Effect of initial particle deposition rate on cake formation during dead-end microfiltration

The relationships between various parameters and membrane fouling are generally well-established in microfiltration. Surprisingly, a non-monotonic correlation between initial particle deposition rate and extent of flux decline was observed during dead-end microfiltration. To understand the underlying mechanism, a fouling model and three-dimensional optical coherence tomography (OCT) scanning were employed. Four different initial deposition rates were imposed using latex particles (3 μm) as the model foulant and polycarbonate track-etched membrane (pore size = 2 μm). The network fouling model revealed that, as the initial deposition rate increased, (i) the probability of deposition on the non-porous part of the membrane (β) increased then decreased; (ii) the initial cake resistance (Rc0) increased monotonically; and (iii) the specific cake resistance (R'c) decreased then increased. The OCT image analysis further informed that (i) the deposited foulants tended to form larger clusters at the lowest and highest initial deposition rates of 12.5 and 31.25 g/m2/h, respectively, but tended to be more dispersed at the intermediate initial deposition rate of 25 g/m2/h; and (ii) the more significant specific cake resistances (R'c) at the lowest and highest initial deposition rates were due to denser cakes stemming from greater particle rearrangement within the cake and greater particle concentration near the membrane, respectively. The network fouling model and experimental OCT characterization served to explain the counter-intuitive flux decline trends, and the insights are valuable towards better understanding of the inevitable membrane fouling phenomena. [Display omitted] •Non-monotonic correlation between initial particle deposition rate and flux decline.•Network fouling model and OCT image analysis used to understand surprising behavior.•Cake resistance decreased then increased with initial particle deposition rate.•Different mechanisms underlie the highest cake resistances.•Larger deposited clusters linked to higher cake resistances.