The pressure and materials gaps in heterogeneous catalysis often complicate the extrapolation of results from surface science experiments over single crystals to real catalysis at elevated pressures and polycrystalline samples. Previous ammonia oxidation studies reported ca. 100% NO selectivity and the absence of N2O on RuO2(110) in ultrahigh vacuum (UHV) at 530 K, p(NH3) = 10-7 mbar, and O2/NH3 = 20 (Wang, Y.; Jacobi, K.; Schone, W.-D.; Ertl, G. J. Phys. Chem. B 2005, 109, 7883). Differently, our steady-state and transient experiments over polycrystalline RuO2 at ambient pressure reveal that N2 is the predominant product. The NO selectivity was as low as 6% at O2/NH3 = 2 and reached a maximum of 65% at the highest temperature (773 K) and effective oxygen-to-ammonia ratio of 140, whereas the maximum N2O selectivity was 25% at 100% NH3 conversion. Density functional theory simulations of the competing paths leading to NO, N2O, and N2 over RuO2(110) and RuO2(101) at different coverages by O- and N-containing species provided insights into the selectivity differences between the extreme operation regimes. Comparison between the (101) and (110) facets reveals that the materials effect is not likely to explain the different product distribution. Instead, the pressure effect (8 orders of magnitude higher at ambient pressure than in UHV) does. Whereas NO is formed by the direct reaction of coadsorbed N and O atoms, N2 can be formed through two different routes: direct N + N recombination or N2O decomposition. The second path is only likely at high pressures because it implies more diffusion steps of surface species, which are highly unlikely at low coverage. Thus, the main pressure effect is to facilitate alternative routes for N2 formation.
