Adsorption coverage dependence

A more dramatic type of restmctiiring occurs with the adsorption of alkali metals onto certain fee metal surfaces [39]. In this case, multilayer composite surfaces are fomied in which the alkali and metal atoms are intemiixed in an ordered stmcture. These stmctiires involve the substitution of alkali atoms into substrate sites, and the details of the stmctiires are found to be coverage-dependent. The stmctiires are influenced by the repulsion between the dipoles fomied by neighbouring alkali adsorbates and by the interactions of the alkalis with the substrate itself [40]. 

Let us consider that Ed corresponding to a peak on the desorption curve is coverage dependent, while kd (and thus the adsorption entropy) remains constant. (For the variability of kd see Section II.A.) When seeking the required function Ed (6) we refer to Eq. (8) in which the term exp (— Edf RT) exhibits the greatest variability. A set of experimental curves of the desorption rate with different initial populations n,B must be available. When plotting ln(— dn,/dt) — x ln(n ) vs 1/T, we obtain the function Ed(ne) from the slope, for the selected n, as has been dealt with in Section V. In the first approximation which is reasonable for a number of actual cases, let us take a simple linear variation of Ed with n ... 

Zero-order desorption occurs if the rate of desorption does not depend on the adsorption coverage, as seen with relatively large silver islands on a ruthenium surface (Fig. 7.7), where the Ag atoms desorb from the edges of the island. As the 0" term in Eq. (12) vanishes, the curves exhibit a clearly recognizable exponential shape on the leading side. Such situations are rare. 

In the case of monolayer adsorption, a limiting adsorption value exists that is attained when the surface is covered completely by particles of a given substance (i.e., at full monolayer coverage). The limiting adsorption value depends on the effective surface area Sj taken up by one particle 1/5. This parameter characterizes the number of sites that can be occupied by adsorbed particles on a given surface. 

The interpretation of these coverage-dependent effects involves such ideas as island formation, mixed domains of CO and 0, the mobility of CO and 0 and the adsorption of CO on oxygen-covered regions (1,18,26, 27). A deeper understanding of the roles of these processes comes from isothermal experiments involving pressure transients. 

The bottom spectrum was obtained by cycling the electrode in CO-free SnCl /HjSO solution to ensure formation of a partial Sn adlayer and then replacing the cell contents with CO-saturated solution. The v(C0) band is still observed, which shows that the Sn adatoms do not saturate the surface even in the absence of competitive CO adsorption. The intensity and frequency of the v(C0) band have both decreased, which confirms that the CO adlayer is only partially complete. There is no evidence for a change in v(C0) beyond that expected for the coverage dependence expected in acid solution. This shows that there is no strong interaction between adsorbed CO molecules and neighboring Sn adatoms, in support of the assumptions used in the adatom oxidation model discussed above. 

From the fit one obtains values of and a. Note how the electronic polarizability of the adsorbed molecules gives the absorptance a nonlinear coverage dependence. However, there exist several systems that do not follow Eqs. (2) and (3). This can be caused either by a coverage dependent change in the electronic structure, that is an additional chemical shift, or because the system exhibits clustering or the molecules occupy more the one adsorption site, since the theory assumes a random filling of the adsorbate lattice. 

Figure 13.5 Adsorption-desorption of ammonia at 280 " C on a model V2O5—WO3/TiO2 catalyst Dashed lines, inlet NH3 concentration triangles, outlet NH3 concentration solid lines, model fit with Temkin-type coverage dependence. Adapted from ref. [3].

In Eq. 38, the partial surface coverage, 0i, of the surfactant is defined as 01 = ri/Foo, where Ao is the surface excess of surfactant at saturation. K is the adsorption constant, which is a function of the surfactant and counterion adsorptions. The dependence is usually linear, yielding K = Ki + K20.1, where Ki and A are the equilibrium adsorption constants of the surfactant ions and their counterions. 

A further question regarding the adsorption registry is whether it depends on adsorption coverage, i.e., on density of adatoms this is relevant to the effects of adatom-adatom interactions. The situation is illustrated by a limited set of results, namely those for quarter-monolayer and half-monolayer adsorption of 0, S, Se and Te on Ni(lOO) in p(2 X 2) and c(2 X 2) periodicities the adsorption site is found not to depend on coverage in these cases (the nearest adatom-adatom distances are 4.90 and 3.46 A for the two coverages, respectively, compared with the largest adatom diameter of about 2.7 A for Te). 

The differences observed in the adsorption isotherm are also qualitatively and quantitatively significant for the entropy. It has been recently shown that the isotherm of adsorption of an ideal adsorbate on a heterogeneous surface can be appreciably improved by taking into account the exact form of S from Eq. (7) instead of the approximate one arising from F-H theory [22], Results for the coverage dependence of the chemical potential (adsorption isotherm) and entropy per site are shown in figs. 1-2 for various fc-mer s sizes and interaction energies [attractive (w<0) as well as repulsive (w>0)]. 

In this chapter, recent results are discussed In which the adsorption of nitric oxide and its Interaction with co-adsorbed carbon monoxide, hydrogen, and Its own dissociation products on the hexagonally close-packed (001) surface of Ru have been characterized using EELS (13,14, 15). The data are interpreted In terms of a site-dependent model for adsorption of molecular NO at 150 K. Competition between co-adsorbed species can be observed directly, and this supports and clarifies the models of adsorption site geometries proposed for the individual adsorbates. Dissociation of one of the molecular states of NO occurs preferentially at temperatures above 150 K, with a coverage-dependent activation barrier. The data are discussed in terms of their relevance to heterogeneous catalytic reduction of NO, and in terms of their relationship to the metal-nitrosyl chemistry of metallic complexes. 

The most common method used for the determination of surface area and pore size distribution is physical gas adsorption (also see 1.4.1). Nitrogen, krypton, and argon are some of the typically used adsorptives. The amount of gas adsorbed is generally determined by a volumetric technique. A gravimetric technique may be used if changes in the mass of the adsorbent itself need to be measured at the same time. The nature of the adsorption process and the shape of the equilibrium adsorption isotherm depend on the nature of the solid and its internal structure. The Brunauer-Emmett-Teller (BET) method is generally used for the analysis of the surface area based on monolayer coverage, and the Kelvin equation is used for calculation of pore size distribution. 

As a result of these various effects the strength of the adsorption may depend on the degree of occupation. The present section deals with physical adsorption phenomena at high degrees of occupation. Chemisorption phenomena at high degrees of coverage will be dealt with in Sec. IX. 

---