Mild and efficient copper-catalyzed cyanation of aryl iodides and bromides

Article abstract of DOI:10.1002/chem.200400979

An efficient copper-catalyzed cyanation of aryl iodides and bromides is reported. Our system combines catalytic amounts of both copper salts and chelating ligands. The latter, which have potential nitrogen- and/or oxygen-binding sites, have never previously been used in this type of reaction. A protocol has been developed that enables the cyanation of aryl bromides through the copper-catalyzed in situ production of the corresponding aryl iodides using catalytic amounts of potassium iodide. Aryl nitriles are obtained in good yields and excellent selectivities in relatively mild conditions (110°C) compared with the Rose-nmund-von Braun cyanation reaction. Furthermore, the reaction is compatible with a wide range of functional groups including nitro and amino substituents. The protocol reported herein involves two main innovations: the use of catalytic amounts of ligands and the use of acetone cyanohydrin as the cyanating agent in copper-mediated cyanation reactions.

Full text of DOI:10.1002/chem.200400979

FULL PAPER
Mild and Efficient Copper-Catalyzed Cyanation of Aryl Iodides and
Bromides
Henri-Jean Cristau,*[a] Armelle Ouali,[a] Jean-Francis Spindler,[b] and Marc Taillefer*[a]
In memory of Dr. Gennady Vasilievich Dolgushin, Head of the Organic Synthesis Laboratory of A. E. Favorsky Irkutsk
Institute of Chemistry
Abstract: An efficient copper-catalyzed
cyanation of aryl iodides and bromides
is reported. Our system combines cata-
lytic amounts of both copper salts and
chelating ligands. The latter, which
situ production of the corresponding
aryl iodides using catalytic amounts of
potassium iodide. Aryl nitriles are ob-
tained in good yields and excellent se-
lectivities in relatively mild conditions
(1108C) compared with the Rose-
nmund–von Braun cyanation reaction.
Furthermore, the reaction is compati-
ble with a wide range of functional
groups including nitro and amino sub-
stituents. The protocol reported herein
involves two main innovations: the use
of catalytic amounts of ligands and the
use of acetone cyanohydrin as the cy-
anating agent in copper-mediated cyan-
ation reactions.
have
potential
nitrogen-
and/or
oxygen-binding sites, have never previ-
ously been used in this type of reac-
tion. A protocol has been developed
that enables the cyanation of aryl bro-
mides through the copper-catalyzed in
Keywords: acetone cyanohydrin
·
aryl halides · copper · cyanation ·
homogeneous catalysis
Introduction
mides and chlorides with sodium cyanide, potassium cyanide
or acetone cyanohydrin. The preparation of aromatic nitriles
from aryl halides using nickel salts in conjunction with mi-
crowave irradiation has also been recently described.[10] The
usefulness of this reaction is however limited because of the
high cost and toxicity of the equimolar amounts of nickel
derivatives that are required.
Palladium-catalyzed cyanodehalogenation of aryl halides
has also been reported.[11] Palladium catalysts are in general
more tolerant towards a variety of functional groups than
nickel ones but suffer from poor reliability. Sodium and po-
tassium cyanide are generally used as cyano group sources[12]
but many variants have been developed: zinc cyanides,[13] tri-
methylsilyl cyanides,[14] tributyltin cyanides,[15] dialkylcyano-
boronates,[16] copper cyanide[17] and potassium hexacyanofer-
rate(ii)[18] can also promote the cyanation of aryl halides to
afford the corresponding aryl nitriles. Interestingly, the use
of acetone cyanohydrin, already known to be a suitable cy-
anating agent in the nickel-catalyzed cyanation of aryl hal-
ides, has recently been extended to reactions that involve
palladium as the catalyst.[19] However, the high cost of this
metal and the need to use expensive and toxic phosphines
as ligands make the development of methods involving
other metals quite attractive.
Aryl nitriles are stable and valuable intermediates in organic
synthesis because of the versatile transformations of the ni-
trile function.[1] Thus they constitute building blocks for the
synthesis of biologically active molecules,[2] polyamides,[3]
and metallophthalocyanine precursors.[4] Moreover aryl and
heteroaryl nitriles are present in numerous dyes, agrochemi-
cals[5] and pharmaceuticals.[6]
Various methods for the synthesis of aryl nitriles have
been reported. One of the most convenient is based on the
transition-metal-mediated displacement of aromatic halides
by the cyanide ion.[7]
Cassar[8] and Sakakibara[9] and their co-workers used
nickel complexes as catalysts in the cyanation of aryl bro-
[a] Prof. Dr. H.-J. Cristau, A. Ouali, Dr. M. Taillefer
Laboratoire de Chimie Organique, UMR CNRS 5076
Ecole Nationale Supꢀrieure de Chimie de Montpellier
8 rue de l’Ecole Normale, 34296 Montpellier Cedex 5 (France)
Fax : (+33)467-144-319
E-mail: cristau@cit.enscm.fr
taillefe@cit.enscm.fr
[b] Dr. J.-F. Spindler
Rhodia Organique Fine, Centre de Recherches de Lyon (CRL)
85 avenue des Frꢁres Perret, BP 62, 69192 Saint-Fons Cedex (France)
Chem. Eur. J. 2005, 11, 2483 – 2492
DOI: 10.1002/chem.200400979
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2483

Finally, the direct reaction between aryl halides and
copper cyanides to give the corresponding aryl nitriles (the
Rosenmund–von Braun reaction) has been known for over
80 years since the work of Pongratz.[7,20] This method re-
quires both stoichiometric quantities of CuCN and high re-
action temperatures (>1508C). Furthermore product isola-
tion is complicated by the need to separate the resulting
copper halide. Recently aryl cyanides have been obtained at
lower temperatures (90–1308C) in ionic liquids and from
aryl iodides in the presence of either two equivalents or cat-
alytic amounts of CuCN.[21] However, the isolated yields are
moderate because of the difficulty of isolating the nitriles.
Cyanation of aryl halides with an excess or at least stoichio-
metric amounts of CuCN has also been achieved by micro-
wave-heating in water[22] or in DMSO.[23] The copper-cata-
lyzed cyanation of aryl- and heteroaryl bromides in toluene
at around 110–1308C has also been reported.[24] However,
this method requires the use of a ten-fold amount of the
rather expensive ligand N,N’-dimethylethylenediamine rela-
tive to the copper salt. This feature is thus less attractive for
large-scale applications. As a consequence, the search for
more practical methods involving copper catalysis appear
necessary to meet the demands of contemporary chemical
synthesis, that is, less waste and the use of catalytic process-
es wherever possible. We report herein our contribution to
this search, presenting a method for the copper-catalyzed cy-
anation of aryl halides which has two main innovations: 1)
the use of catalytic amounts of not only the copper precata-
lyst but also of the ligand and 2) the use of acetone cyano-
hydrin as the cyanide source. To the best of our knowledge
acetone cyanohydrin has until now only been employed in
nickel- or palladium-catalyzed cyanation reactions.
In a previous paper we presented a general and efficient
method for the N-arylation of pyrazoles with aryl bromides
and iodides that involved the use of inexpensive copper-
based catalytic systems.[25] We noticed that bi-, tri- or tetra-
dentate ligands with nitrogen and/or oxygen chelating atoms
considerably accelerate the copper-catalyzed arylation reac-
tions of pyrazoles. Aryl- or heteroarylpyrazoles were thus
obtained with high yields and selectivities under extremely
mild conditions. We also report-
set of experiments performed in DMF as solvent, we investi-
gated the cyanation of iodobenzene in the presence of cop-
per(i) catalysts (10 mol% CuI or 5 mol% Cu2O) and cata-
lytic amounts of a set of bi-, tri- or tetradentate ligands
(Figure 1). These additives comprise only nitrogen-binding
sites (ligands 1, 2, 4, and 6) or combine nitrogen- and
oxygen-binding sites (ligands 3, 5, 7, and 8). In all cases, the
yields were not affected by the nature of the copper precata-
lyst, either cuprous iodide or cuprous oxide. In contrast, the
nature of the ligand significantly influenced the efficiency of
the catalytic system since after 24 h yields of benzonitrile
ranged from 48 to 94%. We identified five ligands with vari-
ous structural features (1–5) that promote the arylation re-
action with yields in excess of 75%. Ligand 1 comprises two
pyridine-type binding sites, whereas 2 and 4 combine both
pyridine- and imine-type binding sites. Oxime-type ligand 3
and ligand 5 combine nitrogen and oxygen as potential che-
lating atoms.
In the following experiments, we focused our attention on
1,10-phenanthroline (1,10-phen, 1) and N,N-dimethyl-N’-pyr-
idin-2-ylmethyleneethane-1,2-diamine (Dmeda-Py-Al, 2),
which promote the arylation reaction to give yields of 94
and 83%, respectively. Since benzonitrile can be obtained
from iodobenzene in an excellent yield, we decided to
extend our methodology to the cyanation of less expensive
and more challenging aryl bromides.
Copper-catalyzed cyanation of bromobenzene: The reaction
conditions used in the cyanation of iodobenzene were ap-
plied to the cyanation of bromobenzene. At first the cross-
coupling reaction failed. However we overcame this prob-
lem and obtained aryl nitriles through the catalyzed produc-
tion of iodobenzene from bromobenzene and iodide salts
and its subsequent in situ cyanation.[29] We first checked that
displacement of the bromide ion by the iodide ion took
place under the conditions used in the cyanation of iodoben-
zene when using 1,10-phen 1 as the ligand (Scheme 1,
[Eq. (1)], a)). The rapid formation of iodobenzene was
indeed observed, 55% of bromobenzene being converted
during the first four hours of the reaction.
ed that phenols,[26] amides, car-
bamates, nitrogen heterocycles
and malonic acid derivatives[27]
can be arylated at moderate
temperatures with similar cata-
lytic systems.[28] In this report
we extend the scope of such
catalytic systems to the cyana-
tion of aryl iodides and bro-
mides.
Results and Discussion
Scheme 1. Cyanation of bromobenzene: halide exchange followed by in situ cyanation of iodobenzene. Gener-
al conditions: a) PhBr (0.5 mmol), KI (0.5 mmol), CuI (0.05 mmol), 1 (0.10 mmol), DMF (300 mL), 48 h,
1108C; b) PhI (0.5 mmol), KBr (0.5 mmol), CuI (0.05 mmol), 1 (0.10 mmol), DMF (300 mL), 48 h, 1108C; c)
Copper-catalyzed cyanation of
iodobenzene: In a preliminary KCN yields were determined by GC with 1,3-dimethoxybenzene as the internal standard.
2484
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2483 – 2492

Copper-Catalyzed Cyanation of Aryl Iodides and Bromides
FULL PAPER
Figure 1. Effect of various ligands on the copper-catalyzed coupling reaction of potassium cyanide and iodobenzene. General conditions: iodobenzene
(0.5 mmol), potassium cyanide (0.55 mmol), CuI (0.05 mmol), ligand (0.10 mmol), DMF (300 mL), 24 h, 1108C. Yields were determined by GC with 1,3-
dimethoxybenzene as the internal standard.
We also demonstrated (Scheme 1, [Eq. (2)]) that the final
composition was the same when starting from either an
equimolar mixture of bromobenzene and potassium iodide
(a) or an equimolar mixture of iodobenzene and potassium
bromide (b): the final mixture was composed of about 20%
PhBr and 80% PhI. Note that these results are in agreement
with the existence of a thermodynamic equilibrium between
the two aromatic halides.[30] Therefore it was expected that
the addition of cyanide ion, which would trap iodobenzene,
would allow the formation of benzonitrile with the simul-
taneous regeneration of potassium iodide (Scheme 1,
[Eq. (1)], c)). As a consequence catalytic amounts of KI
might be sufficient to promote the cyanation of PhBr.
(0.1 equiv), ligand (0.2 equiv), potassium iodide (0.5 equiv),
and solvent (DMF) were introduced simultaneously. Un-
fortunately, neither halide exchange nor cyanation took
place under these conditions. However, this was not surpris-
ing because deactivation of nickel- or palladium-based cata-
lytic systems by an excess of dissolved cyanide is a well-
known phenomenon.[9b,12k,13h,14b,19]
KCN was then introduced into the reaction mixture only
once a sufficient quantity of PhI had been generated.
During the first stage, bromobenzene was converted into io-
dobenzene by using potassium iodide (0.5 equiv) in the pres-
ence of ligand 1 or 2 (Scheme 2). The composition of both
reaction mixtures before the addition of KCN, given in
Scheme 2, shows that the halide exchange is quicker with 1
as ligand than with 2. In the second stage KCN was intro-
The cyanation was first performed as a one-pot reaction:
bromobenzene, potassium cyanide (1.1 equiv), copper iodide
Chem. Eur. J. 2005, 11, 2483 – 2492
www.chemeurj.org
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2485

H.-J. Cristau, M. Taillefer et al.
Scheme 2. Cyanation of PhBr by using potassium cyanide after halide ex-
change. General conditions: bromobenzene (0.5 mmol), potassium iodide
(0.25 mmol), CuI (0.05 mmol), ligand (0.10 mmol), DMF (300 mL), 6 h,
1108C and then addition of potassium cyanide (0.55 mmol). Yields were
determined by GC with 1,3-dimethoxybenzene as the internal standard.
It was arbitrarily decided to add KCN after 6 h and this reaction time has
not been optimized.
duced to allow cyanation of the iodobenzene previously gen-
erated. Note that halide exchange also occurred during the
cyanation reaction since the yields of PhCN are higher than
the yields of PhI estimated before the addition of KCN. The
process being faster in the presence of 1 than with 2, we
therefore focused our attention on 1,10-phen in the follow-
ing experiments.
Thus, this protocol allows the synthesis of benzonitrile
from bromobenzene in a satisfying yield (78%) after 48 h at
1108C. Moreover, the selectivity was excellent (99%) since
the only by-product detected by GC was iodobenzene. How-
ever, it is not easy to add KCN, which is a solid that is only
slightly soluble in DMF, to the reaction mixture at 1108C.
The use of a liquid cyanation agent, acetone cyanohydrin,
was therefore investigated (Scheme 3). This cyanation re-
Scheme 4. Mechanism proposed for the halide exchange–cyanation pro-
cess.
Cycle A: This cycle corresponds to the thermodynamic equi-
librium between aryl bromide and aryl iodide. The first step
involves the oxidative addition of bromobenzene to the cat-
alytically active copper species [CuI] which is proposed to
be a copper(i)–ligand complex.[27] The resulting copper(iii)
III
À
À
intermediate (Ph [Cu ] Br) undergoes nucleophilic substi-
tution of the copper-bound bromide by an iodide ion from
III
À
À
potassium iodide to give Ph [Cu ] I. The third step in-
volves the reductive elimination of iodobenzene and the
subsequent regeneration of the active copper(i) species.
Cycle B: This cycle involves cy-
anohydrin and describes the
synthesis of benzonitrile itself.
In the first step the copper(iii)
III
À
À
intermediate (Ph [Cu ] I) is
trapped by the cyanide ion,
thus displacing the equilibrium
between PhBr and PhI in cycle
À
A. The resulting product Ph
Scheme 3. Cyanation of PhBr with acetone cyanohydrin. General conditions: bromobenzene (0.5 mmol), po-
III
À
tassium iodide (0.25 mmol), CuI (0.05 mmol), ligand 1 (0.10 mmol), DMF (300 mL), 6 h, 1108C followed by ad- [Cu ] CN then allows, by re-
dition of acetone cyanohydrin (0.55 mmol) and tributylamine (0.6 mmol). Yields were determined by GC with
1,3-dimethoxybenzene as the internal standard.
ductive elimination, the forma-
tion of the expected benzoni-
trile and the regeneration of
agent, used in the presence of a base, is cheap, commercially
available on an industrial scale,[19,31] and, to the best of our
knowledge, it has never been used in copper-catalyzed cyan-
ation reactions.
As in the case of KCN, the introduction of acetone cyano-
hydrin and tributylamine to the reaction mixture at the start
of the reaction led to deactivation of the catalyst since
halide exchange did not occur. This difficulty was once
again overcome by delaying the addition of the cyanation
reagent (Scheme 3). Benzonitrile was successfully obtained
from bromobenzene with a yield of 80% and an excellent
selectivity (>98%). A possible mechanism for this reaction
is presented in Scheme 4.
the active copper species [CuI]. The cyanide ion is generated
in situ from both cyanohydrin and tributylamine and then
consumed according to the availability of the intermediate
III
À
À
Ph [Cu ] I. Another interesting feature of our protocol is
the regeneration of the iodide ion (cycle B). Only 0.5 equiv-
alents (based on the bromobenzene) of potassium iodide
were used. A lower loading of KI is possible but a decrease
in the reaction rate is observed.
As briefly mentioned above, the deactivation of catalysts
by an excess of dissolved cyanide ions has already been
demonstrated in the nickel- and palladium-mediated cyana-
tion of aryl bromides and chlorides.[9b,12k,13h,14b,19] Several
methods to avoid such poisoning have been proposed in the
2486
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2483 – 2492

Copper-Catalyzed Cyanation of Aryl Iodides and Bromides
FULL PAPER
literature (Scheme 5). As with nickel-based catalysts, the use
of solvents with lower cyanide solubility permitted the cyan-
ation of heteroaromatic halides to proceed.[9b] In palladium-
mediated cyanation reactions, a method involving the con-
tinuous addition of the cyanation agent (acetone cyanohy-
drin[19] or trimethylsilyl cyanide (TMSCN)[14b]) to the reac-
tion mixture was proposed for the cyanodehalogenation of
aryl halides. The aim of both of these protocols was to keep
the concentration of dissolved cyanide ions low to prevent
poisoning of the catalyst. The addition of a further reagent
can also prevent inhibition of the active catalyst. For exam-
ple, the use of zinc acetate allowed the palladium-catalyzed
cyanation of aryl bromides; however the role of this additive
is not yet clearly understood.[13h] The use of N,N,N’,N’-tetra-
methylethylenediamine (TMEDA) or 1,1’-methylenedipiper-
idine (MDP) as co-catalyst has also been reported. These
amines should be able to substitute cyanide ions on the pal-
ladium center, thus allowing the cyanation of aryl bromides
and chlorides.[12k,14b,19] Similarly, the recently reported
copper-catalyzed cyanation of aryl bromides seemed to re-
quire ten-fold more of the ligand N,N’-dimethylethylenedia-
mine (DMEDA) than the copper salt.[24] These unusual reac-
tion conditions may be aimed at countering the binding of
cyanide ions to copper in order to avoid catalyst deactiva-
tion.
We have demonstrated that high concentrations of cya-
nide ions did not prevent the copper-catalyzed cyanation of
iodobenzene, benzonitrile being quantitatively obtained
even though cyanide ions were present from the start of the
reaction (Figure 1). Since the cyanation of bromobenzene
did not take place under the same reaction conditions, we
considered obtaining the corresponding aryl nitrile by the
catalyzed production of iodobenzene from bromobenzene
and potassium iodide. However, we noticed that introduc-
tion of the cyanating agent (potassium cyanide or cyanohy-
drin) from the beginning of the reaction prevented the
halide exchange from taking place. To explain this phenom-
cycle A). Thus we overcame the poisoning of our copper-
based catalyst by introducing the cyanide source once a suf-
À
ficient amount of iodobenzene and/or intermediate Ph
III
À
[Cu ] I had been generated in the reaction mixture. As a
consequence, the cyanide ions should essentially be used to
cyanate the previously formed iodobenzene (Scheme 4,
cycle B). The resulting decrease in the concentration of the
cyanide ions should thus limit the formation of
(1Àx)À
[Cu](CN)x
complexes, whereas the number of copper
species able to afford oxidative addition of bromobenzene
to [CuI] should increase thus enabling halogen exchange fol-
lowed by cyanation (Scheme 4). Note that our protocol al-
lowed the cyanide source to be introduced at once without
requiring the addition of a further reagent to avoid poison-
ing.
Copper-catalyzed cyanation of substituted aryl bromides:
We were interested in applying our catalytic system to the
cyanation of a variety of aryl halides. It has been shown that
acetone cyanohydrin can quantitatively convert iodobenzene
into benzonitrile in DMF after 48 h at 908C. We then fo-
cused our attention on more challenging aryl bromides.
Thus the cyanation of aryl bromides substituted with differ-
ent electron-donating and -withdrawing groups was investi-
gated by using acetone cyanohydrin in the presence of cata-
lytic amounts of copper, ligands and potassium iodide
(Table 1). Note that to facilitate the isolation of the aryl ni-
triles, tributylamine can be replaced by a more volatile terti-
ary amine (diisopropylethylamine, for example), which can
be removed more easily from the crude mixture before puri-
fication by column chromatography.
The expected aryl nitriles were synthesized whatever the
nature of the substituent and the rates of the reactions were
not significantly affected by electronic effects. Thus very
good yields of aryl nitriles were obtained from the electron-
poor p-bromobenzotrifluoride, m-nitrobromobenzene or p-
bromoacetophenone (Table 1, entries 2, 5 and 7) and also
from the electron-rich p-methylbromobenzene and m-me-
thoxybromobenzene (Table 1, entries 6 and 8).
enon we assumed that copper(i)–cyanide complexes
(1Àx)À
[Cu](CN)x
(x=1–4), which should not promote the oxi-
dative addition of bromobenzene, were formed in situ. The
affinity of copper(i) for cyanide ions has indeed already
been reported.[32] Such complexes would be formed to the
detriment of the active copper species [CuI] (Scheme 4,
In these examples, with the exception of the nitro sub-
stituent, cyanation was totally selective with respect to the
cyanide ion and almost totally selective with respect to the
aryl bromides. Indeed, the only by-products detected by GC
Scheme 5. Reported methods that avoid the deactivation of catalysts by an excess of dissolved cyanide ions.
Chem. Eur. J. 2005, 11, 2483 – 2492
www.chemeurj.org
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2487

H.-J. Cristau, M. Taillefer et al.
Table 1. Cyanation of substituted aryl bromides with acetone cyanohydrin using catalytic amounts of potassi-
um iodide.[a]
ceptable. Our copper-based cat-
alytic system is therefore com-
patible
with
the
nitro
substituent, whereas nickel-cat-
alyzed cyanation of nitro-substi-
tuted aryl halides failed.[9a]
Note that the presence of an
amino substituent also allows
the synthesis of the correspond-
ing nitrile in a very good yield
and with excellent selectivity
(>99%) (Table 1, entry 4).
Indeed the amino group itself is
not arylated by aryl bromides
under the conditions used.
Entry
1
ArX
ArX
Time [h]
Yield [%][b]
Selectivity [%][c]
48
60
80
90 (88)
98
99
9a
9b
9c
9d
9e
9 f
48
60
85
94 (87)
97
99
2
3
4
5
6
48
100 (98)
100
The cyanation of o-bromoto-
luene was more troublesome
and sluggish (40% yield after
48
60
83
98 (70)
97
>99
48
60
82
81 (75)
88[d]
85[e]
70 h
at
1108C,
Table 1,
entry 10). This lack of reactivity
could be due to steric hindrance
in the substrate. However, o-
tolunitrile can be obtained in a
quantitative yield from o-iodo-
toluene at 1108C (Table 1,
entry 11). The cyanation reac-
tion has also been extended to
heteroaryl bromides, for exam-
ple, 2-bromopyridine is quanti-
tatively and selectively convert-
48
60
74
89 (80)
96
99
48
60
92
100 (96)
99
100
7
8
9g
9h
48
60
70
62
81 (79)
97
97
97
>99
48
60
66
98 (94)
99
>99
9
9i
9j
ed
into
2-cyanopyridine
(Table 1, entry 3). Finally, dif-
ferences in the reactivity of aryl
halides in the oxidative addition
to copper(i) active species can
be exploited to obtain a mono-
substituted product from reac-
tions involving di- or trihalo-
benzenes (Table 1, entries 9 and
12). The cyanation of 4-chloro-
48
70
29
43 (40)
99
99
10
11[f]
12
9j
48[g]
98 (95)
>99
48
60
86
96 (86)
97
99
9k
[a] General conditions for GC yields: aryl bromide (0.5 mmol), potassium iodide (0.25 mmol), CuI
(0.05 mmol), ligand 1 (0.10 mmol), DMF (300 mL), 6 h, 1108C followed by in situ addition of acetone cyanohy-
drin (0.55 mmol) and tributylamine (0.6 mmol). General conditions for isolated yields: aryl bromide (5 mmol),
potassium iodide (2.5 mmol), CuI (0.5 mmol), ligand 1 (1.0 mmol), DMF (3 mL), 6 h, 1108C followed by in
situ addition of acetone cyanohydrin (5.5 mmol) and diisopropylethylamine (6.0 mmol). [b] Yields refer to GC
bromobenzene
(Table 1,
entry 12) and 3,5-difluorobro-
mobenzene (Table 1, entry 12)
yields (using 1,3-dimethoxybenzene as internal standard) and yields in parentheses refer to isolated yields; took place exclusively at the
yields are given with respect to aryl bromides. [c] Selectivities are given with respect to aryl bromides. With
bromine position.
the exception of entry 5, only negligible amounts of one by-product (the corresponding aryl iodide) are ob-
served. [d] By-products: 3.5% of PhNO2 and 2.5% of p-NH2C6H4CN. [e] By-products: 10.0% of PhNO2 and
3.5% of p-NH2C6H4CN. [f] General conditions: 2-iodotoluene (0.5 mmol), potassium iodide (0.25 mmol), CuI
Conclusions
(0.05 mmol), ligand 1 (0.10 mmol), acetone cyanohydrin (0.55 mmol), tributylamine (0.6 mmol) and DMF
(300 mL). [g] The reaction time and the temperature were not optimized.
To conclude, herein we propose
were small amounts of aryl iodides corresponding to the
aryl bromides used. Hydrodehalogenation never occurred
during the reactions and by-products resulting from biaryl
coupling were never observed. In the case of 3-nitrobromo-
benzene (Table 1, entry 5), small amounts of hydrodehaloge-
nation (C6H5NO2) and nitro-reduction (BrC6H4NH2) prod-
ucts were detected but the selectivity of 88% remains ac-
a high-yielding, copper-catalyzed method for the cyanation
of aryl iodides and bromides. Our catalytic system involves
not only the use of catalytic amounts of copper salts but
also catalytic amounts of ligands. Aryl nitriles were obtained
from aryl bromides through the copper-catalyzed in situ pro-
duction of the corresponding aryl iodides by using catalytic
amounts of potassium iodide. Furthermore, we have shown
2488
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2483 – 2492

Copper-Catalyzed Cyanation of Aryl Iodides and Bromides
FULL PAPER
General procedure for the cyanation of aryl bromides (5mmol scale):
After standard cycles of evacuation and back-filling with dry and pure ni-
trogen, an oven-dried Radley tube (Carousel “reaction stations
RR98030”) equipped with a magnetic stirring bar was charged with CuI
(95.2 mg, 0.5 mmol), 1,10-phenanthroline (181.2 mg, 1.0 mmol), KI
(415.0 mg, 2.5 mmol) and the aryl bromide (5.0 mmol), if a solid. The
tube was evacuated and back-filled with nitrogen. If a liquid, the aryl
bromide was added by syringe under a stream of nitrogen at room tem-
perature, followed by anhydrous and degassed DMF (3 mL). The tube
was sealed under a positive pressure of nitrogen, and stirred and heated
at 1108C for six hours. Acetone cyanohydrin (502 mL, 5.5 mmol) and dii-
sopropylethylamine (1.05 mL, 6.0 mmol) were then added by syringe at
1108C and the reaction mixture was stirred for 60 h at this temperature.
After cooling to room temperature, the mixture was diluted with diethyl
ether (~50 mL) and filtered through a plug of Celite, the filter cake
being further washed with diethyl ether (~10 mL). The filtrate was
washed twice with water (2ꢃ~30 mL). The aqueous phases were com-
bined and extracted twice with diethyl ether (~30 mL). The organic
layers were combined, dried over MgSO4, filtered and concentrated in
vacuo to yield a brown oil. Excess of diisopropylethylamine was then dis-
tilled and the crude product obtained was purified by silica gel chroma-
tography with hexanes and ethyl acetate as eluent.
that acetone cyanohydrin can be used as a cyanating agent.
To the best of our knowledge, copper-catalyzed cyanation
with cyanohydrin has not been previously reported. This
method avoids deactivation of the catalyst by cyanide ions.
Finally, this protocol is practical on the laboratory scale and
can easily be adapted to an industrial scale.[28d,31] We are cur-
rently working to extend the scope of our cyanation method
to other aromatic compounds such as aryl triflates and espe-
cially aryl chlorides and the results will be reported in due
course.
Experimental Section
General: Column chromatography was performed with SDS 60 A C.C
silica gel (35–70 mm). Thin-layer chromatography was carried out with
Merck silica gel 60 F254 plates. All products were characterized by analy-
sis of their NMR, GC-MS and IR spectra. NMR spectra were recorded
at 208C on Bruker AC 200, DRX-250 and DRX-400 spectrometers work-
General procedure for reactivity comparisons or screening of reaction
conditions (0.5 mmol scale):
1
ing respectively at 200.13, 250.13, and 400.13 MHz for H, at 50.32, 62.90,
and 100.61 MHz for 13C and at 188.31, 236.36, and 376.50 for 19F NMR
spectroscopy. CDCl3 was used as solvent unless otherwise stated. Cou-
pling constants are reported in Hz and chemical shifts in ppm [relative to
TMS for 1H and {1H}13C (d=77.00 ppm for the CDCl3 signal) and to
CFCl3 for {1H}19F]. The first-order peak patterns are indicated as s (sin-
glet), d (doublet), t (triplet) and q (quadruplet). Complex non-first-order
signals are indicated as m (multiplet) and broad signals as br. 13C NMR
signals were assigned by using HMQC and HMBC sequences. Gas chro-
matography-mass spectra (GC-MS) were recorded using an Agilent Tech-
nologies 6890 N instrument with an Agilent 5973 N mass detector (EI)
and a HP5-MS 30 mꢃ0.25 mm capillary apolar column (stationary phase:
5% diphenyldimethylpolysiloxane film, 0.25 mm). GC-MS method: initial
temperature, 458C; initial time, 2 min; ramp, 108CminÀ1; final tempera-
ture, 2508C; final time, 10 min. IR spectra were recorded with a Nicolet
210 FTIR instrument (neat, thin film for liquid products and KBr pellet
or in carbon tetrachloride solution for solid products). FAB+ mass spec-
tra were recorded with a JEOL JMS-DX300 spectrometer (3 keV, xenon)
in a m-nitrobenzyl alcohol matrix. Melting points were determined by
using a Bꢄchi B-540 apparatus and are uncorrected.
Cyanation of aryl bromides with acetone cyanohydrin: The above proce-
dure was applied on
a 0.5 mmol scale and by using tributylamine
(143 mL, 0.6 mmol) instead of diisopropylethylamine. After heating for
the required period of time, the reaction mixture was allowed to cool to
room temperature and was diluted with diethyl ether (5 mL). 1,3-Dime-
thoxybenzene (65 mL) was then added as an internal standard. A small
sample of the reaction mixture was taken and filtered through a plug of
Celite, the filter cake being further washed with diethyl ether. The filtrate
was washed three times with water and analyzed by gas chromatography.
The GC yields were determined by obtaining correction factors using au-
thentic samples of the expected products.
Cyanation of iodobenzene by KCN: After standard cycles of evacuation
and back-filling with dry and pure nitrogen, an oven-dried Radley tube
(Carousel “reaction stations RR98030”) equipped with a magnetic stir-
ring bar was charged with CuI (9.5 mg, 0.05 mmol), ligand (0.1 mmol)
and KCN (35.6 mg, 0.55 mmol). The tube was evacuated and back-filled
with nitrogen. Iodobenzene was added under a stream of nitrogen by sy-
ringe at room temperature, followed by anhydrous and degassed DMF
(300 mL). The tube was sealed under a positive pressure of nitrogen and
stirred and heated at 1108C for 24 h. The reaction mixture was allowed
to cool to room temperature and was diluted with diethyl ether (5 mL).
1,3-Dimethoxybenzene (65 mL) was added as an internal standard. A
small sample of the reaction mixture was taken and filtered through a
plug of Celite, the filter cake being further washed with diethyl ether.
The filtrate was washed three times with water and analyzed by gas chro-
matography. The GC yields were determined by obtaining the correction
factor using an authentic sample of benzonitrile.
Materials
Caution: Safety precautions must be taken with potassium cyanide and
acetone cyanohydrin. All reactions were carried out in 35 mL Schlenk
tubes or in Carousel “reaction stations RR98030” Radley tubes under
pure and dry nitrogen. DMF was distilled under vacuum from MgSO4
and stored, protected from light, on 4 ꢅ activated molecular sieves under
nitrogen. KI (SDS) and KCN (Fluka) were ground to a fine powder. The
former was stored under vacuum at 1008C in the presence of P4O10. The
latter was dried in vacuo and stored under nitrogen. All other solid mate-
rials were stored in the presence of P4O10 in a bench-top desiccator under
vacuum at room temperature and weighed in air. Copper(i) iodide was
purified according to literature procedures[33] and stored protected from
light. 1,10-Phenanthroline (ligand 1) and salicylaldoxime (ligand 8) were
purchased from commercial sources. The latter was recrystallized in pe-
troleum ether prior to use. The following ligands were synthesized ac-
cording to or by adapting literature procedures: 3,[34,35] 6[36] and 7.[37] The
stereochemistry of oxime-type ligands 3 and 8 has not been determined.
The syntheses of ligands 2 and 4 are reported below. All aryl halides, ace-
tone cyanohydrin and amines (tributylamine or diisopropylethylamine)
were purchased from commercial sources (Aldrich, Acros, Avocado,
Fluka, Lancaster). Solids were recrystallized in an appropriate solvent[38]
while liquids were distilled under vacuum and stored under nitrogen.
Special care was taken with liquid aryl iodides: the samples were regular-
ly distilled and stored protected from light. Amines were distilled from
potassium hydroxide.
Cyanation of iodobenzene by acetone cyanohydrin: After standard cycles
of evacuation and back-filling with dry and pure nitrogen, an oven-dried
Radley tube (Carousel “reaction stations RR98030”) equipped with a
magnetic stirring bar was charged with CuI (9.5 mg, 0.05 mmol) and 1,10-
phen (18.0 mg, 0.1 mmol). The tube was evacuated and back-filled with
nitrogen. Iodobenzene (56 mL, 0.5 mmol), acetone cyanohydrin (50 mL,
0.55 mmol) and triethylamine (84 mL, 0.6 mmol) were added under a
stream of nitrogen by syringe at room temperature, followed by anhy-
drous and degassed DMF (300 mL). The tube was sealed under a positive
pressure of nitrogen, and stirred and heated at 908C for 48 h. The reac-
tion mixture was allowed to cool to room temperature and was then di-
luted with diethyl ether (5 mL). 1,3-Dimethoxybenzene (65 mL) was
added as an internal standard. A small sample of the reaction mixture
was taken and filtered through a plug of Celite, the filter cake being fur-
ther washed with diethyl ether. The filtrate was washed three times with
water and analyzed by gas chromatography. The quantitative GC yield
Chem. Eur. J. 2005, 11, 2483 – 2492
www.chemeurj.org
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2489

H.-J. Cristau, M. Taillefer et al.
was determined by obtaining the correction factor using an authentic
sample of benzonitrile.
Synthesis of ligands 2 and 4:
N,N-Dimethyl-N’-(pyridin-2-ylmethylene)ethane-1,2-diamine (Dmeda-Py-
Al) (2): Anhydrous magnesium sulfate (3.6 g, 30.0 mmol) and rac-trans-
1,2-diaminocyclohexane (2.15 mL, 20.0 mmol) were successively added to
a solution of 2-pyridinecarboxaldehyde (1.90 mL, 20.0 mmol) in absolute
ethanol (20 mL). The mixture was stirred for 72 h at room temperature
and filtered through a frit. The solid was discarded and the filtrate was
concentrated in vacuo. The desired product (2.8 g, 78%) was obtained as
Aryl nitrile (9a): The cyanation of bromobenzene (539 mL, 5.0 mmol)
was achieved by following the general procedure. The resulting crude
oily residue was purified by chromatography on silica gel to provide
452 mg (88% yield) of the desired product as a colourless oil. 1H NMR:
d=7.71–7.47 ppm (m, H-3, H-4, H-5, H-6, H-7); 13C NMR: d=132.30 (C-
5), 131.51 (C-3, C-7), 128.65 (C-4, C-6), 118.27 (C-1), 111.77 ppm (C-2);
GC-MS (EI): rt=9.10 min, m/z: 103; Rf =0.31 (hexanes/AcOEt, 95:5).
a brown oil. 1H NMR: d=8.54 (ddd, 3JH5,H6 =4.9 Hz, 4JH4,H6 =1.7 Hz,
3
5JH3,H6 =1.0 Hz, H-1), 8.35 (t, 4JH7,H9 =1.5 Hz, H-7), 7.92 (ddd, JH3,H4
=
=
=
3
8.0 Hz, 4JH3,H5 =1.2 Hz, 5JH3,H6 =1.0 Hz, 1H, H-3), 7.92 (ddd, JH3,H4
3
8.0 Hz, 3JH4,H5 =7.6 Hz, 4JH4,H6 =1.7 Hz, 1H, H-4), 7.25 (ddd, JH4,H5
7.6 Hz, 3JH5,H6 =4.9 Hz, 4JH3,H5 =1.2 Hz, H-5), 3.74 (td, 3JH9,H10 =7.1 Hz,
4JH7,H9 =1.5 Hz, 2H, H-9), 2.61 (t, 3JH9,H10 =7.1 Hz, 2H, H-10), 2.36 ppm
(s, 6H, H-11); 13C NMR: d=162.29 (C-7), 159.96 (C-2), 149.60 (C-6),
136.10 (C-4), 124.29 (C-5), 120.56 (C-3), 61.77 (C-9), 59.79 (C-10),
45.60 ppm (C-11); IR (CCl4): n=3338, 3055, 2973, 2933, 2853, 2822, 2765,
2364, 2336, 1675, 1650, 1583, 1563, 1465, 1431, 1044, 989, 776, 744,
618 cmÀ1: GC-MS: retention time, rt=16.44 min, m/z: 177
Aryl nitrile (9b): The cyanation of 4-bromobenzotrifluoride (700 mL,
5.0 mmol) was achieved by following the general procedure. The result-
ing crude oily residue was purified by chromatography on silica gel to
provide 745 mg (87% yield) of the desired product as a white powder.
M.p. 38–398C (hexanes/AcOEt) [lit.[39]: 36–378C (hexanes)]; 1H NMR:
d=7.84 (m, H-5, H-7), 7.78 ppm (m, H-4, H-6); 13C NMR: d=134.43 (q,
2JC,F =32.7 Hz, C-5), 132.62 (s, C-3, C-7), 126.09 (q, 3JC,F =3.7 Hz, C-4, C-
5
6), 123.00 (q, 1JC,F =273.0 Hz, C-8), 117.36 (s, C-1), 115.99 ppm (q, JC,F
=
1.5 Hz, C-2); 19F NMR: d=(À)63.99 ppm (s, F-8, CF3); GC-MS (EI): rt=
18.50 min, m/z: 171; Rf =0.45 (hexanes/AcOEt, 90:10).
Aryl nitrile (9c): The cyanation of 2-bromopyridine (477 mL, 5.0 mmol)
was achieved by following the general procedure. The resulting crude
oily residue was purified by chromatography on silica gel (eluent: hex-
anes/ethyl acetate, 80:20) to provide 510 mg (98% yield) of the desired
product as a white powder. M.p. 26–278C (hexanes/AcOEt) (lit.:[40] 26–
5
288C); 1H NMR: d=8.75 (ddd, 3JH4,H5 =4.8 Hz, 4JH4,H6 =1.7 Hz, JH4,H7
=
1.0 Hz, H-4), 7.88 (ddd, 3JH6,H7 =7.8 Hz, 3JH5,H6 =7.7 Hz, 4JH4,H6 =1.7 Hz,
1H, H-6), 7.73 (ddd, 3JH6,H7 =7.8 Hz, 3JH5,H7 =1.3 Hz, 5JH4,H7 =1.0 Hz, H-
7), 7.57 ppm (ddd, 3JH5,H6 =7.7 Hz, 3JH4,H5 =4.8 Hz, 4JH5,H7 =1.3 Hz, H-5);
13C NMR: d=150.93 (C-4), 136.99 (C-6), 133.64 (C-2), 128.39 (C-7),
126.88 (C-5), 117.04 ppm (C-1); GC-MS (EI): rt=10.56 min, m/z: 104;
Rf =0.21 (hexanes/AcOEt, 80:20).
trans-1,2-Bis(2’-pyridylideneamino)cyclohexane (Chxn-Py-Al) (4): Anhy-
drous magnesium sulfate (12.65 g, 105.1 mmol) and rac-trans-1,2-diamino-
cyclohexane (4.2 mL, 35.0 mmol) were successively added to a solution
of 2-pyridinecarboxaldehyde (6.66 mL, 70.0 mmol) in absolute ethanol
(50 mL). The mixture was stirred for 20 h at room temperature, refluxed
for 2.5 h and filtered through a frit while still hot. The solid was discarded
and the filtrate was concentrated in vacuo. The residue was recrystallized
in ethanol to provide 8.2 g (80% yield) of the desired product as pale
yellow crystals. M.p. 140–1418C (EtOH); 1H NMR: d=8.54 (ddd,
3JH5,H6 =4.9 Hz, 4JH4,H6 =1.7 Hz, 5JH3,H6 =1.0 Hz, 2H, H-6), 8.30 (brs, 2H,
Aryl nitrile (9d): The cyanation of 4-bromoaniline (860.0 mg, 5.0 mmol)
was achieved by following the general procedure. The resulting crude
oily residue was purified by chromatography on silica gel (eluent: hex-
anes/ethyl acetate, 70:30) to provide 413 mg (70% yield) of the desired
product as a white solid. M.p. 84–858C (hexanes/AcOEt) [lit.:[41] 868C
1
(petroleum ether)]; H NMR: d=7.43 (m, H-3, H-7), 6.67 (m, H-4, H-6),
H-7), 7.87 (ddd, 3JH3,H4 =7.9 Hz, 4JH3,H5 =1.5 Hz, 5JH3,H6 =1.0 Hz, 2H, H-
4.21ppm (brs, H-8, NH2); 13C NMR: d=150.49 (C-5), 133.67 (C-3, C-7),
120.16 (C-1), 114.32 (C-4, C-6), 99.77 ppm (C-2); GC-MS (EI): rt=
16.33 min, m/z: 118; Rf =0.21 (hexanes/AcOEt, 70:30).
5
3), 7.63 (dddd, 3JH,H =7.9 Hz, 3JH3,H4 =7.5 Hz, 4JH4,H5 =1.7 Hz, JH4,H6
=
=
4
0.6 Hz, 2H, H-4), 7.22 (ddd, 3JH4,H5 =7.5 Hz, 3JH5,H6 =4.9 Hz, JH3,H5
1.5 Hz, 2H, H-5), 3.50 (m, 2H, H-9), 1.83 (m, 6H, H-10ax, H-11), 1.40–
1.55 ppm (m, 2H, H-10eq); 13C NMR: d=161.4 (C-7), 154.6 (C-2), 149.2
(C-6), 136.4 (C-4), 124.4 (C-5), 121.3 (C-3), 73.5 (C-9), 32.7 (C-10),
24.3 ppm (C-11); IR (KBr): n=3273, 3071, 3055, 3050, 2941, 2934, 2925,
2865, 2857, 2850, 1644, 1586, 1566, 1467, 1449, 1433, 1372, 1338, 991, 934,
Aryl nitrile (9e): The cyanation of 3-nitrobromobenzene (1.01 g,
5.0 mmol) was achieved by following the general procedure. The result-
ing crude oily residue was purified by chromatography on silica gel
(eluent: hexanes/ethyl acetate, 90:10) to provide 555 mg (75% yield) of
the desired product as a white solid. M.p. 115–1168C (hexanes/AcOEt)
867, 839, 771, 743 cmÀ1
; MS (FAB+, NBA): m/z (%): 293 (100)
4
(lit.:[42] 115–1178C); 1H NMR: d=8.57 (ddd, 4JH3,H5 =2.3 Hz, JH3,H7
=
[M++H], 107 (52), 92 (38), 119 (25), 294 (23) [M++2H], 204 (22), 79
(21), 187 (20), 585 (!) [2M++H].
1.5 Hz, 5JH3,H6 =0.6 Hz, H-3), 8.51 (ddd, 3JH5,H6 =8.3 Hz, 4JH3,H5 =2.3 Hz,
4
4JH5,H7 =1.1 Hz, H-5), 8.03 (ddd, 3JH6,H7 =7.8 Hz, 4JH3,H7 =1.5 Hz, JH5,H7
=
5
1.1 Hz, H-7), 7.77 ppm (ddd, 3JH5,H6 =8.3 Hz, 3JH6,H7 =7.8 Hz, JH3,H6
=
Aryl nitriles: The following numbering system differs from the conven-
tional one to allow convenient comparison of chemical shifts.
0.6 Hz, H-6); 13C NMR: d=148.19 (C-4), 137.57 (C-7), 130.64 (C-6),
2490
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2483 – 2492

Copper-Catalyzed Cyanation of Aryl Iodides and Bromides
FULL PAPER
127.45 (C-5) 127.20 (C-3), 116.50 (C-2), 114.10 ppm (C-1); GC-MS (EI):
rt=15.19 min, m/z: 148; Rf =0.25 (hexanes/AcOEt, 75:25).
diethyl ether (~10 mL). The filtrate was washed twice with water
(~30 mLꢃ2). The combined aqueous phases were twice extracted with
diethyl ether (2ꢃ~30 mL). The organic layers were combined, dried over
MgSO4, filtered and concentrated in vacuo to yield a brown oil. Diisopro-
pylethylamine was then distilled and the crude product obtained was pu-
rified on silica gel (eluent: hexanes/ethyl acetate, 90:10) to provide
554 mg (95% yield) of the desired product as a colourless oil. 1H NMR:
d=7.60 (ddd, 3JH6,H7 =7.6 Hz, 4JH5,H7 =1.4 Hz, 5JH4,H7 =0.6 Hz, H-7), 7.50
Aryl nitrile (9 f): The cyanation of 4-bromotoluene (855.2 mg, 5.0 mmol)
was achieved by following the general procedure. The resulting crude
oily residue was purified by chromatography on silica gel (eluent: hex-
anes/ethyl acetate, 90:10) to provide 470 mg (80% yield) of the desired
product as a white solid. M.p. 27–288C (hexanes/AcOEt) (lit.:[43] 288C);
1H NMR: d=7.56 (m, H-3, H-7), 7.29 (m, H-4, H-6), 2.36 ppm (s, 3H, H-
8, CH3); 13C NMR: d=143.54 (C-5), 131.81 (C-3, C-7), 129.66 (C-4, C-6),
118.96 (C-1), 109.06 (C-2), 21.62 ppm (C-8); GC-MS (EI): rt=11.34 min,
m/z: 117; Rf =0.36 (hexanes/AcOEt, 90:10).
3
3
(td, JH4,H5
=
3JH5,H6 =7.6 Hz, 4JH5,H7 =1.4 Hz, H-5), 7.33 (ddd, JH4,H5
=
3
7.6 Hz, 4JH4,H6 =1.3 Hz, 4JH4,H7 =0.6 Hz, H-4), 7.28 (td, JH6,H7
=
3JH5,H6
=
7.6 Hz, 4JH4,H6 =1.3 Hz, H-6), 2.56 ppm (s, 3H, H-8, CH3); 13C NMR: d=
141.69 (C-3), 132.48 (C-5), 132.27 (C-7), 130.06 (C-4), 126.06 (C-6),
117.92 (C-1), 112.54 (C-2), 20.23 ppm (C-8); GC-MS (EI): rt=10.68 min,
m/z: 117; Rf =0.57 (hexanes/AcOEt, 90:10).
Aryl nitrile (9g): The cyanation of 4-acetylbromobenzene (995.3 mg,
5.0 mmol) was achieved by following the general procedure. The result-
ing crude oily residue was purified by chromatography on silica gel
(eluent: hexanes/ethyl acetate, 80:20) to provide 698 mg (96% yield) of
the desired product as a pale yellow solid. M.p. 57–588C (hexanes/
AcOEt) [lit.:[44] 58–598C (aqueous ethanol)]; 1H NMR: d=8.07 (m, H-4,
H-6), 7.80 (m, H-3, H-7), 2.67 ppm (s, 3H, H-9, CH3); 13C NMR: d=
196.48 (C-8), 139.82 (C-5), 132.42 (C-4, C-6), 128.61 (C-3, C-7), 117.83
(C-1), 116.28 (C-2), 26.67 ppm (C-9); GC-MS (EI): rt=15.13 min, m/z:
145; Rf =0.21 (hexanes/AcOEt, 80:20).
Acknowledgements
We thank RHODIA Organique Fine and the CNRS for a Ph.D. grant
and financial support. Prof. A. Fruchier is gratefully acknowledged for
helpful discussions and assistance with the NMR analyses.
Aryl nitrile (9h): The cyanation of 3-methoxybromobenzene (633 mL,
5.0 mmol) was achieved by following the general procedure. The result-
ing crude oily residue was purified by chromatography on silica gel
[1] R. C. Larock, Comprehensive Organic Transformations. A guide to
Functional Group Preparations, VCH, New York, 1989.
[2] a) W. Oppolzer, D. A. Roberts, Helv. Chim. Acta 1980, 63, 1703–
1705; b) M. H. Gebs, R. H. Abeles, J. Med. Chem. 1986, 29, 585–
589.
[3] M. Rehan, A. D. Schꢄlter, W. J. Feast, Synthesis 1988, 386–388.
[4] A. G. M. Barret, W. E. Broderick, B. M. Hoffmann, C. S. Velasquez,
J. Org. Chem. 1989, 54, 3233–3234.
[5] C. Tomlin, The Pesticide Manual, 10th ed., Crop Protection Publica-
tions, The Bitish Crop Protection Council & The Royal Society of
Chemistry, UK, 1994, pp. 160, 193, 241, 300, 374, 444, 482, 579, 598,
1045.
[6] A. Kleemann, J. Engel, B. Kutsher, D. Reichert, Pharmaceutical sub-
stances: Syntheses, Patents, Applications, 4th ed., Thieme, Stuttgart,
2001.
[7] G. P. Ellis, T. M. Romney-Alexander, Chem. Rev. 1987, 87, 779–794,
and references cited therein.
[8] a) L. Cassar, J. Organomet. Chem. 1973, 54, C57–C58; b) L. Cassar,
S. Ferrera, M. Foꢆ, Adv. Chem. Ser. 132, Homogeneous catalysis II
1974, 252–273; c) L. Cassar, M. Foꢆ, F. Montanari, G. P. Mainelli, J.
Organomet. Chem. 1979, 173.
[9] a) Y. Sakakibara, F. Okuda, A. Shimoyabashi, K. Kirino, M. Sakai,
N. Uchino, K. Takagi, Bull. Chem. Soc. Jpn. 1988, 61, 1985–1990;
b) Y. Sakakibara, Y. Ido, K. Sasaki, M. Sakai, N. Uchino, Bull.
Chem. Soc. Jpn. 1993, 66, 2776–2778.
[10] R. K. Arvela, N. E. Leadbeater, J. Org. Chem. 2003, 68, 9122–9125.
[11] M. Sundermeier, A. Zapf, M. Beller, Eur. J. Inorg. Chem. 2003,
3513–3526, and references therein.
[12] a) K. Takagi, T. Okamoto, Y. Sakakibara, S. Oka, Chem. Lett. 1973,
471–474; b) A. Sekiya, N. Ishikawa, Chem. Lett. 1975, 277–278;
c) K. Takagi, T. Okamoto, Y. Sakakibara, A. Ohno, S. Oka, N.
Hayama, Bull. Chem. Soc. Jpn. 1975, 48, 3298–3301; d) K. Takagi,
T. Okamoto, Y. Sakakibara, A. Ohno, S. Oka, N. Hayama, Bull.
Chem. Soc. Jpn. 1976, 49, 3177–3180; e) Y. Akita, M. Shimazaki, A.
Ohta, Synthesis 1981, 974; f) J. R. Dalton, S. L. Regen, J. Org.
Chem. 1979, 44, 4443–4444; g) K. Takagi, K. Sasaki, Y. Sakakibara,
Bull. Chem. Soc. Jpn. 1991, 64, 1118–1121; h) T. Okano, M. Iwa-
hara, J. Kiji, Synlett 1998, 243–244; i) B. A. Anderson, E. C. Bell,
F. O. Ginah, K. Harn, L. M. Pagh, J. P. Wepsiec, J. Org. Chem. 1998,
63, 8224–8228; j) M. Sundermeier, A. Zapf, M. Beller, J. Sans, Tetra-
hedron Lett. 2001, 42, 6707–6710; k) M. Sundermeier, A. Zapf, S.
Mutyala, W. Baumann, J. Sans, S. Weiss, M. Beller, Chem. Eur. J.
2003, 9, 1828–1836.
(eluent: hexanes/ethyl acetate, 90:10) to provide 524 mg (79% yield) of
3
the desired product as a colourless oil. 1H NMR:[45] d=7.38 (m, JH5,H6
=
8.4 Hz, 3JH6,H7 =7.6 Hz, 5JH3,H6 =0.7 Hz, H-6), 7.24 (m, 3JH6,H7 =7.6 Hz,
4
4JH3,H7 =1.4, 4JH5,H7 =1.0 Hz, H-7), 7.14 (m, 4JH3,H5 =2.6 Hz, JH3,H7
=
1.4 Hz, 5JH3,H6 =0.7 Hz, H-3), 7.14 (m, 3JH5,H6 =8.4 Hz, 4JH3,H5 =2.6 Hz,
4JH3,H7 =1.0 Hz, H-5), 3.83 ppm (s, 3H, H-9, CH3); 13C NMR: d=159.43
(C-4), 129.86 (C-6), 124.22 (C-7), 119.33 (C-5), 118.71 (C-1), 116.37 (C-3),
111.98 (C-2), 55.32 ppm (C-8); GC-MS (EI): rt=13.27 min, m/z: 133;
Rf =0.33 (hexanes/AcOEt, 90:10).
Aryl nitrile (9i): The cyanation of 3,5-difluorobromobenzene (576 mL,
5.0 mmol) was achieved by following the general procedure. The result-
ing crude oily residue was purified by chromatography on silica gel
(eluent: hexanes/ethyl acetate, 90:10) to provide 653 mg (94% yield) of
the desired product as white needles. M.p. 85–868C (hexanes/AcOEt)
[lit.:[45] 83–858C (hexanes)]; 1H NMR:[45] d=7.23 (m, 4JH3,H7 =8.5 Hz,
3JH,F =7.7 Hz, 4JH5,H3 =4JH5,H7 =2.3 Hz, 5JH,F =1.1 Hz, H-3, H-7), 7.12 ppm
(m, 3JH5,F4 =3JH5,F6 =8.7 Hz, 4JH5,H3 =4JH5,H7 =2.3 Hz, H-5); 13C NMR:[45]
4
d=162.83 (dd, 1JC,F =252.2 Hz, 3JC,F =12.6 Hz, C-4, C-6), 116.45 (t, JC,F
=
4
3.4 Hz, C-1), 115.61 (m, 2JC,F =20.3 Hz, JC,F =8.6 Hz, C-3, C-7), 114.30 (t,
2
3JC,F =11.6 Hz, C-2), 109.39 ppm (t, JC,F =24.9 Hz, C-5); 19F NMR:[45] d=
105.69 ppm (m, 3JH,F =8.7 Hz, 3JH,F =7.7 Hz, 4JF,F =1.3 Hz , 5JH,F =1.1 Hz,
F-4, F-6); GC-MS (EI): rt=7.16 min, m/z: 139; Rf =0.41 (hexanes/
AcOEt, 90:10).
Aryl nitrile (9k): The cyanation of 4-chlorobromobenzene (957.3 mg,
5.0 mmol) was achieved by following the general procedure. The result-
ing crude oily residue was purified by chromatography on silica gel
(eluent: hexanes/ethyl acetate, 90:10) to provide 588 mg (86% yield) of
the desired product as white needles. M.p. 91–928C (hexanes/AcOEt)
1
(lit.:[46] 948C); H NMR: d=7.64 (m, H-3, H-7), 7.50 ppm (m, H-4, H-6);
13C NMR: d=139.39 (C-5), 133.26 (C-3, C-7), 129.56 (C-4, C-6), 117.82
(C-1), 110.66 ppm (C-2); GC-MS (EI): rt=12.06 min, m/z: 137; Rf =0.43
(hexanes/AcOEt, 90:10).
Aryl nitrile (9j): After standard cycles of evacuation and back-filling with
dry and pure nitrogen, an oven-dried Radley tube (Carousel “reaction
stations RR98030”) equipped with a magnetic stirring bar was charged
with CuI (95.2 mg, 0.5 mmol) and 1,10-phenanthroline (181.2 mg,
1.0 mmol). The tube was evacuated and back-filled with nitrogen. 2-Iodo-
toluene (636 mL, 5.0 mmol), acetone cyanohydrin (502 mL, 5.5 mmol) and
diisopropylethylamine (1.05 mL, 6.0 mmol) were added under a stream
of nitrogen by syringe at room temperature, followed by anhydrous and
degassed DMF (3 mL). The tube was sealed under a positive pressure of
nitrogen, and stirred and heated at 1108C for 48 h. After cooling to room
temperature, the mixture was diluted with diethyl ether (~50 mL) and fil-
tered through a plug of Celite, the filter cake being further washed with
[13] a) D. M. Tschaen, R. Desmond, A. O. King, M. C. Fortin, B. Pipik,
S. King, T. R. Verhoeven, Synth. Commun. 1994, 24, 887–890; b) T.
Chem. Eur. J. 2005, 11, 2483 – 2492
www.chemeurj.org
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2491

H.-J. Cristau, M. Taillefer et al.
Okano, J. Kiji, Y. Toyooka, Chem. Lett. 1998, 425–426; c) P. E. Mali-
gres, M. S. Waters, F. Fleitz, D. Askin, Tetrahedron Lett. 1999, 40,
8193–8195; d) F. Jin, P. N. Confalone, Tetrahedron Lett. 2000, 41,
3271–3273; e) M. Alterman, A. Hallberg, J. Org. Chem. 2000, 65,
7984–7989; f) A. W. Shaw, J. deSolms, Tetrahedron Lett. 2001, 42,
7173–7176; g) J. Ramnauth, N. Bhardwaj, P. Renton, S. Rakhit, S. P.
Maddaford, Synlett 2003, 14, 2237–2239; h) R. Chidambaran, Tetra-
hedron Lett. 2004, 45, 1441–1444.
A. Ouali, WO 03/101966, 2003; e) M. Taillefer, H.-J. Cristau, P. P.
Cellier, J.-F. Spindler, Patent priority number FR 03 10253, 2004.
[29] We developed this strategy more than two years ago and published
the results as a patent (see ref. [28d]). A copper-catalyzed halide ex-
change-cyanation of aryl bromides has been described but this pro-
cedure involved stoichiometric amounts of ligand with respect to the
aryl bromide (see ref. [24]).
[30] Such a thermodynamic equilibrium has been reported in the litera-
ture: A. Klapars, S. L. Buchwald, J. Am. Chem. Soc. 2002, 124,
14844–14845.
[31] R. J. H. Gregory, Chem. Rev. 1999, 99, 3649–3682.
[32] F. A. Cotton, G. Wilkinson, Advanced Inorganic Chemistry, 5th ed.,
Wiley, New York, 1988, 755–775.
[33] R. K. Dieter, L. A. Silks III, J. R. Fishpaugh, M. E. Kastner, J. Am.
Chem. Soc. 1985, 107, 4679–4692.
[34] C. C. Hach, C. V. Banks, H. Dielh, Org. Synth., 1963, Coll. Vol IV,
230–232.
[35] C. Bischoff, R. Ohme, J. Prakt. Chem. 1973, 315, 505–509.
[36] O. Exner, J. M. Kliegman, J. Org. Chem. 1971, 36, 2014–2015.
[37] P. Grammaticakis, C. R. Acad. Sci., Paris, C. R. Seances Acad. Sci.,
Ser. C 1970, 270, 76–79.
[38] D. D.Perrin„ W. L. F. Armarego, D. R. Perrin, Purification of Labo-
ratory Chemicals, 3rd ed., Pergamon Press, New York, 1985.
[39] J. Berstein, H. L. Yale, K. Losee, M. Holsing, J. Martins, W. A. Lott,
J. Am. Chem. Soc. 1951, 73, 906–912.
[40] P. Dastidar, T. N. G. Row, B. R. Prasad, C. K. Subramanian, S. Bhat-
tacharya, J. Chem. Soc., Perkin Trans. 2 1993, 2409–2411.
[41] R. G. Wallace, J. M. Barker, L. Michael, Synthesis 1990, 1143–1144.
[42] B. P. Bandgar, S. R. Jagtap, S. B. Ghodeshwar, P. P. Wadgaonkar,
Synth. Commun. 1995,25, 2993–2998.
[14] a) N. Chatani, T. Hanafusa, J. Org. Chem. 1986, 51, 4714–4716;
b) M. Sundermeier, S. Mutyala, A. Zapf, A. Spannenberg, M.
Beller, J. Organomet. Chem. 2003, 684, 50–55.
[15] C. Yang, M. Williams, Org. Lett. 2004, 6, 2837–2840.
[16] B. Jiang, Y. Kan, A. Zhang, Tetrahedron 2001, 57, 1581–1584.
[17] T. Sakamoto, K. Ohsawa, J. Chem. Soc., Perkin Trans. 1 1999, 2323–
2326.
[18] T. Schareina, A. Zapf, M. Beller, Chem. Commun. 2004, 1388–1389.
[19] M. Sundermeier, A. Zapf, M. Beller, Angew. Chem. 2003, 115,
1700–1703; Angew. Chem. Int. Ed. 2003, 42, 1661–1664.
[20] a) A. Pongratz, Monatsh. Chem. 1927, 48, 585–591; b) J. Lindley,
Tetrahedron 1984, 40, 1433–1456.
[21] J. X. Wu, B. Beck, R. X. Ren, Tetrahedron Lett. 2002, 43, 387–389.
[22] R. K. Arvela, N. E. Leadbeater, H. M. Torenius, H. Tye, Org.
Biomol. Chem. 2003, 1, 1119–1121.
[23] M. Ghaffarzadeh, M. Bolourtchian, M. Reza Halvagar, M. Hosseini,
J. Chem. Res. 2003, 814–815.
[24] J. Zanon, A. Klapars, S. L. Buchwald, J. Am. Chem. Soc. 2003, 125,
2890–2891.
[25] H.-J. Cristau, P. P. Cellier, J.-F. Spindler, M. Taillefer, Eur. J. Org.
Chem. 2004, 695–709.
[26] H.-J. Cristau, P. P. Cellier, J.-F. Spindler, M. Taillefer, Org. Lett.
2004, 6, 913–916.
[43] J. C. Feng, B. Liu, L. Dai, N. S. Bian, Synth. Commun. 1998, 28,
3765–3768.
[44] C. S. Marvel, C. G. Overbergen, J. Am. Chem. Soc. 1945, 67, 2250–
2252.
[27] H.-J. Cristau, P. P. Cellier, J.-F. Spindler, M. Taillefer, Chem. Eur. J.
2004, 10, 5607–5622.
[28] Note that these results, including those presented in this article,
have already been published as patents: Rights to the commercial
use of the catalytic systems presented in this work are held by
Rhodia Recherches under the following patents: a) M. Taillefer, H.-
J. Cristau, P. P. Cellier, J.-F. Spindler, FR 2833947, 2003; [Chem.
Abstr. 2003, 139, 69290]; b) M. Taillefer, H.-J. Cristau, P. P. Cellier,
J.-F. Spindler, WO 03/050885, 2003; c) M. Taillefer, H.-J. Cristau,
P. P. Cellier, J.-F. Spindler, FR 2840303, 2003; [Chem. Abstr. 2004,
140, 16744]; d) M. Taillefer, H.-J. Cristau, P. P. Cellier, J.-F. Spindler,
[45] The spectra of compounds 9h and 9i are not first-order: “m” has
been used for every non first-order signals. The chemical shifts and
coupling constants are obtained from the exact analysis of the spin
system with gnmr, (Cherwell Scientific Publishing Ltd, Oxford,
1997).
[46] S. S. Elmorsy, A. A. S. El Ahi, H. Soliman, F. A. Amer, Tetrahedron
Lett. 1995, 36, 2639–2640.
Received: September 24, 2004
Published online: February 15, 2005
2492
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2483 – 2492