Normal and inverse primary kinetic deuterium isotope effects for C-H bond reductive elimination and oxidative addition reactions of molybdenocene and tungstenocene complexes: Evidence for benzene σ-complex intermediates

Article abstract of DOI:10.1021/ja027670k

The overall reductive elimination of RH from the ansa-molybdenocene and- tungstenocene complexes [Me₂Si(C₅Me₄)₂]Mo(Ph)H and [Me₂Si(C₅Me₄)₂]W(R)H (R = Me, Ph) is characterized by an inverse primary kinetic isotope effect (KIE) for the tungsten system but a normal KIE for the molybdenum system. Oxidative addition of PhH to {[Me₂Si(C₅Me₄)₂]M} also differs for the two systems, with the molybdenum system exhibiting a substantial intermolecular KIE, while no effect is observed for the tungsten system. These differences in KIEs indicate a significant difference in the reactivity of the hydrocarbon adducts [Me₂Si(C₅Me₄)₂]M(RH) for the molybdenum and tungsten systems. Specifically, oxidative cleavage of [Me₂Si-(C₅Me₄)₂]M(RH) is favored over RH dissociation for the tungsten system, whereas RH dissociation is favored for the molybdenum system. A kinetics analysis of the interconversion of [Me₂Si(C₅Me₄)₂]W(CH₃)D and [Me₂Si(C₅Me₄)₂]W (CH₂D)H, accompanied by elimination of methane, provides evidence that the reductive coupling step in this system is characterized by a normal KIE. This observation demonstrates that the inverse KIE for overall reductive elimination is a result of an inverse equilibrium isotope effect (EIE) and is not a result of an inverse KIE for a single step. A previous report of an inverse kinetic isotope effect of 0.76 for C-H reductive coupling in the [Tp]Pt(CH₃)H₂ system is shown to be erroneous. Finally, a computational study provides evidence that the reductive coupling of [Me₂Si(C₅Me₄)₂]W(Ph)H proceeds via the initial formation of a benzene σ-complex, rather than an η²-π-benzene complex.

Full text of DOI:10.1021/ja027670k

Published on Web 01/11/2003
Normal and Inverse Primary Kinetic Deuterium Isotope Effects
for C-H Bond Reductive Elimination and Oxidative Addition
Reactions of Molybdenocene and Tungstenocene
Complexes: Evidence for Benzene σ-Complex Intermediates
David G. Churchill, Kevin E. Janak, Joshua S. Wittenberg, and Gerard Parkin*
Contribution from the Department of Chemistry, Columbia UniVersity,
New York, New York 10027
Received July 12, 2002 ; E-mail: parkin@chem.columbia.edu
Abstract: The overall reductive elimination of RH from the ansa-molybdenocene and -tungstenocene
complexes [Me2Si(C5Me4)2]Mo(Ph)H and [Me2Si(C5Me4)2]W(R)H (R ) Me, Ph) is characterized by an inverse
primary kinetic isotope effect (KIE) for the tungsten system but a normal KIE for the molybdenum system.
Oxidative addition of PhH to {[Me2Si(C5Me4)2]M} also differs for the two systems, with the molybdenum
system exhibiting a substantial intermolecular KIE, while no effect is observed for the tungsten system.
These differences in KIEs indicate a significant difference in the reactivity of the hydrocarbon adducts
[Me2Si(C5Me4)2]M(RH) for the molybdenum and tungsten systems. Specifically, oxidative cleavage of [Me2Si-
(C5Me4)2]M(RH) is favored over RH dissociation for the tungsten system, whereas RH dissociation is favored
for the molybdenum system. A kinetics analysis of the interconversion of [Me2Si(C5Me4)2]W(CH3)D and
[Me2Si(C5Me4)2]W(CH2D)H, accompanied by elimination of methane, provides evidence that the reductive
coupling step in this system is characterized by a normal KIE. This observation demonstrates that the
inverse KIE for overall reductive elimination is a result of an inverse equilibrium isotope effect (EIE) and is
not a result of an inverse KIE for a single step. A previous report of an inverse kinetic isotope effect of 0.76
for C-H reductive coupling in the [Tp]Pt(CH3)H2 system is shown to be erroneous. Finally, a computational
study provides evidence that the reductive coupling of [Me2Si(C5Me4)2]W(Ph)H proceeds via the initial
formation of a benzene σ-complex, rather than an η2-π-benzene complex.
Introduction
comes from many sources. In addition to low-temperature
spectroscopic and room-temperature flash kinetics studies,3,5 the
The oxidative addition and reductive elimination of C-H
bonds at a transition metal center are reactions that are of
fundamental relevance to the functionalization of hydrocarbons.1
An important discovery for this class of reaction was that
hydrocarbon σ-complexes [M](σ-HR), i.e. complexes in which
the hydrocarbon is coordinated to the metal by three-center, two-
electron M‚‚‚H-C interactions,2 play an important role as key
intermediates.3,4 Evidence for the existence of these σ-complexes
principal evidence for σ-complex intermediates is derived from
the use of isotopic labeling experiments: specifically, (i) the
observation of deuterium exchange between hydride and alkyl
sites, e.g. [M](CH3)D f [M](CH2D)H, and (ii) the measurement
of kinetic isotope effects (KIEs).6 In this paper, we report a
series of kinetic and thermodynamic deuterium isotope effects
pertaining to the reactions of ansa-molybdenocene and -tung-
stenocene complexes of the type [Me2Si(C5Me4)2]M(R)H (M
) Mo, R ) Ph; M ) W, R ) Me, Ph). This study demonstrates
that the KIEs are highly dependent on the nature of the metal,
thereby indicating that the energy surface for reductive elimina-
tion of RH varies significantly for two otherwise similar systems.
(1) For recent reviews, see: (a) Crabtree, R. H. J. Chem. Soc., Dalton Trans.
2001, 2437-2450. (b) ActiVation and Functionalization of Alkanes; Hill,
C. L., Ed.; John Wiley and Sons: New York, 1989. (c) Arndtsen, B. A.;
Bergman, R. G.; Mobley, T. A.; Peterson, T. H. Acc. Chem. Res. 1995,
28, 154-162. (d) Ryabov, A. D. Chem. ReV. 1990, 90, 403-424. (e) Jones,
W. D.; Feher, F. J. Acc. Chem. Res. 1989, 22, 91-100. (f) Special issue:
J. Organomet. Chem. 1996, 504 (1-2). (g) Shilov, A. E.; Shul’pin, G. B.
Chem. ReV. 1997, 97, 2879-2932. (h) Labinger, J. A.; Bercaw, J. E. Nature
2002, 417, 507-514.
(2) Buchanan, J. M.; Stryker, J. M.; Bergman, R. G. J. Am. Chem. Soc. 1986,
108, 1537-1550, and ref 28 therein.
(3) (a) Hall, C.; Perutz, R. N. Chem. ReV. 1996, 96, 3125-3146. (b) Crabtree,
R. H. Chem. ReV. 1995, 95, 987-1007. (c) Crabtree, R. H. Angew. Chem.,
Int. Edit. Engl. 1993, 32, 789-805. (d) Metal Dihydrogen and σ-Bond
Complexes: Structure, Theory, and ReactiVity; Kubas, G. J., Ed.; Kluwer
Academic/Plenum Publishers: New York, 2001.
Results and Discussion
(1) Synthesis of [Me2Si(C5Me4)2]W(Me)H and [Me2Si-
(C5Me4)2]W(Ph)H. We are presently interested in defining how
a [Me2Si] ansa bridge modulates the reactivity of permethylated
metallocene complexes by comparing the chemistry of Cp*2-
(4) Although hydrocarbon σ-complexes were originally formulated for an
interaction of a transition metal with a single C-H bond, it has been noted
that other coordination modes are possible, involving the simultaneous
interaction with two or three C-H bonds. In this paper we use the term
[M](σ-HR) to refer generally to σ-complexes without specifying the exact
coordination mode, since this is often unknown.
(5) For a recent example of a σ-complex that has been characterized by NMR
spectroscopy, see: Geftakis, S.; Ball, G. E. J. Am. Chem. Soc. 1998, 120,
9953-9954.
(6) For recent reviews, see: (a) Jones, W. D. Acc. Chem. Res., in press. (b)
Bullock, R. M.; Bender, B. R. Isotope Methods in Homogeneous Catalysis.
In Encyclopedia of Catalysis; Horva´th, I. T., Ed.; Wiley: New York, 2003.
9
10.1021/ja027670k CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 1403-1420
1403

A R T I C L E S
Churchill et al.
Scheme 1
Scheme 2
MXn and [Me2Si(C5Me4)2]MXn systems.7 As part of the course
of these studies, we desired to investigate the influence of a
[Me2Si] ansa bridge on the reductive elimination of RH from
complexes of the type [Me2Si(C5Me4)2]M(R)H (M ) Mo, W)
and thereby obtain more insight into the mechanism of this
fundamental reaction. While the molybdenum phenyl hydride
complex [Me2Si(C5Me4)2]Mo(Ph)H has previously been syn-
thesized by photolysis of [Me2Si(C5Me4)2]MoH2 in benzene,7b
the tungsten complexes [Me2Si(C5Me4)2]W(Me)H and [Me2Si-
(C5Me4)2]W(Ph)H were previously unknown. Access to these
complexes is, nevertheless, provided by the sequence illustrated
in Scheme 1. The molecular structures of [Me2Si(C5Me4)2]WH2,
[Me2Si(C5Me4)2]WCl2, [Me2Si(C5Me4)2]WMe2, and [Me2Si-
(C5Me4)2]W(Ph)H, as determined by X-ray diffraction, are
illustrated in Figure 1.
(2) Reductive Elimination of Methane from [Me2Si-
(C5Me4)2]W(Me)H: Intra- and Intermolecular C-H Bond
Cleavage Reactions. Previous studies have demonstrated that
the methyl hydride complex Cp*2W(Me)H undergoes reductive
elimination of methane to give the tuck-in complex Cp*(η6-
C5Me4CH2)WH as a result of intramolecular trapping of the
tungstenocene intermediate [Cp*2W].8 The ansa-bridged com-
plex [Me2Si(C5Me4)2]W(Me)H likewise reductively eliminates
methane in cyclohexane to give [Me2Si(η5-C5Me4)(η6-C5Me3-
CH2)]WH (Scheme 2). In view of the fact that elimination of
methane from [Me2Si(C5Me4)2]W(Me)H yields the tuck-in
(7) See, for example: (a) Churchill, D. G.; Bridgewater, B. M.; Parkin, G. J.
Am. Chem. Soc. 2000, 122, 178-179. (b) Churchill, D.; Shin, J. H.; Hascall,
T.; Hahn, J. M.; Bridgewater, B. M.; Parkin, G. Organometallics 1999,
18, 2403-2406. (c) Lee, H.; Desrosiers, P. J.; Guzei, I.; Rheingold, A. L.;
Parkin, G. J. Am. Chem. Soc. 1998, 120, 3255-3256. (d) Shin, J. H.; Parkin,
G. Chem. Commun. 1999, 887-888. (e) Lee, H.; Bonanno, J. B.; Hascall,
T.; Cordaro, J.; Hahn, J. M.; Parkin, G. J. Chem. Soc., Dalton Trans. 1999,
1365-1368. (f) Zachmanoglou, C. E.; Docrat, A.; Bridgewater, B. M.;
Parkin, G.; Brandow, C. G.; Bercaw, J. E.; Jardine, C. N.; Lyall, M.; Green,
J. C.; Keister, J. B. J. Am. Chem. Soc. 2002, 124, 9525-9546.
(8) Parkin, G.; Bercaw, J. E. Organometallics 1989, 8, 1172-1179.
9
1404 J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003

Primary KIE for Mo and W Systems
A R T I C L E S
Figure 1. Molecular structures of [Me2Si(C5Me4)2]WH2, [Me2Si(C5Me4)2]WCl2, [Me2Si(C5Me4)2]WMe2, and [Me2Si(C5Me4)2]W(Ph)H.
Scheme 3
Table 1. Rate Constant and Activation Parameters for Reductive
Elimination of Methane from [Me2Si(C5Me4)2]W(Me)H in C6D6
T/°C
k
obs/s-1
T/°C
k
obs/s-1
100
121
3.74(5) × 10-6 a
140
159
1.35(3) × 10-4
4.93(7) × 10-5
7.95(16) × 10-4
∆Hq ) 27.3(2.6) kcal mol-1; ∆Sq ) 10(6) e.u.
a kCp* ) 1.13(6) × 10-4 s-1 for Cp*2W(Me)H (Parkin, G.; Bercaw, J.
E. Organometallics 1989, 8, 1172-1179).
(CD3)D gives [Me2Si(η5-C5Me4)(η6-C5Me3CH2)]WH. These
experiments support a reductive elimination mechanism for
[Me2Si(C5Me4)2]W(Me)H, in accord with that proposed for
Cp*2W(Me)H.8
Although the reductive elimination of methane from
Cp*2W(Me)H and [Me2Si(C5Me4)2]W(Me)H bear similarities,
there are several interesting differences between the two systems.
First, the ansa bridge substantially inhibits the reductive
elimination of methane, with kansa/kCp* ) 0.03 at 100 °C (Table
1).9 This inhibitory effect of the [Me2Si] bridge is noteworthy
in the sense that the same bridge actually promotes reductive
elimination in the related tantalum system.7d
complex [Me2Si(η5-C5Me4)(η6-C5Me3CH2)]WH, consideration
needs to be given to the possibility that the transformation
involves an alternative mechanism involving the direct abstrac-
tion of a ring methyl hydrogen. However, decisive evidence
that the reaction does indeed involve initial reductive elimination
is provided by deuterium labeling studies. Specifically, ther-
molysis of [Me2Si(C5Me4)2]W(CH3)D liberates CH3D and forms
[Me2Si(η5-C5Me4)(η6-C5Me3CH2)]WH, rather than giving
CH4 and [Me2Si(η5-C5Me4)(η6-C5Me3CH2)]WD (Scheme 3).
Likewise, elimination of methane from [Me2Si(C5Me4)2]W-
A second interesting difference between the two systems is
observed using benzene as solvent. Thus, whereas reductive
elimination of methane from Cp*2W(Me)H gives only the tuck-
(9) Green has reported an even more dramatic effect, with [Me2C(C5H4)2]W-
(Me)H being stable to elimination of methane under the conditions studied.
See: (a) Labella, L.; Chernega, A.; Green, M. L. H. J. Chem. Soc., Dalton
Trans. 1995, 395-402. (b) Chernega, A.; Cook, J.; Green, M. L. H.;
Labella, L.; Simpson, S. J.; Souter, J.; Stephens, A. H. H. J. Chem. Soc.,
Dalton Trans. 1997, 3225-3243.
9
J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003 1405

A R T I C L E S
Churchill et al.
Figure 2. Free energy surface relating the products of intramolecular and intermolecular benzene C-H bond activation by {[Me2Si(C5Me4)2]W} at 100 °C,
illustrating that intramolecular C-H bond activation is kinetically favored, but that the [Me2Si(η5-C5Me4)(η6-C5Me3CH2)]WH product is thermodynamically
unstable with respect to conversion to [Me2Si(C5Me4)2]W(Ph)H. ∆Gq for reductive elimination of C6D6 at 100 °C is estimated from the value at 182 °C,
assuming that ∆Sq is in the range +15 to -15 e.u. The value of 6.8 kcal mol-1 for ∆G was not measured directly but is obtained via a thermodynamic cycle.
Scheme 4
in complex Cp*(η6-C5Me4CH2)WH, the ansa-bridged tung-
stenocene intermediate {[Me2Si(C5Me4)2]W} is also trapped
by benzene to give the phenyl hydride complex [Me2Si(C5-
Me4)2]W(Ph)H, thereby providing another example of how a
[Me2Si] ansa bridge promotes intermolecular C-X (X ) H,
C, S) bond activation reactions.7a-c However, a kinetics analysis
of the relative amounts of [Me2Si(η5-C5Me4)(η6-C5Me3-
CH2)]WH and [Me2Si(C5Me4)2]W(Ph)H as a function of time
demonstrates that although intermolecular oxidative addition of
benzene is thermodynamically favored, intramolecular C-H
bond cleavage within {[Me2Si(C5Me4)2]W} to give [Me2Si(η5-
C5Me4)(η6-C5Me3CH2)]WH is actually kinetically favored, as
illustrated in Figure 2. Independent kinetics experiments that
(i) monitor the conversion of [Me2Si(η5-C5Me4)(η6-C5Me3-
CH2)]WH to [Me2Si(C5Me4)2]W(C6D5)D in C6D6 (Scheme 4)
and (ii) determine the barrier for reductive elimination of
benzene from [Me2Si(C5Me4)2]W(C6D5)D (described in more
detail below) allow the free energy surface shown in Figure 2
to be determined. The molybdenum complex [Me2Si(η5-C5-
Me4)(η6-C5Me3CH2)]MoH7b also reacts rapidly with benzene
at room temperature to give [Me2Si(C5Me4)2]Mo(Ph)H, thereby
indicating that intermolecular C-H bond activation of benzene
by the molybdenocene species {[Me2Si(C5Me4)2]Mo} is also
thermodynamically more favored than intramolecular C-H bond
activation.
(3) Kinetic Isotope Effects and Isotope Scrambling for
[Me2Si(C5Me4)2]W(CH3)H and Its Isotopologues. Before
discussing in more detail the reductive elimination of methane
from [Me2Si(C5Me4)2]W(Me)H, it is pertinent to discuss some
general aspects concerning the mechanisms of reductive elimi-
nation and oxidative addition of C-H bonds in view of the
existence of σ-complex intermediates. Specifically, it is impor-
tant to emphasize that the terms “reductive elimination” (re)
and “oxidative addition” (oa) do not correspond to elementary
steps. The terms reductive elimination and oxidative addition
were introduced at a point in time when σ-complex intermediates
were unknown, so additional terms are required to describe
adequately the overall mechanism.6a Thus, as illustrated in
Scheme 5 and Figure 3, reductive elimination consists of
reductive coupling (rc) followed by dissociation (d), while the
microscopic reverse, oxidative addition, consists of ligand
association (a) followed by oxidative cleavage (oc).
The rate constant for irreversible reductive elimination (i.e.
ka ) 0) is a composite of the rate constants for reductive
9
1406 J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003

Primary KIE for Mo and W Systems
A R T I C L E S
Figure 3. Free energy surface for reductive elimination of RH from a metal
center via a σ-complex intermediate.
Figure 4. Free energy surface for reductive elimination of RH from a metal
center via a σ-complex intermediate for which kd . koc.
Scheme 5
coupling (krc), oxidative cleavage (koc), and dissociation (kd),
namely, kobs ) krckd/(koc + kd). Correspondingly, the form of
the energy surface depends on the relative values of the
individual rate constants. There are two limiting situations: (i)
if kd . koc, the reductive coupling step becomes rate determining
(Figure 4), and (ii) if kd , koc, the dissociation of RH becomes
rate determining (Figure 5). The former situation with kd . koc
is characterized by kobs ) krc, whereas the latter situation with
kd , koc is characterized by kobs ) krckd/koc. Evidence to
distinguish which of the two limiting energy surfaces corre-
sponds better to the system under investigation may be provided
by measurement of the primary KIE. For example, if the
reductive coupling step is rate determining (Figure 4) and the
reductive coupling step exhibits a normal KIE (i.e. kH/kD > 1),10
Figure 5. Free energy surface for reductive elimination of RH from a metal
center via a σ-complex intermediate for which kd , koc.
one would predict a normal KIE for overall reductive elimina-
tion. On the other hand, if dissociation of RH is rate determining
(Figure 5), the KIE for reductive elimination is a composite,
i.e. kH/kD ) [krc(H)/krc(D)][kd(H)/kd(D)]/[koc(H)/koc(D)]. If the isotope
effect for dissociation of RH (i.e. [kd(H)/kd(D)]) is considered to
be close to unity,6a the isotope effect on reductive elimination
(10) Melander, L.; Saunders, W. H., Jr. Reaction Rates of Isotopic Molecules;
John Wiley and Sons: New York, 1980.
9
J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003 1407

A R T I C L E S
Churchill et al.
Figure 6. Origin of an inverse kinetic isotope effect for reductive elimination. The inverse kinetic isotope effect is essentially a consequence of an inverse
equilibrium isotope being transferred to the rate-determining step.
Table 2. Kinetic Isotope Effect for Elimination of Methane from
[Me2Si(C5Me4)2]W(CH3)H (kH) and [Me2Si(C5Me4)2]W(CD3)D (kD)
at 100 °C
is dominated by the ratio of the isotope effect on reductive
coupling to that of oxidative addition, i.e. [krc(H)/krc(D)]/[koc(H)
/
k
oc(D)]. This ratio is identical to the equilibrium isotope effect
solvent
kH/s-1
kD/s-1
kH/kD
(EIE) for interconversion of [M](R)H and [M](σ-RH), which
would be predicted to be inverse because deuterium prefers to
be located in the stronger bond, i.e C-D versus M-D.11 On
this basis, an inVerse KIE would be predicted for the overall
reductive elimination. In essence, the inverse EIE is transferred
to the rate-determining step, and the overall inverse KIE for
reductive elimination does not require an inverse effect for a
single step (Figure 6).
C6D6
C6D12
3.74(5) × 10-6
8.35(10) × 10-6
0.45(3)
0.47(3)
4.68(17) × 10-6
1.00(3) × 10-5
2). Notably, the KIE is more inverse than that for the Cp*2W-
(Me)H system (0.70).8,19 As described above, the observed
inverse KIE is consistent with interconversion between [Me2-
Si(C5Me4)2]W(Me)H and the σ-complex [Me2Si(C5Me4)2]W-
(σ-MeH) prior to rate-determining elimination of methane.20
Further evidence in support of a σ-complex intermediate is
provided by the observation of H/D exchange between the
hydride and methyl sites of the isotopologue [Me2Si(C5-
Me4)2]W(CH3)D. Specifically, [Me2Si(C5Me4)2]W(CH3)D is
observed to convert to [Me2Si(C5Me4)2]W(CH2D)H, prior to
the elimination of methane (Scheme 6). Examples of such
isotope exchange reactions are well-known.6 A particularly
striking example was reported by Green, who demonstrated that
interconversion of (i) [Me2C(C5H4)2]W(CH3)D and [Me2C-
(C5H4)2]W(CH2D)H and also (ii) [Me2C(C5H4)2]W(CD3)H and
[Me2C(C5H4)2]W(CD2H)D occurred with no loss of methane.9b
Keinan has likewise reported an interesting example in which
[Tp]Pt(CH3)H2 undergoes isotopic exchange with CD3OD to
give [Tp]Pt(CD3)D2 by a mechanism that was proposed to
involve initial H/D exchange of the hydride site with the solvent,
followed by exchange via a σ-complex intermediate.21,22 Keinan
claimed that this was the first example of an alkyl hydride
complex that underwent isotopic scrambling without concomi-
tant liberation of alkane;21 however, Keinan’s report is predated
The above discussion indicates that either a normal or an
inverse KIE for overall reductive elimination would be predicted
on the basis of whether the first or second step is rate
determining. Indeed, examples of both normal12 and inverse13
KIEs for reductive elimination of RH are known.6 As an
illustration, complexes that exhibit normal KIEs include (Ph3P)2-
Pt(Me)H (3.3),12a (Ph3P)2Pt(CH2CF3)H (2.2),12b and (Cy2PCH2-
CH2PCy2)Pt(CH2But)H (1.5),12c while examples of complexes
that exhibit inverse KIEs include Cp*Ir(PMe3)(C6H11)H (0.7),2
Cp*Rh(PMe3)(C2H5)H (0.5),13d Cp2W(Me)H (0.75),14 Cp*2W-
(Me)H (0.70),8 [Cp2Re(Me)H]+ (0.8),15 [(Me3tacn)Rh(PMe3)-
(Me)H]+ (0.74),16 (tmeda)Pt(Me)(H)(Cl) (0.29),17 [TpMe ]Pt-
2
18
(Me)2H (0.81), and [TpMe ]Pt(Me)(Ph)H (e0.78).
2
Reductive elimination of methane from [Me2Si(C5Me4)2]W-
(CH3)H and [Me2Si(C5Me4)2]W(CD3)D is also characterized by
a substantial inverse KIE of 0.45(3) in benzene at 100 °C (Table
(11) Wolfsberg, M. Acc. Chem. Res. 1972, 5, 225-233.
(12) (a) Abis, L.; Sen, A.; Halpern, J. J. Am. Chem. Soc. 1978, 100, 2915-
2916. (b) Michelin, R. A.; Faglia, S.; Uguagliati, P. Inorg. Chem. 1983,
22, 1831-1834. (c) Hackett, M.; Ibers, J. A.; Whitesides, G. M. J. Am.
Chem. Soc. 1988, 110, 1436-1448.
(13) For the first reports of inverse KIEs for elimination of RH (R ) H,a Ph,b
Cyc,d), see: (a) Howarth, O. W.; McAteer, C. H.; Moore, P.; Morris, G. E.
J. Chem. Soc., Dalton Trans. 1984, 1171-1180. (b) Jones, W. D.; Feher,
F. J. J. Am. Chem. Soc. 1985, 107, 620-630. (c) Stryker, J. M.; Bergman,
R. G. J. Am. Chem. Soc. 1986, 108, 1537-1550 (d) Periana, R. A.;
Bergman, R. G. J. Am. Chem. Soc. 1986, 108, 7332-7346.
(14) Bullock, R. M.; Headford, C. E. L.; Hennessy, K. M.; Kegley, S. E.; Norton,
J. R. J. Am. Chem. Soc. 1989, 111, 3897-3908.
(19) Furthermore, the KIE for [Me2Si(C5Me4)2]W(Me)H is also more inverse
than that for Cp2W(Me)H. See ref 14.
(20) Calculations on (CpR)2M(σ-MeH) derivatives indicate that the methane in
such complexes is bound asymmetrically with a single M-H-C interaction.
See: (a) Green, J. C.; Jardine, C. N. J. Chem. Soc., Dalton Trans. 1998,
1057-1061. (b) Green, J. C.; Harvey, J. N.; Poli, R. J. Chem. Soc., Dalton
Trans. 2002, 1861-1866. (c) Su, M.-D.; Chu, S.-Y. J. Phys. Chem. A 2001,
105, 3591-3597.
(15) Gould, G. L.; Heinekey, D. M. J. Am. Chem. Soc. 1989, 111, 5502-5504.
(16) Wang, C. M.; Ziller, J. W.; Flood, T. C. J. Am. Chem. Soc. 1995, 117,
1647-1648.
(21) (a) Lo, H. C.; Haskel, A.; Kapont, M.; Keinan, E. J. Am. Chem. Soc. 2002,
124, 3226-3228. (b) Iron, M. A.; Lo, H. C.; Martin, J. M. L.; Keinan, E.
J. Am. Chem. Soc. 2002, 124, 7041-7054.
(22) Bercaw has also reported that (tmeda)Pt(CH3)2 undergoes isotopic exchange
with CD3OD via a mechanism that was proposed to involve a σ-complex
intermediate. See ref 17.
(17) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1996, 118,
5961-5976.
(18) Jensen, M. P.; Wick, D. D.; Reinartz, S.; White, P. S.; Templeton, J. L.;
Goldberg, K. I. Submitted for publication.
9
1408 J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003

Primary KIE for Mo and W Systems
A R T I C L E S
Scheme 6
Table 3. Rate Constants for Isotope Exchange between [Me2Si(C5Me4)2]W(CH3)D and [Me2Si(C5Me4)2]W(CH2D)H and Elimination of
Methanea
krc(H)/krc(D) refined
krc(H)/krc(D) ) 0.45
krc(H)/krc(D) ) 1.00
krc(H)/krc(D) ) 2.00
krc(H)/krc(D) ) 5.00
krc(H)/krc(D) ) 10.00
k
k
k
k
rc(D)/10-5 s-1
rc(H)/10-5 s-1
rc(H)/krc(D)
3.87(10)
5.38(52)
1.4(2)
1.84(24)
0.52(3) koc*(D)
4.0(7)
5.60(28)
2.52(13)
0.45 (fixed)
0.49(4)
0.27(3) koc*(D)
3.3(4)
4.16(8)
4.16(8)
1 (fixed)
1.28(6)
0.46(3) koc*(D)
3.9(2)
3.66(6)
3.38(8)
16.9(4)
5 (fixed)
6.92(52)
0.64(6) koc*(D)
4.2(4)
3.29(8)
32.9(8)
10 (fixed)
13.9(1.2)
0.67(7) koc*(D)
4.2(4)
7.32(16)
2 (fixed)
2.72(13)
0.57(3) koc*(D)
4.1(2)
b
oc*(H)/koc*(D)
kd
Kc
a The different simulations indicate the sensitivity of the derived values as a function of the ratio krc(H)/krc(D) b koc*(H) and koc*(D) correspond to the composite
.
rate constants for oxidative coupling of C-H and C-D bonds within the equilibrating {[Me2Si(C5Me4)2]W(CH3D)} species; koc*(D) is arbitrarily fixed as
unity (1 s-1) for the purpose of the simulation. c K ) [[Me2Si(C5Me4)2]W(CH2D)H]/[[Me2Si(C5Me4)2]W(CH3)D] ) 3 [koc*(H)/koc*(D)]/ [krc(H)/krc(D)].
by Green’s study.9b Furthermore, it is worth emphasizing that
there are other examples of alkyl hydride complexes for which
the barrier for exchange between the hydride and alkyl sites is
considerably less than the barrier for alkane elimination.6 For
example, Heinekey has reported that site exchange within [Cp2-
Re(CD3)H]+ is much more rapid than reductive elimination, with
being modeled by a single species, {[Me2Si(C5Me4)2]W-
(CH3D)}, the oxidative cleavage rate constants are composites
for the rapidly equilibrating mixture of isotopomeric σ-com-
plexes and are thus referred to as koc*(H) and koc*(D) in order to
distinguish them from the values for a single species.30
The simulation generates the fundamental rate constants that
are listed in Table 3, and a graphical representation of the
simulation is presented in Figure 7. In view of correlation
between pairs of rate constants, it is important to evaluate the
uniqueness of the solution and the reliability of the derived
results. Since we are particularly interested in the KIE for
reductive coupling, i.e. krc(H)/krc(D), we performed simulations
to determine the sensitivity of the fit to this value. The results
of simulations in which the ratio krc(H)/krc(D) is fixed to
predetermined values are illustrated in Figure 8. It is evident
from these simulations that the fits are sensitive to the ratio
barriers that differ by ca. 6 kcal mol-1 23
Likewise, Flood has
.
reported that H/D exchange between hydride and hexyl ligands
in the rhodium complex {(1,4,7-tacn)Rh(hexyl)D[P(OMe)3]}+
is significantly more rapid than elimination of hexane.24
Furthermore, Girolami has reported an extremely interesting
example of a complex for which exchange between the hydride
and methyl sites is rapid on the NMR time scale, namely,
[Cp*Os(dmpm)(CH3)H+].25
A kinetics analysis of the isotopic exchange within [Me2Si-
(C5Me4)2]W(CH3)D, and the concomitant reductive elimination
of methane, was performed using the KINSIM/FITSIM simula-
tion program.26 The procedure used is analogous to that recently
(23) Gould, G. L.; Heinekey, D. M. J. Am. Chem. Soc. 1989, 111, 5502-5504.
(24) Flood, T. C.; Janak, K. E.; Iimura, M.; Zhen, H. J. Am. Chem. Soc. 2000,
122, 6783-6784.
(25) Gross, C. L.; Girolami, G. S. J. Am. Chem. Soc. 1998, 120, 6605-6606.
(26) (a) Barshop, B. A.; Wrenn, R. F.; Frieden, C. Anal. Biochem. 1983, 130,
134-145. (b) Zimmerle, C. T.; Frieden, C. Biochem. J. 1989, 258, 381-
387.
reported by Jones for the [TpMe ]Rh(L)(Me)X system (L )
2
CNCH2But; X ) H, D).27,28 Specifically, since the σ-complex
intermediates are not observed spectroscopically, it is not
possible to determine absolute rate constants for oxidative
cleavage or dissociation and only relative values may be derived;
thus, the value for koc*(D) was arbitrarily set as unity (1 s-1) for
the purpose of the simulation.29 Also following Jones’ procedure,
the simulation assumed rapid interconversion between the
various σ-complex intermediates, such that they were modeled
by a single species {[Me2Si(C5Me4)2]W(CH3D)} with a single
rate constant for the dissociation of methane (kd), as illustrated
in Scheme 6. Furthermore, since σ-complex intermediates are
(27) Northcutt, T. O.; Wick, D. D.; Vetter, A. J.; Jones, W. D. J. Am. Chem.
Soc. 2001, 123, 7257-7270.
(28) Furthermore, Bergman has performed similar simulations to analyze the
kinetics of diastereomer interconversion of Cp*Ir(PMe3)(2,2-dimethyl-
cyclopropyl)H via a σ-complex intermediate. See: Mobley, T. A.; Schade,
C.; Bergman, R. G. Organometallics 1998, 17, 3574-3587.
(29) It should be noted that a degree of correlation between the derived rate
constants still exists even when one of the rate constants is fixed.
(30) As such, the value of koc(H*)/koc(D*) [1.84(24)] derived from the simulation
does not correspond to the isotope effect for oxidative cleavage, koc(H)
oc(D), since it also includes the equilibrium constant for the interconversion
of the isotopomeric σ-complexes.
/
k
9
J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003 1409

A R T I C L E S
Churchill et al.
Figure 7. Kinetic simulation of isotope exchange within [Me2Si(C5Me4)2]W(CH3)D and reductive elimination of methane. Error bars correspond to 10%
uncertainty. Since the σ-complex [Me2Si(C5Me4)2]W(σ-CH3D) is not spectroscopically detectible, it is only possible to determine relative, and not absolute,
rate constants for reactions pertaining to this species. For this reason, koc*(D) was arbitrarily set as unity (1 s-1).
Figure 8. Kinetic simulations of isotope exchange within [Me2Si(C5Me4)2]W(CH3)D and reductive elimination of methane in which the ratio krc(H)/krc(D) is
fixed to values of 0.45, 1.00, 2.00, and 5.00. Error bars correspond to 10% uncertainty. Each of these simulated fits are noticeably worse than that in which
krc(H)/krc(D) is allowed to freely refine to a value of 1.4 (Figure 7), as illustrated by the poor fits for the combined products of reductive elimination (solid
squares). Importantly, the simulation in which the ratio krc(H)/krc(D) is fixed to an inverse value of 0.45 is particularly poor, indicating that the observed kinetic
isotope effect of 0.45 for elimination of methane from [Me2Si(C5Me4)2]W(CH3)H and [Me2Si(C5Me4)2]W(CD3)D is not the result of an inverse kinetic
isotope effect on the reductive coupling step and thereby supports the proposal that the inverse kinetic isotope effect for the overall reductive elimination of
methane is a consequence of the inverse equilibrium isotope for formation of σ-complex [Me2Si(C5Me4)2]W(σ-MeH) being transferred to the rate-determining
step.
krc(H)/krc(D), especially with respect to the combined products of
reductive elimination, i.e. [Me2Si(C5Me4)2]W(Ph)H and [Me2-
Si(C5Me4)(C5Me3CH2)]WH. In particular, the fit is poor when
an inverse KIE is imposed on krc(H)/krc(D), as illustrated by the
simulation in which krc(H)/krc(D) ) 0.45, so chosen because this
is the KIE observed for the overall reductive elimination of
methane from [Me2Si(C5Me4)2]W(CH3)H and [Me2Si(C5-
by a normal isotope effect, with a value of 1.4(2). Assuming
that secondary effects do not play a dominant role in the
reductive coupling of [Me2Si(C5Me4)2]W(CH2D)H, the value
of 1.4(2) provides an estimate of the primary KIE for reductive
coupling of [Me2Si(C5Me4)2]W(Me)X (X ) H, D) to form the
σ-complex intermediate [Me2Si(C5Me4)2]W(σ-XMe).
The rate constants corresponding to the best fit (Table 3)
permit construction of the free energy surface illustrated in
Figure 9, of which there are several notable features. Firstly,
the barrier for elimination of methane from the σ-complex is
ca. 0.5 kcal mol-1 greater than that for forming the σ-complex
Me4)2]W(CD3)D. Deviations are also observed when the krc(H)
/
krc(D) is fixed at values >1.4 (the best fit value), but not as badly
as when the value is forced to be inverse. Thus, the data
convincingly support the notion that krc(H)/krc(D) is characterized
9
1410 J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003

Primary KIE for Mo and W Systems
A R T I C L E S
Figure 9. Free energy surface for interconversion of [Me2Si(C5Me4)2]W(CH3)D and [Me2Si(C5Me4)2]W(CH2D)H and elimination of methane at 100 °C.
Note that for each pair of isotopomers, it is the one with deuterium attached to the carbon in a terminal fashion that is the lower in energy.
Figure 10. Enthalpic rationalization of the asymmetry of the free energy surface for isotope exchange between [Me2Si(C5Me4)2]W(CH3)D and [Me2Si-
(C5Me4)2]W(CH2D)H (Figure 9). Inclusion of the zero point energy levels on the symmetric electronic energy surface (i.e. the enthalpy at 0 K) results in
isotopomers with deuterium attached to carbon being lower in energy, and hence results in asymmetry on the enthalpy surface. Furthermore, a 3:1 statistical
factor also plays a role in further stabilizing [Me2Si(C5Me4)2]W(CH2D)H relative to [Me2Si(C5Me4)2]W(CH3)D on the free energy surface.
from [Me2Si(C5Me4)2]W(CH3)D. Secondly, the free energy
surface for the interconversion of [Me2Si(C5Me4)2]W(CH3)D
and [Me2Si(C5Me4)2]W(CH2D)H is asymmetric. Thus, the free
energy of [Me2Si(C5Me4)2]W(CH2D)H is lower than that of
[Me2Si(C5Me4)2]W(CH3)D due to (i) a zero point energy factor
that favors the isotopomer in which deuterium is located on
carbon rather than tungsten and (ii) a 3:1 statistical factor that
favors the deuterium being located on the methyl, rather than
hydride, site. Similar considerations also apply to the iso-
topomers of the σ-complex intermediates, namely, [Me2Si-
(C5Me4)2]W(σ-DCH3) and [Me2Si(C5Me4)2]W(σ-HCH2D), and
their corresponding transition states. The asymmetry of the free
energy surface is thus clearly rationalized by considering the
impact of zero point energy values and statistical factors on
the symmetric electronic energy surface (i.e. the enthalpy at
0 K) for isotopic exchange (Figure 10).
It is important to note that if the reaction involved only
interconversion of [Me2Si(C5Me4)2]W(CH3)D and [Me2Si(C5-
Me4)2]W(CH2D)H, without competitive loss of methane, the
derived value of krc(H)/krc(D) would be indeterminate, with
multiple solutions existing. For example, consider the standard
kinetics analysis for the approach-to-equilibrium conversion of
A to B.31 The analysis yields two independent parameters,
namely, the rate constant for the approach to equilibrium (kapp
)
and the equilibrium constant (K): kapp is the sum of the forward
and reverse rate constants, i.e. kapp ) kf + kr, while K is their
ratio, i.e. K ) kf/kr. Thus, if both kapp and K are known, the
individual values of kf and kr can be determined. As defined,
however, kf and kr are composite rate constants and so it is not
possible for the analysis to distinguish whether or not an
(31) Chemical Kinetics and Reaction Mechanisms, 2nd ed.; Espenson, J. H.,
Ed.; McGraw-Hill: New York, 1995.
9
J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003 1411

A R T I C L E S
Churchill et al.
Figure 11. Illustration that approach to equilibrium kinetics is incapable of providing evidence for an intermediate. Consequently, it is indeterminate
whether the first or second step is rate determining if an intermediate were to be present.
Figure 12. Illustration that the relative rates of isomerization and elimination of methane from a σ-complex intermediate are dependent on whether the first
or second step of the isomerization mechanism is rate determining, for given values of ∆G, ∆Gq , and ∆Gq
.
-MeH
f
Scheme 7
intermediate exists. Consequently, should an intermediate exist,
it is not possible to distinguish whether the first step or the
second step is rate determining. Thus, the three situations
illustrated in Figure 11 are kinetically indistinguishable on the
basis of the analysis of the approach-to-equilibrium intercon-
version.
For the present example involving the interconversion of
[Me2Si(C5Me4)2]W(CH3)D and [Me2Si(C5Me4)2]W(CH2D)H via
a σ-complex intermediate, K, kf, and kr are functions of koc*(H)
,
k
oc*(D), krc(H), and krc(D). Specifically, K ) 3[koc*(H)/koc*(D)]/[krc(H)
/
k
rc(D)] ) 3[KIEoc*]/[KIErc], from which it is evident that it is
not possible to determine the KIE for reductive coupling unless
both the equilibrium constant for the reaction and the KIE for
oxidative cleavage are known, i.e. [KIErc] ) 3[KIEoc*]/K.
It must be emphasized that the inability to determine the KIE
for reductive coupling in such a situation is not a result of the
exchange reaction proceeding to an equilibrium. For example,
even if the forward rate of isomerization of [Me2Si(C5Me4)2]W-
(CD3)H to [Me2Si(C5Me4)2]W(CD2H)D were to be compared
with the forward rate of isomerization of [Me2Si(C5Me4)2]W-
(CH3)D to [Me2Si(C5Me4)2]W(CH2D)H, the data would still not
allow for determination of the KIE for reductive coupling. The
steady-state analyses of these interconversions are presented in
Scheme 7, clearly demonstrating that the KIE for reductive
coupling (KIErc) cannot be determined uniquely from the
observed KIE measurement (KIEobs) but requires a knowledge
of the KIE for oxidative cleavage (KIEoc*). In the absence of a
value for KIEoc*, it is again not possible to determine a value
for KIErc.
reductive elimination of methane provides an additional kinetic
constraint. In essence, the relative rates of methane elimination
(kd) and the formation of [Me2Si(C5Me4)2]W(CH2D)H (koc*(H)
)
from the σ-complex intermediate allows the relative energy of
the transition state for oxidative-cleavage/reductive-coupling to
be ascertained, as illustrated for the general situation in Figure
12. Thus, whereas a kinetics investigation of the interconversion
[Me2Si(C5Me4)2]W(CH3)D and [Me2Si(C5Me4)2]W(CH2D)H
alone is incapable of determining whether the first or second
step is rate determining, once the relative energy of the second
transition state is known, it becomes possible to determine the
KIE for reductive coupling. The complete simulation therefore
becomes sensitive to the ratio of krc(H)/krc(D), whereas a simulation
for the interconversion of only [Me2Si(C5Me4)2]W(CH3)D and
[Me2Si(C5Me4)2]W(CH2D)H would be insensitive, because the
ratio koc*(H)/koc*(D) would merely adjust to maintain a constant
value for K.
It is, therefore, important to describe why the KIE for
reductive coupling in the [Me2Si(C5Me4)2]W(CH3)D/[Me2Si-
(C5Me4)2]W(CH2D)H system is no longer indeterminate if the
interconversion is competitive with loss of methane. The ability
for the simulation to extract rate constants for the reductive
coupling steps is made possible because the barrier for overall
9
1412 J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003

Primary KIE for Mo and W Systems
A R T I C L E S
The observation of a normal kinetic isotope effect for
reductive coupling within [Me2Si(C5Me4)2]W(Me)H is signifi-
cant because it supports the notion that the inverse nature of
the KIE for the reductive elimination of methane is not a
manifestation of an inverse KIE for a single step in the
transformation but is rather associated with an inverse equilib-
rium isotope effect. Of direct relevance to this issue, Jones has
recently demonstrated that the EIE for the interconversion of
Me2
[TpMe ]Rh(L)(Me)X and [Tp ]Rh(L)(σ-XMe) is inVerse be-
cause the KIE for oxidative cleavage is greater than that for
reductive coupling, with both values being normal.27
2
Although the notion that the reductive coupling of a methyl-
hydride complex is characterized by a normal primary kinetic
deuterium isotope effect is in line with the common understand-
ing of KIEs,6 it has recently been proposed that the reductive
coupling for [Tp]Pt(Me)H2 is characterized by an inVerse KIE.21
The experimental study involved comparing (i) the rate constant
for deuterium incorporation into the methyl group of [Tp]Pt-
(CH3)H2 when dissolved in CD3OD (kD) with (ii) the rate
constant for hydrogen incorporation into the methyl group of
[Tp]Pt(CD3)D2 when dissolved in CD3OH (kH). The mechanism
proposed involved rapid H/D exchange between the solvent and
hydride site, followed by a slower isomerization via a σ-complex
Figure 13. Illustration that the forward rate for exchange of a [M(CH3)D]
complex is faster than that for a [M(CD3)H] derivative on the basis of zero
point energy differences on the electronic energy surface (i.e. the enthalpy
at 0 K). For the [M(CH3)D] system the zero point energies are illustrated
with bold lines, while the [M(CD3)H] system the zero point energies are
illustrated with curly lines. For [M(CH3)D], it is the first transition state
(corresponding to reductive coupling) that is rate determining, whereas for
[M(CD3)H] it is the second transition state (corresponding to oxidative
cleavage) that becomes rate determining because hydrogen, rather than
deuterium, is now attached to carbon on a terminal site in the transition
state.
intermediate. The observed kinetic isotope effect, (kH/kD)obs
)
KIEobs ) 0.76, was proposed to correspond to that for reductive
coupling. However, as described above, it is not possible to
extract the KIE for reductive coupling from such a measurement
because the KIE for oxidative cleavage is unknown.32 Assigning
the observed KIE to that for reductive coupling is only possible
if KIEoc* is unity.
proposal that the KIE for reductive coupling had been deter-
mined to be inverse with a value of 0.76.21 In an effort to
determine what value of KIEoc* would cause KIErc to be e1,
the authors plotted KIErc versus KIEoc* for the function KIErc
The observed inverse KIE for [Tp]Pt(CH3)H2 can be readily
explained on the basis of normal KIEs for both reductive
coupling and oxidative cleavage steps, as illustrated in Figure
13. In essence, whereas the barrier for isotope exchange of [Tp]-
Pt(CH3)D2 is equal to the barrier for reductive coupling, the
barrier for isotope exchange of [Tp]Pt(CD3)H2 is greater than
the barrier for reductive coupling, since it is now the second
step that becomes rate determining, as illustrated in Figure 13.
The KIE for isotope exchange of [Tp]Pt(CD3)H2 is thus a
composite and is not that for reductive coupling.33 Thus, the
reported “kinetic isotope effect” for reductive coupling of
[Tp]Pt(CH3)X2 (X ) H, D) is not actually a KIE for a single
step, but more closely corresponds to the equilibrium constant
for the exchange reaction because it is merely the ratio of the
forward and reverse rate constants. An inverse equilibrium
constant would be expected for such a reaction since deuterium
prefers to be located in the stronger bond, i.e. C-D versus
Pt-D.11
) KIEobs[(KIEoc*) + 9]/[(KIEoc*)
-1 + 9], where KIEobs has the
experimental value of 0.76.34 On the basis of this plot, an inverse
value of KIErc would be predicted for a value of KIEoc* < 3.25.
However, the expression used, KIErc ) KIEobs[(KIEoc*) +
9]/[(KIEoc*)
-1 + 9], is incorrect. Specifically, the constant “9”
is erroneous and should be “3”, as illustrated in Scheme 7. The
correct value of 3 for the first exchange corresponds to a
statistical factor for cleaving the C-H versus C-D bond in a
rapidly equilibrating mixture of M(σ-CH3D) σ-complexes (and
correspondingly that for cleaving the C-D versus C-H bond
in a mixture of M(σ-CD3H) σ-complexes); the incorrect value
of 9 is a consequence of erroneously applying this statistical
factor twice.34,36 Using the correct statistical factor of 3 for the
first exchange, an inverse value of KIErc would be predicted
only for a value of KIEoc* < 1.72 which is substantially less
than the proposed value of 3.25 (Figure 14). Furthermore, the
originally reported value of 0.76 for KIErc is only obtained when
KIEox* ) 1. It is, therefore, evident that the experiment purported
to determine an inverse kinetic isotope effect of 0.76 for the
reductive coupling of [Tp]Pt(Me)X2 (X ) H, D) is erroneous32
and that the system does not provide the claimed unprecedented
opportunity to study the initial step of reductive coupling in
alkyl hydride compounds.34
Although the fact that KIEobs is a function of both KIErc and
KIEoc* was not considered in the original papers,21 an expression
that directly relates KIEobs to KIErc and KIEoc has subsequently
been presented, namely, KIEobs ) KIErc[(KIEoc*) )
-1 + 9]/[(KIEoc*
+ 9].34,35 This expression clearly invalidates the original
(4) Kinetic Isotope Effects for Reductive Elimination of
Benzene from [Me2Si(C5Me4)2]M(Ph)H (M ) Mo, W). In
(32) For further criticism of the proposed KIE for reductive coupling, see ref
6a.
(33) It should also be pointed out that, even if the reductive coupling for [Tp]-
Pt(CH3)H2 were to be characterized by an inverse kinetic isotope effect,
the problem would still exist because isotopic exchange of [Tp]Pt(CH3)D2
would then be a composite.
(36) It should also be noted that the analysis described in ref 34 (eq 17) considers
only the first cycle of exchange, i.e. Pt-CH3 to Pt-CH2D and Pt-CD3 to
Pt-CD2H, whereas the experimental study examined complete exchange.
While a more complex analysis is required to determine the exact
relationship between KIErc and KIEoc* for a given value of KIEobs, the value
of KIErc will, nevertheless, remain indeterminate unless KIEoc* is known.
(34) Lo, H. C.; Haskel, A.; Kapont, M.; Keinan, E. J. Am. Chem. Soc. 2002,
124, 12626.
(35) The symbols used here differ slightly from those of ref 34. Specifically,
kiered ) KIErc and kieox ) KIEoc*
.
9
J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003 1413

A R T I C L E S
Churchill et al.
Figure 14. Relationship between KIErc and KIEoc* for comparison of the
first exchange for M(CH3)D and M(CD3)H systems for a fixed value of
KIEobs ) 0.76. The two lines show the relationships for statistical factors
of x ) 3 and x ) 9, thereby demonstrating that an inverse KIErc for the
correct statistical factor of 3 is only observed if KIEoc* < 1.72. Furthemore,
a value of KIErc ) 0.76 is only observed when KIEoc* ) 1.
Scheme 8
Table 4. Rate Constants for Reductive Elimination of Benzene
from [Me2Si(C5Me4)2]W(C6H5)H and [Me2Si(C5Me4)2]W(C6D5)D in
C6X6 (X ) H, D) and C6X6/C6D12 at 182 °C
[C X ]/M
kH/s-1
kD/s-1
kH/kD
6
6
neata
0.22
6.49(14) × 10-7
1.15(3) × 10-6
0.56(3)
0.65(10)
5.35(27) × 10-7
8.17(84) × 10-7
a 11.1 M C6H6; 11.3 M C6D6.
Figure 15. Comparison of the free energy surfaces for reductive elimination
of benzene from [Me2Si(C5Me4)2]M(C6H5)H and [Me2Si(C5Me4)2]M-
(C6D5)D in C6D12 (M ) Mo at 80 °C, W at 182 °C). Energies in kcal
addition to studying the reductive elimination of methane, we
have also investigated the reductive elimination of benzene from
[Me2Si(C5Me4)2]W(Ph)H. Whereas the reductive elimination of
methane is irreversible, the reductive elimination of benzene is
reversible and may therefore be conveniently studied by
observing isotopic exchange with C6D6 solvent (Scheme 8). A
kinetics analysis indicates that the barrier for reductive elimina-
tion of benzene is substantially higher than that for methane;
for example, kMeH/kPhH ≈ 104 at 182 °C.37
mol-1
.
complex mediates the oxidative addition and reductive elimina-
tion of benzene from Cp*Rh(PMe3)(Ph)H.38,39 With respect to
the viability of a σ-complex intermediate, Bergman has postu-
lated that η2-π-benzene coordination is not part of the reaction
coordinate for the oxidative addition of benzene to [Cp*Ir-
(PMe3)] and has proposed that a benzene σ-complex is
responsible for benzene precoordination.40,41 Density functional
theory (DFT) calculations (described in more detail below)
indicate that reductive coupling of [Me2Si(C5Me4)2]W(Ph)H
The KIE for reductive elimination of benzene from [Me2Si-
(C5Me4)2]W(C6H5)H and [Me2Si(C5Me4)2]W(C6D5)D is inverse,
with kH/kD ) 0.65(10) at 182 °C (Table 4). By analogy with
the KIE for reductive elimination of methane [0.45(3) at 100
°C], the inverse isotope effect for reductive elimination of
benzene is also consistent with the existence of an intermediate
with an intact C-H bond prior to rate-determining loss of
benzene (Figure 15); further evidence for an intermediate,
discussed in more detail below, is the observation that the
intermolecular and intramolecular KIEs for oxidative addition
of benzene are different. Two possibilities exist for this
intermediate with an intact C-H bond, namely, (i) an η2-π-
complex in which the benzene coordinates by a CdC double
bond and (ii) a σ-complex in which the benzene coordinates
via a C-H bond. In support of the former possibility, Jones
and Perutz have provided considerable evidence that an η2-π-
(38) See, for example: (a) Jones, W. D.; Feher, F. J. J. Am. Chem. Soc. 1986,
108, 4814-4819. (b) Chin, R. M.; Dong, L. Z.; Duckett, S. B.; Partridge,
M. G.; Jones, W. D.; Perutz, R. N. J. Am. Chem. Soc. 1993, 115, 7685-
7695. (c) Belt, S. T.; Dong, L.; Duckett, S. B.; Jones, W. D.; Partridge, M.
G.; Perutz, R. N. J. Chem. Soc., Chem. Commun. 1991, 266-269. (d) Belt,
S. T.; Duckett, S. B.; Helliwell, M.; Perutz, R. N. J. Chem. Soc., Chem.
Commun. 1989, 928-930. (e) Chin, R. M.; Dong, L.; Duckett, S. B.; Jones,
W. D. Organometallics 1992, 11, 871-876.
(39) For other recent examples of η2-benzene complexes, see: (a) Reinartz, S.;
White, P. S.; Brookhart, M.; Templeton, J. L. J. Am. Chem. Soc. 2001,
123, 12724-12725. (b) Johansson, L.; Tilset, M.; Labinger, J. A.; Bercaw,
J. E. J. Am. Chem. Soc. 2000, 122, 10846-10855. (c) Meiere, S. H.; Brooks,
B. C.; Gunnoe, T. B.; Sabat, M.; Harman, W. D. Organometallics 2001,
20, 1038-1040.
(40) Peterson, T. H.; Golden, J. T.; Bergman, R. G. J. Am. Chem. Soc. 2001,
123, 455-462.
(41) Further support for a benzene σ-complex intermediate is provided by the
structural characterization of a derivative of a bidentate phosphine ligand
that exhibits an η2-H-Ar interaction, see: Vigalok, A.; Uzan, O.; Shimon,
L. J. W.; Ben-David, Y.; Martin, J. M. L.; Milstein, D. J. Am. Chem. Soc.
1998, 120, 12539-12544.
(37) The rate constant for [Me2Si(C5Me4)2]W(Me)H at 182 °C was determined
by extrapolation.
9
1414 J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003

Primary KIE for Mo and W Systems
A R T I C L E S
Figure 16. Explanation for a normal kinetic isotope effect for elimination of RH from molybdenum but an inverse kinetic isotope effect for tungsten. A
normal isotope effect is observed for molybdenum because the rate-determining step is reductive coupling. For clarity, no distinction is made between a
benzene σ-complex and an η2-π-benzene complex for the situation in which R ) Ph.
Table 5. Rate Constants Pertaining to the Reaction of
results directly in the formation of the benzene σ-complex [Me2-
[Me2Si(η5-C5Me4)(η6-C5Me3CH2)]WH with C6D6 at 100 °C
Si(C5Me4)2]W(σ-HPh) but that this species is thermodynamically
k
k
unstable with respect to isomerization to the η2-π-complex [Me2-
5.99(32) × 10-6 s-1 ktuck/kPhD 15.1(1.3) M
a
Si(C5Me4)2]W(η2-π-C6H6).
kuntuck
b
ktuck/{kPhD[C6D6]} 1.34(11)
kobsd
2.56(35) × 10-6 s-1
A most interesting difference in the reductive elimination of
benzene from [Me2Si(C5Me4)2]W(Ph)H and [Me2Si(C5Me4)2]-
Mo(Ph)H is that an inverse KIE is not observed for the
molybdenum system. Specifically, reductive elimination of
benzene from [Me2Si(C5Me4)2]Mo(Ph)H is characterized by a
normal KIE, with kH/kD ) 1.14(10). It is important to emphasize
that the observation of a normal kinetic isotope for reductive
elimination does not imply the nonexistence of either a
σ-complex or an η2-π-benzene intermediate for the molybdenum
system, but rather indicates that the reductive coupling step is
the rate-determining step of the reductive elimination reaction.
In contrast, the reductive coupling step is not rate determining
for the tungsten system, and a subsequent step (which could be
either isomerization to the η2-π-benzene complex or dissociation
of benzene) becomes rate determining (Figure 16).
a ktuck/kPhD is derived from the value ktuck/kPhD[C6D6] using 11.3 M as
b
the concentration of C6D6. kobs for the conversion of [Me2Si(η5-C5Me4)(η6-
C5Me3CH2)]WH to [Me2Si(C5Me4)2]W(C6D5)D is defined by kuntuckkPhD[C6D6]/
{ktuck + kPhD[C6D6]}, i.e. kuntuck/{1 + (ktuck/kPhD[C6D6])}.
Table 6. Rate Constantsa for the Reaction of
[Me2Si(η5-C5Me4)(η6-C5Me3CH2)]MoH with C6D6
T/°C
k
obs/s-1
T/°C
k
obs/s-1
5
15
4.21(4) × 10-5
25
35
6.18(16) × 10-4
1.86(4) × 10-4
2.01(9) × 10-3
∆Hq ) 21.2(5) kcal mol-1; ∆Sq ) 2(2) e.u.
a kobs ) kuntuckkPhD[C6D6]/{ktuck + kPhD[C6D6]}.
elimination of benzene from [Me2Si(C5Me4)2]Mo(Ph)H and
[Me2Si(C5Me4)2]W(Ph)H are different, we have probed the
nature of the energy surface in more detail by studying the
microscopic reverse. Specifically, KIEs for oxidative addition
of benzene to the metallocenes {[Me2Si(C5Me4)2]M} were
determined by competition experiments using the tuck-in
complexes [Me2Si(η5-C5Me4)(η6-C5Me3CH2)]MH (M ) Mo,
W) as sources of the metallocenes {[Me2Si(C5Me4)2]M}, as
illustrated in Scheme 4. In this regard, it is important to note
that the temperatures required to generate {[Me2Si(C5Me4)2]M}
from [Me2Si(C5Me4)(C5Me3CH2)]MH (e.g. 100 °C for M ) W)
are much lower than those for effecting reductive elimination
of benzene from [Me2Si(C5Me4)2]M(Ph)H (e.g. 182 °C for M
) W), so that competition experiments with C6H6/C6D6 mixtures
allow the intermolecular KIEs for oxidative addition to be
determined in the absence of thermodynamic equilibration
(Tables 5 and 6).
It is interesting to contrast the normal KIE for the reductive
coupling step of [Me2Si(C5Me4)2]Mo(Ph)H with the inverse KIE
that Jones has reported for Cp*Rh(PMe3)(Ph)H of 0.52.38a
A
possible rationalization of the difference is that the KIE for
reductive coupling of [Me2Si(C5Me4)2]Mo(Ph)H corresponds
more to that for formation of a σ-complex intermediate, whereas
the KIE for Cp*Rh(PMe3)(Ph)H corresponds to the formation
of an η2-π-benzene intermediate. If the formation of an η2-π-
benzene complex in the rhodium system were to be preceded
by the reversible formation of a σ-complex intermediate, the
observed isotope effect for reductive coupling would then be a
composite of the two steps. Alternatively, it is possible that the
reductive coupling of Cp*Rh(PMe3)(Ph)H to give the η2-π-
benzene intermediate occurs in a single step, such that the KIE
would not be expected to be the same as that for forming a
σ-complex intermediate.
An independent assessment of the KIEs for the oxidative
addition and reductive elimination of benzene is provided by
measurement of the equilibrium constants for the isotope
exchange reaction between [Me2Si(C5Me4)2]M(Ph)H and C6D6
(5) Kinetic Isotope Effects for Oxidative Addition of
Benzene to {[Me2Si(C5Me4)2]M} (M ) Mo, W). To substanti-
ate the proposal that the rate-determining steps for reductive
9
J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003 1415

A R T I C L E S
Churchill et al.
Figure 17. Different kinetic isotope effects for oxidative addition of benzene to {[Me2Si(C5Me4)2]Mo} and {[Me2Si(C5Me4)2]W}. A significant kinetic
isotope effect is observed for addition to {[Me2Si(C5Me4)2]Mo} because the rate-determining step involves cleavage of the C-H(D) bond, whereas no
isotope effect is observed for {[Me2Si(C5Me4)2]W} since the rate-determining step involves coordination of benzene. However, an intramolecular kinetic
isotope effect is observed for reaction of C6H3D3 with {[Me2Si(C5Me4)2]W} because it now probes the product-forming step, rather than the rate-determining
step. For molybdenum, the inter- and intramolecular kinetic isotope effects are similar because they probe the same transition state.
Table 7. Isotope Effectsa for [Me2Si(C5Me4)2]M(Ph)H (T ) 80 °C
for Mo and 182 °C for W)
Table 8. Kinetic Isotope Effects for Oxidative Addition of Benzene
by Competitive Trapping of {[Me2Si(C5Me4)2]M} (M ) Mo, W)
Mo
W
M
C X
T/°C
kH/kD
6
6
k
re(H)/kre(D)
1.14(10)
2.1(2)
2.6(2)
0.54
0.65(10)
1.0(1)
1.4(2)
0.65
Intermolecular
[koa(H)/koa(D)
[koa(H)/koa(D)
Kcalc ) [kre(H)/kre(D)]/[koa(H)/koa(D)
Kobs
]
]
Mo
Mo
Mo
W
1:1 C6H6/C6D6
1:1 C6H6/C6D6
1:1 C6H6/C6D6
1:1 C6H6/C6D6
0
3.1(1)
2.7(1)
2.1(2)a
1.0(1)
inter
25
80
100
intra
]
inter
0.53(2)
0.66(5)
Intramolecular
a Since the rate-determining steps are different for Mo and W, the
observed composite rate constants for reductive elimination [kre ) krckd/
(koc + kd)] and oxidative addition [koa ) kakoc/(kd + koc)] correspond to
different approximations, as illustrated in Figures 4 and 5. The only direct
Mo
Mo
Mo
W
1,3,5-C6H3D3
1,3,5-C6H3D3
1,3,5-C6H3D3
1,3,5-C6H3D3
0
3.2(1)
3.0(2)
2.6(2)a
1.4(2)
25
80
100
correspondences with single rate constants are as follows: kre(H)/kre(D)
rc(H)/krc(D) (for Mo) and [koa(H)/koa(D) ) ka(H)/ka(D) (for W).
)
k
]
inter
a Extrapolated.
(Scheme 8). Specifically, the equilibrium constant (K) for this
reaction corresponds directly with the EIE for reductive elimina-
tion of benzene from [Me2Si(C5Me4)2]M(C6H5)H and [Me2Si-
Although the intermolecular C6H6/C6D6 competition experi-
ment is incapable of probing the step involving oxidative
cleavage of the C-H bond for the tungsten system, such
information can be obtained by an intramolecular competition
experiment employing selectively deuterated benzene, i.e. 1,3,5-
C6H3D3. Intramolecular KIEs of this type have the potential of
specifically probing the C-H cleavage step. However, it must
be emphasized that if the cleavage step occurs after the rate-
determining step, the intramolecular KIE only corresponds to
the C-H cleavage step if the H/D benzene sites in the inter-
mediate are either equivalent or interchange rapidly. If the
H/D benzene sites in the intermediate are not capable of
being rendered equivalent, the intramolecular KIE would then
correspond to the intermolecular KIE. The observation that
the intramolecular KIE of 1.4(2) for addition of benzene to
{[Me2Si(C5Me4)2]W} is distinct from the intermolecular KIE
of 1.0(1) (Tables 7 and 8) indicates clearly that the isotope
effects are probing different steps and thereby indicates the
existence of an intermediate (Figure 17). The requirement that
H/D benzene sites are equivalent may be achieved by either (i)
the initial formation of a benzene η2-π-complex or (ii) facile
interconversion between the benzene σ-complex and the η2-π-
complex. For both situations, the intramolecular KIE corre-
(C5Me4)2]M(C6D5)D, as defined by KH/KD ) [kre(H)/kre(D)]/[koa(H)
/
k
]
oa(D) inter. Examination of Table 7 indicates that there is very
close correspondence between the two methods, thereby sub-
stantiating the derived KIEs.
Significantly, the oxidative addition of benzene to the
molybdenocene and tungstenocene fragments is characterized
by very different KIEs (Table 7). Thus, a substantial KIE is
observed for oxidative addition to the molybdenocene species,
whereas no KIE is observed for oxidative addition to the
tungstenocene analogue (Table 7). These KIEs are consistent
with the energy surfaces illustrated in Figures 15 and 17.
Specifically, a significant KIE is observed for addition to {[Me2-
Si(C5Me4)2]Mo} because the rate-determining step involves
cleavage of the C-H bond, whereas no isotope effect is
observed for {[Me2Si(C5Me4)2]W} since the rate-determining
step only involves coordination of benzene (Figure 17); by
comparison to the KIE for C-H cleavage, coordination that
does not involve significant weakening of the C-H bond (be it
via a benzene σ-complex or a η2-π-complex) would not be
expected to result in a significant KIE.
9
1416 J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003

Primary KIE for Mo and W Systems
A R T I C L E S
sponds specifically to the transformation from the species that
has equivalent H/D benzene sites, i.e. an η2-π-benzene complex.
Therefore, the intramolecular KIE of 1.4(2) for the addition of
1,3,5-C6H3D3 to {[Me2Si(C5Me4)2]W} reflects a composite of
the η2-π-benzene to σ-complex isomerization and oxidative
cleavage. Jones has also reported intermolecular (kH/kD ) 1.05)
and intramolecular (kH/kD ) 1.4) KIEs for oxidative addition
of benzene to the {Cp*Rh(PMe3)} fragment that are very similar
to those for {[Me2Si(C5Me4)2]W} and has likewise interpreted
the difference between intermolecular and intramolecular KIEs
as providing evidence for an η2-π-benzene intermediate.38a
An intramolecular KIE of 2.6(2) is also observed for oxidative
addition of 1,3,5-C6H3D3 in the molybdenum system, which also
corresponds to a transformation from the η2-π-benzene complex,
i.e. a composite of isomerization to the σ-complex and oxidative
cleavage. In contrast to the tungsten system, however, where
no intermolecular KIE is observed, a comparable intermolecular
KIE is observed for the molybdenum system because the two
KIEs probe the same C-H bond cleavage transition state (Table
7 and Figure 17).42 In this regard, Bergman has reported that
oxidative addition of benzene to the {Cp*Ir(PMe3)} fragment
also exhibits similar intramolecular (kH/kD ) 1.15) and inter-
molecular (kH/kD ) 1.22) KIEs.40
addition of the C-H bond of benzene to {[Me2Si(C5Me4)2]M},
illustrates a significant difference in the energy surface for the
molybdenum and tungsten systems. It is likely that this
difference will also be observed for other [Me2Si(C5Me4)2]M-
(R)H derivatives. The principal distinction is concerned with
the relative barriers for oxidative cleavage within the hydro-
carbon adduct [Me2Si(C5Me4)2]M(RH), be it a σ-complex or
η2-π-complex, Versus dissociation of RH. To a certain extent,
the relative barriers for oxidative cleavage within [Me2Si(C5-
Me4)2]M(RH) versus dissociation of RH reflect the difference
in energies of the [Me2Si(C5Me4)2]M(R)H and {[Me2Si(C5-
Me4)2]M} species, which is greater for tungsten due to the
stronger W-X versus Mo-X bonds.45,46 Stabilization of {[Me2-
Si(C5Me4)2]M} would be expected to reduce the barrier for
dissociation of RH from [Me2Si(C5Me4)2]M(RH), with the result
that the transition state for C-H oxidative cleavage becomes
the highest point on the energy surface (Figure 18, right-hand
side). Correspondingly, stabilization of [Me2Si(C5Me4)2]M(R)H
would be expected to lower the barrier for oxidative cleavage
within [Me2Si(C5Me4)2]M(RH), with the result that the transition
state for dissociation of RH becomes the highest point on the
energy surface (Figure 18, left-hand side). Since oxidative
addition to {[Me2Si(C5Me4)2]W} is more exothermic than that
to {[Me2Si(C5Me4)2]Mo}, the above provides a simple rational-
ization for a change in the rate-determining step and a difference
in the KIEs for two otherwise very similar systems.
The equilibrium constant for the interconversion of [Me2Si-
(C5Me4)2]Mo(C6H2D2)D and [Me2Si(C5Me4)2]Mo(C6H2D3)H
[1.96(8)]43 corresponds to the EIE for C-H versus C-D
cleavage of C6H3D3 by {[Me2Si(C5Me4)2]Mo}. Likewise, this
intramolecular EIE corresponds directly to the EIE for the
conversion of the η2-π-complex to the phenyl hydride.44
Correspondingly, the EIE for the reverse reaction, i.e. the
formation of [Me2Si(C5Me4)2]Mo(η2-π-C6H6) from [Me2Si(C5-
Me4)2]Mo(Ph)H, is 0.51(2). This value is in accord with the
inverse EIEs reported by Jones for the interconversion of (i)
Cp*Rh(PMe3)(Ph)H and Cp*Rh(PMe3)(η2-π-C6H6) [KH/KD )
(6) Computational Evidence That a Benzene σ-Complex
Precedes the Formation of an η2-π-Benzene Complex during
Reductive Elimination of Benzene from [Me2Si(C5Me4)2]W-
(Ph)H. To address the issue of whether it is a benzene
σ-complex or an η2-π-complex that is on the direct energy
surface for reductive coupling of [Me2Si(C5Me4)2]W(Ph)H, a
series of geometry optimization calculations that progressively
couple the C-H bond were performed. The result of these linear
transit calculations was the generation of a σ-complex inter-
mediate [Me2Si(C5Me4)2]W(σ-C6H6), as illustrated in Figures
19 and 20. However, although the calculations indicate that the
σ-complex [Me2Si(C5Me4)2]W(σ-C6H6) is the intermediate that
results directly from the reductive coupling step, this species is
calculated to be unstable with respect to the η2-π-benzene
complex [Me2Si(C5Me4)2]W(η2-π-C6H6), as illustrated in Figure
20. In view of this result, it is important to consider the
possibility that the η2-π-benzene adduct [Me2Si(C5Me4)2]W-
(η2-π-C6H6) and the phenyl hydride derivative [Me2Si(C5-
Me4)2]W(Ph)H may interconvert directly without the interme-
diacy of the σ-complex. To address the feasibility of a direct
pathway between [Me2Si(C5Me4)2]W(η2-π-C6H6) and [Me2Si-
(C5Me4)2]W(Ph)H, a series of geometry optimization calcula-
tions that result in the direct oxidative cleavage of the C-H
bond of [Me2Si(C5Me4)2]W(η2-π-C6H6) were performed by
progressively bringing one of the benzene hydrogen atoms
attached to the carbon atoms involved in the π-interaction
towards the tungsten center. These calculations clearly demon-
strated that the transition state for the direct transformation of
[Me2Si(C5Me4)2]W(η2-π-C6H6) to [Me2Si(C5Me4)2]W(Ph)H is
prohibitively high. In contrast, a much lower energy pathway
Me2
0.37]38a and (ii) [TpMe ]Rh(L)(Me)H and [Tp ]Rh(L)(σ-HMe)
2
[KH/KD ) 0.5].27
The combined study of the kinetic and equilibrium isotope
effects pertaining to the reductive elimination of benzene from
[Me2Si(C5Me4)2]M(Ph)H, and the microscopic reverse, oxidative
(42) In this regard, it is worth noting that even though the intramolecular and
intermolecular KIEs probe the same transition state, they are not expected
to be identical because the intramolecular KIE specifically probes only the
oxidative cleavage step (because that is where the partitioning occurs),
whereas the intermolecular isotope effect will also be influenced to a certain
degree by the isotope effects on the rate constants for association and
dissociation of benzene, since koa ) kocka/(koc + kd). Furthermore, inter-
and intramolecular kinetic isotope effects are not necessarily expected to
be equal because factors other than zero point energy differences influence
kinetic isotope effects. Specifically, the kinetic isotope effect is convention-
ally described in terms of a product of four factors, namely, KIE ) SYM‚
MMI‚EXC‚ZPE (where SYM is the symmetry factor, MMI is the mass
moment of inertia term, EXC is the excitation term, and ZPE is the zero
point energy term), and although the SYM factor is the same for the inter-
and intramolecular comparison, the other terms are not the same. In
particular, the MMI term associated with the different masses and moments
of inertia of C6H6 and C6D6 is 1.5, whereas this value is unity for the
intramolecular C6H3D3 comparison because only one type of benzene
molecule is involved. The MMI factor is, however, mitigated by the EXE
term, which approaches (1/MMI)(νqH/νqD) at high temperature. In view of
these additional and competing factors, intra- and intermolecular kinetic
isotope effects for the same fundamental reaction are not expected to be
identical. See, for example: Slaughter, L. M.; Wolczanski, P. T.; Klinck-
man, T. R.; Cundari, T. R. J. Am. Chem. Soc. 2000, 122, 7953-7975.
(43) K ) [[Me2Si(C5Me4)2]Mo(C6H2D3)H]/ [[Me2Si(C5Me4)2]Mo(C6H2D2)D].
(44) The overall equilibrium constant for oxidative addition of benzene is a
composite of binding and activation steps. However, since only one type
of η2-π-benzene complex exists in the intramolecular competition experi-
ment, the intramolecular EIE corresponds directly to the activation steps,
i.e. the interconversion of [Me2Si(C5Me4)2]Mo(η2-π-C6H6) and [Me2Si-
(C5Me4)2]Mo(Ph)H.
(45) Hascall, T.; Rabinovich, D.; Murphy, V. J.; Beachy, M. D.; Friesner, R.
A.; Parkin, G. J. Am. Chem. Soc. 1999, 121, 11402-11417.
(46) For example, the W-H bond in Cp2WH2 (74.3 kcal mol-1) is 12.9 kcal
mol-1 stronger than the Mo-H bond in Cp2MoH2 (61.4 kcal mol-1). See:
Dias, A. R.; Martinho Simo˜es, J. A. Polyhedron 1988, 7, 1531-1544.
9
J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003 1417

A R T I C L E S
Churchill et al.
Figure 18. Rationalization of the different energy surfaces for Mo and W. The relative barriers for dissociation of R-H versus oxidative cleavage of a
σ-complex are proposed to be influenced by the exothermicity of the overall transformation, with oxidative addition to tungsten being more exothermic than
that to molybdenum.
Figure 19. Geometry optimized structures of [Me2Si(C5Me4)2]W(Ph)H, [Me2Si(C5Me4)2]W(σ-HPh), and [Me2Si(C5Me4)2]W(η2-π-C6H6) and transition
states for their interconversions (hydrogen atoms on methyl groups are omitted for clarity).
is calculated for a two-step sequence in which the η2-π-benzene
complex [Me2Si(C5Me4)2]W(η2-π-C6H6) first rearranges to the
σ-complex [Me2Si(C5Me4)2]W(σ-C6H6) and then undergoes
subsequent oxidative cleavage of the C-H bond. The calcula-
tions, therefore, indicate that an η2-π-benzene complex is not
on the direct energy surface for reductive coupling and oxidative
cleavage of benzene in this system.
Although the η2-π-benzene intermediate is not on the direct
energy surface for the oxidative cleavage and reductive coupling
of the C-H bond, such a species may play an important role in
the oVerall reductive elimination and oxidative addition process,
as described by Jones and Perutz.38 Specifically, since the η2-
π-benzene complex is thermodynamically more stable than the
σ-complex, it is possible that η2-π-benzene coordination may
provide a kinetically more favored interaction than that involving
a σ-interaction. For such a situation, the overall oxidative
addition of benzene to the metal center would proceed via a
sequence involving initial η2-π-benzene coordination, followed
by isomerization to a σ-complex, and subsequent C-H oxidative
cleavage. Correspondingly, microscopic reversibility dictates that
reductive elimination would involve reductive coupling to give
a σ-complex, isomerization to an η2-π-benzene complex,
followed by dissociation.
The notion that a benzene σ-complex may play an important
role in oxidative addition and reductive elimination of aromatic
C-H bonds has been previously considered.39b,40,47 In particular,
9
1418 J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003

Primary KIE for Mo and W Systems
A R T I C L E S
Figure 20. Calculated (B3LYP) energy surface for conversion of [Me2Si(C5Me4)2]W(Ph)H to [Me2Si(C5Me4)2]W(η2-π-C6H6) via [Me2Si(C5Me4)2]W(σ-
HPh). Energies are in kcal mol-1 and correspond to the enthalpies at 0 K, and are uncorrected for zero point energy.
Bercaw has stated that it is difficult to envision how an η2-π-
benzene complex could undergo direct C-H oxidative addition
or conversely how the coupling of the C-H bond within a
phenyl hydride complex would lead directly to an η2-π-benzene
species.39b For the present system, it is clear that steric
interactions between the benzene ligand and the [Me2Si(C5-
Me4)2] ligand severely inhibit the direct oxidative cleavage of
the C-H bond within the η2-π-benzene complex [Me2Si(C5-
Me4)2]W(η2-π-C6H6). Indeed, the structure of [Me2Si(C5-
Me4)2]W(η2-π-C6H6) is strained by comparison to other η2-π-
benzene complexes because the cyclopentadienyl rings prevent
the benzene ligand from coordinating in a perpendicular fashion
to the tungsten center. For example, the W-(η2-π-C6H6) hinge
angle in [Me2Si(C5Me4)2]W(η2-π-C6H6) (127.1°) is substantially
greater than the corresponding experimental values for {[η2-
Oxidative addition of PhH to {[Me2Si(C5Me4)2]M} also differs
for the two systems, with the molybdenum system exhibiting a
substantial intermolecular KIE, while no effect is observed for
the tungsten systems. These differences in KIEs for the molyb-
denum and tungsten systems indicate a significant difference
in the reactivity of the hydrocarbon adducts [Me2Si(C5Me4)2]-
M(RH). Specifically, oxidative cleavage of [Me2Si(C5Me4)2]-
M(RH) is favored over RH dissociation for the tungsten
system, whereas RH dissociation is favored for the molybdenum
system. The origin of the difference between the molybdenum
and tungsten systems is rationalized in terms of the differ-
ence in energies of the [Me2Si(C5Me4)2]M(R)H and {[Me2Si-
(C5Me4)2]M} species, which is greater for tungsten due to the
stronger W-X versus Mo-X bonds. Stabilization of {[Me2Si-
(C5Me4)2]M} would be expected to reduce the barrier for
dissociation of RH from [Me2Si(C5Me4)2]M(RH), with the result
that the transition state for C-H oxidative cleavage becomes
the highest point on the energy surface.
TpMe ]Pt(η -π-C6H6)H}{B[C6H3(CF3)2]4} (106.2°)39a and TpRe-
(CO)(MeIm)(η2-π-C6H6) (111.8°).39c Oxidative cleavage of the
C-H bond in [Me2Si(C5Me4)2]W(η2-π-C6H6) naturally requires
the tungsten center to approach the C-H bond which is
prohibited by the orientation of the η2-π-benzene ligand. In
contrast, the orientation of the benzene in the σ-complex [Me2-
Si(C5Me4)2]W(σ-C6H6) is much more suited to the cleavage of
the C-H bond since the W-H interaction already exists, and
a smooth transition to the phenyl hydride complex [Me2Si(C5-
Me4)2]W(Ph)H is made possible.
2
2
In addition to the KIEs, evidence for the methane σ-complex
intermediate [Me2Si(C5Me4)2]W(σ-HMe) is provided by the
observation of H/D exchange between the hydride and alkyl
sites of [Me2Si(C5Me4)2]W(CH3)D prior to reductive elimination
of methane. Furthermore, a kinetics analysis of the intercon-
version of [Me2Si(C5Me4)2]W(CH3)D and [Me2Si(C5Me4)2]W-
(CH2D)H, with competitive elimination of methane, provides
evidence that the reductive coupling step in this system is
characterized by a normal KIE. This observation demonstrates
that the inverse KIE for overall reductive elimination is the result
of an inverse equilibrium isotope effect between [Me2Si(C5-
Me4)2]W(Me)X and [Me2Si(C5Me4)2]W(σ-XMe) (X ) H, D)
and is not the result of an inverse KIE for a single step. In this
regard, the previous report of an inverse kinetic isotope effect
Summary
In summary, the reductive elimination of RH from [Me2Si-
(C5Me4)2]M(R)H is characterized by an inverse KIE for the
tungsten system, but a normal KIE for the molybdenum system.
(47) Lavin, M.; Holt, E. M.; Crabtree, R. H. Organometallics 1989, 8, 99-
104.
9
J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003 1419

A R T I C L E S
Churchill et al.
refined by a quadratic synchronous transit (QST) calculation. The role
of the η2-π-benzene complex [Me2Si(C5Me4)2]W(η2-π-C6H6) on the
energy surface was evaluated by two sets of linear transit scans: (i) to
address the feasibility of the interconversion of the benzene η2-π-
complex and σ-complex, a series of geometry optimization calculations
that progressively increase the W‚‚‚C distance for one of the carbon
atoms were performed, thereby resulting in the transformation to the
σ-complex intermediate [Me2Si(C5Me4)2]W(σ-HPh), as illustrated in
Figure 20; (ii) to address the feasibility of a direct pathway between
[Me2Si(C5Me4)2]W(η2-π-C6H6) and [Me2Si(C5Me4)2]W(Ph)H, a series
of geometry optimization calculations that progressively decrease the
W‚‚‚H distance were performed, thereby resulting in the transformation
to the phenyl hydride, as illustrated in Figure 20. These calculations
clearly demonstrate that the pathway for the direct transformation of
Me2Si(C5Me4)2]W(η2-π-C6H6) to [Me2Si(C5Me4)2]W(Ph)H is prohibi-
tively high in energy compared to the two-step sequence in which the
η2-π-benzene complex [Me2Si(C5Me4)2]W(η2-π-C6H6) first isomerizes
to the σ-complex [Me2Si(C5Me4)2]W(σ-C6H6).
of 0.76 for C-H reductive coupling in the [Tp]Pt(CH3)H2
system has been shown to be erroneous.
Finally, a computational study provides evidence that the
reductive coupling of [Me2Si(C5Me4)2]W(Ph)H proceeds via the
initial formation of a benzene σ-complex, rather than an η2-π-
benzene complex, even though the latter species is more stable.
Experimental Section
The syntheses and characterization of all the compounds used in
this study and the techniques used to measure kinetic and equilibrium
isotope effects are described in the Supporting Information.
Computational Details
All calculations were carried out using DFT as implemented in the
Jaguar 4.1 suite of ab initio quantum chemistry programs. Geometry
optimizations were performed with the B3LYP functional and the
6-31G** basis set for C, H, and Si, while the transition metals were
represented using the Los Alamos LACVP** basis set that includes
relativistic effective core potentials. The energies of the optimized
structures were reevaluated by additional single point calculations on
each optimized geometry using the cc-pVTZ(-f) basis set that includes
a double set of polarization functions. A modified version of LACVP**,
designated as LACV3P**, was used for silicon and the transition metals.
The nature of the intermediate for the reductive coupling of the
phenyl hydride complexes [Me2Si(C5Me4)2]W(Ph)H was determined
by performing a series of geometry optimizations as a function of the
C‚‚‚H separation. The result of these calculations was the generation
of the σ-complex intermediate [Me2Si(C5Me4)2]W(σ-HPh) that could
be freely geometry optimized. The geometry of the transition state was
Acknowledgment. We thank the U.S. Department of Energy,
Office of Basic Energy Sciences (Grant DE-FG02-93ER14339)
for support of this research and Professor W. D. Jones, Professor
R. G. Bergman, Dr. R. M. Bullock, and the reviewers for helpful
comments.
Supporting Information Available: Experimental details
(PDF) and crystallographic CIF files. This material is available
free of charge via the Internet at http://pubs.acs.org.
JA027670K
9
1420 J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003