High temperature steam reforming of methanol over Cu/ZnO/ZrO2 catalysts

https://doi.org/10.1016/j.apcatb.2009.06.023Get rights and content

Abstract

Methanol steam reforming to hydrogen and carbon dioxide is catalyzed at 400 °C over Cu/ZnO/ZrO2 (Cu content, 30 wt.%) prepared by a coprecipitation method. Conventional binary catalysts of Cu/ZnO and Cu/ZrO2 have also been prepared by the same method. The activity of Cu/ZnO/ZrO2 is stable in comparison with the binary catalysts and a commercial Cu/ZnO/Al2O3. The selectivity to carbon monoxide by-produced over Cu/ZnO/ZrO2 is fairly low. The particle sizes of Cu and ZnO for Cu/ZnO/ZrO2 are significantly smaller than those for Cu/ZnO, and the BET surface area of Cu/ZnO/ZrO2 is 2–3 times as high as that of Cu/ZnO. The Cu particle size of Cu/ZnO/ZrO2 is stable during the reaction, but growth of ZnO and ZrO2 particles is accompanied with the deactivation.

Introduction

Polymer electrolyte fuel cells (PEFCs) have been regarded as compact and efficient power generators for stationary and mobile applications. Hydrogen is the most potential fuel for PEFCs, but there are problems in its storage.

Methanol is a capable feedstock of hydrogen processors for fuel cells. Selective methanol steam reforming is usually carried out below 300 °C over copper catalysts such as Cu/ZnO/Al2O3 for hydrogen production. Since the catalysts are easily deactivated and less selective (by-production of unfavorable carbon monoxide) at the higher temperature, the reactor cannot be directly heated with combustion gas; therefore, thermal oil is usually employed for the heat supply [1]. This is an obstacle in development of a simple and compact hydrogen processor. In addition, the activity of the conventional catalyst is insufficient for the compact reformer. For example, Choi and Stenger reported that no complete conversion of methanol can be attained over a commercial Cu/ZnO/Al2O3 at 250 °C with the GHSV as small as 1100 h−1 [2]. High operation temperature will realize the direct heat supply and large heat flux to the reactor. The reaction rate is generally large at high temperatures and it will also contribute to the compact reformer. Hence, a new active catalyst durable at the high temperature is required for the efficient reactor.

Removal of carbon monoxide is another problem in the hydrogen processor. The hydrogen fuel for PEFCs must not contain carbon monoxide above 10 ppm, which poisons the catalytic anode [3]. Selective oxidation of carbon monoxide is often carried out to remove carbon monoxide to the level less than 10 ppm [4], but an excessive amount of oxygen is required in the process and it consumes hydrogen. The CO selectivity is usually increased at the high reaction temperature. Membrane reactors comprised of hydrogen separation membranes and reforming catalysts are proposed for the production of pure hydrogen from methanol [5], [6], [7], [8], [9], [10] and it is a solution for removal of carbon monoxide. Han et al. reported the thermal efficiency of the membrane reactor as high as 82% at 350 °C [11]. Although palladium membranes are the most promising on the viewpoint of the hydrogen permeability and selectivity, the formation of β-Pd hydride is a risk to damage the membrane at the operation temperature less than 300 °C [12]. In addition, the permeability at 300 °C is ca. 60% of that at 400 °C when the activation energy is 15 kJ mol−1 [13]. Thus, the operation temperature of the membrane reactor should be higher than 300 °C, and it meets the requirement of the compact reformer that we are going to integrate with the membrane. Catalysts containing transition metals such as nickel and iridium mainly produce hydrogen and carbon monoxide in the reforming at around 400 °C [14], [15]. While pure hydrogen is obtained through a pinhole-free membrane, defects usually exist on thin membranes and they decrease the selectivity ratio of hydrogen to other gases and result in contamination of hydrogen with carbon monoxide. Thus, the by-production of carbon monoxide is unfavorable even if the hydrogen purification is carried out with the palladium membrane.

The ternary system of Cu/ZnO/ZrO2 was first developed for the methanol synthesis from hydrogen and carbon dioxide [16], [17], [18], [19]. Agrell et al. reported methanol steam reforming on Cu(32 wt.%)/ZnO(40 wt.%)/ZrO2 at temperatures less than 300 °C [20]. By-production of carbon monoxide was significantly small in comparison with Cu/ZnO or Cu/ZnO/Al2O3, and the activity was appreciably high. Matter et al. reported that the activity of Cu(29 wt.%)/ZnO(28 wt.%)/ZrO2 was the highest at 300 °C or below among the catalysts examined by them [21]. Since the selectivity to carbon monoxide usually increases with an increase in the reaction temperature, suppression of CO formation is important in the high temperature reforming above 300 °C as well as the durability of the catalyst.

In the present work, we examined the activity of copper catalysts with the Cu content of 30 wt.% to find the promising catalysts durable at 400 °C. It is shown that Cu/ZnO/ZrO2 is fairly stable with a low CO selectivity in comparison with conventional catalysts such as Cu/ZnO, Cu/ZrO2, and Cu/ZnO/Al2O3.

Section snippets

Experimental

A series of copper-based catalysts were prepared by coprecipitation from a 0.5-M aqueous mixture of Cu(NO3)2·3H2O (Wako Pure Chemical, S grade), Zn(NO3)2·6H2O (Wako, S), and/or ZrO(NO3)2·2H2O (Wako, 1st) with addition of an aqueous solution of Na2CO3 (0.5 M) under vigorous stirring at 80 °C. After filtration, the precipitate was dispersed in distilled water and washed at room temperature for ca. 0.5 h. The procedure was repeated for several times until the pH value and conductivity of the filtrate

High temperature steam reforming of methanol over conventional catalysts

Steam reforming of methanol was carried out over a commercial Cu/ZnO/Al2O3 catalyst at 400 °C for 420 min. The methanol conversion decreased with an increase in a time period of the reaction (Fig. 1). The selectivity to carbon monoxide increased gradually from 5.8% to 7.7%. On the other hand, the CO selectivity produced with Cu/ZnO decreased to 4.4% after 420-min on-stream and the methanol conversion was always higher than that of Cu/ZnO/Al2O3 by ca. 10%. The activity of Cu/ZrO2 decreased steeply

Deactivation of Cu/ZnO

Carbon monoxide is considered as a secondary product of the steam reforming of methanol over Cu/ZnO/Al2O3 by the reverse water gas-shift (WGS) reaction (CO2 + H2  CO + H2O) [27], [28]. In this mechanism CO production is suppressed when methanol conversion is low. However, the CO selectivity increased gradually during the reaction with a commercial Cu/ZnO/Al2O3 catalyst at 400 °C despite decrease in the methanol conversion (see Fig. 1). Hence, it is supposed that the nature of the active sites changes

Conclusions

Addition of zirconium oxide to Cu/ZnO improves the activity and stability of the copreciptated catalyst in the methanol steam reforming at 400 °C. The CO selectivity with Cu/ZnO/ZrO2 is lower than that with Cu/ZnO, while the selectivity decreases with a decrease in the methanol conversion. The higher BET surface area can be produced by the addition of zirconium oxide. The particle sizes of Cu and ZnO for Cu/ZnO/ZrO2 are considerably smaller than those for Cu/ZnO. Growth of the ZnO particles can

References (42)

  • Y. Choi et al.

    Appl. Catal. B

    (2002)
  • P. Marques et al.

    J. Power Sources

    (2006)
  • Y.-M. Lin et al.

    Catal. Today

    (2001)
  • A. Basile et al.

    Int. J. Hydrogen Energy

    (2006)
  • B.K.R. Nair et al.

    Chem. Eng. Sci.

    (2006)
  • F. Gallucci et al.

    Int. J. Hydrogen Energy

    (2007)
  • A. Iulianelli et al.

    J. Membr. Sci.

    (2008)
  • J. Han et al.

    Int. J. Hydrogen Energy

    (2002)
  • C. Qi et al.

    J. Power Sources

    (2007)
  • J. Jia et al.

    Appl. Catal. A

    (2008)
  • Y. Ma et al.

    Appl. Catal. A

    (1998)
  • Y.-W. Suh et al.

    Catal. Today

    (2000)
  • J. Słoczyński et al.

    Appl. Catal. A

    (2003)
  • F. Arena et al.

    J. Catal.

    (2007)
  • J. Agrell et al.

    J. Catal.

    (2003)
  • P.H. Matter et al.

    J. Catal.

    (2004)
  • Z.G. Wu et al.

    J. Non-Crystall. Solids

    (2003)
  • J. Agrell et al.

    J. Power Sources

    (2002)
  • H. Purnama et al.

    Appl. Catal. A

    (2004)
  • B.A. Peppley et al.

    Appl. Catal. A

    (1999)
  • A. Mastalir et al.

    J. Catal.

    (2005)
  • Cited by (72)

    View all citing articles on Scopus
    View full text