Supported Co-based perovskites as catalysts for total oxidation of methane

https://doi.org/10.1016/j.apcata.2004.11.005Get rights and content

Abstract

Supported LaCoO3 perovskites with 2, 5, 10, 15, 20 and 30 wt.% loading were prepared by impregnation of a Ce0.8Zr0.2O2 support (40 m2 g−1) with: (i) a solution of La and Co nitrates and (ii) a “citrate” solution, namely containing La and Co nitrates, and citric acid. All precursors were decomposed and calcined at 700 °C for 5 h. XRD investigations indicated the formation of a pure perovskite phase only if citrates were used. These materials were tested as catalysts for methane combustion in the temperature range 300–700 °C. All catalysts showed a lower T50 (the temperature at which the conversion level of methane is 50%) than the Ce0.8Zr0.2O2 support or non-supported LaCoO3. The activity increased continuously with the perovskite loading. The samples prepared from citrates were slightly more active than from nitrates. This is due to a more homogeneous surface, as indicated by XPS measurements. The presence of a well-characterized perovskite phase (as opposed to highly dispersed elements) seems necessary for good activity. A higher reaction rate per perovskite weight is observed for low loadings when compared to bulk LaCoO3, but the variation with perovskite loading presents a breakpoint, suggesting complex interactions in the catalysts or in the oxidation mechanism.

In spite of the experimental impossibility to evaluate the area developed by the supported perovskite, an approximative approach strongly suggests a synergy between the support and supported species.

Introduction

Perovskites have been frequently studied as substitutes for presently used total oxidation catalysts. The reason is that perovskite-type oxides (mixed oxides with a specific structure, of general formula ABO3, where lanthanide elements usually are at the A-site position and first row transition metals at the B-site) are less expensive and often thermally more stable than noble metals. Some formulations were reported to possess similar or higher activity than Pt-supported catalysts [1], [2]. The practical applications of complete oxidation catalysts are rapidly expanding, in the field of either energy production or environmental protection. In this last case, demands are very diverse, with respect to resistance to inhibitors or poisons, heating value of the feed, nature and size of the equipment to use, etc. For this reason, noble metals cannot satisfy the demands in all cases and continuing efforts are made to adapt perovskites to new niches. The major limitation of perovskites results from their reduced surface area. Aside high temperature techniques that lead to materials having low surface areas, several methods starting from precursors (nitrates, acetates, citrates) are used but necessitate relatively high temperatures for reasons not yet fully elucidated. In particular, despite the fact that the amorphous “citrate method” produces homogeneous high-surface area perovskites [3], [4], the use of these solids is limited due to their ability to sinter.

The case of effluents with low heating value, and/or with compositions varying with time, is particularly interesting. Reverse-flow reactors are advantageous in several situations. If the heating value is low, reverse-flow may be a substitute for external heating. If the gas composition changes with time, the use of special reactors, i.e. reverse-flow reactors, can make unnecessary a re-ignition step. This was the incentive of this research that was undertaken in a consortium associating a major steel company and four university laboratories. In that case, the amount of gas to purify was not considerable, and the composition, actually very complex, changed with time because of the nature of the plant operation. In addition, the composition was complex, making it difficult the use of a noble metal, because of its sensitivity to competitive adsorption of pollutants and additional effects of steam, CO and CO2, present in high proportions. A specific approach for taking advantage of perovskites was thus attempted, namely deposition on a thermally stable oxide. With this option, the use of a reverse-flow reactor was particularly well adapted, especially if the support had a high heat capacity, a necessary prerequisite for adiabatic operation with a feed of variable composition.

The present study was conducted in this context. The objective was to explore the potential as supports of chemical compositions known in ceramic science such as those containing Ce and Zr. In principle, oxides containing Ce and/or Zr are suitable for our purpose, because they possess adequate physical properties, especially heat capacity, and do not react with cations very frequently used for environmental control purposes, e.g. La, Co, Mn or Fe [21], [22], [23], [24], [25], [26].

Among the requirements for obtaining active supported catalysts, probably the most important one is that the active phase should not react with the support yielding an inactive material as transition metals do on alumina [5]. Gallagher et al. [6], [7] and Johnson et al. [8], [9] reported the preparation of cordierite-supported perovskites belonging to the LaMnO3 family by freeze drying, spray drying and precipitation. The high activity in the case of freeze drying was attributed to a more homogeneous distribution of the oxides on the support surface. However, experiments showed that the transition metal elements most frequently used for imparting oxidation activity to perovskite, like Co, reacted with the commonly used supports alumina or silica yielding inactive compounds such as spinels [10], [11], [12]. A pre-coating of Al-containing supports with La2O3 seems effective in preventing or retarding the loss of Co cations by migration into the bulk of the carrier oxide [13], [14], [15]. The use of an aluminum spinel which is less reactive versus Co [12], [14], [15], [16] is also possible. Literature is not very abundant concerning supported perovskites but the use of specially shaped supports (inorganic fibers, hollows cylinders or metallic foams) was extensively studied [2], [17], [18], [19]. Thus, Klvana et al. [2], [17] reported good activity for methane combustion and increased resistance to mercaptans by supporting La0.66Sr0.34Ni0.3Co0.7O3 on inorganic fibers based on Al, Si, Zr and Y oxides. In this case also, the chemical nature of the supports is an important factor. Fujii et al. [20] used LaCoO3 or La0.8Sr0.2CoO3 and observed that the reaction rate for propane oxidation varied greatly with the nature of the oxide carrier. High catalytic activities, exceeding even the one of bulk La0.8Sr0.2CoO3, were obtained by using ZrO2 and CeO2. The authors did not propose any explanation for this difference in behavior. This study concluded that other materials of conventional composition were not interesting as supports. Concerning ZrO2, it was also found that it enhanced the activity of LaNiO3 [21], [22] for NOx reduction with CO and that of LaCoO3 and LaFeO3 [23], [24] for CO oxidation.

Account taken of the specific objective of this work, literature data constituted an encouragement to search for suitable perovskite/support systems.

We have shown previously that the thermal decomposition of citrate precursors gives perovskite-type oxides at relatively low temperatures, such as 600–700 °C [27], [28]. Zhang et al. [29] also reported that the citrate process leads to perovskites with small particle size and suggested that this process was promising for the preparation of supported perovskites [30], [31]. The aim of the present study was to prepare LaCoO3 perovskites supported on Ce0.8Zr0.2O2 and to evaluate their activity for methane combustion. The reason of the selection of Ce0.8Zr0.2O2 as support is explained at the beginning of the next section. Several preparation methods were used in order to see how the physico-chemical properties and the catalytic activity depended on the synthesis route.

Section snippets

Preparation of the support

The support used for this study was Ce0.8Zr0.2O2. The reasons of this selection from a large number of CeO2–ZrO2 mixed oxides of different compositions were presented elsewhere [32]. Briefly, this composition had the largest surface area (40 m2 g−1), was the most stable thermally and the most homogeneous among all investigated materials. The Ce0.8Zr0.2O2 solid solution was prepared by the “citrate method” [3], [32] and was calcined at 700 °C. The calcined material was pressed at 5 t cm−2 into

XRD

For loading levels below 15 wt.%, XRD measurements showed no other phases than the support (cubic Ce0.8Zr0.2O2), both for CIT- and NIT-samples, whatever the composition and amount of deposited perovskite. Characteristic lines of the perovskite phase appeared at 15 wt.% loading. Patterns corresponding to 15 and 20 wt.% LaCoO3 impregnated on Ce0.8Zr0.2O2 are shown in Fig. 1, Fig. 2, respectively. Only two phases could be seen: the support and the perovskite. Actually, an overlap occurs between the

Discussion

The first aim of the discussion is to compare the various physico-chemical data concerning the catalysts in order to describe their texture in the most precise way possible, account taken of experimental limitations. A second part of the discussion will then be aimed at explaining the improved catalytic activity when LaCoO3 is deposited on Ce0.8Zr0.2O2. The supported catalysts (especially CIT-series) will allow and do deserve more extensive consideration. Smaller sections will concern other

Conclusions

Supported LaCoO3/Ce0.8Zr0.2O2 catalysts possess a higher activity than their non-supported counterparts in the catalytic oxidation of methane. The calculation of the last section points to some phase cooperation between the perovskite and Ce0.8Zr0.202. But taking only the experimental results used for constructing Fig. 7, one can note that CIT–10 wt.% LaCoO3 having an SSA of 28.4 m2 g−1 behaves as a hypothetical unsupported LaCoO3 having an SSA of 173 m2 g−1, and CIT-30 wt.% (24.6 m2 g−1) is equivalent

Acknowledgements

We thank the Commission of the European Union for the financial support of this research under the contract ENV4-CT97-0599.

References (42)

  • H. Arai et al.

    Appl. Catal.

    (1986)
  • D. Klvana et al.

    Catal. Today

    (1999)
  • P. Courty et al.

    Powder Tech.

    (1973)
  • P.K. Gallagher et al.

    Mater. Res. Bull.

    (1974)
  • P. Arnoldy et al.

    J. Catal.

    (1985)
  • P.E. Marti et al.

    Appl. Catal. B: Environ.

    (1994)
  • R. Burch et al.

    Appl. Catal. A: Gen.

    (2001)
  • D. Klvana et al.

    Appl. Catal. A: Gen.

    (1997)
  • A.K. Ladavos et al.

    Catal. Today

    (1993)
  • A.K. Ladavos et al.

    Appl. Catal. B: Environ.

    (1993)
  • S. Colonna et al.

    J. Mol. Catal. A: Chem.

    (2002)
  • S. Cimino et al.

    J. Catal.

    (2002)
  • S. Cimino et al.

    Appl. Catal. B: Environ.

    (2002)
  • D. Ferri et al.

    Appl. Catal. B: Environ.

    (1998)
  • J. Kirchnerova et al.

    Appl. Catal. A: Gen.

    (2002)
  • H.M. Zhang et al.

    Appl. Catal.

    (1988)
  • J.W. Geus et al.

    Catal. Today

    (1999)
  • P. Vidmar et al.

    J. Catal.

    (1997)
  • D. Terribile et al.

    Catal. Today

    (1999)
  • D.E. Petrakis et al.

    Thermochim. Acta

    (1992)
  • K.S. Chan et al.

    Appl. Catal. A: Gen.

    (1994)
  • Cited by (52)

    • Double perovskites La <inf>2</inf> MMnO <inf>6</inf> as catalyst for propane combustion

      2017, Journal of Asian Ceramic Societies
      Citation Excerpt :

      Supported noble metal oxides, particularly Pd oxides, are excellent catalysts for low temperature combustion but noble metals are expensive and have tendency to deactivation due to sintering, decomposition or undesirable interaction with supports [5]. Therefore transition metal oxides and mixed oxides have now been established as an inexpensive alternative to precious metal and noble metals containing materials [9]. Diverse transition metals mixed oxides of specific structures, such as spinels, perovskites, pyrochlores, and also hexaaluminate have been reported as suitable catalysts for HC combustion due to oxygen ion conductivity.

    • Effect of catalytically active Ce<inf>0.8</inf>Gd<inf>0.2</inf>O<inf>1.9</inf> coating on the heterogeneous combustion of methane within MgO stabilized ZrO<inf>2</inf> porous ceramics

      2017, Combustion and Flame
      Citation Excerpt :

      Through the vacancy transport, the MSZ lattice facilitates the oxidation of adsorbed compounds, by enabling the transport of O2− ions to adsorbates which bind to the weak acid base sites of the MSZ surface [14,15]. The deposition of catalytically active Gd doped CeO2 on ZrO2 based porous ceramic supports can improve the stability of CeO2 nanopowders for powder loadings less than 15 wt% on the support surface [27]. Under such low catalyst concentrations, the support will prevent loosely distributed powders from sintering and coarsening at high operational temperatures [28] and, thus, the high catalytic activity of CeO2 during methane combustion will be preserved.

    • Cu/Mn-based mixed oxides derived from hydrotalcite-like precursors as catalysts for methane combustion

      2014, Applied Catalysis A: General
      Citation Excerpt :

      Therefore, extensive research is also carried out into methane combustion over less active but much cheaper catalysts, based on transition metal oxides and related systems. Investigated materials include single and mixed metal oxides, both bulk and supported, frequently possessing thermally stable structures of spinel, perovskite or hexaaluminate type [e.g. 4,11–47]. A number of works address the role of the catalyst preparation method and the nature of the oxide precursor [15,19,20,30,31,33,40–43].

    View all citing articles on Scopus
    View full text