An insight into copper catalyzed allylation of alkyl zinc halides. Comparison of reactivity profiles for catalytic and stoichiometric alkylzinc-copper reagents

Article abstract of DOI:10.1016/j.jorganchem.2009.01.026

The γ-selective allylation of catalytic and stoichiometric alkylzinc-cuprates have been kinetically studied. The reactivity profiles generated by allylation reactions of n-butylzinc chloride catalyzed by CuX compounds (X = I, Br, Cl, CN, SCN) and also cat

Full text of DOI:10.1016/j.jorganchem.2009.01.026

Journal of Organometallic Chemistry 694 (2009) 1890–1897
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
An insight into copper catalyzed allylation of alkyl zinc halides. Comparison
of reactivity profiles for catalytic and stoichiometric alkylzinc–copper reagents
*
Ender Erdik , Melike Koçog˘lu
Ankara University, Science Faculty, Beßsevler, Ankara 06100, Turkey
a r t i c l e i n f o
a b s t r a c t
Article history:
The c-selective allylation of catalytic and stoichiometric alkylzinc–cuprates have been kinetically stud-
Received 20 October 2008
Received in revised form 19 December 2008
Accepted 16 January 2009
Available online 23 January 2009
ied. The reactivity profiles generated by allylation reactions of n-butylzinc chloride catalyzed by CuX
compounds (X = I, Br, Cl, CN, SCN) and also catalyzed by n-butylzinc–copper reagents and di n-butyl-
zinc–copper reagents were evaluated. Reactivity profiles for allylation of stoichiometric n-butylzinc–cop-
per reagents and di n-butylzinc–copper reagents were also prepared. All CuX compounds have been
screened for the preparation of Grignard reagent derived n-butylzinc–copper reagents and di n-butyl-
zinc–copper reagents.
Keywords:
Allylic coupling
The evaluation of the profiles indicates that the active catalyst might be RCu(X)ZnCl and also to some
Alkylzinc halides
Reactivity profiles
Kinetics of allylation
Alkylzinc–copper reagents
Catalytic allylation
degree, R2CuZnCl ꢀ ZnClX, which both could favor formation of
elimination of -allyl Cu (III) complex formed at vinylic terminal to give
slow isomerization to -allyl Cu (III) complex formed at allylic terminal to give
allylation mechanism of zinc cuprates, the role of counter ion, ZnCl+ has been discussed.
Ó 2009 Elsevier B.V. All rights reserved.
c
-product. All data supports the reductive
-allylated product with a quite
-allylated product. In the
r
c
r
a
1. Introduction
was reported to be
cuprates and alkyl cyanozinc cuprates was also reported to take
place with high -selectivity [29,34–51].
The mechanism of copper catalyzed allylation of alkyl Grignard
reagents and also dialkylcuprates derived from alkyllithium and
Grignard reagents has been discussed in a number of reports.
According to the mechanism proposed by Goering and co-workers
[52–55] and confirmed by Backvall et al. [21] (Scheme 2), initial
complexation of RCu(X)MgBr to the allylic double bond gives an al-
kene-Cu(I) complex A and an oxidative addition anti to the leaving
c-selective [25,29–33]. Allylation of dialkyl zinc
Transition metal catalyzed or mediated allytic alkylation is a
highly useful synthetic methodology for C–C bond formation.
[1–3] In recent years, controlling the regio- and stereochemistry
of alkyl–allyl coupling using allylic substrates has received much
attention. Several methods have been developed for regioselective
allylation of alkyl Grignard [4–8] and alkylzinc reagents [9–11] un-
der copper catalysis or stoichiometric cuprates derived from these
reagents (Scheme 1) [12,13].
c
The regiochemistry in the copper catalyzed allylation of alkyl
Grignard reagents is controlled by the structure of the Grignard re-
agent, the allylic substrate, copper catalyst and the solvent as well
as reaction conditions. [14–21] It is commonly accepted that the
reaction parameters governing formation of a dialkylcuprate as
group takes place at vinylic terminal. The
intermediate B thus formed can undergo reductive elimination to
give -product or isomerize to the -allyl-Cu(III) complex interme-
diate D formed at allylic terminal, presumably via a -allyl-Cu(III)
complex intermediate C. Reductive elimination of D would give the
-product. The overall regiochemistry of the allylation is deter-
mined by the relative rates of elimination and isomerization of
the complex B. If X group on cuprate is an electron withdrawing
group (X = halide, CN), reductive elimination is fast relative to
r-allyl-Cu(III) complex
c
r
p
an intermediate give
governing formation of an alkylcopper as an intermediate give
-substitution predominantly [20,21]. In the light of the reported
a
-substitution whereas reaction conditions
a
c
results, the regiochemical outcome of the allylation of stoichiome-
tric alkylcopper and dialkylcuprate reagents are mainly consistent
with those of copper catalyzed alkyl Grignard and – lithium
reagents [22–27].
However, the regiochemistry in the uncatalyzed [24,28] and
copper catalyzed allylation of alkylzinc halides and dialkylzincs
isomerization and
kyl group, isomerization predominates over reductive elimination
and elimination of D gives -substitution. Reductive elimination
of -allyl-Cu(III) complexes B and D is stereospecific and occurs
c-substitution is observed. In the case of X = al-
a
r
with retention of configuration.
Recently, Nakamura reported [56] a study showing that forma-
tion of
a-product occurs directly from the
p-allyl-Cu(III) complex
intermediate C via an enyl [
3, pathway a).
r + p]-type transition state (Scheme
* Corresponding author. Tel.: +90 312 212 67 20/1031; fax: +90 312 223 23 95.
E-mail address: erdik@science.ankara.edu.tr (E. Erdik).
0022-328X/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2009.01.026

E. Erdik, M. Koçog˘lu / Journal of Organometallic Chemistry 694 (2009) 1890–1897
1891
Scheme 1.
Scheme 2.
Nakamura also reported [56] that Lewis acid favors anti and
-selective reaction. These reactions may occur directly via an enyl
]-type complex intermediate E without going through a
-allyl-Cu(III) complex intermediate C (Scheme 3, pathway b).
Nakamura, recently used density functional calculations to study
the origin of the contrasting regioselectivities in allylic substitution
of MeCu(CN)Li and Me2CuLi. [57] Bertz and Ogle have been able to
2. Results and discussion
c
[
r + p
In our long term investigation on the reactions of homo and
mixed organozinc reagents, we found that the copper catalyzed
allylation of n-butylmagnesium bromide 1 derived n-butylzinc
chloride 2 and di n-butylzinc 3 with E-crotyl chloride 4 in THF, took
p
place with an a:c ratio not higher than 12:88 (Scheme 4, pathway
prepare and characterize the first examples of both
r-allyl- and p-
a) [60]. We also observed that reaction temperature, addition mode
of the allylic substrate, CuX catalyst and its concentration and
N-donor and O-donor solvents such as TMEDA, HMPA and diglyme
allyl-Cu(III) complexes and to study their reactions by using rapid
injection NMR spectroscopy [58].
However, a mechanism for copper catalyzed allylation of dial-
kylzincs has been discussed only in one report [24].
as cosolvents did not change the
c-regioselectivity. We also ob-
tained similar results in the copper catalyzed allylation of organo-
zinc reagents 2 and 3 prepared by in situ transmetallation of
n-butylmagnesium bromide 1 (Scheme 4, pathway b). Stoichiome-
tric n-butylcopper n-BuCu(X)ZnCl (8) and di n-butylcuprate n-Bu2-
CuZnCl (9) reagents prepared from n-butylzinc chloride (2) and
Our interest in controlling the regiochemistry of allylation of
alkylmagnesium-copper reagents [59] and alkylzinc–copper re-
agents [60,61] by changing the reaction conditions prompted us
to carry out a kinetic study for the mechanism of allylic coupling
of alkylzinc–copper reagents. We evaluated reactivity profiles gen-
erated by allylation reactions of catalytic and stoichiometric mono-
and di n-butylzinc–copper reagents prepared using various Cu(I)
compounds.
CuX compounds 5 also gave
c-allylated products with an a:c ratio
of 10:90 in their reaction with E-crotyl chloride 4 in THF.
If we assume that the allylation of alkylzinc–copper reagents
may take place according to the mainly accepted mechanism
(Scheme 2, M+ = ZnCl+) proposed by Goering and co-workers [52–
55] and Backvall et al. [21] for allylation of alkylmagnesium-copper
reagents, then we should find support for the following points:
Here we report our results on the alkylzinc–copper nucleophiles
as active catalysts in the copper catalyzed
alkylzinc halides.
c-selective allylation of

1892
E. Erdik, M. Koçog˘lu / Journal of Organometallic Chemistry 694 (2009) 1890–1897
Scheme 3.
Scheme 4.
1. In the Cu catalyzed
c
-allylation of alkylzinc reagents, reductive
-allyl-Cu(III)
-allyl-Cu(III) complex D. Then, X group
3. In the allylation of zinc cuprates, it may be probable that isom-
elimination seems to occur predominantly from
complex B rather than
r
erization of complex B to D via C may not be slow compared to
reductive elimination in the case of X = R; however complex D
r
on the zinc cuprate seems to be mainly halide, CN or SCN rather
than alkyl, or in other words RCu(X)ZnCl rather than R2CuZnCl
seems to form as catalytic species.
may not form and complex C is expected to give c-allylated
product in some way.
2. However, stoichiometric alkylcopper-zinc reagents, i.e. not only
RCu(X)ZnCl (X = halide, CN, SCN), but also R2CuZnCl, which is
This study was undertaken to critically examine this assump-
tion, i.e. in the Cu catalyzed allylation of alkylzinc reagents, cuprate
nucleophile species seems to be RCu(X)ꢁ or R2Cuꢁ which are both
expected to give
products. This indicates that in the allylation of zinc cuprates,
isomerization rate of -allyl-Cu(III) complex B to D does not
a-allylated product, afford mainly c-allylated
responsible for c-substitution.
r
For this purpose, we carried out a kinetic study and we com-
pared the reactivity profiles [62,63] for the allylation of n-butylzinc
chloride, n-BuZnCl (2) with E-crotyl chloride (4) catalyzed by CuX
predominate over reductive elimination even in the case of
X = R on the contrary to the allylation of magnesium cuprates.

E. Erdik, M. Koçog˘lu / Journal of Organometallic Chemistry 694 (2009) 1890–1897
1893
Scheme 5.
(5). We also prepared the reactivity profiles for allylation of stoi-
chiometric n-butylcopper n-BuCu(X)ZnCl (8) and di n-butylcuprate
n-Bu2CuZnCl (9). In order to gain some insight into the catalytic
species, we used these preformed n-butylcopper reagents 8 and 9
also as catalysts in the allylation of n-BuZnCl (2). The list of the
reactions in the kinetic study are given in Scheme 5. CuX com-
pounds (X = I, Br, Cl, CN, SCN) 5a–e were used as catalysts in the
allylation of n-butylzinc chloride (2) (Reactions 1a–e). n-Butyl-
zinc–copper reagents, i.e. n-BuCu(X)ZnCl (8a–e) and n-Bu2-
CuZnCl ꢀ ZnClX (9a–e) were prepared using CuX compounds and
used as catalysts (Reaction 2a–e and Reaction 3a–e, respectively)
and as stoichiometric reagents to be allylated (Reaction 4a–e and
Reaction 5a–e).
(ii) A similarity between the profiles of reaction 1 and reac-
tion 3 would reveal that the catalytic alkylcopper inter-
mediate might be R2Cuꢁ rather than RCu(X)ꢁ. However,
this result also indicates that R2Cuꢁ undergoes to reduc-
tive elimination in order to give
c-substitution although
it is expected to undergo isomerization and give
a-
substitution.
(iii) In the presence of excess n-BuZnCl (2), the catalyst
n-BuCu(X)ZnCl (8) may not have time to undergo allylation
(reaction 2); but reacts faster with n-BuZnCl (2) to form
n-Bu2CuZnCl (9) as catalytic intermediate for allylation
(reaction 3) (1). In this case the catalytic species in reaction
2 and reaction 3 are not different and the reactivity profiles
for these reactions should be similar
For reactivity profiles, the progress of allylation reactions 1–5
was followed at room temperature in time. Instead of conver-
sion-time profiles, we prepared
profiles since the relative amount of
90% in reactions 1–5, is almost constant during the reaction.
The amount of -product was monitored by GC analysis. How-
c
-allylated product-time, [R-
c]-t
c
-product, which is 80–
ð1Þ
c
(iv) An agreement is expected between the profiles of reaction 3
and 5 since the active n-butylzinc–copper reagents to be
allylated are same.
ever, due to the heterogeneous reaction, the reproducibility for
the GC analysis of samples taken from the reaction mixture con-
taining the internal standard seemed low. So, a definite number
of reactions were carried out in different flasks and each reaction
was quenched at appropriate times. This method, although time
consuming led to accurate and reproducible results in GC analy-
sis. For catalytic reactions, 1.1 mmoles of n-BuZnCl (2) was al-
lowed to react with 1 mmole of E-crotyl chloride (4) in the
presence of 0.20 mmoles of CuX catalyst 5 in THF at room tem-
perature. For stoichiometric reactions, 1.1 mmoles of n-Bu-
Cu(X)ZnCl (8) or n-Bu2CuZnCl (9) was allowed to react with
1 mmole of E-crotyl chloride in THF at room temperature. The
reactions are rather fast and according to the kinetic results, a
reaction time of 60 min was chosen to guarantee complete ally-
First, we compared the reactivity profiles for allylation reac-
tions 1a–5a, 1b–5b, 1c–5c, 1d–5d and 1e–5e. In Table 1, we gave
a list of reactions which generate similar profiles.
Before evaluating the superimpossibility and similarity of the
profiles, we also determined some numerical values for the cat-
alytic efficiency of n-butylzinc–copper reagents at 20 mol% Cu(I)
catalyst (5, 8 or 9) loading in reactions 1–3, respectively. We
used the kinetic data for this purpose. The determined values
correspond to the ratio of the maximum% yield of
c-product
to the reaction time (min) in which this yield was obtained (y
values). We calculated this ratio for also stoichiometric n-butyl-
zinc–copper reagents (z values) in order to make a comparison
between their reactivities in reactions 4 and 5. y Values may
be regarded as TOF values for allylation of catalytic n-butyl-
zinc–copper reagents and we think that y and z values can be
also used to get an idea for relative reactivity of catalytic and
stoichiometric n-butylzinc–copper reagents. y Values were found
to be 1.2–1.6 and z values were found to be 1.3–1.8, except 0.7
for the allylation of n-BuCu(CN)ZnCl (4d). The error on these
values does not exceed 10%.
lation for all regents. The amount of
c-product values used in
reactivity profiles are an average of at least three parallel reac-
tions. The reactivity profiles for reactions 1–5 were given in
Fig. 1 and they were collected to compare the allylation reactiv-
ity of catalytic and stoichiometric n-butylzinc–copper reagents
prepared using the same CuX compound. The reactivity profiles
for the allylation of n-butylzinc–copper reagents prepared using
CuI (5a), CuBr (5b), CuCl (5c), CuCN (5d) and CuSCN (5e) were
given in Fig. 1a–e, respectively.
In the kinetic study we reasoned that one or more of the follow-
ing results should be obtained from the comparison of the reactiv-
ity profiles:
We also listed the allylation reactions of catalytic and stoichi-
ometric n-butylzinc–copper reagents which have equal y and z val-
ues (Table 2).
(i) As n-BuCu(X)ꢁ is the catalytic alkylcopper intermediate
As seen, the similarity of the reactivity profiles (given in Table
1) and the maximum c-product yield to time ratios (y and z values)
which is assumed to be responsible for
c-substitution in
(given in Table 2) are mostly in agreement.
the Cu catalyzed allylation of alkylzinc reagents according
to Scheme 2, the reactivity profiles of reaction 1 should be
similar to those of reaction 2 and possibly those of reaction
4 for each CuX a–e.
Below, the similarities and differences in the reactivity profiles
of allylation reactions 1–5 were evaluated for each CuX compound
5a–e (X = I a, Br b, Cl c, CN d, SCN e).

1894
E. Erdik, M. Koçog˘lu / Journal of Organometallic Chemistry 694 (2009) 1890–1897
a
b
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
1
5
0.9
3
2
1
0.8
5
3
2
1
4
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
4
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
tmin
tmin
c
d
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
5
5
1
2
3
3
2
1
4
4
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
tmin
tmin
e
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
5
3
2
1
4
0
10
20
30
40
50
60
70
tmin
Fig. 1. Kinetic study (
room temperature. (The amount of
c
-allylated product-time curves) for the allylation reaction of catalytic and stoichiometric n-butylzinc–copper reagents with E-crotyl chloride in THF at
-allylated product (mmol) is given versus time (min) in the allylation with 1 mmol of E-crotyl chloride 4.) (The reagents for the profiles:
c
1. s, 2. N, 3. j, 4. ꢀ, 5. d). The reagents to be allylated: 1. n-BuZnCl (2)/20 mol% CuX (5); 2. n-BuZnCl (2)/20 mol% n-BuCu(X)ZnCl (8); 3. n-BuZnCl (2)/20 mol% n-
Bu2CuZnCl ꢀ ZnClX (9); 4. n-BuCu(X)ZnCl (8); 5. n-Bu2CuZnCl ꢀ ZnClX (9a). (a) Copper source is CuI (5a), (b) copper source is CuBr (5b), (c) copper source is CuCl (5c), (d) copper
source is CuCN (5d), (e) copper source is CuSCN (5e).
2.1. Reactions 1a–5a
2a and 3a) but a different profile with CuI (5a) catalysis (reaction
1a). y Values also support the similarity of active catalysts in reac-
tions 2a and 3a and may provide a support for formation of 8a as
active catalyst from precatalyst CuI in catalytic allylation.
The profiles generated by allylation of n-BuZnCl (2) with the
catalysis of n-BuCu(I)ZnCl (8a) and n-Bu2CuZnCl (9a) (reactions
2a and 3a, respectively) are quite similar. This indicates that the
same catalytic species are present in these reactions. However,
the difference in the profile with the catalysis of CuI (5a) (reaction
1a) possibly originates from the difference in the ease of formation
of the catalytic species, and this profile agrees well with the profile
of the stoichiometric allylation of n-BuCu(I)ZnCl (8a) (reaction 4a).
This indicates that the active catalyst, formed from the precatalyst
CuI (5a) is possibly n-BuCu(I)ZnCl (8a) species. Then, in reaction
2a, n-BuCu(I)ZnCl (8a) catalyst seems to react first with excess
n-BuZnCl (2) to give n-Bu2ZnCl (9a) as catalytic species rather than
allylation. This explanation supports the observation of similar
profiles in the allylation with 8a and 9a as catalysts (reactions
2.2. Reactions 1b–5b
The results are in well accordance with the results outlined in
Section 2.1.
2.3. Reactions 1c–5c
The superimpossible profiles obtained from the allylation of
n-BuZnCl (2) with the catalysis of CuCl (reaction 1c) and
n-BuCu(Cl)ZnCl (8c) (reaction 2c) indicates that the active catalyst
is 8c.

E. Erdik, M. Koçog˘lu / Journal of Organometallic Chemistry 694 (2009) 1890–1897
1895
Table 1
2.4. Reactions 1d–5d
Allylation reactions of catalytica and stoichiometric n-butylzinc–copper reagentsb
which generate superimpossible and similar reactivity profiles (reactions are given in
Scheme 5).
The profiles generated by allylation of n-BuZnCl (2) with the
catalysis of CuCN (5d) (reaction 1d), n-BuCu(CN)ZnCl (8d) (reac-
tion 2d) and also n-Bu2Cu(CN)(ZnCl)2 (9d) (reaction 3d) are super-
impossible indicating that the active catalysts might be the same. A
larger deviation of the profile generated by the stoichiometric ally-
lation of n-BuCu(CN)ZnCl (8d) (reaction 4d) can also be accounted
for by the different structure of the catalytic cuprate species. Then,
we can conclude that the active catalyst might not be
n-BuCu(CN)ZnCl (8d), but can be n-Bu2Cu(CN)(ZnCl)2 (9d). How-
ever, Nakamura and Yoshikai demonstrated [64,65] that reaction
of 2 equiv. of RLi with 1 equiv. of CuCN does not give R2Cu(CN)Li2
(Lipshutz cuprates) and these higher cyanocuprates exist as R2Cu-
Li ꢀ LiCN (cyano-Gilman cuprates). Then the large discrepancy be-
tween the profiles generated by the allylation of stoichiometric
CuX
Reactions which generate superimposible (or similar) reactivity
profiles
CuI (5a)
CuBr (5b)
CuCl (5c)
CuCN (5d)
CuSCN
2 and 3c; 1 and 4
2 and 3; 1 and 4
1 and 2
1, 2 and 3
1 and 4c; 3 and 5
(5e)
a
Reactions 1, 2 and 3 were carried out by reacting n-BuZnCl (2) (1.1 mmoles)
with E-crotyl chloride (4) (1 mmole) in the presence of CuX (5) (0.20 mmoles),
n-BuCu(X)ZnCl (8) (0.20 mmoles) and n-Bu2CuZnCl (9) (0.20 mmoles), respectively
in THF at room temperature.
b
Reactions
4 and 5 were carried out by reacting n-BuCu(X)ZnCl (8) or
n-Bu2CuZnCl (9) (1.1 mmoles) with E-crotyl chloride (4) (1 mmole) in THF at room
cuprates,
n-BuCu(CN)ZnCl
(8d)
(reaction
4d)
and
temperature.
c
The profiles are not superimposible, but quite similar.
n-Bu2Cu(CN)(ZnCl)2 (9d) (reaction 5d) supports the view that CN
group is not resting on copper in 9d and 9d may exist as
n-Bu2CuZnCl ꢀ ZnCl(CN) [16]. The equality of y values for catalytic
allylation reactions (reactions 1d, 2d and 3d) to z value for stoichi-
ometric allylation of n-Bu2Cu(CN)(ZnCl)2 reacting as n-Bu2CuZnCl
Table 2
Allylation reactions of catalytic and stoichiometric n-butylzinc–copper reagents
which yield equal values for maximum
values) (reactions are given in Scheme 5).
c
-product yield to time ratios (ya and zb
(9d) (reaction 5d) rather than
z value for allylation of
Reactions which give equal ya and zb valuesc
n-BuCu(CN)ZnCl (reaction 4d) also provides another support for
formation of n-Bu2CuZnCl type cuprate from CuCN (reaction 3d)
rather than n-BuCu(CN)ZnCl (reaction 2d) as active catalyst.
CuX
CuI (5a)
1 and 2c; 2 and 3; 1 and 4
1 and 2c; 2 and 3; 1 and 4
1 and 2; 2 and 3c; 1 and 4c
1, 2 and 3; 1, 2, 3 and 5c
1 and 2c; 2 and 3c; 3 and 5
CuBr (5b)
CuCl (5c)
CuCN (5d)
CuSCN (5e)
2.5. Reactions 1e–5e
a
The difference between the profiles obtained from the allylation
of n-BuZnCl (2) with the catalysis of CuSCN (5e) (reaction 1e),
n-BuCu(SCN)ZnCl (8e) (reaction 2e) and n-Bu2Cu(SCN)(ZnCl)2
(9e) (reactions 3d) indicates the difference between the structure
of active catalyst in each reaction. However, the similarity between
the profiles generated by allylation of n-BuZnCl (2) with the catal-
ysis of CuSCN (5e) (reaction 1e) and stoichiometric allylation of
y = (max. %yield of
c
-product)/(reaction time, min.) for allylation of n-BuZnCl
(2) in the presence of 20 mol% catalyst (CuX (5), n-BuCu(X)ZnCl (8) or n-Bu2CuZnCl
(9)).
b
z = (max. %yield of
c-product)/(reaction time, min.) for allylation of
n-BuCu(X)ZnCl (8) and n-Bu2CuZnCl (9).
c
The values are equal in the error limit of 10%.
Scheme 6.

1896
E. Erdik, M. Koçog˘lu / Journal of Organometallic Chemistry 694 (2009) 1890–1897
n-BuCu(SCN)ZnCl (8e) (reaction 4e) may indicate the formation of
8e, as catalytic species from precatalyst CuSCN (5e).
Earlier studies by Lipshutz on the use of CuSCN for the forma-
tion lithium cuprates [66] also showed that while CuI serves as a
precursor to lower order cuprates R2CuLi, CuSCN may actually be
forming a higher order mixed species R2Cu(SCN)Li2.
The similarity between the profiles generated by allylation of
n-BuZnCl (2) with the catalysis of 9e (reaction 3e) and stoichiom-
etric allylation of 9e (reaction 5e) can be reasonably explained
niques [67]. Quantitative GC analysis were performed on a Thermo
Finnigan gas chromatograph equipped with a ZB-5 capillary col-
umn packed with phenyl-polysiloxane using internal standard
technique. THF was distilled from sodium benzophenone dianion;
n-butyl bromide and E-crotyl chloride were obtained commercially
and purified using literature procedures. Mg turnings for Grignard
reagents was used without purification. ZnCl2 (Aldrich) was dried
under reduced pressure and used as a THF solution prepared prior
to use. CuI [68], CuBr [69], CuCl [69], CuCN [70] and CuSCN [68]
were purified according to the published procedures. n-Butylmag-
nesium bromide n-BuMgBr (1) was prepared in THF according to
standard procedure and its concentration was found by titration
prior to use [71]. n-Butylzinc chloride, n-BuZnCl (2) was prepared
by dropwise addition of 1 mol equiv. of n-BuMgBr (1) in THF to a
solution of 1 mol equiv. of ZnCl2 in THF at 0 °C and stirring at that
temperature for 20 min. n-Butylcopper reagents, n-BuCu(X)ZnCl
(X = I, Br, Cl, CN, SCN) 8a–e and chlorozinc di n-butylcuprate re-
agents, n-Bu2CuZnCl ꢀ ZnClX 9a–e were prepared by addition of
1 mol equiv. of CuX to 1 mol or 2 mol equiv. of freshly prepared
n-BuZnCl (2), respectively in THF at 0 °C and stirring at that tem-
perature for 20 min [10,11,13].
by the formation of
a
higher order mixed cuprate
n-Bu2Cu(SCN)(ZnCl)2 (9e) from n-BuZnCl (2)/CuSCN (5e) and at
least suggest that to some degree there are differences between
CuCN- and CuSCN-derived higher order cuprates.
In summary, (i) the reactivity profiles generated by the allyla-
tion of n-BuZnCl (2) with the catalysis of CuX (reaction 1) are
quite similar to those generated by the allylation of stoichiometric
n-BuCu(X)ZnCl (8) (reaction 4) in the case of X = I, Br and SCN. In
the case of X = Cl, the similarity of the profiles generated by the
allylation of n-BuZnCl (2) with the catalysis of CuX (reaction 1)
and with the catalysis of n-BuCu(X)ZnCl (8) (reaction 2) is ob-
served. Then we can conclude that the active catalyst seems to
be RCu(X)ZnCl (8) which would favor formation of c-product.
(ii) The similarity of the profiles generated by the allylation of
n-BuZnCl (2) with the catalysis of n-BuCu(X)ZnCl (8) (reaction 2)
or n-Bu2CuZnCl (9) (reaction 3) in the case of X = I, Br and CN indi-
cates that the active catalyst might be also n-Bu2CuZnCl (8) formed
from the reaction of n-BuCu(X)ZnCl (8) with excess n-BuZnCl (2).
Then, not only n-BuCu(X)ZnCl (8), but also n-Bu2CuZnCl (9) is ex-
3.2. Procedure for kinetic study of catalytic and stoichiometric
allylation of n-butylzinc–copper reagents
CuX (X = I, Br, Cl, CN, SCN) (5) catalyzed allylation of n-BuZnCl 2
with E-crotyl chloride 4.
In a flame-dried and two-necked flask equipped with septum
caps and a stirring bar, n-BuZnCl 2 (1.1 mmol) was prepared from
n-BuMgBr (1) (1.1 mmol in 1.1 ml of THF) and ZnCl2 (1.1 mmol in
1.7 ml of THF) at 0 °C. CuX (5) (0.2 mmol) was added at 0 °C and
E-crotyl chloride (4) (1 mmol, 0.1 ml) was added at room temper-
ature. A number of reactions was carried out in different flasks and
each reaction was stirred at room temperature for appropriate
time, i.e. 2, 4, 6, 8, 10, 15, 20, 30, 40, 50 and 60 min. Each reaction
was quenched with aqueous NH4Cl solution. The aqueous phase
was extracted with Et2O. After evaporation of the solvent to a con-
venient volume, internal standard was added and aliquots were
analyzed by GC.
pected to give c-allylation.
(iii) However, in the case of X = CN, the dissimilarity of the pro-
files generated by the allylation of n-BuZnCl (2) with the catalysis
of CuCN (5d) (reaction 1) or n-BuCu(CN)ZnCl (8d) (reaction 2)
with the profile generated by the allylation of stoichiometric
n-BuCu(CN)ZnCl (reaction 4) indicates that the active catalyst
might be n-Bu2CuZnCl.
According to these findings, our assumption that the allylation
of alkylzinc–copper reagents takes place according to the mecha-
nism outlined in Scheme 2 seems correct. We obtained all of the
results that we expected from the reactivity profiles and we found
support to propose the reductive elimination of
complex B to give -allylated product in the case of X = halide,
CN, SCN or alkyl with a quite slow isomerization of complex B to
complex D to give -allylated product in the case of X = alkyl, e.g.
r
-allyl-Cu(III)
In the reactions catalyzed by n-BuCu(X)ZnCl (8) and
n-Bu2CuZnCl ꢀ ZnClX (9) the same procedure was applied. Except,
these reagents were firstly prepared before the reaction.
n-BuCu(X)ZnCl (8) was prepared from n-BuZnCl (2) [0.2 mmol, pre-
pared from n-BuMgBr (1) (0.2 mmol in 0.2 ml of THF) and ZnCl2
(0.2 mmol in 0.3 ml of THF) at 0 °C] and CuX (0.2 mmol) at 0 °C.
Secondly, n-BuZnCl 2 (1.1 mmol in 2.8 ml THF) was prepared at
0 °C and added to freshly prepared n-BuCu(X)ZnCl (8). E-Crotyl
chloride (4) (1 mmol, 0.1 ml) was added at room temperature.
For the preparation of n-Bu2CuZnCl ꢀ ZnClX (9) as a catalyst,
n-BuZnCl (2) (0.4 mmol in 1 ml of THF) and CuX (0.2 mmol) were
used.
c
a
k2 ꢂ k1. So, in the allylation of alkylzinc–copper reagents, the
counter ion ZnCl+ is expected to be responsible for much slower
isomerization of complex B resulting in
c-allylation of both
RCu(X)ZnCl and R2CuZnCl (Scheme 6). We may think that zinc cup-
rates possibly can not react as solvent separated ion pairs [64] due
to the less polar character of Cu–ZnCl bond compared to Cu–MgBr
bond. Then, ZnCld+, which is a strong Lewis acid can possibly de-
crease
r
-donor character of R group, which is responsible for the
isomerization of g1
-
r
-allyl-Cu(III) complex B to g3
-p-allyl-Cu(III)
Allylation of n-BuCu(X)ZnCl (8) and n-Bu2CuZnCl ꢀ ZnClX (9)
complex C, but instead can give rise an easier reductive elimination
with E-crotyl chloride (4).
of RCu similar to elimination of CuX.
n-BuCu(X)ZnCl (8) was prepared from n-BuZnCl (1.1 mmol in
2.8 ml of THF) and CuX (1.1 mmol) at 0 °C. E-Crotyl chloride
(1 mmol, 0.1 ml) was added at room temperature. A number of
reactions was carried out in different flasks and the same
procedure was applied for kinetic study. For the preparation of
n-Bu2CuZnCl ꢀ ZnClX (9), n-BuZnCl (2.2 mmol in 5.6 ml of THF)
and CuX (1.1 mmol) were used.
We hope that results of our kinetic study on the allylation of
catalytic and stoichiometric CuX (X = I, Br, Cl, CN, SCN) derived
alkylzinc–cuprates contributes to a better understanding for the
role of alkylzinc–copper nucleophiles as active catalyst in the
c-selective allylic coupling of alkylzinc halides.
3. Experimental
3.1. General
Acknowledgement
All reactions were carried out under a nitrogen atmosphere in
oven dried glassware using standard syringe-septum cap tech-
We thank Ankara University Research Fund (Grant No. 2005-
07-05-096) for financial support.

E. Erdik, M. Koçog˘lu / Journal of Organometallic Chemistry 694 (2009) 1890–1897
1897
[34] H. Stadtmüller, R. Lentz, C.E. Tucker, T. Stüdeman, W. Dörner, P. Knochel, J. Am.
Chem. Soc. 115 (1993) 7027.
References
[35] P. Knochel, T.S. Chou, J. Jubert, D. Rajagopal, J. Org. Chem. 58 (1993) 588.
[36] S.A. Rao, T.S. Chou, F. Schipor, P. Knochel, Tetrahedron 48 (1992)
2025.
[37] S.A. Rao, P. Knochel, J. Am. Chem. Soc. 114 (1992) 7579.
[38] M.J. Dunn, R.F.W. Jackson, J. Chem. Soc., Chem. Commun. (1992) 319.
[39] Y. Yamamoto, Y. Chounan, M. Tanaka, I. Ibuka, J. Org. Chem. 57 (1992) 1024.
[40] H.P. Knoees, M.T. Furlong, M.J. Rozema, P. Knochel, J. Org. Chem. 56 (1991)
5974.
[41] P. Knochel, S.A. Rao, J. Am. Chem. Soc. 112 (1990) 6146.
[42] T.S. Chou, P. Knochel, J. Org. Chem. 55 (1990) 4791.
[43] C. Retherford, T.S. Chou, R.M. Shelkun, P. Knochel, Tetrahedron Lett. 31 (1990)
1833.
[44] S.A. Rao, C.E. Tucker, P. Knochel, Tetrahedron Lett. 31 (1990) 7575.
[45] H.G. Chen, J. Gage, S.D. Barrett, P. Knochel, Tetrahedron Lett. 31 (1990) 1829.
[46] P. Knochel, N. Jeong, M.J. Rozema, M.C.P. Yeh, J. Am. Chem. Soc. 111 (1989)
6474.
[47] T.N. Majid, M.C.P. Yeh, P. Knochel, Tetrahedron Lett. 30 (1989) 5069.
[48] P. Knochel, M.C.P. Yeh, S.C. Berk, J. Talbert, J. Org. Chem. 53 (1988) 2390.
[49] M.C.P. Yeh, P. Knochel, Tetrahedron Lett. 29 (1988) 2395.
[50] G. Courtois, L. Miginiac, J. Organomet. Chem. 133 (1987) 285.
[51] E.-I. Nakamura, K. Sekiya, H. Oshino, I. Kuwajima, J. Am. Chem. Soc. 109 (1987)
8056.
[52] T.L. Underiner, S.D. Paisley, J. Schmitter, L. Lesheski, H.L. Goering, J. Org. Chem.
54 (1989) 2369.
[53] C.C. Tseng, S.J. Yen, H.L. Goering, J. Org. Chem. 51 (1986) 2892.
[54] C.C. Tseng, S.D. Paisley, H.L. Goering, J. Org. Chem. 51 (1986) 2884.
[55] H.L. Goering, S.S. Kantner, E.P. Seitz Jr., J. Org. Chem. 50 (1985) 5495. and
earlier references cited therein.
[56] M. Yamanaka, S. Kato, E.-I. Nakamura, J. Am. Chem. Soc. 126 (2004)
6287.
[57] N. Yoshiaki, S.-L. Zhang, E.-I. Nakamura, J. Am. Chem. Soc. 130 (2008) 12862.
[58] E.R. Bartholomew, S.H. Bertz, S. Cope, M. Murphy, C.A. Ogle, J Am. Chem. Soc.
130 (2008) 11244.
[59] E. Erdik, M. Koçog˘lu, Tetrahedron Lett. 48 (2007) 4211.
[60] E. Erdik, M. Koçog˘lu, Appl. Organomet. Chem. 20 (2006) 290.
[61] M. Koçog˘lu, E. Erdik, in preparation.
[62] S.H. Bertz, A. Chopra, M. Eriksson, C.A. Ogle, P. Seagle, Chem. Eur. J. 5 (1999)
2680.
[63] T.Zs. Nagy, A. Csâmpai, A. Kotschy, Tetrahedron 61 (2005) 9767.
[64] E.-I. Nakamura, S. Mori, Angew. Chem., Int. Ed. 39 (2000) 3750.
[65] E.-I. Nakamura, N. Yoshikai, Bull. Chem. Soc. Jpn. 77 (2004) 1.
[66] B.H. Lipshutz, J.A. Kozlowski, R.S. Wilhelm, J. Org. Chem. 48 (1983) 546.
[67] J. Leonard, B. Lygo, G. Procter, Advanced Practical Organic Chemistry, Blackie,
London, 1995.
[68] R.N. Keller, H.D. Wynocoff, Inorg. Synth. 2 (1946) 1.
[69] D.D. Perrin, W.L.F. Armarego, Purification of Laboratory Chemicals, Pergamon
Press, Oxford, 1988.
[1] A. Meijere, F. Diederich (Eds.), Metal Catalyzed Cross Coupling Reactions,
Wiley, 2004.
[2] M. Beller, C. Bolm (Eds.), Transition Metals for Organic Synthesis, 2nd ed.,
Wiley-VCH, Weinheim, 2004.
[3] B.H. Lipshutz, in: M. Schlosser (Ed.), Organometallics in Synthesis. A Manual,
Wiley, Chichester, 2002.
[4] G.A. Silverman, P.A. Rakita (Eds.), Handbook of Grignard Reagents, Marcel
Dekker, New York, 1996.
[5] A. Kar, N.P. Argada, Synthesis (2005) 2995.
[6] B.M. Trost, D.L. vanDrenken, Chem. Rev. 96 (1996) 395.
[7] B.H. Lipshutz, S. Sengupta, Org. React. 41 (1992) 135.
[8] R.M. Majid, Tetrahedron 36 (1980) 1901.
[9] Z. Rappoport, I. Marek (Eds.), The Chemistry of Organozinc Compounds, Wiley,
Chichester, 2006.
[10] P. Knochel, P. Jones (Eds.), Organozinc Reagents. A Practical Approach, Oxford
University Press, Oxford, 1999.
[11] E. Erdik, Organozinc Reagents in Organic Synthesis, CRC Press, New York, 1996.
[12] N. Krause (Ed.), Modern Organocopper Chemistry, Wiley-VCH, Weinheim,
2001.
[13] R.J.K. Taylor (Ed.), Organocopper Reagents.
University Press, 1994.
A Practical Approach, Oxford
[14] C.A. Falciola, K. Tissot-Crosset, A. Alexakis, Angew. Chem., Int. Ed. 45 (2006)
5995.
[15] T. Yamazaki, H. Umetani, T. Kitazume, Tetrahedron Lett. 38 (1997) 6705.
[16] E.S. Person, M. Kluveren, D.M. Grove, J.E. Backvall, Chem. Eur. J. 1 (1995) 351.
[17] A. Yanagisawa, N. Nobuyoshi, H. Yamamoto, Tetrahedron 50 (1994) 6017.
[18] A. Yanagisawa, N. Nomura, H. Yamamoto, Synlett (1993) 689.
[19] A. Yanagisawa, Y. Noritake, N. Nomura, H. Yamamoto, Synlett (1991) 251.
[20] H.L. Goering, T.L. Underiner, J. Org. Chem. 56 (1991) 2563. and earlier
references cited therein.
[21] J.E. Backvall, M. Sellen, B. Grant, J. Am. Chem. Soc. 112 (1990) 6615. and earlier
references cited therein.
[22] K.A. Woerpel, J.H. Smitrovich, J. Am. Chem. Soc. 120 (1998) 12988.
[23] V. Calo, V. Fiandanase, A. Nacci, A. Salimati, Tetrahedron 50 (1994) 7383. and
earlier references cited therein.
[24] M. Arai, M. Kawasuji, E. Nakamura, Chem. Lett. (1993) 357.
[25] N. Fujii, Nakai, H. Habshita, H. Yohizawa, T. Ibuka, F. Garrido, A. Mann, Y.
Chounan, Y. Yamamoto, Tetrahedron Lett. 34 (1993) 4227.
[26] T.L. Underiner, S.D. Paisley, J. Schmitter, L. Lesheski, H.L. Goering, J. Org. Chem.
54 (1989) 2370. and earlier references cited therein.
[27] T. Ibuka, M. Tanaka, S. Nishii, Y. Yamamoto, J. Am. Chem. Soc. 111 (1989) 4864.
[28] W. Tückmantel, K. Oshima, H. Nozaki, Chem. Ber. 119 (1986) 1581.
[29] E.-I. Nakamura, K. Sekiya, M. Arai, S. Aoki, J. Am. Chem. Soc. 111 (1989) 3091.
[30] K. Sekiya, E. Nakamura, Tetrahedron Lett. 29 (1988) 5155.
[31] D.J. Burton, L.G. Sprague, J. Org. Chem. 54 (1989) 613.
[32] H. Ochiai, Y. Tamaru, K. Tsubaki, Z-I. Yoshida, J. Org. Chem. 52 (1987)
4418.
[70] J. Barber, J. Chem. Soc. 1 (1943) 79.
[71] J.H. Watson, J.F. Eastham, J. Organomet. Chem. 9 (1967) 165.
[33] E.-I. Nakamura, S. Aoki, K. Sekiya, H. Oshima, I. Kuwojima, J. Am. Chem. Soc.
109 (1987) 8056.