SURFACES OF MOLECULES 227 



The energy involved in the close approach of ions to within molecular 

 dimensions is far greater than lO kT — at least when the medium is one 

 of low dielectric constant. Thus we are not surprised at the fact that sub- 

 stances built up of ions are solid bodies which melt and boil only at very 

 high temperatures, and are soluble only in liquids of high dielectric 

 constant. 



We have found that the energies involved in the approach of dipole and 

 non-polar molecules (having values for \i of about lO"^*, and for a of lo"-^) 

 to within a distance of 3 X 10"^ cm. is approximately equal to kT (T = 

 300°K). Since a given molecule will usually be in contact with about a 

 dozen others, it is to be expected that the energy necessary to separate a 

 molecule from all its neighbors would be of the order of magnitude of 

 10 kT. In fact, we can reasonably expect values larger than this because of 

 the fact that the greatest effects will be produced in the deformation of the 

 molecule on its side facing the adjacent molecule (r = 3^2 molecular 

 diameter), the electric field there being about eight times greater than at 

 the center of the molecule. 



We thus have an explanation of the fact that liquids whose molecules 

 have the properties we have postulated will, in general, have boiling points 

 considerably above room temperature. When the molecules are in contact 

 the energies of the non-polar molecules are not very greatly different from 

 those of the dipole molecules — at least are of the same order of magnitude. 

 In this we have an explanation of the fact that the boiling points of liquids 

 containing dipole molecules are only moderately higher than those of non- 

 polar substances. It is possible, for example, to have liquids whose mole- 

 cules are dipoles, such as anhydrous hydrogen chloride, which boil at tem- 

 peratures far below room temperature. 



Our analysis of the known electrical forces causing the interactions 

 of molecules thus leads us inevitably to the conclusion that the forces in- 

 volved in holding together the molecules of organic liquids are acting almost 

 wholly between molecules in contact, and that in general no serious error 

 would be made in neglecting all forces which act at greater distances. This 

 conclusion is, I believe, thoroughly justified in the case of all liquids built 

 up of non-polar molecules. It applies with reasonable accuracy to most 

 liquids having molecules containing dipoles. But in general, where liquids 

 contain free ions the range of the forces is much greater, and we must 

 then take into account the Coulomb forces which act at greater distances. 



For many years, in connection with studies of adsorption, surface 

 tension, and the kinetics of heterogeneous reactions, the writer has found 

 it extremely useful to explain these phenomena in terms of actions between 

 molecules in contact. From an empirical point of view the results have 

 justified this method of attack. In many ways this conception is in accord 



