Importance of the oxygen bond strength for catalytic activity in soot oxidation

https://doi.org/10.1016/j.apcatb.2016.01.068Get rights and content

Highlights

  • The oxygen bond strength is found to be very important for the activity of a catalyst in soot oxidation.

  • The activites of different catalysts outline two different volcano curves for intimate and loose contact.

  • Linear Brønsted-Evans-Polanyi relationships are observed for both intimate contact and loose contact.

Abstract

The oxygen bond strength on a catalyst, as measured by the heat of oxygen chemisorption, is observed to be a very important parameter for the activity of the catalyst in soot oxidation. With both intimate contact between soot and catalyst (tight contact) and with the solids stirred loosely together (loose contact) the rate constants for a number of catalytic materials outline a volcano curve when plotted against their heats of oxygen chemisorption. However, the optima of the volcanoes correspond to different heats of chemisorption for the two contact situations. In both cases the activation energies for soot oxidation follow linear Brønsted-Evans-Polanyi relationships with the heat of oxygen chemisorption. Among the tested metal or metal oxide catalysts Co3O4 and CeO2 were nearest to the optimal bond strength in tight contact oxidation, while Cr2O3 was nearest to the optimum in loose contact oxidation. The optimum of the volcano curve in loose contact is estimated to occur between the bond strengths of α-Fe2O3 and α-Cr2O3. Guided by an interpolation principle FeaCrbOx binary oxides were tested, and the activity of these oxides was observed to pass through an optimum for an FeCr2Ox binary oxide catalyst, which exhibited a rate constant at 550 °C that was 2.3 times higher than the one for pure α-Cr2O3 and 29 times higher than the one for pure α-Fe2O3.

Introduction

Soot particles in the exhaust from diesel vehicles are likely to cause lung cancer and to affect the climate both locally and globally [1], [2], [3], [4], [5], [6]. The soot particles are therefore typically removed from the exhaust gas by filtration through a ceramic filter [7], [8], [9]. It is necessary with periodic regeneration of the filter, where the filter temperature is increased, and the soot is oxidized. The growing back pressure due to the soot deposits and the temperature increase required for filter regeneration are associated with increased fuel consumption [10]. To limit the increase in fuel consumption it is desirable to develop soot oxidation catalysts that can lower the regeneration temperature—ideally down to the typical temperature of the exhaust gas [7], [11]. Here it is a challenge that the heterogeneously catalyzed soot oxidation is a gas/solid/solid interaction, where the contact between soot and catalyst is very important for the catalytic activity [12], [13], [14]. In tests, where soot and catalyst are crushed together (so-called tight contact), the oxidation occurs at a significantly lower temperature, than when soot and catalyst are stirred together (so-called loose contact) [12], [13], [14]. TEM studies by Gardini et al [15] on Ag/soot mixtures have indicated that tight contact corresponds to an extensive interface between primary particles of catalyst and soot, whereas loose contact corresponds to fewer contact points at the interface between coagulates of catalyst particles and coagulates of soot particles. In several experiments [11], [16], [17], [18], [19], [20] with oxidation of soot particles filtered from gas streams by a catalytic filter part of the soot oxidation has been observed to peak at a relatively low temperature in the range characteristic of tight contact with the catalyst, while another part of the soot oxidation has been observed to peak at a higher temperature more characteristic of loose contact with the catalyst. Hence an understanding of both tight and loose contact oxidation may be relevant for real filter applications.

To promote the development of improved catalysts that enable soot oxidation to take place at lower temperatures it is important to identify the parameters that determine the catalytic activity. The surface area of the employed catalyst is known to be of importance [21], [22], [23], but an improved understanding of the factors determining the intrinsic activity of a catalytic material would be beneficial. Boreskov et al. [24] proposed that for metal oxide catalysts the oxygen bond strength on the catalyst surface, as measured by the heat of oxygen chemisorption, is a determining factor for activity in reactions involving oxygen activation. This has been observed to be the case in oxygen activation reactions such as 16O2/18O2 isotopic exchange [24] and oxidation of CH4 [24], C3H6 [25], H2 [24], [25], [26], [27] and CO [28], [29], [30]. The activity in the catalytic oxidation of CO on thin metal oxide films has also been correlated to the activation energy for oxygen desorption [31], which scales with the heat of chemisorption. Furthermore, the heat of chemisorption has also been found to be among the factors that play a role for the selectivity in oxidation of benzaldehyde [32] and methanol [33]. It is therefore relevant to investigate, if the heat of oxygen chemisorption can explain the trends in catalytic activity for soot oxidation, and that is the topic of the present work.

Section snippets

Catalysts used for screening experiments

The catalysts used in the screening studies were bulk metals or metal oxides. In the cases of γ-Fe2O3, α-Fe2O3, V2O5, Au, Pd and Pt commercially acquired samples were used. The suppliers and purities of the used samples are listed in Table S1 in the Supplementary information. In the cases of CeO2, Co3O4, MnOx, ZnO, CuO, Cr2O3 and TiO2 the oxide samples were prepared by flame spray pyrolysis according to the method described elsewhere [23]. This preparation method results in highly crystalline

The materials

The identities of the employed catalytic materials have been verified by XRD. The diffraction patterns are shown in Figs. S1–S6 in the Supplementary information. The materials are generally phase pure with the exception of MnOx and to a minor extent TiO2. The MnOx sample shows signs of several different phases (MnO2, Mn2O3 and Mn3O4). The TiO2 sample is primarily in the anatase phase, but minor traces of a rutile phase is visible. The surface areas of the employed catalysts have been determined

Conclusion

The oxygen bond strength on the surface of a catalyst as measured by the heat of oxygen chemisorption is observed to be a very important parameter for the activity of the catalyst in soot oxidation. With both intimate contact between soot and catalyst (tight contact) and with the solids stirred loosely together (loose contact) the rate constants (for a reaction order of 2/3 in carbon) of a number of catalytic materials outline a volcano curve when depicted against their heats of oxygen

Acknowledgements

Financial support from The Danish Council for Strategic Research (DSF) is gratefully acknowledged (Grant No. 2106-08-0039). We thank Helge Kildahl Rasmussen from DTU Physics for aid in connection with the XRD measurements.

References (111)

  • D.B. Kittelson

    J. Aerosol Sci.

    (1998)
  • J. Neeft et al.

    Fuel Process. Technol.

    (1996)
  • A.M. Stamatelos

    Energy Convs. Manage.

    (1997)
  • J. Neeft et al.

    Chem. Eng. J.

    (1996)
  • J. Neeft et al.

    Appl. Catal. B

    (1996)
  • J. Neeft et al.

    Stud. Surf. Sci. Catal.

    (1995)
  • D. Gardini et al.

    Appl. Catal. B

    (2016)
  • P.A. Kumar et al.

    Chem. Eng. J.

    (2012)
  • K. Shimizu et al.

    Appl. Catal. B

    (2010)
  • K. Klier

    J. Catal.

    (1967)
  • R. Strobel et al.

    Adv. Powder Technol.

    (2006)
  • A. Setiabudi et al.

    Appl. Catal. B

    (2004)
  • H. Jung et al.

    Combust. Flame

    (2005)
  • J. Im et al.

    Chemosphere

    (2008)
  • A. Braun et al.

    Environ. Sci. Technol.

    (2007)
  • M. Bowker et al.

    Surf. Sci.

    (2012)
  • R. Dziembaj

    J. Solid State Chem.

    (1978)
  • R. Dziembaj

    J. Solid State Chem.

    (1978)
  • B.L. Yang et al.

    J. Catal.

    (1982)
  • D.H. Parker et al.

    Surf. Sci.

    (1989)
  • M. Haruta et al.

    J. Catal.

    (1989)
  • M. Haruta

    Catal. Today

    (1997)
  • N. Lopez et al.

    J. Catal.

    (2004)
  • N. Saliba et al.

    Surf. Sci.

    (1998)
  • H. Tsai et al.

    Surf. Sci.

    (2003)
  • C.J. Tighe et al.

    Combust. Flame

    (2012)
  • S.R. Kelemen et al.

    Carbon

    (1985)
  • B. Marchon et al.

    Carbon

    (1988)
  • S.S. Barton et al.

    Carbon

    (1972)
  • A. Bueno-Lopez et al.

    J. Catal.

    (2005)
  • G. Mul et al.

    J. Catal.

    (1998)
  • S. Simonsen et al.

    J. Catal.

    (2008)
  • E. Aneggi et al.

    J. Catal.

    (2014)
  • P. Mars et al.

    Chem. Eng. Sci. Special Suppl.

    (1954)
  • A.A. Balandin

    Adv. Catal.

    (1969)
  • G.R. Hennig

    J. Inorg. Nucl. Chem.

    (1962)
  • D.W. McKee

    Carbon

    (1970)
  • V.A. Shvets et al.

    J. Catal.

    (1972)
  • K. Yamazaki et al.

    Appl. Catal. A

    (2014)
  • E. Baumgarten et al.

    Appl. Catal.

    (1988)
  • G. Mul et al.

    Carbon

    (1998)
  • C. Ris

    Inhal. Toxicol.

    (2007)
  • W.L. Chameides et al.

    Science

    (2002)
  • R.A. Kerr

    Science

    (2013)
  • B. Frank et al.

    Angew. Chem. Int. Ed.

    (2013)
  • M.O. Andreae et al.

    Science

    (2013)
  • B.A.A.L. Van Setten et al.

    Catal. Rev. Sci. Eng.

    (2001)
  • J. Adler

    Int. J. Appl. Ceram. Technol.

    (2005)
  • B.W.L. Southward, S. Basso, SAE paper 2008-01-0481,...
  • A.G. Konstandopoulos, S. Lorentzou, C. Pagkoura, K. Ohno, K. Ogyu, T. Oya, SAE paper 2007-01-1950,...
  • Cited by (35)

    • Is fighting against pollutants possible with critical raw material free perovskites?

      2022, Catalysis Today
      Citation Excerpt :

      It is accepted that the reaction takes place at the interface between catalyst and soot, following a Mars-Van Krevelen mechanism. In loose contact soot oxidation, the most important role is played by oxygen adsorbed on the surface of the catalyst [72], so the comparison of these results can provide insights on the kind of oxygen that a catalyst can easily dispose. In tight contact, samples with 10 % Sr doping do not show significant differences in the CO2 formation profile.

    View all citing articles on Scopus
    View full text