EVAPORATION OF ATOMS 343 



areas from the vapor phase is thus ii(^„i — da), where dai is the maximum 

 vakie of da- Equating the rates of evaporation and condensation we have 



Langmuir has derived a general equation for the Hfe of adatoms which 

 may readily be put in the form ^" 



r/aa,= l.nX\0-''M'^TM-'e'"' (15) 



where Tm is the mean temperature over the range of validity of the 

 equation. Taking Tm = 800° and M — 133 (the atomic weight of Cs) and 

 eliminating Oai/Xa between this equation and Eq. (14) we obtain 



\n[A0ax-ea)/9a'] = (>S.S-b/T. 

 With common logarithms this becomes 



logio [M(^ai-<?.)/^a] = 28.58 -B'/r. (16) 



In these equations 6ai and B' should be constants the value we now 

 wish to determine from our experimental data on ^obs as a function of |,i 

 and T for very dilute caesium films. These data are given in the first three 

 columns of Table II. 



TABLE II 



Relation of Adsorption on Active and on Normal Areas for Dilute 



Caesium Films 



A series of trials has shown that the best agreement of these data with 

 Eq. (16) is obtained if we take dai = 0.005 as found in the preliminary 

 analysis which led to Eq. (12). 



Since \i can be expressed as a function of 6„ and T by Eqs. (6) to (10) 



or by Table I, we can by a series of approximations (or graphically) 



determine On for each set of experimental values of fx and T. The values 



calculated in this way are given in the 6th column of Table II. Columns 



4 and 5 contain values of Aa and B^ used in these calculations. 



^° See reference 2. The above Eq. (15) is obtained by combining Eqs. (5), (6), 

 (7) and (27) on pages 2799 and 2806. 



