- Email: [email protected]
Anaerobic co-digestion of sewage sludge, food waste and yard waste: Synergistic enhancement on process stability and biogas production
Journal Pre-proofs Anaerobic co-digestion of sewage sludge, food waste and yard waste: Synergistic enhancement on process stability and biogas production Lan Mu, Lei Zhang, Kongyun Zhu, Jiao Ma, Muhammad Ifran, Aimin Li PII: DOI: Reference:
S0048-9697(19)35422-1 https://doi.org/10.1016/j.scitotenv.2019.135429 STOTEN 135429
To appear in:
Science of the Total Environment
Received Date: Revised Date: Accepted Date:
29 August 2019 20 October 2019 6 November 2019
Please cite this article as: L. Mu, L. Zhang, K. Zhu, J. Ma, M. Ifran, A. Li, Anaerobic co-digestion of sewage sludge, food waste and yard waste: Synergistic enhancement on process stability and biogas production, Science of the Total Environment (2019), doi: https://doi.org/10.1016/j.scitotenv.2019.135429
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Elsevier B.V. All rights reserved.
Anaerobic co-digestion of sewage sludge, food waste and yard waste: Synergistic enhancement on process stability and biogas production
Lan Mu, Lei Zhang*, Kongyun Zhu, Jiao Ma, Muhammad Ifran, Aimin Li
Key Laboratory of Industrial Ecology and Environmental Engineering, School of Environmental Science and Technology, Dalian University of Technology, Linggong Road 2, Dalian 116024, PR China *Corresponding author. Tel.: +86 411 8470 7448; fax: +86 411 8470 6679 E-mail: [email protected] (L. Zhang)
Abstract Anaerobic co-digestion (co-AD) could be a more sustainable waste management solution by sharing the existed anaerobic digestion (AD) facilities and generating more biogas energy. In this study, a series of co-AD of different urban derived organic wastes (sewage sludge-SS, food waste-FW, yard waste-YW) were conducted in a semi-continuous mode, and the corresponding dynamic evolutions of microbial community structure were followed by using real-time quantitative polymerase chain reaction (qPCR). As for co-AD of two feedstocks, introduction of SS (25%, VS basis) in FW significantly improved the process stability and archaea/total microbe ratio (from 0.4% to 17.1%), which might be due to the regulating effect of abundant trace metals in SS; co-AD of SS (25%, VS basis) with YW improved the methane yield by 1
2.04 times than AD of YW only together with higher methane contents (57.4±1.3% vs. 50.9±2.2%); in co-AD of FW and YW, synergistic effects in terms of increased methane production (3.4%-19.1%) were observed, which was correlated with more robust growth of both bacteria and archaea. As for co-AD of three feedstocks, high methane yields of 314.9±17.1 mL/g VS were achieved with a reliable stability. These findings could provide some fundamental and technical information for the co-treatment of urban derived organic wastes in centralized AD facilities.
Keywords: Anaerobic digestion, food waste, sewage sludge, yard waste, co-digestion
2
1. Introduction Municipal solid waste (MSW) from social and industrial activities has increased over the last few years. Globally, the MSW generation was recorded 2.01 billion tons in 2016 and is expected to reach up to 3.40 billion tons in 2050 (The World Bank, 2018). Recently, in order to reduce the MSW generation and increase the resource recycling efficiency, the waste sorting policy is becoming more popular. Thus high percentage of organic waste streams of MSW, such as food waste (FW), sewage sludge (SS), yard waste (YW) is separately collected. Therefore, for these perishable wastes with high moisture content, landfill and incineration is considered less attractive because of leachate discharge, land occupation, greenhouse gas (GHG) emission, low calorific value and high pollutants emission risk (Xiong et al., 2019; Mak et al., 2018; Zhang et al., 2018b). Similarly, composting is also considered of less public interest due to low benefit of the fertilizer products (Bollon et al., 2013). In contrast, AD is becoming a more desirable option for sustainable management of these biodegradable wastes with high moisture due to fuel energy recovery and low carbon footprint (Jain et al., 2015). So far, extensive works have been undertaken in evaluating substrate types, operating parameters, pre-treatment methods on process performance of AD process (Mata-Alvarez et al., 2014; Jain et al., 2015; Komilis et al., 2017; Wei et al., 2018). The extensive practical experience indicated that commercialization of centralized biogas plants is still facing different technical and economic challenges related to process efficiency and stability, especially for anaerobic mono-AD of individual
3
feedstock. For example, AD of SS often encountered by low methane yields due to the recalcitrant properties of microbial cell wall and extracellular biopolymers. Although the methane production could be improved by thermal, mechanical and/or chemical pretreatments, the high pretreatment costs limited its application (Carrere et al., 2016). YW, which contained a compact structure among cellulose, hemi-cellulose and lignin, impeded the attack of anaerobes, thus often resulted in long digestion period and low biogas yield (Kang et al., 2018). In contrast, AD of FW presented excellent organics degradation rate and methane productivity. However, process unbalance often occurred as indicated by high level of short chain fatty acids (SCFAs) and low pH values (Tampio et al., 2014; Yong et al., 2015; Zhang et al., 2013). These researches revealed that good biodegradability, balanced macro-, micro-nutrients, high buffering capacity are some essential properties required to obtain highly efficient and stable process performance (Mancini et al., 2018; Thanh et al., 2016). However, it is difficult to meet in AD of individual substrates because they may lack certain characteristic that limits its efficacy in AD. In order to increase process performance and share the waste treatment facility, co-AD of different substrates is an emerging practice. By combining the characteristics among substrates, co-AD greatly increased the process performance and stability. For instance, in farm anaerobic digester, corncob stock was co-digested with animal manure, by which the methane productivity was increased by 11.4 times (Wu et al., 2010). In wastewater treatment plants (WWTPs), FW was co-digested with SS in existing digesters to improve the economic viability with more energy 4
generation and saving investment of wastewater treatment (Wannapokin et al., 2018; Fitamo et al., 2016). For other cases, some synergistic effects were also found in co-AD of different waste streams, e.g. FW and leachate of MSW incineration plant (Zhang et al., 2015a), piggery wastewater and FW (Zhang et al., 2011). The improved process performance could be attributed to the balanced macro- and micro-nutrients, the diluting of the biological inhibitors, and/or the increased biodegradable organic loading (Zhang et al., 2011; Zhang et al., 2015a). In another beneficial aspect, co-AD strategy could exploit this surplus capacity of exiting digesters at local site, by which waste transportation cost would be reduced, and the additional energy obtained from biogas combustion could be delivered to the electricity network or on-site to compensate the energy consumption of waste treatment plant operation (Fitamo et al., 2016). Although great progresses have been achieved in co-AD of different wastes, most cases focused on evaluating the preliminary feasibility of process performance, and some conclusive results were only drawn from batch experiments other than the continuous mode, which were often adopted in practical application and more suitable to provide information regarding positive or negative synergistic effects (Konrad K. et al., 2015). In addition, it is still lack to systematically correlate the feedstock property, synergistic effect and underlying microbial mechanism in a certain scenario. Considering these limitations, in this study, we took organic waste management in urban area as a typical case, and three distinctive and the largest amount organic wastes (i.e. SS, FW and YW) were investigated as co-substrates in semi-continuous 5
AD. These three substrates possessed the distinctive properties in term of organics components, biodegradability and nutrients, thus resulting in some unexplored interactive effects. The investigation on anaerobic digestion of these three substrates had strong application background and potential scientific value. Specifically, the objectives of this study are: (i) to comprehensively characterize the three representative urban organic wastes (YW, SS and YW), (ii) to investigate their interactive effects through semi-continuous co-AD of paired and three substrates, (iii) to analyze the dynamic evolutions of microbial community structure using real-time quantitative polymerase chain reaction (qPCR), and explore the underlying microbial mechanisms.
2. Materials and methods 2.1 Feedstock and inoculum Food waste used in this study was taken from the cafeteria of Dalian University of Technology, China. The raw FW was homogenized using a crusher, and resulting slurry was passed through a 14-mesh sieve (with a hole size of 1.4 mm×1.4 mm) and stored at -20 °C in refrigerator. The frozen FW was thawed at 4 °C and a certain amount of tap water was added to obtain proper feedstock for anaerobic reactors. Phoenix tree leaves, a typical YW, were collected from the campus of Dalian University of Technology and dried in an oven at 40 °C for 48 hours. The dried phoenix tree leaves were crushed using an electric breaker and passed through a 20-mesh sieve (with a hole size of 0.8 mm × 0.8 mm) for later use. The SS as a 6
feedstock was taken from the sludge dewatering unit of a wastewater treatment plant (WWTP) located in Dalian, China. Then, it was transported to laboratory within one hour in a sealed plastic bag, and stored at 4 °C in refrigerator before use. Anaerobic sludge as a seed sludge was taken from an anaerobic digester treating SS located in Dalian city. 2.2 Experimental 2.2.1 General procedure of BMP tests and semi-continuous experiments Biochemical methane potential (BMP) tests of three substrates were carried out according to Zhang et al. (2015b). A 500-mL Schott Duran bottle served as a reactor with a working volume of 300 mL. Initially, 200 mL of seed sludge and 100 mL of substrate (including 3 g.0 VS equivalent substrates and tap water) were poured into a reactor under a nitrogen gas sweeping condition, later the reactor was capped with a silica gel stopper and purged using nitrogen for another 5 min to create an oxygen free atmosphere. After purging, the reactor was reversely placed in a 37 °C shaking incubator at 150 rpm. BMP tests lasted for 60 days until a negligible amount of methane generation was observed. For semi-continuous reactors, the startup phase was same with the BMP tests. After 4 days, the reactor was run in a semi-continuous mode by discharging 15 mL of the digestate and feeding with 15 mL sole or mixed substrates according to Fig. 1 and Table 3 once a day. All the reactors were operated at an organic loading rate (OLR) of 4.0 g VS/L day and a hydraulic retention time (HRT) of 20 days. 2.2.2 Experimental design 7
Anaerobic co-digestion of sewage sludge and food waste Initially, a reactor named as R1, was run with FW as the sole feedstock (Fig. 1). On Day 28, when upset occurred as indicated by the VFAs accumulation, the digestate in R1 was divided into three reactors, named as R1-1, R1-2, R1-3, which were fed by different substrates. R1-1 was operated as control for mono-AD of FW. In R1-2, a certain amount of trace elements were added together with FW to examine the effect of trace elements on long-term performance of AD of FW. The trace metals in the feedstock were 100 mg/L of Fe2+, 2.0 mg/L of Co2+, 10.0 mg/L of Ni2+ and 5.0 mg/L of Mo2+. In order to evaluate the co-AD performance, R1-3 was fed with a mixture of FW and SS (3:1, VS basis). Anaerobic co-digestion of sewage sludge and yard waste Two reactors, named as R2 and R3 were run in a parallel with YW as a sole feedstock. R2 was run as a control by feeding YW only during the whole digestion period. As for R3, on Day 39, the feeding substrate was changed to a mixture of YW and SS (VS basis, 3:1), and named as R3’. Anaerobic co-digestion of food waste and yard waste For evaluating the performance of co-AD of YW and FW, three reactors named as R4, R5 and R6 were fed with mixture of YW and FW with ratio of 3:1(VS basis) for R4 and 2:2 (VS basis) for R5 and 1:3 (VS basis) for R6. Anaerobic co-digestion of sewage sludge, food waste and yard waste In order to investigate the feasibility of co-AD of YW, SS and FW, on Day 37-39, half of R4, R5 or R6 was taken out and transferred to a new reactor, which was named 8
as R4-1, R5-1 and R6-1, respectively. A quarter of feedstocks for previous R4, R5 and R6 were replaced by SS on VS basis, and used as feedstocks for R4-1, R5-1 and R6-1, respectively. The resulting feedstock compositions for R4-1, R5-1 and R6-1 were the mixtures of YW, FW and SS with the ratios of 9:3:4 (VS basis) and 6:6:4 (VS basis) and 3:9:4 (VS basis), respectively.
2.3 Analytical methods 2.3.1 General methods Total solid content (TS) was determined by drying the sample at 105 °C for 24 h in an oven (101A-3E, SLIW, China) and volatile solid content (VS) was measured by burning the sample at 550 °C for 5 h in a muffle furnace (SX2-2.5-12, PULUO, China) (APHA, 2005). pH was monitored using a digital pH meter (PB-10, Sartorius, Germany). C, H and N were measured by an elemental analyzer (Vario EL III, Elementar, Germany). The content of O (% TS) was estimated by subtracting the total percentage of C, H, N and ash from 100%. Alkalinity was measured through titration using hydrochloric acid (1.0 mol/L). The metals in substrate were extracted by microwave assisted digestion using a mixture of HCl + HF + H2O2 (5:3:2). The trace metals in extracted solution were analyzed by Inductively Coupled Plasma Optical Emission Spectrometer
(ICP-OES, Opima 2000 DV, Perkinelmer, America) and
inductively coupled plasma mass spectrometry (ICP-MS, 7900X, Agilent, Japan). 9
Cellulose, hemi-cellulose and lignin were determined according to Van Soestn method (Van Soest et al., 1991) using a semi-automatic meter (FIWE3, VELP, Italy). All these measurements were determined in triplicates. The biogas composition in the headspace of reactor was determined using a gas chromatograph (GC-7900, TECHEPMP, China) equipped with a thermal conductivity detector (TCD) and TDX-01 packed column. The detail GC operating conditions and methodology for biogas generation was described in our previous publications work (Mu et al., 2018; Zhang et al., 2015b). As for VFAs analysis, another gas chromatograph (GC-7900, TECHEPMP, China) was set up, which was installed with a capillary column (DB-FFAP, 30 m × 0.25 mm × 0.25 μm) and a flame ionization detector (FID). The detail sample pretreatment and VFAs calibration could be found in our previous work (Mu et al., 2018). 2.3.2 Real-time quantitative polymerase chain reaction (qPCR) At the end of semi-continuous experiments (Day 51 of R1-1 and Day 76 of the others), the fermentation broth of reactors was collected respectively for qPCR analysis. Two universal primers of 338F/518R (Klammer et al., 2008) and 787F/915F (Shin et al., 2010) targeting bacteria and archaea were used in this study (Table 1). DNAs of sample were extracted and purified using the Fast DNA SPIN Kit for Soil (Tiangen Biochemical Technology Co. LTD, China). Real-time qPCR was conducted on an ABI StepOnePlus qPCR system (Applied Biosystems, USA) with a SYBR Green approach used to measure the 16S rRNA gene copy numbers. Twenty microliter of mixture, including 10 μL 2 × SYBR® Premix Ex Taq II (TaKaRa), 0.4 μL ROX 10
Reference Dye (TaKaRa), 0.4 μM of corresponding primer, 1 μL DNA template and 1 μL nuclease-free water, was prepared for real-time qPCR analysis. The qPCR was performed according to the programme: initial denaturation at 95 °C for 30 s; 40 cycles of denaturation at 95 °C for 10 s; annealing at 57 °C for 30 s; final extension at 72 °C for 30 s. The standards curves used for calculating the copy numbers of genes were constructed according to previous studies (Ruijter et al., 2013; Traversi et al., 2012). For each measurement, standard curve was run together and sterile water was used as the negative control. The standard curves for bacteria and archaea are shown in Table 1.
S0048-9697(19)35422-1 https://doi.org/10.1016/j.scitotenv.2019.135429 STOTEN 135429
To appear in:
Science of the Total Environment
Received Date: Revised Date: Accepted Date:
29 August 2019 20 October 2019 6 November 2019
Please cite this article as: L. Mu, L. Zhang, K. Zhu, J. Ma, M. Ifran, A. Li, Anaerobic co-digestion of sewage sludge, food waste and yard waste: Synergistic enhancement on process stability and biogas production, Science of the Total Environment (2019), doi: https://doi.org/10.1016/j.scitotenv.2019.135429
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Elsevier B.V. All rights reserved.
Anaerobic co-digestion of sewage sludge, food waste and yard waste: Synergistic enhancement on process stability and biogas production
Lan Mu, Lei Zhang*, Kongyun Zhu, Jiao Ma, Muhammad Ifran, Aimin Li
Key Laboratory of Industrial Ecology and Environmental Engineering, School of Environmental Science and Technology, Dalian University of Technology, Linggong Road 2, Dalian 116024, PR China *Corresponding author. Tel.: +86 411 8470 7448; fax: +86 411 8470 6679 E-mail: [email protected] (L. Zhang)
Abstract Anaerobic co-digestion (co-AD) could be a more sustainable waste management solution by sharing the existed anaerobic digestion (AD) facilities and generating more biogas energy. In this study, a series of co-AD of different urban derived organic wastes (sewage sludge-SS, food waste-FW, yard waste-YW) were conducted in a semi-continuous mode, and the corresponding dynamic evolutions of microbial community structure were followed by using real-time quantitative polymerase chain reaction (qPCR). As for co-AD of two feedstocks, introduction of SS (25%, VS basis) in FW significantly improved the process stability and archaea/total microbe ratio (from 0.4% to 17.1%), which might be due to the regulating effect of abundant trace metals in SS; co-AD of SS (25%, VS basis) with YW improved the methane yield by 1
2.04 times than AD of YW only together with higher methane contents (57.4±1.3% vs. 50.9±2.2%); in co-AD of FW and YW, synergistic effects in terms of increased methane production (3.4%-19.1%) were observed, which was correlated with more robust growth of both bacteria and archaea. As for co-AD of three feedstocks, high methane yields of 314.9±17.1 mL/g VS were achieved with a reliable stability. These findings could provide some fundamental and technical information for the co-treatment of urban derived organic wastes in centralized AD facilities.
Keywords: Anaerobic digestion, food waste, sewage sludge, yard waste, co-digestion
2
1. Introduction Municipal solid waste (MSW) from social and industrial activities has increased over the last few years. Globally, the MSW generation was recorded 2.01 billion tons in 2016 and is expected to reach up to 3.40 billion tons in 2050 (The World Bank, 2018). Recently, in order to reduce the MSW generation and increase the resource recycling efficiency, the waste sorting policy is becoming more popular. Thus high percentage of organic waste streams of MSW, such as food waste (FW), sewage sludge (SS), yard waste (YW) is separately collected. Therefore, for these perishable wastes with high moisture content, landfill and incineration is considered less attractive because of leachate discharge, land occupation, greenhouse gas (GHG) emission, low calorific value and high pollutants emission risk (Xiong et al., 2019; Mak et al., 2018; Zhang et al., 2018b). Similarly, composting is also considered of less public interest due to low benefit of the fertilizer products (Bollon et al., 2013). In contrast, AD is becoming a more desirable option for sustainable management of these biodegradable wastes with high moisture due to fuel energy recovery and low carbon footprint (Jain et al., 2015). So far, extensive works have been undertaken in evaluating substrate types, operating parameters, pre-treatment methods on process performance of AD process (Mata-Alvarez et al., 2014; Jain et al., 2015; Komilis et al., 2017; Wei et al., 2018). The extensive practical experience indicated that commercialization of centralized biogas plants is still facing different technical and economic challenges related to process efficiency and stability, especially for anaerobic mono-AD of individual
3
feedstock. For example, AD of SS often encountered by low methane yields due to the recalcitrant properties of microbial cell wall and extracellular biopolymers. Although the methane production could be improved by thermal, mechanical and/or chemical pretreatments, the high pretreatment costs limited its application (Carrere et al., 2016). YW, which contained a compact structure among cellulose, hemi-cellulose and lignin, impeded the attack of anaerobes, thus often resulted in long digestion period and low biogas yield (Kang et al., 2018). In contrast, AD of FW presented excellent organics degradation rate and methane productivity. However, process unbalance often occurred as indicated by high level of short chain fatty acids (SCFAs) and low pH values (Tampio et al., 2014; Yong et al., 2015; Zhang et al., 2013). These researches revealed that good biodegradability, balanced macro-, micro-nutrients, high buffering capacity are some essential properties required to obtain highly efficient and stable process performance (Mancini et al., 2018; Thanh et al., 2016). However, it is difficult to meet in AD of individual substrates because they may lack certain characteristic that limits its efficacy in AD. In order to increase process performance and share the waste treatment facility, co-AD of different substrates is an emerging practice. By combining the characteristics among substrates, co-AD greatly increased the process performance and stability. For instance, in farm anaerobic digester, corncob stock was co-digested with animal manure, by which the methane productivity was increased by 11.4 times (Wu et al., 2010). In wastewater treatment plants (WWTPs), FW was co-digested with SS in existing digesters to improve the economic viability with more energy 4
generation and saving investment of wastewater treatment (Wannapokin et al., 2018; Fitamo et al., 2016). For other cases, some synergistic effects were also found in co-AD of different waste streams, e.g. FW and leachate of MSW incineration plant (Zhang et al., 2015a), piggery wastewater and FW (Zhang et al., 2011). The improved process performance could be attributed to the balanced macro- and micro-nutrients, the diluting of the biological inhibitors, and/or the increased biodegradable organic loading (Zhang et al., 2011; Zhang et al., 2015a). In another beneficial aspect, co-AD strategy could exploit this surplus capacity of exiting digesters at local site, by which waste transportation cost would be reduced, and the additional energy obtained from biogas combustion could be delivered to the electricity network or on-site to compensate the energy consumption of waste treatment plant operation (Fitamo et al., 2016). Although great progresses have been achieved in co-AD of different wastes, most cases focused on evaluating the preliminary feasibility of process performance, and some conclusive results were only drawn from batch experiments other than the continuous mode, which were often adopted in practical application and more suitable to provide information regarding positive or negative synergistic effects (Konrad K. et al., 2015). In addition, it is still lack to systematically correlate the feedstock property, synergistic effect and underlying microbial mechanism in a certain scenario. Considering these limitations, in this study, we took organic waste management in urban area as a typical case, and three distinctive and the largest amount organic wastes (i.e. SS, FW and YW) were investigated as co-substrates in semi-continuous 5
AD. These three substrates possessed the distinctive properties in term of organics components, biodegradability and nutrients, thus resulting in some unexplored interactive effects. The investigation on anaerobic digestion of these three substrates had strong application background and potential scientific value. Specifically, the objectives of this study are: (i) to comprehensively characterize the three representative urban organic wastes (YW, SS and YW), (ii) to investigate their interactive effects through semi-continuous co-AD of paired and three substrates, (iii) to analyze the dynamic evolutions of microbial community structure using real-time quantitative polymerase chain reaction (qPCR), and explore the underlying microbial mechanisms.
2. Materials and methods 2.1 Feedstock and inoculum Food waste used in this study was taken from the cafeteria of Dalian University of Technology, China. The raw FW was homogenized using a crusher, and resulting slurry was passed through a 14-mesh sieve (with a hole size of 1.4 mm×1.4 mm) and stored at -20 °C in refrigerator. The frozen FW was thawed at 4 °C and a certain amount of tap water was added to obtain proper feedstock for anaerobic reactors. Phoenix tree leaves, a typical YW, were collected from the campus of Dalian University of Technology and dried in an oven at 40 °C for 48 hours. The dried phoenix tree leaves were crushed using an electric breaker and passed through a 20-mesh sieve (with a hole size of 0.8 mm × 0.8 mm) for later use. The SS as a 6
feedstock was taken from the sludge dewatering unit of a wastewater treatment plant (WWTP) located in Dalian, China. Then, it was transported to laboratory within one hour in a sealed plastic bag, and stored at 4 °C in refrigerator before use. Anaerobic sludge as a seed sludge was taken from an anaerobic digester treating SS located in Dalian city. 2.2 Experimental 2.2.1 General procedure of BMP tests and semi-continuous experiments Biochemical methane potential (BMP) tests of three substrates were carried out according to Zhang et al. (2015b). A 500-mL Schott Duran bottle served as a reactor with a working volume of 300 mL. Initially, 200 mL of seed sludge and 100 mL of substrate (including 3 g.0 VS equivalent substrates and tap water) were poured into a reactor under a nitrogen gas sweeping condition, later the reactor was capped with a silica gel stopper and purged using nitrogen for another 5 min to create an oxygen free atmosphere. After purging, the reactor was reversely placed in a 37 °C shaking incubator at 150 rpm. BMP tests lasted for 60 days until a negligible amount of methane generation was observed. For semi-continuous reactors, the startup phase was same with the BMP tests. After 4 days, the reactor was run in a semi-continuous mode by discharging 15 mL of the digestate and feeding with 15 mL sole or mixed substrates according to Fig. 1 and Table 3 once a day. All the reactors were operated at an organic loading rate (OLR) of 4.0 g VS/L day and a hydraulic retention time (HRT) of 20 days. 2.2.2 Experimental design 7
Anaerobic co-digestion of sewage sludge and food waste Initially, a reactor named as R1, was run with FW as the sole feedstock (Fig. 1). On Day 28, when upset occurred as indicated by the VFAs accumulation, the digestate in R1 was divided into three reactors, named as R1-1, R1-2, R1-3, which were fed by different substrates. R1-1 was operated as control for mono-AD of FW. In R1-2, a certain amount of trace elements were added together with FW to examine the effect of trace elements on long-term performance of AD of FW. The trace metals in the feedstock were 100 mg/L of Fe2+, 2.0 mg/L of Co2+, 10.0 mg/L of Ni2+ and 5.0 mg/L of Mo2+. In order to evaluate the co-AD performance, R1-3 was fed with a mixture of FW and SS (3:1, VS basis). Anaerobic co-digestion of sewage sludge and yard waste Two reactors, named as R2 and R3 were run in a parallel with YW as a sole feedstock. R2 was run as a control by feeding YW only during the whole digestion period. As for R3, on Day 39, the feeding substrate was changed to a mixture of YW and SS (VS basis, 3:1), and named as R3’. Anaerobic co-digestion of food waste and yard waste For evaluating the performance of co-AD of YW and FW, three reactors named as R4, R5 and R6 were fed with mixture of YW and FW with ratio of 3:1(VS basis) for R4 and 2:2 (VS basis) for R5 and 1:3 (VS basis) for R6. Anaerobic co-digestion of sewage sludge, food waste and yard waste In order to investigate the feasibility of co-AD of YW, SS and FW, on Day 37-39, half of R4, R5 or R6 was taken out and transferred to a new reactor, which was named 8
as R4-1, R5-1 and R6-1, respectively. A quarter of feedstocks for previous R4, R5 and R6 were replaced by SS on VS basis, and used as feedstocks for R4-1, R5-1 and R6-1, respectively. The resulting feedstock compositions for R4-1, R5-1 and R6-1 were the mixtures of YW, FW and SS with the ratios of 9:3:4 (VS basis) and 6:6:4 (VS basis) and 3:9:4 (VS basis), respectively.
2.3 Analytical methods 2.3.1 General methods Total solid content (TS) was determined by drying the sample at 105 °C for 24 h in an oven (101A-3E, SLIW, China) and volatile solid content (VS) was measured by burning the sample at 550 °C for 5 h in a muffle furnace (SX2-2.5-12, PULUO, China) (APHA, 2005). pH was monitored using a digital pH meter (PB-10, Sartorius, Germany). C, H and N were measured by an elemental analyzer (Vario EL III, Elementar, Germany). The content of O (% TS) was estimated by subtracting the total percentage of C, H, N and ash from 100%. Alkalinity was measured through titration using hydrochloric acid (1.0 mol/L). The metals in substrate were extracted by microwave assisted digestion using a mixture of HCl + HF + H2O2 (5:3:2). The trace metals in extracted solution were analyzed by Inductively Coupled Plasma Optical Emission Spectrometer
(ICP-OES, Opima 2000 DV, Perkinelmer, America) and
inductively coupled plasma mass spectrometry (ICP-MS, 7900X, Agilent, Japan). 9
Cellulose, hemi-cellulose and lignin were determined according to Van Soestn method (Van Soest et al., 1991) using a semi-automatic meter (FIWE3, VELP, Italy). All these measurements were determined in triplicates. The biogas composition in the headspace of reactor was determined using a gas chromatograph (GC-7900, TECHEPMP, China) equipped with a thermal conductivity detector (TCD) and TDX-01 packed column. The detail GC operating conditions and methodology for biogas generation was described in our previous publications work (Mu et al., 2018; Zhang et al., 2015b). As for VFAs analysis, another gas chromatograph (GC-7900, TECHEPMP, China) was set up, which was installed with a capillary column (DB-FFAP, 30 m × 0.25 mm × 0.25 μm) and a flame ionization detector (FID). The detail sample pretreatment and VFAs calibration could be found in our previous work (Mu et al., 2018). 2.3.2 Real-time quantitative polymerase chain reaction (qPCR) At the end of semi-continuous experiments (Day 51 of R1-1 and Day 76 of the others), the fermentation broth of reactors was collected respectively for qPCR analysis. Two universal primers of 338F/518R (Klammer et al., 2008) and 787F/915F (Shin et al., 2010) targeting bacteria and archaea were used in this study (Table 1). DNAs of sample were extracted and purified using the Fast DNA SPIN Kit for Soil (Tiangen Biochemical Technology Co. LTD, China). Real-time qPCR was conducted on an ABI StepOnePlus qPCR system (Applied Biosystems, USA) with a SYBR Green approach used to measure the 16S rRNA gene copy numbers. Twenty microliter of mixture, including 10 μL 2 × SYBR® Premix Ex Taq II (TaKaRa), 0.4 μL ROX 10
Reference Dye (TaKaRa), 0.4 μM of corresponding primer, 1 μL DNA template and 1 μL nuclease-free water, was prepared for real-time qPCR analysis. The qPCR was performed according to the programme: initial denaturation at 95 °C for 30 s; 40 cycles of denaturation at 95 °C for 10 s; annealing at 57 °C for 30 s; final extension at 72 °C for 30 s. The standards curves used for calculating the copy numbers of genes were constructed according to previous studies (Ruijter et al., 2013; Traversi et al., 2012). For each measurement, standard curve was run together and sterile water was used as the negative control. The standard curves for bacteria and archaea are shown in Table 1.