Zero coverage sticking probabilities

The initial sticking probability is sensitively dependent on the gas-metal system, spanning the range 1 s0 10-7, although for H2, 02 and CO and N2 on metals and at substrate temperatures where chemisorption does occur, sticking probabilities are generally between 0.1 and 1. Trends across the periodic table are examined in Fig. 16 and 17, where the most reliable data for polycrystalline ribbons, films or foils of the transition 

Adsorption parameters initial sticking probability, s0, (at 300 K unless otherwise indicated) and saturation coverage, Nmax The variation of s with N is indicated by the letters A—E, illustrated in Fig. 22 (p. 56). 

Much has been written about the high reactivity of steps at metal surfaces and there are several examples in chemisorption. For example, the 110 plane of W has a low sticking probability ( 3 x 10 3) for nitrogen chemisorption at 300 K, but on the stepped 320 plane, with 110 terraces, s0 = 0.70. On the other hand, a singular, step-free W 100 surface is also very reactive (s0 = 0.59) and it has been suggested that the reactivity of stepped planes such as 320 is related to the structural similarity of the steps to the 100 plane itself [47]. A further dramatic illustration of the dependence of s0 on the presence of steps is demonstrated by the work of Salmeron et al. [353] on the stepped Pt surface. Using a molecular beam source, they demonstrated that the reactivity for dissociative chemisorption of H2 (as indicated by the formation of HD from a mixture of H2 and D2) was 7 times higher when the beam was directed at the steps than when it was directed over the steps, as shown in Fig. 20. 

Several studies have been made of the dependence of s0 on crystal and gas temperature. Since these results are critically important in establishing the adsorption mechanism, the detailed presentation of results is left until 

A few studies have been made of the dependence of s0 on incidence angle using molecular beam techniques. Apart from the study on a stepped surface illustrated in Fig. 20, relatively small effects have been noted. King and Wells [46] found no dependence (within 0.01) forN2 on polycrystalline W and on W 100. Steinbriichel and Schmidt [359] found significant variations only for H2 on W 100 (Fig. 21) and, in this case, it is probably again reasonable to assume that the results are dominated by step effects. Recently, Bickley et al. [360] have compared accurate 

---

Figure 6 shows the variation in the relative surface concentration of adsorbed N atoms with N2 exposure at 693 K for the Fe(llO), Fe(lOO), and Fe(lll) surfaces from which data for the initial (i.e., extrapolated to zero coverage) sticking coefficient was found to vary from 7 x 10 to 2 x 10 to 4 x 10 for the respective surfaces (21). The value for Fe(lll) is of the same order of magnitude as that derived by Emmett and Brunauer (22) for the doubly promoted catalyst. Even more remarkably, these numbers are also of the same order as the reaction probabilities derived in Somoijai and co-workers work (26). This agreement shows that kinetic parameters derived from single-crystal studies are transferable across the pressure gap (rate measurements were performed at 20 bar, whereas that on adsorption kinetics at 10 mbar ) and are also consistent with the behavior of a real catalyst. 

The probability for sticking is known as the sticking coefficient, S. Usually,. S decreases with coverage. Thus, the sticking coefficient at zero coverage, the so-called initial sticking coefficient,. S q, reflects the interaction of a molecule with the bare surface. 

Prior to a discussion of adsorption kinetic rate laws and mechanistic theories, we review the very large amount of data that has been gathered in recent years. We consider first the values reported in the literature for the sticking probabilities at zero coverage, s0, for a variety of gas—metal systems secondly, the variations of s0 with gas temperature, Tg, and surface temperature, Ts and thirdly, the dependence of s on surface coverage. 

Experimental evidence for the existence of intrinsic precursor states is rather more difficult to come by. The common observation that the initial sticking probability, s0, often decreases with increaing substrate temperature is consistent with the existence of such a state, as discussed here. Indirect evidence is also provided by molecular beam studies, for example, Hayward and Walters [401] (for H2 on W 001 ) and Engel [402] (for 02 on Pd 111 ) have observed scattered particle intensity distributions which, even at a fractional coverage in the chemisorbed layer close to zero, exhibit a strong directional lobe in the specular direction superimposed on a cosine law distribution. The specular lobe clearly contains molecules scattered at the first collision, while the cosine law component is most readily attributed to the particles which are trapped in the precursor state and then scattered back into the gas phase. Of... 

---