)]or Me1 with [IrC1(CO)(PPh3),],failed to give detectable ionic intermediates. The observation of ionic intermediates provides strong evidence for the sN2 mechanism of oxidative addition. during oxintermediate, fuc-[PtMe3(SMe2),(CD3CN)]+I-, idative addition of Me1 to c i ~ - [ P t M e ~ ( S M ein~ CD3CN )~] to give fuc-[PtMe31(SMe2),]was therefore surprising7and led us to make a more general study of the conditions under which such cationic intermediates could be detected. The results are reported below. Introduction The sN2 mechanism of oxidative addition of alkyl halides was first proposed on the bases of the kinetic order and the similarity of the activation parameters, for reaction of Me1 with tr~ns-[IrCl(CO)(PPh~)~], to those for the Menschutkin reaction.' An sN2 mechanism should lead to inversion of configuration for chiral alkyl halides, and this has been confirmed in some cases.2 However, free radical chain and non-chain mechanisms are also important in many oxidative additions, and there have been difficulties in distinguishing between the possible mechanisms in some case^.^^^ The oxidative addition of Me1 to square-planar da complexes is of especial interest because of its role in the Monsanto process for acetic acid.4 In some cases the reaction is accelerated in the presence of iodide, due to a preequilibrium involving formation of a five-coordinate 18-electron complex (eq 1,phen = 1,lO-phenathroline, cod = 1,5-~yclooctadiene).~ I-, = C Results and Discussion The scope of the reaction of eq 2 was investigated. Me, ,Pt< hl e L L +RX - Me, CD3CN I N *,L i K CIr(phenNcod)I+ IrI(phen)tcod)l -1- I 1 hi Me1 R2 M c I . C I r ( p hen) (cod ) M e 1 2 + I - fPSt -I- (1) C I r I M e ( p h e n ) t c o d )If Both [Ir(phen)(cod)]+and [IrI(phen)(cod)]reacted by the SN2mechanism, but the ratio k z / k l = 7. The proposed intermediate [Ir(phen)(c~d)Me]~+ was not identified presumably because iodide addition to give the product [IrIMe(phen)(cod)]+was very fast.5 The observation of such cationic intermediates would be good evidence for the sN2 mechanism of oxidative addition, since they are not expected in free radical or concerted mechanisms of reaction.,% However, numerous attempts to identify such species from 16-electron precursors have been unsuccessfu1.2-6 The observation by low-temperature NMR of an (1) Chock, P. B.; Halpern, J. J . Am. Chem. Soc. 1966,88, 3511. (2) For excellent recent reviews, see: Stille, J. K. In The Chemistry of the Metal-Carbon Bond; Hartley, F. R.; Patai, S., Eds.; Wiley: New York, 1985; Chapter 9. Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G . Principles and Applications of Organotransition Metal Chemistry; University Science Books: Mill Valley, CA, 1987;Chapter 5. (3) Lappert, M. F.;Lednor, P. W. Adu. Organomet. Chem. 1976, 14, 345. (4) Forster, D. J . Chem. SOC.,Dalton Trans. 1979, 1639. Hickey, C. E.; Maitlis, P. M. J . Chem. Soc., Chem. Commun. 1984, 1609. (5) de Waal, D. J. A,; Gerber, T. I. A.; Louw, W. J. Inorg. Chem. 1982, 21, 1259. (6) Cationic intermediates are often observed in oxidative addition of 18-electron complexes such as [Ir(q-C,H,)(CO)(PPh,)]. Hart-Davis, A. J.; Graham, W. A. G. Inorg. Chem. 1970,9, 2658; 1971, 10, 1653. The reactions were carried out at low temperature in an NMR tube using CD3CNsolvent, monitoring the progress of the reaction at -40, -20, and 0 "C. When RX = Me1 and L2 = 2,2'-bipyridine, the spectrum at -40 "C contained peaks due to both 2a and 3a (but none due to la). On further warming peaks due to 2a decayed and pure 3a was p r e ~ e n t . ~The . ~ initial oxidative addition was shown to be trans by use of CD31as reagent. Initially, neither the intermediate 2a' nor the product 3a' showed resonances at 6 0.47 or 0.58 for the methylplatinum resonance trans to CD3CN or I, respectively (Figure 1). However, on further reaction 2a' decayed to give 3a' and scrambling of the CH3 and CD3 groups occurred, this scrambling occurring faster for 2a' than for 3a' (Figure 1). The intermediate 2a is characterized by a high value of 2J(PtCH,) = 78 Hz for the methyl group trans to CD3CN, which has a low trans i n f l ~ e n c e . ~ Very similar results were obtained with cis-[PtMe,(PMe2Ph)z](lb) though it reacts less rapidly than la. In this case reaction occurred at -20 "C to give 2b, and then, on warming to 0 "C, complex 3b was formed. This reaction zyxwvuts zyxw (7) Puddephatt, R. J.; Scott, J. D. Organometallics 1985, 4 , 1221. (8) Clegg, D. E.; Hall, J. R.; Swile, G. A. J . Organomet. Chem. 1972, 38, 403. (9) Jawad, J. K.; Puddephatt, R. J. J. Chem. SOC.,Dalton Trans. 1977, 1466. (10)Appleton, T. G.; Clark, H. C.; Manzer, L. E. Coord. Chem. Reu. 1973, 10, 335. (11) Ruddick, J. D.; Shaw, B. L. J . Chem. SOC.A 1969, 2801. 0276-733318712306-2548$0l.50/0 0 1987 American Chemical Society zyxw zy zy zyxwvutsr zyxwvutsrqpo zyxwvut Organometallics, Vol. 6, No. 12, 1987 2549 Oxidative Addition Reactions of Alkyl Halides Table I. 'H and 31PNMR Dataa R trans to X Me trans to N [PtMeAbpy)1 ( 1 4 [PtMe31(bpy)l ( 3 4 [PtMe.EtIlbDv)l (3d) * . . .",> (JP~H) (JHH) 0.58 (76), Me 1.39 (72) (7.6). CHa 0.01 (68) (7.6), CH; 2.75 (921, CHzPh 6.3 (m)-6.6 (m), Ph 2.80 (93), CH2Ph 6.3 (m)-6.6 (m), Ph 0.47 (78), Me 0.54 (80), Me 0.64 (80), Me 0.60 (801, Me 2.84 (96), CHzPh [PtMezBzCl(bpy)] (3e) [PtMezBzBr(bpy)] (3c) [PtMe3(CD&N)(bpy)1I ( 2 4 [PtMe3(CD3OD)(bpy)1IWb [PtMe3(CH3I)(bpy)1I(2fY [PtMe3(CD3COCD3)(bpy)]I(2eY [PtMezBz(CD3CN)(bpy)]Br(2c) MePt6(H) (JPtH) (JPH) [Pt(CD3)z(PMe2Ph)z1 (lb') [Pt(CD3)zMeI(PMeziPh)2] (3b') 0.63 (70) (7.6) [Pt(CD3)zMe(CD3CN)(PMezPh)z]I (2b') 0.59 (75) (8.6) bpy resonances 6(H3), OL), W M , 8.22, 8.19, 7.59, 9.17 8.66, 8.17, 7.71, 8.95 8.55, 8.29, 7.82, 9.04 1.32 (70) 8.20, 8.04, 7.55, 8.66 1.40 (70) 8.17, 8.01, 7.51, 8.63 1.14 (68) 1.13 (66) 1.01 (71) 1.17 (68) 1.30 (67) 8.52, 8.27, 7.82, 8.87 Solvent CD3CN unless otherwise stated. J values in hertz. PMe2Ph6 ( W (JP~H) (JPH) 1.41 (20) (7.9) 1.63 (12) (9) 1.70 (11) (5.3) 1.51 (11) (5.3) 1.56 (11) (5.7) W) (JPtP) -12.29 (1842) -36.28 (1200) -46.51 (1192) 6(H) (JPH) 0.98 (5.7) 1.14 (5.3) [IrClI(CH3)(CO)(PPh3121 [I~Iz(CH~)(CO)(PP~~)ZI a 6(H) (JPtH) 0.89 (86) 1.36 (70) 1.35 (71) zyxwvutsrq WU Solvent CD30D. "Solvent (CD3)&0. 1 2a.S :CD3CN 30 lb, L:PMe2Ph 2b,S:CD3CN 3b 2a a) 2b lb I ,& - Figure 1. 'H NMR spectra (200 MHz) during the reaction of CDBIwith [PtMez(bpy)] in CD3CN: (a) before addition of CD31; (b) spectrum a t -40 "C; (c) spectrum a t -20 "C; (d) spectrum a t 0 "C. Figure 2. 31P(1H)NMR spectra (121.5 MHz) during the reaction (a) a t -40 "C; (b) of Me1 with c i ~ - [ P t M e ~ ( P M e ~ Pinh )CD,CN: ~] a t -20 "C; (c) at 0 "C; (d) a t 20 "C. was monitored by both 'H and 31P NMR spectroscopy in CD3CN were also unsuccessful. Again reaction occurred (Figure 2). The value of lJptp fell from 1842 Hz for l b at 0 OC to give the final products 3d and 3e. Qualitatively, to 1200 Hz for 2b and 1192 Hz for 3b, the changes being it seems that the ionic intermediates can be detected only typical of those expected on oxidation of platinum(I1) to when oxidative addition occurs at -20 "C or lower. Above platinum(IV).12 Oxidative addition was again trans, as this temperature, the reaction of 2 to give 3 occurs rapidly, shown by reaction of Me1 with ~~S-[P~(CD~),(PM~~P~)~], and hence 2 cannot be detected. There is good evidence and no scrambling of CH3 and CD3 groups occurred at that the reactions of l a with Et1 and PhCHzCl occur by room temperature in this system. the SN2mechanism,14 and so the ionic intermediate 2 is No ionic intermediate analogous to 2 could be detected expected to be formed. on reaction of trans-[IrC1(CO)(PPh3),]with Me1 in CD3Now, knowing that only the most reactive alkyl halides in combination with the most reactive transition-metal CN. No reaction was observed at temperatures up to -20 complexes were likely to give detectable ionic intermediOC, and at 0 OC reaction occurred to give [IrClIMeThis complex reacted further at room (CO)(PPh3),] ates, the reaction of PhCHzBr with la was studied.15 The ionic intermediate 2c was detected at -40 "C (Table I), and temperature to give [IrIzMe(CO)(PPh3)2],which was prepared independently from [IrC1(CO)(PPh3),]with Me1 further reaction to give 3c was complete at 0 "C. Only trans addition was observed in both 2c and 3c. in the presence of NaI. Attempts to detect ionic intermediates in reactions of Et1 or PhCHzCl to [PtMe,(bpy)] .12313 (12) Pidcock, A. Adu. Chem. Ser. 1982, No. 196, 1. (13) Stang, P. J.; Schiavelli, M. D.; Chenault, H. K.; Breidegam, J. L. n . n n n n m n + n l l ; n o lClP.4 9 11'2'2 (14) Monaghan, P. K.; Puddephatt, R. J. Inorg. Chim. Acta 1983, 76, T nnn LLJI. (15) Datta, D.; Sharma, G . T. Inorg. Chem. 1987,26,329. Schrauzer, G. N.; Deutsch, E. J . Am. Chem. SOC.1969, 91, 3341. zyxwvutsr zyxwvuts zyxwvutsrq 2550 Organometallics, Vol. 6, No. 12, 1987 Crespo a n d P u d d e p h a t t Solvents Which Give Ionic Intermediates. A survey of some polar solvents was made to determine which gave detectable ionic intermediates in eq 3. 2a 2d a 2' - Our attempts to identify ionic intermediates in oxidative addition to Vaska's iridium(1) complex have been unsuccessful. This is due to a slower rate of oxidative addition for the iridium(1) complex and, most probably, a higher rate of ligand substitution for the iridium(II1) cationic intermediate.'P2J6 Nevertheless, the direct observation of ionic intermediates in oxidative addition to platinum(II), together with the strong similarity in solvent effects on rates and in activation parameters for oxidative addition of methyl iodide to iridium(1) and platinum(I1) systems, strongly supports the s N 2 mechanism in both cases for reactions involving methyl iodide. 12 9 13,17 zyxwvutsrqponmlk zyxwvutsrq S CD3Ch 5 CD3CD 8 lCD3ZCC S ' e1 3c - The intermediate was most easily detected when S = CD3CN. However, when S = CD30D, the ionic intermediate 2d could be detected at -80 OC along with 3a. On warming, 2d reacted further to give 3a. When S = (CD&CO, two intermediates were detected at -80 "C and are thought to be 2e and 2f. The relative abundances of 2e and 2f were found to change as the concentration of Me1 was changed, 2f being more abundant at higher concentration of MeI. This situation presumably arises because Me1 is a better ligand for platinum than is acetone. Both intermediates had disappeared at -20 O C , and only 3a was present. NMR data are given in Table I. Conclusions This work has shown that ionic intermediates can be detected by low-temperature NMR spectroscopy during oxidative addition of alkyl halides to platinum(I1) complexes, provided the oxidative addition occurs at about -20 O C or lower. It is essential that the oxidative addition step should be faster than the subsequent substitution of halide for solvent in the ionic intermediate (eq 2 and 3), if the intermediate is to be detectable. It is then clear that the cationic intermediate is formed as a result of kinetic control, providing good evidence for the s N 2 mechanism of oxidative addition. We consider it likely that the solvent coordination (eq 2 and 3) occurs synchronously with the oxidative addition step, thus contributing to the large negative entropies of activation observedgand reducing the incipient positive charge buildup on the platinum center. The other experimental evidence, for example the trans stereochemistryof addition, is consistent with but does not prove this hypothesis. For octahedral platinum(1V) complexes the ligand substitution step is expected to occur by a dissociative (D or Id) me~hanism.'~Normally such substitutions in platinum(IV) complexes are very slow, but in this case it should be accelerated due to the high trans influence of the methyl group trans to solvent (eq 2 and 3) and the poor coordinating ability of the solvent molecules. As expected, the ligand substitution occurs more rapidly for the oxygen donor solvents, S = CD30D or acetone-d,, than for the nitrogen donor, S = CD3CN. I zyxw zyxw 1 9 Experimental Section NMR spectra were recorded by using Varian XL200 (lH) or XL300 (31P)NMR spectrometers, and chemical shifts are referenced to Mel& ('H) or (Me0)3P0 (31P). The complexes [PtMe2(bpy)]and c i ~ - [ P t M e ~ ( P M e ~ Pwere h ) ~ ]prepared by literature methods."J8 Most of the final products of oxidative addition were known complexes and were identified by comparison of the NMR spectra (Table I) with those of authentic Samp l e ~ . ~ , ~ , The ~ , ~NMR , ~ ~ spectral , ~ ~ , ~ studies ~ showed that the products were formed in essentially quantitative yields (e.g. Figures 1 and 2), and they were isolated after crystallization in yields of a t least 70%. The following new complexes were prepared. [PtC1Me2(C€12Ph)(bpy)].To a solution of [PtMez(bpy)](10 mg) in acetone (10 mL) was added PhCH2Cl (10 NL). After 15 min the product was isolated by evaporation of the solvents and purified by recrystallization from CH2Clz/pentane;yield 75%. Anal. Calcd for C,9H21C1N2Pt:C, 44.9; H, 4.2; N, 5.5. Found: C, 44.5; H, 4.4; N, 5.0%. [PtBrMez(CH2Ph)(bpy)]was prepared in a similar way; yield 78%. Anal. Calcd for C19H21BrN2Pt:C, 41.3; H, 3.8; N, 5.1. Found: C, 41.2; H, 3.9; N, 5.3. Detection of Intermediates. Typically a solution of [PtMe2(bpy)](5 mg) in CD3CN (0.5 mL) in an NMR tube was cooled to -45 "C, and CD31(3 fiL) was added. The tube was then placed in the cooled probe of the NMR spectrometer, and spectra were recorded at -40, -20,0, and 20 "C. Results are given in Table I and Figures 1 and 2. Acknowledgment. We thank NSERC (Canada) for financial support. Registry No. la, 52594-52-2; lb, 24917-48-4; lb', 69721-14-8; 2a, 11063829-4;2a', 110638-30-7;2b, 110638-31-8;2b', 110638-32-9; 2c, 110661-47-7;2d, 110638-33-0;2e, 110638-34-1;2f, 110638-35-2; 3a, 38194-05-7; 3a', 110638-36-3;3b, 24833-69-0; 3b', 110661-48-8; 3c, 62343-16-2; 3d, 62342-98-7; 3e, 110638-37-4; trans-[IrCl(CO)(PPh,),], 15318-31-7; IrClIMe(CO) (PPh,),, 24315-50-2; Ir12Me(CO)(PPh3)z, 110716-19-3. (16) Langford, C. H.; Gray, H. B. Ligand Substitution Processes; W. A. Benjamin: New York, 1965. (17) Jawad, J. K.; Puddephatt, R. J. J.Organomet. Chem. 1976,117, 297. (18) Monaghan,P. K.; Puddephatt, R. J. Organometallics 1984,3,444.