Highly active supported palladium catalyst for the regioselective synthesis of
2-arylpropionic acids by carbonylation
S. Jayasree, A. Seayad and R. V. Chaudhari*
Homogeneous Catalysis Division, National Chemical Laboratory, Pune- 411 008, India. E-mail: rvc@ems.ncl.res.in
Received (in Cambridge, UK) 30th March 1999, Accepted 5th May 1999
Table 1 Carbonylation of 1-(4-isobutylphenyl)ethanol (p-IBPE) to give
A catalyst system consisting of supported palladium in the
Ibuprofen using supported palladium and platinum catalystsa
presence of phosphine ligands, TsOH and LiCl catalyses the
carbonylation of 1-arylethanols to 2-arylpropionic acids
with significantly improved activity and regioselectivity; the
catalyst can be recycled with no loss in activity and
selecivity.
Conver- Selec-
Catalyst
Ligand
t/h
sion (%) tivity (%) TOF/h21
1% Pd/C
PPh3
PPh3
PPh3
PPh3
4.2 96
5.5 92
5.8 90
99.2
99.5
99.0
99.2
99.3
98.5
99.5
99.5
99.0
98.5
3375
2475
2285
550
1% Pd/g-alumina
1% Pd/H-ZSM-5
1% Pt/C
Pd metalb
1% Pd/C
Carbonylation of 1-arylethanols is of great interest since it
provides an environmentally benign route1 for the synthesis of
anti-inflammatory drugs such as Ibuprofen, Naproxen etc. The
importance of this route can be clearly demonstrated from the
commercialisation of an Ibuprofen process at Texas in 1992 by
Hoechst Celanese Corporation based on the carbonylation of
1-(4-isobutylphenyl)ethanol using a palladium complex cata-
lyst.2 The desired high selectivity ( > 98%) for Ibuprofen was
achieved only at very high pressures3 (16–34 MPa). This
homogeneous palladium catalyst system gave lower TOF†
(50–100 h21) and selectivity ( < 68%) at lower pressures ( ~ 6.8
MPa).4 Additives such as CuCl2 were reported to improve the
selectivity under similar conditions.5 In another study, a
biphasic catalyst with a water soluble palladium complex6 has
been reported but in this case also, very low reaction rates (TOF
= 2.3 h21) and lower Ibuprofen selectivity (70%) were
observed.
Here we report, for the first time, highly active and selective
supported palladium and platinum catalyst systems for the
synthesis of 2-arylpropionic acids from the corresponding
1-arylethanols. Supported palladium combined with PPh3, LiCl
and TsOH provides very high TOF (3375 h21) and more
importantly high selectivity (99.5%) for 2-arylpropionic acids
at lower pressures (5.4 MPa). The activity as well as the
selectivity was substantially higher compared to previous
reports under identical conditions. The catalyst was recycled
four times with no loss in activity and selectivity, attaining a
total TON of 55,000. These results indicate that this new
catalyst system (Pd-C/PPh3/TsOH/LiCl) is highly efficient and
provides significant improvement over the current state of the
art for the synthesis of 2-arylpropionic acids.
In a typical experiment, the supported catalyst,‡ the substrate,
the phosphine ligand, LiCl, TsOH, H2O and the solvent (methyl
ethyl ketone) were charged into a stirred pressure reactor and
the reaction was carried out at 5.4 MPa of CO partial pressure
at 115 ºC under 1100 rpm for a specified time. The reaction
mixture was analysed by gas chromatography§ and the products
were further characterized by GC-MS and NMR.
Typical results for the carbonylation of 1-(4-isobutylphenyl)-
ethanol (p-IBPE) using different catalysts, supports, promoters
and ligands are presented in Table 1. Pt/C provided compar-
atively lower activity than Pd/C. Other supports such as g-
alumina and H-ZSM-5 were also useful. Different acid and
halide sources were checked and were found to not significantly
effect the catalytic activity. The nature of the phosphine ligand
has a strong influence on the catalytic activity, as evidenced
from Table 1. No reaction was observed with diphos ligands
such as dppb, unlike the catalyst system reported by Ali et al. for
the carbonylation of olefins.7
24
8
90
97
PPh3
P(p-Tol)3
P(p-FPh)3 5.9 95
P(Cy)3
PPh3
90
4.5 94
3062
2384
308
1676
2272
1% Pd/C
1% Pd/C
24
8
6
50
91
93
1% Pd/Cc
1% Pd/Cd
a
PPh3
Conditions: p-IBPE (14.04 mmol), catalyst (10 mg), phosphine (0.1908
mmol), TsOH (5.6052 mmol), LiCl (5.6052 mmol), H2O (67 mmol), methyl
ethyl ketone (21 ml), PCO = 5.4 MPa, T = 115 ºC. b Pd metal (2 mg). c HCl
(5.6052 mmol) instead of TsOH and LiCl. d Bu4NCl instead of LiCl.
samples were withdrawn during reaction and also after the end
of the reaction (on cooling of the contents). The analysis of the
liquid phase, after the reaction, for Pd content by atomic
absorption spectroscopy showed < 0.1 ppm of Pd in solution,
but the sample withdrawn under the reaction conditions showed
significant leaching of Pd (1.5 ppm Pd). This indicates that the
reaction is more likely to be homogeneously catalysed and the
enhanced activity and selectivity is due to the combination of Pd
with PPh3 and promoters like TsOH, and LiCl. At the same
time, readsorption of Pd onto the support after the reaction
facilitates repeated recycling. Thus, the proposed catalyst
system offers significant advantages, not only with respect to
activity and selectivity, but also to catalyst recycling.
Intermediate sampling for the carbonylation of p-IBPE
showed that the reaction proceeds through the formation of
4-isobutylstyrene and 1-(4-isobutylphenyl)ethyl chloride as the
intermediates. Here the question was, which intermediate is
undergoing carbonylation? To make this point clear, the
carbonylation of 4-isobutylstyrene was performed in the
absence of LiCl. The reaction was found to be very slow (TOF
= 300 h21) with very poor Ibuprofen selectivity (65%). But in
H+
H+/Cl–
OH
Cl
R
R
R
–H2O
2
3
1
Pd-Support/PPh3
CO/H2O
CO2H
Bui
a R =
R
4
b R = Ph
c R =
+
CO2H
R
It was important to understand whether the reaction occurs
heterogeneously or homogeneously (by soluble complexes of
Pd formed in situ). For this purpose, liquid phase reaction
MeO
5
Scheme 1
Chem. Commun., 1999, 1067–1068
1067