Surface Composition under Reaction Conditions

The AIF3 surfaces discussed so far have all been clean and consisted only of A1 and F ions. It is, however, well known that AIF3 will hydrolyse and adsorb water to form a siuface that also contains O and H ions. In Section 6.2.2 we showed how the energy of such a surface can be calculated as a function of H2O, HF and H2 chemical potentials. If the surface energy of aU the structures that could conceivably occur are calculated then the lowest energy surfaces can be plotted as a function of H2O, HF and H2 chemical potential. 

Competition between hydroxylation and fluorination leads to a number of stable phases. Relatively small changes in reaction conditions can alter the surface very significantly. At 300 K the (1 X 1) 3F + H2O termination is predicted to be the thermodynamically stable surface at most relevant HF and H2O partial pressures. The reactive A1 ions on this 

It is important to note that the phase diagram is only based on thermodynamic considerations. The kinetic barriers to phase transitions are not considered. For instance, there is likely to be a considerable barrier to the transition from the (1 x l)tothe( 2x y 2) structure as it requires the cleavage and formation of several Al-F bonds. a-AlFa is usually synthesized at elevated temperatures, at which the ( 2 x 2) 3F termination will dominate. It may be that the transition to the (1 x 1) 3F+H2O termination upon cooling to room temperature is kinetically hindered. Conversely, catalytically active HS-AIF3 is synthesized using sol-gel methods that proceed at lower temperatures [32,33]. Under these conditions it is speculated that structures that are similar to those found on the (1 x 1) 3F termination form. 

At 300 K a range of terminations can be expected, depending on the HF and H2O partial pressures. Under normal laboratory conditions, three H2O molecules are predicted to adsorb on the 3F termination. At 600 K, a large number of different tominations occur. Under typical reaction conditions, for instance 20% humidity and an HF partial pressure between 10 and 10 atm two H2O molecules are predicted to adsorb on the 3F termination. 

---

In general, differentiation between ensemble and electronic effects is rarely clear and often there is a contribution from both effects. However, a major goal of heterogeneous catalytic investigations using bimetallic catalysts is to establish the surface composition under reaction conditions and to attempt to... 

The charge regulation model can be used successfully for interpretating pH poten-tiometric titration data, thus enabling the determination of intrinsic equilibrium constants of surface reactions and the prediction of surface composition under various conditions. The following example demonstrates this. 

We have already mentioned that fundamental studies in catalysis often require the use of single crystals or other model systems. As catalyst characterization in academic research aims to determine the surface composition on the molecular level under the conditions where the catalyst does its work, one can in principle adopt two approaches. The first is to model the catalytic surface, for example with that of a single crystal. By using the appropriate combination of surface science tools, the desired characterization on the atomic scale is certainly possible in favorable cases. However, although one may be able to study the catalytic properties of such samples under realistic conditions (pressures of 1 atm or higher), most of the characterization is necessarily carried out in ultrahigh vacuum, and not under reaction conditions. 

The second approach is to study real catalysts with in situ techniques such as infrared and Mossbauer spectroscopy, EXAFS and XRD, under reaction conditions, or, as is more often done, under a controlled environment after quenching of the reaction. The in situ techniques, however, are not sufficiently surface specific to yield the desired atom-by-atom characterization of the surface. At best they determine the composition of the particles. 

Before any attempt to establish a correlation between the surface structure of the oxidized alloys and their CO conversion activity one must stress that the surface composition of the samples under reaction conditions may not necessarily be Identical to that determined from ESCA data. Moreover, surface nickel content estimates from ESCA relative Intensity measurements are at best seml-quantlta-tlve. This can be readily rationalized If one takes Into consideration ESCA finite escape depth, the dependence of ESCA Intensity ratio... 

The catalytic properties of a surface are determined by its composition and structure on the atomic scale. Hence, it is not sufficient to know that a surface consists of a metal and a promoter, say iron and potassium, but it is essential to know the exact structure of the iron surface, including defects, steps, etc., as well as the exact locations of the promoter atoms. Thus, from a fundamental point of view, the ultimate goal of catalyst characterization should be to look at the surface atom by atom, and under reaction conditions. The well-defined surfaces of single crystals offer the best likelihood of atom-by-atom characterization, although occasionally atomic scale information can be obtained from real catalysts under in situ conditions as well, as the examples in Chapter 9 show. 

The comparison of catalytic properties was made under identical reaction conditions, among three important candidate catalysts, namely, the Pt/y-Al203, Au/a-Fe203, and Cu Ce, x02 y systems [50], The catalytic tests were performed in the reactant feed containing CO, H2, C02, and HzO — the so-called reformate fuel. The effects of the presence of both C02 and H20 in the reactant feed on the catalytic performance (activity and selectivity) of these catalysts as well as their stability with time under reaction conditions have been studied. The composition of the prepared samples and their BET specific surface areas are presented in Table 7.6. The results obtained with the three catalysts in the presence of 15 vol% COz and of both 15 vol% COz and 10 vol% H20 in the reactant feed (with contact time wcat/v = 0.144 g sec/cm3 and X = 2.5) are shown in Figure 7.12. For comparison, the corresponding curves obtained under the same conditions but without water vapor in the feed are also shown in Figure 7.12. 

It is clear that equilibrium measurements of surface thermodynamics cannot predict surface composition under the dynamic conditions of catalytic oxidation. Nevertheless, such measurements will provide a sounder base than bulk thermodynamics for understanding the surface chemistry and permit working backward, from direct measurements of surface chemistry during reaction, to predictions concerning the microenvironment at the surface under reaction conditions. 

A stated objective of many of the reported studies of the catalytic properties of alloys has been to elucidate the significance of the band structure of the metallic phase (i.e., the energy levels of the d electrons) in determining the energetics of reaction (i.e., the value of E). While significant correlations of the values of E with band structures have been found in several instances [e.g., (25,255)], the interpretation of results is not always straightforward (237) and it may be necessary to incorporate due allowance for other factors that may exert some control over the mechanisms of reactions. Such factors include the possible presence of more than one alloy phase (207), dissolution of hydrogen in the alloy (207), and the composition and disposition of elements in the active outer surface of the alloy under reaction conditions (28,113,208). 

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. 

Impressive theoretical progress has been made in the prediction of fundamental modifications of catalyst surface structures and compositions as a function of the chemical potential of the environment in relatively simple cases (N0rskov et al., 2006 Reuter and Scheffler, 2002 Stampfl et al., 2002). This level of dynamic analysis with either single crystals or realistic polycrystalline catalyst materials has not yet been attained experimentally and certainly not in experiments with XRD under reaction conditions. There are no investigations that provide quantitative links between the phases and texture of a catalyst with its performance. All investigations discussed here can at best provide evidence relating the catalytic activity with a phase or a defect structure of a phase. 

The cleanliness and single crystallinity of electrode surfaces are not assumed even if the preparative steps outlined above are followed. The verification or identification of initial, intermediate, and final interfacial stmctures and compositions is an essential ingredient in our studies. The interfacial characterization methods employed to date have been conveniently classified in terms of whether they are conducted under reaction conditions (in situ) or outside the electrochemical cell (ex situ). In situ methods here consisted of cychc voltammetry (CV), EC-STM and DBMS. Ex situ methods included LEED, AES, and HREELS. 

It can be seen, then, that while the surface composition of a mixed metal catalyst is of critical importance to the outcome of a given reaction, there is little that may be said concerning the optimum surface concentration for a particular reaction. Even if such a prediction could be made it would be difficult to design a catalyst having the prescribed surface composition under the reaction conditions used. Much more needs to be done to optimize the use of such mixed metal catalysts, particularly in synthetically useful reactions. 

Clearly, size-selected cluster catalysts will play a key role in the future of model catalysis and will be an important tool in developing a detailed understanding of size effects in catalysis. Improvements in characterization under reaction conditions are needed to study the stability of these systems. In addition, exciting new possibilities for examining the effects of surface loading (number density) and alloy composition exist which will drive the field forward in the next decade. 

The photoelectron microscopy results described above were necessarily acquired in vacuum. Under such conditions there is no doubt that the Na is present on the Pt surface as sodium metal. However, under reaction conditions, this cannot be the case it is to be expected that the alkali would be present as a submonolayer quantity of surface compound, and indeed this is just what is observed. Furthermore, also in accordance with expectation, the nature of the alkali promoter compound is dependent on the composition of the gas atmosphere. Figure 7 shows postreaction XPS and XANES spectra acquired from Pt/Na P" alumina EP samples after exposure to reaction conditions and without exposure to laboratory atmosphere for the Pt-catalyzed reactions NO-fpropene [Figure 7(a)] and 02-fpropene [Figure 7(b)], respectively. In the first case the promoter phase consists of a mixture of NaN02 and NaNOs, in the second case it consists of Na2C03. This is important... 

---