Palladium-catalyzed cyanation of aryl bromides promoted by low-level organotin compounds

Article abstract of DOI:10.1021/ol049621d

(Equation Presented) A novel method for palladium-catalyzed cyanation of aryl bromides promoted by low-level tri-n-butyltin chloride or cyanide is described. The method features low catalyst loading and mild reaction conditions. KCN is used as the cyanide source. Only trace levels of the tri-n-butyltin compound are required to achieve high conversion and yield in the cyanation of aryl bromides, iodides, and triflates.

Full text of DOI:10.1021/ol049621d

ORGANIC
LETTERS
2
004
Vol. 6, No. 17
837-2840
Palladium-Catalyzed Cyanation of Aryl
Bromides Promoted by Low-Level
Organotin Compounds
2
Chunhua Yang* and J. Michael Williams
Department of Process Research, Merck Research Laboratories, Merck & Co., Inc.,
P.O. Box 2000, Rahway, New Jersey 07065
chunhua_yang@merck.com
Received March 1, 2004
ABSTRACT
A novel method for palladium-catalyzed cyanation of aryl bromides promoted by low-level tri-n-butyltin chloride or cyanide is described. The
method features low catalyst loading and mild reaction conditions. KCN is used as the cyanide source. Only trace levels of the tri-n-butyltin
compound are required to achieve high conversion and yield in the cyanation of aryl bromides, iodides, and triflates.
Cyanation of aryl halides is a common and useful transfor-
mation in organic synthesis. Not only are products containing
the nitrile group biologically important, but the nitrile is
medium. In this way, the oxidative insertion is fast relative
to the reaction leading to an inactive form of Pd. The
difference in rate is a consequence of the high aryl halide
concentration and low cyanide salt concentration. For
1
also valuable in installation of functionalities such as
aldehydes, amines, amidines, tetrazoles, acids, and acid
4
i,j,l,m
5
example, Zn(CN)2
and CuCN, which have limited
2
derivatives. For many years, the only method for cyanation
solubility in the organic solvents used for cyanation, have
demonstrated utility in this regard.
of an aryl halide required stoichiometric CuCN and harsh
3
conditions. Since the discovery that Pd catalyzes the
While these methods represented significant improvements
over earlier procedures, substantial Zn- or Cu-containing
waste is produced. As an alternative to transition metal
conversion of an aryl halide to a nitrile, methods have been
developed that offer milder reaction conditions resulting in
fewer side-products.4 It was recognized that cyanide
interferes with the prerequisite reaction of Pd(0) with the
aryl halide. To make the reactions catalytic in palladium, it
was essential to devise a means of delivering cyanide for
the reaction with Pd(II) following oxidative insertion while
minimizing the exposure of Pd(0) to cyanide. One approach
makes use of cyanide salts with poor solubility in the reaction
-7
(4) (a) Tagagi, K.; Okamoto, T.; Sakakibara, Y.; Oka, A. Chem. Lett.
1973, 471. (b) Sekiya, A.; Ishikawa, N. Chem. Lett. 1975, 277. (c) Dalton,
J.; Regen, S. J. Org. Chem. 1979, 44, 4443. (d) Kosugi, M.; Kato, Y.;
Kiuchi, K.; Migita, T. Chem. Lett. 1981, 69. (e) Akita, Y.; Shimazaki, M.;
Ohta, A. Synthesis 1981, 974. (f) Chatani, N.; Hanafusa, T. J. Org. Chem.
1986, 51, 4714. (g) Takagi, K.; Sasaki, K.; Sakakibara, Y. Bull. Chem.
Soc. Jpn. 1991, 64, 1118. (h) Andersson, Y.; Langstrom, B. J. Chem. Soc.,
Perkin Trans. 1 1994, 1395. (i) Tschaen, D. M.; Desmond, R.; King, A.
O.; Fortin, M. C.; Pipik, B.; King, S.; Verhoeven, T. R. Synth. Commun.
1994, 24, 887. (j) Okano, T.; Kiji, J.; Toyooka, Y. Chem. Lett. 1998, 425.
(k) Tsuji, Y.; Kusui, T.; Kojima, T. Sugiura, Y.; Yamada, N.; Tanaka, S.;
Ebihara, M.; Kawmur, T. Organometallics 1998, 17, 4835. (l) Maligres, P.
E.; Waters, M. S.; Fleitz, F.; Askin, D. Tetrahedron Lett. 1999, 40, 8193.
(m) Jin, F.; Confalone, P. N. Tetrahedron Lett. 2000, 41, 3271. (n) Jiang,
B.; Kan Y.; Zhang, A. Tetrahedron 2001, 57, 1581.
(1) (a) Uehling, D. E.; Nanthakumar, S. S.; Croom, D.; Emerson, D. L.;
Leitner, P. P.; Luzzio, M. J.; McIntyre, G.; Morton, B.; Profeta, S.; Sisco,
J.; Sternbach, D. D.; Tong, W.-Q.; Vuong, A.; Besterman, J. M. J. Med.
Chem. 1995, 38, 1106. (b) Nagamura, S.; Kobayashi, E.; Gomi, K.; Saito,
H. Bioorg. Med. Chem. 1996, 4, 1379. (c) Kleemann, A.; Engel, J.; Kutscher,
B.; Reichert, D. Pharmaceutical Substance: Synthesis, Patents, Applications,
4
th ed.; Georg Thieme: Stuttgart, 2001.
2) (a) Rappoport, Z. The Chemistry of the Cyano Group; Interscience
(5) (a) Anderson, B. A.; Bell, E. C.; Ginah, F. O.; Harn, N. K.; Pagh, L.
M.; Wepsiec, J. P. J. Org. Chem. 1998, 63, 8224. (b) Allentoff, A. J.;
Markus, B.; Duelfer, T.; Wu, A.; Jones, L.; Ciszewska, G.; Ray, T. J.
Labelled Compd. Radiopharm. 2000, 43, 1075. (c) Sakamono T.; Ohsawa,
K. J. Chem. Soc., Perkin Trans. 1 1999, 2323.
(
Publishers: London, 1970. (b) Larock, R. C. ComprehensiVe Organic
Transformations; Wiley-VCH: New York, 1988.
(3) Ellis, G. A.; Romney-Alexander, T. M. Chem. ReV. 1987, 87, 779.
1
0.1021/ol049621d CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/20/2004

cyanides, it was reported that KCN can be used if 0.2 equiv
of TMEDA is added. Evidence was presented indicating
compound needed. Using KCN as the stoichiometric cyanide
source, we could control the level of Bu SnCN in the reaction
6
3
that the diamine prevents deactivation of the Pd catalyst.
Recently, acetone cyanohydrin was used as a cyanide source
along with TMEDA; the concentration of cyanide in the
reaction was kept low by slow addition of the cyanohydrin
by charging a relatively small amount of a tri-n-butyltin
halide. In the cyanation reaction of an aryl bromide,
Bu
3
SnBr would be generated, which would then react with
SnCN to the catalytic cycle. Direct
solid KCN to return Bu
3
7
using a syringe pump. There remain opportunities for
inhibition of the catalyst by KCN would be minimized by
using a solvent that does not appreciably dissolve the salt.
Our thinking is illustrated in Scheme 1.
devising milder and more practical methods for general use
from bench to manufacturing scales. Here we report a unique
Pd-catalyzed cyanation for aryl and heteroaryl bromides
promoted by Bu
Early attempts to use Me
cyanation of aryl halides, in analogy to the Stille coupling,
3
SnCl and using KCN as a cyanide source.
3
SnCN for palladium-catalyzed
Scheme 1. Proposal for Pd-Catalyzed Cyanation of Aryl
Halides
were not successful.1
Pd is required. The catalyzed reaction fails as a result of the
rapid reaction of Pd(0) with Me SnCN terminating the
catalytic cycle. Reaction of Pd(0) with Me SnCN is much
faster than insertion into the C-I bond of an ArI. Our own
observations using Bu SnCN are consistent with the results
b,4c,8
It was found that stoichiometric
3
3
4c
3
reported. When we attempted cyanation of 1-bromonaph-
thalene using 10 mol % tris(dibenzylideneacetone)dipalladium-
(
0) [Pd
of the desired product was formed. If, however, we treated
-bromonaphthalene with 1 equiv of Pd (dba) with t-Bu
and then added 1 equiv of Bu SnCN, cyanation was
completed in less than 10 min at 80 °C. Clearly, Bu SnCN
2 3 3 3
(dba) ] with t-Bu P and 1 equiv of Bu SnCN, none
1
2
3
3
P
3
3
With 0.5 mol % Pd
2 3 3
(dba) with (t-Bu P), 0.5 mol %
interferes with the requisite oxidative insertion but does
transfer cyanide to Pd(II) with subsequent reductive elimina-
tion to give the nitrile.
It had been proposed that the relative rates in the reaction
3
of Pd(0) with ArI and Me SnCN could be made favorable
Bu SnCl or Bu SnCN, and 1.5 equiv of KCN in acetonitrile
3
3
at 80 °C, 1-bromonaphthalene was efficiently converted to
the corresponding nitrile. A series of experiments were
performed to determine the optimum charges. The ratio of
Bu
conversion. It was found that very little Bu
to improve both the rate and the conversion. The optimum
molar ratio of Bu SnCl to Pd (dba) was about 1:3.7; at
3
SnCl to Pd catalyst had a dramatic effect on reaction
for the desired reaction if the concentration of the tin cyanide
were sufficiently low throughout the reaction. It was reported
3
SnCl is required
that slow addition of Me
afforded a modest yield (40%) in the cyanation of an aryl
iodide. We found that slow addition of Bu SnCN to a
reaction containing 0.25 mol % Pd (dba) with t-Bu P as a
3
SnCN using a syringe pump
3
2
3
higher and lower ratios, the reaction was slower and did not
go to completion.
Under optimal conditions (0.25 mol % Pd
3
2
3
3
2 3
(dba) , 1.25 mol
ligand gave 12% conversion of 1-bromonaphthalene to the
nitrile.
%
t-Bu P, and 0.07 mol % Bu SnCl in acetonitrile at 80 °C),
3
3
the conversion for 1-bromonaphthalene was 54% in 1 h with
complete conversion achieved in 6 h. In a control experiment
without Bu SnCl, only 2% conversion was observed after 1
3
h with 54% conversion after 17 h.
In this reaction, we found t-Bu
racemic-2,2′-bis(diphenylphosphino)-1,1′-binaphthyl (BI-
NAP), 9,9-dimethyl-4,5-bis(diphenyl-phosphino) xanthene
Xantphos), 1,1-bis(diphenylphosphino)ferrocene (dppf),
These results validated the premise, but we felt that there
would be problems inherent in this approach. Although
optimization of the addition time might improve conversion,
the addition time affording complete conversion would be
highly dependent on the substrate. For use at production
scale, the reaction would depend on the tight control of
integral parameters (time, temperature, addition rate), making
consistent performance difficult to achieve. Finally, a
stoichiometric amount of the tin compound is required.
P10 to be superior to
3
(
P(o-Tol) , P(o-furyl) , and PPh . With a ratio of Pd to t-Bu P
3
3
3
3
between 1:1 and 1:2.5, complete conversion was observed
in cyanation of 1-bromonaphthalene, but when the ratio of
Formation of Bu
3
SnCN in the reaction of a tri-n-butyltin
halide with KCN suggested an alternate means of controlling
the level of Bu SnCN while reducing the amount of tin
9
Pd to t-Bu
3
P was 1:3, the conversion was only 40%. With
3
t-Bu P as a ligand, aryl bromides were generally converted
3
to the nitrile at 50 °C. With other ligands, a somewhat higher
temperature (80 °C) was required.
With these encouraging results in hand, we applied this
new protocol to a range of substrates, including aryl
(
6) (a) Sundermeier, M.; Zapf, A.; Mutyala, S.; Baumann, W.; Sans, J.;
Weiss, S.; Beller, M. Chem. Eur. J. 2003, 8, 1828. (b) Sundemerier, M.;
Zapf, A.; Beller, M.; Sans, J. Tetrahedron Lett. 2001, 42, 6707.
(
7) Sundermerier, M.; Zapf, A.; Beller, M. Angew. Chem., Int. Ed. 2003,
2, 1661.
8) Kingsbury, W. D.; Boehm, J. C.; Jakas, D. R.; Holden, K. G.; Hecht,
4
(
S. M.; Gallagher, G.; Caranfa, M. J.; McCabe, F. L.; Faucette, L. F.;
Johnson, R. K.; Hertzberg, R. P. J. Med. Chem. 1991, 34, 98.
(10) (a) Otsuka, S.; Yoshida, T.; Matsumoto, M.; Nakatsu, K. J. Am.
Chem. Soc. 1976, 98, 5850. (b) Rithner, C. D.; Bushweller, C. H. J. Am.
Chem. Soc. 1985, 107, 7823.
(9) Tanaka, M. Tetrahedron Lett. 1980, 21, 2959.
2838
Org. Lett., Vol. 6, No. 17, 2004

bromides, heteroaryl bromides, an aryl iodide, and aryl
triflate. To ensure good performance with a wide range of
substrates, we increased the catalyst loading to 0.5 mol %
which are considered deactivated or slightly deactivated
substrates for cyanation. Good conversion was achieved with
substrates containing fluoride, trifluoromethyl, methyl ketone,
ester, chloride, and hydroxymethyl (entries 5-8, 13-14)
functionality. Aryl bromides substituted with a deactivating
methoxyl group (entries 9, 15) provided moderate yields.
To improve the yield for 3-bromoanisole, we adjusted the
amount of tributyltin chloride and found that the yield
improved dramatically (entry 15). It will be necessary to fine-
3
and the corresponding Bu SnCl charge to 0.14 mol %. The
reactions were conducted at 80 °C and stopped after 17 h.
Table 1 shows that a variety of cyanation substrates are
converted to the corresponding nitriles using these conditions.
Table 1. Pd-Catalyzed Cyanation Results
3
tune the amount Bu SnCl for some substrates. Cyanation of
an aryl iodide (entry 12) was completed at ambient temper-
ature in less than 3 h. Even after tuning the amount of ligand
and Bu SnCl for cyanation of 3-bromopyridine, the yield was
3
only 46%. For heteroaryl bromides, the chelating ligand
Xantphos afforded higher yields. With 3-bromopyridine as
a substrate, we further optimized the amount of ligand and
Bu
Using 1 mol % Pd and a Pd-to-Xantphos ratio of 1:0.5,
with 0.26 mol % Bu SnCl, the conversion increased to 97%
entry 17). The results show that conversion is lower when
3
SnCl.
3
(
3
the Bu SnCl charge is reduced. It is worth noting that the
optimal Pd to Xantphos ratio is 1:0.5, which is unusual for
a bidentate ligand. When the ratio of Pd (dba) to Xantphos
2
3
is 1:2 (Pd to Xantphos ratio of 1:1), the reaction conversion
was lower. It appears that the dba ligand remains associated
with the Pd through the catalytic cycle. However, no reaction
2 3
is observed using Pd (dba) without a phosphine ligand.
We applied these conditions in the cyanation of 3-bro-
moquinoline and a quantitative yield was obtained (entry 19).
We also used these conditions for a nitro-substituted aryl
bromide and aryl triflate. Both the nitroaryl bromide and
triflate (entries 10, 11) were converted to the corresponding
nitrile smoothly. An aryl triflate, which is considered a
1
1
difficult substrate in Pd-catalyzed cyanation, performed
well. A lower yield was observed using Pd (dba) /t-Bu P.
2
3
3
As with the aryl iodide, a heteroaryl iodide (entry 18) was
converted to the nitrile at ambient temperature in 3 h with a
quantitative yield. Unfortunately, 2-bromopyridine gave a
moderate yield using the general method. We found,
however, that dppf is a good ligand for the cyanation of
2
-bromopyridine. Using dppf rather than Xantphos with
conditions otherwise the same gave 96% yield in the
cyanation of 2-bromopyridine (entry 16).
Although it was gratifying to see that our reasoning had
led to such a general method, in recognizing the extent of
this success, we had to question the validity of our premise.
In order for this method to be as general as it proved to be
and to work so well at very low catalyst loading, the reaction
of Pd(0) with Bu
must be very slow, inconsistent with what had been reported
for Me SnCN. It has been reported that Me SnCl reacts with
3
SnCN removing Pd from the catalytic cycle
a
Reaction conditions: 1.0 equiv of 1-aryl bromide, 1.5 equiv of KCN,
acetonitrile (0.94 mL/mmol), 0.0014 equiv of Bu3SnCl, 0.005 equiv of
Pd2(dba)3, 0.025 equiv of t-Bu3P, 80 °C, 17 h. Reaction conditions: 1.0
equiv of heteroaryl bromide, 1.5 equiv of KCN, acetonitrile (0.94 mL/mmol),
.0027 equiv of Bu3SnCl, 0.005 equiv of Pd2(dba)3, 0.005 equiv of
Xantphos, 80 °C, 17 h. Reaction conditions: 22 °C, 3 h. Performed with
.0007 equiv of Bu3SnCl. Performed with 0.0014 equiv of Bu3SnCl and
.005 equiv of dppf.
b
3
3
KCN in the presence of a crown ether to form an ate
0
12
complex. If an ate complex were formed on reaction of
c
d
e
0
0
(11) (a) Aburaki, S.; Okuyama, S.; Hoshi, H.; Kamachi, H.; Nishio, M.;
Hasegawa, T.; Masuyoshi, S.; Iimura, S.; Konishi, M.; Oki, T. J. Antibiotics
1993, 46, 1447. (b) Sonesson, C.; Waters, N.; Svensson, K.; Carlsson, A.;
Smith, M. W.; Piercey, M. F.; Meier, E.; Wikstron, H. J. Med. Chem. 1993,
Excellent yields were observed for aryl bromides and
36, 3188.
methyl- or phenyl-substituted aryl bromides (entries 1-4),
(12) Johnson, S. E.; Knobler, C. B. Organometallics 1992, 11, 3684.
Org. Lett., Vol. 6, No. 17, 2004
2839

3 3
Bu SnCl or Bu SnCN with KCN, the transfer of cyanide to
Pd(II), as required for cyanation, might be favored, and the
reactions removing Pd from the catalytic cycle might be
suppressed.
Scheme 2. Catalytic Cycle for Pd-Catalyzed Cyanation of
Aryl Bromides Promoted by Tin Ate Complexes
NMR experiments provided evidence that ate complexes
are formed in the cyanation reaction. On treating Bu
with an excess of solid KCN in acetonitrile-d at ambient
temperature, the observed H and 13C spectra did not
correspond to spectra obtained for commercial Bu SnCN.
Likewise, treatment of Bu SnCN with excess solid KCN
3
SnCl
3
1
3
3
1
13
resulted in a substantial change in the H and C spectra.
These results are consistent with the formation of ate
complexes. We are working to further characterize these
complexes and the equilibria involved. These results will be
reported in due course.
other promoters are being explored to further broaden the
scope of this method to include aryl chlorides.
On the basis of these observations, we propose the catalytic
cycle depicted in Scheme 2.
In conclusion, we have shown that organotin compounds
can be used at very low levels to promote the palladium-
catalyzed cyanation of aryl bromides. The reaction conditions
are mild, and the catalyst loading is 0.5 mol % or lower.
KCN is used as a cyanide source in place of Cu and Zn
salts with advantages in cost and water solubility for workup.
It was found that tin ate complexes are formed under the
conditions of the reaction, and these complexes are proposed
to be key to the success of the method. With this discovery,
Acknowledgment. We would like to acknowledge Pro-
fessor Barry Trost for useful suggestions and Ms. Lisa
DiMichele for NMR support.
Supporting Information Available: General experimen-
tal procedure details. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL049621D
2840
Org. Lett., Vol. 6, No. 17, 2004