Liquid phase glucose hydrogenation to D-glucitol over an ultrafine Ru-B amorphous alloy catalyst

Article abstract of DOI:10.1016/S1381-1169(03)00008-6

A Ru-B amorphous alloy catalyst in the form of ultrafine particles was prepared by chemical reduction of RuCl₃ with borohydride in aqueous solution, whose amorphous structure was confirmed by XRD, DSC, and SAED. Heating pretreatment resulted in the rapid crystallization and the deep decomposition of the Ru-B amorphous alloy as well as the abrupt decrease in the surface area due to the gathering of small particles at high temperature. XPS spectra revealed that partial electrons transferred from the alloying B to the metallic Ru in the as-prepared Ru-B sample. In comparison with other catalysts, the as-prepared Ru-B amorphous catalyst exhibited excellent activity and perfect selectivity to D-glucitol as well as superior lifetime during the liquid phase glucose hydrogenation, showing its potential application in industrial process. The higher activity of the Ru-based catalysts than that of other metal catalysts, such as Co-B and Ni-B amorphous catalysts as well as Raney Ni catalysts, demonstrated that the metallic Ru was more active than both metallic Ni and Co for the glucose hydrogenation. Meanwhile, the Ru-B amorphous catalyst exhibited higher activity than its corresponding crystallized Ru-B and pure Ru powder catalysts, showing the promoting effects of both the amorphous structure and the electronic interaction between the metallic Ru and the alloying B, which was briefly discussed based on the kinetic studies and various characterizations.

Full text of DOI:10.1016/S1381-1169(03)00008-6

Journal of Molecular Catalysis A: Chemical 200 (2003) 213–221
Liquid phase glucose hydrogenation to d-glucitol over an
ultrafine Ru-B amorphous alloy catalyst
Haibing Guoa, Hexing Lia,∗, Jian Zhua, Wanhua Yea, Minghua Qiaob, Weilin Daib
a
Department of Chemistry, Shanghai Normal University, Shanghai 200234, PR China
b
Department of Chemistry, Fudan University, Shanghai 200433, PR China
Received 26 September 2002; received in revised form 6 November 2002; accepted 12 December 2002
Abstract
A Ru-B amorphous alloy catalyst in the form of ultrafine particles was prepared by chemical reduction of RuCl3 with
borohydride in aqueous solution, whose amorphous structure was confirmed by XRD, DSC, and SAED. Heating pretreatment
resulted in the rapid crystallization and the deep decomposition of the Ru-B amorphous alloy as well as the abrupt decrease
in the surface area due to the gathering of small particles at high temperature. XPS spectra revealed that partial electrons
transferred from the alloying B to the metallic Ru in the as-prepared Ru-B sample. In comparison with other catalysts, the
as-prepared Ru-B amorphous catalyst exhibited excellent activity and perfect selectivity to d-glucitol as well as superior
lifetime during the liquid phase glucose hydrogenation, showing its potential application in industrial process. The higher
activity of the Ru-based catalysts than that of other metal catalysts, such as Co-B and Ni-B amorphous catalysts as well as Raney
Ni catalysts, demonstrated that the metallic Ru was more active than both metallic Ni and Co for the glucose hydrogenation.
Meanwhile, the Ru-B amorphous catalyst exhibited higher activity than its corresponding crystallized Ru-B and pure Ru
powder catalysts, showing the promoting effects of both the amorphous structure and the electronic interaction between the
metallic Ru and the alloying B, which was briefly discussed based on the kinetic studies and various characterizations.
© 2003 Elsevier Science B.V. All rights reserved.
Keywords: Glucose; d-Glucitol; Hydrogenation; Ru-B amorphous catalyst
1. Introduction
employed owing to their high activity and low cost
[7,8].
d-Glucitol is widely used in industry as a starting
material for the Vitamin C synthesis, an emulsifier for
fatty acid ester production, an intermediate for drug
design, and as an additive in food, cosmetic and paper
products [1–5]. The natural d-glucitol can be refined
from red seaweed and many fruits of the plant family
Rosaceae [6], but large-scale production of d-glucitol
always depends on the catalytic hydrogenation of glu-
cose, in which Raney Ni catalysts are most frequently
However, Raney Ni is fragile and easy to suffer poison-
ing. In addition, the preparation of Raney Ni by alkali
leaching Ni–Al alloy always causes environmental
∗
Corresponding author. Tel.: +86-21-643-22930.
E-mail address: hexing-li@shtu.edu.cn (H. Li).
1381-1169/03/$ – see front matter © 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S1381-1169(03)00008-6

214
H. Guo et al. / Journal of Molecular Catalysis A: Chemical 200 (2003) 213–221
pollution. Due to the industrial requirements and en-
vironmental considerations, there is a strong driving
force to design new environmentally friendly catalysts
which are more powerful and more resistant in glucose
hydrogenation [9,10]. Ru-based catalysts are potential
in the glucose hydrogenation; however, their high cost
is posing a problem. Thus, more active Ru-based cata-
lysts should be designed in order to reduce the cost of
the catalyst. Recently, the metal–metalloid amorphous
alloy catalysts have caused much attention owing to
their higher activity, better selectivity and stronger
sulfur resistance during various hydrogenation reac-
tions [11–14]. However, the glucose hydrogenation
over Ru-based amorphous alloy has never been re-
ported so far. In this paper, we report an ultrafine Ru-B
amorphous alloy catalyst with the object to compare
its performance with other Ru-based catalysts and
other metal-based catalysts including Ni-B and Co-B
amorphous catalysts as well as Raney Ni during liquid
phase glucose hydrogenation to d-glucitol. Correla-
tion of the catalytic properties to both the structural
and electronic characteristics will also be discussed
briefly.
aqueous solution at 373 K. Raney Ni was available
commercially from a d-glucitol production factory in
China and was used without additional activation.
2.2. Catalyst characterization
The composition and surface area of the as-prepared
catalysts was analyzed by inductively coupled plasma
(ICP, Jarrell-As Scan 2000). The surface morphology
and particle sizes of the Ru-B catalyst were observed
by means of transmission electron micrography (TEM,
JEM-2010). The amorphous structure was determined
by both the X-ray powder diffraction (XRD, Bruker
AXS D8-Advance with Cu K␣ radiation) and the se-
lective area electronic diffraction (SEAD, JEM-2010).
Thermal stability was examined by diffraction scan-
ning calorimetry (DSC Perkin-Elmer) performed un-
der N2 atmosphere at the heating rate of 10 K/min. The
surface electronic states were characterized by X-ray
photoelectron spectroscopy (XPS, Perkin-Elmer PHI
5000C ECSA, Al K␣ radiation). To avoid the surface
oxidation, the Ru-B samples were dried in situ in Ar
atmosphere. All the binding energy (BE) values were
calibrated using C1S = 284.6 eV as a reference. The
BET surface area was determined by N2 adsorption at
77 K using ASAP 2010 Micromeritics.
2. Experimental
2.1. Catalyst preparation
2.3. Activity test
The Ru-B amorphous catalyst was prepared by
chemical reduction of RuCl3 with 2.0 M KBH4 con-
taining 0.2 M NaOH in aqueous solution at room tem-
perature. Under stirring, excess KBH4 (molar ratio of
BH4−/Ru3+ = 5 : 1) was added dropwise into RuCl3
solution. During the reaction, a lot of bubbles were
released and the black solid was formed. After bub-
ble release ceased, the solution was kept stirring for
another 1 h to ensure the complete reduction of Ru3+
ions. Then, the black solid was filtrated and washed
repetitively with distilled water until pH = 7. Since the
as-prepared Ru-B sample was extremely active, it was
stored in water to avoid its oxidation. The correspond-
ing crystallized Ru-B sample was obtained by treating
the fresh Ru-B sample at 873 K for 2 h in N2 flow.
For comparison, Co-B and Ni-B amorphous catalysts
were also prepared by in the similar way as mentioned
above. The pure Ru powder catalyst was prepared by
chemical reduction of RuCl3 by NH2NH2 in NaOH
The hydrogenation of glucose was carried out
in a 200 ml stainless autoclave charged with 50 ml,
50 wt.% glucose aqueous solution and 0.3 g catalyst at
P(H2) = 4.0 MPa, T = 353 K. The solution was vig-
orously stirred at 1200 rpm which was high enough
to eliminate the diffusion effect as discussed below.
According to the drop of hydrogen pressure within
the first hour, the initial activity of the as-prepared
catalysts was calculated using the ideal gas equation.
Both the specific activity, i.e. the hydrogen uptake rate
per gram Ru or other metals (Rm, mmol h−1 g−1Ru)
and the areal activity, i.e. the hydrogen uptake rate per
m2 of the BET surface area (RS, mmol h−1 m−2) were
employed to express the activity. The RS could display
the difference in the nature of the active sites since
the dispersion effect has been excluded. In concept,
the active surface area, detected by either hydrogen or
CO chemisorption, should be used in calculating RS.
However, we found it difficult to determine the active

H. Guo et al. / Journal of Molecular Catalysis A: Chemical 200 (2003) 213–221
215
surface area of the Ru-B amorphous catalyst since
during the hydrogen chemisorption measurement,
heating treatment at more than 573 K was essential
to clean the catalyst surface which may have resulted
in the crystallization. Thus, for the unsupported cat-
alysts like the ultrafine Ru-B amorphous alloy parti-
cles, the BET surface area was roughly used instead
of the active surface area in calculating RS. During
the hydrogenation, the products were analyzed by a
gas chromatography equipped with a 25 m OV 101
capillary column and a FID. Besides glucose, only
d-glucitol was detected under the present conditions
regardless of the catalysts, showing that the selectivity
to d-glucitol is perfect (100%). The glucose conver-
sion was determined by means of chemical titration
with Fehlings agent [15].
when the Ru-B sample was treated at the temperature
below 473 K. However, when the Ru-B sample was
treated at the temperature > 473 K, the original broad
peak disappeared and various diffractional peaks ap-
peared, as shown in Fig. 1b. The number and strength
of the diffractional peaks increased gradually with the
increase of treating temperature and reached the max-
imum at 873 K, as shown in Fig. 1c,d. Thus, one can
conclude that the crystallization of the Ru-B amor-
phous alloy began at 473 K and reached completion
at 873 K. Only metallic Ru phases were found on the
XRD pattern of the Ru-B sample after crystallization,
indicating a deep decomposition of the Ru-B alloy dur-
ing the heating treatment. The amorphous characteris-
tic of the Ru-B sample was further confirmed by means
of SAED. As shown in Fig. 2, the fresh Ru-B sam-
ple displayed various diffractional circles indicative
of the amorphous structure [18], which disappeared
after being treated at 873 K. Furthermore, the DSC
analysis also demonstrated that the Ru-B amorphous
alloy was thermodynamically metastable. During the
heating treatment, the Ru-B amorphous alloy would
undergo spontaneous crystallization, corresponding
to a strong exothermic peak, as shown in Fig. 3.
Fig. 4 shows the XPS spectra of the fresh Ru-B
amorphous catalyst. One can see that nearly all the
Ru species were present in the metallic state corre-
sponding to two XPS peaks at binding energy (BE) of
279.8 eV in Ru3d and 284.0 eV in Ru3d levels, re-
3. Results and discussion
Fig. 1 shows the XRD patterns of the Ru-B sample
treated at different temperatures in N2 flow for 2 h.
As shown in Fig. 1a, the fresh Ru-B sample exhibited
only one broad peak around 2θ = 44◦, similar to
those observed by using the fresh Ni-B and Co-B sam-
ple, implying that the as-prepared Ru-B sample was
present in a typical amorphous structure [16,17]. No
significant change in the XRD pattern was observed
3/2
5/2
spectively. As the peak of metal Ru in Ru3d just cov-
3/2
ers the peak of C1S [19], only the peak in Ru3d level
5/2
was employed in the following discussion. Unlike Ru
species, the B species in the Ru-B amorphous catalyst
were present in three states. As can be seen from the
B1S level, the peak at BE of 188.3 eV was indicative
of the elemental B, while the peaks corresponding to
BE of 191.5 eV and 193.0 eV were attributed to the
−
oxidized B in the forms of BO2 and B2O3, respec-
tively [20]. In comparison with the standard BE val-
ues of the pure Ru metal (280.04 eV) and the pure B
(188.3 eV), it was obvious that the BE of the metallic
Ru in the Ru-B amorphous catalyst shifted negatively
by 0.2 eV while the alloying B shifted positively by
1.2 eV. Chen also observed the positive BE shift of
the alloying B in the Ni-B amorphous alloy [11]. He
believed that partial electrons transferred from the
alloying B to oxygen in the neighboring B2O3 rather
than to the metallic Ni since the electronegativity of B
Fig. 1. XRD patterns of the Ru-B sample treated at different
temperatures in H2 atmosphere, each for 2 h, respectively: (a)
as-received; (b) 473 K; (c) 673; (d) 873 K.

216
H. Guo et al. / Journal of Molecular Catalysis A: Chemical 200 (2003) 213–221
Fig. 2. SAED patterns of the Ru-B sample: (a) as-received; (b) after being treated at 873 K for 2 h in H2 atmosphere.
is much higher than that of Ni. However, when KBH4
was decomposed into the mixture of the free B and
the B2O3 species by lowering the pH of the KBH4 so-
lution, the XPS spectra did not display the BE shift of
the free B species. This means that the positive shift
in the BE value of the B occurred only in the presence
of metallic Ni, rather than in the presence of O atoms
in the neighboring B2O3. Thus, as reported previously
by Okamoto et al. [21], we concluded that partial elec-
trons transferred from B to metallic Ru in the Ru-B
amorphous alloy, making Ru electron-enriched while
B electron-deficient. This was reasonable by consid-
ering the assumption that the bonding electrons of the
B occupied the vacant d-orbitals of metallic Ru [22].
The much less BE shift of the metallic Ru than that
of the alloying B in the Ru-B amorphous alloy was
possibly owing to the relatively greater atomic weight
of the Ru atom comparing to that of the B atom [20].
Fig. 3. DSC curves of the Ru-B amorphous catalyst.

H. Guo et al. / Journal of Molecular Catalysis A: Chemical 200 (2003) 213–221
217
did not depend on the stirring speed. Therefore, a stan-
dard stirring speed of 1200 rpm was used subsequently.
At such a stirring speed, the influence of the mass of
the Ru-B amorphous catalyst on the glucose conver-
sion (after reaction for 1 h) was also investigated. The
proportional relationship clearly demonstrated the ab-
sence of external mass transfer limitation. In addition,
since the as-prepared Ru-B particles were very small
with average size around 30–50 nm and nearly porous-
less, it could be considered that there was no influence
of internal diffusion in the liquid phase hydrogenation
of glucose [23]. These results suggested that mass
transfer effects could be neglected under the present
conditions and the hydrogenation of glucose was con-
trolled by the intrinsic kinetics of the reaction. Thus,
the kinetic behaviors in Fig. 5 could be understood by
considering the difference in the adsorption strength
between glucose and hydrogen on the Ru-B amor-
phous catalyst. As the glucose molecule was strongly
adsorbed, it reached to saturated adsorption rapidly
even at low concentration. Thus, the change in the
glucose concentration in the liquid phase did change
its adsorption amount on the catalyst and thus, did
not affect the rate of the surface hydrogenation. Only
at very low glucose concentration, the hydrogenation
rate increased with the increase of glucose concen-
tration in the liquid phase since its surface adsorption
was unsaturated. Similarly, the glucose hydrogenation
exhibited first-order with respect to hydrogen pressure
since the adsorption of hydrogen on the Ru-B amor-
phous catalyst was relatively weak and could not reach
saturated adsorption under the present reaction condi-
tions. According to the Langmuir isothermal equation,
the increase in the hydrogen pressure would increase
the adsorption amount of hydrogen on the catalyst
surface and in turn, increase the hydrogenation rate.
As shown in Table 1, both the specific activity
(Rm) and the areal activity (RS) during the liquid
phase glucose hydrogenation changed in the order of
fresh Ru–B > crystallized Ru–B > pure Ru > fresh
Ni–B > fresh Co–B > Raney Ni. The following
conclusions could be drawn from the above results:
Fig. 4. XPS spectra of the as-prepared Ru-B catalyst.
Fig. 5 shows that the initial rate increased linearly
with the increase of H2 pressure from 1.0 to 4.0 MPa,
implying that the glucose hydrogenation was first-
order with respect to hydrogen pressure. However, the
influence of glucose concentration on the hydrogena-
tion rate was relatively complex. With the increase of
glucose concentration, the initial rate first increased
proportionally and then reached a plateau at glucose
concentration >40 wt.%. In order to examine the mass
transport problems, preliminary rate measurements
were performed on the fresh Ru-B amorphous catalyst
at hydrogen pressure of 4.0 MPa and glucose concen-
tration of 50 wt.%. First, the dependence of the rate
upon the stirring speed was studied, which revealed
that there was a plateau above 1000 rpm where the rate
1. The Ru-based catalysts exhibit much higher activ-
ity than both the Ni- and Co-based catalysts, about
8–9 times higher than Ni-B and Co-B amorphous
catalysts, and 50 times higher than Raney Ni cat-
alyst. Meanwhile, the Ru-B amorphous catalyst

218
H. Guo et al. / Journal of Molecular Catalysis A: Chemical 200 (2003) 213–221
Fig. 5. Dependence of the initial specific activity (Rm) on: (a) P(H2); (b) glucose concentration during the glucose hydrogenation over
Ru-B amorphous alloy catalyst. Reaction conditions: 0.3 g catalyst, 50 ml 50 wt.% glucose aqueous solution, T = 353 K, P(H2) = 4.0 MPa,
stirring rate = 1200 rpm.
could be used repetitively for more than 13 times
without significant deactivation while the Co-
or Ni-based catalysts could be repetitively used
only for less than six times, showing its excel-
lent durability under the present conditions. Thus,
the Ru-B amorphous catalyst might be used in
industrial production of d-glucitol via glucose hy-
drogenation owing to its higher activity and longer
lifetime which could compensate for its relatively
higher cost. The higher activity of the Ru-B amor-
phous catalyst than Ni-based catalysts could be
interpreted in terms of their activation energies.
According to the dependence of the initial rate of
glucose hydrogenation on the reaction temperature
from 343 to 383 K, as shown in Fig. 6, the apparent
activation energy was determined as 14.5 kJ/mol

H. Guo et al. / Journal of Molecular Catalysis A: Chemical 200 (2003) 213–221
219
Table 1
Structural properties and catalytic behaviors of the as-prepared catalystsa
Catalyst
Composition (at.%)
SBET (m2 g−1
)
Rm (mmol h−1 g−1 cat.)
Rs (mmol h−1 m−2
)
Conversion (%)
Fresh Ru-B
Fresh Ni-Bb
Fresh Co-Bb
Crystallized Ru-Bc
Pure Ru
Ru88.9B11.1
Ni75.0B25.0
Co76.0B24.0
Ru88.9B11.1
Ru
12.7
17.8
25.3
5.7
8.1
106
504.2
60.3
50.6
57.9
57.0
10.6
39.7
3.4
2.0
10.2
7.0
0.10
95.1
58.8
47.8
23.1
22.2
16.7
Raney Nid
Ni
a Reaction conditions: 0.3 g catalyst, 50 ml, 50 wt.% glucose aqueous solution, T = 353 K, P(H2) = 4.0 MPa, stirring rate = 1200 rpm,
reaction time = 2.0 h.
b Hydrogenation at T = 373 K.
c Obtained by treating the Ru-B sample at 873 K under H2 flow for 2 h.
d1.0 g Raney Ni was used.
by means of the Arrhenius equation, which was
only about a quarter of that obtained over Raney
Ni catalyst (64.8 kJ/mol), showing that the metallic
Ru was more active than the metallic Ni.
of the Ru-B sample remained almost unchanged
after crystallization. Thus, the higher activity of
the Ru-B amorphous catalyst than its crystallized
counterpart could be understood by considering the
following factors. Firstly, the TEM picture revealed
that the Ru-B amorphous alloy was present in the
form of spherical particles with average size around
30–50 nm. After being treated at 873 K for 2 h, big
lumps appeared owing to the gathering of small
particles. These results demonstrated that the crys-
tallization of the Ru-B amorphous alloy caused an
abrupt decrease in SBET (see Table 1), which might
2. The Ru-B amorphous alloy catalyst exhibited
much higher activity than both the crystallized
Ru-B catalysts and the pure Ru powder catalyst.
According to ICP analysis, no significant change
in the bulk composition of the Ru-B catalyst was
observed after crystallization. Furthermore, by ran-
dom sampling 10 particles, the selected area EDX
also demonstrated that the average composition
Fig. 6. Dependence of the initial specific activity (Rm) on the reaction temperature during the glucose hydrogenation over Ru-B amorphous
alloy catalyst. Other reaction conditions are given in Fig. 5.

220
H. Guo et al. / Journal of Molecular Catalysis A: Chemical 200 (2003) 213–221
be an important factor responsible for the decrease
On one hand, more free Ru active sites were left
which could be used for the hydrogen adsorption.
Therefore, more hydrogen molecules could be
adsorbed dissociatively on the Ru-B amorphous
catalyst, which may increase the hydrogenation
rate taking into account that the glucose hydro-
genation was first-order with respect to hydrogen
while zero-order with respect to glucose under the
present reaction conditions. On the other hand, the
electropositive boron attracted the electron pair of
in the hydrogenation activity (Rm) after crystalliza-
tion. Examining the SBET and Rm values, one can
immediately find that the Rm of the Ru-B amor-
phous catalyst was nearly nine times higher than
that of the corresponding crystallized Ru-B while
the former SBET was only two times as that of the
latter, showing that the change in the surface area of
the Ru-B catalyst after crystallization was not a cru-
cial factor accounting for the decrease of its hydro-
genation activity. Thus, the nature of the Ru active
sites must play a key role in determining the activity
of different Ru-based catalysts which could be dis-
cussed based on the intrinsic activity (RS), as listed
in Table 1. By comparing the RS values, it was obvi-
ous that the Ru active sites in the Ru-B amorphous
alloy were much more active than those in either
the pure Ru powder catalyst or the corresponding
crystallized Ru-B catalyst. From the viewpoint of
structural effect, the higher RS of the fresh Ru-B
catalyst might be attributed to the promoting ef-
fect of the amorphous characteristics, such as the
short-range ordering but long-range disordering
structure and the highly unsaturated coordination
of the Ru active sites in the Ru-B amorphous alloy.
These factors have been claimed to be favorable
for most hydrogenation reactions [24,25]. From
the viewpoint of electronic effect, the higher RS of
the fresh Ru-B catalyst might be attributed to the
promoting effect of the electron transfer between
metallic Ru and alloying B, making Ru slightly
electron-enriched while B electron-deficient, as
confirmed by the aforementioned XPS spectra.
According to the adsorption model of the glucose
molecule, the high electron density on the metallic
Ru might weaken the strength for glucose adsorp-
=
oxygen of the C O bond, which was then polarized
and more easily hydrogenated. After being crystal-
lized at 873 K, the hydrogenation activity decreased
abruptly since the electronic interaction diminished
due to the deep composition of Ru-B alloy (see the
XRD patterns). Similarly, the above discussions
could also explain the higher activity of the Ni-B
amorphous alloy catalyst than that of Raney Ni or
the corresponding crystallized Ni-B catalysts.
3. Although the Ru-B alloy had completely decom-
posed after being treated at 873 K, the crystallized
Ru-B catalyst still exhibited higher RS than the
pure Ru powder catalyst. Thus, we concluded that
the presence of the free B and the oxidized B
species might also be favorable for the glucose hy-
drogenation, as observed by Chen and coworkers
during the liquid phase hydrogenation of nitroben-
zene over supported Pt catalysts [27]. However,
no quantitative discussion could be conducted at
present without an exact knowledge of the metal
surface coverage by these promoters and without
determination of exact surface content of the free
B and the oxidized B species.
4. Conclusions
=
tion since it might repel the C O group due to
the presence of lone electron pair on the O atom.
Meanwhile, the electron-deficient B might adsorb
glucose molecules via a side-bonding model [26],
as shown in the following diagram:
The above experiments showed that the ultra-
fine Ru-B amorphous catalyst exhibited excellent
higher activity than other Ru-based catalysts includ-
ing the crystallized Ru-B and the pure Ru powder
catalysts owing to its unique amorphous structure
and surface electronic characteristics. The activity
of the as-prepared Ru-B amorphous catalyst was
nearly 10 times as that of the Raney Ni catalyst, or
the Ni-B amorphous alloy catalyst. Meanwhile, the
Ru-B amorphous catalyst also exhibited much longer
lifetime. These promoting factors could sufficiently

H. Guo et al. / Journal of Molecular Catalysis A: Chemical 200 (2003) 213–221
221
compensate the higher cost of the Ru-B amorphous
alloy catalyst, implying that it may be used instead
of Raney Ni in the glucose hydrogenation to produce
d-glucitol. Before its application in real industrial
processes, chemical engineering studies must be con-
ducted and suitable reactor design must be optimized.
Such studies are being considered in the future.
[7] F.B. Bezhanov, D.V. Sokolskii, N.K. Nadirov, S. Khando-
dzhaev, Kinet. Catal. 12 (1971) 217.
[8] N.I. Shcheglov, D.V. Sokolskii, Nauk. Akad. Nack. Kazakh.
SSR 5 (1959) 92.
[9] J. Volf, J. Pasek, Stud. Surf. Sci. Catal. 27 (1986) 105.
[10] B.J. Arena, Appl. Catal. A: Gen. 87 (1992) 219.
[11] Y. Chen, Catal. Today 44 (1993) 3.
[12] M. Shibata, T. Masumoto, Prep. Catal. 4 (1987) 353.
[13] A. Molnar, G.V. Smith, M. Bartok, Adv. Catal. 36 (1989) 329.
[14] H.X. Li, W.J. Wang, J.F. Deng, J. Catal. 191 (2000) 257.
[15] H.X. Li, H. Li, W.J. Wang, J.F. Deng, Chem. Lett. (1999)
629.
Acknowledgements
[16] S. Yoshida, H. Yamashita, T. Funabiki, T. Yonezawa, J. Chem.
Soc., Faraday Trans. I 80 (1984) 1435.
[17] H. Yamashita, M. Yoshikawa, T. Funabiki, S. Yoshida, J.
Chem. Soc., Faraday Trans. I 81 (1985) 2485.
[18] S. Klein, J.A. Martens, R. Parton, K. Vercruysse, P.A. Jacobs,
W.F. Maier, Catal. Lett. 38 (1996) 209.
This work was supported by the National Natural
Science Foundation of China (29973025) and by the
Natural Science Foundation of Shanghai Science and
Technology Committee.
[19] R. Schulz, A. van Neste, P.A. Zielinski, S. Boily, F.
Czerwinski, J. Szpunar, J. Kaliaguine, Catal. Lett. 35 (1995)
89.
References
[20] H. Li, H.X. Li, W.L. Dai, W.J. Wang, Z.G. Fang, J.F. Deng,
Appl. Surf. Sci. 152 (1999) 25.
[21] Y. Okamoto, Y. Nitta, T. Imanaka, S. Teranishi, J. Chem.
Soc., Faraday Trans. 1 75 (1979) 2027.
[22] T. Imanaka, Y. Nitta, S. Teranishi, Bull. Chem. Soc. Jpn. 46
(1973) 1134.
[23] T. Furusawa, J.M. Smith, A. I. Ch. E. J. 20 (1974) 88.
[24] A. Baiker, Faraday Discuss. Chem. Soc. 87 (1989) 239.
[25] J.F. Deng, H.X. Li, W.J. Wang, Catal. Today 51 (1999) 121.
[26] H.X. Li, X.F. Chen, M.H. Wang, Y.P. Xu, Appl. Catal. A:
Gen. 225 (2001) 117.
[1] K. van Gorp, E. Boerman, C.V. Cavenaghi, P.H. Berben,
Catal. Today 52 (1999) 349.
[2] D.J. Cram, G.S. Hammond, Organic Chemistry, second ed.,
McGraw-Hill, New York, 1964, p. 647.
[3] H.X. Li, H.X. Li, M.H. Wang, Appl. Catal. A: Gen. 207
(2001) 129.
[4] X.H. Mu, X. Wang, B.N. Bao, X.T. Shu, E.Z. Min, Chin.
Fine. Chem. 16 (1999) 41.
[5] P. Gallezot, P.J. Cerino, B. Blance, G. Fleche, P. Fuertes, J.
Catal. 146 (1994) 93.
[27] C.P. Li, Y.W. Chen, Y.J. Wang, Appl. Catal. A: Gen. 119
(1994) 185.
[6] R.L. Lohmar, W. Pigman (Eds.), The Carbohydrates,
Academic Press, New York, 1957.