Evaluation of mechanical membrane cleaning with moving beads in MBR using Box–Behnken response surface methodology

Abstract

Incorporation of moving beads into membrane bioreactor (MBR) has been suggested as an effective membrane fouling control because moving beads can mechanically remove bio-cakes on the membrane surface without additional equipment and energy input. As the efficiency of fouling control is dependent on factors associated with moving beads, however, the design of experiment was applied to find optimum condition for the effective mechanical cleaning with moving beads in MBR. Bead diameter (mm), bead number, and aeration rate (m3/h) were selected as independent design parameters. Based on batch test results, the correlation between the detachment efficiency of bio-cakes and three design parameters was established using Box–Behnken methodology. When all three design parameters at their optimal conditions (beadopt) were extended to the continuous lab-scale MBR, membrane filterability increased by three times, compared with that in the control MBR without bead. On the other hand, each of five experimental sets with two optimal and one random parameter (beadrandom) showed less membrane filterability by 9–80%, respectively, compared with beadopt. The parameter C (aeration rate) affected most significantly as it is associated with not only bead movement but also shear induced by air bubbles.