First synthesis of β-keto sulfoxides by a palladium-catalyzed carbonylative suzuki reaction

Article abstract of DOI:10.1021/ol0518737

(Chemical Equation Presented) An unprecedented palladium-catalyzed three-component cross-coupling reaction between α-bromo sulfoxide, carbon monoxide, and aromatic boronic acids provides a new and efficient approach to the synthesis of β-ketosulfoxides. T

Full text of DOI:10.1021/ol0518737

ORGANIC
LETTERS
2005
Vol. 7, No. 21
4669-4672
First Synthesis of
a Palladium-Catalyzed Carbonylative
Suzuki Reaction
â-Keto Sulfoxides by
Mercedes Medio-Simo´n, Cristian Mollar, Nuria Rodr´ıguez, and Gregorio Asensio*
Departamento de Qu´ımica Orga´nica, UniVersidad de Valencia,
AVda Vicent Andres Estelles s/n, 46100 Burjassot, Spain
gregorio.asensio@uV.es
Received August 4, 2005
ABSTRACT
An unprecedented palladium-catalyzed three-component cross-coupling reaction between
r-bromo sulfoxide, carbon monoxide, and aromatic
boronic acids provides a new and efficient approach to the synthesis of -ketosulfoxides. The reaction takes place under mild conditions with
â
a wide range of variously substituted aryl and heteroaryl boronic acids. The carbonylative cross-coupling reaction is strongly favored over
competing direct cross-coupling and homocoupling processes, except with boronic acids carrying strong electron-withdrawing substituents.
The cross-coupling reaction is considered one of the most
straightforward and general methods for the formation of
carbon-carbon bonds.1 In this area, the transition-metal-
catalyzed three-component cross-coupling reaction between
organometallic reagents, carbon monoxide, and organic
halides is now considered a useful tool for ketone synthesis.2
Aryl, 1-alkenyl, 1-alkynyl, allyl, benzyl, and alkyl halides
have been shown to be suitable electrophiles for this
carbonylative cross-coupling reaction.3 â-Hydride elimination
associated with alkyl halides with an sp3 carbon containing
â-hydrogens does not proceed in the carbonylation reaction
because the insertion of carbon monoxide produces the
corresponding acylpalladium(II) halides. While various or-
ganometallic reagents including magnesium,4 aluminum,5
silicon,6 tin,7 and boron derivatives8 have been reported to
undergo carbonylative coupling, organoboron reagents, which
are generally nontoxic and thermally, air-, and moisture-
stable, offer an obvious practical advantage compared to
other cross-coupling processes. Due to these advantages, the
Suzuki palladium-catalyzed cross-coupling reaction of or-
ganoboron reagents with organic halides or pseudohalides
to form biaryl derivatives has emerged over the past two
decades as an extremely powerful tool in organic synthesis.9
The palladium-catalyzed three-component cross-coupling of
aryl halides, carbon monoxide, and aromatic boronic acids,
which is closely related to the Suzuki reaction, also provides
convenient access to diaryl ketones.10 However, the main
drawback of the carbonylative Suzuki reaction often lies in
the formation of significant amounts of the direct coupling
(7) (a) Kang, S. K.; Yamaguchi, T.; Kim, T.-H.; Ho, P.-S. J. Org. Chem.
1996, 61, 9082. (b) Davies, S. G.; Pyatt, D.; Thomson, C. J. Organomet.
Chem. 1990, 387, 381. (c) Echevarren, A. M.; Stille, J. K. J. Am. Chem.
Soc. 1988, 110, 1557. (d) Stille, J. K. Angew. Chem., Int. Ed. Engl. 1986,
25, 508.
(8) (a) Ishiyama, T.; Murata, M.; Suzuki, A.; Miyaura, N. J. Chem. Soc.,
Chem. Commun. 1995, 295. (b) Ishiyama, T.; Miyaura, N.; Suzuki, A.
Tetrahedron Lett. 1991, 32, 6953. (c) Ishiyama, T.; Miyaura, N.; Suzuki,
A. Bull. Chem. Soc. Jpn. 1991, 64, 1999. (d) Wakita, Y.; Yasunaga, T.;
Akita, M.; Kojima, N. J. Organomet. Chem. 1986, 301, C17.
(9) (a) Bellina, F.; Carpita, A.; Rossi, R. Synthesis 2004, 2419. (b) Suzuki,
A. J. Organomet. Chem. 1999, 576, 147.
(1) (a) Tsuji, J. Transition Metal Reagents and Catalysts; Wiley:
Chichester, 2000. (b) Miyaura, N. Cross-Coupling Reactions; Springer
-Verlag: Berlin, Heidelberg, 2002.
(2) Ishiyama, T.; Kizaki, H.; Hayashi, T.; Suzuki, A.; Miyaura, N. J.
Org. Chem. 1998, 63, 4726 and references therein.
(3) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457.
(4) Yamamoto, T.; Kohara, T.; Yamamoto, A. Chem. Lett. 1976, 1217.
(5) Bumagin, N. A.; Ponomaryov, A. B.; Beletskaya, I. P. Tetrahedron
Lett. 1985, 26, 4819.
(6) (a) Hatanaka, Y.; Fukushima, S.; Hiyama, T. Tetrahedron 1992, 48,
2113. (b) Hatanaka, Y.; Hiyama, T. Synlett 1991, 845.
(10) (a) Couve-Bonnaire, S.; Carpentier, J.-F.; Morteux, A.; Castanet.
Y. Tetrahedron 2003, 59, 2793. (b) Maerten, E.; Hassouna, F.; Couve-
Bonnaire, S.; Morteux, A.; Carpentier, J.-F.; Castanet. Y. Synlett 2003, 1874.
10.1021/ol0518737 CCC: $30.25
© 2005 American Chemical Society
Published on Web 09/27/2005

product without carbon monoxide insertion, particularly with
electron-deficient aryl halides.
Table 1. Palladium-Catalyzed Three-Component Suzuki
We have recently described a novel Suzuki-Miyaura cross-
coupling reaction of R-bromo sulfoxides with aromatic
boronic acids.11a,b While the usual procedure allows the
formation of Csp2-Csp2 bonds through the Suzuki cross-
coupling reaction, the procedure described by us yields the
much less usual11c-e formation of a Csp3-Csp2 bond. We
report here the new palladium-catalyzed three-component
cross-coupling of R-bromo sulfoxides, carbon monoxide and
aromatic boronic acids. Our present report widens the scope
of the palladium-catalyzed carbonylative Suzuki reaction,
since for the first time we show that R-bromo sulfoxides
are suitable electrophiles for this type of reaction. In addition,
this method represents a new approach for the synthesis of
â-keto sulfoxides. Three general methods have been reported
in the literature for the synthesis of â-keto sulfoxides, namely,
i) the reaction of R-sulfinyl anions derived from sulfoxides
with esters or nitriles,12 ii) the oxidation of â-thioethers,13
and iii) the addition of ketone enolates to sulfinate esters.14
The first method is used most often, in particular for the
synthesis of chiral acyclic â-keto sulfoxides, which are
widely used in asymmetric synthesis.15 Although the syn-
thesis of many enantiomerically pure â-keto sulfoxides has
been reported using this method, a major drawback is that
only a moderate conversion of the substrate can be achieved
due to quenching of the R-sulfinyl anion by the reaction
product present in the medium, in addition to limitations in
the number of functional groups compatible with the organo-
lithium reagents. Thus, the development of alternative
methods of synthesis to overcome these limitations should
be of major interest.
Cross-Coupling
yieldsa (%)
entry
2
R
t (h)
3
4
1
2
3
4
5
a
C6H5
3
2
2
2
2
3
3
3
3
3
3
3
1
3
15
18
80b
95
88b
85b
88b
93c
54
45b
77
33b
0
0
0
0
0
0
0
9
0
23
0
b
c
d
e
b
f
4-MeOC6H4
3-MeOC6H4
2-MeOC6H4
2-MeC6H4
4-MeOC6H4
4-BrC6H4
2-BrC6H4
4-CF3C6H4
2-CF3C6H4
2-CH3COC6H4
3-NO2C6H4
3-thienyl
2-thienyl
Me
6
7
8
9
g
h
i
j
k
l
m
n
o
10
11
12
13
14
15
16
0
55b
99
72
5d
0
0
0
0
Bu
0
0
a Isolated yields. b Conversion > 95% calculated from reacted sulfoxide
1. c 5 Mol % of catalyst used. d Detected by H-NMR.
1
insertion, diphenyl arising from the homocoupling reaction
of 2a, benzophenone resulting from the carbonylative homo-
coupling reaction of the boronic acid 2a, and the sulfoxide
resulting from dehalogenation, was not significant. The
reaction took place under conditions that are considered
unfavorable for the carbonylative cross-coupling of aryl
halides. In particular, the reaction took place at moderate
temperatures under atmospheric pressure, in contrast to the
carbonylative cross-coupling reaction of aryl halides in which
high temperatures and an overpressure of carbon monoxide
are needed to promote the insertion process. The best
An initial experiment was performed with R-bromo
sulfoxide 1, phenyl boronic acid 2a, CsF and Pd(PPh3)4 (10%
mol) as a catalyst under an atmospheric pressure of carbon
monoxide (balloon) (Scheme 1).16 The reaction took place
(11) (a) Rodr´ıguez, N.; Cuenca, A.; Ram´ırez de Arellano, C.; Medio-
Simo´n, M.; Peine, D.; Asensio, G. J. Org. Chem. 2004, 69, 8070. (b)
Rodr´ıguez, N.; Cuenca, A.; Ram´ırez de Arellano, C.; Medio-Simo´n, M.;
Asensio, G. Org. Lett. 2003, 5, 1705. (c) Kirchhoff, J. H.; Netherton, M.
R.; Hills, I. D.; Fu, G. C. J. Am. Chem. Soc. 2002, 124, 13662. (d) Netherton,
M. R.; Dai, C.; Neuschutz, K.; Fu, G. C. J. Am. Chem. Soc. 2001, 123,
10099. (e) Cardenas, D. J. Angew. Chem., Int. Ed. 1999, 38, 3018.
(12) (a) Vleggaar, R.; Zeevaart, J. G. Tetrahedron Lett. 1999, 40, 9301.
(b) Solladie´, G.; Fre´chou, C.; Demailly, G.; Greck, C. J. Org. Chem. 1986,
51, 1912. (c) Annunciata, R.; Cinquini, M.; Cozzi, F. J. Chem. Soc., Perkin.
Trans. 1 1979, 1687.(d) Corey, E. J.; Chaykovsky, M. J. Am. Chem. Soc.
1964, 86, 1639.
Scheme 1
(13) (a) Lattanzi, A.; Bonadies, F.; Schiavo, A.; Seetri, A. Tetrahedron:
Asymmetry 1998, 9, 2619. (b) Breitschuh, R.; Seebach, D. Synthesis 1992,
1170. (c) Ohta, H.; Kato, Y.; Tsuchihashi, G. J. Org. Chem. 1987, 52, 2735.
(14) (a) Carren˜o, M. C.; Garc´ıa-Ruano, J. L.; Pedregal, C.; Rubio, A. J.
Chem. Soc., Perkin Trans. 1 1989, 1335. (b) Carren˜o, M. C.; Garcia-Ruano,
J. L.; Rubio, A. Tetrahedron Lett. 1987, 28, 4861. (c) Monteiro, H. J.; De
Souza, J. P. Tetrahedron Lett. 1975, 16, 921.
(15) (a) Garc´ıa-Ruano, J. L. Top. Curr. Chem. 1999, 204, 1. (b) Carren˜o,
M. C. Chem. ReV. 1995, 95, 1717. (c) Hanquet, G.; Colobert, F.; Lanners,
S.; Solladie´, G. ArkiVoc 2003, Vii, 328.
(16) Representative Procedure. A mixture of R-bromo sulfoxide 1 (0.4
mmol), boronic acid 2 (0.8 mmol), CsF (1.6 mmol), and Pd(PPh3)4 (0.04
mmol) was added to a flask fitted with a reflux condenser and a septum
inlet. The flask was flushed with carbon monoxide and then charged with
THF (6 mL). The mixture was stirred at 60 °C under an atmospheric pressure
of carbon monoxide. After the appropriate reaction time, the mixture was
cooled at room temperature, quenched with water (10 mL), and extracted
with diethyl ether (2 × 15 mL) and dichloromethane (3 × 15 mL). The
combined organic extracts were dried with Na2SO4 and evaporated under
reduced pressure.
to completion after 2 h and the corresponding keto sulfoxide
3a was isolated in 80% yield (Table 1, entry 1). Formation
of the expected side products, i.e., benzyl sulfoxide 4a arising
from the direct cross-coupling reaction without carbonyl
4670
Org. Lett., Vol. 7, No. 21, 2005

palladium-to-ligand ratio (1:4) for the three-component
coupling with the R-bromo sulfoxide 1 is unusual among
carbonylative reactions with aryl halides. With these com-
pounds, deviation from a 1:2 ratio reduces the activity of
the catalyst since the coordination sphere of palladium is
likely to be too hindered to allow good activity. These facts
prompted us to perform a survey of the reaction with a series
of representative boronic acids (Table 1). Aryl boronic acids
2b-e substituted with electron-donating groups (Table 1,
entries 2-5) were found to be very active in the carbonyl-
ative cross-coupling. The position of the electron-donating
group in the aromatic ring did not induce noticeable
differences in the chemoselectivity or extent of conversion
(Table 1, entries 2-5). Thus, good yields were obtained with
the relatively hindered o-substituted boronic acids 2d and
2e (Table 1, entries 4 and 5). Significant amounts of side
products were not detected in any case. Aryl boronic acids
substituted with electron-donating groups gave the corre-
sponding keto sulfoxides 3b-e in high yields after short
reaction times.
carbonyl species before the migration step becomes more
difficult.18
Even the absence of a direct cross-coupling product can
be explained by the same effect, since coordination of the
acyl group of the boronic acid might hinder the reductive
elimination step. Heteroaryl boronic acids 2l and 2m, which
had failed to give the direct cross-coupling reaction,11 reacted
under carbonylative conditions to give the corresponding
â-keto sulfoxides (Table 1, entries 13 and 14). The behavior
of boronic acid 2l was similar to that observed for electron-
rich aryl boronic acids, whereas 2m behaved like an electron-
poor aryl boronic acid. Alkyl boronic acids 2n and 2o were
also tested but unfortunately failed to give the carbonylative
Suzuki reaction product (Table 1, entries 15 and 16). In the
reaction of 2n, small amounts of the corresponding product
3n were detected by NMR analyses of the crude reaction
mixture.
The results obtained in the present carbonylative Suzuki
reaction are likely not due to the following usually proposed
catalytic cycle: Oxidative addition of the organic halide to
palladium(0) (step 1), migratory insertion of carbon mon-
oxide (step 2), transmetalation of the arylboronic acid (step
3), and reductive elimination (step 4). However, if it is
assumed to occur through a permutation of steps 2 and 3,
i.e., carbon monoxide insertion takes place after transmeta-
lation, the experimental results are nicely explained. In fact
if carbon monoxide insertion did occur in the second step
after oxidative addition, we should expect that a change in
the electronic properties of the boronic acid would not
produce any significant difference in the selectivity of the
carbonylative Suzuki product compared to the direct cross-
coupling reaction. Moreover, the carbonylative cross-
coupling product should be favored with substituted boronic
acids with electron-attracting groups, since in this case
transmetalation (third step) is slow and accordingly the
competitive direct Suzuki coupling is hindered (Scheme 2).
When the amount of catalyst was reduced to 5 mol %,
boronic acid 2b reacted to completion after 3 h (Table 1,
entry 6). Reactions were complete much faster in the case
of carbonylative cross-coupling than in the case of direct
cross-coupling of related substrates, consistent with the
known effect of carbon monoxide in accelerating reductive
elimination.17 Aryl boronic acids bearing an electron-
withdrawing group in the aromatic ring 2f-k (Table 1,
entries 7-12) were less reactive than the parent 2a or
electron-rich aryl boronic acids 2b-e. Therefore, the car-
bonylation reaction proceeded more slowly and the corre-
sponding keto sulfoxides 3f-i and 3k were produced in
moderate yields. The chemoselectivity for the carbonylative
cross-coupling was still favored, but the selectivity progres-
sively decreased as the electron-attracting character of the
para substituent in the aryl boronic acid increased (see Table
1, entries 7 and 9). However, the presence of a meta
substituent with strong electron-attracting character did not
favor the formation of the direct coupling product 4.
Nevertheless, the yield of carbonylative product 3k was only
moderate in this case (Table 1, entry 12).
Scheme 2
The carbonylative Suzuki reaction failed with 2-acetyl-
phenyl boronic acid and 2j (Table 1, entry 11) and proceeded
slowly with 2-trifluoromethylphenyl boronic acid (2i) (Table
1, entry 10). The finding that the o-acetyl and o-trifluoro-
methyl groups, respectively, precluded or retarded the
reaction seems to be related to the coordinating ability of
the substituent (acting as a ligand) and not to electronic or
steric effects. In fact, the reaction took place with other ortho-
substituted boronic acids such as 2d, 2e, and 2g which lack
heteroatoms or carry heteroatoms in a position that does not
allow coordination with the metal (Table 1, entries 4, 5, and
8). Hence, the deceleration or inhibition of the reaction may
be related to the presence of fluorine or oxygen atoms that
might occupy a place in the coordination sphere of the metal.
The required displacement of one ligand for carbon monoxide
in the square-planar plane to form a four-coordinate cis-
By the usual mechanism in our case, the carbonylative
cross-coupling should be favored for boronic acids bearing
electron-attracting groups, which is just the opposite of the
experimental results. Conversely, if carbon monoxide inser-
tion takes place after transmetalation, the aryl group can play
(17) Hegedus, L. S. Transition Metals in the Synthesis of Complex
Organic Molecules, 2nd ed.; University Sciences Books: Sausalito, 1999.
(18) Lindsell, W. E.; Palmer, D. D.; Preston, P. N.; Rosair, G. M.
Organometallics 2005, 24, 1119.
Org. Lett., Vol. 7, No. 21, 2005
4671

a role in controlling carbon monoxide insertion. It is known
from the carbonylative Suzuki reaction of aryl halides that
carbon monoxide insertion occurs faster with electron-rich
aryls than with aryl groups with electron-withdrawing
substituents. Thus, direct cross-coupling without carbon
monoxide insertion is easier with electron-poor aryls, and
the selectivity for the carbonylative coupling decreases. We
observed the same trend with the different boronic acids
assayed. As a consequence, it appears that carbon monoxide
insertion is also controlled by an aryl group, but now the
aryl group is provided by the boronic acid. Consequently,
the hypothesis that insertion occurs after transmetalation
accounts for our experimental results.
of the textbook catalytic cycle seems to be necessary for
these substrates to account for the results obtained with
different aryl and heteroaryl boronic acids. Hence, the
chemoselectivity between carbonylative cross-coupling prod-
ucts and direct cross-coupling products in the reaction of
R-bromo sulfoxides seems to depend on the electronic
character of the aryl group present in the palladium complex.
The influence of the aryl group on the control of the
migratory insertion of carbon monoxide suggests that this
process might occur after the transmetalation step. Thus, the
established sequence for the catalytic cycle that explains the
carbonylative cross-coupling reactions of aryl halides might
be different in the case of R-bromo sulfoxides.
In conclusion, we have described the first example of a
palladium-catalyzed three-component cross-coupling reaction
with R-bromo sulfoxides as electrophiles. The reaction takes
place under very mild conditions and with high selectivity
for a wide range of boronic acids. This method does not
require an overpressure of carbon monoxide or special
palladium ligands. The mildness of the reaction conditions
allows us to obtain â-keto sulfoxides containing functional
groups such as -NO2 that would not be available using
alternative procedures from the literature such as the addition
of the lithium R-sulfinyl anion to esters. Thus, a wide array
of â-keto sulfoxides or sulfinyl aromatic ketones function-
alized in the ketone ring are readily available. Concerning
the reaction mechanism of the carbonylative Suzuki cross-
coupling reaction with R-bromo sulfoxides, a modification
Acknowledgment. Financial support from the Spanish
Direccio´n General de Investigacio´n Cient´ıfica y Te´cnica
(BQU2003-00315) and the Generalitat Valenciana (Grupos
2003-121) is gratefully acknowledged. N.R thanks the
Spanish Ministerio de Educacion for a fellowship. We
gratefully acknowledge the SCSIE (Universidad de Valencia)
for access to the instrumental facilities.
Supporting Information Available: Experimental details
for the synthesis and characterization of compounds in Table
1. This material is available free of charge via the Internet
at http://pubs.acs.org.
OL0518737
4672
Org. Lett., Vol. 7, No. 21, 2005