Linear Solvation Energy Relationships. Local Empirical Rules -- or Fundamental Laws of Chemistry? A Reply to the Chemometricians.

Abstract

Sjostrom and Wold have suggested that, rather than being considered as combinations of fundamental effects, linear free energy, relationships (LFERs) and linear solvation energy relationships (LSERs) should be regarded as local empirical models of similarity or locally valid linerizations of complicated relationships. In reply to the chemometricians, it is shown that hundreds of physicochemical properties and reactivity parameters of many diverse types are described by equations of the following forms: For solubility properties of multiple solutes in single solvents, or distribution between pairs of solvents, XYZ=XYZo+m.hivin.V2/100+s.pi.*2+a(.alpha.m)2+b(.beta.m)2, and for effects of multiple solvents on single indicators or combinations of reactants, XYZ=XYZo+h(.delta.H2)1+s.pi.*1+a.alpha.1+b.beta.1, where .hivin.V is the solute molar volume, .delta.H is the solvent Hildebrand solubility parameter, and .pi.*, .alpha., and .beta. are the solvatochromic parameters that scale solute and solvent dipolarity/polarizability, hydrogen bond donor acidity, and hydrogen bond acceptor basicity. Evidence is offered that, if the chemometricians are correct, the term local must be stretched to include every area in chemistry and many in biology, where physicochemical, biological, toxicological, and pharmacological properties depend on interactions between solutes and solvents.