Physical organic chemistry of transition metal carbene complexes. 21.1 Kinetics and mechanism of hydrolysis of (CO)5m=C(SR)AR (M = CR and W; R = CH3 and CH3CH2CH2; AR = C6H5 and 3-CIC6H4) in Aqueous Acetonitrile. Important differences relative to complexes with alkoxy leaving groups

Article abstract of DOI:10.1021/ja002371l

A kinetic study of the hydrolysis of (CO)₅M=C(SMe)Ph (M = Cr and W) in 50% MeCN-50% water (v/v) at 25 °C over a pH range from 1.7 to 14.2 is reported. The reaction occurs in two stages: the first is formation of (CO)₅M=C(O^-)Ph or (CO)₅M=C(OH)Ph while the second stage leads to the formation of PhCH=O and (CO)₅MOH^-. This paper is concerned with the first stage. The rate-pH profiles are complex and consistent with a mechanism (Scheme 1) that involves water/OH^- addition to the substrate to form a tetrahedral intermediate (T^-_OH), followed by product formation via five potential pathways whose relative importance depends on the pH. A more limited study of the reactions of (CO)₅M=C(SCH₂CH₂CH₃)Ph (M = Cr and W) and (CO)₅M=C(SMe)C₆H₄-3-C1 (M = Cr and W) with OH^- is also reported. The main focus of the discussion is aimed at understanding the reactivity differences between (CO)₅M=C(SMe)Ph and the corresponding methoxy analogues studied earlier. This understanding is in large measure based on an analysis, of how the intrinsic rate constants are affected by the interplay of steric, inductive, and π-donor effects and the potential imbalances of these effects at the transition state. It is also shown that the much lower sensitivity to H⁺-catalysis of RS^- compared to RO^- leaving group departure from the tetrahedral intermediate is responsible for the more complex rate-pH profile for the hydrolysis of (CO)₅M=C(SMe)Ph than for the hydrolysis of the methoxy analogue.

Full text of DOI:10.1021/ja002371l

J. Am. Chem. Soc. 2000, 122, 12441-12446
12441
Physical Organic Chemistry of Transition Metal Carbene Complexes.
21.1 Kinetics and Mechanism of Hydrolysis of (CO)5MdC(SR)Ar
(M ) Cr and W; R ) CH3 and CH3CH2CH2; Ar ) C6H5 and
3-ClC6H4) in Aqueous Acetonitrile. Important Differences Relative to
Complexes with Alkoxy Leaving Groups
Claude F. Bernasconi* and Gabriel S. Perez
Contribution from the Department of Chemistry and Biochemistry, UniVersity of California,
Santa Cruz, California 95064
ReceiVed June 30, 2000
Abstract: A kinetic study of the hydrolysis of (CO)5MdC(SMe)Ph (M ) Cr and W) in 50% MeCN-50%
water (v/v) at 25 °C over a pH range from 1.7 to 14.2 is reported. The reaction occurs in two stages: the first
is formation of (CO)5MdC(O-)Ph or (CO)5MdC(OH)Ph while the second stage leads to the formation of
PhCHdO and (CO)5MOH-. This paper is concerned with the first stage. The rate-pH profiles are complex
and consistent with a mechanism (Scheme 1) that involves water/OH- addition to the substrate to form a
tetrahedral intermediate (T-OH), followed by product formation via five potential pathways whose relative
importance depends on the pH. A more limited study of the reactions of (CO)5MdC(SCH2CH2CH3)Ph (M )
Cr and W) and (CO)5MdC(SMe)C6H4-3-Cl (M ) Cr and W) with OH- is also reported. The main focus of
the discussion is aimed at understanding the reactivity differences between (CO)5MdC(SMe)Ph and the
corresponding methoxy analogues studied earlier. This understanding is in large measure based on an analysis
of how the intrinsic rate constants are affected by the interplay of steric, inductive, and π-donor effects and
the potential imbalances of these effects at the transition state. It is also shown that the much lower sensitivity
to H+-catalysis of RS- compared to RO- leaving group departure from the tetrahedral intermediate is responsible
for the more complex rate-pH profile for the hydrolysis of (CO)5MdC(SMe)Ph than for the hydrolysis of the
methoxy analogue.
Introduction
shown to be consistent with the mechanism described by
Scheme 1 (X ) O); k-H1 and kH2 refer to H+-catalyzed OH- and
RX- departure from T-OH, respectively, kH3 to H+-catalyzed
A kinetic study of the hydrolysis of the Fischer type transition
metal carbene complexes 1-M,2 2-M2, and 3-Cr in 50%
RX- departure from T2- , and ki2 to intramolecular acid
OH
catalysis of RX- departure from TO-H by the OH group. The
rate law, given by
kobsd ) k1H O + k1OH[OH-] + k1B[B]
(1)
2
indicated that formation of the tetrahedral intermediate, TO-H
,
MeCN-50% water has recently been reported.3 As had been
shown by Aumann et al.,4 the reaction occurs in two stages.
The first stage involves conversion of the alkoxy carbene
complex into the corresponding hydroxy complex, e.g., 4-M,
by water attack at low pH and by OH- attack at high pH is rate
limiting; the kB1 [B] term refers to general base catalysis by
buffer bases.5
The reason Scheme 1 includes an intramolecular pathway
(ki2) as well as a pathway through the dianionic intermediate
T2O-H is that these additional pathways provide faster conversion
of TO-H to products than the k2H O + kHaH reactions and explain
2
+
2
why T-OH does not accumulate to detectable levels during the
reaction.6
the second stage leads to benzaldehyde and (CO)5MOH-. The
reported kinetic study dealt with the first stage. The results were
(3) Bernasconi, C. F.; Flores, F. X.; Kittredge, K. W. J. Am. Chem. Soc.
1997, 119, 2103.
(1) Part 20: Bernasconi, C. F.; Whitesell, C.; Johnson, R. A. Tetrahedron
2000, 56, 4917.
(2) When using the symbols 1-M, 2-M, etc. both the Cr and W derivatives
will be meant. When only one of the derivatives is referred to, we will use
the symbols 1-Cr, 1-W, 2-Cr, etc.
(4) Aumann, R.; Hinterding, P.; Kru¨ger, C.; Goddard, R. J. Organomet.
Chem. 1993, 459, 145.
(5) For simplicity, the buffer terms for all steps (kB1 [B], kBH[BH],
-1
kB2 H[B], and kB3 H[BH]) have been omitted from Scheme 1.
10.1021/ja002371l CCC: $19.00 © 2000 American Chemical Society
Published on Web 11/30/2000

12442 J. Am. Chem. Soc., Vol. 122, No. 50, 2000
Bernasconi and Perez
Scheme 1
In the present paper we report a detailed kinetic study of the
hydrolysis of the thiomethyl analogues to compound 1-M, i.e.,
5-M, and a more limited study of 6-M as well as the n-propyl
derivatives 7-M. It will be shown that the same mechanistic
scheme applies for these reactions as for the hydrolysis of 1-M
and 2-M, but that for the thioalkyl carbene complexes changes
in pH induce multiple changes in the rate-limiting step not
observed with the alkoxy derivatives. These changes in rate-
limiting step can be attributed to the much weaker response to
acid catalysis of thioalkyl compared to the alkoxy leaving
groups.
Figure 1. Time-dependent absorption spectra of (CO)5CrdC(SMe)-
Ph (5-Cr) in 50% MeCN-50% water at pH 14.0. Scans taken every
20 s.
Results
All experiments were performed in 50% MeCN-50% water
(v/v) at 25 °C and an ionic strength of 0.1 M maintained with
KCl. The reaction manifests itself by a distinct change in the
absorption spectrum as shown in Figure 1 for a representative
example. The presence of sharp isosbestic points indicates a
clean reaction. For 5-M, rates were measured under pseudo-
first-order conditions in KOH and HCl solutions, and in
triethylamine, N-methylmorpholine, and acetate buffers, span-
ning a pH range from 1.7 to 14.2. Figure 2 shows the rate-pH
profiles. The rate constants derived from the experiments in the
triethylamine and N-methylmorpholine buffers represent values
that were extrapolated to zero buffer concentration.
For 6-M and 7-M rate measurements were restricted to KOH
solutions. Plots of kobsd vs [KOH] were linear with intercepts
that were too small for an accurate determination.
Discussion
Mechanism. As mentioned in the Introduction, the reason
Figure 2. Rate-pH profiles for the hydrolysis of 5-Cr (O) and 5-W
(b).
the ki2 step and the pathway via T2- were added to the scheme
OH
for the reactions of 1-M and 2-M (X ) O in Scheme 1) was
that they are necessary to explain why T-OH does not ac-
cumulate to observable levels at high pH.3 This conclusion was
reached based on an estimate of KO1 H, the equilibrium constant
for formation of T-OH by addition of OH- to 1-M, and an
estimate of the kO1 H/k2H O ratio. These estimates led to the
2
conclusion that, at high pH, KO1 H[OH-] >(.) 1 and k1OH[OH-]/
-
kH2 O >(.) 1; these are the two necessary conditions for TOH to
2
become observable. The additional pathways must therefore lead
-
to a more rapid conversion of TOH to products than the k2H O
2
(6) T-OH would accumulate to detectable levels at high pH without these
additional pathways because the equilibrium of the first step is expected to
pathway, turning T-OH into a steady-state intermediate.
The situation for 5-M (X ) S in Scheme 1) is probably
different in that the two additional pathways are not necessary
favor T-OH over the reactants and k1OHaOH . k2H O, as discussed in detail in
2
-
ref 3.

Transition Metal Carbene Complexes
J. Am. Chem. Soc., Vol. 122, No. 50, 2000 12443
Table 1. Rate Constants for OH- and Water Attack on Fischer Carbene Complexes in 50% MeCN-50% Water at 25 °C
kO1 H, M-1 s-1
kH1 O, s-1
kP1rS, M-1 s-1
KP1rS, M-1
2
carbene complex
(CO)5CrdC(OMe)Ph (1-Cr)
(CO)5WdC(OMe)Ph (1-W)
(CO)5CrdC(SMe)Ph (5-Cr)
(CO)5WdC(SMe)Ph (5-W)
(CO)5CrdC(SMe)C6H4-3-Cl (6-Cr)
(CO)5WdC(SMe)C6H4-3-Cl (6-W)
(CO)5CrdC(SPr)Ph (7-Cr)
(CO)5WdC(SPr)Ph (7-W)
26.6a
2.9 × 10-3 a
1.34 × 104 b
4.21 × 104 b
5.33 ×102 c
1.58 ×103 c
1.06 × 104 b
1.35 × 104 b
1.10 ×106 c
4.75 × 106 c
26.3a
2.8 × 10-3 a
(1.27 ( 0.03) ×10-1
1.00 ( 0.03
(1.03 ( 0.06) × 10-3
(3.24 ( 0.05) × 10-3
(3.40 ( 0.10) × 10-1
2.04 ( 0.10
(7.25 ( 0.17) × 10-2
(3.10 ( 0.10) × 10-1
a Reference 3. b Reference 9. c Reference 12.
to turn T-OH into a steady-state intermediate. This is because
one would expect the kO1 H/k2H O ratio to be smaller than that for
2
X ) O for two reasons: kO1 H is lower (Table 1), as discussed in
more detail below, and kH2 O is expected to be higher because
2
MeS- is a better leaving group that MeO-.9 This means that
the kO1 H[OH-]/kH2 O ratio is probably <(,)1 even at the highest
2
pH (14.2) used in this study, ensuring that T-OH is a steady-state
intermediate irrespective of the contribution of the other
pathways. Nevertheless, since the additional pathways are crucial
for the alkoxy carbene complexes they should be included in
the scheme for the thioalkyl complexes as well, irrespective of
their relative contribution to the rate of conversion of T-OH to
products (more on this below).
apply; the kHaH term is again negligible and omitted from eq
7.
+
2
kH2 O . kO1 H[OH-] and k-1aH+ . k-H21O
(6)
(7)
H
2
kH2 O + k2i + KTa kH3 O/aH+ + KT kH . kH a
2
2
+
H
a
3
-1
In relating our kinetic results to Scheme 1 we note an
important contrast with the kinetic behavior of the alkoxy
complexes. For these latter, nucleophilic attack by OH- or water
is rate limiting over the entire pH range3 while for the thioalkyl
complexes the reaction undergoes several pH-dependent changes
in rate-limiting step. These changes are apparent from the
complex rate-pH profiles (Figure 2). We offer the following
interpretation for the four regions of the rate-pH profiles; the
general expression of kobsd and its derivation are given in the
Supporting Information.10
Region III (pH ∼6 to ∼10). In this region eq 6 remains valid
but eq 7 changes to eq 8, which means that the nucleophilic
k-1aH+ . kH O + ki2 + KaT kH3
(8)
2
2
attack step is now a rapid preequilibrium and hence kobsd is given
H
by eq 9, with KH1 O ) k1H O/k-1 being the equilibrium constant
2
2
for nucleophilic attack by water. The fact that the slope of the
KH1 O
2
Region I (pH J12.5 for 5-Cr, pH J11.5 for 5-W). In this
pH-range, OH- attack is rate limiting, i.e., the observed pseudo-
first-order rate constant, kobsd, is given by
2
kobsd
)
(kH2 O + ki2 + KTa k3H)
(9)
aH
+
rate-pH profile is approximately unity indicates that neither
the kHaH term nor the KaT k3H O/aH term contribute signifi-
2
kobsd ) kO1 H[OH-]
(2)
+
+
2
cantly to kobsd in this region and hence these terms have been
omitted from eqs 8 and 9.
This means that the relationships shown in eqs 3 and 4 apply.
Region IV (pH j6). Below pH 6 the kHaH term finally
Note that in eq 4 the kHaH term has been omitted because, as
+
+
2
2
becomes important; it now represents the dominant pathway
shown below, it is negligible down to pH 6.
for the rate-limiting collapse of T-OH (eq 10), eq 8 becomes eq
11, and eq 9 becomes eq 12.
kO1 H[OH-] . kH1 O and k-H1O . k-1aH
(3)
(4)
H
2
2
+
kH2 aH+ . kH O + kj2 + KaT kH3
(10)
(11)
(12)
2
k2H O + k2i + KTa kH3 O/aH+ + KT kH . k-H21O
2
2
2
a
3
k-1aH+ . kHa
+
H
Region II (pH ∼10 to ∼12.5 (5-Cr) or ∼10 to ∼11.5 (5-
W)). Nucleophilic attack by water is rate limiting, eq 5. The
2
kobsd ) KH1 O kH2
2
kobsd ) kH2 O + k1B[B]
(5)
2
Could the Reaction be Concerted? A referee has raised the
possibility that the hydrolysis of 5-M, 6-M, and 7-M might not
involve a tetrahedral intermediate (T-OH) but could conceivably
be concerted. This suggestion is based on theoretical calculations
which indicate that tetrahedral intermediates formed by attach-
ment of thiolate ions to carbonyl or acyl compounds are much
less stable than those formed by attachment of OH-.11 The fol-
lowing observations are inconsistent with a concerted reaction.
kB1 [B] term refers to buffer base catalyzed water attack, most
likely via a class n concerted reaction7 whose transition state is
shown as 8-M. The rate law implies that relationships 6 and 7
(7) Jencks, W. P. Chem. Soc. ReV. 1981, 10, 345.
(8) Bernasconi, C. F.; Ketner, R. J.; Brown, S. D.; Chen, X.; Rappoport,
Z. J. Org. Chem. 1999, 64, 8829.
(9) Bernasconi, C. F.; Kittredge, K. W.; Flores, F. X. J. Am. Chem. Soc.
1999, 121, 6630.
(10) See paragraph concerning Supporting Information at the end of this
paper.
(11) (a) Arad, D.; Langridge, R.; Kollman, P. A. J. Am. Chem. Soc. 1990,
112, 491. (b) Shokhen, M.; Arad, D. J. Mol. Mod. 1996, 2, 399.

12444 J. Am. Chem. Soc., Vol. 122, No. 50, 2000
Bernasconi and Perez
the ki2 and KaT kH3 terms may very well be negligible (e.g. in eq
9).
(1) Tetrahedral intermediates derived from nucleophilic attack
on Fischer carbene complexes are much more stable than those
derived from nucleophilic attack on esters. In fact, these
intermediates are so stable that, under the right conditions, they
accumulate to detectable levels, even those that contain alkyl
thio groups, e.g. 10-M (R ) n-Pr, HOCH2CH2, MeO2CCH2-
CH2, MeO2CCH2).9
Structure-Reactivity Relationships. Table 1 summarizes
the kO1 H and kH2O values for 1-Cr, 1-W, 5-Cr, and 5-W; also
-1
included are the kO1 H values for 6-Cr, 6-W, 7-Cr, and 7-W.
kO1 H: MeO vs MeS as Leaving Group. The reactivity of
the MeS complexes toward OH- is significantly lower than
that of the methoxy derivatives; for example, the kO1 H(5-Cr)/
kO1 H(1-Cr) ratio is 4.77 × 10-3 and the k1OH(5-W)/k1OH(1-W)
ratio is 3.80 × 10-2. These results are reminiscent of the lower
reactivity of the thiomethyl compared to the methoxy carbene
complexes toward thiolate ions;9,12 for example, for the reaction
with n-PrS-, the kP1rS(5-Cr)/k1PrS(1-Cr) ratio is 4.00 × 10-2 and
the k1PrS(5-W)/k1PrS(1-W) ratio is 3.75 × 10-2; for the reaction
with HOCH2CH2S-, the corresponding ratios are 2.93 × 10-2
and 4.52 × 10-2, respectively.
(2) The pH-rate profile requires the postulate of an inter-
mediate because, if a reaction does not involve an intermediate,
there cannot be changes in rate-limiting steps.
There are several factors that may affect these reactivity ratios.
They include steric, inductive, and π-donor effects; for the
reactions of 1-M with OH- they may also include an anomeric
effect.13 Specifically, the steric effect is expected to lower the
reactivity of 5-M relative to 1-M due to the larger size of the
MeS group;16 the stronger electron-withdrawing inductive effect
of the MeO group18 should enhance the reactivity of 1-M rela-
tive to that of 5-M; the anomeric effect may increase the reac-
tivity of 1-M toward oxyanion nucleophiles; the stronger
π-donor effect of the MeO group compared to that of the MeS
group20 should lower the reactivity of 1-M relative to that of
5-M.
How can one evaluate the relative importance of these factors?
For the reactions of 1-M and 5-M with thiolate ions, valuable
insights into this question were obtained by comparing the effect
of changing leaving group not only on the rate but on the
equilibrium constants for nucleophilic addition (KP1rS) as well.
It is therefore useful to briefly review the situation for these
reactions.12 The KP1rS values showed a trend that was opposite
to that for kP1rS, i.e., K1PrS was larger for the thiomethyl than for
the methoxy derivatives, e.g. KP1rS(5-Cr)/K1PrS(1-Cr) ) 1.04 ×
102 or KP1rS(5-W)/KP1rS(1-W) ) 3.52 × 102. These opposite
trends in the rate and equilibrium constants imply that the
intrinsic rate constants (ko)21 for the thiomethyl derivatives are
substantially lower than those for the methoxy derivatives. This
is an indication that the relative contributions of the steric,
inductive, and π-donor effects in affecting the rate constants
are different from those that affect the equilibrium constants,
i.e., the transition states of these reactions are imbalanced in
the sense that the development of these factors is not synchro-
nous with bond formation.22
Why Are the Rate-pH Profiles Different for Complexes
with Thio and Alkoxy Leaving Groups? For the hydrolysis
of 1-M, 2-M, and 3-Cr there are no changes from rate-limiting
nucleophilic attack at high pH to rate-limiting collapse of TO-H
at low pH, i.e., nucleophilic attack is rate limiting over the entire
rate-pH profile. The key feature responsible for the difference
between the two types of carbene complexes is that oxyanion
(OH- as well as RO-) departure from TO-H is very sensitive to
H+-catalysis while for thiolate ion departure the sensitivity to
H+-catalysis is much weaker. Hence, for the hydrolysis of the
alkoxy carbene complexes, H+-catalysis of conversion of T-OH
to products and of conversion back to reactants becomes
important in similar pH ranges. This means that lowering the
pH affects the rates of conversion of T-OH to reactants and
products in a comparable way so that no change in rate-limiting
steps occurs. In contrast, for the hydrolysis of the thioalkyl
carbene complexes, kH2 is so small that at pH values where the
H
H
+
+
k-1aH term starts to become important, the k2 aH term is still
negligible. Hence in the pH range from 6 to 10 where
H
+
H+-catalysis of OH- departure (k-1aH ) is already strong, H -
+
catalysis of RS- departure (kHaH ) is insignificant which ren-
+
2
H
-
+
ders the k-1aH process to be faster than the conversion of TOH
to products and leads to the observed change in rate-limiting
step. As the pH is lowered further, the kHaH term eventually
+
2
H
+
becomes significant but the k-1aH has also increased so that
the collapse of T-OH to products remains rate limiting.
The lower sensitivity to H+-catalysis of thiolate compared
to alkoxide ion departure has been observed in other reactions
and is a general phenomenon; a detailed discussion has been
presented in connection with the study of the acid-catalyzed
breakdown of alkoxide and thiolate ion adducts of benzylidene
Meldrum’s acid type adducts.8 The main reason for this reduced
sensitivity appears to be the lower basicity of RS- compared
to RO- or OH- ions. However, the mechanism for H+-catalysis
is probably the same for all leaving groups and most likely
represents a concerted reaction with a transition state such as
8-M (B ) H2O) or 9-M.
(12) Bernasconi, C. F.; Ali, M. J. Am. Chem. Soc. 1999, 121, 11384.
(13) In the present context, the anomeric effect14 refers to the stabilization
exerted by geminal oxygen atoms,15 e.g., in dialkoxy adducts such as T-OH
(X ) O).
(14) (a) Kirby, A. G. The Anomeric Effect and Related Stereoelectronic
Effects of Oxygen; Springer-Verlag: Berlin, 1983. (b) Schleyer, P. v. R.;
Jemmis, E. D.; Spitznagel, G. W. J. Am. Chem. Soc. 1985, 107, 6393.
(15) (a) Hine, J.; Klueppl, A. W. J. Am. Chem. Soc. 1974, 96, 2924. (b)
Wiberg, K. B.; Squires, R. R. J. Chem. Thermodyn. 1979, 11, 773. (c)
Harcourt, M. P.; More O’Ferrall, R. A. Bull. Soc. Chim. Fr. 1988, 407.
(16) Taft’s17 steric substituent constants, Es, are -0.55 for MeO and
-1.07 for MeS, respectively.
The low sensitivity of thiolate ion departure to acid catalysis
has implications regarding the relative importance of the
intramolecular pathway (kj2) and the pathway via T2- in
(17) Unger, S. H.; Hansch, C. Prog. Phys. Org. Chem. 1976, 12, 91.
OH
19
(18) σF ) 0.30 and 0.20 for MeO and MeS, respectively.
converting T-OH to products. Just as is the case for kH2 , k2i and kH3
are expected to be much smaller for X ) S than for X ) O,
(19) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165.
(20) σR ) -0.43 and -0.15 for MeO and MeS, respectively.
19
while kH3 O, just as kH2 O, should be much larger for X ) S than
2
2
(21) The intrinsic rate constant for a reaction with a forward rate constant
kf and a reverse rate constant kr is defined as ko ) kf ) kr when Kf ) kf/kr
) 1.
for X ) O. Hence the KaT k3H O/aH term is probably quite sig-
nificant and perhaps dominant at high pH (e.g., in eq 4) while
2
+
(22) For more details see ref 12.

Transition Metal Carbene Complexes
J. Am. Chem. Soc., Vol. 122, No. 50, 2000 12445
may not be as large as the KP1rS(5-M)/KP1rS(1-M) ratios for two
reasons. (a) The steric effect is probably smaller because OH-
is a smaller nucleophile than the thiolate ions. (b) The anomeric
effect should mainly increase KO1 H(1-M) but not K1OH(5-M).
These two factors should also enhance kO1 H(1-M) relative to
kO1 H(5-M), potentially making the k1OH(5-M)/kO1 H(1-M) ratios
smaller than the k1PrS(5-M)/kP1rS(1-M) ratios. However, these
factors should be partially or completely offset because the
reduced steric effect will be less effective in decreasing the
intrinsic rate constants for 5-M relative to those for 1-M. Indeed,
the actual kO1 H(5-W)/kO1 H(1-W) ratio (3.80 × 10-2) is almost
the same as the k1PrS(5-W)/k1PrS(1-W) ratio (4.00 × 10-2), while
the k1OH(5-Cr)/k1OH(1-Cr) ratio (4.77 × 10-3) is somewhat
lower than the k1PrS(5-Cr)/k1PrS(1-Cr) ratio (3.75 × 10-2).
A summary of the specific interpretation of the dependencies
of KP1rS and kP1rS on the leaving group is as follows. The fact
that KP1rS is larger for the thiomethyl carbene complexes
implies that the stronger reactant-stabilizing π-donor effect of
the methoxy group compared to the thiomethyl group is the
dominant factor on KP1rS, overriding the steric and inductive
effects. The lower intrinsic rate constants and hence lower k1PrS
values for the thiomethyl derivatives are mainly a consequence
of the development of the steric effect at the transition state
being more advanced than bond formation.23
Another contribution to the differences in intrinsic rate
constants is the stronger inductive effect of the methoxy group;
this inductive effect increases the intrinsic rate constant more
for the methoxy derivative than for the thiomethyl derivative
because, at the transition state, the negative charge, which in
the intermediate is delocalized into the CO ligands (T-OH in
Scheme 1), is mainly localized on the metal (11-M).25
kH1 O: MeO vs MeS as Leaving Group. The reactivity of
2
the MeS complexes toward water is very similar to that of the
MeO derivatives: the k1H O(5-Cr)/kH1 O(1-Cr) ratio is 0.355, the
2
2
kH1 O(5-W)/kH1 O(1-W) ratio is 1.16. This contrasts with the
reactions of OH- and thiolate ions discussed above, where 5-M
is substantially less reactive than 1-M. At first sight, this result
would appear to be in clear violation of the Reactivity-
Selectivity Principle27 or the Hammond29-Leffler30 postulate.
The following interpretation shows that the results are actually
understandable in the context of the Hammond-Leffler pos-
tulate.
2
2
Regarding the possible contribution of the π-donor effect to
the difference in the intrinsic rate constants, the situation is more
complex because there are two opposing interaction mechanisms
that partially or completely offset each other. One is the loss of
resonance stabilization of the reactants that is expected to run
ahead of bond formation and hence lower the intrinsic rate
constant.23 The π-donor ability of oxygen being stronger than
that of sulfur,20 there will be a stronger reduction of ko for 1-M
than for 5-M. The other interaction mechanism is the preorga-
nization of the structure of the (CO)5M moiety in 5-M or 1-M
toward its electronic configuration in the adducts that results
from the π-donor effect (11-M). As a consequence, the lag in the
charge delocalization into the CO ligands at the transition state25
is reduced and the intrinsic rate constant for 1-M is not as
strongly depressed by the effect associated with this lag as that
for 5-M. Observations regarding the effect of changing π-donor
strength in the deprotonation of alkoxyalkyl carbene complexes27
suggest that the two factors essentially cancel each other, i.e.,
the π-donor effect on the intrinsic rate constant is negligible.
Can the reasoning regarding the relative rate constants for
the reactions of 1-M and 5-M with thiolate ions be applied in
explaining why OH- reacts more slowly with 5-M than with
1-M? The answer is yes. As is the case for KP1rS, the equilibri-
um constants for OH- addition, KO1 H, are expected to be higher
for 5-M than for 1-M, although the KO1 H(5-M)/K1OH(1-M) ratios
The equilibrium constants for water addition, KH1 O, are
2
related to KO1 H for OH- addition by eq 13, with Kw being the
ionic product of the solvent. Hence the KH1 O(5-M)/K1H O(1-M)
2
2
KH1 O ) KO1 HKw
(13)
2
ratios must be equal to the KO1 H(5-M)/K1OH(1-M) ratios. In the
previous section it was argued that these ratios are >(.)1
although probably not as large as the KP1rS(5-M)/K1PrS(1-M)
ratios (100 to 350); and the fact that the kO1 H(5-M)/k1OH(1-M)
ratios are ,1, i.e., do not follow the trend of the equilibrium
constant ratios, was attributed to transition state imbalances.
For the water reaction, the kH1 O(5-M)/k1H O(1-M) ratios of
approximately unity do not follow the trend in the KO1 H(5-M)/
KO1 H(1-M) ratios either. This must again be the result of
transition state imbalances but the less dramatic discrepancy
between the rate and equilibrium constant ratios suggests that
2
2
these imbalances are smaller and hence have a smaller effect
on k1H O than on k1OH or k1PrS
.
2
A plausible reason for the smaller imbalances is that, due to
the lower reactivity of water, the transition state is later in the
sense that bond formation between the nucleophile and the
carbene carbon is more advanced than in the reactions with OH-
or thiolate ions, as predicted by the Hammond-Leffler postulate.
This has the effect of reducing the imbalance between the
development of the steric effect or the delocalization of negative
charge into the carbonyl ligands on one hand and bond formation
on the other, i.e., the impact on the intrinsic rate constants and
hence on the actual rate constants is reduced.
(23) This is a manifestation of the principle of nonperfect synchronization
(PNS).24 With respect to the factors relevant to the present study, the PNS
states that the intrinsic rate constant decreases if (a) the development of a
product destabilizing factor (e.g., a steric effect) runs ahead of bond changes
at the transition state, (b) a product stabilizing factor (e.g., resonance) lags
behind bond changes, or (c) the loss of a reactant stabilizing factor is ahead
of bond changes (π-donor effect).
(24) (a) Bernasconi, C. F. Acc. Chem. Res. 1987, 20, 301. (b) Bernasconi,
C. F. Acc. Chem. Res. 1992, 25, 9. (c) Bernasconi, C. F. AdV. Phys. Org.
Chem. 1992, 27, 119.
It should be noted that this is a conclusion that not only
pertains to the reactions discussed in this paper but is the general
validity. In other words, a more advanced transition state will
(25) This lag of the charge delocalization behind bond changes is another
example of a transition state imbalance. The resulting PNS23 effect is a
major factor why the intrinsic rate constants for reactions of Fischer carbene
complexes with nucleophiles or bases (deprotonation of carbene complexes
with protons on the R-carbon) are relatively small.26
(28) (a) Pross, A. AdV. Phys. Org. Chem. 1977, 16, 69. (b) Giese, B.
Angew. Chem., Int. Ed. Engl. 1977, 16, 125. (c) Rappoport, Z., Ed. Isr. J.
Chem. 1985, 26, 303-428 (RSP-issue).
(29) Hammond, G. S. J. Am. Chem. Soc. 1955, 77, 334.
(30) Leffler, J. E.; Grunwald, E. Rates and Equilibria of Organic
Reactions; Wiley: New York, 1963; p 156.
(26) For a review, see: Bernasconi, C. F. Chem. Soc. ReV. 1997, 26,
299.
(27) Bernasconi, C. F.; Leyes, A. E. J. Am. Chem. Soc. 1997, 119, 5169.

12446 J. Am. Chem. Soc., Vol. 122, No. 50, 2000
Bernasconi and Perez
always leave less room for imbalances of any kind. This is best
appreciated by considering the limiting case where the transition
state is almost completely product-like since for such a transition
state all processes will have approached 100% completion.
(4) For the reaction with OH-, the MeS complexes are
kinetically much less reactive than the MeO derivatives yet the
equilibrium constants for intermediate formation are probably
larger for the MeS than for the MeO complexes. This indicates
that the intrinsic rate constants are much lower for the reactions
of the MeS complexes than for the MeO complexes due to
imbalances in the development of the steric effects and the
delocalization of the negative charge into the CO ligands.
(5) For the reaction with water, the MeS and MeO complexes
show comparable kinetic reactivities. This can be understood
in terms of a later transition state than in the reactions with
OH-, which leads to smaller imbalances.
kO1 H: Effect of Metal, Size of Leaving Group, and Phenyl
Substituent. As has been noted before, the tungsten carbene
complexes are usually slightly more reactive than their chro-
mium counterparts.9,12,26,31 This is particularly true for the
thioalkyl carbene complexes,11 as is borne out by the results
for the reactions of 5-M, 6-M, and 7-M with OH- and water.
As discussed elsewhere, the higher reactivity of the tungsten
derivatives results from a combination of a somewhat greater
thermodynamic driving force as well as a greater intrinsic rate
constant for adduct formation.9
Experimental Section
Materials. 5-Cr and 5-W were synthesized as described previously12
by reacting 1-Cr and 1-W, respectively, with NaSMe in a benzene-
methanol mixture. The same procedure was used to prepare 6-Cr and
6-W from (CO)5MdC(OMe)C6H4-3-Cl (M ) Cr and W) with the latter
complexes being available from a previous study.31 1H NMR (CDCl3)
for 6-W: δ 2.16 (s, 3H, CH3S), 6.61 (d, 1H, Ar), 6.73 (s, 1H, Ar),
In some cases, e.g. for the reactions of 1-Cr and 1-W with
OH- and water, the reactivities of the chromium and tungsten
complexes are about the same.3 The reasons for this are not
entirely clear.
The dependence on the RS group, 7-M vs 5-M, shows a small
reduction in kO1 H for the carbene complexes with the n-PrS
group, indicating a steric effect. The somewhat larger k1OH
values for 6-M compared to 5-M (kO1 H(6-Cr)/k1OH(5-Cr) )
2.68, kO1 H(6-W)/kO1 H(5-W) ) 2.04) indicate stabilization of
the incipient negative charge on the transition state by the
electron-withdrawing m-Cl group. These ratios are similar to
the kR1 O(8-W)/kR1 O(1-W) ratio of 3.01 for the reactions with
CF3CH2O- in 50% MeCN-50% water, where 8-W ) (CO)5Wd
C(OMe)C6H4-3-Cl.31
1
7.37 (m, 2H, Ar). H NMR (CDCl3) for 6-Cr: δ 2.33 (s, 3H, CH3S),
6.53 (d, 1H, Ar), 6.65 (s, 1H, Ar), 7.23 (d, 1H, Ar), 7.40 (t, 1H, Ar).
13C NMR (CDCl3) for 6-W: δ 28.26 (CH3S), 116.46, 126.51, 129.39
and 158.52 (Ar), 196.34 (CO, cis), 208.08 (CO, trans), 324.21 (dC).
13C NMR (CDCl3) for 6-Cr: δ 28.54 (CH3S), 117.01, 126.90, 130.27,
and 158.09 (Ar), 215.43 (CO, cis), 228.58 (CO, trans), 358.89 (dC).
7-Cr and 7-W were available from a previous study.9
Instrumentation, Procedures, and Kinetic Experiments. NMR
spectra were recorded on a Bruker 250 MHz instrument. UV spectra
were obtained on a Hewlett-Packard 8562 diode array spectrophotom-
eter. Kinetic experiments were performed on a Perkin-Elmer Lambda
2 or the 8562 diode array spectrophotometer. pH measurements were
made with an Orion Research 611 digital pH meter and calibrated with
standard aqueous buffers at pH 4.00, 7.00, and 10.00. Actual pH values
were calculated by adding 0.18 to the measured pH, according to Allen
and Tidwell.32 All rates were measured under pseudo-first-order
conditions as described under Results; the reactions were monitored at
466 and 450 nm, respectively, corresponding to the λmax of the Cr and
W complexes, respectively. For the slowest reactions the method of
initial rates was used where kobsd was determined by using eq 14 where
Conclusions
(1) The hydrolysis of 5-M and, presumably, that of 6-M and
7-M as well proceed by the mechanism shown in Scheme 1,
the same mechanism postulated for the hydrolysis of alkoxy
carbene complexes (1-M). However, the ki2 and KaT k3H path-
ways are probably negligible while the kH2 O pathway should be
2
much more important than for 1-M.
(2) The rate-limiting step changes from nucleophilic attack
slope
by OH- (region I of the pH rate profiles) and by water (region
kobsd
)
(14)
Ao - A∞
-
II) to breakdown of TOH via the combined kH2 O, k2i , and KaT k3H
2
pathways (region III) and finally to the kH2 pathway (region
IV). This contrasts with the hydrolysis of 1-M where nucleo-
philic attack by OH- (high pH) or water (low pH) is rate limiting
over the entire pH range.
slope is the slope of a plot of absorbance, A, vs time for the first 1-5%
of reaction, and Ao is the initial absorbance and A∞ the absorbance after
the reaction is complete. This latter was obtained from runs in KOH
solutions.
(3) The key feature responsible for the difference between
how the hydrolysis of the two types of complexes respond to
pH changes is that OH- and MeO- departure from TO-H is very
sensitive to H+-catalysis while for RS- departure the sensitivity
to H+-catalysis is much weaker.
Acknowledgment. This research was supported by Grant
No. CHE-9734822 from the National Science Foundation.
G.S.P. was supported by NIH/MARC GM07910.
Supporting Information Available: General rate expression
for Scheme 1 (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
(31) Bernasconi, C. F.; Garc´ıa-R´ıo, L. J. Am. Chem. Soc. 2000, 122,
3821.
(32) Allen, A. D.; Tidwell, T. T. J. Am. Chem. Soc. 1987, 109, 2774.
JA002371L