3D modeling of PVDF membrane aging using scanning electron microscope and OpenCV image analysis

Abstract

The aging of membranes compromises their integrity and reduces the quality of filtered water. In the present study, we constructed three-dimensional (3D) models of polyvinylidene fluoride (PVDF) membranes to evaluate the effects of membrane aging. Virgin and used PVDF membrane images were obtained by field emission scanning electron microscopy (FE-SEM), and the size distribution of the PVDF spherulites was analyzed using a Python-OpenCV program. Assuming a random distribution, the 3D membrane models were assessed in hydraulic and particle tracing simulations using COMSOL Multiphysics®. Compared to virgin membranes, the PVDF spherulites in the 10-year-old used membranes were enlarged and sloughed off owing to repeated filtration and backwashing. The porosity, filtration flux, and particle retention of the used membranes varied significantly more than those of the virgin membranes. Although our results were constrained by the size of the 3D membrane models and overall small sample size, our study provides an empirical basis for the 3D modeling of membranes from SEM images and our method could be applied for the optimization of membrane structure and integrity for water treatment and purification.