Assessment of the degradation efficiency of full-scale biogas plants: A comparative study of degradation indicators

Accepted Manuscript Assessment of the degradation efficiency of full-scale biogas plants: A comparative study of degradation indicators Chao Li, Ivo Achu Nges, Wenjing Lu, Haoyu Wang PII: DOI: Reference:

S0960-8524(17)31273-7 http://dx.doi.org/10.1016/j.biortech.2017.07.157 BITE 18577

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

20 May 2017 22 July 2017 26 July 2017

Please cite this article as: Li, C., Nges, I.A., Lu, W., Wang, H., Assessment of the degradation efficiency of fullscale biogas plants: A comparative study of degradation indicators, Bioresource Technology (2017), doi: http:// dx.doi.org/10.1016/j.biortech.2017.07.157

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.

Assessment of the degradation efficiency of full-scale biogas plants: a comparative study of degradation indicators Chao Li a,b,c*, Ivo Achu Nges a, Wenjing Lu c, Haoyu Wang d a

Division of Biotechnology, Center for Chemistry and Chemical Engineering, Lund University, Naturvetarvägen 14, 22241, Lund, Sweden b

Nova Skantek Environmental Technology (Beijing) Co., Ltd, Beijing 100027, China

c

School of Environment, Tsinghua University, Beijing, 100084, China

d

State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin, 150090, China

Corresponding author: Li Chao P.O. Box 124, SE-22100 Lund, Sweden Email: [email protected] Division of Biotechnology, Center for Chemistry and Chemical Engineering, Lund University TEL: +46 (0) 2228193

Abstract Increasing popularity and applications of the anaerobic digestion (AD) process has necessitated the development and identification of tools for obtaining reliable indicators of organic matter degradation rate and hence evaluate the process efficiency especially in full-scale, commercial biogas plants. In this study, four biogas plants (A1, A2, B and C) based on different feedstock, process configuration, scale and operational performance were selected and investigated. Results showed that the biochemical methane potential (BMP) based degradation rate could be use in incisively gauging process efficiency in lieu of the traditional degradation rate indicators. The BMP degradation rates ranged from 70 to 90 % wherein plants A2 and C showed the highest throughput. This study, therefore, corroborates the feasibility of using the BMP degradation rate as a practical tool for evaluating process performance in full-scale biogas processes and spots light on the microbial diversity in full-scale biogas processes.

Keywords: Degradation efficiency, Biochemical methane potential, Microbial diversity, Chicken manure, Kitchen waste, Municipal solid waste

1. Introduction Biomass energy continues to play an important role in power generation and sustainable development. For instance, animal manure, agricultural and kitchen wastes have become an important resource for biogas (methane) production via anaerobic digestion (AD) in countries such as China and Sweden (Zhou et al., 2017). China’s agricultural sector is growing rapidly, resulting in increasing amounts of agricultural wastes such as straw, livestock and poultry manure, and organic wastes from the agroindustry, which can be used to generate biogas. There is a clear trend away from smaller household farming towards larger, intensive farms and consequently the production of large amounts of manure; and centralization, which allows for easier collection of this manures (Li et al., 2017). In order to ensure a competitive return on investment considering the increasing cost of feedstock, it is critically important to ensure a stable, efficient biogas (methane) production so much so that methane potential should approach the ultimate potential of the feedstock i.e. full utilisation of feedstock and a minimum residual methane potential (Ruile et al., 2015). Feedstock/digestate characterization parameters of AD such as the volatile solids (VS) degradation rate which is achieved by measurements of VS concentration in the feed and the digestate are used as routine process parameter (degradation indicator) in most full-scale biogas plants (Weiland, 2010). VS is envisaged as the total concentration of organic matter and not all of this may be bio-available for the AD process (Schievano et al., 2011). Other parameters commonly used to describe the concentration of wastewaters are the chemical oxygen demand (COD), biochemical oxygen demand (BOD) and total organic carbon (TOC). The BOD and COD constitutes the organic compounds in both biodegradable and non-biodegradable forms under aerobic conditions. The BOD is a common measurement of the consumed oxygen by microorganisms to decompose or oxidise organic matter and has its widest application in wastewater treatment (Liu et al., 2004). On the other hand, TOC is the amount carbon bound in an organic compound (Bouallagui et al., 2003). These parameters can represent the (partial) organic content of the feedstock and ratio between inputs and outputs are often used as indicators of process yield (Hartmann & Ahring, 2005) or to track the fate of organic carbon (Gao et al., 2011). However, the above process parameters can only

provide limited information about the organic matter degradation or oxidation under one-sample-only circumstance, and it may, therefore, be insufficient to assess the true process efficiency (Schievano et al., 2010). An approach to gain a substantial knowledge about the quality of substrates, as well as its biodegradability is by means of biochemical methane potential (BMP) test of feedstock and that of the digestate (residual methane potential, RMP) (Schievano et al., 2011). The biodegradability embodies the potential to produce methane under anaerobic conditions from organic materials. Besides the quantitative information concerning the methane potential of the substrate, the BMP experimental protocol may also include the degradation kinetics information (VDI 4630, 2016). In an attempt to evaluate process efficiency under practical conditions, it is best to employ an authentic and incisive indicator that would provide fruitful and experimentally determined process information. Numerous bench-scale experimental data have been published on BMP of feedstock for process optimisation in full-scale biogas plants (Asam et al., 2011; Koch et al., 2016; Linke et al., 2013). To our knowledge, only one study has been published wherein a related the BMP-based indicator, the methane volume achieved via BMP assay, was projected as a practical process indicator to assess degradation efficiency in full-scale biogas plants (Schievano et al., 2011). The BMP and residual methane production (RMP) was also hinted in a German standard a viable means to evaluate process performance (VDI 4630, 2016). Considering the practical importance of the BMP as an AD process indicator, it is therefore, worth the continuance especially when presented side-by-side with the other traditional process parameters commonly used in assessing degradation efficiencies or process performance. The AD process is highly complex and dynamic that makes the assessment of process efficiency a big challenge. This has led to an increased need for process assessment, microbial dynamics and operating knowledge. Methane, renewable energy source, origins from various type of organic matters via a multi-step anaerobic decomposition process involving a well-organized community of various microbial populations(Ghasimi et al., 2015a). Besides substrates characteristics and operational conditions, microorganisms and microbial community structure can greatly affect the AD process(Wang et al., 2015). Therefore, in other to understand the degradation process systematically,

an investigative look at microbial community structure is vital to provide valuable information for the performance of the AD process. Knowledge about the microbial community structures coupled with proper assessment of process efficiency could be a way forward towards an understanding of the full potential of the biogas plants. Therefore, in this study, four full-scale biogas plants of different process technologies and fed with different feedstocks were adopted and investigated in terms of process efficiencies and microbial community dynamics. The process efficiencies were assessed and compared, for the first time, both via the traditional indicators which are based on VS, COD, BOD, TOC, as well as the new BMP-based indicator. Microbial community analyses were presented and discussed in relation to the process performances towards a deeper understanding of the full potential of the biogas plants. Since maximizing substrate degradation is necessary for both an economical and environmental stand point, this study may be a way forward assisting feasibility studies, bio-wastes management, evaluating technologies for reduction of greenhouse gas emission.

2. Materials and methods 2.1. Configurations and operating conditions of full-scale biogas plants Four mesophilic (37 ºC) full-scale biogas plants were surveyed over a period of in a bid to gain insights into the average feedstock characteristics and operating conditions (Table 1). In general, the biogas plants were divided into pre-digestion, primary digestion and post digestion steps. The pre-digestion step can be split into three parts: feedstock collection, mixing, and sand removal. The first biogas plant termed A1 and

the second biogas plant termed Plant A2 are located in the same chicken farm in Penglai city (Shandong province, China) that utilises the chicken manure mixed with the wastewater as feedstock. The A1 consisted of eight digesters that were divided into two identical production lines. The A2 worked as the extended project which constituted twelve digesters with three production lines (Li et al., 2017). Each production line consisted of three primary digesters which were operated in parallel and one secondary digester which acted as collecting tank (Table 1).

The third biogas plant termed Plant B is located in Tsingtao (Shandong province, China). This biogas plant utilised homogenised thermal pre-treated kitchen waste as feedstock. The waste was collected separately from 4 municipalities after the removal of indigestible materials, such as chopsticks, shell waste, glasses and metals. Before thermal pre-treatment, the kitchen waste was shredded into particles with an average size of less than 10mm. Prior to feeding into the reactor, kitchen waste was diluted with wastewater produced in a 1:1 volume ratio to ensure liquidity, ease feeding and discharge. The process consisted of 2 digesters operated in parallel (Table 1). The fourth biogas plant termed plant C is also located in Tsingtao (Shandong province, China) and fed with the organic fraction of the municipal solid waste (OFMSW). The biodegradable fraction sorted by the pre-sorting lines of the MSW plant. The OFMSW varied widely and consisted of food waste (vegetable waste and fruit peels from the local market), kitchen waste and yard trimming (leaves or grasses). These reactors' volumes, loading rates, solid retention time (SRT) and all other conditions and parameters describing the AD processes of the four full-scale plants are listed in Table 1.

Table 1. Process parameters of the full-scale biogas plants (A1, A2, B and C) surveyed during the study (n = 5).

2.2. Sample collection To minimalize sample dilution and hence misleading methane yields, feedstock and digestate samples were collected from the digesters every 2-day for 10 days (n=5 for each plant; 20 in total) during a steady-state operation of the process. The feedstock samples were collected from pipes conveying into the (primary) digesters. Digestate samples after the AD processes were collected from circulating pipes of the (secondary) digesters. Approximately 1000 ml of sample was collected in plastic bottles and stored in the refrigerator (av. 4 ºC) prior to use. All samples were collected during active mixing in the process in a bid to ensure homogeneity.

2.3. Biochemical methane potential and residual methane potential The BMP and residual methane potential (RMP) tests were performed with the aid of the automatic methane potential test system or AMPTS II (Bioprocess Control AB, Sweden). The inoculums were collected from final effluent of the respective digesters i.e. from digester 4 for plants A1&A2, mixture of digesters 1&2 for plant B and digester 1 for plant C. Prior to the BMP tests, the inoculum was incubated at 37 oC in the water baths for 7 days following the protocol to decrease the endogenous biogas production (VDI 4630, 2016). The inoculum to substrate ratio (based on grams of volatile solids, VS) was set at 2:1. The experimental protocol was performed according to another study (Li et al., 2017). The incubation time ranged from 35 to 40 days and the processes were terminated when the daily methane production was less 1% of the total. For the RMP test, 400 ml of digestate was put in Schott bottles, flushed with nitrogen for 60 seconds and immediately sealed. The operation protocol was performed same as with the BMP test. All the tests were performed in triplicates under mesophilic (37 ºC) conditions were continued until no more gas production was observed.

2.4. Analytical methods Chemical characteristics of the feedstocks and output digestates were determined in triplicate for each sample. The TS, VS, TOC, total nitrogen (TN), organic nitrogen (ON), ammonium-nitrogen (N-NH4+), total phosphorous (TP), total alkalinity were determined according to the standard method (APHA, 2005). Volatile fatty acids (VFAs) were analysed by gas chromatography (Agilent 7890A, Agilent Technologies Inc, USA) in a 5-times diluted bulk sample after filtration through a nylon type, 0.45 µm filter as recommended in standards methods (APHA, 2005). The COD was measured by the method of potassium dichromate oxidation (APHA, 2005). The BOD5 determinations were carried out with the aid of the OxiTop® system (WTW, Weilheim, Germany) by using the manometric respirometric wherein activated sludge from a WWTP was used as inoculum (Roppola et al., 2007). Outliers in the BMP and RMP test were determined and eliminated with the aid of the Grubb test (P≤ 0.05). Statistical

differences between the process BMPs were determined by one-way analysis of variance (ANOVA) at a 95% confidence interval.

2.5 Definition of the organic matter degradation rate The organic matter degradation rates were defined in terms of BMP, VS, COD, BOD and TOC. After the fundamental methane potential test, the BMP degradation rate (BDR) was calculated according to the equation below. BDR (%) = ((BMP – RMP) / BMP) *100

(1)

Where BDR is the BMP degradation rate, BMP is the biochemical methane potential and RMP is the residual methane potential. The VS, COD, BOD and TOC degradation rates were calculated as the ratio between the difference in the inlet and outlet divided by the input following the same rationale in equation 1.

2.6 DNA extraction, PCR amplification DNA was extracted according to the instructions of the soil total DNA isolation kit (OMEGA Bio-tek, USA). 16S rRNA genes segments were amplified using bar-coded primer pairs of 338F (5’ACTCCTACGGGAGGCAGCA-3’) and 806R (5’- GGACTACHVGGGTWTCTAAT-3’) for bacteria and Arch519F (5’- CAGCCGCCGCGGTAA-3’) and Arch915R (5’- GTGCTCCCCCGCCAATTCCT3’) for archaea. The PCR amplification program for bacteria contained an initial denature at 95 oC for 3 min, followed by 27 cycles of denaturing at 95oC for 30s, annealing at 55oC for 30s, and extension at 72oC for 45s. The thermal cycling for archaea was similar to that for bacteria except that the cycles were 32 rather than 27. Purified PCR products were sent to Majorbio Technology Co., LTD (Shanghai, China) for Illumina Miseq PE300 sequencing.

2.7 High-Throughput 16S rRNA Gene Sequencing and Analysis

Bacteria and archaea sequencing reads were all assigned to each sample according to the unique barcode of each sample. The barcode and primers were then removed. Pairs of reads from the original DNA fragments were merged using FLASH (Magoč & Salzberg, 2011). The overlapped reads were thereafter filtered using QIIME quality filters. Default settings for the Illumina processing in QIIME were used. PCR chimeras were filtered out using UCHIME (Edgar et al., 2011). The average lengths of all of the clean reads were 468 bp and 396 bp for bacteria and archaea sequences, respectively. The taxonomic classification of the sequences of each sample was carried out individually using the RDP Classifier. The sequences to different taxonomy levels were assigned at the bootstrap cutoff of 50% as suggested by the RDP (Wang et al., 2007) . In addition, the diversity and richness indices were calculated using the relevant RDP pipeline modules (Zhang et al., 2015). The average lengths of all clean reads were between 394 to 430 bp and 400 to 405 bp for bacteria and Archaea respectively. The average sequencing depths were about 31,000 to 45,000 and 10,000 to 25,000 clean reads for the bacteria and Archaea community analysis respectively. The rarefaction curves were generated at 3% cut-off for both the bacteria and archaea communities are shown in Figure I at supplementary data. The curves of bacteria have completely reached a plateau; meantime the curves of Archaea also approached a plateau, indicating that these sequencing depths were enough to cover the whole microbial diversity for each sample.

3. Results and discussion 3.1. Characterization of feedstock and digestate Tables 2 to 4 presents the characteristics of the feedstocks and digestates from all the four biogas plants. The kitchen waste (plant B) showed the higher VS, COD, BOD and TOC values as compared to chicken manure (Plants A1 and A2) and OFMSW (plant C). In all the biogas plants, the VS values were surprisingly higher than COD values. However, it is worth mentioning that COD determination can be interfered with by the inorganic substances such as nitrite, chloride and hydrogen peroxide leading to erroneous results (Zhang et al., 2007). Meantime, underestimation of VS values in VFAs,

alcohol and ammonia-laden feedstock are not uncommon (Kreuger et al., 2011). As compared to VS and COD, the BOD values were understandably far lower. Whereas VS and COD are gotten via aggressive physical and chemical methods, BOD is gotten via a less aggressive biological method wherein not all the organic matter may be bio-available thereby warranting the low value. In fact, the seeding materials used in BOD estimation are nonspecific bacteria that may not be able to biodegrade some of the recalcitrant compounds present in the waste water (Kumar et al., 2010). The TOC is far lower than the other organic matter quantifying parameters, which is because only the organic carbon bound to the feedstock is quantified. The feedstock in biogas plants A1 and A2 (chicken manure) showed high levels of nitrogen (N) and NH4-N as compared to biogas plants B and C which had kitchen waste and OFMSW as substrates. As expected, the VS, COD, BOD, TOC and BMP (RMP) all decreased after the AD process as significantly fraction should have evolved as biogas, H 2S, NH3 etc.

Table 2．Characterization of the feedstock, intermediates and the digestate of the full-scale biogas plant A1

There was a dramatic increase in N, phosphorous (P) and NH4-N in digestate of plant B probably as results of pre-termal hydrolysis which may have contributed to the increased degradation of proteins thereby releasing compounds such as N and P. In general, macronutrients such P and potassium (K) in most of these AD process did not decrease substantially meanwhile N in the form of NH4-N were noted to increase in the digestate. The final pH values were alkaline in nature in all the plants, and within the optimal range for the methanogenic activity (Chen et al., 2008) though the free ammonia levels in plants, A1 and A2 under the prevailing digester pH may be inhibitory to the methanogenic process (Chen et al., 2008). The total VFAs were as high as 4.5 g/l, though the VFA/TA ratio was, in general, less than 0.4 which is compatible with stable methanogenic milieu as suggested in other studies e.g. (Lindorfer et al., 2008).

Table 3. Characterization of the feedstock, intermediate and the digestate of the full-scale biogas plants A2

Table 4．Characterization of the feedstock and the digestates of the full-scale biogas plants B and C

3. 2. Methane potential and residual methane potential Figure 1 shows the batch methane production from the feedstocks (BMP) and digestates (RMP) as related to biogas plants A1, A2, B and C all based per g VS. The methane production rates were fast for all three kinds of feedstocks. All the processes reached a plateau after about 10 days of incubation. The BMP of chicken manure, kitchen waste and OFMSW were 557 ± 26.3 (A1), 400 ± 3.8 (A2), 514 ± 8.4 (B) and 348 ± 23.6 (C) ml/gVS respectively. The BMPs of the various feedstocks achieved in the present study are in line with yields reported in other studies (Khalid et al., 2011) except for chicken manure which was exceptionally high (Li et al., 2013). The high yields could be as results of the exceptional high VFAs concentrations which are precursors of methane production in the manure (Tables 2&3). It may also be explained by the underestimation of the VS values and hence a concomitant overestimation of the methane yields in the VFAs laden manure. The difference in the BMP between before A1 and A2 could be explained by the extended pre-digestion applied in plant A1. As expected, the digestate showed a slower degradation process and lower methane potential or RMP (Figure 1a&1b). The RMP were 60.9 to 139, 28.9 to 50.6, 46.8 to 80.5 and 108 ml/gVS for plant A1, A2, B and plant C respectively, which corresponded to 10.9 to 25.0%, 7.2-12.7%, 9.1-15.7% and 31.0% of the BMP of the feedstock (Figure 2).

Figure 1. The cumulative methane production of feedstocks, digestates from plant A1 (a), plant A2 (b), plant B (c), plant C (d).

The low RMP values as compared to the BMP can be explained by the easily biodegradable organic fraction was converted to biogas in the primary digesters, thus, the organic fraction in the digestate which could be refractory in nature was hardly converted to methane especially at short SRT.

Depending on the feedstock and its characteristics, the SRT may have a big impact on the RMP. For example, the RMP from plant A1 (SRT of 30 days) was about two times higher than that of Plant A2 (SRT of 45 days). The digestate from plant B also showed a lower RMP (p ≤ 0.05) though with the longest SRT of 54 days, while plant C showed the highest RMP (p ≤ 0.05) with the shortest SRT of 17 days. It has been reported that the SRT may have a huge impact on the degradation efficiency than other AD process parameters Ruile et al. (2015). In this study, the SRT of 17 days in plant C resulted in the highest (p ≤ 0.05) RMP which could reach 31% of the BMP value (Figure 2). It should be reiterated though that OFMSW may contain recalcitrant compounds which are not easily hydrolysed especially under a short SRT (Nges & Liu, 2010).

Figure 2. Percentage of the residual methane potential for all the biogas plants investigated (P is primary digester and S is secondary digester)

3.3. Degradation indicators and process efficiency Many parameters and approaches have been used to characterise feedstocks and evaluate the degradation efficiency in AD Processes (Chen et al., 2008; Schievano et al., 2011; Somasiri et al., 2008). The VS degradation rate is well established and it is the traditional degradation indicator as per process efficiency or performance (Chen et al., 2008). In some instances, TOC, COD and BOD have been used to characterised the feedstock (wastewater and low strength waste) and digestate, hence as the indicator of degradation rates during the AD processes (Appels et al., 2008; Somasiri et al., 2008; Steyer et al., 2002). In this study, the BMP and RMP were determined as feedstock and digestate characteristics from where the BMP degradation rate, BDR (Eq. 1) was used as degradation rate indicator side-by-side the other traditional rate indicators. Figure 3 shows all the organic degradation rates in terms of VS, COD, BOD, TOC and BMP in the four full-scale biogas plants and for the digesters. The VS and COD degradation rates in plant A1 were about 60% while BOD and BMP degradation rates were about 80%. TOC degradation rates in primary digesters ranged from 74 to 83%. However, the TOC degradation rate in secondary digester was 74%.

In plant A2, COD, BOD and BMP degradation rates were on average 85% while VS degradation rate was 70%. TOC showed a very high degradation efficiency that reached to 98% in all the digesters. In plant B, both digesters showed a good degradation profile wherein the degradation rate based on all five indicators hovered around 90% (Figure 4c). In plant C, TOC based degradation rate was 80% while the VS, COD, BOD and BMP based degradation rates were 50, 70, 70, and 72%, respectively (Figure 4d). Paying cognisance to the fact that process performance should be judged on final effluent (digestate), special attention was paid on the secondary digesters in all four biogas plants. Amongst the degradation rate indicators, BMP was in the same range or higher than VS, COD and BOD except in plants B and C wherein COD and TOC based indicators were much higher.

Figure 3. Presentation of the various degradation rates as indicators during the processes: plant A1 (a), plant A2 (b), plant B (c), plant C (d).

The organic matter degradation is often used to measure process efficiency in AD and the VS degradation rate is a typical parameter for AD process evaluation (Hartmann & Ahring, 2005). The VS degradation rate can be estimated as the ratio between the VS in the feedstock and digestate (Schievano et al., 2011) and in some instances, the SRT has been factored in VS degradation rate wherein an empirical equation allows the estimation of the VS degraded (Appels et al., 2008). In the present study, the VS degradation rates in all digesters of the biogas plants showed the lowest values (Figure 3). This is probably because VS is only a quantitative presentation of organic matter and not all of this may be bio-available to the microbial consortium in the AD process (Schievano et al., 2011). Also, quantitative measurements of organic matter in terms of COD may be less specific, since it measures everything that can be chemically oxidised (Gao et al., 2011) while TOC estimates the organic matter content only in form of carbon. Generally, VS, TOC and COD give no information about the state of the biological process. However, BOD which is a measure of the dissolved oxygen consumed by microorganisms during the oxidation of the feedstock (Butkus & Manous, 2005), can be assumed to be closely related to biodegradability. However, BOD values and hence BOD degradation rates may be misleading

considering the non-specificity of seeding microorganisms (Kumar et al., 2010). It should be noted that for all the processes, between 10 to 50% of organic matter based on VS, TOC, COD and BOD were estimated as undegraded (Figure 3). It is plausible to state that this was a result of the refractory nature of the feedstock (Nges & Liu, 2010) as biogas plant C (OFMSW) showed the poorest organic matter degradation. It may as well be as a result of the inefficiency of AD process to degrade the biodegradable fraction (Schievano et al., 2011). However, no definite hypothesis can be put forth from the quantitative analysis of the organic matter alone. In fact, the VS analysis, as well as TOC and COD, provide only a quantitative measurement of the organic matter, while nothing tells about its biodegradability under the anaerobic conditions (Schievano et al., 2010). Thus, it may be inaccurate to use these indicators to evaluate the organic matter degradation rate in an AD process. On the other hand, the BMP-based indicator (BDR) considers both quantitative (TS and VS contents) and qualitative aspects of the organic sample (nature of organic molecules) as it concerns only the degradable fraction of sample (Schievano et al., 2011). This should therefore be considered as a better indicator to represent the organic matter degradation rate and process efficiency in full scale biogas plants. Worth mentioning is that biogas plants often generate digestate of high refractory content of organic matter with high potential RMP (BMP of digestate) (Nges & Björnsson, 2012). In such instances, the BDR may be even more vital for evaluating the efficiency the process. The BDR (BMP-based degradation rate) in the secondary digester of biogas Plants A2 (chicken manure) and B (kitchen waste) showed the highest organic matter degradation efficiency in terms of the reduction in organic content (˃ 90%) while plants A1 (chicken manure) and C (OFMSW) showed the lower organic degradation efficiency i.e organic content reduction was 72%. These results corresponded understandably to the BMP and RMP values of the processes. The implication of these findings was threefold;

(i) The BDR relates to actual fraction of the organic matter that can be converted to renewable energy in form of methane. Herein, its reduction incisively described efficiency (bioconversion of feedstock to methane) the biogas or AD process. There was less than 10% of the degradable fraction of organic matter in plants A2 and B while remaining degradable fraction in plants A1 and C were as high as 28%. (ii) The BDR values demonstrated that there was a fraction of organic matter that could not be digested under the prevailing reactor conditions. For example, the BDR values (72 to 90%) were higher than the VS degradation values (50 to 80%) indicating that about half of the remaining fraction of VS (20 to 50%) was not degraded under the BMP test (remaining fraction of BDR was 10 to 28%). This results are in agreement with those published in another study (Schievano et al., 2011). (iii) The BDR values also demonstrated that there was a substantial amount of degradable organic matter in effluents (digestates) of plants A1 and C which could be further transformed into methane. Paying cognisance to the fact that digestate are often deposited in open tanks prior to final discharge, measure should be taken to check and abate the potential methane formation and escape into the atmosphere. Methane is greenhouse gas and the BDR can be used therefore as a tool to check and abate global warming. It is worth mentioning that process efficiency have been gauged by comparing the ultimate or experimental methane yield with the theoretical methane yield of the substrate (Møller et al., 2004). However, calculation of theoretical methane yield of a heterogenous substrate may be complicated and theoretical methane yield as well as VS or COD does not allude to the actual biodegradation of the substrate under the prevailing reactor conditions. It should be noted that the BDR values were corroborated by the Achaea richness evaluated in terms of the Shannon diversity index as presented below (section 3.4). The high BDR in plant A1 and A2 also showed high Shannon diversity index values. It is plausible to state that processes with predominant acetoclastic methanogens (Methanosaeta) i.e. biogas plants B and C; and hydrogenotrophic methanogens (Methanocelleus) i.e. biogas plant A1 and A2 performed equally well. These results indicated that the performance of an AD process is not necessarily governed by the type of dominant

methanogens, neither by a specified methanogenic pathway, i.e. acetoclastic and hydrogenotrophic pathways. There is a growing body knowledge which opines that hydrogenotrophic methane production can perform as well as acetoclastic methane production (Sasaki et al., 2011). This phenomenon can be explained by the fact when acetoclastic methanogens are inhibited by high levels of VFAs and ammonia, the acetate is easily converted to hydrogen and carbon dioxide by syntrophic oxidizers which are then converted to biogas via hydrogenotrophic methanogens (Sun et al., 2014) with the concomitant alleviation of acetoclastic inhibition.

3.4．Microbial community The main of studying the diversity was to buttress the fact that feedstock characteristic could be directly liaised to process performance (degradation efficiency). The high-throughput sequencing results showed big differences in bacterial communities amongst biogas plants A1, A2, B and C. The predominant bacteria genus in the four digesters of plant A1 were Clostridium, which belongs to phylum Firmicutes. The predominant Clostridiaceae in the secondary digester (21%) was lower than that in the three primary digesters (relative abundance 31%, 32% and 34% for digester 1, 2 and 3, respectively), which shared very similar structures of bacterial genera (Figure 4a). Ruminiclostridium dominated both the primary and secondary digester(s) in plant A1 with nearly the same abundance of 8%-9% (Figure 4a). The members of Clostridium are known to produce cellulosomes which degrade recalcitrant microcrystalline cellulose and yield short-chain compounds for acetogens and methanogens, and they are ubiquitous in many anaerobic digesters (Ziganshin et al., 2011) (Zhang et al., 2016). These results were also mirrored in plant A2, which was fed with the same feedstock type and operated by using much alike SRT (more than 40 days) (Figure 4a). The significantly reduced abundance of Clostridiaceae in the secondary digester might be related to the longer SRT and change of substrates, which was fed with the effluent from the primary digester. The main function of many Ruminiclostridium strains is to decompose large molecules, such as proteins and lipids into short-chain fatty acids and it can also decompose cellulose, starch and other polysaccharides. In plant B, the

predominant bacterial genus was Bacteroides that belongs to the phylum Bacteroidetes (29%-38%), followed by Proteiniphilum (18%-23%) (Figure 4a). Bacteroides, like Clostridium, is another ubiquitous and versatile bacteria across a substantial number of AD reactors (Ghasimi et al., 2015b) with functions such as hydrolysis, fermentation and generating acetate (Weiss et al., 2010). The members of the genus Proteiniphilum are acidogenic, sugar-fermenting, saccharolytic and proteolytic bacteria which produces primary products of propionate, acetate and succinate (Cardinali-Rezende et al., 2012). The syntrophic bacteria, Syntrophomonas, owned a relatively high abundance of 5% in each Plant B digester, which was higher than the values of many digesters (Ghasimi et al., 2015b). Plant C owned a unique bacterial community that is obviously different from Plant A1, Plant A2 and Plant B (Figure 4a). The predominant bacteria of Plant A1, A2 and B, Clostridium and Bacteroides, only had a relative abundance of 1% and 7% in Plant C. Intriguingly, there are no obvious dominant bacterial genera in Plant C, and the genera that had the highest abundance, Alistipes and Runella, only had the relative abundance of about 9% of each. Both Alistipes and Runella belongs to the phylum Bacteroidetes. The former is known to ferment carbohydrates to produce propionic and succinic acids (Svensson et al., 2007), and they are often involved in acidogenesis (Díaz et al., 2007), while the latter utilises lactate as a substrate to produce acetate and propionate in the fermentation processes (Yi et al., 2014).

Figure 4. Distribution of bacteria (a) and archaea (b) from different plants at the genus levels.

The predominant archaea in the four digesters of plant A1 and plant A2 were Methanoculleus of family Methanomicrobiaceae (Figure 4b) with a relative abundance of 58%-85%. Methanomassiliicoccus (4%-18%) and Methanobrevibacter (1%-18%) were other methanogens in plant A1 and plant A2. It is obvious that Plant A was dominated by hydrogenotrophic methanogens which utilises H2 and CO2 as the main substrates to produce methane (Shin et al., 2010). On the other hand, Plant B and C were both dominated by the acetoclastic methanogen Methanosaeta, which belongs to the family

Methanosaetaceae. Its relative abundance ranged from 67% to 74% in plant B, and 55% in Plant C (Figure 4b). The similar predominant archaea in plants B and C can be explained by the similar feedstock characteristics. It should be noted that the feedstock in plant C included a considerable amount of kitchen waste which was the sole substrate in plant B. Methanosaeta has been demonstrated as a strict acetoclastic methanogen striving at low VFA and low ammonia levels (Ghasimi et al., 2015a; Wang et al., 2015). This agreed with the low VFAs and ammonia concentrations in plant B and C (Table 1). The ammonia concentration in plant A was two times higher than that in plant B, while it was eight times higher than that in plant C. Also, the VFAs concentration in plant A was about a hundred times higher than those in plants B and C. It has been reported that high ammonia or VFAs concentrations can promote the dominance of Methanoculleus and a likely shift from acetoclastic methanogenesis to hydrogenotrophic methanogenesis via a syntrophic acetate-oxidizing process (Werner et al., 2014). Although correlations between microbial community and degradation efficiency have been reported (Koo et al., 2017; Shin et al., 2016), the microbial structures that are responsible for a given biodegradation efficiency index are not fully understood for biogas plants. A further survey involving long-term monitoring of how microbial community can help develop high degradation efficiency would be performed for the furtherance of the present study.

4. Conclusions In evaluating process performance in biogas plants, the BMP-based indicator could be used in lieu of traditional indicator such as VS and COD. While VS or COD are only quantitative measurements of organic matter, the BMP relates to both quantitative (VS, COD) and qualitative (type of molecules) aspect of the feedstock (biodegradable fraction). Furthermore, the BMP-based indicator represents the actual energy content of the AD process and can, therefore, be used conveniently to assess the performance of full-scale biogas processes. This study also demonstrated that feedstock type and its degradation products had an unyielding impact on microbial diversity.

Acknowledge This research was financially supported by the National “Twelfth Five-Year” Plan for Science & Technology Support Program (2014BAC24B01, 2014BAD24B01).

References 1. APHA. 2005. Total, fixed, and volatile solids in solid and semisolid samples. In Standard methods for the examination of water and wastewater. 21st edition. Edited by Eaton AD, Clesceri LS, Rice EW, Greenberg AE, Franson MA. Baltimore: American Public Health Association/American Water Works Association/Water Environment Federation. 2. Appels, L., Baeyens, J., Degrève, J., Dewil, R. 2008. Principles and potential of the anaerobic digestion of waste-activated sludge. Progress in energy and combustion science, 34(6), 755-781. 3. Asam, Z.-u.-Z., Poulsen, T.G., Nizami, A.-S., Rafique, R., Kiely, G., Murphy, J.D. 2011. How can we improve biomethane production per unit of feedstock in biogas plants? Applied Energy, 88(6), 2013-2018. 4. Bouallagui, H., Ben Cheikh, R., Marouani, L., Hamdi, M. 2003. Mesophilic biogas production from fruit and vegetable waste in a tubular digester. Bioresource Technology, 86(1), 85-89. 5. Butkus, M.A., Manous, J.D. 2005. Biochemical Oxygen Demand. in: Water Encyclopedia, John Wiley & Sons, Inc. 6. Cardinali-Rezende, J., Colturato, L.F., Colturato, T.D., Chartone-Souza, E., Nascimento, A.M., Sanz, J.L. 2012. Prokaryotic diversity and dynamics in a full-scale municipal solid waste anaerobic reactor from start-up to steady-state conditions. Bioresource technology, 119, 373-383. 7. Chen, Y., Cheng, J.J., Creamer, K.S. 2008. Inhibition of anaerobic digestion process: A review. Bioresource Technology, 99(10), 4044-4064. 8. Díaz, C., Baena, S., Fardeau, M.-L., Patel, B. 2007. Aminiphilus circumscriptus gen. nov., sp. nov., an anaerobic amino-acid-degrading bacterium from an upflow anaerobic sludge reactor. International journal of systematic and evolutionary microbiology, 57(8), 1914-1918. 9. Edgar, R.C., Haas, B.J., Clemente, J.C., Quince, C., Knight, R. 2011. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics, 27(16), 2194-2200. 10. Gao, D.-w., Tao, Y., An, R., Fu, Y., Ren, N.-q. 2011. Fate of organic carbon in UAFB treating raw sewage: impact of moderate to low temperature. Bioresource technology, 102(3), 2248-2254. 11. Ghasimi, D.S., Tao, Y., de Kreuk, M., Abbas, B., Zandvoort, M.H., van Lier, J.B. 2015a. Digester performance and microbial community changes in thermophilic and mesophilic sequencing batch reactors fed with the fine sieved fraction of municipal sewage. Water research, 87, 483-493. 12. Ghasimi, D.S.M., Tao, Y., de Kreuk, M., Abbas, B., Zandvoort, M.H., van Lier, J.B. 2015b. Digester performance and microbial community changes in thermophilic and mesophilic sequencing batch reactors fed with the fine sieved fraction of municipal sewage. Water Research, 87, 483-493.

13. Hartmann, H., Ahring, B.K. 2005. Anaerobic digestion of the organic fraction of municipal solid waste: Influence of co-digestion with manure. Water Research, 39(8), 1543-1552. 14. Khalid, A., Arshad, M., Anjum, M., Mahmood, T., Dawson, L. 2011. The anaerobic digestion of solid organic waste. Waste Management, 31(8), 1737-1744. 15. Koch, K., Plabst, M., Schmidt, A., Helmreich, B., Drewes, J.E. 2016. Co-digestion of food waste in a municipal wastewater treatment plant: comparison of batch tests and full-scale experiences. Waste Management, 47, 28-33. 16. Koo, T., Shin, S.G., Lee, J., Han, G., Kim, W., Cho, K., Hwang, S. 2017. Identifying methanogen community structures and their correlations with performance parameters in four full-scale anaerobic sludge digesters. Bioresource technology, 228, 368-373. 17. Kreuger, E., Nges, I.A., Björnsson, L. 2011. Ensiling of crops for biogas production: effects on methane yield and total solids determination. Biotechnology for biofuels, 4(1), 44. 18. Kumar, A., Dhall, P., Kumar, R. 2010. Redefining BOD:COD ratio of pulp mill industrial wastewaters in BOD analysis by formulating a specific microbial seed. International Biodeterioration & Biodegradation, 64(3), 197-202. 19. Li, C., Strömberg, S., Liu, G., Nges, I.A., Liu, J. 2017. Assessment of regional biomass as co-substrate in the anaerobic digestion of chicken manure: Impact of co-digestion with chicken processing waste, seagrass and Miscanthus. Biochemical Engineering Journal, 118, 1-10. 20. Li, Y., Zhang, R., Liu, X., Chen, C., Xiao, X., Feng, L., He, Y., Liu, G. 2013. Evaluating methane production from anaerobic mono-and co-digestion of kitchen waste, corn stover, and chicken manure. Energy & fuels, 27(4), 2085-2091. 21. Lindorfer, H., Corcoba, A., Vasilieva, V., Braun, R., Kirchmayr, R. 2008. Doubling the organic loading rate in the co-digestion of energy crops and manure–a full scale case study. Bioresource technology, 99(5), 1148-1156. 22. Linke, B., Muha, I., Wittum, G., Plogsties, V. 2013. Mesophilic anaerobic co-digestion of cow manure and biogas crops in full scale German biogas plants: A model for calculating the effect of hydraulic retention time and VS crop proportion in the mixture on methane yield from digester and from digestate storage at different temperatures. Bioresource technology, 130, 689-695. 23. Liu, J., Olsson, G., Mattiasson, B. 2004. On-line monitoring of a two-stage anaerobic digestion process using a BOD analyzer. Journal of biotechnology, 109(3), 263-275. 24. Magoč, T., Salzberg, S.L. 2011. FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics, 27(21), 2957-2963. 25. Møller, H.B., Sommer, S.G., Ahring, B.K. 2004. Methane productivity of manure, straw and solid fractions of manure. Biomass and Bioenergy, 26(5), 485-495. 26. Nges, I.A., Björnsson, L. 2012. High methane yields and stable operation during anaerobic digestion of nutrient-supplemented energy crop mixtures. Biomass and Bioenergy, 47(0), 62-70. 27. Nges, I.A., Liu, J. 2010. Effects of solid retention time on anaerobic digestion of dewatered-sewage sludge in mesophilic and thermophilic conditions. Renewable Energy, 35(10), 2200-2206.

28. Roppola, K., Kuokkanen, T., Rämö, J., Prokkola, H., Heiska, E. 2007. Comparison study of different BOD tests in the determination of BOD 7 evaluated in a model domestic sewage. Journal of Analytical Methods in Chemistry, 2007. 29. Ruile, S., Schmitz, S., Mönch-Tegeder, M., Oechsner, H. 2015. Degradation efficiency of agricultural biogas plants–A full-scale study. Bioresource technology, 178, 341-349. 30. Sasaki, D., Hori, T., Haruta, S., Ueno, Y., Ishii, M., Igarashi, Y. 2011. Methanogenic pathway and community structure in a thermophilic anaerobic digestion process of organic solid waste. Journal of Bioscience and Bioengineering, 111(1), 41-46. 31. Schievano, A., D’Imporzano, G., Malagutti, L., Fragali, E., Ruboni, G., Adani, F. 2010. Evaluating inhibition conditions in high-solids anaerobic digestion of organic fraction of municipal solid waste. Bioresource Technology, 101(14), 5728-5732. 32. Schievano, A., D’Imporzano, G., Orzi, V., Adani, F. 2011. On-field study of anaerobic digestion full-scale plants (Part II): New approaches in monitoring and evaluating process efficiency. Bioresource technology, 102(19), 8814-8819. 33. Shin, S.G., Han, G., Lim, J., Lee, C., Hwang, S. 2010. A comprehensive microbial insight into two-stage anaerobic digestion of food waste-recycling wastewater. water research, 44(17), 4838-4849. 34. Shin, S.G., Koo, T., Lee, J., Han, G., Cho, K., Kim, W., Hwang, S. 2016. Correlations between bacterial populations and process parameters in four full-scale anaerobic digesters treating sewage sludge. Bioresource technology, 214, 711-721. 35. Somasiri, W., Li, X.-F., Ruan, W.-Q., Jian, C. 2008. Evaluation of the efficacy of upflow anaerobic sludge blanket reactor in removal of colour and reduction of COD in real textile wastewater. Bioresource Technology, 99(9), 3692-3699. 36. Steyer, J.P., Bouvier, J.C., Conte, T., Gras, P., Harmand, J., Delgenes, J.P. 2002. On-line measurements of COD, TOC, VFA, total and partial alkalinity in anaerobic digestion processes using infra-red spectrometry. Water Science and Technology, 45(10), 133-138. 37. Sun, L., Müller, B., Westerholm, M., Schnürer, A. 2014. Syntrophic acetate oxidation in industrial CSTR biogas digesters. Journal of Biotechnology, 171, 39-44. 38. Svensson, L.M., Björnsson, L., Mattiasson, B. 2007. Enhancing performance in anaerobic high-solids stratified bed digesters by straw bed implementation. Bioresource technology, 98(1), 46-52. 39. VDI 4630. 2016. Fermentation of organic materials – Characterization of the substrate, sampling,collection of material data, fermentation tests. . VDI Handbuch Energietechnik. 40. Wang, H., Tao, Y., Gao, D., Liu, G., Chen, C., Ren, N., van Lier, J.B., de Kreuk, M. 2015. Microbial population dynamics in response to increasing loadings of pre-hydrolyzed pig manure in an expanded granular sludge bed. Water research, 87, 29-37. 41. Wang, Q., Garrity, G.M., Tiedje, J.M., Cole, J.R. 2007. Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Applied and environmental microbiology, 73(16), 52615267. 42. Weiland, P. 2010. Biogas production: current state and perspectives. Applied microbiology and biotechnology, 85(4), 849-860.

43. Weiss, S., Tauber, M., Somitsch, W., Meincke, R., Mulla, H., Berg, G., Guebitz, G.M. 2010. Enhancement of biogas production by addition of hemicellulolytic bacteria immobilised on activated zeolite. Water Research, 44(6), 1970-1980. 44. Werner, J.J., Garcia, M.L., Perkins, S.D., Yarasheski, K.E., Smith, S.R., Muegge, B.D., Stadermann, F.J., DeRito, C.M., Floss, C., Madsen, E.L. 2014. Microbial community dynamics and stability during an ammoniainduced shift to syntrophic acetate oxidation. Applied and environmental microbiology, 80(11), 33753383. 45. Yi, J., Dong, B., Jin, J., Dai, X. 2014. Effect of increasing total solids contents on anaerobic digestion of food waste under mesophilic conditions: performance and microbial characteristics analysis. 46. Zhang, J., Cai, X., Qi, L., Shao, C., Lin, Y., Zhang, J., Zhang, Y., Shen, P., Wei, Y. 2015. Effects of aeration strategy on the evolution of dissolved organic matter (DOM) and microbial community structure during sludge bio-drying. Applied microbiology and biotechnology, 99(17), 7321-7331. 47. Zhang, X.D., Tao, Y., Hu, J.M., Liu, G., Spanjers, H., van Lier, J.B. 2016. Biomethanation and microbial community changes in a digester treating sludge from a brackish aquaculture recirculation system. Bioresource Technology, 214, 338-347. 48. Zhang, Z., Yuan, Y., Fang, Y., Liang, L., Ding, H., Jin, L. 2007. Preparation of photocatalytic nano-ZnO/TiO2 film and application for determination of chemical oxygen demand. Talanta, 73(3), 523-528. 49. Zhou, Y., Li, C., Nges, I.A., Liu, J. 2017. The effects of pre-aeration and inoculation on solid-state anaerobic digestion of rice straw. Bioresource Technology, 224, 9. 50. Ziganshin, A.M., Schmidt, T., Scholwin, F., Il’inskaya, O.N., Harms, H., Kleinsteuber, S. 2011. Bacteria and archaea involved in anaerobic digestion of distillers grains with solubles. Applied microbiology and biotechnology, 89(6), 2039-2052.

Figure 1. The cumulative methane production of feedstocks, digestates from plant A1 (a), plant A2 (b), plant B (c), plant C (d).

Figure 2. Percentage of the residual methane potential for all the biogas plants investigated (P is primary digester and S is secondary digester)

Figure 3. Presentation of the various degradation rates as indicators during the processes: plant A1 (a), plant A2 (b), plant B (c), plant C (d).

(a)

(b)

Figure 4. Distribution of bacteria (a) and archaea (b) from different plants at the genus levels.

Table 1. Process parameters of the full-scale biogas plants (A1, A2, B and C) surveyed during the study (n = 5). Parameters/substrates Substrate Process

Units

3000*2 110 ±35

Plant C OFMSW CSTR Solid-liquid separation 1 digestion unit 1 production line 10000 600 ± 32

1.23±0.3

1.5± 0.44

3.06 ± 0.51

0.97±0.4

0.85±0.3

1.2± 0.4

1.13±0.4

d d o C

30 10 37±2 8.3 0.35±0.05

45 10 37±2 7.8 0.29±0.04

54

17

37±1 7.9 0.1±0.04

34±1 7.6 0.08±0.03

g/L

4.3±0.8

4.2±0.54

0.83±0.21

0.45±0.11

Pretreatment Numbers & type of digesters Total volume Loading rate (w/w) Loading rate (TS) Organic loading rate (VS) SRT (primary) SRT (secondary) Temperature pH VFA/TA ratio Total VFA concentration

m3 t/d kgm-3d1

Plant A1 Chicken manure CSTR

Plant A2 Chicken manure CSTR

Plant B Kitchen waste CSTR

Sand removal tank

Sand removal tank

Thermal hydrolysis

8 digestion units 2 production lines, each with 3 primary digesters 3300*8 557 ±45

12 digestion units 3 productions lines, each with 3 primary digesters 3300*12 539 ±63

1.43±0.5

2 digestion unit 2 digesters in parallel

-3 -

kgm d 1

Table 2．Characterization of the feedstock, intermediates and the digestate of the full-scale biogas plant A1 Digestate

Intermediates Parameters

Units

feedstock

TS VS BMP NH4+-N TN ON TOC COD BOD VFA TA TK TP pH

% % Nml g-1 VS mg/l mg/l mg/l mg/l mg/l mg /l mg/l g/l mg/l mg/l mg/l

4.63±0.21 3.16±0.13 557±26.3 3094 ±125.3 4542 ±214.5 1448 ±137.2 9533 ±196.6 32198±956 21000±212 25801 ±1158 9.2 ±0.35 1885 ±143 298 ±4.5 6.5 ±0.4

Digester 1

Digester 2

Digester 3

Digester 4

2.76±0.02 1.31±0.02 106±4.4 4088 ±184.4 5006 ±156.9 917 ±74.3 864±27.0 12164 ±393 4250±184 4341 ±351 11.6 ±0.17 1992 ±119 264 ±3.1 8.3 ±0.2

2.7±0.01 1.22±0.06 60.9±5.2 4045 ±238.6 5103 ±182.5 1058 ±65.9 1042±18.4 9302 ±730 2000±148 3867 ±137 12.8±0.32 2138 ±262 344 ±9.5 8.4 ±0.5

2.78±0.04 1.25±0.09 116±5.3 4024 ±271.5 4924 ±251.7 900 ±91.3 1197±46.0 12402 ±449 3750±198 4635 ±249 11.8 ±0.94 2027 ±179 223 ±5.3 8.5 ±0.4

2.83±0.03 1.41±0.02 139±6.8 3829 ±126.9 5055 ±329.4 1226 ±88.4 1249±24.7 12164 ±505 5125±190 43.1 ±0.62 12.1 ±0.48 1940 ±152 237 ±3.6 8.4 ±0.3

Table 3. Characterization of the feedstock, intermediate and the digestate of the full-scale biogas plants A2 Parameters TS VS BMP NH4+-N TN ON TOC COD BOD VFA TA TK TP pH

Units

Feedstock

% % Nml g-1 VS mg/l mg/l mg/l mg/l mg/l mg /l mg /l g/l mg/l mg/l mg/l

6.13±0.11 4.26±0.06 400±3.76 3482±72.1 4145±7.16 559±72.4 12288±31.8 33700±565 24000±707 21891 ±2107 9.35±0.10 3323±180 294±4.94 6.35±0.04

Intermediates Digester 1 3.73±0.01 1.86±0.02 50.1±2.06 4001±98.2 5035±150 1023±180 2298±12.0 6525±106 4850±70 4188±332 14.65±0.42 3207±88.8 275 ±2.85 7.92±0.01

Digester 2 3.51±0.03 1.71±0.02 40.4±6.3 4213±95.3 4737±214 551±234 2221±15.6 5875±176 4850±212 4747±52.3 12.6±0.14 3105±79.5 290±4.28 7.91±0.01

Digester 3 3.45±0.07 1.67±0.04 50.6±12.6 4347±49.1 5233±28.6 877±56.8 2374±19.1 5550±212 4400±282 4443±66.5 13.4±0.12 3143±56. 317±8.67 7.92±0.02

Digestate Digester 4 3.30±0.05 1.58±0.03 28.9±1.90 4583±72.1 5101±71.6 505±102 1914 ±25.5 5075±176 3350±212 52.2±0.71 14.7±0.43 3147±60.7 325±3.27 7.89±0.01

Table 4．Characterization of the feedstock and the digestates of the full-scale biogas plants B and C Parameters TS VS BMP NH4+-N TN ON TOC COD BOD VFA TA TK TP pH

Units % % Nml g-1 VS mg/l mg/l mg/l mg/l mg/l mg /l mg/l g/l mg/l mg/l mg/l

Feedstock Plant B 8.21±0.04 6.67±0.01 514±8.4 196 ±41.4 1742 ±115.8 1546± 87.4 17518 ±161.2 76547±28888 35000±282 520±42.5 13.8±1.05 692 ±45.7 424.5 ±5.32 4.4 ±0.2

Digestate 1 Plant B 2.05±0.04 0.83±0.01 80.5±4.2 2402 ±123.8 2833 ±146.9 431 ± 67±8.1 5414±585 1125±63.6 692±33.8 15.7 ±2.01 804 ±63.8 94.4 ±6.77 7.9 ±0.5

Digestate2 Plant B 2.58±0.19 1.06±0.15 46.8±2.2 2142 ±89.6 3002 ±139.0 860 ±45.3 96±2.5 8810±726 1000±170 882±12.5 16.6 ±1.89 832 ±52.7 80.3 ±3.95 7.9 ±0.4

Feedstock Plant C 3.95±0.05 1.51±0.05 348±23.6 369 ±28.9 651 ±78.3 281±31.4 2231±78.1 11808±1403 4125±162 911±64.3 8.7 ±1.12 1544 ±101 76.3 ±8.31 6.9 ±0.5

Digestate Plant C 2.91±0.13 0.75±0.09 108±14.2 542±47.1 713 ±61.2 170 ±22.0 439±19.8 3966±1064 1375±56.6 467±38.2 10.6 ±2.05 1288 ±73 54.1 ±4.86 7.6 ±0.6

Graphical Abstract

Biogas plants

Highlights



Four full-scale mesophilic biogas plants were assessed for degradation efficiency.



Biogas plants vary in the feedstock, reactor volume and process configuration.



BMP was projected as a reliable parameter to access process efficiency



Microbial community structure closely liaised to feedstock type



The BMP degradation rate aligned positively with microbial community structure.