Chapter II — 13 — Structure of Water 



The immediate effect of breaking down the relatively empty ice-like 

 structure of water I would be a decrease in volume to water II (maximum 

 density) followed by an increase of a more normal type to water III where 

 the increase in spacing due to thermal agitation would more than com- 

 pensate for the change from the quartz-like to the close-packed structure. 

 The fact that several forms of ice occur under different pressures and that 

 these vary in density confirms the x-ray evidence that the structure of water 

 is not a close-packed one. Trihydrol, dihydrol, and hydrol, as visualized 

 by the earlier workers have no direct structural analogy with water I, II, 

 and ///. Bernal and Fowler explain, by means of geometrical internal 

 structure of the liquid, physical properties which the hydrol theory at- 

 tempted to explain in terms of hypothetical molecules. Ice, they conclude, 

 has a crystalline structure in which the H nuclei and hence the orientation 

 of the molecules are fixed. Every molecule is surrounded by four others 

 in a tetrahedron with the H nuclei opposite two of the neighbors and one 

 of the H nuclei of each of the remaining neighbors lying opposite a nega- 

 tive corner of the original molecule. This forms the simplest possible 

 regular physical structure. It is illustrated in Figure 7. 



Pauling (1939) differs from Bernal and Fowler stating that if the 

 water molecules in ice are oriented in a definite way so that a unique con- 

 figuration could be assigned to the crystal there should be no entropy. The 

 fact that ice retains appreciable entropy even at very low temperature in- 

 dicates that the molecules retain some freedom for motion and that the 

 crystals can exist in a number of configurations. This is further borne 

 out by the fact that above about 200° K. the dielectric constant of ice is of 

 the same order of magnitude as that of water. 



Bernal and Fowler conclude that the magnitudes of the dielectric 

 constants of water and ice can be explained by molecular rotation, but only 

 if the majority of the water molecules are not entirely free to respond by 

 orientation to the electric field. They propose that its unique properties 

 are due not only to its dipole character, but even more to the geometrical 

 structure which is the simplest that can form extended four-coordinated 

 networks. Studying ionic solutions they postulate that the strongly polariz- 

 ing ions H*, Li*, Na"^, all divalent and trivalent positive ions, and OH" and 

 F- are hydrated ; whereas NH4+, Rb*, Cs"^, and most negative ions are not. 

 The degree of ionic hydration, they conclude, depends mainly on the ionic 

 radius, and is the same in solutions as in crystals. In view of their dipole 

 character it seems to us that all ions should be hydrated, the effects of 

 NH4"*', Rb*, Cs*, and most negative ions upon the activity of water being 

 relatively less than the effects of those listed above as being strongly hy- 

 drated. 



Concerning the high mobility of hydrogen and hydroxyl ions, Bernal 

 and Fowler arrive at a new theory. Assuming H* to exist in solutions as 

 {OU-^y, they suggest that a proton may jump from one water molecule to 

 another along the coordinated network when favorable configurations oc- 

 cur. Similarly for OH" a proton under the influence of an electric field 

 may pass from H2O to the (OH)- and the systems can separate again 

 with the ion and neutral system interchanged. Thus with models built al- 

 most entirely from absorption and diffraction data, Bernal and Fowler 

 have postulated a structure for water that accounts for many of its unusual 

 properties. They picture water as an organized system of irregular, four- 

 coordinated molecular groups exhibiting three different intermolecular ar- 



