The gas—solid surface interaction potential

A major breakthrough in the study of gas—solid interactions was the development by Lennard-Jones [20] in 1924 of a potential curve for the interaction. For a gas—metal system where the interaction is strong enough to form a chemisorbed species, it was shown that an incoming molecule passes through two minima, the first a broad, shallow well (the physisorption well attributed to van der Waals forces) and the second a deeper well corresponding to the formation of a chemical bond. 

The physisorption well was originally calculated by summing the 6 12 potentials between the incoming species and each surface atom 

The potential energy well for chemisorption is associated with the more familiar chemical bond, although the valence band of the solid provides many unique features, and is reviewed by Grimley [41]. The experimental distinction between physisorbed and chemisorbed states is now readily made by photoemission studies of the combined adsorbate-adsorbent system, thus (thankfully) committing the otherwise rather theological discussions of borderline cases to past history. Chemisorption heats (see ref. 42) usually lie within the range 30 q 600 kj mole-1 and measured adatom—adsorbent atom equilibrium distances are usually very close to those observed in solid state of molecular analogues. (Such measurements are obtained by LEED or, more accurately, by surface EXAFS.) 

Two schematic combined potential energy wells for the interaction of a gaseous species with a surface are shown in Fig. 2, illustrating the importance of the crossover point of the chemisorption and physisorption wells for adsorption and desorption kinetics. In the first case, adsorption is activated in the second, it is non-activated. (There are, in fact, only a few well-documented cases of activated chemisorption.) Recently, Lundqvist et al. [43] have made detailed calculations of the potential interaction between H2 and a magnesium surface which substantiate the presence of two minima. Their work is reviewed elsewhere [44]. It must be borne in mind that diagrams such as Fig. 2 grossly oversimplify the 

---