Objective To measure the toxicity of phenol, aniline, and their derivatives to algae and to assess, model and predict the toxicity using quantitative structure-activity relationship (QSAR) method. Methods Oxygen production was used as the response endpoint for assessing the toxic effects of chemicals on algal photosynthesis. The energy of the lowest unoccupied molecular orbital (ELUMO) and the energy of the highest occupied molecular orbital (E) Were obtained from the ChemOffice 2004 program using the quantum chemical method MOPAC, and the frontier orbital energy gap (ΔE) was obtained. Results The compounds exhibited a reasonably wide range of algal toxicity. The most toxic compound was α-naphthol, whereas the least toxic one was aniline. A two-descriptor model was derived from the algal toxicity and structural parameters:logl/EC50=0.268logKow-1.006ΔE+11.769 (n=20,r2=0.946). This model was stable and satisfactory for predicting toxicity. Conclusion Phenol aniline, and their derivatives axe polar narcotics. Their toxicity is greater than estimated by hydrophobicity only, and addition of the frontier orbital energy gap ΔE can significantly improve the prediction of logKow-dependont models.