We established three-dimensional quantitative structure-activity relationship (3D QSAR) pharmacophore models of ultraviolet (UV) absorption intensities for phthalic acid esters (PAEs). Substituent positions were obtained using the optimum pharmacophore model, and hydrophobic groups were used to generate PAE derivatives from the priority pollutants. The UV spectra of the PAE derivatives were calculated using the same method, and were used to screen for PAE derivatives with greatly enhanced UV absorption intensities. The derivatization and solvation effects on identification of PAEs and derivatives were considered. The results showed that the Hypo1 pharmacophore model had good predictive abilities and there were 14 PAE derivatives generated. The maximum UV absorption intensities of three PAE derivatives were 121.85%, 105.20%, and 191.11% respectively, higher than those of the corresponding original PAEs. Analysis of the solvation effect showed that the UV peaks red shifted, and the degree of red shift increased with increasing solvent polarity. The minimum differential wave numbers of the UV peaks after PAE derivatization and solvation were higher than the minimum resolution of the UV spectrometer (0.1 nm). A two-dimensional QSAR analysis of the enhancement mechanism can reveal the derivatization and solvation enhancement. The logK_ow values of DNOP derivatives were predicted to lower by 25.55%-34.12%.