Quantitative Structure-Activity Relationship (QSAR) analysis of functionalized triblock copolymers with applications as dehydrating agents of crude oil

QSAR analysis was applied to obtain the water removal efficiency (WRE) values of triblock copolymers functionalized with tertiary amines, employed as demulsifier agents of crude oil. Taking into account the evaluation of F- and G-series reported by Zamora et al., the H-series was synthesized based on a triblock copolymer of 1920 g mol(-1), and evaluated in CM1 and CM2 crude oils (7.5 and 18.7 degrees API). In addition, quantum and QSAR parameters, physicochemical properties of crude oil were included in the QSAR analysis. 66 WRE values were considered in the training set. Q(2) value was established to weigh up an equation as acceptable, achieving eight equations, T1 - T8, for the training set. About external validation, F-, G- and H-series were assessed in CM3 heavy crude oil (16.9 degrees API), considering an overall of 33 WRE values for the validation. For the external validation, the coefficient of the prediction set (R-EXT(2)), the predictive squared correlation coefficient (Q(Fn)(2)), the metric value (r(m)(2)), the concordance correlation coefficient (CCC), and the regression line slopes (k and k') criteria were obtained. Only two equations, T1 and T8, complied with threshold established in the validation criteria. About r(m)(2) criterion, T8 equation gave a better correlation, 0.62, than that of T1 equation, 0.58. However, both equations could be optimal to calculate the water removal efficiency. Regarding physicochemical parameters of crude oil, number average molecular mass and run-off temperature were the unique parameters showing a clear effect over the calculated water removal efficiency.