Facile preparation of supported noble metal nanoparticle catalysts with the aid of templating surfactants in mesostructured materials

Article abstract of DOI:10.1016/j.jcat.2010.07.025

Quaternary ammonium surfactants were generally used as templates in the synthesis of mesoporous materials, and the surfactants were removed by extraction or calcination to give the space of mesopores. Here with the cetyltrimethylammonium bromide templated mesostructured materials, we directly utilized the surfactants as capturing and stabilizing agents to prepare supported noble metal nanoparticle catalysts. The Pd-supported catalysts exhibited high and stable activity in hydrogenation of allyl alcohol (with a remarkably high turnover frequency of 5676 h^-1) and aerobic oxidation of benzyl alcohol. During the recycle use of the catalyst, though surfactants were gradually leached into the reaction medium, the amount of Pd in the catalyst remained constant within experimental error. It is facile and energy-saving to use directly the as-synthesized mesostructured materials as support for noble metal nanoparticles without further calcination or further modification treatment. This method is versatile to prepare other noble metal (such as Au and Pt) supported catalysts by altering the metal precursors.

Full text of DOI:10.1016/j.jcat.2010.07.025

Journal of Catalysis 275 (2010) 140–148
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Facile preparation of supported noble metal nanoparticle catalysts with the aid of
templating surfactants in mesostructured materials
*
Hu Wang, Jin-Gui Wang, Zhu-Rui Shen, Yu-Ping Liu, Da-Tong Ding, Tie-Hong Chen
Institute of New Catalytic Materials Science, Key Laboratory of Advanced Energy Materials Chemistry (MOE), College of Chemistry, Nankai University, Tianjin 300071, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Quaternary ammonium surfactants were generally used as templates in the synthesis of mesoporous
materials, and the surfactants were removed by extraction or calcination to give the space of mesop-
ores. Here with the cetyltrimethylammonium bromide templated mesostructured materials, we
directly utilized the surfactants as capturing and stabilizing agents to prepare supported noble metal
nanoparticle catalysts. The Pd-supported catalysts exhibited high and stable activity in hydrogenation
of allyl alcohol (with a remarkably high turnover frequency of 5676 hꢀ1) and aerobic oxidation of ben-
zyl alcohol. During the recycle use of the catalyst, though surfactants were gradually leached into the
reaction medium, the amount of Pd in the catalyst remained constant within experimental error. It is
facile and energy-saving to use directly the as-synthesized mesostructured materials as support for
noble metal nanoparticles without further calcination or further modification treatment. This method
is versatile to prepare other noble metal (such as Au and Pt) supported catalysts by altering the metal
precursors.
Received 12 May 2010
Revised 24 July 2010
Accepted 27 July 2010
Available online 30 August 2010
Keywords:
Quaternary ammonium surfactant
Hybrid mesostructured materials
Palladium
Allyl alcohol hydrogenation
Aerobic selective oxidation
Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction
toxic and volatile organic solvents, energy-wasting in calcination
and high costs of the reactants, large-scale applications of some re-
Mesoporous materials containing noble metal nanoparticles
have attracted much attention owing to their potential application
in catalysis [1–3]. Many strategies including impregnation [4,5],
deposition–precipitation [6], in situ encapsulation [7–9], chemical
vapor deposition (CVD) [10,11], and surface functionalization
schemes [12–15] were reported to prepare supported metal nano-
particles on mesoporous materials. The impregnation method has
been widely used because of its easy manipulation; however, it re-
sults in uncontrolled growth of metal particles with relatively large
size. CVD is an efficient method for preparing nanoparticles or thin
films, but the precursors are limited, and the devices are compli-
cated. Au-supported MCM-41 was prepared by a modified deposi-
tion–precipitation method with the aid of ethylenediamine and
surfactant-containing support, but a calcination procedure was fol-
lowed to remove the occluded surfactant templates [16]. In the
surface functionalization schemes, functional groups such as thiol
or amino groups can be immobilized onto the inner walls of mes-
opores by grafting process with a silane coupling reagent. The
grafted functional groups act as strong ligands to anchor the corre-
sponding metal ions into the channels of mesoporous materials.
However, due to some disadvantages, such as tedious synthesis
and post-treatment procedures of the supports, employment of
ported method for preparing supported noble metal catalysts are
still limited. Facile, mild and environmentally benign preparation
method is still a challenge.
Cationic surfactants, such as cetyltrimethylammonium bromide
(CTAB), have been prevalently used as templates for the synthesis
of mesoporous materials. Generally, the templating surfactant mi-
celles were removed by calcination or extraction to give the mes-
oporous voids for hosting other guest species. However, in the
respect of metal nanoparticle supporting, removal of the surfac-
tants would be less cost-efficient because alkylammonium ions
have been proved to be good surface-stabilizing agents in the prep-
aration and morphological control of noble metal nanoparticles
[17–19].
Recently, as-synthesized MCM-41 was applied as basic catalyst
in the Knoevenagel and Claisen–Schmidt condensation reactions.
The catalyst showed higher catalytic activity than MCM-41 modi-
fied with Cs2O or aminopropylsilyl group and exhibited long-term
catalytic stability [20]. Kubota et al. used the as-synthesized MCM-
41 catalyst for Knoevenagel condensation and Michael reaction
with high catalytic activity and stability [21,22]. Although the
channels of as-synthesized MCM-41 were filled with surfactants,
the catalyst still exhibited high activity, and the excellent perfor-
mance was ascribed to the basic Si-O- sites presented at the
pore-mouth because the mesopores were occluded by the quater-
nary ammonium surfactants [20,23–25].
* Corresponding author. Fax: +86 22 23507975.
E-mail address: chenth@nankai.edu.cn (T.-H. Chen).
0021-9517/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2010.07.025

H. Wang et al. / Journal of Catalysis 275 (2010) 140–148
141
Here, we report a facile, novel and cost-effective approach to
prepare supported noble metal nanoparticles with as-synthesized
mesostructured materials by utilizing directly the templating sur-
factants as capturing and stabilizing agents. By the adsorption of
noble metal precursors and reduction, the noble metal nanoparti-
cles were confined and stabilized by the alkylammonium surfac-
tants within the mesopores. Because the mesopores were
occluded by the surfactants, the noble metal precursors would
preferentially form complex with the surfactants near the opening
of the mesopores, and after reduction the noble metal nanoparti-
cles would be inclined to locate near the entrance of the mesop-
ores. In hydrogenation of allyl alcohol and aerobic oxidation of
benzyl alcohol, the as-prepared Pd-loaded catalysts exhibited en-
hanced activity and recycling stability.
Gold was deposited on the calcined MCM-41 by homogenous
deposition–precipitation method using urea as the precipitating
agent. In a typical procedure, 0.3 g urea was dissolved in 30 mL
of 1.73 mmol/L HAuCl4 solution at the room temperature. MCM-
41 support is then added to this solution. The temperature of the
resulting slurry was increased gradually to 90 °C and maintained
for 4 h. The solid was filtered, washed several times with distilled
water, dried and then calcined at 300 °C for 4 h in static air. The
catalyst was denoted as Au/M41-DP.
2.4. Catalytic reaction tests
ð1Þ
Catalytic hydrogenation of allyl alcohol (formula 1) was per-
formed in a 50-mL, three-neck, round-bottomed flask at room tem-
perature (30 1 °C). H2 was bubbled through a plastic pipe at the
bottom of the solution at an invariable rate controlled by a valve.
Suspended catalyst in 25 mL of ethanol was bubbled with H2 for
30 min before adding 10 mmol allyl alcohol. In the recycled uses
of the catalysts, 20 mmol allyl alcohol was used in the reaction.
2. Experimental
2.1. Preparation of Pd/as-synthesized MCM-41 catalysts
Preparation of as-synthesized MCM-41(as-M41): In a typical
synthesis, CTAB (0.85 g) was dissolved in a mixture of water
(200 g) and ammonium hydroxide (14.3 g, 30%). Tetraethoxysilane
(4.37 g) was added to this homogeneous solution with vigorous
stirring for 10 min and then heated at 80 °C for 2 h. The resultant
solid was filtered off and then extensively washed with distilled
water and then dried at 120 °C for 24 h. As-synthesized mesostruc-
tured amorphous titania (denoted as-T) and as-synthesized meso-
structured titanium phosphate (denoted as-TP) were synthesized
by quaternary ammonium surfactants as template according to
the published procedure [26,27].
ð2Þ
The liquid phase oxidation of benzyl alcohol (formula 2) was
performed in a three-necked flask with a reflux condenser under
magnetic stirring. The catalyst was added into the benzyl alcohol,
and the mixture was heated by oil bath to reaction temperature,
and O2 flow (25 ml/min) was bubbled into the liquid.
Preparation of Pd/as-M41: Typically, 1 g as-M41 was suspended
in 25 mL aqueous H2PdCl4 solution of desired concentration and
stirred at room temperature for 30 min. The resulted precipitate
was filtered and washed with distilled water for several times.
Without further drying procedure, the obtained light orange pow-
der was reduced with a freshly prepared solution of NaBH4 (0.1 M)
at room temperature. The sample was filtered, washed extensively
with distilled water, and dried at 60 °C to obtain black power (de-
noted as Pd/as-M41). We also prepared Au/as-M41 and Pt/as-M41
catalysts by the same method with HAuCl4 and H2PtCl6 solution as
the noble metal precursors, respectively.
O
O
C
CH
CH2
TBHP
CH
CH2
H
+
else
+
ð3Þ
The selective oxidation of styrene (formula 3) was performed in
a three-necked flask with a reflux condenser under magnetic stir.
The Au-supported catalyst (100 mg) was added to the toluene
(10 mL) solution containing 40 mg of styrene and 200 mg TBHP
(tert-butyl hydroperoxide). Dodecane (30 mg) was added as inter-
nal standard substance. The mixture was heated by oil bath to
80 °C.
After the reactions, the catalyst was separated by centrifuga-
tion, and the liquid products were analyzed by a gas chromato-
graph (GC-7800) with a flame ionization detector, using a FFAP
column and N2 as carrier gas.
2.2. Preparation of diamine-functionalized MCM-41 supported Pd
catalyst
The as-M41 was calcined at 550 °C in air for 5 h to remove the
surfactant. One gram of calcined MCM-41 was stirred vigorously in
50 mL dry toluene containing 1.0 mmol of N-[3-(trimethoxysilyl)
propyl] ethylenediamine. This solution was heated to 110 °C for
10 h. The powder was collected by filtration, washed with 2-propa-
nol and dried at 100 °C. The obtained functionalized MCM-41 was
denoted as F-M41. Pd/F-M41 catalyst was prepared in a similar
way of the Pd/as-M41 catalyst preparation.
2.5. Characterization methods
The Pd (or Au) content was determined by inductively coupled
plasma (ICP) optical emission spectrometry using a Thermo Jarrell-
Ash ICP-9000(N + M). Powder X-ray diffraction (XRD) patterns
were recorded on a Bruker D8 Focus with Cu Ka radiation (40 kV,
40 mA). Transmission electron microscopy (TEM) and energy-dis-
persive X-ray spectroscopy (EDX) measurements were taken using
a Philips Tecnai F20 microscope. All samples subjected to TEM
measurements were ultrasonically dispersed in the ethanol and
dropped on copper grids. X-ray photoelectron spectroscopy (XPS)
measurements were taken with a Kratos Axis Ultra DLD spectrom-
2.3. Preparation of catalysts by methods of wetness impregnation and
deposition–precipitation
For comparison, palladium was deposited on the MCM-41 by
wetness impregnation method. For the 0.5%Pd/M41-imp catalyst,
an aqueous solution of H2PdCl4 (4 mL, 0.012 mol/L) was added to
the MCM-41 (1.0 g). The resulting powder was dried at 110 °C for
24 h and calcined in air at 500 °C for 6 h. The catalyst was denoted
as Pd/M41-imp.
eter employing a monochromated Al
Ka X-ray source (hm =
1486.6 eV). Thermogravimetry (TG) was performed with a Rigaku
thermogravimetry–differential thermal analysis analyzer in air
with the heating rate of 20 K/min.

142
H. Wang et al. / Journal of Catalysis 275 (2010) 140–148
based on ICP measurement, and this value was in good agreement
3. Results and discussion
3.1. Preparation of Pd/as-M41 samples
with the designed loading of 1.0% in the PdCl24ꢀ adsorption process.
Without calcination, modification or using organic solvent, the
preparation method reported here is facile and energy-saving. This
method is versatile to prepare other supported nanosized noble
metal catalysts by altering the metal precursors, such as Au/as-
M41 and Pt/as-M41. Meanwhile, other mesostructured materials,
such as titania and titanium phosphate, were synthesized tem-
plated by quaternary ammonium surfactants to support noble me-
tal nanoparticles by the same scheme.
As shown in Scheme 1, at room temperature the surfactants in
the channels of as-M41 adsorbed PdCl24ꢀ ions specifically and
quickly from the aqueous solution of H2PdCl4. After reduction by
NaBH4 solution at room temperature, well-dispersed Pd nanoparti-
cles were formed and hosted by the as-M41. During the prepara-
tion process, when the as-M41 was added into H2PdCl4 solution
under stirring, within several minutes the color of as-M41 changed
from white to light orange (Fig. 1). After filtration, the filtrate be-
came transparent and colorless. For the case of 1.0% Pd loading,
ICP analysis indicated that the concentrations of Pd in the starting
solution and in the filtrate were 380 ppm and 2.7 ppm, respec-
tively, indicating that over 99% of the total PdCl24ꢀ ions were ad-
sorbed in as-M41. The coordination between the quaternary
ammonium cations and PdCl42ꢀ anions would be responsible for
the rapid adsorption of PdCl24ꢀ. When the calcined MCM-41 was
used in the above procedure, it did not adsorb PdCl24ꢀ and the color
of the solution and the powder did not change.
3.2. Characterization of Pd/as-M41 samples
As shown in Fig. 2, the low-angle XRD patterns of as-M41, 1%Pd/
as-M41 before reduction and 1%Pd/as-M41 after reduction dis-
played three diffraction peaks of 2D hexagonal mesostructure,
respectively. The wide-angle XRD pattern of 1%Pd/as-M41
(Fig. 3a) did not display diffraction peaks of Pd, implying that the
size of Pd nanoparticles could be too small to be detected by
XRD. If the Pd/as-M41 was calcined at 500 °C for 2 h, PdO particles
in relatively larger size were formed and gave rise to the diffraction
peaks in the XRD pattern (Fig. 3b).
It is conceivable that the PdCl24ꢀ ions would coordinate mainly
with the CTA+ groups located at the opening of the meso-channels
(pore-mouth). The penetration of the PdCl24ꢀ ions deep inside the
pore would be less favorable because the mesopores were oc-
cluded with surfactants, and there were enough CTA+ groups pres-
ent near the entrance of pore to host the PdCl24ꢀ ions. Therefore, the
Pd nanoparticles formed after reduction would locate near the
pore-mouth of the mesopores. Furthermore, the Pd nanoparticles
were confined and stabilized by the surfactant micelles in the mes-
opores, and thus the size of the nanoparticles would be small. The
loading of palladium in the catalyst after reduction was 1.09%
c
b
a
2
3
4
5
6
2 Theta/degree
Scheme 1. Illustration of the preparation of Pd/as-M41 catalyst. RT means room
Fig. 2. Low-angle XRD patterns of (a) as-M41, (b) Pd/as-M41 before reduction, (c)
temperature.
Pd/as-M41 after reduction.
Fig. 1. Photographs of preparation procedure of CTAB-templated mesostructured materials supported noble metal nanoparticle catalysts.

H. Wang et al. / Journal of Catalysis 275 (2010) 140–148
143
the Pd nanoparticles would result in low contrast in the TEM
images so they could not be clearly distinguished from the surfac-
tant/silica matrix. Similar phenomenon was also reported by Lu
et al. for supported Pd clusters in mesoporous carbon, and due to
the small particle size no visible Pd clusters could be found in
the carbon framework by TEM analysis [28]. In the TEM images
with higher magnification, occasionally only a few Pd nanoparti-
cles could be identified which were located inside the mesopores
(Fig. 4c and d, indicated by arrows). The dispersion of palladium
clusters inside the as-M41 could be proved by EDX analysis. As
shown in Fig. 5 and Fig. S1 (in Supporting Information), though
there were no observable Pd nanoparticles in TEM images, Pd sig-
nal could be well detected in the EDX spectra collected from differ-
ent local areas. To further prove the presence of Pd clusters in the
catalyst, the Pd/as-M41 was calcined and TEM revealed that many
Pd (actually PdO) nanoparticles appeared on the external surface of
the MCM-41 particles (Fig. 6), due to the aggregation and out-
transfer of the dispersed Pd clusters during burn of the surfactants.
In this case, the diffraction peaks of PdO appeared in the wide-an-
gle XRD pattern (Fig. 3b). According to above results, in the Pd/as-
M41 catalyst the size of the Pd nanoparticles (or clusters) was
small and well dispersed.
a
b
*
*
40
10
20
30
50
60
2 Theta/degree
Fig. 3. Wide-angle XRD patterns of (a) Pd/as-M41 and (b) Pd/as-M41 after
calcination at 500 °C. The asterisks indicate the diffraction peaks of PdO.
The dispersion of the Pd nanoparticles in Pd/as-M41 catalyst
was analyzed by TEM analysis. As shown in Fig. 4, at low magnifi-
cation (Fig. 4a and b) there were hardly any visible Pd nanoparti-
cles, indicating that the Pd nanoparticles could be small, in
agreement with the result of the XRD pattern. The small size of
The confinement effect of the mesopores on Pd nanoparticles
during the reduction of the surfactant-hosted PdCl24ꢀ could be
proved by a control experiment. PdCl24ꢀ/CTAB solution was pre-
pared, and after NaBH4 reduction a homogenous black solution
was obtained. The Pd particles were then observed by TEM, and
Fig. 4. TEM image of 1.0%Pd/as-M41, the arrows indicate a few observable Pd nanoparticles.

144
H. Wang et al. / Journal of Catalysis 275 (2010) 140–148
Fig. 5. TEM images and EDX spectrum (collected from the circle area) of the 1.0% Pd/as-M41 catalyst. The arrow in (b) indicates the area deformed by the irradiation of the
electron beam after EDX measurement.
Fig. 6. TEM images of Pd/as-M41catalyst after calcination.
the average particle size was about 8 nm (Fig. S2, Supporting Infor-
mation), which was much larger than that of the Pd nanoparticles
located within the as-M41.
Before reduction of the sample, the XPS spectrum of Pd 3d
(Fig. 7a) displayed two peaks at 342.3 eV (Pd 3d3/2) and 336.9 eV
(Pd 3d5/2), which were ascribed to Pd2+. Fig. 7b shows the Pd 3d

H. Wang et al. / Journal of Catalysis 275 (2010) 140–148
145
reaction, the conversion was only 15% but with 100% selectivity to-
ward 1-propyl alcohol. After 540 min of reaction, the conversion
reached 100% with an excellent selectivity of 99% and a good TOF
a
b
of 867 hꢀ1
.
In general, surface functionalization is a prevalent strategy for
the preparation of supported Pd/MCM-41 catalyst [31]. The inter-
nal surface of MCM-41 was generally modified by grafting silane
coupling agents. For comparison, diamine-functionalized MCM-
41 supported Pd catalyst (0.5%Pd/F-M41) was prepared and used
in the hydrogenation reaction of allyl alcohol. After 45 min of reac-
tion, the conversion was only 64%. It took longer time (70 min)
than that for 0.5%Pd/as-M41 (45 min) when the conversion
reached 100%. The lower activity of the 0.5%Pd/F-M41 sample
could be attributed to the longer diffusion path for the reactants
to reach the active sites, which located deeper inside the mesop-
ores. Because hydrogen was involved in the liquid reaction system,
the gas phase would be difficult to diffuse into the meso-channel.
In contrast, the high activity of 0.5%Pd/as-M41 catalyst in the
hydrogenation reaction of allyl alcohol could be attributed to the
active sites presented near the pore-mouth.
3
4
6
3
4
4
3
4
2
3
4
0
3
3
8
3
3
6
334
Binding Energy (eV)
346
344
342
340
338
336
334
Aerobic oxidation of alcohols has been intensively studied in re-
cent years due to its economic and environmental advantages, and
different noble metal supported catalysts were tested on this reac-
tion [32]. Kaneda and coworkers [33,34] reported that phenyleth-
anol and benzyl alcohol could be oxidized by hydroxyapatite
supported Pd nanoparticles with high turnover frequencies. Hutch-
ings and coworkers [35,36] showed that TiO2-supported Au–Pd al-
loy nanocrystals give significantly enhanced activity for alcohol
oxidation compared with monometallic Au or Pd catalysts. Prati
and coworkers [37,38] also investigated alloy effect of Au–Pd and
Au–Pt catalysts in alcohol oxidation. Corma and coworkers have
shown that gold nanoparticles transform nanocrystalline cerium
oxide from a stoichiometric oxidant into a catalytic material for
the oxidation of alcohols with high TOFs and selectivities [39–
41]. Here, the catalytic activities of the Pd/as-M41 catalysts were
measured by aerobic oxidation of benzyl alcohol, and the results
were listed in Table 2. The main product was benzaldehyde; how-
ever, further oxidation of benzaldehyde could occur to give benzoic
acid, and benzylbenzoate was produced by esterification of benzoic
acid and benzyl alcohol. The catalysts exhibited good activities un-
der base- and solvent-free condition. The 0.5%Pd/as-M41 gave a
high TOF (>1000 hꢀ1). Comparison with the results between Entry
1, 7, and 8, the 0.5%Pd/F-M41 and 0.5%Pd/M41-imp catalysts pre-
sented lower activity than that of the 0.5%Pd/as-M41 sample in
oxidation of benzyl alcohol.
Binding Energy (eV)
Fig. 7. Pd 3d XPS spectra of 1.0%Pd/as-M41: (a) before NaBH4 reduction and (b)
after reduction.
XPS spectrum of Pd/as-M41, and the peaks at 340.8 eV (Pd 3d3/2
)
and 335.5 eV (Pd 3d5/2) were ascribed to Pd0 in the catalyst. The
shoulders at 342.7 eV (Pd 3d3/2) and 337.3 eV (Pd 3d5/2) were
due to surface oxidation of the palladium nanoparticles upon expo-
sure in air. Fig. S3a (Supporting Information) shows the XPS spec-
trum of Cl 2p of Pd/as-M41 before reduction, in which obvious Cl
2p signal was observed. The formation of [CTA+]2[PdCl2ꢀ] complex
4
has been verified by other literature [17,29]. In the XPS spectrum of
Pd/as-M41, no Cl could be detected (Fig. S3b), indicating that the
Clꢀ anions were washed out of the catalyst after reduction.
3.3. Catalytic experiments
Using 0.5%Pd/as-M41 catalyst as a representative, catalytic
activity was tested by a model hydrogenation reaction of allyl alco-
hol to 1-propyl alcohol, and the results were listed in Table 1. A
blank experiment carried out without catalyst showed no conver-
sion, even after 12 h of hydrogen treatment. The 0.5%Pd/as-M41
catalyst gave a remarkably high turnover frequency (TOF) (up to
5676 hꢀ1) with the 100% conversion for 45 min. For the 0.5
wt%Pd/MCM4-imp prepared by impregnation method, the conver-
sion was only 80% for 45 min. For comparison, the TOF was
2944 hꢀ1 for the Pd/diamine-functionalized mesopolymer working
under the same reaction condition [30]. The reaction speed was
relatively slow with Pt/as-M41 as catalyst, and after 45 min of
In the Knoevenagel and Claisen–Schmidt condensation reac-
tions, the as-synthesized MCM-41 catalyst exhibited higher cata-
lytic activity than that of the calcined MCM-41 modified with
Table 2
Performance of the catalysts in the aerobic oxidation of benzyl alcohol.a
Table 1
Entry
Catalyst
Time (h)
Conv. (%)
Selec.b (%)
TOFc (hꢀ1
)
Performance of the catalysts in the hydrogenation of allyl alcohol.a
1
2
0.5%Pd/as-M41
1.0%Pd/as-M41
0.5%Pd/as-TP
1.0%Au/as-TP
0.5%Pd/as-T
1.0%Au/as-T
0.5%Pd/F-M41
0.5%Pd/M41-imp
6
5
6
6
6
6
6
6
79
84
78
40
77
63
31
16
92
90
92
91
90
60
87
92
1359
867
1341
637
1325
1003
533
Entry Catalyst
Time (min) Conv. (%) Selec.b (%) TOFc (hꢀ1
)
1
2
3
4
6
7
8
0.5%Pd/as-M41
45
30
100
100
100
100
100
64
78
80
99
75
78
74
80
5676
4257
867
3
4d
5
1.0%Pd/as-M41
0.5%Pt/as-M41
0.5%Pd/as-TP
0.5%Pd/as-T
0.5%Pd/F-M41
0.5%Pd/M41-imp
540
45
45
45
45
6d
7
5676
5676
3632
4541
8
275
80
O2 flow, 25 mL minꢀ1; 100 °C; solvent-free reactions; catalyst, 0.1 g; benzyl
alcohol, 48.5 mmol.
The by-products were benzoic acid and benzylbenzoate.
The TOF was calculated by the moles of substrates converted per mole of Pd per
hour.
a
H2 flow, 40 mL minꢀ1; 30 °C; 25 mL ethanol as solvent; catalyst, 0.05 g; allyl
alcohol, 10 mmol.
The by-product was acetone.
The TOF was calculated by the moles of substrates converted per mole of Pd per
hour.
a
b
b
c
c
d
The Entry 4, 6, K2CO3, 0.1 g.

146
H. Wang et al. / Journal of Catalysis 275 (2010) 140–148
Table 4
Cs2O or aminopropylsilyl group. Because the basic sites present in-
side the channels were inaccessible for the reactants, the high
activity was attributed to the active sites present at the pore-
mouth [20]. In our case, considering the different catalytic perfor-
mance in the two model reactions between 0.5%Pd/as-M41 and
0.5%Pd/F-M41 catalysts, the latter suffered from the disadvantage
of prolonged reaction time and lower activity. This result might
be due to different location of Pd nanoparticles in the two catalysts.
The active sites near the pore-mouth would exhibit better perfor-
mance than those located in the interior of the mesopores. The
resistance of mass transport was eliminated when the reaction
took place at pore-mouth. Therefore, the reactant (benzyl alcohol
or allyl alcohol) could be easy to reach the Pd active sites. In addi-
tion, when gas phase was involved in the liquid reaction system,
the gas was difficult to diffuse into the meso-channels, and this
is another factor resulting in lower catalytic performance of ami-
no-functionalized M41 supported Pd catalyst.
Recycling performance of the 0.5%Pd/as-M41a catalyst in allyl alcohol hydrogenation
and solvent-free aerobic oxidation of benzyl alcohol.
Catalytic cycle
Hydrogenation of allyl
alcoholb
Oxidation of benzyl alcoholc
Conv. (%)
Selec. (%)
Conv. (%)
Selec. (%)
1
2
3
4
5
6
7
8
67
65
68
67
65
62
64
66
78
77
75
79
75
78
76
77
78
76
80
79
79
77
80
78
92
93
90
92
92
93
90
91
a
The loading of Pd in the catalyst was 0.54% by ICP analysis.
Reaction condition: H2 flow, 40 mL minꢀ1; 30 °C; 25 mL ethanol as solvent;
b
catalyst, 0.05 g; allyl alcohol, 20 mmol; time, 45 min.
c
Reaction condition: O2 flow, 25 mL minꢀ1; 100 °C; solvent-free reactions; cat-
alyst, 0.1 g; benzyl alcohol, 48.5 mmol; time, 6 h.
Supported gold catalysts are promising in many reactions such
as aerobic oxidation of alcohol [39–41], selective oxidation of sty-
rene [42,43], and propylene epoxidation with O2 and H2 [44–46].
Here, the catalytic performance of 0.25% Au/as-M41 was tested
by selective oxidation of styrene in toluene at 80 °C with anhy-
drous TBHP as oxidant, and the results were listed in Table 3.
The catalytic data of Au/as-M41, Au/as-T, and Au/as-TP were more
or less the same, representing the catalytic behavior of the surfac-
tant-hosted gold nanoparticles. The conversion of styrene was sim-
ilar to that by Au25 clusters on hydroxyapatite at the same reaction
condition [42]. However, the selectivities of products were close to
the result of thiolate-protected Au nanoclusters catalyst with O2 as
the oxidant [43]. The catalyst prepared by deposition–precipitation
method (0.25%Au/M41-DP) was also tested, and it exhibited lower
conversion than that of 0.25%Au/as-M41.
alcohol conversion and benzaldehyde selectivity were essentially
constant within experimental error (Table 3). When the CTAB-tem-
plated titania was used as the substrate (0.5%Pd/as-T), benzyl alco-
hol conversion and benzaldehyde selectivity were essentially
constant within experimental error (Table S1 in Supporting
Information).
The Pd/as-M41 catalyst was an inorganic–organic hybrid meso-
structured material, and the possible leaching of CTA+ surfactants
and Pd nanoparticles during the reactions must be investigated,
even though the catalyst exhibited stable activity in repeated uses.
As reported by other groups, who used as-synthesized MCM-41 as
catalyst in the Michael addition and Knoevenagel condensation
reaction [21,25], partial leaching of the surfactants was observed
in the reaction process. Kubota et al. [21] used as-synthesized
MCM-41 in Michael addition reaction at 30 or 80 °C with benzene
as solvent, and the weight loss after third run was less than 2% or
5% of the initial weight, respectively. According to Cardoso and
coworkers, when as-synthesized MCM-41 was used in the Knoeve-
nagel condensation reaction at 50 °C in the presence of toluene as
solvent, the loss of mass was 53.5% in the recovered catalyst after
the fourth use, while the value for the fresh catalyst was 54.2%
[25]. Here to gain better insight into the stability of Pd/as-M41 cat-
alyst, TG and ICP analyses were used to determine the weight loss
and palladium content in the fresh and reused catalysts. Though
the 0.5%Pd/as-M41 catalyst exhibited good activity in allyl alcohol
hydrogenation and benzyl alcohol selective oxidation reactions, in
order to enhance the veracity of the analysis results, 1.0%Pd/as-
M41 sample was selected on purpose in the quantitative analysis.
The stable performance after 8 runs of recycling tests for the
1.0%Pd/as-M41 catalyst was displayed in Table S2 in Supporting
Information.
As shown in Table 5, the weight loss of the fresh 1.0%Pd/as-M41
catalyst was 35.8%, which was assigned to the decomposition of
CTA+ surfactants. The loading of palladium was 1.09% for the fresh
catalyst by ICP measurement (Table 5, Entry 1). In the reaction of
allyl alcohol hydrogenation, the weight loss of the reused catalyst
became 30.7% and 27.3% after recycling uses for 4 and 8 times,
respectively. For aerobic oxidation of benzyl alcohol, the weight
loss of the reused catalyst was 33.3%, 22.2%, and 21.4% after recy-
cling uses for 1, 4, and 8 times, respectively. The continuous weight
loss of the catalysts indicated gradual leaching of the surfactants
during the reactions. However, the amount of palladium in the
recovered catalyst became 1.18% and 1.22% after 4 and 8 runs in
hydrogenation, and 1.12%, 1.34%, and 1.37% after 1, 4, and 8 runs
in oxidation, respectively. It is not surprising to see the increasing
Pd content in the reused catalysts, due to the gradual leaching of
3.4. Stability tests of the catalysts
In order to verify the catalytic stability of Pd/as-M41, the cata-
lyst was repeatedly used in the hydrogenation of allyl alcohol and
aerobic oxidation of benzyl alcohol, respectively. In hydrogenation
of allyl alcohol, the substrate was doubled so that the conversion
was less than 100%, and by this way the effect of the recycled cat-
alysts could be compared. After one run of catalytic reaction, the
catalyst was separated from the reaction mixture by centrifuga-
tion, thoroughly washed with ethanol, and then reused as catalyst
for next run under the same condition. As shown in Table 4, the
catalyst remained similar activity and selectivity after eight runs,
indicating the reusability and stability of the catalyst in hydroge-
nation reaction of allyl alcohol. For the 0.5%Pd/as-M41 catalyst,
the conversion was 67% after 45 min, and the TOF became as high
as 7606 hꢀ1
.
In the recycling experiments of aerobic oxidation of benzyl alco-
hol, 0.5%Pd/as-M41 catalyst exhibited high stability and reusabil-
ity. After recycling use of the catalyst for eight times, the benzyl
Table 3
Performance of the Au catalysts in selective oxidation of styrene with TBHP as
oxidant.a
Catalysts
Conv. (%)
Selec. (%)
Styrene epoxide
Benzaldehyde
Elseb
0.25%Au/as-M41
0.25%Au/T
0.25%Au/TP
93
90
89
70
22
25
28
7
73
68
66
88
5
7
6
5
0.25%Au/M41-DP
a
TBHP, 200 mg; 80 °C; 10 mL toluene as solvent; catalyst, 0.1 g; styrene, 40 mg;
reaction time, 10 h.
b
Else products include benzoic acid and acetophenone.

H. Wang et al. / Journal of Catalysis 275 (2010) 140–148
147
Table 5
eight consecutive reactions for two model reactions. The as-syn-
thesized mesostructured materials have a great advantage as the
support of the noble metal nanoparticles, because they can be used
as support directly, without further calcination or modification.
Due to the facile preparation process and high catalytic perfor-
mance, the hybrid catalysts would be promising in practical appli-
cations at relative low temperature.
Weight loss by TG and Pd content of 1.0%Pd/as-M41 catalysts before and after
reaction.
Loss of mass (%)a
Pd (%)b
Pd(%)-SiO2
c
Entry
Catalyst
1
2
3
4
5
6
1.0%Pd/as-M41
35.8
30.7
27.3
33.3
22.2
21.4
1.09
1.18
1.22
1.12
1.34
1.37
1.70
1.70
1.68
1.68
1.72
1.74
1.0%Pd/as-M41-H-4d
1.0%Pd/as-M41-H-8
1.0%Pd/as-M41-O-1e
1.0%Pd/as-M41-O-4
1.0%Pd/as-M41-O-8
Acknowledgments
a
b
c
Weight loss from TG curve between 150–600 °C.
Pd weight percent in the catalyst detected by ICP measurement.
Calculated Pd weight percent in the catalyst by deduction of the weight loss of
This work was supported by NSFC (20873070, 20973095),
National Basic Research Program of China (2009CB623502), RFDP
(200800550007), NCET of Ministry of Education (NCET-07-0448)
and MOE (IRT-0927).
the organic species. The oxidation of Pd and dehydroxylation of the silica during the
TG measurements were not taken into account.
d
H-n: the catalyst was reused in hydrogenation of allyl alcohol for n times.
O-n: the catalyst was reused in aerobic oxidation of benzyl alcohol for n times.
e
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jcat.2010.07.025.
the surfactants. For easy comparison, the Pd content was calibrated
by deduction of the weight loss of the organic species, and the cal-
culated Pd content values were listed in Table 5. It could be clearly
found that the Pd content in the reused catalysts remained un-
changed within experimental error, indicating that there was al-
most no Pd leaching in the reaction system even after repeated
uses. This point was in good agreement with the experiment, i.e.,
when extra allyl alcohol (10 mmol) was added into the filtrate after
extraction of the catalyst, no further conversion occurred after
45 min at the same reaction condition as was detected by GC. This
indicated that no observable leaching of the Pd nanoparticles into
the solution during the reaction.
References
[1] J.M. Campelo, D. Luna, R. Luque, J.M. Marinas, A.A. Romero, ChemSusChem 2
(2009) 18.
[2] R.J. White, R. Luque, V.L. Budarin, J.H. Clark, D.J. Macquarrie, Chem. Soc. Rev. 38
(2009) 481.
[3] A. Taguchi, F. Schueth, Micropor. Mesopor. Mater. 77 (2005) 1.
[4] X. Chen, H.Y. Zhu, J.C. Zhao, Z.F. Zheng, X.P. Gao, Angew. Chem. Int. Ed. 47
(2008) 5353.
[5] A. Fukuoka, H. Araki, J. Kimura, Y. Sakamoto, T. Higuchi, N. Sugimoto, S. Inagaki,
M. Ichikawa, J. Mater. Chem. 14 (2004) 752.
[6] P. Haider, A. Baiker, J. Catal. 248 (2007) 175.
As discussed above, in the Pd/as-M41 the Pd nanoparticles were
stabilized by the CTA+ surfactants. The gradual loss of the surfac-
tants during the catalytic reactions would probably result in the
partial leaching of the Pd nanoparticles; however, the Pd content
in the catalyst remained unchanged. To explain this point, the
black solution of CTAB-stabilized Pd nanoparticles was prepared
as described previously. When proper amount of as-M41 (corre-
sponding to 1 wt% Pd loading) was added in the solution, it was
found that the CTAB-stabilized Pd nanoparticles were adsorbed
by as-M41, and the black solution became colorless (Fig. S4 in Sup-
porting Information). This implies though some Pd nanoparticles
could be leached together with the surfactants from the Pd/as-
M41 during the catalytic reactions, those disassociated surfac-
tant-stabilized Pd nanoparticles in the reaction solvent could be
adsorbed back by the catalyst, while the extracted surfactants were
irreversibly lost. This could explain why the loss of the surfactants
would not influence the stability of the Pd nanoparticles in the
catalyst.
[7] U. Junges, W. Jacobs, I. Voigt-Martin, B. Krutzschc, F. Schueth, J. Chem. Soc.
Chem. Commun. (1995) 2283.
[8] P. Krawiec, E. Kockrick, P. Simon, G. Auffermann, S. Kaskel, Chem. Mater. 18
(2006) 2663.
[9] A.K. Prashar, R.P. Hodgkins, R. Kumar, R.N. Devi, J. Mater. Chem. 18 (2008)
1765.
[10] P. Serp, P. Kalck, R. Feurer, Chem. Rev. 102 (2002) 3085.
[11] T. Ishida, M. Nagaoka, T. Akita, M. Haruta, Chem. Eur. J. 14 (2008) 8456.
[12] J. Sun, D. Ma, H. Zhang, X. Liu, X. Han, X. Bao, G. Weinberg, N. Pfander, D. Su, J.
Am. Chem. Soc. 128 (2006) 15756.
[13] C. Yang, H. Shen, K. Chao, Adv. Funct. Mater. 12 (2002) 143.
[14] L. Li, J.L. Shi, L.X. Zhang, L.M. Xiong, J.N. Yan, Adv. Mater. 16 (2004) 1079.
[15] C. Wang, G. Zhu, J. Li, X. Cai, Y. Wei, D. Zhang, S. Qiu, Chem. Eur. J. 11 (2005)
4975.
[16] M.P. Mokhonoana, N.J. Coville, A.K. Datye, Catal. Lett. 135 (2010) 1.
[17] H. Lee, S.E. Habas, S. Kweskin, D. Butcher, G.A. Somorjai, P. Yang, Angew. Chem.
Int. Ed. 45 (2006) 7824.
[18] T.K. Sau, C.J. Murphy, J. Am. Chem. Soc. 126 (2004) 8648.
[19] H. Wang, S. Deng, Z. Shen, J. Wang, D. Ding, T. Chen, Green Chem. 11 (2009)
1499.
[20] L. Martins, W. Holderich, P. Hammer, D. Cardoso, J. Catal. 271 (2010) 220.
[21] Y. Kubota, H. Ikeya, Y. Sugi, T. Yamada, T. Tatsumi, J. Mol. Catal. A: Chem. 249
(2006) 181.
[22] Y. Kubota, Y. Nishizaki, H. Ikeya, M. Saeki, T. Hida, S. Kawazu, M. Yoshida, H.
Fujii, Y. Sugi, Micropor. Mesopor. Mater. 70 (2004) 135.
[23] A.C. Oliveira, L. Martins, D. Cardoso, Micropor. Mesopor. Mater. 120 (2009)
206.
4. Conclusions
[24] L. Martins, D. Cardoso, Micropor. Mesopor. Mater. 106 (2007) 8.
[25] L. Martins, T.J. Bonagamba, E.R. Azevedo, P. Bargiela, D. Cardoso, Appl. Catal. A:
Gen. 312 (2006) 77.
[26] T. Jiang, Q. Zhao, M. Li, H. Yin, J. Hazar. Mater. 159 (2008) 204.
[27] A. Bhaumik, S. Inagaki, J. Am. Chem. Soc. 123 (2001) 691.
[28] A.H. Lu, W.C. Li, Z. Hou, F. Schuth, Chem. Commun. (2007) 1038.
[29] A.K. Prashar, R.P. Hodgkins, J.N. Chandran, P.R. Rajamohanan, R.N. Devi, Chem.
Mater. 22 (2010) 1633.
[30] R. Xing, Y. Liu, H. Wu, X. Li, M. He, P. Wu, Chem. Commun. (2008) 6297.
[31] B. Karimi, A. Zamani, S. Abedi, J.H. Clark, Green Chem. 11 (2009) 109.
[32] T. Mallat, A. Baiker, Chem. Rev. 104 (2004) 3037.
[33] K. Mori, K. Yamaguchi, T. Hara, T. Mizugaki, K. Ebitani, K. Kaneda, J. Am. Chem.
Soc. 124 (2002) 11572.
In summary, a facile method was presented to prepare sup-
ported noble metal catalysts by using as-synthesized mesostruc-
tured materials as the host. Other than the role of templating for
mesopores, the surfactants played a new role of capturing and sta-
bilizing agents to prepare supported noble metal nanoparticles.
The specific interaction between the alkylammonium surfactants
and PdCl24ꢀ at room temperature offered an easy control of Pd load-
ings. The Pd/as-M41 catalyst exhibited higher catalytic activity in
hydrogenation of allyl alcohol and aerobic oxidation of benzyl alco-
hol reactions than that of the amino-functionalized MCM-41 sup-
ported Pd catalyst. Partial leaching of surfactants was observed in
the recycling uses of the catalyst. However, the catalyst still
showed high activity because no obvious leaching of Pd was ob-
served in the process. The result was confirmed by the recycling
uses of the catalyst, which exhibited stable catalytic activity in
[34] K. Mori, T. Hara, T. Mizugaki, K. Ebitani, K. Kaneda, J. Am. Chem. Soc. 126
(2004) 10657.
[35] D.I. Enache, J.K. Edwards, P. Landon, B.S. Espriu, A.F. Carley, A.A. Herzing, M.
Watanabe, C.J. Kiely, D.W. Knight, G.J. Hutching, Science 311 (2006) 362.
[36] N. Dimitratos, J.A. Lopez-Sanchez, D. Morgan, A.F. Carley, R. Tiruvalam, C.J.
Kiely, D. Bethell, G.J. Hutchings, Phys. Chem. Chem. Phys. 11 (2009) 5142.
[37] N. Dimitratos, A. Villa, D. Wang, F. Porta, D. Su, L. Prati, J. Catal. 244 (2010) 113.

148
H. Wang et al. / Journal of Catalysis 275 (2010) 140–148
[38] A. Villa, N. Janjic, P. Spontoni, D. Wang, D. Su, L. Prati, Appl. Catal. A 364 (2009)
221.
[43] Y. Zhu, H. Qian, M. Zhu, R. Jin, Adv. Mater. 22 (2010) 1915.
[44] N.S. Patil, B.S. Uphade, P. Jana, S.K. Bharagava, V.R. Choudhary, J. Catal. 223
[39] A. Abad, P. Concepcion, A. Corma, H. Garcia, Angew. Chem. Int. Ed. 44 (2005)
4066.
(2004) 236.
[45] J. Huang, T. Takei, T. Akita, H. Ohashi, M. Haruta, Appl. Catal. B 95 (2010)
[40] A. Abad, C. Almela, A. Corma, H. Garcia, Chem. Commum. (2006) 3178.
[41] A. Abad, A. Corma, H. Garcia, Chem. Eur. J. 14 (2008) 212.
[42] Y. Liu, H. Tsunoyama, T. Akita, T. Tsukuda, Chem. Commun. 46 (2010) 550.
430.
[46] T. Liu, P. Hacarlioglu, S.T. Oyama, M.F. Luo, X.R. Pan, J.Q. Lu, J. Catal. 267 (2009)
202.