Preparation of an alumina-wall tube reactor and its catalytic performance in the steam-reforming of methanol

Article abstract of DOI:10.1246/cl.2007.1342

The inner wall of a narrow-bore aluminum tube was successfully oxidized to give a thin alumina wall by electrolysis. A Cu-Zn-loaded alumina-wall tube reactor afforded high performance in the steam reforming of CH₃OH to supply H₂. Copyright

Full text of DOI:10.1246/cl.2007.1342

1342
Chemistry Letters Vol.36, No.11 (2007)
Preparation of an Alumina-wall Tube Reactor and Its Catalytic Performance
in the Steam-reforming of Methanol
Fumihiko Naka-o, Na-oki Ikenaga, Takanori Miyake, and Toshimitsu SuzukiÃ
Department of Chemical Engineering, Kansai University and High Technology Research Center, Suita, Osaka 564-8680
(Received August 13, 2007; CL-070865; E-mail: tsuzuki@ipcku.kansai-u.ac.jp)
The inner wall of a narrow-bore aluminum tube was suc-
cessfully oxidized to give a thin alumina wall by electrolysis.
A Cu–Zn-loaded alumina-wall tube reactor afforded high per-
formance in the steam reforming of CH3OH to supply H2.
100 nm
20 µm
Al2O3
The steam-reforming reaction of methanol (SRM) has
attracted much attention as a promising method for a compact
hydrogen production system for a fuel cell. Because of the large
endothermicity of this type of reaction,1 a packed bed reactor
was found to exhibit heat-transfer limitations and temperature
gradients in the reactor. This heat-transfer limitation can lead
to lowered catalytic productivity. To avoid this problem, use
of a wall-tube reactor is proposed.2,3 The formation of a porous
alumina thin layer on an aluminum plate by anodic oxidation is
an established technique.4 However, until now, there have been
no reports on the anodic oxidation of the inner wall of a thin
aluminum tube. This paper provides a first report on the anodic
oxidation of a thin aluminum tube and its application to the
SMR.
Al
(a) Cross section
(b) Inner surface
Figure 1. SEM images of the oxidized aluminum tube. Prepara-
tion conditions: constant voltage of 30-V DC at 18 ꢀC in 0.6 M
(COOH)2 electrolyte for 9 h.
50
40
30
20
10
0
A straight aluminum tube, 1 m in length with 3-mm outer
and 2-mm inner diameters, was used. The aluminum tube was
pretreated by washing, first with benzene and then aqueous
NaOH, rinsing it in distilled water, and washing it again with
HNO3 solution.
Anodic oxidation of the inner wall of the pretreated tube was
carried out by inserting a Teflon-coated 0.56-mm copper wire,
with the Teflon coat partially stripped off, into the tube, and this
was used as a cathode electrode. The anodic oxidation was car-
ried out by following the literature.5 Preoxidation was achieved
by circulating the electrolyte (0.6 M oxalic acid) inside the tube
at a flow rate of 25 mL/min, and an electric potential of 30-V DC
was applied between the cathode and aluminum tube. After the
first oxidation, the inner surface was exposed to a mixture of
6.0 wt % H3PO4 and 1.8 wt % H2CrO4 at 60 ꢀC for 15 min to re-
move the oxidized alumina layer. The second anodic oxidation
was carried out under the same condition as above for 1–9 h.
The obtained tube was washed with distilled water, and calcined
at 350 ꢀC under flowing air.
Figure 1 shows scanning electron micrograph (SEM) images
of the cross section and the inner surface of the oxidized alumi-
num tube. Relatively uniform pores sized 20 to 30 nm were
observed, and the surface area (BET, N2) of oxidized inner wall
was ca. 20 m2/m-tube.
As seen in Figure 2, thickness of the alumina layer increased
to 40 mm linearly with increasing oxidation time up to 9 h.
Oxidation time of 3 h was selected to maintain the mechanical
strength of the tube.
0
2
6
8
10
4
Oxidation time/h
Figure 2. Relation between thickness of alumina film and
oxidation time. Preparation conditions constant voltage of 30-
V DC at 18 ꢀC in 0.6 M (COOH)2 electrolyte.
for 2–12 h, the solution inside the tube was drained off, followed
by calcination in air flow at 350 ꢀC for 6 h.
The reforming reaction was carried out by feeding CH3OH
together with Ar as a sweep gas at 200–300 ꢀC, using an effec-
tive tube length of 0.8 m (heated zone). As shown in Table 1,
with the bare alumina tube (Run 1), only a trace amount of
DME was obtained. With a loading level of 2.8 mg/m-tube of
combined weight of Cu and Zn oxides, methanol conversion
of 23.9% was observed. If the catalyst adsorption time was pro-
longed from 2 to 12 h, the loading level increased from 2.8 to
4.5 mg/m-tube at 8 h, with an increase in CH3OH conversion
to 60.2% (Run 5). Since elongation of the catalyst loading time
did not significantly increase the amount of loading, catalyst
loading process was repeated as shown in Runs 6–9. With in-
creasing repetition number, catalyst loading increased. When
impregnation procedures were repeated four times, the loading
level reached to 12.4 mg/m-tube, leading to methanol conver-
Catalysts were prepared by the equilibrium adsorption
method. An aqueous solution of 0.25 M Cu and 0.25 M Zn
nitrates was introduced into the wall tube reactor. After standing
Copyright ꢀ 2007 The Chemical Society of Japan

Chemistry Letters Vol.36, No.11 (2007)
Table 1. Results of SRM using wall tube reactora
1343
Catalyst preparation
Impregnation
W=Fc
/g h molÀ1
CH3OH
conv./%
H2
/mmol hÀ1
selectivity/%
CO CO2
Run
Catalyst
time/h
repetition
weight/mg mÀ1b
1d
2d
3d
4d
5d
0
2
4
8
12
0
1
1
1
1
0
0
0.2
23.9
38.7
58.7
60.2
0
0.0
3.2
2.1
2.0
2.2
0
2.8
3.3
4.5
4.5
0.10
0.12
0.17
0.17
28.3
36.8
70.1
70.1
91.5
96.5
96.5
97.3
6
7
8
9e
8
8
8
8
1
2
3
4
4.5
6.2
8.3
12.4
0.17
0.23
0.31
0.46
28.2
78.1
89.8
45.5
22.7
62.9
72.3
36.2
0.3
0.8
1.3
0.0
99.6
99.2
98.6
99.9
10
11
12
13
14
8
8
8
8
8
4
4
4
4
4
12.4
12.4
12.4
12.4
12.4
0.46
0.28
0.20
0.15
0.12
98.8
97.2
82.7
73.3
61.3
75.8
126
151
187
213
2.0
2.1
1.6
1.1
1.3
97.9
97.9
98.3
98.8
98.6
c
aReaction temperature 250 ꢀC, time 1 h, Ar 30 mL/min, Steam/Carbon 1.0. bTotal weight of CuO and ZnO. W=F catlyst weight
e
(g)/feed methanol (mol/h). dReaction temperature 300 ꢀC. Reaction temperature 200 ꢀC.
decrease in the W=F to 0.12 gÁcat/mol (feed rate: 107 mmol/h)
100
300
decreased in the methanol conversion to 61.3%. Owing to
an open tubular reactor, at a high space velocity, a part of
fed methanol might have passed through without contacting on
active catalyst. At 250 ꢀC, space time yield of this tube was
estimated as ca. 50 mmol H2/h cm3 from the result of Run 11.
This value is much higher than reported results,6 although the
catalyst and conditions are different.
80
60
40
20
250
200
150
All these results clearly indicate that SRM using a wall tube
reactor afforded high CH3OH conversion to give high H2 yield
with a low CO selectivity.
100
50
0
This work was partly supported by ‘‘High-Tech Research
Center Project’’ for Private Universities: matching fund subsidy
from MEXT, 2007–2011.
0
0.6
0.4
0.0
0.2
W/F / g h mol−1
Figure 3. Effect of W=F on steam reforming of methanol. Tem-
perature 250 ꢀC, Reaction time 1 h, Steam/Carbon , conver-
sion; , H2 yield. Product selectivity: , CO2; , CO; , DME.
References
1
C. Cao, G. Xia, J. Holladay, E. Jones, Y. Wang, Appl. Catal.,
A 2004, 262, 19.
sion of 45.5 and 98.8% at 200 and 250 ꢀC, respectively, with a
very low CO selectivity at 200 ꢀC (Runs 9 and 10).
The effect of the contact time (W=F) of the methanol
was examined at a constant CuO + ZnO loading, by varying
methanol feed rate. As shown in Figure 3, at a W=F of
0.46 g h/mol (feed rate: 26.8 mmol/h), methanol conversion of
98.8% and H2 yield of 75.8 mmol/h were obtained, while a
2
3
A. Karim, J. Bravo, A. Datye, Appl. Catal., A 2005, 282, 101.
A. Karim, J. Bravo, D. Gorm, T. Conant, A. Datye, Catal.
Today 2005, 110, 86.
4
J. C. Ganley, K. L. Riechmann, E. G. Seebauer, R. I. Masel,
J. Catal. 2004, 227, 26.
5
6
H. Masuda, K. Fukuda, Science 1995, 268, 1466.
X. Zhang, P. Shi, J. Mol. Catal. A: Chem. 2003, 194, 99.
Published on the web (Advance View) October 13, 2007; doi:10.1246/cl.2007.1342