par Jecko, Thierry;Sutcliffe, Brian T. ;Woolley, Guy R.G.
Référence Journal of Physics A: Mathematical and Theoretical, 48, 44, 445201
Publication Publié, 2015-10
Article révisé par les pairs
Résumé : Recently there has been a renewed interest in the chemical physics literature of factorization of the position representation eigenfunctions {φ} of the molecular Schrödinger equation as originally proposed by Hunter in the 1970s. The idea is to represent φ in the form φχ where χ is purely a function of the nuclear coordinates, while φ must depend on both electron and nuclear position variables in the problem. This is a generalization of the approximate factorization originally proposed by Born and Oppenheimer, the hope being that an 'exact' representation of φ can be achieved in this form with φ and χ interpretable as 'electronic' and 'nuclear' wavefunctions respectively. We offer a mathematical analysis of these proposals that identifies ambiguities stemming mainly from the singularities in the Coulomb potential energy.