Energy-Efficient PFAS Defluorination via Reactor-Scale VUV/UVC Boron Nitride Photocatalysis for Semiconductor Fabrication Wastewater

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Boron nitride (BN) is a commercially available photocatalyst with demonstrated activity for degrading per- and polyfluoroalkyl substances (PFAS), but reactor-scale performance has not been reported. Here, we evaluate BN photocatalysis (BN + UV) in a 16 L recirculating reactor equipped with low-pressure mercury lamps emitting vacuum ultraviolet (VUV, 185 nm) and UV–C (254 nm) radiation. Using perfluorooctanoic acid (PFOA) as a model compound at mg/L concentrations, BN + UV achieved >99.8% removal within 1 h and 88% defluorination after 7 h, outperforming direct photolysis (UV-only; 92% removal, 30% defluorination) and TiO2 photocatalysis (TiO2 + UV; 63% removal, 10% defluorination) under identical conditions. A reactor-scale kinetic model yielded an intrinsic illuminated-zone rate constant (kA) of 110 h–1 for BN + UV, 11-fold greater than photolysis and 24-fold greater than TiO2 + UV. BN + UV correspondingly presented the lowest electrical energy per order removed (EEO = 1.82 kWh m–3) and per order defluorinated (EEOD = 6.68 kWh m–3) based on lamp power. Furthermore, BN + UV degraded a 13-compound PFAS mixture in simulated semiconductor fabrication wastewater at μg/L levels. Long-chain compounds were completely removed within 24 h, while perfluorobutanesulfonic acid (PFBS) persisted, consistent with the known recalcitrance of short-chain sulfonates. These results demonstrate the energy efficiency of BN photocatalysis for destructive PFAS treatment in complex industrial matrices.
Funding Information
  • Directorate for Technology, Innovation and Partnerships (TI-2314154)
  • Division of Engineering Education and Centers (EEC-1449500)
  • Bureau of Reclamation (R23AS00282)
  • Strategic Environmental Research and Development Program (ER22-3258)
  • ALTRA-SANEXEN
  • WaTER Institute, Rice University

This publication has 39 references indexed in Scilit: