Making and breaking of chemisorption bonds

Abstract

A brief review of recent contributions from surface physics to the understanding of chemisorption and desorption is given. The first question that is addressed is: What happens to a molecule when itcomes close to a metal surface? On the basis of self-consistent calculations for atoms [1,2] and molecules [3] on simple metal surfaces general trends and concepts are described and a scenario for H 2 on Mg(0001) is presented [4]. Except for the physisorption occuring at distances ∼ 3 A ̊ or larger outside the surface the molecule-surface interaction is characterized by substantial redistributions of the electrons. Molecular electron energy levels are shifted and broadened. In particular, the antibonding affinity level of H 2 is shifted to lower energies. The spatial redistribution of the electrons is beneficial for an increased electron-proton interaction, and thereby for the chemisorption energy of H, i.e. for the constituents of H 2. The intramolecular bond between the constituents, on the other hand, is weakened on the surface by at least one order of magnitude and at some instances even becomes repulsive. The reason for this breaking of the molecular bond is the downward shift at the antibonding molecular-orbital resonance allowing its partial filling with conduction electrons from the metal [5]. Also for the dynamic aspects of the chemisorption, the variation of electron structure upon the molecular approach to the surface is of key importance. For instance, there can be electronic mechanisms for the dissipation of the molecular translational energy, required for the trapping of the molecule [6]. For the emission of ions, occuring in, e.g., sputtering, the variation of the electron structure of the emitted particle is also very important [7]. The second question concerns the desorption of adsorbed species upon electron or photon impact. A brief review is given of proposed mechanisms for electron- and photon-stimulated desorption, ESD and PSD, respectively [8]. The Redhead-Gomer-Menzel mechanism for ESD [9] assumes the incoming electron to excite a valence electron contributing to the adsorbate-substrate bond into an antibonding orbital. This changes the attractive potential energy curve of the adsorbate into a repulsive one, and the adparticle finds it energetically favourable to leave the surface, i.e., to desorb. The mechanism of Knotek and Feibelman [10] originated from studies of O + desorption from TiO 2, where the O + ion yield was found to increase considerably at incident electron energies that coincide with ionization potentials of core electrons of the metal ion. Their mechanism is an Auger decay model, where after the creation of a core hole there is a deexcitation by an interatomic Auger process, where due to the suddenness of the process and the Coulomb repulsion between electrons the Auger electron is likely to be followed by a second emitted electron. The resulting O + ion ends up on a repulsive potential-energy curve and desorbs. The mechanism has later been extended to intraatomic Auger processes and PSD and successfully applied to selective breaking of bonds [8].