G. Liang et al. / Journal of Catalysis 309 (2014) 468–476
469
production by using H4SiW12O40 and Ru/C, and 63% yield of isosor-
bide could be obtained from the fibrillar cellulose within 1 h [16].
All the above results are really promising; nevertheless, the
selective hydrolytic hydrogenation of cellulose into polyols is still
a big challenge. One significant disadvantage is the use of large
amount of noble metal catalyst, making those approaches too
expensive for the large-scale application in industry. Recently,
attention is drawn to the non-noble metal catalysts, including Ni,
Cu, and tungsten carbide catalysts because of the fast turnover
rates, availability, and low cost. Zhang et al. studied the catalytic
performance of Ni-promoted tungsten catalysts on cellulose
hydrogenation; notably, ethylene glycol (EG) was obtained with
61% yield at 245 °C [17]. In their following research, tungsten car-
bide on mesoporous carbon (MC) and Ni-W/SBA-15 were demon-
strated to be also effective for catalyzing cellulose and produced
EG with the highest yield of 72.9% [18,19]. Conventional Ni-based
catalysts have always been used as effective catalysts for aqueous-
phase hydrogenolysis of sorbitol to produce glycerol, ethylene gly-
col, and propanediol. Unfortunately, supported Ni catalysts exhib-
ited inferior performance in the transformation of cellulose into
sorbitol. The attempts to increase activity and selectivity of Ni-
based catalysts for the cellulose hydrogenation still encounter
great difficulties. Therefore, the modified Ni-based catalysts are de-
signed to promote the yield of sorbitol by suppressing the subse-
quent hydrogenolysis reaction. Sels et al. obtained 50% yield of
sorbitol at 92% conversion of cellulose over the reshaped Ni parti-
cles on carbon nanofibers [20], their further study indicated that
the metal active sites and the acidic functional group on Ni/CNFs
should be properly balanced, and the 7.5 wt% Ni/CNFs with a rela-
microcrystalline cellulose over Ni catalysts, including the hydroge-
nation of cellobiose and glucose, as well as the dehydrogenation of
sorbitol, were discussed in order to figure out the correlation
between the catalytic performance and the nature of the catalyst.
The kinetic study demonstrated that the hydrogenation/
dehydrogenation ability of Ni catalysts had significant influence
on the catalytic activity and selectivity of Ni catalysts in converting
cellulose into hexitols.
2. Experimental
2.1. Materials
Microcrystalline cellulose (relative crystallinity of about 74.6%,
Alfa Aesar) was dried at 70 °C for 24 h before use. Ni(NO3)2Á9H2O
(AR) was purchased from Sinopharm Chemical Reagent, SiO2
(Sigma–Aldrich), TiO2 (Sigma–Aldrich),
c-Al2O3, bentonite, and
kieselguhr (Sinopharm Chemical Reagent) were used as received.
ZSM-5 (NKF-5, H type, Si/Al = 25, 38, 50), HY, and USHY were
purchased from the Catalyst Plant of Nankai University. Nano
CuO was obtained from Aladdin.
2.2. Catalyst preparation
The supported Ni catalysts were prepared by a modified incip-
ient impregnation method. In detail, a series of supports (Al2O3,
SiO2, ZSM-5, bentonite, kieselguhr, and TiO2) were immersed in
an aqueous Ni(NO3)2 solution with a certain concentration at room
temperature. The mixture was treated under ultrasonic condition
for 0.5 h and the solid suspension was formed. Then, the solid sus-
pension was dried at 70 °C with stirring. During this process, the
solvent was evaporated slowly until the solid suspension was dried
to powder again. This process will take about 3 h; after that, the
mixture was vacuum dried at 60 °C for 12 h, followed by calcina-
tions under Ar atmosphere at 450 °C for 2 h with heating rate of
5 °C minÀ1. Prior to reaction or characterization, all the catalysts
were reduced under H2 atmosphere for 2 h at appropriate temper-
À1
tively high amount of Ni surface atoms (26:9 mmol g ) and low
cat:
À1
density of Brønsted acid sites (0.02 mmol
g
) gave 76% yield
H+
of hexitols at a cellulose conversion of 93% [21]. Zhang et al.
reported that nickel phosphides supported on activated carbon
and SiO2 were effective for conversion of cellulose to sorbitol,
and about 48% yield of sorbitol was reached at complete conver-
sion [22]. Moreover, they developed various Ni-based bimetallic
catalysts using mesoporous carbon (MC) as support; more recently,
nearly 60% yield of sorbitol was obtained over the Ir–Ni/MC
catalyst [23]. In addition, they also developed a binary catalyst
system composed of tungstic acid and Raney Ni which produced
65% of ethylene glycol [24]. 20% Ni/ZnO was reported to be one
of the most effective catalysts among a series of the supported Ni
catalysts, with it 70% yield of glycols (1,2-propanediol, ethylene
glycol, 1,2-butanediol, and 1,2-hexanediol), which was obtained
at complete cellulose conversion [25]. The above studies suggested
that the support has a significant effect on the product selectivity
in cellulose hydrogenation over the Ni based catalysts. However,
the real role of Ni particles on the hydrolytic hydrogenation was
still unclear as the whole process involved hydrolysis, hydrogena-
tion, and hydrogenolysis steps was rarely studied. Specially, the
relationship between the activity and the nature of hydrogenation
catalysts in conversion of cellulose was still elusive.
For production of hexitols, a desired Ni catalyst should promote
the hydrogenation of C@O bond in aldose or polysaccharide, but
retard the further hydrogenolysis of hexitols or parallel reactions
including aqueous phase reforming of polyols. We previously
reported that Ni/ZSM-5 could efficiently catalyze cellulose into
hexitol with about 91% selectivity at 48.6% conversion at 230 °C
[26]. We demonstrated the Ni/ZSM-5 catalyst with petaloid-like
nickel particles could not only favor the hydrogenation of the
glucose formed, but also suppress the further hydrogenolysis of
sorbitol, leading to a high yield of sorbitol. However, there is still
little knowledge about the effect of the Ni catalysts on the conver-
sion of cellulose to sorbitol. Therefore, it is essential to figure out
the key factors that determine the production of hexitol. In this
paper, the basic reaction steps in the hydrolytic hydrogenation of
atures with a heating rate of 5 °C minÀ1
.
2.3. Catalyst characterization
Powder X-ray diffraction of the samples was recorded on a
Bruker D8 Advance X-ray diffractometer with a Cu K source
(k = 0.154 nm) in the 2h range 10–80° with a scan speed of
10° minÀ1
a
.
H2-temperature-programmed reduction (H2-TPR), H2-tempera-
ture-programmed desorption (H2-TPD), and NH3-temperature-
programmed desorption (NH3-TPD) were conducted on a Tianjin
XQ TP-5080 chemisorption instrument with a thermal conductiv-
ity detector (TCD). As for H2-TPR, 30 mg of fresh catalyst was
loaded into a quartz reactor. Before H2-TPR, the catalysts were
heated at 150 °C for 30 min in nitrogen and then cooled to room
temperature. The sample was reduced in a 10% H2/N2 flow with a
heating rate of 10 K minÀ1. The effluent gas was analyzed with a
thermal conduction detector (TCD). As for the H2-temperature-
programmed desorption, 100 mg of each catalyst was loaded into
a quartz reactor and then reduced with a H2 flow at appropriate
temperature. After reduction, the reactor was cooled down to room
temperature and then the catalyst sample was maintained under
H2 flow for 30 min. Following the desorption step, the reactor
was flushed with N2 for 2 h to reach a stable background. At last,
H2-TPD was carried out with N2 at a flow rate of 30 mL minÀ1
and a temperature ramp rate of 10 °C minÀ1. The process of NH3-
TPD was same as that of H2-TPD.