Recently steps and surface imperfections were proven by Dahl et al. [Phys. Rev. Lett. 83 (1999) 1814] to have an all-dominating effect on the N2 dissociation on Ru(0 0 0 1). In this paper the dissociative sticking of CH4 on the close-packed surfaces of Ni and Ru has been investigated in order to clarify to what extent a similar effect is present in these systems. The apparent activation energies found were 74±10 and 51±6 kJ/mol on Ni(1 1 1) and Ru(0 0 0 1) respectively. On neither Ru(0 0 0 1) nor Ni(1 1 1) did we observe a significant decrease of the overall sticking coefficient when the steps were passivated with Au. On Ni(1 1 1) sputtering the surface at 500 K with no subsequent annealing increased the initial sticking but at higher coverages this effect vanished. We interpret these results as due to the steps on both surfaces being blocked by carbon species leaving them inaccessible to impinging CH4 molecules. We conclude that in these types of experiments steps play a minor role for CH4 dissociation on Ru(0 0 0 1) and Ni(1 1 1). In a temperature programmed oxidation reaction we observe that carbon from the steps of Ru(0 0 0 1) leaves the surface at lower temperatures than carbon from the terraces. Finally we have observed a large promotion of the sticking probability of CH4 on Ru(0 0 0 1) by increasing the surface temperature at a fixed gas temperature thus lending support to a mechanism where the dissociation of methane takes place over Ru atoms displaced normal to the surface.

In this paper, results from an experimental study of the growth and reactivity of Fe overlayers on Ru(0 0 0 1) in combination with density functional theory (DFT) calculations of nitrogen dissociation on closed-packed Fe surfaces are presented. Based on these, it is suggested that the N2 dissociation on Fe(1 1 0) and Fe/Ru(0 0 0 1) surfaces is dominated by atomic steps/defects. By DFT we calculate that the activation barrier for N2 dissociation on Fe/Ru(0 0 0 1) is 36 kJ/mol lower than for Fe(1 1 0). Neither in the thermal nor the molecular beam experiments do we observe any sign of this huge activity difference between Fe overlayers on Ru(0 0 0 1) and Fe(1 1 0). From thermal data we extract an apparent activation barrier for N2 dissociation on Fe/Ru(0 0 0 1) of 28±3 kJ/mol which is significantly lower than that calculated by DFT (71 kJ/mol) on the terrace, but in good agreement with that calculated for a step site on the same surface (39 kJ/mol). The low thermal barrier and the similarity between N2 activation on Fe/Ru(0 0 0 1) and Fe(1 1 0) strongly indicates that steps and/or defects dominate the reaction on both Fe(1 1 0) and Fe/Ru(0 0 0 1). Temperature programmed desorption curves indicated a nitrogen induced reconstruction of the surface.

Using supersonic molecular beams the sticking coefficient of N-2 on Ru(0001) has been measured as a function of translational and vibrational energy. The results obtained are in striking contrast to sticking coefficients obtained under thermal conditions where steps on the Ru(0001) surface were found to dominate the dissociative sticking of N-2. The results reported here suggest that the beam is probing the terrace atoms and thus a reaction pathway, which is not viable under conditions relevant for ammonia synthesis.

The dissociative chemisorption of CH 4 on Ni overlayers on Ru(0001) has been investigated. It is found that the initial sticking probability at T =530 K is approximately a factor of 20–30 higher on a pseudomorphic overlayer of Ni than on Ni(111) and a factor of two higher than on Ru(0001) illustrating the unique properties of metal-on-metal systems. The effect of enhanced reactivity is primarily ascribed to electronic effects induced by a straining of the Ni overlayer. The enhanced reactivity towards CH 4 is accompanied by new features in the thermal desorption spectra of CO. The reactivity of the system depends strongly on the annealing temperature. Molecular beam experiments at high translational energy are qualitatively different from thermal data showing a monotonic decrease of the CH 4 sticking probability as Ni is added.

Using adsorption experiments and density functional calculations we show that ${\mathrm{N}}_{2}$ dissociation on the Ru(0001) surface is totally dominated by steps. The measured adsorption rate at the steps is at least 9 orders of magnitude higher than on the terraces at 500 K, and the corresponding calculated difference in activation energy is 1.5 eV. The low barrier at the step is shown to be due to a combination of electronic and geometrical effects. The consequences for Ru as a catalyst for ammonia synthesis are discussed.