The cyanation of aromatic halides catalyzed by nickel(0) complexes generated in situ. III. Kinetic and mechanistic aspects

Article abstract of DOI:10.1246/bcsj.77.1013

The cyanation of chlorobenzene catalyzed by [Ni(PPh₃) _n], which was generated in situ from [NiBr₂(PPh ₃)₂]-Zn-PPh₃ (Ni:Zn:P = 1:3:2 molar ratio), was carried out at 60 °C using KCN or NaCN in hexamethyl phosphoric triamide (HMPA). Kinetic analyses revealed that the reaction rates in the KCN-HMPA system were zero order for both of chlorobenzene and nickel concentrations, and two-thirds order for KCN. The determination of the kinetic order for NaCN was difficult due to its high solubility in HMPA since excess cyanide ion inhibited the reaction. Zinc bromide which was formed with [Ni(PPh₃)_n] from [NiBr₂(PPh₃) ₂]-Zn-PPh₃, acted as an important cocatalyst for the successful cyanation by comparison with [Ni(cod)₂] (cod = 1,5-cyclooctadiene)-PPh₃ system. Based on these experimental results, a reaction mechanism was proposed and discussed.

Full text of DOI:10.1246/bcsj.77.1013

ꢀ
2004 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 77, 1013–1019 (2004) 1013
The Cyanation of Aromatic Halides Catalyzed by Nickel(0) Complexes
#
Generated In Situ. III. Kinetic and Mechanistic Aspects
ꢀ
Takuji Ueda, Mutsuji Sakai, and Kentaro Takagi
Yasumasa Sakakibara, Ken Sasaki, Fumio Okuda, Akihiro Hokimoto,
1
Department of Polymer Science and Engineering, Kyoto Institute of Technology,
Matsugasaki, Sakyo-ku, Kyoto 606-8585
1
Department of Chemistry, Faculty of Science, Okayama University, Tsushima, Okayama 700-8530
Received November 21, 2003; E-mail: ksasaki@ipc.kit.ac.jp
The cyanation of chlorobenzene catalyzed by [Ni(PPh3)n], which was generated in situ from [NiBr2(PPh3)2]–Zn–
ꢁ
PPh3 (Ni:Zn:P = 1:3:2 molar ratio), was carried out at 60 C using KCN or NaCN in hexamethyl phosphoric triamide
(
HMPA). Kinetic analyses revealed that the reaction rates in the KCN–HMPA system were zero order for both of chloro-
benzene and nickel concentrations, and two-thirds order for KCN. The determination of the kinetic order for NaCN was
difficult due to its high solubility in HMPA since excess cyanide ion inhibited the reaction. Zinc bromide which was
formed with [Ni(PPh3)n] from [NiBr2(PPh3)2]–Zn–PPh3, acted as an important cocatalyst for the successful cyanation
by comparison with [Ni(cod)2] (cod = 1,5-cyclooctadiene)–PPh3 system. Based on these experimental results, a reaction
mechanism was proposed and discussed.
The transition metal-catalyzed cyanation of aromatic halides
by alkali cyanides has received much attention as a method of
nickel(0) species in the absence of substrates. Therefore, the
following two procedures were adopted for the kinetic studies
(see experimental for detail). Procedure A: The sequential ad-
dition of the aryl halide followed by solid cyanide (within
one minute). Procedure B: The sequential addition of cyanide
followed by the aryl halide. Procedure A could be used success-
fully even for higher solubility systems such as NaCN–HMPA
1
preparation of the corresponding nitriles. In previous papers,
2
3
we reported cyanations of aromatic and vinylic halides by
MCN (M = K, Na) that took place in the presence of a catalytic
amount of [Ni(PPh3)n] generated in situ from [NiBr2(PPh3)2]–
Zn–PPh3 (Chart 1). As a noteworthy result, the cyanation of ar-
omatic halides proceeds smoothly by simple mixing of the re-
actants and the catalytic system in MCN–dipolar aprotic sol-
vents with low cyanide solubility, such as KCN in hexamethyl-
ꢂ3
ꢁ
(461 mmol dm at 60 C), but Procedure B could be used only
for low solubility systems such as KCN–HMPA (1.5 mmol
ꢂ3
ꢁ
dm at 60 C).
2
phosphoric triamide (HMPA), or KCN or NaCN in MeCN. On
The effect of the PPh3 amount in the catalyst component was
examined prior to kinetic study in both systems. In the KCN–
HMPA system using Procedure B, yields of benzonitrile in-
creased lineally after a short induction period with the ratio
of added PPh3/Ni = 1 or 2, as shown in Fig. 1. Higher ratios
of external PPh3/Ni resulted in a longer induction period ac-
companied by a decrease in the reaction rate of the steady-state
nitrile formation.
On the other hand, using NaCN–HMPA by Procedure A, the
effect of PPh3 was not so remarkable in contrast with that in the
KCN–HMPA system: Higher ratios of added PPh3/Ni, from 2
to 12, brought about only a slight decrease in the rates (Fig. 2).
Thus, the added PPh3/Ni ratio of 2 was adopted for further in-
vestigation of both systems, and the reaction rates were deter-
mined from linear region of time-dependent yields after the in-
duction period.
further investigation, we have found that the rate of cyanation
of chlorobenzene in the KCN–HMPA system is faster than that
of the dissolution of KCN. In this article, the cyanation of chlo-
robenzene using two systems which have remarkably different
cyanide solubility, KCN–HMPA and NaCN–HMPA, was in-
vestigated from a kinetic viewpoint to elucidate the reaction
mechanism. In addition, the role of zinc salts on the reaction
was also examined using [Ni(cod)2] (cod = 1,5-cycloocta-
diene)–PPh3 as the catalyst.
Results and Discussion
As reported in our previous papers,#,2 the cyanation of aryl
halides by [NiBr2(PPh3)2]–Zn–PPh3 is highly sensitive to both
the cyanide solubility and order of addition of the reagents,
since free cyanide ion in solution formed less reactive cyano-
The Effect of Substituents on Chlorobenzene in the Reac-
tion Rates. The relative rates of the cyanation, RArCl=RPhCl,
were determined by measuring the initial rates of both the indi-
vidual and competitive reactions with the KCN–HMPA system
using Procedure B (Table 1). In the competitive reaction, the
relative rates of Y-substituted chlorobenzenes correlated linear-
ly with the Hammet’s ꢀ values in the range from ꢀ ¼ ꢂ0:27
Chart 1.
Published on the web May 6, 2004; DOI 10.1246/bcsj.77.1013

1
014 Bull. Chem. Soc. Jpn., 77, No. 5 (2004)
Cyanation of Aromatic Halides
Fig. 1. Effect of PPh3 addition on cyanation in KCN–
HMPA. Carried out by Procedure B. [NiBr2(PPh3)2], 1
mmol (Ni:Zn:P = 1:3:0–8); PhCl, 30 mmol; KCN, 33
mmol; HMPA, 12 cm ; Temp, 60 C. Added PPh3/Ni
ratio: - -, 0; - -, 1.0; - -, 2.0; - -, 4.0; - -, 8.0.
Fig. 2. Effect of PPh3 addition on cyanation in NaCN–
HMPA. Carried out by Procedure A. [NiBr2(PPh3)2], 1
mmol (Ni:Zn:P = 1:3:2–12); PhCl, 30 mmol; NaCN, 33
mmol; HMPA, 12 cm ; Temp, 60 C. Added PPh3/Ni
ratio: - -, 2.0; - -, 4.0; - -, 12.0.
3
ꢁ
3
ꢁ
Table 1. Relative Rates of Cyanation of Y-Substituted Chlorobenzenes
Relative rate (RArCl=RPhCl)
aÞ
bÞ
Competitive
Y
ꢀ
Individual
KCN–HMPA
NaCN–HMPA
cÞ
KCN–HMPA
p-OMe
p-Me
m-Me
H
ꢂ0:27
ꢂ0:17
ꢂ0:07
0
0.90
0.99
1.0
0.41
0.64
0.96
1
cÞ
1
1
m-OMe
p-F
p-CF3
þ0:12
þ0:34
þ0:54
1.1
0.93
1.1
1.2
3.0
6.9
a) Carried out by Procedure B. [NiBr2(PPh3)2], 1 mmol (Ni:Zn:P = 1:3:2); ArCl, 30 mmol;
3
ꢁ
MCN, 30 mmol; ArCl–HMPA, 15 cm (total volume); Temp, 60 C. b) Under conditions as
described above, except that each 15 mmol of ArCl and PhCl was used. c) Carried out by
Procedure A.
(
(
p-OMe) to ꢀ ¼ þ0:54 (p-CF3). The resultant positive ꢁ value
þ1:4) indicated that the relative reactivity in the competitive
systems. As standard conditions, [PhCl] = 2.0 M (1 M = 1
ꢂ3
ꢂ2
mol dm ), MCN = 3:0 ꢃ 10 mol (equivalent to chloroben-
ꢂ2
reactions reflected the relative rates of the oxidative addition
of the aryl halides to Ni(0), which were accelerated by elec-
tron-withdrawing substituents.4 In the individual reactions,
however, the relative rates did not vary with the ꢀ values, being
about 1.0 in the range of ꢂ0:27 (Y = p-OMe) to þ0:54 (p-
CF3). A similar insensitivity to the substituents was also ob-
served in the individual reactions using the NaCN–HMPA sys-
tem with a high cyanide solubility. These results suggest that
the oxidative addition of chlorobenzenes to the Ni(0) species,
which plays the main role in the competitive reactions, is not
the rate-determining step in the over all catalytic reaction in
both systems.
zene), [Ni] = 6:7 ꢃ 10 M (PhCl/Ni = 30), total volume of
3
PhCl and HMPA = 15 cm , using Procedure B for the KCN–
,5
HMPA system and Procedure A for the NaCN–HMPA.
KCN–HMPA System: As shown in Fig. 3, the initial rate
for the formation of benzonitrile was independent of the initial
concentration of chlorobenzene, which ranged from 0.67 to 2.7
M. On the other hand, the rate increased non-linearly with the
amount of KCN. The slope of the log[rate] vs log[KCN] plots
revealed that the rates obey a two-thirds order relationship with
the amount of KCN, as illustrated in Fig. 4. A linear relation-
ship was observed between the rates and the concentrations
of the nickel catalyst (Fig. 5).
Kinetics. The rate measurements were made for the cyan-
ation of chlorobenzene in the KCN–HMPA and NaCN–HMPA
Eq. 1 shows the rate expression determined experimentally,
where [Ni]0 stands for the initial concentration of nickel cata-

Y. Sakakibara et al.
Bull. Chem. Soc. Jpn., 77, No. 5 (2004) 1015
Fig. 3. Plot of rate against chlorobenzene concentration.
ꢂ2
Fig. 5. Plot of rate against nickel concentration. Carried out
ꢂ2
Carried out by Procedure B. [Ni], 6:7 ꢃ 10 M (Ni:Zn:P
by Procedure B. [PhCl], 2.0 M; KCN, 3:0 ꢃ 10 mol;
ꢂ2
ꢁ
3
=
1:3:2); KCN, 3:0 ꢃ 10 mol; PhCl–HMPA, 15 cm
3
ꢁ
PhCl–HMPA, 15 cm ; Temp, 60 C.
(
total volume); Temp, 60 C.
Fig. 4. Plot of log(Rate) against log(KCN/mol). Carried out
ꢂ2
by Procedure B. [Ni], 6:7 ꢃ 10 M; [PhCl], 2.0 M; PhCl–
Fig. 6. Reaction zone in NaCN–HMPA system. Carried out
by Procedure A. [NiBr2(PPh3)2], 1 mmol (Ni:Zn:P =
1:3:2); PhCl, 30 mmol; Temp, 60 C. - - -, Border line be-
tween reaction and no reaction; , Reaction proceeds
3
ꢁ
HMPA, 15 cm ; Temp, 60 C.
ꢁ
lyst and [MKCN] is the molar amount of KCN in the reaction
mixture.
smoothly; ꢃ, Reaction dose not proceed.
2
=3
Rate ¼ d½PhCNꢄ=dt ¼ k½Niꢄ ½MKCNꢄ
ð1Þ
0
2=3
ranges, the rate was zero order with respect to chlorobenzene
and first order with respect to the nickel catalyst. However,
the kinetic order for NaCN was lower than that for
Here, ½MKCNꢄ is proportional to the surface area of KCN in
the solid state. The initial rate of the dissolution of KCN
ꢂ2
3
(
(
3 ꢃ 10 mol) into a mixture of PhCl (3 cm ) and HMPA
3
2=3
ꢂ4
ꢂ1
½MNaCNꢄ owing to the effects of high cyanide ion concentra-
tion compared with the KCN–HMPA system. The reaction rate
in NaCN–HMPA was only about three times faster than that in
KCN–HMPA in spite of the high solubility (ca. 300 times).
12 cm ) was determined to be 3:0 ꢃ 10 M min from the
6
solubility vs time plots, which is far smaller (1/10–1/30) than
ꢂ3
ꢂ3
the rates of formation of benzonitrile (2:8 ꢃ 10 –9:1 ꢃ 10
ꢂ1
M min , Fig. 5). This result suggests that the cyanide ions
which contribute to the reaction come directly from the solid
KCN (KCNS).
[Ni(cod)2]–PPh3 Catalyst and Effect of ZnBr2 on the
Reactivity. In order to clarify the effect of ZnBr2 on the reac-
tion, the cyanation of chlorobenzene was attempted using a
modified Procedure B using a [Ni(cod)2]–PPh3 catalyst, which
forms [Ni(PPh3)n] without accompanying the formation of
ZnBr2 in [NiBr2(PPh3)2]–Zn–PPh3 (Eqs. 2 and 3).8
NaCN–HMPA System: In this system, the cyanation was
restricted to a narrow zone of NaCN concentration, and pro-
ceeded smoothly only in the zone below the dotted line in the
presence of 1 mmol of Ni catalyst and 30 mmol of PhCl, as
7
shown in Fig. 6 Therefore, kinetic measurements were made
only in the restricted range of [PhCl] = 2.0–3.3 M, NaCN =
[
NiBr2(PPh3)2] þ Zn+PPh3ꢂ! [Ni(PPh3)n] þ ZnBr2
Ni(cod)2] þ PPh3ꢂ![Ni(PPh3)n]
ð2Þ
ð3Þ
[
0
.010–0.030 mol, and [Ni] = 0.033–0.13 M (Table 2). In these

1
016 Bull. Chem. Soc. Jpn., 77, No. 5 (2004)
Cyanation of Aromatic Halides
aÞ
Table 2. Kinetic Results of the Cyanation of Chlorobenzene in NaCN–HMPA System
2=3
[
PhCl]
NaCN
/mol
(NaCN/mol)
/M
[Ni]
/M
Rate
/M min
ꢂ1
/M
2
2
3
.0
.7
.3
0.030
0.030
0.030
0.10
0.10
0.10
0.067
0.067
0.067
0.020
0.019
0.020
2
2
2
.0
.0
.0
0.010
0.020
0.030
0.046
0.074
0.10
0.067
0.067
0.067
0.012
0.016
0.020
bÞ
2
2
2
.0
.0
.0
0.030
0.030
0.030
0.10
0.10
0.10
0.033
0.067
0.13
ꢁ
—
0.020
0.035
3
a) Carried out by Procedure B. PhCl–HMPA, 15 cm ; Temp, 60 C. b) No reaction.
Fig. 7. Effect of ZnBr2 addition on the rate of cyanation in
KCN–HMPA with [Ni(cod)2]–PPh3 catalyst. Carried out
ꢂ2
Fig. 8. Effect of ZnBr addition on the cyanation in NaCN–
2
by a modified Procedure A. [[Ni(cod)2]], 6:7 ꢃ 10
M
ꢂ2
HMPA with [Ni(cod) ]–PPh catalyst. Under conditions as
2
(
HMPA, 15 cm ; Temp, 60 C.
Ni:P = 1:4); [PhCl], 2.0 M; KCN, 3 ꢃ 10 mol; PhCl–
3
3
ꢁ
described in Fig. 7 except that NaCN was used instead of
KCN. ZnBr2/Ni ratio: - -, 0; - -, 0.5; - -, 1.0; - -, 3.0.
As shown in Fig. 7, with the KCN–HMPA system, the addition
of ZnBr2 accelerated the reaction remarkably. The reaction rate
increased linearly with an increase of the ZnBr2/Ni ratio from 0
that the reaction proceed via cyanide anion abstraction from the
solid KCN surface, as mentioned above.
ꢂ3
ꢂ1
to 1. The rate at the ratio of 1 (5:0 ꢃ 10 M min ) was almost
In NaCN–HMPA, the cyanation proceeded fairly smoothly
without the addition of ZnBr2, in contrast to the KCN–HMPA
system (Fig. 8). This implies that the dissolved cyanide ions
largely contribute to the reaction. On the other hand, the addi-
tion of ZnBr2 in ZnBr2/Ni = 0.5 or 1.0 accelerated the reaction
at an early stage, but the reaction stopped at about 50% and
60% conversion, respectively. Furthermore, a large amount of
ZnBr2, for example, ZnBr2/Ni = 3, inhibited rather than accel-
erated the reaction. Similarly an equimolar addition of ZnBr2
into the [NiBr2(PPh3)2]–Zn–PPh3 catalyst system inhibited
the reaction remarkably. One possible interpretation for these
effects is any adduct formation of PPh3 with ZnBr2, which
would result in a decrease in free PPh3 in solution leading to
the same as that for the [NiBr2(PPh3)2]–Zn–PPh3 catalyst
ꢂ3
ꢂ1
(
6:6 ꢃ 10 M min ), and was 25 times faster than that in
the absence of ZnBr2. However, the rate at the ratio of 3 was
somewhat lower compared to that at the ratio of 1. Furthermore,
with the [NiBr2(PPh3)2]–Zn–PPh3 catalyst, the reaction rate
was scarcely accelerated by an equimolar addition of ZnBr2.
These results suggest that the acceleration of the reaction is
due to the action of any adduct formed by ZnBr2 with a nickel
intermediate, [PhNiCl(PPh)2] (Eq. 4).9
[PhNiCl(PPh3)2] þ ZnBr2ꢂ! [Ph-Ni(PPh3)2Cl{ZnBr2] ð4Þ
A cation exchange of KCN to Zn(CN)2 could be considered
as a possible role of the Zn salt, which has been proposed for
the Pd catalyzed cyanation using cocatalyst (CuX2 or ZnX2).10
Indeed, the efficiency of Zn(CN)2 as a cyanide source has been
generally accepted for the Pd catalyzed reaction.11 However,
this possibility would not be the case for the present Ni cata-
lyzed reaction on the basis of the kinetic result, which indicated
ꢂ
a less active nickel(0) species, such as [Ni(CN)(PPh3)3] from
ꢂ
[Ni(PPh3)3] with CN . As strong support for the above view, it
was observed that the reaction proceeds smoothly even in
NaCN–HMPA with the [NiBr2(PPh3)2]–Zn–PPh3 catalyst in
the case of further addition of PPh3 with ZnBr2 (2:1 molar
ratio), but does not proceed in the case of addition of ZnBr2

Y. Sakakibara et al.
Bull. Chem. Soc. Jpn., 77, No. 5 (2004) 1017
Scheme 1. Reaction mechanism.
alone.12
Cycle A would play the main role in the KCN–HMPA sys-
tem, because the dissolution rate of KCN and the reaction rate
with the [Ni(cod)2]–PPh3 catalyst in the absence of ZnBr2 is
very slow. The oxidative addition of chlorobenzene to
[Ni(PPh3)3] (step (a)) is known to be very fast.4 In accord-
ance with this, the overall rate of the cyanation was independ-
ent of the concentrations of chlorobenzene (Fig. 3) and of the
kinds of substituted chlorobenzenes. However, significant dif-
ferences in the relative rate of substituted chlorobenzenes in
the competitive reaction imply that their ability towards the ox-
idative addition becomes a dominant factor.4 In the next step
(b), [PhNiCl(PPh3)] forms an adduct with ZnBr2 (Eq. 4) and the
Mechanistic Consideration. On the basis of the experi-
mental results, the mechanism as shown in Scheme 1 is pro-
posed as the most plausible one for the cyanation. The nickel
catalyst, [Ni(PPh3)4], which dissociates into [Ni(PPh3)3] in so-
,15
13,14
lution,
with Zn dust in the presence of PPh3. As reported previously,
is easily formed by the reduction of [NiBr2(PPh3)2]
the presence of excess cyanide ion formed a less reactive
ꢂ
2
cyanonickel(0) complex, [Ni(CN)(PPh3)3] , and therefore it
is necessary for the successful cyanation in NaCN–HMPA to
adopt Procedure A under limited conditions (Fig. 6). On the
other hand, in the KCN–HMPA system, the cyanation proceeds
smoothly by any procedure attempted.
The paths of cyanation are shown by two catalytic cycles, A
and B in Scheme 1. In both cases, the reaction is initiated by the
oxidative addition of chlorobenzene to [Ni(PPh3)3] (step (a)).
The addition product, [PhNiCl(PPh3)2], is converted into
,5
ꢂ
adduct draws out CN from the surface of solid KCN by an in-
teraction between the Zn atom and KCN. Here, assuming that
this step (c) is the rate-determining step, the derivable rate
ꢂ
equation is in accord with Eq. 1. Subsequently, the CN coor-
dinated adduct is converted into [PhNi(CN)(PPh3)2] via a cya-
no-bridged nickel complex16 (steps (d), (e)). In connection with
the last step (f), being a so-called reductive elimination, it is
well known that phosphorus ligands accelerate the reductive
[PhNi(CN)(PPh3)2] via the reaction with MCNS in the presence
of ZnBr2 as promoter (steps (b)–(e) in Cycle A) or via the reac-
ꢂ
0
0
tion with the dissolved CN (steps (b ), (c ) in Cycle B). In the
last step, [PhNi(CN)(PPh3)2] undergoes reductive elimination
to give benzonitrile and [Ni(PPh3)3] (step (f)), and the catalytic
reaction is recycled.
17
elimination of nitriles from cyanoarylnickel and cyanoalkyl-
nickel18 complexes. In our case, however, large amounts of tri-
phenylphosphine retarded the reaction in the early stage rather

1
018 Bull. Chem. Soc. Jpn., 77, No. 5 (2004)
Cyanation of Aromatic Halides
than contributing to the rate enhancement (Fig. 1). This sug-
gests that the step (f) is not the rate-determining step. The ox-
idative addition of benzonitrile to the Ni(0) species appears to
ied as indicated in Figs. 3, 4, and 5 for the KCN–HMPA system
and in Table 1 (and Fig. 6 as reference experiment) for the
NaCN–HMPA system. The reaction rates were measured in a
similar manner as above.
be difficult compared with that of chlorobenzene.1
9,20
In fact,
General Procedure for the Cyanation Catalyzed by [Ni-
cod)2]–PPh3–ZnBr2. The reaction was carried out using a modi-
we ascertained that benzonitrile failed to undergo an oxidative
addition under our reaction conditions.
(
fied Procedure A described below. After a prescribed amount of
ZnBr2 (0–3 mmol) was placed in a flask filled with nitrogen, a so-
lution of [Ni(cod)2] (2.75 g, 1.0 mmol) and PPh3 (1.05 g, 4.0
In the NaCN–HMPA system, Cycle B would participate in
the reaction together with Cycle A, given that the reaction pro-
ceeds fairly smoothly in the absence of ZnBr2 (cf. Fig. 8).
3
mmol) in HMPA (12 cm ) was added, and the mixture was then
ꢁ
ꢂ
[
PhNiCl(PPh3)2] reacts with the dissolved CN to give
kept at 60 C for 30 min while stirring. Chlorobenzene (3.38 g,
0 0
[PhNi(CN)(PPh3)2] (steps (b ), (c )) probably via a CN-coordi-
3
0 mmol) was added, followed immediately by MCN (30 mmol)
from the MCN container. The reaction mixture was kept at 60
21,22
nated intermediate (so-called association path).
tution can be assumed to be a rapid reaction judging from the
The substi-
ꢁ
C. The variation in yield with time was followed by GC analysis
Figs. 7 and 8).
fact that the Cl substitution of trans-[(p-C6H4Me)NiCl(PEt3)2]
(
ꢂ
21
by CN in EtOH proceeds rapidly. With this system, both the
ꢂ
dissolution of NaCN and the abstraction of CN from the sur-
face of KCNS by the [PhNiCl(PPh3)2–ZnBr2] adduct appear to
act as the rate-determining step.
The present work was partially supported by a Grant-in-Aid
for Scientific Research (No. 254158) from the Ministry of
Education, Culture, Sports, Science and Technology.
Experimental
References
Materials. HMPA and chlorobenzene were distilled after dry-
ing over Molecular Sieve 13X and the former was stored in the
presence of the drying agent. The other substituted chlorobenzenes
from commercial sources were used without further purification.
KCN and NaCN were ground by the use of a mortar and were
sieved to obtain 350–250 mm (40–60 mesh) fractions. Zn dust,
PPh3, ZnBr2 (anhydride), and [Ni(cod)2] were commercially avail-
able. [NiBr2(PPh3)2] was prepared by a literature method.23
General Procedure for the Cyanation Catalyzed by [NiBr2-
#
Part II. Y. Sakakibara, Y. Ido, K. Sasaki, M. Sakai, and N.
Uchino, Bull. Chem. Soc. Jpn., 66, 2776 (1993).
For reviews, G. P. Ellis and T. M. Ronney-Alexander,
1
Chem. Rev., 87, 779 (1987); V. V. Grushin and H. Alper, Chem.
Rev., 94, 1047 (1994).
2
Part I. Y. Sakakibara, F. Okuda, A. Shimobayashi, K.
Kirino, M. Sakai, N. Uchino, and K. Takagi, Bull. Chem. Soc.
Jpn., 61, 1985 (1988).
(
mmol), Zn dust (200 mg, 3.0 mmol), and PPh3 (520 mg, 2.0 mmol)
PPh3)2]–Zn–PPh3. Procedure A: [NiBr2(PPh3)2] (740 mg, 1.0
3
K. Kunitake, K. Sasaki, and M. Sakai, Bull. Chem. Soc. Jpn., 68,
3137 (1995).
Y. Sakakibara, H. Enami, H. Ogawa, S. Fujimoto, H. Kato,
3
were placed in a 50 cm flask equipped with a magnetic stirrer, a
three-way cock, and an MCN container connected by means of a
ground-glass joint. After replacing the air with nitrogen, HMPA
4
2572.
5
(1979).
6
M. Foa and L. Cassar, J. Chem. Soc., Dalton Trans., 1975,
T. T. Tsou and J. K. Kochi, J. Am. Chem. Soc., 101, 6319
Measurement of the solubility of MCN was described in our
ꢁ
was added by a syringe. The mixture was then kept at 60 C for
3
0 min while stirring, chlorobenzene (3.38 g, 30 mmol) and naph-
thalene (0.2 g, GC internal standard) were added, followed by
MCN (33 mmol) (within one minute) from the MCN container.
previous paper. See; K. Takagi, T. Okamoto, Y. Sakakibara, A.
Ohno, S. Oka, and N. Hayama, Bull. Chem. Soc. Jpn., 49, 3177
(1976).
ꢁ
The reaction mixture was kept at 60 C. Small aliquots of the mix-
ture were withdrawn at appropriate time intervals, poured into a
small amount of CH2Cl2, and, after bubbling with air, subjected
to GC analysis (3 m EGS column, He carrier gas, 150 C).
Procedure B: This procedure was the same as Procedure A,
except that the order of the addition of chlorobenzene and MCN
was reversed.
7
The reaction proceeded even in the presence of large
ꢁ
amounts of NaCN when HMPA is diluted with a large amount of
chlorobenzene as reactant. For example, 90 mmol of chloroben-
3
zene (about 9 cm ) reacted completely with 100 mmol of NaCN
3
ꢁ
in 12 cm of HMPA at 60 C for 6 h to give 95% benzonitrile
and 2% biphenyl. This should be noted from a synthetic viewpoint.
Relative Rate of Substituted Chlorobenzenes. The reaction
was carried out using Procedure B. In all cases, the total volume
8
C. A. Tolman, D. W. Reutter, and W. C. Seidel, J. Organo-
met. Chem., 117, C30 (1976).
For examples of 1:1 adducts of Ni complexes with
3
of substrate and HMPA was adjusted to 15 cm . In the competitive
reaction, a mixture of equimolar amounts of two chlorides was
used. The initial rates were calculated from the slope of the time
to yield plots at an early stage of the reaction (Table 1).
Kinetic Measurement. With KCN–HMPA, the reaction was
carried out using Procedure B, while with the NaCN–HMPA sys-
tem, it was carried out using Procedure A. The standard reaction
mixture was composed of 740 mg (1.0 mmol) of [NiBr2(PPh3)2],
9
CH3AlCl2, and Me2Mg, see; P. W. Jolly and G. Wilke, ‘‘The
Organic Chemistry of Nickel,’’ Academic Press, New York
(1975), Vol. 2, Chap. 1, pp. 18–21; W. Kaschube, K. R. Porschke,
K. Angermund, C. Kruger, and G. Wilke, Chem. Ber., 121, 1921
(1988); An acceleration effect by the complexation of organozinc
species in the nickel-catalyzed conjugate addition/aldol reaction
was also reported in K. Subburaj and J. Montogomery, J. Am.
Chem. Soc., 125, 11210 (2003).
10 B. A. Anderson, E. C. Bell, F. O. Ginah, N. K. Harn, L. M.
Pagh, and J. P. Wepsiec, J. Org. Chem., 63, 8224 (1998).
11 D. M. Tschaen, R. Desmond, A. O. King, M. C. Fortin, B.
Pipik, S. King, and T. R. Verhoeven, Synth. Commun., 24, 887
2
00 mg (3.0 mmol) of Zn dust, 520 mg (2.0 mmol) of PPh3,
3
or 1.47 g (30 mmol) of NaCN, and 12 cm of HMPA (total volume
.38 g (30 mmol) of chlorobenzene, 1.95 g (30 mmol) of KCN
3
3
of chlorobenzene and HMPA was adjusted to 15 cm ). The reac-
ꢁ
tion was carried out at 60 ꢅ 1 C. The amounts of chlorobenzene,
MCN, and [NiBr2(PPh3)2]–Zn–PPh3 (Ni:Zn:P = 1:3:2) were var-

Y. Sakakibara et al.
Bull. Chem. Soc. Jpn., 77, No. 5 (2004) 1019
(
(
1994).
When the reaction of chlorobenzene (30 mmol) with NaCN
position; W. P. Fehlhammer and M. Fritz, Chem. Rev., 93, 1243
(1993).
17 G. Favero, M. Gaddi, A. Morvillo, and A. Turco, J. Orga-
nomet. Chem., 149, 395 (1978).
1
30 mmol) was carried out by Procedure A in HMPA (12 cm ) at
2
3
ꢁ
6
0 C in the presence of ZnBr2 (1 mmol) and [NiBr2(PPh3)2]–Zn–
PPh3 (Ni, 1 mmol; 1:3:2) the yield of benzonitrile was lower than
0% at 6 h. But, by further addition of 2 mmol of PPh3 the reaction
proceeded smoothly to give 84% benzonitrile.
C. A. Tolman, W. C. Seidel, and D. H. Gerlach, J. Am.
Chem. Soc., 94, 2669 (1972).
R. Mynott, A. Mollbach, and G. Wilke, J. Organomet.
Chem., 199, 107 (1980).
J. F. Fauvarque and F. Pfluger, J. Organomet. Chem., 208,
19 (1981).
It is reported that CN migrates between two metals bridging
18 R. J. Mckinney and D. C. Roe, J. Am. Chem. Soc., 107, 261
(1985).
19 G. Favero, A. Morvillo, and A. Turco, J. Organomet.
Chem., 162, 99 (1978).
20 G. W. Parshall, J. Am. Chem. Soc., 96, 2360 (1974).
21 L. Cassar, S. Ferrara, and M. Foa, ‘‘Homogeneous Catalysis
II,’’ Adv. Chem. Ser. 132, Am. Chem. Soc. (1974), p. 252.
22 M. Cusumano and V. Ricevuto, J. Chem. Soc., Dalton
Trans., 1978, 1682.
1
1
3
1
4
1
5
4
1
6
23 K. Yamamoto, Bull. Chem. Soc. Jpn., 27, 501 (1954).