Continuous supercritical low-temperature methanol synthesis with n-butane as a supercritical fluid

Article abstract of DOI:10.1246/cl.2008.790

A process of supercritical low-temperature methanol synthesis from syngas containing CO₂ was carried out at 443 K and 60 bar. The 2-butanol and n-butane was used as catalytic solvent and supercritical medium, respectively. The results showed that the total carbon conversion, especially the CO ₂ conversion of the methanol synthesis was increased significantly under the supercritical condition. Copyright

Full text of DOI:10.1246/cl.2008.790

790
Chemistry Letters Vol.37, No.7 (2008)
Continuous Supercritical Low-temperature Methanol Synthesis
with n-Butane as a Supercritical Fluid
Prasert Reubroycharoen,ꢀ1 Jun Bao,2 Yi Zhang,3 and Noritatsu Tsubakiꢀ3
1Department of Chemical Technology, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand
2National Synchrotron Radiation Lab., University of Science and Technology of China, Hefei, Anhui 230026, P. R. China
3Department of Applied Chemistry, School of Engineering, University of Toyama, Toyama 930-8555
(Received April 1, 2008; CL-080332; E-mail: prasert.r@chula.ac.th, tsubaki@eng.u-toyama.ac.jp)
A process of supercritical low-temperature methanol syn-
lower Tc was employed to accompany ROH where the latter
acted like an entrainer. The results showed that the reaction
rate of low-temperature methanol synthesis can be improved
significantly under the supercritical conditions.
thesis from syngas containing CO2 was carried out at 443 K
and 60 bar. The 2-butanol and n-butane was used as catalytic
solvent and supercritical medium, respectively. The results
showed that the total carbon conversion, especially the CO2
conversion of the methanol synthesis was increased significantly
under the supercritical condition.
The Cu/ZnO catalyst with Cu/Zn molar ratio of 1 was
prepared by the coprecipitation from copper nitrate and zinc
nitrate, using sodium carbonate as precipitant.8 The obtained
precursor was reduced at 473 K for 13 h by flowing 5% H2 in
N2 and passivated by 2% O2 in Ar. The reaction was carried
out in a continuous flow fixed-bed reactor. The feed gas was
CO/H2/CO2/Ar = 31.6/60.7/4.8/2.9. The balance gas was
helium in the gas-phase reaction. 2-Butanol and n-butane were
used as the solvent and SCF, respectively. Detailed information
on the experimental setup and the product analysis has been
reported elsewhere.10
Supercritical fluid (SCF) as a new technology has unique
properties such as superior solubility and transfer characteristics.
The use of SCF in heterogeneous reactions can increase the re-
action rates because the products and reaction heat can be re-
moved out rapidly while the multiphase interface is eliminated
and the molecules diffusivity is increased. Moreover, better se-
lectivity can be achieved in supercritical reactions owing to
the possibility of uncoupling process variables and the deactiva-
tion of catalyst can be mitigated through better heat and mass
transfer.1,2
Methanol, as one of the primary chemicals, is commercially
produced via the ICI process under extreme conditions, such as
573 K and 100 bar. Under these conditions, one-pass conversion
is only 15–25% because of severe thermodynamic limitation
of this exothermic reaction.3,4 Zhu et al. conducted the ICI
methanol synthesis under supercritical conditions with n-hexane
as the supercritical medium.5 The conversions and methanol
mole fraction are obviously higher than those of gas–solid
reaction. Similar results were also reported by Zhong et al.6
Han et al. simulated the methanol synthesis in supercritical n-
hexane using the Monte Carlo (MC) method.7 The results indi-
cated that the catalyst activities depended not only on the num-
ber of active sites, but also on the ratio of different adsorbed spe-
cies on the catalyst surface.
The applicability of ester, alcohol, ketone, and saturated hy-
drocarbon as supercritical medium was studied by Zhong et al.
The results showed that only saturated hydrocarbon has prefera-
ble appetency to methanol under the supercritical condition.11 In
this work, the n-butane was selected as the supercritical medium
due to its low critical point (Tc ¼ 425:2 K, Pc ¼ 37:7 bar).
Additionally, n-butane in the supercritical phase at the range
of reaction temperature has favorable stability and preferable
diathermancy. The reaction results are shown in Table 1. In
general, the catalytic conversions are relatively low owing to
the low partial pressure of feed gas and alcohol solvent. Only
CO and CO2 were the carbon-containing gas components after
reaction. The reaction without 2-butanol and n-butane had a
low conversion. The intermediate formic acid was the only
product and no methanol was detected, which might be due to
the low temperature and short contact time. The total carbon
conversion increased when the supercritical n-butane was intro-
duced. With only the presence of 2-butanol, the intermediate es-
ter which was formed by the esterification as described in step
(2),9 was the only liquid product and no methanol was detected
because of the short contact time. When conducting the reaction
in the supercritical n-butane as in Reaction No. 4, the total
carbon conversion increased from 2.6 to 4.2%. At a higher
W=F value of 30 gꢁhꢁmolꢂ1, the methanol selectivity of the super
critical reaction was high up to 98.4% and the total carbon
conversion was 8.6%, much higher than that of the reaction
without n-butane. The 2-butanol, acted as the catalytic solvent,
was almost not consumed during the reaction, except only a little
was converted to the intermediate ester. These findings indicated
that the utilization of SCF in the low-temperature-methanol
synthesis increased the reaction rate significantly.
Previous studies of our group have developed a new route
of low-temperature methanol synthesis.8,9 This process consists
of three steps: (1) water–gas shift reaction; (2) esterification
reaction; and (3) hydrogenation of ester to form methanol and
alcohol. Herein, ROH acts as a catalytic solvent, which partici-
pates in the reaction but is not consumed by the overall reaction.
CO þ H2O ¼ (HCOOH) ¼ CO2 þ H2
HCOOH þ ROH ¼ HCOOR þ H2O
HCOOR þ 2H2 ¼ CH3OH þ ROH
CO þ 2H2 ¼ CH3OH
ð1Þ
ð2Þ
ð3Þ
ð4Þ
Through the new path, methanol can be produced with high
one-pass conversion and specific selectivity at 423–443 K and
30 bar. To investigate the promotional effects of SCF on this
low-temperature–catalysis process, supercritical n-butane with
The catalytic solvent effect of 1-butanol and 2-butanol on
the activity was compared in Table 2. 2-Butanol exhibited higher
conversion and selectivity than that of the 1-butanol. It is sug-
Copyright Ó 2008 The Chemical Society of Japan

Chemistry Letters Vol.37, No.7 (2008)
Table 1. Reaction results for supercritical low-temperature methanol synthesisa
791
Partial pressure/bar
Conversion/%
Selectivity/%
HCOOR HCOOH
W=F
No.
Syngas
He
2-Butanol
n-Butane
/gꢁhꢁmolꢂ1
CO
CO2
Total Cb
MeOH
1
2
3
4
5
6
10
10
10
10
10
10
50
10
40
0
40
0
0
0
10
10
10
10
0
40
0
40
0
10
10
10
10
30
30
2.3
ꢂ0.1
1.2
46.2
2.2
6.0
2.6
4.2
5.4
8.6
0
0
0
0
95.5
98.4
0
0
100
100
0
5.5 ꢂ16.2
100
100
4.5
1.6
ꢂ2.1
45.7
ꢂ1.2
40.2
0
0
0
6.4
3.8
40
aReaction conditions: T ¼ 443 K, P(total) = 60 bar, P(CO/H2/CO2) = 10 bar, Catalyst weight: 0.5 g. bTotal carbon conver-
sion = CO conv. ꢃ 31.6/(31.6 + 4.8) + CO2 conv. ꢃ 4.8/(31.6 + 4.8).
of the reaction, since the dead volume of the reactor and the
Table 2. Effects of 1-butanol and 2-butanol on the methanol
ice-trap were filled by the pressurized feed gas, the apparent
synthesis activitya
conversions were low but increased gradually. After reaction
Conversion/%
Selectivity/%
MeOH Ester
for 10 h, the dilution effect disappeared and the conversions were
stable. It is noted that with only the presence of 2-butanol, CO2
conversion was negative, while in the supercritical n-butane,
CO2 conversion was high up to 53.5%. From the reaction steps
(1–4), CO was converted to CO2 firstly through the water–gas
shift reaction, step (1), and then CO2 was converted to methanol
via other steps. This suggested that without supercritical n-
butane, the formation rate of CO2 was higher than its consump-
tion rate, resulting in the accumulation of CO2 and the negative
conversion. Under supercritical conditions, the high CO2 con-
version indicated that the conversion rate of CO2 was increased
more significantly than that of CO, which might be attributed
to the fact that the nonpolar CO2 has higher solubility in
the nonpolar supercritical n-butane due to the rule of ‘‘like
dissolves like.’’
Alcohol
CO
CO2
Total C
2-Butanol
1-Butanol
11.0
3.8
53.5
64.2
16.6
11.7
97.7
94.3
2.3
5.7
aReaction conditions: T ¼ 443 K, P(total) = 60 bar, P(CO/H2/
CO2) = 10 bar, P(n-butane) = 40 bar; P(alcohol) = 10 bar;
.
Catalyst weight: 1.5 g, W=F ¼ 90 gꢁhꢁmolꢂ1
10
(a)
5
0
-5
-10
-15
60
In summary, the introduction of supercritical n-butane in
the low-temperature methanol synthesis can increase the
catalytic conversion, especially the CO2 conversion significant-
ly. The present study provides an experimental foundation
for the industrial application of supercritical low-temperature
methanol synthesis.
(b)
50
40
30
20
10
0
0
3
6
9
12
15
18
References
1
Time on stream/h
P. E. Savage, S. Gopalan, T. I. Mizan, C. J. Martino, E. E.
Brock, AIChE. J. 1995, 41, 1723.
Figure 1. Variations of CO, CO2, and total carbon conversions
with reaction time. Reaction conditions: T ¼ 443 K, P(total) =
60 bar, P(CO/H2/CO2) = 10 bar, P(2-butanol) = 10 bar,
P(He or n-butane) = 40 bar, W=F(CO/H2/CO2) = 90 gꢁhꢁ
molꢂ1, Catalyst weight: 1.5 g. (a) With only the presence of 2-
butanol; (b) with the coexistence of 2-butanol and supercritical
n-butane. ( ) CO conv.; ( ) CO2 conv.; ( ) Total carbon conv.
2
M. Shirai, C. V. Rode, E. Mine, A. Sasaki, O. Sato, N.
Hiyoshi, Catal. Today 2006, 115, 248.
3
4
I. Wender, Fuel Process. Technol. 1996, 48, 189.
G. H. Graaf, P. Sijtsema, E. J. Stamhuis, G. Joosten, Chem.
Eng. Sci. 1986, 41, 2883.
5
6
7
S.-H. Zhang, T. Li, B.-C. Zhu, Chem. Eng. Sci. 2006, 61,
1167.
T. Jiang, Y. Niu, B. Zhong, Fuel Process. Technol. 2001, 73,
175.
X. Zhang, B. Han, Y. Li, B. Zhong, S. Peng, J. Supercrit
Fluids 2002, 23, 169.
N. Tsubaki, M. Ito, K. Fujimoto, J. Catal. 2001, 197, 224.
J. Zeng, K. Fujimoto, N. Tsubaki, Energy Fuels 2002, 16, 83.
gested that different alcohol affected the esterification reaction,
step 2, by both electronic and spatial effects.9 High electron den-
sity of oxygen atom in branched ROH makes it more efficient for
the ROH to attack the carbon atom of HCOOCu, the intermedi-
ate of step (2). Low steric hindrance in linear ROH is favorable
to the nucleophilic attack in the esterification reaction. As a
balanced effect between electronic factor and spatial factor,
2-butanol exhibited higher promoting effect.
8
9
10 P. Reubroycharoen, Y. Yoneyama, T. Vitidsant, N. Tsubaki,
Fuel 2003, 82, 2255.
11 B. Zhong, 3rd National Symposium on Supercritical Fluids
Technology and Applications, 2000, p. 139.
Figure 1 shows the variations of CO, CO2, and total carbon
conversions with reaction time. For comparison purpose, the
reaction without n-butane was also tested. At the initial stage
Published on the web (Advance View) June 21, 2008; doi:10.1246/cl.2008.790