Hydrogen production by steam reforming of ethanol using cobalt and nickel catalysts supported on strontium titanate

Article abstract of DOI:10.1246/cl.2005.668

The steam reforming of ethanol over Co and Ni catalysts supported on SrTiO₃ was carried out at 823 K and atmospheric pressure. It was found that Co/SrTiO₃ showed a high catalytic activity and long-term stability and remarkably inhibited carbon deposition. Copyright

Full text of DOI:10.1246/cl.2005.668

668
Chemistry Letters Vol.34, No.5 (2005)
Hydrogen Production by Steam Reforming of Ethanol
Using Cobalt and Nickel Catalysts Supported on Strontium Titanate
Kohei Urasaki, Kazuhisa Tokunaga, Yasushi Sekine, Eiichi Kikuchi, and Masahiko Matsukata
Department of Applied Chemistry, School of Science and Engineering, Waseda University,
3-4-1, Okubo, Shinjuku, Tokyo 169-8555
(Received December 16, 2004; CL-041546)
The steam reforming of ethanol over Co and Ni catalysts
873 K in a hydrogen stream. After the reduction with hydrogen,
supported on SrTiO3 was carried out at 823 K and atmospheric
pressure. It was found that Co/SrTiO3 showed a high catalytic
activity and long-term stability and remarkably inhibited carbon
deposition.
a mixture of ethanol, steam and argon was fed into the reactor at
823 K. The flow rate of ethanol, steam and argon were 4.5, 45,
and 50 cm3 minꢀ1 (STP), respectively. The products were
analyzed by means of a gas chromatograph (Shimadzu GC-
14B. The hydrogen yield was defined as
out
in
in
Hydrogen yield ¼ FH
ꢁ 100=ð3 ꢁ FC H OH þ FH O
Þ
is the formation rate of hydrogen, FC H OH is the
2
2
5
2
In recent years, ethanol produced from renewable biomass
resources has attracted considerable attention as an energy
source alternative to fossil fuels.1,2 Since the concentration of
ethanol in biomass-derived aqueous solution is 12–15 wt %, en-
ergy intensive and rather expensive distillation processes has
been used for recovering highly concentrated ethanol. Instead,
hydrogen manufacture by the steam reforming of ethanol has
been proposed to be an effective way for the utilization of the bi-
omass-derived ethanol without distillation.
The number of publications on the catalytic steam reforming
of ethanol has substantially been increasing.3–11 Although Rh
seems to be most catalytically active for the steam reforming
of ethanol,5–7 the use of a noble metal is considered to be disad-
vantageous mostly due to an economical reason. On the other
hand, sintering of Ni and Co and coke formation often cause de-
activation of supported Ni and Co catalysts.8–11
out
in
where FH
feed rate of ethanol, and FH O is the feed rate of water. The
2
2
5
in
2
maximum hydrogen yield calculated is 46% when both the lev-
els of ethanol conversion and the selectivity to carbon dioxide
reach 100% (C2H5OH þ 3H2O ! 2CO2 þ 6H2).
We tested the activities of Co and Ni catalysts supported on
SrTiO3, MgO and ꢀ-Al2O3. Figure 1 and Table 1 show the levels
of ethanol conversion as a function of time on stream and the
yields of products, respectively. The catalytic activities signifi-
100
80
60
One of the approaches to improve the stabilities of the Co
and Ni catalysts is the selection and modification of catalyst sup-
port.9–11 We previously reported that Ni/SrTiO3 shows a high
catalytic activity and durability under the conditions of the steam
reforming of methane at 1073 K because the lattice oxygen in
SrTiO3 possibly promotes the oxidation of CHx fragments and
hinders the production of inactive carbonaceous species.12 Ni
catalysts supported on some perovskite type-oxides such as
SrTiO3 have been reported for partial oxidation of methane13,14
and dry reforming of methane.15
40
20
0
100
200
300
Time on stream / min
Figure 1. Steam reforming activities of supported Ni and Co
catalysts. Catalyst: , Co/MgO; , Ni/MgO; , Co/SrTiO3;
, Ni/SrTiO3; , Co/ꢀ-Al2O3.
Based on these results, in this study, we investigated the cat-
alytic activities and stabilities of Co and Ni catalysts supported
on SrTiO3 for the steam reforming of ethanol and compared with
those of Co and Ni catalysts supported on MgO or ꢀ-Al2O3,
which have been studied by several researchers.3,4,6–9,11
Table 1. Carbon-containing product yield for steam reforming
ethanol over supported Ni and Co catalysts
aTOS
Yield/%
Catalyst
/min
CO
CO2
CH4
C2H4
MgO (Kanto Chemical Co., Inc, BET = 46.4 m2 g-catꢀ1),
ꢀ-Al2O3 (JRC-ALO-2, the Catalysis Society of Japan, BET =
280 m2 g-catꢀ1) and SrTiO3 were used as the support. SrTiO3
was prepared by physically mixing TiO2 (anatase) and SrCO3
and then calcining the resultant mixture at 1423 K for 10 h.
The BET specific surface area was 7.7 m2 g-catꢀ1. Co and Ni
catalysts were prepared by an impregnation method using these
supports and an aqueous solution of Co(NO3)2 and Ni(NO3)2, re-
spectively. The amount of Co or Ni loaded was 5 wt % as metal.
Catalytic activities were tested in a continuous flow reactor
with a fixed bed of catalyst at 823 K and atmospheric pressure.
Before the activity test, the catalyst was reduced for 1 h at
Co/MgO
Co/MgO
Ni/MgO
Ni/MgO
Co/ꢀ-Al2O3
Co/ꢀ-Al2O3
Co/SrTiO3
Co/SrTiO3
Ni/SrTiO3
Ni/SrTiO3
55
300
55
300
60
180
55
300
55
13.9
11.1
10.7
8.6
6.3
6.1
22.4
21.0
19.1
16.4
21.5
15.7
21.5
14.5
4.4
4.3
3.2
6.0
4.1
2.2
2.0
7.8
8.5
9.8
6.2
0
0
0
0
69.1
86.5
0
0
0
3.8
28.9
38.2
27.2
23.0
300
0
aTime on stream
Copyright Ó 2005 The Chemical Society of Japan

Chemistry Letters Vol.34, No.5 (2005)
669
200
150
100
50
30
25
20
15
10
5
0
Ni/MgO Co/MgO Ni/SrTiO3 Co/SrTiO3 Ni/γ-Al2O3
Figure 3. Amounts of coke deposited on Co and Ni catalysts af-
ter time course experiments of ( ) 2 h and ( ) 5 h.
0
0
20
40
60
80
100
hand, Co and Ni catalysts supported on SrTiO3 showed much
longer-term stability. One of the crucial reasons for the deactiva-
tion of catalysts in the steam reforming of ethanol is the coverage
of active sites with deposited carbon. The amounts of carbon de-
posited during the reaction were quantified by the combustion of
carbon in a stream containing 30% O2. Figure 3 shows the
amounts of carbon deposited after the activity test for 2 and
5 h. Carbon was deposited on the catalysts mostly up to 2 h on
stream. We, thus, consider that the decrease in the activities of
Co/MgO, Ni/MgO, and Ni/SrTiO3 in the early stage of reaction
is mainly due to coking on catalysts and the blockage of the
active sites.
The amounts of carbon deposited on Co/SrTiO3 and Ni/
SrTiO3 were much smaller than those on Ni/MgO, Co/MgO,
and Co/ꢀ-Al2O3. In particular, the amounts of carbon deposited
on Co/SrTiO3 after the activity test for 5 h were extremely
small, 9 mg g-catꢀ1. As described above, we have reported that
in the case of the steam reforming of methane over supported Ni
catalysts at 1073 K, Ni catalysts using perovskites as support
suppress the formation of inactive carbonaceous species. In the
present study, we presume that SrTiO3 similarly acts to inhibit
coke formation at a lower temperature employed for the steam
reforming of ethanol.
Ethanol conversion / %
Figure 2. Hydrogen yield as functions of ethanol conversions
of over supported Co and Ni catalysts. Catalyst: , Co/MgO;
, Ni/MgO; , Co/SrTiO3; , Ni/SrTiO3; , Co/ꢀ-Al2O3.
Maximum hydrogen yield at a given ethanol conversion was
shown as the dotted line.
cantly depended on the kind of catalyst support. Although Co/ꢀ-
Al2O3 gave the highest levels of conversion of ethanol among
the catalysts tested, the yield of ethylene was extremely high
as shown in Table 1, resulting in small amounts of hydrogen
formed. It is known that dehydration of ethanol on an acidic
ꢀ-Al2O3 support let to the selective formation of ethylene.3,4,9–11
On the other hand, the products over other catalysts tested were
composed of hydrogen, carbon dioxide, carbon monoxide and
methane. Other products such as oxygenated compounds or
ethylene were not detected.
Co/MgO and Ni/MgO featured similar levels of conver-
sion. While the levels of ethanol conversion over Co/MgO
and Ni/MgO after 15 min on stream were 54 and 49%, respec-
tively, both catalysts deactivated rapidly in the first hour of reac-
tion. After the reaction for 300 min, the levels of ethanol conver-
sion over Co/MgO and Ni/MgO decreased to 30 and 27%, re-
spectively.
We found that SrTiO3 is suitable as a support of Co and Ni,
compared with conventional ꢀ-Al2O3 and MgO. Although the
conversion level over Ni/SrTiO3 decayed from 60.3% after
15 min to 45.6% after 300 min, the rate of deactivation was re-
duced in comparison with Ni/MgO. It should be noted that the
catalytic activity of Co/SrTiO3 gradually increased with time
on stream: the level of ethanol conversion increased from 60%
after 15 min to 68% after 60 min and level off.
Figure 2 shows the hydrogen yield as a function of ethanol
conversion. The maximum hydrogen yield at a given ethanol
conversion is calculated when the selectivity to carbon dioxide
reaches 100%. All the catalysts used show a similar trend except
for Co/ꢀ-Al2O3 producing ethylene as the main product. The
hydrogen yields over Co catalysts are slightly higher than that
over Ni catalysts at a given level of ethanol conversion. The se-
lectivities to methane over Co/SrTiO3 and Co/MgO were ca. 12
or 10%, respectively, while Ni/SrTiO3 and Ni/MgO gave slight-
ly higher selectivities to methane of ca. 14 and 15%, respective-
ly. The lower selectivities to methane over Co catalysts caused
higher yield of hydrogen.
References
1
L. R. Lynd, J. H. Cushman, R. J. Nichols, and C. E. Wyman, Science,
251, 1318 (1991).
2
3
L. F. Brown, Int. J. Hydrogen Energy, 26, 381 (2001).
F. Haga, T. Nakajima, K. Yamashita, S. Mishima, and S. Suzuki,
Nippon Kagaku Kaishi, 1997, 1.
4
5
6
7
8
9
F. Haga, T. Nakajima, H. Miya, and S. Mishima, Catal. Lett., 48, 223
(1997).
P. Y. Sheng, A. Yee, G. A. Bownmaker, and H. Idriss, J. Catal., 208,
393 (2002).
S. Cavallaro, N. Mondello, and S. Freni, J. Power Sources, 102, 198
(2001).
D. K. Liguras, D. I. Kondarides, and X. E. Verykios, Appl. Catal., B,
43, 345 (2003).
S. Freni, S. Cavallaro, N. Mondello, L. Spadaro, and F. Frusteri, Catal.
Commun., 4, 259 (2003).
J. Llorca, N. Homs, J. Sales, and P. R. Piscina, J. Catal., 209, 306
(2002).
10 A. N. Fatsikostas and X. E. Verykios, J. Catal., 225, 439 (2004).
11 M. S. Batista, R. K. S. Santos, E. M. Assaf, J. M. Assaf, and E. A.
Ticianelli, J. Power Sources, 124, 99 (2003).
12 K. Urasaki, Y. Sekine, S. Kawabe, E. Kikuchi, and M. Matsukata,
Appl. Catal., A, in press.
13 K. Takehira, T. Shishido, and M. Kondo, J. Catal., 207, 307 (2002).
14 H. Provendier, C. Petit, C. Estournes, S. Libs, and A. Kiennemann,
Appl. Catal., A, 180, 163 (1999).
15 T. Hayakawa, S. Suzuki, J. Nakamura, T. Uchijima, S. Hamakawa, K.
Suzuki, T. Shishido, and K. Takehira, Appl. Catal., A, 183, 273 (1999).
As described above, the catalytic activities of Co/MgO and
Ni/MgO decreased rapidly although these catalysts showed high
catalytic activities in the early stage of reaction. On the other
Published on the web (Advance View) April 9, 2005; DOI 10.1246/cl.2005.668