Effect of dopants on the performance of CuO–CeO 2 catalysts in methanol steam reforming Joan Papavasiliou a,b , George Avgouropoulos a , Theophilos Ioannides a, * a Foundation for Research and Technology-Hellas, Institute of Chemical Engineering and High Temperature Chemical Processes (FORTH/ICE-HT), P.O. Box 1414, GR-26504 Patras, Greece b Department of Chemical Engineering, University of Patras, GR-26504 Patras, Greece Received 13 February 2006; received in revised form 30 June 2006; accepted 10 July 2006 Available online 17 August 2006 Abstract Steam reforming of methanol was carried out over a series of doped CuO–CeO 2 catalysts prepared via the urea–nitrate combustion method. XRD analysis showed that at least part of the dopant cations enter the ceria lattice. The addition of various metal oxide dopants in the catalyst composition affected in a different way the catalytic performance towards H 2 production. Small amounts of oxides of Sm and Zn improved the performance of CuO–CeO 2 , while further addition of these oxides caused a decrease in catalyst activity. XPS analysis of Zn- and Sm-doped catalysts showed that increase of dopant loading leads to surface segregation of the dopant and decrease of copper oxide dispersion. The addition of oxides of La, Zr, Mg, Gd, Yor Ca lowered or had no effect on catalytic activity, but led to less CO in the reaction products. Noble-metal modified catalysts had slightly higher activity, but the CO selectivity was also significantly higher. # 2006 Elsevier B.V. All rights reserved. Keywords: Copper oxide; Cerium oxide; Hydrogen production; Methanol; Steam reforming; Combustion method 1. Introduction Solid polymer fuel cells (SPFCs), which consume hydrogen and oxygen to produce electricity, appear to be a clean and efficient energy solution for both mobile and stationary applications. The use of hydrogen for mobile applications is hindered by problems of storage, safety and refueling. Alternatively, H 2 can be produced onboard from various liquid fuels, such as methanol. Methanol is a strong candidate fuel, since it is readily available and can be catalytically converted into a H 2 -rich gas at moderate temperatures (200–300 8C). It has a high H/C ratio and no C–C bonds, hence minimizing the risk for coke formation. Production of hydrogen from methanol is also attractive because of the relatively low selectivity to byproducts, such as carbon monoxide and methane compared to alkane or higher alcohol reforming processes [1]. Moreover, methanol can be produced from renewable sources and, as a consequence, may be considered as a sustainable energy carrier [2]. Ultra-small methanol processors in combination with fuel cells have been proposed as Li-battery substitutes in demanding applications [3]. Copper-based catalysts, especially with the composition Cu–ZnO–(Al 2 O 3 ), have been widely used for generating hydrogen from methanol via the steam reforming (SRM) process [4–13]. The activity and CO selectivity are greatly dependent on catalyst morphology (high copper dispersion is desirable) and the redox properties of the catalyst [2,4,14]. For the same process we have recently examined copper– manganese spinel oxide catalysts, prepared via the combustion method [15]. These catalysts showed high catalytic activity and selectivity towards hydrogen production. On the other hand, cerium oxide has attracted much attention in environmental catalysis due to its high-oxygen storage capacity and the ability of fast transfer of bulk oxygen to its surface. Additionally, ceria can affect the valence state of various metal oxides. Prereduced, coprecipitated Cu/CeO 2 catalysts were found to be more active in the SRM process than the state-of-the-art Cu/Zn/Al 2 O 3 catalysts [16]. Recently, we reported on the synthesis of CuO– CeO 2 catalysts via the urea–nitrate combustion method and their catalytic performance for the SRM reaction [14]. These catalysts were produced in a single step in ready-to-use form. The resulting material characteristics and their catalytic www.elsevier.com/locate/apcatb Applied Catalysis B: Environmental 69 (2007) 226–234 * Corresponding author. Tel.: +30 2610 965264; fax: +30 2610 965223. E-mail address: [email protected](T. Ioannides). 0926-3373/$ – see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.apcatb.2006.07.007
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Joan Papavasiliou a,b, George Avgouropoulos a, Theophilos Ioannides a,*a Foundation for Research and Technology-Hellas, Institute of Chemical Engineering and High Temperature Chemical
Processes (FORTH/ICE-HT), P.O. Box 1414, GR-26504 Patras, Greeceb Department of Chemical Engineering, University of Patras, GR-26504 Patras, Greece
Received 13 February 2006; received in revised form 30 June 2006; accepted 10 July 2006
Available online 17 August 2006
Abstract
Steam reforming of methanol was carried out over a series of doped CuO–CeO2 catalysts prepared via the urea–nitrate combustion method.
XRD analysis showed that at least part of the dopant cations enter the ceria lattice. The addition of various metal oxide dopants in the catalyst
composition affected in a different way the catalytic performance towards H2 production. Small amounts of oxides of Sm and Zn improved the
performance of CuO–CeO2, while further addition of these oxides caused a decrease in catalyst activity. XPS analysis of Zn- and Sm-doped
catalysts showed that increase of dopant loading leads to surface segregation of the dopant and decrease of copper oxide dispersion. The addition of
oxides of La, Zr, Mg, Gd, Y or Ca lowered or had no effect on catalytic activity, but led to less CO in the reaction products. Noble-metal modified
catalysts had slightly higher activity, but the CO selectivity was also significantly higher.
CuCe0.90Ca0.10 (^), CuCe0.95Gd0.05 (&), commercial ($). Activity–selectivity curves of unmodified CuO–CeO2 (open circle) catalyst as well as the CO selectivity
equilibrium curve (dotted line) are also presented.
CuO–CeO2 and showed enhanced reducibility with TPR peaks
shifted to lower temperatures.
3.2. Dopants with a negligible effect
Mg, Zr, La and Gd dopants belong in this category. The
addition of these dopants at a molar ratio of 0.05 (0.10 for Zr)
almost did not affect the catalytic activity, but lowered the CO
selectivity, especially at high temperatures (Fig. 4a, c, d and g).
A direct relationship between the physicochemical character-
istics of these catalysts (as obtained from BET, XRD and TPR
measurements) and their catalytic performance is not obvious.
For example, Mg-, Zr- and La-doped catalysts had lower
surface area than CuO–CeO2, while the Gd-doped sample had a
larger one. The TPR a-peak was smaller in intensity in these
doped catalysts and, in the case of Zr- and Mg-doped catalysts,
J. Papavasiliou et al. / Applied Catalysis B: Environmental 69 (2007) 226–234232
Fig. 4. (Continued ).
a third high-temperature b3-peak appeared implying a higher
fraction of less active copper oxide species. Gd- and La-doped
catalysts, on the other hand, had TPR peaks shifted to 10–20 8Chigher temperatures. The Rh-containing catalyst had also
similar activity to CuO–CeO2, but a much higher CO selectivity
close to its equilibrium value.
3.3. Dopants with a negative effect
Ca- and Y-doped catalysts had lower activity than the
unmodified CuO–CeO2 catalyst (Fig. 4g). The decrease in the
activity correlates well with the TPR results in the case of the
Ca-doped sample, since there is a shift of all peaks by 50 8C to
higher temperatures (the largest shift among all catalysts).
Comparison of TPR profiles of the Y-doped and the undoped
catalyst, on the other hand, shows that the Y-doped catalyst has
a smaller area of the a-peak.
The results of this work indicate, first of all, that the
combustion method employed for the synthesis of doped CuO–
CeO2 catalysts leads to incorporation of at least part of the
dopant cations into the lattice of CeO2. In most of the cases, the
presence of the dopant did not enhance the reducibility of
J. Papavasiliou et al. / Applied Catalysis B: Environmental 69 (2007) 226–234 233
Fig. 5. Activity and selectivity for SRM over CuCe0.95Zn0.05 (triangles),
CuCe0.95Sm0.05 (circles) and CuO–CeO2 (rectangles) catalysts for different
W/F ratios: 0.103 g s cm�3 (solid symbols), 0.257 g s cm�3 (open symbols) and
0.686 g s cm�3 (open-crossed symbols). Operating conditions: 5% MeOH,
H2O/MeOH = 1.5.
copper oxide species, the two exceptions being Sm- amd Zn-
doped catalysts (increase of a-peak in TPR), which also showed
increased catalytic activity. The performance of Zn- and Sm-
doped catalysts in methanol steam reforming was further
examined by variation of the W/F ratio and the results are given
in Fig. 5. A common feature of both doped catalysts at all W/F
ratios is that their CO selectivity is noticeably lower than the
one of undoped CuO–CeO2 at all temperatures. The improved
performance of the Zn-doped catalyst was observed for all
W/F ratios employed, while the Sm-doped catalyst showed
higher activity than the undoped catalyst only at the
intermediate W/F ratio.
The addition of a larger amount of dopant leads always to
decrease of catalytic activity. One explanation for this could be
the segregation of the dopant oxide on the catalyst surface (as
found by XPS). The dopant oxide may partially cover copper
oxide crystallites or disturb the interaction between copper and
ceria leading to decrease of copper oxide dispersion. As far as
CO selectivity is concerned, the most pronounced effect is
found in the case of Pd and Rh-containing catalysts, for which
CO selectivity is remarkably (three times or more) higher and
quite close to WGS equilibrium. To further investigate this
behavior, the performance of CuO–CeO2 and Pd–CuO–CeO2
catalysts in methanol decomposition was examined. It was
found that, in the absence of steam, methanol conversion over
CuO–CeO2 is almost zero at 280 8C and becomes 7% and 20%
at 300 and 320 8C, respectively. Taking into account that
methanol conversion is �90% at 280 8C in the presence of
steam, it can be inferred that methanol decomposition is not an
important pathway on CuO–CeO2 under steam reforming
conditions. The Pd–CuO–CeO2 catalyst was more active than
CuO–CeO2 in methanol decomposition achieving methanol
conversion of 5%, 13% and 35% at 280, 300 and 320 8C,
respectively. This means that methanol decomposition is not an
important pathway at least at temperatures lower than 300 8C.
As the CO selectivity is close to equilibrium over the Pd and Rh
containing CuO–CeO2 catalysts even at temperatures at
which methanol decomposition does not take place (i.e. at
T < 280 8C), it can be concluded that this is due to the high
activity of these metals (when supported on CeO2) for the WGS
and the reverse WGS reaction [33,34]. On the other hand, CO
selectivity decreases with the addition of dopant cations and
this is desirable for the use of CuO–CeO2 catalysts in fuel cell
applications.
4. Conclusions
Doping of CuO–CeO2 catalysts with small amounts of
oxides of Sm and Zn improves their catalytic performance in
methanol steam reforming, while doping with oxides of La, Zr,
Mg, Gd, Y or Ca lowers or has negligible effect on catalytic
activity. All doped catalysts produce less CO than CuO–CeO2.
Addition of larger amounts of dopant leads always to a decrease
of catalytic activity. Pd and Rh-containing catalysts have
similar (Rh) or higher (Pd) activity compared to CuO–CeO2,
but their CO selectivity is significantly higher and close to WGS
equilibrium. At least part of dopant cations gets incorporated
into the CeO2 lattice leading to solid solution formation, while
increase of dopant loading leads to its surface segregation and
decrease of copper oxide dispersion. As the activity of CuO–
CeO2 catalysts is, in most cases, not improved by doping, it
appears that the effect of surface segregation outweighs the
effect of the modification of the ceria lattice by dopant
incorporation.
Acknowledgement
The work was carried out in the frame of EPAN E-25 project
with funding from the General Secretariat for Research and
Technology (Ministry of Development, Greece).
References
[1] L.F. Brown, Int. J. Hydrogen Energy 26 (2001) 381.
[2] J. Agrell, H. Birgersson, M. Boutonnet, I. Melian-Cabrera, R.M. Navarro,