Surface coverage and heat of adsorption

While the surface of clean metal films appears to be homogeneous with regard to heat of adsorption and surface coverage (the latter within the limits of size of different crystallographic sites), the rate of hydrogenation of ethylene is markedly dependent on the crystal parameter. 

A most important feature of the models upon which adsorption isotherm equations, such as those above, are based is a characteristic assumption relating to heat of adsorption and surface coverage. Several factors merit consideration in this respect. 

It remains to be determined to what extent the dye adsorption technique is applicable to other substrates. No evidence was obtained for Pseudocyanine adsorption to Mn02, Fe2Os or to pure silver surfaces, although this dye can be bound to mica, lead halides, and mercury salts with formation of a /-band (61). Not only cyanines but other dye classes can yield surface spectra which may be similarly analyzed. This is specifically the case with the phthalein and azine dyes which were recommended by Fajans and by Kolthoff as adsorption indicators in potentio-metric titrations (15, 30). The techniques described are also convenient for determining rates and heats of adsorption and surface concentrations of dyes they have already found application in studies of luminescence (18) and electrophoresis (68) of silver halides as a function of dye coverage. 

While a knowledge of surface mobility is of great interest in physical adsorption, it becomes essential in chemisorption phenomena. For instance in calorimetric work a curve of differential heats of adsorption versus surface coverage will be horizontal if adsorption is localized but shows the customary slope from high to low values of the heat of adsorption if the adsorbed layer is mobile Furthermore if a chemisorbed intermediate takes part in a surface reaction (crystal growth, corrosion, catalysis), it is essential to know whether, after adsorption anywhere on the surface, it can migrate to a locus of reaction (dislocation, etch pit, active center). Yet here again, while Innumerable adsorption data have been scrutinized for their heat values, very few calculations have been made of the entropies of chemisorbed layers. A few can be found in the review of Kemball (4) and in the book of Trapnell (11). 

From the experimental adsorption isothenns both heats of adsorption and surface areas can be derived. For nonlinear adsorption isothenns the heat of adsorption varies with surface coi r e and results are expressed as an isosteric heat of adsorption, at a specified coverage, a. 

The following notes and symbols will be used in the other tables as well T, adsorption temperature Si/Al, silicon to aluminum ratio q, differential heat of adsorption n, surface coverage < inai < location of the maximum distribution of sites in the site energy distribution plot, with letters indicating the relative number of sites under the peak L, large I, intermediate S, small. 

The foregoing example of the decarburization of iron is especially simple. In many cases, however, complications occur—first, because of partial coverage of the surface by one or more adsorbed species, and second, because of a heterogeneity of the surface involving different types of adsorption sites as indicated by a strong dependence of the heat of adsorption on surface coverage. 

Another Important concept introduced by Taylor was that of heterogeneity of surface-active centers.(25-26) This stemmed from observation of R. N. Pease that minute amounts of carbon monoxide, much smaller than the amount necessary to cover the surface, were sufficient to poison the surface of a copper catalyst. Taylor proposed that there were active centers on the surface while others argued that nickel impurities segregated preferentially on the surface and acted as catalyst. The variation of the heats of adsorption with surface coverage as determined by R. Beebe was used as evidence supporting the concept of active centers. In spite of the contradictory interpretation of the same experimental data, the concept of active centers has been a fruitful one. It inspired Imaginative research in the field of metal and oxide catalysis and has its present day expression in sophisticated surface physics studies. Subsequent work by coworkers of Turkevich at Princeton refined the nature of active centers in monodisperse metal particles and crystalline oxide catalysts. 

The adsorption of cationic polymers such as nonylphenol diethylamine, NP triethylamine, NP tetraethyl pentamine. Octyl phenol (OP) diethylene triamine, and OP tetraethylene pentamine from aqueous solutions on an activated carbon showed a two-stage adsorption. " low adsorptions the molecules were postulated to have flat orientation on the surface of the carbon. A reorientation of the molecules occurred in the second stage as the adsorption increased, until at saturation the molecules were almost perpendicular to the surface. The variation of heats of adsorption with surface coverage showed two inflections (Figure 7.26). 

Fig. 4.11 Heat of adsorption versus surface coverage [Reproduced with permission from Masel, I. R., Principles of Adsorption and Reaction on Solid Surfaces, John Wiley Sons, Inc., New York, (1996) 3, p. 249]...

At low values of the bulk concentration Bcy surface coverage is proportional to this concentration, but at high values it tends toward a limit of unity. This equation was derived by Irving Langmuir in 1918 with four basic assumptions (1) the adsorption is reversible (2) the number of adsorption sites is limited, and the value of adsorption cannot exceed A° (3) the surface is homogeneous aU adsorption sites have the same heat of adsorption and hence, the same coefficient B and (4) no interaction forces exist between the adsorbed particles. The rate of adsorption is proportional to the bulk concentration and to the fraction 1-9 of vacant sites on the surface = kjil - 9), while the rate of desorption is proportional to the fraction of sites occupied Vj = kjd. In the steady state these two rates are equal. With the notation kjk = B, we obtain Eq. (10.14). 

Decrease of the heat of chemisorption with surface coverage. This is a fairly general phenomenon in adsorption on metals and of great importance in relation to catalysis, since catalytic activity tends to depend inversely on the heat of adsorption. 

Boudart (26) suggests that the presence of the electrical double layer produced by the surface dipoles can account for the observed fall in the heat of adsorption and change in work function as the surface coverage is increased. Furthermore, assuming that the dipole interaction is negligible, as will be the case for small surface coverages, the heat of adsorption and work function changes should be related by the equation... 

---