Answers are grounded in this paper and cite their sources.
A quantitative structure–property relationship (QSPR) model was developed to predict equilibrium adsorption coefficients ( log k , L/kg) for 83 organic pollutants, including phenols, PAHs, and chlorophenols. The adsorption dataset used in this study originates from measurements on single-walled carbon nanotubes (SWCNTs); the model does not incorporate explicit SWCNT physicochemical properties. Instead, the model relies solely on structural descriptors of the pollutants. Using multiple linear regression with descriptor expression n M = n C + 3.80 n Cl + 6.47 n Br − 3.04 n OH − 25.5 n CHO and fragment corrections, the model achieved r 2 = 0.963 ( n = 83 ), RMSE = 0.232, and MAE = 0.185, showing better performance than five benchmark models on the available comparison data. Increased values of n C , n Cl , and n Br enhanced adsorption through π - π stacking and hydrophobic interactions, whereas n OH and n CHO decreased adsorption due to polarity and steric hindrance. Internal validation and external prediction supported the robustness of the model within the evaluated dataset. Although it excludes nanotube-specific attributes such as chirality, surface area, and environmental conditions, the new model can be applied to estimate adsorption behaviors of organic pollutants on SWCNTs within similar aqueous low-concentration systems and was further applied to 400 external compounds as a prediction within the applicability domain rather than a true experimental validation.
No takes yet. Share an insight, caveat, or question.
Keshavarz et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: