Angewandte
Chemie
cocatalysts, such as Na2CO3, K2CO3, and K3PO4, to increase
catalytic activity and neutralize acidic byproducts (HCl, HBr
or HI).[1] In the absence of such bases, the homogeneous
Suzuki reaction mediated by Pd(NO3)2·xH2O resulted in a
catalytic conversion no greater than 7%. The marked differ-
ence in the catalytic activities of Pd(NO3)2·xH2O and the
Pd2+-exchanged zeolites indicates that, in the case of the
zeolite catalysts, the catalytic reaction occurred due to the
intrinsic basicity of Al-rich zeolite frameworks.[7,27]
Based on these results, we propose that the catalytic
reaction is carried out by a molecular Pd0 species that is
generated in situ under the reaction conditions. The Pd0
species is immediately re-oxidized into PdII by O2, before
agglomeration to Pd0 nanoparticles can take place
(Scheme 1). According to this mechanism, O2 in the reaction
In the absence of external base, the mesoporous zeolites
exhibited high reusability under the reaction conditions with
an aryl halide/Pd molar ratio of 100:1. The initial activity was
completely maintained during five recycling experiments (see
the Supporting Information, Figure S2), and no leaching of Pd
was detected by inductively coupled plasma (ICP) emission
spectroscopy. With higher aryl halide/Pd molar ratios, how-
ever, excessive generation of acid byproducts (HCl, HBr and
HI) caused the destruction of zeolite frameworks, leading to a
significant loss of catalytic activity and reusability. The
destruction of zeolite frameworks could be prevented by
addition of an external base. For example, the mesoporous
zeolites were highly active even at an aryl halide/Pd molar
ratio of 1000:1, when Na2CO3 was added as an external base
(Table 1, entry 2). The catalyst also completely maintained
catalytic activity during five recycling experiments.
The presence of O2 in the reaction media played a crucial
role in catalyst reusability. When the Suzuki reaction (Table 1,
entry 3) was carried out under an N2 atmosphere instead of
air, catalytic conversion gradually decreased to 52% after the
5th cycle (91% in the first cycle).To investigate the role of O2,
the chemical state of Pd in the mesoporous sodalite was
analyzed by X-ray absorption near-edge structure (XANES)
and extended X-ray absorption fine structure spectroscopy
(EXAFS) studies, before and after the Suzuki coupling
(Table 1, entry 3) in N2 and air. The PdII zeolite showed
exactly the same XANES pattern before and after the Suzuki
reaction in air (see the Supporting Information, Figure S3),
which was similar to that of Pd(NO3)2·xH2O. According to the
EXAFS analysis, no peaks corresponding to the nearest Pd–
Pd distance in the metallic state (approximately 2.8 ꢀ) were
detected in the radial distribution (see the Supporting
Information, Figure S3). This result indicated that the Pd2+
ions remained highly stabilized against metal reduction when
exchanged onto the zeolitic framework, without agglomer-
ation to Pd0 nanoparticles. This observation is remarkable,
whereas Pd0 nanoparticles or Pd black (bigger agglomerates)
were often detected after reactions in both homogeneous and
heterogeneous systems.[6,9,14–18] Contrary to the result in air,
the reaction in N2 produced Pd0 nanoparticles. The XANES
patterns indicated a change from PdII to Pd0 states during the
reaction, and a distinct peak corresponding to the Pd–Pd
distance occurred at 0.28 nm in the radial distribution
function. The results indicated that O2 in air can effectively
suppress the reduction of PdII and subsequent agglomeration
into Pd0 nanoparticles. Bubbling of O2 in the reaction mixture
also stabilized the PdII state during the Suzuki reaction
(Table 1, entry 3), although the catalytic conversion (34%)
decreased compared to the reaction in air (95%).
Scheme 1. Proposed reaction mechanism for Pd2+-exchanged zeolite
catalysts under aerobic atmosphere.
medium prevents the formation of less reactive Pd0 nano-
particles by shifting the equilibrium toward PdII. The equi-
librium seems to be highly shifted toward the PdII state as a
result of the strong electrostatic stabilization of Pd2+ ions by
the negatively charged zeolite surface. However, an exces-
sively high concentration of O2 (or O2 pressure in the gas
phase) can suppress the in situ generation of the molecular
Pd0 species, causing decreased activity.
In this mechanistic consideration, the solubility of the
in situ-generated molecular Pd0 species is a very important
issue. Zeolites have long been used by Djakovitch and
Koehler for aryl-coupling reactions.[6] In many cases, although
Pd is initially supported on the zeolite, it is not clear whether
the Pd was leached to solution so that the reactions were
catalyzed homogeneously by the dissolved Pd species.[6–9]
Traces of soluble Pd were detected in a sufficient concen-
tration to afford catalytic activity. Conversely, the present Pd
catalyst seemed to catalyze the coupling reactions in a truly
heterogeneous manner. This conclusion was based on the
following observations. 1) During the reaction, we removed
the Pd-exchanged zeolite by microfiltration and allowed the
reaction to continue. The conversion stopped increasing
immediately (see the Supporting Information, Figure S4). If
a trace amount of soluble Pd species had been the active
species, the catalytic conversion should have increased
further, even after the removal of the zeolite catalyst. 2) We
tested the coupling reactions in the presence of poly(4-
vinylpyridine),[28] which is well known as a solid poison to trap
homogeneous Pd species in the solution phase through
chelation. However, no significant change in catalytic activity
was detected (see the Supporting Information, Figure S4).
3) Mesoporous zeolites exhibited much higher activity than
the solely microporous NaX zeolite catalyst, especially for
reactions involving bulky molecular species. If the reactions
Angew. Chem. Int. Ed. 2009, 48, 3673 –3676
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3675