The Path To Molecular Orbital Theory

Abstract

The early history of the development of molecular orbital (MO) theory is reviewed. Aided by analogies to atomic spectra and based on evidence from molecular spectra in connection with the old quantum theory, a classification of electronic states of many diatomic molecules was effected early in 1926. This classification was clarified and extended with the advent of quantum mechanics, using the new concept of molecular orbitals (not so named until 1932): Hund, Mulliken. The bonding power of electrons in MOs was discussed, and with the help of the LCAO approximation, MOs were classified as bonding or antibonding. Herzberg proposed that one half the number of bonding less half the number of antibonding electrons is equal to the number of chemical bonds in not too polar diatomic molecules. As an alternative to the use of general or non-localized MOs, Hund showed the usefulness of localized σ and π MOs in describing the structures of single and multiple chemical bonds. The close correspondence of a pair of electrons in a localized MO to G. N. Lewis’s earlier concept of an electron-pair bond is pointed out. The ‘semi-united-atom’ MO concept, e.g. for N2and CO, is related to Langmuir’s earlier description of the structure of these molecules. The description of the structures of polyatomic molecules using non-localized MOs, the criterion of maximum overlap in MO theory, the electronegativity scale, and the prediction of MO ionization potentials are discussed briefly. © 1970, Walter de Gruyter. All rights reserved.