Y. Li et al. / Applied Catalysis A: General 522 (2016) 13–20
19
addition of Cl could surely increase the activity of Pd(N)-Re/SBA-
15. For mono-component Pd and Re catalysts, the influence of HCl
addition was also tested. The activity of mono-component Pd or
Re catalyst was very low, and as shown in Table S2, the addition
of HCl affected little on the activity of Pd(N)/SBA-15 or Re/SBA-15.
1,2-PD increased from 35.6% to 70.4%, which was quite similar with
the selectivity to 1,2-PD on Pd(C)/SBA-15 (72.2%).
(mol) of HCl increased from 0 to 5 times of that of Pd(OAc)2, the
conversion of glycerol over these catalysts increased from 18.0%
to 59.5%, while the selectivity to 1,2-PD became lower. To further
examine the influence of HCl addition on the distribution of prod-
ucts, reaction times over the catalysts were changed to maintain
conversions of glycerol in the range of 18–27% (Table S3). It can
be seen that using the catalysts prepared with the addition of HCl
could increase the activity of Pd-Re while the selectivity to 1,2-PD
was decreased. For Pd(N)-Re/SBA-15, the addition of HCl decreased
the selectivity to 1,2-PD from 55.7% to 51.3%, while the addition of
HCl decreased the selectivity to 1,2-PD from 67.4% to 50.5% over
Pd(A)-Re/SBA-15 (Table S3). In all these catalysts, the selectivities
to 1,3-PD were all maintained at 7–12%. In glycerol hydrogenoly-
sis, 1,2-PD could undergo further hydrogenolysis reaction to 1-PO.
The increase of the activity of Pd-Re catalyst with Cl addition in
hydrogenolysis of OH group might be favorable to convert 1,2-PD
into 1-PO. This might be the reason of lower selectivity to 1,2-PD
and higher selectivity to 1-PO.
Furthermore, H2-TPR was also used to characterize the reduc-
tion properties of Pd(A)-Re/SBA-15 prepared with different
amounts of HCl addition (Fig. 2(c)). The addition of HCl could also
change the reduction behaviors of Pd(A)-Re/SBA-15. When the ratio
of Pd:HCl was 5:1, the reduction peak at 130 ◦C in the TPR profile
of Pd(A)-Re/SBA-15 prepared with HCl addition was not as sharp
as that of without HCl addition, while the intensity of the reduc-
tion peak at 330 ◦C became stronger. As the addition amount (mol)
of HCl increased to 5 times of that of Pd(OAc)2, the TPR profile of
Pd(A)-Re/SBA-15 was very like that of Pd(C)-Re/SBA-15. This result
was also as similar as that of Pd(N)-Re/SBA-15 prepared with HCl
addition.
From Fig. 3(b), the crystal structures of Pd(A)-Re/SBA-15 pre-
pared with different amounts of HCl addition had no obviously
differences except for the sample of Pd:HCl = 1:5. For Pd(A)-Re/SBA-
15 prepared with Pd:HCl = 1:5, the Pd◦ diffraction peaks in the XRD
pattern were weaker than those of others. This result was similar
with that of Pd(N)-Re/SBA-15. HR-TEM was also used to character-
ize the morphologies of Pd(A)-Re/SBA-15 prepared with different
amounts of HCl addition, and the results are shown in Fig. 5 (The
performed in Fig. S4). Really, the addition of HCl during the prepara-
tion of the catalysts also increased the dispersion of Pd-Re particles
addition (Fig. 5(a)), most of the Pd-Re particles grew on the outside
of the support, and the average particle size was 17 nm, which was
much larger than that of Pd(C)-Re/SBA-15. When a little amount
of HCl was added into Pd(A)-Re/SBA-15 during the preparation of
the catalyst (Pd:HCl = 5:1, Fig. 5(b)), the average particle size of
Pd-Re was 10 nm and particles dispersed more homogenously. At
the same time, the Pd-Re particles grew along the SBA-15 chan-
nels and their shapes became rod like. Further increasing the HCl
When Pd:HCl was 1:5 (Fig. 5(d)), the average Pd-Re particle size of
Pd(A)-Re/SBA-15 was 4 nm. These results were also as similar as
that of Pd(N)-Re/SBA-15 prepared with HCl addition.
The actual amounts of Cl in the catalysts were characterized by
XRF, and the results are listed in Table 3. For Pd(N)-Re/SBA-15 pre-
pared with Pd:HCl = 1:5 addition, the actual amount of Cl on the
catalyst was 3.76% after the sample was evaporated, while only
1.51% of Cl remained on the catalyst after it was dried at 110 ◦C
overnight. Furthermore, when the sample was reduced at 250 ◦C,
the Cl amount on the catalyst further decreased to only 0.17%. For
Pd(A)-Re/SBA-15 prepared with Pd:HCl = 1:5 addition, it had simi-
lar results. That is, during the treatment steps of evaporation, drying
and reduction, part of the Cl component on the catalyst lost. Only
0.12% of Cl remained on the catalyst after the reduction at 250 ◦C.
The reduction behaviors of Pd(N)-Re/SBA-15 prepared with dif-
ferent amounts of HCl addition are shown in Fig. 2(b). It clearly
showed that when the amount of HCl was small (Pd:HCl = 5:1), the
intensity of reduction peak at 130 ◦C in TPR profile was stronger,
while the reduction peak at 200 ◦C shifted to 210 ◦C. As the addi-
tion amount of HCl increased, the intensity of reduction peak at
130 ◦C became stronger, but the reduction peak at 210 ◦C became
weaker. When the addition amount of HCl was 2 times of that of
Pd(NO3)2, the TPR profile of the Pd(N)-Re/SBA-15 was very like that
of Pd(C)-Re/SBA-15. When the amount (mol) of HCl was 5 times of
that of Pd(NO3)2, the main reduction peak in the TPR profile was at
130 ◦C, while the reduction peak at 210 ◦C was almost disappeared.
The differences of reduction behaviors of these Pd(N)-Re/SBA-15
samples might be due to the addition of Cl component, and this
would finally result in the different activities of Pd(N)-Re/SBA-15.
To further investigate the effect of HCl, mono-component Pd and Re
catalysts were also prepared with HCl addition, respectively. From
the TPR results of Pd(N)/SBA-15 with HCl addition, the reduction
peak at 130 ◦C was observed (Fig. S3(a)). On the other hand, for the
TPR profiles of Re/SBA-15 samples (Fig. S3(b)), the reduction tem-
perature of Re/SBA-15 with HCl addition decreased from 330 ◦C
These results showed that the Cl would also affect the reduction
behaviors of ReOx, and finally also influence the performance of
Pd-Re catalysts.
Fig. 3(a) shows the XRD patterns of Pd(N)-Re/SBA-15 prepared
with different amounts of HCl addition. When the samples were
reduced at 250 ◦C, all Pd(N)-Re/SBA-15 showed similar Pd◦ diffrac-
tion peaks in the XRD patterns. When the addition amount (mol) of
HCl was 5 times of that of Pd(NO3)2, the Pd◦ diffraction peaks in the
The morphologies of Pd(N)-Re/SBA-15 prepared with differ-
tion (Fig. 4(a)), Pd-Re particles agglomerated into very large ones,
while Pd and Re components were both existed in the particles
according to EDX analysis. After a certain amount of HCl was added
(Pd:HCl = 5:1, Fig. 4(b)), Pd-Re particles on the sample were still
agglomerated, but the dispersion of these particles were clearly bet-
of HCl increased to Pd:HCl = 1:1 (Fig. 4(c)), Pd-Re particles on the
sample were obviously dispersed better and grew along the SBA-15
channels, and particle sizes of Pd-Re were in the range of 4–5 nm.
Further increasing the amount (mol) of HCl to 5 times of that
of Pd(NO3)2 (Fig. 4(d)), Pd-Re particles dispersed more homoge-
nously, and the particle sizes of Pd-Re particles were also 4–5 nm.
From these results, it clearly indicated that the addition of HCl
would surely decrease the particle size and increase the dispersion
In order to further make sure the effect of Cl, Pd(A)-Re/SBA-15
catalysts were also prepared with HCl addition during an impreg-
nation process. The reaction results using these catalysts are listed
in Table 2 (No. 9–14). Similar as the case of Pd(N)-Re/SBA-15, for
Pd(A)-Re/SBA-15 without HCl addition, it had the lowest activity
(conversion of glycerol was 18.0%), while it had the highest selec-
tivity to 1,2-PD (67.4%). The addition of HCl into Pd(A)-Re/SBA-15
also increased the activity of the catalysts. As the addition amount