High-throughput and combinatorial development of multicomponent catalysts for ethanol steam reforming

https://doi.org/10.1016/j.apcata.2010.05.041Get rights and content

Abstract

Noble metal-free MgAl2O4 supported multicomponent catalysts for steam reforming of ethanol have been designed by means of combinatorial tools and high-throughput approaches. Additionally, the catalysts have been investigated using high-throughput temperature programmed technique. MgAl2O4 supported Ni-based catalysts have been found to be the most effective in hydrogen production. It has been revealed that practically no noble metal is required in order to achieve high hydrogen yield. At 500 °C a four-component catalyst containing Ni, Co, Ce and Mo has resulted in 4.4 mol hydrogen per mole of ethanol. In this system Ni is considered as the active metal, while Co, Ce and Mo are promoters. The results revealed strong synergism between Ni and Co leading to high activity in hydrogen production. In the temperature range between 320 and 370 °C pronounced coke formation via the Boudouard reaction was observed over the two-component Ni–Co catalyst. The addition of Ce to NiCo/MgAl2O4 system proved to be advantageous, resulting in further improvement in hydrogen yield and suppression of coke formation. The presence of Mo as the fourth component hindered all reactions of CO in the temperature range between 320 and 370 °C. Thus, the contribution of reactions, such as WGRS, CO methanation and coke formation were reduced as well. The four-component catalyst has been proved to be highly effective in hydrogen production at higher temperatures. Its advantage is the strong suppression of the formation of carbonaceous deposition. The analysis of results shows also that some of the catalysts compositions can be promising for ethanol pre-reforming producing both hydrogen and methane from ethanol for Molten Carbonate Fuel Cells.

Graphical abstract

Noble metal-free MgAl2O4 supported multicomponent catalysts for steam reforming of ethanol have been designed by means of combinatorial tools and high-throughput approaches. At 500 °C a four-component catalyst containing Ni, Co, Ce and Mo has resulted in 4.4 mol hydrogen per mole of ethanol. Its advantage is the strong suppression of the formation of carbonaceous deposition.

  1. Download : Download high-res image (86KB)
  2. Download : Download full-size image

Introduction

Hydrogen seems to be a promising energy carrier in the future, which can be converted to electric power with high efficiency in fuel cells. In order to support sustainable hydrogen economy, it is crucial to produce hydrogen, renewably. Hydrogen production by bio-ethanol reforming has the potential of being nearly CO2 neutral, since the produced CO2 is consumed for the biomass source growth, thus offering a nearly closed carbon loop.

Depending on the catalysts used, working at very low contact times, two primary reactions take place yielding acetaldehyde or ethylene, according to the following reactions [1], [2], [3]:C2H5OH  CH3CHO + H2C2H5OH  CH2CH2 + H2O

Dehydrogenation of ethanol via reaction (R1), takes place on Cu and Ni catalysts [1], [2], [3]. Acidic sites, such as the acidic sites of alumina, are responsible for the dehydration of ethanol via reaction (R2). At higher contact times, both the initial and intermediate products can easily react further. Nevertheless, whatever the steam:ethanol ratio is, the total conversion of ethanol at T > 500 K is predicted according to the thermodynamics [4]. The following additional reactions were considered:CH3CHO  CH4 + COCO + H2O  CO2 + H2CH3CHO + H2O  2CO + 3H2CO + 3H2  CH4 + H2OCO + H2  H2O + C2CO  CO2 + CCH4  2H2 + C

Accordingly, acetaldehyde, by decarboxylation (reaction (R3)), gives methane and carbon monoxide while steam reforming of acetaldehyde leads to CO and hydrogen (reaction (R5)). Water gas shift reaction (WGSR) (reaction (R4)) results in CO2 and hydrogen. CO methanation (reaction (R6)) is thermodynamically feasible below T < 700 K, while the reverse reaction (methane steam reforming) takes place at higher temperatures [4], [5]. It has to be mentioned that, the presence of CO in the effluent gas should be suppressed when the produced hydrogen is used in PEM fuel cells. In order to minimize CO formation, in case of high steam:ethanol ratios, temperatures at T > 700 K are suitable. Above 700 K WGSR is favored, which lead to reduced CO content [4]. There are many processes reported leading to coke deposition and removal, such as carbon gasification (reaction (R7)), Boudouard reaction (R8) and methane decomposition (reaction (R9)). According to thermodynamic calculations at higher temperatures and higher steam:ethanol ratios the removal of carbonaceous deposits is favorable. It has been observed that, at H2O/EtOH ratios above 3, the graphitic carbon deposition is practically zero above 600 K [4], [5], [6].

Beside thermodynamic issues discussed above, kinetic promotion and hindrance of different processes helps to improve selectivity towards hydrogen. For this reason, different active metals, various methods of preparation, numerous supports and additives have already been tested. The appropriate catalyst should work at as low temperature as possible, it should be active in WGSR, while it has to inhibit reactions, such as coke formation, CO production and CO methanation.

It has been found that noble metals are active catalyst components for ethanol reforming. Among the Al2O3 supported noble metals, Rh containing catalyst has proven to be the most active [7], [8], [9]. Over Al2O3 supported noble metal catalysts the formation rate of hydrogen decreased and that for ethylene increased in time on stream, while the conversion was almost constant [10]. This finding was explained by the inhibiting effect of the surface acetate species. This effect is weakened by increasing the concentration of water, the metal content of the catalyst and the reaction temperature. The activity of Al2O3 supported catalyst decreased in the order of Rh > Pd > Ni = Pt. Upon using CeO2 or ZrO2 as supports the formation of ethylene is negligible and the order of the catalyst activity changes to Pt  Rh > Pd [7]. Dehydration of ethanol can be depressed by adding K to neutralize the acidic support as was found in the case of Pt/Al2O3 [11], or by using basic supports, i.e. La2O3 and MgO [12]. Depositing Rh on MgAl2O4 exhibited higher basicity than alumina-supported Rh, resulting in improved catalyst stability [13].

It is a general observation that the nickel containing catalysts are more active than the supported noble metals [14]. It has been evidenced that Ni has high activity for C–C bond and O–H bond breaking and also for hydrogenation reaction of CO. The addition of alkali species modifies the character of interaction between adsorbed species and the metallic Ni, further enhancing its steam reforming activity. However, like Rh, Ni is less active for WGSR. Since Cu favors dehydrogenation and WGSR, the combination of Ni and Cu shows high steam reforming activity and high selectivity to hydrogen [3]. In addition, mixing Cu with noble metal, such as Rh may also improve hydrogen production due to enhanced WGSR by Cu. Like other catalysts, Ni-based catalysts also suffer from coke formation as well as metal sintering, leading to considerable performance degradation during long-term operation. Thus, addition of properly selected additives is needed to improve the stability of catalysts. All these results strongly indicate that ethanol reforming requires development of multicomponent catalysts.

In this study we describe the design and testing of multicomponent heterogeneous catalysts for ethanol steam reforming using combinatorial tools and high-throughput approaches. The main goal is to reduce the noble metal content while maintaining high activity and selectivity towards hydrogen production. The role of different active metals and modifiers is also discussed in this paper.

Section snippets

Objective function

Ideally, the overall process of the ethanol steam reforming takes place according to reaction (R10), where hydrogen and CO2 are considered as the desired products.CH3CH2OH + 3H2O  6H2 + 2CO2However, as a consequence of numerous side reactions shown before, other products different from hydrogen and CO2 are also formed as by-products.

In the beginning of the optimization, i.e. in the design of the second catalyst generation, the objective function was created using the molar flow rate of hydrogen

Preparation of catalysts

The catalysts were prepared parallel by means of a liquid dispensing robot in Syncore reactor (BÜCHI Labortechnik AG, Switzerland) using the rack for 24 glass reaction vessels. The following precursor compounds were used: Pt(NH3)4(NO3)2 (Aldrich, 99.99%), Pd(NO3)2·2H2O (Fluka, >98.0%), HAuCl4·3H2O (Aldrich, 99.99%), Ni(NO3)2·6H2O (Fluka, >98.5%), Co(NO3)2·6H2O (Fluka, >98.0%), Cu(NO3)2·3H2O (Fluka, >99.0%), Zn(NO3)2·6H2O (Fluka, >99.0%), La(NO3)3·6H2O (Fluka, >99.0%), (NH4)2Ce(NO3)6 (Fluka,

Catalyst library design

Based on results obtained in preliminary high-throughput temperature programmed study [18] MgAl2O4 has been selected as a support for catalyst library design. In order to design a catalyst library an appropriate reaction temperature was selected. Eventually, the forthcoming optimization was done at 350 °C, which is explained by the relatively large difference in the hydrogen production between different catalysts at this temperature. This approach is favorable when the establishment of a

Conclusions

Hydrogen production by means of steam reforming of ethanol over MgAl2O4 supported catalysts has been investigated at 350 °C applying combinatorial and high-throughput approaches. Additional high-throughput results were obtained at higher temperatures. The design and testing of five catalyst generations resulted in highly active noble metal free catalysts containing Ni, Co and Ce. Upon using MgAl2O4 support no formation of ethylene was observed indicating that in the first step of ethanol

Acknowledgements

Partial financial support by OTKA (grant K 77720 to A.T.) is greatly acknowledged. The authors thank to Mr. Lajos Végvári (Combitech-Nanotech Kft.) for his help using HRS as a catalyst library optimization tool.

References (24)

  • F.J. Mariño et al.

    Int. J. Hydrogen Energy

    (1998)
  • F.J. Mariño et al.

    Appl. Catal. A: Gen.

    (2003)
  • V. Mas et al.

    Int. J. Hydrogen Energy

    (2006)
  • A.L. Alberton et al.

    Catal. Today

    (2007)
  • E.Y. Garcia et al.

    Int. J. Hydrogen Energy

    (1991)
  • J.P. Breen et al.

    Appl. Catal. B: Environ.

    (2002)
  • A. Erdőhelyi et al.

    Catal. Today

    (2006)
  • M. Dömök et al.

    Appl. Catal. B: Environ.

    (2007)
  • S. Cavallaro et al.

    Appl. Catal. A: Gen.

    (2003)
  • F. Aupretre et al.

    J. Catal.

    (2005)
  • C. Diagne et al.

    Catal. Commun.

    (2002)
  • A. Tompos et al.

    Appl. Catal. A: Gen.

    (2008)
  • Cited by (26)

    View all citing articles on Scopus
    View full text