Scholarlink

<p>The increasing complexity of modern wastewater streams enriched with recalcitrant pharmaceuticals, PFAS, and synthetic dyes poses a significant challenge for conventional membranes, which often face trade-offs among permeability, selectivity, and fouling resistance. To address these limitations, we report an interfacial nanostructure engineering strategy for sulfonated polyethersulfone (SPES) hybrid membranes by co-embedding a hydrophobic deep eutectic solvent (HES) and nanoporous holey graphene (HGN) via non-solvent-induced phase separation. Systematic variation of HGN loading identified a 0.2–0.3 wt% formulation window that produced highly porous (~ 80%) and hydrophilic (water contact angle ~ 60°) SPES/HES/HGN architectures, with improved mechanical integrity arising from combined HES-mediated matrix restructuring and low-loading HGN interfacial reinforcement. Within this range, the membrane exhibited a pure-water flux of ~ 90 LMH at 3.4&#xa0;bar, high rejection of antibiotics (up to ~ 97%), nearly complete dye removal (~ 100%), initial adsorptive PFOA removal (73% at 1&#xa0;h), and effective reduction of organic compounds, nutrients, mono- and divalent ions in real municipal wastewater from Abu Dhabi (UAE). Structure–performance analysis identified bulk porosity and surface hydrophilicity as the primary parameters governing permeability, fouling reversibility, and pharmaceutical rejection. Based on these relationships, we developed an interpretable data-driven surrogate model using a neural network ensemble to predict flux and multi-solute rejection within the experimentally sampled operating domain as functions of operating pressure, pH, and feed concentration. This integrated experimental–computational framework, deployed through an interactive predictive dashboard, provides a predictive platform for rational design and optimization of next-generation hybrid membranes for advanced wastewater treatment.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Interfacial nanostructure engineering of holey graphene–eutectic solvent hybrid membranes with data-driven surrogate modeling for advanced water purification

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年8月5日

查看全文
  • Anjali Goyal,
  • Tarek Lemaoui,
  • Ahmad S. Darwish,
  • Mahendra Kumar,
  • Fawzi Banat,
  • Hassan A. Arafat,
  • Shadi W. Hasan,
  • Inas M. Al Nashef

摘要

The increasing complexity of modern wastewater streams enriched with recalcitrant pharmaceuticals, PFAS, and synthetic dyes poses a significant challenge for conventional membranes, which often face trade-offs among permeability, selectivity, and fouling resistance. To address these limitations, we report an interfacial nanostructure engineering strategy for sulfonated polyethersulfone (SPES) hybrid membranes by co-embedding a hydrophobic deep eutectic solvent (HES) and nanoporous holey graphene (HGN) via non-solvent-induced phase separation. Systematic variation of HGN loading identified a 0.2–0.3 wt% formulation window that produced highly porous (~ 80%) and hydrophilic (water contact angle ~ 60°) SPES/HES/HGN architectures, with improved mechanical integrity arising from combined HES-mediated matrix restructuring and low-loading HGN interfacial reinforcement. Within this range, the membrane exhibited a pure-water flux of ~ 90 LMH at 3.4 bar, high rejection of antibiotics (up to ~ 97%), nearly complete dye removal (~ 100%), initial adsorptive PFOA removal (73% at 1 h), and effective reduction of organic compounds, nutrients, mono- and divalent ions in real municipal wastewater from Abu Dhabi (UAE). Structure–performance analysis identified bulk porosity and surface hydrophilicity as the primary parameters governing permeability, fouling reversibility, and pharmaceutical rejection. Based on these relationships, we developed an interpretable data-driven surrogate model using a neural network ensemble to predict flux and multi-solute rejection within the experimentally sampled operating domain as functions of operating pressure, pH, and feed concentration. This integrated experimental–computational framework, deployed through an interactive predictive dashboard, provides a predictive platform for rational design and optimization of next-generation hybrid membranes for advanced wastewater treatment.

Graphical Abstract