Document Type : Research Paper
Nuclear Science and Technology Research Institue, Tehran, Iran
Three dimensional (3D) and two dimensional (2D) quantitative structure–property relationship (QSPR) models were established to computer-aided design of new phosphorus-containing podands extractants of the uranyl cation. GRIND methodology, where descriptors are derived from GRID molecular interaction fields (MIF), and Dragon generated descriptors were used to perform QSPR modelling of the distribution coefficient (log D). The best model for 3D-QSPR has been obtained with R2=0.93 and Q2 =0.79. Some simple 2D-QSPR models, able to correlate and predict the log D are developed. The final models satisfied a set of rigorous validation criteria and performed well in the prediction of an external test set. The results reveal the role of size and steric hindrance of hydrophobic part of extractant molecule, as well as the importance of electrostatics and charge transfer interactions in distribution coefficient of uranyl cation. This information could be very useful to design the most efficient ligands and find new extractants for uranyl ion extraction.