An efficient computational framework for predicting particle capture in heterogeneous filtration systems

Heterogeneous filters outperform uniform media by capturing a wider range of particle sizes, enhancing efficiency and longevity. This study introduces a computational method to predict particle penetration and capture in arbitrarily structured heterogeneous filters, considering variations in fiber size, orientation, and distribution. Unlike traditional homogeneous models, it discretizes filters along the flow direction and incorporates fiber-scale interactions. Validated against experimental and numerical data, the framework applies to diverse filtration systems, including air purification and vegetation-based filtering. The method was utilized to analyze filtration under binary and log-normal size distributions, showing heterogeneous filters improve capture efficiency (over 90%) and reduce clogging.