Liquid phase hydrogenation of benzalacetophenone: Effect of solvent, catalyst support, catalytic metal and reaction conditions

Article abstract of DOI:10.1016/S1872-2067(10)60249-5

Innovative catalysts based on a "porous glass" support material were developed and investigated for the reduction of benzalacetophenone. The easy preparation conditions and possibility to use different metals (e.g. Pd, Pt, Rh) for impregnation gave a broa

Full text of DOI:10.1016/S1872-2067(10)60249-5

CHINESE JOURNAL OF CATALYSIS
Volume 32, Issue 8, 2011
Online English edition of the Chinese language journal
Cite this article as: Chin. J. Catal., 2011, 32: 1312–1322.
RESEARCH PAPER
Liquid Phase Hydrogenation of Benzalacetophenone: Effect
of Solvent, Catalyst Support, Catalytic Metal and Reaction
Conditions
1
,
1
1
2
Achim STOLLE *ˈChristine SCHMÖGER , Bernd ONDRUSCHKA , Werner BONRATH ,
3
3
Thomas F. KELLER , Klaus D. JANDT
1
Institute for Technical Chemistry and Environmental Chemistry (ITUC), Friedrich-Schiller University Jena, Lessingstrasse 12, D-07743 Jena,
Germany
2
R&D Chemical Process Technology, DSM Nutritional Products, P.O. Box 2676, CH-4002 Basel, Switzerland
3
Institute of Materials Science and Technology (IMT), Friedrich-Schiller University Jena, Löbdergraben 32, D-07743 Jena, Germany
Abstract: Innovative catalysts based on a “porous glass” support material were developed and investigated for the reduction of benzal-
acetophenone. The easy preparation conditions and possibility to use different metals (e.g. Pd, Pt, Rh) for impregnation gave a broad variety
of these catalysts. Hydrogenation experiments with these supported catalysts were carried out under different hydrogen pressures and
temperatures. Porous glass catalysts with Pd as the active component gave chemoselective hydrogenation of benzalacetophenone, while Pt-
and Rh-catalysts tended to further reduce the carbonyl group, especially at elevated hydrogen pressures and temperatures. Kinetic analysis
of the reactions revealed these had zero order kinetics, which was independent of the type of porous glass support and solvent used.
Key words: chalcone; hydrogenation; porous glass; supported catalyst
The reduction of unsaturated bonds is important in the syn-
theses of fine chemicals such as vitamins, flavours, and fra-
grances [1,2]. The chemical industry often employs gaseous
hydrogen as the reducing reagent due to its excellent atom
efficiency. Hydrogenation is used in the processing of vegeta-
ble oils and fats and in the crude oil and coal industries [3–5].
In contrast, on the laboratory scale, catalytic transfer hydro-
genation (CTH) is often the method of choice [6] due to less
hazardous handling and convenient laboratory equipment.
Formic acid (or its salts), hydrazine, or cyclohexene derivatives
are common reducing reagents, and are most often used in
connection with a heterogeneous or homogeneous catalyst (e.g.
Ni-alloy, Ru-complexes) [7,8]. The use of gaseous hydrogen
and a heterogeneous catalyst in the liquid phase would com-
bine their advantages. The easy recovery of catalyst, good yield
and selectivity, and a high atom efficiency of the hydrogenation
connection with microwave irradiation is unusual and only a
few papers on it have been published to date [6,9,10]. Part of
the following work used reduction with gaseous hydrogen in a
microwave apparatus. A new reactor named QRS (quartz-
reactor-system) was developed to carry out hydrogenation
under moderate pressures (” 0.8 MPa) and temperatures (” 200
°C) on the laboratory scale. Commercially available steel
autoclaves were used for liquid phase hydrogenation under
higher pressures (” 5.0 MPa).
The model substrate used for the liquid phase hydrogenation
was 1,3-diphenylprop-2-en-1-one (1; Scheme 1), often called
benzalacetophenone or chalcone. Chalcones are essential in the
pharmaceutical industry because of their antibiotic or bacte-
riostatic properties [11]. They are part of a family of aromatic
ketones with two aromatic groups bridged by an enone linkage.
Usually they are prepared by base- or acid-catalyzed aldol
condensation of aromatic aldehydes and the respective aceto-
phenones [11,12]. The solvent-free preparation with the use of
with H are some of the advantages.
2
The use of gaseous hydrogen for reduction reactions in
Received 23 March 2011. Accepted 2 May 2011.
*
Corresponding author. Tel: +49-3641-948413; Fax: +49-3641-948402; E-mail: Achim.Stolle@uni-jena.de
Copyright © 2011, Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier BV. All rights reserved.
DOI: 10.1016/S1872-2067(10)60249-5

Achim STOLLE et al. / Chinese Journal of Catalysis, 2011, 32: 1312–1322
ach/Germany). The remaining support materials, substrates,
O
O
OH
+
H2
+
H2
metal precursors, and solvents were from Sigma-Aldrich. The
chemicals were used without further purification. The purities
of the solvents and starting materials were checked by gas
chromatography (GC) prior to use.
Ph
Ph
Ph
Ph
Ph
Ph
Ph
1
a
1b
1
+
3H2
ꢀH O
2
+
H2
Ph
O
Ph
Ph
1d
In a round bottom flask (100 ml), the metal component (0.09
mmol) was dissolved in an adequate amount of solvent (50 ml;
Table 2). The support material (1 g; Table 1) was added to the
mixture. After equilibration (stirring, 30 min), the solvent was
removed in vacuo. Subsequently. the catalyst precursor was
calcined for 2 h at 300–450 °C (corresponding to the decom-
position temperature of the metal compound, Table 2) in a
muffle furnace “mls 1200 pyro” (MLS GmbH, Leutkirch,
Germany) resulting in a catalyst with a metal loading of 0.09
mmol/g. Calcination temperatures above 450 °C cannot be
realized, because the porous texture of the porous glasses.
BET-analyses showed that higher temperatures lead to a sig-
nificant decrease of specific surface area. In contrast, longer
reaction times (4 or 6 h) show no influence on the porosity of
the porous glass support.
Ph
Cyx
1e
+3H2
Cyx
1c
+
H2
ꢀ
H O
2
+
H2
Ph = Phenyl
Cyx = Cyclohexyl
+3H2
Ph
Cyx
Cyx
1
f
1g
Scheme 1. Reaction scheme for the reduction of benzalacetophenone
1) with H and supported metal catalysts.
(
2
acidic clays is also possible [13]. Stoichiometric amounts of
benzaldehyde and acetophenone are the raw materials for the
preparation of 1. The reduction of the C-C double bond in 1
leads to the saturated ketone 1,3-diphenylpropan-1-one (1a),
which is an intermediate for the Friedländer quinoline synthe-
sis [14]. It is also an interesting building block for enol car-
bonates [15].
The aim of this study was to determine the influence of
various reaction parameters in the liquid phase hydrogenation
of benzalacetophenone catalyzed by metal-loaded porous
glasses. Various catalytically active metals and support mate-
rials used at different reaction temperatures and hydrogen
pressures were tested. To cover the required temperature and
pressure range, three different apparatuses were used.
The preparation of catalysts with two metal components
used a modification of the procedure described above. The
catalyst Ce/Pd/TP(I) was prepared by one impregnation step.
Here, both metal components (each with 0.09 mmol) were
dissolved in 50 ml solvent. After removal of the solvent, the
catalyst precursors were also calcined at the corresponding
temperature. The catalyst Ce/Pd/TP(II) was prepared by double
impregnation. After impregnation with one metal component
(0.09 mmol) and calcination, a second impregnation step fol-
lowed with the other metal component (also 0.09 mmol). The
impregnations were performed in the same way as described
above.
1
Experimental
1
.1 Preparation of the catalysts
Table 1 summarizes the important parameters (e.g. surface
area, pore size) of the support materials for the preparation of
the catalysts. The metal precursors and corresponding solvents
employed are listed in Table 2. The porous glass supports
1.2 Characterization of the catalysts
TGA analyses of the metal precursors were carried out with
(
Table 1, entry 1–8) were purchased from VitraBio (Stein-
a Shimadzu DTG-60 simultaneous DTA-TG apparatus
Table 1 Selected parameters of the support materials
2
3
Entry
Support name
Notation
TA
d
50/μm
d
p
/nm
51
50
49
55
55
45
S
V
/(m /g)
V
p
/(mm /g)
®
1
2
3
4
5
6
7
8
9
TRISOPOR
50–90
89
98
75
96
94
85
20
14
1215
1321
1115
1213
1256
759
®
TRISOPOR
TB
TC
TPA
TP
100–200
®
TRISOPOR
315–500
®
TRISOPERL
50-100
®
TRISOPERL
100–200
®
TRISOPOR
A
1250
®
TRISOPOR
B
800–2500
173
300
711
®
TRISOPOR
C
800–2500
664
silica
silica
SiA
SiB
Si1
Al1
35–45
17–23
26–34
—
110–180
70–170
500–600
50
600–1000
600–1000
—
1
0
1
35–45
1
silica
105–200
110
1
2
alumina
—
—
d
50 = average particle diameter, d
p
= average pore size, S
V
p p V p
= average specific surface area, V = average specific pore volume; d , S , and V were calcu-
lated from data of 6-point BET measurements and are in agreement with the data provided by the manufactures.

Achim STOLLE et al. / Chinese Journal of Catalysis, 2011, 32: 1312–1322
Table 2 Metal precursors, the required solvents, and calcination temperatures for the preparation of the various catalysts
a
b
c
Catalyst
Metal precursor
Pd(AcO)
PdCl
Ce(acac) + Pd(AcO)
Ce(acac) /Pd(AcO)
Pt(acac)
RhCl ·H
Rh(acac)
RuCl ·H
Ru(acac)
mpLit /°C
Solvent
T /°C
d
Pd/Support
Pd-A/TP
2
205 (dec.)
675 (dec.)
131/205
131/205
249–252
100
dichloromethane
HClaq (5%)
dichloromethane
dichloromethane
acetone
300
450
300
300
300
450
300
450
300
300
300
450
300
300
300
2
Ce/Pd/TP(I)
Ce/Pd/TP(II)
Pt/TP
3
2
3
2
2
Rh/TP
3
2
O
acetone
Rh-A/TP
Ru/TP
3
263–264
>500 (dec.)
260 (dec.)
269–271
211 (dec.)
56
acetone
3
2
O
acetone
Ru-A/TP
Ir/TP
3
acetone
Ir(acac)
3
acetone
Co/TP
Co(acac)
3
acetone
Ni/TP
Ni(NO
Ag(hfacac)(COD)
HAuCl ·3H
Fe(acac)
3
)
2
·6H
2
O
acetone
Ag/TP
122–124
254 (dec.)
acetone
Au/TP
4
2
O
acetone
Fe/TP
3
180
acetone
a
b
c
Melting point of metal precursor according to literature (confirmed by TGA). Solvent used for catalyst preparation. Calcination temperature (calcination
d
time: 2 h inclusive preheating time 20 min). All supports from Table 1. dec. – Decomposition.
(
(
T-ramp: 10 °C/min). Transmission electron microscopy
liquid phase hydrogenation reactions were carried out at at-
mospheric pressure and room temperature in a shaking appa-
ratus “HS 501 digital” (IKA Labortechnik, Staufen, Germany).
Every 20 min (after 60 min reaction time every 30 min), a
sample (500 μl) was taken. The hydrogenation process was
observed by the consumption of hydrogen and by analyzing the
samples with GC-FID and GC-MSD.
TEM) micrographs were recorded with a transmission elec-
tron microscope H7500 from Hitachi (120 kV, LAB6 filament).
The scanning electron microscopy (SEM) images were meas-
ured using a scanning electron microscope DSM 940 (Carl
Zeiss). Optical microscopy pictures were recorded with a Stemi
2
000-C microscope (Carl Zeiss; magnification: (6.5–50)×; cold
light source KL 1500 LCD) in combination with a digital
camera JVC GZ-MC200E (2 mega pixel; 10× optical zoom).
1.3.2 Reactions in setup B
6-Point BET measurements were carried out with an Auto-
sorb1 (Quantachrome) using N as sorbent gas and a heating
Benzalacetophenone (1, 7.5 mmol, 1.56 g), 100 μl hexa-
decane (internal standard), catalyst (200 mg, 0.24 mol%), and
ethyl acetate (30 ml) were put into a quartz reactor vessel. The
vessel was placed in a microwave equipment (multisynth, MLS
GmbH, Leutkirch, Germany) and connected to the hydrogen
reservoir. The use of a gas inlet allowed the direct flushing of
hydrogen into the reaction mixture. The gas was continuously
let off with a gas outlet valve to get a uniform gas flow (0.2
ml/min) through the reaction mixture. The liquid phase hy-
drogenation reactions were carried out at various pressures
(0.2–0.8 MPa), temperatures (20–50 °C), and reaction times
(0–40 min). The reaction temperature was measured and con-
trolled using an ACT-CE sensor placed directly in the reaction
solution. The reaction temperature of 25 °C was maintained by
cooling the reactor with pressurized air. The hydrogenation
process was observed by analyzing the samples with GC-FID
and GC-MSD.
2
temperature of 350 °C. X-ray photoelectron spectroscopy
(XPS) spectra were recorded with a Quantum 2000 (PHI Co.)
employing a focused monochromatic Al K source (1486.7 eV)
D
for excitation. The pass energy was set to 23.5 eV and as ref-
erence, the C 1s peak of the C–C bond was used.
1
.3 Reduction reactions
Three different experimental setups were required for the
reduction reactions because different temperatures and pres-
sures were used. The following equipments were used. Setup
A: room temperature (25 °C) and atmospheric pressure. Setup
B: pressure ” 0.8 MPa, temperature ” 50 °C. Setup C: pressure
”
5.0 MPa, room temperature (25 °C).
1
.3.1 Reactions in setup A
Benzalacetophenone (1, 15 mmol, 3.12 g) and 200 μl
1.3.3 Reactions in setup C
hexadecane (internal standard) were dissolved (60 ml solvent)
and put into a Schlenk flask. After adding the catalyst (400 mg,
Reduction at higher pressures were carried out in a com-
mercial steel autoclave (Berghoff, Germany) equipped with a
100 ml PTFE insert. The reaction was controlled with pressure
and temperature sensors equipped with the sealed autoclave.
0
.24 mol%), a sample was taken (500 μl). The Schlenk flask
was connected to the hydrogen reservoir by a plug valve. The
reaction vessel was flushed with hydrogen (400 ml × 3). The

Achim STOLLE et al. / Chinese Journal of Catalysis, 2011, 32: 1312–1322
Benzalacetophenone (1, 7.5 mmol, 1.56 g), 100 μl hexadecane
internal standard), catalyst (200 mg, 0.24 mol%), and ethyl
(b)
(a)
(
acetate (30 ml) were put into a Teflon vessel inside the steel
autoclave. The reaction vessel was flushed with nitrogen (1.0
MPa u 3). Afterwards the desired hydrogen pressure was set
up. The liquid phase reduction reactions were carried out at
different pressures (1.0–5.0 MPa) and reaction times (30–60
min). The hydrogenation process was observed by analyzing
the samples with GC-FID and GC-MSD.
magnification: 10h
magnification: 5h
Fig. 1. Optical microscopy images of porous glasses TB (a) and TP (b).
materials prior to impregnation with the metal precursor. TB
consisted of irregular aspheric particles with sharp broken
edges, while the TP particles were regular and spherical. Re-
garding the particle shapes of TB and TP, the outer particle
surface areas were higher for the aspheric particles than for the
ball-shaped particles. Due to the porous structure of these
1
.3.4 Analysis of the reduction reaction
The analyses of the reaction mixtures were carried out with
GC-FID (Agilent 6890) and GC-MSD (Agilent 6890 N/5973
MS).
Conditions of GC-FID. Column: HP 5, 30 m × 0.32 mm ×
materials, the influence of the outer surface on S was rather
V
0
.25 μm; Carrier gas: H , 68.9 kPa; program: 60 °C (hold 1
2
small due to the high pore volume (V ; Table 1). The dispersion
p
min), 25 °C/min up to 280 °C (hold for 12 min); injector
temperature: 280 °C; detector temperature: 300 °C.
of the particles sizes was 100 ” d50 ” 200 μm.
The detailed structure of the catalysts (porous glass im-
pregnated with metal) was determined by SEM (Fig. 2) and
TEM (Fig. 3). Catalysts Pd/TB and Pd/TP were chosen for
measurement. The photos revealed the open porous structure
Conditions of GC-MSD. Column: HP 5, 30 m × 0.32 mm ×
0
.25 μm; Carrier gas: He, 68.9 kPa; program: 60 °C (hold 1
min), 25 °C/min up to 280 °C (hold for 12 min); injector
temperature: 280 °C; EI: 70 eV.
and the SiO network of the catalyst. The Pd clusters (5 nm ”
2
Conversion, yield, and selectivity were calculated based on
corrected peak areas using hexadecane as the internal standard.
All results were averages of at least two experiments.
cluster size ” 50 nm) were clearly seen on the surface (bright
points in the SEM and dark points in the TEM micrographs).
The measurements revealed that there was no complete cov-
erage of the surface. The dispersion of the Pd particles on the
surface could not be found because of the irregular spherical
structure of the support material and its nano-porous structure.
The micrographs indicated a narrow pore size distribution and
no hierarchical porosity.
2
Results and discussion
2
.1 Catalyst characterization
Porous glasses have been applied as the support material for
XPS was used to identify the oxidation state of the metals
and to deduce the catalytically active form of the catalysts.
XPS data of selected catalysts that showed good catalytic
activity (cf. following sections) are listed in Table 3. The dif-
Pd catalysts in the hydrogenation of alkenes and alkynes
17–19], Suzuki-Miyaura [20], and Mizoroki-Heck
[
cross-coupling [20,21] and for the partial hydrogenation of
citral to citronellal [19]. The catalysts on the porous glasses or
support materials with similar properties were characterized
ference in binding energies (E ) between the oxidation states of
B
®
with different methods. The support materials TRISOPOR
®
(
Table 1, entries 1–3 and 6–8) and TRISOPERL (Table 1,
entries 4 and 5) were the focus of these investigations. In par-
ticular, the two types of nano-structured glasses, TB and TP,
were examined because most of the following liquid phase
hydrogenation was carried out with catalysts on these supports.
The two materials differ only in their external form but not in
1
ȝ
m
100 nm
Fig. 2. SEM micrographs of Pd/TP.
the parameters of d50 and S (Table 1, entries 2 and 5). Both
V
support materials are commercially available and have been
prepared by the same method (alkaline leaching of
phase-separated borosilicate glasses) [22–27].
First, the particles were examined with optical microscopy.
The optical resolution of this method allowed the retrieval of
information on the geometric texture of the particles, but it was
not enough to see the pore structure of the isolated particles (dp
2
00 nm
100 nm
Fig. 3. TEM micrographs of Pd/TB.
”
60 nm). Figure 1 shows the photos of the two porous support

Achim STOLLE et al. / Chinese Journal of Catalysis, 2011, 32: 1312–1322
Table 3 XPS data of selected catalyst prepared by wet impregnation of the support material TP
Binding energy of catalysts (eV)
Core level
a
Catalyst
b
c
d
Before calcination
336.8
341.7
336.5
341.8
336.4
341.8
885.7
905.8
336.4
341.9
885.7
905.8
71.8
After calcination
336.1
341.3
336.2
341.5
336.0
341.4
885.8
905.8
336.0
341.2
885.8
905.6
71.8
After application
335.3
340.3
335.9
341.2
335.5
340.5
885.7
905.7
335.5
340.8
885.7
905.6
72.8
Pd/TP
Pd 3d5/2
Pd 3d3/2
Pd 3d5/2
Pd 3d3/2
Pd 3d5/2
Pd 3d3/2
Ce 3d5/2
Ce 3d3/2
Pd 3d5/2
Pd 3d3/2
Ce 3d5/2
Ce 3d3/2
Pt 4f7/2
Pd-A/TP
Ce/Pd/TP(I)
Ce/Pd/TP(II)
Pt/TP
Pt 4f5/2
75.2
75.2
76.0
Rh/TP
Rh-A/TP
Rh 3d5/2
Rh 3d3/2
Rh 3d5/2
Rh 3d3/2
308.4
312.7
308.7
313.4
308.3
312.8
309.3
313.4
308.2
312.8
307.6
312.2
a
b
c
Metal loading 0.09 mmol/g. Impregnation with metal precursors listed in Table 2. Calcination for 2 h in air atmosphere (for temperature see Table 2).
d
Hydrogenation of 1 at room temperature and atmospheric pressure (15 mmol 1, 0.24 mol% metal/TP, 60 ml ethyl acetate).
the metal was rather small ('E § 2–3 eV). For this reason, the
catalysts were analyzed before and after calcination (time and
temperature data given in Table 2) as well as after their use in
the hydrogenation of 1 (Scheme 1). This procedure ensured
comparable measurement conditions with respect to the
metal-coordination sphere and surface concentration of the
metal. Thus the results obtained can be compared and inter-
preted with a better accuracy by comparison with bulk metal
compounds of different oxidation states. The XPS data shown
in Table 3 did not indicate any metal-support interaction. All
values were typical for metals supported on oxide materials
like alumina and silica. Distinguishing between the metal lay-
ers in contact with the support material and those on the surface
of the metal particle was not possible. Therefore, no effect of
the support on the oxidation state was found.
was probably due to incorporation into the SiO network of the
2
glass carrier. A bonding to Pd did not occur since the shift for
the Pd 3d was similar to that observed for Pd/TP (Table 3).
In case of Pt/TP, two characteristic peaks at 71.8 (Pt 4f )
7
/2
and 75.2 eV (Pt 4f ) were present (Table 3) [30,31]. A shift to
5
/2
higher E was detected after the catalyst was used in the re-
B
duction reaction, which showed a partial change in oxidation
state had taken place. This oxidation state was not studied in
detail because all the possible Pt oxidation states (+2, +4) were
active in the hydrogenation reaction.
The Rh-containing catalysts, Rh/TP and Rh-A/TP, differed
only in the metal precursor used, which were RhCl ·H O and
3
2
Rh(acac) , respectively. The characteristic E for the Rh signals
3
B
were at 308 and 313 eV for the Rh 3d5/2 and 3d3/2 core levels,
respectively (Table 3) [32]. The E of both peaks and catalysts
B
The interpretation of the Pd/TP XPS data set is clearly de-
scribed in the literatures [14–27]. The corresponding Ce con-
taining catalysts (Ce/Pd/TP(I) and Ce/Pd/TP(II)) showed, in
showed no significant change before and after calcination. That
is, the oxidation state was not changed by calcination in air.
The use of Rh-A/TP in a hydrogen atmosphere gave a shift to
case of Pd, the same shift of E as Pd/TP. The difference be-
smaller E , which indicated the oxidation state change from +3
B
B
tween the two catalysts was in their preparation method. In the
case of Ce/Pd/TP(I), both metals were co-impregnated from
one solution, while in the other case, Pd was deposited before
to Rh(0) for the spent catalyst. The catalyst Rh/TP did not show
this shift after it was used. This difference was attributed to the
precursor compound RhCl used, which left residual Cl on the
3
wet impregnation with Ce(acac) . Thus, an additional calcina-
surface and inhibited the reduction from +3 to Rh(0).
3
tion step was used to anchor the Pd before the decoration with
Ce. A comparison of the Ce binding energies showed no sig-
nificant change in E for the two standard Ce peaks (Ce 3d
2.2 Liquid phase hydrogenation at room temperature and
atmospheric pressure
B
5/2
and Ce 3d ) [28,29]. The constant E indicated no incorpora-
3
/2
B
tion of the Ce component during the use of these catalysts.
Furthermore, the broad peaks obtained showed that two oxi-
dation states of Ce (+3 and +4) existed on the surface, which
The liquid phase hydrogenation of 1 was chosen as the
model reaction (Scheme 1). The molecule structure has dif-
ferent unsaturated groups: i) a conjugated C-C double bond, ii)

Achim STOLLE et al. / Chinese Journal of Catalysis, 2011, 32: 1312–1322
a carbonyl group, and iii) aromatic C-C double bonds. The
ble rates for the porous glass supports and conventional support
materials. The activities were nearly the same for the porous
glass systems (Table 4, entries 1–5) and silica-based catalysts
(Table 4, entries 9–11). The lowest rate constants and conver-
sions were found with the porous glass catalysts Pd/A, Pd/B,
and Pd/C. It was probable that increasing the particle sizes in
degree of reduction and reaction rate of the hydrogenations
depended on several parameters like the type of metal and
support material, hydrogen pressure, and reaction temperature.
First, the influence of support material in the liquid phase
hydrogenation of 1 at room temperature and atmospheric
pressure was studied (Table 4). In this case, the reduction ex-
periments were carried out with palladium as the active metal.
Thus, a comparison of the selected support materials in the
reduction of 1 was possible. The various support materials are
listed in Table 1. Mainly, porous glasses were used (Table 1,
entries 1–8). They belong to the group of leached phase sepa-
conjunction with decreasing S decreased mass transport and
V
metal dispersion on the catalyst surface (Table 1) resulted in
decreased reaction rates. Particularly good results were ob-
tained with the systems Pd/TB and Pd/TP. A conversion of 75%
was obtained after 120 and 150 min, respectively. The differ-
ence between the TB and TP supports was the geometrical
shape of the particles (Fig. 1). The difference in particle mor-
phology had no direct influence on the hydrogenation as shown
by the similar initial reaction rates (Table 4) and kinetic profiles
(Fig. 4). Thus, the similarity with respect to the physical
properties (d , d , S and V ; cf. Table 1) showed they were
rated borosilicate glasses (SiO content • 96 %; confirmed with
2
EDX) and offer the advantages as the support of chemical
inertness, flexible texture, and narrow pore size distribution
[
17–23]. In addition, three different silica gels (Table 1, entries
–11) and one alumina (Table 1, entry 12) were used to com-
9
5
0
p
V
p
pare conventional support materials with the relatively un-
known porous glass systems of TB and TP. When 0.24 mol%
Pd (catalyst loading 0.09 mmol/g) was used, all the experi-
ments gave the same reduction product 1,3-diphenylpropan-
equal when applied as the support material in catalysis. The
100
8
0
0
1
-one (1a, Scheme 1). In all cases, the hydrogenation was
chemoselective under the selected reaction conditions since
only the conjugated C-C double bond was reduced. Further-
more, the support material had no influence on the selectivity
but only on the reaction rate of the hydrogenation.
6
40
20
0
Pd/TB
Pd/TP
Pd/Si1
Pd/Al1
Pd/B
The rate constants k and k were calculated from the time
1
2
dependent hydrogen uptake curves for the hydrogenation re-
action. The use of a pseudo-zero order of reaction gave k1,
which indicated that the reaction was diffusion controlled and
the reaction gas hydrogen was the limiting reactant at the be-
ginning of the process. The reaction became substrate con-
0
50
100
150
200
250
Time (min)
trolled (k ) with the decrease of substrate concentration, thus
2
Fig. 4. Conversion of 1 by selected supported Pd catalysts at room
temperature and atmospheric pressure. Reaction conditions are the same
as in Table 4.
the reaction order changed to pseudo-first order kinetics. The
comparison of k for the different catalysts revealed compara-
1
Table 4 Rate constants (k
1
and k
2
) and time for 25%, 50%, and 75% conversion of 1 using various supported Pd-catalysts
Time for conversion of 1 (min)
a
b
b
–1
Entry
Catalyst
k
1
/(mlH2/min)
2
k /min
2
5%
50%
85
75%
145
120
100
140
150
420
300
375
100
120
145
80
1
2
3
4
5
6
7
8
9
Pd/TA
Pd/TB
Pd/TC
Pd/TPA
Pd/TP
Pd/A
2.705
3.014
2.751
2.779
2.844
0.930
1.240
0.940
2.802
2.738
1.944
3.672
0.0025
0.0020
0.0018
0.0024
0.0026
0.0008
0.0009
0.0030
0.0008
0.0025
0.0032
0.0011
30
20
20
30
30
80
40
75
25
30
40
20
60
60
70
75
240
175
220
60
Pd/B
Pd/C
Pd/Si1
Pd/SiA
Pd/SiB
Pd/Al1
1
0
1
70
1
90
1
2
45
a
Reaction conditions: substrate 1 15 mmol, catalyst 0.4 g (0.24 mol% Pd), ethyl acetate 60 ml, T = 25 °C, p(H
2
) = ambient pressure. Pd loading 0.09
b
mmol/g. The rate constants k
1
and k
2
were estimated with linear regression. Following acceptances are made: liquid-phase hydrogenation is a reaction of
pseudo zero-order and become a reaction of pseudo first-order at the end of reaction time.

Achim STOLLE et al. / Chinese Journal of Catalysis, 2011, 32: 1312–1322
time versus conversion curves of 1 for selected catalysts are
Table 5 Rate constants and time for 25%, 50%, and 75% conversion of
shown in Fig. 4. The curves showed complete conversion of 1
after 210 min for all the catalysts except Pd/B. In summary, the
results demonstrated that the porous glasses can be used as
support material in heterogeneous catalysis, especially in liquid
phase hydrogenation [17–19]. The porous glasses TB and TP
were used as the standard support for the catalysts and used in
the following reactions.
1
using various solvents
Time for conversion
a
a
–1
Solvent
k
1
/(mlH2/min)
k
2
/min
of 1 (min)
2
5%
50%
—
75%
—
Chloroform
Dioxane
1.076
2.017
2.092
3.014
3.113
3.945
4.236
0.0046
210
65
40
20
20
25
15
*
210
90
—
Acetone
0.0043
0.0023
0.0042
0.0028
0.0032
150
120
150
120
105
Further experiments used the variation of different parame-
ters during the liquid phase hydrogenation under room tem-
perature and atmospheric pressure. First, the metal loading of
the catalyst system Pd/TB was varied in the range of 0.1–2 wt%
Pd. The curves of time versus conversion of 1 with 0.4 g of
catalyst are shown in Fig. 5(a). The catalysts with a Pd loading
of 1 wt% and higher gave complete conversion within 210 min.
A decrease of the Pd amount on the support led to a decrease in
the conversion of 1 to 1a since less active metal centers were
available. For this reason, all subsequent experiments were
carried out with a metal loading of 1 wt% Pd. These results
were confirmed by the corresponding turnover frequencies
Ethyl acetate
2-Propanol
Methanol
Ethanol
60
70
60
50
Reaction conditions: substrate 1 15 mmol, catalyst 0.4 g Pd/TB (0.24
mol% Pd, Pd loading 1 wt% Pd), solvent 60 ml, T = 25 °C, p(H
2
) = am-
a
bient pressure. The rate constants k
1
and k
2
were estimated with linear
regression. Following acceptances are made: liquid-phase hydrogenation
is a reaction of pseudo zero-order and become a reaction of pseudo
b
first-order at the end of reaction time. Over the complete chosen time
range the curve progression is linear with the same slope (pseudo-zero
order reaction).
(
TOF, Fig. 5(b)). The highest TOF was obtained with 1 wt%
time. In the course of the reduction of 1, the TOF decreased in
–
1
Pd. At lower Pd loadings, less active centers were present,
while in the opposite direction, no improvement was seen with
an excess of catalytic centers. A similar relationship was shown
with the TOF for the conversion of 1 plotted against reaction
all cases and reached 150 h .
When the liquid phase hydrogenation of 1 was performed in
different solvents, interesting effects were observed. The sol-
vents and rate constants (k and k ) for hydrogenation with
1
2
Pd/TB at room temperature and atmospheric pressure are
summarized in Table 5. The chemoselective hydrogenation of
the C-C double bond in 1 with the Pd catalyst was not affected
by the solvent. 1a was the only product. The polarity and hy-
drophilicity of the solvents increased from chloroform to
ethanol (Table 5). These effects influenced the reaction rate and
1
00
(a)
7
5
2
5
0
5
0
0.1 wt% Pd
0.5 wt% Pd
1.0 wt% Pd
2.0 wt% Pd
led to an increase of the rate constant k with increasing solvent
1
polarity. This was attributed to the fact that a polar media has
good hydrogen absorption capacity. Thus, the gas-liquid dif-
fusion barrier for the mass transport of molecular hydrogen to
the active centers of the catalyst was lower. The decrease in the
mass transport limitation of the solvents was shown by an
2
2
1
1
50
00
50
00
0.1 wt% Pd
(b)
increase of k . Complete conversion of 1 was detected after 210
1
0.5 wt% Pd
1.0 wt% Pd
2.0 wt% Pd
min for acetone, ethyl acetate, 2-propanol, methanol, and
ethanol. The three alcohols showed the highest initial hydro-
genation rates, but all polar solvents showed complete con-
version within the chosen time range. The solvents dioxane and
chloroform (both non-polar) showed less hydrogen absorption
capacity. This had a direct influence on the reaction rate for the
reduction of 1. Only 50% (dioxane) and 25% (chloroform)
conversion of 1 were obtained after a reaction time of 210 min.
5
0
This effect was manifested in the smaller rate constants k with
1
0
these solvents. The purity of the solvent played an important
role in the liquid phase hydrogenation. No reduction of 1 was
detected using solvents with a purity < 98%. The small
amounts of impurities inhibited the hydrogen absorption ca-
0
50
100
150
200
Time (min)
Fig. 5. Conversion of 1 (a) and hydrogen uptake TOF (b) using catalyst
Pd/TB with different metal loadings at room temperature and atmospheric
pressure. Reaction conditions are the same as in Table 4.
pacity or/and contaminated the catalyst. The rate constant k2

Achim STOLLE et al. / Chinese Journal of Catalysis, 2011, 32: 1312–1322
(a)
(b)
Pd/TP
1
00
1.0
0.8
0.6
0.4
0.2
0.0
Ce/Pd/TP(II)
Ce/Pd/TP(I)
Pt/TP
8
6
4
2
0
0
0
0
0
Pd-A/TP
Pd/TP
Pd-A/TP
Ce/Pd/TP(II)
Ce/Pd/TP(I)
Pt/TP
0
50
100
150
200
250
300
0
50
100
150
200
250
Time (min)
Time (min)
Fig. 6. Time dependence of the conversion of 1 and molar fraction of 1a using various metals on porous glass at room temperature and atmospheric
pressure. Reaction conditions: 15 mmol 1, 0.24 mol% metal/TP, metal loading = 0.09 mmol/g, 60 ml ethyl acetate.
was in the same range for all the solvents and indicated the
change from the diffusion controlled regime to reaction con-
trol. This is accompanied by a change of reaction order from
was observed when the experimental results were compared
with those of the Ce-free Pd-catalyst Pd/TP. These results are in
agreement with the observations from the XPS data described
above regarding the change in oxidation state of Ce and Pd in
the bimetallic catalysts (Table 3). Neither dehydration nor
reduction of the aromatic double bond was found. The catalysts
Rh/TP, Rh-A/TP, Ru/TP, Ru-A/TP, Au/TP, Ag/TP, Ir/TP,
Co/TP, and Fe/TP showed no catalytic activity under the con-
ditions employed.
pseudo-zero to pseudo-first order. In the range of k the sub-
2
strate 1 was the limiting reactant and not the reaction gas hy-
drogen like before (range of k1).
Finally, the choice of solvent had an important influence on
the reaction rate and conversion of the reactant. The solvent
must have good hydrogen absorption capacity to get optimal
results for the hydrogenation. All following experiments were
carried out with ethyl acetate because of its polar character and
hydrophilic properties.
2.3 Liquid phase hydrogenation under moderate
temperature (” 50 °C) and pressure (” 0.8 MPa)
Figure 6(a) illustrates the catalyst systems with different
metal components and their activities in the liquid phase hy-
drogenation of 1 under the conditions used (25 °C, atmospheric
pressure). All the catalysts except for Pd-A/TP gave complete
conversion within 210 min. The reason for the low activity of
Reduction reactions under moderate conditions (T ” 50 °C, p
” 0.8 MPa) were carried out in the microwave reactor QRS
described in the experimental section. The porous glass support
used was TP impregnated with different metals (Table 2). All
the catalysts that showed catalytic activity under the selected
conditions are displayed in Table 6. The influence of reaction
temperature on the hydrogenation of 1 with Pd/TP as catalyst at
0.4 MPa is shown in Fig. 7. The reduction was also chemose-
lective since only 1a was found in the product. Reaction at
room temperature (25 °C) gave a maximum conversion of
Pd-A/TP was the metal precursor PdCl employed for the
2
preparation of this catalyst. The calcination temperature of the
catalyst (450 °C) was below the decomposition temperature of
PdCl (675 °C; see Table 2). Therefore, traces of chloride re-
2
mained on the surface and contaminated the active centers of
the catalyst. Thus, in situ reduction of the Pd(II)-species to the
catalytic active Pd(0) was suppressed, which was shown in the
oxidation state of Pd deduced from the XPS data (Table 3).
This led to the observed decrease in catalytic activity. A similar
behavior was observed for the change of the oxidation state for
Table 6 Conversion of 1 due to the catalyst and reaction temperature
Conversion of 1 (%)
Catalyst
2
5 °C
73
50 °C
99
Pd/TP
RhCl (Rh/TP; Table 3). All the catalysts reduced chemoselec-
3
Pd-A/TP
Ce/Pd/TP(I)
Ce/Pd/TP(II)
Pt/TP
14
18
tively the conjugated C-C double bond at room temperature
and atmospheric pressure. The only exception was Pt/TP. Fig-
ure 6(b) shows the molar fraction of the product 1a. In the case
of Pt/TP, a decrease of molar fraction after 50 min was de-
tected, indicating the occurrence of a consecutive reaction
pathway. The analysis showed the reduction of the C-O double
bond and the formation of the product 1,3-diphenylpropan-1-ol
92
99
65
99
58
99
Rh/TP
40
99
Rh-A/TP
17
99
Reaction conditions: 7.5 mmol 1, 30 ml ethyl acetate, 0.24 mol% M/TP,
metal loading = 0.09 mmol/g, pH2 = 0.8 MPa, reaction time 40 min, Mul-
tiSynth 400 W, ramp time 2 min.
(
1b; Scheme 1). No influence of Ce on the hydrogenation of 1

Achim STOLLE et al. / Chinese Journal of Catalysis, 2011, 32: 1312–1322
1
00
(
(
(
1)
2)
3)
(4)
1
00
1
00
8
0
8
0
8
0
60
6
0
0
6
4
0
4
0
40
2
0
2
0
20
0
0
10
10
50
0
0
5
20
20
40
0
4
30
30
4
0
2325
4
0
Fig. 7. Conversion of 1 at different temperatures. Reaction conditions:
.5 mmol 1, 30 ml ethyl acetate, 0.2 g Pd/TP (0.24 mol% Pd), pH2 = 0.4
Fig. 8. Conversion of 1 with various catalysts at different conditions.
(1) t = 30 min, pH2 = 3.0 MPa; (2) t = 30 min, pH2 = 5.0 MPa; (3) t = 90
min, pH2 = 3.0 MPa; (4) t = 90 min, pH2 = 5.0 MPa. Other conditions: 7.5
mmol 1, 30 ml ethyl acetate, 0.24 mol% M/TP, metal loading = 0.09
mmol/g, 25 °C.
7
MPa, continuous mode, reaction time 40 min, MultiSynth 400 W, ramp
time 2 min.
6
5%. The higher temperature and pressure gave complete
the reaction conditions used, a reduction of the aromatic C-C
double bonds was not observed. All the catalysts not listed in
Table 6 showed no catalytic activity (Table 2). These metals are
well known to have catalytic activity only under extreme con-
ditions (often gas phase reduction) or in homogenous catalysis
conversion within 30 or 40 min reaction time at 50 or 40 °C
reaction temperature, respectively. Similar results were found
for the reaction at elevated hydrogen pressure (0.6 MPa) since
only the conversions at low temperature and short reaction time
were slightly enhanced. Thus, the experiments proved that the
experimental set-up was suitable for performing hydrogenation
reactions under microwave irradiation with gaseous hydrogen
as the reducing agent [6,19,34–39]. The change from conven-
tional to microwave-assisted heating did not affect the course
of reaction. Increased conversions at a shorter time (Fig. 6)
were rather attributed to the elevated hydrogen pressure and/or
reaction temperature and not to the microwave. No cooperative
effect in both methods was seen from the spectroscopic analy-
sis of the catalyst and the reaction data.
[
41–43].
2
.4 Liquid phase hydrogenation under higher pressures
(
” 5.0 MPa) and room temperature
After reaction under moderate conditions, an increase of
pressure was studied by performing the model reaction in
stainless steel autoclaves. Figure 8 summarizes the results of
the reduction of 1 with different catalysts. The experiments
were undertaken with an upper pressure limit of 5.0 MPa at
room temperature. All the catalysts except Pd-A/TP showed
complete conversion at 5.0 MPa and 90 min. A reaction time
of 30 min was not sufficient in all cases. The best results
(conversion of 1 > 50%) were achieved with Pd/TP, Pt/TP,
Rh/TP, and Rh-A/TP. The use of Ce-containing catalysts
The catalytic activities with various metal catalysts at two
different reaction temperatures (25 and 50 °C) are compared in
Table 6. The catalysts and metal precursors employed are de-
scribed in Table 2. The reduction was carried out at 0.8 MPa
hydrogen pressure to use the maximum possible pressure in the
QRS microwave system. Therefore, moderate pressure and
temperature conditions were used, which can be compared with
the results described earlier. All the catalysts except Pd-A/TP
gave complete conversion of 1 at 0.8 MPa and 50 °C. At room
temperature (25 °C) only Pd/TP and Ce/Pd/TP(I) gave ac-
ceptable conversions. The influence of temperature was espe-
cially large for the Rh-containing systems (Rh/TP and
Rh-A/TP). The catalysts Pd/TP, Pd-A/TP, Ce/Pd/TP(I), and
Ce/Pd/TP(II) chemoselectively reduced the C-C double bond
and only product 1a was identified. The Rh- and Pt-containing
systems showed the reduction of the C-O double bond too.
During the reaction time (40 min) 2%–4% of 1b were formed
(
Ce/Pd/TP(I) and Ce/Pd/TP(II)) and the use of Pd/TP and
Pd-A/TP gave product 1a only. The reduction of the C-C dou-
ble bond was chemoselective in these cases. The addition of
Ce to the catalyst had no influence on the selectivity, but led
to a decreased conversion, probably due to blocking of the
active Pd centers. The smaller conversion of Pd-A/TP com-
pared to Pd/TP was again attributed to residual chloride on the
surface due to an insufficient calcination temperature. Con-
secutive reactions of 1a were found in the case of Pt/TP,
Rh/TP and Rh-A/TP.
In particular, with the catalyst Pt/TP, different further re-
duction products were detected (Table 7). Subsequent to the
formation of 1a, the carbonyl group in 1a was reduced giving
1b when the hydrogen pressure was increased from 1.0 to 3.0
(
Scheme 1), in agreement with published results [40]. Under

Achim STOLLE et al. / Chinese Journal of Catalysis, 2011, 32: 1312–1322
References
Table 7 Distribution of products for the hydrogenation of 1 and the
dependence on the hydrogen pressure
Reactant and corresponding
1
2
3
Swift K A D. Top Catal, 2004, 27: 143
Molar fraction at different pH2
Monteiro J L F, Veloso C O. Top Catal, 2004, 27: 169
Pérez-Cadenas A F, Kapteijn F, Ziverink M M P, Moulijn J A.
Catal Today, 2007, 128: 13
products
1.0 MPa
3.0 MPa
0
5.0 MPa
0
1
0.16
0.41
0.43
0
1
1
1
1
1
a
b
c
e
f
0.29
0.62
0.07
0.01
0.01
0.33
0.61
0.03
0.02
0.01
4
Tonetto G M, Sánchez J F, Ferreira M L, Damiani D D. J Mol
Catal A, 2009, 299: 88
5
6
Schmidt A, Schomäcker R. J Mol Catal A, 2007, 271: 192
Schmöger C, Stolle A, Bonrath W, Ondruschka B. Curr Org
Chem, 2011, 15: 151
0
0
Reaction conditions: 7.5 mmol 1, 30 ml ethyl acetate, 0.2 g Pt/TP (0.24
7
Mennecke K, Cecilia R, Glasnov T N, Gruhl S, Vogt C, Feldhoff
A, Vargas M A L, Kappe C O, Kunz U, Kirschning A. Adv Synth
Catal, 2008, 350: 717
mol%), 25 °C, reaction time 90 min.
8
9
Sharma A, Kumar V, Sinha A K. Adv Synth Catal, 2006, 348:
MPa. Simultaneously, dehydration of 1b to 1,3-diphenylpro-
pene (1d) as intermediate on hydrogenation increase to
3
54
Pillai U R, Sahle-Demessi E, Varma R S. J Mater Chem, 2002,
2: 3199
1
,3-diphenylpropane (1e). Furthermore, the addition of hy-
drogen to the phenyl rings was detected yielding
-cyclohexyl-1-phenylpropan-1-one (1c) and 3-cyclohexyl-1-
1
1
0
1
Vanier G S. Synlett, 2007: 131
3
1
Chtourou M, Abdelhedi R, Krikha M H, Trabelsi M. Ultrason
Sonochem, 2010, 17: 246
phenylpropane (1f). In addition, the catalysts Rh/TP and
Rh-A/TP also gave small amounts of these products (6 < 3%).
Due to the fast reduction of the C-C double bond in 1d, neither
this intermediate product nor the completely saturated product
12 Wang H S, Zeng J. Can J Chem, 2009, 9: 1209
13 Balini R, Bosica G, Ricciutelli M, Maggi R, Sartori G, Sartorio
R, Righi P. Green Chem, 2001, 3: 178
1
4
5
Cho C S, Ren W X, Yoon N S. J Mol Catal A, 2009, 299: 117
Trost B M, Xu J Y, Schmidt T. J Am Chem Soc, 2009, 131:
1
,3-dicyclohexylpropane (1g) was detected. The performance
1
of liquid phase hydrogenation at 5.0 MPa and room tempera-
ture revealed that the catalysts Ir/TP, Au/TP, Ru/TP, Fe/TP, and
Ru-A/TP were not active under the applied experimental con-
ditions.
1
8343
1
1
1
1
6
7
8
9
Schmöger C, Stolle A, Bonrath W, Ondruschka B, Keller T, Jandt
K D. ChemSusChem, 2009, 2: 77
Bonrath W, Ondruschka B, Schmöger C, Stolle A. WO
2010/020671 A1. 2010
3
Conclusions
Schmöger C, Gallert T, Stolle A, Ondruschka B, Bonrath W.
Chem Eng Technol, 2011, 34: 445
Porous glass can be applied as a support material in het-
Schmöger C, Szuppa T, Tied A, Schneider F, Stolle A,
Ondruschka B. ChemSusChem, 2008, 1: 339
erogeneous catalysis, especially in the liquid phase hydro-
genation of benzalacetophenone. It can be impregnated with
various metal precursors to give a wide variety of metal cata-
lysts. Reactions under higher pressure and temperature can be
carried out without any change in support parameters like
surface area and porosity. With an increase in pressure from
atmospheric to 5.0 MPa, shorter reaction times are possible.
Some catalysts (Rh/TP and Rh-A/TP) showed catalytic activity
only at higher pressures. The influences of the solvent, metal
precursor, and metal loading with catalyst Pd/TB were inves-
tigated under mild reaction conditions (room temperature,
atmospheric pressure).
20 Li J T, Mau A W H, Strauss C R. Chem Commun, 1997: 1275
21 Enke D, Janowski F, Schwieger W. Microporous Mesoporous
Mater, 2003, 60: 19
2
2
Gille W, Enke D, Janowski F. J Phys Chem Solids, 2003, 64:
2
2
0
9
2
3
Weyrich P A, Trevin H, Hoelderich W F, Sachtler W M H. Appl
Catal A, 1997, 163: 31
2
2
4
5
Feuerriegel U, Klose W. Chem Ing Tech, 1998, 70: 449
Yazawa Y, Yoshida H, Takagi N, Komai S I, Satsuma A, Hattori
T. Appl Catal B, 1998, 19: 261
2
6
7
Huang L, Wang Z, Ang T P, Tan J, Wong P K. Catal Lett, 2006,
112: 219
2
Reddy J K, Durgakumari V, Subrahmanyam M, Sreedhar B.
Mater Res Bull, 2009, 44: 1540
Acknowledgments
2
8
9
Korsvik C, Patil S, Seal S, Self W T. Chem Commun, 2007: 1056
Matsumura Y, Kagawa K, Usami Y, Kawazoe M, Sakurai H,
Haruta M. Chem Commun, 1997: 657
2
We gratefully acknowledge Dr. G. Völksch (Otto-Schott In-
stitute for Glass Chemistry, Friedrich-Schiller University Jena,
Germany) for performing the SEM micrographs. AS is thank-
ful to Dr. N. Theissen (MPI Mühlheim) for the preparation of
the TEM micrographs. TFK thanks Ralf Wagner (IMT) for his
help with the XPS measurements.
30
31
32
Contreras A M, Yan X-M, Kwon S, Bokor J, Somorjai G A.
Catal Lett, 2006, 111: 5
Del C Aguirre M, Reyes R, Oportus M, Melian-Cabrera I, Fierro
J L G. Appl Catal A, 2002, 233: 183
Vanier G. WO 2007/103501 A2. 2007

Achim STOLLE et al. / Chinese Journal of Catalysis, 2011, 32: 1312–1322
3
3
3
4
Gustafsson T, Hedenstrom M, Kihlberg J. J Org Chem, 2006, 71:
911
Heller E, Lautenschläger W, Holzgrabe U. Tetrahedron Lett,
005, 46: 1247
Res, 2009, 344: 1020
1
38 Reyes P, Rojas H, Pecchi G, Fierro J. J Mol Catal A, 2002, 179:
293
2
39 Ning J B, Xu J, Liu J, Lu F. Catal Lett, 2006, 109: 175
40 Moura F C C, dos Santos E N, Lago R M, Vargas M D, Araujo
M H. J Mol Catal A, 2005, 226: 243
3
5
6
Piras L, Genesio E, Ghiron C, Taddei M. Synlett, 2008: 1125
Ma Y M, Wei X Y, Zhou X, Cai K Y, Peng Y L, Xie R L, Zong Y,
Wei Y B, Zong Z M. Energ Fuels, 2009, 23: 638
3
41 Chou J, Franklin N R, Baeck S-H, Jaramillo T F, McFarland E
W. Catal Lett, 2004, 95: 107
3
7
Hsieh Y C, Chir J L, Zou W, Wu H H, Wu A T. Carbohydrate