Establishing the mechanisms underpinning solids breakthrough in UASB configured anaerobic membrane bioreactors to mitigate fouling

In this study, the mechanisms for solids breakthrough in upflow anaerobic sludge blanket (UASB) configured anaerobic membrane bioreactors (AnMBRs) have been described to establish design parameters to limit membrane fouling. As the sludge blanket develops, two periods can be identified: (i) an initial progressive enhancement in solids separation provided through sludge blanket clarification, via depth filtration, which sustains downstream membrane permeability; and (ii) sludge blanket destabilisation, which imposed solids breakthrough resulting in a loss in membrane permeability. The onset of sludge blanket destabilisation was identified earlier in the flocculent AnMBR, which was ascribed to an increased gas production, caused by hydrolysis within the sludge blanket at extended solids residence time. Whilst hydrolysis also induced higher gas productivity within the granular AnMBR, solids breakthrough was not evidently observed during this period, and was instead only observed as the sludge blanket approached the UASB overflow. However, solids breakthrough was observed earlier for both reactors when treating wastewater with lower temperatures. This was explained through characterisation of the settling velocity of discrete particles from the sludge blanket of both MBRs; solids washout was evidenced to be induced by the increase in fluid viscosity with a reduction in temperature, which lowered terminal particle settling velocity. Nevertheless, particle settling velocity was comparable for particles from both sludge blankets. We therefore propose that the enhanced stability imparted by the granular AnMBR is due to the higher inertial force of the dense granular sludge. From this study, we suggest that similarly low levels of membrane fouling can be achieved within flocculent AnMBR by managing solids retention time to constrain sludge bed height and excess hydrolysis, together with adopting an upflow velocity based on particle buoyancy at the lowest expected operating temperature. (C) 2020 Elsevier Ltd. All rights reserved.