Nonpolar biphasic catalysis: Sonogashira and Suzuki coupling of aryl bromides and chlorides

Article abstract of DOI:10.1039/b303602b

Palladium-phosphine catalysts which are phase-tagged with soluble poly(4-methylstyrene) can be used for efficient carbon-carbon coupling reactions by nonpolar biphasic catalysis with high recyclability as evidenced by high yields, constant tof of the cata

Full text of DOI:10.1039/b303602b

Nonpolar biphasic catalysis: Sonogashira and Suzuki coupling of aryl
bromides and chlorides
Anupama Datta and Herbert Plenio*
Institut für Anorganische und Physikalische Chemie, TU Darmstadt, Petersenstr. 18, 64287 Darmstadt,
Germany. E-mail: plenio@tu-darmstadt.de
Received (in Cambridge, UK) 2nd April 2003, Accepted 29th April 2003
First published as an Advance Article on the web 21st May 2003
Palladium–phosphine catalysts which are phase-tagged with
soluble poly(4-methylstyrene) can be used for efficient
carbon–carbon coupling reactions by nonpolar biphasic
catalysis with high recyclability as evidenced by high yields,
constant tof of the catalyst and the absence of significant
leaching of the catalyst into the polar product phase.
for the Sonogashira coupling10 and was used together with the
phosphinated polystyrene, CuI and the coupling partners to
form the catalytically active species after deprotonation of the
phosphonium salt by HNiPr2 and reduction of the Pd(II) to
Pd(0). The first task was to find a suitable biphasic solvent
system composed of two room temperature immiscible solvents
to carry out Sonogashira reactions with high activity. Cyclohex-
ane was the obvious choice as the catalyst phase since the Pd
complex with the phosphinated polymer does not dissolve in
less polar alkanes. The critical choice is the polar solvent used
as a product phase since many solvents are known to inhibit the
catalyst activity. Several solvents were tested in combination
with cyclohexane: acetonitrile, DMF, DMA, nitromethane and
DMSO. The biphasic system composed of DMSO–cyclohexane
turned out to be the best choice. In all other solvent mixtures the
coupling yields were found to be moderate to low; in addition
the yields decrease significantly over the consecutive reaction
cycles. Among the various bases tested (Et3N, Na2CO3, K3PO4
and HNi Pr2) in DMSO–cyclohexane HNi Pr2 gives the highest
yield in the Sonogashira coupling reaction.
The results of the Sonogashira coupling experiments are
summarized in Table 1. The overall yields of isolated product
for the different reactions after five cycles are between 79 and
96% per cycle. The phase yields listed in the table correspond to
the amount of product isolated after separation of the product
phase from the catalyst phase. The phase yields only take into
account the amount of product in the catalyst phase and thus
reflect the partition coefficients of the respective products.11
The acetylenes with the more polar groups studied here display
a favorable distribution between the two solvents.
Biphasic catalysis utilizing two room-temperature immiscible
solvents relies on catalysts which are phase-tagged1 to be
soluble only in one of the two solvents,2,3 which is termed
catalyst phase. The product of the catalytic reaction should be
located preferentially in the other solvent, termed product
phase. In this manner the separation of catalyst and product
boils down to a simple phase separation. While the phase
preference of the catalyst can be designed, the products of
catalytic reactions have different partition coefficients, depend-
ing on their chemical nature. Traditionally biphasic catalysts
have been tagged with polar groups [triphenylphosphine-
trisulfonate,4 PEG = poly(ethylene glycol)5] to generate the
respective polar phase preference for polar biphasic catalysis or
with a fluorinated ponytail to be used in fluorous biphasic
chemistry.6 †
While polar biphasic catalysis is best suited for synthesis and
separation of nonpolar products, it is obvious that for polar
substrates the partition coefficient dictates that the separation of
products from the catalyst is facile with catalyst having a
nonpolar phase tag.7,8 We wish to term this nonpolar biphasic
catalysis and want to describe here our work aimed towards
establishing this type of biphasic catalysis.
We have recently demonstrated that Pd(0) complexes with
BnP(1-Ad)2 (Ad = adamantyl), which is a sterically demanding
and electron-rich phosphine, are highly active catalysts suitable
for the room temperature Sonogashira coupling of aryl
bromides.9 Following minor modifications of the peripheral
functional groups this ligand has now been attached to a
nonpolar, linear poly(methylstyrene). The anionic polymer-
isation of 4-methylstyrene was controlled to result in a 5 kDa
polymer, which was reacted with substoichiometric amounts of
Br2 to convert 15% of the –CH3 groups to –CH2Br. Heating
(1-Ad)2PH with the functionalized polymer under toluene
reflux gave the phosphinated polymer in quantitative yield
(Scheme 1).
For the evaluation of catalyst recycling the nearly constant
yields of the 4th and 5th reaction cycle point to the high
performance of the catalyst. The recyclability of a given catalyst
can be demonstrated more convincingly by determining the
turnover frequency (tof) for an incomplete reaction over several
cycles. We tested this for the coupling of 4-bromoacetophenone
and phenylacetylene and found the tof to be virtually constant
(cycle (tof): cycle 1 (220 h21), cycle 2 (220 h21), cycle 3 (215
Table 1 Nonpolar biphasic Sonogashira coupling
We wish to describe here Sonogashira coupling reactions
using Pd catalysts supported on soluble poly(methylstyrene)s.
Pd(PhCN)2Cl2 was shown to be one of the best metal sources
Phase yield (%)/cycle
Extract Yield
R
RA
1
2
3
4
5
(%)a
(%)b
CH3CO
H
OMe
CH3CO
COOEt
Ph
Ph
Ph
91
88
85
97
95
91
87
99
105
98 101
98
91
99 100
2
11
3
17
2
95
92
89
79
96
95
98
89
98
89
Hexyl 82
Ph 95
103 102
104
a Extract corresponds to the amount of product extracted from the catalyst
phase after five cycles. b Yield refers to the amount of product after
chromatographic purification. Conditions: 1.0 mol% Pd(PhCN)2Cl2, 2.0
mol% polymer, 2.0 mol% CuI, HNiPr2, reaction times: 2–12 h depending on
the substrates but constant during cycles.
Scheme 1 Poly(methylstyrene) supported phosphine.
1504
CHEM. COMMUN., 2003, 1504–1505
This journal is © The Royal Society of Chemistry 2003

h21), cycle 4 (210 h21), cycle 5 (210 h21)) over five cycles.
This indicates excellent stability of the catalyst and suggests
that the extent of leaching of the complex catalyst into the
product phase is small.
The same ligand which was used for the Sonogashira
coupling was utilized for Suzuki type coupling reactions,12
while Pd(OAc)2 was found to be a better metal source. The main
results concerning these coupling reactions are summarized in
Table 2.
soluble in the product phase solvent (here DMSO or nitro-
methane). By GC we found that there is only a small loss of
cyclohexane into DMSO (2%) and nitromethane (3.5%).
Furthermore the loss of polymer into the product phase was
checked with 1H NMR spectroscopy. We were not able to detect
any polymer in the crude products. We thus estimate the
polymer leaching to be significantly smaller than 0.5%.
Secondly, the amount of palladium lost into the DMSO or
nitromethane product phase was determined by two independ-
ent methods: total reflection XRF and UV spectrophotometry
using the colorimetric reagent 4,4A-bis(dimethylamino)thio-
benzophenone.13 Results obtained from the two methods were
found to be in good agreement and the retention of the catalyst
in the cyclohexane layer was found to be > 99.8%.
Several biphasic solvent systems were tested; again cyclo-
hexane was combined with various polar solvents: acetonitrile,
DMF, DMA, DMSO and nitromethane. Among the various
solvent mixtures DMSO–cyclohexane and nitromethane–cyclo-
hexane were found to perform equally well, while the other
solvent combinations resulted only in moderate coupling yields
and significant decomposition of the catalyst after the first
reaction cycle. Due to its much lower boiling point nitro-
methane was finally preferred over DMSO. In order to further
optimize the reactions various bases (NaF, Na2CO3, K3PO4 and
HNi Pr2) were tested for the coupling reaction; K3PO4 gave the
highest yield in the Suzuki coupling reaction. It can be seen in
Table 2 that both aryl bromides and chlorides can be Suzuki
coupled and recycled several times in excellent yields. Again
the efficient recyclability of the catalyst is evidenced by the
almost constant and high coupling yields over the various
reaction cycles. Probing the catalyst performance by the more
stringent criterion tof provides evidence of the excellent
stability, as the tof remains almost constant over the reaction
cycles for the coupling of 4-bromoacetophenone and PhB(OH)2
(cycle (tof h21): cycle 1 (180 h21), cycle 2 (182 h21), cycle 3
(175 h21), cycle 4 (168 h21), cycle 5 (165 h21)).
The virtually quantitative retention of the catalyst in the
cyclohexane solvent, which is the catalyst phase, is essential for
the usefulness of biphasic catalysis. Consequently, the product
phase solvent should not strongly solvate Pd(0). For a polar
solvent this condition is less obviously met than for a nonpolar
solvent used as the product phase in polar biphasic catalysis.
However, it should be kept in mind that high catalytic activity
requires the presence of low coordinated Pd(0) species.
Consequently, high catalytic activity in a given product phase
and low leaching into the product phase can go hand in hand.
Thus for the Sonogashira and the Suzuki coupling reactions,
using poly(4-methylstyrene) supported Pd catalysts, the ab-
sence of significant leaching of the catalyst into DMSO or
nitromethane is indicated by the high yields of catalytic
transformations and the constant tof over the reaction cycles.
Another critical parameter concerning leaching of the
catalyst is the amount of catalyst phase solvent (cyclohexane)
This work was supported by the DFG and the Fonds der
Chemischen Industrie. We wish to thank Prof. Dr H. Ortner and
Dipl.-Ing. J. Saroukh for the TXRF measurements.
Notes and references
†
Biphasic Sonogashira reaction. In a Schlenk tube, Pd(PhCN)2Cl2 (5.7
mg, 1.5 mol%) and phosphinated polymer (15% loading, 150 mg, 3 mol%)
were dissolved in cyclohexane (7 ml). CuI (4.5 mg, 3 mol%), HNi Pr2 (0.1
ml), PhC·CH (1.8 mmol), the respective aryl bromide (1.5 mmol) and
DMSO (4 ml) were added and the reaction was stirred at 60 °C till
completion. After cooling to room temperature, the DMSO layer was
evaporated and the remaining crude product purified (see below). Fresh
DMSO, HNi Pr2 and the two substrates were added to the catalyst phase for
the next cycle to start the next reaction. After the last cycle, the product in
cyclohexane was isolated by evaporating cyclohexane. The crude products
from the various cycles were combined and purified by column chromatog-
raphy (silica, cyclohexane–ethyl acetate) to obtain the respective pure
products.
Biphasic Suzuki reaction. In a Schlenk tube, Pd(OAc)2 (3.3 mg, 1.5
mol%) and polymer (150 mg, 3 mol%) were dissolved in cyclohexane (7.0
ml). Aryl halide (1.0 mmol), phenyl boronic acid (183 mg, 1.5 mmol),
K3PO4 (24 mg, 2 mol%) and nitromethane (4 ml) were added to the solution
and stirred at 70 °C until all the starting material was consumed. After
cooling, the nitromethane layer was removed and was evaporated to obtain
the crude product. Two substrates, K3PO4 and nitromethane, were added to
the catalyst phase for the next cycle. The same procedure was repeated for
the five cycles. After the last cycle, the product in cyclohexane was isolated
by evaporating cyclohexane. The crude products from the various cycles
were combined and purified by column chromatography (silica, cyclohex-
ane–ethyl acetate) to obtain the respective pure product.
1 D. P. Curran, Angew. Chem., 1998, 110, 1230 (Angew. Chem., Int. Ed.,
1998, 37, 1174).
2 D. E. Bergbreiter, Chem. Rev., 2002, 102, 3345.
3 Aqueous-Phase Organometallic Chemistry (Eds.: B. Cornils, W. A.
Herrmann), Wiley-VCH, Weinheim, 1998.
Table 2 Nonpolar biphasic Suzuki coupling
4 D. E. Bergbreiter, P. L. Osburn, A. Wilson and E. M. Sink, J. Am. Chem.
Soc., 2000, 122, 9058.
5 C. W. Kohlpaintner, R. W. Fischer and B. Cornils, Appl. Catal., 2001,
221, 219.
Phase yield (%)/cycle
Extract Yield
6 E. de Wolf, G. van Koten and B.-J. Deelman, Chem. Soc. Rev., 1999, 28,
37.
R
X
1
2
3
4
5
(%)a
(%)b
7 (a) D. E. Bergbreiter, P. L. Osburn and J. D. Frels, J. Am. Chem. Soc.,
2001, 123, 11105; (b) D. E. Bergbreiter and C. Li, to be published.
8 E. Bayer and V. Schurig, Chemtech, 1976, 212.
9 A. Köllhofer and H. Plenio, Chem. Eur. J., 2003, 9, 1416.
10 T. Hundertmark, A. F. Littke, S. L. Buchwald and G. C. Fu, Org. Lett.,
2000, 2, 1729.
11 Since the composition of product and catalyst phase can vary slightly in
the course of the reaction, the phase yields can be higher than 100%.
12 J. Hassan, M. Sévignon, C. Gozzi, E. Schulz and M. Lemaire, Chem.
Rev., 2002, 102, 1359.
CH3CO Br
H
OMe
CH3CO Cl
CN Cl
91
88
85
82
82
97
95
91
87
90
105
98
98
91
91
99
101
98
89
92
100
95
98
89
90
6
8
8
7
4
95
92
90
83
86
Br
Br
a Extract corresponds to the amount of product extracted from the catalyst
phase after five cycles. b Yield refers to the amount of product after
chromatographic purification. Conditions: 1.5 mol% Pd(OAc)2, 3.0 mol%
polymer, K3PO4, reaction times: 6–24 h, depending on the substrates but
constant during cycles.
13 K. L. Cheng and B. L. Goydish, Microchem. J., 1966, 10, 158. The
validity of this method was also tested in the presence of small amounts
of phosphinated polymer.
CHEM. COMMUN., 2003, 1504–1505
1505