Anaerobic membrane bioreactor for the treatment of synthetic wastewater to produce energy and water reusable in agricultural sector

Sustainable development and efficient management of water is an increasingly complex challenge in India. Increasing population, growing urbanization, and rapid industrialization combined with the need for raising agricultural production generates competing claims for water.

India accounts for 2.45% of land area and 4% of water resources of the world but represents 16% of the world population. Within the limitations of physiographic conditions, socio-political environment, legal and constitutional constraints and the technology available at hand, the utilizable water resources of the country have been assessed at 1,123 billion cubic meters (BCM), of which 690 BCM is from surface water and 433 BCM from groundwater sources (UNICEF, FAO and SaciWATERs 2013). Out of total water usage, about 85% (688 BCM) of water usage is being diverted for irrigation, which may increase to 1,072 BCM by 2050 (Central Water Commission 2010). Major source for irrigation is groundwater. The water availability for irrigation is expected to reduce in near future. With the present population growth rate (1.9% per year), the population is expected to cross the 1.5 billion mark by 2050. Owing to increasing population and all round development in the country, the per capita average annual freshwater availability has been reducing from 5,177 m³ in 1951 to 1,869 m³ in 2001, and 1,588 m³ in 2010. It is expected to further reduce to 1,341 m³ in 2025 and 1,140 m³ in 2050 (UNICEF, FAO and SaciWATERs 2013). Hence, there is an urgent need for efficient water resource management through enhanced water use efficiency and waste water recycling.

With rapid expansion of cities and domestic water supply, quantity of wastewater is increasing in the same proportion. As per central pollution control board (CPCB) estimates, the total wastewater generation from Class I cities (498) and Class II (410) towns in the country is around 35,558 and 2,696 million liters per day (MLD), respectively. However, the installed sewage treatment capacity is just 11,553 and 233 MLD, respectively, thereby leading to a gap of 26,468 MLD in sewage treatment capacity. The remaining sewage is disposed into the water bodies without any treatment due to which three-fourths of surface water resources are polluted (CPCB 2009; Ministry of Urban Development, Government of India 2012). While the per capita water availability is reducing considerably, the domestic wastewater is polluting water reservoirs further and quality of available water is deteriorating considerably. It is projected that by 2050, about 48.2 BCM (132 billion liters per day) of wastewaters (with a potential to meet 4.5% of the total irrigation water demand) would be generated thereby further widening the gap (Bhardwaj 2005). Thus, overall analysis of water resources indicates that in coming years, there will be a twin edged problem to deal with reduced fresh water availability and increased wastewater generation due to increased population and industrialization. Considering the urgency of preventing pollution of our water bodies and preserving our precious water resources, sewage treatment and reutilization of treated sewage need to be accorded higher priority. The first emphasis should be given to development of 100% treatment capacity and diversion of treated sewage for its utilization in irrigation.

Different treatment technologies are employed in various sewage treatment plants across India. Activated sludge process is the most commonly employed technology, followed by upflow anaerobic sludge blanket and waste stabilization ponds technology. There are number of newer treatment technologies that have come into practice in recent times and becoming more popular for sewage treatment. These include moving bed biofilm reactor/fluidized aerobic bioreactor, sequencing batch reactor, and MBR (National River Conservation Directorate Ministry of Environment & Forests, Government of India 2009).

Conventional treatment technologies generate poor quality of effluent and have many operational issues. Advanced treatment technologies are mostly energy intensive due to aeration cost. The anaerobic treatment system has lesser operating cost than aerobic treatment and therefore anaerobic treatment is more preferable option in case of high-strength effluent; however, it has many operational issues and the quality of effluent is even poorer compared to aerobic treatment.

MBR represent an attractive technological solution which has proven its worth for the efficient treatment of industrial and municipal wastewater (Kang et al. 2002; Kanai et al. 2010; Meng et al. 2009). However, membrane fouling, an intrinsic phenomenon linked to MBR operation, requires periodic chemical cleaning, one of the major drawbacks of this technology. Energy intensive operation of MBR is one of the important factor creating limitation for the deployment of the technology in India. Application of anaerobic processes for the treatment of low-strength wastewater has drawn considerable attention recently due to its inherent benefits compared to aerobic treatment, including less energy consumption, less sludge production, and valuable methane generation. However, maintaining a long sludge retention time is one of the challenges of anaerobic treatment processes due to the slow growth rate of anaerobic micro-organisms (Jaeho & Shihwu 2010).

Considering various lacunas of conventional wastewater treatment technologies and cost implications of advanced technologies in Indian context, a need exist to develop sustainable technologies which can provide recyclable quality treated water with low-operating cost. An anaerobic bioreactor coupled with membrane unit is phrased as anaerobic membrane bioreactor (AnMBR) and could be an effective solution to treat low-strength domestic wastewater. Most of the AnMBRs in wastewater treatments have used the external configuration although over the last few years there has been increased research into submerged AnMBRs (Gimenez et al. 2011; Hu & Stuckey 2006; Jeison & Van Lier 2008; Van Zyl et al. 2008).

In recent years, there has been increase in attention on AnMBRs studies, due to their advantages for sustainable wastewater treatment (Aquino et al. 2006; Hu & Stuckey 2006; Lew et al. 2009). The membranes used in AnMBR facilitate complete retention of biomass in the bioreactor and thus offers possibility of operating the system at high mixed liquor suspended solids concentration. As a consequence, AnMBR offers the complete separation of hydraulic retention time (HRT) and sludge retention time, which facilitates a more flexible control of operation parameters (He et al. 2005). The AnMBRs are also expected to provide more efficient digestion, higher methane production, better effluent quality, and can be smaller in size than conventional anaerobic digesters (Padmasiri et al. 2007). In addition, AnMBRs appear suitable for the treatment of wastewaters with high organic suspended solids content, since particles are confined inside the reactor, allowing their degradation (Fuchs et al. 2003; Harada et al. 1994). Moreover, because permeate through the membrane is free of solids or cells, treated effluent will require fewer post treatment steps if reuse or recycle is of interest.

Much of the laboratory scale studies in AnMBR have been completed by various researchers; however, commercialization of the technology is a major challenge. Long-term performance of AnMBR is needed to be evaluated for actual wastewater treatment (Lin et al. 2013). Studies in scaled up configurations of AnMBR will be helping for identifying the operational issues and risk factors in scaling up of the technology. The objective of this study was to evaluate the long-term performance of bench-scale external AnMBR for the treatment of synthetic wastewater representing domestic wastewater. The performance of AnMBR was evaluated in terms of organic removal efficiency, biogas generation, biogas yield, and membrane filtration. The composition of synthetic wastewater used for the study was determined considering the composition of low to medium strength domestic wastewater shown in Table 1 (Metcalf and Eddy Inc. 2003).

A CSTR type of side stream configuration of AnMBR was used for this study. The bench-scale setup of AnMBR comprises of bioreactor and membrane filtration unit is described in following section. A schematic diagram of bench-scale setup used for this study is shown in Figure 1. Actual photograph of the bench-scale setup is shown in Figure 2.

A 670 l capacity cylindrical bioreactor was designed and fabricated in stainless steel 304. The bioreactor was fully automatic for operation, equipped with pH as well as temperature measurement and control. A limpet coil jacket was provided around the bioreactor vessel for heating or cooling purpose. A water heating bath and pump (make – Raj Pumps Pvt. Ltd., Jodhpur, Type – RBSP SS30) was used for circulating the heated water through jacket coil and maintaining reactor temperature at desired level. Two-stage blade type agitator was provided at the top of the bioreactor for agitation purpose. Flange mounted, 2 HP variable speed motor supplied by Crompton Greaves was used as a drive for agitator. Agitator speed display (make – Electronics Systems and Devices, Pune, model – ESD 9010) and variable frequency drive (Schneider make) for changing the speed of agitator were used in the control panel of the setup. A temperature display along with controller supplied by Electronics Systems and Devices, Pune was provided in the control panel of the unit. A pH probe supplied by Hanna Instruments was mounted at the bottom of bioreactor for the measurement of pH. Acid and alkali dosing arrangement was made in place for the control of pH. Hanna make pH display along with controller was provided in the control panel of the unit. Capacity feed tank of 2,000 l was fabricated in mild steel and used for the storage of feed water.

A drain connection having a ball valve is provided at the bottom of the feed tank for draining and cleaning purpose. A self-priming centrifugal pump (make – Compton Greaves, Model – Ministar –III) was used for the pumping of feed to the bioreactor. A magnetic float operated guided level switch supplied by Pune Techtrol Pvt. Ltd., Pune (Model – FGSO-J12O1WWW) was mounted in the bioreactor for the operation and control of the feed pump. The generated biogas collected in the head space of the bioreactor was taken out of the bioreactor and stored in the variable volume dome arrangement. Stored biogas was burned using a biogas burner on daily basis.

The necessary information of the membrane modules used for the study is given in Table 2 as membrane data sheet.

A skid mounted membrane filtration system was consists of membranes, pumps, flow meters, and pressure gauges. Tubular ultrafiltration membrane modules supplied by Berghof Germany were used in this study. Two Berghof membrane modules were used in series configuration in the filtration unit. A half HP centrifugal pump as shown in Figure 1 was used to push the bioreactor content to membranes. A 3 mm pre-filter was used for the filtration of reactor content before entering it in to membrane modules. Another four HP centrifugal pump was used for the recirculation of the treated water through the membrane modules. Rotameters fabricated in acrylic material were used for the measurement of membrane recirculation flow and the permeate flow. The trans-membrane pressure (TMP) developed across the membrane was measured using pressure gauges installed on feed, reject, and permeate line of the membrane unit. A 100-l working volume tank made up of stainless steel was provided for preparation and storage of chemical solution required for chemical cleaning of the membranes.

Synthetic wastewater representing low to medium strength domestic wastewater was used as feed for the AnMBR. The targeted concentration of chemical oxygen demand (COD) in the synthetic wastewater was 500 mg/l and consists of dextrose monohydrate as a source of organic carbon. Urea and di-ammonium phosphate were added as a source of nitrogen and phosphorus, respectively, along with other salts as micronutrients for the healthy growth of microbial population. The constituents of synthetic wastewater and their concentrations are given in Table 3. The average characteristics of the synthetic wastewater are shown in Table 4.

The bench-scale external AnMBR system was in operation for different experiment but treating similar wastewater before starting this study. The external AnMBR was allowed to stabilize for new organic load before starting the performance evaluation study. Before starting the study, membranes were chemically washed as per standard chemical cleaning protocols suggested by supplier and mentioned elsewhere (Bornare et al. 2014).

Fresh feed was prepared in the feed tank on daily basis. After stabilization of the biological performance, HRT was set at 8 h and maintained constant for throughout the study. Depending upon the mixed liquor volatile suspended solids (MLVSS) in the bioreactor, calculated quantity of the sludge was withdrawn from the bioreactor on daily basis to keep the MLVSS at about 8,000 mg/l and mixed liquor suspended solids (MLSS) at about 12,000 mg/l. The bioreactor temperature was constant and maintained at 37 °C for complete study. The membrane filtration operation was started with the initial flux of 96 l/m² h (LMH) at 0.75 bar TMP. To keep a constant HRT in the bioreactor, the permeate flow rate from the external membrane was always set a little higher than the feed flow rate with recycling the excess permeate because it was practically difficult to precisely keep a constant flux. On sixtieth day of AnMBR operation, the membrane unit was disconnected from the bioreactor for 10 h and chemical cleaning of the membranes was carried as per cleaning protocols mentioned elsewhere (Bornare et al. 2014).

Feed and permeate samples were collected on daily basis for the measurement of pH, COD, and TOC. Feed and permeate sample were collected periodically for the measurement of total suspended solids (TSS). The bioreactor content was analyzed periodically for MLSS and MLVSS. COD, MLSS, and MLVSS were measured according to standard methods for the examination of water and wastewater (APHA 2005). TOC analysis was carried out with Shimadzu make total organic carbon (TOC) analyzer (model: TOC VCPH). Monitoring of TMP was continuous which indicate the extent of membrane fouling. The biogas generated in the bioreactor was measured and analyzed on daily basis. The analysis of the generated biogas for methane composition was carried out using a Thermo fisher scientific make gas chromatography (Chemito, Ceres 800 plus) fitted with a TG-WAXMS column (30 m × 0.25 mm × 0.25 µm) and a flame ionization detector. The nitrogen gas was used as a carrier gas, while the temperatures of the oven, injector port, and the detector were 50, 100, and 100, respectively. Biogas generation was measured through rise in the height of the floating dome used for gas storage. TMP was measured using the pressure gauges fixed on feed, reject and permeate line of the membrane unit.

Even though the targeted COD concentration in the feed was 500 mg/l, the degradation of organics was observed in the feed tank. To minimize the observed natural degradation of feed solution, the feed tank cleaning was carried out on daily basis. The fresh feed solution was prepared on every day during the performance evaluation period of AnMBR. The approximate MLSS of about 12,000 mg/l and MLVSS of about 8,000 mg/l were maintained in the bioreactor for throughout the period of performance evaluation. The feed wastewater organic strength was represented by an average COD and TOC concentration of 365 and 135 mg/l, respectively.

The feed and permeate samples were collected on daily basis for the measurement of COD. The evolution of concentrations of COD in the feed and permeate is shown in Figure 3. The concentrations of COD in the feed and permeate were not constant on every day and found varying around a mean value during investigation.

The feed COD was varied from 300 to 462 mg/l during performance evaluation. The residual concentration of COD in permeate was varied from 12 to 31 mg/l during the investigation. The average concentration of COD in permeate was 19 mg/l during the investigation period of 72 days. The relatively lesser residual concentrations of COD in permeate indicate the effectiveness of AnMBR for wastewater treatment compared to other conventional systems. Hu & Stuckey (2006) reported an average COD concentration in the effluent below 45 mg/l during the treatment of dilute synthetic wastewater at 6 h HRT in submerged AnMBR.

Actually dextrose based synthetic wastewater can be considered as easily biodegradable effluent than the actual low to medium strength domestic wastewater. However, as indicated in Tables 1 and 4, the TSS load of the typical low to medium strength domestic wastewater is considerably higher than the synthetic wastewater. It has indirectly revealed the lesser soluble COD load of low to medium strength domestic wastewater than synthetic wastewater. In other words, the synthetic wastewater used in this study has got higher concentration of easily biodegradable organics and the actual domestic wastewater has got lower concentration of relatively tougher compounds. Considering similar total COD load in both types of wastewater and AnMBR performance for the treatment of synthetic wastewater, AnMBR has got potential to be used for the treatment of actual domestic wastewater. However further studies are required for performance validation.

The obtained COD removal efficiency of AnMBR is shown in Figure 4 and found very stable during complete period of performance evaluation. The COD removal efficiency was varied from 92.33 to 96.61%. The average COD removal efficiency of AnMBR was 94.91% and can be comparable with reported literature. Jaeho & Shihwu (2010) found more than 95% COD removal efficiency for the treatment of synthetic municipal wastewater in AnMBR at 25 °C, whereas Saddoud et al. (2007) reported 90% COD removal efficiency for the treatment of municipal wastewater in AnMBR at 37 °C. The permeate samples were periodically tested for TSS; however, TSS content for the permeate sample was practically found nil for all samples collected during the study. Membrane in AnMBR is an absolute barrier for suspended solids and used ultrafiltration membrane could be one of the reasons for achieving 100% TSS removal in present study. One-hundred percent TSS removal in AnMBR study was also reported by other researchers (Saddoud et al. 2007; Torres et al. 2011).

TOC removal performance of AnMBR is shown in Figure 5. Feed TOC was varied from 102 to 165 mg/l. The average concentrations of TOC in the feed and permeate were 135 and 7 mg/l, respectively. The obtained TOC removal efficiency of AnMBR is shown in Figure 6. The TOC removal efficiency of AnMBR was varied from 91.03 to 96.30% during the study. The average of the achieved TOC removal efficiency in AnMBR was 94.64%. TOC being the parameter for measuring organic content of the sample, the TOC removal performance of AnMBR was matched with the obtained COD removal performance.

After stabilization of biological performance, the biogas generation from the AnMBR was measured on daily basis and the steady state data obtained is indicated in Figure 7. Biogas generation varied from 211 to 296 l/day with average generation of 248 l/day. As indicated in Figure 7, the observed biogas generation was very stable during the complete period of performance evaluation.

Constant process parameters and fixed operating condition could be few reasons for stable biogas generation during the period of investigation. The average biogas yield was 0.44 m³/kg COD removed. The obtained biogas yield was relatively higher than the reported values of conventional anaerobic treatment for sewage. The biogas generation and yield obtained were comparable with the reported literature by Hu & Stuckey (2006). The generated biogas was analyzed for methane composition on periodic basis and found varying from 65 to 82%.

Initially, the membrane filtration operation was started with 96 LMH flux. Since the decline in filtration flux represent the extent of membrane fouling, the decline of the membrane filtration flux during continuous operation of AnMBR was monitored closely and recorded on daily basis. The initial TMP was 0.75 bar set across the membrane. Figure 8 shows the permeate flux evolution during performance evaluation phase of AnMBR. The membrane flux decline was relatively higher in the initial period of 1 week of operation and showed reduction in flux from 96 to 85 LMH. The membrane flux reduction was relatively lower in the next period of 7 weeks of operation. The membrane flux was reduced from 85 to 72 LMH during this period and can be considered as stable flux operation for membrane filtration. The overall flux decline is the clear indication of membrane fouling and therefore the filtration operation was discontinued at the end of fifty-ninth day for chemical cleaning of the membranes.

During first 59 days cycle of membrane filtration, the flux was declined from 96 to 72 LMH resulting into an average flux of 81 LMH. The complete flux recovery was observed after chemical cleaning of the membrane and the average flux of 83 LMH was maintained for complete study.

As indicated in Figure 9, the TMP was increased from 0.75 to 0.97 bar in 59 days of operation before chemical cleaning of the membrane. The rise rate for TMP was relatively lower in the initial 50 days of operation; however, it was higher in last week of operation. Probably after this stage if the filtration operation would have been continued, the membrane might have shown some unexpected range of flux and TMP. The rise in TMP across the membrane was the indication of membrane fouling. The membranes were disconnected from the piping for physical check-up after water flushing during chemical cleaning of the membrane. Actually little deposition of some fibrous material was noticed at the entry point of first membrane. This could be one of the reasons for continuous built up of TMP during operation. Built up of sludge cake from inner side of the membrane may lead to rise in TMP during operation.

The membrane flux of 40 to 60 LMH is very common for external cross flow operation of anaerobic MBR. The obtained flux was found relatively higher than the reported literature. The new set of membrane modules with mesophilic operation of the AnMBR could be the possible reasons for obtained higher flux. The TMP for the operation was well below the allowable limit of the membrane. Figures 8 and 9 indicate that the membrane cleaning is necessary within 2 months operation to keep the membrane flux and TMP within acceptable range. As per literature information, timely chemical cleaning of the membrane not only reduces the operating cost of the system, but also increases the life of the membrane.

The present study demonstrated that the AnMBR can be an efficient treatment technology for low to medium strength domestic wastewater and to obtain high quality treated effluent. AnMBR system can be considered as important potential technology to recover energy in terms of biogas from the treatment of municipal sewage. The TSS load of the actual domestic wastewater is higher than the load available in synthetic wastewater used for present study. However soluble COD load of the used synthetic wastewater is higher than the actual low to medium strength domestic wastewater. Considering the performance of AnMBR for the treatment of synthetic wastewater, it has got potential to be used for the treatment of actual domestic wastewater. The overall degradability pattern of synthetic wastewater and actual domestic wastewater is expected to be comparable considering the basis of similar total COD however further studies are required for performance validation. The findings of the present study for organic removal efficiency and biogas yield are expected to remain same in field testing of AnMBR for the treatment of actual low to medium strength domestic wastewater. Since, membrane fouling pattern of the AnMBR generally depends upon the concentration and properties of TSS available in the feed, the observed membrane fouling pattern of AnMBR in present study may not replicate for the treatment of different type of wastewater.

Average COD and TOC concentrations of 365 mg/l and 135 mg/l, respectively, for feed and 19 mg/l and 7 mg/l, respectively, for permeate were obtained at 8 h HRT in AnMBR. After stabilization of AnMBR at 37 °C, the average TSS, COD, and TOC removals were 100%, 95%, and 95%, respectively. The average biogas yield was 0.44 m³/kg COD removed during the performance evaluation study of AnMBR. The average flux of 83 LMH was maintained for less than 1 bar TMP developed across the membrane during complete study. Thus, AnMBR can be effectively used for the treatment of low to medium strength domestic wastewater. Recyclable quality treated effluent having COD concentration of less than 20 mg/l can be used for agricultural irrigation to reduce the fresh water demand.

The study was part of the project on ‘Development of Anaerobic Membrane Bioreactor for Waste to Energy Solutions', partially funded by Department of Biotechnology, Govt. of India under the BIPP scheme. The authors sincerely thank the Department of Biotechnology for their support.