Sustainable nanofiltration of industrial wastewater using waste-derived-graphene oxide-modified polyamide membranes: machine learning toward a smart portable water treatment unit.

Authors

Montassar T Bouzidi, Mariem M BenSlimene, Tawfik A Saleh

Published in

RSC advances. Sep 03, 2026. Epub Sep 03, 2026.

Abstract

Industrial wastewater treatment poses significant challenges because effluents vary in composition, containing dissolved salts, organic contaminants, and heavy metals. Nanofiltration (NF), employing thin-film composite (TFC) membranes, offers an energy-efficient separation platform capable of rejecting multivalent ions, organics, and dissolved metals. Here, graphene oxide (GO) was synthesized from waste graphite via a modified Hummers' method, functionalized with imidazole groups, and incorporated into polyamide TFC membranes to enhance rejection performance. The membranes were characterized, then evaluated for their efficiency in separating a mixture of pollutants in wastewater. To overcome limitations of one-factor-at-a-time experimentation, filtration experiments spanning ranges of pressure and feed composition were used to build a hybrid machine learning (ML) framework. Target-specific regression models, including k-nearest neighbors (kNN), Extreme Gradient Boosting (XGBoost), and support vector regression (SVR), and random forest (RF), were evaluated and compared based on their predictive accuracy and generalization performance. The optimized models achieved good predictive accuracy: R ² = 0.879 (kNN) for metal-ion permeate concentration, R ² = 0.998 (XGBoost) for organic-pollutant permeate concentration, and R ² = 0.983 (SVR) for salt permeate concentration. Model-guided analysis identified operating regimes that maintained high rejection while preserving acceptable permeate flux. The validated framework was integrated into a portable bench-top NF unit equipped with a web-based predictive interface, enabling real-time guidance under variable feed conditions. This hybrid ML/experimental approach provides a scalable, data-informed pathway for industrial wastewater recycling, advancing dual sustainability through graphite waste valorization and efficient water reuse.

PMID:
42694726
Bibliographic data and abstract were imported from PubMed on 04 Sep 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.