Kinetics of hydrogen atom transfer from (η5-C 5H5)Cr(CO)3H to various olefins: Influence of olefin structure

Article abstract of DOI:10.1021/ja066325i

Treating (η⁵-C₅H₅)Cr(CO)₃H (1) or (η⁵-C₅H₅)Cr(CO)₃D (1-d₁) with an excess of olefin containing the opposite isotope generally leads to H/D exchange, althoug

Full text of DOI:10.1021/ja066325i

Published on Web 12/15/2006
Kinetics of Hydrogen Atom Transfer from (η5-C5H5)Cr(CO)3H
to Various Olefins: Influence of Olefin Structure
Jongwook Choi, Lihao Tang, and Jack R. Norton*
Contribution from the Department of Chemistry, Columbia UniVersity, 3000 Broadway,
New York, New York 10027
Received August 31, 2006; E-mail: jrn11@columbia.edu
Abstract: Treating (η5-C5H5)Cr(CO)3H (1) or (η5-C5H5)Cr(CO)3D (1-d1) with an excess of olefin containing
the opposite isotope generally leads to H/D exchange, although hydrogenation is also observed in some
cases. Application of an appropriate statistical correction to the observed exchange rate gives kH and kD,
the rate constants for H• (D•) transfer from (η5-C5H5)Cr(CO)3H (D) to various olefins. The values of kH and
kD vary appreciably with the substituents on the double bond. Phenyl-substituted olefins accept H• more
readily than do carbomethoxy-substituted olefins, although the latter accept H• more readily than do alkyl-
substituted olefins. A methyl substituent on the incipient radical site increases kH at 323 K by a factor
between 5 and 50. A methyl substituent on the carbon to which the H• is being transferred decreases kH
substantially. On the whole, the rate constants for H• transfer reflect steric effects as well as the stability of
the resulting carbon-centered radicals.
Scheme 1
Introduction
Hydrogen atom (H•) transfers from transition-metal hydrides
to organic molecules play an important role in many reactions.1,2
For example, H• transfer from transition-metal hydrides is
involved in some hydrogenations. In 1977, Sweany and Halpern3
reported strong evidence for H• transfer in the hydrogenation
of R-methylstyrene by HMn(CO)5 (Scheme 1); they observed
chemically induced dynamic nuclear polarization (CIDNP) and
an inverse isotope effect for HMn(CO)5/DMn(CO)5. Since then,
there have been numerous reports of hydrogenations that occur
through H• transfer.2,4
In such a hydrogenation the first H• transfer, from the
transition-metal hydride to an unsaturated carbon-carbon bond,
generates a metalloradical and a carbon-centered radical. A
second H• transfer, from a second equivalent of the hydride to
the intermediate carbon-centered radical, produces the hydro-
genated product.
An H• transfer from a transition-metal hydride to a carbon-
carbon double bond also occurs in catalysis of chain transfer
by metalloradicals during free radical polymerizations.5,6 In eq
1, H• abstraction from a propagating radical by the chain-transfer
catalyst M• generates a vinyl-terminated polymer and a metal
hydride complex. In eq 2, H• transfer from the hydride complex
to monomeric olefin reinitiates the chain. While direct observa-
tion of eq 2 with Co(II) and Mo(III) chain-transfer catalysts
has been hampered by the instability of the corresponding
hydride complexes, we have observed eq 2 directly, initiating
the polymerization of MMA with (η5-C5Ph5)Cr(CO)3H.6
Hydroformylation reactions may also involve H• transfer.
Evidence for such a step in the hydroformylation of styrene by
HCo(CO)4 is offered by an inverse H/D kinetic isotope effect
(k(HCo(CO)4)/k(DCo(CO)4)) and by the absence of an inverse
CO dependence in the rate law.1,4c
(1) Halpern, J. Pure Appl. Chem. 1986, 58, 575-584.
(2) Eisenberg, D. C.; Norton, J. R. Isr. J. Chem. 1991, 31, 55-66.
(3) Sweany, R. L.; Halpern, J. J. Am. Chem. Soc. 1977, 99, 8335-8337.
(4) (a) Roth, J. A.; Orchin, M. J. Organomet. Chem. 1979, 182, 299-311. (b)
Sweany, R. L.; Comberrel, D. S.; Dombourian, M. F.; Peters, N. A. J.
Organomet. Chem. 1981, 216, 57-63. (c) Ungva´ry, F.; Marko´, L.
Organometallics 1982, 1, 1120-1125. (d) Roth, J. A.; Wiseman, P.;
Ruszala, L. J. Organomet. Chem. 1983, 240, 271-275. (e) Connolly, J.
W. Organometallics 1984, 3, 1333-1337. (f) Garst, J. F.; Bockman, T.
M.; Batlaw, R. J. Am. Chem. Soc. 1986, 108, 1689-1691. (g) Thomas, M.
J.; Shackleton, T. A.; Wright, S. C.; Gillis, D. J.; Colpa, J. P.; Baird, M. C.
Chem. Commun. 1986, 312-314. (h) Wassink, B.; Thomas, M. J.; Wright,
S. C.; Gillis, D. J.; Baird, M. C. J. Am. Chem. Soc. 1987, 109, 1995-
2002.
(5) (a) Gridnev, A. J. Polym. Sci. Part A: Polym. Chem. 2000, 38, 1753-
1766. (b) Gridnev, A. A.; Ittel, S. D. Chem. ReV. 2001, 101, 3611-3659.
(c) Heuts, J. P. A.; Roberts, G. E.; Biasutti, J. D. Aust. J. Chem. 2002, 55,
381-398.
(6) Abramo, G. P.; Norton, J. R. Macromolecules 2000, 33, 2790-2792.
9
234
J. AM. CHEM. SOC. 2007, 129, 234-240
10.1021/ja066325i CCC: $37.00 © 2007 American Chemical Society

H• Transfer from (
η
5-C5H5)Cr(CO)3H to Various Olefins
A R T I C L E S
Scheme 2
Figure 1. Olefins studied as H• transfer acceptors.
Despite the importance of metal-to-olefin H• transfer, few
kinetic studies4b,7,8 have been reported. If the rate-determining
step in a hydrogenation is the first H• transfer to the olefin, and
the resulting organic radical receives the second H• (rate constant
k′[M-H]) faster than it transfers the first H• back to M• (rate
constant ktr[M•]) (see Scheme 2), it is simple to determine kH
from the rate at which M-H disappears in the presence of
excess olefin. However, the kinetics become more complicated
when back transfer (ktr[M•]) competes with the second H•
transfer (k′[M-H]); kH cannot then be deduced from the overall
rate of the reaction. Bullock and Samsel have shown how kinetic
simulations can be used to obtain kH in a casesthe reaction
between R-cyclopropylstyrene and (η5-C5H5)Cr(CO)3Hswhere
back transfer is significant.7
Three years ago we obtained rate constants for H• transfer
from (η5-C5R5)Cr(CO)3H (R ) Ph, Me, H) to styrene-d8 and
methyl methacrylate-d5 (MMA-d5) by monitoring H/D ex-
change.9 We found that kH for MMA decreased as the chromium
hydride became more congested, while kH for styrene was less
affected by the steric bulk of the hydride.
Figure 2. Growth of 1 from H/D exchange between 2 (1.0 M) and 1-d1
(0.03 M) at 323 K in C6D6. The curve is calculated with the best-fit kobs
and [1]∞.
In the present paper, we report rate constants for H•/D• transfer
from (η5-C5H5)Cr(CO)3H (1) or (η5-C5H5)Cr(CO)3D (1-d1) to
variously substituted olefins: 1,1-diphenylpropene (2), 1,1-
diphenylethylene (3), R-methylstyrene (4), methyl crotonate (5),
1-octene (6), (E)-2-methylbut-2-enoic acid methyl ester (methyl
tiglate, 7), and 2-methyl-1-heptene (8) (Figure 1). Such transfers
lead to H/D exchange or hydrogenation or both, depending on
the olefin substituent. The kH values thereby obtained should
be useful in predicting synthetic applications of H• transfer
chemistry.
d[Cr-D]
-
) k1[2][Cr-D] ) kobs[Cr-D]
(4)
dt
S1) showed that kobs was a linear function of [2] (Figure 3),
confirming that exchange was first-order in each reactant,
second-order overall, as shown in eq 4. The slope of Figure 3
showed that k1 was 9.8(4) × 10-4 M-1 s-1 at 323 K.
Results
H/D Exchange between (η5-C5H5)Cr(CO)3D (1-d1) and 1,1-
Diphenylpropene (2). When a solution of 1-d1 was treated with
a large excess of 2 in C6D6 at 323 K, an exponential increase
in the hydride resonance of 1 was observed (Figure 2) in the
1H NMR. 2H NMR confirmed that H/D exchange had occurred
(eq 3), as a signal from the vinyl deuterium of 2-d1 appeared
while that from the deuteride resonance of 1-d1 decreased. Only
a trace of the hydrogenation product, 1,1-diphenylpropane, was
observed after an hour or two of reaction.
Extracting kD, the rate constant for the first D• transfer in
Scheme 3, from k1 requires knowledge of the fractional
probability S (eq 5) that H• will return to the chromium radical
after the initial transfer. If we define ktrH and ktrD as the second-
order rate constants for H• transfer and D• transfer, respectively,
from the radical 9-d1 to Cr• () CpCr(CO)3 ), we can write the
•
fractional probability S as in eq 6. The isotope effect ktrH/ktrD
can be estimated, as in our previous paper, by comparing the
rate constant for termination during an MMA-d0 polymerization
with that during an MMA-d5 polymerization;10 the ratio, a CfC
isotope effect for the transfer of H•/D• from an MMA chain-
carrying radical, is 2.93 at 60 °C. Our CfCr isotope effect can
also be estimated from the ratio of the efficiency of a cobalt
catalyst for chain transfer during the polymerization of MMA
to the efficiency of the same catalyst during the polymerization
of MMA-d8; this ratio has an average value of 3.5 between 40
1
Monitoring the increase of the hydride H NMR resonance
under these conditions (excess 2) gave a pseudo-first-order rate
constant, kobs (eq 4). Varying [2] from 0.29 to 1.0 M (Table
(7) Bullock, R. M.; Samsel, E. G. J. Am. Chem. Soc. 1990, 112, 6886-6898.
(8) (a) Bullock, R. M. Isotope Effects in Reactions of Transition Metal
Hydrides. In Transition Metal Hydrides; Dedieu, A., Ed.; VCH: New York,
1991, Chapter 8. (b) Bullock, R. M.; Bender, B. R. Isotope Methodss
Homogeneous. In Encyclopedia of Catalysis, Vol. 4; Horva´th, I. T., Ed.;
Wiley: New York, 2003.
(9) Tang, L.; Papish, E. T.; Abramo, G. P.; Norton, J. R.; Baik, M.; Friesner,
R. A.; Rappe´, A. J. Am. Chem. Soc. 2003, 125, 10093-10102; 2006, 128,
11314.
(10) Ayrey, G.; Wong, D. J. D. Polymer 1975, 16, 623.
9
J. AM. CHEM. SOC. VOL. 129, NO. 1, 2007 235

A R T I C L E S
Choi et al.
Scheme 4
Table 1. Values of kH and Activation Parameters for Isotope
Exchange between 1 and 3-d2 in Toluene-d8
1
T (K)
kH
(
×
10-2 M-1 s-
)
∆
Hq (kcal/mol)
∆
Sq (eu)
263.7
270.2
278.2
283.7
323.0
0.544(7)
1.01(2)a
2.03(8)
2.90(5)
46(6)b
12.0(5)
-23(2)
a Data plotted in Figure S1. b Extrapolation from activation parameters.
Figure 3. Plot of kobs vs [2] for D• transfer from 1-d1 to 2 at 323 K in
C6D6.
Reaction of (η5-C5H5)Cr(CO)3H (1) with 1,1-Diphenyl-
ethylene-2,2-d2 (3-d2) and r-Methylstyrene-d5 (4-d5). H/D
exchange between 1 and an excess of 3-d2 or 4-d5 proved too
rapid to follow by 1H NMR at 323 K. However, kinetic
experiments at lower temperatures were straightforward. Treat-
ment of 1 with an excess of 3-d2 in toluene-d8 at 270 K and
Scheme 3
1
examination of the resulting solution by H NMR showed an
exponential decay in the hydride resonance of 1 (Figure S1),
along with a smooth increase in the resonance (δ 5.41) of the
3-d1 arising from the H/D exchange in Scheme 4. No hydro-
genated product was observed.
and 80 °C.11 Our H•/D• transfer from carbon to chromium is
downhill by 12 kcal/mol,9 which argues for an early transition
state and against the large isotope effect associated with a
thermoneutral transfer and a symmetric transition state.8 The
only reason to think our isotope effect might be large is a recent
investigation by Bullock and co-workers of the reaction between
Monitoring the decay of the hydride resonance under these
conditions (excess 3-d2) gave a pseudo-first-order rate constant,
k
obs, as defined in eq 7. We can assume that H• transfers to
other substrates also obey second-order rate laws like that (eq
4) established for 2; such rate laws have been reported for the
transfer of H• from many hydrides to many olefins.8 Thus, we
can obtain values of k2 and kH, the rate constant for the first H•
transfer in Scheme 4, at various temperatures from expressions
(eqs 8 and 9) similar to those used with 2. The fractional
probability S (eq 8) that D• will return to the chromium radical
after the initial transfer can be estimated as 0.4 from eq 9. In
this case, S is more sensitive to the kinetic isotope effect; a
10% variation in ktrH/ktrD would change S and kH by 6%. The
values of kH at each temperature and the activation parameters
are given in Table 1.
•
1,4-cyclohexadiene and CpOs(CO)2 ; indirect observations are
consistent with an isotope effect between 4 and 20.12 The
Bullock reaction is, however, more nearly thermoneutral than
ours.
k1 ) SkD
(5)
1/2 ktrH[Cr•]
1/2 ktrD[Cr•] + 1/2 ktrH[Cr•]
S )
d[Cr-H]
k
trH/ktrD
-
) k2[3-d2][Cr-H] ) kobs[Cr-H]
k2 ) SkH
(7)
(8)
3
4
dt
)
≈
(6)
1 + ktrH/ktrD
If we estimate the kinetic isotope effect, ktrH/ktrD, as 3 as in
our previous paper,9,13 S will be 0.75 (eq 6). The value of S is
not very sensitive to the kinetic isotope effect; a 10% variation
in ktrH/ktrD would change S only by 3%. Dividing k1 by S gives
a kD of 1.31(5) × 10-3 M-1 s-1 for 2 at 323 K.
2/3 ktrD[Cr•]
2/3 ktrD[Cr•] + 1/3 ktrH[Cr•]
S )
2
2
5
)
≈
(9)
2 + ktrH/ktrD
(11) Gridnev, A. A.; Ittel, S. D.; Wayland, B. B.; Fryd, M. Organometallics
1996, 15, 5116-5126.
(12) Zhang, J.; Grills, D. C.; Huang, K.; Fujita, E.; Bullock, R. M. J. Am. Chem.
Soc. 2005, 127, 15684-15685.
Similarly, 1 was treated with an excess of 4-d5 in C6D6 at
temperatures from 283 to 306 K. A similar exponential decay
of the hydride resonance of 1 was observed, along with growth
of three proton resonances corresponding to the isotopomers
(13) We have used the same estimate (3) of ktrH/ktrD for all transfers of H•/D•
from carbon-centered radicals back to Cr• because the radicals in question
are similar (tertiary) and none of the equations for S is very sensitive to
this isotope effect (see text).
9
236 J. AM. CHEM. SOC. VOL. 129, NO. 1, 2007

H• Transfer from (
η
5-C5H5)Cr(CO)3H to Various Olefins
A R T I C L E S
Scheme 5
of excess 5 is probably the result of a small amount of
hydrogenation.
In contrast to substrates 2, 3, and 4, which form tertiary
radicals (9-d1, 10-d2, and 11-d5 in Schemes 3, 4, and 5) after
H•/D• transfer, substrate 5 will give the secondary radical 15-
d1 (Scheme 6). Thus, the initial D• transfer to 5 is less
thermodynamically favorable than the transfers above, and
therefore slower. However, because 15-d1 is a relatively
uncrowded secondary radical, it accepts a second D• before back
transfer can occur. Thus, “hydrogenation” is the exclusive
outcome of the reaction of 5 with 1-d1. Hydrogenation is also,
as some of us have previously reported,9 the exclusive outcome
of the reaction between CpCr(CO)3H and styrene, which also
generates a secondary radical by the initial H• transfer.
The fast second D• transfer in Scheme 6 contributes a factor
of 2 when we write the rate law (eq 11) in terms of the rate
constant kD for the initial D• transfer. The small amount of
hydrogenation we have observed gives an upper limit on kobs
Table 2. Values of kH and Activation Parameters for Isotope
Exchange between 1 and 4-d5 in C6D6
1
T (K)
kH
(
×
10-2 M-1 s-
)
∆
Hq (kcal/mol)
∆
Sq (eu)
283.2
289.6
295.6
301.1
306.2
323.0
0.31(1)
0.56(2)
0.97(2)a
1.47(2)
2.2(1)
14.1(2)
-20.1(7)
7.9(3)b
(eq 12), so kD at 323 K is e7.1 × 10-7 M-1 s-1
.
a Data plotted in Figure S2. b Extrapolation from activation parameters
d[Cr-D]
-
) 2kD[5][Cr-D] ) kobs[Cr-D]
(11)
(12)
dt
12 (δ 5.02), 13 (δ 5.38), and 14 (δ 1.98) (Scheme 5). The ratio
of the 12, 13, and 14 resonances was 1:1:3, reflecting the
expected equal probability of introducing H into each of the
two vinyl positions and the expected relative probability, 2:3,
of abstracting D• from the CHD2 vs the CD3 of the intermediate
radical 11-d5. We also observed a little of the hydrogenation
product cumene, in amounts implying that between 2 and 10%
of the hydride 1 had been consumed by this side reaction.
For simplicity, we ignored the hydrogenation and calculated
an observed psuedo-first-order rate constant, kobs, at various
temperatures from the decline in the hydride resonance of 1.
We then obtained values of kH, the rate constant for the first H•
transfer in Scheme 5, from an expression (eq 10) for S, the
fractional probability that D• will return to the chromium radical
after the initial transfer, like the expressions used above (eqs 7
and 8). In this case, a 10% variation in ktrH/ktrD would change
S and kH by 4%. The results are shown in Table 2.
[Cr-D]t
[Cr-D]0
0.95 e
(observed) ) exp(-kobst)
Similarly, little reaction occurred when 1-d1 (0.03 M) was
treated with an excess of 1-octene (6, 1.0 M) at 323 K in C6H6.
2H NMR showed that only about 4% of 1-d1 had been consumed
after 24 h. An analysis like that for 5 implies that kD for 6 at
323 K is e2.4 × 10-7 M-1 s-1
.
We then examined the reactivities of 7 and 8 (the methyl
derivatives of 5 and 6) so the incipient radicals would be tertiary
instead of secondary. Treatment of 1-d1 (0.03 M) with an excess
of 7 (the methyl ester of (E)-2-methylbut-2-enoic acid, 1.0 M)
at 323 K in C6H6 gave competing exchange and hydrogenation
2
(Scheme 7). The H NMR spectrum showed consumption of
70% of the 1-d1 in 10 h, with approximately a 50:50 ratio of
2H incorporation into 17/19 (arising from H/D exchange) and
20 (arising from hydrogenation). As shown in Scheme 7, H/D
exchange gave only 17 and 19; no 18 was observed. This result
probably reflects the thermodynamic instability of 18 relative
to 1714 and the reversibility of H•/D• transfer in these reac-
tions.
However, the complexity of the situation makes it impossible
to determine an exact value of kD from the time-dependent NMR
data. Neither ktr nor khyd is known independently. H/D exchange
adds to the number of rate expressions that must be considered;
e.g., hydrogenation is carried out not only by 1-d1 but also by
1-d0 (CpCr(CO)3H). Six rate constants, related by three kinetic
isotope effects (see Scheme S1 in the Supporting Information),
would have to be considered in simulating the operation of
Scheme 7.
5/6 ktrD[Cr•]
5/6 ktrD[Cr•] + 1/6 ktrH[Cr•]
S )
5
5
8
)
≈
(10)
5 + ktrH/ktrD
Reaction of (η5-C5H5)Cr(CO)3D (1-d1) with Methyl Cro-
tonate (5), 1-Octene (6), (E)-2-Methylbut-2-enoic Acid Meth-
yl Ester (7), and 2-Methyl-1-heptene (8). The reaction of 1-d1
(0.03 M) with an excess of 5 (1.0 M) at 323 K in C6D6 was
1
extremely slow, with little change in the H NMR spectrum
over 10 h. 2H NMR showed little decrease (e5%) in the
deuteride resonance of 1-d1 over 10 h in C6H6, and no other
deuterium resonance became observable. When, as a control
experiment, the stability of 1-d1 (CpCr(CO)3D) in C6H6 at 323
K was examined, its deuteride resonance showed no decrease
over the same period of time. Treating a concentrated solution
of 1 (CpCr(CO)3H, 0.1 M) with 0.5 equiv of 5 (0.05 M)
produced a small amount of the hydrogenation product methyl
butyrate after 2 days, so the 5% decrease in 1-d1 in the presence
Fortunately, the two limiting expressions, eqs 13-15 for H/D
exchange and eq 16 for hydrogenation, produce kD values that
differ only by about a factor of 2. Application of eqs 13-15 to
the decline of [1-d1] gives a kD of 3.6 × 10-5 M-1 s-1, which
would be negligibly affected by a 10% variation in ktrH/ktrD
.
(14) E-7 is 1.38 kcal/mol more stable than Z-7: J. Franz, Pacific Northwest
National Laboratories, personal communication.
9
J. AM. CHEM. SOC. VOL. 129, NO. 1, 2007 237

A R T I C L E S
Scheme 6
Choi et al.
Scheme 7
Scheme 8
Application of eq 16 to the same data gives a kD of 1.7 × 10-5
value is presumably lower. A 10% variation in ktrH/ktrD would
have a negligible effect on this S and kD.
M-1 s-1
.
5/6 ktrH[Cr•]
1/6 ktrD[Cr•] + 5/6 ktrH[Cr•]
d[Cr-D]
dt
-
) kobs[Cr-D]
(13)
(14)
S )
5(ktrH/ktrD
)
kobs ) SkD[7]
15
16
)
≈
(17)
1 + 5(ktrH/ktrD
)
4/5 ktrH[Cr•]
1/5 ktrD[Cr•] + 4/5 ktrH[Cr•]
Discussion
S )
Comparison of H• Transfer Rate Constants: Influence
of Olefin Structure. In order to compare the H•-accepting
ability of the various olefins, we need to be able to estimate kH
(from CpCr(CO)3H) when we have measured kD (from CpCr-
(CO)3D). Inverse isotope effects are generally observed for the
transfer of H• from M-H to a carbon-carbon double bond.8,15
The only isotope effect reported for CpCr(CO)3H, by Bullock
et al.,7 is kHCr/kDCr ) 0.45 ( 0.10 for its reaction with
2-cyclopropylpropene at 341 K. It seems appropriate to take
0.45 as a conversion factor, despite the small difference in
temperature,16 and we have done so in Table 3.
4(ktrH/ktrD
)
12
13
)
≈
(15)
(16)
1 + 4(ktrH/ktrD
)
kobs ) 2kD[7]
When 1-d1 was treated with an excess of 8 under the same
conditions used with 6, hydrogenation again competed with H/D
exchange. The H NMR spectra showed that only 23% of the
2
2
1-d1 had been consumed after 11 h and that the ratio of H
incorporation into 22/23 (arising from H/D exchange) and 25
(arising from hydrogenation) was 70:30. The internal double
bond isomer 24 was not observed, presumably because the
internal hydrogen in the radical 21-d1 is sterically hindered (see
Scheme 8).
kHCr
kH ) kD
(18)
kDCr
There have been numerous kinetic studies of free radical
additions to multiple bonds.17,18 However, there is comparatively
little information on the relative rates at which various olefins
Again, the competing reactions make it impossible to
determine kD exactly from the time-dependent NMR data. If
we consider the reaction to be an H/D exchange, calculate an
observed psuedo-first-order rate constant kobs from the decline
in the hydride resonance of 1, and calculate S from eq 17, we
obtain a kD of approximately e7.0 × 10-6 M-1 s-1; the actual
(15) It has been suggested8 that such inverse isotope effects are the consequence
of a two-step mechanism for H• transfer, with a fast H• transfer equilibrium
inside a solvent cage followed by slow diffusion out of the cage.
(16) The effect of the temperature difference is insignificant compared to the
uncertainty in our rate constants.
9
238 J. AM. CHEM. SOC. VOL. 129, NO. 1, 2007

H• Transfer from (
η
5-C5H5)Cr(CO)3H to Various Olefins
A R T I C L E S
Table 3. Rate Constants kH for H• Transfer from
reactivity over that of 6 by a factor of about 30. In general, a
methyl substituent on the incipient radical site increases the H•
transfer rate 5-50 times.
(η5-C5H5)Cr(CO)3H to Variously Substituted Olefins at 323 K
relative
rate
1
kH (× )
10-3 M-1 s-
The additional phenyl substituent in 3 increases its reactivity
over that of styrene by a factor of about 30sa bigger increase
than that produced by the additional methyl substituent in
R-methylstyrene (4), which is only 5 times more reactive than
styrene. These results reflect the ability of phenyl substituents
to stabilize radicals: the replacement of the italicized hydrogen
in Ph(CH3)(H)C-H by phenyl weakens the C-H bond by 4.4
kcal/mol, whereas the replacement of the same hydrogen by
methyl weakens the C-H bond by only 3.4 kcal/mol.20,22 Thus,
the stability of the incipient radical plays an important role in
determining the rate of such H• transfer reactions.
The data in Table 3 also show that a methyl substitutent on
the carbon that receives the H• slows down the rate of H• transfer
significantly: 2 is about 800 times less reactive than 3, and 7
is about 1200 times less reactive than MMA. The effect is
obviously steric and not unexpected from the literature. In their
investigation of the hydrogenation of styrenes by HCo(CO)4,
which occurs by an H• transfer mechanism, Roth and co-
workers4d found little reaction when there were â-methyl
substituents. Similarly Connolly,4e when studying the transfer
of H• from HFe(CO)4SiCl3 to dienes, found no observable
reaction when both ends of the diene were substituted.
2b
0.59 (2)
1
3
4
460 (60)
79 (3)
780
134
5b
6b
7b
e3.2 × 10-4
e1.1 × 10-4
e5 × 10-4
e2 × 10-4
∼0.02
(0.8-1.6) × 10-2
8b
e3.2 × 10-3
15.8 (6)a
14 (3)a
e0.005
27
styrene
MMA
24
a From our previous work.9 b Obtained from measured kD by eq 18.
accept H• from a hydride complex. Roth and co-workers have
reported rate constants for the stoichiometric hydrogenation of
several styrenes by HCo(CO)4.4d Masuyama and co-workers
have recently used competition experiments to compare the
reactivity of alkenes in a reaction that probably involves H•
transfer from Co, the Co(II)-catalyzed formation of trimethylsilyl
peroxides from O2 and Et3SiH.19 However, there has been no
previous direct measurement of rate constants for H• transfer
to olefins with a variety of substituents.
The reactivity of an olefin toward a free radical is largely
determined by three factors: (a) the ∆H of the reaction, (b) the
extent of steric congestion on the radical and the olefin, and (c)
polar effects. We would expect similar factors to determine the
rate at which an olefin can accept H• transfer from a transition-
metal hydride. Comparison of 4, 8, and MMA in Table 3 shows
that additional substituents on propylene affect its reactivity in
the order Ph (4) > CO2Me > alkyl (8), with the reactivity ratio
being about 24,000:4000:1. The sluggishness of 8 can be
attributed to the high energy of the radicals that would be
generated by H• transfer. A tertiary radical with only alkyl
substituents forms a C-H bond with a strength of 96.7 kcal/
mol; the introduction of a carbomethoxy substituent, for example
in (MeO2C)(CH3)2C-H, weakens the bond to 85 kcal/mol,20,21
while the introduction of a phenyl substituent, for example in
Experimental Section
General. All air-sensitive materials were prepared and handled under
an argon or nitrogen atmosphere using standard Schlenk techniques or
an inert atmosphere box. THF, benzene, C6D6, and toluene-d8 were
distilled under N2 from Na/benzophenone. Et2O and hexane were
deoxygenated and dried with two successive activated alumina columns
under Ar.23 Other chemicals were deoxygenated by purging with N2
or degassed by freeze-pump-thaw procedures. Commercial methyl
crotonate (5) was dried by CaH2; 1-octene (6) and 2-methyl-1-heptene
(8) were dried by sodium. Hexamethylcyclotrisiloxane was purified
by vacuum transfer and used as an internal standard in NMR kinetics
experiments.
Materials. 1,1-Diphenylpropene (2),24 1,1-diphenylethylene-2,2-d2
(3-d2),25 (E)-2-methylbut-2-enoic acid methyl ester (methyl tiglate, 7),26
and CpCr(CO)3H (1)27 were prepared as described in the literature.
CpCr(CO)3D. A slight modification of the literature procedure28
was employed. D3PO4 (0.15 mL, 85% in D2O, 99% d) was added to a
solution of 0.22 g of Na[CpCr(CO)3]29 (1 mmol) in 10 mL of THF.
After 45 min of stirring, THF was removed in vacuo, and the remaining
green powder was sublimed onto a dry ice/acetone-cooled sublimation
probe. A 54% yield of CpCr(CO)3D (0.11 g) was obtained. The isotopic
purity was >90%.
r-Methylstyrene-d5 (4-d5) was prepared by a slight modification
of the literature procedure.30 Acetone-d6 was treated with phenylmag-
nesium bromide, and the product was dehydrated with p-toluenesulfonic
acid in warm benzene to afford 4-d5 (>99% D).31
20,22
Ph(CH3)2C-H, weakens it to 84.4 kcal/mol.
As the data in Table 3 show, the methyl substituent in 4
increases its reactivity over that of styrene by a factor of 5; the
methyl substituent in 7 increases its reactivity over that of 5 by
a factor of about 40; the methyl substituent in 8 increases its
(23) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers,
F. J. Organometallics 1996, 15, 1518-1520.
(24) Simes, B. E.; Rickborn, B.; Flournoy, J. M.; Berlman, I. B. J. Org. Chem.
1988, 53, 4613-4616.
(25) Belluci, G.; Chiappe, C.; Moro, G. L. J. Org. Chem. 1997, 62, 3176-
3182.
(17) Fischer, H.; Radom, L. Angew. Chem., Int. Ed. 2001, 40, 1340-1371.
(18) Lalevee, J.; Allonas, X.; Fouassier, J. J. Org. Chem. 2005, 70, 814-819.
(19) Tokuyasu, T.; Kunikawa, S.; McCullough, K. J.; Masuyama, A.; Nojima,
M. J. Org. Chem. 2005, 70, 251-260.
(26) Buckles, R. E.; Mock, G. V. J. Org. Chem. 1950, 15, 680-684.
(27) Keppie, S. A.; Lappert, M. F. J. Organomet. Chem. 1969, 19, P5-P6.
(28) Edidin, R. T.; Sullivan, J. M.; Norton, J. R. J. Am. Chem. Soc. 1987, 109,
3945-3953.
(20) CRC Handbook of Chemistry and Physics, 82nd ed.; Lide, D. R., Ed.; CRC
Press: Boca Raton, FL, 2001-2002.
(29) Behrens, U.; Edelmann, F. J. Organomet. Chem. 1984, 263, 179-182.
(30) Somich, C.; Mazzocchi, P. H.; Ammon, H. L. J. Org. Chem. 1987, 52,
3614-3619.
(21) Engel, P. S.; Chen, Y.; Wang, C. J. Org. Chem. 1991, 56, 3073-3079.
(22) McMillen, D. F.; Golden, D. M. Annu. ReV. Phys. Chem. 1982, 33, 493-
532.
(31) Decrease of deuterium content in 4-d5 was observed under prolonged
reaction times or at high temperatures.
9
J. AM. CHEM. SOC. VOL. 129, NO. 1, 2007 239

A R T I C L E S
Choi et al.
Reaction of CpCr(CO)3H (or D) with Olefins. In preliminary
experiments, the concentrations of the olefins and 1 (or 1-d1) were varied
to determine the reaction time and temperature suited for measuring
kH (or kD) accurately. An excess of olefin (>10-fold over 1 or 1-d1)
was always employed. In most kinetic experiments, C6D6 or C6H6 (for
2H NMR) was used as a solvent. For the reaction of CpCr(CO)3H with
1,1-diphenylethylene-2,2-d2, toluene-d8 was used because of the low
temperatures required.
fit. The activation parameters were obtained from a linear fit to the
Eyring equation, and the values of kH at 323 K were extrapolated from
that equation; the uncertainties given for the extrapolated kH values
take into account covariance between ∆Hq and ∆Sq.
For 2H NMR kinetics in C6H6, the C6H5D resonance was used as an
internal standard,32 with 400 pulses (separated by a delay of 100 ms)
used to obtain each kinetic point.
Acknowledgment. This work was supported by the U.S.
Department of Energy, grant DE-FG02-97ER14807. The authors
are grateful to Dr. James Franz, PNNL, for helpful discussions,
and Jiamin Chin for preliminary work.
NMR Kinetic Measurements. All such experiments were performed
at 300 MHz. A stock solution of 1 or 1-d1 was prepared with the internal
standard (hexamethylcyclotrisiloxane), placed in a J-Young tube, and
frozen under an atmosphere of N2. A solution of the olefin was then
added to the same tube and also frozen under N2. The tube was allowed
to thaw, and its contents were mixed before it was placed in the probe
of the NMR (whose temperature controller was already set to the desired
temperature). Ethylene glycol (for T > 300 K) and methanol (for T <
300 K) standards were used to calibrate the controller.
The height of the hydride or deuteride peak (relative to the internal
standard) was recorded as a function of time. When the reaction was
monitored by 1H NMR, two pulses were generally used for each kinetic
point, with 10 s between the two pulses; however, for the reaction of
1,1-diphenylethylene-2,2-d2 with 1, only a single pulse was used (due
to the speed of the reaction). The kinetic data were fit to an exponential,
and the rate constant and the infinity point were adjusted for the best
Supporting Information Available: Details of (η5-C5H5)Cr-
(CO)3D/1,1-diphenylpropene isotope exchange, details of isotope
exchange between (η5-C5H5)Cr(CO)3H and 1,1-diphenylethyl-
ene-2,2-d2 or R-methylstyrene-d5, and a complete list of the
reactions required for the simulation of Scheme 7. This material
is available free of charge via the Internet at http://pubs.acs.org.
JA066325I
(32) The concentration of deuterium in C6H6 from the natural abundance of 2H
is approximately 0.01 M.
9
240 J. AM. CHEM. SOC. VOL. 129, NO. 1, 2007