Empirical Results Higher Order Properties

As a result of the fortuitous simplifying circumstance mentioned earlier, a molecule can be divided into action and neutral zones. The latter zone usually occupies the bulk of the molecule s size. The word action is more appropriate than reaction as used by many authors, because the concept is applicable even where no reaction occurs. Changes in properties are assumed to occur only as a result of changes in the action zone. Rules can be formulated for the effect of different substituent groups in the action zone. These are the additivity rules, or the rules of group contributions. It must be noted, however, that in the interest of greater accuracy in properties estimation, it may often be necessary to introduce higher order approximations that violate the neutrality of the neutral zone but one pays a price for this an increase in the number of empirical parameters. 

Molecular mechanics methods are also omitted from the present discussion for similar reasons. Although very sophisticated force fields are available for water (including polarizable models), most force fields for weakly bound clusters are essentially 2-body (dimer) potentials that have been adjusted empirically to reproduce bulk-phase properties. This procedure leads to very reliable descriptions of liquid water, but diminishes the quality of results for small clusters. Although force fields that include 3-body interactions are beginning to appear, ° the effects of higher order interactions (4-body, 5-body, etc.) are still untested. Furthermore, the composition of a weakly bound cluster, not just its size, is a major concern with molecular mechanics force fields. The highly refined potentials that have been developed for water are not necessarily transferable to other weak noncovalent systems (methanol, acetone, etc.). 

---