Machine learning predictions of positron binding to molecules
Carregando...
Data
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal de Minas Gerais
Descrição
Tipo
Artigo de periódico
Título alternativo
Primeiro orientador
Membros da banca
Resumo
Machine-learning techniques are used to check the theoretical and experimental predictions of positron binding to general molecules. The bound or unbound character of previous calculations for polar molecules are mostly confirmed. Binding for so far unexplored polar molecules is predicted. For apolar molecules, a formula for the binding energy in terms of isotropic polarizability and ionization potential is obtained, leading to unprecedented agreement with experiments as well as prediction of previously unidentified bound systems. The role of the ionization potential is suggested as a consequence of enhanced formation of virtual positronium at short distances.
Abstract
Assunto
Elétrons
Palavras-chave
Machine learning, Positron complexes, Ionization potential
Citação
Departamento
Curso
Endereço externo
https://journals.aps.org/pra/abstract/10.1103/PhysRevA.102.052808