Effect of hydraulic retention time and sludge recirculation on greenhouse gas emission and related microbial communities in two-stage membrane bioreactor treating solid waste leachate

Bioresource Technology xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Effect of hydraulic retention time and sludge recirculation on greenhouse gas emission and related microbial communities in two-stage membrane bioreactor treating solid waste leachate Nararatchporn Nuansawan, Jarungwit Boonnorat, Wilai Chiemchaisri, Chart Chiemchaisri ⇑ Department of Environmental Engineering, Faculty of Engineering, Kasetsart University, 50 Ngam Wong Wan Road, Chatuchak, Bangkok 10900, Thailand

h i g h l i g h t s
Long term (>500 days) of CH4 and N2O emissions from two-stage MBR was investigated.
More than 90% of CH4 emissions were contributed from first stage anaerobic reactor.
N2O emission were at the same level from anaerobic and aerobic reactors.
Effect of HRT and sludge recirculation was more pronounced for CH4 than N2O.
Microbial diversity and abundance were less when hydraulic loading was increased.

a r t i c l e

i n f o

Article history: Received 14 December 2015 Received in revised form 26 January 2016 Accepted 28 January 2016 Available online xxxx Keywords: Greenhouse gas Hydraulic loading Leachate Membrane bioreactor Microbial diversity

a b s t r a c t Methane (CH4) and nitrous oxide (N2O) emissions and responsible microorganisms during the treatment of municipal solid waste leachate in two-stage membrane bioreactor (MBR) was investigated. The MBR system, consisting of anaerobic and aerobic stages, were operated at hydraulic retention time (HRT) of 5 and 2.5 days in each reactor under the presence and absence of sludge recirculation. Organic and nitrogen removals were more than 80% under all operating conditions during which CH4 emission were found highest under no sludge recirculation condition at HRT of 5 days. An increase in hydraulic loading resulted in a reduction in CH4 emission from anaerobic reactor but an increase from the aerobic reactor. N2O emission rates were found relatively constant from anaerobic and aerobic reactors under different operating conditions. Diversity of CH4 and N2O producing microorganisms were found decreasing when hydraulic loading rate to the reactors was increased. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Disposal of municipal solid wastes in sanitary landfills lead to the formation of landfill leachate, a highly polluted wastewater. Municipal solid waste landfill leachate generally contains high concentrations of dissolved organic matter and nutrients (Kjeldsen et al., 2002). In order to meet strict quality standards for direct discharge of leachate into natural water environment, integrated treatment methods, i.e. combination of chemical, physical and biological steps are required (Wisniowski et al., 2006). Membrane bioreactor (MBR) technology which combined biological wastewater treatment process and advanced physical separation has been developed as a promising technology for treating various types of wastewater including landfill leachate. The ⇑ Corresponding author. E-mail address: [email protected] (C. Chiemchaisri).

membrane separation in MBR allows complete retention of biomass and maintaining high biomass concentration, resulting in an efficient biological digestion system (Ahmed and Lan, 2012). The advantages of MBRs over conventional biological processes are well-known including improvement of effluent quality, good process stability, reduce reactor size by retaining higher biomass or mixed liquor suspended solids (MLSS) concentrations, and lower sludge production (Van Dijk and Roncken, 1997). In order to apply MBR under high organic loading of wastewater, a novel type MBR utilizing inclined-plate separator in first stage anoxic reactor followed by second stage aerobic submerged MBR has been developed (Xing et al., 2006). The unique feature of the system was its capacity to storage majority of biomass in the first stage reactor while maintaining relatively constant biomass concentration in the second reactor for membrane fouling control. Two-stage MBR has also been applied to the treatment of partially stabilized leachate yielding satisfactory results

http://dx.doi.org/10.1016/j.biortech.2016.01.109 0960-8524/Ó 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Nuansawan, N., et al. Effect of hydraulic retention time and sludge recirculation on greenhouse gas emission and related microbial communities in two-stage membrane bioreactor treating solid waste leachate. Bioresour. Technol. (2016), http://dx.doi.org/10.1016/j. biortech.2016.01.109

2

N. Nuansawan et al. / Bioresource Technology xxx (2016) xxx–xxx

(Chiemchaisri et al., 2011). Nevertheless, greenhouse gases (GHGs) can be significantly produced from biological activities during the treatment as methane (CH4) gas could be produced under anaerobic condition during the initial step of treatment (Yan et al., 2014) and recirculation of sludge between the anaerobic and aerobic reactors yield alternate oxygen absence and presence condition which is favorable for nitrous oxide (N2O) production (Aboobakar et al., 2013). Significant emission of CH4 could occur at the anaerobic zone of leachate treatment system under high organic loading (Chiemchaisri et al., 2009). Moreover, high nitrogenous characteristics of leachate could lead to significant N2O emissions soon after raw leachate was aerated (Lin et al., 2008). The production of CH4 and N2O during leachate treatment depends on aerobic and anaerobic conditions in the treatment system and it varied from one treatment technology to the others. There were only few studies conducted on N2O emissions during landfill leachate treatment even though high emissions are anticipated during the treatment of this nitrogen-rich leachate in the past years (Barton and Atwater, 2002). Only recent researches have been reported that several factors could influence the emission of CH4 and N2O during wastewater and leachate treatment. They include the configuration of wastewater treatment process (Yan et al., 2014) with oxygen profile as an important parameter (Aboobakar et al., 2013), wastewater characteristics (Wang et al., 2014; Liu et al., 2014). Nevertheless, the acquired knowledge is still far from complete understanding of GHG emission during wastewater treatment processes as they could be operated under wide range of environmental conditions. For instance, the N2O emission were greatly influenced by dynamic conditions and variation of influencing parameters including oxygen, nitrite and ammonium concentrations during nitrification and denitrification reactions (Kampschreur et al., 2008). Among the factors affecting CH4 and N2O emission during leachate treatment being reviewed, there was still very limited information on the effect of hydraulic retention time (HRT) and sludge recirculation on their emissions especially from the treatment systems operated under high biomass concentration such as MBR. Therefore, this study was carried out to investigate CH4 and N2O gas emission characteristics from the two-stage MBR system incorporating anaerobic and aerobic conditions during the treatment of leachate under different HRT and sludge recirculation during long term operation (>500 days) in order to determine their emission rate during steady operating condition and understand the effect of hydraulic condition on their emissions from the treatment system. Furthermore, detailed characterization of microbial community using molecular biology technique in the two-stage MBR operated under different conditions was carried out for better understanding of the dynamic of GHG producing microorganisms in the system.

2. Methodology 2.1. Experimental system A laboratory scale MBR unit consisting of anaerobic and aerobic reactors with 30 L working volume each was used. The schematic diagram of the experimental system is shown in Fig. 1. In the anaerobic reactor, an inclined tube module (2.5 cm opening) was installed or separating sludge from the mixed liquor so that sludge can be stored inside the system while allowing low mixed liquor suspended solids (MLSS) overflow into the aerobic tank. Anaerobic condition were kept in major part of the reactor especially beneath the inclined tube module. Meanwhile, periodical aeration was supplied at the surface of the reactor above the inclined tube module to maintain dissolved oxygen (DO) at about 0.5 mg/l for odor control purpose. In aerobic tank, a hollow fiber membrane module

(Mitsubishi Rayon, PVDF, 0.4 lm pore size, 0.077 m2 surface area) was used for solid–liquid separation. Intermittent suction (5 min on and 1 min off) was performed for withdrawal of filtrate water through the membrane module. The aeration was continuously supplied to the aerobic reactor by maintaining DO level at 3– 4 mg/l. HRT in each reactor was set at 5 days during 1st and 2nd experiment run and 2.5 days during 3rd and 4th experimental run. Consequently, average membrane permeate flux was 0.1 and 0.2 m3/m2 d at HRT of 5 and 2.5 days respectively. The study was carried out in four experimental runs. During the 1st run (day 1–275) and 3rd run (day 410–496), the system was operated without sludge re-circulation. During the 2nd run (day 276–409) and 4th run (day 497–575), the system was operated with sludge re-circulation from aerobic reactor back to anaerobic reactor at 100% of influent feed flow rate. Under each operating condition, the system was operated until steady condition in terms of water qualities and gas emission rate have been reached. 2.2. Leachate preparation and water quality analyses Raw leachate was obtained from solid waste collection trucks entering a solid waste disposal site in Thailand. The wastewater samples were kept inside glass containers and stored at a temperature of 4 °C. Prior to analysis, the waste water samples were filtered through the glass microfiber filter (GF/C). All leachate sample analysis was performed according to Standard Methods for the Examination of Water and Wastewater (APHA, 2005). The analytical parameters included pH, DO, BOD, COD, TOC, SS, NH3,  TKN, NO 2 and NO3 . In the reactors, MLSS concentrations were monitored. Chemical characteristics of leachate used in this study are shown in Table 1. The leachate used exhibited high organic concentrations in terms of BOD, COD and TOC and acidic in nature. Feeding leachate used in the laboratory scale MBR was prepared by mixing fresh leachate and tap water at about 1:1.5 v/v ratio to maintain BOD and COD concentrations in the feed leachate at about 20,000 and 40,000 mg/l to ensure steady operation and greenhouse gas production along the experimental period. 2.3. Determination of greenhouse gas emission GHG emission was determined on regular basis from the anaerobic and aerobic reactors during the MBR operation. For the anaerobic reactor, a closed-flux chamber was occasionally placed on top of anaerobic reactor to determine greenhouse gas emission from the system. Close flux chamber is made of acrylic plate with 250mm width, 300-mm in length and 100-mm in height. During the measurement, special care was taken to make sure that there are no gas leakage. The coverage area of the chamber was 0.075 m2. In order to determine the emission rate, gas samples from the closed-flux chamber were collected into a 9-ml vial by a gastight syringe at different time intervals (e.g. every 30 min) up to 120 min. Then, gas composition in a vial was analyzed by using a gas chromatograph (GC). For CH4 analysis, GC (Agilent 6890) with thermal conductivity (TCD) and Alltech-CRT column was used. For N2O analysis, GC (Shimadzu Clarus 580) with thermal conductivity (ECD) installed with Heyesep D column was used. Closed flux chamber operated by allowing upward diffusive gas to accumulate in the chamber. As, the area of flux chamber and reactor was equal, the increasing rate of gas in the chamber was used to determine the mass of emitting gas as follows.

F AN ¼ qV DC=Dt

ð1Þ

where FAN = mass of gas emitted from anaerobic reactor; q = density of gas; V = volume of chamber; DC/Dt = gas concentration gradient.

Please cite this article in press as: Nuansawan, N., et al. Effect of hydraulic retention time and sludge recirculation on greenhouse gas emission and related microbial communities in two-stage membrane bioreactor treating solid waste leachate. Bioresour. Technol. (2016), http://dx.doi.org/10.1016/j. biortech.2016.01.109

3

N. Nuansawan et al. / Bioresource Technology xxx (2016) xxx–xxx

Anaerobic reactor Leachate storage tank

Aerobic reactor

P

Treated water storage tank

Fig. 1. Schematic of two-stage MBR system.

Table 1 Chemical characteristics of feeding leachate. Parameters

pH BOD COD TOC SS NH+3-N TKN NO 2 -N NO 3 -N

Fresh leachate

Diluted leachate used in the experiment (average values)

Range

Avg.

Range (Run 1–4)

Run 1

Run 2

Run 3

Run 4

2.9–3.7 41,250–72,500 72,360–91,240 16,750–23,430 9340–12,970 2340–3720 2540–3890 0.5–1.1 0.5–1.0

3.43 60,420 86,460 18,940 10,890 2960 3390 0.8 0.9

4.0–4.9 20,100–22,800 37,800–45,970 7930–8930 7350–8940 2170–2690 2320–2820 0.4–0.6 2.4–3.5

4.5 21,150 42,180 8440 8330 2500 2610 0.48 2.84

4.6 20,800 41,160 8400 8510 2360 2550 0.49 2.64

4.6 21,800 40,380 8380 8670 2160 2550 0.43 2.59

4.2 22,100 40,860 8370 8150 2165 2540 0.46 2.63

Table 2 Effluent qualities and MLSS of anaerobic and aerobic reactors in two-stage MBR during steady operation. Parameters

pH BOD (mg/l) COD (mg/l) TOC (mg/l) NH+3-N (mg/l) TKN (mg/l) NO 2 -N (mg/l) NO 3 -N (mg/l) MLSS (g/l) MLSS change (mg/l d)

Run 1 (day 1–275) HRT 5 day, w/o recirculation Anaerobic

Aerobic

7.4(0.3) 12,945(512) 32,669(2163) 5468(165) 1864(86) 1972(82) 0.30(0.07) 0.63(0.22) 12.55(1.44) +9.1

9.4(0.2) 190(38) 3300(483) 400(88) 330(44.1) 349(33) 0.03(0.02) 0.16(0.04) 20.45(1.91) +0.4

Run 2 (day 276–409) HRT 5 day, with recirculation %Removal 99.1 92.2 95.3 87.3 86.4 92.7 94.3

Run 3 (day 410–496) HRT 2.5 day w/o recirculation

pH BOD (mg/l) COD (mg/l) TOC (mg/l) NH+3-N (mg/l) TKN (mg/l) NO 2 -N (mg/l) NO 3 -N (mg/l) MLSS (g/l) MLSS change (mg/l d)

Anaerobic

Aerobic

6.3(0.2) 16,922(1322) 35,760(2110) 5760(336) 1884(104) 2103(125) 0.30(0.07) 0.86(0.29) 11.90(0.53) +12.4

9.1(0.4) 1730(517) 7654(526) 1480(88) 414(36) 465(49) 0.04(0.02) 0.27(0.08) 21.41(0.06) +3.8

%Removal 92.1 81.0 82.3 80.8 81.7 91.6 89.4

Anaerobic

Aerobic

7.8(0.2) 13,295(1370) 31,672(1512) 5248(136) 1807(72) 2021(68) 0.35(0.07) 0.52(0.08) 12.73(0.72) +8.9

9.3(0.1) 214(25) 3900(626) 526(92) 304(69) 360(40) 0.04(0.01) 0.23(0.05) 19.67(1.19) +0.1

%Removal 99.0 90.4 92.7 87.1 85.8 92.1 91.4

Run 4 (day 497–575) HRT 2.5 day with recirculation Anaerobic

Aerobic

7.0(0.4) 15,791(1004) 36,080(2263) 5660(226) 1944(107) 2162(70) 0.28(0.09) 0.72(0.19) 12.68(0.58) +22.9

8.9(0.3) 2260(342) 8735(730) 1815(105) 489(84) 552(79) 0.05(0.01) 0.3(0.07) 20.70(1.03) 6.3

%Removal 89.7 78.6 78.3 77.4 78.3 89.2 88.5

Note: the numbers show average (SD) values.

The gas emission was measured from anaerobic reactor at different times along the operation period. For the determination of gas emission from aerobic reactor, gas samples were collected from the cover chamber equipped with gas

outlet port. The size of cover chamber were identical to that used in anaerobic reactor. The gas emission was determined from supplied air flow rate and measured outflow gas concentration using the following equation.

Please cite this article in press as: Nuansawan, N., et al. Effect of hydraulic retention time and sludge recirculation on greenhouse gas emission and related microbial communities in two-stage membrane bioreactor treating solid waste leachate. Bioresour. Technol. (2016), http://dx.doi.org/10.1016/j. biortech.2016.01.109

4

F AE ¼ Q air C=A

N. Nuansawan et al. / Bioresource Technology xxx (2016) xxx–xxx

ð2Þ

where FAE = flux of gas emitted from aerobic reactor, Qair = supplied air flow rate, C = outflow gas concentration and A = area of the cover chamber. 2.4. Microbial community analyses PCR-DGGE technique was used for the analysis of microbial community in the two-stage MBR operated under different conditions. Samples are collected from anaerobic and aerobic reactors at day 275 (HRT 5 days w/o sludge recirculation) day 350 (HRT 5 days with sludge recirculation) day 470 (HRT 2.5 days w/o sludge recirculation) and day 560 (HRT 2.5 days with sludge recirculation) and then refrigerated at 80 °C until analysis. The DNAs of bacteria are extracted by DNA extraction kit (Invitrogen, USA). This PCR was performed using 338GC-F and 518R (Invitrogen, USA) for most bacteria 344GC-F and Univ522R (Invitrogen, USA) for methanogen as described in Khemkhao et al. (2012) as universal bacteria primer for bacteria community screening in samples. The combination of these primer had generated a PCR fragment about 180–190 bp which was suitable for subsequent DGGE analysis. The PCR amplification of genes in this study was performed according to the following condition; 50 ll of PCR reaction containing 1 ll of genomic DNA template, 1 PCR buffer, 3 mM MgCl2, 200 lM deoxynucleoside triphosphates (dNTPs), 10 pmol of each primer, and 1 U of Taq DNA polymerase (Qiagen, Germany). The PCR was conducted in a Perkin–Elmer GeneAmp PCR System 9700 (Applied Biosystem, USA). The thermal program was set as followed; initial denaturation at 95 °C for 5 min, followed by 30 cycles of denaturation at 95 °C for 50 s, annealing at 55 °C for 30 s and extension at 72 °C for 40 s. Finally, the extension step was performed at 72 °C for 7 min. The size and amount of PCR products were visualized on a 1.2% (w/v) agarose gel electrophoresis staining with ethidium bromide and estimated for DNA quantity using a NanoDrop 1000 spectrophotometer (Thermoscientific, USA). DGGE analyses of 16s DNA amplified PCR products were performed using the DGGE-2000 system apparatus (CBS Scientific Company, Del Mar, CA) with different ranges of denaturants for PCR products amplification. The polyacrylamide gel concentration of samples was applied on 9% (w/v) polyacrylamide gel in range of 30–70% denaturing gel gradient. The eighty percent of denaturant is corresponded to 5.6 M of urea and 32% (v/v) formamide in 1 TAE. Electrophoresis of both DGGE analysis was performed for 16 h at a constant voltage of 80 V. The temperature was set at a constant temperature of 60 °C. The polyacrylamide gel was stained with SYBR Gold nucleic acid gel stain (Invitrogen, USA) for 15 min before visualized on a UV transillumination and photographs were captured using Biovision CN 1000/26 M (Vilber Lourmat, France). DGGE profiles were observed from obtained band light intensities at different locations. The selected bands were excised from the polyacrylamide gel and re-suspended in 20 ll of MilliQ water and stored overnight at 4 °C. The eluted DNA were used as the templates for re-amplification. The DNA sequences were manually identified using the online BLAST software on the NCBI website (http://www.ncbi.nlm.nih.gov/) and aligned with the GenBank reference. 2.5. Real-time quantitative PCR analysis The quantification methodologies described in Limpiyakorn et al. (2011) were used. For each sample, quantitative real-time PCR was performed with duplicate sets of extracted DNA. For each dilution, quantitative real-time PCR was carried out in duplicate with a Brilliant II SYBR Green QPCR Master Mix (Stratagene, USA) in an Mx3005P instrument (Stratagene, USA). The quantification

of methanogenic bacteria was performed using the primers 344GC-F and Univ522R (Invitrogen, USA). Nitrifying and denitrifying bacteria was detected using the primer set of 967F, 1046R and nirS1F, nirS6R in a cycler (BioRad, Hercules, CA, USA). The standard DNA used was the pGEM-T Easy Vector (Promega, USA) containing the 16S rRNA gene fragment of methanogenic, nitrifying and denitrifying bacteria. To confirm the single target fragment of the PCR amplified products, dissociation curves were analyzed and plotted at the end of every quantitative real-time PCR reaction. To verify the correct amplification of the target microorganisms’ DNA, few clones from the clone libraries constructed from the real-time PCR amplified products were randomly selected for sequencing and the results for every reaction tested verified the correct amplification of the target microorganisms’ DNA. The cell numbers of bacteria were calculated from the quantified numbers of genes on the basis of the numbers gene copies found in isolated bacteria. Per genome, bacteria possessed one copy of bacteria gene.

3. Results and discussion 3.1. Treatment performance of two-stage MBR Table 2 shows the effluent qualities and biomass concentrations in two-stage MBR. During the 1st run, the MBR was operated at HRT of 5 days without sludge recirculation. At steady-state operation, average BOD and COD removals in the system was found to be 99.1% and 92.2% among which 39% and 23% were found removed in the first stage anaerobic reactor. Majority of organic carbon remaining in the effluent were recalcitrant compounds as suggested by low BOD/COD ratio (0.06) as compared to that of the feeding leachate (0.5). Simultaneously, total NH3 and TKN removals were 86–87% in which about 25% were removed in anaerobic reactor. In the anaerobic reactor, some percentages of NH3 and TKN removals could take place from microbial uptake for biomass assimilation in the reactor and simultaneous nitrification and denitrification reactions at the upper part above the inclined tube module where DO concentration was maintained at about 0.5 mg/l. Subsequent treatment was mainly taken place in the aerobic reactor where majority of organic carbon and nitrogen were removed. Meanwhile, the effluent from aerobic reactor (2nd stage) contained negligible concentration of oxidized nitrogen  (NO 2 and NO3 ) suggesting that majority of incoming NH3 and TKN were removed via simultaneous nitrification and denitrification in the aerobic reactor. High biomass (MLSS) concentration (15 to 20 g/L) maintained in the aerobic reactor helped promoting anoxic condition in microbial flocs where denitrification reactor could be promoted even though bulk DO concentration was maintained at high level of 3–4 mg/l. In the aerobic reactor, the nitrification took place at the outer part of the bio-particles where oxygen concentrations was maintained by the diffusion from bulk solution. Meanwhile, the denitrification reaction took place at the inner part of the bio-particles after the oxygen has been consumed up and nitrate was only presented as electron acceptor (Chiemchaisri and Yamamoto, 2005). Recent research also indicated a possibility of having denitrification reaction under aerobic condition and it was also associated with low N2O emission (Zheng et al., 2014). When sludge recirculation was introduced in the 2nd run, BOD removal were maintained at the same level (99%) whereas slightly lower COD and TKN removal efficiencies of 90.4% and 85.8% were observed. Slightly lower COD removal efficiencies observed during introduction of sludge recirculation could be due to an increase in hydraulic loading which adversely affected the retention of recalcitrant organic compounds, particularly humic-like substances, within the MBR (Sanguanpak et al., 2013). As the HRT was shortened from 5 to 2.5 days in the 3rd run, BOD

Please cite this article in press as: Nuansawan, N., et al. Effect of hydraulic retention time and sludge recirculation on greenhouse gas emission and related microbial communities in two-stage membrane bioreactor treating solid waste leachate. Bioresour. Technol. (2016), http://dx.doi.org/10.1016/j. biortech.2016.01.109

5

N. Nuansawan et al. / Bioresource Technology xxx (2016) xxx–xxx

COD and TKN removal were slightly decreased to 92.1%, 81.0% and 81.7%. Then, they were slightly decreased to 89.7%, 78.6% and 78.3% when sludge recirculation was introduced in the 4th run. At these lowered HRT conditions, much higher biodegradable organic (BOD) concentrations (9–10-fold increase) were detected in the effluent of aerobic reactor even though the effluent from first stage anaerobic reactor were only moderately elevated (20–30% increase). The results suggested that sufficient HRT would be required in second stage reactor for polishing the remaining BOD to a lower value. These results reveal that organic and nitrogen removal efficiencies decreased when hydraulic loading to the reactor increased due to the changes in HRT and sludge re-circulation of the reactor operation. Nevertheless, the deterioration in removal efficiencies from the introduction of sludge re-circulation were not significant but they were greater affected when the HRT has been shortened from 5 to 2.5 days, possible contribution from simultaneous increase in organic loading. As the hydraulic loading to the reactor increased, it is expected that the overflow rate of hardly biodegradable matter from the anaerobic reactor to the aerobic reactor as the hydraulic flow to the reactors was increased, yielding higher penetration of dissolved organic substances through the microfiltration membrane submerged in the aerobic reactor. Nevertheless, an increase in hydraulic loading coupling with organic loading increase yielded further reduction in organic and nitrogen removal efficiencies especially in the aerobic stage where majority of BOD, COD and TKN were removed. In most runs, it was also noticed that the nitrified nitrogen were mostly denitrified leaving majority of nitrogen in TKN form in the aerobic reactor. It is also revealed that the treated water qualities are still containing considerable amount of recalcitrant organic compounds and nitrogen therefore further polishing treatment of the effluent from the system would be required for discharging to natural water environment or reuse. During the operation, biomass concentrations in the anaerobic and aerobic reactors were regularly monitored. During the 1st run (day 1–275), MLSS concentrations in the anaerobic reactor below the inclined tube module was increasing from about 11 – 13.5 g/l at an increasing rate of 9.1 mg/l d. At steady operation, average MLSS concentration was kept at 12.55 g/l. Meanwhile, the MLSS in aerobic reactor gradually built up from 15 g/l and became stable at average value of 20.45 g/l. The MLVSS/MLSS ratio of biomass was found to be relatively constant at 0.7–0.8. When sludge recirculation was performed during the 2nd run, MLSS in the aerobic reactor could maintained relatively constant at 19.67 g/l as the excess sludge was recirculated back and kept in the first reactor yielding MLSS increasing rate of 8.9 mg/l d in the anaerobic reactor. During the operation 3rd and 4th run, MLSS in the aerobic reactor was maintained at about 20 g/l. An increase in the hydraulic loading had a short-term adverse impact to the

biomass in the anaerobic reactor during which a reduction in biomass concentration was observed. Nevertheless, the biomass concentration could be restored and net increasing rates of 12.4 and 22.9 mg/l d were observed during the operation without and with recirculation respectively. In two-stage MBR, the biomass concentration in the system was controlled by recirculation of excess biomass back to anaerobic reactor and the accumulation of biomass was allowed at the bottom part of the reactor beneath the inclined tube module thus allowing the system to operate without excess sludge wastage (Chiemchaisri et al., 2011). 3.2. GHG emissions from two-stage MBR Table 3 presents the emissions of CH4 and N2O from anaerobic and aerobic reactors during steady operation of two-stage MBR. During the 1st run (HRT 5 days without recirculation), average surface CH4 emission rates from anaerobic and aerobic reactor surface were 74.4 and 0.73 g/m2 d. This is equivalent to emitted CH4 mass of 5.58 and 0.55 g/d. Meanwhile, N2O emission rates were found at lower level of 0.04 and 0.03 g/d respectively. These results show that both GHGs were mainly emitted from the anaerobic reactor. Similar observation on the CH4 emission trend along the treatment process was reported in Wang et al. (2014). The major source of CH4 emission came from the first reactor where oxidation reduction potential (ORP) below 300 mV was favorable for methanogenesis. Nevertheless, N2O emission profile along anaerobic and aerobic reactors in this study were found different from those reported in previous studies. In this study, comparable N2O emission was found from anaerobic and aerobic reactors even though they were at relatively low range. Previous studies have reported much higher N2O emission from aerobic reactor mainly due to air stripping mechanism (Yan et al., 2014; Wang et al., 2014). The difference in emission profile could be due to unique configuration of anaerobic reactor in which inclined tube separator module is installed. The presence of inclined tube separator resulted in separation of reactor into biomass storage zone beneath the separator where highly concentrated biomass were maintained under anaerobic condition. Meanwhile, N2O production could take place at low solid zone above the separator where DO was maintained at about 0.5 mg/l. In previous research, it was reported that high N2O production was observed under a DO level less than 2 mg/l (Kampschreur et al., 2008) as N2O was produced from denitrification instead of N2 in low oxygen condition (Itokawa et al., 2001). When sludge recirculation was introduced during the 2nd run, CH4 and N2O emission rates in the anaerobic reactor were found decreasing by 17% and 20%. At the same period, the emission of CH4 in aerobic reactor was slightly increased whereas N2O emission was remained relatively constant in the aerobic reactor. An increase in CH4 emission from the aerobic reactor could be

Table 3 Methane and nitrous oxide emissions from anaerobic and aerobic stages of two-stage MBR under different conditions. HRT (d)

Re-circulation

5

Without With Without With

2.5

Anaerobic

Aerobic

Cloaded, g/d

CH4emission, g/d

%C as CH4 gas

Cloaded, g/d

CH4emission, g/d

%C as CH4 gas

50.64 (2.2) 50.43 (14.6) 100.59 (1.3) 100.43 (1.8)

5.58 4.66 4.09 3.85

8.26 6.93 3.05 2.88

32.80 31.48 69.78 68.26

0.55 0.68 0.30 0.37

1.26 1.62 0.32 0.41

(0.5) (0.4) (0.3) (0.4)

Anaerobic

5 2.5

Without With Without With

(1.0) (9.2) (3.8) (2.6)

(0.12) (0.22) (0.75) (0.06)

Aerobic

Nloaded, g/d

N2Oemission, g/d

%N as N2O gas

Nloaded, g/d

N2Oemission, g/d

%N as N2O gas

16.25 15.40 30.30 31.52

0.04 0.03 0.02 0.03

0.16 0.12 0.04 0.06

10.85 10.76 23.98 22.90

0.03 0.03 0.03 0.03

0.18 0.18 0.08 0.08

(0.7) (1.1) (1.1) (1.2)

(0.007) (0.008) (0.003) (0.003)

(1.1) (24.0) (1.97) (3.2)

(0.006) (0.002) (0.002) (0.004)

Remark: the numbers show average (SD) values.

Please cite this article in press as: Nuansawan, N., et al. Effect of hydraulic retention time and sludge recirculation on greenhouse gas emission and related microbial communities in two-stage membrane bioreactor treating solid waste leachate. Bioresour. Technol. (2016), http://dx.doi.org/10.1016/j. biortech.2016.01.109

6

N. Nuansawan et al. / Bioresource Technology xxx (2016) xxx–xxx

Table 4 Microbial community in two-stage MBR detected by DGGE using 338F/512R primers. Anaerobic W/o recirculation

Aerobic With recirculation

W/o recirculation

Closet relative (similarity more than 90%) With recirculation

+,/ + + + /

+,/

+,/ +,/ + +,/ +

/ +,/ +,/ +

+ +,/

+ +,/ + + +,/ +,/ +,/

+,/ +,/ +,/

+ +,/

+ +,/ +,/

+,/ + + + +,/ +,/ +,/ + +,/ + + +,/ +,/ +,/ + / + +,/ +,/ +,/ +,/ +,/ + + +,/ / +,/ +

/ +,/ + + +,/

+ +,/ + +

+ +,/

/

/ +,/

+

/ +,/

+,/ +,/

+,/ + +

+ +,/ +

+,/

/ +,/

+,/ +,/ + + / + +,/ + +,/ +,/ +,/ +,/ + +,/ + +,/

+,/ /

+

/ +,/ +,/

+,/ + +,/ +,/ +,/ +,/ +

+,/ + + / / +,/ +,/

+,/ +,/ +,/

Bacillus subtilis No.66 Bacillus sp. NLA3E Candidatus Liberibacter asiaticus str. Spy62 Candidatus Liberibacter asiaticus str. Spy62 Paracoccus denitrificans Pseudomonas Achromobacter Nitrobacter minogradskyi Nb-255 Thiobacillus denitrificans ATCC25259 Nitrosomonas europaea ATCC19718 Nitrosomonas sp. AL212 Nitrosomonas sp. AL212 Nitrosospira multiformis ATCC25196 Nitrococcus mobilis Nb-231 Pseudomonas putida KT2440 Pseudomonas sp. UW4 Pseudomonas fluorescens Pf0-1 Pseudomonas fluorescens Pf0-1 Nitrosomonas sp. ls79A3 Nitrosomonas eutropha C91 Nitrosospira Nitrobacter sp. Nitrosolobus Nitrosococcus Micrococcus Luteus ATCC4698 Candidatus Nitrospira defluvii Methanococcus aeolicus Methanococcus vannielii Methanococcus burtonii Methanosaeta concilii Methanosaeta soehngenii Methanosaeta harundinacea Methanosaeta thermophile Methanobolus oregonesis Methanobolus psychrophilus Methanosalsum zhilinae Methanosarcina acetivorans Methanosarcina mazii Methanosarcina barkeri Methanosarcina sp. Methanosarcina frista Methanobactorium sp. Methanococciods methylutens Methanoculleus marisnigri Methanoculleus bourgensis Methanogenium cariaci Methanoregula boonei Methanobrevibacter gottschalkii Methanomicrobium mobile Methanomicrobium mobile Methanofollis liminatans Methanosphaerula palustris

Note: + mean their presence at HRT of 5 days, / mean their presence at HRT of 2.5 days.

partially due to increased overflow of anaerobic reactor effluent containing high dissolved CH4 concentration to the aerobic reactor where it was stripped off by aeration. Nevertheless, total combined CH4 emission from both reactors were decreased by 13% through introduction of sludge recirculation. Furthermore, these changes in CH4 and N2O emissions were found relatively at higher degree when compared between to the differences in organic and nitrogen removals between the experimental runs and thus suggesting that an introduction of sludge recirculation of mixed liquor could create less favor condition for CH4 and N2O production in the system. As the HRT was shortened to 2.5 days, CH4 emission was reduced by 27% and 31% during the operation without sludge recirculation (3rd run) and with sludge recirculation (4th run) respectively. Meanwhile, N2O emissions from anaerobic and aerobic reactors was kept relatively constant. Doubling of hydraulic loading by shortening HRT from 5 days (1st run) to 2.5 days (3rd run)

and introduction of sludge recirculation at 100% at HRT of 5 days (2nd run) had dissimilar effect on CH4 emission. Higher reduction in CH4 emission was observed when the organic loading to the system was increased together with hydraulic loading. Even though the mass of organic carbon removed in the system at shorter HRT (2.5 days) was observed, the mass of CH4 emitted from the system operated at HRT of 2.5 days were found lower than those observed at HRT of 5 days. At steady condition, carbon and nitrogen loadings to the system were determined as 50.64 and 16.25 g/d at HRT of 5 days (1st run). During the treatment, fugitive CH4 emission accounted for 8.26% of C loaded to the anaerobic reactor and 1.26% of C loaded to the aerobic reactor respectively. The contribution of CH4 emission from the anaerobic reactor was 91%. Meanwhile, N2O gaseous emission accounted for only 0.16% and 0.18% of N loaded to the anaerobic and aerobic reactors. Frison et al. (2015) reported that low N2O

Please cite this article in press as: Nuansawan, N., et al. Effect of hydraulic retention time and sludge recirculation on greenhouse gas emission and related microbial communities in two-stage membrane bioreactor treating solid waste leachate. Bioresour. Technol. (2016), http://dx.doi.org/10.1016/j. biortech.2016.01.109

N. Nuansawan et al. / Bioresource Technology xxx (2016) xxx–xxx

emission (0.24% of influent nitrogen load) could be achieved during biological nitrogen removal from anaerobic effluents through the operation at sufficient DO level and lower volumetric nitrogen loading than nitrogen removal capacity of the system. Meanwhile, its production on the occurrence of process disturbances such as period of no aeration or nitrification instability (RodriguezCaballero et al., 2014). Introduction of sludge recirculation decreased CH4 emission to 6.93% in anaerobic reactor but increased the emission from aerobic reactor to 1.62% of C loaded while only slightly decreased N2O emission from the anaerobic reactor. When the HRT was reduced to 2.5 days, the percentages of C as CH4 and N as N2O gas were significantly decreased. It can be concluded from this finding that the quiescent hydraulic condition under extended HRT without sludge recirculation could promote CH4 emission especially from the anaerobic reactor while they had little impact on N2O emission from both reactors. 3.3. Microbial communities under different conditions PCR-DGGE technique was used for bacteria analyses to describe microbial community related to GHG production. The profiles of microorganisms analyzed by 338F/518R and 344F/522R primers are summarized in Table 4. The number of identified species for methanogenic bacteria, nitrifying bacteria and denitrifying bacteria were compared between anaerobic and aerobic reactors and during the reactor operation at HRT of 5 and 2.5 days with and without recirculation. Among all experimental conditions, the anaerobic reactor operated at HRT of 5 days without sludge recirculation had highest number of methanogenic bacteria detected (24 species). Meanwhile, 13 species were detected in the aerobic reactor. When sludge recirculation was introduced, the number of methanogenic species were found to be 18 and 16 respectively. Some species (Methanococcus aeolicus, Methanosaeta concilii, Methanobolus oregonesis, Methanobrevibacter gottschalkii) were found disappeared from the system. When the HRT was shortened to 2.5 days, methanogenic species were all found less than that under HRT of 5 days. They were 16 species in the anaerobic reactor and 12 species in the aerobic reactor under no sludge recirculation and decreased to 14 and 13 species when sludge recirculation was introduced. Jang et al. (2015) has reported that microbial diversity in mesophilic anaerobic process operated at HRT of 40 and 20 days was found increased when thermophilic aerobically-digested

7

sludge were introduced to enhance the reactor performance. Nevertheless, the synergy effect between the anaerobic and aerobic reactors was not observed in this study possibly due to relatively short HRT (5 and 2.5 days) used in this study. Less diversity of methanogenic species detected were found corresponding well to the lower CH4 emission observed in the anaerobic reactor. Total 16 nitrifying bacterial species were detected in the system among which 7 species were found in the anaerobic reactor when the system was operated without recirculation. Their diversity in both reactors increased slightly when sludge recirculation was introduced (18 species in total) at HRT of 5 days. For instance, Nitrosomonas europaea which was found associated with an increase in N2O emission under high salinity condition (Liu et al., 2015) was detected when sludge recirculation was introduced in this study. Nevertheless, an increase in hydraulic loading through shortened HRT reduced the diversity of nitrifying bacteria in both reactors. Part of N2O could be produced during nitrification reaction but majority of them would be produced during denitrification (Aboobakar et al., 2013). Similar trend was observed for denitrifying bacteria but the number of species were much less, i.e. 3 and 4 species in total respectively. The microbial community in both reactors contained both nitrifying and denitrifying bacteria but their population diversity were different depending on the reactor operation. Comparing between the reactors, the nitrifying bacteria were more diverse under aerobic condition whereas larger number of denitrifying species were detected under anaerobic condition. When sludge recirculation was practiced, the diversity of nitrifying and denitrifying bacteria increased as they could still survive in both reactors. In addition, 2 species of anaerobic ammonia oxidizing (Anammox) bacteria was detected and they were found in the anaerobic reactor only. They are different from other autotrophic nitrifying bacteria as they do not use oxygen for ammonia oxidation (Jetten et al., 1998). Relative abundance of DNA specific to methanogenic, nitrifying and denitrifying bacteria under different operating conditions are presented in Fig. 2. Among them, methanogenic DNA were found at highest number of copies followed by nitrifying and denitrifying DNA respectively. The abundance of methanogenic DNA (41.6  105 DNA copies/g sludge) were highest in anaerobic reactor operated at HRT of 5 days without sludge recirculation. They were decreased when sludge recirculation was introduced and

Fig. 2. Quantification of DNA specific to methanogenic, nitrifying and denitrifying bacteria in anaerobic and aerobic (MBR) reactors.

Please cite this article in press as: Nuansawan, N., et al. Effect of hydraulic retention time and sludge recirculation on greenhouse gas emission and related microbial communities in two-stage membrane bioreactor treating solid waste leachate. Bioresour. Technol. (2016), http://dx.doi.org/10.1016/j. biortech.2016.01.109

8

N. Nuansawan et al. / Bioresource Technology xxx (2016) xxx–xxx

HRT was shortened to 2.5 days along with lowering CH4 emission observed in the reactor. Considerable amount of methanogenic DNA (12.8  105–19.1  105 DNA copies/g sludge) was also detected in the aerobic reactor. Most of them would not be active as the CH4 emission was found much lower in the aerobic reactor. The nitrifying DNA was found at relatively low level comparing to methanogenic DNA. They were 1.1  105–2.8  105 and 1.6  105–4.9  105 DNA copies/g sludge in the anaerobic and aerobic reactors respectively. The numbers of DNA copies were corresponding to higher diversity of nitrifying microorganisms detected in the aerobic reactor even though the N2O emissions from both reactors were not found different. Meanwhile, the denitrifying DNA was detected at lowest level (0.2  105–0.8  105 DNA copies/g sludge). In most case, higher numbers were found in the anaerobic reactor. Yapsaki et al. (2011) identified and quantified nitrogen-converting organisms in a full-scale leachate treatment plant and found that Nitrosomonas and Nitrospira species as major ammonia oxidizing bacteria (AOB) and nitrite oxidizing bacteria (NOB) and they were predominated over other nitrogenconverting organisms in the system. In this study, the presence of those microorganisms was confirmed together with the information on their relative diversity. From this study, negative effect of increasing hydraulic condition through shortened HRT and sludge recirculation on the diversity and abundance of GHG related microorganisms, i.e. methanogenic, nitrifying and denitrifying species were observed. As the diversity and abundance of those microorganisms were less, CH4 emissions from two-stage MBR were lowered whereas they had insignificant effect on N2O emission. Meanwhile, Yan et al. (2016) reported that internal recycle and could inhibit the N2O reductase gene-containing bacteria in treatment the process and the effect of recycle was more pronounced to the functional gene than the bacterial community composition. 4. Conclusion GHG emissions from two-stage MBR treating solid waste leachate during long term operation (>500 days) were investigated. Highest CH4 and N2O emissions took place from first stage anaerobic reactor operated at HRT of 5 day without sludge recirculation. These emissions accounted for 8.26% of C and 0.16% of N loading into the reactor. An increase in hydraulic loading through sludge recirculation and shortened HRT reduced CH4 emission by 17– 31% while organic and nitrogen removal efficiencies were adversely affected. Analysis of microbial communities suggest less diversity and DNA abundance of CH4 and N2O producing microorganisms when the hydraulic loading was increased. Acknowledgements This research was financially supported by Kasetsart University Research and Development Institute (KURDI) – Thailand. References Aboobakar, A., Cartmell, E., Stephenson, T., Jones, M., Vale, P., Dotro, G., 2013. Nitrous oxide emissions and dissolved oxygen profiling in a full-scale nitrifying activated sludge treatment plant. Water Res. 47, 524–534. Ahmed, F.N., Lan, C.Q., 2012. Treatment of landfill leachate using membrane bioreactor: a review. Desalination 287, 41–54.

APHA, 2005. Standard Methods for the Examination of Water and Wastewater, 21st ed. American Public Health Association/American Water Works Association/ Water Environment Federation, Washington, D.C.. Barton, P.K., Atwater, J.W., 2002. Nitrous oxide emissions and the anthropogenic nitrogen in wastewater and solid waste. J. Environ. Eng. ASCE 128 (2), 137–150. Chiemchaisri, C., Yamamoto, K., 2005. Enhancement of oxygen transfer and nitrogen removal in membrane separation bioreactor for domestic wastewater treatment. Water Sci. Technol. 51 (10), 85–92. Chiemchaisri, C., Chiemchaisri, W., Junsod, J., Threedeach, S., Wicreanarachchi, P.N., 2009. Leachate treatment and greenhouse gas emission in subsurface horizontal flow constructed wetland. Bioresour. Technol. 100, 3808–3814. Chiemchaisri, C., Chiemchaisri, W., Nindee, P., Chang, C.Y., Yamamoto, K., 2011. Treatment performance and microbial characteristics in two-stage membrane bioreactor applied to partially stabilized leachate. Water Sci. Technol. 64 (5), 1064–1072. Frison, N., Chiumenti, A., Katsou, E., Malamis, S., Bolzonella, D., Fatone, F., 2015. Mitigating off-gas emissions in the biological nitrogen removal via nitrite process treating anaerobic effluents. J. Cleaner Prod. 93, 126–133. Itokawa, H., Hanaki, K., Matsuo, T., 2001. Nitrous oxide production in high-loading biological nitrogen removal process under low COD/N ratio condition. Water Res. 35, 657–664. Jang, H.M., Ha, J.H., Park, J.M., Kim, M.S., Sommer, S.G., 2015. Comprehensive microbial analysis of combined anaerobic-thermophilic aerobic process treating high-strength food wastewater. Water Res. 73, 291–303. Jetten, M.S.M., Strous, M., van de Pas-Schoonen, K.T., Schalk, J., van Dongen, L., van de Graaf, A.A., Logemann, S., Muyzer, G., van Loosdrecht, M.C.M., Kuenen, J.G., 1998. The anaerobic oxidation of ammonium. FEMS Microbiol. Rev. 22, 421– 437. Kampschreur, M.J., Tan, N.C.G., Kleerebezem, R., Picioreanu, C., Jetten, M.S.M., van Loosdrecht, M.C.M., 2008. Effect of dynamic process conditions on nitrogen oxides emission from a nitrifying culture. Environ. Sci. Technol. 42 (2), 429–435. Khemkhao, M., Nuntakumjorn, B., Techkarnjanaruk, S., 2012. UASB performance and microbial adaptation during a transition from mesophilic to thermophilic treatment of palm oil mill effluent. Environ. Manage. 103, 74–82. Kjeldsen, P., Balaz, M.A., Rooker, A.P., Baun, A., Ledlin, A., Christensen, T.H., 2002. Present and long-term composition of MSW landfill leachate: a review. Crit. Rev. Environ. Sci. Technol. 32 (4), 297–336. Limpiyakorn, T., Sonthiphand, P., Rongsayamanont, C., Polprasert, C., 2011. Abundance of amoA genes of ammonia-oxidizing archaea and bacteria in activated sludge of full-scale wastewater treatment plants. Bioresour. Technol. 102, 3694–3701. Lin, L., Lan, C.Y., Huang, L.N., Chan, G.Y.S., 2008. Anthropogenic N2O production from landfill leachate treatment. J. Environ. Manage. 87, 341–349. Liu, M., Liu, T., Peng, Y., Wang, S., Xiao, H., 2014. Effect of salinity on N2O production during shortcut biological nitrogen removal from landfill leachate. J. Biosci. Bioeng. 117 (5), 582–590. Liu, M., Yang, Q., Peng, Y., Liu, T., Xiao, H., Wang, S., 2015. Treatment performance and N2O emission in the UASB-A/O shortcut biological nitrogen removal system for landfill leachate at different salinity. J. Ind. Eng. Chem. 32, 63–71. Rodriguez-Caballero, A., Aymerich, I., Poch, M., Pijuan, M., 2014. Evaluation of process conditions triggering emissions of green-house gases from a biological wastewater treatment system. Sci. Total Environ. 493, 384–391. Sanguanpak, S., Chiemchaisri, C., Chiemchaisri, W., Yamamoto, K., 2013. Removal and transformation of dissolved organic matter (DOM) during the treating of partially stabilized leachate in membrane bioreactor. Water Sci. Technol. 68 (5), 1091–1099. Van Dijk, L., Roncken, G.C.G., 1997. Membrane bioreactors for wastewater treatment: the state of the art and new developments. Water Sci. Technol. 35, 35–41. Wang, X., Jia, M., Chen, X., Xu, Y., Lin, X., Kao, C.M., Chen, S., 2014. Greenhouse gas emissions from landfill leachate treatment plants: a comparison of young and aged landfill. Waste Manage. 34, 1156–1164. Wisniowski, J., Robert, D., Surmacz-Gorska, J., Miksch, K., Weber, J.V., 2006. Landfill leachate treatment methods: a review. Environ. Chem. Lett. 4, 51–61. Xing, C.H., Yamamoto, K., Fukushi, K., 2006. Performance of an inclined-plate membrane bioreactor at zero excess sludge discharge. J. Membr. Sci. 275, 175– 186. Yan, X., Han, Y., Li, Q., Sun, J., Su, X., 2016. Impact of internal recycle ratio on nitrous oxide generation from anaerobic/anoxic/oxic biological nitrogen removal process. Biochem. Eng. J. 106, 11–18. Yan, X., Li, L., Liu, J., 2014. Characteristics of greenhouse gas emission in three fullscale wastewater treatment processes. J. Environ. Sci. 26, 256–263. Yapsaki, K., Aliyazicioglu, C., Mertoglu, B., 2011. Identification and quantitative evaluation of nitrogen-converting organisms in a full-scale leachate treatment plant. J. Environ. Manage. 92, 714–723. Zheng, M., He, D., Ma, T., Chen, Q., Liu, S., Ahmad, M., Gui, M., Ni, J., 2014. Reducing NO and N2O emission during aerobic denitrification by newly isolated Pseudomonas stutzeri PCN-1. Bioresour. Technol. 162, 80–88.

Please cite this article in press as: Nuansawan, N., et al. Effect of hydraulic retention time and sludge recirculation on greenhouse gas emission and related microbial communities in two-stage membrane bioreactor treating solid waste leachate. Bioresour. Technol. (2016), http://dx.doi.org/10.1016/j. biortech.2016.01.109