Metal-doped epoxy resins: Easily accessible, durable, and highly versatile catalysts

Article abstract of DOI:10.1016/j.jcat.2008.02.003

Thermosetting epoxy resins, such as the triglycidyl derivative of 4-aminophenol, can be polymerized using molybdenum, palladium, or rhodium complexes as initiators. The resulting metal-doped materials proved to be efficient catalysts for epoxidation, carbon-carbon coupling, hydrogenation, and hydroformylation reactions. Propylene oxide yields of around 50% were obtained in the epoxidation of propylene with tert-butyl hydroperoxide as the oxidant, and biphenyl yields of ≥98% were observed in the Suzuki coupling of iodobenzene with phenylboronic acid. Almost quantitative conversions were accomplished in the hydrogenation of ethyl crotonate, ethyl cinnamate, and croton aldehyde, whereas aldehyde yields of around 16% were obtained in the hydroformylation of 1-octene. Organic-inorganic hybrid catalysts can be obtained in a convenient one-step procedure by the addition of inorganic components to the liquid resins and subsequent polymerization. Different metal species can be combined in one thermoset matrix, affording multifunctional catalysts that can be us in various catalytic liquid-phase transformations. Quantification of metal traces in the reaction mixtures by metal enrichment and atomic spectroscopy demonstrated very low metal losses of the catalysts. The catalysts can be simply recovered by filtration and reused without reconditioning while maintaining stability, activity, and selectivity.

Full text of DOI:10.1016/j.jcat.2008.02.003

Journal of Catalysis 255 (2008) 180–189
www.elsevier.com/locate/jcat
Metal-doped epoxy resins:
Easily accessible, durable, and highly versatile catalysts
Johannes Artner, Harald Bautz, Fengwen Fan, Wilhelm Habicht, Olaf Walter,
∗
Manfred Döring, Ulrich Arnold
ITC-CPV, Forschungszentrum Karlsruhe GmbH, Hermann-von-Helmholtz-Platz 1, D-76344 Eggenstein-Leopoldshafen, Germany
Received 21 December 2007; revised 1 February 2008; accepted 1 February 2008
Available online 10 March 2008
Abstract
Thermosetting epoxy resins, such as the triglycidyl derivative of 4-aminophenol, can be polymerized using molybdenum, palladium, or rhodium
complexes as initiators. The resulting metal-doped materials proved to be efficient catalysts for epoxidation, carbon–carbon coupling, hydro-
genation, and hydroformylation reactions. Propylene oxide yields of around 50% were obtained in the epoxidation of propylene with tert-butyl
hydroperoxide as the oxidant, and biphenyl yields of ꢀ98% were observed in the Suzuki coupling of iodobenzene with phenylboronic acid.
Almost quantitative conversions were accomplished in the hydrogenation of ethyl crotonate, ethyl cinnamate, and croton aldehyde, whereas alde-
hyde yields of around 16% were obtained in the hydroformylation of 1-octene. Organic–inorganic hybrid catalysts can be obtained in a convenient
one-step procedure by the addition of inorganic components to the liquid resins and subsequent polymerization. Different metal species can be
combined in one thermoset matrix, affording multifunctional catalysts that can be us in various catalytic liquid-phase transformations. Quantifi-
cation of metal traces in the reaction mixtures by metal enrichment and atomic spectroscopy demonstrated very low metal losses of the catalysts.
The catalysts can be simply recovered by filtration and reused without reconditioning while maintaining stability, activity, and selectivity.
©
2008 Elsevier Inc. All rights reserved.
Keywords: Epoxidation; Carbon–carbon coupling; Hydrogenation; Hydroformylation; Molybdenum; Palladium; Rhodium; Epoxy resins; Polymer-supported
catalysts; Catalyst recycling
1
. Introduction
ing simultaneously as precursors for catalytically active species
in the resulting polymers [4]. Innumerable monomers and a
series of polymerization techniques afford a wide variety of
catalyst systems applicable in a series of catalytic transforma-
tions, and current research focuses on enantioselective catalysis
using polymer-bound catalysts [5–7], soluble supports [8], mi-
croencapsulated catalysts [9–11], or hybrid organic–inorganic
catalysts [12].
But academic research frequently concentrates on mecha-
nistic investigations, such as the question of metal leaching, the
distinction between heterogeneous and homogeneous catalysis,
and the development of “truly heterogeneous catalysts” [13],
neglecting technological, ecological, and economic criteria,
such as costs or long-term applicability (e.g., in continuously
operating processes). In fact, almost no data on the long-term
performance of polymer-based catalysts exceeding hours or a
few days are available. Keep in mind that a “truly heteroge-
neous catalyst” that exhibits a perfect initial performance can
Modification of organic polymers with catalytically active
species represents one of the most versatile strategies for the
development of efficient catalyst systems that can be easily
separated from reaction mixtures, thus facilitating recycling.
Catalysts can be obtained by attachment of metal species to
already polymerized materials via several established immobi-
lization techniques [1]. Catalytically active species also can be
incorporated in polymer matrices during polymer growth by,
for example, (co)polymerization of metal complexes bearing
ligands with polymerizable moieties [2] or polycondensation
in the presence of metal complexes [3]. In a recent approach,
metal complexes were used as polymerization initiators serv-
*
Corresponding author. Fax: +49 7247 822244.
E-mail address: ulrich.arnold@itc-cpv.fzk.de (U. Arnold).
0021-9517/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2008.02.003

J. Artner et al. / Journal of Catalysis 255 (2008) 180–189
181
deactivate after a few reaction cycles by poisoning, leaching,
or decomposition, whereas a catalyst that exhibits high ini-
tial metal leaching can reach a steady-state performance af-
ter several cycles with negligible metal loss while maintaining
high catalytic activity. Furthermore, heterogeneous and homo-
geneous catalysis can proceed simultaneously, interfering with
each other, and leached metal traces in the nanogram range,
detectable only by accurate metal enrichment techniques, can
contribute significantly to the catalytic activity.
Epoxy resins are widely used high-performance thermosets
with numerous applications such as coatings, composites, or ad-
hesives [14–17]. Monomeric epoxy resins can be polymerized
using metal complexes as polymerization initiators [18–22].
The complexes release ligands (i.e., Lewis bases) on heating,
thus initiating epoxy ring opening, followed by anionic poly-
merization. According to this concept, a wide variety of metal
species can be incorporated in the network of crosslinked epoxy
resins.
Recently, we reported on the use of metal-doped epoxy
resins as catalysts [4,23] obtainable in a convenient time-
and cost-savings one-step procedure using metal complexes as
polymerization initiators and precursors for catalytically ac-
tive species in the polymerized materials. Molybdenum-doped
resins were used as epoxidation catalysts, and mechanistic stud-
ies revealed a superposition of heterogeneous and homoge-
neous catalysis. Compared with other metal catalysts based
on organic polymers [1,10,24–33], unprecedented long-term
activities over periods of months were observed. To demon-
strate the versatility and applicability of this new type of cat-
alyst, we report on a broad extension of this concept. Epoxy
resins doped with molybdenum, palladium, or rhodium species
were prepared and used as recyclable catalysts for epoxidation,
carbon–carbon coupling, hydrogenation, and hydroformylation
reactions.
calorimetry (DSC) data for the polymerization reactions were
e
recorded using a Mettler Toledo DSC822 device with a heat-
◦
−1
ing rate of 10 C min . Measurements were carried out under
nitrogen atmosphere typically using 3–10 mg resin samples.
Polymerized resins were pulverized in a cryogenic grinding
mill at liquid nitrogen temperature. Surface areas were deter-
mined using a Quantachrome NOVA 2000 gas sorption ana-
lyzer by adsorption–desorption of nitrogen at 77 K. Thermo-
gravimetric analyses (TGA) of the catalysts were carried out
e
using a Mettler Toledo TGA/SDTA851 instrument with a heat-
◦
−1
ing rate of 10 C min under air and nitrogen flow. Catalyst
samples of 2–6 mg were used.
Conversions and selectivities were analyzed by GC-FID us-
ing an Agilent 6890N gas chromatograph equipped with a Phe-
nomenex ZB-1 column (60 m × 0.32 mm; 1 µm film thick-
ness). Products were quantified using calibration curves ob-
tained with standard solutions and dodecane as an external
standard. Unknown products were identified by GC-MS us-
ing an Agilent 6890N instrument (J & W Scientific DB5 col-
umn; 30 m × 0.25 mm; 0.25 µm film thickness), coupled with
an Agilent 5973 mass selective detector. In the epoxidation of
propylene, dodecane was added as internal standard to the re-
action mixtures. Propylene oxide yields were based on TBHP
consumption. Yields were determined by TBHP titration and
GC-FID.
Quantifications of metal traces in the reaction solutions were
carried out by inductively coupled plasma-atomic emission
spectroscopy (ICP-AES) and atomic absorption spectroscopy
(AAS) analyses on a Varian Liberty 150 and a Varian Spec-
trAA 800 instrument, respectively. Metals were concentrated
before analysis by evaporation of the solvents and dissolving
of the reaction residues in concentrated HNO . In the case of
3
C–C-coupling reactions, the palladium content of aqueous and
organic phases was determined after metal enrichment in the
reaction residues.
2
. Experimental
®
Rotilabo syringe filters (PTFE, 0.45 µm pore width) were
2
.1. General
used for catalyst separation. In the case of C–C-coupling re-
actions, the catalyst was recovered by filtration of the reaction
mixtures through Whatman GF/C glass fiber paper. In recycling
tests, the catalysts were reused without any reconditioning un-
less stated otherwise.
The epoxy resin N,N-diglycidyl-4-glycidyloxyaniline
(
TGAP) was purchased from Aldrich. Mo(OEt)5 and Pd(PPh3)4
were obtained from Gelest and Strem, respectively. The rhodium
complex Rh(C8H12)[(iPr)2PC6H4CH(OMe)2]Cl was prepared
as described previously [34]. Silicagel type 62 with particle
sizes of 75–250 µm (Aldrich) was used for the preparation of
the resin–silicagel hybrid catalysts. Propylene 2.5, synthesis gas
2.2. Catalyst preparation
(
CO 4.7, H2 6.0, 49.6 vol% CO), and hydrogen 5.0 were ob-
2
.2.1. Preparation of TGAP62.0%/SiO236.4%–Mo0.5%
The epoxy resin TGAP (9.998 g), the initiator Mo(OEt)5
tained from Messer Griesheim, Air Liquide, and Basi, respec-
tively. Anhydrous tert-butyl hydroperoxide (TBHP) in toluene
was prepared by azeotropic drying of TBHP (T-HYDRO so-
lution from Aldrich, 70 wt% TBHP in water) [35] and TBHP
concentrations were determined by iodometric titration [36].
All other substrates, reagents, and solvents were purchased
from Aldrich and used as received.
SEM images were recorded on a Leo 982 digital scan-
ning microscope combined with an Oxford Instruments ISIS
®
(0.265 g), and silicagel (5.860 g) were mixed vigorously at
5
minum mold and polymerized in an oven by raising the temper-
ature successively from 120 to 230 C (120 C for 1 h, 180 C
for 3 h, 200 C for 2 h and 230 C for 1 h). The resulting brittle
material was pulverized manually in a mortar and annealed at
230 C for 4 h. The powder was sieved, and two fractions with
◦
0 C for 15 min. The formulation was transferred to an alu-
◦
◦
◦
◦
◦
◦
3
00 energy-dispersive X-ray (EDX) unit. Differential scanning
particle diameters of 20–150 and 150–500 µm were obtained.

182
J. Artner et al. / Journal of Catalysis 255 (2008) 180–189
2
.2.2. Preparation of TGAP48.3%/SiO250.0%–Mo0.5%
for TGAP48.3%/SiO250.0%–Mo0.5% and TGAP46.7%/SiO250.0%
–
1.0%
A mixture of TGAP (10.000 g), Mo(OEt)5 (0.347 g), and sil-
Mo
: TBHP (50 mmol, 34 wt% in toluene/dodecane cor-
◦
icagel (10.347 g) was stirred vigorously at 50 C for 15 min.
The resulting powder was transferred to an aluminum mold,
and polymerization was carried out in an oven by raising the
temperature successively from 120 to 220 C (120 C for 1 h,
1
2
responding to ca. 15 ml), and catalyst (2 g; particle sizes:
150–300 µm); reaction conditions for TGAP–Mo1 /Pd
.4%
0.5%
:
TBHP (50 mmol, 36 wt% in toluene/dodecane corresponding
to ca. 14 ml), and catalyst (1 g; particle sizes: 150–300 µm).
After replacing the air and saturating the reaction solution with
propylene, a pressure of 8 bar was adjusted. The mixture was
◦
◦
◦ ◦ ◦
40 C for 30 min, 160 C for 30 min, 180 C for 30 min,
◦
◦
00 C for 30 min, and 220 C for 1 h). The powder was an-
◦
◦
nealed at 220 C for 4 h and sieved. Two fractions with particle
diameters of 20–150 and 150–300 µm were obtained.
magnetically stirred for 24 h at 90 C, with an operating pres-
sure of ca. 25 bar. Cylindrical 20 × 6 mm stirring bars were
used at 200 rpm. After cooling to room temperature and expan-
sion of the system, the catalyst was separated by filtration and
reused without reconditioning.
2
.2.3. Preparation of TGAP46.7%/SiO250.0%–Mo1.0%
A mixture of TGAP (10.000 g), Mo(OEt)5 (0.718 g), and sil-
icagel (10.718 g) was homogenized and processed as described
above for the system TGAP4
8.3%
/SiO2
50.0%
–Mo0.5%.
2
.3.2. Suzuki coupling of iodobenzenes with phenylboronic
acid
Iodobenzene (5 mmol), phenylboronic acid (6 mmol), and
K2CO3 (20 mmol) were dissolved in a mixture of dioxane
2
.2.4. Preparation of TGAP–Pd1.0%
The epoxy resin TGAP (8.2 g) was heated to 50 C, and
Pd(PPh3)4 (1.0 g) was added under vigorous stirring. After
◦
1.0%
(
10 ml) and water (10 ml). The catalyst (0.5 g; TGAP–Pd
3
0 min, the temperature was raised, and the mixture was stirred
1.4%
/Pd0.5%
with particle sizes of 150–300 µm or TGAP–Mo
with particle sizes of 20–150 µm) was added, and the mixture
was magnetically stirred for 24 h at 90 C. After cooling to
room temperature, Et2O (10 ml) was added. The catalyst was
recovered by filtration and reused in the following reactions
without reconditioning. The two phases of the filtrate were sep-
arated, and samples for ICP-AES, AAS, and GC analysis were
taken.
◦
for another 10 min at 80 C. The resin was cast into an alu-
minum mold, and a thin (ca. 1 mm) layer was polymerized in
an oven by raising the temperature successively from 120 to
◦
◦ ◦ ◦ ◦
2
00 C (120 C for 1 h, 180 C for 1 h and 200 C for 4 h).
The polymer was cut, milled, sieved, and annealed for 4 h at
00 C. Fractions with particle diameters of 20–150, 150–300,
and 300–500 µm were obtained.
◦
2
2
.2.5. Preparation of TGAP–Rh1.0%
2
.3.3. Hydrogenation reactions
A 250-ml glass autoclave was charged with the substrate (10
The rhodium complex Rh(C8H12)[(iPr)2PC6H4CH(OMe)2]-
◦
Cl (0.537 g) was dissolved at 50 C in TGAP (10.214 g), and
a thin (ca. 1 mm) layer of the formulation was polymerized in
an aluminum mold at 100 C for 1 h, 120 C for 1 h, 140 C for
mmol), 20 ml of solvent (MeOH in the case of ethyl croto-
nate and ethyl cinnamate; toluene in the case of crotonaldehyde
◦
◦
◦
1
.0%
◦ ◦
and cinnamaldehyde), and the catalyst (0.5 g; TGAP–Pd
or
1
h, 160 C for 1 h, and 180 C for 1 h. After demolding, the
TGAP–Mo1 /Pd
.4%
0.5%
with particle sizes of 150–300 µm). Af-
resin plate was crushed and milled. Particle sizes were adjusted
to 20–150 µm by sieving the polymer powder.
ter replacing the air and saturating the reaction solution with
hydrogen, the pressure was raised to 2.5 bar. The mixture was
heated to 60 C and magnetically stirred for 4 h (for ethyl cro-
tonate and ethyl cinnamate) or 24 h (for crotonaldehyde and
cinnamaldehyde). After cooling to room temperature and ex-
pansion of the system, the catalyst was separated by filtration
and reused without reconditioning.
◦
.2.6. Preparation of TGAP–Mo1.4%/Pd0.5%
A mixture of Mo(OEt)5 (0.427 g), Pd(PPh3)4 (0.488 g), and
2
TGAP (8.070 g) was homogenized by ultrasonic treatment and
transferred to an aluminum mold. The formulation was cured
by raising the temperature successively from 120 to 200 C
◦
◦
◦
◦
◦
(
120 C for 1 h, 140 C for 1 h, 160 C for 3 h, 180 C for 1 h,
◦
and 200 C for 1 h). The resulting resin plate (thickness of ap-
proximately 1 mm) was crushed and milled, and the polymer
powder was heated to 200 C for 2 h. Particle diameters of 20–
1
2
2
2.3.4. Hydroformylation of 1-octene
A 300-ml steel autoclave with an integrated gas injec-
◦
tion stirrer and a thermocouple was charged with 1-octene
(0.624 mol) and the catalyst TGAP–Rh1
.0%
(5 g with particle
50, 150–300, and 300–500 µm were adjusted by sieving.
sizes of 20–150 µm). After replacing the air and saturating the
reaction mixture with synthesis gas, the temperature of 110 C
◦
.3. Catalytic procedures
was adjusted, and the pressure was raised to 50 bar. The reaction
was run for 8 h with the impeller operating at 580 rpm, and the
pressure was kept constant at 50 bar by an automated gas doser
(Brooks 5866 pressure controller and Bronkhorst F-231C-FD-
33-V flow regulator). After cooling to room temperature, the
system was purged with nitrogen and the catalyst recovered by
.3.1. Epoxidation of propylene
A 80-ml steel autoclave was charged with TBHP (an-
hydrous solution in toluene), dodecane as an internal stan-
dard for GC analysis and the catalyst. Reaction conditions
for TGAP6
2.0%
/SiO2
36.4%
–Mo
0.5%
: TBHP (50 mmol, 31 wt%
◦
in toluene/dodecane corresponding to ca. 16 ml) and cat-
alyst (1 g; particle sizes: 20–150 µm); reaction conditions
filtration. The catalyst was heated to 160 C for 2 h and used
in a second reaction under the same reaction conditions. In the

J. Artner et al. / Journal of Catalysis 255 (2008) 180–189
183
Scheme 1. Preparation of metal-doped epoxy resins by polymerization of TGAP.
subsequent reactions, the catalyst was reused without thermal
treatment.
material was obtained that could be easily pulverized. Increas-
ing the silicagel content to 50%, catalyst powders were obtained
directly after polymerization, and milling of the thermosets
could be circumvented. A bimetallic system comprising molyb-
2
.3.5. Heck coupling of iodobenzene with ethyl acrylate
denum and palladium labeled as TGAP–Mo1 /Pd
.4%
0.5%
was
Iodobenzene (50 mmol), ethyl acrylate (75 mmol), and
also prepared using a mixture of Mo(OEt)5 and Pd(PPh3)4 as
an initiator.
K2CO3 (100 mmol) were dissolved in N-methylpyrrolidone
1.4%
0.5%
(
50 ml). The catalyst TGAP–Mo
/Pd
(0.5 g with particle
Polymerization reactions were investigated by differential
scanning calorimetry (DSC); strongly exothermic reactions
size 150–300 µm) was added, and the mixture was magnetically
◦
stirred for 24 h at 120 C. After cooling to room temperature,
−
1
with enthalpies >560 J g indicate a high cross-linkage in
the polymers (Fig. 1 and Table 1). In the case of molybde-
num catalysts, neat resins without silicagel were investigated to
the solids were removed by filtration. The catalyst was recov-
ered by washing the solids with H2O and Et2O and reused in
the subsequent reactions without reconditioning. Volatile com-
ponents of the filtrate were removed under reduced pressure,
and the residue was extracted with 100 ml of Et2O three times.
The combined organic layers were washed with 100 ml of H2O
three times and dried over Na2SO4. The solvent was removed
under reduced pressure, and the purity of ethyl cinnamate was
controlled by NMR.
0
.77%
ensure comparability to other catalysts. Thus, TGAP–Mo
,
TGAP–Mo1 , and TGAP–Mo
.0%
2.0%
were the resin components
62.0%
36.4%
of the corresponding hybrid catalysts TGAP
/SiO2
, and TGAP46.7%/
. The onset and peak temperatures of the
–
0.5%
48.3%
50.0%
0.5%
Mo
SiO25
,
TGAP
/SiO2
–Mo
0.0%
1.0%
–Mo
polymerization reactions depend on the amount of Mo(OEt)5 in
the mixture and decrease with increasing amounts of the metal
compound (Fig. 1a and entries 1–3 in Table 1). Two separated
3
. Results and discussion
peaks were observed for TGAP–Pd1
.0%
(Fig. 1b and entry 4 in
◦
Table 1); the less intense peak at 182 C is likely due to decom-
position of Pd(PPh3)4 and/or reaction with the resin before initi-
3
.1. Catalyst preparation and characterization
1.0%
ation of polymerization. The reaction of TGAP–Rh
(Fig. 1b
Several metal-doped resins based on the triglycidyl deriv-
and entry 5 in Table 1) shows three overlapping processes. It is
ative of 4-aminophenol (TGAP) were prepared in a one-step
procedure using Mo(OEt)5, Pd(PPh3)4 and Rh(C8H12)[(iPr)2P-
C6H4CH(OMe)2]Cl as polymerization initiators (Scheme 1).
Metal contents of 0.5, 1.0, or 1.4% were adjusted, and the
mixtures were polymerized at temperatures between 100 and
2
◦
strongly exothermic with a high onset temperature of 177 C.
1.4%
0.5%
A comparison of the bimetallic system TGAP–Mo
/Pd
1.0%
(
Fig. 1b and entry 6 in Table 1) with TGAP–Mo
, TGAP–
shows that the polymerization was
2.0%
1.0%
Mo
, and TGAP–Pd
initiated by Mo(OEt)5 and an onset temperature similar to that
◦
seen for TGAP–Mo1
.0%
was detected.
30 C. The resulting resin plates were milled and sieved. Frac-
tions with particle diameters of 20–300 µm were used for the
catalytic reactions. As shown previously [4,23], catalytic per-
formances depend significantly on catalyst particle sizes, and
catalysts with particle diameters in this range exhibit an opti-
mum balance between low metal leaching and high catalytic
activity. Organic–inorganic hybrid catalysts containing 0.5 and
Catalyst surfaces were investigated by scanning electron mi-
croscopy coupled with energy-dispersive X-ray spectroscopy
1.0%
(SEM-EDX). The surfaces of TGAP–Pd
and TGAP–
1.0%
Rh
appear to be homogeneous (Figs. 2a and 2b), with the
latter showing a uniform metal distribution in the resin ma-
trix with the expected rhodium content of around 1.0 wt%.
In the case of the palladium catalyst, a few palladium con-
glomerates with metal content up to 61 wt% were detected
along with homogeneously dispersed palladium, reflecting the
significantly lower resin solubility of the palladium complex
compared with the rhodium complex. Bright areas and spots
62.0%
36.4%
1
.0% of molybdenum and designated TGAP
/SiO2
, and TGAP46.7%/
were prepared by modification of moly-
bdenum-doped TGAP with 36.4 and 50.0% of silicagel, respec-
–
0.5%
48.3%
50.0%
0.5%
Mo
,
TGAP
–Mo
/SiO2
–Mo
SiO25
0.0%
1.0%
tively. In the case of TGAP6
2.0%
/SiO2
36.4%
–Mo
0.5%
, a brittle

184
J. Artner et al. / Journal of Catalysis 255 (2008) 180–189
Mo1.4%/Pd0.5% with the expected average palladium content of
around 0.5 wt%.
The catalysts are highly cross-linked, nonporous thermosets
that swell slightly in the presence of organic solvents, resulting
in a weight increase of a few percent. Surface area analysis by
nitrogen adsorption–desorption at 77 K revealed low surface ar-
2
−1
eas (around 1 m g ) for catalysts based on neat TGAP. But the
surface areas could be increased significantly by the addition of
2
−1
silicagel; a surface area of 87 m g was found for the organic–
inorganic hybrid system TGAP4
6.7%
/SiO2
50.0%
–Mo
1.0%
. Ther-
mogravimetric analyses (TGA) of the catalyst systems demon-
strated high thermal stability and decomposition starting at
◦
around 300 C regardless of whether the measurement was
carried out under air or nitrogen flow. An initial weight loss
of around 2% was observed up to 150 C, likely due to
(a)
◦
the release of adsorbed water. The silica-modified catalysts
6
2.0%
36.4%
0.5%
46.7%
50.0%
TGAP
Mo
/SiO2
exhibited significantly higher initial weight loss of
around 4% and 5%, respectively.
–Mo
and TGAP
/SiO2
–
1
.0%
3
.2. Catalytic performance of metal-doped epoxy resins
The hybrid catalysts TGAP62.0%/SiO 36.4%–Mo0.5%
,
2
48.3%
/SiO250.0%–Mo
0.5%
46.7%
/SiO250.0%–
TGAP
, and TGAP
1
.0%
Mo
were tested in the epoxidation of propylene with tert-
butyl hydroperoxide (TBHP) as an oxidant (Fig. 3). The cat-
alysts were used in 10 consecutive reactions with no recon-
ditioning. Catalyst recycling was carried out by filtration, and
metal leaching was determined using highly sensitive atomic
emission and atomic absorption spectroscopic techniques af-
ter removal of volatile compounds and metal enrichment in
(
b)
Fig. 1. DSC curves of TGAP containing (a) different amounts of Mo(OEt)
2.0, 1.0 and 0.77% of molybdenum) and (b) TGAP containing a mix-
ture of Mo(OEt) and Pd(PPh ) , Rh(C H )[(iPr) PC H CH(OMe) ]Cl and
5
(
62.0%
36.4%
0.5%
the reaction residues. Using TGAP
propylene oxide yield increased from 35% in the first run to
2% in the second run; however, yields decreased in the sub-
/SiO2
–Mo
,
5
3 4
8
12
2
6
4
2
Pd(PPh ) .
3
4
6
sequent reactions to values of around 25% from run 8 on.
Metal leaching was very low, and the average value per
run was 0.02% of molybdenum initially loaded on the poly-
mer corresponding to 0.1 ppm in the reaction mixture. Us-
Table 1
DSC-data of metal-doped resins
a
b
c
ꢀH
Entry
Metal-doped resin
T
T
Onset
C)
Peak
◦
◦
−1
(
( C)
(J g )
ing TGAP4
8.3%
/SiO2
50.0%
–Mo
0.5%
, a catalyst with the same
0
1
2
.77%
1
2
3
4
5
6
a
TGAP–Mo
TGAP–Mo
TGAP–Mo
237
136
122
257
242
756
567
638
116, 881
1074
molybdenum loading of 0.5% but a higher silicagel content,
a comparatively constant performance was observed through-
.0%
.0%
179, 262
182, 311
196, 258, 297
199
−1
1.0%
out the 10 reactions. Metal leaching was around 0.06% run
TGAP–Pd
159, 289
177
1
.0%
TGAP–Rh
TGAP–Mo
(0.5 ppm), and epoxide yields of around 28% were obtained.
Using the catalyst with a double molybdenum loading of 1.0%,
a steady-state performance with epoxide yields of around 50%
was observed from run 3 on after catalyst activation in the
first two reactions. Metal leaching was slightly higher with
1
.4%
0.5%
/Pd
134
656
Onset temperature of reaction peak.
Peak temperature.
Polymerization enthalpy.
b
c
−
1
an average metal loss of 0.11% run (1.7 ppm). No byprod-
ucts were observed in any of the reactions, and nonproductive
TBHP decomposition, monitored by iodometric titration, was
were visible in the case of the molybdenum-containing catalysts
46.7%
50.0%
1.0%
1.4%
0.5%
TGAP
Figs. 2c and 2d), respectively. Molybdenum contents up to
2% were determined in these areas by EDX, as described
/SiO2
–Mo
and TGAP–Mo
/Pd
−
1
negligible (<1.5% run ). Blank experiments with metal-free
TGAP/SiO2 materials under the same reaction conditions gave
no propylene conversions, and thus a contribution of the cata-
lyst matrix to the catalytic activity can be excluded.
(
4
previously for similar catalyst systems and these agglomer-
ates are due to the formation of molybdenum oxo clusters [4,
So far, only a few catalysts based on organic polymers,
such as molybdenum compounds supported on benzimidazole,
polystyrene, or poly(glycidyl methacrylate) [37] and micelle-
2
3]. Variable palladium contents between 0 and 1.0 wt% were
detected in the resin matrix of the bimetallic catalyst TGAP–

J. Artner et al. / Journal of Catalysis 255 (2008) 180–189
185
(a)
(b)
(c)
(d)
1.0%
1.0%
46.7%
50.0%
1.0%
1.4%
0.5%
Fig. 2. SEM images of (a) TGAP–Pd
, (b) TGAP–Rh
, (c) TGAP
/SiO
–Mo
, and (d) TGAP–Mo
/Pd
.
2
incorporated manganese porphyrin catalysts [38], have been
tested in the epoxidation of propylene. The former afforded re-
markably high propylene oxide yields up to 99.8%, but were
tested for a few hours only.
deposit on catalyst particles on termination of the reaction and
cooling of the reaction mixture to ambient temperature [43].
Considering polymer-supported catalysts for Suzuki cou-
pling, significant progress has been made recently through mi-
croencapsulation of catalytically active metal species. Microen-
capsulated palladium catalysts, so-called “polymer-incarcerated
palladium” catalysts, were prepared using glycidyl-functional-
ized copolymers as a matrix [44]. The catalysts demonstrated
high activity in the coupling of bromobenzenes with boronic
acids and could be recycled [45]. However, they had to be
synthesized in a multistep procedure, including polymer prepa-
ration by radical copolymerization, and, in most cases, a high
catalyst loading of 5 mol% was employed in the coupling
reactions. In another approach, a polyurea–encapsulated pal-
ladium catalyst, designated PdEnCat™, has been successfully
tested as a catalyst for Suzuki cross-coupling reactions [11], and
high yields were achieved in reactions of bromobenzenes with
boronic acids under batch and continuous flow conditions [46].
Using the palladium-doped system TGAP–Pd1.0% as a cat-
alyst for the Suzuki coupling of iodobenzene with phenyl-
boronic acid (Table 2), biphenyl yields of ꢀ98% with very
low palladium leaching (<0.4 ppm in the reaction mixture)
were obtained. Metal loss decreased significantly, from 0.21%
in the first to 0.08% in the fifth run, while catalytic activity
remained unchanged. Control experiments under the same re-
action conditions with substituted iodobenzenes, such as 2- and
4
-iodoanisole, 4-iodotoluene, and 4-iodoacetophenone, con-
firmed cross-coupling, and no homocoupling was observed.
Quantitative conversions were obtained using these substrates.
Significantly lower yields of 32 and 22% were observed using
bromobenzene and chlorobenzene as substrates.
Much research has been devoted to elucidating the nature
of the active species in palladium-catalyzed coupling reac-
tions [39–41]; recent developments have been highlighted pre-
viously [42]. In this context, several control experiments have
been discussed to distinguish between homogeneous cataly-
sis due to leached metal species and heterogeneous catalysis.
With respect to TGAP–Pd1 , hot filtration tests, mercury and
poly(vinyl pyridine) poisoning tests, and redeposition tests have
strongly suggested that catalysis is to a large extent due to
leached metal species exhibiting a distinctive tendency to re-
Quantitative conversions of ethyl crotonate from run 4 on
were observed using TGAP–Pd1
.0%
as the hydrogenation cata-
lyst (Table 3). Almost the same results were obtained with ethyl
cinnamate and crotonaldehyde as substrates; the latter was se-
lectively reduced at the C–C-double bond. Significantly lower
yields of around 83% were obtained using this catalyst in the
hydrogenation of cinnamaldehyde, and selectivities to 3-phenyl
propionaldehyde were 100%. Palladium leaching was very low
.0%
−
1
(<0.007% run , corresponding to values <0.02 ppm in the re-

186
J. Artner et al. / Journal of Catalysis 255 (2008) 180–189
Table 2
1.0%
Catalytic performance of TGAP–Pd
with phenylboronic acid
in the Suzuki coupling of iodobenzene
a
b
Run no.
Biphenyl yield (%)
Pd leached (%)
1
2
3
4
5
98
99
99
98
98
0.21
0.14
0.17
0.09
0.08
a
Iodobenzene (5 mmol), phenylboronic acid (6 mmol), dioxane (10 ml), wa-
ter (10 ml), K CO (20 mmol), and catalyst (0.5 g; particle sizes: 150–300 µm),
2
3
◦
9
0 C, 24 h.
b
Determined by GC analysis.
ity. A comparatively high catalyst loading of 5 mol% was used,
however.
(a)
The rhodium-doped resin TGAP–Rh1.0% was tested as cat-
alyst in the hydroformylation of 1-octene (Table 4), with the
reaction carried out in the neat alkene with a reaction time of
8
h. Metal leaching was high in the first run (4.18%) and de-
creased drastically to around 0.09% (0.6 ppm) in the subsequent
reactions. The aldehyde yield was 59% in the first reaction, and
the product mixture comprised nonanal (43%), 2-methyloctanal
(
36%), 2-ethylheptanal (12%), and 2-propylhexanal (9%). Iso-
merization products of 1-octene, namely 2- and 3-octene, also
were observed. Selectivities were significantly higher in runs
2
2
to 5, and aldehyde mixtures of nonanal, 2-methyloctanal, and
-ethylheptanal in a molar ratio of 72:27:1 were obtained. Alde-
hyde yields in these reactions were around 16%.
Recent developments in the field of polymer-supported
hydroformylation catalysts include rhodium complexes at-
tached to phosphine-functionalized acetals of polyvinyl al-
cohol [34,51] or phosphine-functionalized polystyrene [52].
Almost quantitative conversions of 1-octene or 4-vinylanisole
were reported using a significantly higher catalyst loading com-
pared with the TGAP–Rh1 -catalyzed hydroformylation of
(
b)
Fig. 3. Catalytic performance of different molybdenum hybrid catalysts in the
epoxidation of propylene. (a) Propylene oxide yields and (b) metal leaching.
.0%
6
2.0%
36.4%
0.5%
Reaction conditions for TGAP
/SiO
–Mo
: propylene (8 bar),
2
1
-octene. Rhodium catalysts bound to polystyrene [53] or a flu-
TBHP (50 mmol, 31 wt% in toluene/dodecane), and catalyst (1 g; particle
oroacrylate copolymer [54] through phosphine ligands and their
use in supercritical carbon dioxide also have been described.
High activities with up to quantitative conversions of 1-hexene
or styrene were observed. Additional promising approaches
include dendrimeric phosphine ligands and their rhodium com-
plexes [55–57], asymmetric hydroformylation using highly
cross-linked polystyrene-supported BINAPHOS-Rh(I) com-
plexes [58], and rhodium catalysts supported on amphiphilic
polymers for hydroformylations in water [59,60]. Outstand-
ing results were obtained using the dendrimer-supported cat-
alysts in the hydroformylation of aryl olefins and vinylesters,
with almost quantitative conversions reached at room temper-
ature [56]. Some of these catalysts could be used in up to 10
consecutive reactions, but data on metal leaching are limited.
Compared with TGAP–Rh1 , preparation of these catalysts
is much more elaborate, starting with polymer modification by
coordinating moieties or even ab initio synthesis of appropriate
polymers or linkers between polymer and catalyst species.
4
8.3%
50.0%
0.5%
sizes: 20–150 µm); reaction conditions for TGAP
/SiO
–Mo
: propylene (8 bar), TBHP (50 mmol,
4 wt% in toluene/dodecane), and catalyst (2 g; particle sizes: 150–300 µm),
2
46.7%
50.0%
1.0%
and TGAP
/SiO
–Mo
2
3
9
◦
0 C, 24 h. Yields are based on TBHP consumption (determined by iodomet-
ric titration and GC analysis).
action mixtures) even after reaction times of 24 h, and metal
traces could not be quantified even after metal enrichment and
application of highly sensitive AES and AAS techniques.
With respect to polymer-supported hydrogenation catalysts,
the aforementioned polymer-incarcerated palladium catalysts
exhibited high activity in the hydrogenation of various olefins
and could be recycled [47,48]. Recently, this concept was sig-
nificantly improved by the use of polysilane-supported palla-
dium [49]. Almost quantitative conversions within a few hours
were observed in the hydrogenation of numerous substrates at
room temperature. The polyurea-encapsulated palladium cata-
lyst PdEnCat™ (see above) also has been tested successfully as
hydrogenation catalyst [50]. Metal leaching was extremely low,
and the catalyst could be recovered and recycled 20 times in the
hydrogenation of cyclohexene with no significant loss of activ-
.0%
The bimetallic system TGAP–Mo1 /Pd
.4%
0.5%
was tested
as an epoxidation, C–C-coupling, and hydrogenation catalyst
(Table 5). Epoxide yields between 48 and 73% were observed

J. Artner et al. / Journal of Catalysis 255 (2008) 180–189
187
Table 3
1.0%
a
Catalytic performance of TGAP–Pd
in hydrogenation reactions
Run
no.
Ethyl crotonate
Ethyl cinnamate
Crotonaldehyde
Cinnamaldehyde
b
b
b
b
Yield
Pd leached
(%)
Yield
(%)
Pd leached
(%)
Yield
Pd leached
(%)
Yield
Pd leached
(%)
(%)
(%)
(%)
1
2
3
4
5
96
98
98
>99
>99
<0.004
<0.004
<0.004
<0.003
<0.004
95
99
97
>99
>99
<0.004
<0.004
<0.004
<0.004
<0.004
>99
>99
>99
>99
>99
<0.006
<0.006
<0.006
<0.006
<0.006
88
81
85
79
81
<0.006
<0.006
<0.006
<0.006
<0.007
a
◦
Hydrogenation of ethyl crotonate and ethyl cinnamate: alkene (10 mmol), H (2.5 bar), MeOH (20 ml), and catalyst (0.5 g; particle sizes: 150–300 µm), 60 C,
2
◦
4
2
h; hydrogenation of crotonaldehyde and cinnamaldehyde: alkene (10 mmol), H (2.5 bar), toluene (20 ml), and catalyst (0.5 g; particle sizes: 150–300 µm), 60 C,
2
4 h.
b
Determined by GC analysis.
Table 4
1.0%
a
Catalytic performance of TGAP–Rh
in the hydroformylation of 1-octene
b
b
Run
no.
Aldehyde yield
(%)
Selectivity (%)
Rh leached
(%)
Nonanal
2-Methyl-
octanal
2-Ethyl-
heptanal
2-Propyl-
hexanal
1
2
3
4
5
59
16
19
12
18
43
72
73
73
71
36
27
26
26
28
12
1
1
1
1
9
0
0
0
0
4.18
0.11
0.11
0.04
0.09
a
◦
1
-Octene (624 mmol), H /CO (50 bar), and catalyst (5.0 g; particle sizes: 20–150 µm), 110 C, 8 h.
2
b
Determined by GC analysis.
in the epoxidation of propylene, with an epoxide selectivity of
metal compounds as polymerization initiators, catalytically ac-
tive metal species can be incorporated in the thermoset matrices
with no additional modification of the organic polymer. Thus,
catalysts for epoxidation, carbon–carbon coupling, hydrogena-
tion, or hydroformylation reactions can be prepared. Various
resin monomers and metal compounds can be combined, af-
fording promising potential for optimization. Multifunctional
catalysts for different catalytic reactions can be obtained by
combining several catalytically active components in one ther-
moset matrix. The catalysts demonstrated superior long-term
activity and stability compared with other systems based on
organic polymers and could be recovered and reused with no
reconditioning. Based on previously reported long-term tests,
catalyst lifetimes of months up to years can be expected. Metal
leaching is extremely low and can be controlled by choice of
the resin composition. The liquid resin systems can be easily
modified by, for example, adding inorganic components, thus
yielding organic–inorganic hybrid catalysts after polymeriza-
tion with tunable surface properties. This also makes catalyst
preparation possible on a kg scale. Reactor surfaces and porous
reactor filling materials could be coated with metal-doped resin
layers and integrated in catalytic processes after polymeriza-
tion. The applicability of the catalysts may be limited depend-
ing on the required thermal, mechanical, and chemical resis-
tance of the materials. Another constraint is that the catalyst
performance is only somewhat predictable and tunable, because
the complexes decompose at least partially under the somewhat
harsh polymerization and catalyst preparation conditions. Fur-
ther work, particularly in the characterization and optimization
of the catalyst systems, are needed.
1
0
00%. Molybdenum leaching was very low and decreased from
.14% (1.6 ppm) in the initial reaction to 0.02% (0.2 ppm)
in the fourth and fifth runs. The material also demonstrated
catalytic activity in the Suzuki coupling of iodobenzene with
phenyl boronic acid. The catalyst exhibited an initial activation
period, and the yield of coupling product increased from 89%
in the first run to around 97% in the subsequent reactions. Pal-
ladium leaching decreased significantly from 0.70% (0.7 ppm)
in the initial reaction to around 0.17% (0.2 ppm) from the third
run on. Ethyl cinnamate yields up to 82% were isolated using
this polymer as a catalyst for the Heck coupling of iodobenzene
◦
with ethyl acrylate. The reactions were carried out at 120 C
with NMP as the solvent; significant metal leaching of around
−1
2
.2% run (0.7 ppm) was observed after a metal loss of 10%
(
3.0 ppm) in the first run. This finding suggests that the resin
matrix was affected by the solvent and/or the base at a com-
◦
paratively high temperature of 120 C. In addition, the material
was tested as a hydrogenation catalyst using ethyl crotonate as
the test substrate. The alkene was quantitatively converted to
ethyl butyrate from run 4 on; palladium leaching was very low
(
<0.008%, corresponding to values <0.01 ppm in the reaction
mixtures). Similar catalytic behavior was observed using ethyl
cinnamate as the substrate.
4
. Conclusion
We have described a convenient solvent-free, time- and cost-
saving, one-step procedure for the preparation of a wide va-
riety of catalysts based on thermosetting epoxy resins. Using

188
J. Artner et al. / Journal of Catalysis 255 (2008) 180–189
Table 5
1.4%
0.5%
a
Catalytic performance of TGAP–Mo
/Pd
in epoxidation, C–C-coupling and hydrogenation reactions
b
c
Run no.
Yield (%)
Mo leached (%)
Yield (%)
Pd leached (%)
1
2
3
4
5
73
62
69
55
48
0.14
0.10
0.06
0.02
0.02
89
97
97
97
96
0.70
0.31
0.18
0.15
0.19
d
c
Run no.
Yield (%)
Pd leached (%)
Yield (%)
Pd leached (%)
1
2
3
4
5
80
78
82
64
70
9.96
2.52
2.00
2.20
2.20
97
96
98
>99
>99
<0.007
<0.007
<0.007
<0.008
<0.008
a
Epoxidation of propylene: propylene (8 bar), TBHP (50 mmol, 36 wt% in toluene/dodecane), and catalyst (1 g; particle sizes: 150–300 µm), 80 ml steel
autoclave, 25 bar operating pressure; Suzuki coupling of iodobenzene: iodobenzene (5 mmol), phenylboronic acid (6 mmol), dioxane (10 ml), water (10 ml),
K CO (20 mmol), and catalyst (0.5 g; particle sizes: 20–150 µm); Heck coupling of iodobenzene: iodobenzene (50 mmol), ethyl acrylate (75 mmol), NMP (50 ml),
2
3
K CO (100 mmol), and catalyst (0.5 g; particle sizes: 150–300 µm); hydrogenation of ethyl crotonate: ethyl crotonate (10 mmol), H (2.5 bar), MeOH (20 ml),
2
3
2
and catalyst (0.5 g; particle sizes: 150–300 µm).
b
Yields are based on TBHP consumption (determined by iodometric titration and GC analysis).
Determined by GC analysis.
Isolated yields.
c
d
Acknowledgments
[17] H. Lee, K. Neville (Eds.), Handbook of Epoxy Resins, McGraw–Hill, New
York, 1967.
[18] I. Hamerton, B.J. Howlin, P. Jepson, Coord. Chem. Rev. 224 (2002) 67.
[19] Z. Zhang, C.P. Wong, J. Appl. Polym. Sci. 86 (2002) 1572.
[20] A. Palinsky, M. Döring, H. Dey, J. Wuckelt, EP 0995766 (2000), to Bake-
lite AG.
Financial support was provided by the “Bundesministerium
für Bildung und Forschung“ (BMBF). The authors thank Hex-
ion Specialty Chemicals GmbH for supplying epoxy resins.
[21] J. Brown, I. Hamerton, B.J. Howlin, J. Appl. Polym. Sci. 75 (2000) 201.
References
[22] M. Döring, W. Ludwig, E. Uhlig, S. Wocadlo, U. Müller, Z. Anorg. Allg.
Chem. 611 (1992) 61.
[
[
[
[
[
[
[
23] U. Arnold, F. Fan, W. Habicht, M. Döring, J. Catal. 245 (2007) 55.
24] A.D. Pomogailo, Kinet. Catal. 45 (2004) 67.
25] R. Haag, S. Roller, Top. Curr. Chem. 242 (2004) 1.
26] N.E. Leadbeater, M. Marco, Chem. Rev. 102 (2002) 3217.
27] B. Clapham, T.S. Reger, K.D. Janda, Tetrahedron 57 (2001) 4637.
28] D.C. Sherrington, J. Polym. Sci. Part A Polym. Chem. 39 (2001) 2364.
29] Y.R. de Miguel, E. Brulé, R.G. Margue, J. Chem. Soc. Perkin Trans. 1
[
1] B.M.L. Dioos, I.F.J. Vankelecom, P.A. Jacobs, Adv. Synth. Catal. 348
(2006) 1413.
[
[
2] P. Mastrorilli, C.F. Nobile, Coord. Chem. Rev. 248 (2004) 377.
3] M. Nielsen, A.H. Thomsen, T.R. Jensen, H.J. Jakobsen, J. Skibsted, K.V.
Gothelf, Eur. J. Org. Chem. (2005) 342.
[
[
4] U. Arnold, W. Habicht, M. Döring, Adv. Synth. Catal. 348 (2006) 142.
5] Q.H. Xia, H.Q. Ge, C.P. Ye, Z.M. Liu, K.X. Su, Chem. Rev. 105 (2005)
(2001) 3085.
1603.
[
[
30] Y.R. de Miguel, J. Chem. Soc. Perkin Trans. 1 (2000) 4213.
31] D.C. Sherrington, Catal. Today 57 (2000) 87.
[
[
[
[
6] S. Bräse, F. Lauterwasser, R.E. Ziegert, Adv. Synth. Catal. 345 (2003) 869.
7] Q.H. Fan, Y.M. Li, A.S.C. Chan, Chem. Rev. 102 (2002) 3385.
8] D.E. Bergbreiter, Top. Curr. Chem. 242 (2004) 113.
[32] S.V. Ley, I.R. Baxendale, R.N. Bream, P.S. Jackson, A.G. Leach, D.A.
Longbottom, M. Nesi, J.S. Scott, I. Storer, S.J. Taylor, J. Chem. Soc.
Perkin Trans. 1 (2000) 3815.
[33] S.J. Shuttleworth, S.M. Allin, R.D. Wilson, D. Nasturica, Synthesis 8
(2000) 1035.
9] D.A. Pears, S.C. Smith, Aldrichim. Acta 38 (2005) 23.
[
[
10] S. Kobayashi, R. Akiyama, Chem. Commun. (2003) 449.
11] C. Ramarao, S.V. Ley, S.C. Smith, I.M. Shirley, N. DeAlmeida, Chem.
Commun. (2002) 1132.
[
[
12] M.H. Valkenberg, W.F. Hölderich, Catal. Rev. Sci. Eng. 44 (2002) 321.
13] R.A. Sheldon, M. Wallau, I.W.C.E. Arends, U. Schuchardt, Acc. Chem.
Res. 31 (1998) 485.
[34] M. Ahlmann, O. Walter, M. Frank, W. Habicht, J. Mol. Catal. A 249 (2006)
80.
[35] J.G. Hill, B.E. Rossiter, K.B. Sharpless, J. Org. Chem. 48 (1983) 3607.
[36] K.B. Sharpless, T.R. Verhoeven, Aldrichim. Acta 12 (1979) 63.
[37] D.C. Sherrington, Catal. Today 57 (2000) 87.
[38] J.H.M. Heijnen, V.G. de Bruijn, L.J.P. van den Broeke, J.T.F. Keurentjes,
Chem. Eng. Process. 42 (2003) 223.
[39] V. Farina, Adv. Synth. Catal. 346 (2004) 1553.
[40] R.B. Bedford, C.S.J. Cazin, D. Holder, Coord. Chem. Rev. 248 (2004)
2283.
[
[
[
14] B. Ellis (Ed.), Chemistry and Technology of Epoxy Resins, Blackie Aca-
demic & Professional, London, 1993.
15] C.A. May (Ed.), Epoxy Resins, Chemistry and Technology, Marcel
Dekker, New York, 1988.
16] J.W. Muskopf, S.B. McCollister, in: W. Gerhartz, Y.S. Yamamoto, L.
Kaudy, J.F. Rounsaville, G. Schulz (Eds.), Ullmann’s Encyclopedia of In-
dustrial Chemistry, vol. A9, VCH, Weinheim, 1987, p. 547.

J. Artner et al. / Journal of Catalysis 255 (2008) 180–189
189
[41] A.F. Littke, G.C. Fu, Angew. Chem. Int. Ed. 41 (2002) 4176.
[52] T.-J. Yoon, J.I. Kim, J.-K. Lee, Inorg. Chim. Acta 345 (2003) 228.
[
42] N.T.S. Phan, M. Van Der Sluys, C.W. Jones, Adv. Synth. Catal. 348 (2006)
[53] S.-I. Fujita, S. Akihara, S. Fujisawa, M. Arai, J. Mol. Catal. A 268 (2007)
244.
609.
[
[
[
[
43] F. Fan, Ph.D. thesis, University of Heidelberg, 2007.
[54] I. Kani, R. Flores, J.P. Fackler Jr., A. Akgerman, J. Supercrit. Fluids 31
(2004) 287.
[55] S. Ricken, P.W. Osinski, P. Eilbracht, R. Haag, J. Mol. Catal. A 257 (2006)
78.
[56] S.-M. Lu, H. Alper, J. Am. Chem. Soc. 125 (2003) 13126.
[57] G.E. Oosterom, S. Steffens, J.N.H. Reek, P.C.J. Kamer, P.W.N.M. van
Leeuwen, Top. Catal. 19 (2002) 61.
[58] F. Shibahara, K. Nozaki, T. Matsuo, T. Hiyama, Bioorg. Med. Chem.
Lett. 12 (2002) 1825.
44] K. Okamoto, R. Akiyama, S. Kobayashi, Org. Lett. 6 (2004) 1987.
45] H. Hagio, M. Sugiura, S. Kobayashi, Org. Lett. 8 (2006) 375.
46] C.K.Y. Lee, A.B. Holmes, S.V. Ley, I.F. McConvey, B. Al-Duri, G.A.
Leeke, R.C.D. Santos, J.P.K. Seville, Chem. Commun. (2005) 2175.
47] R. Akiyama, S. Kobayashi, J. Am. Chem. Soc. 125 (2003) 3412.
48] K. Okamoto, R. Akiyama, S. Kobayashi, J. Org. Chem. 69 (2004) 2871.
49] H. Oyamada, R. Akiyama, H. Hagio, T. Naito, S. Kobayashi, Chem. Com-
mun. (2006) 4297.
[
[
[
[
50] N. Bremeyer, S.V. Ley, C. Ramarao, I.M. Shirley, S.C. Smith, Synlett
[59] M.T. Zarka, M. Bortenschlager, K. Wurst, O. Nuyken, R. Weberskirch,
Organometallics 23 (2004) 4817.
(2002) 1843.
[51] M. Ahlmann, O. Walter, J. Organomet. Chem. 689 (2004) 3117.
[60] Y. Uozumi, M. Nakazono, Adv. Synth. Catal. 344 (2002) 274.