Hydrogenolysis of 1,2-propanediol for the production of biopropanols from glycerol

Article abstract of DOI:10.1002/cssc.201000040

Production of propanols from glycerol, which are known as biopropanols, requires catalysts for the hydrogenolysis of 1,2-propanediol, which has been easily derived from glycerol. It is found that the Rh/SiO₂ catalysts modified with ReO_x

Full text of DOI:10.1002/cssc.201000040

DOI: 10.1002/cssc.201000040
Hydrogenolysis of 1,2-Propanediol for the Production of
Biopropanols from Glycerol
Yasushi Amada,[a] Shuichi Koso,[a] Yoshinao Nakagawa,[a, b, c] and Keiichi Tomishige*[a, b, c]
Production of propanols from glycerol, which are known as
biopropanols, requires catalysts for the hydrogenolysis of 1,2-
propanediol, which has been easily derived from glycerol. It is
found that the Rh/SiO2 catalysts modified with ReOx species ex-
hibited high activity and selectivity in the hydrogenolysis of
1,2-propanediol to propanols with low selectivity to degrada-
tion products and high stability. The optimized RhÀReOx/SiO2
(Re/Rh=0.5) catalyst gave high yields of 1-propanol (66%) and
propanols (1-propanol +2-propanol) (85%) in the hydrogenol-
ysis of 1,2-propanediol. In addition, the catalyst was applicable
to the one-pot conversion of glycerol to propanols. The struc-
ture of Rh metal particles with attached ReOx clusters is sug-
gested from the catalyst characterization. It is proposed that
1,2-propanediol hydrogenolysis proceeds by the hydrogenoly-
sis of the alkoxide species on Re with hydrogen species on the
Rh metal surface.
Introduction
Much attention has been given to the transformation of bio-
mass and biomass-derived chemicals to fuel and valuable
chemicals for the sustainable development, and the facility
that integrates the processes is called biorefinery.[1] In the bio-
refinery, glycerol can be regarded as a renewable feedstock be-
cause it is the coproduct of triglyceride saponification and
transesterification with methanol and ethanol to biodiesel fuel
from vegetable oils.[2] A possible target product by the glycerol
transformation is 1,2-propanediol because 1,2-propanediol is
widely used as raw materials for polyester resins, additives for
food, cosmetics, pharmaceuticals, and functional fluids.[3] The
hydrogenolysis of glycerol to 1,2-propanediol have been exten-
sively studied in recent years,[4–16] and high yields have been
obtained in some cases.[6,10,11,16] The yield of 1,2-propanediol
became 89% over Cu/ZnO/Al2O3 catalyst in 8 MPa H2 at
503 K.[6] Ru–Cu bimetallic catalysts showed 100% of glycerol
conversion and 85% yield of 1,2-propanediol in 0.64 MPa H2 at
463 K.[11] In addition, it has been reported that a hydrotalcite
precursor-supported Pt catalyst gave 86% yield of 1,2-pro-
panediol in the glycerol hydrogenolysis in 3.0 MPa H2 at
493 K.[16] These catalysts exhibited high selectivity and yield in
the glycerol hydrogenolysis to 1,2-propanediol, and this cata-
lytic performance is related to the low activity in the overhy-
drogenolysis of 1,2-propanediol. However, overhydrogenolysis
products of 1,2-propanediol, 1- and 2-propanols, are also
useful chemicals. 1-Propanol, which has been produced via hy-
droformylation of ethylene, is used mainly as a solvent, a print-
ing ink, and a chemical intermediate for the production of n-
propyl acetate.[17] 2-Propanol has been produced by the hydra-
tion of propylene and used mainly as a solvent.[18] In contrast
to the glycerol hydrogenolysis, reports on effective catalysts
for the hydrogenolysis of 1,2-propanediol are very limited. It
has been reported that the homogeneous Ru catalyst cata-
lyzed the hydrogenolysis of 1,2-propanediol to 1-propanol at
383 K with a moderate yield of 54%.[19] At the same time, the
system demands an organic solvent such as sulfolane and uses
a homogeneous catalyst that is difficult to separate and recy-
cle.
Herein, the development of heterogeneous catalysts for the
hydrogenolysis of 1,2-propanediol without using organic sol-
vents is reported.
Results and Discussion
Optimization of the additive and its amount
The performance of the catalysts in the hydrogenolysis of 1,2-
propanediol (1,2-PrD) is compared in Table 1. Typical catalysts
for the hydrogenolysis of glycerol such as Rh/C,[20] Ru/C,[13]
Raney Ni,[10] Raney Co, Raney Cu,[5] and Cu-chromite[5] were ap-
plied to the hydrogenolysis of 1,2-propanediol. In the case of
Raney Ni, Raney Co, Raney Cu, and Cu-chromite, the reaction
temperature was increased to 453 K and/or more catalysts
were used because the catalytic activity at 393 K of these cata-
lysts were so low that it was difficult to determine the conver-
sion and selectivity. This tendency indicates that the activity of
these catalysts are very low and not suitable to the hydroge-
nolysis of 1,2-PrD to propanols. In addition, the selectivity to
[a] Y. Amada, S. Koso, Y. Nakagawa, K. Tomishige
Graduate School of Pure and Applied Sciences, University of Tsukuba,
1-1-1, Tennodai, Tsukuba, Ibaraki 305-8573 (Japan)
Fax:(+81)227957214
E-mail: tomi@tulip.sannet.ne.jp
[b] Y. Nakagawa, K. Tomishige
International Center for Materials Nanoarchitectonics Satellite (MANA),
National Institute of Materials Science (NIMS) and University of Tsukuba,
1-1-1, Tennodai, Tsukuba, Ibaraki 305-8573 (Japan)
[c] Y. Nakagawa, K. Tomishige
Present address
School of Engineering, Tohoku University, 6-6-07
Aoba, Aramaki, Aoba-ku, Sendai 980-8579 (Japan)
728
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2010, 3, 728 – 736

Production of Biopropanols from Glycerol
ing reaction temperature for
both catalysts, and Rh/SiO2
showed a larger increase. As a
result, a lower reaction tempera-
ture is more favorable to the
synthesis of propanols from 1,2-
PrD because the selectivity to
propanols is higher.
Table 1. Comparison of catalytic performances in the 1,2-propanediol hydrogenolysis over each catalyst.
Entry Catalyst
Catalyst weight
[g]
T
t
Conv.
Selectivity [%]
[K] [h] [%]
1-
2-
Propane Degradation
PrOH PrOH
1
RhÀReOx/SiO2 (Re/
0.15
393 24 86.6
74
19
6
1
Rh=0.5)
Rh/SiO2
2
3
4
5
6
7
8
9
0.15
0.15
0.15
0.15
1.5
3.0
3.0
0.15
393 24 8.8
393 24 2.8
393 24 8.3
453 24 2.2
453 24 2.1
453 96 0.3
453 96 0.7
393 48 98.2
65
58
38
0
35
29
40
68
22
29
15
2
0
40
0
10
6
12
3
1
0
3
8
Figure 3 shows the effect of
H2 pressure over RhÀReOx/SiO2
(Re/Rh=0.5) and Rh/SiO2. On
both catalysts, 1,2-PrD conver-
sion increased gradually with in-
creasing H2 pressure except the
high pressure range (>6 MPa)
for Re modified catalyst, where
the rate was saturated. There-
fore, the reaction order on H2
pressure around the standard
Rh/C
Ru/C
36
94
64
31
60
2
Raney Ni
Raney Co
Raney Cu
Cu-chromite
RhÀReOx/SiO2 (Re/
Rh=0.5)
0
11
19
Reaction conditions: 20wt% 1,2-propanediol aqueous solution (20 mL), initial H2 pressure (8.0 MPa). Degrada-
tion: ethanol+ethane+methane.
propanols was also low. In contrast, Rh/SiO2, Rh/C, and Ru/C
gave higher conversion than Raney Ni, Raney Co, Raney Cu,
and Cu-chromite even at 393 K. However, the degradation pro-
ceeded significantly on Ru/C and the selectivity to propanols
was not high. In contrast, Rh/SiO2 and Rh/C gave high selectivi-
ty to propanols, and Rh/SiO2 showed the highest activity.
Therefore, Rh/SiO2 exhibited both high activity and selectivity
in the hydrogenolysis of 1,2-propanediol.
conditions (8 MPa) would be zero. Regarding the selectivity,
Rh/SiO2 tends to give higher selectivity to degradation prod-
ucts at lower H2 pressure. In contrast, the effect of H2 pressure
on the selectivity was very small on RhÀReOx/SiO2. The result
suggests that the 1,2-PrD hydrogenolysis can proceed selec-
tively even at low H2 pressure on RhÀReOx/SiO2.
Figure 4 shows the effect of 1,2-PrD concentration on the
hydrogenolysis of 1,2-PrD over RhÀReOx/SiO2. In this experi-
ment, the total amount of 1,2-PrD was fixed and the concen-
tration of 1,2-PrD was adjusted by the amount of H2O as a sol-
vent. When the range of 1,2-PrD concentration was between
5% and 40%, the conversion of 1,2-PrD increased gradually
with increasing 1,2-PrD concentration, and the reaction order
with respect to 1,2-PrD was estimated to be 0.3. The 1,2-PrD
conversion was almost saturated at 40 and 60%. The conver-
sion decreased beyond 60% 1,2-PrD concentration. Small posi-
tive reaction order with respect to 1,2-PrD in the range of low
1,2-PrD concentration and the tendency in the high concentra-
tion range suggest that the adsorption of 1,2-PrD on the cata-
lyst surface is strong. In addition, the significant decrease in
the 1,2-PrD conversion with increasing concentration beyond
60% can be interpreted by the explanation that too high con-
centration can suppress the H2 adsorption on the catalyst sur-
face. In contrast, the conversion increased monotonously with
increasing 1,2-PrD concentration over Rh/SiO2, and the results
represent the positive reaction order with respect to 1,2-PrD.
This positive reaction order indicates weaker interaction of 1,2-
PrD with the catalyst surface. The comparison in the reaction
order with respect to 1,2-PrD between RhÀReOx/SiO2 and Rh/
SiO2 suggests that the modification of Rh/SiO2 with Re makes
the interaction of 1,2-PrD with the catalyst surface strong.
Figure 5 shows the reaction time dependence of the 1,2-PrD
hydrogenolysis over RhÀReOx/SiO2 (Re/Rh=0.5). Yield of 1-
propanol increased remarkably with increasing the reaction
time up to 24 h. The maximum yield of 1-PrOH was 66% at
48 h, and the yield of 1-propanol+2-propanol (1-PrOH+2-
PrOH) also reached 85% at 48 h. At the reaction time longer
than 48 h, the yield of propane increased significantly by the
To enhance the catalytic performance of Rh/SiO2, we modi-
fied Rh/SiO2 with various additives. Here, it is found that Re,
Mo, and W are effective additives. Figure 1 shows dependence
of additive amount on the hydrogenolysis of 1,2-PrD over Rh/
SiO2. The addition of Re, Mo, and W species enhanced the ac-
tivity of 1,2-PrD hydrogenolysis remarkably. Similar behavior in
the promoting effect of the modification on the hydrogenoly-
sis of tetrahydrofurfuryl alcohol and glycerol has been report-
ed.[15,21] In the case of Re modification, the conversion was
maximum at Re/Rh=0.5 and the conversion became 37 times
higher than that on Rh/SiO2. The RhÀReOx/SiO2 (Re/Rh=0.5)
catalyst showed high activity at lower temperature compared
to the catalysts in the literature, such as Cu/ZnO/Al2O3, bimet-
allic Ru–Cu, and Pt/hydrotalcite. Another promoting effect of
Re addition is suppression of the degradation and overhydro-
genolysis to propane.
Similar promoting effects of Mo and W are observed in Fig-
ure 1b andc, and the conversion was maximum at Mo/Rh=
0.063 and W/Rh=0.13, which was smaller than the case of Re/
Rh. From the comparison, it is found that RhÀReOx/SiO2 (Re/
Rh=0.5) showed the highest performance, and the present
study focuses on the investigation of catalytic performance
and catalyst characterization on RhÀReOx/SiO2 (Re/Rh=0.5).
Catalytic performance of RhÀReOx/SiO2 in the 1,2-PrD
hydrogenolysis
The reaction temperature dependence of the 1,2-PrD hydroge-
nolysis over RhÀReOx/SiO2 and Rh/SiO2 is shown in Figure 2.
The selectivity to degradation products increased with increas-
ChemSusChem 2010, 3, 728 – 736
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
729

K. Tomishige et al.
Figure 2. Reaction temperature dependence of the 1,2-propanediol hydro-
genolysis over a) RhÀReOx/SiO2 (Re/Rh=0.5) and b) Rh/SiO2. Reaction condi-
tions: 20 wt% 1,2-propanediol aqueous solution (20 mL), reaction time 4 h
(a) and 24 h (b), initial H2 pressure (8.0 MPa), catalyst (150 mg). Degradation:
ethanol+ethane+methane.
lysts were not added and the acidity of RhÀReOx/SiO2 was
rather weak;[15] the side reactions were not catalyzed at all
judging from no detection of di-n-propyl ether and propylene
glycol propyl ether. This explains the high yield of 1-propanol.
Figure 1. Dependence of the 1,2-propanediol hydrogenolysis over Rh/SiO2
on additive amount of a) Re, b) Mo, and c) W. Reaction conditions: 20 wt%
1,2-propanediol aqueous solution (20 mL), reaction temperature (393 K), re-
action time (6 h), initial H2 pressure (8 MPa), catalyst (150 mg). Degradation:
ethanol+ethane+methane.
To verify the stability of the catalyst, the RhÀReOx/SiO2 (Re/
Rh=0.5) catalyst was used repeatedly, and the results are
shown in Figure 6. The conversion and the selectivity were
almost reproduced over the used catalyst. The slight increase
in the conversion was probably due to the small excess of the
added catalyst for the compensation of catalyst loss in the re-
covery procedure.
overhydrogenolysis of 1-PrOH and 2-PrOH to propane. In the
previous report, the conversion of 1,2-PrD to 1-PrOH can pro-
ceed by the acid-catalyzed dehydration of 1,2-PrD to propio-
naldehyde and the subsequent hydrogenation of propionalde-
hyde to 1-PrOH.[19] In the previous case, HOTf was used as an
acid catalyst. The acid catalyst promoted 1,2-PrD conversion,
however, it also catalyzed ether formation such as di-n-propyl
ether and propylene glycol propyl ether. The side reaction de-
creased the yield of 1-PrOH, therefore, the maximum yield was
as low as 54%. In contrast, in the present case, the acid cata-
As reported previously, RhÀReOx/SiO2 (Re/Rh=0.5) exhibited
high activity of the hydrogenolysis of glycerol.[15] Both the glyc-
erol hydrogenolysis to propanediols and the hydrogenolysis of
the propanediols to propanols can be catalyzed by the RhÀ
ReOx/SiO2 (Re/Rh=0.5) catalyst. Therefore, a one-pot transfor-
mation of glycerol to propanols was attempted. The reaction
730
www.chemsuschem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2010, 3, 728 – 736

Production of Biopropanols from Glycerol
Figure 4. Effect of 1,2-propanediol concentration in the 1,2-propanediol hy-
drogenolysis over a) RhÀReOx/SiO2 (Re/Rh=0.5) and b) Rh/SiO2.Reaction con-
ditions: 1,2-propanediol (4 g), reaction temperature (393 K), reaction time
4 h (a) and 24 h (b), initial H2 pressure (8 MPa), catalyst 50 mg (a) and
150 mg (b). Degradation: ethanol+ethane+methane.
Figure 3. Effect of H2 pressure on the 1,2-propanediol hydrogenolysis over
a) RhÀReOx/SiO2 (Re/Rh=0.5) and b) Rh/SiO2. Reaction conditions: 20 wt%
1,2-propanediol aqueous solution (20 mL), reaction temperature (393 K), re-
action time 6 h (a) and 48 h (b), catalyst (150 mg). Degradation: ethanol+
ethane+methane.
time dependence of the glycerol hydrogenolysis is shown in
Figure 7. After 12 h, the yield of 1,2- and 1,3-PrDs was still
high, and it decreased with increasing reaction time. Maximum
yield of 1-PrOH was 76% at 24 h, and that of 1-PrOH+2-PrOH
was 92% at 24 h. The maximum yield of 1-PrOH and 1-
PrOH+2-PrOH in the glycerol hydrogenolysis was higher than
in the hydrogenolysis of 1,2-PrD. This tendency can be due to
the formation of 1-PrOH via 1,3-PrD. These results indicate that
RhÀReOx/SiO2 (Re/Rh=0.5) is also effective to the one-pot con-
version of glycerol to propanols. It should be noted that the
homogeneous Ru catalyst with HOTf for the conversion of 1,2-
PrD is not active for the deoxygenation of glycerol.[19] One-pot
conversion of glycerol to propanols is a characteristic of the
RhÀReOx/SiO2 catalyst.
Figure 5. Reaction time dependence of the 1,2-propanediol hydrogenolysis
over RhÀReOx/SiO2 (Re/Rh=0.5). * Conversion; & selectivity to 1-propanol;
D selectivity to 2-propanol; * selectivity to propane; ꢁ selectivity to degra-
dation. Reaction conditions: 20 wt% 1,2-propanediol aqueous solution
(20 mL), reaction temperature (393 K), initial H2 pressure (8.0 MPa), catalyst
(150 mg). Degradation: ethanol+ethane+methane.
Catalyst characterization
Figure 8 shows X-ray diffraction (XRD) patterns of RhÀReOx/
SiO2 (Re/Rh=0.5) and Rh/SiO2 after the catalytic use. The peak
around 2q=418 is assigned to Rh metal, and the average size
of Rh metal particles on Rh/SiO2 was estimated to be 3.1Æ
0.3 nm. The pattern of the used RhÀReOx/SiO2 (Re/Rh=0.5)
was almost the same as those of Rh/SiO2 and the fresh catalyst
treated with H2.[15] The peak due to Re species was not detect-
ed at all suggesting that the Re species is highly dispersed.
The transmission electron microscope (TEM) image of RhÀ
ReOx/SiO2 (Re/Rh=0.5) was similar to those before the reac-
tion[21] or after the reaction of glycerol.[15]The average size of
ChemSusChem 2010, 3, 728 – 736
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
731

K. Tomishige et al.
the Rh metal particle was calculated to be 3.1Æ0.2 nm, which
agrees well with the particle size obtained from the XRD pat-
tern.
Figure 9 shows the Re L3-edge extended X-ray absorption
fine structure (EXAFS) spectra of RhÀReOx/SiO2 (Re/Rh=0.5)
Figure 6. Reuse of RhÀReOx/SiO2 (Re/Rh=0.5) in the 1,2-propanediol hydro-
genolysis. Reaction conditions: 20 wt% 1,2-propanediol aqueous solution
(20 mL), reaction temperature (393 K), reaction time (4 h), initial H2 pressure
(8.0 MPa), catalyst (150 mg). Degradation: ethanol+ethane+methane.
Figure 7. Reaction time dependence of the glycerol hydrogenolysis over
RhÀReOx/SiO2 (Re/Rh=0.5). * Conversion; & selectivity to 1-propanol; D
selectivity to 2-propanol; * selectivity to propane; ꢁ selectivity to degrada-
^
tion; selectivity to 1,3-propanediol; selectivity to 1,2-propanediol. Reac-
*
tion conditions: 100 wt% glycerol (4 g), reaction temperature (393 K), initial
H2 pressure (8.0 MPa), catalyst (150 mg). Degradation: ethanol+ethane+me-
thane.
Figure 9. Results of the Re L3-edge EXAFS analysis of RhÀReOx/SiO2. 1) k3-
Weighted EXAFS oscillations. 2) Fourier transform of k3-weighted Re L3-edge
EXAFS, FT range: 30–120 nmÀ1. 3) Fourier filtered EXAFS data (solid line) and
calculated data (dots), Fourier filtering range: 0.129–0.301 nm. a) Re powder,
b) NH4ReO4, c) RhÀReOx/SiO2 (Re/Rh=0.5) after the catalytic use.
Figure 8. XRD patterns of a) RhÀReOx/SiO2 (Re/Rh=0.5) and b) Rh/SiO2 after
the catalytic use.
732
www.chemsuschem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2010, 3, 728 – 736

Production of Biopropanols from Glycerol
after the catalytic use. Table 2 summarizes the curve fitting re-
sults. The curve fitting analysis indicates the presence of the
ReÀO, ReÀRh, and ReÀRe bonds with bond length (R) of 0.210,
Table 2. Curve fitting results of Re L3-edge EXAFS of RhÀReOx/SiO2 (Re/
Rh=0.5) after the catalytic use.
Shells
CN[a]
R [nm][b]
s [nm][c]
DE0 [eV][d] Rf [%][e]
ReÀRh 3.6Æ0.2 0.266Æ0.001 0.0072Æ0.0002 7.3Æ0.8
ReÀRe 2.8Æ0.5 0.272Æ0.001 0.0078Æ0.0011 3.5Æ2.5
0.9
ReÀO
1.3Æ0.4 0.210Æ0.002 0.0089Æ0.0008 9.3Æ3.2
[a] Coordination number. [b] Bond distance. [c] Debye–Waller factor.
[d] Difference in the origin of photoelectron energy between the refer-
ence and the sample. [e] Residual factor. Rh: 4 wt%, Re/Rh=0.5. Fourier
filtering range: 0.129–0.301 nm.
0.266, and 0.272 nm, respectively, and coordination number
(CN) of 1.3, 3.6, and 2.8, respectively. These results are almost
the same as those for the fresh catalyst treated with H2 (ReÀO
with R=0.210 nm and CN=1.4, ReÀRh with R=0.265 nm and
CN=3.7, and ReÀRe with R=0.272 nm and CN=2.7).[15] The re-
sults of Rh K-edge EXAFS analysis are shown in Figure 10 and
Table 3. The curve fitting results were also almost the same as
those for the fresh catalyst treated with H2 (RhÀRh bond with
R=0.267–0.268 nm and CN=10.1, and RhÀRe bond with R=
0.265 nm and CN=1.8).[15] High CN of the RhÀRh bond means
the formation of Rh metal particles and low CN of the ReÀO
bond represents the formation of low valent ReOx species; the
presence of ReÀRe bond means the clustering of the low
valent ReOx species. The reuse experiments, XRD, and EXAFS
data show that the catalytic use does not change the RhÀReOx
structure where Rh metal and low-valent ReOx clusters com-
pose the particle as core and shell, respectively.
Effect of ReOx over Rh metal particles on the reaction
mechanism
The addition of Re to Rh/SiO2 enhanced the activity remarkably
and suppressed the degradation reactions significantly in the
hydrogenolysis of 1,2-PrD. The amount of CO adsorption on
the RhÀReOx/SiO2 catalyst decreased gradually with increasing
Re amount.[15] Because the amount of CO adsorption repre-
sents the number of Rh surface atoms, the additive effect of
ReOx is not the increase of the surface of Rh, but the synergy
between Rh and Re. Judging from the kinetic analysis of 1,2-
PrD hydrogenolysis, in particular, the small positive reaction
order with respect to 1,2-PrD in the low concentration range, it
is expected that the interaction between 1,2-PrD and the cata-
lyst surface is strengthened by the modification with Re. Ac-
cording to the catalyst characterization, the Re species is in the
partially oxidized state. On the other hand, adsorbed methoxy
species is formed by the interaction between methanol and
the ionic Re species.[22] Therefore, the alkoxide of 1,2-PrD is ad-
sorbed on the Re species. In addition, it has been reported
that the interaction of glycerol[15] and tetrahydrofurfuryl alco-
hol[21] with ReOx and MoOx species is rather strong.
Figure 10. Results of the Rh K-edge EXAFS analysis of RhÀReOx/SiO2 (Re/
Rh=0.5). 1) k3-Weighted EXAFS oscillations. 2) Fourier transform of k3-
weighted Rh K-edge EXAFS, FT range: 30–150 nmÀ1. 3) Fourier filtered EXAFS
data (solid line) and calculated data (dots), Fourier filtering range: 0.153–
0.304 nm. a) Rh foil, b) RhÀReOx/SiO2 (Re/Rh=0.5) after the catalytic use.
First effect of ReOx on Rh/SiO2 can strengthen the interaction
between the surface of the catalysts and the alcohols, and the
coverage of the alkoxide on RhÀReOx/SiO2 became higher than
that on Rh/SiO2. High coverage of the alkoxide as a reaction in-
termediate is connected to high activity of 1,2-PrD hydrogenol-
ChemSusChem 2010, 3, 728 – 736
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
733

K. Tomishige et al.
carbocation species, the route to 1-propanol is more favorable
than that to 2-propanol, which can explain the selectivity in
the hydrogenolysis of 1,2-PrD. In addition, the Rh metal en-
semble on the surface of the Rh metal particles becomes
smaller or isolated by the modification with the ReOx species.
The decrease of the ensemble size of Rh atoms can contribute
to the suppression of the degradation reactions. The study to
clarify the nature of the intermediates is currently in progress.
Table 3. Curve fitting results of Rh K-edge EXAFS of RhÀReOx/SiO2 (Re/
Rh=0.5) after the catalytic use.
Shells
CN[a]
R [nm][b]
s [nm][c]
DE0 [eV][d] Rf [%][e]
RhÀRh 10.1Æ0.3 0.268Æ0.001 0.0078Æ0.0001
0.6Æ0.6
0.5
RhÀRe
1.8Æ0.3 0.265Æ0.001 0.0077Æ0.0003 À9.9Æ2.3
[a] Coordination number. [b] Bond distance. [c] Debye–Waller factor.
[d] Difference in the origin of photoelectron energy between the refer-
ence and the sample. [e] Residual factor. Rh: 4 wt%, Re/Rh=0.5. Fourier
filtering range: 0.129–0.301 nm.
Conclusions
Conventional hydrogenolysis catalysts are not applicable to
the hydrogenolysis of 1,2-propanediol to propanols because of
low activity and selectivity.
ysis. On the basis of the EXAFS analysis that the ReOx clusters
are attached on the surface of Rh metal particles, the alkoxide
of 1,2-PrD adsorbed on ReOx species is also located near the
Rh metal surface, which has the ability to supply hydrogen
species for the hydrogenolysis reaction. The synergy between
high coverage of the alkoxide on ReOx and the hydrogen spe-
cies supplied from the neighboring Rh surface can enhance
the activity of hydrogenolysis of 1,2-PrD. On the other hand, it
is thought that the alkoxide of propanols has very low reactivi-
ty from the results of the activity test in the reaction of 1-prop-
anol and 2-propanol. When the reaction conditions are 150 mg
of RhÀReOx/SiO2 (Re/Rh=0.5), 20 mL of 20% aqueous solution
of 1-propanol or 2-propanol, 393 K, 12 h, 3.8% and 2.1% con-
versions were obtained, respectively. Under the same condi-
tions, 57% conversion was obtained in the hydrogenolysis of
1,2-PrD.
The Rh/SiO2 catalysts modified with Re, Mo, and W were
very effective to the hydrogenolysis of 1,2-propanediol, and
the Rh/SiO2 modified with Re showed the highest perfor-
mance. The promoting effect of Re addition to Rh/SiO2 was
maximum at Re/Rh=0.5, and the excess addition of Re de-
creased the catalytic performance. As a result, the hydrogenol-
ysis of 1,2-propanediol was promoted by the synergy between
Rh and Re.
It is characteristic that RhÀReOx/SiO2 (Re/Rh=0.5) main-
tained high selectivity to propanols in the hydrogenolysis of
1,2-propanediol and suppressed the degradation reactions
even under low H2 pressure and high reaction temperature,
whereas Rh/SiO2 showed high selectivity to the degradation
products.
Characterization results of the used catalyst indicated that
catalytic use does not change the structure where ReOx clus-
ters were attached on the surface of Rh metal surface.
The interaction between 1,2-propanediol and the surface of
RhÀReOx/SiO2, in particular, the formation of the alkoxide of
1,2-propanediol on the ReOx species is suggested. The strong
interaction can enhance the coverage of 1,2-propanediol,
which can cause high activity of the hydrogenolysis.
The synergistic effect of Rh and Re in the 1,2-propanediol
hydrogenolysis can be explained by the hydrogenolysis of the
alkoxide species on Re with hydrogen species on the Rh metal
surface. The stability of the carbocation as a reaction inter-
mediate can explain the higher selectivity to 1-propanol than
2-propanol.
Higher reactivity of 1,2-PrD than that of propanols suggests
that the hydrogen species attacks the residual OH group in
1,2-PrD, not the oxygen atom in the alkoxide species. Based on
the above discussions, we propose the reaction mechanism il-
lustrated in Scheme 1. In the hydrogenolysis route to 1-propa-
nol from 1,2-PrD, a reaction intermediate is a secondary carbo-
cation species. Judging from the stability of the intermediate
Experimental Section
Catalyst preparation
The SiO2 (G-6, BET surface area 535 m2 g) supplied by Fuji Silysia
Chemical Ltd was used as a support of the catalysts. A Rh/SiO2 cat-
alyst was prepared by impregnating SiO2 with an aqueous solution
of RhCl3·3H2O (Soekawa Chemical Co., Ltd). After evaporating the
solvent and drying at 383 K for 12 h, they were calcined in air at
773 K for 3 h. The Rh/SiO2 catalyst modified with Re, Mo, and W
was prepared by impregnating Rh/SiO2 after a drying procedure
with an aqueous solution of NH4ReO4 (Soekawa Chemical Co., Ltd),
(NH4)6Mo7O24·4H2O (Wako Pure Chemical Industries, Ltd), and
(NH4)10W12O41·5H2O (Wako Pure Chemical Industries, Ltd). The dried
catalysts were calcined in air at 773 K for 3 h after drying at 383 K
Scheme 1. 1,2-propanediol hydrogenolysis to 1-propanol and 2-propanol on
the interface between ReOx and Rh metal surface over RhÀReOx/SiO2.
734
www.chemsuschem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2010, 3, 728 – 736

Production of Biopropanols from Glycerol
for 12 h. The loading amount of Rh was 4 wt%, and that of the ad-
ditive was represented by the molar ratio of the additive to Rh. All
the catalysts were used in powdery form with granule size of
<100 mesh. Commercial catalysts such as Rh/C (557 m2 g), Ru/C
(485 m2 g), Raney Ni (26 m2 g), Raney Co (15 m2 g), Raney Cu
(14 m2 g), and Cu chromite (16 m2 g) were used as reference cata-
lysts. Rh/C and Ru/C were purchased from Wako Pure Chemical In-
dustries, Ltd, and the loading of noble metals on both catalysts
was 5 wt%. Raney catalysts (Raney Ni, Raney Co, Raney Cu) were
supplied from Sigma–Aldrich Co., Ltd. The Cu-chromite was pur-
chased from N. E. Chemcat, and the content of Cu and Cr in the
catalyst was 67% and 12%, respectively.
quently the adsorption was performed at room temperature. Gas
pressure at adsorption equilibrium was about 1.1 kPa. The sample
weight was about 0.1 g. The dead volume of the apparatus was
about 60 cm3. The adsorption amount of CO is represented as the
molar ratio to Rh and this corresponds to the number of the sur-
face atoms assuming that the stoichiometry of adsorbed CO to sur-
face Rh atom is one. XRD patterns were recorded on a diffractome-
ter (Philips X’pert). Average metal particle size was estimated using
the Scherrer equation.[23] TEM images were taken for determination
of the particle size (JEOL, JEM 2010). The samples were dispersed
in ethanol with supersonic waves and placed on Cu grids under air
atmosphere. Average particle size was calculated by ꢀnidi3/ꢀnidi2
(di: average particle size, ni: number of particle with di).[24]
The EXAFS spectra were measured at the BL01B1 station at SPring-
8 with the approval of the Japan Synchrotron Radiation Research
Institute (JASRI; Proposal No. 2008B1235). The storage ring was op-
erated at 8 GeV, and a Si (111) single crystal was used to obtain a
monochromatic X-ray beam. Two ion chambers for I0 and I were
filled with 85% N2+15% Ar and 50% N2+50% Ar, respectively, for
Re L3-edge measurement. For Rh K-edge measurement, both ion
chambers were filled with Ar. We prepared the sample after the
catalytic use as follows. The catalytic reaction was carried out in an
autoclave. The standard reaction conditions as shown in the reac-
tion of activity tests were applied. After cooling, the wet catalyst
powder was transferred to the measurement cell in a glove bag
filled with nitrogen. The thickness of the cell filled with the
powder was 2 mm to give an edge jump of 0.3 and 0.3 for the Rh
K-edge and Re L3-edge measurement, respectively. The EXAFS data
were collected in transmission mode. For EXAFS analysis, the oscil-
lation was first extracted from the EXAFS data using a spline
smoothing method.[25] Fourier transformation of the k3-weighted
EXAFS oscillation from the k space to the r space was performed
to obtain a radial distribution function. The inversely Fourier fil-
tered data were analyzed using a usual curve fitting method.[26] For
the curve fitting analysis, the empirical phase shift and amplitude
functions for the ReÀO, ReÀRe, and RhÀRh bonds were extracted
from data for NH4ReO4, Re powder, and Rh foil, respectively. Theo-
retical functions for the ReÀRh and RhÀRe bonds were calculated
using the FEFF8.2 program.[27] Analyses of EXAFS data were per-
formed using a computer program (REX2000, ver. 2.5.9; Rigaku
Corp.). Error bars for each parameter were estimated by stepping
each parameter, while optimizing the others parameter, until the
residual factor became two times its minimum value.[28]
Activity tests
Activity tests were performed in a 190 mL stainless steel autoclave
with an inserted glass vessel. An aqueous solution of 1,2-propane-
diol (Wako Pure Chemical Industries, Ltd,>99%) was put into the
autoclaves together with a spinner and an appropriate amount of
catalysts. After sealing the reactors, their air content was purged
by flushing thrice with 1 MPa hydrogen (99.99%; Takachiho Trad-
ing Co. Ltd.). Autoclaves were then heated to 393 K and pressur-
ized to 1 MPa for the reduction pretreatment of the catalysts. The
temperature was monitored using a thermocouple inserted in the
autoclave. After 1 h, the H2 pressure was increased to 8 MPa at the
reaction temperature. During the experiment, the stirring rate was
fixed at 500 rpm (magnetic stirring). After an appropriate reaction
time, the reactors were cooled down and the gases were collected
in a gas bag. The autoclave contents were transferred to vials, and
the catalysts were separated by centrifugation and filtration. The
standard conditions for the reaction were as follows: 393 K reac-
tion temperature, 8.0 MPa initial hydrogen pressure, 24 h reaction
time, 20 wt% 1,2-propanediol aqueous solution, and 150 mg sup-
ported metal catalyst. The parameters were changed appropriately
in order to investigate the effect of the reaction conditions. Details
of the reaction conditions are described in each result. In the activ-
ity test of Raney Ni, Raney Co, Raney Cu, and Cu-chromite, the re-
duction pretreatment temperature and the reaction temperature
were increased to 453 K. The products were analyzed using a gas
chromatograph (Shimadzu GC-2014) equipped with a FID. A TC–
WAX capillary column (diameter 0.25 mmf, 30 m) was used for the
separation. Products were also identified using GC–MS (QP5050,
Shimadzu). The products in the 1,2-propanediol (1,2-PrD) hydroge-
nolysis were 1-propanol (1-PrOH) and 2-propanol (2-PrOH). The
overhydrogenolysis reaction of propanols gave propane. In addi-
tion, the degradation products such as ethanol, ethane, and meth-
ane were detected. The conversion and the selectivity were de-
fined on the C-based calculation in a similar way as reported previ-
ously.[13] The mass balance was also confirmed in each result and
the difference in mass balance was always in the range of the ex-
perimental error. The used catalyst was collected by centrifugation,
washed with excess water, and dried in air. A slight loss (<10% in
weight) was observed during the recovery process and was com-
pensated with fresh catalyst in each reuse experiment. In addition,
the one-pot reaction of glycerol hydrogenolysis to propanols was
also tested. The reaction and analysis conditions were similar to
those in the case of the hydrogenolysis of 1,2-propanediol.
Acknowledgements
This work was in part supported by World Premier International
Research Center (WPI) Initiative on Materials Nanoarchitectonics,
MEXT, and the Industrial Technology Research Grant Program
(No. 08B40001c) of New Energy and Industrial Technology Devel-
opment Organization (NEDO) of Japan. Authors appreciate Dr.
Tokushi Kizuka for the TEM images of the catalyst.
Keywords: alcohols · biomass · heterogeneous catalysis ·
rhenium · rhodium
[1] a) A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107, 2411–2502; b) P.
Gallezot, Green Chem. 2007, 9, 295–302; c) M. Schlaf, Dalton Trans.
2006, 4645–4653; d) R. M. West, E. L. Kunkes, D. A. Simonetti, J. A. Du-
mesic, Catal. Today 2009, 147, 115–125.
[2] C. H. Zhou, J. N. Beltramini, Y. X. Fan, G. Q. Lu, Chem. Soc. Rev. 2008, 37,
527–549.
Catalyst characterization
The amount of CO chemisorption was measured in a high-vacuum
system using a volumetric method. Before adsorption measure-
ments, the catalysts were treated in H2 at 393 K for 1 h. Subse-
ChemSusChem 2010, 3, 728 – 736
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
735

K. Tomishige et al.
[3] A. E. Martin, F. H. Murphy, Kirk-Othmer Encyclopedia of Chemical Technol-
ogy, John Wiley Sons, NY, 2000, DOI: 10.1002/
0471238961.1618151613011820.a01.
[4] C. Liang, Z. Ma, L. Ding, J. Qiu, Catal. Lett. 2009, 130, 169–176.
[5] M. A. Dasari, P. Kiatsimkul, W. R. Sutterlin, G. J. Suppes, Appl. Catal. A
2005, 281, 225–231.
[17] J. D. Unruh, D. Pearson, Kirk-Othmer Encyclopedia of Chemical Technolo-
&
gy,
0471238961.1618151621141821.a01.
[18] J. E. Logsdon, R. A. Loke, Kirk-Othmer Encyclopedia of Chemical Technolo-
gy, John Wiley Sons, NY, 2000, DOI: 10.1002/
0471238961.0919151612150719.a01.
John
Wiley
&
Sons,
NY,
2000,
DOI:
10.1002/
&
[6] L. Huang, Y. Zhu, H. Zheng, Y. Li, Z. Zeng, J. Chem. Technol. Biotechnol.
2008, 83, 1670–1675.
[19] M. Schlaf, P. Ghosh, P. J. Fagan, E. Hauptman, R. M. Bullock, Adv. Synth.
Catal. 2009, 351, 789–800.
[7] M. Balaraju, V. Rekha, P. S. S. Prasad, R. B. N. Prasad, N. Lingaiah, Catal.
Lett. 2008, 126, 119–124.
[20] J. Chaminand, L. Djakovitch, P. Gallezot, P. Marion, C. Pinel, C. Rosier,
Green Chem. 2004, 6, 359–361.
[8] L. C. Meher, R. Gopinath, S. N. Naik, A. K. Dalai, Ind. Eng. Chem. Res.
2009, 48, 1840–1846.
[9] S. Wang, H. Liu, Catal. Lett. 2007, 117, 62–67.
[10] A. Perosa, P. Tundo, Ind. Eng. Chem. Res. 2005, 44, 8535–8537.
[11] T. Jiang, Y. Zhou, S. Liang, H. Liu, B. Han, Green Chem. 2009, 11, 1000–
1006.
[21] a) S. Koso, I. Furikado, A. Shimao, T. Miyazawa, K. Kunimori, K. Tomishige,
Chem. Commun. 2009, 2035–2037; b) S. Koso, N. Ueda, Y. Shinmi, K.
Okumura, T. Kizuka, K. Tomishige, J. Catal. 2009, 267, 89–92.
[22] J. Liu, E. Zhan, W. Cai, J. Li, W. Shen, Catal. Lett. 2008, 120, 274–280.
[23] S. R. Sashital, J. B. Cohen, R. L. Burwell Jr. , J. B. Butt, J. Catal. 1977, 50,
479–493.
[12] J. Wang, S. Shen, B. Li, H. Lin, Y. Yuan, Chem. Lett. 2009, 38, 572–573.
[13] a) Y. Kusunoki, T. Miyazawa, K. Kunimori, K. Tomishige, Catal. Commun.
2005, 6, 645–649; b) T. Miyazawa, Y. Kusunoki, K. Kunimori, K. Tomish-
ige, J. Catal. 2006, 240, 213–221; c) T. Miyazawa, S. Koso, K. Kunimori, K.
Tomishige, Appl. Catal. A 2007, 318, 244–251; d) T. Miyazawa, S. Koso, K.
Kunimori, K. Tomishige, Appl. Catal. A 2007, 329, 30–35.
[14] I. Furikado, T. Miyazawa, S. Koso, A. Shimao, K. Kunimori, K. Tomishige,
Green Chem. 2007, 9, 582–588.
[24] Y. G. Chen, K. Tomishige, K. Yokoyama, K. Fujimoto, J. Catal. 1999, 184,
479–490.
[25] J. W. Cook, D. E. Sayers, J. Appl. Phys. 1981, 52, 5024–5031.
[26] a) K. Okumura, J. Amano, N. Yasunobu, M. Niwa, J. Phys. Chem. B 2000,
104, 1050–1057; b) K. Okumura, S. Matsumoto, N. Nishiaki, M. Niwa,
Appl. Catal. B 2003, 40, 151–159.
[27] A. L. Ankudinov, B. Ravel, J. J. Rehr, S. D. Conradson, Phys. Rev. B 1998,
58, 7565–7576.
[15] a) A. Shimao, S. Koso, N. Ueda, Y. Shinmi, I. Furikado, K. Tomishige,
Chem. Lett. 2009, 38, 540–541; b) Y. Shinmi, S. Koso, T. Kubota, Y. Naka-
gawa, K. Tomishige, Appl. Catal. B 2010, 94, 318–326.
[16] Z. Yuan, P. Wu, J. Gao, X. Lu, Z. Hou, X. Zheng, Catal. Lett. 2009, 130,
261–265.
[28] K. Tomishige, K. Asakura, Y. Iwasawa, J. Catal. 1994, 149, 70–80.
Received: February 15, 2010
Published online on May 6, 2010
736
www.chemsuschem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2010, 3, 728 – 736