Johnson Matthey Technol. Rev., 2018, 62, (3), 316
In Situ and Operando Spectroscopy: A Powerful Approach Towards Understanding Catalysts
Revealing reaction mechanisms to find the correlation between structure and surface composition
- Liliana Lukashuk§
- Institute of Materials Chemistry, Technische Universität Wien, Getreidemarkt 9/BC/01, 1060 Vienna, Austria
- §Present address: Johnson Matthey, PO Box 1, Belasis Avenue, Billingham, Cleveland, TS23 1LB, UK
- Karin Foettinger*
- Institute of Materials Chemistry, Technische Universität Wien, Getreidemarkt 9/BC/01, 1060 Vienna, Austria
The improvement of catalytic processes is strongly related to the better performance of catalysts (higher conversion, selectivity, yield and stability). Additionally, the desired catalysts should meet the requirements of being low cost as well as environmentally and user-friendly. All these requirements can only be met by catalyst development and optimisation following new approaches in design and synthesis. This article discusses three major approaches in the design and development of catalysts: (a) high-throughput synthesis; (b) reaction kinetic studies; (c) in situ and operando spectroscopy for studying catalysts under process conditions. In contrast to approaches based on high-throughput synthesis and reaction kinetic studies, an emerging approach of studying catalysts under process conditions using in situ and operando spectroscopy and transferring the gained knowledge to design of new catalysts or the optimisation of existing catalysts is not yet widely employed in the chemical industry. In this article, examples of using in situ or operando spectroscopy for studying the surface and bulk of catalysts under process conditions are discussed, with an overview of applying in situ X-ray absorption spectroscopy (XAS), in situ infrared (IR) spectroscopy and in situ near-ambient pressure X-ray photoelectron spectroscopy (NAP-XPS) for monitoring the bulk and surface composition of PdZn/ZnO and Pd2Ga/Ga2O3 methanol steam reforming catalysts.
1. Approaches for Catalyst Development
1.1 Discovery of a ‘Lucky Catalyst’ via High-Throughput Synthesis
Research and development of more efficient and sustainable catalysts, either the development of novel catalysts or optimisation of existing ones, is a vital part of industrial and academic research. One of the most common approaches for catalyst development is the synthesis of a large number of catalysts, followed by screening of their catalytic activity (known as high-throughput synthesis) (1, 2). A high-throughput synthesis approach is the most common method of catalyst discovery and optimisation and can be strongly boosted by the combinatorial design of experiments and using robotic equipment to synthesise and test catalysts (3–8). One of numerous catalysts prepared using high-throughput synthesis might be very active for a particular reaction (a ‘lucky catalyst’). However, employing such an approach alone without understanding activity-controlling factors often does not lead to a gain in knowledge or deep understanding of factors driving the activity and selectivity of catalysts (for example the nature of active sites and reaction mechanisms).
Understanding the structure-performance relationships of catalysts is a crucial step for the successful development of the next generation of more efficient catalysts and covers different aspects. Some research lines focus on the preparation of catalysts by varying certain parameters (for example preparation methods, elemental composition or loading) and then correlate catalytic properties of the synthesised materials with their structure, reducibility or tendency to adsorb or desorb the educts and products for a particular reaction. Combining knowledge of a catalyst’s performance with its structure adds substantially to the understanding of catalytic systems and the potential to transfer knowledge to similar processes or products. Yet this approach provides rather a macroscopic picture of the structure-performance relationships of catalysts, omitting the detailed microscopic insights into the activity-controlling factors driving a particular reaction.
1.2 Reaction Kinetic Studies
Another line of research focuses on microscopic insights and explores the nature of the reaction sites of a catalyst and the mechanisms of reactions (9, 10). The commonly used methodology is based on a detailed study of reaction kinetics (reaction rates and rate laws). It takes its roots back to 1921 when the mechanisms of the catalytic action of platinum in the reactions of carbon monoxide oxidation and hydrogen oxidation were described with the help of kinetics by Irving Langmuir and presented at a Discussion of the Faraday Society held in September 1921 at the Institution of Electrical Engineers in London, UK (11). Langmuir’s work was later elaborated by Cyril Hinshelwood. The combined work of Irving Langmuir and Cyril Hinshelwood is known today as the Langmuir-Hinshelwood mechanism and is one of the most common reaction mechanisms describing numerous catalytic reactions (12). According to the Langmuir-Hinshelwood mechanism, a catalytic act is a reaction between two molecules, adsorbed on neighbouring sites (12).
Some years later, in 1938, Dan Eley and Eric Rideal proposed the Eley-Rideal mechanism, in which a catalytic act is a reaction between an adsorbed molecule and a molecule from the gas phase (13). Later works of Mars and van Krevelen on reducible oxides in 1954 suggested another mechanism that today is called the Mars-van-Krevelen type redox mechanism. According to the Mars-van-Krevelen mechanism, one reactant forms a chemical bond with the catalytic surface; the other reactant reacts directly from the gas phase. A vacancy that is formed when the reaction product desorbs was thought to be filled with an atom from the bulk (14). However, nowadays it is generally accepted that the vacancy created by the reaction is filled with the first reactant again.
The type of mechanism depends on the type of catalyst. The Mars-van-Krevelen type redox mechanism typically applies to reducible metal oxide catalysts, where lattice oxygen is involved in the reaction cycle and generation of vacancies is a crucial reaction step. The Eley-Rideal and Langmuir-Hinshelwood mechanisms typically describe non-reducible oxides and metal nanoparticle catalysts, where a network of elementary reaction steps involves adsorption of reactants on the catalyst surface, a catalytic reaction and desorption of products.
As is demonstrated in the review of Rob Berger et al., in the chemical industry knowledge of the reaction kinetics is of primary importance for: (a) catalyst development; (b) process development; (c) process optimisation; and (d) mechanistic research (12). 29% of the utilisation of kinetic data in industry is dedicated to catalyst development (12). For mechanistic studies and advanced catalyst development, a precise description of the reaction rate is required at the level of elementary steps. This calls for advanced kinetic studies such as fast transient kinetic investigations (employing a temporal analysis of product reactor) (15) and advanced kinetic modelling studies (16–18).
1.3 The In Situ and Operando Approach
Although studying the kinetics of reactions is a powerful tool for identifying steps of the reaction cycle and for obtaining hints about active sites, it does not allow a complete picture of the reaction mechanism to be obtained. Often several reaction mechanisms can run in parallel; therefore, an exact identification of the reaction mechanism is not feasible. Moreover, for heterogeneous catalytic reactions, the identification of active sites is rather vague especially for solid catalysts because of their complexity. Often solid catalysts are multicomponent and multielement materials. Furthermore, the structures of solid catalysts are usually not well defined; thus, several sites might serve as active sites and as a result, different reaction pathways might simultaneously run on different reaction sites or in different temperature regimes.
The complementary methodology based on the use of in situ and operando techniques for studying the catalyst under reaction conditions can overcome some of the limitations of the traditional kinetic methodology and provide a broader insight into the complex network of reaction pathways, thereby strongly contributing to understanding the structure-performance relationships of catalysts, the nature of active sites, reaction mechanisms, the structure of working catalysts and, based on the knowledge gained, the design of new catalysts (19–21).
Why is it so important to study catalysts under process conditions? Solid catalysts are very dynamic under working conditions when high temperatures and pressures are applied. Freshly prepared catalysts often undergo structural changes under activation and pretreatment conditions and further under reaction conditions. Thus, simply speaking, catalysts ‘adjust’ to the reaction environment (22). Monitoring such environment-induced changes of catalysts and understanding why they happen should be taken into account during the design of catalysts.
The traditional approach to gain insights into environment-induced changes was and is still in many cases to study catalysts ex situ (meaning outside, off site, out of place). However, this does not allow a clear picture of working catalysts to be obtained (Figure 1). The famous saying of Gabor Somorjai highlights this, stating that knowing the ‘before-reaction part’ or ‘after-reaction part’ “is like studying a life with access only to the prenatal and postmortem states” (23). Moreover, exposing the catalyst to ambient conditions might again change the catalyst’s structural and electronic properties and lead to data misinterpretation (Figure 1).
A more informative way is to monitor the catalyst directly under reaction conditions, in situ (meaning on site, in position, undisturbed) (23). Watching the catalyst under reaction conditions by physical-chemical methods and simultaneously recording catalytic data using a mass spectrometer or a gas chromatograph allows the catalytic properties to be correlated directly with electronic and structural changes in the catalyst. This approach is called ‘operando’ (24–26). An in situ approach is not only used for studying catalysts under working conditions, but can also be applied to the preparation of catalysts. For instance, application of tomographic energy dispersive diffraction imaging during the calcination of [Ni(en)(H2O)4]Cl2/γ-Al2O3 and [Ni(en)3](NO3)2/γ-Al2O3 hydrogenation catalyst bodies was used to study the changes in the structural composition of the catalysts at the macroscopic level (27).
An operando or in situ approach allows one to study the “birth, life and death” of catalysts, a concept that was recently introduced by Bert Weckhuysen (22). According to his concept, “birth” refers to the stage when a catalyst is prepared, as it is essential to understand how catalytic solids are formed; “life” describes how a catalyst functions and performs in a catalytic reaction; “death” explains how a catalyst undergoes deactivation.
Generally, operando investigation of catalysts can include spectroscopic techniques (for example Raman spectroscopy, infrared (IR) spectroscopy, X-ray photoelectron spectroscopy (XPS), XAS, ultraviolet-visible (UV-Vis) spectroscopy and electron paramagnetic resonance (EPR) spectroscopy), but also microscopy (for example environmental transmission electron microscopy or environmental scanning electron microscopy, X-ray diffraction (XRD) and X-ray total scattering). A number of research groups have demonstrated the strength of combining several in situ techniques for the investigation of a range of catalytic systems under reaction conditions to identify dynamic changes to the catalysts under varying conditions (Figure 2) (28–35). Therefore, many efforts and the attention of the operando community are directed towards developing an integrated or combined approach, simultaneously applying several techniques (for example IR, Raman, XRD and XAS), for studying catalysts under reaction conditions.
Such an approach is successfully employed in the BM01B beamlines at the European Synchrotron Research Facility (ESRF), where in situ XRD, XAS and Raman spectroscopy are combined (36). Other synchrotron facilities have also been taking steps in this direction. Table I summarises the most commonly used hard X-ray synchrotron beamlines from the European synchrotron facilities where opportunities for operando and in situ studies of catalysts are provided. Using an integrated approach requires special reaction cells that allow in situ experiments to be carried out either simultaneously or sequentially in the same cell. Therefore, many efforts are directed toward the development of cells that can be used for combined measurements, in particular focusing on the cell design to obtain good agreement between the catalytic activity and selectivity of a catalyst measured in a conventional reactor and in a spectroscopic cell (37–39).
2. The Need to Understand the Surface of Catalysts Under Process Conditions
The increased interest in the in situ and operando approach for studying catalysis is demonstrated by the increased number of research papers where in situ and operando techniques are employed (40). A lot of money is currently invested into the building and improvement of synchrotron facilities to provide dedicated beamlines for in situ and operando studies and the topic regularly appears on the agenda at catalysis conferences. Moreover, a dedicated conference on operando spectroscopy was launched in 2003 in Lunteren, The Netherlands, and successfully takes place every three years.
Despite numerous studies on catalysts using in situ and operando investigations including those using combined techniques, very limited knowledge exists in the literature about the surface composition of catalysts under reaction conditions, as most of the techniques characterise the bulk of the catalysts. The changes in surface composition, however, are the most relevant for explaining catalytic events that take place at catalyst surfaces (41). Importantly to note, the average bulk structure of materials does not represent the surface composition as was demonstrated in a recent study by Karim et al. on the size-dependent redox behaviour of iron, where it was shown that during oxidation iron grows in layers, forming Fe(0)/FeO/Fe3O4/Fe2O3 core-shell structures (42).
Dynamic surface variations under reaction conditions for catalysts are often not taken into account (for example, surface segregation processes). Assumptions are mainly made on the basis of the average bulk structure that leads to misleading conclusions on the active phase, active sites, role of promoters or deactivation mechanisms, thus hindering further optimisation and development of catalysts. As an example, despite intensive studies of high temperature water gas shift (HTWGS) iron-based catalysts, the nature of active sites and the HTWGS reaction mechanism (redox vs. associative) are still subject to ongoing debate. Open questions and contradictory results concern the interaction of chromium oxide with iron oxide (the formation of a Cr-enriched shell around the iron oxide phase, magnetite, vs. strong interaction of Cr3+ with magnetite) as a way of stabilising iron oxide against sintering (28, 29, 43).
To study catalysts, often operando IR spectroscopy is combined with bulk techniques such as XAS, Raman and XRD. Operando IR spectroscopy is a crucial technique for studying surface adsorbed species, particularly important for revealing reaction intermediates (39, 44); however, it does not provide insights into the redox properties of the catalyst surface. One of the most important factors that governs the catalytic activity, selectivity and stability of catalysts in particular for redox reactions is the redox properties (the oxidation state and interplay between oxidation states during the reaction) of the surface.
XPS is one of the most powerful spectroscopic tools for obtaining chemical information about the surfaces of solid materials. It is based on the photoelectric effect theory (the ejection of electrons from the surface in response to incident photons). This technique is also known as electron spectroscopy for chemical analysis (ESCA). The ESCA term was introduced in 1967 (45) by the Swedish physicist Kai Siegbahn, who in 1954 recorded the first high energy resolution XPS spectrum of cleaved NaCl (46), to highlight the importance of obtaining chemical information like oxidation states for materials investigated with XPS. In 1981 Siegbahn was awarded the Nobel Prize in Physics as acknowledgment “for his contribution to the development of high-resolution electron spectroscopy”.
Traditionally, XPS is known to be an ultra-high vacuum (UHV) technique that can only operate at pressures below 10−9 mbar because of the strong scattering of photoelectrons by gas molecules. However, the first ‘non-UHV’ XPS experiments were performed in the Kai Siegbahn research group (47, 48). In 1973 Kai Siegbahn together with his son Hans Siegbahn published a study in which the first spectrum of a liquid (formamide) was presented (47). In 1979 Wyn Roberts published a study entitled “A ‘High-Pressure’ Electron Spectrometer for Surface Studies”, where he described the development of the first photoelectron spectrometer capable of studying solid surfaces in the presence of gas up to 1.3 mbar (49). A new type of photoemission spectrometer, based on several differential pumping stages with electrostatic lenses between the sample surface and the photoelectron detection, was developed by the groups from the Advanced Light Source, Berkeley, California, USA, and the Fritz Haber Institute of the Max Planck Society, Berlin, Germany (50). A schematic representation of the differential pumping system of the NAP-XPS setup from the Innovative Station for In Situ Spectroscopy (ISISS) beamline, BESSY II, is given in Figure 3. This setup enables operation at pressures close to ambient (up to 20 mbar).
Although these pressures are still far away from pressures used in industrial fixed-bed flow reactors, NAP-XPS experiments can model the reaction environment and allow the dynamic variations of a catalyst surface and its catalytic properties to be studied (52–56). Today, NAP-XPS is the most powerful technique for studying the surface of redox catalytic materials under process conditions (57), particularly when employing synchrotron radiation as a light source. Using the tuneability of synchrotron radiation, different information depths can be studied by NAP-XPS ranging from ~0.5 nm to 3 nm, which is not achievable with routinely available lab-based XPS. Table II summarises the most commonly used soft X-ray synchrotron beamlines, where opportunities for operando and in situ NAP-XPS, ambient pressure XPS (AP-XPS) and XAS studies of catalysts are provided.
To gain improved understanding of redox-based catalytic systems, one can combine in situ NAP-XPS or near edge X-ray absorption fine structure (NEXAFS) for surface-specific chemical information with in situ Raman, XRD or XAS studies which provide information on local coordination environment, structure and oxidation state of the bulk of a material.
3. In Situ and Operando Studies Providing Information on Surface and Bulk Properties of Working Catalysts
There are only a few operando studies reported in the literature in which XAS, XRD or Raman and NAP-XPS were performed on the same catalytic system, giving valuable insights into bulk versus surface properties. As examples, recent studies by Israel Wachs’ group demonstrate the importance of employing a surface sensitive technique like NAP-XPS for studying HTWGS and shed light on iron-based HTWGS catalysts in combination with in situ XAS and Raman (28, 29). Looking at data from the respective techniques in isolation could lead to wrong conclusions. Examples from our own work at the Institute of Materials Chemistry, TU Wien, Austria, comprise preferential CO oxidation (PROX) on Co3O4 and CeO2-Co3O4 catalysts studied by operando NAP-XPS and XAS (30). During PROX both the bulk and surface of Co3O4 were fully oxidised up to 523 K, while in CO the surface reduction of Co3O4 occurred from 373 K. Preferential oxidation of CO to CO2 in an excess of H2 was suggested to follow predominantly the Mars-van-Krevelen mechanism. Promotion of Co3O4 with 10 wt% CeO2 increased the reduction temperatures in CO and H2 and enhanced the PROX activity. Since CeO2 is a less active material, this can only be explained by higher activity of the Co-O-Ce ensembles and creation of oxygen vacancies as detected by NAP-XPS in Ce 3d spectra.
In another recent work, we utilised NAP-XPS to study the surface composition of Cu-promoted Ni/ZrO2 catalysts in methane atmosphere at mbar pressures and relevant reaction temperatures. An important result from this work was that the surface composition changed from a Cu-enriched state after reduction to a Ni-enriched surface in methane at T>~700 K which was the active state (58). The surface segregation of Ni0 was driven by the strong interaction between Ni and carbon and was corroborated by density functional theory (DFT) calculations (59).
4. Operando FTIR, XAS and NAP-XPS Investigation of Methanol Steam Reforming on Pd-Based Intermetallic Compounds: Surface vs. Bulk
An exemplary study will be discussed here in more detail to illustrate the insights into catalyst structure and composition under reaction conditions and the reaction mechanism gained from the operando approach using surface sensitive methods. In the discussed work we aimed at obtaining a detailed understanding of the fundamental processes occurring at the surface and in the bulk of ZnO and Ga2O3 supported Pd nanoparticles applied as methanol steam reforming (MSR) catalysts (60–69). In addition, the work should contribute to the improvement and development of active, selective and stable catalysts for the production of clean (CO-free) hydrogen, which can be used for example in proton exchange membrane fuel cells.
Our goal was to understand the selectivity of the desired MSR reaction (Equation (i)) producing CO2 and H2 and the unwanted methanol decomposition (MDC) (Equation (ii)) yielding CO and H2, by combining information on surface and bulk composition, structure and electronic properties under reaction conditions obtained by operando XAS, XRD and NAP-XPS with mechanistic understanding obtained by Fourier transform infrared (FTIR) spectroscopy.
4.1 Active and Selective Phase Under Reaction Conditions: Operando XAS and NAP-XPS
By in situ quick-extended X-ray absorption fine structure (QEXAFS) spectroscopy we followed in real time how the Pd/ZnO nanoparticles dynamically adapt to the surrounding atmosphere (62), which presented the first direct proof for the formation of PdZn alloy in methanol/water starting from Pd/ZnO without prior high temperature reduction. PdZn alloying occurred via spillover and reduction of the ZnO by H2 generated in the reaction and was accompanied by a simultaneous change in reactivity, from an unselective state characteristic of metallic Pd to a highly selective state due to the formation of a PdZn intermetallic compound (IMC). The XAS near-edge region is shown in Figure 4 and resembles the progressive transition from the structural and electronic properties characteristic of Pd to those of the tetragonal Pd:Zn IMC with a 1:1 stoichiometry occurring with time on stream in methanol/water. By combining time resolved in situ QEXAFS with FTIR spectroscopy we could clarify the processes occurring when switching between reducing (methanol, H2) and oxidising (alloy formation and decomposition with ZnO island formation on the Pd nanoparticle surface) atmospheres (62). The studies on powder catalysts were complemented by surface science work on single crystal-based UHV grown model systems (PdZn surface alloys), which were in excellent agreement with the PdZn/ZnO powder catalysts and DFT calculations (67).
For the related system Pd/Ga2O3 the combination of techniques of different surface sensitivity, namely in situ XRD, synchrotron-based in situ XPS, FTIR and XAS, were used to identify the formation of the MSR-selective intermetallic Pd2Ga phase under relevant reaction conditions as well as upon reduction at medium temperature around 573–673 K (63, 65). One of the major advantages of XAS in catalysis is its suitability for in situ studies, due to the high penetrating power of the hard X-rays used. Fitting the extended fine structure function identified Pd2Ga as the active phase present under MSR conditions (65). Interestingly, different intermetallic phases exhibited a different reactivity (63) with Pd2Ga showing the best catalytic performance.
Since the catalytic properties of a material are mainly determined by its composition and the structure and properties of the surface and near-surface region, we have further characterised the Pd2Ga/Ga2O3 catalyst by applying synchrotron radiation-based NAP-XPS as a surface sensitive technique. Upon reduction in H2 or in the H2 produced from methanol a species attributed to zerovalent Ga evolved in the Ga 3d binding energy range shifted by 2.2 eV relative to Ga3+, whereas the Pd 3d signal was shifted by ca. 0.5 eV to higher values upon IMC formation because of the charge transfer from Ga to Pd (Figure 5). The composition of the surface was in very good agreement with the Pd:Ga stoichiometry 2:1 (Figure 5). Depth profiling by varying the incident photon energy and thereby the kinetic energy of emitted photoelectrons reveals equal relative amounts of Ga0 and Pd with a ratio of about 2:1 and therefore uniform distribution in the experimentally accessible depth range (63).
4.2 Stability and Origin of Unwanted CO Formation: Operando FTIR
Special attention was paid to the stability of the PdZn and Pd2Ga surfaces. Importantly, in situ FTIR spectroscopic results reveal for the very first time the limited stability of the intermetallic surfaces (PdZn and Pd2Ga) under MSR reaction conditions (61, 63). Besides intermetallic PdZn and Pd2Ga the presence of domains of metallic Pd catalysing the unwanted methanol decomposition to CO and H2 was detected in both catalysts (Figure 6). Metallic Pd and Pd present within the IMC can be distinguished by CO adsorption. While linear, bridge and hollow bonded CO are characteristic for adsorption on metallic Pd with vibrational bands at around 2090 cm−1, 1990 cm−1 and 1940 cm−1, CO adsorbs on only on-top sites on Pd in PdZn and Pd2Ga (61–63, 67). In addition, a shift of the on-top band to ~2070 cm−1, because of the charge transfer, is observed for PdZn (62, 67).
The instability of the intermetallic surface in methanol/water has been suggested as the main factor limiting MSR selectivity and was attributed to the detrimental effect of CO. The degradation of the IMC surface was found to strongly depend on the reaction temperature (61) being more critical at lower reaction temperatures (Figure 6). Thus, we suggest the presence of a dynamic equilibrium between IMC formation by H2 and decomposition by CO. As a consequence, CO formation in MSR could not be fully avoided. Understanding the origin of unwanted CO is a first step towards the development of strategies to avoid CO formation and is therefore a highly important result.
4.3 Insights into the Reaction Mechanism: Operando FTIR
Vibrational spectroscopy was applied to investigate the reaction mechanism of MSR over Pd2Ga/Ga2O3 (64). The mechanistic study explored the interaction of the catalyst with methanol and water by combining temperature-programmed, steady state and concentration modulation FTIR experiments. A different surface chemistry on low temperature reduced (unselective) Pd/Ga2O3 and high temperature reduced (selective) Pd2Ga/Ga2O3 was detected. Formation of intermediate surface formate species was observed at much lower temperatures on Pd2Ga/Ga2O3 (Figure 7) and was attributed to reactive oxygen sites in the Ga2O3 surface modified by the high temperature reduction and IMC formation. The presence of H2O strongly affected the stability of the formates.
A reaction mechanism involving the interconversion of formate species was proposed, as described in more detail elsewhere (64). This work provides important new insights into the reaction mechanism of MSR and emphasises the crucial role of the oxide support. The reaction seems to predominantly proceed on the reduced Ga2O3 surface near the IMC nanoparticles or at the interface and is promoted by the intermetallic particles. An additional and necessary effect of IMC formation is the slowdown of the unwanted MDC to CO and H2 on metallic Pd.
4.4 Reactivity of the Oxide Support
Concerning the reactivity of the oxide support, special attention was paid to the role of oxygen vacancies. This was investigated by applying different reducing pretreatments with subsequent quantification of vacancy sites by titration with O2. This approach was successfully applied to study the defect chemistry of Ga2O3 (68, 69). Strongly related to the work on MSR, the reactivity of Ga2O3 for reverse water gas shift (RWGS), a follow-up reaction of the MSR products CO2 and H2 and potential source of unwanted CO, was investigated by combining FTIR spectroscopy and temperature-programmed methods. By looking in detail at the interaction of Ga2O3 in different defective states in terms of nature and concentration of vacancies with H2, CO, CO2, water and their mixtures, the reactivity of Ga2O3 for water gas shift (WGS) and RWGS was explored. Ga2O3 is catalytically active for these reactions, with the reaction most likely proceeding via oxygen vacancies. Ga2O3 on its own (without metal promotion) is capable of H2 activation; formed oxygen vacancies can then be re-oxidised by CO2 yielding CO in the catalytic RWGS reaction.
The molecular level understanding of the reaction mechanism, the factors controlling selectivity and the correlation of catalyst structure and composition with catalytic properties enable a more scientific approach to catalyst optimisation on a rational basis. Only the combination of various operando methods providing complementary information on the composition of the catalyst surface, the reacting species and the catalytic properties allowed an almost complete picture of the surface chemistry of PdZn/ZnO and Pd2Ga/Ga2O3 MSR catalysts to be obtained.
5. Perspectives: A Combined Approach for the Successful Development of the Next Generation of More Efficient Catalysts
The work described in this paper clearly demonstrates the importance of combining several operando techniques in studying catalysts under steady state as well as dynamic conditions for revealing reaction mechanisms and for finding correlations between the structure and surface composition and the activity and selectivity of catalysts. In particular surface-sensitive information is crucial. Moreover, the presented results highlight the great potential of in situ NAP-XPS for obtaining redox and structural information on the (near) surface changes of catalytic materials for different information depths.
For the successful development of the next generation of more efficient catalysts, it is essential to combine three approaches (Figure 8): (a) high-throughput synthesis in combination with correlating catalytic performance with the electronic and geometric structure of synthesised materials (structure-performance relationship); (b) reaction kinetic studies; (c) in situ and operando investigation of catalysts under reaction conditions. Using such a combined approach and interconnecting the obtained information is a powerful research tool that can boost the research and development of more efficient catalysts with the desired catalytic properties.
U. Rodemerck, and M. Baerns, ‘High-Throughput Experimentation in the Development of Heterogeneous Catalysts: Tools for Synthesis and Testing of Catalytic Materials and Data Analysis’, in “Basic Principles in Applied Catalysis”, ed. M. Baerns, Vol. 75, Springer-Verlag, Berlin, Germany, 2004, pp. 259–279 LINK https://doi.org/10.1007/978-3-662-05981-4_7
R. A. Paciello, ‘High Throughput Screening of Homogeneous Catalysts: Selected Trends and Applications in Process Development’, in “Applied Homogeneous Catalysis with Organometallic Compounds”, 3rd Edn., eds. B. Cornils, W. A. Hermann, M. Beller and R. Paciello, Vols. 1–4, Ch. 16, Wiley-VCH Verlag GmbH and Co KGaA, Weinheim, Germany, 2018, pp. 1085–1096 LINK https://doi.org/10.1002/9783527651733.ch16
K. Yaccato, R. Carhart, A. Hagemeyer, M. Herrmann, A. Lesik, P. Strasser, A. Volpe, H. Turner, H. Weinberg, R. K. Grasselli, C. J. Brooks and J. M. Pigos, Comb. Chem. High Throughput Screening, 2010, 13, (4), 318 LINK https://doi.org/10.2174/138620710791054286
A. Hagemeyer, A. Lesik, G. Streukens, A. F. Volpe, H. W. Turner, W. H. Weinberg, K. Yaccato and C. Brooks, Comb. Chem. High Throughput Screening, 2007, 10, (2), 135 LINK https://doi.org/10.2174/138620707779940965
S. Cypes, A. Hagemeyer, Z. Hogan, A. Lesik, G. Streukens, A. F. Volpe, W. H. Weinberg and K. Yaccato, Comb. Chem. High Throughput Screening, 2007, 10, (1), 25 LINK https://doi.org/10.2174/138620707779802788
H. Wennemers, Comb. Chem. High Throughput Screening, 2001, 4, (3), 273 LINK https://doi.org/10.2174/1386207013331138
Z. Q. Zheng and X. P. Zhou, Comb. Chem. High Throughput Screening, 2011, 14, (3), 147 LINK https://doi.org/10.2174/138620711794728725
M. H. Fonseca and B. List, Curr. Opin. Chem. Biol., 2004, 8, (3), 319 LINK https://doi.org/10.1016/j.cbpa.2004.04.013
M.-F. Reyniers and G. B. Marin, Annu. Rev. Chem. Biomol. Eng., 2014, 5, 563 LINK https://doi.org/10.1146/annurev-chembioeng-060713-040032
D. G. Blackmond, Angew. Chem. Int. Ed., 2005, 44, (28), 4302 LINK https://doi.org/10.1002/anie.200462544
I. Langmuir, Trans. Faraday Soc., 1922, 17, 621 LINK https://doi.org/10.1039/TF9221700621
R. J. Berger, E. H. Stitt, G. B. Marin, F. Kapteijn and J. A. Moulijn, CATTECH, 2001, 5, (1), 36 LINK https://doi.org/10.1023/A:1011928218694
W. H. Weinberg, Acc. Chem. Res., 1996, 29, (10), 479 LINK https://doi.org/10.1021/ar9500980
C. Doornkamp and V. Ponec, J. Mol. Catal. A: Chem., 2000, 162, (1–2), 19 LINK https://doi.org/10.1016/S1381-1169(00)00319-8
A. C. van Veen, D. Farrusseng, M. Rebeilleau, T. Decamp, A. Holzwarth, Y. Schuurman and C. Mirodatos, J. Catal., 2003, 216, (1–2), 135 LINK https://doi.org/10.1016/S0021-9517(02)00100-8
A. A. Hakeem, R. S. Vásquez, J. Rajendran, M. Li, R. J. Berger, J. J. Delgado, F. Kapteijn and M. Makkee, J. Catal., 2014, 313, 34 LINK https://doi.org/10.1016/j.jcat.2014.02.010
A. A. Hakeem, M. Li, R. J. Berger, F. Kapteijn and M. Makkee, Chem. Eng. J., 2015, 263, 427 LINK https://doi.org/10.1016/j.cej.2014.10.104
S. K. Wilkinson, L. G. A. van de Water, B. Miller, M. J. H. Simmons, E. H. Stitt and M. J. Watson, J. Catal., 2016, 337, 208 LINK https://doi.org/10.1016/j.jcat.2016.01.025
B. M. Weckhuysen, Phys. Chem. Chem. Phys., 2003, 5, (20), 4351 LINK https://doi.org/10.1039/B309650P
A. J. Foster and R. F. Lobo, Chem. Soc. Rev., 2010, 39, (12), 4783 LINK https://doi.org/10.1039/C0CS00016G
R. Raja, M. E. Potter and S. H. Newland, Chem. Commun., 2014, 50, (45), 5940 LINK https://doi.org/10.1039/C4CC00834K
B. M. Weckhuysen, Nat. Sci. Rev., 2015, 2, (2), 147 LINK https://doi.org/10.1093/nsr/nwv020
B. M. Weckhuysen, Chem. Commun., 2002, (2), 97 LINK https://doi.org/10.1039/B107686H
B. M. Weckhuysen, Phys. Chem. Chem. Phys., 2003, 5, (20), vi LINK https://doi.org/10.1039/B309654H
M. A. Bañares, Catal. Today, 2005, 100, (1–2), 71 LINK https://doi.org/10.1016/j.cattod.2004.12.017
M. A. Bañares, M. O. Guerrero-Pérez, J. L. G. Fierro and G. G. Cortez, J. Mater. Chem., 2002, 12, (11), 3337 LINK https://doi.org/10.1039/B204494C
L. Espinosa-Alonso, M. G. O’Brien, S. D. M. Jacques, A. M. Beale, K. P. de Jong, P. Barnes and B. M. Weckhuysen, J. Am. Chem. Soc., 2009, 131, (46), 16932 LINK https://doi.org/10.1021/ja907329j
C. J. Keturakis, M. Zhu, E. K. Gibson, M. Daturi, F. Tao, A. I. Frenkel and I. E. Wachs, ACS Catal., 2016, 6, (7), 4786 LINK https://doi.org/10.1021/acscatal.6b01281
M. Zhu, T. C. R. Rocha, T. Lunkenbein, A. Knop-Gericke, R. Schlögl and I. E. Wachs, ACS Catal., 2016, 6, (7), 4455 LINK https://doi.org/10.1021/acscatal.6b00698
L. Lukashuk, K. Föttinger, E. Kolar, C. Rameshan, D. Teschner, M. Hävecker, A. Knop-Gericke, N. Yigit, H. Li, E. McDermott, M. Stöger-Pollach and G. Rupprechter, J. Catal., 2016, 344, 1 LINK https://doi.org/10.1016/j.jcat.2016.09.002
T. A. Wezendonk, V. P. Santos, M. A. Nasalevich, Q. S. E. Warringa, A. I. Dugulan, A. Chojecki, A. C. J. Koeken, M. Ruitenbeek, G. Meima, H.-U. Islam, G. Sankar, M. Makkee, F. Kapteijn and J. Gascon, ACS Catal., 2016, 6, (5), 3236 LINK https://doi.org/10.1021/acscatal.6b00426
K. Föttinger and G. Rupprechter, Acc. Chem. Res., 2014, 47, (10), 3071 LINK https://doi.org/10.1021/ar500220v
A. Brückner and E. Kondratenko, Catal. Today, 2006, 113, (1–2), 16 LINK https://doi.org/10.1016/j.cattod.2005.11.006
M. Brandhorst, S. Cristol, M. Capron, C. Dujardin, H. Vezin, G. Le bourdon, and E. Payen, Catal. Today, 2006, 113, (1–2), 34 LINK https://doi.org/10.1016/j.cattod.2005.11.008
J. A. Anderson, Z. Liu and M. F. García, Catal. Today, 2006, 113, (1–2), 25 LINK https://doi.org/10.1016/j.cattod.2005.11.007
‘Station Bending Magnet 31: BM 31’, European Synchrotron Radiation Facility, Grenoble, France: http://www.esrf.eu/UsersAndScience/Experiments/CRG/BM01/bm01b (Accessed on 14th May 2018)
G. L. Chiarello, M. Nachtegaal, V. Marchionni, L. Quaroni and D. Ferri, Rev. Sci. Instrum., 2014, 85, (7), 074102 LINK https://doi.org/10.1063/1.4890668
S. J. Tinnemans, J. G. Mesu, K. Kervinen, T. Visser, T. A. Nijhuis, A. M. Beale, D. E. Keller, A. M. J. van der Eerden and B. M. Weckhuysen, Catal. Today, 2006, 113, (1–2), 3 LINK https://doi.org/10.1016/j.cattod.2005.11.076
A. Vimont, F. Thibault-Starzyk and M. Daturi, Chem. Soc. Rev., 2010, 39, (12), 4928 LINK https://doi.org/10.1039/B919543M
A. Chakrabarti, M. E. Ford, D. Gregory, R. Hu, C. J. Keturakis, S. Lwin, Y. Tang, Z. Yang, M. Zhu, M. A. Bañares, I. E. Wachs, Catal. Today, 2017, 283, 27 LINK https://doi.org/10.1016/j.cattod.2016.12.012
J. Dou, Z. Sun, A. A. Opalade, N. Wang, W. Fu and F. Tao, Chem. Soc. Rev., 2017, 46, (7), 2001 LINK https://doi.org/10.1039/C6CS00931J
W. Karim, A. Kleibert, U. Hartfelder, A. Balan, J. Gobrecht, J. A. van Bokhoven and Y. Ekinci, Sci. Rep., 2016, 6, 18818 LINK https://doi.org/10.1038/srep18818
M. A. Edwards, D. M. Whittle, C. Rhodes, A. M. Ward, D. Rohan, M. D. Shannon, G. J. Hutchings and C. J. Kiely, Phys. Chem. Chem. Phys., 2002, 4, (15), 3902 LINK https://doi.org/10.1039/B202347B
O. Marie, P. Bazin, and M. Daturi, ‘Vibrational Spectroscopic Studies of Catalytic Processes on Oxide Surfaces’, in “Spectroscopic Properties of Inorganic and Organometallic Compounds: Techniques, Materials and Applications”, eds. J. Yarwood, R. Douthwaite and S. B. Duckett, Vol. 42, The Royal Society of Chemistry, Cambridge, UK, 2012, pp. 34–103 LINK https://doi.org/10.1039/9781849732833-00034
K. Siegbahn, C. N. Nording, A. Fahlman, R. Nordberg, K. Hamrin, J. Hedman, G. Johansson, T. Bermark, S. E. Karlsson, I. Lindgren and B. Lindberg, “ESCA: Atomic, Molecular and Solid State Structure Studied by Means of Electron Spectroscopy”, Almquist and Wiksell, Uppsala, Sweden, 1967
K. Siegbahn and K. Edvarson, Nucl. Phys., 1956, 1, (3), 137 LINK https://doi.org/10.1016/0029-5582(56)90015-3
H. Siegbahn and K. Siegbahn, J. Electron Spectrosc. Relat. Phenom., 1973, 2, (3), 319 LINK https://doi.org/10.1016/0368-2048(73)80023-4
K. Siegbahn, Rev. Mod. Phys., 1982, 54, (3), 709 LINK https://doi.org/10.1103/RevModPhys.54.709
R. W. Joyner, M. Wyn Roberts and K. Yates, Surf. Sci., 1979, 87, (2), 501 LINK https://doi.org/10.1016/0039-6028(79)90544-2
A. Knop-Gericke, E. Kleimenov, M. Hävecker, R. Blume, D. Teschner, S. Zafeiratos, R. Schlögl, V. I. Bukhtiyarov, V. V. Kaichev, I. P. Prosvirin, A. I. Nizovskii, H. Bluhm, A. Barinov, P. Dudin and M. Kiskinova, ‘X-Ray Photoelectron Spectroscopy for Investigation of Heterogeneous Catalytic Processes’, in “Advances in Catalysis”, Ch. 4, Vol. 52, Elsevier Inc, Amsterdam, The Netherlands, 2009, pp. 213–272 LINK https://doi.org/10.1016/S0360-0564(08)00004-7
‘ISISS Station: Innovative Station for In Situ Spectroscopy’, Helmholtz Center Berlin for Materials and Energy, Berlin, Germany: https://www.helmholtz-berlin.de/pubbin/igama_output?modus=einzel&sprache=en&gid=1671 (Accessed on 15th May, 2018)
O. Karslýođlu, and H. Bluhm, ‘Ambient-Pressure X-ray Photoelectron Spectroscopy (APXPS)’, in “Operando Research in Heterogeneous Catalysis”, eds. J. Frenken and I. Groot, Springer International Publishing, Switzerland, 2017, pp. 31–57
M. Salmeron and R. Schlögl, Surf. Sci. Rep., 2008, 63, (4), 169 LINK https://doi.org/10.1016/j.surfrep.2008.01.001
H. Bluhm, M. Hävecker, A. Knop-Gericke, M. Kiskinova, R. Schlögl and M. Salmeron, MRS Bull., 2007, 32, (12), 1022 LINK https://doi.org/10.1557/mrs2007.211
D. E. Starr, Z. Liu, M. Hävecker, A. Knop-Gericke and H. Bluhm, Chem. Soc. Rev., 2013, 42, (13), 5833 LINK https://doi.org/10.1039/C3CS60057B
F. Tao, S. Zhang, L. Nguyen and X. Zhang, Chem. Soc. Rev., 2012, 41, (24), 7980 LINK https://doi.org/10.1039/C2CS35185D
D. E. Starr, H. Bluhm, Z. Liu, A. Knop-Gericke, and M. Hävecker, ‘Application of Ambient-Pressure X-ray Photoelectron Spectroscopy for the In-situ Investigation of Heterogeneous Catalytic Reactions’, in, “In-Situ Characterization of Heterogeneous Catalysts”, eds. J. A. Rodriguez, J. C. Hanson and P. J. Chupas, Ch. 12, John Wiley and Sons Inc, Hoboken, USA, 2013, pp. 315–343 LINK https://doi.org/10.1002/9781118355923.ch12
A. Wolfbeisser, B. Klötzer, L. Mayr, R. Rameshan, D. Zemlyanov, J. Bernardi, K. Fottinger and G. Rupprechter, Catal. Sci. Technol., 2015, 5, (2), 967 LINK https://doi.org/10.1039/C4CY00988F
A. Wolfbeisser, G. Kovács, S. M. Kozlov, K. Föttinger, J. Bernardi, B. Klötzer, K. M. Neyman and G. Rupprechter, Catal. Today, 2017, 283, 134 LINK https://doi.org/10.1016/j.cattod.2016.04.022
K. Föttinger and G. Rupprechter, Acc. Chem. Res., 2014, 47, (10), 3071 LINK https://doi.org/10.1021/ar500220v
K. Föttinger, Catal. Today, 2013, 208, 106 LINK https://doi.org/10.1016/j.cattod.2012.12.004
K. Föttinger, J. A. van Bokhoven, M. Nachtegaal and G. Rupprechter, J. Phys. Chem. Lett., 2011, 2, (5), 428 LINK https://doi.org/10.1021/jz101751s
A. Haghofer, K. Föttinger, F. Girgsdies, D. Teschner, A. Knop-Gericke, R. Schlögl and G. Rupprechter, J. Catal., 2012, 286, 13 LINK https://doi.org/10.1016/j.jcat.2011.10.007
A. Haghofer, D. Ferri, K. Föttinger and G. Rupprechter, ACS Catal., 2012, 2, (11), 2305 LINK https://doi.org/10.1021/cs300480c
A. Haghofer, K. Föttinger, M. Nachtegaal, M. Armbrüster and G. Rupprechter, J. Phys. Chem. C, 2012, 116, (41), 21816 LINK https://doi.org/10.1021/jp3061224
K. Fottinger, ‘PdZn Based Catalysts: Connecting Electronic and Geometric Structure with Catalytic Performance’, in “Catalysis”, eds. J. J. Spivey, Y.-F. Han and K. M. Dooley, Vol. 25, The Royal Society of Chemistry, Cambridge, UK, 2013, pp. 77–117 LINK https://doi.org/10.1039/9781849737203-00077
C. Weilach, S. M. Kozlov, H. H. Holzapfel, K. Föttinger, K. M. Neyman, and G. Rupprechter, J. Phys. Chem. C, 2012, 116, (35), 18768 LINK https://doi.org/10.1021/jp304556s
M. Armbrüster, M. Behrens, K. Föttinger, M. Friedrich, É. Gaudry, S. K. Matam and H. R. Sharma, Catal. Rev.: Sci. Eng., 2013, 55, (3), 289 LINK https://doi.org/10.1080/01614940.2013.796192
L. Li, B. Zhang, E. Kunkes, K. Föttinger, M. Armbrüster, D. S. Su, W. Wei, R. Schlögl and M. Behrens, ChemCatChem, 2012, 4, (11), 1764 LINK https://doi.org/10.1002/cctc.201200268
Glossary of Techniques
ambient pressure photoelectron spectroscopy
ambient pressure X-ray photoelectron spectroscopy
density functional theory
electron paramagnetic resonance
electron spectroscopy for chemical analysis
extended X-ray absorption fine structure
Fourier transform infrared
hard X-ray photoelectron spectroscopy
high-resolution X-ray diffraction
inelastic X-ray scattering
near-ambient pressure high energy X-ray photoelectron spectroscopy
near ambient pressure X-ray photoelectron spectroscopy
near edge X-ray absorption fine structure
| || |
Pair distribution function
photoemission electron microscopy
resonant inelastic X-ray scattering
resonant x-ray emission spectroscopy
small angle X-ray scattering
X-ray absorption near edge structure
X-ray absorption spectroscopy
X-ray magnetic circular dichroism
X-ray magnetic linear dichroism
X-ray photoelectron spectroscopy
Liliana Lukashuk is a Research Scientist at Johnson Matthey Plc, UK. Liliana obtained her PhD in chemistry from TU Wien, Vienna, Austria. At TU Wien she worked in the research group of Professor Karin Foettinger and Professor Guenther Rupprechter, focusing on operando NAP-XPS, XAS, XRD and FTIR spectroscopy studies of oxidation catalysts. Since 2015 Liliana has been working at Johnson Matthey’s Technology Centre in Chilton, UK, supporting external research programmes and the Efficient Natural Resources business sector. Her expertise lies in research and development with a background in design and synthesis of catalysts, characterisation of catalytic materials under process conditions using operando spectroscopic techniques, identifying structure-functionality relationships, reaction kinetics and kinetic modelling for development and optimisation of existing and novel catalyst products and processes.
Karin Foettinger is Associate Professor at TU Wien. She obtained her PhD from TU Wien in 2005, after which she carried out two years’ postdoctoral research with Professor Rupprechter and received a research fellowship at the Max-Planck-Society, Fritz Haber Institute, Berlin, Germany, with Professor Robert Schlögl. She has been head of the group for Technical Catalysis since 2008 and completed her Habilitation in 2013. In recent years, she has broadened her experience in catalysis by several research stays abroad, including at the University Tarragona, Spain, with Professor Francesc Medina; at the University of Glasgow, UK, with Professor David Lennon; and at the ETH Zürich with Professor Jeroen A. van Bokhoven. Karin is Chair of the Austrian Catalysis Society. Her main research interests are centred around establishing structure-performance relations, identifying reaction mechanisms and active sites by spectroscopic characterisation of catalysts under reaction conditions.