H. Gu et al. / Catalysis Communications 41 (2013) 65–69
67
The AuPd-OMC-1 and AuPd-OMC-2 samples show three diffrac-
tion lines positioned at an angle 2θ = 39.9ο, 46.4ο, and 67.8ο, which
could be indexed as Pd (111), (200), and (220) reflections. No
peaks indexed as Au or alloy reflections were found in AuPd-OMC-1
and AuPd-OMC-2 samples. It is reported that the diffraction peak cen-
tered at 2θ value intermediate between the Au (111) and the Pd
(111) is an indication of alloy formation [10]. Besides the predomi-
nant presence of separate dispersed Au and Pd NPs, the presence of
Au-Pd alloy NPs should also be taken into consideration. Moreover,
compared to the Pd-OMC sample, AuPd-OMC-1 and AuPd-OMC-2
samples show wider peaks, suggesting the presence of smaller NPs.
In contrast, the AuPd/OMC sample has sharp peaks positioned at an
angle 2θ = 38.2ο, indicating the formation of large Au NPs. No evi-
dence for the formation of alloy AuPd NPs and monometallic Pd par-
ticles was found in the AuPd/OMC catalyst, probably indicating that
the Pd NPs are small in size and highly dispersed.
The average particle sizes of Au, and Pd NPs on supported AuPd
catalysts, as calculated by XRD or TEM, are listed in Table 1. It can
be seen that AuPd-OMC-1 and AuPd-OMC-2 samples have smaller
Pd NPs than the Pd-OMC sample. It has been reported that the two
metal components will help in decreasing the particle size of the
active components [8]. The AuPd-OMC-1 and AuPd-OMC-2 samples
have a similar average particle size as those of Au and Pd NPs. The
increase in the metal loading did not induce the growth and aggrega-
tion of the metal NPs. In addition, the AuPd/OMC sample showed the
largest Au NPs due to the poor interaction between the gold precur-
sors and the OMC surface as per an earlier report [10].
to the smaller particle size of this sample. Thus it can be concluded
that the gold content in AuPd-OMC-1 and AuPd-OMC-2 catalysts did
not alloy with the Pd to form the bimetallic metal NPs, instead it
formed mono-metallic Au NPs in the mesoporous carbon channels.
These Au NPs are smaller in size and homogeneously dispersed, so
that no peaks indexed as Au were found in the wide angle XRD pat-
tern of AuPd-OMC-1 and AuPd-OMC-2 catalysts. As calculated from
the TEM images, the average particle size of the Au NPs is 2.5 nm.
It is commonly accepted that Au NPs grow more easily than Pd
NPs to become large size NPs during a thermal treatment [16,10]. In
general, it is difficult to obtain highly dispersed Au NPs due to their
low melting point. The functionalization of amino groups may be
the main reason for the unusual dispersion of Au and Pd NPs in this
study. As discussed in the previous work [14], the amino groups im-
part high thermal stability to the Au NPs during the thermal treat-
ment. Unfortunately, as shown in the TEM images, the amino
groups could not protect the Pd NPs from sintering. However, the
intrinsic mechanism underlying these results still requires further
study.
Clearly, in both of AuPd-OMC-1 and AuPd-OMC-2 samples, sepa-
rate Au and Pd NPs were found to be dispersed on the OMC support.
The Au NPs were attached to the pore walls of the OMC support,
while the Pd NPs stayed on the outer surface of the OMC support.
The synthetic route for the hydrogenation of CMA is depicted in
Fig. 5. It can result in various product distributions, depending on
the active site of the metal, solvent, and promoters [17,18]. The turn
over frequency (TOF) value, a measure of the average catalytic activ-
ity, was determined from the total number of Pd and Au atoms
involved in the reactions for all the AuPd bimetallic catalysts [12].
In order to develop the applications for the AuPd-OMC catalysts,
hydrogenation of CMA was carried out under atmospheric pressure
at 40 °C. Reaction without the use of H2 gas was also carried out, as
a control reaction, to ensure that the active hydrogen for CMA hydro-
genation originated from the used H2 gas and not from the
isopropanol. As shown in Table 1, the monometallic Au-OMC catalyst
is inactive under the given reaction conditions. The monometallic
Pd-OMC catalyst only gave a conversion of 4.5%; and a low selectivity
of 60.2%. Although the AuPd/OMC catalyst shows a conversion similar
to the Pd-OMC, a better selectivity could be obtained. Both the AuPd
bimetallic catalysts show excellent activity compared to the mono-
metallic Pd catalyst and AuPd/OMC catalysts. The AuPd-OMC-1 cata-
lyst gives the highest TOF of 120 h−1 which is around 3 times
higher than the monometallic Pd-OMC catalyst. Moreover, the selec-
tivity of HCAL was maintained at about 90% until the end of the reac-
tion. Due to the higher metal loading, approximately 100% conversion
was achieved in 2 h when AuPd-OMC-2 was used as the catalyst.
The typical TEM images of the Au-OMC, Pd-OMC, AuPd-OMC-1
and AuPd-OMC-2 are shown in Fig. 4. All the catalysts have regular
mesochannels and highly dispersed metal NPs, which is consistent
with the results from XRD analyses. Also, some larger NPs can be
seen in the Pd-OMC, AuPd-OMC-1, and AuPd-OMC-2 images, imply-
ing the sintering or aggregation of the metal components in this cat-
alyst. The Au-OMC sample has a narrow particle size distribution of
2.3
of 15
0.5 nm, while the Pd-OMC has a wide particle size distribution
5 nm. In the AuPd-OMC catalysts, the metal NPs have
2–20 nm particle size. According to the EDX results, the larger metal
NPs in AuPd-OMC-1 and AuPd-OMC-2 mainly consists of the Pd spe-
cies. The aggregation of Pd catalysts has been recognized and report-
ed previously [15]. Considering that the average pore diameters of the
carbon channels are below 6 nm, the large Pd crystallites cannot
enter into the pores, thus being forced to locate on the external sur-
face of the OMCs. A similar EDX-TEM analysis was also made for the
NPs below 5 nm in AuPd-OMC-1 and AuPd-OMC-2 catalysts. The
lattice spacing is 0.235 nm that is consistent with the (111) plane of
metallic gold (0.235 nm). Unfortunately, the EDX analysis failed due
Table 1
a
Catalytic activity of AuPd-OMC catalysts for the hydrogenation of CMA under atmospheric pressure
.
Catalysts
Metal loading
(wt.%)
d
TOF
Conversion
(%)
Selectivity
(%)
f
(nm)
(h−1
)
Au
Pd
Au
Pd
HCAL
HCOL
COL
Pd-OMC
0
0.25
0.2
0.2
–
8.2b
6.5b
45
120
119
105
98
4.5
19.9
39.8
96.7
94.5
4.0
60.2
96.2
94.2
88.0
46.9
95.5
–
24.2
2.1
3.1
9.4
45.2
4.2
–
15.6
1.7
2.7
2.6
7.9
3.3
–
AuPd-OMC-1
AuPd-OMC-1d
AuPd-OMC-1e
AuPd-OMC-2
AuPd/OMC
Au-OMC
2.5c
0.95
0.25
3.5
1.4
0.2
0
2.5c
7.0b
2.5c
–
10.0b
2.3b
12
0
0
a
Reaction conditions: 0.5 g CMA, 0.1 g catalyst, 20 mL isopropyl alcohol as solvent, 40 °C, atmospheric pressure, reaction for 2 h.
b
c
d
e
f
Calculated from XRD patters with Scherrer equation.
Calculated according to TEM images.
Reaction for 4 h.
Reaction for 11 h.
Determined from the total number of Pd and Au atoms on the catalysts.