Comparison of membrane biofouling in nitrification and denitrification for the membrane bioreactor (MBR).

Batch filtration tests were conducted to compare the characteristics of membrane biofouling with regard to nitrification and denitrification. A Modified Fouling Index (MFI) was obtained using a stirred cell tester. The denitrification assays showed higher membrane fouling rates than the nitrification assays. The fouling became worse, not only due to pore blocking resistance, but also from cake layer resistance after denitrification. The Extracellular Polymeric Substances (EPS) concentration and relative hydrophobicity were decreased after denitrification, resulting in floc deterioration. The floc deterioration was assumed to have increased the cake layer resistance in the filtration test. The protein Soluble Microbial Products (SMP) concentration, portion of high molecular weight in carbohydrate SMP and relative hydrophobicity were increased after denitrification, which was assumed to cause membrane pore blocking. The changes in the EPS and SMP characteristics were the main fouling parameters in denitrification.     

Impact of dissolved oxygen concentration on membrane filtering resistance and soluble organic matter characteristics in membrane bioreactors

This study investigated the impact of dissolved oxygen (DO) concentration on membrane filtering resistance, soluble organic matter (SOM) and extracellular polymeric substance (EPS) characteristics in a membrane bioreactor (MBR). A laboratory-scale MBR was operated under DO limited (0.2 mg L(-1) DO) and fully aerobic (3.7 and 5.4 mg L(-1) DO) conditions. Membrane filtering resistance was determined for the mixed liquor suspended solids (MLSS) and for resuspended microbial biomass after removing SOM. Regardless of the DO concentration, the cake resistance (Rc) was approximately 95 percent of the total resistance (Rt). The membrane cake resistance was found to decrease significantly after removing the SOM. The total resistance caused by the resuspended biomass was 29 percent of that caused by the MLSS under DO limited conditions, while the total resistance caused by resuspended biomass was 41 to 48 percent of that caused by the MLSS under fully aerobic conditions. Under DO limited conditions, SOM in the MLSS contained a larger amount of high molecular weight compounds, leading to higher cake resistance than under fully aerobic conditions. There was significant variation in the molecular weight fractions of the EPS, with no clear relationship with DO concentration. There was also no distinct relationship between membrane filtering resistance and molecular weight fraction of the EPS.     

Influence of extracellular polymeric substances on nitrogen removal in an intermittently-aerated membrane bioreactor.

Influence of EPS on fouling of intermittent aeration MBR reactor (denitrification MBR) was investigated changing intermittent aeration cycle (10 minute-cycle and 120 minute-cycle) in laboratory-scale reactors using synthetic wastewater. EPS were extracted from bacterial cells using cation resin method and molecular weight fractioning of EPS was conducted using gel chromatography. In both of the reactors, nitrogen removal rate was almost 100% after 50th day although DO concentration was not very high during the aerated phase because of accumulation of nitrifying bacteria in the reactors. In the 120 minutes-cycle reactor, trans-membrane pressure increased more rapidly than in the 10 minutes-cycle reactor. The reason might be that EPS of more than 1000 kDa, which are the main fouling substances, are produced more rapidly in the 120 minute-cycle condition. It was also found that three peaks at around 100 kDa, 500 kDa and 2000 kDa are prominent in EPS in intermittent-aeration MBR irrespective of cycle and higher molecular weight EPS are decomposed to smaller molecular weight EPS on membrane surface.     

Characteristics of membrane fouling in submerged membrane bioreactor under sub-critical flux operation.

Recently, the membrane bioreactor (MBR) process has become one of the novel technologies to enhance the performance of biological treatment of wastewater. Membrane bioreactor process uses the membrane unit to replace a sediment tank, and this can greatly enhance treatment performance. However, membrane fouling in MBR restricts its widespread application because it leads to permeate flux decline, making more frequent membrane cleaning and replacement necessary, which then increases operating and maintenance costs. This study investigated the sludge characteristics in membrane fouling under sub-critical flux operation and also assessed the effect of shear stress on membrane fouling. Membrane fouling was slow under sub-critical flux operation. However, as filamentous microbes became dominant in the reactor, membrane fouling increased dramatically due to the increased viscosity and polysaccharides. A close link was found between membrane fouling and the amount of polysaccharides in soluble EPS. The predominant resistance was the cake resistance which could be minimized by increasing the shear stress. However, the resistance of colloids and solutes was not apparently reduced by increasing shear stress. Therefore, smaller particles such as macromolecules (e.g. polysaccharides) may play an important role in membrane fouling under sub-critical flux operation.     

Fouling cake layer in a submerged anaerobic membrane bioreactor treating saline wastewaters: curse or a blessing?

The treatment of inhibitory (saline) wastewaters is known to produce considerable amounts of soluble microbial products (SMPs), and this has been implicated in membrane fouling; the fate of these SMPs was of considerable interest in this work. This study also investigated the contribution of SMPs to membrane fouling of the; (a) cake layer/biofilm layer, (b) the compounds below the biofilm/cake layer and strongly attached to the surface of the membrane, (c) the compounds in the inner pores of the membrane, and (d) the membrane. It was found that the cake/biofilm layer was the main reason for fouling of the membrane. Interestingly, the bacteria attached to the cake/biofilm layer showed higher biodegradation rates compared with the bacteria in suspension. Moreover, the bacteria attached to the cake layer showed higher amounts of attached extracellular polysaccharides (EPS) compared with the bacteria in suspension, possibly due to accumulation of the released EPS from suspended biomass in the cake/biofilm layer. Molecular weight (MW) analysis of the effluent and reactor bulk showed that the cake layer can retain a large fraction of the SMPs in the reactor and prevent them from being released into the effluent. Hence, while cake layers lead to lower fluxes in submerged anaerobic membrane bioreactors (SAMBRS), and hence higher costs, they can improve the quality of the reactor effluent.     

Comparison of mixed liquor filterability measured with bench and pilot-scale membrane bioreactors.

Parallel experimental tests to measure mixed liquor filterability for submerged membrane bioreactors were conducted over a six month period using three ZW-500 pilot plants and a ZW-10 lab-scale filterability apparatus. Non-air sparged conditions during the tests yielded operation behaviour that was equivalent to dead-end filtration. The fouling resistance increased linearly with the intercepted mass until a critical point was reached at which point significant cake compression was induced and the resistance began to increase exponentially. Although the point of cake compression appears to be dependent on the membrane module design, similar resistance per unit solid mass intercepted per unit area (R(mass)) values were observed when the same mixed liquor was filtered. Coupled with the established correlation between the R(mass) and the critical flux, it is suggested that the filterability test results from a side-stream, lab-scale module may be used to predict fouling potential in a full scale MBR wastewater treatment system without interrupting the full-scale MBR operation.     

Quantitative analysis of biological effect on membrane fouling in submerged membrane bioreactor.

The objective of this study is to investigate solids concentration and extracellular polymeric substance (EPS) effects on the membrane fouling in the submerged membrane bioreactor. The relationship between the solids retention time (SRT) and the amount of EPS is observed in three lab-scale MBRs. Additionally, the EPS effect on membrane fouling is quantified by calculating the specific cake resistance (alpha) using an unstirred batch cell test. By observing the sludge over a long period under various SRT scenarios, a wide range of EPS and membrane fouling data is obtained. These observations provide sufficient evidence of the functional relationship between SRT, EPS and alpha. As SRT decreases, the amount of EPS bound in sludge floc becomes higher in the high MLSS condition (> 5,000 mg/L). The amount of EPS in the sludge floc has positive influence on alpha. A sigmoid trend between EPS and alpha is observed and the functional relationship obtained by dimensional analysis is consistent with the experimental results.     

Effect of internal recycle rate on the high-strength nitrogen wastewater treatment in the combined UBF/MBR system.

An anaerobic/aerobic system combining an anaerobic upflow-sludge bed filter (UBF) and an aerobic membrane bioreactor (MBR) was operated to enhance organic and nitrogen removal efficiency. The internal recycle rate, which is one of the most important operation factors that affects overall removal efficiency, was varied from 100% to 300% of the influent flow. Under these conditions, the overall removal efficiencies of organic and nitrogen and characteristics of membrane fouling in the combined system treating the synthetic wastewater including high concentration of organics and nutrients were studied. As a result, nitrogen removal efficiency was increased to 67% when the internal recycle rate was 300% of influent flow rate. As the internal recycle ratio increased from 100% to 200%, protein content decreased by 17% and carbohydrate content increased by 12%. However, there was no remarkable difference in total extracellular polymeric substances (EPS) content. At the high recycle rate of 300%, the surface charge of sludge was decreased while hydrophobicity (specific ultraviolet absorbance, SUVA) was increased. The differences in SUVA and surface charge were 11% and 1%, respectively. It is concluded that SUVA and EPS composition were important parameters affecting membrane fouling in the combined system.     

The role of fouling mechanisms in a membrane bioreactor.

The present study has aimed to quantify the role of pore blocking and cake layer in a laboratory scale hollow fibre membrane module in submerged configuration, The membrane reactor (MBR) was fed with raw wastewater, only screened with a 2-mm sieve, collected from the Palermo WWTP. The MBR was characterised by an operating volume of 190 L and equipped with an aeration system located on the bottom of the reactor. The MBR operated for 65 days. The permeate was extracted by imposing a constant flux through the membrane (21 Lh(-1) m(-2)). The results confirm the importance of pore blocking control during start-up. In particular, it provides a rapid irreversible fouling that takes place at the beginning of the filtration process, before the deposition mechanism. Therefore, low suspended solids concentration in the initial phase causes a fast irreversible fouling. This circumstance creates the need for more frequent chemical cleaning after start-up without inoculum. Finally, the results underline that the cake has a mainly reversible feature.     

Occurrence of EPS in activated sludge from a membrane bioreactor treating municipal wastewater.

EPS are supposed to be among the causes of membrane fouling in membrane bioreactors (MBR). In this work they are measured as total proteins and total polysaccharides. Theoretical and empirical considerations of biomass membrane filtration lead to the conclusion that EPS in the water phase is decisive for the filterability of activated sludge. In this study therefore different ways of separating the water phase from the biomass are investigated, where a simple filtration over a paper filter turned out to be sufficient. Subsequently, a simple batch test set up was used to investigate the influence of substrate conditions on the amount of EPS in the water phase. Dilution of the biomass does not result in changes. Dilution together with substrate addition leads to an increase both in proteins and polysaccharides. Replacement of the water phase leads to no significant changes in protein concentration, but polysaccharide concentration may vary considerably. This phenomenon is more pronounced after replacement of the water phase and substrate addition.     

Behavior of micro-particles in monolith ceramic membrane filtration with pre-coagulation.

This paper is intended to clarify the characteristics unique to monolith ceramic membranes with pre-coagulation by referring to the behavior of micro-particles. Flow analysis and experiments have proved that monolith ceramic membranes show a unique flow pattern in the channels within the element, causing extremely rapid flocculation in the channel during dead-end filtration. It was assumed that charge-neutralized micro-particles concentrated near the membrane surface grow in size due to flocculation, and as a result, coarse micro-particles were taken up by the shearing force to flow out. As the dead end points of flow in all the channels are located near the end of the channels with higher filterability, most of the flocculated coarse particles are formed to a columnar cake intensively at the dead end point. Therefore cake layer forming on the membrane other than around the dead end point is alleviated. This behavior of particle flocculation and cake formation at the dead end point within the channels are unique characteristics of monolith ceramic membranes. This is why all monolith ceramic membrane water purification systems operating in Japan do not have pretreatment equipment for flocculation and sedimentation.     

Comparison of linear alkylbenzene sulfonates removal in conventional activated sludge systems and membrane bioreactors.

The potential of a membrane bioreactor (MBR) and a conventional activated sludge (CAS) system to remove polar micropollutants was evaluated using linear alkylbenzene sulfonates (LAS) as model components. Removal efficiencies over 97% were achieved in both reactor systems. The appearance of biological breakdown metabolites and the respirometric response of the sludges to LAS addition indicated that LAS removal was due to biodegradation, rather than sorption phenomena. The effect of operational variables, such as hydraulic retention time, LAS composition and hydrophobicity of the membrane used in the MBR, was negligible in the range tested. A stepwise increase in LAS influent concentration resulted in higher residual effluent concentrations but did not change the procentual removal efficiency. Because an increase in LAS and SPC effluent concentration occurred to a larger extent in the CAS than in the MBR under similar operating conditions, MBRs may turn out to be be more robust with respect to biological degradation of micropollutants than CAS.     

Continuous biological ferrous iron oxidation in a submerged membrane bioreactor.

Microbial oxidation of ferrous iron may be available alternative method of producing ferric iron, which is a reagent used for removal of H2S from biogas. In this study, a submerged membrane bioreactor (MBR) system was employed to oxidize ferrous iron to ferric iron. In the submerged MBR system, we could keep high concentration of iron-oxidizing bacteria and high oxidation rate of ferrous iron. There was membrane fouling caused by chemical precipitates such as K-jarosite and ferric phosphate. However, a strong acidity (pH 1.75) of solution and low ferrous iron concentration (below 3000 mg/I) significantly reduced the fouling of membrane module during the bioreactor operation. A fouled membrane module could be easily regenerated with a 1 M of sulfuric acid solution. In conclusion, the submerged MBR could be used for high-density culture of iron-oxidizing bacteria and for continuous ferrous iron oxidation. As far as our knowledge concerns, this is the first study on the application of a submerged MBR to high acidic conditions (below pH 2).     

Optimization of operational factors of a membrane bioreactor with gravity drain.

A new membrane bioreactor with gravity drain for municipal wastewater treatment was tested and its operational factors were investigated in this study. These factors include pressure head, MLSS, aeration intensity (an air flow rate per unit floor area) and temperature. Results of batch experiments showed that a critical pressure head of the MBR was 0.85-1.5 m-H2O. At a pressure head of 0.85 m-H2O, statistical analysis of batch experiments showed that aeration intensity significantly affected membrane flux, and the MLSS had no impact on membrane flux under a temperature of 22.0 +/- 1.0 degrees C. Results of the long-term continuous experiment showed that temperature significantly affected membrane flux. The impact of temperature on membrane flux in this case was about 4-10 times of that analyzed by using a classical cake layer model. During this experiment, the average removal efficiencies of COD, BOD5 and NH4+-N were over 85%, 97% and 94%, respectively.     

Optimising the operation of a MBR pilot plant by quantitative analysis of the membrane fouling mechanism.

In order to optimize some operational conditions of MBR systems, a MBR pilot plant equipped with a submerged hollow fibre membrane module was employed in this study. The pilot MBR was fed with real municipal wastewater and the filtration flux, backwashing interval, aeration frequency and temperature were varied. A filtration flux below 25 I/m2h is generally recommended, at below this flux, the MBR operated at sub-critical flux conditions, the filter cake was minimized and membrane fouling was mainly attributed to the membrane pore blocking. Moreover, the membrane fouling, at below 25 I/m2h, was more reversible to backwashing; above this value, backwashing became less efficient to clean the membrane. Less frequent backwashing (e.g. 600 s filtration/45 s backwashing) decreased the amount of fouling irreversible to backwashing and its performance was superior to that of frequent backwashing (e.g. 200 s filtration/15 s backwashing). The MBR suffered more fouling at low temperature conditions (e.g. at 13-14 degrees C) than at high temperature conditions (e.g. at 17-18 degrees C). A conceptual model was built up and successfully interpreted this temperature effect.     

Direct force measurement of bacteria adhesion on metal in aqueous media.

The adhesion of bacteria on metal surfaces in aqueous media and the development of biofilm and resultant biofouling are important phenomena in both the natural environment and engineering systems. This work reports on the use of a force microscopy technique to measure bacterial metal adhesion by two anaerobic sulphate-reducing bacteria (Desulfovibrio desulfuricans and a local marine isolate) and an aerobe (Pseudomonas sp.). Using a modified bacteria tip, the atomic force microscope was able to quantify the attraction and repulsion force in the nano-Newton range between the bacteria cell and metal surface in aqueous media. Results show that increasing surface hydrophobicity of the metal, and increasing the ionic strength of the aqueous medium both enhance the adhesion force. The adhesion forces were also influenced by the physiological properties of the bacterium, such as the bacterial surface charges and hydrophobicity.     

活性污泥胞外聚合物EPS的影响因素研究

胞外聚合物EPS对活性污泥的表面特性、生物絮凝、沉降及脱水性能等具有重要的影响,主要对污泥负荷Ns,DO,温度,Ca2+,pH及废水水质等因素对EPS的影响进行了研究。试验结果表明:Ns,pH及废水水质对EPS的含量、成分有显著的影响;DO,温度对EPS的影响较小;同时随着进水Ca2+浓度的增加,污泥的EPS,脱氢酶及OUR增加。     

MBR fouling control and permeate quality enhancement by polyaluminium chloride dosage: a case study

Scope of this study was to evaluate the effectiveness of a metal salt (polyaluminium chloride, PACl) dosage into a pilot-scale MBR (membrane bioreactor) in terms of fouling control and permeate quality enhancement, especially with reference to specific textile macro-pollutants (dyes and surfactants). The pilot plant was fed with a mixed domestic-textile wastewater (textile wastewater accounted for 65% of total flow and for 70% of total chemical oxygen demand, COD, load) and operated for 7.5 months without flux enhancers (step 1) and 3 months with the addition of PACl (step 2). The optimum dose was defined performing a jar-test campaign between step 1 and step 2 (12.5 mg gMLSS(-1) that corresponds to 0.4 g d(-1)). The addition of PACl resulted in a significant decrease of the filtration resistance due to cake layer formation (R(c), -65.4%) and of the irreversible fouling rate, evaluated as the average variation per unit time of the filtration resistance due to foulants adsorption on membrane pore wall (FR, -45.3%). As for permeate quality, removal rates related to total phosphorus and textile macro-parameters such as colour and anionic surfactants, increased by +64, +16 and +7%, respectively. No significant effect was observed on COD, non-ionic surfactants and nitrogen compounds removal.     

Influence of suspension viscosity and colloidal particles on permeability of membrane used in membrane bioreactor (MBR).

In this study, pilot scale experiments were carried out to examine membrane fouling occurring in membrane bioreactors (MBR) with or without pre-treatment (coagulation/sedimentation). Especially, the influence of suspension viscosity and dissolved organic matter (DOM) on membrane fouling was investigated. The pre-coagulation/sedimentation process improved the performance of a MBR in terms of membrane permeability by controlling irreversible fouling and formation of thick cake layer. The upper limit of MLSS concentration for an efficient operation in MBR without pre-treatment was suggested to be around 10 g/L based on the measurement of suspension viscosity. In this study, it was difficult to directly relate membrane fouling to DOM detected in the membrane chamber. A series of laboratory scale dead-end filtration experiments was carried out to investigate which fractions in biomass suspension would be the most influential in the deterioration of membrane permeability. Based on the dead-end tests, it was shown that the deterioration of membrane permeability was mainly caused by the colloidal particle fraction in the biomass suspension.     

Function of dynamic membrane in self-forming dynamic membrane coupled bioreactor.

The Self-Forming Dynamic Membrane Coupled Bioreactor (SFDMBR), which uses coarse pore-sized material to separate solid and liquid in bioreactors, has some advantages compared with MBR using micro-/ultra-filtration membranes, for example, low module cost and high flux. The cake layer and gel layer formed on the surface and in the pores of the material during filtration played an important role, called self-forming dynamic membrane (DM), which mainly consisted of activated sludge. In this study, the function of DM in pollutant removal was investigated. It was found that DM could remove some organic matter (12.6 mg L(-1) on average) and total nitrogen (3.01 mg L(-1) on average) in the supernatant. Colloids and organic nitrogen were partly removed by DM while DOC, ammonia nitrogen and nitrate nitrogen removal by DM varied from negative to positive, which resulted from the combination of various biological activities, e.g. nitrification, biological utilization and so on. DO concentration in DM decreased with the depth and reached zero at about 1.5-2.5 mm depth. The organic degradation activity and nitrification activity of the biomass suspended in the bioreactor were higher than those of the biomass in the cake layer, which might be caused by the low DO concentration and low organic pollutant content in DM.