Enhancement of reaction rates for catalytic benzaldehyde hydrogenation and sorbitol dehydration in water solvent by addition of carbon dioxide

Article abstract of DOI:10.1007/s12039-014-0582-3

The effect of pressured carbon dioxide on heterogeneous hydrogenation of benzaldehyde and homogeneous dehydration of sorbitol in water solvent was studied. Initial hydrogenation rates of benzaldehyde over a charcoal-supported palladium catalyst in water at 313 K were enhanced by the addition of carbon dioxide. The initial rate increased with an increase in carbon dioxide pressure and became a maximum at 5 MPa. Dehydration of sorbitol proceeded in water phase at 500 K and initial dehydration rates were enhanced by addition of 30 MPa of carbon dioxide.

Full text of DOI:10.1007/s12039-014-0582-3

J. Chem. Sci. Vol. 126, No. 2, March 2014, pp. 395–401. ꢀc Indian Academy of Sciences.
Enhancement of reaction rates for catalytic benzaldehyde hydrogenation
and sorbitol dehydration in water solvent by addition of carbon dioxide
a,b,∗
a
a
MASAYUKI SHIRAI
, OSAMU SATO , NORIHITO HIYOSHI and
a
ARITOMO YAMAGUCHI
Research Center for Compact Chemical System, National Institute of Advanced Industrial Science and
Technology (AIST), 4-2-1 Nigatake, Miyagino, Sendai, 983-8551, Japan
a
b
Department of Chemistry and Bioengineering, Faculty of Engineering, IWATE University, 4-3-5 Ueda,
Morioka, Iwate, 020-8551, Japan
e-mail: mshirai@iwate-u.ac.jp
MS received 8 July 2013; revised 24 December 2013; accepted 26 December 2013
Abstract. The effect of pressured carbon dioxide on heterogeneous hydrogenation of benzaldehyde and
homogeneous dehydration of sorbitol in water solvent was studied. Initial hydrogenation rates of benzaldehyde
over a charcoal-supported palladium catalyst in water at 313 K were enhanced by the addition of carbon diox-
ide. The initial rate increased with an increase in carbon dioxide pressure and became a maximum at 5 MPa.
Dehydration of sorbitol proceeded in water phase at 500 K and initial dehydration rates were enhanced by
addition of 30 MPa of carbon dioxide.
Keywords. High-pressure carbon dioxide; high-temperature liquid water; carbonic acid.
1
. Introduction
by removal of carbon dioxide with depressurization
procedure.
We have succeeded to producing valuable chemicals
with efficient chemical engineering in the twentieth
century. However, we have to develop more efficient
techniques for production of sustainable chemicals.
Also, the technique should be more environmentally -
In this study, we report the application of high-
pressured carbon dioxide to the heterogeneous hydro-
genation of benzaldehyde over a charcoal-supported
ruthenium catalyst in water solvent and to the homoge-
neous dehydration of sorbitol in aqueous phase.
1
benign technique. In fine chemicals industries, organic
+
−
+
2−
CO +H O ꢀ H CO ꢀ H +HCO ꢀ 2H +CO
solvents and inorganic acids are used in the enhance-
ment of chemical reaction rates. Generally, organic
solvents are hazardous and toxic, and much energy
is required to remove organic solvents by distillation
2
2
2
3
3
3
(
1)
processes. Also, cost-consuming troublesome proce- 2. Experimental
dures are needed for neutralization of acid, and salts
as by-products should be wasted. Application of high- 2.1 Hydrogenation of benzaldehyde
pressure carbon dioxide and water including supercriti-
A charcoal-supported palladium catalyst (palladium
loading: 5 wt%) was purchased (Wako Pure Chemicals
Industries Ltd.) and used as received. Benzaldehyde
was also purchased (Wako Pure Chemicals Industries
Ltd.) and used without purification.
cal conditions are studied as an alternative to organic
2
,3
solvent and inorganic acid.
Savage’s group pro-
posed that a system with high-temperature liquid water
under high-pressure carbon dioxide was effective for
4
acid-catalysed reactions. Protons from carbonic acid
Hydrogenation of benzaldehyde was carried out with
molecules, which are formed by the reactions of water
and carbon dioxide, can play a role as acid catalysts
in the system (eq. 1). Carbonic acid is weak and its
application in acid reactions is limited; however, neu-
tralization processes after the reaction can be eliminated
5
a high-pressure reactor made of stainless steel. A dia-
gram of the reaction system is depicted in figure 1.
6
Hydrogenation procedure is as follows. After loading
of weighted amounts of catalyst, benzaldehyde, mag-
netic stirrer and distilled water into the reactor (inner
3
volume: 50 cm ), the inside of the reactor was purged
∗
For correspondence
with argon to remove air and the reactor was heated
395

396
Masayuki Shirai et al.
with an oil bath. After the reactor was heated to the iced water bath for cooling quickly. After depressuriza-
desired temperature, 1 MPa of hydrogen was introduced tion, the slurry in the reactor was filtrated and products
and then quickly carbon dioxide was added through a were recovered with acetone.
liquid pump. Carbon dioxide pressure was estimated by
Quantitative analysis of products and unreacted ben-
subtracting hydrogen pressure from the total pressure. zaldehyde was conducted by a gas chromatograph
We defined the start of the reaction as the time when the equipped with a flame ionization detector (Agilent GC
introduction of carbon dioxide was finished. After the System 6890N) and a DB-WAX capillary column. Con-
desired reaction time, the reactor was submerged into an version and yield are defined as follows:
Conversion(%)=(1−(the amount of benzaldehyde unreacted (mol ))/(initial amount of benzaldehyde (mol )))×100.
Yield (%) = (the amount of each product obtained (mol ))/(initial amount of benzaldehyde (mol )) × 100.
2.2 Dehydration of sorbitol
into a sand bath at 500 K for a given reaction time,
and submerged into a water bath for cooling quickly
Sorbitol, which was purchased from Wako Pure Chemi- to ambient temperature after the reaction. We estimated
cals Industries Ltd., was used without purification.
that the inside pressure at 500 K was 46 MPa based
Dehydration was carried out with a tube reactor made on Charles’s law. After depressurization by opening the
7
of a SUS 316. A diagram of the reactor is depicted in valve, the mixture of the reactant and liquid products
8
figure 2. Dehydration procedure is as follows. After was taken out from the reactor with distilled water.
3
loading of 3 cm of aqueous sorbitol solution
Quantitative analysis of liquid products and unre-
−3
(
1.0 mol·dm ) into the tube reactor (inter volume: acted sorbitol was conducted by a high performance
3
6
cm ), the inside of the reactor was purged with argon liquid chromatograph (GILSON, Inc.) with an evapo-
to remove air. Dehydration under pressured carbon rative light scattering detector (model 300S ELSD,
dioxide was operated as follows. After purging the SofTA Corporation) equipped with a Carbomix Ca NP5
inside with argon, the reactor was heated with a water column (Sepac Technologies Inc.). The products were
bath (323 K). After purging with carbon dioxide, identified by comparing the retention times with those
30 MPa of carbon dioxide was introduced into the reac- of the standard materials. Conversion and yield are
tor through a liquid pump. The reactor was submerged defined as follows;
Conversion (%) = (1 − (the amount of sorbitol unreacted (mol)) / (initial amount of sorbitol (mol))) × 100.
Yield (%) = (the amount of each product obtained (mol)) / (initial amount of sorbitol (mol)) × 100.
mixture of benzaldehyde, water, and Pd/C at room tem-
3
. Results and discussion
perature showed that two liquid phases of benzaldehyde
and water were formed and Pd/C existed in benzalde-
hyde phase. Condensation of benzaldehyde to Pd/C and
3.1 Hydrogenation of benzaldehyde
Table 1 shows conversion and selectivity values of high hydrogen solubility to organic phase may cause the
benzaldehyde hydrogenation over Pd/C for 30 min enhancement of the hydrogenation yield by addition of
at 313 K. GC analysis showed that products were water.
only benzyl alcohol and toluene. Hydrogenation of
We also studied the effect of carbon dioxide on the
benzaldehyde to benzyl alcohol and the following hydrogenation over Pd/C in water. Addition of 3 MPa
dehydroxylation of benzyl alcohol to toluene proceeded carbon dioxide showed 53.4% conversion (table 1, entry
under the reaction conditions (scheme 1). Conversion 5), indicating that carbon dioxide enhanced hydro-
under the neat (non-solvent) condition was 12.7% over genation rate. Solubility data of carbon dioxide in
Pd/C for 30 min (table 1, entry 1). Conversion was water showed that pH of water decreased with an
3
changed by the addition of 5 cm of solvent to the increase in carbon dioxide pressure and became a con-
hydrogenation system. The conversion order was water stant value of 3 under more than 3 MPa of carbon
6
,9
>
ethanol > neat > n-heptane. Visual observation of the dioxide. Enhancement of benzyl alcohol yield by

CO2 pressure effect on hydrogenation and dehydration
h)
397
(
T
(d)
(
g)
P (
i)
(e)
(
f)
(
c)
(
a)
(b)
Figure 1. Apparatus for hydrogenation. (a) Hydrogen cylinder, (b) carbon dioxide
cylinder, (c) liquid pump, (d) oven, (e) reactor, (f) catalyst, (g) back pressure regulator
and (h) thermo couple.
addition of water and carbon dioxide would suggest that Benzaldehyde conversion in the presence of 5 cm3
hydrogenation was enhanced by protons derived from of aqueous acetic acid solution (pH = 3) was 54.7%
carbonic acid. We also tested the effect of adding acetic after 30 min, which was close to that in the pres-
3
3
acid (0.05 cm ) on the hydrogenation of benzaldehyde ence of 5 cm of water and 3 MPa of carbon
over the Pd/C catalyst at 313 K (table 1, entry 6). dioxide (pH = 3) (53.4%), indicating that protons pro-
moted hydrogenation. We have reported that addition
of water and carbon dioxide enhanced initial reaction
rates for acetophenone hydrogenation over a charcoal-
supported palladium catalyst and discussed the proba-
bility that protons on palladium metal sites interacted
with oxygen atoms on ketone groups in acetophenone
molecules resulted in enhancement of initial reaction
rates. It is probable that enhancement of the rate of
benzaldehyde hydrogenation by adding carbon dioxide
or acetic acid would result from interactions between
oxygen atoms of aldehyde groups in benzaldehyde
molecules and protons on the palladium surfaces.
We also studied the carbon dioxide pressure effect
on the hydrogenation over Pd/C. Yields for 30 min at
3
13 K are shown in figure 3. Yields of benzyl alcohol
for 30 min increased from 46.4% (0.1 MPa of argon
atmosphere) to 65.6% (under 5 MPa of carbon diox-
ide) with an increase in carbon dioxide pressure up
to 5 MPa and decreased with an increase in carbon
dioxide pressure above 5 MPa. The main product was
benzyl alcohol; however, minor by-product toluene,
which could be obtained by dehydroxylation of benzyl
alcohol, was formed. The yield of by-product toluene
was less than 1% under less than 5 MPa of carbon
dioxide; however, by-product yield increased with
increasing carbon dioxide pressure and it became 1.8%
Figure 2. Apparatus for dehydration in high-temperature
water.

398
Masayuki Shirai et al.
Table 1. Hydrogenation of benzaldehyde. a
Entry
Solvent b
Conversion(%)
Selectivity (%)
Benzyl alcohol
Toluene
1
2
3
4
5
6
Neat (non-solvent)
n-C7H16
C2H5OH
12.7
6.3
25.4
46.4
53.4
54.7
99.2
97.7
99.2
98.7
98.7
98.4
0.8
2.3
0.8
1.3
1.3
1.6
H2O
c
H2O + CO2
H2O + CH3COOHd
a
Hydrogen pressure 1 MPa, reaction temperature 313 K, reaction time 30 min, Pd/C 0.003 g, initial benzaldehyde 0.63 g
b
c
d
3
Solvent 5 cm
3
Water 5 cm and carbon dioxide pressure 3 MPa
3
3
Water 5 cm and acetic acid 0.05 cm
under 15 MPa of carbon dioxide. This result indicates yield was only 2.2% at 60 min. The benzyl alco-
that higher carbon dioxide pressure is unfavourable. It hol yield; however, decreased at higher conversion of
could be explained by the fact that conversion of ben- benzaldehyde. Benzyl alcohol and toluene yields were
zaldehyde increased with an increase in carbon dioxide 86.2 and 13.7%, respectively, at 99% benzaldehyde
pressure up to 3 MPa because of the increase in proton conversion for 90 min. To obtain high yield of benzyl
amounts; however, conversion increased further under alcohol, prolonged reaction time is undesirable.
more than 3 MPa and then, decreased under more than
5
MPa of carbon dioxide. We confirmed that two liquid
3
.2 Dehydration of sorbitol
phases were formed in the mixture of benzaldehyde and
water in air at room temperature by direct visual obser-
vation; however, the phases could be changed under
high pressure carbon dioxide. It is probable that ben-
zaldehyde and water could be soluble in supercriti-
cal carbon dioxide phase (critical point 304.3 K and
A reaction profile for the treatment of aqueous d-
−3
sorbitol solution (sorbitol concentration: 1.0 mol·dm )
at 500 K under argon atmosphere is shown in
figure 5(a). An yield of 1.8% of 1,4-d-anhydrosorbitol
(1,4-AHSO) and 0.1% of 2,5-d-anhydrosorbitol (2,5-
7.38 MPa). Further investigation by direct phase obser-
AHSO) was obtained at 1 h. Yield of 1,4-AHSO
vation under high-pressure condition will be done for
detailed understanding.
The reaction profile of benzaldehyde hydrogenation
over Pd/C under 5 MPa of carbon dioxide, which
showed the maximum yield as a function of reac-
tion time, was studied with the batch reaction system
1
00
8
0
0
0
(
figure 4). At the initial stage, in fact, when we immedi-
6
4
ately cooled the reactor to stop the hydrogenation reac-
tion right after introducing 5 MPa of carbon dioxide,
13.3% of benzyl alcohol yield was obtained. Benzalde-
hyde conversion and benzyl alcohol yield increased
with an increase in time and became 93.0% and
×10
20
0
91.0%, respectively, at 60 min. The by-product toluene
0
2
4
6
8
10
12
14
16
Carbon dioxide pressure (MPa)
CHO
CH OH
CH3
2
Figure 3. Carbon dioxide pressure effect on catalytic
hydrogenation activity of benzaldehyde over charcoal-
supported palladium catalyst in water (◦: benzyl alcohol and
+
H2
+
H2
-
H O
2
ꢁ
: toluene). Amount of toluene is multiplied by 10. Water
3
Benzaldehyde
Benzyl alcohol
Toluene
solvent 5 cm , catalyst 0.003 g, reaction time 30 min, reac-
tion temperature 313 K, hydrogen pressure 3 MPa and initial
benzaldehyde 6.0 mmol.
Scheme 1. Hydrogenation of benzaldehyde.

CO2 pressure effect on hydrogenation and dehydration
399
100
figure 5(a) shows that dehydration of sorbitol proceeded
in liquid water at lower temperature of 500 K.
A reaction profile of aqueous d-sorbitol solution
8
0
0
−3
(
sorbitol concentration: 1.0 mol·dm ) at 500 K in the
presence of carbon dioxide, which was introduced as
0 MPa at 323 K in the reactor and reached to 46 MPa
8
3
at 500 K, is shown in figure 5(b). 1.4-AHSO (4.1%
yield) and 2,5-AHSO (0.6%) were obtained at 1 h,
respectively. Both yields increased with increase in
reaction time and 21.7% of 1,4-AHSO and 2.9% of
4
2
0
0
2
,5-AHSO were obtained for 7 h treatment. Isosorbide
was also obtained from 1.5 h and the yield became
.2% at 7 h. Figure 5 shows that addition of 30 MPa
0
7
0
20
40
60
80
100
of carbon dioxide enhanced dehydration reaction of
sorbitol at 500 K, in which carbon dioxide partial
pressure was 46 MPa at the reaction condition. We
Time (min)
Figure 4. Catalytic hydrogenation of benzaldehyde over
charcoal-supported palladium catalyst in water (ꢂ: benzalde- also studied the effect of introduction of 10 MPa of
3
hyde, ◦: benzyl alcohol and ꢁ: toluene). Water solvent 5 cm , carbon dioxide separately, and no effect was observed.
catalyst 0.003 g, reaction temperature 313 K, hydrogen pres-
We have reported that introduction of 10 MPa of car-
sure 3 MPa, carbon dioxide pressure 5 MPa and initial
bon dioxide at 323 K was not effective for sorbitol
benzaldehyde 6.0 mmol.
10
dehydration at 523 K. It was revealed that much
higher pressure of 30 MPa is need for enhancement
of rate of sorbitol dehydration. We have studied
increased with increase in time and 15.9% yield was intramolecular dehydration behaviour of several types
obtained after 7 h. In addition, 2.2% of isosorbide, of diols and triols in aqueous phase between 473 K
which could be obtained by intramolecular dehydra- and 573 K and showed that intramolecular dehydration
tion of 1,4-AHSO, was obtained for the treatment of proceeded via S 2 type mechanism, in which proton
N
aqueous sorbitol solution for 7 h. Yield of 2,5-AHSO adsorbed on reactant polyols followed by cyclization by
11
did not increase so high and was only 0.9% after 7 h. a backside attack of the other hydroxyl group. We also
Three kinds of anhydrosorbitol could be formed by showed that intramolecular dehydration of diols and
intramolecular dehydration from sorbitol (scheme 2); triols to cyclic ethers in water at 523 K were enhanced
however, 1,5-d-anhydrosorbitol was not obtained. We by addition of 10 MPa of carbon dioxide at 323 K and
have reported that sorbitol dehydration proceeded in that order of dehydration rates depended on the degree
10
water in the absence of inorganic acid at 523 K;
of hydroxyl groups (tertiary > secondary > primary
40
30
20
10
4
3
2
1
0
0
0
0
0
0
0
2
4
6
8
10
0
2
4
6
8
10
(a)
Time (h)
(b)
Time (h)
Figure 5. Dehydration of sorbitol in water under (a) argon (0.1 MPa, introduced at
room temperature) and (b) carbon dioxide (30 MPa, introduced at 323 K) at 500 K.
(
◦: 1,4-anhydrosorbitol, ꢁ: 2,5-anhydrosorbitol, and ꢂ: isosorbide). Initial sorbitol
−3
concentration was 1.0 mol·dm
.

400
Masayuki Shirai et al.
HO
H
H
OH
OH
OH
O
O
HO
OH
HO
- H
2
O
- H O
2
H
O
OH
OH
-
H O
HO
OH
OH
2
d-sorbitol
1,4-anhydro-d-sorbitol (1,4-AHSO)
isosorbide
O
HO
OH
HO
OH
-
H O
2
2
,5-anhydro-d-sorbitol (2,5-AHSO)
HO
O
HO
HO
,5-anhydro-d-sorbitol (1,5-AHSO)
OH
1
Scheme 2. Dehydration of sorbitol.
alcohols), by the study of dehydration of 1,4- 4. Conclusion
butanediol, 2,5-hexanediol, and 2,5-dimethyl-2,5-
12,13
We studied the effect of pressured carbon dioxide
on heterogeneous hydrogenation of benzaldehyde and
homogeneous dehydration of sorbitol in aqueous phase.
Initial hydrogenation rates of benzaldehyde over a
charcoal-supported palladium catalyst in water at 313 K
were enhanced by the addition of carbon dioxide. The
initial rate increased with an increase in carbon dioxide
pressure and became a maximum at 5 MPa. Dehydra-
tion of sorbitol in aqueous phase proceeded at 500 K
and the initial dehydration rates were enhanced by the
addition of 30 MPa of carbon dioxide.
hexanediol.
1,4-AHSO was obtained by dehydra-
tion of one water molecule from first position (primary
alcohol) and fourth position (secondary alcohol) of
hydroxyl groups in sorbitol molecules and 2,5-AHSO
was obtained by dehydration of one water molecule
from second position (secondary alcohol) and fifth
position (secondary alcohol) of hydroxyl groups in
sorbitol molecules. From the results of 1,4-butanediol,
2,5-hexanediol, and 2,5-dimethyl-2,5-hexanediol, it
was supposable that dehydration of sorbitol in water
could be enhanced by addition of 10 MPa of carbon
dioxide; however, it did not. Sorbitol molecules are
symmetric in polarity because a sorbitol molecule
consists of six carbon atoms and each carbon atom has
Acknowledgements
one hydroxyl group, which would be insensitive to the This study was supported by Special Coordina-
protonation; then the high pressured carbon dioxide tion Funds for Promoting Science and Technology
was not so effective. Although we did not calculate ‘Development of Sustainable Catalytic Reaction Sys-
the amount of protons, we can image that the amount tem using Carbon dioxide and Water’.
of proton in water phase under 10 MPa of carbon
dioxide is not so different from that under 30 MPa,
indicating that other factors except for proton would
References
contribute to the enhancement of dehydration. Molar
amount of 30 MPa of carbon dioxide is estimated to be
1
2
3
4
. Anastas P T and Zimmerman J B 2003 Environ. Sci.
Technol. 37 94A
. Osada M, Sato T, Watanabe M, Shirai M and Arai K
14
about one-third of water molecules in the reactor. We
will carry out phase observation and theoretical calcu-
lation of solvation geometry under reaction conditions
for understanding the effect of high-pressure carbon
dioxide.
2006 Comb. Sci. Tech. 178 537
. Arai M, Fujita S and Shirai M 2009 J. Supercritical
Fluids 47 351
. Hunter S E, Ehrenberger C E and Savage P E 2006 J.
Org. Chem. 71 6229

CO2 pressure effect on hydrogenation and dehydration
401
5
6
7
8
9
. Rode C V, Joshi U D, Sato O and Shirai M 2003 Chem. 10. Yamaguchi A, Hiyoshi N, Sato O and Shirai M 2011
Commun. 1960 Green Chem. 13 873
. Hiyoshi N, Sato O, Yamaguchi A and Shirai M 2011 11. Yamaguchi A, Hiyoshi N, Sato O and Shirai M 2011
Chem. Commun. 47 11546 ACS Catal. 1 67
. Osada M, Sato O, Watanabe M, Arai K and Shirai M 12. Yamaguchi A, Hiyoshi N, Sato O, Bando K K and Shirai
006 Energy Fuels 20 930 M 2009 Green Chem. 11 48
. Yamaguchi A, Hiyoshi N, Sato O, Rode C V and Shirai 13. Yamaguchi A, Hiyoshi N, Sato O and Shirai M 2012
M 2008 Chem. Lett. 37 926 Catal. Today 185 302
. Carroll J J, Slupsky J D and Mather A E 1991 J. Phys. 14. Hiyoshi N, Sato O, Yamaguchi A, Rode C V and Shirai
Chem. Ref. Data 20 1201 M 2013 J. Jpn. Petrol. Inst. 56 165
2