Characterization of membrane pore size using statistical model by facile method of air bubbling in liquid media

Abstract

Research has shown the correlation between porous membrane pore size and bubble formation including bubble size, frequency, and operating parameters. However, attempts to create a governing equation to establish this correlation have often suffered from low accuracy due to variable interactions. Therefore, it is necessary to identify the significant effects of membrane pore size on bubble formation to establish effective correlations. This study aimed to identify the significant effect of porous membrane pore size on bubble formation using analysis of variance (ANOVA) to establish a mathematical model that predicts porous membrane pore size using bubble formation. To achieve this, various porous membranes were fabricated by varying hydrophilic silica loadings. The resulting membranes were characterized regarding pore size and employed in an aeration device under variable air flow rates and inlet pressures to obtain bubble size and frequency data. JMP software developed a statistical model to characterize membrane pore sizes. The results show a significant effect of bubble formation information on pore size, with a p-value of 0.0089, R² of 0.96, and root mean squared error (RMSE) value of 0.02. Validation of the model using three membranes demonstrated minor deviations of 0–6.5 %, emphasizing the model's effectiveness in predicting membrane pore size.