ScienceWeek

CHEMISTRY: ON THE DESIGN OF CATALYSTS

The following points are made by Charles T. Campbell (Nature 2004 432:282):

1) Chemical conversions driven by catalysts are essential to modern society. But we must do better: sustaining industrial and economic growth, while protecting the environment, will necessitate developing more efficient processes. In particular, there is a pressing need for new solid catalysts for mixed-phase, or heterogeneous, reactions. Such reactions tend to involve large volumes, and they are the basis of, for example, more efficient ways of reducing factory and automotive emissions, and of producing cleaner fuels such as hydrogen or methanol.

2) A way forward lies in predicting how the details of a catalyst's surface structure control key kinetic parameters in the reaction mechanism. One such parameter is the activation barrier, which, if known for the rate-controlling elementary steps, allows the relevant rates to be calculated. These in turn enable accurate predictions of both the rate of production of the desired products and the branching ratios to undesired products, knowledge of both of which is essential to ensure a successful outcome in terms of energy efficiency and environmental impact.

3) Reuter et al[1] have demonstrated just how close science has come to using ab initio theoretical methods to calculate net catalytic reaction rates on complex solid surfaces. They applied an elegant method that they call "ab initio statistical mechanics", which involves two stages -- first, using first-principles quantum mechanics to calculate the activation barriers and transition-state vibrational frequencies for all the relevant elementary surface reactions; second, to couple these through statistical mechanical methods involving transition-state theory and kinetic Monte Carlo simulations of the reaction process. Vibrational frequencies for all the intermediates and transition states, also calculated quantum mechanically, are used to incorporate entropy considerations into the calculated rate constants. Monte Carlo simulations, which are in widespread use in science, allow a highly complicated system to be sampled efficiently in a number of random configurations, the upshot being a description of the system as a whole. Kinetic Monte Carlo takes this a step further, allowing efficient sampling of the tremendous range of different timescales necessary to describe all the different elementary steps, and thus simulate the kinetics of the whole system.

4) In this way, Reuter et al[1] were able to calculate net catalytic reaction rates on solid surfaces under conditions similar to those used in industrial processes. Other workers have used a related approach to investigate different types of phenomena[2], but the application to catalytic reaction rates is a particularly demanding test. Reuter et al[1] have achieved impressive accuracy in calculating the steady-state rates of the catalytic oxidation of carbon monoxide (a reaction performed by the catalytic converters of automobiles) over a model ruthenium oxide (RuO2) catalyst. Their rates agree almost perfectly with excellent experimental rate measurements, performed by a different group at the same institute[3], over a wide range of reaction conditions. The beauty of the method is that it allows one to identify which elementary steps control the reaction rates and how their rates are affected by reaction conditions and, in principle, by surface structure. This is just what is needed to guide the development of better catalysts.[3-5]

References (abridged):

1. Reuter, K., Frenkel, D. & Scheffler, M. Phys. Rev. Lett. 93, 116105 (2004)

2. Ovesson, S., Bogicevic, A. & Lundqvist, B. I. Phys. Rev. Lett. 83, 2608-2611 (1999)

3. Wang, J., Fan, C. Y., Jacobi, K. & Ertl, G. J. Phys. Chem. B 106, 3422-3427 (2002)

4. Boudart, M. Kinetics in Chemical Processes (Prentice-Hall, Engelwood Cliffs, NJ, 1968)

5. Boudart, M. & Tamaru, K. Catal. Lett. 9, 15 (1991)

Nature http://www.nature.com/nature

--------------------------------

Related Material:

CHEMISTRY: ON NANOPARTICLE GOLD CATALYSIS

The following points are made by Charles T. Campbell (Science 2004 306:234):

1) Most pollution from US automobiles is emitted in the first 5 min after startup. This is because Pt- or Pd-based catalysts currently used in automobile exhaust cleanup are inactive below about 200 deg C. Gold-based catalysts present a potential solution to this cold-startup problem. Gold nanoparticles dispersed across the surfaces of certain oxides have been shown to be amazingly active and selective as catalysts for a variety of important reactions. There is intense interest in these catalysts for carbon monoxide oxidation, because they are active at room temperature. Interestingly, the low-temperature gold catalysts are totally inactive unless the gold is in the form of particles smaller than ~8 nm in diameter (1-3). Though gold nanoparticles have been perhaps the most widely studied catalyst system in the last 2 to 3 years, the structure of the active site has remained elusive.

2) The problem has been that the active sites are on or near tiny gold particles that are themselves difficult to structurally characterize, and the gold-coated surface is very heterogeneous and thus structurally ill-defined. Chen and Goodman (4) have produced a highly active model gold catalyst where the gold is incorporated in a crystalline film, spread uniformly over a Ti2O3 surface like icing on a cake. The coated surface is therefore amenable to structural elucidation with quantitative low-energy electron diffraction and other surface crystallographies. The gold appears to be a pure, crystalline film, two atomic layers thick, with an epitaxial relationship to the underlying oxide support, itself a crystalline thin film of Ti2O3. The authors prepared this Ti2O3 as an ultrathin film on the (112) surface of a molybdenum single crystal using elegant synthetic strategies pioneered previously (5). The very high catalytic turnover rate for this gold film raises the possibility of an approximately 50-fold improvement in the performance of realistic high-area catalysts.

3) Previous work on oxide-supported gold nanoparticle catalysts has provided evidence used to support a wide range of active-site structures. Some researchers have proposed that the active sites are on the surface of the oxide (usually defects), possibly modified by the presence of nearby gold, and function together with sites on the gold nanoparticles. Others attribute the catalytic activity entirely to the presence of neutral gold atoms on the gold nanoparticles. These neutral atoms differ from atoms on bulk gold in three ways that might enhance their catalytic activity: (a) They have fewer nearest-neighbor atoms (that is, a high degree of coordinative unsaturation) and also possibly a special bonding geometry to other gold atoms that creates a more reactive orbital. (b) They exhibit quantum size effects that alter the electronic band structure of gold nanoparticles (3). (c) They undergo electronic modification by interactions with the underlying oxide that cause partial electron donation to the gold cluster. Another proposal is that positively charged gold ions on the oxide support are the key to the catalytic activity of these gold catalysts.

4) The Chen and Goodman study (4) marks an important step toward identification of the active site. Although the atomic resolution crystal structure of this highly active gold thin film has not yet been determined, the authors have provided strong evidence for the broad features of its structure using a powerful combination of surface analysis techniques (qualitative low-energy electron diffraction, x-ray photoelectron and Auger electron spectroscopies, high-resolution electron energy loss spectroscopy, and low-energy ion-scattering spectroscopy). Their results imply that the active site, at least for low-temperature CO oxidation, involves gold atoms that are nearly electrically neutral and bound to the surface via Au-Au covalent bonds and Au-Ti bonds.

References (abridged):

1. M. Haruta, Catal. Today 36, 153 (1997)

2. T. Hayashi, K. Tanaka, M. Haruta, J. Catal. 178, 566 (1998)

3. M. Valden, X. Lai, D. W. Goodman, Science 281, 1647 (1998)

4. M. S. Chen, D. W. Goodman, Science 306, 252 (2004)

5. M. C. Wu, J. S. Corneille, C. A. Estrada, J. W. He, D. W. Goodman, Chem. Phys. Lett. 182, 472 (1991)

Science http://www.sciencemag.org

--------------------------------

Related Material:

CHEMISTRY: ON NANOPARTICLE CATALYSTS

The following points are made by V. Johanek et al (Science 2004 304:1639):

1) Heterogeneous catalysts (1) have surfaces with highly specific structural and chemical properties that vary on the nanometer scale, primarily with respect to particle size, particle structure, support, and promotors. All of these aspects are carefully optimized for the reaction of interest (2). As a result, structurally and chemically complex surfaces are obtained that are extremely difficult to characterize in desired detail. Hence, molecular-level knowledge on heterogeneously catalyzed reactions is scarce, a fact which for a long time led to the depiction of heterogeneous catalysis as "black magic" (3).

2) In the last decade, however, enormous progress has been made toward a fundamental and detailed understanding of the underlying processes. This development has been driven by three main factors: (i) advances in experimental techniques now provide greatly improved microscopic (4,5) and in situ spectroscopic information, (ii) the development of model catalysts has made available catalyst surfaces with reduced complexity and extremely well-controlled structure and composition, and (iii) the development of first-principles calculations and phenomenological simulation schemes has greatly advanced the theoretical understanding of the microscopic details.

3) One of the much-discussed mysteries in heterogeneous catalysis are so-called size effects, i.e., the dependence of the global reaction kinetics on the size of the active particles. Size effects are a common phenomenon and are typically taken advantage of in catalyst optimization. Unfortunately, their exact origin remains highly ambiguous in most cases. Among the most common explanations are those that have their basis in the assumption that small particles expose specific sites with modified adsorption properties. The reason for these modifications could either be of geometrical nature (e.g., edge, corner, or other irregular adsorption sites) or of electronic origin (e.g., electron confinement effects in small aggregates or interaction with the support).

4) However, there are other pure nanoscale effects, which exclusively arise as a consequence of the limited dimension of the active particle, i.e., without any modification of the individual adsorption site. One example are so-called communication effects originally discussed by Zhdanov and Kasemo (1997). These effects arise from coupling of the kinetics between different nanofacets, occurring via surface diffusion, and may play a critical role for the global kinetics.

5) In summary: The authors demonstrate that coverage fluctuations on catalyst particles can drastically alter their macroscopic catalytic behavior. Scrutinizing the occurrence of kinetic bistabilities, the authors demonstrate by molecular beam experiments on model catalysts that macroscopically observable bistabilities vanish completely with decreasing particle size, as previously predicted by theory. The effect is attributed to fluctuation-induced transitions between two kinetic reaction regimes, with a transition rate controlled by both particle size and surface defects. The authors suggest these results indicate that fluctuation-induced effects represent a general phenomenon affecting the reaction kinetics on nanostructured surfaces.

References (abridged):

1. G. Ertl, H. Knoezinger, J. Weitkamp, Eds., Handbook of Heterogeneous Catalysis (VCH, Weinheim, Germany, 1997)

2. A. T. Bell, Science 299, 1688 (2003)

3. R. Schloegl, Angew. Chem. Int. Ed. Engl. 32, 381 (1993)

4. J. Wintterlin, Adv. Catal. 45, 131 (2001)

5. T. W. Hansen et al., Science 294, 1508 (2001)

Science http://www.sciencemag.org

ScienceWeek http://scienceweek.com