A detailed modelling and simulation framework is developed using COMSOL Multiphysics to solve mass and energy balance equations that describe transport in porous media. The model incorporates adsorbent hydrodynamics, mass transfer, and thermal effects to replicate realistic operating conditions. PFAS adsorption onto polymer matrices is characterized using Langmuir and Freundlich isotherms, allowing for an accurate representation of adsorption behavior.
A Bayesian optimization algorithm is implemented in MATLAB to optimize process performance, interfacing with COMSOL for iterative evaluation. This approach is well-suited for complex, computationally expensive optimization problems, leveraging a Gaussian process surrogate model to explore and refine system parameters efficiently. Sensitivity analyses assess the robustness of optimized solutions under varying conditions while key operational factors such as contact time, process stream composition, flow rate, and system configuration are systematically explored. Additionally, the impact of polymer adsorbent properties on overall system performance is evaluated to ensure feasibility across different semiconductor wastewater treatment scenarios.
This integrated approach provides valuable insights into the trade-offs between cost and environmental sustainability, supporting the development of scalable and effective PFAS treatment solutions. The findings from this work offer semiconductor manufacturers a data-driven framework for designing optimized treatment systems that meet regulatory standards while minimizing operational costs and environmental footprint.