SURFACE CHEMISTRY 39 



filament is heated to a temperature above i,700°K in the presence of carbon 

 dioxide at low pressure, one molecule of carbon monoxide is liberated from 

 every molecule of carbon dioxide (See Ref. 2, p. 11 54). If this carbon 

 monoxide is pumped out from the bulb and the filament is then heated 

 to 2,300°K, a volume of carbon monoxide is slowly liberated equal to that 

 originally formed from the dioxide. This proves that each carbon dioxide 

 molecule which comes in contact with the filament and reacts, loses one 

 oxygen atom to the filament and thus produces a carbon monoxide molecule. 

 The oxygen atoms thus transferred to the filament form a monatomic film 

 of oxygen atoms covering the surface, these atoms being chemically com- 

 bined with the carbon atoms with which they are in contact, presumably by 

 double bonds. This adsorbed film on the filament is very stable at 1,700°, 

 but when heated to 2,300°, the film is destroyed, not by the evaporation of 

 the oxygen atoms, but by the breaking of the bonds between the carbon 

 atoms which are attached to the oxygen and the underlying carbon 

 atoms with which they are in contact. Thus the oxygen escapes as carbon 

 monoxide. 



From this point of view we see that we can look upon the adsorbed film 

 on the carbon filament either as consisting of adsorbed oxygen, or as an 

 adsorbed film of oriented carbon monoxide molecules, or as carbonyl 

 radicals which are chemically attached by their carbon atoms to the under- 

 I\ing carbon atoms in the filament. It is to be expected that the properties 

 of the adsorbed film would be very different if it consisted of carbon mon- 

 oxide molecules attached to the underlying surface through their oxygen 

 atoms. These experiments thus led to the concept that the properties of 

 adsorbed films should in general depend on the orientation of the molecules 

 or radicals in the film. It seemed to the writer (in 1915) that evidence of 

 such orientation could be found among the data on the surface tensions of 

 pure liquids. 



Surface Energies of Pure Liquids (17) (18). 



If a prism of liquid having a cross section of i sq. cm. is divided into 2 

 parts by an imaginary plane perpendicular to the axis of the prism, and 

 the 2 portions of the liquids are then separated, there has been an increase 

 in the surface of the liquid of 2 sq. cm. The total energy theoretically 

 required per sq. cm. of new surface to bring about this separation may be 

 called the total surface energy y^- This is related to the free surface energy 

 y, which is equivalent to the surface tension of the liquid, by the relation 



By measurements of the surface tension, the surface energy Yo can thus 

 be measured. This quantity represents the potential energy of the molecules 



