Machine Learning Guided Polyamide Membrane with Exceptional Solute–Solute Selectivity and Permeance

Abstract

Designing polymeric membranes with high solute−solute selectivity and permeance is important but technically challenging. Existing industrial interfacial polymerization (IP) process to fabricate polyamide-based polymeric membranes is largely empirical, which requires enormous trial-and-error experimentations to identify optimal fabrication conditions from a wide candidate space for separating a given solute pair. Herein, we developed a novel multitask machine learning (ML) model based on an artificial neural network (ANN) with skip connections and selectivity regularization to guide the design of polyamide membranes. We used limited sets of lab-collected data to obtain satisfactory model performance over four iterations by introducing human expert experience in the online learning process. Four membranes under fabrication conditions guided by the model exceeded the present upper bound for mono/divalent ion selectivity and permeance of the polymeric membranes. Moreover, we obtained new mechanistic insights into the membrane design through feature analysis of the model. Our work demonstrates a ML approach that represents a paradigm shift for high-performance polymeric membranes design.

MatGPT: A Vane of Materials Informatics from Past, Present, to Future

“…AI technology has also received more and more attention in the design and development of environmentally friendly materials, such as the development of adsorbents for capture pollutant particles and toxic elements in groundwater, and the separation membrane materials for the separation of pollutants in wastewater. [ 174,175 ]…”

“…AI technology has also received more and more attention in the design and development of environmentally friendly materials, such as the development of adsorbents for capture pollutant particles and toxic elements in groundwater, and the separation membrane materials for the separation of pollutants in wastewater. [174,175] As shown in Figure 15, Wang et al combined deep NN and transfer learning, and used 7000 adsorption energies calculated from first-principles as a training set to construct a model to explore the potential of 2D g-C 3 N 4 material as an adsorbent for HMIs (Pb 2+ , Hg 2+ , Cd 2+ ). The results show that HMIs are more inclined to adsorb at the center hole of the material than at the edge site.…”

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

“…Recently, Deng et al developed an ANN-based ML for designing polyamide membranes with a limited set of experimental laboratory data. The developed approach with feature selection indicated a paradigm shift design . In this study, the formation of the PDA-s-UiO-66-NH 2 separating layer was accomplished through (a) the introduction of a polydopamine (PDA) layer onto the ceramic membrane surface to enhance its compatibility with the MOF, (b) the in situ growth of the MOF on the PDA-coated ceramic membrane using a solvothermal process, and (c) the self-assembly of the UiO-66-NH 2 sheets onto the surface of the ceramic membrane facilitated by PDA, resulting in the formation of an exceptionally permeable separating layer on the ceramic membrane.…”

“…The developed approach with feature selection indicated a paradigm shift design. 38 In this study, the formation of the PDA-s-UiO-66-NH 2 separating layer was accomplished through (a) the introduction of a polydopamine (PDA) layer onto the ceramic membrane surface to enhance its compatibility with the MOF, (b) the in situ growth of the MOF on the PDA-coated ceramic membrane using a solvothermal process, and (c) the self-assembly of the UiO-66-NH 2 sheets onto the surface of the ceramic membrane facilitated by PDA, resulting in the formation of an exceptionally permeable separating layer on the ceramic membrane. The first part of this study involved the practical application of the PDA-s-UiO-66-NH 2 separating layer for the separation of oil-in-water emulsions stabilized by surfactants.…”

Smart Organic–Inorganic Polyoxomolybdates in Forward Osmosis for Antiviral-Drug Wastewater Treatment and Drug Reclamation

“…Membrane processes such as reverse osmosis and nanofiltration have also been used to treat medical wastewater owing to their easy operation, low secondary pollution, and small use of chemicals. − However, these membrane techniques rely heavily on applied pressures to drive the separation process. This not only considerably increases energy input but also exacerbates membrane fouling and thus damages membrane separation performance.…”