NSC-678515

To effectively characterize natural zeolite powder (ZP) and faujasite zeolite (FAU) as adsorbents to remove a wide variety of organic micropollutants, quantitative structure-activity relationship (QSAR) models for the adsorption of zeolites were developed. For this purpose, batch isotherms were performed to measure the adsorption affinity (K_d) between zeolite and organic micropollutants, and the measured K_d values were used as a dependent variable in the QSAR modeling. In the modeling, the concept of a linear free energy relationship (LFER) was employed and used either empirically measured or in silico calculated descriptors. Modeling results based on empirical descriptors showed that log K_d values for ZP could be predicted with R² = 0.949 and standard error (SE) = 0.137 log units, and for FAU, R² = 0.895 and SE = 0.144 log units. A test set was used to validate the models developed by the training set. The predictabilities of the models for the test set were R² = 0.907 and SE = 0.209 log units for ZP and R² = 0.784 and SE = 0.236 log units for FAU, indicating that the models have reasonable robustness and predictability. Also, we showed that in silico-based descriptors could be applied to the prediction. These findings may help determine the general coverage of ZP and FAU zeolites and identify suitable applications.