J . Am. Chem. SOC.1991, 113, 1225-1236 achieve this, it is necessary to carefully control the temperature and nature of the colloid surface coating. This latter condition is well met by using citrate-reduced sols. Adsorbed citrate ions coat the silver surface and confer a urotein-comuatible surface. This organic layer may prevent the heme approac'hing too closely to the silver surface, especially when a pH < p l is used to activate a Drotein/colloid interaction. This is critical in avoiding denatuiation by heme extraction, in cytochrome P-420 formZion, or in avoiding undesirable low- to high-spin-state conversion. The
1225
main features of the spectra of each of four P-450sstudied are summarized in Table I. That the cytochromes P-450 are relatively intact and biologically active on the sol surface is exemplified by the resDonse to substrate of the adsorbed Droteins. Acknowledgment. We thank the Science and Engineering Research (U*K.) for financial sup!Jort. Registry No. Ag, 7440-22-4; cytochrome P450, 9035-51-2; benzphetamine, 156-08-1; citric acid, 77-92-9: heme, 14875-96-8.
Intrinsic Barriers to Atom Transfer: Self-Exchange Reactions of CpM( CO),X/CpM( C0)3- Halide Couples Carolyn L. Schwarz, R. Morris Bullock, and Carol Creutz* Contribution from the Department of Chemistry, Brookhaven National Laboratory, Upton, New York 1 1 973. Received January 31, 1990
Abstract: Rate constants and activation parameters were measured for the self-exchange of CPM(CO)~-with CPM(CO)~X
in CD3CN solvent: CpM(CO)3- + CpM(CO),X
+ CpM(C0)3X
+ CpM(CO)3-
The self-exchange reactions were followed by 'HNMR spectroscopy: For the X = I complexes, standard line width measurements yicld ( M = Mo) k(298) = 1.5 X IO4 M-I s-I (4H* = 6.4 (f0.4) kcal mol-', 4S* = -18 (k1.5) cal K-' mol-') and (M = W) k(298) = 4.5 X IO3 M-' s-' (4H' = 7.5 (hO.1)kcal mol-', 4S* = -16.8 (hO.5) cal K-' mol-'). For the X = Br complexes, magnetization-transfer experiments yield (M = Mo) k(298) = 1.6 X IO' M-' s-I (4H* = 12.1 (h4.5)kcal mol-', pS* = -12 ( A I S ) cal K-' mol-') and ( M = W) k(298) = 1.5 M-'s-' (4H' = 15.1 (k5.2) kcal mol-', 4S* = -7 (f16) cal K-' mol-'); 'H NMR longitudinal relaxation times TI for the Cp groups of the reactants are typically 40 s. The X = CI systems were studied by conventional techniques, with the rates of "transfer" of Cp-ds from (Cp-d,)W(CO)< to CpW(CO)3CIbeing monitored; for M = Mo, k(298) = 9.0 X IO-* M-'s-' (4H' = 18.9 (kl.0) kcal mol-', 4S* = 0 (f4) cal K-' mol-') and for M = W, M-' s-l (4H* = 17.7 (k3.3) kcal mol-', 4S* = -1 1 (*I I ) cal K-' mol-'). For X = C'H,, the CpWk(298) = 2.1 X (C0)3-/CpW(CO),CH3 self-exchange rate constant, also determined by monitoring rates of "transfer" of Cp-d, from (CpM-' s-I a t 335 K. The latter self-exchange reaction is discussed in terms of the ds)W(CO); to CPW(CO)~X,is = I X intrinsic barrier for oxidative addition to the anion. For the X = halogen exchanges, the role of the M(1) ("metal radical") state is considered, and it is concluded that the latter isovalent state is not an intermediate for these systems. However, the isovalent state may serve to stabilize the transition state for two-electron transfer between the metal centers through configuration intcraction. Our results, taken with those for other X = halogen systems, indicate that effective transfer of X+ may be intrinsically rapid when both reactants are 18-electron species and steric factors are favorable.
Introduction The concept of factoring the kinetic barrier to a reaction into intrinsic and thermodynamic components has proven remarkably powerful for outer-sphere electron-transfer reactions.' In recent years, this approach has also been applied to proton-transfer, methyl (CH3+)-transfer, and other "atom"-transfer reaction^.^,^ In these applicatons, a crucial parameter is the free energy barrier to the self-exchange (4C0 = 0) reaction, 4C*,,. While 4G*,, values have been inferred from kinetic data for net (4C0# 0) reactions, only rarely (with the exception of outer-sphere electron-transfer reactions) have the intrinsic barriers been evaluated by direct study of the self-exchange process. It is becoming apparent that transition-metal complexes are excellent substrates for such studies. In this paper we describe our studies of "atomtransfer" self-cxchangc reactions of CpM(CO),-/CpM(CO),X couples. In our experiments a range of IH N M R techniques provide a probe of the self-exchange process analogous to that ~
~~~
( I ) Sutin, N . f r o g . Inorg. Chem. 1983, 30, 441. (2) Albery, W. J . Ann. Rev. Phys. Chem. 1980, 31, 221. (3) (a) Riveros. J. M.; Jose, S. M.; Takashima, K. Adu. Phys. Org. Chem. 1985, 21, 197. (b) Pelerite, M. J.; Braumann, J. I . In Mechanistic Aspects of Inorganic Reactions: Rorabacher, D. B., Endicott, J . F., Eds.; ACS Symposium Series; American Chemical society: Washington, DC, 1982; Vol. 198, p 81; J. Am. Chem. Soc. 1983, 105, 2612.
0002-7863/91/ 15 13- I225$02.50/0
of radioisotopes in the earliest self-exchange measurements. Following Taube4 we adopt the term "atom transfer" for reactions in which an atom originating in either the oxidizing or reducing agent is transferred to the reaction partner so that in the activated complex both oxidizing and reducing centers are attached to the atom being transferred. In this sense, "atom transfer" is intended as a broad reaction class and is not restricted to reactions in which a single neutral atom is transferred (for example, a hydrogen atom or halogen atom abstraction). Reactions falling within this class include one-electron and twoelectron inner-sphere electron-transfer reaction^,^ halogen- and hydrogen-atom abstractions, hydride-transfer reactions, and certain proton-transfer and nucleophilic substitution reactions. With, perhaps, one6 exception, all known two-electron-transfer reactions (4) Taube, H. I n Mechanistic Aspects of Inorganic Reactions; Rorabacher, D. B., Endicott, J. F., Eds.; ACS Symposium Series; American Chemical Society: Washington, DC, 1982; Vol. 198, p 151. ( 5 ) (a) Taube H. Electron Transfer Reactions of Complex Ions in Solution; Academic Press: New York, 1970. (c) Endicott, J. F. f r o g . Inorg. Chem. 1983, 30, 141. (d) Haim, A. Ibid. 273. ( 6 ) (a) Dodson, R. W. I n Mechanistic Aspects of Inorganic Reactions; ACS Symposium Series; Rorabacher, D. B., Endicott, J . F., Eds.: American Chemical Society: Washington, DC. 1982; Vol. 198, p 132. (b) Schwarz, H. A.; Comstock, D.; Yandell, J. K.; Dodson, R. W. J . Phys. Chem. 1974, 78, 488.
0 1991 American Chemical Society
Schwarz et al.
1226 J . Am. Chem. SOC.,Vol. 113, No. 4, 1991
in solution involve atom transfer. Such reactions are central to many catalytic changes effected by transition-metal centers and, indeed, to much chemistry in general. Self-exchange studies of only a few atom-transfer couples have been made. One of the earliest and most comprehensive was of the Pt(II)/Pt(IV)7-9 exchange (eq 1). However, these systems are complicated by
+
t r a n ~ - P t ~ " ( N H ~ ) ~ C l , *CI+
+ PtlI(NH3)42+ [CIPt(NH3)4CIPt(NH3)4CIJ 3 +
[CIPt(NH3)4CIPt(NHj)4Cl]3+ F= Pt11(NH3),2++ CI- trans-Pt'V(NH3)4C122+ (1)
+
the intrinsic coordination number mismatch of the 16-electron Pt( 11) and 18-electron R(IV) complexes, since electron "exchange" requires loss of two ligands from reactant Pt(1V) and the gain of two ligands by reactant Pt(I1). Recently self-exchange rate constants were determinedi0for a series of 18-electron ruthenium and osmium MCp, complexes (eq 2) where X = I or Br and Cp
= cyclopentadienide;the pressure dependence of the rate constants for M = Ru, X = Br has also been reported." These systems, which (at least formally) involve two-electron transfer coupled to atom transfer, have distinct advantages over previously studied two-electron systems since they are not complicated by possible solvent coordination and differing electron counts. The protontransfer self-exchange studies of 18-electron metal hydrides and their 18-electron, conjugate bases (e.g., eq 3; M = Cr, Mo, W) reported by Norton and colleaguesI2 also fall within the broad definition of atom transfer. Recently oxo-transfer self-exchange CpMo(C0)3- + CpM1'(C0)3H CpM"(CO),H
+ CpMo(CO),-
(3)
reactions have been studied,13 and a reversible three-electron transfer between manganese porphyrins accompanied by nitrogen-atom transfer has been r e ~ 0 r d e d . l ~ In the present work we have measured rate constants and activation parameters for the self-exchange reactions of metal anions with their metal halide or methyl complexes (eq 4) with CpMo(CO)3-
+ CpM"(CO),X
+
CpM"(CO),X
+ CPMO(CO)~-(4)
M = W or Mo and X = CI, Br, or I (the M = Cr members of the series being omitted because the C P C ~ ( C O ) ~complexes X are not stable). These I8-electron, d6-d4 systems are isoelectronic with the Ru and Os MCp, systemsI0 in eq 2 and are also related to the proton exchanges in eq 3 (eq 4, X = H). As will be seen, the self-exchange rate constants determined for eq 4 in this study span many orders of magnitude and require the exploitation of a wide range of IH NMR techniques, including line broadening, magndtiyation transfer, and conventional kinetic techniques. Experimental Section Materials. PPN [CpMo(CO),] (PPN = bis(tripheny1phosphine)iminium cation and C p = cyclopentadienyl anion) was prepared by a
(7) (a) Basolo, F.;Wilks, P. H.; Pearson, R. G.; Wilkins, R. G. J . Inorg. Nucl. Chem. 1958, 6, 161. (b) Basolo, F.; Morris, M. L.; Pearson, R. G. Discuss. Faraday SOC.1960, 29, 80. (8) Cox, L. T.; Collins, S. 8.; Marin, D. S. J . Inorg. Nucl. Chem. 1961, 17, 383. (9) Mason, W. R.; Johnson, R. C. Inorg. Chem. 1965, 4, 1258. (IO) Smith, T. P.; lverson, D. J.; Droege, M. W.; Kwan, K. S.;Taube, H. Inorg. Chem. 1987, 26, 2882. ( I I ) Kirchner, K.; Dodgen, H. W.; Wherland, S.; Hunt, J . P. Inorg. Chem. 1989, 28, 604. (12) Edidin, R. T.: Sullivan, J. M.; Norton, J. R. J. Am. Chem. SOC.1987, 109, 3945. (13) Holm, R. H. Chem. Reo. 1987.87, 1401. (14) Woo, L. K.; Goll, J. G. J . Am. Chem. SOC.1989, / I / , 3755
slight modification of the literature method,,' with NaKi6 instead of Na/Hg being used to reduce the dimer [ C ~ M O ( C O ) ~ PPNICpW]~. (CO),] was prepared from Na[CpW(CO),] and PPNCI as described for the Mo a n a l o g ~ e . 'Deuterated ~ cyclopentadiene, CsD6, was prepared by a slight modification of the literature procedure," by using three exchanges of cyclopentadiene with deaerated NaOD/D,O. The C5D6 was converted to [ ( C P - ~ ~ ) M ( C O ) , ] ~ ,( 'M * = Mo, W). Reduction of the dimers followed by counterion exchange gave PPN[(Cp-d,)M(CO),] (M = Mo, W). The iodides, CpM(CO),I (M = Mo, W), were prepared by 1, oxidation of [CpM(CO),], in CH2CI2.l9 The bromides, CpM(CO),Br, were prepared20 from the reaction of CpM(CO),H with CHBr, instead of CBr,. The chlorides, CpM(CO),CI, were prepared from CpM(CO),H and CC1,.2' CpW(CO),CH, was prepared by reaction of CH,I and Na[CpW(CO),] in THF.21 CD3CN (Aldrich) was dried over P,Oi0, then vacuum transferred, and stored over 3A molecular sieves. Sample Preparation. Samples were prepared in one of two ways. (I) A weighed sample of the compound was placed in an N M R tube in an argon drybox. On a vacuum line, CD,CN was transferred into the N M R tube and the tube was flame sealed. The volume of the solution was calculated from the height of solvent in the N M R tube.22 (2) I n an argon drybox, known quantities of compound were placed in a volumetric flask and diluted to the mark with CD3CN. A portion was transferred by pipet to an N M R tube. The tubes were then flame sealed under vacuum. Prepared N M R tubes were stored in a -81 "C freezer until used. 'H NMR Measurements. The N M R data (300 MHz) were collected on a Bruker AM-300 N M R equipped with an Aspect 3000 computer and a VT-I000 variable temperature unit. The variable temperature unit was calibrated with MeOH (0.1% HCI)*, and ethylene glycolz4 over the temperature range 235-350 K. The unit was calibrated with the appropriate solution, before each N M R run. The N M R instrument was allowed to temperature equilibrate for 2 h prior to any single accumulation experiments in order to maintain an unvarying lock. The lock unit was found to be intolerant of room temperature changes greater than f 1 "C for the single-accumulation experiments. Line broadening experiments were performed on the iodo systems over a range of temperatures and varying CpM(CO),l concentration with constant PPN[CpM(CO),] concentration. Line widths were determined by measuring the full width at half-height (FWHH) of the C p resonances. The line widths in the absence of exchange for the individual compounds were 0.5 f 0.2 Hz, and their variation with temperature was found to be negligible. Dioxane was used as an internal line width standard. Magnetization-transfer experiments were performed on the bromide systems with a DANTE pulse sequence.2s This pulse sequence has the form
Dl-(-P ,-D~-)"-T-FI D where D , is a long delay of 5 times the longest TI value, PI is I / n of the 180" pulse width, D, is a short delay, T is a variable delay, and FID is the free induction decay. The pulse sequence selectively inverts by excitation one of the resonances; then, the observed pulse being delayed by a specified interval T , the change in magnetization of the other exchange-coupled resonance is followed with time. I n this study n = I O was used. This resulted in a coarse digitization parameterization26and possibly some small errors in the baseline, but none were readily apparent. The 180° pulse width was measured prior to each run. A 90" pulse width was used in collecting the FID. The short delay D, was I ms, and the 7 value varied over a range of 0.5-400 s. Only one accumulation was taken at each delay because of the long TI values and the long delays needed between data points. The intensity data were input into a fourparameter nonlinear least-squares fit program written by Perkin2' and (15) Watson, P. L.; Bergman, R. G. J . Am. Chem. Soc. 1979, 101, 2055. (16) Ellis, J. E.; Flom, E. A . J . Urganomef. Chem. 1975, 99, 263. (17) Gallinella, E.; Mirone, P. J . Labelled Compds. 1971, 7 . 183. (18) Birdwhistell, R.; Hackett, P.; Manning, A . R. J . Urganomef.Chem. 1978, 157, 239. (19) (a) Abel, E. W.;Singh, A,; Wilkinson, G. J . Chem. SOC.1960, 1321. (b) Burckett-St. Laurent, J, C. T. R.; Field, J. S.; Haines, R . J.; McMahon, M . J . Organomet. Chem. 1978, 153, C19. (20) Sloan, T. E.; Wojcicki, A. Inorg. Chem. 1968, 7, 1268. (21) Piper. T. S.; Wilkinson, G. J . Inorg. Nircl. Chem. 1956, 3, 104. (22) Bryndza, H . E.; Calabrese, J. C.; Marsi, M.; Roe, D. C.; Tam, W.; Bercaw, J . E. J . Am. Chem. Soc. 1986, 108, 4805. (23) Van Geet, A. L. Anal. Chem. 1970, 42,679. (24) Van Geef, A . L. Anal. Chem, 1968, 40, 2227. (25) Morris, G.A,: Freeman, R. J . Magn. Reson. 1978, 29, 433. ( 2 6 ) Wu, X.-L.; Xu, P.; Friedrich, J.; Freeman, R. J . Magn. Reson. 1988, 81, 206.
J. Am. Chem. Soc., Vol. 113, No. 4, 1991 1227
Self- Exchange Reactions of CpM(C0)3X/CpM(CO)3Halide Couples modified by Brunschwig. An important input parameter is the true TI value for each component at each temperature. These were determined separately with use of an inversion recovery pulse sequence on the Aspect 3000. The T,’s were evaluated by using the TI fit program on the NMR, or, in some cases, the Tl’swere calculated from plots2*of In [[(a)- I ( f ) ] vs T , where the slope is - l / T l The chloro systcms were monitored by conventional N M R techniques. Thc growth of the C p resonance for CPM(CO)~-was followed with time after mixing CpM(CO),CI with the perdeuterated C p complex, (Cpd5)M(CO)c (eq 5, M = Mo, W). This method was also used to investigate thc rate of methyl exchange, X = C H 3 with M = W. (Cp-d,)M(CO),-
+ CpM(C0)’X * ( C P - ~ S ) M ( C O )+ J C P M ( C O ) ~ -(5)
Results The room-temperature 300-MHz proton N M R spectra of the complexes in CD3CN exhibit a single resonance for the cyclopentadienyl protons. The resonances of the anion and halo complexes are well separated. The anion resonances lie at 5.04 ppm for Mo and at 5.08 for W. The resonances for the halo complexes CpM(CO),X lie, respectively, at 5.83, 5.72, and 5.73 ppm for X = I, Br, and CI when M = Mo and at 5.82, 5.84, and 5.84 ppm for X = I , Br, and CI when M = W. The second-order rate constants k,, and the activation parameters for the CpM(CO),-/CpM(CO),X self-exchange reactions were measurcd with use of three different N M R techniques. C P M ( C O ) ~ -+ CpM(CO),X
CpM(CO),X
[CpM(COkI, M 0.06 14
0.06 1 3
0.0635
0.061 5
+ CpM(CO)3-
(4)
Line Broadening. The relatively rapid exchange rates for the CpM(CO),I systems were best determined by standard line width measurements. The spectrum of a solution 0.05 M in CpM(CO),I and in (PPN)[CpM(CO),] in CD3CN at rmm temperature shows two very broad and overlapping resonances. As the solution is cooled, the peaks sharpen and separate (Figure I ) . For both the W and Mo complexes, the corrected line width (experimental line width minus the line width in the absence of exchange) increases with increasing temperature, placing these systems in the slowexchange regime.29 The analysis of the data thus takes the simple form given in eq 6.,O In eq 6, AuI,2M.MIis the F W H H of CpM-
(CO),- ( M ) in the presence of a given concentration [MI] of CpM(CO),I (MI) and bll2M is the F W H H of CpM(CO),- in is the F W H H the absence of CpM(CO),I. Similarly, AUI/~MI,M of CpM(CO),I in the presence of a given concentration [MI of CpM(CO),-, and is the F W H H of C P M ( C O ) ~ in I the absencie of CpM(CO),-. Corrected line widths obtained at a single temperature were plotted against the concentration of the reaction partner, and k,, was determined from the slope. Line width data are summarized in Table I, and ke, values extracted from the concentration dependences of the line widths are listed in Table 11. Figure 2 shows the behavior of CpW(CO),-and CpW(C0)J resonances as a function of CpW(CO)31concentration at 290 K. Small amounts of impurities which form as the solution ages over the course of the measurements do not interfere with the determination. Magnetization Transfer. The rates of exchange in the CpM(CO),Br and CPM(CO)~CIsystems were too slow for measurement by a simple line broadening technique. However, the CpM(CO),Br systems did prove amenable to study by magnetization-transfer techniques, although the data are less precise than those obtained from the other techniques used here. The magnetization-transfer or spin-saturation-transfer method is a (27) Perkin, T. Ph.D. Thesis, California Institute of Technology, 1981. (28) Harris, R. K. Nuclear Magnetic Resonance Spectroscopy: A Physicochemical View; Pitman: London, 1983. (29) Swift, T. J.: Connick, R. E. J . Chem. Phys. 1962. 37, 307. (30) McConnell, H. M.; Berger, S. B. J . Chem. Phys. 1957, 27, 230.
[CpM(CO)311, temp,” M K M = Mo 0.0621 235 246 257 269 279 0.0333 235 246 257 269 279 0.0101 235 246 257 269 279 0.0070 235 246 257 269 279
AuIp(CpM(CO)3-),* Hz 12.4 22.4 50.4 89.4 122.4 8.6 12.1 23.7 41.5 61.0 2.1 1 3.27 5.68 11.4 18.9 1.33 2.15 4.62 8.46 13.8
M = W 0.0489
& kcx
Table I . Corrected Line Widths AullZ Obtained for PPN[CpM(CO),] in CD3CN in the Presence of CpM(CO),I as a Function of Temperature
0.0460
246 257 269 279 290 300 0.0488 0.0200 246 257 269 279 290 300 0.0485 0.0076 246 257 269 279 290 300 0.0492 0.0030 246 257 269 279 290 300 OTemperature calculated from calibration curve. rected by line width in the absence of exchange.
Table 11. Bimolecular Rate Constants for the CpM(CO),-/CpM(CO),I Self-Exchange M = Mo temp, K k,,, M-I s-I 235 0.64 x 103 246 1.2 x 10’ 257 2.6 X 10’ 269 4.6 x 103 279 6.2 x 103 290 300
3.67 7.84 15.6 25.8 45.0 71.7 1.66 3.46 6.46 11.4 19.5 31.6 0.79 1.51 2.63 4.89 8.02 12.4 0.21 0.5 1 0.97 1.76 2.30 3.38 Line width cor-
M = W
kex9M-’ s-I 0.25 x 103 0.53 x 103 1.1 x 103 1.7 x 103 3.1 x 103 4.9 x 103
double resonance N M R technique initially developed by Forsen and Hoffman3’ for measuring rates of exchange between magnetically inequivalent sites within a single molecule. Associated with each site in a molecule are a characteristic lifetime for each spin state and a time constant associated with the recovery of populations expected at thermal equilibrium. The time constant associated with the restoration of the resonance intensity is known (31) Forsen, S.: Hoffman, R. A. J . Chem. Phys. 1963, 11, 2892.
Schwarz et ai.
1228 J . Am. Chem. SOC.,Vol. 113, No. 4, 1991
n
Table 111. TI Values Determined for PPN[CpM(CO),] and CpM(CO)3BrComplcxcs in CD,CN
cam pou nd (PPN)[CPM~(CO),I
temp, K
TI, s 39 44 44 55 43 42 48 38 50 61 57 50 60 36 39 42 45
294 30 I 306 314 294 298 309 314 298 310 313 319 326 310 313 320 326
CpMo(CO),Br
(P~")[CPW(CO),I
Cp W(CO) Br
n
I1
The sequence of events occurring in a magnetization-transfer experiment involving a bimolecular exchange is summarized in Scheme 1. In the magnetization-transfer sequence, A is selectively
Scheme I A
-
-
*
'-
'1
~
6.2
1
~
6.0
,
5.8
~
275 290 295
-
h
- &
4
~
5.6 5.4 PPM
,
5.2
'
1
5.0
300 ~
1
'
1
'
1
'
A*
4.8
Figure 1. Stacked plot of N M R spectra from a linebroadening experiment with CpW(CO),I (0.046 M) ( I ) and CpW(CO),- (0 049 M) (11) in CD3CN at different temperatures. Small peaks are due to impurities.
as the longitudinal relaxation time, TI. When chemical exchange occurs, the nuclei are transferred from one (magnetically inequivalent) site to another. The exchange takes place without relaxation occurring in the transfer process. Consequently, when the exchange rate is comparable to the relaxation rates at both sites, complete saturation of one resonance results in partial saturation of the other. Saturation can be obtained by a "soft" pulse used to selectively negate or invert one of the exchanging resonances. In the DANTE sequence used in this study, saturation occurs as the excited resonance is nulled. When it is allowed to relax, the signal intensity a t the other site recovers. TI values determined in this study for CpM(CO)< and CpM(CO)3Br arc listed in Table 111. They are typically of the order of 40 s, and l / T l is of the order of 2.5 X IOy2 s-I. Thus if the self-cxchangc time scale can be brought into this range, either by choosing the appropriate concentration level or temperature, magnetization transfer should be observed. Figure 3 illustrates the behavior observed for 3 mM molybdenum complexes at 29 "C.
B*
pulse
A*
IITIA
IITIB
A
B
excited to A * (i). In the presence of B, exchange (ii) occurs with a bimolecular rate constant kex to yield partial saturation of B (B*). Similarly, since B is undergoing self-exchange with A, the B* population can be depleted through (iii). In parallel, both A* and B* decay to A and B by their respective time constants, T I , and TIB(iv and v). Scheme 1 yields eqs 7 and 8 for [A*] and [B*]. The solution to the time-dependent Bloch equations de-
scribing the magnetization at each site yields eq 9, which contains the two characteristic time constants XI and X, (eq IO). C2 is a constant of integration, and the X expression contains TI values for both A and B as well as k A and kB.32 The intensity data and (32) Doherty, N . M.; Bercaw, J . E. J . Am. Chem. SOC.1985, 107, 2670.
Self- Exchange Reactions of CpM(CO),X/CpM(CO)3-Halide Couples
J . Am. Chem. Soc., Vol. I 13, No. 4, 1991
1229
I
t
5.70 PPH
5.10
PPU
5.00
Figure 3. Stacked plot of magnetization-transfer results for c ~ M o ( C 0 ) ~ B(0.0025 r M) ( I ) and CpMo(CO)< (0.0036 M) (11) in CD3CN at 303 K. Each spectrum was obtained at a different value of T : the value of T at the bottom of the plot is 0.0002 s, and at the top it is 250 s. Table IV. Magnetization-Transfer Rate Constants for PPN[CpM(CO),]/CpM(CO),Brin CDICN 14
12
10
-
M = Mo
-
-
0.0036 0.0036 0.0034 0.0034 0.0036 0.0037 0.0038
0.0023 0.0026 0.0070 0.0018 0.0025 0.0019 0.0027
0.0032 0.0032 0.0032 0.0032 0.0032 0.0034 0.0032
0.0034 0.0067 0.0092 0.0034 0.0034 0.0033 0.0033
293 298 302 302 303 309 313
0.019 0.043 0.30 0.040 0.044 0.047 0.11
308 310 310 313 318 315 326
.015 ,017 .030 ,029 ,026 ,030 .044
8.2 1.7 4.4 2.2 1.7 2.5 4.4
X
IO'
x IO' x IO' X IO' X 10)
x 101
M = W
8-
4.4 2.5 3.2 8.9 7.5 8.8 1.3 X IO'
rime, sec
Figure 4. Fit of CpMo(C0)F intensity data from magnetization-transfer experiment with CpMo(CO)< and CpMo(CO),Br in CD3CN at 303 K.
other parameters were fit to eq 9 to yield values of the pseudofirst-order rate constant kA. AMz = A M ,
+
-I. ,
,
,
5.1
'
I
5.2
510
PPM
4:B
4:6
Figure 5. Stacked plot of exchange results for CpW(CO),CI (0.057 M) and (Cp-dS)W(CO),- (0.040 M) in CD,CN at 298 K. Plot shows the growth of CpW(C0);.
A typical data set and fit are shown in Figure 4. The bimolecular exchange rate constants were determined from the fitted pseudo-first-order rate constants corrected for the concentration
Schwarz et al.
1230 J . Am. Chem. Soc., Vol. 113, No. 4, 1991
I
1
-7
4 -11
-15
3
3.2 3.4 3.6
3.8
4.2
4
I\ 3
4.4
w-Cl
1 3.2
3.4 3.6
3.8
4.2
4
4.4
1O/T ) ~ XCpW(CO),-/CpW(CO),X, where X = I (m), Br Figure 6. Plots of In (k,,/7') vs 1/T for C ~ M O ( C O ) ~ ' / C ~ M O ( C Oand
(O), and
CI (A).
Table V. Rate Constants for Reaction of PPN[(Cp-d,)M(CO),]
Table VI. Activation Parameters and Self-Exchange Rate Constants
with CpM(CO),CI in CD3CN
at 298 K for CDM(CO),- or ( C D - ~ ~ M ( C O with ) , - CuM(CO),X"
[(Cp-d,)M(CO),-l, M 0.056 0.065 0.065 0.074 0.058 0.037 0.047 0.057 0.036 0.066
[CPM(CO)~CII, temp, M K M = Mo 0.056 258 0.065 268 0.067 268 0.072 279 0.058 284 M = W 0.035 287 0.058 298 0.039 298 0.032 308 0.048 318
6.7 3.0 2.8 1.2 2.4
S-I
x 10-5 6.0 x 10-4 X
2.3 X IO-'
x 10-4 2.1 x 10-3 x 10-4 8.3 x 10-3 X
5.1 X 7.1 X 3.3 x 10-4 3.1 x 10-3 1.3 X 1.4 X 2.6 x 10-4 3.9 x 10-3 2.1 X 1.9 X
(Cp-dS)M(CO)j- + C p M ( C 0 ) j X F= ( C P - ~ S ) M ( C ~+) ~CpM(CO),X (5) kobsd = kex([M-l + [MC1l) (1 1) The errors on the derived bimolecular rate constants are generally & I O % but larger for the X = Br systems studied by the magnetization-transfer method. For all three types of experiments activation parameters were determined from plots of In (k,,/T) vs 1 / T as shown in Figure 6, and values for k,, at 298 K, calculated from the least-squares fit of the data, are given in Table VI. As expected, the errors on the activation parameters for the X = Br systems are much greater than for the others. In a single exploratory experiment, the exchange of (Cp-d5)W(CO)3- with CPW(CO)~(CH,)was followed in CD3CN at 335 K. The half-life for eq 12 with [CpW(C0)3CH3]= 0.010 M and
(33) McKay, H. A. C. Nature 1938, 142, 997; J . Am. Chem. SOC.1943,
a
as*,
AH', kcal mol-'
1.5 X IO4
6.4 (f0.4)
cal K-I mol-' -17.8 (f1.5)
16
12.1 (f4.5)
-12 (*IS)
9.0 X
18.9 ( f l . O ) 0 ( f 4 )
4.5 x 103 1.5 2.1 X IO-'
7.5 (f0.1) -16.8 (f0.5) 15.1 ( f 5 . 2 ) -7 (f16) 17.7 ( f 3 . 3 ) -11 ( A l l )
2.1 X
of the reaction partner. I n most cases, C P M ( C O ) ~ Bwas ~ irradiated so that A = CpM(CO),Br and B = C P M ( C O ) ~ - .The range of temperatures over which these reactions could be studied was limited by a slow rate of exchange at lower temperatures and decomposition of the reactants to the more insoluble dimer species at higher temperatures. The results for X = Br are given in Table IV. Conventional Techniques. The rates of exchange were determined for the C P M ( C O ) ~ C Isystems by conventional N M R techniques. Solutions containing both CpM(CO)3CI and the perdeuterated complex (Cp-d,)M(CO)c were monitored with time until the reaction had reached equilibrium (eq 5 ) and k,, was The semilog plots of the intensity-time evaluated from eq 1 data were linear for the 4-5 half-lives monitored (12-18 time points). Figure 5 shows the N M R results for one of the slow exchange experiments, and all of the results are summarized in Table V.
65, 702.
k,,(298 K), M-I s-1
k,,, M-'
k , s-I
Values in parentheses are standard deviations.
[(Cp-d,)W(CO),-] = 0.012 M was approximately 1 month. The exchange rate constant was estimated to be 8 ( f l ) X IO4 M-' SKI.
+ C P W ( C O ) ~ C H+~ CpW(CO)3- + (Cp-dS)W(C0)3CH3
(Cp-dS)W(C0)3-
(12)
As a probe of the substitution kinetics of compounds in this series, the potential reaction of PMe3 with CPM(CO)~-in CD3CN was investigated by 'HNMR: at room temperature, no reaction to form phosphine-substituted complexes was observed over a 7 month period ([PMe3]/5 = [CpM(CO),-] > 3 mM). Discussion The CpM(C0)f and CpM(CO),X complexes used in this study have structures based on the three- and four-legged piano stool. In the course of the self-exchange the X group is transferred from one "leg" site to another, and only small changes in the intramolecular CpM(CO), distances and angles ensue: In the anion,34" the OC-M-CO angles and Cp-M-CO angles are typically 86 and 128', respectively; in CPM(CO)~X,both angles are somewhat smaller (for X = C1, 78 and 110-125', respectively).34b The Mo-CI and W-CI distances are 2.498 ( 1 ) and 2.490 (2) A.34b
e e I
M
0
c'"'I\co C 0
I
O C H M
l
l'\c 0"
x 0
Homolytic bond dissociation enthalpies for CpMo(CO),X are35 72.4 (CI), 60.5 (Br), 51.8 (I), and 47 (CH,) kcal mol-'. Met(34) (a) For the Mo complex: Adams, M.A,; Folting, K.; Huffman, J. C.; Caulton, K . G. Inorg. Chem. 1979, 18, 3020. (b) Bueno, C.; Churchill, M. R. Inorg. Chem. 1981, 20, 2197. (35) Nolan, S. P.; De La Vega, R. L.; Mukerjee, S.L.; Gonzalez, A. A,; Zhang, K.; Hoff, C. D.Polyhedron 1988, 7, 1491.
Sew- Exchange Reactions of CpM(CO)3X/CpM(CO)3-Halide Couples
J . Am. Chem. SOC.,Vol. 113, No. 4, 1991
1231
Table VII. Properties of CpM(CO)? Species M reaction Cr Mo W CPM(CO)~+ e- = CPM(CO)~EO, V vs sce' -0.38 -0.078 -0.022 CpM(CO)3f + 2e' = CpM(CO),Eo, V, vs sceb (-0.37) (-0.39) log K (pK, M-H)' 13.3 13.9 16.1 c~M(co)J- + H+ = C~M(CO);H CpM(CO),' + CH3I --c CpM(CO),CHj + Ik (M-I s-')" 7.5 x 10-2 1.5 2.4 CpM(C0)JH = CpM(CO)3 + H A H o , kcal mol-' 62 70 73 AH'. kcal mol-' 15.8' 32.5' 56.18 ICDM(CO),I? = ZCDM(CO), "Tilset, M.; Parker, V. D. J . Am. Chem. SOC.1989, I l l , 671 I ; 1990, 112, 2843. Estimated from the anodic peak for the anion and the cathodic peakfor the cation reported by Kadish, K. M.; Lacombe, D. A,; Anderson, J. E. Inorg. Chem. 1986, 25, 2246
