Comparison of four types of membrane bioreactor systems in terms of shear stress over the membrane surface using computational fluid dynamics.

Membrane bioreactors (MBRs) have been used successfully in biological wastewater treatment to solve the perennial problem of effective solids-liquid separation. A common problem with MBR systems is clogging of the modules and fouling of the membrane, resulting in frequent cleaning and replacement, which makes the system less appealing for full-scale applications. It has been widely demonstrated that the filtration performances in MBRs can be greatly improved with a two-phase flow (sludge-air) or higher liquid cross-flow velocities. However, the optimization process of these systems is complex and requires knowledge of the membrane fouling, hydrodynamics and biokinetics. Modern tools such as computational fluid dynamics (CFD) can be used to diagnose and understand the two-phase flow in an MBR. Four cases of different MBR configurations are presented in this work, using CFD as a tool to develop and optimize these systems.

Analytical and numerical modelling of Newtonian and non-Newtonian liquid in a rotational cross-flow MBR.

Fouling is the main bottleneck of the widespread use of MBR systems. One way to decrease and/or control fouling is by process hydrodynamics. This can be achieved by the increase of liquid cross-flow velocity. In rotational cross-flow MBR systems, this is attained by the spinning of, for example, impellers. Validation of the CFD (computational fluid dynamics) model was made against laser Doppler anemometry (LDA) tangential velocity measurements (error less than 8%) using water as a fluid. The shear stress over the membrane surface was inferred from the CFD simulations for water. However, activated sludge (AS) is a non-Newtonian liquid, for which the CFD model was modified incorporating the non-Newtonian behaviour of AS. Shear stress and area-weighted average shear stress relationships were made giving error less that 8% compared with the CFD results. An empirical relationship for the area-weighted average shear stress was developed for water and AS as a function of the angular velocity and the total suspended solids concentration. These relationships can be linked to the energy consumption of this type of systems.

Experimental and CFD simulation studies of wall shear stress for different impeller configurations and MBR activated sludge.

Membrane bioreactors (MBRs) have been used successfully in biological wastewater treatment for effective solids-liquid separation. However, a common problem encountered with MBR systems is fouling of the membrane resulting in frequent membrane cleaning and replacement which makes the system less appealing for full-scale applications. It has been widely demonstrated that the filtration performances in MBRs can be improved by understanding the shear stress over the membrane surface. Modern tools such as computational fluid dynamics (CFD) can be used to diagnose and understand the shear stress in an MBR. Nevertheless, proper experimental validation is required to validate CFD simulation. In this work experimental measurements of shear stress induced by impellers at a membrane surface were made with an electrochemical approach and the results were used to validate CFD simulations. As good results were obtained with the CFD model (<9% error), it was extrapolated to include the non-Newtonian behaviour of activated sludge.

Analysis of shear stress and energy consumption in a tubular airlift membrane system.

Application of a two-phase slug flow in side-stream membrane bioreactors (MBRs) has proven to increase the permeate flux and decrease fouling through a better control of the cake layer. Past literature has shown that the hydrodynamics near the membrane surface have an impact on the degree of fouling by imposing high shear stress near the surface of the membrane. Previously, shear stress histograms (SSH) have been introduced to summarize results from an experimental setup developed to investigate the shear stress imposed on the surface of a membrane under different two-phase flow conditions (gas and liquid) by varying the flow of each phase. Bimodal SSHs were observed, with peaks corresponding to the shear induced by the liquid and gas flow respectively. In this contribution, SSHs are modelled using simple empirical relationships. These are used to identify the two-phase flow conditions that optimize fouling control. Furthermore, the total energy consumption of the system was estimated based on the two-phase pressure drop. It was found that low liquid and high gas flow rates (ratio of approx. 4) balanced the peaks and minimized the energy consumption.

Investigation of the effect of viscosity on slug flow in airlift tubular membranes in search of a sludge surrogate.

The behaviour of three different liquid-gas slug flows (water, carboxymethyl cellulose and activated sludge) in a vertical tube was studied using a high speed camera (HSC). Experiments were performed using different flow rates and two tube diameters (6.3 and 9.9 mm). The observed difference in behaviour of the ascending gas slugs can be explained by the difference in viscosity of the fluids (Newtonian and non-Newtonian). Moreover, it was observed that the degree of coalescence of gas slugs is lower for non-Newtonian liquids and they behave like a succession of slugs without actually coalescing into a single larger gas slug. Finally, gas slug rising velocities were also extracted, but no subsequent difference in the rising velocities of the different fluids was found.

Analysis of two-phase air-water annular flow in U-bends.

This paper presents an experimental and numerical study of gas-liquid annular flow in horizontal 180 U-bends. The paper aims to study the effect of bend curvature radius and superficial gas velocity in the liquid film's behavior and annular flow characteristics. The study is divided into three sections. The first section corresponds to the experimental methodology and results. The second section compresses the validation of the computational fluid dynamic (CFD) model with the experimental results. Finally, the last section presents the CFD estimation of additional variables that cannot be acquired with the existing experimental setup. The experimental results provide an initial understanding of the multiphase mixture obtained using optical techniques (i.e., High-Speed Filming (HSF) analysis). The comparison between the experiments and the numerical simulations is presented, and a reasonable agreement is observed between both approaches. Finally, additional results such as film distribution and rotation before and after the bend are extracted from the CFD simulations.

Activated sludge rheology: a critical review on data collection and modelling.

Rheological behaviour is an important fluid property that severely impacts its flow behaviour and many aspects related to this. In the case of activated sludge, the apparent viscosity has an influence on e.g. pumping, hydrodynamics, mass transfer rates, sludge-water separation (settling and filtration). It therefore is an important property related to process performance, including process economics. To account for this, rheological behaviour is being included in process design, necessitating its measurement. However, measurements and corresponding protocols in literature are quite diverse, leading to varying results and conclusions. In this paper, a vast amount of papers are critically reviewed with respect to this and important flaws are highlighted with respect to rheometer choice, rheometer settings and measurement protocol. The obtained rheograms from experimental efforts have frequently been used to build viscosity models. However, this is not that straightforward and a lot of errors can be detected with respect to good modelling practice, including fair model selection criteria, qualitative parameter estimations and proper model validation. These important steps are however recurrently violated, severely affecting the model reliability and predictive power. This is illustrated with several examples. In conclusion, dedicated research is required to improve the rheological measurements and the models derived from them. At this moment, there is no guidance with respect to proper rheological measurements. Moreover, the rheological models are not very trustworthy and remain very "black box". More insight in the physical background needs to be gained. A model-based approach with dedicated experimental data collection is the key to address this.

Critical review of membrane bioreactor models--part 2: hydrodynamic and integrated models.

Membrane bioreactor technology exists for a couple of decades, but has not yet overwhelmed the market due to some serious drawbacks of which operational cost due to fouling is the major contributor. Knowledge buildup and optimisation for such complex systems can heavily benefit from mathematical modelling. In this paper, the vast literature on hydrodynamic and integrated MBR modelling is critically reviewed. Hydrodynamic models are used at different scales and focus mainly on fouling and only little on system design/optimisation. Integrated models also focus on fouling although the ones including costs are leaning towards optimisation. Trends are discussed, knowledge gaps identified and interesting routes for further research suggested.