Elsevier

Advances in Catalysis

Volume 52, 2009, Pages 213-272
Advances in Catalysis

Chapter 4 X‐Ray Photoelectron Spectroscopy for Investigation of Heterogeneous Catalytic Processes

https://doi.org/10.1016/S0360-0564(08)00004-7Get rights and content

Abstract

X‐ray photoelectron spectroscopy (XPS) is commonly applied for the characterization of surfaces in ultrahigh vacuum apparatus, but the application of XPS at elevated pressures has been known for more than 35 years. This chapter is a description of the development of XPS as a novel method to characterize surfaces of catalysts under reaction conditions. This technique offers opportunities for determination of correlations between the electronic surface structures of active catalysts and the catalytic activity, which can be characterized simultaneously by analysis of gas‐phase products. Apparatus used for XPS investigations of samples in reactive atmospheres is described here; the application of synchrotron radiation allows the determination of depth profiles in the catalyst, made possible by changes in the photon energy. The methods are illustrated with examples including methanol oxidation on copper and ethene epoxidation on silver. Correlations between the abundance of surface oxygen species and yields of selective oxidation products are presented in detail. Further examples include CO adsorption and methanol decomposition on palladium and CO oxidation on ruthenium.

Introduction

The surface science approach to heterogeneous catalysis takes advantage of simplified model versions of catalysts and reactions. Model reactions have been investigated under well‐defined ultrahigh‐vacuum (UHV) conditions with powerful methods that have allowed chemical and structural characterization down to the atomic level. Some investigations, carried out with complementary spectroscopic and structural surface science techniques have yielded detailed information about (1) elementary steps of surface reactions, such as molecular dissociation, adsorption, desorption, and interactions between the species present on surfaces, (2) responses of surface geometric and electronic structures to the presence of atoms and molecules that are participants in catalytic reactions, (3) effects of structural imperfections and controlled amounts of adsorbed atoms such as those used often as additives in the preparation of real catalysts, and (4) surface dynamics, such as propagation of reaction fronts and the related formation of stationary patterns, which may act as local chemical microreactors, etc. Such information provides basic knowledge about the elementary steps of surface reactions used in modeling reactivity and mechanisms of catalytic processes. Theoretical predictions of elementary steps, reaction barriers, the nature of transition states, and dynamics of surface processes based on DFT electronic structure calculations or Monte‐Carlo simulations have been used to guide the experiments.

The major drawbacks of this “bottom ups” approach are the so‐called “materials gap,” and “pressure gap.” Technological catalysts are morphologically complex multicomponent materials, often with microscopic dimensions, whereas the fundamental surface‐science investigations focus on the catalytic behavior of well‐structured and, most often, single‐crystal metal, alloy, or oxide samples. Some attempts have already been made to bridge the materials gap by preparing microstructured and nanostructured model supported catalysts, made by using lithography or dispersing catalyst nanoparticles or microparticles on supports (Bäumer and Freund, 1999, Goodman, 1995, Graham et al., 1994, Günther et al., 2004, Schütz et al., 1999). The size dependence of the geometric and electronic structures of the nanomaterials implies the need for quantitative understanding the effects of the size of these structures on surface reactivity, an important issue for tailoring the catalyst properties that is still in infancy (Aballe et al., 2004, Valden et al., 1998).

Realistic operating pressures in catalytic reactions are orders of magnitude higher than those used in most surface‐science experiments, and the chemical potential of the gas, usually neglected in the UHV experiments, becomes a significant contribution to the free energy of the surface layer. This pressure difference implies that the structures monitored under unrealistic UHV conditions do not necessarily involve the structures that play important roles in catalytic reactions. Significant changes in the morphology of the catalysts can occur as a consequence of dynamic structural rearrangements induced by the molecules participating in the reaction. The compositional and structural changes of the catalyst surface may exert dramatic effects not only on the catalyst performance, but also on the reaction mechanism (Stampfl et al., 2002).

The first attempts to bridge the pressure gap were ex situ experiments, allowing direct transfer of samples between a UHV station and a high‐pressure reaction cell; thus, UHV surface‐science techniques were coupled with reaction experiments carried out under realistic conditions (Rodriguez and Goodman, 1991). This strategy permits determination of the structure and composition of a model catalyst (usually a clean single crystal metal, alloy, or oxide, which may or may not be modified by controlled amounts of adsorbates) before and after the catalytic reaction to be correlated with the catalytic activity measured in the cell at a relatively high pressure.

An example illustrating the value of this approach is the work demonstrating the pressure gap effect in CO oxidation on Ru(0001). By use of ex situ TPD, STM, LEED, and XPS characterization, it was shown that the Ru(0001) surface, which appears to be inactive in surface‐science investigations, is in reality a more efficient catalyst than platinum, because the active oxygen‐rich phase cannot form under UHV conditions (Böttcher et al., 1997, Ogletree et al., 2002, Over et al., 2001).

The limitation of the ex situ approach is that the composition and structure of the catalyst surface are not investigated under reaction conditions. This limitation prevents the post-mortem analysis of the catalyst under UHV to determine an assessment of the surface restructuring, intermediate species, segregation of specific components, etc., that are characteristic of reaction conditions.

Thus, there is a strong motivation to implement techniques to characterize catalytic surfaces during the reaction and to elucidate their active phases. The first such reaction investigations were carried out with optical absorption, diffraction, and structural techniques that work at atmospheric pressures; the techniques included are IR spectroscopy (Beitel et al., 1996, Szanyi et al., 1994), SFG spectroscopy (Davis and Barteau, 1990, Su et al., 1996), XAS (Knop‐Gericke et al., 1998), SXRD (Peters et al., 2001), and STM (Hendriksen and Frenken, 2002, McIntyre et al., 1994).

X‐ray photoelectron spectroscopy (XPS) has been recognized as one of the best analytical methods for probing composition and electronic structure of solid surfaces and interfaces. However, it is based on monitoring emitted photoelectrons, which imposes a limitation on the operating pressures; consequently, substantial time has passed before this technique was applied in reaction experiments in the near‐atmospheric pressure range.

The section that follows is a review of the development of equipment for carrying out XPS at relatively high pressures. Although more than 35 years have passed since the first such XPS experiments were performed, only a few XPS investigations of catalysts under reaction conditions have yet been published. Subsequent sections provide experimental details and examples of XPS investigations of catalysts in the working state. Examples concerned with the interaction of CO and methanol with the Pd(111) surface demonstrate how XPS is useful for investigation of the pressure dependence of the structure of adsorption sites. The example of ethene epoxidation on silver illustrates a correlation of the catalytic activity with the abundance of oxygen species on the surface. XPS showed that in methanol oxidation catalyzed by copper a subsurface oxygen species is involved in the selective formation of formaldehyde.

Section snippets

History of XPS Applied at Substantial Pressures

The basic elements of the equipment for XPS with samples in reactive atmospheres are shown in Figure 1. The X‐ray source (1) can be a conventional X‐ray tube or a synchrotron radiation facility. The thin X‐ray window (2) separates the volume of the X‐ray source from the sample cell (4). X‐rays from the source pass through the X‐ray window, hit the sample (3), and induce the emission of photoelectrons. After traveling through the sample cell, a fraction of the photoelectrons reach the entrance

VG ESCALAB Photoelectron Spectrometer

The group working at the Boreskov Institute used a VG ESCALAB photoelectron spectrometer (Boronin et al., 1988, Joyner and Roberts, 1979a, Joyner and Roberts, 1979b), shown in Figure 3; it consists of three main chambers (analyzer and two preparation chambers), each pumped by a separate diffusion pump providing a background pressure lower than 5 × 10−10 mbar. A higher pressure is created in a gas cell, which is inserted inside the analyzer chamber of the spectrometer through the left

Interaction of CO with Pd(111)

CO on Pd(111) was chosen for an initial XPS investigation with the equipment operated at the Boreskov Institute described earlier, because it has been characterized with numerous vibrational spectroscopy techniques (Bourguignon et al., 1998, Bradshaw and Hoffmann, 1978, Gates and Kesmodel, 1983, Hoffmann, 1983, Loffreda et al., 1999, Morkel et al., 2003, Ohtani et al., 1987, Ozensoy et al., 2002, Rupprechter et al., 2002, Tüshaus et al., 1990), allowing comparison of the results with literature

Dehydrogenation and Oxidation of Methanol on Pd(111)

The dehydrogenation (decomposition) of methanol to give CO and H2 on supported catalysts has attracted much attention because of its practical relevance for methanol‐fueled vehicles or heat‐recovery techniques (Conrad et al., 1978, Matsumura et al., 1997, Shiozaki et al., 1999, Usami et al., 1998, Wickham et al., 1991, and references cited therein). Although methanol decomposition occurs on palladium‐containing catalysts with high selectivities for CO + H2, the activity is not satisfactory and

Ethene Epoxidation Catalyzed by Silver

Silver catalysts for ethene epoxidation have been investigated extensively (Bal'zhinimaev, 1999, Bukhtiyarov et al., 1999, Bukhtiyarov et al., 1999, Campbell, 1985, Campbell and Paffett, 1984, Grant and Lambert, 1985, van Santen and de Groot, 1986, van Santen and Kuipers, 1987), but mostly with high‐vacuum surface science techniques and not under reaction conditions. Removal of the reaction gas mixtures by evacuation, which is a usual step in the post-reaction analysis, can lead to the

Methanol Oxidation Catalyzed by Copper

Elemental copper can be used as an unsupported catalyst for the oxidative dehydrogenation of alcohols to their respective aldehydes. There are two main reaction paths: partial oxidation to formaldehyde and total oxidation to carbon dioxide, which is thermodynamically favored. The oxidation of methanol has been investigated by both classical UHV surface‐science techniques and synchrotron‐based X‐ray absorption spectroscopy with catalysts in reactive atmospheres. In room‐temperature UHV–X‐ray XPS

CO Oxidation Catalyzed by Ru(0001)

Following the pioneering work demonstrating a high activity of the so‐called O‐rich Ru(0001) surface for catalysis of CO oxidation (Böttcher and Niehus, 1999a, Böttcher and Niehus, 1999b, Böttcher et al., 1997), the accepted interpretation is that rutile RuO2(110), formed on the Ru(0001) surface under realistic oxidation conditions, is the catalytically active phase (Kim et al., 2001, Ogletree et al., 2002, Over and Seitsonen, 2002). These findings have been followed by many experimental and

Outlook

This review gives an overview of the application of XPS in the characterization of surfaces of working catalysts at pressures in the millibar range. The presented examples clearly showed that in situ XPS represents an appropriate tool for the investigation of metastable species formed on the surface of an active catalyst just under reaction conditions. The still limited number of publications in this field shows clearly the disadvantage that only a few groups have access to this method, because

Acknowledgments

The authors thank the staff of BESSY for help in carrying out experiments. V.I.B. and V.V.K. gratefully acknowledge the Russian Foundation for Basic Research (grants 04–03–32667 and 06–03–33020) for partial support of this work.

References (167)

  • C. Au et al.

    Chem. Phys. Lett.

    (1979)
  • C. Backx et al.

    Surf. Sci.

    (1983)
  • A.K. Bhattacharya et al.

    Surf. Sci.

    (1988)
  • A.I. Boronin et al.

    Surf. Sci.

    (1988)
  • A. Böttcher et al.

    Surf. Sci.

    (2002)
  • B. Bourguignon et al.

    Surf. Sci.

    (1998)
  • M. Bowker et al.

    Surf. Sci.

    (1980)
  • A.M. Bradshaw et al.

    Surf. Sci.

    (1978)
  • G. Broden et al.

    Surf. Sci.

    (1976)
  • V.I. Bukhtiyarov et al.

    J. Catal.

    (1994)
  • V.I. Bukhtiyarov et al.

    Surf. Sci.

    (1994)
  • C.T. Campbell

    J. Catal.

    (1985)
  • C.T. Campbell

    Surf. Sci.

    (1985)
  • C.T. Campbell et al.

    Surf. Sci.

    (1984)
  • C.I. Carlisle et al.

    Surf. Sci.

    (2000)
  • J.‐J. Chen et al.

    Surf. Sci.

    (1995)
  • H. Conrad et al.

    Surf. Sci.

    (1974)
  • H. Conrad et al.

    Surf. Sci.

    (1978)
  • M.L. Cubeiro et al.

    J. Catal.

    (1998)
  • J.L. Davis et al.

    Surf. Sci.

    (1990)
  • D.L. Doering et al.

    J. Catal.

    (1982)
  • H. Fellner‐Feldegg et al.

    J. Electron Spectrosc. Relat. Phenom.

    (1975)
  • T. Fuhrmann et al.

    Chem. Phys. Lett.

    (2004)
  • J.A. Gates et al.

    J. Catal.

    (1983)
  • R.B. Grant et al.

    J. Catal.

    (1985)
  • B.L. Henke et al.

    Atomic Data and Nuclear Data Tables

    (1993)
  • F.M. Hoffmann

    Surf. Sci. Rep.

    (1983)
  • Z. Hussain et al.

    J. Electron Spectrosc. Relat. Phenom.

    (1996)
  • V. Johánek et al.

    Appl. Surf. Sci.

    (2000)
  • R. Joyner et al.

    Surf. Sci.

    (1979)
  • R.W. Joyner et al.

    Chem. Phys. Lett.

    (1979)
  • E. Kleimenov et al.

    Surf. Sci.

    (2005)
  • A. Knop‐Gericke et al.

    Nucl. Instr. Meth. A

    (1998)
  • G.A. Kok et al.

    Surf. Sci.

    (1983)
  • N. Kruse et al.

    Surf. Sci.

    (1990)
  • K.Y. Kung et al.

    Surf. Sci. Lett.

    (2000)
  • D. Loffreda et al.

    Surf. Sci.

    (1999)
  • V. Matolin et al.

    Surf. Sci.

    (1990)
  • V. Matolin et al.

    Surf. Sci.

    (1991)
  • K. McCrea et al.

    Surf. Sci.

    (2001)
  • L. Aballe et al.

    Phys. Rev. Lett.

    (2004)
  • M. Abate et al.

    Surf. Interface Anal.

    (1992)
  • B.S. Bal'zhinimaev

    Kinet. Catal.

    (1999)
  • X. Bao et al.

    Phys. Rev. B

    (1996)
  • M. Barber et al.

    Faraday Discuss. Chem. Soc.

    (1972)
  • M.A. Barteau et al.
  • M. Bäumer et al.

    Surf. Sci. Rep.

    (1999)
  • G.A. Beitel et al.

    J. Chem. Phys.

    (1996)
  • H. Bluhm et al.

    J. Phys. Chem. B

    (2004)
  • H. Bluhm et al.

    J. Phys.Cond. Matter

    (2002)
  • Cited by (251)

    • Smart nanomaterials based on metals and metal oxides for photocatalytic applications

      2023, Advances in Smart Nanomaterials and their Applications
    • Selective oxidation by mixed metal nanoparticles

      2023, Comprehensive Inorganic Chemistry III, Third Edition
    View all citing articles on Scopus
    View full text