We have used density-functional theory to compute the activation energy for the dissociation of NO on two physical and two hypothetical systems: unstrained and strained Rh͑100͒ surfaces and monolayers of Rh atoms on strained and unstrained MgO͑100͒ surfaces. We find that the activation energy, relative to the gas phase, is reduced when a monolayer of Rh is placed on MgO, due both to the chemical nature of the substrate and the strain imposed by the substrate. The former effect is the dominant one, though both effects are of the same order of magnitude. We find that both effects are encapsulated in a simple quantity which we term as the "effective coordination number" ͑n e ͒; the activation energy is found to vary linearly with n e . We have compared the performance of n e as a predictor of activation energy of NO dissociation on the above-mentioned Rh surfaces with the two well-established indicators, namely, the position of the d-band center and the coadsorption energy of N and O. We find that for the present systems n e performs as well as the other two indicators.Catalysts are vital to the present-day chemical industry and are also important in the search for alternative and cleaner energy sources. However, most catalysts that are in use today have been developed by a process of trial and error, and replacing this by a program of rational design is one of the grand challenges in chemistry today. Theoretical calculations using density-functional theory ͑DFT͒ can help provide insight and guiding principles since they can supply detailed microscopic information about the various elementary steps in a reaction. 1 They also allow one to explore hypothetical systems that, even if impossible to create in the laboratory, can help one to discern patterns that can then provide guidelines for the design process.In this Brief Report, we focus on one particular reaction: the dissociation of NO, which is one of the steps in converting undesirable NO x gases to N 2 , for example, in catalytic converters in automobiles; Rh has been shown to be one of the best catalysts for this process. [2][3][4] In catalytic converters, small particles of the metal catalyst are placed on a ceramic substrate. This situation is quite different from the ideal and clean single-crystal metal surfaces that have traditionally been studied using the techniques of surface science. In recent years, however, there has been increasing attention paid to more realistic situations, such as the presence of asperities and defects, finite particle sizes, and the influence of the substrate. The substrate can affect the operation of a catalyst in various ways since it changes the environment of the metal catalyst atoms: they can have different neighbors, and may also be strained due to the presence of the substrate.Here we consider four model systems that enable us to explore the effects of changing the local environment of catalyst atoms: ͑a͒ a Rh͑100͒ surface, ͑b͒ a Rh͑100͒ surface where both the surface and substrate have been stretched ͑by 9.9%͒ to the larg...