A Hybrid Response Surface Methodology and Machine Learning Framework for Quantifying Effects of Physicochemical Parameters on PFAS Distribution

Authors

• Harsh V. Patel, North Carolina A&T University
• Jazmin Green, North Carolina A&T University
• Hyoshin Park, Old Dominion UniversityFollow
• Stephanie Luster-Teasley Pass, North Carolina A&T University
• Renzun Zhao, North Carolina A&T University

Document Type

Article

Publication Date

2026

DOI

10.1021/acsestwater.5c01162

Publication Title

ACS ES&T Water

Volume

Advance online publication

Pages

12 pp.

Abstract

Predicting PFAS adsorption across diverse adsorbents and environmental matrices remains challenging because adsorbent physicochemical properties, PFAS molecular descriptors, and operational conditions simultaneously influence adsorption. This study develops and evaluates a unified hybrid modeling framework that integrates Response Surface Model (RSM) with machine-learning algorithms to quantify how six key variables, surface area, Log K_ow, pH_pzc, pK_a, log dose, and log-initial concentration, affect PFAS distribution coefficients (Log K_d). A data set of more than 1000 adsorption observations spanning 15 PFAS compounds, multiple adsorbent types, and a broad operational range was compiled and preprocessed using mode imputation and log transformation. Model performance was evaluated using an 80/20 split and Leave-One-PFAS-Out (LOPO) validation. Gradient Boosting performed best in the 80/20 scenario (R² = 0.93; RMSE = 0.25), whereas Random Forest achieved the highest performance under LOPO validation (R² = 0.30; RMSE = 0.78), highlighting the challenge of compound-wise generalization. Feature-importance analyses consistently identify log-dose and log-initial concentration as key predictors, followed by surface area and pHpzc. Partial-dependence analysis and RSM surfaces revealed nonlinear relationships and interaction effects among adsorption drivers. Overall, the framework provides a transparent approach for predicting PFAS adsorption while disentangling the coupled roles of adsorbent properties, PFAS chemistry, and operational conditions.

Rights

© 2026 The Authors.

This publication is licensed under Creative Commons Attribution 4.0 International (CC-BY 4.0).

ORCID

0000-0002-1490-5404 (Park)

Original Publication Citation

Patel, H. V., Green, J., Park, H., Luster-Teasley Pass, S., & Zhao, R. (2026). A hybrid response surface methodology and machine learning framework for quantifying effects of physicochemical parameters on PFAS distribution. ACS ES&T Water. Advance online publication. https://doi.org/10.1021/acsestwater.5c01162

Repository Citation

Patel, Harsh V.; Green, Jazmin; Park, Hyoshin; Pass, Stephanie Luster-Teasley; and Zhao, Renzun, "A Hybrid Response Surface Methodology and Machine Learning Framework for Quantifying Effects of Physicochemical Parameters on PFAS Distribution" (2026). Engineering Management & Systems Engineering Faculty Publications. 277.
https://digitalcommons.odu.edu/emse_fac_pubs/277