Reactivity of mixed organozinc and mixed organocopper reagents. Part 4: A kinetic study of group transfer selectivity in C-C coupling of mixed diorganocuprates

Article abstract of DOI:10.1002/poc.1567

The competitive rate data and Taft relationships for the coupling of bromomagnesium n-butyl (substituted phenyl) cuprates with alkyl bromides show that selective n-butyl transfer can be explained by an oxidative addition mechanism. Taft reaction constants

Full text of DOI:10.1002/poc.1567

Research Article
Received: 5 December 2008,
Revised: 16 March 2009,
Accepted: 2 April 2009,
Published online in Wiley InterScience: 29 July 2009
(www.interscience.wiley.com) DOI 10.1002/poc.1567
Reactivity of mixed organozinc and mixed
organocopper reagents. Part 4: a kinetic study
of group transfer selectivity in C—C coupling
of mixed diorganocuprates
a
a
¨
*
Ender Erdik and Duygu Ozkan
The competitive rate data and Taft relationships for the coupling of bromomagnesium n-butyl (substituted phenyl)
cuprates with alkyl bromides show that selective n-butyl transfer can be explained by an oxidative addition
mechanism. Taft reaction constants also show that the residual group FG-C6H4 in the mixed cuprate
n-Bu(FG-C6H4)CuMgBr changes the ability of the copper nucleophile to react with the electrophile RBr. These results
provide support for the commonly accepted hypothesis regarding the dependence of the R1 group transfer ability on
the strength of R2—Cu bond in reactions of R1R2CuMgBr reagents. Copyright ß 2009 John Wiley & Sons, Ltd.
Keywords: alkylation of mixed cuprates; competitive kinetics; mixed magnesium cuprates; Taft plots
derived magnesium cuprates do not have the same reactivity as
INTRODUCTION
lithium cuprates in some reactions; however, they are more
readily available and thermally stable.[22] Mixed magnesium
cuprates of R1R2CuMgBr type[22–24] and RRRTCuMgBr type[23,25]
have been found as successful alternatives to mixed lithium
cuprates. Mixed cuprates RR(RT)nCuMn (M ¼ Li, MgBr) (n ¼ 2,3)[12]
Organocuprates, R2CuM (M: Li, MgBr), represent a class of
organometallic reagents most used in organic syntheses because
of their reactivity toward carbon electrophiles.[1,2] However, in the
reaction of R2CuM, only one organic group is transferred to the
electrophile. The solutions for this problem are (i) the use of
mixed diorganocuprates of R1R2CuM type, in which one of the R1
and R2 groups has a lower transfer rate than the other and (ii) the
use of mixed organocuprates of RRRTCuM type composed of
one transferable group RT together with the residual (non-
transferable or dummy) group RR. For the first type of mixed
cuprates, organic groups bearing sp-C or sp2C such as alkynyl,[3–5]
aryl,[5,6] 2-thienyl,[7,8] and cyano groups[9] have been found to
have a lower transfer rate. Selectivity in organic group transfer in
reactions of mixed cuprates R1R2CuM has been investigated in
detail by Whitesides,[5] House,[10] and Posner.[11] The relative rate
of organic group transfer is known to be in the order of
n-Bu ꢀ s-Bu ꢀ t-Bu » Ph alkynyl and vinyl Me » alkynyl. However,
for the mixed cuprates R1R2CuLi.LiCN[12–14] the observed
preference for Me over vinyl in substitution reactions gets
reversed in 1,4-addition reactions.[13] Recently, Knochel has
shown that mixed functionalized aryl cuprates containing
Me3CCH2 or Me2CPhCH2 groups selectively transfer the functio-
nalized aryl groups.[15] For the second type of mixed cuprates,
organic groups such as RS,[16] R2N,[17,18] and R3P[17–19] have also
been used as residual groups.
or their cyano analogs RRRTCuLi.MCN (M ¼ MgBr,[12,13] Na[26]
)
have been reported to increase the group transfer selectivity in
some reactions.
In mixed diorganocuprates of the R1R2CuM type, the organic
group selectivity to transfer to an electrophile is considered to be
a function of R1—E or R2—E formation in the reductive
elimination of the Cu(III) intermediate R1R2ECu, which is formed
in the oxidative addition step (Scheme 1).[1] Oxidative addition is
the rate determining step in the substitution of diorganocuprates
with alkyl halides.[1] However, studies on their 1,4-addition
to enones indicated that the rate determining step may also
be a reductive elimination of the Cu(III) intermediate step
(Scheme 1).[27,28]
A widely accepted hypothesis for the transfer selectivity of the
R1 or R2 group to an electrophile is that the group which has a
stronger bond to Cu in R1R2CuE acts as the group of lower
transfer selectivity.[1,5] However, detailed studies have also been
reported regarding the controlling factors for the group transfer
selectivity of mixed lithium cuprates in their 1,4-addition
reactions.[29]
We are currently working on the reactivities of mixed
organocopper and mixed organozinc reagents and controlling
factors of group transfer selectivity in their reactions.[30,31] To the
Among the atom-economic mixed diorganocuprates, the most
attractive appeared to use cyanocuprates RCu(CN)Li or
R(2-thienyl)CuLi[7] and their cyano analogs R(2-thienyl)
CuLi.LiCN.[8,20] Bertz introduced a trimethylsilylmethyl (TMSM)
group as a residual group and reported that the mixed cuprates
R(TMSM)CuLi[21] are more reactive than the corresponding
homocuprates R2CuLi. A number of reactions of mixed lithium
cuprates of R1R2CuLi type[5,10–15] and also RRRTCuLi type[6–9,16–21]
have been reported. It is well known that the Grignard reagent
*
Correspondence to: E. Erdik, Faculty of Science, Ankara University, Be¸sevler,
Ankara, Turkey.
E-mail: erdik@science.ankara.edu.tr
¨
E. Erdik, D. Ozkan
Faculty of Science, Ankara University, Be¸sevler, Ankara, Turkey
a
J. Phys. Org. Chem. 2009, 22 1148–1154
Copyright ß 2009 John Wiley & Sons, Ltd.

KINETIC STUDY OF GROUP TRANSFER SELECTIVITY
would be affected by the possible equilibrium between
di-n-butyl cuprate 1aa and diphenylcuprate 1bb to form
n-butylphenylcuprate 1ab.
We used CuI as the transmetallation reagent for the
preparation of mixed cuprate 1ab and carried out the alkylation
reaction with a ratio of 3:1 for 1ab:2 at room temperature for 3 h
(optimized conditions). The total yield and the product ratio 3:4
were found to be 92% (97:3), 95% (100:0), 74% (98:2), and 81%
(98:2) using the mixed cuprate 1ab prepared by the Methods A1,
A2, B1, and B2, respectively. As expected, Methods A1 and A2 gave
a higher and almost equal total yield of coupling, and the original
bonding of n-butyl group to Cu or Mg was found to be not
important. We used the mixed cuprate 1ab prepared according
to Method A2 for further experiments. The lower yields obtained
using the mixed cuprate 1ab prepared by Methods B1 and B2 may
indicate that the n-butyl transfer is affected by the extent of the
equilibrium between 1aa and 1bb to yield 1ab. The background
yields for alkylation of the mixed cuprate 1ab was found to be
71% for di-n-butylcuprate 1aa and 3% for diphenylcuprate 1bb. It
is surprising that the n-butyl group in the mixed cuprate 1ab is
alkylated with a higher yield than the n-butyl group in
di-n-butylcuprate 1aa.
Scheme 1.
best of our knowledge, the control of group transfer selectivity in
reactions of R1R2CuMgBr type mixed cuprates has not been
investigated in detail so far.
Here we report results of our competitive kinetic studies and
Taft correlations for the alkylation of mixed n-butyl (substituted
phenyl) cuprates in THF to provide a kinetic support for the
hypothesis of the dependence of the R1 group transfer ability on
the strength of the R2—Cu bond in reactions of R1R2CuMgBr
reagents.
RESULTS AND DISCUSSION
(ii) We screened a number of CuX compounds for transme-
tallation of n-butylmagnesium bromide to obtain n-butylcopper
n-Bu(X)CuMgBr used for the preparation of n-butylphenylcuprate
1ab. The use of CuBr, CuCl, and CuSCN lowered the total yield to
58–84% with almost complete n-butyl selectivity (3:4 ¼ 99:1).
However, CuCN changed both the total yield and transfer ability
of the n-butyl group since the mixed cuprate 1ab-CN was
alkylated with a total yield 59% and a lower n-butyl selectivity
(3:4 ¼ 75:25). Using CuCN.2LiCl increased the total yield to 84%
with a 3:4 ratio of 81:19. However, the obvious change obtained
with the mixed cyanocuprate, which is expected to exist as
n-BuPhCuMgBr.MgBr(CN), may seem reasonable since lithium
analogs of mixed cyanocuprates R1R2CuLi.LiCN have already
been reported to exhibit alternative selectivity for organic group
transfer.[12–14]
In the coupling reaction, we also used n-BuPh2Cu(MgBr)2.MgBrI
1ab2 and its cyano analog, n-BuPh2Cu(MgBr)2.MgBr(CN)
1ab2-CN, which are higher order mixed cuprates prepared using
CuI and CuCN, respectively, expecting that 1ab2-CN would give a
lower yield and n-butyl selectivity than that obtained with 1ab2,
similar to the results obtained with n-BuPhCuMgBr.MgBrI 1ab
and n-BuPhCuMgBr.MgBr(CN) 1ab-CN as stated above. As
expected, coupling of 1ab2 resulted in a total yield of 50%
and n-butyl selectivity of 3:4 ¼ 84:16, whereas 1ab2-CN coupled
with a low n-butyl selectivity of 3:4 ¼ 70:30, but with a higher
yield of 79%. As seen, the higher order cuprate 1ab2 containing
two residual Ph groups gave a much lower coupling yield of 50%
compared to 95% obtained with the lower order cuprate 1ab.
(iii) We carried out the alkylation reaction of mixed cuprate 1 in
THF using a coordinating co-solvent, a Lewis base or acid as an
additive. We had already reported success in the control of group
transfer selectivity by changing the solvent or using an additive in
the acylation of n-butylphenylzinc[30] and other mixed diorga-
nozincs[31] in THF. However, using THF:HMPA (1:1), THF:DMPU
(1:1), THF:NMP (1:1), or THF:diglyme (1:1) as solvent for the
coupling of the mixed cuprate 1ab did not make an appreciable
change in the total yield, i.e., 70–90% compared to 92% yield in
THF. The complete transfer selectivity of the n-Bu group also did
not change. However, the n-butyl transfer yield lowered to 19%
for the reaction in THF:TMEDA (1:1). Use of TMSCI, MgCl2 and also
As model mixed magnesium cuprates, we chose n-butyl
(substituted phenyl) cuprates, in which the n-butyl group has
a much higher transfer selectivity than the phenyl group in their
reactions. We selected the n-butyl–n-alkyl coupling reaction to
test our hypothesis that the reaction rate of the n-butyl group
depends on the strength of substituted phenyl—Cu bond.
In order to find out suitable reaction conditions for the kinetic
studies of alkylation of n-butyl (substituted phenyl) cuprates, we
first carried out a brief investigation to see how the reaction
conditions affect the group transfer selectivity in C—C coupling
reactions. We chose the alkylation of bromomagnesium n-butyl
(phenyl) cuprate 1 with n-pentyl bromide 2 in THF at room
temperature as a model reaction (Scheme 2).
We focused on the following parameters: (i) preparation
method of the mixed cuprates, reaction temperature and time,
(ii) Cu(I) compounds used for Mg ! Cu transmetallation, and
(iii) co-solvents and additives. The relative transfer ability of n-Bu
and Ph groups was determined by quantitative GC analysis using
authentic samples of 3 and 4.
(i) We prepared n-butylphenylcuprate 1ab by using two
different methods. In Method A, n-butylmagnesium bromide or
phenylmagnesium bromide was first transmetallated to obtain
the corresponding monocopper reagent and then allowed to
react with the other Grignard reagent. Either phenylcopper or
n-butylcopper was used as the organocopper reagent (Methods
A1 and A2, respectively).
In Method B, di-n-butylcuprate 1aa and diphenylcuprate 1bb
were mixed; either di-n-butyl cuprate or diphenylcuprate was
added to the other cuprate reagent (Method B1 and Method B2,
respectively).
Method A1 and Method A2 were tested to see if the organyl
group originally bonded to Cu or Mg could make a change in the
relative transfer ability of n-butyl and phenyl groups. Methods B1
and B2 were tried to find if the transfer ability of the groups
Scheme 2.
J. Phys. Org. Chem. 2009, 22 1148–1154
Copyright ß 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

¨
E. ERDIK AND D. OZKAN
Scheme 3.
n-Bu3P and bipyridyl as additives did not lead to a change in the
total yield (85–99%) and n-Bu transfer ability.
Scheme 4.
In summary, we observed that the complete transfer selectivity
of the n-butyl group in the reaction of n-BuPhCuMgBr 1ab with
n-pentyl bromide 2 in THF does not change by reaction
conditions. We also tested substituted phenyl groups as residual
groups in mixed n-butylcuprate. We found that the alkylation of
mixed cuprates, n-Bu (FG-C6H4)CuMgBr (FG: 4-Me, 3-MeO, 4-Br),
takes place with a total yield of 47, 65, and 100%, respectively,
with n-butyl selectivity higher than 95% compared to the total
yield of 92% with n-BuPhCuMgBr 1ab.
This result led to the idea that the transferable character of
n-Bu group would change by changing the residual character of
the substituted phenyl group. In accordance with the commonly
accepted hypothesis for coupling of mixed cuprates, we would
expect a significant effect of the strength of the FG-C6H4-Cu bond
on the transfer rate of the n-Bu group (Scheme 3). In the
formation of the Cu(III) intermediate A in the rate determining
oxidative addition step, the FG-C6H4 group may form a stronger
bond in the Cu(III) intermediate to stay on the Cu atom in
reductive elimination. As the functional group is expected to
affect the strength of the FG-C6H4-Cu bond, it would also affect
the formation rate of the Cu(III) intermediate or, in other words,
the n-butyl transfer rate of 1.
In the kinetic study of the coupling reaction of
n-butyl(phenyl)cuprate 1ab with n-PentBr 2a, we found the
absolute rate measurements to be complicated and time
consuming due to the heterogeneous reaction. So, we used
the competitive kinetics method, in which a pair of R1Br(n-PentBr)
2a and R2Br reacted with
a
limited amount of
n-butyl(phenyl)cuprate 1ab (Scheme 4). As alkyl bromides R2Br
2b-f, we chose a number of prim-alkyl bromides 2a-d, t-butyl
bromide 2e, homobenzyl bromide 2f, and benzyl bromide 2g.
Taft reaction constants were found for the reactions of a number
of n-butyl (substituted phenyl) cuprates 1ab–1ae by collecting
their competitive rate data. Taft correlation was also found for the
alkylation of bromomagnesium di-n-butylcuprate 1aa.
In the competitive reaction of A1 and A2 with subequimolar B
to give the products C1 and C2, assuming a first-order reaction in
A, relative reactivities of A1 and A2 are calculated using Eqn
(1)[37,38]
We assumed, therefore, that the n-Bu-n-Pent coupling rate of
n-Bu(FG-C6H4)CuMgBr reagents with n-PentBr would be affected
by the electronic effects of the functional group on the residual
group FG-C6H4. This study was undertaken to critically examine
this assumption. For this purpose, we carried out a kinetic study in
three steps. First, we found the rate constants of the coupling of
n-Bu(FG-C6H4)CuMgBr reagents with n-PentBr and other alkyl
bromides to see the dependence of the reaction rate on the
residual group, FG-C6H4. Secondly, we applied Taft methodology
to find the electronic effects in these reactions, and thirdly, we
investigated the dependence of Taft reaction constants of
n-Bu(FG-C6H4)CuMgBr reagents on the residual group, FG-C6H4.
The reactions of lithium diorganocuprates with alkyl and aryl
halides have been shown to be of first order in the cuprate and in
the organyl halide and take place by displacement of the leaving
group in the electrophile, E ¼ R—A (Scheme 1) with diorgano-
cuprate nucleophile to form a triorganyl Cu(III) intermediate in
the rate determining step. There are a few reported works on the
kinetics of the coupling of homocuprates.[32–35] However, the
kinetics of reactions of mixed cuprates have not been
investigated so far. Hence, our findings regarding the rate
constants and Taft reaction constants for mixed cuprate–alkyl
coupling reactions might provide support for the substitution
mechanism of mixed cuprates as well as for our assumption
regarding the group transfer ability of these reagents.
ð1Þ
where k2/k1 is the competitive rate ratio (competition con-
stant[39]), [A]t and [A]0 are the concentrations of the competitive
reactants (A1 and A2) at the time t (a suitable reaction time or the
end of the reaction) and their initial concentrations, respectively.
Among the reported methods[37–50] for competitive kinetics, Eqn
(2) was also frequently used for calculation of competitive rate
ratios,
k2 %yield of C2
¼
(2)
%yield of C1
k1
where the yield of products C1 and C2 are measured at a suitable
reaction time or at the end of the reaction.[46–48]
For calculating the rate ratios for the competitive coupling of
alkyl bromides R1Br and R2Br with a mixed cuprate, we adapted
Eqn (1) as follows:
k2 log ð½R2Brꢁ1=½R2Brꢁ0Þ
¼
(3)
log ð½R1Brꢁ1=½R1Brꢁ0Þ
k1
Bertz had recently reported[36] the observation of a Cu(III)
intermediate in the reaction of Me2CuLi.LiX (X ¼ I, SCN, SPh, CN)
with ethyl iodide in THF and had also reported that any process
that replaces a residual group with a transferable one (i.e., from
X ¼ SPh to X ¼ Me) may result in diminished yield of the coupled
product. This result also supports our idea.
where [RBr]t and [RBr]0 are the amounts of competing reactants
at time t and t ¼ 0, respectively. However, being interested in
the competitive formation rate of coupling products and
also assuming small perturbations of side reactions, we made
the simplification [RBr]t ¼ [RBr]0 ꢂ [n-Bu-R]t and rearranged
Eqn (3) in terms of the coupling products n-Bu-R1 and n-Bu-R2
www.interscience.wiley.com/journal/poc
Copyright ß 2009 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 1148–1154

KINETIC STUDY OF GROUP TRANSFER SELECTIVITY
as follows:
alkyl bromides R1Br and R2Br with a mixed cuprate 1ab would
allow us to collect reliable data.
k2 log ð½R2Brꢁ0 ꢂ ½nꢂBuR2ꢁt=½R2Brꢁ0Þ
We optimized the scale for competition to be
[R1Br]:[R2Br]:[1] ¼ 7:7:1 and we found the amount of n-BuR1
and n-BuR2 by GLC analysis. Carrying out the reaction in the
presence of an internal standard, taking 4–8 samples at 5 or
10 min intervals and quenching the samples, and calculating the
amount of products in each sample and taking the average of
k2/k1 values did not give reproducible results, possibly owing to
the heterogeneous reaction. Therefore, we applied the method
that we used in the competitive amination study of phenyl
carbanions,[51,52] i.e., 4–8 competition experiments were carried
out in different flasks in the presence of an internal standard and
the reactions were quenched at different time points. This
method of analysis gave more reproducible results in the
evaluation of reactivity profiles for competitive alkylation. We also
used Eqn (5) to calculate the competitive rate ratios and found
that these values are in accordance with those found by using
Eqn (4). However, taking the average k2/k1 values found by Eqn (5)
led to a lower mean deviation of 4–8% than that found by Eqn (4),
both are in the error limit of GLC analysis.
We carried out experiments for the alkylation kinetics of
n-butyl (phenyl) cuprate 1ab, n-butyl (4-tolyl) cuprate 1ac, n-butyl
(p-anisyl) cuprate 1ad, and n-butyl (4-bromophenyl) cuprate 1ae
with competing alkyl bromides 2a–g in THF at room temperature.
The competitive rate ratios k2/k1 are given in Table 1.
The competitive rate ratios k2/k1 were also found for the
alkylation kinetics of di-n-butyl cuprate 1aa and are given in
*
Table 1. Taft inductive substituent constants s and steric
¼
log ð½R1Brꢁ0 ꢂ ½nꢂBuR2ꢁt=½R2Brꢁ0Þ
k1
À
Á
log 1 ꢂ ð½nꢂBuR2ꢁt=½R2Brꢁ0Þ
À
Á
¼
(4)
log 1 ꢂ ð½nꢂBuR1ꢁt=½R1Brꢁ0Þ
We have already used a similar equation successfully in the
competitive amination kinetics of phenyl and substituted phenyl
carbanions derived from Grignard reagents,[51,52] stoichiometric
magnesium–copper reagents[51] and catalytic zinc-copper
reagents.[52] We have also used the competitive rate date in
the application of Hammett methodology for the amination
reactions.[51,52]
With the same simplication, Eqn (2) can be adapted as follows:
k2 %yield of nꢂBuR2
¼
(5)
%yield of nꢂBuR1
k1
with competing R1Br (n-PentBr) and R2Br.
Before collecting the rate data leading to a suitable reactivity
profile, we carried out preliminary experiments to see whether
the alkylation of mixed cuprate 1ab would meet the require-
ments of the competitive kinetics method. As we could arrange
the experimental conditions to keep the amounts of the products
n-BuR1 (n-C9H20) and n-BuR2 much higher than the experimental
error during the progress of the reaction and since we also
observed that the difference between the reaction rates of R1Br
and R2Br are not large, we decided that competitive coupling of
Table 1. Competitive rate ratios for the alkylation of bromomagnesium di-n-butylcuprate 1aa, n-butyl phenylcuprate 1ab, and
n-butyl (substituted phenyl) cuprates 1ac–1ae with alkyl bromides 2b–f in the presence of n-pentyl bromide 2a in THF at 25 8C
nꢂBuRRMgBr þ R1Br þ R2Br ꢂꢂꢂꢂꢂꢂ! nꢂBuR1 þ nꢂBuR2
ꢃ
2a
2bꢂg THF; 25 C
3a
3bꢂg
1
RR ¼ nꢂBu HꢂC6H4 4ꢂMeC6H4 3ꢂMeOC6H4 4ꢂBrC6H4
a
c
e
b
d
R1 ¼ nꢂPent
a,b
k2/k1
RR
c
R2
s
Esc
H-C6H4 b
4-MeC6H4 c
3-MeOC6H4 d
4-BrC6H4 e
n-Bu a
*
n-Pent 2a
n-Hept 2b
n-Oct 2c
i-Bu 2d
t-Bu 2e
PhCH2CH2 2f
PhCH2 2g
ꢂ0.16
ꢂ0.17
ꢂ0.17
ꢂ0.125
ꢂ0.30
0.08
ꢂ0.40
ꢂ0.17
ꢂ0.17
ꢂ0.93
ꢂ1.64
ꢂ0.55
—
1.00d
1.14
1.05
0.14
0.08
0.68
23
1.00d
1.34
1.07
0.18
0.11
0.69
1.00d
1.15
—
0.16
0.14
0.69
51
1.00d
1.18
—
0.13
0.13
0.69
65
1.00d
1.21
0.92
0.13
0.18
0.70
76
0.215
84
a k2/k1 ¼ kR2Br/kR1Br
.
b The average of k2/k1 values (4–8 experiments) was calculated from the competition experiments using Eqn (5). The relative error of
the values does not exceed 8%.
c
*
Inductive substituent s and steric constants Es to use in the Taft–Ingold equations are taken from Reference [53].
d By definition, R2Br ¼ R1Br.
J. Phys. Org. Chem. 2009, 22 1148–1154
Copyright ß 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

¨
E. ERDIK AND D. OZKAN
constants Es for alkyl groups to be used in the Taft equation and
Taft–Ingold equation are also given in Table 1.[53–56]
First, we compared the competitive ratios for alkylation of
bromomagnesium n-butyl (substituted phenyl) cuprates 1ab–ae
with the same alkyl bromide. As seen, the rate ratio diminishes in
the order of 1ac > 1ad or 1ae > 1ab in the alkylation with
n-HeptBr 2b and PhCH2Br 2g, and in the order of
1ac > 1ad > 1ab ꢀ 1ae in the alkylation with i-BuBr 2c. The rate
ratio has been found to be constant in the alkylation of 1ab–ae
with PhCH2CH2Br 2f for a reason as yet unknown. In the alkylation
with sterically hindered t-BuBr 2e, however, the highest rate ratio
was obtained with 1ad and 1ae.
Figure 1. Taft plot for the competitive coupling kinetics of phenyl
(4-tolyl) cuprate 1ac with alkyl bromides in THF at 25 8C
As we assumed the oxidative addition to be the rate
determining step in the coupling of mixed cuprates, as in
the coupling of homocuprates, we tried to explain the role of the
substituted phenyl ligand group on the competitive rate ratio. In
the reactions of monoorganocoppers R1Cu(Z)LMgBr and diorga-
Taft
plot
for
competitive
alkyl
coupling
of
n-Bu(4-MeC6H4)CuMgBr 1ac is given in Fig. 1. A reasonably
good linear relationship was obtained for each reaction except
for the points for coupling partners, i-BuBr 2d and PhCH2CH2Br 2f.
These points were found to be lying significantly outside the
linear plots. The linearity of Taft plots supports the assumption of
nocuprates R1 CuLMgBr.MgBrZ prepared from 1 or 2 equivalents
2
of R1MgBr and 1 equivalent of CuZ or CuZ.L (Z ¼Cl, Br, I, CN, and
L ¼ Ligand), Z and L are known to favor oxidative addition in the
coupling with RA if they are s-donor–p-acceptor ligands.[57,58]
However, reductive elimination of Cu(III) intermediates, R1RCu(Z)L
a
first-order reaction in the cuprate n-BuRRCuMgBr and
or R1 RCuL, which are formed by coupling of R1Cu(Z)LMgBr and
n-Bu2CuMgBr for the calculation of competitive rate ratios. The
2
R1 CuLMgBr.MgBrZ, respectively is expected to be fast with
reaction constants r are all positive, as expected, i.e., these
*
2
s-acceptor ligands.[58,59] If in the coupling of n-butyl (substituted
phenyl) cuprates 1ab, ac, and ae, the s-donor properties of
substituted phenyl ligands for a soft acid center Cu(I) decrease in
the order of 4-MeC6H4 > c C6H5 b > 4-BrC6H4 e, then the rate
ratios of 1ab, ac, and, ae for coupling with the same alkyl bromide
are expected to decrease in the same order. As expected, in
the alkylation of mixed n-butylcuprates we found the highest
rate ratio for the cuprate containing 4-MeC6H4 c as the residual
group and the lowest rate ratio for the cuprate containing C6H5
b or 4-BrC6H5 e as the residual groups. However, s-donor
properties of FG-C6H4 groups will decrease the ease of reductive
elimination of n-Bu(FG-C6H4)RCu intermediates to give n-BuR and
then formation of n-BuR will take place possibly with a lower
yield in the case of n-Bu(4-MeC6H4)RCu than that in the case
of n-Bu(C6H5)RCu and n-Bu(4-BrC6H4)RCu. In the coupling of
n-Bu(FG-C6H4)MgBr reagents with n-PentBr, the yield of n-C9H20
was found to be 47, 92, and 100% using 4-MeC6H4, C6H5, and
4-BrC6H5 as the residual groups, respectively. These results
support our assumption that the coupling rate of a mixed cuprate
changes, depending on the residual group.
reactions are accelerated by electron-attracting alkyl groups in
RBr that help in the formation of the n-BuRRRCu intermediate
between a copper nucleophile n-BuRCudꢂ and electrophile Rdþ in
*
the rate determining step. The magnitudes of r for mixed
cuprates n-Bu(FG-C6H4)CuMgBr show that the amount of
negative charge developed at the reaction center, R-Br, changes
depending on the residual group, FG-C6H4.[53] However, both
electron donating and electron-attracting substituents on the
residual group of the copper nucleophile, increase the sensitivity
of the reaction to change in the R group of RBr.
Due to the downward deviation of i-BuBr 2d from the linear
*
*
r –s plot, we subjected the competitive rate ratios to the two
parameter Taft–Ingold equation to determine whether the steric
effect is important in the coupling of mixed n-butylcuprates
1ab–ae with alkyl bromides. However, correlations gave negative
*
inductive reaction constants, r ¼ ꢂ0.307 ꢂ (ꢂ0.437) and steric
reaction constants, d ¼ 0.778 ꢂ 1.039 with a far greater scatter of
points than that obtained with the Taft equation. So, instead of
evaluating the steric effect, we applied a transformation to
change the two-parameter plot to a one-parameter plot.[54] This
approach gave a satisfactory result with positive inductive
The competitive rate ratios for n-Bu2CuMgBr 1aa were found to
be higher than those obtained for n-Bu(C6H5)CuMgBr 1ab, but
comparison with those obtained for 1ac–ae did not give a
satisfactory result.
Secondly, the competitive rate ratios for the alkylation of mixed
n-butylcuprates 1ab–ae and also alkylation of di-n-butylcuprate
1aa were subjected to the one parameter version of the Taft
*
reaction constants r ¼ 0.451 ꢂ 0.683 for mixed n-butylcuprates
*
1ab–ae and r ¼ 0.369 for di-n-butylcuprate. These results also
provide support for our assumption that the residual group in the
mixed cuprate changes the ability of the copper nucleophile to
react with the electrophile.
In order to propose a reaction pathway for n-butyl transfer in
the coupling of mixed n-butylcuprates with alkyl bromides, we
assumed a heterodimer contact ion pair structure for bromo-
magnesium cuprates similar to lithium cuprates and assumed
pathways similar to the mechanisms proposed by Nakamura and
Yoshikai[14,32] and also recently by Bertz[36] for the coupling of
lithium diorganocuprates.
* *
equation, log(k2/k1) ¼ r s rather than its two parameter version,
[53–56]
* *
the Taft–Ingold equation, logk2/k1 ¼ r s þ dEs.
The suc-
cessful use of the one parameter Taft equation was already
reported in the copper(I) catalyzed acylation of Grignard
reagents.[60] The plots of log(k2/k1) against the Taft inductive
*
substituent constants s for the coupling of n-Bu(FG-C6H4)
CuMgBr reagents 1ab–ae and n-Bu2CuMgBr 1aa gave positive
reaction constants: for n-Bu(FG-C6H4)CuMgBr reagents,
We may suggest that the oxidative addition of a mixed cuprate
can take place through the formation of the transition state A
according to the mechanism proposed by Nakamura and
Yoshikai (Scheme 5).[14,32] The R1R2RCu intermediate is formed
after the rate determining halide displacement step as the
*
r ¼ 4.044 (r ¼ 0.921) (FG ¼ H); 4.951 (r ¼ 0.995) (FG ¼ 4-Me);
5.521 (r ¼ 0.966) (FG ¼ 3-MeO); 4.775 (r ¼ 0.957) (FG ¼ 4-Br); and
*
for n-Bu2CuMgBr r ¼ 4.619 (r ¼ 0.949).
www.interscience.wiley.com/journal/poc
Copyright ß 2009 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 1148–1154

KINETIC STUDY OF GROUP TRANSFER SELECTIVITY
Scheme 5.
Scheme 6.
unstable transient species B and C which have a T-shape
geometry with the fourth ligand (solvent or halide). In the case of
coupling of mixed cuprates n-Bu(FG-C6H4)CuMgBr, R2 was
assumed to be FG-C6H4 for the elimination of n-BuR. However,
it seemed to us less speculative to propose a mechanism
involving the reactive intermediate proposed by Bertz
(Scheme 6).[36] Oxidative addition of n-Bu(FG-C6H4)CuMgBr can
take place through the formation of the R1R2RCu intermediate as
transient species D. Assumption of the formation of D may help
to explain the difference in the ease of reductive elimination of
n-BuR in the case of different X ligands.
In conclusion, the competitive rate data and Taft relationships
obtained in this study show that selective n-butyl transfer in the
coupling of bromomagnesium n-butyl (substituted phenyl)
cuprates with alkyl bromides can be explained by an oxidative
addition mechanism.
Taft reaction constants for the coupling of mixed cuprates
n-Bu(FG-C6H4)CuMgBr show that the amount of negative charge
developed at the reaction center RBr by the copper nucleophile
changes depending on the residual group FG-C6H4, and provide
support for our assumption that the residual group in the mixed
cuprate changes the ability of the copper nucleophile to react
with the electrophile. These results also support the commonly
accepted hypothesis regarding the dependence of the R1 group
transfer ability on the strength of R2—Cu bond in reactions of
R1R2CuMgBr reagents.
EXPERIMENTAL
All reactions were carried out under dry nitrogen atmosphere
using oven-dried glassware. Reagents and solvents were handled
by using standard syringe–septum cap techniques.[61] Quanti-
tative GLC analyses were performed on a Thermo Finnigan gas
chromatograph equipped with a ZB-5 capillary column packed
phenyl–polysiloxane using an internal standard technique. THF
was distilled from sodium benzophenone dianion. Pure HMPA,
NMP, DMPU, and diglyme were distilled just before use. n-BuBr,
C6H5Br, 4-MeC6H4Br, 3-MeOC6H4Br, and 4-BrC6H4Br were
obtained commercially and purified using literature procedures.
Mg turnings for Grignard reagents were used without purifi-
cation. CuI,[62] CuBr,[63] CuCl,[63] CuCN,[64] and CuSCN[62] were
purified according to the published procedures. Grignard
reagents, n-BuMgBr, and FG-C6H4MgBr (FG ¼ H, 4-Me, 3-MeO,
4-Br) were prepared in THF according to the standard procedure
and their concentrations were found by titration prior to use.[65]
n-Bu2CuMgBr 1aa was prepared by the addition of 2 molar
equivalent of n-BuMgBr to a suspension of 1 molar equivalent of
J. Phys. Org. Chem. 2009, 22 1148–1154
Copyright ß 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

¨
E. ERDIK AND D. OZKAN
CuI in THF at ꢂ20 8C and stirring at that temperature for
15–30 min. Mixed n-butyl (phenyl) cuprates n-Bu (FG-C6H4)
CuMgBr 1ab–ae (FG ¼ H b, 4-Me c, 3-MeO d, 4-Br e) were
prepared according to the method A2. To n-BuCu prepared by the
addition of 1 mol equiv of n-BuMgBr to a suspension of CuI in THF
at ꢂ20 8C was added 1 molar equivalent of FG-C6H4MgBr at this
temperature with continuous stirring for 15–30 min.
[18] S. H. Bertz, G. Dabbagh, J. Org. Chem. 1984, 49, 1119.
[19] S. F. Martin, J. R. Fishpaugh, J. M. Power, D. M. Giulando, R. A. Jones, C.
M. Nunn, A. H. Cowley, J. Am. Chem. Soc. 1988, 110, 7226.
[20] S. H. Bertz, G. Miao, M. Eriksson, Chem. Commun. 1996, 815–816.
[21] S. H. Bertz, M. Eriksson, G. Miao, P. S. Synder, J. Am. Chem. Soc. 1996,
118, 10906–10907.
[22] D. E. Bergbreiter, J. M. Killough, J. Org. Chem. 1976, 41, 2750–2753.
[23] D. E. Bergbreiter, G. M. Whitesides, J. Org. Chem. 1975, 40, 779–782.
[24] S. M. H. Kabir, M. T. Rahman, J. Organomet. Chem. 2001, 619, 31–35.
[25] F. Leyendecker, F. Jesser, Tetrahedron Lett. 1980, 21, 1311–1314.
[26] B. H. Lipshutz, E. L. Ellsworth, Tetrahedron Lett. 1988, 29, 893–898.
[27] D. E. Frantz, D. A. Singleton, J. P. Snyder, J. Am. Chem. Soc. 1997, 119,
3383–3384.
The general procedure for the competitive coupling of a pair of
n-PentBr 2a and 2b–g with a cuprate is given below for the
coupling of n-butyl (phenyl) cuprate 1ab with n-PentBr 2a and
PhCH2Br 2g. The scale for competition is 1:7:7 for cuprate:
2a:2b–g. To a flame-dried two-necked and round bottom flask
equipped with a septum cap and a stirring bar, CuI (0.5 mmol) in
3 cm3 of THF was added at ꢂ20 8C and n-BuMgBr (0.5 mmol) was
placed by a syringe and the mixture was stirred for 2 min. To the
light yellow n-BuCu suspension was added PhMgBr (0.5 mmol) by
syringe at ꢂ20 8C. Heterogeneous solution of n-BuPhMgBr 1ab
was stirred at that temperature for 15 min. To the cuprate 1ab, a
mixture of n-PentBr 2a (3.5 mmol) and PhCH2Br (3.5 mmol)
prepared separately was added rapidly. The flask was removed
from the cooling bath and the reaction mixture was stirred
at 25 8C for an appropriate time (5, 10, 15,. . ., 40 min or 10, 20, 30,
40, 50 min or 15, 30, 45, 60 min). The internal standard,
di-n-pentylether (0.9918 mmol, 0.2 cm3) was added and the
mixture was hydrolyzed with saturated NH4Cl solution containing
20% NH3. The aqueous phase was extracted with diethyl ether
and the product mixture was analyzed by GLC.
[28] M. Yamanaka, E. Nakamura, J. Am. Chem. Soc. 2001, 123, 5675–5681.
[29] M. Yamanaka, E-i. Nakamura, J. Am. Chem Soc. 2005, 127, 4697–4706.
¨
¨
[30] E. Erdik, O. Omu¨r Pekel, J. Organometal Chem. 2008, 693, 338–342.
¨
¨
[31] E. Erdik, O. Omu¨r Pekel, Tetrahedron Lett. 2009, 50, 1500–1503.
[32] E. Nakamura, S. Mori, Angew. Chem. Int. Ed. 2000, 39, 3750–3771.
[33] C. R. Johnson, G. A. Dutra, J. Am. Chem. Soc. 1973, 95, 7783–7788.
[34] R. G. Pearson, C. D. Gregory, J. Am. Chem. Soc. 1976, 98, 4088–4094.
[35] W. J. Spanenberg, B. E. Snell, M. C. Su, Microchem. J. 1993, 47, 79–89.
[36] S. H. Bertz, S. Cope, D. Donna, M. Murphy, C. A. Ogle, Angew. Chem. Int.
Ed. 2007, 46, 7082–7085.
[37] H. L. Goering, S. Kantner, E. P. Seitz,, Jr., J. Org. Chem. 1985, 50,
5495–5499.
[38] A. Guijarro, R. D. Rieke, Angew. Chem. Int. Ed. 1998, 37, 1679–1681.
[39] L. Shi, Y. Chu, P. Knochel, H. Mayr, Angew. Chem. Int. Ed. 2008, 47,
202–204.
[40] H. Yamataka, N. Miyano, T. Hanafusa, J. Org. Chem. 1991, 56,
2573–2575.
[41] H. Yamataka, N. Fujimoro, Y. Kawafuji, T. Hanafusa, J. Am. Chem. Soc.
1967, 109, 4305–4308.
[42] H. Yamataka, T. Matsuyama, T. Hanafusa, J. Am. Chem. Soc. 1989, 131,
4912–4918.
[43] C. H. Ho, T. C. Lau, New J. Chem. 2000, 859–863.
[44] A. Guijarro, D. M. Rosenberg, R. D. Rieke, J. Am. Chem. Soc. 1999, 121,
4155–4167.
[45] L. Keinicke, P. Fristrup, P.-O. Norby, R. Madsen, J. Am. Chem. Soc. 2005,
127, 15756–15761.
Acknowledgements
We thank the Turkish Scientific and Technical Research Council
(grant no. TBAG 106T644) for the financial support.
[46] J. A. McPhee, J. E. Dubois, Tetrahedron Lett. 1972, 467–470.
[47] M. J. Aurell, M. J. Banuls, R. Mestres, E. Monuz, Tetrahedron 2001, 57,
1067–1074.
[48] F. Kynes, K. E. Bergmann, J. T. Welch, J. Org. Chem. 1998, 63,
2824–2828.
REFERENCES
[49] M. Makosza, O. Lobanova, A. Kwast, Tetrahedron 2004, 60,
2577–2581.
[50] K. M. Maclin, H. G. Richey, J. Org. Chem. 2002, 67, 4370–4371.
[1] N. Krause, Modern Organocopper Chemistry, Wiley-VCH, Weinheim,
2002.
[2] R. J. K. Taylor, Organocopper Reagents: A Practical Approach, Oxford
University Press, Oxford, 1994.
[3] E. J. Corey, D. J. Beames, J. Am. Chem. Soc. 1972, 94, 7210–7211.
[4] H. O. House, M. J. Umen, J. Org. Chem. 1973, 38, 3893–3901.
[5] W. H. Mandeville, G. M. Whitesides, J. Org. Chem. 1974, 39, 400–405.
[6] F. Aravalo, L. Castedo, B. R. Fernandez, A. Mourino, L. Sarandeses,
Chem. Lett. 1988, 745–748.
[7] H. Malmberg, M. Nilsson, C. Ullenius, Tetrahedron Lett. 1982, 23,
3823–3826.
[8] B. H. Lipshutz, J. A. Kozlowski, D. A. Parker, S. L. Nguyen, K. E.
McCarthy, J. Organomet. Chem. 1985, 285, 437–447.
[9] J. P. Gorlier, L. Hamon, J. Levisalles, J. Wagnon, J. Chem. Soc. Chem.
Commun. 1973, 88–89.
˘
ˆ
[51] E. Erdik, F. Eroglu, D. Kahya, J. Phys. Org. Chem. 2005, 18, 950–956.
¨
¨
[52] E. Erdik, O. Omu¨r, Appl. Organomet. Chem. 2005, 29, 887–893.
[53] P. Zuman, R. C. Patel, Techniques in Organic Reaction Kinetics, Wiley,
New York, 1984.
[54] N. S. Isaacs, Physical Organic Chemistry, Longmans, Harlow, 1987.
[55] J. H. Espenson, Chemical Kinetics and Reaction Mechanisms, Mc-Graw
Hill, Toronto, 1995.
[56] F. Ruff, J. G. Csizmadia, Organic Reactions, Equilibria, Kinetics and
Mechanism, Elsevier, New York, 1994.
[57] F. D. Pruchnik, Organometallic Chemistry of the Transition Elements
(Chapter 11), Plenum Press, New York, 1990.
[58] J. P. Collman, L. S. Hegedus, Principles and Applications of Organo-
transition Metal Chemistry (Chapter 14.1), University Science Books,
Mill Valley, CA, 1987.
[10] H. O. House, C.-Y. Chu, J. M. Wilkins, M. J. Umen, J. Org. Chem. 1975, 40,
1460–1469.
[59] D. F. Shriver, P. W. Atkins, C. H. Langford, Inorganic Chemistry (Chapter
6.3), Oxford University Press, London, 1992.
[11] G. H. Posner, C. E. Whitten, J. J. Sterling, D. J. Brunelle, Tetrahedron Lett.
1974, 30, 2591–2594.
[60] J. A. McPhee, M. Boussu, J.-E. Dubois, J. Chem. Soc. Perkin III 1974,
1525–1530.
[61] J. Leonard, B. Lygo, G. Procter, Advanced Practical Organic Chemistry,
Blackie, London, 1995.
[62] R. N. Keller, H. D. Wycoff, Inorg. Synth. 1946, 2, 1–4.
[63] D. D. Perrin, W. L. F. Armarego, Purification of Laboratory Chemicals,
Pergamon Press, Oxford, 1988.
[12] B. H. Lipshutz, R. S. Wilhelm, J. A. Kozlowski, Tetrahedron 1984, 40,
5005–5038.
[13] B. H. Lipshutz, Synthesis 1987, 325–341.
[14] E-i. Nakamura, N. Yoshikai, Bull. Chem. Soc. Jpn 2004, 77, 1–12.
[15] P. Knochel, C. Piazza, Angew. Chem. Int. Ed. 2002, 41, 3263–3265.
[16] S. H. Bertz, G. Dabbagh, J. Chem. Soc. Chem. Commun. 1982,
1030–1032.
[64] H. J. Barber, J. Chem. Soc. 1943, 1, 79.
[65] C. H. Watson, J. F. Eastham, J. Organomet. Chem. 1967, 9, 165–168.
[17] S. H. Bertz, G. Dabbagh, G. M. Villacorta, J. Am. Chem. Soc. 1982, 104,
5824–5826.
www.interscience.wiley.com/journal/poc
Copyright ß 2009 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 1148–1154