New progress in the cobalt-catalysed synthesis of aromatic organozinc compounds by reduction of aromatic halides by zinc dust

Article abstract of DOI:10.1016/S0040-4039(03)01595-8

The experimental conditions of the cobalt-catalysed synthesis of aromatic organozinc species from the corresponding bromides have been optimised. The preparation of numerous aromatic organozinc reagents under these optimised conditions is reported, with y

Full text of DOI:10.1016/S0040-4039(03)01595-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 6417–6420
New progress in the cobalt-catalysed synthesis of aromatic
organozinc compounds by reduction of aromatic halides
by zinc dust
a
a,
b
a
Igor Kazmierski, Corinne Gosmini, * Jean-Marc Paris and Jacques P e´ richon
a
Laboratoire d’Electrochimie, Catalyse et Synth e` se Organique, UMR 7582, Universit e´ Paris XII-CNRS, 2, Rue Henri Dunant,
4320 Thiais, France
9
b
Rhodia, Research Center of Lyon, 85, rue des Fr e` res Perret, BP 62, 69192 Saint-Fons Cedex, France
Received 27 May 2003; revised 26 June 2003; accepted 26 June 2003
Abstract—The experimental conditions of the cobalt-catalysed synthesis of aromatic organozinc species from the corresponding
bromides have been optimised. The preparation of numerous aromatic organozinc reagents under these optimised conditions is
reported, with yields between 67 and 99%.
©
2003 Elsevier Ltd. All rights reserved.
7
Organozinc species, especially in the aromatic series,
are useful reagents for the synthesis of polyfunctional
compounds when used in conjunction with the proper
transition-metal catalyst, because of their broad spec-
in DMF–pyridine or acetonitrile–pyridine mixtures.
Moreover, a process using pure acetonitrile as solvent
and cobalt(II) bromide as catalyst has been developed,
permitting the efficient electrosynthesis of organozinc
1
8
trum of tolerated functional groups. However, their
species from aryl bromides.
preparation from aromatic halides requires mild meth-
ods when the aromatic ring bears sensitive functions,
among which nitriles, ketones or esters. These methods
are based on the use of an activated zinc as the reduc-
ing agent. Whereas aryl iodides can readily react with
These last electrochemical processes have enabled the
discovery of an easy, reproducible chemical synthesis of
aromatic organozinc compounds from functionalised
9
aryl bromides or iodides (ArꢀX). These aromatic
2
3
zinc dust in polar or ethereal solvents, the conversion
of the less reactive aryl bromides to their organozinc
counterparts must be realised with a more active zinc,
prepared by the reduction of zinc(II) halides by naph-
halides are reduced in pure acetonitrile by zinc dust
activated by a small quantity of trifluoroacetic acid, in
the presence of a cobalt(II) halide as catalyst. In the
same study, it has been evidenced that proportions of
by-products ArꢀH and ArꢀAr can be significatively
reduced when, in a ‘pre-treatment’ preliminary step, a
catalytic quantity of bromobenzene as an additive
reacts with the activated zinc dust and the cobalt(II)
catalyst (Scheme 1).
4
talene/lithium (the Rieke zinc ). Still another method
has recently been described by Knochel and co-work-
5
ers: the aromatic organozinc species is prepared by a
metal–halogen exchange between an aryl bromide and
isopropylmagnesium chloride, subsequently forming the
organozinc species by transmetallation with a zinc(II)
halide, at low temperature.
The current work is devoted to the optimisation of this
reaction, including investigation of the effect of other
halides than bromobenzene in the preliminary step in
Research in our laboratory has focused in the past few
years on electrochemical methods of synthesis of
organozinc species from aromatic bromides or chlo-
rides, catalysed either by nickel complexes in dimethyl-
6
formamide, or less harmful and cheaper cobalt halides
Scheme 1. Chemical synthesis of functionalised organozinc
*
Corresponding
author.
Fax:
+33-(0)1-4978-1148;
e-mail:
gosmini@glvt-cnrs.fr
species (Ref. 9).
0
040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)01595-8

6
418
I. Kazmierski et al. / Tetrahedron Letters 44 (2003) 6417–6420
order to further reduce the proportion of by-products,
concentration of catalyst, quantity of zinc dust and the
effect of the presence of zinc(II) halide.
is also possible to increase the proportion of AllylCl
(entry 4) and thus to further increase the yield of the
reaction, diminishing the total of by-products to less
than ten percent of the final yield.
Another worry was the addition of zinc(II) bromide,
which is a very hygroscopic salt and can therefore
increase the ArH yield. However, addition of this salt
Results and discussion
9
The reactions have been carried out using the proce-
was found necessary to the reaction. As it has already
9
dure described earlier and p-bromoanisole as test-com-
been pointed out, the organic halide used as the addi-
tive presumably reacts in our medium following the
same sequence as the aryl bromide. The probable reac-
pound. The results of the optimisation of the reaction
are reported in Table 1.
9
tion mechanism, as it has been described earlier, is
The effect of the additive has been checked out by
comparing preliminary steps with (entry 2) and without
PhBr (entry 1). Indeed, using an additive decreases the
quantity of the reduction by-product ArH, PhBr being9
converted to PhH. However, as previously reported,
with a higher proportion of PhBr the reaction tends to
slow down and a significative amount of PhZnBr is
formed. It thus occured to us that PhBr was not
reactive enough to act as an additive, and that the
by-products of the preliminary step could disturb fur-
ther coupling reactions. Other additives, such as benzyl
chloride, a-chloroacetyl propionate and allyl bromide
or chloride have consequently been tried. Among these,
allyl chloride has proved to be the most interesting in
terms of reactivity and by-products, being fully con-
verted into propene, 1,5-hexadiene and the addition
shown in Scheme 2.
Zn(II) salts are formed in situ by the oxydation of zinc
dust, and are afterwards engaged in the transmetalla-
tion reaction with the organocobalt(II) species. In the
case of allyl chloride, the allylzinc species readily reacts
with the solvent, thus liberating zinc(II) ions. Our
hypothesis therefore was that with 0.3 equiv. of AllylCl,
no further addition of ZnBr is necessary. This was
verified (entry 5) as results are very similar whether
2
ZnBr is used or not.
2
We then wondered if a preliminary step of 30 min was
needed when using AllylCl as the additive, as it is more
reactive than PhBr. After checking that the one-step
reaction (entry 6) results in a slightly worse yield and
more by-products, different lengths were tried (entries 7
and 8). It appeared that a 5 min preliminary step (entry
7) was sufficient in order to obtain high yields and
reasonable by-product proportions. The next step was
to try and reduce the proportion of the cobalt catalyst,
1
0
product of allylzinc species on acetonitrile, all of those
compounds being easily separated from aromatic or
heteroaromatic coupling products. The result is that
organozinc synthesis using AllylCl (entry 3) proceeds
faster and with a slightly better yield than with PhBr. It
Table 1. Optimisation of the reaction with 4-bromoanisole (GC yields)
Entry
Additive
equiv.)
Prel. step
(min)
Zn
(equiv.)
CoBr2
(equiv.)
ZnBr2
(equiv.)
Reaction
time (min) (%)
ArZnBr
ArH
(%)
ArAr
(%)
Conv.
(%)
a
(
1
2
3
–
30
3
3
3
0.1
0.1
0.1
0.1
0.1
0.1
15
20
15
73
78
83
26
16
9
1
5
8
100
100
100
PhBr 0.15 30
AllylCl 30
.15
0
4
5
6
7
8
9
AllylCl 0.3 30
AllylCl 0.3 30
3
3
3
3
3
3
3
0.1
0.1
0.1
0.1
0.1
0.05
0.05
0.1
–
–
–
–
20
25
10
14
25
4
91
91
86
89
88
83
89
7
6
4
6
8
2
3
10
6
4
5
100
100
100
100
100
100
99
AllylCl 0.3
AllylCl 0.3
0
5
AllylCl 0.3 20
AllylCl 0.3
AllylCl
5
5
–
–
12
7
1
1
1
0
25
3
0
.15
AllylCl
.15
AllylCl
.15
1
5
5
1.5
1.5
0.05
0.05
–
–
60
40
83
89
13
5
4
6
100
100
0
2b
0
a
The obtained arylzinc halides are converted into the corresponding aryl iodide by addition of iodine.
Conc. of ArBr=1 M.
b
Scheme 2. Supposed reaction sequence of the synthesis of organozinc species from aryl bromides (Ref. 9).

I. Kazmierski et al. / Tetrahedron Letters 44 (2003) 6417–6420
6419
these conditions, both the rate of the reaction and the
yield in ArZnBr were satisfying.
The reaction conditions of entry 12 were consequently
applied to various aromatic bromides bearing a wide
array of functional groups (Scheme 3). Results are
reported in Table 2.
Scheme 3. Chemical synthesis of functionalised organozinc
species with optimised conditions.
It is of interest to notice that in the para series of
aromatic compounds (entries 13–19), no clear differ-
ence in kinetics or yield (86–94%) are observed except
for 4%-bromoacetophenone (entry 19), which reacts
faster and gives rise to more dimeric product. The
synthesis of organozinc species can also be achieved in
the meta and ortho series, as the two examples of
but when using 5% of CoBr instead of 10% (entry 9),
2
the reaction was considerably slowed down. Conse-
quently, the proportion of AllylCl was adapted to that
of the cobalt catalyst by bringing it to 3 equiv. versus
CoBr2 (entry 10), and the reaction then proceeded
smoothly, and surprisingly nearly as fast as with 10%
CoBr2.
3
-bromobenzonitrile (entry 20) and 2-bromobenzoni-
trile (entry 21) show. This last aryl bromide reacts
faster than its meta and para counterparts. With a
poorly reactive heteroaromatic bromide such as 3-bro-
mothiophene (entry 22), however, the starting product
could not be entirely converted into the corresponding
organozinc reagent. Strangely enough, carrying the
reaction with a greater proportion of catalyst (10%
instead of 5%) with the appropriate changes in AllylCl
and zinc dust quantities resulted in a faster 3-bromo-
thiophene conversion but availed no change in yield.
The last step was to diminish the proportion of zinc
dust, as it was used in large excess versus aryl bromide.
It was also decided that this proportion would be based
upon the total quantity of halogenated compounds
(
aryl bromide and allyl chloride), as it would permit
further modification of the proportion of CoBr and/or
2
AllylCl while always keeping the same amount of zinc
dust for the formation of the organozinc species. After
a few attempts, it was found that the minimal propor-
tion of zinc that can be used is 1.3 equiv. versus the
total quantity of halogenated compounds, which
amounts, in the case of entry 11, to 1.5 equiv. versus
aryl bromide. It is supposed that the 0.3 overstoichio-
metric equivalents of zinc dust are the part which
cannot be activated by our method. The reaction still
proceeded with a good yield, but slower than in the
case of 3 equiv. of zinc dust. This prompted us to try
and work at a somewhat higher concentration of aryl
bromide, switching from 0.75 to 1 M (entry 12). In
Other compounds of a special interest to us are the
diorganozinc species formed by the reaction of dibro-
moaromatic compounds. These syntheses have already
11
been achieved by electrochemical methods as well as
9
with the initial reaction conditions. It has thus been
tried to apply the optimised conditions from entry 12 to
the synthesis of the diorganozinc compound of 1,4-
dibromobenzene (Table 3).
Table 2. Synthesis of mono organozinc species from aryl bromides and 3-bromothiophene using optimised conditions (GC
yields)
Entry
ArBr
Reaction time (min)
ArZnBr (%)a
ArH (%)
ArAr (%)
Conv. (%)
13
14
15
16
17
18
19
20
21
22
Ph-Br
40
40
60
60
40
60
15
60
30
60
86
89
94
86
94
91
69
96
99
67
10
5
4
5
0
2
5
2
0
6
6
2
6
6
2
26
1
1
5
100
100
100
97
100
95
100
99
100
77
p-MeOPhBr
p-FPhBr
p-NCPhBr
p-EtO CPhBr
2
p-CF PhBr
3
p-MeCOPhBr
m-NCPhBr
o-NCPhBr
3-Bromothiophene
5
a
The obtained arylzinc halides are converted into the corresponding aryl iodide by addition of iodine.
Table 3. Synthesis of diorganozinc species from 1,4-dibromobenzene (GC yields)
Entry
Zn
Reaction time
(h)
BrZnPhZnBr
(%)
BrPhZnBr
(%)
PhZnBr
(%)
BrZnPh–PhZnBr
(%)
Conv.
(%)
a
a
a
a
(equiv.)
2
2
3
4
1.5
2.8
3b
3
30
78
23
6
4
6
5
5
66
99
a
The obtained arylzinc halides are converted into the corresponding aryl iodide by addition of iodine.
No conversion beyond 1 h.
b

6
420
I. Kazmierski et al. / Tetrahedron Letters 44 (2003) 6417–6420
References
1. (a) Knochel, P.; Singer, R. D. Chem. Rev. 1993, 93,
2117–2188; (b) Knochel, P.; Almena Prea, J. J.; Jones, P.
Tetrahedron 1998, 54, 8275–8319.
2
3
4
. Majid, T. N.; Knochel, P. Tetrahedron Lett. 1990, 31,
4413–4416.
Scheme 4. Cross-coupling reactions of organozinc bromides
with 4-iodoanisole.
. Ikegami, R.; Koresawa, A.; Shibata, T.; Takagi, K. J.
Org. Chem. 2003, 68, 2195–2199.
. Zhu, L.; Wehmeyer, R. M.; Rieke, R. D. J. Org. Chem.
1994, 56, 1445–1453.
With 1.5 equiv. of zinc (entry 23), only 66% of the
original dibromide is converted, and the diorganozinc
species BrZnArZnBr is formed along with the mono
organozinc compound BrArZnBr. The only by-prod-
ucts of a significant amount are the dimeric diorgano-
zinc species BrZnAr–ArZnBr and the semi-reduced
ArZnBr, whose formation had already been reported
earlier. As it was clear that there was not enough zinc
dust to form predominantly the diorganozinc com-
pound, another experiment was carried out with addi-
tional 1.3 equiv. zinc dust versus aryl bromide (entry
5. Boudier, A.; Bromm, L. O.; Lotz, M.; Knochel, P.
Angew. Chem., Int. Ed. 2000, 39, 4414–4435.
6. Sibille, S.; Ratovelomanana, V.; P e´ richon, J. J. Chem.
Soc., Chem. Commun. 1992, 283–284.
7. Gosmini, C.; Rollin, Y.; N e´ d e´ lec, J. Y.; P e´ richon, J. J.
Org. Chem. 2000, 65, 6024–6026.
8. Fillon, H.; Le Gall, E.; Gosmini, C.; P e´ richon, J. Tetra-
hedron Lett. 2002, 43, 5941–5944.
9
. (a) Fillon, H.; Gosmini, C.; P e´ richon, J. J. Am Chem.
Soc. 2003, 125, 3867–3870; (b) Fillon, H.; Gosmini, C.;
P e´ richon, J. Patent Application 01/08880, France, July 4,
2
4), the other parameters remaining unchanged. In this
case, the starting molecule is fully converted, yielding
8% of the expected diorganozinc compound in 3 h.
7
2001.
1
0. This property of allylzinc species has been reported ear-
As an application of the synthesis of aromatic organo-
zinc species, we also chose to realise the palladium-
catalysed cross-coupling reactions of the organozinc
species obtained from 3- and 4-bromobenzonitrile
lier. See: (a) Blaise, E. E. C. R. Acad. Sci. 1904, 138,
284–286; (b) Rollin, Y.; Derien, S.; Dunach, E.; Gebe-
henne, C.; P e´ richon, J. Tetrahedron 1993, 49, 7723–7732.
1. Fillon, H.; Gosmini, C.; Nedelec, J. Y.; P e´ richon, J.
Tetrahedron Lett. 2001, 42, 3843–3846.
1
(
entries 16 and 20) with 4-iodoanisole (1 equiv. versus
starting aryl bromide), in the optimised conditions of
entry 12 (Scheme 4). The corresponding biaryls, 4%-
methoxy-3-cyanobiphenyl
1
2
12. General procedure for the synthesis of aromatic organo-
zinc species (entry 20) and their palladium-catalysed
cross-coupling with 4-iodoanisole: (a) Preliminary step:
1
3
and 4%-methoxy-4-cyano-
1
3
biphenyl were isolated in 70 and 64% yield, respec-
tively.
zinc dust <10 mm (1.47 g; 1.49 equiv.) and CoBr (164
2
mg, 0.05 equiv.) are introduced under argon in a reaction
vessel. 15 mL Of acetonitrile, allyl chloride (0.185 mL;
0.15 equiv.), dodecane (0.2 mL; reference for GC) and 50
mL trifluoroacetic acid are added and the reaction is
stirred for 5 min. (b) Synthesis of the organozinc reagent:
3-bromobenzonitrile (2.73 g; 15 mmol) is added to the
reaction, which is carried over a 1 hour period, until
complete conversion of the starting aryl bromide. (c)
Cross-coupling: 4-iodoanisole (3.51 g; 1 equiv.) and
PdCl (PPh ) (105 mg; 0.01 equiv.) are introduced in the
reaction vessel, and the reaction is stirred until complete
conversion of the organozinc reagent (1 h). The reaction
is then quenched with 40 mL HCl 1.5 M and extracted
with 3×40 mL ether. The crude product is purified by
flash chromatography on silica gel, eluted with pentane/
ether 97/3 to 96/4, yielding 2.20 g (70%) of 4%-methoxy-3-
cyanobiphenyl (identified by GC/MS and NMR by
comparison to Ref. 13).
The conclusion of this work is that aromatic bromides
can be efficiently converted into the corresponding
organozinc species in acetonitrile using cobalt(II) bro-
mide and zinc dust, under more inexpensive conditions
than those described in a previous work. It has been
proved that the use of a slight excess of zinc dust and a
small quantity of catalyst, along with a short prelimi-
nary step requiring allyl chloride as an additive and a
somewhat higher concentration, can lead to good to
excellent yields in organozinc species. Furthermore,
these functionalised organometallics can be readily cou-
pled with aryl iodides using a palladium(II) catalyst.
New applications of this synthetic process are being
currently developed and will be reported in further
works.
2
3 2
Acknowledgements
1
3. GC/MS and NMR analyses were compared to those
found in the following reference: Hassan, J.; Hathroubi,
C.; Gozzi, C.; Lemaire, M. Tetrahedron 2001, 57, 7845–
7855.
We gratefully acknowledge the financial support pro-
vided by Rhodia. I. Kazmierski thanks Rhodia for a
scholarship.