Reactivity of mixed organozinc and mixed organocopper reagents: 11. Nickel-catalyzed atom-economic aryl-allyl coupling of mixed (n-alkyl)(aryl)zincs

Article abstract of DOI:10.1002/aoc.3192

Group selectivity in the allylation of mixed (n-butyl)(phenyl)zinc reagent can be controlled by changing reaction parameters. CuCN-catalyzed allylation in tetrahydrofuran (THF)-hexamethylphosphoric triamide is n-butyl selective and also γ-selective in the presence of MgCl₂, whereas CuI-catalyzed allylation in THF in the presence of n-Bu₃P takes place with a n-butyl transfer:phenyl transfer ratio of 23:77 and an α:γ transfer ratio of phenyl of 76:24. NiCl₂(Ph₃P) ₂-catalyzed allylation in the presence of LiCl is phenyl selective with an α:γ ratio of 65:35. The reaction of methyl- or n-butyl(aryl)zinc reagents with an allylic electrophile in THF at room temperature in the presence of NiCl₂(Ph₃P)₂ catalyst and LiCl as an additive provides an atom-economic alternative to aryl-allyl coupling using diarylzincs. Copyright

Full text of DOI:10.1002/aoc.3192

Full Paper
Received: 22 May 2014
Revised: 30 June 2014
Accepted: 1 July 2014
Published online in Wiley Online Library: 28 July 2014
(wileyonlinelibrary.com) DOI 10.1002/aoc.3192
Reactivity of mixed organozinc and mixed
organocopper reagents: 11. Nickel-catalyzed
atom-economic aryl–allyl coupling of mixed
(n-alkyl)(aryl)zincs†,‡
Melike Kalkan*
Group selectivity in the allylation of mixed (n-butyl)(phenyl)zinc reagent can be controlled by changing reaction parameters.
CuCN-catalyzed allylation in tetrahydrofuran (THF)–hexamethylphosphoric triamide is n-butyl selective and also γ-selective in
the presence of MgCl2, whereas CuI-catalyzed allylation in THF in the presence of n-Bu3P takes place with a n-butyl transfer:phenyl
transfer ratio of 23:77 and an α:γ transfer ratio of phenyl of 76:24. NiCl2(Ph3P)2-catalyzed allylation in the presence of LiCl is phenyl
selective with an α:γ ratio of 65:35. The reaction of methyl- or n-butyl(aryl)zinc reagents with an allylic electrophile in THF at room
temperature in the presence of NiCl2(Ph3P)2 catalyst and LiCl as an additive provides an atom-economic alternative to aryl–allyl
coupling using diarylzincs. Copyright © 2014 John Wiley & Sons, Ltd.
Keywords: alkylarylzincs; mixed diorganozincs; allylation; Ni catalyst
on the reaction parameters, such as solvent and catalyst. They also
Introduction
developed new atom-economic procedures for C–alkyl,[40–42,44,45]
C–N[39] and C–acyl[37,38,45,46] coupling reactions using mixed diorga-
Transition metal-catalyzed reactions of organozinc reagents RZnX
and R2Zn are commonly used in organic synthesis due to their high
tolerance to functional groups and efficiency toward many
electrophiles.[1–3] Diorganozincs, R2Zn, are more reactive than
monoorganozincs, RZnX; however, R2Zn reagents are not atom-
economic and they transfer only one of the R groups efficiently to
an electrophile. The problem has been solved by developing mixed
diorganozincs, R1R2Zn, in which the R2 group has a lower transfer
rate than the R1 group.[4–18]
Recently, mixed diorganozincs, RRRTZn, which have one transfer-
able group RT together with the residual group RR, have been devel-
oped for synthetic purposes.[19,20] Whereas mixed diorganozincs,
RRRTZn, have found widespread use in asymmetric 1,2-addition[19–22]
and 1,4-addition reactions,[23–26] their C–C[27–29] and C–heteroatom
coupling reactions[9] are rare. Mixed alkylzinc-derived zinc cuprate
reagents were also used.[27]
nozincs RRRTZn.
In connection with these studies, we focused our interest on
exploring the effect of reaction parameters on the group selectivity
and regioselectivity in the reactions of mixed alkylarylzincs with
allylic halides, and also on developing an atom-economic procedure
for aryl–allyl coupling using mixed arylzincs.
Transition metal-catalyzed allylic substitution is one of the most
valuable methods for the formation of C―C bonds.[47] Until now,
great progress has been made in controlling the regiochemistry
and enantioselectivity of copper-catalyzed alkylzinc–allyl coupling
reactions. Alkylzinc reagents undergo exclusively γ-selective
allylation in the presence of a Cu catalyst whereas α-selective
allylation is observed in the presence of a Ni or Pd catalyst.
However, less attention has been given to the regioselective and
enantioselective allylic substitution of arylzinc reagents.
The widely accepted hypothesis for the group selectivity of
mixed cuprates R1R2CuLi is that the group in R1R2CuLi that has a
stronger bond to Cu acts as the group of lower selectivity.[30–33]
However, theoretical studies of Yamanaka and Nakamura on the
substitution[34] and 1,4-addition reactions[35] of R(X)CuLi reagents
(X= tert-Bu, alkynyl, CN, RS) showed that the group selectivity is
controlled by several factors, such as thermodynamic stability and
kinetic reactivity of the triorganocopper(III) intermediates formed
in these reactions. Peculiar properties of RR groups in RRRTCuLi
reagents in the formation of π-complexes of cuprates in their
1,4-addition reactions have been also reported by Bertz et al.[36]
However, Erdik and co-workers in their serial work[37–46] on the re-
activity and group selectivity of mixed organometallics reported that
the group selectivity of mixed diorganozincs R1R2Zn,[37–39,41–43,45]
mixed diorganocuprates R1R2CuM (M = MgBr,[40,44,45] ZnCl[46]) and
Cu-catalyzed mixed triorganozincates R1(R2)2ZnMgBr[41,45] depends
Recently, in our kinetic study of the reaction of diorganozincs with
allyl bromide in THF at 25°C, it was observed[43] that the reaction with
n-BuPhZn results in quantitative yield with a ratio of 13:87 for n-Bu
transfer:Ph transfer, and allylation of n-Bu2Zn and Ph2Zn respectively
gives the corresponding allylated products in a quantitative total yield.
Herein, successful results on the control of group selectivity and
regioselectivity in the allylation of alkylarylzinc reagents allowing an
atom-economic allylic coupling of alkylarylzincs are reported.
*
Correspondence to: Melike Kalkan, Ankara University, Science Faculty, Beşevler,
Ankara 06100, Turkey. E-mail: mkalkan@ankara.edu.tr
Ankara University, Science Faculty, Beşevler, Ankara 06100, Turkey
†
‡
The number of the serial work has been used with the permission of Ender Erdik.
This paper is dedicated to my PhD supervisor Prof. Dr. Ender Erdik.
Appl. Organometal. Chem. 2014, 28, 725–732
Copyright © 2014 John Wiley & Sons, Ltd.

M. Kalkan
Results and Discussion
As a model reaction, allylation of (n-butyl)(phenyl)zinc (1ab) with
(E)-crotyl chloride (2) in THF was selected. Firstly, the effect of
reaction parameters on the group selectivity and regioselectivity
in the reaction was investigated (Scheme 1).
Magnesium-based organozinc reagents, i.e. n-butyl- and
phenylmagnesium bromides to be transmetallated, were used.[37]
Phenylzinc chloride prepared by transmetallation of phenyl-
magnesium bromide with ZnCl2 was allowed to react with n-
butylmagnesium bromide and this one-pot successive procedure
leads to formation of (n-Bu)(Ph)Zn. The same procedure can be also
applied by reacting n-butylzinc bromide with phenylmagnesium
bromide. Mixing equimolar amounts of n-Bu2Zn (1a2) and Ph2Zn
(1b2) also favors the formation of 1ab. However, it was already
Scheme 1. Group selective allylation of n-BuPhZn reagent (1ab) with (E)-crotyl
chloride 2 in THF.
Table 1. Group selectivity and regioselectivity in the allylation of n-butylphenylzinc reagent n-BuPhZn (1) with (E)-crotyl chloride (2) in THF in the pres-
ence of Cu, Pd and Ni catalystsa
Entry
Catalyst (mol%)
Coupling
yield (%)b
Group selectivityc
n-Bu coupling:Ph coupling
Regioselectivity
6a:7ad
6b:7be
1
CuI (5)
98
81
42:58
42:58
5:95
6:94
5:95
4:96
3:97
4:96
5:95
0:100
3:97
0:100
3:97
4:96
5:95
6:94
2:98
8:92
—
53:47
43:57
46:54
29:71
42:58
35:65
48:52
55:45
55:45
56:44
42:58
51:49
76:24
66:34
56:44
70:30
67:33
58:42
56:44
59:41
2
CuI (10)
3
CuI (20)
CuI (100)f
93
44:56
4
99
48:52
5
CuCN (5)
96
61:39
6
CuCN (100)f
CuX (5)g
92
57:43
7
85–88
96
40:60 to 44:56
40:60
8
CuOAc (5)
CuX2 (5)h
9
92–97
~100
~100
99
39:61 to 45:55
42:58
10
11
12
13
14
15
16
17
18
19
20
CuBr.Me2S (5)
CuCN.2LiCl (5)
CuOTf.C6H6 (5)
CuI (5), n-Bu3P (5)
CuI (5), Ph3P (5)
CuCN (5), n-Bu3P (5)
CuCN (5), Ph3P (5)
Pd(OAc)2 (5)
NiCl2 (5)
62:38
48:52
~100
~100
90
23:77
32:68
52:48
98
39:61
24
0:100
45
0:100
—
NiCl2 (5), Ph3P (5)
NiCl2(Ph3P)2 (2.5)
82
0:100
—
76
0:100
—
aAll the data are the average of at least two experiments. The reactions were carried out on a 2mmol scale according to the conditions indicated by the
above equation, unless otherwise specified. Molar ratio of 1ab:2 was optimized to be 1.1:1.
bThe sum of GC yields of n-Bu coupling products (6a and 7a) and Ph coupling products (6b and 7b).
cThe ratio of GC yields of (6a+ 7a) and (6b+ 7b).
dThe ratio of GC yields of 6a and 7a.
eThe ratio of GC yields of 6b and 7b.
fStoichiometric cuprate reagent derived from n-BuPhZn and CuI (or CuCN) was prepared in situ.
gX = Cl, Br and SCN.
hX = Cl, OAc, OTf and acac.
wileyonlinelibrary.com/journal/aoc
Copyright © 2014 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2014, 28, 725–732

Reactivity of mixed organozinc and mixed organocopper reagents: 11
found that one-pot successive Mg to Zn transmetallation reactions
are more practical for the preparation of magnesium-based
diorganozincs.[37] In addition, the group originally attached to Zn,
which is n-Bu or Ph, does not change the relative transfer ability of
these groups. The reaction was carried out by adding allylic
substrate 2 to the organozinc reagent 1ab in THF in the presence
of a transition metal catalyst and/or organic catalyst. α-Selective and
γ-selective transfer abilities of n-Bu and Ph groups were determined
by finding the GC yields of coupled products 6a, 7a, 6b and 7b.
The reaction temperature and time were optimized to be room tem-
perature and 1 h, respectively. As transition metal catalysts, the mostly
used Cu, Ni and Pd catalysts for organozinc reagents were studied.
Donor solvents were used as co-solvents and some Lewis base and
Lewis acid reagents were also tested as additives.[37,45,48]
Uncatalyzed allylation of 1ab with 2 in THF at 25°C resulted in a
yield lower than 20%. The effects of Cu, Ni and Pd catalysts on the
group selectivities and regioselectivities of the allylation are
summarized in Table 1. With the optimized 5 mol% of CuI catalyst
(entry 1), quantitative yield of coupling was obtained. The group
selectivity, i.e. n-Bu transfer:Ph transfer ratio, is around 40:60 and
regioselectivities, i.e. α:γ ratios for n-Bu transfer and Ph transfer,
are 5:95 and 53:47, respectively. As expected, coupling of n-Bu
group gives preferential γ-selectivity even with the use of the mixed
1ab. Using one equiv. CuI, i.e. preparation in situ of cuprate, did not
lead to any change in the total coupling yield, and in the group
selectivity and also in the regioselectivity of n-Bu transfer. However,
Ph transfer results in a quite lower α:γ ratio (entry 4).
The source of copper has been reported to have a strong effect
on the reactivity and regioselectivity of catalyst in Cu-catalyzed
allylic alkylation with organometallic reagents.[49] So a number of
Cu(I) and Cu(II) salts (entries 5–9) and Cu(I) catalysts complexed with
a Lewis base such as CuBr.Me2S (entry 10) or in the presence of a
Lewis acid such as LiCl (entry 11) or a Lewis base such as R3P (entries
13–16) were also examined as Cu catalysts.
The use of CuCN in catalytic and stoichiometric amounts could
drive the allylation of 1ab quantitatively (entries 5 and 6) and
gave somewhat better group selectivity (n-Bu transfer:Ph transfer
ratio = 61:39). Other CuX catalysts (X = Cl, Br, SCN) (entry 7) did not
markedly change either group selectivity or regioselectivities;
however, CuOAc gave quantitative yield for total coupling (entry 8).
CuX2 catalysts (X = Cl, OAc, OTf, acac) (entry 9) led to quantitative
total coupling yields with n-Bu transfer:Ph transfer ratios and
also regioselectivities being identical to those obtained with
CuX catalysts. Allylation of 1ab with CuBr.Me2S (entry 10), with
CuCN.2LiCl (entry 11) or with CuOTf.C6H6 (entry 12) resulted in
quantitative total coupling with no marked change in the n-Bu
transfer:Ph transfer ratios compared to the ratios obtained with
non-coordinated catalysts CuCN and CuX (entries 5 and 7). It is
noteworthy that, in the presence of CuI and n-Bu3P or Ph3P as an
additive, 1ab can be allylated with a higher Ph transfer (n-Bu
transfer:Ph transfer ratio = 23:77 or 32:68) and with a better regioselec-
tivity in the Ph transfer (α:γ ratio = 76:24 or 66:34) (entries 13 and 14).
Similar results were obtained when using n-Bu3P or Ph3P as additives
in the CuCN-catalyzed allylation (entries 15 and 16).
Table 2. Effect of N- and O-donor co-solvents and Lewis bases on selectivity and regioselectivity in the allylation of n-BuPhZn (1) with (E)-crotyl chloride (2)
in the presence of CuI or CuCN catalysta
Entry
Catalyst
Solventb
Coupling yield (%)c
Group selectivityd
Regioselectivity
n-Bu coupling:Ph coupling
6a:7ae
6b:7bf
1
CuI
THF–NMP
89
95
84
88
100
96
12
87
84
79
84
72:28
82:18
74:26
80:20
48:52
44:56
58:42
39:61
32:68
80:20
85:15
8:92
7:93
7:93
7:93
6:94
5:95
0:100
3:97
4:96
8:92
7:93
80:20
63:37
67:33
47:53
79:21
74:26
100:0
58:42
82:18
65:35
62:38
2
CuI
THF–HMPA
3
CuI
THF–DMPU
4
CuI
THF–diglyme
THF–DMF
5
CuI
6
CuI
THF–DMSO
7
CuI
THF–TMEDA
THF and 2,2-bipyridylg
THF and urotropineg
THF–NMP
8
CuI
9
CuI
10
11
CuCN
CuCN
THF–HMPA
aMolar ratio of 1ab:2 is 1.1:1.
bTHF–co-solvent volume ratio was optimized to be 4:1. In the case of HMPA, decreasing this ratio to 2:1 decreased the coupling to 71%.
cThe sum of GC yields of n-Bu coupling products (6a and 7a) and Ph coupling products (6b and 7b).
dThe ratio of GC yields of (6a+ 7a) and (6b+ 7b).
eThe ratio of GC yields of 6a and 7a.
fThe ratio of GC yields of 6b and 7b.
g1 mol equiv. vs. 2 was used.
Appl. Organometal. Chem. 2014, 28, 725–732
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. Kalkan
As seen, all Cu catalysts are unsuccessful in group selective
allylation of 1ab with 2. The highest n-Bu transfer was obtained in
the presence of CuCN catalyst (n-Bu transfer:Ph transfer ratio = 61:39)
(entry 5). The highest Ph transfer was observed in the presence
of CuI catalyst with n-Bu3P as an additive (n-Bu transfer:Ph transfer
ratio= 23:77) (entry 13).
Allylation of 1ab with 2 in THF in the presence of Pd(Ph3P)4, PdCl2
or PEPPSI–IPr catalysts under the optimized conditions did not pro-
ceed to completion with a yield higher than 7%. Use of Pd(OAc)2 as
catalyst also gave quite a low yield; however, Ph selective coupling
was observed (entry 17). Allylation of 1ab with Ni catalysis provided
a surprising result. The use of NiCl2 (entry 18), NiCl2 and Ph3P (entry
19) or NiCl2(Ph3P)2 (entry 20) as catalysts resulted in Ph selective
allylation with moderate to high yields, but with a pretty low α:γ
ratio of around 60:40.
CuI-catalyzed allylation of 1ab with 2 in THF was also carried out
in the presence of N- and O-donor solvents and Lewis base reagents
as additives (Table 2). Screening of N-methyl-2-pyrrolidone (NMP),
hexamethylphosphoric triamide (HMPA), N,N-dimethylpropylene urea
(DMPU), N,N-dimethylformamide (DMF), diethyleneglycol dimethyl
ether (diglyme) and dimethylsulfoxide (DMSO) as co-solvents
(entries 1–6) in the optimized THF–co-solvent ratio of 4:1 led to
the finding that n-Bu transfer is improved in a coordinating solvent
except DMF and DMSO (entries 5 and 6). HMPA, being the solvent
of choice, gave total coupling with a yield of 95% and n-Bu transfer:
Ph transfer ratio of 82:18. As observed in the reaction of 1ab with
allyl bromide,[45] N,N,N′,N′-tetramethylethylenediamine (TMEDA)
decreased both the total coupling yield and n-Bu transfer (entry
7). 2,2′-Bipyridyl and urotropine were also tested (entries 8 and 9)
as Lewis bases in optimized equimolar amounts to allylic substrate
2. It was interesting to see that urotropine, in contrast to donor
solvents, increased Ph transfer selectivity. All Lewis bases resulted
in a higher α:γ ratio in Ph transfer. Additional screening of HMPA
and NMP in the presence of CuCN catalyst (entries 10 and 11) led
to similar group selective and regioselective outcomes of the
reaction as obtained in the presence of CuI. Non-coordinating
solvents such as toluene as a co-solvent (2:1) did not change the
total coupling (95%) and n-Bu transfer:Ph transfer ratio (33:67).
The effect of MgCl2 and LiCl as Lewis catalysts was also investi-
gated in the CuI- and CuCN-catalyzed allylation of 1ab in THF
(Table 3). In fact, the reaction medium is not Mg-free, but Li-free.
Under the optimized conditions, MgCl2 or LiCl were used in
equivalent amounts vs. substrate in the presence of donor solvents.
As seen, in the CuI-catalyzed allylation, the addition of MgCl2 increases
the n-Bu transfer:Ph transfer ratio from 42:58 (Table 1, entry 1) to 67:33
(Table 3, entry 1). In the presence of NMP or HMPA and MgCl2, the
n-Bu transfer:Ph transfer ratio reached 87:13 (entries 2 and 3). The
addition of LiCl did not result in a group selectivity higher
(entries 4–6) than those obtained in the presence of MgCl2.
Meanwhile, CuCN-catalyzed allylation was completed with the
highest n-Bu transfer:Ph transfer ratio using NMP (or HMPA)
and MgCl2, i.e. 98:2 (entries 8 and 9, respectively), or using HMPA
and LiCl, i.e. 86:14 (entry 12).
Allylation of stoichiometric iodozinc n-butyl phenylcuprates,
n-BuPhCuZnI, prepared from 1 equiv. n-BuPhZn and 1 equiv. CuI
in THF, was also carried out to prove the formation of catalytic
Table 3. Optimization of conditions in the CuI- and CuCN-catalyzed allylation of n-BuPhZn (1) with (E)-crotyl chloride (2) in THF in the presence of
a co-solvent and/or Lewis acida
Entry
Catalyst
Co-solvent, Lewis acidb
Coupling yield (%)c
Group selectivityd
Regioselectivity
n-Bu coupling:Ph coupling
6a:7ae
6b:7bf
1
CuI
—, MgCl2
NMP, MgCl2
HMPA, MgCl2
—, LiCl
93
87
89
96
83
89
93
71
81
88
89
94
67:33
87:13
87:13
67:33
82:18
87:13
76:24
97:3
6:94
7:93
6:94
6:94
6:94
6:94
8:92
7:93
6:94
6:94
6:94
7:93
39:61
55:45
50:50
56:44
60:40
50:50
27:73
50:50
50:50
48:52
53:47
38:62
2
CuI
3
CuI
4
CuI
5
CuI
NMP, LiCl
HMPA, LiCl
—, MgCl2
NMP, MgCl2
HMPA, MgCl2
—, LiCl
6
CuI
7
CuCN
CuCN
CuCN
CuCN
CuCN
CuCN
8
9
98:2
10
11
12
76:24
81:19
86:14
NMP, LiCl
HMPA, LiCl
aMolar ratio of 1ab:2 is 1.1:1.
b1 mol equiv. to 2 was used.
cThe sum of GC yields of n-Bu coupling products (6a and 7a) and Ph coupling products (6b and 7b).
dThe ratio of GC yields of (6a+ 7a) and (6b+ 7b).
eThe ratio of GC yields of 6a and 7a.
fThe ratio of GC yields of 6b and 7b.
wileyonlinelibrary.com/journal/aoc
Copyright © 2014 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2014, 28, 725–732

Reactivity of mixed organozinc and mixed organocopper reagents: 11
cuprates as intermediates in the CuI-catalyzed allylations of 1ab
reagents. The stoichiometric cuprate afforded 94% total coupling
with a ratio of 51:49 for n-Bu transfer:Ph transfer and γ-selective
n-Bu transfer. The use of LiCl as an additive increased the n-Bu
transfer:Ph transfer ratio to 73:27 and the use of HMPA as a co-
solvent and LiCl resulted in exclusively n-Bu transfer (91:9). Similarity
of these results to the data in Table 1 (entry 1), Table 2 (entry 2) and
Table 3 (entry 6), which were obtained in the allylation of n-BuCuZn
in the presence of CuI catalyst, provides a proof for the expected in
situ formation of catalytic mixed cuprates.
Donor solvents, Lewis base and Lewis acid additives were also
used in the Ni-catalyzed Ph selective allylation of 1ab with 2 with
the aim of finding better coupling yield and/or better regioselec-
tivity (Table 4). The use of donor solvents (entries 1–4) such as
NMP, HMPA, DMPU, diglyme, DMF and DMSO is ineffective in alter-
ing the Ph coupling yield; however, using HMPA, DMF or DMSO
resulted in a small increase in the α:γ ratio (entries 1, 3 and 4).
Urotropine and MgCl2 decreased the coupling yield (entries
6 and 7). The use of LiCl resulted in the highest coupling yield,
i.e. 83%, for Ph transfer with an α:γ ratio of 65:35 (entry 8). In
the presence of HMPA and LiCl, the α:γ ratio increased to
73:27; however, this increase in regioselectivity is within the
limit of experimental error.
(1 equiv. vs. 2). Total coupling yield is 81% and 94% and
n-butyl transfer:Ph transfer ratio is 98:2 and 86:14, respec-
tively (Table 3, entries 9 and 12). The n-butyl transfer takes
place with γ-selectivity.
(ii) The allylation in the presence of CuI (5 mol%) catalyst and
n-Bu3P (5mol%) (Table 1, entry 13) or urotropine (1mol
equiv.) (Table 2, entry 9) in THF takes place with a total
coupling yield of ca 100% and 84% and with a n-Bu trans-
fer:Ph transfer ratio of 23:77 and 32:68, respectively. Phenyl
transfer results in moderate α-selectivity (α:γ = 76:24 and
82:18, respectively).
(iii) The allylation takes place with complete Ph transfer in the
presence of NiCl2(Ph3P)2 (2.5mol%) catalyst in THF with a
yield of 76% and an α:γ ratio of 59:41 (Table 1, entry 20).
Using LiCl (1 equiv. vs. 2) increases the coupling yield to
83% and the α:γ ratio to 65:35 (Table 4, entry11).
After having determined the reaction conditions for n-alkyl or
aryl transfer to an allylic substrate using a mixed n-alkylarylzinc
reagent, we also found the coupling yields and regioselectivity in
the allylation of homodiorganozincs, n-Bu2Zn (1a2) and Ph2Zn
(1b2), with 2 under the optimized conditions (Table 5). Uncatalyzed
allylation of 1a2 and 1b2 did not take place. Coupling yields of 1a2
in the CuI- or CuCN-catalyzed reactions (entries 1 and 2) are not
different from those of 1ab. Coupling in the presence of a donor
solvent resulted in somewhat higher yields (entries 3 and 4). Ni
catalyst did not allow an alkyl–allyl coupling (entry 5).
Optimization of the effects of reaction parameters to control the
group selectivity and regioselectivity in the Cu- and Ni-catalyzed
allylation of 1ab with 2 has revealed the following points:
(i) The allylation is n-Bu group-selective in the presence of CuCN
(5 mol%) catalyst in THF–HMPA (4:1) and MgCl2 or LiCl
Allylation of 1b2 takes place quantitatively in the presence of
both Cu and Ni catalysts (entries 6–8). Carrying out the coupling
Table 4. Effect of Lewis bases and Lewis acids on the group selectivity and regioselectivity in the allylation of n-BuPhZn
(1) with (E)-crotyl chloride (2) in THF in the presence of NiCl2(Ph3P)2 catalysta: optimization of conditions
Entry
Cosolvent,
Lewis acidb
Coupling
yield (%)c
Group selectivityd
n-Bu coupling:
Ph coupling
Regioselectivity
6a:7ae
6b:7bf
1
2
3
4
5
6
7
8
9
10
HMPA
77
68
78
68
74
40
35
83
72
52
0:100
0:100
3:97
—
—
—
—
—
—
—
—
—
—
68:32
59:41
71:29
69:31
54:46
60:40
54:46
65:35
73:27
74:26
Diglyme
DMF
DMSO
0:100
0:100
0:100
0:100
0:100
3:97
2,2′-Bipyridylb
Urotropineb
—, MgCl2
—, LiCl
HMPA, LiCl
DMF, LiCl
4:96
aMolar ratio of 1ab:2 is 1.1:1.
b1 mol equiv. vs. 2 was used.
cThe sum of GC yields of n-Bu coupling products (6a and 7a) and Ph coupling products (6b and 7b).
dThe ratio of GC yields of (6a+ 7a) and (6b+ 7b).
eThe ratio of GC yields of 6a and 7a.
fThe ratio of GC yields of 6b and 7b.
Appl. Organometal. Chem. 2014, 28, 725–732
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. Kalkan
Table 5. Allylation of n-Bu2Zn (1a2) and Ph2Zn (1b2) with (E)-crotyl chloride (2) in the presence of Cu or Ni catalyst in THFa: effect of co-solvent
Entry
R2Zn
Catalyst
Co-solvent
Coupling
yield (%)b
Regioselectivity
6a:7ac
6b:7bd
1
2
3
4
5
6
7
8
9
10
n-Bu2Zn
n-Bu2Zn
n-Bu2Zn
n-Bu2Zn
n-Bu2Zn
Ph2Zn
CuI
—
—
82
92
7:93
7:93
7:93
6:94
—
CuCN
CuI
HMPA
Diglyme
—
91
CuI
99
NiCl2(Ph3P)2
CuI
—
—
100
100
100
80
60:40
42:58
56:44
55:45
75:25
Ph2Zn
CuCN
NiCl2(Ph3P)2
CuI
—
Ph2Zn
—
Ph2Zn
HMPA
Diglyme
Ph2Zn
CuI
85
aMolar ratio of 1a2:2 and 1b2:2 is 1.1:1.
bThe sum of GC yields of n-Bu coupling products (6a and 7a) and Ph coupling products (6b and 7b).
cThe ratio of GC yields of 6a and 7a.
dThe ratio of GC yields of 6b and 7b.
Table 6. Ni-catalyzed coupling of n-butylarylzincs or methylarylzincs with allylic substrates in THF in the presence of LiCla
Entry
R1
R2
R, A (2a, 2b, 2c)
Coupling
yield (%)b
Regioselectivity
6:7c
1
n-Bu
n-Bu
Me
C6H5
Me, Cl
80
77
80
76
85
95
69
77
82
63
72
79
72:28
75:25
66:34
73:27
71:29
80:20
86:14
76:24
60:40
56:44
67:33
56:44
2
4-MeC6H4
3-MeC6H4
4-tert-BuC6H4
4-MeOC6H4
3-MeOC6H4
C6H5
Me, Cl
3
Me, Cl
4
n-Bu
n-Bu
n-Bu
Me
Me, Cl
5
Me, Cl
6
Me, Cl
7
C6H5, Cl
C6H5, OAc
C6H5, OAc
C6H5, OAc
C6H5, OAc
C6H5, OAc
8
Me
C6H5
9
Me
4-MeC6H4
3-MeC6H4
4-MeOC6H4
3-MeOC6H4
10
11
12
Me
Me
Me
aReactions were run with a 1.1:1 molar ratio of 1:2. General reaction conditions: R1R2Zn reagent (10 mmol) in THF, allylic substrate (9mmol), NiCl2(Ph3P)2
(0.25 mol), LiCl (9mmol) at room temperature for 1 h. See Experimental section.
bIsolated yield of product mixture of 6 and 7.
cThe α:γ ratio was determined using 500 MHz 1H NMR analysis and also using GC analysis.
wileyonlinelibrary.com/journal/aoc
Copyright © 2014 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2014, 28, 725–732

Reactivity of mixed organozinc and mixed organocopper reagents: 11
in the presence of a donor solvent led to lower yields (entries 9 and
10) as observed in the yield of Ph transfer of 1ab reagents.
This study showed that in the Cu-catalyzed allylation of 1ab, both
n-Bu and Ph group are transferred to the allylic substrate 2 in a ratio
of ca 1:1 with total quantitative coupling yield. However, reaction
parameters (transition metal catalyst, Lewis base and/or Lewis acid)
can be adjusted for selective n-butyl transfer or phenyl transfer to
take place with moderate to high regioselectivities.
As a result, the use of mixed (n-alkyl)(aryl)Zn reagents in allylation
allows coupling of either n-alkyl or aryl group. CuCN (5mol%)-cata-
lyzed allylation of mixed (n-alkyl)PhZn reagents in THF–HMPA (4:1)
with MgCl2 (1 equiv.) at 25°C seems a prominent γ-selective proce-
dure for n-alkyl–allyl coupling (Table 3, entry 9). This protocol pro-
vides an atom-economic alternative to allylation using (n-alkyl)2Zn
reagents if the n-alkyl group is cost-sensitive.
of group selectivity of diorganocuprates on the reaction
parameters.
(ii) A new and simple procedure for aryl–allyl coupling using Ni-
catalyzed allylation of methyl- or n-butyl(aryl)zincs provides
an atom-economic alternative to allylic coupling of diarylzincs.
Further studies are underway with the aim of increasing the
functional group tolerance and the regioselectivity in the allylation
of mixed arylzincs as well as investigating the mechanism of the
organic catalysts in the group selectivity.
Experimental
General
For the particular interest in developing an atom-economic
procedure for aryl–allyl coupling using mixed n-Bu(aryl)zincs,
two procedures can be applied: allylation in the presence of
CuI (5 mol%) catalyst and n-Bu3P (5 mol%) as an additive in
THF (Table 1, entry 13); or allylation in the presence of NiCl2
(Ph3P)2 (2.5 mol%) catalyst and LiCl (1 equiv.) as an additive
in THF (Table 4, entry 11). These both give good to high yields.
NiCl2(Ph3P)2-catalyzed allylation is more advantageous, being
aryl group-selective with an α:γ ratio of 65:35. This protocol
provides an atom-economic alternative to allylation using
(aryl)2Zn reagents if the aryl group is cost-sensitive.
All reactions were carried out in oven-dried glassware under a
positive pressure of nitrogen using standard syringe–septum cap
techniques.[50] GC analyses were performed using a Thermo
Finnigan gas chromatograph equipped with a ZB-5 capillary
column packed with phenylpolysiloxane using the internal stan-
dard technique. THF was distilled from sodium benzophe-
nonedianion. Alkyl bromides and bromobenzene were obtained
commercially and purified using literature procedures. Mg turnings
for Grignard reagents were used without further purification. ZnCl2
(Aldrich) was dried under reduced pressure at 100°C for 2 h and
used as a THF solution. CuI was purified according to the literature
procedure, dried under reduced pressure at 60–90°C for at least 1 h
and kept under nitrogen.[51] CuCN was purified according to a pub-
lished procedure.[52] Ni and Pd catalysts were used without further
purification. HMPA, NMP, DMPU, diglyme and TMEDA were distilled
under reduced pressure and kept over molecular sieves under
nitrogen. Toluene was distilled and kept over sodium under nitro-
gen. MgCl2 and LiCl were dried under reduced pressure.
Grignard reagents RMgBr (R= n-Bu, C6H5, FG-C6H4 (FG = 3-MeO,
4-MeO, 3-Me, 4-Me, 3-Br, 4-Br)) were prepared in THF by standard
methods and their concentrations were found by titration before
use.[53] For the preparation of (n-alkyl)(aryl)zinc reagents R1R2Zn
(R1 = Me, n-Bu; R2 = C6H5, FG-C6H4),[37] arylzinc chlorides R2ZnCl
were reacted with n-BuMgBr (or MeMgCl). R2ZnCl were prepared
by addition of arylmagnesium bromide (10 mmol) to ZnCl2
(10 mmol) in THF (10 ml) at À20°C with stirring at that temperature
for 15 min. To freshly prepared R2ZnCl reagent (10 mmol), n-
BuMgBr (or MeMgCl) (10 mmol) in THF was added dropwise and
the mixture was stirred at À20°C for another 15 min.
Thus, with the optimized conditions in hand, potential synthetic
transfer of aryl groups of (n-butyl)(aryl)zincs to allylic substrates was
investigated. Instead of (n-butyl)(phenyl)zinc, (methyl)(phenyl)zinc
was also examined in the allylation, which did not change the
outcome of the reaction. So, both n-butyl and methyl groups were
used as residual groups in the mixed alkylarylzincs to give aryl–allyl
coupling in the presence of NiCl2(Ph3P)2 as catalyst and LiCl. The
results are summarized in Table 6. The data are averages of at
least two independent experiments.[50] The products were fully
1
characterized using H NMR analysis, and the α:γ ratios of the
1
product mixtures were determined using H NMR analysis and
also by GC analysis. As allylic substrates, (E)-crotyl chloride, (E)-
cinnamyl chloride and (E)-cinnamyl acetate were screened. The
coupling yields are high (63–95%) and α:γ ratios for the coupling
product vary between 56:44 and 86:14 depending on the
coupling partners.
Coupling of alkyl- and methoxy-substituted aryl groups with 2
(entries 2–6) and with cinnamyl substrates 2b and 2c (entries 8–13)
proceeded smoothly and provided the target compounds. Side
products, such as methyl- and methoxy-substituted biphenyls could
not be completely suppressed. However, coupling of halogeno- and
acyl-substituted aryl groups was discouraged due to low yields of
products with formation of side products.
Synthetic Procedure for Ni-Catalyzed Allylation of (Alkyl)(aryl)
zincs R1R2Zn in the Presence of LiCl
To the prepared R1R2Zn reagent (10 mmol), NiCl2[(Ph)3P]2
(0.25 mmol, 0.16 g) was added at À20°C and stirred at that temper-
ature for another 15 min. LiCl (9mmol, 0.425g) was added and then
allylic substrate (9mmol) was added dropwise at À20°C. The
mixture was stirred at room temperature for 1 h. The mixture was
hydrolyzed with saturated NH4Cl solution. The aqueous layer was
extracted with ether. The organic combined solutions were dried
with Na2SO4, concentrated and subjected to silica gel column
chromatography with petroleum ether. A mixture of α- and
γ-products was obtained as a light yellow liquid. The ratio of
α-product to γ-product was determined using GC analysis and also
using 500 MHz 1H NMR analysis.
Conclusions
A range of reaction parameters were examined with the aim of
controlling the group selectivity and regioselectivity in the allylation
of (n-butyl)(phenyl)zinc. The following were demonstrated:
(i) In the Cu-catalyzed allylation of mixed (n-butyl)(phenyl)zinc,
group selectivity can be controlled by changing the reaction
parameters, and either the n-butyl or phenyl group can be
transferred to the allylic substrate. These findings also provide
another support for Erdik’s hypothesis of the dependence
Appl. Organometal. Chem. 2014, 28, 725–732
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. Kalkan
K. E. McCarthy, J. Organomet. Chem. 1985, 285, 437. e) J. P. Gorlier,
L. Hamon, J. Levisalles, J. Wagnon, J. Chem. Soc. Chem. Commun.
1973, 88. f) H. O. House, C.-Y. Chu, J. M. Wilkins, J. M. Umen, J. Org.
Chem. 1975, 40, 1460. g) G. H. Posner, C. E. Whitten, J. J. Sterling,
D. J. Brunelle, Tetrahedron Lett. 1974, 30, 2591. h) B. H. Lipshutz,
R. S. Wilhelm, J. A. Kozlowski, Tetrahedron 1984, 40, 5005. i)
B. H. Lipshutz, Synthesis 1987, 325.
Acknowledgements
The author thanks the Turkish Scientific and Technical Research
Council (grant no. KBAG 112T886) for financial support. I am grateful
to my PhD supervisor, Prof. Dr. Ender Erdik, for her help as a consul-
tant in this project and also in the preparation of this paper.
[33] a) W. H. Mandeville, G. M. Whitesides, J. Org. Chem. 1974, 39, 400. b)
S. H. Bertz, G. Dabbagh, J. Chem. Soc. Chem. Commun. 1982, 1030. c)
S. H. Bertz, G. Dabbagh, G. M. Villacorta, J. Am. Chem. Soc. 1982, 104,
5824. d) S. H. Bertz, G. Dabbagh, J. Org. Chem. 1984, 49, 1119. e)
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. f)
S. H. Bertz, M. Eriksson, G. Miao, P. S. Synder, J. Am. Chem. Soc. 1996,
118, 10906.
[34] M. Yamanaka, E. Nakamura, J. Am. Chem. Soc. 2005, 127, 4697.
[35] M. Yamanaka, E. Nakamura, J. Am. Chem. Soc. 2001, 123, 1703.
[36] S. H. Bertz, R. A. Hardin, M. D. Murphy, C. A. Ogle, J. D. Richter,
E. A. Thomas, Angew. Chem. Int. Ed. 2012, 51, 2681.
References
[1] Z. Rappoport, I. Marek (Eds), The Chemistry of Organozinc Compounds,
Patai Series, Wiley-VCH, Chichester, 2007.
[2] P. Knochel, Organozinc Reagents. A Practical Approach (Ed.: P. Jones),
Oxford University Press, Oxford, 1999.
[3] E. Erdik, Organozinc Reagents in Organic Synthesis, CRC Press, New
York, 1996.
[4] E. Laloe, W. Srebnik, Tetrahedron Lett. 1994, 35, 5587.
[5] J. B. Johnson, P. Yu, P. Fink, E. A. Bercot, T. Rovis, Org. Lett. 2006, 8, 4307.
[6] P. Wipf, S. Ribe, J. Org. Chem. 1998, 63, 6454.
[37] E. Erdik, Ö. Ö. Pekel, J. Organometal. Chem. 2008, 693, 338.
[38] E. Erdik, Ö. Ö. Pekel, Tetrahedron Lett. 2009, 50, 1501.
[39] E. Erdik, Ö. Ö. Pekel, M. Kalkan, Appl. Organometal. Chem. 2009, 23, 245.
[40] E. Erdik, D. Özkan, J. Phys. Org. Chem. 2009, 22, 1148.
[41] E. Erdik, E. Z. Serdar, J. Organometal. Chem. 2012, 703, 1.
[42] Ö. Ö. Pekel, E. Erdik, Tetrahedron Lett. 2011, 52, 7087.
[43] E. Erdik, M. Kalkan, J. Phys. Org. Chem. 2013, 26, 256.
[44] E. Erdik, D. Özkan, React. Kinet. Mech. Catal. 2013, 108, 293.
[45] E. Erdik, F. Eroglu, M. Kalkan, Ö. Ö. Pekel, D. Özkan, E. Z. Serdar,
J. Organometal. Chem. 2013, 745–746, 235.
[46] Ö. Ö. Pekel, E. Erdik, J. Organometal. Chem. 2014, 751, 644.
[47] a) R. M. Magid, Tetrahedron 1980, 36, 1901. b) B. H. Lipshutz,
S. Sengupta, Org. React. 1992, 41, 135. c) E. Negishi, F. Liu, in
Handbook of Organopalladium Chemistry for Organic Synthesis (Ed.:
E. Negishi), Wiley, New York, 2002, Chaps III.2.9 and III.2.10; d)
T. Takahashi, K. Kanno, in Modern Organonickel Chemistry (Ed.: Y.
Tamaru), Wiley-VCH, Weinheim, 2005, Chap. 2.3; e) R. Shintani,
T. Hayashi, in Modern Organonickel Chemistry (Ed.: Y. Tamaru), Wiley-
VCH, Weinheim, 2005, Chap. 9.2; f)A. Alexakis, C. Malan, L. Louise,
K. Tissot-Croset, D. Polet, Chimia Int. J. Chem. 2006, 60, 124. g)
C. A. Falciola, A. Alexakis, Chem. Eur. J. 2008, 14, 10615.
[48] a) L. Shi, Y. Chu, P. Knochel, H. Mayr, Org. Lett. 2009, 11, 3502. b)
N. Boudet, J. R. Lachs, P. Knochel, Org. Lett. 2007, 9, 5525. c) C.-Y. Liu,
H. Ren, P. Knochel, Org. Lett. 2006, 8, 617. d) C.-Y. Liu, P. Knochel, Org.
Lett. 2005, 7, 2543. e) H. Ren, A. Krasovskiy, P. Knochel, Org. Lett. 2004,
6, 4215.
[7] W. Oppolzer, R. N. Radinov, Helv. Chim. Acta 1992, 75, 170.
[8] B. H. Lipshutz, W. V. Randall, Tetrahedron Lett. 1999, 40, 2871.
[9] E. Hupe, P. Knochel, Org. Lett. 2001, 3, 127.
[10] C. Bolm, N. Herman, J. P. Hildebrand, K. Muniz, Angew. Chem. Int. Ed.
2000, 39, 3465.
[11] S. Özçubukçu, F. Schmidt, C. Bolm, Org. Lett. 2005, 7, 1407.
[12] J. Rudolph, C. Bolm, P. O. Norby, J. Am. Chem. Soc. 2005, 127, 1548.
[13] M. Fontes, X. Verdaguer, L. Sola, M. A. Pericas, A. Riera, J. Org. Chem.
2004, 69, 2532.
[14] M. Schinnerl, M. Seitz, A. Kaiser, O. Reiser, Org. Lett. 2001, 3, 4259.
[15] J. G. Kim, P. H. Walsh, Angew. Chem. Int. Ed. 2006, 45, 4175.
[16] W. Oppolzer, R. N. Radinov, J. Am. Chem. Soc. 1993, 115, 1593.
[17] M. Srebnik, Tetrahedron Lett. 1991, 32, 2449.
[18] S. Niwa, K. J. Soai, J. Chem. Soc., Perkin Trans. 1 1990, 937.
[19] S. Berger, F. Langer, C. Lutz, P. Knochel, T. A. Mobley, C. K. Reddy, Angew.
Chem. Int. Ed. 1997, 36, 1496.
[20] C. Lutz, P. Jones, P. Knochel, Synthesis 1999, 312.
[21] C. Lutz, P. Knochel, J. Org. Chem. 1997, 62, 7895.
[22] J. F. Trawerse, A. H. Hoveyda, M. Snapper, Org. Lett. 2003, 5, 3273.
[23] P. Jones, P. Knochel, J. Chem. Soc., Perkin Trans. 1 1997, 3117.
[24] P. Jones, C. K. Reddy, P. Knochel, Tetrahedron 1998, 54, 1471.
[25] C. K. Reddy, A. Devasagaraj, P. Knochel, Tetrahedron Lett. 1996,
37, 4495.
[26] A. Rimkus, N. Sewald, Org. Lett. 2002, 4, 3289.
[27] E. Hupe, M. I. Calaza, P. Knochel, J. Organometal. Chem. 2003, 680, 136.
[28] A. E. Jensen, P. Knochel, J. Org. Chem. 2002, 67, 79.
[29] D. Soorukram, P. Knochel, Org. Lett. 2004, 6, 2409.
[30] N. Krause, Modern Organocopper Chemistry, Wiley, Weinheim, 2000.
[31] Z. Rappoport, I. Marek (Eds), The Chemistry of Organocopper Com-
pounds, Patai Series, Wiley, Chichester, 2009.
[32] a) H. O. House, M. J. Umen, J. Org. Chem. 1973, 38, 3893. b) F. Aravalo,
L. Castedo, B. R. Fernandez, A. Mourino, L. Sarandeses, Chem. Lett. 1988,
745. c) H. Malmberg, M. Nilsson, C. Ullenius, Tetrahedron Lett. 1982, 23,
3823. d) B. H. Lipshutz, J. A. Kozlowski, D. A. Parker, S. L. Nguyen,
[49] a) H. Malda, A. W. Van Zijl, L. A. Arnold, B. L. Feringa, Org. Lett. 2001, 3,
1169. b) A. Alexakis, K. Crosset, Org. Lett. 2002, 4, 4147. c) W. J. Shi,
L.-X. Wang, Y. Fu, S. F. Zhu, Q. L. Zhou, Tetrahedron Asymm 2003, 14, 3867.
[50] J. Leonard, B. Lygo, G. Procter, Advanced Practical Organic Chemistry,
Blackie, London, 1995.
[51] R. N. Keller, H. D. Wyncoff, Inorg. Synth. 1946, 2, 1.
[52] H. J. Barber, J. Chem. Soc. 1943, 1, 79.
[53] S. C. Watson, J. F. Eastham, J. Organometal. Chem. 1967, 9, 165.
wileyonlinelibrary.com/journal/aoc
Copyright © 2014 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2014, 28, 725–732