Deciphering the stereoselectivity of Claisen rearrangements: joint density functional theory and machine learning models

• Ano de defesa: 2024
• Autor(a) principal: Oliveira, Ana Gabriela Coelho
• Orientador(a): Oliveira, Heibbe Cristhian Benedito de
• Banca de defesa: Oliveira, Heibbe Cristhian Benedito de, Alonso, Christian Gonçalves, Muniz, Aline Silva, Silveira Neto, Brenno Amaro da, Oliveira, Guilherme Colherinhas de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Goiás
• Programa de Pós-Graduação: Programa de Pós-graduação em Química (IQ)
• Departamento: Instituto de Química - IQ (RMG)
• País: Brasil
• Palavras-chave em Português: Rearranjos de Claisen; Estereosseletividade; Estado de transição; Teoria do funcional de densidade; Decomposição de energia; Aprendizado de máquina supervisionado
• Palavras-chave em Inglês: Claisen rearrangements; Stereoselectivity; Transition state; Density functional theory; Energy decomposition; Supervised machine learning
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::QUIMICA
• Link de acesso: http://repositorio.bc.ufg.br/tede/handle/tede/13368

In the present study, the stereoselectivity of Claisen Rearrangements was addressed, focusing on the influence of two distinct electron-withdrawing groups and eight different substituents in three variants of the rearrangement: Hurd, Eschenmoser, and Johnson. Using the Curtin-Hammett principle, the energies of reactions, products, and transition states were calculated using the M062X/def2TZVPP theory level. The results indicate that kinetic effects predominantly govern the reaction equilibrium. A key aspect of our investigation involved applying Shubin’s energy decomposition analysis to the optimized transition states. This approach highlighted the significant influence of the electrostatic component on stereoselectivity, revealing its predominance over the quantum and steric components. Moreover, each transition state was divided into four fragments: the electron-withdrawing groups (Ester and Nitrile), the specific Hurd/Esch/John group (H, NMe2, and OEt), various substituents (alkyl and aryl), and the central fragment. This fragmentation allowed for a comprehensive analysis of the dipole moments of the groups and non-covalent interactions, providing insights into the electrostatic forces driving the rearrangement process. In addition, Supervised Machine Learning algorithms were employed, focusing on the analysis of electronic and geometric datasets related to the transition states. The results obtained not only elucidate the mechanisms underlying the stereoselectivity of Claisen Rearrangements but also provide a subtle understanding of the interaction between different molecular components, establishing new perspectives in advanced applications in organic synthesis.