NMR Theoretical-Experimental Study for Structural Elucidation of Cardanol Derivatives, Homolycorines and cycloheptenones.
Name: LAYLA ROSARIO BARBOSA
Type: PhD thesis
Publication date: 27/08/2018
Advisor:
Name |
Role |
|---|---|
| VALDEMAR LACERDA JUNIOR | Advisor * |
Examining board:
| Name |
Role |
|---|---|
| ÁLVARO CUNHA NETO | Internal Examiner * |
| ARLAN DA SILVA GONÇALVES | Internal Examiner * |
| LUIZ HENRIQUE KENG QUEIROZ JÚNIOR | External Examiner * |
| PEDRO ALVES BEZERRA MORAIS | External Examiner * |
| VALDEMAR LACERDA JUNIOR | Advisor * |
Summary: This study analyzed nine organic compounds of the following classes: a) semi-synthetic derivatives of cardanol from LCC; b) homolycorines from plants of the family Amaryllidaceae; c) cycloheptenones intermediates in the synthesis of guaianes. The NMR spectra of these compounds were analyzed employing unidimensional techniques (1H NMR, 13C NMR, DEPT-135) and bidimensional (gCOSY, gHMBC, gHSQC and gNOESY). The experimental data obtained from the attribution of chemical displacement values of 1H NMR and 13C NMR and spin-spin coupling (J) were compared with theoretical data obtained through computational calculations, which analyzed the standard mean deviation (MD), standard deviation (SD) and linear correlation coefficient (R) of each compound. Systematic studies were carried out in order to validate de efficiency of the solvent effect during the optimization stage (comparing two different calculation routines) and to evaluate which method GIAO or CSGT showed better correlation with the experimental data. Also, the study evaluated if model B3LYP/cc-pVTZ showed good results during the shielding tensor (σ) calculation and spin-spin coupling (J) constant calculation. For the homolycorines the calculations were performed to evaluate the position of the substituent on C2 (α or β oriented). the mean deviation, standard deviation and linear correlation coefficient values were better for α-substituent for compounds 4 and 5 and for β-substituent for compound 6. Based on the correlation between theoretical and experimental data, it was possible to unmistakably attribute the NMR signals of the study compounds. We can also conclude that the model used (B3LYP/cc-pVTZ) was efficient to calculate the shielding tensors and coupling constant. When analyzing the solvent efficiency during the optimization stage (comparing Routines 1 and 2), we concluded that Routine 2 (which disregards solvent effect) showed to be effective to describe the chemical displacements of 13C NMR and 1H NMR and the coupling constants JH,H, which are associated to lower computational costs. If both methods for NMR property calculation are compared, we can conclude that CSGT was effective to describe the chemical displacements of 13C NMR and 1H NMR, which are associated to higher cost-effectiveness.
