Mechanism of the nucleophilic substitution of acyl electrophiles using lithium organocuprates

Article abstract of DOI:10.1002/adsc.200800060

The mechanism of nucleophilic substitution reaction at an sp² carbon center of a thioester or an acid chloride with a lithium organocuprate reagent has been investigated. Density functional calculations indicated that the thioester undergoes oxidative addition of the C-S bond to the copper(I) atom through a three-centered transition state to afford an organocopper(III) intermediate, which gives the product through reductive elimination of the alkyl and the acyl groups. On the other hand, the acid chloride loses a chloride anion very easily when it interacts with the cuprate, because the chloride anion is captured by a lithium(I) cation rather than a copper(I) atom. ¹³C kinetic isotope effect (KIE) experiments showed excellent agreement with computational predictions for the thioester reaction, but suggested that the nucleophilic displacement transition state of the acid chloride occurs much more advanced than the calculations predict.

Full text of DOI:10.1002/adsc.200800060

FULL PAPERS
DOI: 10.1002/adsc.200800060
Mechanism of the Nucleophilic Substitution of Acyl Electrophiles
using Lithium Organocuprates
a
a
a,
Naohiko Yoshikai, Ryoko Iida, and Eiichi Nakamura *
a
Department of Chemistry, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan
Fax : (+81)-3-5800-6889; e-mail: nakamura@chem.s.u-tokyo.ac.jp
Received: January 26, 2008; Published online: April 15, 2008
This paper is dedicated to Prof. Andreas Pfaltz, my long-time best friend, on the occasion of his 60th birthday,
in honor of his outstanding contribution to the field of transition metal catalysis.
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: The mechanism of nucleophilic substitu- anion is captured by a lithium(I) cation rather than a
2
13
tion reaction at an sp carbon center of a thioester or copper(I) atom. C kinetic isotope effect (KIE) ex-
an acid chloride with a lithium organocuprate re- periments showed excellent agreement with compu-
agent has been investigated. Density functional cal- tational predictions for the thioester reaction, but
culations indicated that the thioester undergoes oxi- suggested that the nucleophilic displacement transi-
dative addition of the CÀS bond to the copper(I) tion state of the acid chloride occurs much more ad-
atom through a three-centered transition state to vanced than the calculations predict.
afford an organocopper(III) intermediate, which
A
H
R
N
gives the product through reductive elimination of
the alkyl and the acyl groups. On the other hand, the Keywords: cuprates; density functional calculations;
acid chloride loses a chloride anion very easily when isotope effects; nucleophilic substitution; reaction
it interacts with the cuprate, because the chloride mechanisms
Introduction
placement of the CÀS bond. While the catalytic var-
iants developed later have overridden stoichiometric
[
9,10]
Nucleophilic substitution of a carboxylic acid chloride, cuprate acylation,
it still finds applications as a
[
11]
or its equivalent, with an organometallic reagent is highly reliable transformation.
one of the most basic methods used to synthesize ke-
Despite its synthetic utility, the mechanism of orga-
[1,2]
tones.
While organozinc reagents were the first or- nocuprate reactions with acyl electrophiles has not
ganometallic reagents employed in such transforma- been studied, while other organocopper transforma-
tions, they quickly fell out of favor because of their tions, such as conjugate additions and alkylation reac-
[
3]
[12,13]
extremely limited scope. On the other hand, since tions, have been extensively studied by both our
[
4]
[12,14]
the first report by Gilman in 1936 followed more re- and other research groups.
cently by Whitesides in 1967, organocopper re- that the mechanism of the acylation of R CuLi is not
These studies suggest
[
5]
2
agents, and organocuprate reagents (R CuLi) in par- merely a simple “addition-elimination” sequence of
2
À
ticular, have shown a much wider scope in the substi- the R anion, but involves a Cu(I)/Cu
tution of acid chlorides. The mild, yet sufficient, nu- cess.
cleophilicity of organocuprates allows a high compati-
A
H
R
U
G
[
6]
An organocuprate transformation related to these
bility with functional groups such as ketones, nitriles, acylation reactions would be the substitution of an al-
[
15]
and esters, and they can be used in the synthesis of a kenyl halide. Our recent study on this reaction sug-
[
7]
variety of natural products. Shortly after the reports gested two mechanisms of the halide displacement on
2
[16]
on the acid chloride reaction, Rosenblum et al. re- the sp carbon atom (Scheme 1).
One mechanism
ported on the substitution of a thioester with an orga- involves an insertion of the copper(I) atom into the
[8]
nocuprate. This reaction further expanded the utility carbon-halogen bond followed by the reductive elimi-
of organocopper reagents in ketone synthesis, since nation of the resulting organocopper(III) intermedi-
classical organozinc reagents do not undergo a dis- ate (a three-centered mechanism, path a). In the
AHCTREUNG
Adv. Synth. Catal. 2008, 350, 1063 – 1072
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1063

Naohiko Yoshikai et al.
FULL PAPERS
Scheme 1. The two pathways for the substitution of vinyl bromide with lithium organocuprate.
other mechanism, the halide anion eliminates by itself be entirely different from each other. Thus, the reac-
without an interaction with the copper atom, and is tion of the thioester involves a three-centered oxida-
trapped by the lithium cation (an eliminative mecha- tive addition/reductive elimination sequence, while
nism, path b). Experimental measurements and theo- the acid chloride follows the eliminative reaction
12
13
retical predictions of the C/ C kinetic isotope effect pathway. The origin of the contrasting reaction path-
indicate that the latter mechanism occurs in the ways will be discussed in more detail.
actual reaction.
We envisioned that such a mechanistic dichotomy
(
i.e., a three-centered versus eliminative mechanism) Results and Discussion
2
could also exist in the substitution reaction on the sp
carbon atom of acyl electrophiles. In this paper, we Reaction of [Me CuLi] and S-Methyl Thioacetate
2
2
report on theoretical and experimental studies on the
nucleophilic substitution reactions of a thioester and First, we examined the reaction pathway for the sub-
an acid chloride with a lithium organocuprate. The re- stitution of S-methyl thioacetate (2a) with [Me CuLi]2
2
action pathways of these electrophiles turned out to (1) (Scheme 2, Figure 1). The dimeric organocuprate
Scheme 2. Reaction pathway for the substitution of S-methyl thioacetate with [Me CuLi] . Values in parentheses refer to the
2
2
À1
potential energy (in kcalmol ) relative to (1 + 2a). The change in energy is shown above the reaction arrows.
1064
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1063 – 1072

Mechanism of the Nucleophilic Substitution of Acyl Electrophiles
FULL PAPERS
À1
Figure 1. Potential energy profile (in kcalmol ) for the sub-
stitution of S-methyl thioacetate with [Me CuLi] . Values in
2
2
À1
italic refer to the Gibbs free energies (kcalmol ) at 298 K.
1
and 2a initially form the complex 4a by an interac-
tion between the lithium and the oxygen atoms (DE=
1
À1
3.8 kcalmol ). Complex 4a then undergoes oxida-
Figure 2. Structure of the complexes and the TS in the reac-
tive addition of the CÀS bond to the copper atom tion of [Me CuLi] and S-methyl thioacetate. The values
2
2
°
through a three-centered transition state (TS) (DE = refer to the bond lengths in Š (plain text) and natural
À1
2
4.7 kcalmol for TS4a-5a) to give the square-planar charges (underlined).
acyldimethylcopper(III) intermediate 5a. The subse-
A
H
R
U
G
quent reductive elimination of the methyl and the
acyl groups (TS5a-PD) occurs smoothly with an activa-
À1
À1
tion energy of 2.4 kcalmol . The overall process is bilization of 5.1kcalmol . Then, the Cu atom ap-
À1
expectedly highly exothermic (À30.6 kcalmol ). Note proaches the C=O bond to form the p-complex 4a’·2S
°
À1
that an eliminative TS for the CÀS bond cleavage through TS4a-4a’·2S (DE =15.6 kcalmol ). This pro-
À1
(
such as path b in Scheme 1) was not found. The cess is endothermic by 14.8 kcalmol . The p-complex
strong affinity of the sulfur atom for the copper atom 4a’·2S undergoes three-centered oxidative addition of
°
should make the eliminative mode of the CÀS cleav- the CÀS bond to the Cu atom (TS
·2S, DE =
4
a’-5a
À1
age much more unfavorable than the three-centered 3.2 kcalmol ) to form a square-planar Cu
ACHTREUNG
mode (i.e., TS4a-5a).
mediate 5a·2S. The triorganocopper
ACHTREUNG
The structures of the complexes and the TS are 5a·2S smoothly° undergoes reductive elimination
À1
shown in Figure 2. The CÀS bond substantially elon- (TS5a-PD·2S, DE =5.6 kcalmol ) to give the final
gates on passing from complex 4a to TS
(a 20% in- products. The energy profile of the overall pathway
a-5a
4
crease from 1.77 to 2.13 Š), and the forming CuÀC (Figure 3) indicates that the CÀS bond cleavage is the
acyl) (1.97 Š) and the CuÀS (2.37 Š) bond distances rate-determining step of the reaction.
in TS are very close to those in the oxidative addi- The structure of the intermediates and the TS are
(
4a-5a
tion product 5a (CuÀC=1.93 Š and CuÀS=2.32 Š). shown in Figure 4. Upon p-complexation, both the
Natural population analysis indicates a formal oxida- CÀS and the CÀO bonds elongate (CÀS bond=
tion of the Cu atom (+0.44 in 4a and +0.86 in 5a) 1.77 Š in 4a·2S and 1.86 Š in 4a’·2S, and the CÀO
and the displacement of the SMe group as a thiolate bond=1.22 Š in 4a·2S and 1.33 Š in 4a’·2S). In a par-
anion (+0.26 in 4a and À0.50 in 5a). The structures of allel manner, the LiÀO (acyl) bond distance becomes
the Cu(III) intermediate 5a and its reductive elimina- considerably shorter (2.01Š in 4a·2S and 1.82 Š in
A
C
H
T
R
E
U
N
G
[12–14]
tion TS (TS5a-PD) are similar to the related cases.
4a’·2S). These structural changes should reflect the
Based on the above reaction pathway, we examined back-donation from the Cu atom to the C=O p*/CÀS
the effect of solvent coordination to the lithium atom s* mixed orbital. In fact, the positive charge on the
using the model complex 1·2S (S=Me O). As shown Cu atom increases (+0.42 to +0.92), while the nega-
2
in Scheme 3, the reaction pathway is essentially the tive charges increase on the carbonyl oxygen (À0.65
same as the reaction pathway without any coordinat- to À0.87), the carbonyl carbon (+0.41to 0.00) and
ing solvent molecules (Scheme 2) except for the pres- the SMe group (+0.26 to +0.02). The oxidative addi-
ence of a p-complex 4a’·2S. First, the complex 4a·2S tion TS (TS4a’-5a·2S) takes a similar structure to the TS
forms through a lithium-oxygen interaction with a sta- without any Me O molecules (see TS
in Figure 2)
2
4a-5a
Adv. Synth. Catal. 2008, 350, 1063 – 1072
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1065

Naohiko Yoshikai et al.
FULL PAPERS
Scheme 3. Reaction pathway for the substitution of S-methyl thioacetate with [Me CuLi] ·2S (S=Me O). Values in parenthe-
2
2
2
À1
ses refer to the potential energy (in kcalmol ) relative to (1·2S + 2a). The change in energy is shown above the reaction
arrows.
À1
Figure 3. Potential energy profile (in kcalmol ) for the substitution of S-methyl thioacetate with [Me CuLi] ·2S (S=Me O).
2
2
2
À1
Values in italic refer to the Gibbs free energies (kcalmol ) at 298 K.
except for the small difference in the CuÀS and LiÀS Reaction of [Me CuLi] and Acetyl Chloride
2
2
bond distances (CuÀS=2.46 Š versus 2.37 Š, and LiÀ
S=2.75 Š versus 3.07 Š). The structures of the result- Next, we examined the reaction pathway for the sub-
ing Cu(III) intermediate, 5a·2S, and the following re- stitution of acetyl chloride (3a) with [Me CuLi] (1)
ductive elimination TS (TS5a’-PD·2S) are very close to (Scheme 4, Figure 5). First, the dimeric organocuprate
ACHTREUNG
2
2
those without any coordinating solvents (Figure 2).
1 and 3a form a complex 8a through the lithium-
oxygen interaction having a stabilization energy of
À1
9
.4 kcalmol . Complexation between the Cu atom
1066
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1063 – 1072

Mechanism of the Nucleophilic Substitution of Acyl Electrophiles
FULL PAPERS
Figure 4. Structure of complexes and TS in the reaction of [Me CuLi] ·2S and S-methyl thioacetate. The values refer to the
2
2
bond lengths in Š (plain text) and natural charges (underlined).
and the C=O bond takes place through TS8 with an
The structure of the intermediates and TS are
a-9a
À1
activation energy of 16.8 kcalmol to give the p-com- shown in Figure 6. The CÀCl bond elongates consider-
plex 9a. The formation of 9a from 8a is endothermic ably on p-complex formation (an increase of 20%,
À1
by 15.9 kcalmol . The p-complex 9a undergoes a 1.79 Š in 8a and 2.15 Š in 9a). Accordingly, both the
practically spontaneous CÀCl bond cleavage through negative charge on the Cl atom (À0.03 in 8a and
an eliminative TS (TS
). The potential surface À0.42 in 9a) and the positive charge on the Cu atom
9
a-10a
around the CÀCl bond cleavage is extremely flat, and (+0.44 in 8a and +0.84 in 9a) increase. In accordance
the calculated activation barrier is low at 0.006 kcal with the extremely low activation barrier for the CÀ
À1
mol . The eliminative CÀCl bond cleavage leads to Cl bond cleavage, the structural changes on passing
the formation of the triorganocopper
A
H
R
U
G
are very small. For example, the
a-10a
9
ate 10a. Rearrangement of the complex 10a gives a CÀCl bond length increases by only 2% (2.15 Š to
more stable copper(III) complex 11a, which under- 2.20 Š). The acyldimethylcopper complex 10a has a
goes facile reductive elimination through TS highly distorted square-planar structure, where the
DE =4.2 kcalmol ) to give the final products. The carbonyl oxygen acts as an internal ligand to the Cu-
AHCTREUNG
1
1a-PD
°
À1
(
overall transformation is expectedly highly exother-
A
T
E
N
À1
mic (À46.7 kcalmol ). Note that all attempts to is indicated by the short CuÀO bond distance
locate a three-centered insertion TS (such as TS
Scheme 2) were unsuccessful.
in (2.29 Š) and the small CuÀCÀO bond angle (92.68).
On the other hand, in the other acyldimethylcopper
intermediate, 11a, the chloride anion serves as a
4
a-5a
Adv. Synth. Catal. 2008, 350, 1063 – 1072
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1067

Naohiko Yoshikai et al.
FULL PAPERS
Scheme 4. Reaction pathway for the substitution of acetyl chloride with [Me CuLi] . Values in parentheses refer to the po-
2
2
À1
tential energy (in kcalmol ) relative to (1 + 3a). The change in energy is shown above the reaction arrows.
À1
Figure 5. Potential energy profile (in kcalmol ) for the substitution of acetyl chloride with [Me CuLi] . Values in italic refer
2
2
À1
to the Gibbs free energy (kcalmol ) at 298 K.
ligand to Cu
A
C
H
T
R
E
U
N
G
(III) (CuÀCl=2.35 Š) to form the CÀCl bond cleavage failed. Thus, approaches of the
square-planar structure, and the oxygen atom lies out Cu atom to the C=O bond and the Li atom to the Cl
of the plane. This structural arrangement is suitable atom result in a spontaneous cleavage of the CÀCl
for the reductive elimination (TS11a-PD), where the C= bond. A theoretical study on an extremely fast organ-
O p orbital can participate in the formation of a CÀC ic reaction tends to result in a very flat potential
[13]
bond.
energy surface, and sometimes even in the absence of
[
17]
Based on the above reaction pathway, we examined a saddle point. Considering that the reaction of the
the effect of solvent coordination to the lithium acid chloride is almost instantaneous, the absence of
atoms using the model complex 1·2S. However, all at- the potential energy barrier for the present case is not
tempts to locate a p-complex or a transition state for very surprising. As discussed in more detail later, we
1068
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1063 – 1072

Mechanism of the Nucleophilic Substitution of Acyl Electrophiles
FULL PAPERS
Figure 6. Structure of intermediates and TS in the reaction of [Me CuLi] and acetyl chloride. The values refer to the bond
2
2
lengths in Š (plain text) and natural charges (underlined).
speculate that the spontaneous collapse of the acid tution reactions, we carried out measurements of the
1
1
2
3
13
chloride is a sign of a late nature of the actual CÀCl
C/ C kinetic isotope effect (KIE) using quantitative
bond cleavage TS.
C NMR analysis (see Supporting Information for ex-
[
16,18]
perimental details).
Thus, the reaction of the thio-
ester 2b and the acid chloride 3b with Me CuLi was
performed to a 93% and 86% conversion, respectively
2
Kinetic Isotope Effect Studies
[
Eq. (1)], and the unreacted starting material was re-
12 13
To obtain further information on the CÀX bond covered from the reaction mixture. The C/ C KIE
13
cleavage (oxidative addition) step in the above substi- data were calculated from a comparison of the C in-
Adv. Synth. Catal. 2008, 350, 1063 – 1072
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1069

Naohiko Yoshikai et al.
FULL PAPERS
tegrations of the starting and the recovered materials.
Note that, in the case of the acid chloride 3b, the re-
action was quenched by the addition of MeOH, and
the unreacted material was recovered as the corre-
12
13
Figure 7. Experimental C/ C KIE values for: (a) the sub-
stitution of thioester 2b and (b) acid chloride 3b with
Me CuLi. The values in the parentheses indicate experimen-
2
sponding methyl ester. The para-methyl carbon atom tal errors in the last digit.
13
was taken as the internal standard for the C NMR
integration.
The experimental KIE values are shown in
Figure 7. In both cases, a significant KIE value was
found on the carbonyl carbon atom (1.034 for 2b and
1
.024 for 3b), while the KIE values on the other
carbon atoms were either negligible or very small
1.001–1.005). Thus, these KIE profiles should reflect
(
the CÀX bond cleavage step as the rate-determining
step of the reaction.
To evaluate the reaction pathways obtained by the
12
13
above theoretical studies, the C/ C KIE values in
the CÀX bond cleavage TS were calculated for both
the model and the real substrates (i.e., 2a/2b and 3a/
3
b). Figure 8 shows the calculated KIE values in the
CÀS bond cleavage step for the model substrate 2a
and the real substrate 2b. The KIE value on the car-
bonyl carbon atom of 2a was ca. 1.050 for both the
non-solvated (TS4a-5a in Figure 8a) and the solvated
1
2
13
Figure 8. (a)–(c) Calculated C/ C KIE values for the CÀS
bond cleavage TS (a=TS4a-5a, b=TS
·2S, and c=TS
4a’-5a
)
4b-5b
(
TS4a’-5a·2S, Figure 8b) models, and much larger than
and (d) structure of TS4b-5b
.
the experimental value (1.034). Note that the similar
KIE values for TS4a-5a and TS4a’-5a·2S would come
from their similar three-centered structure (especially
the CÀS bond length, see Figure 2 and Figure 4). On
the other hand, the corresponding KIE value for the
real substrate 2b (1.034, Figure 8c) was in excellent
agreement with the experimental value (1.034,
Figure 7). The significant difference in the KIE values
for the model and the real substrates reflects the
structural difference in the CÀS bond cleavage TSs
(
TS
vs. TS
: see Figure 2 and Figure 8d). Thus,
4b-5b
4a-5a
TS4b-5b can be regarded as being a much later TS than
TS
as seen from the much more advanced CÀS
4
a-5a
bond cleavage (CÀS bond length=2.13 Š in TS
4
a-5a
and 2.52 Š in TS4 , and the natural charge on the
b-5b
SR group is À0.07 in TS4 and À0.28 in TS4b-5b).
a-5a
Figure 9 shows the calculated KIE values of the CÀ
Cl bond cleavage step for the model substrate 3a
(
TS9a-10a) and the real substrate 3b (TS9b-10b). The KIE
values on the carbonyl carbon were 1.053 and 1.051
for TS9a-10a and TS9b-10b, respectively. The similar mag-
nitude of the KIE values for TS9a-10a and TS9b-10b is in
accordance with their similar structures and charge
distributions (see Figure 6 and Figure 9c). These
1
2
13
Figure 9. (a) and (b) Calculated C/ C KIE values for the
values are much larger than the experimental value CÀCl bond cleavage TS (a=TS
and b=TS9b-10b) and (c)
9
a-10a
(
1.024, Figure 7).
the structure of TS
.
b-10b
9
1070
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1063 – 1072

Mechanism of the Nucleophilic Substitution of Acyl Electrophiles
FULL PAPERS
Scheme 5. Mechanisms of the oxidative addition of acyl CÀS
and CÀCl bonds to organocuprates.
Figure 10. Chemical models used in the computational stud-
ies on the substitution reactions of thioesters and acid chlor-
ides with organocuprates.
The considerable discrepancy between the experi-
mental and the calculated KIE values should be the
problem of the computational side, that is, the insuffi- Experimental Section
cient model and method (e.g., the absence of ethereal
solvent molecule), which should have led to the inap- Computational Methods and Models
propriate estimation of the degree of the CÀCl bond
All the calculations were performed using the Gaussian 03
cleavage. The smaller experimental KIE value than
the calculated one suggests two mechanistic possibili-
ties.
[21]
program package.
employed using the B3LYP hybrid functional. The struc-
tures were optimized using a basis set (denoted as 631SDD)
Density functional theory (DFT) was
[22]
[19]
Thus, the CÀCl bond cleavage in the actual
transition state is much less advanced (i.e., early TS) consisting of the SDD basis set consisting of a double-z va-
or much more advanced (i.e., late TS) than predicted lence basis set and the Stuttgart–Dresden effective core po-
[
23]
[24]
by the calculation. We think that the latter possibility tential (ECP) for Cu, and the 6–31G(d) basis set for
the remaining atoms. The method and the basis sets used
is more likely in light of the following points: (1) The
here are known to give reliable results for the structure and
reaction of the acid chloride with the organocuprate
[25]
reactivity of organocopper reagents.
were adequately characterized using normal coordinate
analysis. Intrinsic reaction coordinate (IRC) analysis was
carried out for the major reaction pathway to confirm that
all the stationary points were smoothly connected to each
other. The kinetic isotope effect (KIE) was calculated using
Stationary points
takes place almost instantaneously, (2) The unsolvat-
ed organocuprate model (1) gives the extremely flat
potential surface of the CÀCl bond cleavage
[26]
(
Figure 5) and the solvated model (1·2 S) effects the
spontaneous collapse of the acid chloride, and (3) the
[27]
potential surface analysis of an extremely fast reac- Bigeleisen-Mayerꢁs equation with the Wigner tunnel cor-
[
28]
tion often underestimates the degree of the bond re- rection using frequencies scaled by a factor of 0.9614. Nat-
[
17]
organization in the actual transition state.
ural population analysis was performed at the same level as
[29]
that used for the geometry optimization.
We employed dimeric lithium dimethyl cuprate
[
Me CuLi] (1) as the model organocuprate, and S-methyl
2
2
Conclusions
thioacetate (2a) and acetyl chloride (3a) as model substrates
Figure 10). The effect of the coordination of the ethereal
(
In summary, we have addressed the mechanisms of nu-
cleophilic substitution reactions to acyl electrophiles,
such as thioesters and acid chlorides, using a lithium
solvent was modeled by a single Me O molecule on the lithi-
um atom (1·2S). The critical step (i.e., the displacement of
the leaving group) of the reaction of S-ethyl p-methylbenzo-
2
organocuprate. Computations and experimental studies thioate (2b) or p-toluoyl chloride (3b) was also studied to
indicate that the reaction of a thioester takes place evaluate the experimental KIE data.
through a three-centered oxidative addition of the CÀS
bond to the Cu atom, while an acid chloride undergoes
eliminative CÀCl bond cleavage through Lewis acid ac-
Acknowledgements
tivation of the Cl atom with the Li atom (Scheme 5).
Such mechanistic differences originate from the nature
of the leaving group. The thioester favors a three-cen-
tered pathway due to the strong affinity of the “soft”
We wish to thank the Ministry of Education, Culture, Sports,
Science and Technology of Japan for financial support
(
Grant-in-Aid for scientific research (No. 18105004 for E.N
thiolate anion to the “soft” copper atom. On the other and No. 19029008 for N.Y.) and Global COE Program for
hand, the “hard” chloride anion prefers to interact Chemistry Innovation).
with the “hard” lithium cation, and making the elimi-
native pathway more feasible. Finally, this study shows
that the dichotomy of the “three-centered” and “elimi-
References
native” pathways is a critical issue in explorations into
2
the mechanism of the oxidative addition of a C
A
H
R
U
G
[
1] Comprehensive Organic Transformations, 2nd edn.,
Ed.: R. C. Larock), Wiley-VCH, Weinheim, 1999.
10
[16]
bond to a d or related transition metal complex,
and in the design of a catalytic cycle involving such an
(
[2] R. K. Dieter, Tetrahedron 1999, 55, 4177–4236.
[3] D. A. Shirley, Org. React. 1958, 8, 28–58.
[
20]
elementary process.
Adv. Synth. Catal. 2008, 350, 1063 – 1072
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1071

Naohiko Yoshikai et al.
FULL PAPERS
[
4] H. Gilman, J. M. Straley, Recl. Trav. Chim. Pays-Bas
[16] N. Yoshikai, E. Nakamura, J. Am. Chem. Soc. 2004,
126, 12264–12265.
1
936, 55, 821.
[
[
5] G. M. Whitesides, C. P. Casey, J. San Filippo Jr, E.
Panek, Trans. N. Y. Acad. Sci. 1967, 29, 572–574.
6] a) G. H. Posner, C. E. Whitten, Tetrahedron Lett. 1970,
[17] a) A. E. Keating, S. R. Merrigan, D. A. Singleton, K. N.
Houk, J. Am. Chem. Soc. 1999, 121, 3933–3938;
b) D. T. Nowlan III, D. A. Singleton, J. Am. Chem. Soc.
2005, 127, 6190–6191.
5
3, 4647–4650; b) C. Jallabert, N. T. Luong-Thi, H. Riv-
iere, Bull. Soc. Chim. Fr. 1970, 797–798; c) G. H.
Posner, C. E. Whitten, P. McFarland, J. Am. Chem. Soc.
[18] a) D. A. Singleton, A. A. Thomas, J. Am. Chem. Soc.
1995, 117, 9357–9358; b) D. E. Frantz, D. A. Singleton,
J. P. Snyder, J. Am. Chem. Soc. 1997, 119, 3383–3384.
1
972, 94, 5106–5108.
[
[
[
19] L. C. S. Melandar, W. H. Saunders, Reaction Rates of
[
7] a) S. Masamune, M. Hirama, S. Mori, S. A. Ali, D. S.
Garvey, J. Am. Chem. Soc. 1981, 103, 1568–1571;
b) P. R. McGuirk, D. B. Collum, J. Am. Chem. Soc.
Isotopic Molecules, Wiley, New York, 1980.
20] N. Yoshikai, H. Mashima, E. Nakamura, J. Am. Chem.
Soc. 2005, 127, 17978–17979.
1
982, 104, 4496–4497; c) D. V. Patel, F. VanMiddles-
21] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuse-
ria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery,
Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam,
S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M.
Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Na-
katsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Po-
melli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A.
Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski,
S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B.
Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford,
J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith,
M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian 03,
Revision C.02, Gaussian, Inc., Wallingford CT, 2004.
22] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652;
b) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37,
worth, J. Donaubauer, P. Gannett, C. J. Sih, J. Am.
Chem. Soc. 1986, 108, 4603–4614.
[
[
8] R. J. Anderson, C. A. Henrick, L. D. Rosenblum, J.
Am. Chem. Soc. 1974, 96, 3654–3655.
9] Metal-catalyzed Cross-couplingReactions , 2nd edn.,
(Ed.: A. deMeijere, F. Diederich), Wiley-VCH, New
York, 2004.
[
10] For example: a) H. Tokuyama, S. Yokoshima, T. Yama-
shita, T. Fukuyama, Tetrahedron Lett. 1998, 39, 3189–
3
192; b) Y. Zhang, T. Rovis, J. Am. Chem. Soc. 2004,
1
26, 15964–15965; c) H. Yang, H. Li, R. Wittenberg,
M. Egi, W. Huang, L. S. Liebeskind, J. Am. Chem. Soc.
007, 129, 1132–1140; d) J. M. Villalobos, J. Srogl, L. S.
2
Liebeskind, J. Am. Chem. Soc. 2007, 129, 15734–15735.
11] a) W. Zhang, R. G. Carter, A. F. T. Yokochi, J. Org.
Chem. 2004, 69, 2569–2572; b) W. Zhang, R. G. Carter,
Org. Lett. 2005, 7, 4209–4212.
[
[
12] a) E. Nakamura, S. Mori, Angew. Chem. 2000, 112,
3
3
902–3924; Angew. Chem. Int. Ed. 2000, 39, 3750–
771; b) E. Nakamura, N. Yoshikai, Bull. Chem. Soc.
Jpn. 2004, 77, 1–11, and references cited therein.
13] a) E. Nakamura, S. Mori, M. Nakamura, K. Morokuma,
J. Am. Chem. Soc. 1997, 119, 4887; b) E. Nakamura, S.
Mori, K. Morokuma, J. Am. Chem. Soc. 1997, 119,
[
[
7
85–789.
23] M. Dolg, U. Wedig, H. Stoll, H. Preuss, J. Chem. Phys.
987, 86, 866–872.
24] W. J. Hehre, L. Radom, P. v R. Schleyer, J. A. Pople,
Ab Initio Molecular Orbital Theory; John Wiley &
Sons, Inc., New York, 1986, and references cited there-
in.
[
[
4
900; c) E. Nakamura, S. Mori, K. Morokuma, J. Am.
1
Chem. Soc. 1998, 120, 8273; d) S. Mori, E. Nakamura,
Chem. Eur. J. 1999, 5, 1534; e) E. Nakamura, M. Yama-
naka, J. Am. Chem. Soc. 1999, 121, 8941–8942; f) S.
Mori, E. Nakamura, K. Morokuma, J. Am. Chem. Soc.
2
000, 122, 7294; g) E. Nakamura, M. Yamanaka, N.
Yoshikai, S. Mori, Angew. Chem. 2001, 113, 1989–
992; Angew. Chem. Int. Ed. 2001, 40, 1935–1938;
h) M. Yamanaka, E. Nakamura, Organometallics 2001,
0, 5675; i) M. Yamanaka, S. Kato, E. Nakamura, J.
[
25] M. Yamanaka, A. Inagaki, E. Nakamura, J. Comput.
Chem. 2003, 24, 1401–1409.
26] a) K. Fukui, Acc. Chem. Res. 1981, 14, 363–368; b) C.
Gonzalez, H. B. Schlegel, J. Chem. Phys. 1989, 90,
1
[
2
2
154–2161; c) C. Gonzalez, H. B. Schlegel, J. Phys.
Am. Chem. Soc. 2004, 126, 6287–6293; j) M. Yamana-
ka, E. Nakamura, J. Am. Chem. Soc. 2005, 127, 4697–
Chem. 1990, 94, 5523–5527.
27] a) J. Bigeleisen, M. G. Mayer, J. Chem. Phys. 1947, 15,
[
4
706.
2
61–267; b) J. Bigeleisen, M. Wolfsberg, Adv. Chem.
[
14] For recent literature: a) S. H. Bertz, S. Cope, D.
Phys. 1958, 1, 15–76.
Dorton, M. Murphy, C. A. Ogle, Angew. Chem. 2007,
[28] A. P. Scott, L. Radom, J. Phys. Chem. 1996, 100,
16502–16513.
1
19, 7212–7215; Angew. Chem. Int. Ed. 2007, 46, 7082–
7
085; b) S. H. Bertz, S. Cope, M. Murphy, C. A. Ogle,
[29] a) A. E. Reed, R. B. Weinstock, F. Weinhold, J. Chem.
Phys. 1985, 83, 735–746; b) A. E. Reed, L. A. Curtiss,
F. Weinhold, Chem. Rev. 1988, 88, 899–926; c) NBO
Version 3.1in the Gaussian 03 package implemented
by E. D. Glendening, A. E. Reed, J. E. Carpenter, F.
Weinhold, University of Wisconsin: Madison, WI, 1990.
B. J. Taylor, J. Am. Chem. Soc. 2007, 129, 7208–7209;
c) T. Gärtner, W. Henze, R. M. Gschwind, J. Am.
Chem. Soc. 2007, 129, 11362–11363.
[
15] E. J. Corey, G. H. Posner, J. Am. Chem. Soc. 1967, 89,
3
911.
1072
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1063 – 1072