No abstract

Membrane bioreactors (MBRs) are becoming more suitable alternatives for conventional wastewater treatment devices. The performance of a pressure-driven MBR is dominantly affected by the hydrodynamic conditions of the system. This study was conducted to investigate various hydrodynamic characteristics including shear stress, cross-flow velocity, and membrane fouling resistance, using computational fluid dynamics (CFD). Simulation of two-and three-phase flow for a flat-sheet submerged membrane module was carried out, and the results were compared with the experimental data. The CFD simulation was implemented to analyze the fluid-flow pattern, shear stress on membrane surfaces, and cross-flow velocity between membranes at various mixed liquor suspended solid concentrations in the bioreactor. It was shown that the cross-flow velocity plays an important role in the membrane fouling and determination of the critical particle diameter. To achieve an optimal operating condition, the critical particle diameter was calculated at different air flow rates and permeate fluxes. The CFD results showed that the outermost membranes are more prone to fouling because of the lower shear stress on their surface as well as the lower cross-flow velocity between them and the module wall. Moreover, the effect of the air bubble diameter on the air and liquid shear stress was investigated to determine an optimal bubble size.

In the present attempt, a 3D simulation using a commercially available computational fluid dynamics package (FLUENT 6.3.26), and a metabolic model were used to investigate hydrodynamics and production of polyhydroxybutyrate (PHB) in an airlift bioreactor, performed by Tavares et al. (L. Z. Tavares, E. S. da Silva, J. G. C. Pradella. Biochem. Eng. J. , 2004; 18, 21-31). An Eulerian approach was applied to model the gas-liquid interactions. In order to account for the combined effect of bubble breakup and coalescence in the bioreactor, a population balance model implemented in the software was used. Biosynthesis of PHB in the bioreactor was examined and maximum forward reaction rate values of thiolase, reductase, and synthase steps were determined as V 1 (thiolase) = 1.8 mM/min, V 1 (reductase) = 400 mM/min, and V 1 (synthase) = 380 mM/min to predict the experimental PHB production rate by Tavares et al. suitably. A simplified reaction was considered for PHB production. Moreover, gas holdup, liquid velocity vectors, shear stress, and volumetric oxygen transfer coefficient were investigated. Also, molar concentration profiles of PHB and glucose within the bioreactor were obtained.values [HSCOA] = 100 μM[ACOA] = 1000 μM[NADP+] = 100 μM[NADPH] = 300 μM P. MAVADDAT ET AL. Asia-Pacific Journal of Chemical Engineering 564

A B S T R A C TThis study presents a closed-loop process for forming pre-coated dynamic membrane used in membrane bioreactor. Powder-activated carbon (PAC) particles are suspended by using axial-flow agitator whilst simultaneously deposited upon the filter medium by cross-flow filtration. The revolution speed of agitator is varied in the range of 150-450 rpm. Computational fluid dynamic is used to calculate velocity and shear rate adjacent to membrane surface. The force balance of the particles considers the driving force of dynamic membrane formation, drag force due to negative pressure-driven of suction pump, lift force due to mechanical mixing, and adhesive force due to particle-particle interactions. The effect of two major terms including drag and lift forces on the porosity of dynamic membrane is systematically investigated. The lower drag force results in the higher porosity of dynamic membrane. The results also show that the denser dynamic membrane is formed when the lift force becomes lower. The porosity of dynamic membrane is ranged from 0.38 to 0.46 while the thickness of dynamic membrane is altered between 100 and 500 μm. A dimensionless immobilized parameter is derived to predict the formation of pre-coating dynamic membrane. The results obtained from modeling show that dynamic membranes formed at higher values of immobilization parameter have a more cohesive structure. The results obtained from the experiments and the model reveal that the thickness of 400 μm of PAC layer is considered as a stability threshold thereafter the structure of dynamic membrane becomes loose and unstable. The rejection capacity of dynamic membrane is also evaluated at different thicknesses by using Formazin solution. The results reveal that the turbidity of filtrate decreases with increasing the thickness of dynamic membrane. The results display that rejection capacity of pre-coating dynamic membrane is comparable to that of microfiltration membrane.

A three-phase computational fluid dynamics (CFD) simulation was carried out in a full-scale membrane bioreactor to investigate the effect of sparger configuration on various hydrodynamic parameters.