The hydrogenation of o-chloronitrobenzene (o-CNB) was
employed as a probe reaction to test the activity and selectivity
of the Ag/SiO2 catalysts (Scheme 2) and the results are summarized
in Table 1. At the initial reaction pressure of 0.5 MPa, the
conversion of o-CNB was only 20% (entry 1). Increasing the
reaction pressure to 2.0 MPa, the o-CNB conversion increased to
100% dramatically (entry 2), at the same time, the product
selectivity to o-chloroaniline reached 100%, indicating that
Ag/SiO2 is a good catalyst for selective hydrogenation of o-CNB
to o-chloroaniline. Further increasing the reaction pressure to
4.0 MPa, no changes in conversion or selectivity were observed
(entry 3). When the reaction temperature was decreased from
140 uC to 100 uC, the conversion of o-CNB decreased to 35.9%
with the selectivity of the target product (o-chloroaniline) being
80.7%, meanwhile, nitrobenzene, a dechlorinated byproduct, was
formed (entry 4). It can be interpreted that the hydrogenation of
the nitro group and hydrogenolysis of the C–Cl bond are
competitive reactions at 100 uC, however, the hydrogenation of
the nitro group is more sensitive to temperature, thus it is mostly
accelerated at 140 uC and the selectivity to o-chloroaniline is
improved significantly. The study on the influence of reaction time
showed that the conversion could reach 100% in 1.5 h (entry 5).
The recycled Ag/SiO2 catalyst showed that the conversion of
o-CNB decreased to 56.3% (entry 6). Further experiments proved
that the organic fragments were not removed from the particle
interface in the reaction process, and a small amount of H2O had
no influence on the activity of the Ag/SiO2 catalyst, thus the
reasons for the lost activity of the recycled Ag/SiO2 catalyst must
be related to other factors. The first possible reason is the
aggregation of the silver particles: the TEM study (TEM image not
shown here) of a recovered catalyst shows that the silver particles
with a size of ca. 15.0–17.0 nm are bigger in contrast to a fresh
catalyst. In addition, the XRD study shows that in the case of the
recovered catalyst (graph b in Fig. 1B), the (111), (200), (220) and
(311) diffraction lines of metallic silver are present, confirming that
aggregation of silver particles took place during the reaction
process. It is known that the particle size of the catalysts has a
great influence on the catalyst activity;16 we believe that the silver
particle size plays a key role in the CNB hydrogenation catalyzed
by silver. Another possible reason is the leaching of silver particles
from the support. This assumption could be proved by the fact
that 7.7% conversion of o-CNB was obtained (entry 7) when the
reaction solution (catalyst was filtered out after one cycle) with an
additional equivalent of substrate underwent the same procedure
under the same conditions, indicating some leached sliver
nanoparticles were present in the reaction solution. It should also
be noted that the selectivity in this case dropped to 94.5%,
implying that the support made a contribution to the improvement
in the selectivity of the desired product. Pure CH2O–SiO2 was also
tested and no hydrogenation reactions were observed (entry 8),
which clearly proves that silver particles are the active component
in the o-CNB hydrogenation reaction. For comparison, an
Ag/SBA-15-imp catalyst with the size of the silver nanoparticles
similar to the Ag/SiO2 catalysts, prepared by the traditional
wetness impregnation method, was also tested under identical
hydrogenation conditions, showing only 40.3% of the o-CNB was
converted (entry 9).
In order to improve the stability of the silver nanoparticles on
the support, after some trial-and-error experiments, we developed
a simple method for the pretreatment of Ag/SiO2 to enhance the
interaction between the support and the silver nanoparticles. It has
been found that after treating the Ag/SiO2 at 300 uC in H2 for 2 h,
the leaching of silver particles can be avoided, though the
conversion of o-CNB suffered and dropped to 15.3% under the
same reaction conditions as used for fresh Ag/SiO2 (entry 10 in
Table 1). Furthermore, the XRD pattern of the heat-treated
catalyst showed that the silver particles were bigger in comparison
to the fresh catalyst, as can be seen in graph c of Fig. 1B, the
intensity of the characteristic diffraction lines (111), (200), (220)
and (311) of metallic silver becomes more significant. However,
when the treatment temperature was further decreased to 200 uC
(denoted as Ag/SiO2-200), the TEM and XRD studies showed
little change in the silver nanoparticles in comparison to the fresh
catalyst, but in this case the catalytic hydrogenation of o-CNB
over Ag/SiO2-200 catalyst showed that the high activity and
selectivity were maintained. More tests show that the recovered
catalyst (Ag/SiO2-200) could be used repeatedly for consecutive
reactions, and no significant loss in catalytic efficiency was found
(as shown in the ESI{), implying that the leaching of silver
particles was effectively avoided.
Scheme 2 Selective hydrogenation of o-chloronitrobenzene over Ag/
SiO2 catalyst to o-chloroaniline.
Table 1 Selective hydrogenation of o-chloronitrobenzene over fresh
Ag/SiO2 catalyst (dried only at 120 uC; reaction conditions: 0.1 g of the
catalyst, 0.5 g of the substrate and 25 mL of ethanol)
Entry t/h T/ uC P/MPa Conversion (%) Selectivity (%)
1
2
3
4
3
3
3
3
140
140
140
100
0.5
2.0
4.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
20
100
100
35.9
100
56.3
7.7
0
40.3
15.3
100
100
100
80.7a
5
1.5 140
100
100
6b
7c
8d
9e
10f
a
3
3
3
3
3
140
140
140
140
140
94.5a
Following the catalytic hydrogenation of o-CNB presented
above, the hydrogenations of other CNBs over the Ag/SiO2-200
catalyst were also conducted, of which the detailed results are
shown in the ESI.{ When nitrobenzene was used as a substrate, the
conversion was 93.9% and aniline was the only product. The
reason for the fact that the byproduct of hydrogenation of o-CNB
is nitrobenzene instead of aniline may be because the reduction
rate of nitrobenzene is slower than that of o-CNB. The Ag/SiO2-
200 catalyst also showed high activity and selectivity for
0
100
100
b
Nitrobenzene is a byproduct. Recovered catalyst after the 1st
c
reaction cycle (after entry 2). Catalyst was filtered off after the
reaction (entry 2) and the substrate was added for the second
reaction cycle. Reaction performed over CH2O–SiO2. Ag/SBA-15-
d
e
f
imp catalyst prepared by a wetness impregnation method. Ag/SiO2
catalyst treated at 300 uC in H2 for 2 h.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 5298–5300 | 5299
Chen, Yangying
