Preferential CO oxidation promoted by the presence of H2 over K-Pt/Al2O3

Article abstract of DOI:10.1039/b416565a

In preferential CO oxidation in H₂-rich gas, K-Pt/Al ₂O₃ (K/Pt =10) was very effective in decreasing CO concentration below 10 ppm in the 375-410 K range, and the turnover frequency of the K-Pt/Al₂O₃

Full text of DOI:10.1039/b416565a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Preferential CO oxidation promoted by the presence of H2 over
K–Pt/Al2O3
Yuji Minemura,a Shin-ichi Ito,a Toshihiro Miyao,b Shuichi Naito,b Keiichi Tomishigea and Kimio Kunimori*a
Received (in Cambridge, UK) 28th October 2004, Accepted 5th January 2005
First published as an Advance Article on the web 26th January 2005
DOI: 10.1039/b416565a
and 75 vol% H2 (He balanced). The effluent gas was analyzed
using an on-line gas chromatograph (GC) system equipped with a
TCD detector. In addition, the concentration at ppm level of CO
in the effluent gas was determined using FID-GC equipped with a
methanator. The activity was evaluated on the basis of CO and O2
conversions, which can be calculated on the basis of CO and O2
concentrations in the reactant gas and the effluent gas. The
selectivity of CO oxidation is defined as the ratio of O2
consumption for the CO oxidation to total O2 consumption. In
addition, we also measured the catalyst bed temperature directly
using a thermocouple. No temperature increase was observed at all
during the reaction. Therefore under these reaction conditions, the
thermal effect, which is due to exothermic reaction of CO and H2
oxidation, can be neglected. The catalysts were characterized by H2
and CO adsorption measurements and by transmission electron
microscope observation. The amounts of irreversible adsorption of
H2 (H/Pt) and CO (CO/Pt) were measured at room temperature in
a vacuum system (sample weight: 0.15 g, dead volume: 65 cm3) by
the volumetric method.10
In preferential CO oxidation in H2-rich gas, K–Pt/Al2O3
(K/Pt 5 10) was very effective in decreasing CO concentration
below 10 ppm in the 375–410 K range, and the turnover
frequency of the K–Pt/Al2O3 was 20 times as high as that of
Pt/Al2O3 at 353 K; furthermore, the activity of CO oxidation
was promoted drastically by the presence of H2.
Polymer electrolyte fuel cells (PEFC) can generate electricity
without polluting the environment, however, they demand high-
purity hydrogen as fuel. Especially when hydrogen is produced by
reforming of methanol and hydrocarbons, the gas always contains
some amount of CO, which poisons for the fuel cell. Therefore, the
hydrogen production system must be equipped with a CO removal
system. Although the CO concentration can be decreased via water
gas shift reaction, the stream produced usually contains about
y1–2 vol% CO. The performance of PEFC with the conventional
Pt catalyst seriously decreased with a poisoning of only 10 ppm of
CO.1,2 Regarding the CO poisoning, it has been reported that the
PtRu alloy has higher resistance (100 ppm) than Pt catalyst.3
However, in any case, stringent removal of CO is necessary, and
one promising method is preferential oxidation of CO in H2-rich
gas with air. Supported noble metals, such as Pt,4–8 Pd, Rh9,10 have
been reported. We report that a K-promoted Pt/Al2O3 catalyst is
highly efficient for CO removal down to 10 ppm level by
preferential CO oxidation in H2-rich gas compared to Pt/Al2O3. It
is also mentioned that the presence of H2 promoted drastically CO
oxidation over K–Pt/Al2O3.
Fig. 1 shows the reaction temperature dependence of the
performance of CO oxidation in H2-rich gas over K–Pt/Al2O3
catalysts. The CO conversion over unpromoted Pt/Al2O3 catalyst
was much lower than those over K–Pt/Al2O3 catalysts. The
selectivity of CO oxidation was in the range of 10–30% over Pt/
Al2O3, and this indicates that the oxidation of CO is less
preferential than H2 oxidation. On the other hand, the addition
of K to Pt/Al2O3 enhanced the CO conversion and selectivity of
CO oxidation dramatically. The additive effect of potassium was
greatly dependent on the amount of additive, and the optimum
amount was K/Pt 5 10. In addition, the CO concentration in the
effluent gas over K–Pt/Al2O3 (K/Pt 5 10) is also shown in Fig. 1(a).
It is interesting that CO concentration was maintained below
10 ppm in the 375 K–410 K range. Especially from the practical
view of hydrogen production for PEFC, this result can be
important. In our experiments, the feeding ratio of CO to O2 was
1/1, and this corresponds to the twice of the stoichiometric O2
amount. This is related to about 50% of CO oxidation selectivity.
The stoichiometric condition (CO/O2 5 2) can be ideal in terms
that the loss of H2 by oxidation and higher efficiency. However,
the CO conversion and the selectivity of CO oxidation over K–Pt/
Al2O3 (K/Pt 5 10) under the stoichiometric feeding conditions
(0.2 vol% CO, 0.1 vol% O2 and 75 vol% H2 balanced with He)
were 58.0% and 57.6% at 388 K, and the CO concentration in the
effluent gas is not so low as that shown in Fig. 1. This result
indicates that our feeding ratio (CO/O2 5 1) is necessary to
decrease the CO concentration in the effluent gas to the suitable
level.
Al2O3 (JRC-ALO-4 from Japan Reference Catalyst (JRC); BET
surface area, 170 m2 g21) was used as a support material; it was
calcined at 873 K for 3 h under atmosphere. Pt/Al2O3 catalyst was
prepared by impregnation of Al2O3 support with an aqueous
solution of Pt(NO2)2(NH3)2. The sample was dried at 383 K for
12 h, and then calcined in air at 773 K for 3 h. The loading amount
of Pt was 2 wt.%. Potassium-promoted Pt/Al2O3 catalyst (denoted
as K–Pt/Al2O3) was prepared by impregnation with an aqueous
solution of Pt(NO2)2(NH3)2, followed by drying at 383 K for 12 h,
and then impregnation of an aqueous solution of KNO3. After
these procedures, the calcination was carried out in air at 773 K for
3 h. Potassium loading is denoted as the molar ratio of K/Pt in
parentheses. These catalysts were reduced with hydrogen at 773 K
for 1 h in a catalyst bed reactor just before the activity test.
CO oxidation in H2-rich gas was carried out in a fixed-bed flow
reaction system at atmospheric pressure using 100 mg of the
catalyst at the total flow rate of 100 cm3 min21 (STP) (GHSV 5
30 000 h21). The feed stream contained 0.2 vol% CO, 0.2 vol% O2
*kunimori@ims.tsukuba.ac.jp
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 1429–1431 | 1429

View Article Online
K–Pt/Al2O3 (K/Pt 5 10) was determined to be 2.0 nm by TEM
observation, and this suggests that the dispersion can be 0.55, and
this is larger than the estimation from the adsorption measure-
ments. From this difference, it is suggested that the surface of Pt
metal particles are covered with potassium compound. Similar
behavior was also observed on K–Rh/USY catalysts.12
TOF(turnover frequency) values of CO2 formation in H2-rich
gas are also listed in Table 1. In this case, TOF was determined on
the basis of CO adsorption. It is possible that CO is adsorbed on
potassium oxide to give carbonate species. CO is mainly adsorbed
on Pt surface atoms compared to potassium ions, and this is
confirmed by FTIR observation of CO adsorption over K–Pt/
Al2O3. This indicates that potassium oxide is minor species on the
catalyst, and it is expected that potassium species is present as
potassium carbonate. In addition, CO was not adsorbed on
potassium carbonate as reported previously.13 TOF increased
drastically with increasing the additive amount of potassium in the
range of K/Pt 5 5–10. On the other hand, by more addition, TOF
decreased, and the optimum was observed at K/Pt 5 10.
Especially, it should be noted that the TOF of K–Pt/Al2O3
(K/Pt 5 10) was 20 times higher than that of Pt/Al2O3 in terms
of CO conversion. This indicates that the addition of potassium to
Pt/Al2O3 dramatically promotes the preferential oxidation of CO
in H2-rich gas.
Fig. 2 shows the reaction temperature dependence of catalyst
performance in CO oxidation (CO + O2) and preferential
oxidation of CO (CO + O2 + H2) over K–Pt/Al2O3 (K/Pt 5 0
and 10). In the CO oxidation (CO + O2) without the presence
of H2, K–Pt/Al2O3 exhibited higher CO oxidation activity than
Pt/Al2O3. Furthermore, in the case of Pt/Al2O3, from the
comparison between CO + O2 and CO + O2 + H2 reactions, the
addition of H2 promoted CO oxidation in the temperature range
below 413 K, however, at higher temperature, the presence of H2
drastically inhibited CO oxidation. This is because oxidation of H2
is more preferential than CO oxidation. On the other hand,
regarding K–Pt/Al2O3 (K/Pt 5 10), the presence of H2 promoted
CO oxidation drastically. The promoting effect of the H2 presence
on K–Pt/Al2O3 was much more significant than that on Pt/Al2O3.
Fig. 1 Reaction temperature dependence of (a) CO conversion and (b)
selectivity of CO oxidation in CO oxidation in H2-rich gas over K–Pt/
Al2O3 catalysts (K/Pt 5 0, 5, 10, 15 and 20). CO concentration in the
effluent gas over K–Pt/Al2O3 (K/Pt 5 10) (6). Reaction conditions:
0.2 vol% CO, 0.2 vol% O2 and 75 vol% H2 (He balance).
Pt dispersions estimated from the adsorption amounts of H2
and CO are listed in Table 1. In the case of Pt/Al2O3, the
dispersion can be determined to be about 0.5, which corresponds
to the particle size 2 nm based on the previous report.11 The
adsorption amounts of H2 and CO decreased with increasing the
additive amount of potassium. On the other hand, we also
measured XRD patterns of reduced K–Pt/Al2O3 catalysts, and the
peak assigned to Pt metal was not observed at all on all the K–Pt/
Al2O3 catalysts. This indicates that the size of Pt metal particles
is below 3 nm on all the catalysts. Average particle size of Pt over
Table 1 Pt dispersion and TOF of CO oxidation in H2-rich gas over
K–Pt/Al2O3
Adsorptiona
TEM
K–Pt/Al2O3K/Pt Particle size/nm H/Pt CO/Pt CO2 formationb
103TOF/s21
0
5
10
15
20
a
—
—
2.0
—
—
0.70 0.52
0.45 0.41
0.38 0.40
0.25 0.30
0.19 0.27
1.5
11
33
29
16
Fig. 2 Reaction temperature dependence of CO conversion in CO
oxidation (without H2) and preferential oxidation of CO over K–Pt/Al2O3
(K/Pt 5 0 and 10). K/Pt 5 0 (%,n), K/Pt 5 10 (&,m,#). Reaction
conditions: 0.2 vol% CO, 0.2 vol% O2 and 75 vol% H2 balanced with He
(%,&), 0.2 vol% CO and 0.2 vol% O2 balanced with He (n,m), 0.2 vol%
CO, 0.2 vol% O2, 75 vol% H2 and 5 vol% H2O balanced with He (#).
b
Irreversible adsorption amount. TOF of CO2 formation at 353 K
is calculated on the basis of CO/Pt. Reaction conditions are the
same in Fig. 1.
1430 | Chem. Commun., 2005, 1429–1431
This journal is ß The Royal Society of Chemistry 2005

View Article Online
From the observed behavior, it is thought that the additive effect
of potassium on preferential oxidation of CO has two aspects: one
is the effect on the CO + O2 oxidation and the other is the effect of
H2 presence. According to the previous report,14,15 the promoting
effect of alkali ion on the O2 adsorption can be explained by
changing the binding energies of adsorbed oxygen, and it can
promote CO oxidation. On the other hand, the effect of H2
presence on CO oxidation can not be explained by water gas shift
reaction (CO + H2O A CO2 + H2), which is induced by the H2O
production from H2 oxidation. This is because Pt/Al2O3 and K–Pt/
Al2O3 exhibited very low activity of water gas shift reaction. One
interpretation can be due to the role of the hydroxyl group (OH),
which can be formed from H2 and O2. It has been reported that
the OH group can promote CO oxidation on Pt (111).15,16 In fact,
the effect of H2 presence on CO oxidation over Pt/Al2O3 was
observed in low temperature range, although it was not so
significant. In contrast, the effect over K–Pt/Al2O3 was much more
remarkable. At present, it is thought that the coverage of OH
group can be increased over K–Pt/Al2O3 due to the interaction
with potassium species, which covers Pt metal surface. Another
possibility is that H2O2, which can be formed from the reaction
between H2 and O2, can promote the preferential CO oxidation.
Promoting mechanism of H2 presence over K–Pt/Al2O3 is not
clear at present, and further investigation is necessary for the
elucidation of this mechanism. In terms of practical case, the effect
of the presence of steam in the reactant gas is important. In fact,
under our basic feeding conditions (0.2 vol% CO, 0.2 vol% O2 and
75 vol% H2 balanced with He) as shown in Fig. 1, about 0.1%
steam was formed when CO and O2 conversions are almost 100%.
Under these conditions, K–Pt/Al2O3 was very stable. This suggests
that the effect of low steam pressure is small. In addition, we
investigated the effect of higher steam pressure (5 vol% H2O), and
the result is also shown in Fig. 2. The addition of 5 vol% steam to
CO + O2 + H2 decreased the CO conversion level over K–Pt/Al2O3
(K/Pt 5 10). The temperature, where CO conversion reached 90%,
became 40 K higher by the addition of steam. This indicates that
high concentration of steam can decrease the catalytic activity of
preferential oxidation of CO. On the other hand, this poisoning
effect of steam was reversible, and the high activity was completely
reproduced in the reaction when the steam addition stopped. This
indicates that the catalyst structure was very stable even after
exposing to steam. Although the details are not shown here, the
additive effect of steam is small in CO + H2 + O2 over Pt/Al2O3.
The important point is that the effect of potassium is very
remarkable even under the presence of steam with high pressure.
Yuji Minemura,a Shin-ichi Ito,a Toshihiro Miyao,b Shuichi Naito,b
Keiichi Tomishigea and Kimio Kunimori*a
aInstitute of Materials Science, University of Tsukuba, 1-1-1 Tennodai,
Tsukuba, Ibaraki, 305-8573, Japan.
E-mail: kunimori@ims.tsukuba.ac.jp; Fax: +81 29 855 7440;
Tel: +81 29 853 5026
bDepartment of Applied Chemistry, Faculty of Engineering, Kanagawa
University, 3-27-1, Rokkakubashi, Kanagawa-ku, Yokohama, 221-8686,
Japan
Notes and references
1 B. Ho¨hlein, M. Boe, J. Bøgild-Hansen, P. Bro¨ckerhoff, G. Colsman,
B. Emonts, R. Menzer and E. Riedel, J. Power Sources, 1996, 61, 143.
2 B. Rohland and V. Plzak, J. Power Sources, 1999, 84, 183.
3 M. Iwase and S. Kawatsu, in Proton Conducting Membrane Fuel Cells 1,
ed. S. Gottesfeld, G. Halpert and A. Landgrebe, Electrochemical
Society, Pennington, NJ, 1995, p. 12.
4 A. Manasilp and E. Gulari, Appl. Catal. B, 2002, 37, 17.
5 H. Igarashi, H. Uchida, M. Suzuki, Y. Sasaki and M. Watanabe, Appl.
Catal. A, 1997, 159, 159.
6 I. H. Son and A. M. Lane, Catal. Lett., 2001, 76, 3.
7 O. Korotkikh and R. Farrauto, Catal. Today, 2000, 62, 249.
8 X. Liu, O. Korotkikh and R. Farrauto, Appl. Catal. A, 2002, 226, 293.
9 S. H. Oh and R. M. Sinkevitch, J. Catal, 1993, 142, 254.
10 S. Ito, T. Fujimori, K. Nagashima, K. Yuzaki and K. Kunimori, Catal.
Today, 2000, 57, 247.
11 E. Romeo-Pascual, A. Larrea, A. Monzo´n and R. D. Gonza´lez, J. Solid
State Chem., 2002, 168, 343.
12 H. Tanaka, S. Ito, S. Kameoka, K. Tomishige and K. Kunimori, Appl.
Catal. A, 2003, 250, 255.
13 S. Ito, H. Tanaka, Y. Minemura, S. Kameoka, K. Tomishige and
K. Kunimori, Appl. Catal. A, 2004, 273, 295.
14 N. Pavlenko, P. P. Kostrobij, Y. Suchorski and R. Imbihl, Surf. Sci.,
2001, 489, 29.
15 I. N. Yakovkin, V. I. Chernyi and A. G. Naumovets, Surf. Sci., 1999,
442, 81.
16 J. Bergeld, B. Kasemo and D. V. Chakarov, Surf. Sci., 2001, 495,
L815.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 1429–1431 | 1431