Vol. 13, Issue 6 (2025)

The dissociation constant (pK_a) of acids is one of the most widely used physicochemical parameters, both in industry and in fundamental research. It allows to understand numerous chemical and biological phenomena such as polymer formation, adsorption, metabolism, distribution and biological transport. Several methods can be used to determine pK_a values. This study was carried out using QSPR (Quantitative Structure-Property Relationship) modelling. Each model was developed from a series of sets of three quantum descriptors, calculated in the gaseous state at the B3LYP level with the LANL2DZ basis set, using multiple linear regression. The descriptors are the energies of the E_HOMO (energy of the highest occupied molecular orbital) and E_LUMO (energy of the lowest unoccupied molecular orbital) orbitals, the dipole moment, and the energy gap ?E (energy difference between E_HOMO and E_LUMO). Indeed, based on experimental pK_a values for benzoic, fluoro-benzoic, chloro-benzoic and bromo-benzoic acids and their isomers, four reliable, efficient and robust QSPR models were developed to theoretically predict the pK_a values of these acids in aqueous solution. In addition to their high predictive power for theoretical pK_a values, these models offer the possibility of explaining the chemical phenomena that accompany the dissociation reaction of these acids, in which the solvent plays an important role.