Obtención de Relaciones Cuantitativas Estructura-Propiedad en Cristales Líquidos

Abstract

Sixty nine thermotropic liquid crystal molecules were analyzed by different chemometric methods, including Principal Component Analysis (PCA), Genetic algorithms and Quantitative Structure-Property Relationships (QSPR). Mathematical models to predict nematic transition temperatures (T_N) were derived. This is the first time that Principal Component Analysis was used to predict liquid crystal properties, leading to the derivation of local (specific) models for different molecule sets. Results indicate that local models have higher prediction capabilities than global models, which is consistent to our initial assumption. This methodology will be used in the study of liquid crystals used in pharmaceutical applications.