Hydrogen production by partial oxidation of methanol over Au/CuO/ZnO catalysts

Article abstract of DOI:10.1016/j.molcata.2007.07.002

Hydrogen production by partial oxidation of methanol (POM) (CH₃OH + 0.5O₂ → 2H₂ + CO₂) was examined over Au/CuO/ZnO (Au 3 wt.%, Cu 37 wt.% and Zn 60 wt.%) catalysts, prepared by the co-precipitation method. The activity of Au/CuO/ZnO catalysts has been compared with that of CuO/ZnO (Cu 40 wt.% and Zn 60 wt.%) catalysts. The catalysts were characterized by XRD, TEM and TPR analyses. The Au/CuO/ZnO catalysts are more active and exhibit higher hydrogen selectively with smaller amount of CO compared to the CuO/ZnO catalysts. The enhanced activity of Au/CuO/ZnO catalyst is due to the strong interaction between Au and CuO species as evidenced by TPR analysis. The influence of calcination temperature on the catalytic activity of Au/CuO/ZnO catalyst depends on crystalline phases of support and particle size of Au and CuO present on the catalysts. The optimum calcination temperature is 573 K. The catalytic performance at various reaction temperatures shows that at 448 K the reaction sets on with O₂ completely consumed. A high hydrogen selectivity of 97.7% is attained at 523 K with 100% methanol conversion. The undesired by-product, CO is formed in very small amount throughout the temperature range studied.

Full text of DOI:10.1016/j.molcata.2007.07.002

Journal of Molecular Catalysis A: Chemical 276 (2007) 184–190
Hydrogen production by partial oxidation of methanol
over Au/CuO/ZnO catalysts
∗
Hsien-Chang Yang, Feg-Wen Chang , L. Selva Roselin
Department of Chemical and Materials Engineering, National Central University, Chungli 32001, Taiwan
Received 7 August 2006; received in revised form 25 June 2007; accepted 3 July 2007
Available online 10 July 2007
Abstract
Hydrogen production by partial oxidation of methanol (POM) (CH
Cu 37 wt.% and Zn 60 wt.%) catalysts, prepared by the co-precipitation method. The activity of Au/CuO/ZnO catalysts has been compared with
that of CuO/ZnO (Cu 40 wt.% and Zn 60 wt.%) catalysts. The catalysts were characterized by XRD, TEM and TPR analyses. The Au/CuO/ZnO
catalysts are more active and exhibit higher hydrogen selectively with smaller amount of CO compared to the CuO/ZnO catalysts. The enhanced
activity of Au/CuO/ZnO catalyst is due to the strong interaction between Au and CuO species as evidenced by TPR analysis. The influence of
calcination temperature on the catalytic activity of Au/CuO/ZnO catalyst depends on crystalline phases of support and particle size of Au and
CuO present on the catalysts. The optimum calcination temperature is 573 K. The catalytic performance at various reaction temperatures shows
3
OH + 0.5O
2
→ 2H
2 2
+ CO ) was examined over Au/CuO/ZnO (Au 3 wt.%,
that at 448 K the reaction sets on with O
conversion. The undesired by-product, CO is formed in very small amount throughout the temperature range studied.
2007 Elsevier B.V. All rights reserved.
2
completely consumed. A high hydrogen selectivity of 97.7% is attained at 523 K with 100% methanol
©
Keywords: Gold catalyst; Copper oxide; Zinc oxide; Partial oxidation of methanol; Hydrogen
1
. Introduction
activity and stability of the Au/TiO2 system is not high enough,
modification of the catalyst for improved activity and stabil-
ity is needed. Our recent report has proved that the catalytic
activity and stability of supported gold catalysts for hydro-
gen generation is improved by using a composite support [11].
Even though the catalytic activity and stability of the gold
catalyst was improved by using the composite support, fur-
ther improvement in the activity is necessary. In the present
work, the catalytic activity of Au/CuO/ZnO catalysts has been
studied for the formation of hydrogen by POM using oxygen
as oxidant. The catalyst characterizations included are X-ray
diffraction (XRD), transmission electron microscopy (TEM)
and temperature-programmed reduction (TPR). Attention has
been paid to the optimization of calcination temperature and
reaction temperature in order to achieve complete conversion of
methanol for the selective formation of hydrogen.
Hydrogen production by partial oxidation of methanol
(
(
POM) is of great interest for the development of fuel cell
proton exchange membrane fuel cell, PEMFC). Hydrogen-
powered fuel cells generate electricity much more efficiently
than fossil fuel without discharging any polluting by-products
such as NOx, SOx, etc. [1,2]. Previous studies on POM to pro-
duce hydrogen showed that Cu-based catalysts such as Cu/ZnO
and Cu/ZnO/Al2O3 were active [3–7]. Unfortunately, these cat-
alysts produce considerable amount of carbon monoxide as a
by-product. CO is a poison to fuel cell anodes. If the carbon
monoxide level exceeds a few ppm, the precious metal-based
anode electrocatalyst of the fuel cell is deactivated. The discov-
ery of the exceptionally high activity of supported Au catalysts
for low temperature CO oxidation [8,9] aroused interest in test-
ing these catalysts for POM to produce hydrogen. Our previous
investigation on POM showed that Au/TiO2 catalyst was active
for selective formation of hydrogen [10]. Since the catalytic
2. Experimental
2.1. Catalyst preparation
∗
The weight percentage of each metal in the most active
Corresponding author. Tel.: +886 3 4227151x34202; fax: +886 3 4252296.
E-mail address: fwchang@cc.ncu.edu.tw (F.-W. Chang).
CuO/ZnO catalysts for POM reported earlier is 40 wt.% Cu and
1
381-1169/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2007.07.002

H.-C. Yang et al. / Journal of Molecular Catalysis A: Chemical 276 (2007) 184–190
185
6
0 wt.% Zn [3]. Therefore, the CuO/ZnO catalysts with the same
composition were prepared in this study. For Au/CuO/ZnO cat-
alysts, the weight percentages of Au, Cu and Zn were fixed
as 3, 37 and 60 wt.%, respectively. The Au/CuO/ZnO catalysts
were prepared by the co-precipitation method [9]. Typically,
an aqueous solution of Cu(NO3)2·2.5H2O (0.25 M), Zn(NO3)2
(
0.25 M), and HAuCl4 (0.1 M) (Fluka) were mixed together in
a volume ratio of 1.5:2.4:0.1, with stirring, at 343 K. Aqueous
Na2CO3 (0.1 M) was added dropwise to the mixed nitrate solu-
tion until pH 8.5 was attained. The precipitate formed in this
process was aged at 343 K for 2.5 h. The precipitate was recov-
ered by filtration and washed with hot deionized water (343 K)
−
+
until complete elimination of Cl and Na ions. Then the sample
was dried in air at 373 K for 24 h to give a material designated as
the catalyst precursor, which was subsequently calcined in air at
different temperatures for 3 h. CuO/ZnO catalyst was prepared
using similar preparation procedures.
Fig. 1. XRD patterns of Au/CuO/ZnO catalysts before calcination, after cal-
cination at different temperatures and after POM reaction: (a) uncalcined; (b)
573 K; (c) 673 K; (d) after POM reaction at 523 K for 2 h.
2
.2. Characterization
The total flow was 60 ml/min with the O2/CH3OH molar ratio
of 0.5. The reaction products were analyzed on-line using two
gas chromatographs (GC) equipped with thermal conductivity
detector and porapak Q and carbosieve S-II columns.
X-ray powder diffraction (XRD) patterns of the catalyst sam-
ples were obtained in an X-ray diffractometer (Bruker, Model
D8A) operated at 40 kV and 40 mA using Cu K␣ radiation with
◦
◦
◦ −1
˚
a wavelength of 1.5406 A, from 5 to 80 at a rate of 0.05 s .
Transmission electron microscopy (TEM) studies were per-
formed on a JEOL TEM-2000FXII instrument operated at
3. Results and discussion
1
60 kV. To obtain suitable samples for TEM characterization,
the powders were dispersed in ethanol by ultrasonication. A
drop of the solution was then deposited on to a thin carbon film
supported on a copper microgrid and left to dry at 333 K. For
each catalyst, about 100 Au particles were measured in order to
determine a statistically justified average particle size and size
distribution.
Temperature-programmed reduction (TPR) of the catalyst
was performed in a U-shaped micro-reactor made of quartz,
surrounded with a furnace controlled by a programmed heating
system. Prior to the TPR experiment, 20 mg of the catalyst sam-
ple was pretreated under flowing Ar (20 ml/min) at 523 K for
3.1. XRD
XRDwasusedtoidentifythecrystallinephasespresentonthe
Au/CuO/ZnO catalysts before and after calcination at different
temperatures. Fig. 1 shows the XRD patterns of Au/CuO/ZnO
catalysts before and after calcination at 573 and 673 K. The
pattern of the Au/CuO/ZnO catalyst calcined at 573 K after
exposure to POM reaction at 523 K for 2 h is also inserted.
The XRD of the uncalcined sample (Fig. 1a) exhibited the
presence of hydrozincite (Zn (CO3)2(OH) ) and aurichalcite
(Zn3Cu2(CO3)2(OH) )compounds[3]. Forexample, theintense
5
6
6
4
5 min. After the pretreatment, the sample was cooled to room
lines at 2θ = 13.0, 24.3 and 47.5 are characteristic of hydroz-
◦
◦
◦
temperature. A reducing gas composed of 5% H2 plus 95% Ar
was employed at a flow rate of 20 ml/min, with a heating ramp
of 10 K/min from 373 to 773 K. The amount of the consumed
H2 was determined by a thermal conductivity detector (TCD).
incite and those at 2θ = 34.1 , 41.9 and 50.1 are characteristic
of aurichalcite. The sample calcined at 573 K undergo ther-
mal decomposition, resulting in the formation of a mixture of
poorly crystallized CuO and ZnO phases as observed by the
diminished peak intensities of the hydrozincite and aurichalcite
compounds and appearance of peaks corresponding to ZnO and
CuO (Fig. 1b). The crystallinity of the ZnO and CuO phases
increased with increasing calcination temperature to 673 K
(Fig. 1c). Fig. 1d illustrates the XRD patterns of Au/CuO/ZnO
catalyst calcined at 573 K after POM at 523 K for 2 h. Upon reac-
tion with methanol and oxygen at 523 K, the line characteristic
2
.3. Catalytic activity
Partial oxidation of methanol (POM) was carried out using
an apparatus, which has been described in detail elsewhere [10].
The reaction was carried out at atmospheric pressure and at
temperatures between 423 and 548 K, using a U-shaped microre-
actor made of quartz (i.d. = 4 mm). The reactor was located in
a temperature-programmable furnace with a type K thermocou-
ple placed in the center of the catalyst bed. Typically, 20 mg of
catalyst was used in each experiment. Methanol was fed into
the pre-heater by means of a Cole-Parmer liquid pump (Model:
0
◦
of Cu appears at 2θ = 43.0 . This can be interpreted to mean
that CuO is reduced to metallic Cu during the reaction. Wang
et al. [12] have also observed the reduction of CuO species in
Cu/SiO2 catalysts to metallic copper in a similar type of reaction.
As seen from Fig. 1, for all the samples, the XRD analysis did
◦
7
7120-30) at a rate of 0.924 ml/h. The oxygen and argon (dilu-
not reveal the presences of any oxidic Au species (2θ = 25.5 ,
30.2 and 32.5 ) or metallic Au species (2θ = 38.2 , 44.4 and
◦
◦
◦
◦
ent) flows were adjusted by Brooks 5850E mass flow controllers.

1
86
H.-C. Yang et al. / Journal of Molecular Catalysis A: Chemical 276 (2007) 184–190
◦
6
4.5 ), suggesting that gold particles are highly dispersed on the
after exposure to POM. Fig. 2a shows the TEM image of the
Au/CuO/ZnO catalyst before calcination. During the drying
procedure, aurichalcite and the gold precursor thermally decom-
posed to a small extent to form CuO and metallic Au. Deposited
gold particle appear as small black spots in the TEM photo-
graph and are uniformly dispersed on the support. The gold and
copper crystallites are sphere in shape with an average diameter
of 2.5 and 6.2 nm, respectively. As shown in Fig. 2b, the TEM
image of the Au/CuO/ZnO catalyst calcined at 573 K indicates
an increase in the number of gold and copper crystallites. Zhang
et al. [14] reported that in gold catalyst supported on zinc oxide,
gold oxide species are decomposed to metallic gold when the
sample was calcined at 513 K. Therefore during calcination, the
gold precursor decomposed to metallic gold, and the hydroz-
incite and aurichalcite were decomposed to form CuO. In the
figure, the large particles are CuO particles, which exhibited an
surface, that the Au peaks are overlapped by hydrozincite and/or
CuO, or that the corresponding gold particles are too small to
be detected by the instrument. According to Chang et al. [13],
the limitations of XRD for gold crystallites correspond to a par-
ticle size smaller than 5 nm. TEM technique was subsequently
employed to confirm the Au particle size in the catalysts.
3
.2. TEM
The particle size of gold and copper species and their level of
dispersion on the support in Au/CuO/ZnO catalysts before and
after calcination at different temperatures were examined by
transmission electron microscope (TEM) study. Fig. 2 presents
the TEM pictures of uncalcined and calcined Au/CuO/ZnO cat-
alyst samples (573 and 673 K) and the sample calcined at 573 K
Fig. 2. TEM images of Au/CuO/ZnO catalysts before calcination, after calcination at different temperatures and after POM reaction: (a) uncalcined (200k×); (b)
73 K (300k×); (c) 673 K (200k×); (d) after POM reaction at 523 K for 2 h (150k×).
5

H.-C. Yang et al. / Journal of Molecular Catalysis A: Chemical 276 (2007) 184–190
187
average diameter of 9.5 nm. The small gold particles appear to
be fairly homogenously distributed over the support and have
an average diameter of 3.4 nm. This result clearly indicates that
at higher calcination temperature both the size and the number
of gold particles are increased. Fig. 2c shows the TEM image
of the Au/CuO/ZnO catalyst calcined at 673 K. When the cata-
lyst is calcined at 673 K, the gold and CuO particles grow to an
average size of 4.6 and 13 nm, respectively. At this temperature,
aggregation of copper particles becomes larger due to sintering
and is accompanied by changes in their shape from a spherical
to irregular shape. Therefore, at higher calcination temperature,
gold and copper particles sinters, and so, the number of gold
and copper crystallites is decreased. It implies that sintering
of the catalyst takes place at 673 K; consequently, the number
of active sites is decreased. Fig. 2d shows the TEM image of
Au/CuO/ZnO catalyst calcined at 573 K after exposure to POM
at 523 K for 2 h. The particle size of Au and Cu are increased
from 3.4 to 5.4 nm and from 9.5 to 27 nm, respectively. The
POM reaction results in the increase in the size of Au and Cu
crystallites.
perature for CuO in CuO/ZnO and Au/CuO/ZnO are different.
The reduction peak of CuO in the Au/CuO/ZnO sample shifts
to a lower temperature compared with the CuO/ZnO sample
(Fig. 3b and c). This shows that Au promotes CuO reduction,
which occurs at lower temperatures, in accordance with pre-
vious work [19]. Cabrera et al. [20] pointed out that metallic
Pd exhibited a certain promotion effect on the CuO reduction
in palladium–copper–zinc catalysts. The enhanced reducibil-
ity by the metal in the catalysts is reported in other catalysts
[21–23]. Scir e` et al. [22] have reported that reduction of Fe2O3
enhanced by Au in Au/Fe2O3 catalysts. The enhanced reducibil-
ity of Fe2O3 has been explained in terms of ability to weaken
the Fe–O bond. The metal in Fe2O3 could lead to decrease in the
strength of the Fe–O bond located nearby the metal atom [23].
Similar phenomenon is occurring here; the Cu–O bond is weak-
ened by the presence of metal Au. Therefore, we may propose
that there is some sort of interaction between Au and CuO in
the Au/CuO/ZnO catalysts, which can enhance the reducibility
of CuO.
3.4. Catalytic activity
3
.3. TPR
The catalytic activity and product distribution in the partial
Fig. 3 illustrates the TPR patterns of Au/ZnO, CuO/ZnO
oxidation of methanol (POM) over Au/CuO/ZnO catalysts were
studied with an O2/CH3OH molar ratio of 0.5. Hydrogen and
carbon dioxide were the main products, while water and carbon
monoxide were the by-products. Methanol conversion, hydro-
gen and carbon monoxide selectivity are defined, respectively,
as follows:
and Au/CuO/ZnO catalysts calcined at 573 K. The Au/ZnO
sample (Fig. 3a) does not exhibit any signal which indicates
that gold is present only in a metallic state after calcination at
73 K, in agreement with previous reports [14]. The TPR pat-
tern of CuO/ZnO (Fig. 3b) shows a major peak at 500 K along
with two shoulder peaks at 460 and 525 K. The major peak at
5
5
00 K corresponds to the reduction of CuO to metallic copper.
CH3OH conversion (%)
The low temperature shoulder peak at 460 K is assigned to the
reduction of CuO to Cu and the high temperature shoulder
peak at 525 K is assigned to the reduction of Cu to metallic
copper [15–17]. The TPR pattern of the Au/CuO/ZnO sample
+
= (moles of methanol consumed/moles of methanol fed)
+
×
100,
H2 selectivity (%)
(moles of H2 produced/moles of methanol consumed× 2)
100,
CO selectivity (%)
(moles of CO produced/moles of methanol consumed)
100.
(
Fig. 3c) exhibits a broad peak with two unresolved signals at
86 and 500 K. The two TPR signals are related to the two step
reduction (Cu(II) → Cu(I) → Cu(0)) as suggested by Fang et al.
18]. It is noteworthy that both the shape and reduction tem-
4
=
×
[
=
×
Fig. 4 shows the comparison of catalytic activity of CuO/ZnO
and Au/CuO/ZnO catalysts for partial oxidation of methanol
(POM) at 523 K. These two catalysts showed similar high
methanol conversion (100%). However, they differ in their prod-
uct distribution. The Au/CuO/ZnO catalysts displayed higher
hydrogen selectivity with lower carbon monoxide selectivity
compared to that over CuO/ZnO catalysts. The hydrogen selec-
tivity is 97.7 and 84.6%, respectively for Au/CuO/ZnO and
CuO/ZnOcatalysts. ItisinterestingtonotethattheAu/CuO/ZnO
catalyst produced negligible amount of CO. However, signif-
icant amount of CO was observed over CuO/ZnO catalysts,
being consistent with the results reported in the previous stud-
ies on supported copper catalysts [3,24,25]. The CO selectivity
Fig. 3. TPR curves of Au/ZnO, CuO/ZnO and Au/CuO/ZnO catalysts calcined
at 573 K: (a) Au/ZnO catalyst; (b) CuO/ZnO catalyst; (c) Au/CuO/ZnO catalyst.

1
88
H.-C. Yang et al. / Journal of Molecular Catalysis A: Chemical 276 (2007) 184–190
Fig. 5. Effect of calcination temperature on methanol conversion, hydrogen
selectivity and carbon monoxide selectivity for partial oxidation of methanol
over Au/CuO/ZnO catalysts (reaction temperature, 523 K).
Fig. 4. Catalytic performance of CuO/ZnO and Au/CuO/ZnO catalysts for
methanol conversion, hydrogen selectivity and carbon monoxide selectivity for
partial oxidation of methanol (calcination temperature, 573 K; reaction temper-
ature, 523 K; reaction time, 10 min).
and CuO particles. The mean particle size of gold increases
from 3.4 to 4.6 nm and CuO size increases from 9.5 to 13 nm
when the calcination temperature increases from 573 to 673 K
(Fig. 2b and c). High temperature heating results in sintering
of gold and copper particles and large particles are formed by
coalesces of smaller particles. Highly dispersed Au particles are
more active for methanol oxidation than for hydrogen oxidation,
while large Au particles are more active for hydrogen oxida-
tion [27]. The carbon monoxide selectivity over Au/CuO/ZnO
catalysts decreases in the following order: uncalcined > cal-
cined at 673 K > calcined at 573 K. The performance of the gold
supported zinc oxide catalysts showed that the formation of
hydrozincite led to the decrease in activity, because the inter-
action of CO and oxygen species might be suppressed by the
carbonate-like species formed during reaction [12,14]. During
POM, the carbon monoxide formed by methanol decomposi-
tion and/or reverse water gas shift (RWGS) is subsequently
oxidized to carbon dioxide, since these catalysts are reported
to be highly active for CO oxidation [9]. However, this pro-
cess has been restricted by the presence of hydrozincite [26].
This causes formation of larger amount of carbon monoxide
over the uncalcined catalyst sample. Moreover, for CO oxidation
using supported gold catalysts, the particle size of gold plays an
important role, smaller gold particles being highly active. Thus,
the influence of the calcination effect on hydrogen selectivity
depends on crystalline phases of support and particle size of Au
present on the Au/CuO/ZnO catalysts. This leads to a conclu-
sion that the Au/CuO/ZnO catalyst calcined at 573 K provides
high hydrogen selectivity with low carbon monoxide selectivity.
It is interesting to note that these catalysts are stable during cat-
alytic tests. In our previous study on POM, we have inferred that
the deactivation of Au/TiO2 catalyst for hydrogen production is
explained in terms of particle growth of Au during the course
of the reaction [10]. In the present study, the Au particle size in
the Au/CuO/ZnO catalyst increases from 3.4 to 5.4 nm after 2 h
catalytic test (Fig. 2b and d). XRD analyses of the catalysts after
over Au/CuO/ZnO and CuO/ZnO catalysts is 1.5 and 10.7%,
respectively. In our recent studies on POM, we have shown that
Au/TiO2 and Au/TiO2–Fe2O3 catalysts are active in produc-
ing hydrogen with small amount of CO [10,11]. Therefore, the
enhanced activity of Au/CuO/ZnO catalyst to produce hydrogen
with negligible amount of CO is due to the strong interaction
between Au and CuO species as evidenced by TPR studies
(Fig. 3b and c). In this catalyst, partial oxidation of methanol
was mainly found to take place leading to high hydrogen and
carbon dioxide selectivities. Moreover, Au/CuO/ZnO catalysts
are reported to be an efficient catalyst for CO oxidation [9].
Therefore, in Au/CuO/ZnO catalysts, the undesired by-product
CO formed even in small amount during POM through methanol
decomposition and/or reverse water gas shift reactions is read-
ily oxidized to carbon dioxide. Since Au/CuO/ZnO catalysts
give higher hydrogen selectivity with negligible amount of CO
compared to CuO/ZnO catalysts, a systematic study has been
undertaken using Au/CuO/ZnO catalysts to optimize calcina-
tion temperature and reaction temperature for POM to produce
hydrogen.
Fig. 5 shows the effect of calcination temperature on
methanol conversion, hydrogen selectivity and carbon monox-
ide selectivity for partial oxidation of methanol at 523 K. All
the catalysts show similar activity towards methanol conversion.
As can be seen from the figure, methanol conversion is 100%
and independent of reaction time. But it is clear that calcina-
tion conditions significantly influence the performance of the
catalyst towards hydrogen formation. The hydrogen selectivity
over Au/CuO/ZnO catalysts decreases in the following order:
calcined at 573 K > calcined at 673 K > uncalcined. Thus, the
optimum calcination temperature is 573 K. The lower activity
of the uncalcined catalyst is due to the presence of hydrozincite
as evidenced from XRD analysis (Fig. 1a). It was reported that
the amount of active gold species available on the catalyst sur-
face could be decreased in the presence of hydrozincite phase
0
reaction show the presence of new species Cu (Fig. 1d). This
[
26]. The lower activity of the catalyst at higher calcination tem-
can be interpreted to mean that CuO is reduced during the reac-
tion. Metallic copper is reported as active species for the POM
peratures (673 K) may be due to the presence of larger gold

H.-C. Yang et al. / Journal of Molecular Catalysis A: Chemical 276 (2007) 184–190
189
CO + H2O → CO2 + H2
(4)
According to this reaction sequence, if the entire CO formed in
the MD is consumed in the WGS, coincidence in the proportion
of products by SRM and by MD–WGS could be expected. The
MD–WGS scheme has been proposed by several authors for
SRM reaction over copper catalysts [28,29]. MD and reverse
water gas shift (RWGS) (Eq. (5)) is believed to occur and pro-
duce CO when no more oxygen and water are available [5]:
CO2 + H2 → CO + H2O
(5)
Nevertheless, the CO selectivity is very low compared to that
observed over supported Cu catalysts [3,4]. The CO formed by
MD and RWGS can be converted to CO2 by WGS (Eq. (4))
and/or by CO oxidation (Eq. (6)):
Fig. 6. Effect of reaction temperature on methanol conversion, oxygen conver-
sion, hydrogen selectivity and carbon monoxide selectivity for partial oxidation
of methanol over Au/CuO/ZnO catalysts (calcination temperature, 573 K; reac-
tion time, 10 min).
CO + 0.5O2 → CO2
(6)
The presence of a certain proportion of CO in the reaction prod-
uct at high temperature suggests that the rate of MD is high
when the reaction temperature is above 523 K. It is important to
note that, in the present study, complete conversion of methanol
with very high hydrogen selectivity (97.7%) with small amount
of carbon monoxide (∼1.5%) is detected at 523 K. In our pre-
vious studies we have shown that a Au/TiO2 catalyst produced
a maximum of 35% hydrogen selectivity with 100% methanol
conversion at 583 K and a Au/TiO2–Fe2O3 catalyst produced
reaction [3,12]. Therefore the Au/CuO/ZnO catalyst is consid-
ered as sustainable active catalyst for formation of hydrogen by
POM.
The effect of reaction temperature on the catalytic perfor-
mance of the Au/CuO/ZnO catalysts is depicted in Fig. 6. It can
be observed that, the reaction sets on at 448 K. The O2 conver-
sion reaches 100% at 448 K, methanol conversion reaches 100%
at 498 K and hydrogen selectivity reaches 97.7% at 523 K. CO
selectivity is very low throughout the temperature range studied;
the CO selectivity increases from 1.1 to 3.2%, when the reac-
tion temperature increases from 448 to 548 K. The conversion
and selectivity data illustrate the participation of several reac-
tions during POM. During POM, methanol quickly consumes
excess oxygen leading to highly exothermic methanol combus-
tion (MC) reaction (Eq. (1)) to occur and produce water and
carbon dioxide:
7
5
6% hydrogen selectivity with 95% methanol conversion at
48 K [10,11]. Therefore, the current results clearly indicate
that improved catalytic performance can be achieved by using
Au/CuO/ZnO catalysts for hydrogen production by partial oxi-
dation of methanol.
4. Conclusions
CH3OH + 1.5O2 → 2H2O + CO2
(1)
Production of hydrogen by partial oxidation of methanol
was studied over Au/CuO/ZnO and CuO/ZnO catalysts. The
Au/CuO/ZnO catalyst is more active and exhibited higher hydro-
gen selectively with smaller amount of CO than the CuO/ZnO
catalysts. The enhanced activity of Au/CuO/ZnO catalyst is due
to the strong interaction between Au and CuO species. The cat-
alytic activity strongly depends on the calcination and reaction
temperatures. The optimum calcination temperature for hydro-
gen selectivity is 573 K. The influence of calcination on the
activity suggests that the activity depends on the crystalline
phases of support and particle size of Au and CuO present on the
catalysts. The effect of reaction temperature on the performance
of the Au/CuO/ZnO catalysts was studied in the temperature
range of 423–548 K. Both methanol conversion and hydrogen
selectivity increase with increasing reaction temperature. Com-
plete conversion of methanol is observed above 498 K. A high
hydrogen selectivity of 97.7% with very low CO selectivity of
At 448 K, water selectivity was 18.1%. With increasing reac-
tion temperature, water selectivity is decreased with consequent
increaseinhydrogenselectivity. Thiscanbeexplainedbyassum-
ing that water formed by MC is consumed by steam reforming
of methanol (SRM) and produces hydrogen and carbon dioxide
(
Eq. (2)):
CH3OH + H2O → 3H2 + CO2
(2)
It is known that copper catalysts perform a variety of catalytic
functions depending on its oxidation state. The metallic copper
is active for SRM, while Cu2+ as CuO promotes MC producing
mainly water and carbon dioxide [7]. At low reaction tempera-
ture copper in Au/CuO/ZnO catalyst is present as copper oxide.
With increasing reaction temperature some of the copper species
are reduced to metallic copper. At 523 K, the selectivity towards
hydrogen formation is 97.7%, which suggests that water is con-
sumed by SRM. The SRM (Eq. (2)) can be considered as a
combination of methanol decomposition (MD) (Eq. (3)) and
water gas shift (WGS) (Eq. (4)):
1.5% is observed at 523 K. The present study suggests that the
Au/CuO/ZnO catalyst is an excellent catalyst for selective and
effective production of hydrogen, making it possible to utilize
this catalyst in the application of proton exchange membrane
fuel cell.
CH3OH → CO + 2H2
(3)

1
90
H.-C. Yang et al. / Journal of Molecular Catalysis A: Chemical 276 (2007) 184–190
[14] J. Zhang, Y. Wang, B. Chen, C. Li, D. Wu, X. Wang, Energy Convers.
Acknowledgements
Manage. 44 (2003) 1805.
[
[
[
15] E.D. Guerreiro, O.F. Gorriz, J.B. Rivarola, L.A. Arrfia, Appl. Catal. A 165
The authors express thanks to the Ministry of Economical
Affairs of Taiwan for its financial support under contract number
(
1997) 259.
16] B. Lindstr o¨ m, L.J. Pettersson, P. Govind Menon, Appl. Catal. A 234 (2002)
11.
9
3-EC-17-A-09-S1-022.
1
17] H. Oguchia, T. Nishiguchia, T. Matsumotoa, H. Kanaia, K. Utania, Y.
Matsumurab, S. Imamura, Appl. Catal. A 281 (2005) 69.
18] X. Fang, S. Yao, Z. Qing, F. Li, Appl. Catal. A 161 (1997) 129.
19] H.L. Lian, M.J. Jia, W.C. Pan, W.X. Zhang, D.Z. Jiang, Chem. Res. Chin.
U 22 (2006) 99.
References
[
[
[
[
1] B. Lindner, K. Sjomstrom, Fuel 63 (1984) 1485.
2] J. Appleby, F.R. Foulkes, Fuel Cell Handbook, Van Nostrand Reinhold,
New York, 1989, p. 177.
[
[
20] I.M. Cabrera, M.L. Granados, J.L.G. Fierro, Catal. Lett. 79 (2002) 165.
21] A. Venugopal, J. Aluha, D. Mogano, M.S. Scurrell, Appl. Catal. A 245
[
3] L. Alejo, R. Lago, M.A. Pe n˜ a, J.L.G. Fierro, Appl. Catal. A 162 (1997)
(
2003) 149.
22] S. Scir e` , S. Minic o` , C. Crisafulli, S. Ganvango, Catal. Commun. 2 (2001)
29.
2
81.
[
[
[
[
[
[
[
[
[
4] J. Agrell, K. Hasselbo, K. Jansson, S.G. J a¨ r a˚ s, M. Boutonnet, Appl. Catal.
A 211 (2001) 239.
2
23] S. Minic o` , S. Scir e` , C. Crisafulli, R. Maggiore, S. Ganvango, Appl. Catal.
B 28 (2000) 245.
24] J. Agrell, M. Boutonnet, I.M. Cabrera, J.L.G. Fierro, Appl. Catal. A 253
[5] J. Agrell, M. Boutonnet, J.L.G. Fierro, Appl. Catal. A 253 (2003) 213.
[
6] R.M. Navarro, M.A. Pe n˜ a, J.L.G. Fierro, J. Catal. 212 (2002) 112.
[
7] J. Agrell, H. Birgersson, M. Boutonnet, I. Meli a´ n-Cabrera, R.M. Navarro,
J.L.G. Fierro, J. Catal. 219 (2003) 389.
(
2003) 201.
25] D. Wang, Z. Haob, D. Chenga, X. Shi, C. Hu, J. Mol. Catal. A 200 (2003)
29.
26] G.Y. Wang, W.X. Zhang, H.L. Lian, D.Z. Jiang, T.H. Wu, Appl. Catal. A
39 (2003) 1.
[
[
8] M. Haruta, N. Yamada, T. Kobayashi, S. Iijima, J. Catal. 15 (1989) 301.
9] G.J. Hutchings, M.R.H. Siddiqui, A. Burrows, C.J. Kiely, R. Whyman, J.
Chem. Soc., Faraday Trans. 93 (1997) 187.
2
2
[
[
10] F.W. Chang, H.Y. Yu, L.S. Roselin, H.C. Yang, Appl. Catal. A 290 (2005)
38.
11] F.W. Chang, H.Y. Yu, L.S. Roselin, H.C. Yang, T.C. Ou, Appl. Catal. A
02 (2006) 157.
27] T.V. Choudhary, C. Sivadinarayana, C.C. Chusuei, A.K. Datye, J.P. Fackler,
D.W. Goodman, J. Catal. 207 (2002) 247.
28] J.C. Amphlett, R.F. Mann, R.D. Weir, Can. J. Chem. Eng. 66 (1988)
1
3
950.
[
[
12] Z. Wang, W. Wang, G. Lu, Int. J. Hydrogen Energy 28 (2003) 151.
13] C.K. Chang, Y.J. Chen, C. Yeh, Appl. Catal. A 174 (1998) 13.
29] E. Santacesaria, S. Carra, Appl. Catal. 5 (1983) 345.