Prediction of organic contaminant rejection by nanofiltration and reverse osmosis membranes using interpretable machine learning models

Zhu, Tengyi; Tao, Cuicui; Chen, Wenxuan; Cheng, Haomiao

doi:10.1016/j.scitotenv.2022.159348

Cited by 32 publications

(7 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, Zhu et al. employed four different ML (MLR, SVM, ANN, kNN) and ensemble algorithms for contaminant removal efficiency by NF/RO membrane with a prediction efficiency range of 92.4–99.5% . in comparison with the current study, our findings showed similar predictive skills agreement with 99% accuracy.…”

Section: Resultscontrasting

confidence: 79%

“…For instance, Zhu et al employed four different ML (MLR, SVM, ANN, kNN) and ensemble algorithms for contaminant removal efficiency by NF/RO membrane with a prediction efficiency range of 92.4−99.5%. 67 in comparison with the current study, our findings showed similar predictive skills agreement with 99% accuracy. Another scenario was developed by Ignacz, and Szekely, for modeling solute rejection in organic solvent nanofiltration using traditional ML and deep learning approaches.…”

Section: Machine Learning Modelsupporting

confidence: 87%

See 1 more Smart Citation

Design and Machine Learning Prediction of In Situ Grown PDA-Stabilized MOF (UiO-66-NH₂) Membrane for Low-Pressure Separation of Emulsified Oily Wastewater

Usman,

Abba,

Baig

et al. 2024

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

Significant progress has been made in designing advanced membranes; however, persistent challenges remain due to their reduced permeation rates and a propensity for substantial fouling. These factors continue to pose significant barriers to the effective utilization of membranes in the separation of oil-in-water emulsions. Metal−organic frameworks (MOFs) are considered promising materials for such applications; however, they encounter three key challenges when applied to the separation of oil from water: (a) lack of water stability; (b) difficulty in producing defect-free membranes; and (c) unresolved issue of stabilizing the MOF separating layer on the ceramic membrane (CM) support. In this study, a defect-free hydrolytically stable zirconium-based MOF separating layer was formed through a two-step method: first, by in situ growth of UiO-66-NH 2 MOF into the voids of polydopamine (PDA)-functionalized CM during the solvothermal process, and then by facilitating the self-assembly of UiO-66-NH 2 with PDA using a pressurized dead-end assembly. A stable MOF separating layer was attained by enriching the ceramic support with amines and hydroxyl groups using PDA, which assisted in the assembly and stabilization of UiO-66-NH 2 . The PDA-s-UiO-66-NH 2 −CM membrane displayed air superhydrophilicity and underwater superoleophobicity, demonstrating its oil resistance and high antifouling behavior. The PDA-s-UiO-66-NH 2 −CM membrane has shown exceptionally high permeability and separation capacity for challenging oil-in-water emulsions. This is attributed to numerous nanochannels from the membrane and its high resistance to oil adhesion. The membranes showed excellent stability over 15 continuous test cycles, which indicates that the developed MOFs separating layers have a low tendency to be clogged by oil droplets during separation. Machine learning-based Gaussian process regression (GPR) models as nonparametric kernel-based probabilistic models were employed to predict the performance efficiency of the PDA-s-UiO-66-NH 2 −CM membrane in oil-in-water separation. The outcomes were compared with the support vector machine (SVM) and decision tree (DT) algorithm. This efficiency includes various metrics related to its separation accuracy, and the models were developed through feature engineering to identify and utilize the most significant factors affecting the membrane's performance. The results proved the reliability of GPR optimization with the highest prediction accuracy in the validation phase. The average percentage increase of the GPR model compared to the SVM and DT model was 6.11 and 42.94%, respectively.

show abstract

Section: Resultscontrasting

confidence: 79%

Section: Machine Learning Modelsupporting

confidence: 87%

Design and Machine Learning Prediction of In Situ Grown PDA-Stabilized MOF (UiO-66-NH₂) Membrane for Low-Pressure Separation of Emulsified Oily Wastewater

Usman,

Abba,

Baig

et al. 2024

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

show abstract

“…We employed the XGBoost algorithm to construct the model, primarily due to its superior interpretability and comparable predictive accuracy to other complex algorithms on small data sets. , Notably, data from the same study could involve the same or similar hidden information . To prevent potential data leakage, we adopted grouped random splitting and seed assessment during model development (Figure S3).…”

Section: Methodsmentioning

confidence: 99%

“…Machine learning (ML), a data-driven approach relying on mining patterns, correlations, and trends within data, has shown substantial potential in membrane separation, including removal prediction, − process control, − knowledge discovery, , membrane material design, , and other applications. Nevertheless, due to constraints in data quantity and quality, the direct application of ML algorithms is limited in identifying TrOC rejection mechanisms by PA membranes, especially concerning charge effects and hydrophobic interactions .…”

Section: Introductionmentioning

confidence: 99%

Understanding Rejection Mechanisms of Trace Organic Contaminants by Polyamide Membranes via Data-Knowledge Codriven Machine Learning

Wang,

Zeng,

Dai

et al. 2024

Environ. Sci. Technol.

View full text Add to dashboard Cite

Data-driven machine learning (ML) provides a promising approach to understanding and predicting the rejection of trace organic contaminants (TrOCs) by polyamide (PA). However, various confounding variables, coupled with data scarcity, restrict the direct application of data-driven ML. In this study, we developed a data-knowledge codriven ML model via domainknowledge embedding and explored its application in comprehending TrOC rejection by PA membranes. Domain-knowledge embedding enhanced both the predictive performance and the interpretability of the ML model. The contribution of key mechanisms, including size exclusion, charge effect, hydrophobic interaction, etc., that dominate the rejections of the three TrOC categories (neutral hydrophilic, neutral hydrophobic, and charged TrOCs) was quantified. Log D and molecular charge emerge as key factors contributing to the discernible variations in the rejection among the three TrOC categories. Furthermore, we quantitatively compared the TrOC rejection mechanisms between nanofiltration (NF) and reverse osmosis (RO) PA membranes. The charge effect and hydrophobic interactions possessed higher weights for NF to reject TrOCs, while the size exclusion in RO played a more important role. This study demonstrated the effectiveness of the dataknowledge codriven ML method in understanding TrOC rejection by PA membranes, providing a methodology to formulate a strategy for targeted TrOC removal.

show abstract

“…ML models have already been demonstrated for NF ( Fetanat et al, 2021 ;Hu et al, 2021 ;Jeong et al, 2021 ;Hosseinzadeh et al, 2022 ;Ignacz et al, 2023 ;Lee et al, 2023 ;Zhu et al, 2023 ), RO ( Santos et al, 2007 ;Yangali-Quintanilla et al, 2009 ;Jeong et al, 2021 ;Zhu et al, 2023 ), and gas separation ( Barnett et al, 2020 ;Yang et al, 2022 ) applications ( Galinha and Crespo 2021 ). Currently, downstream predictive models usually output the desired parameters such as incompatibilities, material permeability, or any other material or process parameter.…”

Section: Inverse Design To Transform Discovery Of Nanofiltration Memb...mentioning

confidence: 99%

Data-driven future for nanofiltration: Escaping linearity

Ignácz¹,

Beke²,

Székely³

2023

Journal of Membrane Science Letters

View full text Add to dashboard Cite

Prediction of organic contaminant rejection by nanofiltration and reverse osmosis membranes using interpretable machine learning models

Cited by 32 publications

References 68 publications

Design and Machine Learning Prediction of In Situ Grown PDA-Stabilized MOF (UiO-66-NH₂) Membrane for Low-Pressure Separation of Emulsified Oily Wastewater

Design and Machine Learning Prediction of In Situ Grown PDA-Stabilized MOF (UiO-66-NH₂) Membrane for Low-Pressure Separation of Emulsified Oily Wastewater

Understanding Rejection Mechanisms of Trace Organic Contaminants by Polyamide Membranes via Data-Knowledge Codriven Machine Learning

Data-driven future for nanofiltration: Escaping linearity

Contact Info

Product

Resources

About

Prediction of organic contaminant rejection by nanofiltration and reverse osmosis membranes using interpretable machine learning models

Cited by 32 publications

References 68 publications

Design and Machine Learning Prediction of In Situ Grown PDA-Stabilized MOF (UiO-66-NH2) Membrane for Low-Pressure Separation of Emulsified Oily Wastewater

Design and Machine Learning Prediction of In Situ Grown PDA-Stabilized MOF (UiO-66-NH2) Membrane for Low-Pressure Separation of Emulsified Oily Wastewater

Understanding Rejection Mechanisms of Trace Organic Contaminants by Polyamide Membranes via Data-Knowledge Codriven Machine Learning

Data-driven future for nanofiltration: Escaping linearity

Contact Info

Product

Resources

About

Design and Machine Learning Prediction of In Situ Grown PDA-Stabilized MOF (UiO-66-NH₂) Membrane for Low-Pressure Separation of Emulsified Oily Wastewater

Design and Machine Learning Prediction of In Situ Grown PDA-Stabilized MOF (UiO-66-NH₂) Membrane for Low-Pressure Separation of Emulsified Oily Wastewater