Significant effect of base on the improvement of selectivity in the hydrogenation of benzoic acid over NiZrB amorphous alloy supported on γ-Al2O3

Article abstract of DOI:10.1039/c5cy00115c

This study presents a facile way to improve the selectivity for cyclohexanecarboxylic acid by adding a base in the hydrogenation of benzoic acid over a non-noble metal NiZrB amorphous alloy supported on γ-Al₂O₃. It is found that alkali metal carbonates exhibit an excellent selectivity improvement from 50.3% to a range of 93.5-95.7%, with the conversion of benzoic acid higher than 92.3%. Even a very small amount of K₂CO₃ (1 mol% benzoic acid) was efficient for improving the selectivity for cyclohexanecarboxylic acid. In addition, a lower reaction temperature was beneficial to the improvement of selectivity. Based on the results of temperature programmed desorption of NH₃ and inductively coupled plasma analysis, the improvement of selectivity in the presence of a base was attributed to the neutralization of the acidic sites on the surface of the catalyst by the in situ generated potassium benzoate, inhibiting the hydrodeoxygenation of carbonyl and resulting in a high selectivity for cyclohexanecarboxylic acid.

Full text of DOI:10.1039/c5cy00115c

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Significant effect of base on the improvement of
selectivity in the hydrogenation of benzoic acid
over NiZrB amorphous alloy supported on
γ-Al2O3†
Cite this: DOI: 10.1039/c5cy00115c
a
a
a
a
a
Xin Wen,ab Yingying Cao, Xianliang Qiao, Libo Niu, Li Huo and Guoyi Bai
*
This study presents a facile way to improve the selectivity for cyclohexanecarboxylic acid by adding a base
in the hydrogenation of benzoic acid over a non-noble metal NiZrB amorphous alloy supported on
γ-Al2O3. It is found that alkali metal carbonates exhibit an excellent selectivity improvement from 50.3% to
a range of 93.5–95.7%, with the conversion of benzoic acid higher than 92.3%. Even a very small amount of
K2CO3 (1 mol% benzoic acid) was efficient for improving the selectivity for cyclohexanecarboxylic acid. In
addition, a lower reaction temperature was beneficial to the improvement of selectivity. Based on the
results of temperature programmed desorption of NH3 and inductively coupled plasma analysis, the
improvement of selectivity in the presence of a base was attributed to the neutralization of the acidic sites
on the surface of the catalyst by the in situ generated potassium benzoate, inhibiting the hydro-
deoxygenation of carbonyl and resulting in a high selectivity for cyclohexanecarboxylic acid.
Received 26th January 2015,
Accepted 10th April 2015
DOI: 10.1039/c5cy00115c
www.rsc.org/catalysis
controlled ruthenium-based catalysts.17 Another approach is
to select a suitable support to apply the interaction between
1. Introduction
Selective hydrogenation of a targeted functional group in a
molecule having several reducible functional groups is of
great utility in chemical processes.1–5 Particularly, the hydro-
genation of aromatic substrates aiming at manufacturing sat-
urated or partially saturated molecules has attracted much
interest in the chemical and pharmaceutical industry, due to
it being environmentally benign, having readily available suit-
able feedstocks and being the most straightforward
approach.5–10 In general, aromatic ring hydrogenation usually
requires harsh reaction conditions, increasing the potential
for the hydrogenation of other reducible moieties, which then
leads to low chemoselectivity.11,12 One approach to enhance
the selectivity is to modify the surface of catalysts.13 For
instance, the selective hydrogenation of aromatic and hetero-
aromatic compounds, such as quinoxalines,14 quinolines and
indoles,15,16 in order to obtain their relative saturated or par-
tially saturated molecules, can be achieved by using ligand-
the catalyst and the support, which improves the selectivity
for the hydrogenation of the aromatic ring. For example,
hydroxy aromatic derivatives, such as phenol, could be selec-
tively hydrogenated to the corresponding ketones by palla-
dium nanoparticles supported on hydrophilic carbon,18 meso-
porous graphitic carbon nitride19 or nitrogen-functionalized
ordered mesoporous carbon.20,21 It is worth noting that a sim-
ple way to enhance the selectivity for cyclohexanone in the
hydrogenation of phenol has been established by Han and
his coworkers.22 A Lewis acid (AlCl3, ZnCl2 or InCl3) was
added as a dual-functional additive to increase the selectivity
and activity of the Pd/C catalyst, because Lewis acid could not
only activate the phenyl ring but also suppress its further
hydrogenation.
Selective hydrogenation of benzoic acid (BA) to cyclo-
hexanecarboxylic acid (CCA), which is an important organic
intermediate for the synthesis of praziquantel and
ansatrienin, has been studied in previous reports.23–26 Most
of the catalysts used in this transformation were noble metal
catalysts.24–26 However, their high price has limited their fur-
ther industrial applications. Recently, we have prepared a
non-noble metal catalyst Ni–Zr–B–PEGIJ800), which showed
both good activity and selectivity for CCA in the selective
hydrogenation of BA using water as solvent.23 However, the
turnover number of Ni–Zr–B–PEGIJ800) was relatively low due
to the following two reasons: (i) the Ni active species leached
a Key Laboratory of Chemical Biology of Hebei Province, College of Chemistry and
Environmental Science, Hebei University, Baoding 071002, China.
E-mail: baiguoyi@hotmail.com; Fax: +86 312 5937102; Tel: +86 312 5079359
b Hebei Chemical and Pharmaceutical College, Shijiazhuang 050026, China.
E-mail: wenxin767@126.com; Fax: +86 311 85110024; Tel: +86 311 85110352
† Electronic supplementary information (ESI) available: FTIR analysis of
catalysts, time–conversion plot, data for recycle runs, characterization of the
used catalyst (TEM and XRD) and GC-MS analysis of products. See DOI: 10.1039/
c5cy00115c
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
into water during the long time stirring at high tempera-
tures;27 (ii) the formation of nickel hydroxide during the reac-
tion caused a rapid deactivation of the Ni-based catalyst.28,29
Whereas when we selected aprotic solvents in the selective
hydrogenation of BA to CCA, the selectivity for CCA was
markedly lowered due to the decrease in solvent polarity.23
In a continuing attempt to achieve a higher selectivity
using a non-noble metal catalyst in the selective hydrogena-
tion of BA to CCA, we herein demonstrate a facile and effec-
tive approach to improve the selectivity for CCA over a NiZrB
amorphous alloy supported on γ-Al2O3 ijNiZrB/Al2O3] just by
adding small amounts of bases in cyclohexane. To the best of
our knowledge, this might be the first report on selectivity
improvement by tuning the catalyst via the in situ generated
intermediate during the reaction. Therefore, a possible mech-
anism was proposed based on the necessary characterization.
Furthermore, the influence of parameters such as solvent,
reaction temperature, and ratio of base to BA was also
investigated.
2.3. Catalyst activity test
Selective hydrogenation of BA was conducted as follows: BA
(3.0 g), solvent (60 mL), catalyst (1.0 g), and base were mixed
in a stainless steel autoclave (100 mL) equipped with an elec-
tric heating system and a magnetically driven mechanical
stirrer. The reaction system was filled with H2 to 3.0 MPa sev-
eral times by evacuation to displace residual air, and then
pressured with H2 to 5.0 MPa and heated to the requisite
temperature. Upon reaching the required temperature, the
reaction was performed by stirring the reaction mixture vigor-
ously. When the reaction was complete, the reaction mixture
was allowed to cool to room temperature, and separated by
filtration. The solution was analyzed by gas chromatography
(GC) using a 30 m SE-54 capillary column. The product was
identified using gas chromatography-mass spectroscopy (GC-
MS) on a Thermo Finnigan Polaris-Q spectrometer. The used
catalyst was denoted as “NiZrB/Al2O3–u–base.” When a base
was not added, the used catalyst was denoted as
“NiZrB/Al2O3–u.”
2. Experimental
2.1. Catalyst preparation
2.4. Comparative experiments
Comparative experiments were carried out using the same
method as that of BA hydrogenation but without adding any
catalyst, instead excess additive was used – half of the
amount of BA. For a typical run: BA (3.0 g), cyclohexane (60
mL), and K2CO3 or K2SO4 (50 mol% BA) were allowed to react
at 423 K under 5 MPa hydrogen pressure for 4 h, and then
treated by the same process as described above. The experi-
ments were denoted as CE–K2CO3 and CE–K2SO4,
respectively.
Unless otherwise noted, all chemicals were of analytical
reagent grade and used without further purification.
NiZrB/Al2O3 was prepared by an impregnation–reduction
method. The support γ-Al2O3 (2.7 g) was first heated at 773 K
for 4 h and then impregnated with a mixed solution of 1.234
g of NiCl2·6H2O, 0.055 g of ZrIJNO3)4·5H2O (molar ratio Ni : Zr
= 40 : 1), 0.5 g of poly(ethylene glycol) (average molecular
weight: 800 g mol−1, PEG800) and 2 mL of water under ultra-
sound for 20 min. After the resulting paste was dried at 393
K for 2 h, the light yellow precursor was ground into powder
and then placed in an ice bath and reduced by adding 20 mL
of 1.0 M aqueous KBH4 containing 0.25 M NaOH dropwise
under vigorous stirring. When bubbles ceased to appear, the
precipitate was filtered, washed with water several times until
the washings were pH 7, and then further washed with abso-
lute ethanol three times to replace residual water. Further-
more, an unsupported NiZrB amorphous alloy modified with
PEG(800) was prepared by a chemical reduction method as
described previously.23
3. Results and discussion
3.1. Catalyst characterization
The physical properties of NiZrB/Al2O3, NiZrB and γ-Al2O3 are
summarized in Table 1. From the ICP analysis, it was found
that the contents of Ni and Zr in both NiZrB/Al2O3 and NiZrB
were about 40 : 1, which was almost identical to the feed
ratio. The Ni loading of NiZrB/Al2O3 was only 11.0 wt%,
much lower than that of NiZrB. Thus, it is reasonable that
NiZrB/Al2O3 has a lower H2-chemisorption value than NiZrB.
In fact, it has a higher H2-chemisorption value for each Ni
atom due to the dispersive effect of γ-Al2O3. The BET surface
area, pore volume and pore size of NiZrB/Al2O3 markedly
decreased compared with those of the pure γ-Al2O3 support,
indicating that some NiZrB amorphous alloy particles occu-
pied the surface and the pores of the γ-Al2O3 support due to
its porous nature.30
XRD analysis of the prepared NiZrB/Al2O3 catalyst showed
three characteristic peaks, which were all attributed to
γ-Al2O3 (Fig. 1); in contrast, no peak related to the amorphous
alloy was detected in this catalyst, probably because the dif-
fraction peak of NiZrB in NiZrB/Al2O3 was obscured by the
peaks of γ-Al2O3 at about 45 °C.30–33
2.2. Catalyst characterization
The Brunauer–Emmett–Teller (BET) surface area was esti-
mated using a Micromeritics Tristar II 3020 surface area and
pore
analyzer.
H2-chemisorption
and
temperature
programmed desorption of NH3 (NH3-TPD) were performed
using a TP-5000 instrument supplied by Xianquan Ltd. X-ray
diffraction (XRD) patterns were recorded on a Bruker D8-
ADVANCE X-ray diffractometer. Transmission electron
microscopy (TEM) was performed using a FEI Tecnai G2 F20
microscope. Fourier transform infrared spectroscopy (FTIR)
was carried out using a Bruker Vertex 70 spectrometer. Induc-
tively coupled plasma analysis (ICP) was performed using a
Varian Vista-MPX spectrometer.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
Table 1 Structural properties of the amorphous alloy catalysts and γ-Al2O3
Compositiona
(atomic ratio)
Ni loading
(wt%)
H2-chemisorption
Surface area
Pore volume
Pore size
(nm)
Catalyst
(cm3 g−1
)
(m2 g−1
)
(cm3 g−1
)
NiZrB/Al2O3
NiZrBb
γ-Al2O3
Ni1.00Zr0.026B0.30
Ni1.00Zr0.025B0.40
—
11.0
90.2
0
0.15
0.45
0
154
50
202
0.25
0.14
0.42
6.5
10.9
8.3
a
Based on ICP results. b According to the results of Ni–Zr–B–PEGIJ800) in ref. 23.
The TEM image of NiZrB/Al2O3 is shown in Fig. 2a. It was
obvious that the NiZrB amorphous alloy was well dispersed
on the γ-Al2O3 support with the particle size in the range of
7.0–21.0 nm (Fig. 2b). The mean particle size of the NiZrB
amorphous alloy was 11.4 nm, which was much smaller than
that of the unsupported NiZrB amorphous alloy (20.8 nm), as
reported in our previous work.23 On the other hand, no evi-
dence for the existence of PEG(800) on the surface of
NiZrB/Al2O3 can be found from the TEM results. Thus, we
suggest that most of the PEG(800) has been removed during
the washing process of preparation, as supported by the FTIR
spectra (Fig. S1, ESI†) and in agreement with our previous
work.34 Its main role was to prevent the agglomeration of the
amorphous alloy particles in the reduction process, which
resulted in more active Ni species.
complex, protecting the carboxyl group from hydrogenation
and then improving the selectivity for CCA. Unfortunately, a
terribly low conversion (1.1%) was obtained when AlCl3 was
added in cyclohexane, although the selectivity for CCA can
reach 99.9% (entry 3). Subsequently, some bases and salts
were also tested as additives to improve the selectivity in
cyclohexane. Interestingly, bases showed a remarkable effect
on the selectivity for CCA. In the case of alkali metal carbon-
ates (entries 4–7), the selectivities for CCA were markedly
improved from 50.3% to a range of 93.5–95.7%. In the case
of other bases (entries 8–14), the selectivities for CCA were
also over 80% except for triethylamine, which was only
56.7%, just a little higher than that obtained when no addi-
tive was used. In contrast, in the case of a neutral salt, such
as K2SO4, the selectivity for CCA was only 47.7% (entry 15),
much lower than those of alkali metal carbonates. Thus, we
can conclude that bases, especially for alkali metal carbon-
ates, favored the selectivity for CCA in cyclohexane over
NiZrB/Al2O3.
3.2. Effect of bases
Table 2 shows the results of the selective hydrogenation of
BA to CCA over NiZrB/Al2O3 under different conditions. As
can be seen, the conversion of BA was 99.9%, but the selectiv-
ity for CCA was only 50.3% in cyclohexane without any addi-
tive (entry 1). Although the selectivity for CCA was increased
to 82.5% in water, the conversion was only 29.6% (entry 2),
which might be due to the low Ni content in the catalyst.
According to the observation of Han and his coworkers, the
coordination of the Lewis basic CO group and Lewis acid
inhibited further the hydrogenation of cyclohexanone in the
hydrogenation of phenol.22 Thus, for our reaction system, the
carboxyl group of BA and the Lewis acid might form a
To further study the effect of base on the selectivity for
CCA in the hydrogenation of BA, K2CO3 was chosen in
Fig. 1 XRD patterns of amorphous alloy catalysts (a) NiZrB and (b)
NiZrB/Al2O3.
Fig. 2 TEM image (a) and size distribution (b) of NiZrB/Al2O3.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 2 Effect of additives in the selective hydrogenation of BA over
Table 3 Effect of reaction temperature
NiZrB/Al2O3
Temperature
(K)
Time
(h)
Conversion
(%)
Selectivity
(%)
Yield
(%)
Entry
Conversion
of BA (%)
Selectivity
for CCA (%)
Yield of
CCA (%)
Entry
Additives
1
2
3
4
423
398
373
373
4
4
4
16
99.9
82.8
30.5
82.7
95.3
99.5
100
95.2
82.4
30.5
82.5
1
—
—
AlCl3
Li2CO3
Na2CO3
K2CO3
Cs2CO3
KOAc
99.9
29.6
1.1
50.3
82.5
99.9
93.5
93.8
95.3
95.7
93.9
89.3
93.7
80.3
90.3
86.5
56.7
47.7
50.2
24.4
1.1
2a
3
99.8
4
5
6
7
8
9
10
11
12
13
14
15
99.9
99.9
99.9
92.3
96.7
99.4
99.9
99.9
99.9
99.9
99.9
99.9
93.4
93.7
95.2
88.3
90.8
88.8
93.6
80.2
90.2
86.4
56.6
47.7
Reaction conditions: BA (3.0 g), cyclohexane (60 mL), NiZrB/Al2O3
(1.0 g, 11 wt% Ni), K2CO3 (5 mol% BA), initial PIJH2) = 5 MPa, and
reaction time 4 h.
NaHCO3
C6H5COOK
NaOH
KOH
These results indicated that a lower temperature would bene-
fit the orientation of the aromatic ring facing towards the cat-
alyst or decrease the reaction rate towards the carboxyl hydro-
deoxygenation product, thus inhibiting the generation of
toluene. Finally, 423 K was chosen as the optimal reaction
temperature in view of the transformation efficiency.
Fig. 4 shows the effect of the amount of K2CO3 in the
hydrogenation of BA. It was found that when the amount of
K2CO3 decreased from 15 mol% to 1 mol%, the conversion of
BA remained over 98.0% and the selectivity for CCA remained
over 95.0%, higher than that observed in our previous work
both in water (93.6%) and cyclohexane (42.7%) over the
NiZrB catalyst.23 With a further decrease in the amount of
K2CO3 to 0.1 mol%, the selectivity for CCA decreased to
75.8%, which was still higher than that observed when no
additive was added (50.3%). Thus, we can conclude that even
a small amount of K2CO3 (1 mol%) is efficient for improving
the selectivity for CCA in the hydrogenation of BA in
cyclohexane.
Furthermore, the turnover number (molar ratio of trans-
formed BA to Ni atom) of NiZrB/Al2O3 was 13.1, much higher
than that of the NiZrB catalyst (3.0).23 We ascribed this to the
more well-dispersed Ni active sites in NiZrB/Al2O3, as demon-
strated by the TEM and H2-chemisorption results. Therefore,
the observation of this phase of study led to a conclusion that
the addition of base and the γ-Al2O3 support exhibited a
robust effect on improving the selectivity and activity of
NiZrB/Al2O3 in the hydrogenation of BA, respectively. It must
PAASb
NIJC2H5)3
K2SO4
Reaction conditions: BA (3.0 g), cyclohexane (60 mL), NiZrB/Al2O3
(1.0 g, 11.0 wt% Ni), additive (5 mol% BA), 423 K, initial PIJH2) = 5
a
b
MPa, and reaction time 4 h. Water as the solvent. Polyacrylic
acid sodium.
consideration of its catalytic efficiency and cost. The effects
of solvent, reaction temperature and amount of K2CO3 were
then investigated. The effect of solvent on the selectivity for
CCA was first studied and the results are shown in Fig. 3.
With the addition of K2CO3, no matter what kind of solvent
was used, the selectivities for CCA were higher than 95%.
Thus, the catalytic system involving both base and
NiZrB/Al2O3 displays a good solvent profile for selectivity. On
the other hand, the best conversion of BA was obtained in
cyclohexane, probably due to its aprotic and non-polar
nature. Thus, cyclohexane was the solvent of choice for the
hydrogenation of BA.
The effect of reaction temperature on the selectivity for
CCA in the presence of K2CO3 is shown in Table 3. The selec-
tivity for CCA was nearly 100% with the reaction temperature
decreasing from 423 K to 373 K; at the same time, the conver-
sion of BA decreased to 30.5% (entries 1–3), and it can reach
82.7% if the reaction time is prolonged to 16 h (entry 4).
Fig. 3 Effect of different solvents: (A) cyclohexane, (B) tert-butanol,
(C) tetrahydrofuran, (D) ethyl acetate, and (E) dioxane. Reaction condi-
tions: BA (3.0 g), solvent (60 mL), NiZrB/Al2O3 (1.0 g, 11 wt% Ni), K2CO3
(5 mol% BA), 423 K, initial PIJH2) = 5 MPa, and reaction time 4 h.
Fig. 4 Effect of the amount of K2CO3. Reaction conditions: BA (3.0 g),
cyclohexane (60 mL), NiZrB/Al2O3 (1.0 g, 11 wt% Ni), base IJK2CO3), 423
K, initial PIJH2) = 5 MPa, and reaction time 4 h.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
be noted that the activity of the catalyst gradually decreased
during the recycle in cyclohexane (Fig. S2 and S3, ESI†),
although it was still much better than that in water. We can
deduce that the decrease of catalyst activity in cyclohexane
should be due to the transformation of the active nickel spe-
cies from the unstable amorphous alloy to the relatively sta-
ble crystalline, as proven by the TEM (Fig. S4, ESI†) and XRD
(Fig. S5, ESI†) results. In contrast, Ni leaching,27 the forma-
tion of nickel hydroxide28,29 and hydrate boehmite35,36 are
believed to be the main reasons for the deactivation of the
nickel-based catalysts in water. On the other hand, it is nec-
essary to clarify the role of K2CO3 in the improvement of
selectivity of BA hydrogenation.
Fig. 6 FTIR spectra of (a) the precipitate of CE–K2CO3, (b) PB, (c) BA,
and (d) precipitate of CE–K2SO4.
3.3. Mechanism insight into the selectivity improvement
by bases
curve b),37 together with the disappearance of the characteris-
tic peaks of BA (Fig. 6, curve c). Thus, PB can be in situ gener-
ated from BA and K2CO3 under the reaction conditions. In
contrast, in a comparative experiment using excess K2SO4
instead of K2CO3 IJCE–K2SO4), no PB can be identified from
the FTIR spectrum of its precipitate (Fig. 6, curve d). Curve d
was believed to be a combination of the characteristic peaks
of BA and K2SO4 (typical at 1118 cm−1 and 613 cm−1), proving
that PB could not be generated from BA and K2SO4.
On the other hand, the FTIR spectra of the products pre-
pared in the absence (Fig. 7, curve a) and presence (Fig. 7,
curve b) of K2CO3 are similar, suggesting that the addition of
K2CO3 did not affect the purity of the product. Further ICP
analysis showed that the content of potassium in the
obtained product was only 0.025 mol% of CCA, which can be
ignored and in agreement with the FTIR results (Fig. 7). Com-
bined with the above results, we can conclude that PB was in
situ generated from BA and K2CO3 under the reaction condi-
tions, and then adsorbed on the catalyst surface. Then, it will
probably tune the acid–base properties of the catalyst and
improve the selectivity for CCA.
FTIR spectroscopy measurements were first carried out to
investigate the mechanism of selectivity improvement. As can
be seen, the FTIR spectra of NiZrB/Al2O3, NiZrB/Al2O3–u, and
NiZrB/Al2O3–u–K2CO3 were markedly different (Fig. 5). For
the fresh NiZrB/Al2O3, the peaks at about 3400 cm−1 and
1641 cm−1 were assigned to the stretching and deformation
vibration absorption of water, and the peaks at about 1397
cm−1 and 816–568 cm−1 were assigned to the absorption of
Al–O. For the NiZrB/Al2O3–u catalyst, which underwent hydro-
genation in the absence of K2CO3, some new peaks at 2931
cm−1, 1582 cm−1, 1446 cm−1 and 980 cm−1 appeared, which
can be assigned to the adsorbed CCA on the used catalyst.
Interestingly, apart from these new peaks, other new peaks at
3075 cm−1, 1610 cm−1, 1390 cm−1 and 709 cm−1 appeared in
the FTIR spectrum of NiZrB/Al2O3–u–K2CO3, which was used
in the presence of K2CO3. These new peaks were similar to
the characteristic peaks of potassium benzoate (PB).37 So,
some PB might be in situ generated under the reaction
conditions.
To confirm the formation of PB, a comparative experiment
was then carried out using excess K2CO3 under the same
reaction conditions in the absence of the catalyst IJCE–
K2CO3). The FTIR spectrum of the precipitate of CE–K2CO3
(Fig. 6, curve a) is consistent with that of pure PB (Fig. 6,
Fig. 8 shows the NH3-TPD profiles of NiZrB/Al2O3–u and
NiZrB/Al2O3–u–K2CO3. For NiZrB/Al2O3–u (Fig. 8, curve a),
there were two peaks at about 510 K and 850 K, whereas the
Fig. 7 FTIR spectra of the products obtained from the hydrogenation
of BA in the absence (a) and in the presence of K2CO3 (b) in
cyclohexane.
Fig. 5 FTIR spectra of (a) NiZrB/Al2O3, (b) NiZrB/Al2O3–u, and (c)
NiZrB/Al2O3–u–K2CO3.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
was added during the reaction, PB will be in situ generated
from BA and K2CO3, and loaded on the surface of γ-Al2O3. It
can neutralize some acidic sites on the catalyst surface,
inhibiting the hydrodeoxygenation of carbonyl38–40 and
resulting in a high selectivity for CCA (Fig. 9). It can be fur-
ther proved by the poor selectivity improvement performace
of K2SO4 (Table 2, entry 15), which cannot generate PB with
BA, as demonstrated by the previous CE–K2SO4 comparative
experiment. Thus, the formation of PB and its neutralization
of the acidic sites on the catalyst surface should play an
important role in the CCA selectivity improvement in the
hydrogenation of BA.
Fig. 8 NH3-TPD profiles of amorphous alloy catalysts (a) NiZrB/Al2O3–u
and (b) NiZrB/Al2O3–u–K2CO3.
4. Conclusions
In summary, we have first shown the appreciable positive
effect of bases on the improvement of selectivity for CCA in
the selective hydrogenation of BA over the NiZrB/Al2O3 cata-
lyst. The presence of even a very small amount of K2CO3 in
the hydrogenation of BA resulted in a substantial increase in
the selectivity for CCA. PB was proven to be in situ generated
from BA and K2CO3 under the reaction conditions, and then
adsorbed on the catalyst surface. Thus, it can neutralize the
acidic sites on the catalyst surface, inhibiting the hydro-
deoxygenation of carbonyl and resulting in a high selectivity
for CCA. The selective hydrogenation of BA exhibited a better
solvent profile by base addition over the NiZrB/Al2O3 catalyst.
These observations will also be helpful for the improvement
of selectivity in other reactions.
peak at 510 K disappeared and the peak at 850 K moved to a
lower temperature (770 K) when K2CO3 was used during the
hydrogenation process (Fig. 8, curve b), so that NiZrB/Al2O3–
u–K2CO3 has nearly no weakly acidic sites and has a lower
acidity of strong acidic sites compared with NiZrB/Al2O3–u.
In combination with the FTIR results, we ascribed this to the
neutralization of the acidic sites by the in situ generated PB,
which then accounted for the high selectivity for CCA in
cyclohexane. Although there have been some reports on the
selectivity improvement by the modulation of the acid–base
properties of the catalysts during their preparation,38–40 it
might be the first observation that selectivity improvement
can be achieved by tuning the catalyst via the in situ gener-
ated intermediate during the reaction.
It is really interesting that a high selectivity for CCA was
obtained in cyclohexane, a non-polar solvent, by the addition
of base in this reaction system, in comparison with our
prvious study.23 Indeed, the orientation of the BA molecule is
irregular on the surface of NiZrB/Al2O3 in cyclohexane,23,24 as
shown in Fig. 9. In the case when no base is added, both the
benzene ring and carbonyl of BA will be adsorbed on the sur-
face of the catalyst. Thus, BA can be hydrodeoxygenated to
form toluene, as proven by GC-MS (Fig. S6, ESI†), in the pres-
ence of twin oxygen vacancies,41,42 resulting in a low selectiv-
ity for CCA (Fig. 9). On the contrary, if the base like K2CO3
Acknowledgements
Financial support from the National Natural Science Founda-
tion of China (21376060), the Natural Science Foundation of
Hebei Province (B2014201024), and the Youth Foundation of
Hebei Educational Committee (QN2014069) is gratefully
acknowledged.
Notes and references
1 R. V. Jagadeesh, A. E. Surkus, H. Junge, M. Pohl, J. Radnik, J.
Rabeah, H. Huan, V. Schünemann and M. Beller, Science,
2013, 342, 1073–1076.
2 B. Peng, X. Yuan, C. Zhao and J. A. Lercher, J. Am. Chem.
Soc., 2012, 134, 9400–9405.
3 X. Mou, B. Zhang, Y. Li, L. Yao, X. Wei, D. S. Su and W.
Shen, Angew. Chem., Int. Ed., 2012, 51, 2989–2993.
4 P. G. W. A. Kompio, A. Brückner, F. Hipler, G. Auer, E.
Löffler and W. Grünert, J. Catal., 2012, 286, 237–247.
5 M. Tamura, K. Tokonami, Y. Nakagawa and K. Tomishige,
Chem. Commun., 2013, 49, 7034–7036.
6 T. Yu, J. Wang, X. Li, X. Cao and H. Gu, ChemCatChem,
2013, 5, 2852–2855.
7 A. M. Rasero-Almansa, A. Corma, M. Iglesias and F. Sánchez,
Green Chem., 2014, 16, 3522–3527.
8 H. J. Chen, H. W. Liu, W. Liao, H. B. Pan, C. M. Wai, K. H.
Chiu and J. F. Jen, Appl. Catal., B, 2012, 111, 402–408.
Fig. 9 Proposed mechanism for the hydrogenation of BA with or
without K2CO3.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
9 V. Mévellec, A. Roucoux and E. Ramirez, Adv. Synth. Catal.,
2004, 346, 72–76.
26 X. Xu, M. Tang, M. Li, H. Li and Y. Wang, ACS Catal.,
2014, 4, 3132–3135.
10 Y. Gong, P. Zhang, X. Xu, Y. Li, H. Li and Y. Wang, J. Catal.,
2013, 297, 272–280.
27 G. Bai, H. Dong, Z. Zhao, H. Chu, X. Wen, C. Liu and F. Li,
RSC Adv., 2014, 4, 19800–19805.
11 Y. G. Zhou, Acc. Chem. Res., 2007, 40, 1357–1366.
12 J. A. Widegren and R. G. Finke, J. Mol. Catal. A: Chem.,
2003, 191, 187–207.
13 H. U. Blaser, C. Malan, B. Pugin, F. Spindler, H. Steiner and
M. Studer, Adv. Synth. Catal., 2003, 345, 103–151.
14 Q. A. Chen, D. S. Wang, Y. G. Zhou, Y. Duan, H. J. Fan, Y.
Yang and Z. Zhang, J. Am. Chem. Soc., 2011, 133,
6126–6129.
28 W. Lin, H. Cheng, J. Ming, Y. Yu and F. Zhao, J. Catal.,
2012, 291, 149–154.
29 H. Li, Y. Zhao, C. Gao, Y. Wang, Z. Sun and X. Liang, Chem.
Eng. J., 2012, 181, 501–507.
30 X. Wen, Z. Zhao, Y. Cao, H. Dong, C. Liu, H. Chu and G. Bai,
Res. Chem. Intermed., 2014, DOI: 10.1007/s11164-014-1744-0.
31 H. Li, D. Zhang, G. Li, Y. Xu, Y. Lu and H. Li, Chem.
Commun., 2010, 46, 791–793.
15 F. R. Gou, W. Li, X. Zhang and Y. M. Liang, Adv. Synth.
Catal., 2010, 352, 2441–2444.
32 D. G. Tong, W. Chu, P. Wu and L. Zhang, RSC Adv., 2012, 2,
2369–2376.
16 D. S. Wang, Q. A. Chen, W. Li, C. B. Yu, Y. G. Zhou and X.
Zhang, J. Am. Chem. Soc., 2010, 132, 8909–8911.
17 S. Urban, N. Ortega and F. Glorius, Angew. Chem., Int. Ed.,
2011, 50, 3803–3806.
33 Y. Pei, G. Zhou, N. Luan, B. Zong, M. Qiao and F. Tao,
Chem. Soc. Rev., 2012, 41, 8140–8162.
34 G. Bai, Z. Zhao, L. Niu, H. Dong, M. Qiu, F. Li, Q. Chen and
G. Chen, Catal. Commun., 2012, 23, 34–38.
18 P. Makowski, R. D. Cakan, M. Antonietti, F. Goettmann and
M. M. Titirici, Chem. Commun., 2008, 999–1001.
19 Y. Wang, J. Yao, H. Li, D. Su and M. Antonietti, J. Am. Chem.
Soc., 2011, 133, 2362–2365.
35 R. M. Ravenelle, J. R. Copeland, W. G. Kim, J. C. Crittenden
and C. Sievers, ACS Catal., 2011, 1, 552–561.
36 G. Lefevre, M. Duc, P. Lepeut, R. Caplain and M. Fédoroff,
Langmuir, 2002, 18, 7530–7537.
20 Z. Li, J. Liu, C. Xia and F. Li, ACS Catal., 2013, 3, 2440–2448.
21 J. Zhong, J. Chen and L. Chen, Catal. Sci. Technol., 2014, 4,
3555–3569.
37 M. Gasgnier, J. Mater. Sci. Lett., 2001, 20, 1259–1262.
38 T. Yokoyama and N. Yamagata, Appl. Catal., A, 2001, 221,
227–239.
22 H. Liu, T. Jiang, B. Han, S. Liang and Y. Zhou, Science,
2009, 326, 1250–1252.
39 A. Saadi, Z. Rassoul and M. M. Bettahar, J. Mol. Catal. A:
Chem., 2000, 164, 205–216.
23 G. Bai, X. Wen, Z. Zhao, F. Li, H. Dong and M. Qiu, Ind. Eng.
Chem. Res., 2013, 52, 2266–2272.
40 D. Procházková, P. Zámostný, M. Bejblová, L. Červený and J.
Čejka, Appl. Catal., A, 2000, 332, 56–64.
24 J. A. Anderson, A. Athawale, F. E. Imrie, F. M. McKenna, A.
McCue, D. Molyneux, K. Power, M. Shand and R. P. K. Wells,
J. Catal., 2010, 270, 9–15.
41 N. Perret, X. Wang, J. J. Delgado, G. Blanco, X. Chen, C. M.
Olmos, S. Bernal and M. A. Keane, J. Catal., 2014, 317, 114–125.
42 F. Dury, V. Misplon and E. M. Gaigneaux, Catal. Today,
2004, 91, 111–116.
25 H. J. Wang and F. Y. Zhao, Int. J. Mol. Sci., 2007, 8, 628–634.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.