SURFACE CHEMISTRY 49 



require that a large number of adatoms must be contained in a second 

 layer. Now these adatoms in the second layer cannot be in contact with 

 the solid surface on which the primary adsorption occurs, and therefore 

 for these atoms A2 should differ from the value of ^i that applies to the 

 adatoms in the first layer. Since A. occurs in the exponent of Eq. (13), a 

 relatively small change in X is sufficient to cause a marked difference in 

 the value of v. 



Thus when the adsorbed substance is not nearly identical in its proper- 

 ties with the substance on which adsorption occurs, we should expect the 

 value of V for atoms in the first and atoms in the second layer to be 

 greatly different from one another. Naturally two cases may arise: V2 may 

 be either greater or less than Vi. (Ref. 28, pp. 281 1 and 2815.) 



Case I. V2<vi. If V2 is less than Vi, the atoms in the second layer are 

 held by stronger forces to those in the first layer than these latter are by 

 the atoms of the underlying solid. There is therefore a tendency for the 

 atoms in the first layer to form clusters and on these clusters the second 

 and third, etc., layers begin to form long before the whole of the first 

 layer is covered (27). Such phenomena are of common occurrence. For 

 example, when mercury, cadmium or iodine is vaporized in vacuum and is 

 condensed on a glass surface, at a not too low temperature, discrete crys- 

 tals of the condensed substances form on the glass. This is a direct indi- 

 cation that these atoms exert greater forces on one another than they do 

 on the underlying glass. 



If the glass surface is maintained at a temperature so high that these 

 nuclei do not form, then the glass surface appears to reflect all the incident 

 atoms, for they evaporate from the glass surface far more rapidly than 

 they do from a cadmium surface at the same temperature. Such observa- 

 tions were made by R. W. Wood, who attempted to explain them by a 

 large reflection coefficient. He believed that experiments indicated that 

 molecules of mercury incident at low pressures on cold glass surfaces 

 were completely reflected at temperatures above — 90°C, but were com- 

 pletely condensed below that temperature. 



From our present viewpoint such a critical temperature has nothing 

 to do with a true reflection but is dependent on the fact that with a given 

 stream of mercury or cadmium atoms there is a definite temperature 

 below which evaporation of the individual adatoms cannot take place fast 

 enough to prevent accumulation sufficient to form a monatomic film, so 

 that all further evaporation at these low temperatures ceases. Experiments 

 show conclusively that at all temperatures the mercury atoms condense, 

 but that, if the temperature conditions are favorable, single atoms re- 

 evaporate from the glass at fairly low temperatures. 



It is evident therefore that those cases in which V2 is less than Vi are 



