Bacterial community structures in MBRs treating municipal wastewater: relationship between community stability and reactor performance

Bacterial community structures in pilot-scale conventional membrane bioreactors (CMBRs) and hybrid MBRs (HMBRs) which were combined with pre-coagulation/sedimentation were analyzed by polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) and fluorescence in situ hybridization (FISH) techniques. The results were compared with the community structure in a full-scale activated sludge (AS) process treating the same municipal wastewater. The Dice index (Cs) of similarity analysis of DGGE banding patterns demonstrated that the microbial community in AS was more similar to those in CMBR1 and CMBR2 than HMBR1 and HMBR2. This suggested that influent wastewater composition had a larger impact on bacterial community structures. Long-term community structure changes in the HMBRs and CMBRs were monitored and analyzed over 240 days by Non-metric multidimensional scaling (NMDS) analysis of DGGE banding patterns. The NMDS analysis revealed that both HMBRs and CMBRs had marked changes in community structures during the first about 100 days. Thereafter the perpetual fluctuations of bacterial community structures were observed in both HMBRs and CMBRs, even though the stable MBR performances (the performance was measured as membrane permeability and removal of dissolved organic carbon, DOC) were achieved. These results suggest that not only the stability, but also the adequate dynamics ("flexibility") of the bacterial community structure are important for the stable performance of the MBRs treating complex municipal wastewater.     

Saline sewage treatment using a submerged anaerobic membrane reactor (SAMBR): Effects of activated carbon addition and biogas-sparging time

This study investigated the performance of a submerged anaerobic membrane reactor (SAMBR) treating saline sewage under fluctuating concentrations of salinity (0-35 g NaCl/L), at 8 and 20 h HRT, with fluxes ranging from 5-8 litres per square metre per hour (LMH). The SAMBRs attained a 99% removal of Dissolved Organic Carbon (DOC) with 35 g NaCl/L, while removal inside the reactor was significantly lower (40-60% DOC). Even with a sudden drop in salinity overall removal recovered quickly, while the recovery inside the reactor took place at a slower rate. This highlights the positive effect of the membrane in preventing the presence of high molecular weight organics in the effluent while also retaining biomass inside the reactor so that they can rapidly acclimatize to salinity. The reduction of continuous biogas sparging to intervals of 10 min ON and 5 min OFF resulted in a slight increase in transmembrane pressure (TMP) by 0.025 bar, but also resulted in an increase in effluent DOC removal and inside the SAMBR by 10% and 20%, respectively. The addition of powdered activated carbon (PAC) resulted in a decrease in the TMP by 0.070 bar, and an increase in DOC removal in the reactor and effluent by 30% and 5%, respectively. The PAC dramatically decreased the high molecular weight organics in the reactor over a period of 72 h. SEM pictures of the membrane and biomass before and after addition of PAC revealed a remarkable reduction of flocks on the membrane surface, and a reduction inside the reactor of soluble microbial products (SMPs). Finally, Energy Dispersive X-ray (EDX) analysis of the membranes pores and biofilm highlighted the absence of organic matter in the inner pores of the membrane.     

Solids characterisation in an anaerobic migrating bed reactor (AMBR) sewage treatment system

The aim of this work was to characterise the solids in an anaerobic sewage treatment process. Hindered settling velocity, particle size distributions (PSD), influent and effluent COD(P)/SS and discrete settling velocity distributions were all measured. The anaerobic migrating bed reactor (AMBR) solids were mainly flocculent and had a settling rate equivalent to a good settling activated sludge ( approximately SSVI=60 mL/g). The PSD of the anaerobic solids were very different to PSD for activated sludge flocs, with the anaerobic solids having a modal size an order of magnitude smaller than activated sludge, but a range of particle sizes being two orders of magnitude larger. There was a far greater range in size and structure in the anaerobic solids. The anaerobic process solids were primarily feed solids undergoing VSS destruction (hydrolysis). The biological mass was small. The solids seemed to retain their size as the volatiles were degraded and the density decreased ('skeletons' of the influent particulates). The small fraction of slowly settling solids, which have been identified to have a similar modal size but lower density than the mixed solids in the reactor, pose a solids retention time (SRT) control problem when relying on settling alone for solids retention.     

Treatment of easily degradable wastewater in a stirred anaerobic sequencing batch biofilm reactor

The main objectives of this study were to evaluate the performance of an anaerobic sequencing batch reactor when subjected to a progressive increase of influent glucose concentration and to estimate the kinetic parameters of glucose degradation. The reactor was initially operated in 8-h cycles, treating glucose in concentrations of 500, 1000 and 2000 mg l(-1). No glucose was detected in the effluent under these three conditions. The reactor showed operating stability when treating a glucose concentration of approximately 500 mg l(-1), with filtered chemical oxygen demand (COD) removal efficiencies varying from 93% to 97%. Operational instability occurred in the operation with glucose concentrations of approximately 1000 and 2000 mg l(-1), caused mainly by a production of extracellular polymeric substances (EPS), which led to hydrodynamic and mass transfer problems in the reactor. The mean volatile acid concentration values in the effluent were approximately 159+/-72 and 374+/-92 mg l(-1), respectively. A first-order model was adjusted to glucose concentration profiles and a modified model, including a residual concentration of substrate, was adjusted to COD temporal profiles. To check the formation of EPS, the reactor was operated in 3-h cycles with concentrations of 500 and 1000 mg l(-1). The purpose of this step was to discover if the production of EPS resulted from the biomass's exposure to a low concentration of substrate over a long period of time. Thus, it was hypothesized that a reduction of the time cycle would also reduce the exposure to low concentrations. However, this hypothesis could not be confirmed because large amounts of EPS were formed already under the first operational condition, using approximately 500 mg l(-1) of glucose in the influent, thus indicating the fallacy of the hypothesis. The production of EPS proved to depend on the organic volumetric load applied to the reactor.     

Impact of ambient conditions on SMP elimination and rejection in MBRs

The widespread application of the membrane-assisted activated sludge process is restricted by membrane fouling, which increases investment and operating costs. Soluble microbial products (SMPs) are currently considered as the major cause of membrane fouling in membrane bioreactors (MBRs). This study aims at elucidating and quantifying the effects of varying environmental conditions on SMP elimination and rejection based on findings in a pilot MBR and in well-defined lab trials. Several factors are thought to influence the concentration of SMP and their fouling propensity in one way or the other, but findings are often inconsistent or even contradictory. Here, SMP loading rate was found to have the greatest effect on SMP elimination and thus on concentration in the MBR. The degree of elimination decreased at very low DO and low nitrate concentrations. On average, 75% of influent SMP were eliminated in both pilot and lab trials, with the elimination of polysaccharides (PS) mostly above 80%. Rejection of SMP components by the used membrane (PAN, 37nm) ranged mainly from 20% to 70% for proteins and from 75% to 100% for PS. Especially protein rejection decreased at higher temperatures and higher nitrification activity. The increased fouling rates at lower temperatures might therefore partly be explained by this increased rejection. Apparently, mainly the nitrite-oxidising community is responsible for the formation for smaller SMP molecules that can pass the membrane.     

Biodegradation of distillery spent wash in anaerobic hybrid reactor

A lab-scale anaerobic hybrid (combining sludge blanket and filter) reactor was operated in a continuous mode to study anaerobic biodegradation of distillery-spent wash. The study demonstrated that at optimum hydraulic retention time (HRT), 5 days and organic loading rate (OLR), 8.7kgCOD/m(3)d, the COD removal efficiency of the reactor was 79%. The anaerobic reduction of sulfate increases sulfide concentration, which inhibited the metabolism of methanogens and reduced the performance of the reactors. The kinetics of biomass growth i.e. yield coefficient (Y,0.0532) and decay coefficient (K(d), 0.0041d(-1)) was obtained using Lawrence and McCarty model. However, this model failed in determining the kinetics of substrate utilization. Bhatia et al. model having inbuilt provision of process inhibition described the kinetics of substrate utilization, i.e. maximum rate of substrate utilization (R,1.945d(-1)) and inhibition coefficient values (K(i),0.032L/mg). Modeling of the reactor demonstrated that Parkin and Speece, and Bhatia et al. models, both, could be used to predict the effluent substrate concentration. However, Parkin and Speece model predicts effluent COD more precisely (within +/-2%) than Bhatia et al. model (within +/-5-20%) of the experimental value. Karhadkar et al. model predicted biogas yield within +/-5% of the experimental value.     

Influence of the agitation rate on the treatment of partially soluble wastewater in anaerobic sequencing batch biofilm reactor

This work reports on the influence of the agitation rate on the organic matter degradation in an anaerobic sequencing batch reactor, containing biomass immobilized on 3 cm cubic polyurethane matrices, stirred mechanically and fed with partially soluble soymilk substrate with mean chemical oxygen demand (COD) of 974+/-70 mg l(-1). Hydrodynamic studies informed on the homogenization time under agitagion rates from 500 to 1100 rpm provided by three propeller impellers. It occurred very quickly compared to the total cycle time. The results showed that agitation provided good mixing and improved the overall organic matter consumption rates. A modified first-order kinetic model represented adequately the data in the entire range of agitation rate. The apparent first-order kinetic constant for suspended COD rose approximately 360% when the agitation rate was changed from 500 to 900 rpm, whereas the apparent first-order kinetic constant for soluble COD did not vary significantly.     

Effect of anaerobic reactor process configuration on useful energy production

The effect of reactor process configuration on anaerobic production of useful energy (hydrogen and methane) from a complex substrate was investigated for the following reactor systems: suspended growth, two-phase mixed, two-stage mixed, upflow anaerobic sludge blanket (UASB) reactor, and two-phase UASB. The mixed two-phase and two-stage configurations yielded the highest specific energy productions of 13.3 and 13.4 kJ/g COD fed, respectively. Reactor process configuration influenced microbial pathways in acidogenic reactors in that butyrate was the predominant volatile acid in phased configurations, whereas acetate was predominant in the staged configuration. The UASB reactor achieved the highest average daily energy production per reactor volume of 101 kJ/L reactor-d. All reactor configurations achieved high COD removals on the order of 99%. However, hydrogen represented only 3% of the total energy produced by the two-phase mixed and two-phase UASB configurations. Theoretical analysis revealed that the maximum specific energy production by the two-phase suspended-growth configuration is only 9% higher than that for a single-stage mixed reactor. Consequently, the production of hydrogen from complex substrates in these process configurations does not seem to be justifiable solely from an energy point of view. Instead, it is suggested that phased anaerobic systems should be considered primarily for improved process stability whereas resultant hydrogen production is of secondary benefit.     

Impacts of hydrodynamic shear force on nucleation of flocculent sludge in anaerobic reactor

The Sludge granulation in an anaerobic reactor consists of two steps: nucleation and maturation of nuclei. Nucleation as the starting point is of particular importance. In this paper, the nucleation of flocculent sludge as seed under weak, strong and violent hydrodynamic shear conditions is studied with an original quantitative method, and then the satisfactory linear correlations between the average sludge diameters and the operation time during the nucleation are demonstrated. Nucleation under strong shear conditions with a shear rate of about 8.28 s⁻¹, corresponding to superficial liquid and gas velocities of 2.66 and 0.24 m h⁻¹, develops fastest compared to weak shear conditions with a shear rate of about 0.04 s⁻¹ and violent shear conditions with a shear rate of about 12.42 s⁻¹ with the average augmentation rate of average sludge diameter of 0.57, 0.40 and 0.41 μm day⁻¹ respectively. One of the major mechanisms of the shear force on nucleation is that a high shear force accelerates the extracellular protein secretion of sludge. Although high extracellular protein content benefits nucleation, it is also shown that the extracellular proteins over-produced above around 80.5 mg gVSS⁻¹ leads nuclei to weaken and inhibit nucleation. So the violent shear force would result in disruption and wash-out of nuclei. However, the high extracellular polymers could intensify the shear force by raising the viscosity in the reactor, thus, in practice, it is important to monitor the shear conditions and extracellular protein content of sludge simultaneously in high rate reactors for stable operation.     

Effect of linear alkylbenzene sulphonates (LAS) on the anaerobic digestion of sewage sludge

Batch anaerobic biodegradation tests with different alkylbenzene sulphonates (LAS) at increasing concentrations were performed in order to investigate the effect of LAS homologues on the anaerobic digestion process of sewage sludge. Addition of LAS homologues to the anaerobic digesters increased the biogas production at surfactant concentrations 5-10 g/kg dry sludge and gave rise to a partial or total inhibition of the methanogenic activity at higher surfactant loads. Therefore, at the usual LAS concentration range in sewage sludge, no adverse effects on the anaerobic digesters functioning of a wastewater treatment plant (WWTP) can be expected. The increase of biogas production at low surfactant concentrations was attributed to an increase of the bioavailability and subsequent biodegradation of organic pollutants associated with the sludge, promoted by the surfactant adsorption at the solid/liquid interface. When the available surfactant fraction in the aqueous phase instead of the nominal surfactant concentration was used to evaluate the toxicity of LAS homologues, a highly significant relationship between toxicity and alkyl chain length was obtained. Taking into account the homologue distribution of commercial LAS in the liquid phase of the anaerobic digesters of a WWTP, an EC(50) value of 14 mg/L can be considered for LAS toxicity on the anaerobic microorganisms.     

Submerged anaerobic membrane bioreactor for low-strength wastewater treatment: Effect of HRT and SRT on treatment performance and membrane fouling

Three 6-L submerged anaerobic membrane bioreactors (SAnMBRs) with solids retention times (SRTs) of 30, 60 and infinite days were setup for treating synthetic low-strength wastewater at hydraulic retention times (HRTs) of 12, 10 and 8 h. Total COD removal efficiencies higher than 97% were achieved at all operating conditions. Maximum biogas production rate was 0.056 L CH₄/g MLVSS d at an infinite SRT. A shorter HRT or longer SRT increased biogas production due to increased organic loading rate or enhanced dominancy of methanogenics. A decrease in HRT enhanced growth of biomass and accumulation of soluble microbial products (SMP), which accelerated membrane fouling. A drop in carbohydrate to protein ratio also inversely affected fouling. At 12-h HRT, the effect of SRT on biomass concentration in SAnMBRs was negligible and membrane fouling was controlled by variant surface modification due to different SMP compositions, i.e., higher carbohydrate and protein concentrations in SMP at longer SRT resulted in higher membrane fouling rate. At 8 and 10-h HRTs, infinite SRT in SAnMBR caused highest MLSS and SMP concentrations, which sped up particle deposition and biocake/biofilm development. At longer SRT, lower extracellular polymeric substances reduced flocculation of particulates and particle sizes, further aggravated membrane fouling.     

Simulation of the treatment performance of outdoor subsurface flow constructed wetlands in temperate climates

Numerical models are a means to increase the understanding of the processes occurring in the "black box" constructed wetland. Once reliable models for constructed wetlands are available they can be also used for evaluating and improving existing design criteria. The paper shows simulation results for outdoor experimental subsurface vertical flow constructed wetlands using CW2D, a multi-component reactive transport module developed to simulate transport and reactions of the organic matter, nitrogen and phosphorus in subsurface flow constructed wetlands. The surface area of the experimental vertical flow bed was 20 m(2). The organic load applied was 27 g COD m(-2) d(-1) (corresponding to a specific surface area of 3 m(2) per person). The aim of the work is to calibrate the model for temperature dependency that has been implemented in CW2D. Water temperature during the investigation period varied between 4 degrees C and 18 degrees C. The measured effluent concentrations during summer could be simulated using the standard CW2D parameter set when the flow model was calibrated well. However, the increasing effluent concentrations at low temperatures could not be simulated with the standard CW2D parameter set where temperature dependencies are considered only for maximum growth, decay, and hydrolysis rates. By introducing temperature dependencies for half-saturation constants for the hydrolysis and nitrification processes it was possible to simulate the observed behaviour. The work presented is a step on the way to validate the CW2D module. Model validation is a necessary step before numerical simulation can be finally used in practice, e.g. for checking existing design guidelines.     

Investigation of the sludge reduction mechanism in the anaerobic side-stream reactor process using several control biological wastewater treatment processes

To investigate the mechanism of sludge reduction in the anaerobic side-stream reactor (SSR) process, activated sludge with five different sludge reduction schemes were studied side-by-side in the laboratory. These are activated sludge with: 1) aerobic SSR, 2) anaerobic SSR, 3) aerobic digester, 4) anaerobic digester, and 5) no sludge wastage. The system with anaerobic SSR (system #2) was the focus of this study and four other systems served as control processes with different functions and purposes. Both mathematical and experimental approaches were made to determine solids retention time (SRT) and sludge yield for the anaerobic SSR process. The results showed that the anaerobic SSR process produced the lowest solids generation, indicating that sludge organic fractions degraded in this system are larger than other systems that possess only aerobic or anaerobic mode. Among three systems that involved long SRT (system #1, #2, and #5), it was only system #2 that showed stable sludge settling and effluent quality, indicating that efficient sludge reduction in this process occurred along with continuous generation of normal sludge flocs. This observation was further supported by batch anaerobic and aerobic digestion data. Batch digestion on sludges collected after 109 days of operation clearly demonstrated that both anaerobically and aerobically digestible materials were removed in activated sludge with anaerobic SSR. In contrast, sludge reduction in the aerobic SSR process or no wastage system was achieved by removal of mainly aerobically digestible materials. All these results led us to conclude that repeating sludge under both feast/fasting and anaerobic/aerobic conditions (i.e., activated sludge with anaerobic SSR) is necessary to achieve the highest biological solids reduction with normal wastewater treatment performance.     

High-rate anaerobic treatment of Fischer-Tropsch wastewater in a packed-bed biofilm reactor

This study investigates the anaerobic treatment of an industrial wastewater from a Fischer-Tropsch (FT) process in a continuous-flow packed-bed biofilm reactor operated under mesophilic conditions (35 °C). The considered synthetic wastewater has an overall chemical oxygen demand (COD) concentration of around 28 g/L, mainly due to alcohols. A gradual increase of the organic load rate (OLR), from 3.4 gCOD/L/d up to 20 gCOD/L/d, was adopted in order to overcome potential inhibitory effects due to long-chain alcohols (>C6). At the highest applied OLR (i.e., 20 gCOD/L/d) and a hydraulic retention time of 1.4 d, the COD removal was 96% with nearly complete conversion of the removed COD into methane. By considering a potential of 200 tCOD/d to be treated, this would correspond to a net production of electric energy of about 8 × 10⁷ kWh/year.During stable reactor operation, a COD balance and batch tests showed that about 80% of the converted COD was directly metabolized through H₂⁻ and acetate-releasing reactions, which proceeded in close syntrophic cooperation with hydrogenotrophic and acetoclastic methanogenesis (contributing to about 33% and 54% of overall methane production, respectively). Finally, energetic considerations indicated that propionic acid oxidation was the metabolic conversion step most dependent on the syntrophic partnership of hydrogenotrophic methanogens and accordingly the most susceptible to variations of the applied OLR or toxicity effects.     

Treatment of raw domestic sewage in an UASB reactor

The treatment of raw domestic sewage at ambient temperatures in an upflow anaerobic sludge blanket (UASB) reactor with a volume of 120 l. and a height of 1.92 m was studied. The sewage had an average BOD₅ of 357 mg l⁻¹ and COD of 627 mg l⁻¹. Approximately 75% of the organic materials were in the suspended fraction. The sewage temperature ranged from 18 to 28°C during the experimental period. The reactor operated continuously for 9 months and assessed self-inoculation and raw domestic sewage purification. The unit was started without inoculum and ran during the entire experimental period with a hydraulic retention time of 4 h. During the experiment, a sludge bed build-up was observed. At the end of the experimental period, the predominance of spherical granular particles up to 6–8 mm in diameter was evident.

After a 4-month operation, it was observed that the inoculation/acclimatization steps had been concluded. Removal efficiencies of BOD₅ = 78%, COD = 74% and TSS = 72% were obtained. A typical gas production factor of 80 l kg⁻¹ COD added was observed and the CH₄ content of the biogas was 69%.     

New approach to control the methanogenic reactor of a two-phase anaerobic digestion system

A new control strategy for the methanogenic reactor of a two-phase anaerobic digestion system has been developed and successfully tested on the laboratory scale. The control strategy serves the purpose to detect inhibitory effects and to achieve good conversion. The concept is based on the idea that volatile fatty acids (VFA) can be measured in the influent of the methanogenic reactor by means of titration. Thus, information on the output (methane production) and input of the methanogenic reactor is available, and a (carbon) mass balance can be obtained. The control algorithm comprises a proportional/integral structure with the ratio of (a) the methane production rate measured online and (b) a maximum methane production rate expected (derived from the stoichiometry) as a control variable. The manipulated variable is the volumetric feed rate. Results are shown for an experiment with VFA (feed) concentration ramps and for experiments with sodium chloride as inhibitor.     

Rheological and fractal characteristics of granular sludge in an upflow anaerobic reactor

The rheological and fractal characteristics of the granular sludge in an upflow anaerobic sludge blanket (UASB) reactor were investigated in this study. The influences of sludge concentration and temperature on the rheological properties of the granular sludge were evaluated, and the Bingham model was adopted to describe its rheology. In addition, image analysis was used to determine the sludge fractal dimension. The results indicate that the UASB granular sludge showed a shear-thinning behavior. The relationships between the limiting viscosity and the sludge concentration, as well as the limiting viscosity and temperature could be respectively modeled using an exponential equation and Arrhenius equation well. The Bingham model was able to adequately describe the rheology of the granular sludge. The fractal dimension of the granular sludge, 2.79+/-0.03, was larger than that of some other aggregates, suggesting that the granular sludge were more compact and denser. Furthermore, the relationship between rheological and fractal properties of the granular sludge could be properly described with the model proposed by Shih et al. [1990. Scaling behavior of the elastic properties of colloidal. Phys. Rev. A 42, 4772-4779].     

Nylon fibers as supporting media in anaerobic hybrid reactors: it's effects on system's performance and microbial distribution

The performances of three anaerobic hybrid reactors with various nylon fiber densities per packed bed volume (33, 22, and 11 kg/m(3) in R1, R2, and R3, respectively) as supporting media were evaluated through their ability to remove organic compounds in cassava starch wastewater. In addition, the distributions of non-methanogenic and methanogenic population in the reactors were investigated. During a 6-month operation, the organic loading rate was increased in stepwise from 0.5 to 4.0 kg COD/m3/day and the hydraulic retention time (HRT) shortened to 5.4 days. The COD removal efficiency was more favorable in R1 (87%) and R2 (84%) than in R3 (70%). The total biomass in the reactors with greater nylon fiber densities was also higher and increased from 20.4 to 67.3 g VSS and to 57.5 g VSS in R1 and R2, respectively. When the HRT was further shortened to 3 days, however, the efficiency of both reactors demonstrated a declining trend and reached 74% in R1 and 61% in R2. The distribution of microbial populations involved in the reactors was determined using the Most Probable Number technique. The result showed the lowest number of methanogens in R3 which correlated well to its relatively low efficiency. The number of non-methanogens in all reactors was, nonetheless, comparable. By shortening the HRT to 3 days, the methanogenic population in R2 diminished in both attached and suspended biomass whereas a slight reduction was detected only in the attached biomass of R1.     

Investigation of green roof thermal performance in temperate climate: A case study of an experimental building in Florianópolis city, Southern Brazil

Green roofs have been investigated as a bioclimatic strategy to improve the energy efficiency of buildings. Quantitative data on this subject are still needed for many specific climatic conditions. This paper deals with the investigation of the green roof thermal performance of an experimental single-family residence in Florianópolis (SC, Brazil), a southern city with a temperate climate. Field measurements during a warm period (01-March-2008-07-March-2008) and during a cold period (25-May-2008-31-May-2008) included internal air temperature of rooms, internal and external surface temperature of three types of roofs (green, ceramic and metallic), heat fluxes through these roofs, green roof's temperature profile, water volumetric content in substrate layer and meteorological data. During the warm period, the green roof reduced heat gain by 92-97% in comparison to ceramic and metallic roofs, respectively, and enhanced the heat loss to 49 and 20%. During the cold period, the green roof reduced heat gain by 70 and 84%, and reduced the heat loss by 44 and 52% in comparison to ceramic and metallic roofs, respectively. From the derived data it has been confirmed that green roof contributes to the thermal benefits and energy efficiency of the building in temperate climate conditions.     

Effect of process configuration and substrate complexity on the performance of anaerobic processes

The roles of substrate complexity (molecular size of the substrate) and process configuration in anaerobic wastewater treatment were investigated to determine optimal methanogenic technology parameters. Five substrates (glucose, propionate, butyrate, ethanol, and lactate) plus a mixed waste (60% carbohydrate, 34% protein, and 6% lipids) were studied under five reactor configurations: batch-fed single-stage continuous stirred tank reactor (CSTR), continuously fed single-stage CSTR, two-phase CSTR, two-stage CSTR, and single-stage upflow anaerobic sludge blanket (UASB). The substrate feed concentration was 20,000 mg/L as COD. The solids retention time (SRT) and hydraulic retention time (HRT) in the CSTR reactors were 20 d, while HRT in the UASB was 2 d. All reactors were operated for at least 60 d (equal to 3SRT). Substrate complexity was observed to be less significant under two-phase, two-stage and UASB reactor configurations. Two-phase CSTR, two-stage CSTR, and single-stage UASB configurations yielded the lowest effluent chemical oxygen demands (130-550, 60-700, and 50-250 mg/L, respectively). The highest effluent chemical oxygen demands were detected when feeding glucose, propionate, and lactate to continuously fed single-stage CSTRs (10, 400, 9900, and 4700 mg/L COD, respectively) and to batch-fed single-stage CSTRs (11, 200, 2500, and 2700 mg/L COD, respectively). Ironically, the one stage CSTR--most commonly utilized in the field--was the worst possible reactor configuration.