Descriptor-based analysis applied to HCN synthesis from NH3 and CH4.

The design of solid metal catalysts using theoretical methods has been a long-standing goal in heterogeneous catalysis. 2] Recent developments in methodology and computer technology as well as the establishment of a descriptor-based approach for the analysis of reaction mechanisms and trends across the periodic table allow for the fast screening for new catalytic materials and have lead to first examples of computational discoveries of new materials. The underlying principles of the descriptor-based approach are the existence of relations between the surface electronic structure, adsorption energies and activation barriers that result in volcanoshaped activity plots as function of simple descriptors, such as atomic binding energies or the d-band center. 6, 9-17] Linear scaling relations have been established between the adsorption energies of hydrogen-containing molecules such as CHx, NHx, OHx and SHx and the C, N O and S adsorption energies on transition-metal surfaces. Transition-state energies have also been shown to scale linearly with adsorption energies in a similar fashion. 16, 19, 20] Recently, a single transition state scaling relation has been identified for a large number of C-C, C-O, C-N, N-O, N-N, and O-O coupling reactions. The scaling relations provide a powerful tool for the investigation of reaction mechanisms and the prediction of potential energy surfaces. They limit the number of independent variables to a few, typically adsorption energies of key atoms. Using this information as input to a microkinetic model provides an understanding of trends in catalytic activity across the transition metals. In most cases a volcano-shaped relation between activity and the key variables, the descriptors, is observed. In the present paper we will provide an example of the approach outlined above and show how one can obtain an understanding of activity/selectivity trends of a reaction with just a few new calculations. We take the synthesis of hydrogen cyanide (HCN), a versatile synthetic building block in organic chemistry, as an example. For the catalytic production of HCN from methane (CH4) and ammonia (NH3) there exist two main processes: the Andrussow and the Degussa (BMA) process. Both processes use Pt gauze catalysts and are run at temperatures up to 1573 K. The main reason for the high reaction temperature is the endothermicity (ΔH = 251 kJ/mol(HCN)) of the reaction CH4 + NH3 → HCN + 3 H2, whereas the competing side reaction 2 NH3 → N2 + 3 H2 is less endothermic (ΔH = 46 kJ/mol(NH3)). Hence, NH3 decomposition is thermodynamically favored at low temperatures. In contrast to the Degussa process where the necessary heat is supplied externally, O2 is added to the feed of the Andrussow process and its reaction with H2 supplies the necessary energy to drive the endothermic HCN formation. The industrially used Pt/Rh gauze catalysts reach their optimal activity only after an extended activation period under reaction conditions. SEM and STM characterization of the catalyst have shown that during this activation period the catalysts surface area is increased (roughening) and random faceting occurs. As an approximation of the active site on this surface, we use the stepped (211) facet, which has also been shown to dominate the catalytic activity in other reactions.. Pt is the active catalytic component of the Pt/Rh catalyst, while the role of Rh is to increase the long term stability. We consider the following elementary steps, including 9 different routes for C-N coupling between CHx and NHy with x,y = 0,1,2, where X* denotes an adsorbed species and * represents an empty surface site: