Dynamic and kinetic aspects of the adsorption of acrylonitrile onSi(001)−2×1

Abstract

Using scanning tunnelling microscopy (STM), photoelectron and photoabsorption spectroscopies, we have examined how acrylonitrile $({\mathrm{H}}_{2}\mathrm{C}\mathrm{C}\mathrm{H}\text{\ensuremath{-}}\mathrm{C}\mathrm{N})$ reacts with the $\mathrm{Si}(001)\text{\ensuremath{-}}2\ifmmode\times\else\texttimes\fi{}1$ surface for coverages ranging from $\ensuremath{\sim}{10}^{12}\phantom{\rule{0.3em}{0ex}}\text{molecules}∕{\mathrm{cm}}^{2}\phantom{\rule{0.3em}{0ex}}\text{to}\phantom{\rule{0.3em}{0ex}}\ensuremath{\sim}{10}^{14}\phantom{\rule{0.3em}{0ex}}\text{molecules}∕{\mathrm{cm}}^{2}$. At $300\phantom{\rule{0.3em}{0ex}}\mathrm{K}$, in the very low coverage regime (below ${10}^{13}\phantom{\rule{0.3em}{0ex}}\text{molecules}∕{\mathrm{cm}}^{2}$), filled- and empty-state STM images show that the molecule bridges, via its $\ensuremath{\beta}$ carbon and nitrogen ends, two silicon dangling bonds, across the trench separating two dimer rows. A cumulative-double-bond unit $(\mathrm{C}\mathrm{C}\mathrm{N})$ is formed. The $300\phantom{\rule{0.3em}{0ex}}\mathrm{K}$ STM image results from the dynamic flipping of the molecule between two equivalent equilibrium positions, which can be seen when the molecular motion is slowed down at $80\phantom{\rule{0.3em}{0ex}}\mathrm{K}$. For coverages larger than ${10}^{13}\phantom{\rule{0.3em}{0ex}}\text{molecules}∕{\mathrm{cm}}^{2}$, for which STM does not show ordered adsorption any more, the adsorption kinetics were observed in real-time using valence band photoemission and resonant Auger yield, associated with N $1s$ x-ray absorption spectroscopy (NEXAFS). At $300\phantom{\rule{0.3em}{0ex}}\mathrm{K}$, these techniques point to a situation more complex than the one explored by STM at very low coverage. Three species (cyano-bonded, vinyl-bonded, and cumulative-double-bond species) are detected. Their distribution does not vary with increasing coverage. All dimerization-related surface states are quenched at saturation. The uptake rates versus coverage relationship points to the presence of a mobile precursor. Finally, the paper discusses a possible mechanism leading to the formation of cross-trench $\mathrm{C}\mathrm{C}\mathrm{N}$ unit at low coverage, and the reasons why the product branching ratio changes with increasing coverage.