Bridging the pressure and material gap in heterogeneous catalysis

Over the past decades, surface science has gathered an unprecedented wealth of information about the structural and electronic properties of surfaces, the dynamics of clean and adsorbate covered surfaces and surface processes. The resulting profound understanding of surfaces and surface processes does not, however, automatically translate into an equally good understanding of heterogeneous catalysis as one might perhaps expect. Progress in that field has been limited mainly by two big obstacles known as the pressure and material gap in heterogeneous catalysis. The more than 13 orders of magnitude in pressure between UHV experiments and typical industrial applications and the difference in complexity between single crystal surfaces and catalysts used in so-called “real catalysis” cannot be easily bridged. Instead, concepts and strategies are required in order to successfully bridge these gaps.

In the so-called single crystal approach, this bridge is established by simply extrapolating the results from UHV/single crystal experiments to industrial conditions. A number of examples exists where this approach has been successful, the most well known being the ammonia synthesis over iron-based catalysts via the Haber–Bosch process. More frequent, however, are the cases where this extrapolation does not work, and the reasons for this failure are quite general in nature. Leaving aside the problems of heat and mass transport, which arise at higher pressure and which can be solved by chemical engineering, the pressure/material gap problem reduces to three essential points:

(i) Thermodynamic and kinetic barriers. At elevated pressures new phases may form that do not exist at lower pressure for thermodynamic reasons. If these phases determine the catalytic activity, then any extrapolation of UHV data to atmospheric pressure conditions will naturally fail. Similarly, reaction pathways that are insignificant at low pressure may become kinetically dominant at high pressure. In these cases an extrapolation of UHV data to higher pressures will also provide incorrect results.

(ii) Catalyst as dissipative structure. By definition, any heterogeneously catalyzed reaction represents a non-equilibrium system (a “dissipative structure”). The structure and chemical composition of a catalyst in operation will be largely determined by dynamical processes, while the static conditions typically applied in surface science become increasingly irrelevant with increasing rate and pressure. Self-organization will result in a spatiotemporal organization of catalytic processes, and deviations from the ideal structures stabilized by non-equilibrium conditions like strain or structural defects may determine the catalytic activity.

(iii) Lack of in situ methods and suitable model systems. From (i) and (ii), it follows that experiments under surface science conditions are of limited use for the understanding of catalysis. Instead, in situ experiments showing a catalyst in operation are required. Only these experiments allow one to establish a reliable structure–function relation. Since most surface analytical tools rely on electrons, in situ experiments under industrial conditions have been quite rare. A second problem is the structural complexity of most catalysts, which prevents an interpretation of data, i.e. the material gap problem. The strategy here is to build suitable model systems, i.e. to reduce the complexity of the system while still retaining those features of the real catalyst that are decisive for the catalytic reaction.

In the development of in situ methods suitable for catalytic reactions, significant progress has been achieved in recent years: with differentially pumped electron spectrometers, electron spectroscopy has been extended up to 1 mbar and methods utilizing photons like non-linear optical spectroscopy are surface sensitive while applicable at any pressure. Despite the experimental progress, a full mechanistic picture which includes the identification of the catalytically active phase and the active sites as well as the determination of the relevant reaction pathways would still remain, in most cases, an elusive goal without the support from theory. From the time when the catalytic system was represented by an isolated adsorbed molecule at T = 0 K, enormous progress has been made. With density functional theory (DFT), realistic catalytic systems can now be simulated and, since the chemical potential has become a parameter in these calculations, the pressure gap can be bridged in simulations.

The experimental and theoretical tools alone, however, do not suffice to solve the pressure/material gap problem; in addition, it takes a cooperative effort of scientists from different disciplines, from Surface Science, Heterogeneous Catalysis, Quantum Chemistry, and Chemical Engineering in order to be successful. This cooperation is demonstrated in this special issue of PCCP, which should provide an overview of the various concepts and strategies that have been applied to bridge the pressure and material gap in catalysis. The papers gathered in this issue illustrate the problems in this field as well as the progress that can be achieved by an interdisciplinary approach.

R. Imbihl, R. J. Behm and R. Schlögl.

---