## Revista CON-CIENCIA

*versión impresa* ISSN 2310-0265

#### Resumen

CHOQUE, Rigoberto y NOGALES, Jorge. **Study of molecular properties of glyphosate using chemical-quantum **** computational methods**.

*Rev.Cs.Farm. y Bioq*[online]. 2019, vol.7, n.1, pp.41-56. ISSN 2310-0265.

**Abstract** In this research work a computational study of glyphosate was carried out, calcu-lating in detail the molecular properties of it. Through this work, the characterization of the glyphosate herbicide was achieved, presenting values of link lengths and angles as well as chemical properties of interest in QSAR studies, of energies and bioactivi-ty obtained by quantum mechano methods using molecular, semi-empirical and ab-me-chanical mechanics. initio The results obtained were analyzed by comparing the experimental data of glyphosate with those obtained computationally demonstrating a good correlation. The structure of glyphosate was plotted using the Hyperchem inter-face; This was subjected to geometric opti-mization calculations initially of Molecular Mechanics to obtain the most stable structure, subsequently all structures were ana-lyzed using the PM3 semi-empirical method to obtain more reliable energy and geom-etry values. For the *ab initio *method, the 3-21G base was used.The energy minimiza-tion of the structures was developed by the Fletcher-Reeves method of conjugate gradi-ent up to a RMS (root-mean-square) gradi-ent value less than 0,100 Kcal / (A°mol) or a maximum of 500 iteration cycles , the SCF convergence limitused was0,001 Kcal /mol. The molecular properties required in QSAR (Activity structure relationships) and bioac-tivity prediction were performed in the Molinspiration computer chemo software. The ordering of the numerical data, the graphs, regressions and the ANOVA were performed in Excel 2015. The values found for link length are very satisfactory, it was found that the F statistic (0,1574) is less than the critical value (2,479) and the P is greater than 0.05, accepting the null hypothesis, the ab-solute mean deviation was 0.27Å. The re-gression gave r^{2} coefficients of 0,89-0,86 and 0,76 for the semi-empirical method, molecular mechanics and *ab initio *respectively. For the angles, the values found are also satisfactory, the statistical F (0,226) critical value (2,43) and P is greater than 0,05, absolute mean deviation found was 5,02 degrees

**Palabras clave
:
**Glyphosate; Molecular mechanics; semiempirical; *Ab initio*; Computational chemistry; herbicide.