Efficient conversion of D-glucose into D-sorbitol over MCM-41 supported Ru catalyst prepared by a formaldehyde reduction process

Article abstract of DOI:10.1016/j.carres.2011.04.037

Ru/MCM-41 catalyst prepared by an impregnation-formaldehyde reduction method showed higher catalytic activity and sorbitol selectivity than other catalysts in the hydrogenation of glucose. SEM and XRD indicated the partial surface properties of Ru/MCM-41. Moreover, Ru dispersion and Ru surface area of Ru/MCM-41 were determined by pulse chemisorption, and the result further proved that Ru/MCM-41 had higher catalytic activity. A catalyst recycling experiment demonstrated that Ru/MCM-41 was a better catalyst and it could be reused three or four times. A speculated mechanism was proposed to illustrate the detailed process of d-glucose hydrogenation to produce sorbitol.

Full text of DOI:10.1016/j.carres.2011.04.037

Carbohydrate Research 346 (2011) 1327–1332
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Note
Efficient conversion of D-glucose into D-sorbitol over MCM-41 supported Ru
catalyst prepared by a formaldehyde reduction process
⇑
Jun Zhang, Lu Lin , Junhua Zhang, Jianbin Shi
State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou, Guangdong 510640, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Ru/MCM-41 catalyst prepared by an impregnation–formaldehyde reduction method showed higher cat-
alytic activity and sorbitol selectivity than other catalysts in the hydrogenation of glucose. SEM and XRD
indicated the partial surface properties of Ru/MCM-41. Moreover, Ru dispersion and Ru surface area of
Ru/MCM-41 were determined by pulse chemisorption, and the result further proved that Ru/MCM-41
had higher catalytic activity. A catalyst recycling experiment demonstrated that Ru/MCM-41 was a better
catalyst and it could be reused three or four times. A speculated mechanism was proposed to illustrate
Received 26 February 2011
Received in revised form 16 April 2011
Accepted 26 April 2011
Available online 3 May 2011
Keywords:
Ru/MCM-41
Hydrogenation
the detailed process of
D-glucose hydrogenation to produce sorbitol.
Ó 2011 Elsevier Ltd. All rights reserved.
D-Glucose
Mechanism
1
. Introduction
additional costs. Therefore, researchers begin to look for new cata-
lysts that could prevent metal and support leaching in the acidic
and chelating reaction medium. Furthermore, these new catalysts
should be resistant to metal sintering and poisoning.
D
-Sorbitol formed by the reduction of the free aldehyde group of
-glucose has been widely used in the food industry, as well as
drugs, cosmetics, toothpaste and so on. Moreover, -sorbitol is an
D
D
In recent years, low ruthenium loading catalysts have been
found to replace nickel-based catalysts, which show much less or
1
important intermediate for the production of
The source of raw materials for preparation of
abundant in the pulp and paper industry. In addition,
L
-ascorbic acid.
-sorbitol is very
-sorbitol
5
,10,11
D
no deactivation.
Although ruthenium is much more expensive
D
than nickel, it may be of interest for industrial uses as it is more ac-
tive than nickel, and this catalyst shows no leaching. Hoffer et al.
studied selective hydrogenation of D-glucose by carbon supported
can be further degraded into lower polylols that are downstream
products in the petrochemical industry,2 which could increase
the added value of
The catalytic hydrogenation of
been well studied in the literature. Experimental results showed
that -glucose is converted to -sorbitol according to the Lang-
muir–Hinshelwood rate law where the reaction between adsorbed
D
-sorbitol (see Scheme 1).
Ru catalysts when compared with that of Raney-type Ni.12 The
authors found that the Ru/C catalysts displayed higher activities,
and that Ru did not leach into the reaction solution. The activity
of the catalyst was proportional to the Ru surface area and inde-
pendent of the preparation method. The Cr/Fe promoted Raney-
type Ni catalyst exhibited the highest activity, but significant Fe
was detected in the reaction mixture. The catalytic activity of Ru/
supported catalysts partially depended on the organic promoters,
D-glucose into
D-sorbitol has
D
D
3
D
-glucose and dissociated hydrogen was rate determining. The
4
–6
most commonly used catalyst for this process is Raney nickel,
which usually is activated with lye. Moreover, Raney nickel could
7
13
14
be formed with different addition agents such as Cr, Fe, and Mo.
precursors (ruthenium acetate
and ruthenium trichloride )
1
0,12
15
16
However, Ni catalysts show deactivation after recycling several
and supports (activated carbon,
Al
2
O
3
2
and SiO ). Relevant
times during the hydrogenation of
D
-glucose. Like other hexitols,
research showed that the addition of organic promoters not only
increased the adsorption amount of ruthenium, but also increased
the adsorption strength of ruthenium on the support, and im-
proved the surface properties of the Ru/C catalysts. Finally, the cat-
alytic hydrogenation activity was enhanced.
sorbitol also forms chelate complexes with nickel (or Fe, Cu) easily
in aqueous solution; these complexes are stable under alkaline
conditions, which results in nickel being left in the sorbitol
6
,8,9
solution.
Most important is that the nickel must be completely
removed from the sorbitol solution when it is used in the food
industry, or as medicine or cosmetics, which results in high
Conventional impregnation and the incipient wetness tech-
niques are the methods reported most frequently for the prepara-
tion of supported ruthenium catalysts. In this paper, MCM-41
molecular sieves are selected as the carrier, and the catalyst, Ru/
MCM-41, has been prepared by an impregnation–formaldehyde
⇑
Corresponding author. Tel.: +86 186 2809 9093; fax: +86 020 22236719.
E-mail address: lclulin@scut.edu.cn (L. Lin).
0
008-6215/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2011.04.037

1
328
J. Zhang et al. / Carbohydrate Research 346 (2011) 1327–1332
Sawdust
Wood
Extraction
Hemicellulose
Cellulosans and Polyuronides)
Lignin
Cellulose
(
6 10 5
(C H O )n
n:7000-10000
O H+
2
H O H+
H
2
CHO
OH
H
CHO
H
CHO
CHO
CHO
OH
H
H
HO
HO
H
HO
HO
H
H
OH
H
HO
H
OH
H
H
HO
H
+... + HO
H
H
+
+
H
OH
OH
OH
H
H
OH
OH
OH
OH
OH
CH
2
OH
H
H
2
CH OH
CH
2
OH
CH
OH
CH
OH
2
2
D-Xylose
D-Galactose
D-Mannose
D-Glucose
D-Glucose
Human liver
H
2
H+ -3H O
H
2
2
H+ -3H
O
2
CHO
H
O
2
2
2
CH OH
H
HO
H
OH
HO
CH
2
OH
H
O
H
H
HO
H
OH
H
CHO
HO
H
OH
H
CH
2
OH
COOH
OH
H
OH
OH
H
OH
H
H
HO
H
OH
H
H
HO
H
O
OH
HO
CHO
2
CH OH
CH
2
OH
CH
OH
2
OH
OH
OH
D-Xylitol
Furfural
D-Glucose D-Mannitol
H
H
OH
H
2
CH
2
OH
CH
2
OH
D-Sorbitol
D-Gluconic acid 5-Hydroxymethylfurfural
O
2
CH OH
O
Hydrogenolysis
2
Furfuryl alcohol
O
HO
COOH
Glycol 1,2-Propanediol PDO Glycerine C4-C6 Polyols 5-Hydroxymethyl-2-furoic acid
Scheme 1. Reaction network of the preparation and conversion of
D-sorbitol.
reduction method, which is very simple and practical in compari-
son with the hydrogen reduction process, which is conducted at
high temperatures and pressures. In contrast, the Ru/MCM-41 cat-
alyst can be produced in large quantities by using a formaldehyde
reduction method and it exhibits higher catalytic activity after
being recycled several times. Regeneration involved only simple
drying, without other complicated processing. Part of the unre-
.
Ru
a)
duced catalyst could also be reduced when they are used in H
atmosphere conditions. In the present work, the Ru/MCM-41
catalyst has been applied in the hydrogenation of -glucose to
2
(
D
D-
.
sorbitol, and its activity and selectivity are compared with those
of other catalysts such as Ru/C. Pulse chemisorption, X-ray diffrac-
tion (XRD) and SEM are used to characterize the texture and sur-
face properties of catalyst.
(
b)
0
10
20
30
40
50
60
70
Fig. 1 shows the XRD patterns of MCM-41 and Ru/MCM-41. The
MCM-41 carrier was composed of silica and alumina with the Si/Al
ratio of 27. As shown in line (b), only a dispersion peak appeared at
about 43° in Ru/MCM-41 prepared by formaldehyde reduction pro-
cess without high temperature processing. The reason for this has
to do with Ru loading, if the Ru loading is less than 5 wt %, Ru could
hardly be observed as it was covered by the support as determined
by XRD. Because the actual content of Ru in Ru/MCM-41 was
2θ, degrees
Figure 1. Powder XRD patterns of the carrier and catalyst: (a) MCM-41 and (b) Ru/
MCM-41.
lite diffraction peak, and a distinct peak from the oxidation state
of Ru did not appear.
Fig. 2 gives a scanning electron micrograph at high magnifica-
tion showing the texture of Ru/MCM-41(fresh and used several
times) and Ru/C(fresh). As can be seen from the picture (a–d), there
was only little active Ru lost after being used once. Active Ru was
stably loaded in the support MCM-41, which did not show signifi-
cant leaching after being recycled several times. When Ru/MCM-41
3
.80 wt %, only a small peak appeared at about 43° in Ru/MCM-
4
1. In related articles, if the catalyst was treated with high
temperatures, the original dispersion peak disappears gradually
and a series of crystalline phase peaks appeared.16 When compar-
ing line (a) with line (b), Ru/MCM-41 just had an obvious crystal-

J. Zhang et al. / Carbohydrate Research 346 (2011) 1327–1332
1329
Figure 2. Scanning electron micrographs of catalysts: (a) Ru/MCM-41(fresh), (b) Ru/MCM-41(used once), (c) Ru/MCM-41(used twice), (d) Ru/MCM-41(used four times), and
e and f) Ru/C(fresh).
(
9
2
catalyst was used four times, there was remarkable agglomeration.
Comparing Fig. 2(a) with Fig. 2(e or f), the results showed that Ru
was uniformly distributed in support MCM-41 compared to
activated carbon. From the above analysis, the basic characteristic
of catalyst Ru/MCM-41 was stable and better than those of other
catalysts.
1
9
00
5
88
Sorbitol yield
Glucose conversion
84
Table 1 gives the main characteristics of the Ru/MCM-41 and
Ru/C catalysts. The Ru/MCM-41(fresh) catalyst showed the highest
80
2
metal dispersion, metallic surface area (m /g sample) and metallic
90
85
2
surface area (m /g metal) when compared with those of other cat-
7
7
6
2
2
alysts. The metal dispersion, metallic surface area (m /g sample)
2
and metallic surface area (m /g metal) of Ru/MCM-41 catalyst de-
creased obviously after used being twice or four times. This was
partly because a small amount of active Ru leached from the cata-
lyst and some particles agglomerated when recycling times in-
creased. Comparison of characteristic data of Ru/MCM-41(fresh)
with Ru/C(fresh) and Ru/MCM-41 (fresh) showed much higher cat-
alytic activity. The above experimental results were in agreement
with the outcome shown in Fig. 5, and the activity of the catalyst
1
2
3
4
5
6
Reaction time,h
Figure 3. Effect of reaction time on the hydrogenation of
0.25 g Ru/MCM-41, 25 mL aqueous solution, 120 °C, 3 MPa H2.
D-glucose; 2.5 g D-glucose,
converted. At the early period of the hydrogenation, glucose was
translated into sorbitol and the sorbitol yield reached the highest
level at 1.5 h. The sorbitol yield declined as the reaction pro-
gressed. Perhaps the generated sorbitol is isomerized into mannitol
during this process. In the meantime, a little sorbitol was degraded
into lower alcohols such as ethylene glycol, glycerin, propanediol
and other C4–C6 polyols. Hence, the optimum reaction time was
1.5 h.
2
was proportional to the dispersion of Ru and its surface area (m /
2
g sample and m /g metal).
Different influencing factors were investigated in the conver-
sion of
D
-glucose into
D
-sorbitol catalyzed by Ru/MCM-41. The ef-
-glucose are
fects of reaction time on the hydrogenation of
D
shown in Fig. 3. The glucose conversion increased quickly when
the reaction time was prolonged; at 3 h glucose was completely
Table 1
Characteristics of the applied catalysts
2
2
Catalyst
Metal dispersion (%)
Metallic surface area (m /g sample)
Metallic surface area (m /g metal)
Ru/MCM-41(fresh)
1.2587
0.2438
0.2135
0.7592
0.2299
0.0445
0.0390
0.1387
4.5980
0.8908
0.7801
2.7735
Ru/MCM-41(used twice)
Ru/MCM-41(used four times)
Ru/C(fresh)

1
330
J. Zhang et al. / Carbohydrate Research 346 (2011) 1327–1332
9
7
6
4
3
0
5
0
5
0
1
9
8
7
6
5
00
0
80
60
40
20
0
0
0
Sorbitol yield
0
Glucose conversion
0
1
00
110
120
130
140
0
1
2
3
4
Reaction temperature, C
Recycling times of catalyst
Figure 4. Effect of reaction temperature on the hydrogenation of
D-glucose; 2.5 g
D-glucose, 0.25 g Ru/MCM-41, 25 mL aqueous solution, 2 h, 3 MPa H
2
.
Figure 6. Effect of recycling times of catalyst on the hydrogenation of
D-glucose;
2
.0 g -glucose, 0.2 g Ru/MCM-41, 20 mL aqueous solution, 120 °C, 2 h, 3 MPa H .
D
2
Fig. 4 illustrates the influence of reaction temperature on sorbi-
sorbitol yield was not found to reduce significantly. This result
showed that Ru could be well loaded in MCM-41 support, and it
could maintain higher catalytic activity and led to the conclusion
that the Ru/MCM-41 catalyst had better activity and stabilization
than that of other catalysts.
Comparing the catalytic activity of Ru/MCM-41 with other cat-
alysts (Fig. 7) Ru/MCM-41 exhibited the highest catalytic activity
under the same reaction condition; alumel showed the lowest cat-
alytic activity. The catalytic activity of these catalysts ranged in
descending order: Ru/MCM-41 > Ru/C > Pd/C > Nickel powder >
alumel. The sorbitol yield and glucose conversion catalyzed by
Ru/MCM-41 were the highest, and the product selectivity also
was the highest. Ru/C revealed relatively high catalytic perfor-
mance, the sorbitol selectivity was up to 59.7%. While catalyst
Pd/C and Nickel powder showed lower sorbitol selectivity, this
may be due to the weaker catalytic ability of Pd and Ni by compar-
ison with Ru. As determined from the above analysis, the Ru/MCM-
tol yield and glucose conversion. At the initial stages, sorbitol yield
increased steeply with a rise of temperature. When the tempera-
ture was higher, glucose underwent several degradation pathways,
which produced some other products, leading to a decrease in sor-
bitol yield. The principal reaction was accomplished as the temper-
ature reached 120 °C; the sorbitol yield was as high as 92.13% at
1
25 °C.
The relationship between catalyst usage and sorbitol yield or
glucose conversion are indicated in Fig. 5. When there was no cat-
alyst added, little sorbitol was produced during the reaction. The
sorbitol yield increased with an increase in the amount of catalyst,
when further increasing catalyst amount, the yield did not change
significantly, and glucose had already been converted completely.
In this experiment, the amount of reactant glucose was constant,
when excessive dosage of catalyst was employed, no high conver-
sion rate could be found in this experiment. Hence the sorbitol
yield was relatively stable when the dosage of catalyst was
4
1 was a good catalyst in glucose hydrogenation.
2
0 wt %.
The Ru/MCM-41 catalysts were recycled after each experiment
Glucose hydrogenation over Ru/MCM-41 catalyst is a gas–
liquid–solid three-phase catalytic reaction. As shown in Scheme
by filtering the final solution and washing with de-ionized water.
Catalysts were only dried at 50 °C in vacuum oven without other
complex processing. As shown in Fig. 6, the catalytic activity of
Ru/MCM-41 decreased slightly after being reused once; the sorbi-
tol yield decreased from 83.13% to 68.21%. The main reason was
that some reaction products could be bonded and accumulated
on the surface of catalyst, which led to a decline in catalyst activity.
Therefore, after recycling the catalyst three or four times, the
2
, the mechanism of reducing glucose with H
2
on the catalyst
was spread from
surface is proposed to proceed as follows: First, H
air to the liquid membrane. Then H
interface and it spread from the liquid membrane to glucose in the
liquid phase. It is supposed that H did not react with the carbonyl
group of glucose, but instead was adsorbed by the active centers of
2
2
dissolved in the gas–solution
2
100
Sorbitol yield
Glucose conversion
8
6
0
0
1
00
1
8
6
4
00
0
8
0
0
0
0
6
4
2
40
0
20
0
Sorbitol yield
0
Glucose conversion
0
20
0
5
10
15
20
25
30
Catalyst usage,%
Figure 5. Effect of catalyst usage on the hydrogenation of
-glucose, 25 mL aqueous solution, 120 °C, 2 h, 3 MPa H
D-glucose; 2.5 g
Figure 7. Effect of different catalysts on the hydrogenation of
D-glucose; 2.5 g
D
2
.
D-glucose, catalyst usage: 10%, 25 mL aqueous solution, 120 °C, 2 h, 3 MPa H .
2

J. Zhang et al. / Carbohydrate Research 346 (2011) 1327–1332
1331
Ru
H2
CH OH
2
O
H
H +
OH
Ru
H
CH O
HO
HO
2
O
OH
OH
Ru-H
HO
HO
OH
CHO
OH
H
HO
H
H
OH
OH
Hydrogen transfer
H
CH OH
2
H
Ru
CH O
2
O _
O
H
H
OH
Ru
C
HO
HO
H
O
OH
H
H
HO
H
H
H2
OH
OH
H
Ru-H
CH2OH
CH O H
2
OH
OH
H
HO
HO
Hydrogen transfer
OH
H
H
_
O
OH
H
H
Ru
C
C
H
O
H
H
Ru-H
H
HO
H
OH
H2
HO
H
H
OH
OH
OH
OH
H
H
CH 2OH
CH2OH
Scheme 2.
the catalyst, producing activated H on the catalyst. Finally, glucose
reacts with the activated H on the surface of catalyst, which is an
irreversible reaction, and then the product desorbs from the cata-
lyst and diffuses into the liquid phase. Therefore, glucose hydroge-
lytic reaction, in which the carbonyl group in glucose reacted with
activated H as the main step.
nation proceeds through
adsorption on the active centers of the catalyst, to produce acti-
vated H. Finally, the carbonyl group in glucose reacted with the
H
2
dissolution,
H
2
diffusion,
H
2
2. Experimental
2.1. Materials
activated H on the surface of catalyst to produce D-sorbitol.
In conclusion, the Ru/MCM-41 catalyst prepared by an impreg-
nation–formaldehyde reduction method showed higher catalytic
activity in glucose hydrogenation, and exhibited greater product
selectivity than other catalysts such as Ru/C, alumel, Ni powder,
and Pd/C. Experimental results showed that the highest sorbitol
yield, 94.43%, could be obtained at 120 °C when the amount of cat-
alyst was 20% (based on glucose) and the reaction time was 2 h;
the glucose conversion was as high as 100%. SEM revealed that
Ru is well dispersed on the surface of MCM-41, the only dispersion
peak appeared at about 43° in Ru/MCM-41 from XRD spectrum.
Ru/C (5 wt %), alumel (Ni content: 47%) and 5 wt % Pd/C were
purchased from Aladdin Reagent Limited Company. A.P. reagents,
including nickel powder and RuCl were purchased from Tianjin
3
Kermel Chemical Reagents Limited Company. Ru/MCM-41
(5 wt %) catalyst was prepared by an impregnation–formaldehyde
reduction method. MCM-41 molecular sieve (alumina content:
3%) was employed as support material to prepare the homemade
3
Ru/MCM-41 catalyst. RuCl was selected as the precursor of Ru.
First, RuCl was dissolved with 5 wt % HCl, then MCM-41 was
3
added. After being immersed for a while, the pH value of the solu-
tion was adjusted to 5–7 by 10 wt % NaOH, which was conducive to
formaldehyde reduction. After completion of reduction, the ob-
tained product was filtered, washed with demineralized water un-
2
Moreover, the results of Ru dispersion and Ru surface area (m /g
2
sample and m /g metal) of Ru/MCM-41 further proved that Ru/
MCM-41 had a higher catalytic activity than that of other catalysts.
A catalyst recycling experiment demonstrated that Ru/MCM-41
sustained high catalytic activity after reusing several times. A pro-
3
til chloride was washed out as detected by AgNO . The product was
then dried overnight in a vacuum oven at 50 °C. During the whole
process, the color of the sample changed from rufous to dark green,
then it was light gray after reduction.
posed mechanism for
D-glucose hydrogenation over Ru/MCM-41
catalyst to sorbitol involved a gas–liquid–solid three-phase cata-

1
332
J. Zhang et al. / Carbohydrate Research 346 (2011) 1327–1332
2
.2. Catalyst characterization
Acknowledgments
Powder X-ray powder diffraction (XRD) was performed in a
Bruker D8 Advance diffractometer using CuK radiation. The oper-
The authors are grateful to the financial support from the Nat-
ural Science Foundation of China (50776035, U0733001), the Foun-
dation of Scientific Research for Universities (20070561038) from
the Ministry of Education of China, National High Technology Pro-
ject (863 project) (2007AA05Z408), National Key R&D Program
(2007BAD34B01) and the National Key Basic Research Program
(2010CB732201) from the Ministry of Science and Technology of
China.
a
ating voltage and current were 40 kV, 40 mA, respectively. The step
À1
length was 0.02° with scanning rate of 2° min . The distribution of
Ru particles in catalyst Ru/MCM-41 and Ru/C were determined
with a scanning electron microscope (SEM) Nova Nano SEM 430
operating at 10 kV.
The properties of active Ru on support MCM-41 were deter-
mined by means of pulse chemisorption using Micromeritics Auto
chem II 2920. The pretreatment of samples included the following
steps: (1) heating up to 120 °C at the rate of 10 °C/min, waiting for
References
1.
2.
3.
4.
Gallezot, P.; Cerino, P. J.; Blanc, B.; Flèche, G.; Fuertes, P. J. Catal. 1994, 146,
3–102.
Li, G. J.; Cai, H. G.; Liu, Y. Q.; Su, T. F.; Hebei, J. Ind. Sci. Technol. 2009, 26,
497–499.
5
min; (2) then heating up to 220 °C with the same temperature
9
rate speed, staying 5 min; (3) heating up to 220 °C at the same
heating rate, leaving for an additional 5 min; (4) changing gas flow
2
Perrard, A.; Gallezot, P.; Durand, R.; Cédric, B.; Coq, B.; Trens, P. Appl. Catal., A
with 10% H
2
and Ar, gas velocity of 20 cm /min; (5) reducing the
2007, 331, 100–104.
temperature to 50 °C at the rate of 20 °C/min. After these pro-
cesses, then the experiment began. The obtained results are col-
lected in Table 1.
Schimpf, S.; Louis, C.; Claus, P. Appl. Catal., A 2007, 318, 45–53.
5. Hoffer, B. W.; Crezee, E.; Devred, F.; Mooijman, P. R. M.; Sloof, W. G.; Kooyman,
P. J.; Van Langeveld Kapteijn, A. D. F.; Moulijn, J. A. Appl. Catal., A 2003, 253,
4
37–452.
Kusserow, B.; Schimpf, S.; Claus, P. Adv. Synth. Catal. 2003, 345, 289–299.
7. Gallezot, P.; Nicolaus, N.; Flèche, G.; Fuertes, P.; Perrard, A. J. Catal. 1998, 180,
1–55.
6.
2
.3. Catalytic reaction
5
8
9
.
.
Arena, B. J. Appl. Catal., A 1992, 87, 219–229.
Tukac, V. Collect. Czech. Chem. Commun. 1997, 62, 1423–1428.
The conversion of glucose to sorbitol was performed with high
pressure reactor (PARR 5500), Ru/MCM-41 catalyst and 10 wt %
glucose was added to the reactor, and then the reaction was carried
10. Crezee, E.; Hoffer, B. W.; Berger, R. J.; Makkee, M.; Kapteijn, F.; Moulijn, J. A.
Appl. Catal., A 2003, 251, 1–17.
11. van Gorp, K.; Boermana, E.; Cavenaghi, C. V.; Berben, P. H. Catal. Today 1999, 52,
2
out at several temperatures under 3 MPa H with a stirring rate of
349–361.
5
00 rpm for a certain time. After the reaction, the solid catalyst was
12. Hoffer, B. W.; Crezee, E.; Mooijman, P. R. M.; Van Langeveld, A. D.; Kapteijn, F.;
separated by filtration, and was regenerated by vacuum drying for
the next use. The liquid products were analyzed by IC (Dionex ICS-
Moulijn, J. A. Catal. Today 2003, 79–80, 35–41.
1
1
1
3. Shen, Y. F.; Lei, Y. J.; Shen, S. W.; Pan, Z. F. Precious Metal 2009, 30, 40–44.
4. Xu, S. K.; Wang, X. Y.; Liang, L. Z. J. Henan Univ. Technol. 2006, 27, 67–69.
5. Eisenbeis, C.; Guettel, R.; Kunz, U.; Turek, T. Catal. Today 2009, 147, S342–S346.
3
000) with CarboPac™ PA1 column. The eluents were NaOH and
CH COONa with a flow rate of 0.25 mL/min. The sample loop had
a volume of 50 L. The column temperature was 30 °C.
16. Wu, Y. D.; Guo, H. B.; Wan, Y.; Wang, M. H.; Li, H. X. Chin. J. Catal. 2004, 25,
33–536.
3
5
l