150 Variation of Surface Energy with Temperature. [Mar. 23, 



III. " On the Variation of Surface Energy with Temperature." 

 By William Ramsay, Ph.D., F.R.S., and John Shields, 

 B.Sc., Ph.D. Received March 14, 1893. 



(Abstract.) 



It is shown that a close analogy exists between the equation for 

 gases, 



pv = RT, 



and an equation expressing the relation of surface energy to tempera- 

 ture, 



fyS = /ct, 



where 7 is surface tension ; s, surface ; k, a constant ; and t, tempe- 

 rature measured downwards from a point about 6° below the 

 critical point of the fluid. As the origin of T in the gaseous equation 

 is where p = 0, so the origin of t should be where 7 = 0. Correct- 

 ing the above equation so that 7 shall represent the number of degrees 

 measured downwards from the critical point, the equjation becomes 



7s = K (j—d). 



But even this equation does not express the whole truth. For at 

 temperatures less than 30° below the critical temperature, the 

 relation between surface energy and temperature is not a rectilinear 

 one ; a correction is therefore introduced in the form of a second 

 term, which becomes insignificant at temperatures more than 

 25° or 30° t ; it is 



7S = KT — Kd (1 — 10 _XT ). 



The liquids examined were : ether, methyl formate, ethyl acetate, 

 carbon tetrachloride, benzene, chlorobenzene, acetic acid, and methyl 

 and ethyl alcohols ; in fact, the only ones for which data are avail- 

 able. For, in order to calculate 7 from the rise in a capillary tube, 

 it is necessary to know the density of the orthobaric liquid and gas ; 

 and reliable data exist only for these liquids, and for a few others 

 which resemble them closely, e.g., fluorobenzene, bromobenzene, &c. 

 Also to calculate s, i.e., molecular surface, it is necessary to know the 

 molecular volume of the liquid, and to raise it to the power § rds. 

 Hence == s, or molecular surface; i.e., it is possible to compare 

 different liquids on the surfaces of which equal numbers of molecules 

 lie. 



Measurements were made at —89*8°, the boiling point of nitrous 

 oxide under atmospheric pressure, with ether, methyl formate, ethyl 

 acetate, and the two alcohols ; the other substances are solid at that low 



