Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

From Naturally Occurring to Engineered Nano-Biochar: Properties, Feasibility, and Challenges for Sustainable Environmental Remediation.

Environmental science & technology·2026
Same author

Resolving Nitrous Oxide Source Ambiguity in Ammonia-Oxidizing Archaea-Dominated Nitrification: Mechanistic Modeling of the Hybrid Pathway and Coupled Denitrification.

Environmental science & technology·2026
Same author

Knowledge-Based Feature Selection Substantially Enhances Data-Driven Wastewater Treatment Modeling.

Environmental science & technology·2026
Same author

1,3-Dichloro-5,5-dimethylhydantoin (DCDMH)-Driven Sludge Pretreatment for Organic Carbon Valorization: Mechanistic Insights into Controlled Oxidative Disruption and Hormesis-Mediated Metabolic Reshaping.

Environmental science & technology·2026
Same author

Sustainable Plant-Based Electrodes for Electrocatalytic Conversion of Small Molecules: From Biomass to Functional Electrodes.

Exploration (Beijing, China)·2026
Same author

Turning Motion into Methane: Electromagnetic Induction in Microbial Aggregates Enhances Wastewater Resource Recovery.

Environmental science & technology·2026

Related Experiment Video

Updated: Jul 7, 2025

Layer-by-layer Synthesis and Transfer of Freestanding Conjugated Microporous Polymer Nanomembranes
09:09

Layer-by-layer Synthesis and Transfer of Freestanding Conjugated Microporous Polymer Nanomembranes

Published on: December 15, 2015

9.4K

Machine learning for layer-by-layer nanofiltration membrane performance prediction and polymer candidate exploration.

Chen Wang1, Li Wang2, Hanwei Yu1

  • 1School of Civil and Environmental Engineering, University of Technology Sydney, Sydney, New South Wales, 2007, Australia.

Chemosphere
|December 27, 2023
PubMed
Summary

Machine learning models accurately predict layer-by-layer (LBL) nanofiltration (NF) membrane performance. This approach identified 23 promising polymer candidates for fabricating high-performance NF membranes.

Keywords:
Layer-by-layer membraneMachine learningNanofiltrationPermeabilityPolymerSelectivity

More Related Videos

Three-Dimensionally Printed Microfluidic Cross-flow System for Ultrafiltration/Nanofiltration Membrane Performance Testing
10:19

Three-Dimensionally Printed Microfluidic Cross-flow System for Ultrafiltration/Nanofiltration Membrane Performance Testing

Published on: February 13, 2016

11.4K
Electrophoretic Crystallization of Ultrathin High-performance Metal-organic Framework Membranes
07:45

Electrophoretic Crystallization of Ultrathin High-performance Metal-organic Framework Membranes

Published on: August 16, 2018

10.0K

Related Experiment Videos

Last Updated: Jul 7, 2025

Layer-by-layer Synthesis and Transfer of Freestanding Conjugated Microporous Polymer Nanomembranes
09:09

Layer-by-layer Synthesis and Transfer of Freestanding Conjugated Microporous Polymer Nanomembranes

Published on: December 15, 2015

9.4K
Three-Dimensionally Printed Microfluidic Cross-flow System for Ultrafiltration/Nanofiltration Membrane Performance Testing
10:19

Three-Dimensionally Printed Microfluidic Cross-flow System for Ultrafiltration/Nanofiltration Membrane Performance Testing

Published on: February 13, 2016

11.4K
Electrophoretic Crystallization of Ultrathin High-performance Metal-organic Framework Membranes
07:45

Electrophoretic Crystallization of Ultrathin High-performance Metal-organic Framework Membranes

Published on: August 16, 2018

10.0K

Area of Science:

  • Materials Science
  • Chemical Engineering
  • Computational Chemistry

Background:

  • Nanofiltration (NF) membrane performance is crucial for water treatment and separation processes.
  • Predicting NF membrane performance and identifying optimal polymer candidates traditionally involves extensive experimentation.
  • Developing efficient and accurate methods for NF membrane design is essential for advancing separation technologies.

Purpose of the Study:

  • To establish machine learning (ML)-based models for predicting layer-by-layer (LBL) nanofiltration (NF) membrane performance.
  • To explore and identify novel polymer candidates for high-performance LBL NF membrane fabrication.
  • To provide a computational framework for accelerating the discovery of advanced NF materials.

Main Methods:

  • Four ML models (linear, random forest, boosted tree, eXtreme Gradient Boosting) were developed to predict membrane permeability and selectivity.
  • The Shapley Additive exPlanations (SHAP) method was used to analyze the influence of fabrication conditions and polymer structure on membrane performance.
  • Polymers were represented using Morgan fingerprints, and a database search identified potential candidates based on SHAP-derived reference fingerprints.

Main Results:

  • The eXtreme Gradient Boosting (XGBoost) model achieved high prediction accuracy for membrane permeability (R²: 0.99) and selectivity (R²: 0.80).
  • SHAP analysis revealed key atomic groups influencing membrane permeability and selectivity, enabling the construction of reference polymer fingerprints.
  • 204 potential polymers were screened, leading to the selection of 23 promising polymer candidates for LBL NF membrane fabrication.

Conclusions:

  • ML models, particularly XGBoost, offer a powerful tool for accurate LBL NF membrane performance prediction.
  • The SHAP-based approach effectively links polymer structure to membrane performance, guiding the rational design of new materials.
  • This study presents a novel computational strategy for accelerating the exploration and discovery of high-performance polymers for NF applications, with publicly available code.