Performance and design considerations for an anaerobic moving bed biofilm reactor treating brewery wastewater: Impact of surface area loading rate and temperature.

Three 4 L anaerobic moving bed biofilm reactors (AMBBR) treated brewery wastewater with AC920 media providing 680 m2 protected surface area per m3 of media. Different hydraulic retention times (HRT; 24, 18, 12, 10, 8 and 6 h) at 40% media fill and 35 °C, as well as different temperatures (15, 25 and 35 °C) at 50% media fill and 18 h HRT were examined. Best performance at 35 °C and 40% media fill was observed when HRT was 18 h, which corresponded with 92% removal of soluble COD (sCOD). Organic loading rates (OLR) above 24 kg-COD m-3d-1 decreased performance below 80% sCOD removal at 35 °C and 40% media fill. The reason was confirmed to be that surface area loading rates (SALR) above 50 g-sCOD m-2d-1 caused excessive biofilm thickness that filled up internal channels of the media, leading to mass transfer limitations. Temperature had a very significant impact on process performance with 50% media fill and 18 h HRT. Biomass concentrations were significantly higher at lower temperatures. At 15 °C the mass of volatile solids (VS) was more than three times higher than at 35 °C for the same OLR. Biofilms acclimated to 25 °C and 15 °C achieved removal of 80% sCOD at SALR of 10 g-sCOD m-2d-1 and 1.0 g-sCOD m-2d-1, respectively. Even though biomass concentrations were higher at lower temperature, biofilm acclimated to 25 °C and 15 °C performed significantly slower than that acclimated to 35 °C.

Removal of Soluble Phosphorus from Surface Water Using Iron (Fe-Fe) and Aluminum (Al-Al) Electrodes.

The removal of soluble phosphorus using iron and aluminum electrodes was studied in water samples from the Red River, a hyper-eutrophic stream in Winnipeg, Canada. Four trials were conducted: (I) mixed batch with 150-900 mA applied for 1 min to 1 L, (II) stagnant batch with 600-900 mA applied for 1 min to 1 L, and (III and IV) continuously stirred-tank reactor with 6.25-10 min hydraulic retention times and constant 900 mA. Maximum soluble phosphorus removals of 70-80% were observed in mixed batch, and there was no significant difference between aluminum and iron electrodes (P value of 0.0526-0.9487). Aluminum electrodes performed significantly worse than iron electrodes under higher hydraulic loads, with iron removing >70% soluble phosphorus and aluminum <40% (P values of 0.0035-0.0143). The estimated cost of consumables, reported per million liters of water treated, to remove 70% soluble phosphorus from eutrophic waters with 0.35 g m-3 soluble phosphorus would include 5-17.5 USD electricity costs and material costs of 5.3-12.2 USD for iron and 39.2 USD for aluminum.

Granulation of activated sludge under low hydrodynamic shear and different wastewater characteristics.

Five reactors were operated with low upflow superficial air velocities (0.41cmmin-1) in order to observe granulation on synthetic wastewaters with different characteristics: 1) 340mg-CODL-1; 2) 630mg-CODL-1; and 3) 1300mg-CODL-1. Stable granulation was only observed under low hydrodynamic shear for low-strength wastewater. 55-70% of soluble chemical oxygen demand (COD) was utilized before aeration and 91% COD, 62% total nitrogen (TN), and 96% total phosphorus (TP) were removed from the low-strength wastewater. Although medium-strength wastewater did generate granules they rapidly acquired a filamentous surface layer that resulted in decreased performance and loss of nitrification. 94% COD, 30% TN, and 85% TP were removed from the medium-strength wastewater. The high-strength wastewater did not develop granules and 85% COD was removed. Results demonstrated that high shear force was not required for granulation. Rather, granulation depended on multiple parameters to out-select rapidly growing aerobic microorganisms.

Potential of hydrolysis of particulate COD in extended anaerobic conditions to enhance biological phosphorous removal.

The effect of anaerobic hydrolysis of particulate COD (pCOD) on biological phosphorous removal in extended anaerobic condition was investigated through (i) sequencing batch reactors (SBR)s with anaerobic hydraulic retention time (HRT) of 0.8, 2, and 4 h; (ii) batch tests using biomass from a full scale biological nutrient removal (BNR) plant; and (iii) activated sludge modeling (BioWin 4.1 simulation). The results from long-term SBRs operation showed that phosphorus removal was correlated to the ratio of filtered COD (FCOD) to total phosphorus (TP) in the influent. Under conditions with low FCOD/TP ratio (average of 20) in the influent, extending anaerobic HRT to 4 h in the presence of pCOD did not significantly improve overall phosphorous removal. During the period with high FCOD/TP ratio (average of 37) in the influent, all SBRs removed phosphorous completely, and the long anaerobic HRT did not have negative effect on overall phosphorous removal. The batch tests also showed that pCOD at different concentration during 4 h test did not affect the rate of anaerobic phosphorus release. The rate of anaerobic hydrolysis of pCOD was significantly low and extending the anaerobic HRT was ineffective. The simulation (BioWin 4.1) of SBRs with low influent FCOD/TP ratio showed that the default kinetics of anaerobic hydrolysis in ASM2d overestimated phosphorous removal in the SBRs (high anaerobic hydrolysis of pCOD). The default anaerobic hydrolysis rate in BioWin 4.1 (ten times lower) could produce similar phosphorous removal to that in the experiment. Results showed that the current kinetics of anaerobic hydrolysis in ASM2d could lead to considerable error in predicting phosphorus removal in processes with extended anaerobic HRT. Biotechnol. Bioeng. 2016;113: 2377-2385. © 2016 Wiley Periodicals, Inc.

Modification of activated sludge properties caused by application of continuous and intermittent current.

This study investigated the impact of direct current (DC) field on the activated sludge properties for potential improvement of the biological as well as membrane treatment processes. Three mixed-liquor suspended solids (MLSS) concentrations (5,000, 10,000 and 15,000 mg/l) were subjected to current densities (CD) ranging from 5 to 50 A/m² at five electrical exposure modes (time-ON/time-OFF). The results showed that CD between 15 and 35 A/m² increased the filterability of the sludge more than 200 times when compared with the untreated reference sludge. The average removals of protein, polysaccharides and organic colloids from the sludge supernatant at this range of CD were 43%, 73% and 91%, respectively, while the average reduction of the specific resistance to filtration (SRF) was 4.8 times higher. The changes of sludge properties depended on the current density, electrical exposure mode and the MLSS concentration. At CD of 25 A/m² and MLSS below 10,000 mg/l, shorter time-OFF was needed in each electrical cycle, while more time-OFF was needed at higher MLSS concentrations. It was concluded that proper application of the DC field could improve biomass in terms of its dewaterability and the removal of SMP, which are highly correlated to membrane fouling in the submerged membrane electro-bioreactor (SMEBR).

Impact of free ammonia on anammox rates (anoxic ammonium oxidation) in a moving bed biofilm reactor.

Using a bench scale moving bed bioreactor (MBBR), the effect of free ammonia (FA, NH(3), the un-ionized form of ammonium NH(4)(+)) concentration on anoxic ammonium oxidation (anammox) was evaluated based on the volumetric nitrogen removal rate (NRR). Although, a detailed microbial analysis was not conducted, the major NRR observed was assumed to be by anammox, based on the nitrogen conversion ratios of nitrite to ammonium and nitrate to ammonium. Since the concentration of free ammonia as a proportion of the total ammonia concentration is pH-dependent, the impact of changing the operating pH from 6.9 to 8.2, was investigated under constant nitrogen loading conditions during continuous reactor operation. Furthermore, the effect of sudden nitrogen load changes was investigated under constant pH conditions. Batch tests were conducted to determine the immediate response of the anammox consortium to shifts in pH and FA concentrations. It was found that FA was inhibiting NRR at concentrations exceeding 2 mg N L(-1). In the pH range 7-8, the decrease in anammox activity was independent of pH and related only to the concentration of FA. Nitrite concentrations of up to 120 mg N L(-1) did not negatively affect NRR for up to 3.5 h. It was concluded that a stable NRR in a moving bed biofilm reactor depended on maintaining FA concentrations below 2 mg N L(-1) when the pH was maintained between 7 and 8.

Electrically enhanced MBR system for total nutrient removal in remote northern applications.

Thousands of sparsely populated communities scatter in the remote areas of northern Canada. It is economically preferable to adopt the decentralized systems to treat the domestic wastewater because of the vast human inhabitant distribution and cold climatic conditions. Electro-technologies such as electrofiltration, elctrofloatation, electrocoagulation and electrokinetic separation have been applied in water and conventional wastewater treatment for decades due to the minimum requirements of chemicals as well as ease of operation. The membrane bioreactor (MBR) is gaining popularity in recent years as an alternative water/wastewater treatment technology. However, few studies have been conducted to hyphenate these two technologies. The purpose of this work is to design a novel electrically enhanced membrane bioreactor (EMBR) as an alternative decentralized wastewater treatment system with improved nutrient removal and reduced membrane fouling. Two identical submerged membranes (GE ZW-1 hollow fiber module) were used for the experiment, with one as a control. The EMBR and control MBR were operated for 4 months at room temperature (20 ± 2 °C) with synthetic feed and 2 months at 10 °C with real sewage. The following results were observed: (1) the transmembrane pressure (TMP) increased significantly more slowly in the EMBR and the interval between the cleaning cycles of the EMBR increased at least twice; (2) the dissolved chemical oxygen demand (COD) or total organic carbon (TOC) in the EMBR biomass was reduced from 30 to 51%, correspondingly, concentrations of the extracellular polymeric substances (EPS), the major suspicious membrane foulants, decreased by 26-46% in the EMBR; (3) both control and EMBR removed >99% of ammonium-N and >95% of dissolved COD, in addition, ortho-P removal in the EMBR was >90%, compared with 47-61% of ortho-P removal in the MBR; and (4) the advantage of the EMBR over the conventional MBR in terms of membrane fouling retardation and phosphorus removal was further demonstrated at an operating temperature of 10 °C when fed with real sewage. The EMBR system has the potential for highly automated control and minimal maintenance, which is particularly suitable for remote northern applications.

Acetate and propionate impact on the methanogenesis of landfill leachate and the reduction of clogging components.

Synthetic leachate with different initial concentrations of acetate (500-2500 mg HAc/L) and propionate (500-3500 mg HPr/L) was treated with active biomass acclimated to landfill leachate under anaerobic mesophilic conditions for 72 h. Methanogenesis was observed within all samples during the first 48 h. The greatest removal of acetate (80-100%) and propionate (15-35%) was achieved in tests with initial concentrations ranging from 1500 to 190 0mg HAc/L and from 1000 to 1800mgHPr/L. Concurrent with the removal of acids, pH increased between 0.3 and 0.45 units, to above the threshold pH for precipitation of CaCO(3). Therefore, some 50-70% of dissolved Ca(2+) was removed from solution. This study suggests that by using an equalization tank (prior to the anaerobic digester) to maintain the acetate and propionate concentrations to within their optimum range would help to alleviate clogging through the removal of dissolved Ca(2+) and maximizing CH(4) production.

Influence of electric current on bacterial viability in wastewater treatment.

Minimizing the influence of electric current on bacterial viability in the electro-technologies such as electrophoresis and electrocoagulation is crucial in designing and operating the electric hybrid wastewater treatment system. In this study the biomass from a membrane bioreactor (MBR) was subjected to constant direct current and the bacterial viability was monitored against electrical intensity, duration as well as the spatial vicinity related to the electrodes. It was found that the bacterial viability was not significantly affected (less than 10% of death percentage) when the applied electric current density (CD) was less than 6.2 A/m2 after 4 h. The percentage of live cell dropped by 15% and 29% at CD of 12.3 A/m2 and 24.7 A/m2, respectively. The pH of electrolytic biomass fluid has shifted to alkaline (from nearly neutral to around pH 10) at CD above 12.3 A/m2, which could have been the contributing factor for the bacterial inactivation. The temperature change in the electrolytic media at all current densities during 4 h of experiment was less than 2 °C, thus temperature effects were negligible. Bacteria experienced different micro-environments in the electrochemical reactor. Bacterial cells on the cathode surface exhibited highest death rate, whereas bacteria outside the space between electrodes were the least affected. It was concluded that in an electro-technology integrated wastewater treatment process, sufficient mixing should be used to avoid localized inactivation of bacterial cells.

Autotrophic ammonium removal from reject water: partial nitrification and anammox in one-reactor versus two-reactor systems.

Removal of total nitrogen from anaerobically digested sludge reject water by a fully autotrophic process in either one- or two-reactor systems was compared. Autotrophic nitrogen removal is currently most often applied in the one-reactor system primarily in the DEMON configuration. The two-reactor systems had a similar nitrogen removal rate to the one-reactor systems. It was evident that the limiting step was partial nitrification. Increase in partial nitrification can be difficult in a one-reactor system, where the overall conditions, such as the oxygen concentration or substrate concentrations, would progress to full nitrification. The partial nitrification of the two-reactor system may be improved by adjusting key parameters: the un-ionized form of the substrates, the oxygen concentration or the solids residence time. The impact of two process configurations on the operational stability and process performance was presented based on two cases--DEMON and the SHARON-ANAMMOX process.

Importance of the operating pH in maintaining the stability of anoxic ammonium oxidation (anammox) activity in moving bed biofilm reactors.

Two bench-scale parallel moving bed biofilm reactors (MBBR) were operated to assess pH-associated anammox activity changes during long term treatment of anaerobically digested sludge centrate pre-treated in a suspended growth partial nitrification reactor. The pH was maintained at 6.5 in reactor R1, while it was allowed to vary naturally between 7.5 and 8.1 in reactor R2. At high nitrogen loads reactor R2 had a 61% lower volumetric specific nitrogen removal rate than reactor R1. The low pH and the associated low free ammonia (FA) concentrations were found to be critical to stable anammox activity in the MBBR. Nitrite enhanced the nitrogen removal rate in the conditions of low pH, all the way up to the investigated level of 50mg NO(2)-N/L. At low FA levels nitrite concentrations up to 250 mg NO(2)-N/L did not cause inactivation of anammox consortia over a 2-days exposure time.

VFA generation from waste activated sludge: effect of temperature and mixing.

The success of enhanced biological phosphorus removal (EBPR) depends on the constant availability of volatile fatty acids (VFAs). To reduce costs, waste streams would be a preferred source. Since VFAs were shown to vary in the incoming sewage and fermentate from primary sludge the next available source is waste activated sludge (WAS). The opportunity is particularly good in plants where WAS is stored before shipment. Little information is however available on the rate of VFA release from such sludge, especially at the lower temperatures and under the storage conditions typically found in colder climates. Bench-scale batch tests were performed to investigate the effect of temperature and requirement for mixing on VFA generation from WAS generated in full scale non-EBPR wastewater treatment plant. WAS fermentation was found highly temperature-dependent. Hydrolysis rate constant (k(h)) values of 0.17, 0.08 and 0.04 d⁻¹ at 24.6, 14 and 4°C were obtained, respectively. Arrhenius temperature coefficient was calculated to be 1.07. It took 5 d to complete hydrolysis at 24.6°C, 7 d at 14°C, and 9 d at 4°C. The fermentation lasted for 20 d. At 24.6°C the mixed reactor reached 84% of the overall VFA production only in 5 d. When temperature dropped to 14 and 4°C, the ratio of VFA production at day 10 to overall VFA production in the mixed reactor were 62% and 48%, respectively. The overall VFA-COD concentration in the non-mixed reactors was much lower than the mixed reactors. The information is important for the designer as there was uncertainty with the effect of temperature and mixing on sludge fermentation.

Effect of sludge type on the fermentation products.

Primary sludge (PS), waste activated sludge (WAS) and a mixture of WAS and PS were fermented at 21 degrees C. The sludge was collected from two plants: the biological nutrient removal (BNR) West End Water Pollution Control Center (WEWPCC) and carbon-removal only South End WPCC (SEWPCC). The PS fermentation predictably generated a significantly higher amount of soluble COD than WAS. WAS fermentation released considerable amounts of phosphate and ammonium. Co-fermentation of WAS with PS enhanced soluble COD production and increased the release of phosphate and ammonium. The semi-continuous flow fermenters showed that regardless of the sludge source, with a similar total COD load, there was no significant difference in soluble COD production observed during co-fermentation between the two plants. Volatile fatty acids were the primary components of the soluble COD generated during fermentation. 20-22% volatile solids destruction was achieved due to sludge fermentation. The WEWPCC sludge released a higher concentration of phosphate than the SEWPCC sludge - the latter originating from a non-BNR process. Fermentation of combined PS and WAS sludge generated a concentration of phosphate high enough to allow phosphorus recovery as struvite at both plants.

Enhancing biological phosphorus removal with glycerol.

An enhanced biological phosphorus removal process (EBPR) was successfully operated in presence of acetate. When glycerol was substituted for acetate in the feed the EBPR process failed. Subsequently waste activated sludge (WAS) from the reactor was removed to an off-line fermenter. The same amount of glycerol was added to the WAS fermenter which led to significant volatile fatty acids (VFA) production. By supplying the system with the VFA-enriched supernatant of the fermentate, biological phosphorus removal was enhanced. It was concluded that, if glycerol was to be used as an external carbon source in EBPR, the effective approach was to ferment glycerol with waste activated sludge.

Biomass fermentation to augment biological phosphorus removal.

A combination of a lab scale biological phosphorus removal sequencing batch reactor (called mother reactor) and a side-stream biomass fermenter was setup. It was found that when fermented biomass was recirculated back into the mother reactor as volatile fatty acid (VFA) supplement, the phosphate concentration in the effluent decreased from 6 in the control reactor to 4.5 mgL(-1) in the effluent from mother reactor. The addition of the fermentation effluent into the mother reactor increased the phosphate and ammonium loads and resulted in deterioration of nitrification. Phosphorus removal and nitrification improved when the fermented biomass was separated from the liquid phase using an up-flow system, followed by the addition of MgO to the supernatant to precipitate phosphate and ammonium. Phosphorus removal was further improved by delaying the time of VFA addition into mother reactor during the anaerobic period as soon as denitrification ceased. Biomass fermentation was found to generate 157 mg VFA-COD by fermenting 1g of biomass at a solids retention time of 5d. Acetate (78% of generated COD) and propionic acid (10%) were the major components of the produced VFA. It was concluded that biomass fermentation to augment a biological nutrient removal process can be effective if generated phosphate and ammonia are removed, e.g. through struvite precipitation.

Nutrient removal in an electrically enhanced membrane bioreactor.

Integration of the membrane bioreactor (MBR) into wastewater treatment facilities has gained popularity in recent years due to increasingly stringent discharge permits. However, up to now no research has been conducted on the combination of nitrification, denitrification and electrochemical phosphorus removal into a MBR system. In this study a novel electrically enhanced MBR (EMBR) system was used. Without pH adjustment and external carbon source supplementation, using synthetic feed, ammonium-nitrogen was completely eliminated; COD, total nitrogen and ortho-phosphorus were removed by 94.3%, 77% and 86.6%, respectively. The power consumption was 0.22 kW/m(3) of the influent synthetic wastewater. With a control MBR run in parallel, the applied voltage gradient of 1.82 V/cm did not exhibit adverse influence on the microbial growth. This system has the potential to achieve phosphorus removal through alternating the direct current intensity.

Long-term operation of membrane biofilm reactors for nitrogen removal with autotrophic bacteria.

Efficient gas delivery and biofilm development on membrane fibers in a membrane biofilm reactor (MBfR) would be well suited to autotrophic nitrification and denitrification using hydrogen. Total nitrogen removal in a two-step MBfR system incorporating sequential nitrification and hydrogen-driven autotrophic denitrification was investigated in order to achieve nitrogen removal by autotrophic bacteria alone. This study also aimed at the long-term stable operation, which proved difficult in previous studies due to excessive biofilm accumulation in autotrophic denitrification systems. Consecutive operation of nitrification and autotrophic denitrification lasted 230 days. Average specific nitrification rate of 1.87 g N/m(2) d was achieved and the performance was very stable throughout the experimental periods. Nitrification performance from this study showed comparable rates to previous studies although this work was conducted at slightly lower temperature. Batch tests confirmed the presence of nitrifiers from the effluent of the nitrification reactor, which reattached to the biofilm in the denitrification reactor leading to further nitrification. Performance of autotrophic denitrification was maintained stably throughout the experimental periods, however biofilm control by nitrogen sparging was required for process stability. Average specific denitrification rate of 1.41 g N/m(2) d and a maximum specific denitrification rate of 2.04 g N/m(2) d was maintained. This study showed that, with an appropriate biofilm control plan, stable long-term operation of a fully autotrophic MBfR system for total nitrogen removal was possible without major membrane cleaning procedures.

Waste activated sludge fermentation: effect of solids retention time and biomass concentration.

Laboratory scale, room temperature, semi-continuous reactors were set-up to investigate the effect of solids retention time (SRT, equal to HRT hydraulic retention time) and biomass concentration on generation of volatile fatty acids (VFA) from the non-methanogenic fermentation of waste activated sludge (WAS) originating from an enhanced biological phosphorus removal process. It was found that VFA yields increased with SRT. At the longest SRT (10d), improved biomass degradation resulted in the highest soluble to total COD ratio and the highest VFA yield from the influent COD (0.14g VFA-COD/g TCOD). It was also observed that under the same SRT, VFA yields increased when the biomass concentration decreased. At a 10d SRT the VFA yield increased by 46%, when the biomass concentration decreased from 13g/L to 4.8g/L. Relatively high nutrient release was observed during fermentation. The average phosphorus release was 17.3mg PO(4)-P/g TCOD and nitrogen release was 25.8mg NH(4)-N/g TCOD.

Nitrifying genera in activated sludge may influence nitrification rates.

Sequencing batch reactors were acclimated under aerobic and alternating anoxic/aerobic conditions. Greater nitrification rates in the alternating reactor were investigated by comparing environmental conditions. In the alternating reactor, pH, alkalinity, oxygen, and nitrite were higher at the onset of aerobic nitrification. Kinetic studies and batch tests, with biomass developed under aerobic and alternating conditions, revealed that these factors were insufficient to explain the divergent nitrification rates. Nitrifying genera vary in nitrification kinetics and sensitivity to environmental conditions. Nitrosospira and Nitrospira spp. could dominate in aerobic reactors, as they are adapted to low nitrite and oxygen conditions. Nitrosomonas and Nitrobacter spp. are better competitors with abundant substrates and have higher nitrite tolerance, so they could excel under alternating conditions. This theoretical explanation is consistent with the kinetics and environmental conditions in these reactors and argues for using alternating treatment, because the harsh conditions select for populations with inherently faster nitrification rates.

Impact of shear force on the biofilm structure and performance of a membrane biofilm reactor for tertiary hydrogen-driven denitrification of municipal wastewater.

Hydrogen-driven denitrification using a hollow-fiber membrane biofilm reactor (MBfR) was evaluated for operation in tertiary wastewater treatment. Specific objectives were to evaluate the impact of different levels of shearing stress caused by mixing and nitrogen sparging on the biofilm structure and denitrification rates. Applying high shear force proved to be effective in improving denitrification rates by reducing the thickness of the biofilm. With intensive mixing a biofilm thickness of approximately 800 microm was maintained, while additional nitrogen sparging could further reduce the biofilm thickness to approximately 300 microm. The highest denitrification rates of 0.93 gN/m(2)d were obtained when biofilm thickness was lower than 500 microm. Lower extracellular polymeric substances (EPS) accumulation and carbohydrates to protein ratio observed in thinner biofilms allowed for higher nitrate removal in the system. No significant sloughing of biomass or change in total and soluble COD in the final effluent was observed under steady-state conditions.