S.J. You et al. / Applied Catalysis A: General 466 (2013) 161–168
167
Table 5
Cellulose conversion and polyol yields in the presence of various catalyst systems.
Catalysts
Conversion (%)
Yield (%)
Sorbitol
Mannitol
Erythritol
Ethylene glycol
1,2-Propanediol
Ni/W/SiO2(R)a
Ni/W/SiO2(P)a
Ni/W/SiO2(O)a
Ni/SiO2(40) + W
99
100
100
100
100
1.9
2.8
–
1.2
0.6
1.8
0.7
–
0.4
0.1
1.9
1.7
–
0.7
0.9
29.9
26.7
–
2.6
2.7
6.1
5.4
–
0.7
0.3
b
Ni/SiO2(40) + WO3b
a
Reaction conditions: weight of cellulose = 500 mg; volume of water = 30 mL; weight of catalyst = 50 mg; reaction temperature = 518 K; reaction time = 2 h; and initial H2
pressure = 6 MPa.
b
Reaction conditions: weight of cellulose = 500 mg; volume of water = 30 mL; weight of nickel catalyst = 50 mg; weight of tungsten catalyst = 50 mg; reaction tempera-
ture = 518 K; reaction time = 2 h; and initial H2 pressure = 6 MPa.
solvent properties (pH, oxidation potential, chelating properties of
molecules) and the bulk properties and surface properties of the
solid catalyst [31].
Thus, based on the previous and present data, the following reac-
tion scheme can be considered, where first, cellulose is hydrolyzed
into glucose/oligosaccharide by H+ ions generated in hot water.
The formed sugar intermediates can then be transformed into low-
molecular-weight polyols over reduced NiW bimetallic catalysts
3.3. Effect of degree of surface oxidation
through hydrogenolysis in a H -rich atmosphere. In the presence
2
The catalytic activity compared among the various
of fully oxidized tungsten oxides, the sugar intermediates can then
be transformed into organic acids via a series of reactions such as
hydrolysis, elimination, and rearrangement [32].
Ni/W/SiO (40) catalysts with different surface oxidation degrees
2
is shown in Table 5. The highest polyols yield was obtained
over Ni/W/SiO (R). The polyol yield increased with increasing
2
fraction of the metallic Ni and W. The product distribution was
dependent on the fraction of the metallic Ni and W. The yields of
ethylene glycol, lactic acid, and acetic acid were 30%, 0%, and 0%,
4
. Conclusion
The primary crystalline size of W increased on increasing the
respectively, over Ni/W/SiO (R), while they were 0%, 22%, and 3%,
2
average pore size of support and reduction temperature for the
Ni/W/SiO2 catalysts containing 5 wt.% Ni and 25 wt.% W. However,
the amount of surface acid site decreased on increasing the average
pore size of support and reduction temperature. A certain fraction
of Ni and W can be oxidized in the passivation step at room tem-
perature. Therefore, the bulk crystalline W metal is covered with
a certain fraction of WO2 and WO3 in Ni/W/SiO2 catalysts. The
catalytic activity is inversely proportional to the crystal size of W
metal. A Ni/W/SiO2 catalyst containing reduced and oxidized Ni
and W species is effective for the hydrogenolysis of cellulose into
respectively, over Ni/W/SiO (O). An explanation for the varying
2
yields is that the Ni/W/SiO (R) catalyst containing reduced and
2
oxidized Ni and W species is effective for the hydrogenolysis of
cellulose into low-molecular-weight polyols. On the other hand,
Ni/W/SiO (O) catalyst containing only oxidized Ni and W species
2
produces organic acids via a series of reactions such as hydrolysis,
elimination, and rearrangement [32]. Chambon et al. [33] carried
out cellulose conversion using solid acids and obtained high yields
of lactic acid, levulinic acid, and formic acid.
The hydrogenolysis of cellulose into polyols involves C-C bond
cleavage and C O bond cleavage via dehydrogenation, retro-
aldol condensation, decarbonylation, dehydration, hydrogenation
low-molecular-weight polyols. On the other hand, a Ni/W/SiO cat-
2
alyst containing only oxidized Ni and W species produces organic
acids via a series of reactions such as hydrolysis, elimination, and
rearrangement. The combined catalyst systems comprising Ni/SiO2
[
34,35]. The key steps in C C bond cleavage are dehydrogena-
tion, decarbonylation, and retro-aldol condensation at metal sites
35]. Dehydration is effective at C O bond cleavage on acid sites,
and W species (W or WO ) showed lower polyol yields compared
3
[
to the Ni/W/SiO2 catalysts, thus indicating that the NiW bimetallic
catalyst has a remarkable synergistic effect on the hydrogenolysis
of cellulose.
and hydrogenation occurs on metal site [35]. For Ni/W/SiO2 cata-
lysts, metallic tungsten, oxidized tungsten, and metallic nickel are
responsible for C C bond cleavage, C O bond cleavage, and hydro-
genation, respectively. That’s why the balance and location of metal
site and acid site are critical to increase polyols yields.
Acknowledgements
In our previous work [26], the hydrogenolysis of cellulose over
Ni/W/SiO -Al O catalysts was examined. In Ni/W/SiO -Al O cat-
This work was financially supported by a grant from the Indus-
trial Source Technology Development Programs (10033099) of the
2
2
3
2
2
3
alysts, a large amount of acid sites of SiO -Al O causes smaller
2
2
3
W particle size and larger amount of acid sites, compared with
Ni/W/SiO2 catalysts in this study. Although Ni/W/SiO -Al O cata-
2
2
3
lysts have smaller W particle size than Ni/W/SiO2 catalysts, these
catalysts provide lower polyols yield owing to lots of acid sites. This
Appendix A. Supplementary data
implies that Ni/W/SiO catalysts have more proper balance of metal
2
Supplementary data associated with this article can be
site and acid site than Ni/W/SiO -Al O ones.
2
2
3
To understand the synergistic effect of Ni and W in this reac-
2
013.06.053.
tion, the combined catalyst systems composed of Ni/SiO (40) and
2
W species (W or WO ) were compared under the same reaction
3
conditions (Table 5). The amount of chemisorbed H for Ni/SiO (40)
References
2
2
was determined to be 13.1 mol/gcat, which corresponds to 3.1% Ni
dispersion. In both the cases, the total yield to polyols was 5–6%. The
lactic acid yield was 7% and 15% for Ni/SiO + W and Ni/SiO + WO
systems, respectively, indicating that NiW bimetallic catalyst had
a remarkable synergistic effect on the hydrogenolysis of cellulose.
[
[
1] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
2] M. FitzPatrick, P. Champagne, M.F. Cunningham, R.A. Whitney, Bioresour. Tech-
nol. 101 (2010) 8915–8922.
2
2
3
[3] R. Rinaldi, F. Schüth, Energy. Environ. Sci. 2 (2009) 192–196.
[4] A. Fukuoka, P.L. Dhepe, Angew. Chem. Int. Ed. 45 (2006) 5161–5163.