Biomethanation efficiency of para-grass in piggery wastewater in single stage and temperature phased anaerobic systems

Biocatalysis and Agricultural Biotechnology 15 (2018) 254–263

Contents lists available at ScienceDirect

Biocatalysis and Agricultural Biotechnology journal homepage: www.elsevier.com/locate/bab

Biomethanation efficiency of para-grass in piggery wastewater in single stage and temperature phased anaerobic systems

T

Kanyarat Saritpongteerakaa,b, Sumeth Dechrugsac,d, Duangporn Kantachotee, Wen-Hsing Chenf, ⁎ Sumate Chaiprapata,g, a

Department of Civil Engineering, Faculty of Engineering, Prince of Songkla University, Hat Yai Campus, Hat Yai, Songkhla 90110, Thailand Center of Excellence on Energy Technology and Environment, Postgraduate Education and Research Development Office (PERDO), Bangkok 10400, Thailand c National Center of Excellence for Environmental and Hazardous Waste Management, Chulalongkorn University, Bangkok 10330, Thailand d CPF (Thailand) Public Company Limited, Bangkok 10120, Thailand e Department of Microbiology, Faculty of Science, Prince of Songkla University, Hat Yai Campus, Hat Yai, Songkhla 90110, Thailand f Department of Environmental Engineering, National Ilan University, Yilan 260, Taiwan g PSU Energy Systems Research Institute (PERIN), Prince of Songkla University, Songkhla, Hat Yai Campus, Songkhla 90110, Thailand b

A R T I C LE I N FO

A B S T R A C T

Keywords: Co-digestion Grass Manure Solid concentration Thermophilic Mass transfer

Effects of para-grass (PG) addition to pig manure (PM) digester were evaluated at organic loadings 0.10–3.76 gVS/L.d by different PG mixing ratios 0–8% in mesophilic single-stage (MS) and temperature-phased anaerobic digestion (TPAD, comprising T1 and M2 reactors) systems. Results showed equivalent methane production between MS and TPAD until 4%PG mix. Highest biogas yields obtained were 271.7 and 264.0 m3/tondry for MS and TPAD, respectively. Even with intense VFAs accumulation, the acidification yield in T1 was less than M2 because of continuous conversion of VFAs to CH4. Only at higher loading (8%PG), reactor staging by temperature was justified for this co-digestion as TPAD exhibited a superior performance and lesser mass transfer impediment. Para-grass addition to PM digester shifted the domination of bacterial strains whereas achaea were steady. Higher microbial diversity and some evolving hydrolytic bacteria observed in T1 could contribute to greater system stability at high solid loading. An addition of small front-end thermophilic tank should be considered when expansion of the existing MS system is planned to process higher solid.

1. Introduction Energy and environment have become the important elements in human activities. A large portion of energy we consume is non-renewable or renewable from food-derived origin, whose productions always cause negative impacts on the environment. Wastes derived energy can have a two-fold benefit where environmental clause is taken care of while producing clean energy. Waste produced from agro-industry is endlessly growing as to serve more food to our growing population. Due to its large quantity, extensive treatment or conversion is critically required prior to circulate it back to the environment. Swine farm is one of the major agro-businesses in Thailand involving over 180,000 farmers with, in year 2017, approximately 10,191,784 pigs raised generating an estimate of 5–6 billion kilograms of manure in a

year (DLD, 2017). This manure, rich in organics and nutrients, needs practical handling. Raising swine in a modern time is energy intensive due to the animal house cooling/heating and waste management. Coupling waste treatment with energy production posed clear advantages both in environmental and economic aspects. Anaerobic digestion (AD) of animal manure delivers biogas (containing 50–70% CH4) for use as fuel for electricity and heat generation while the separated solid digestate can be sold to crop farmers and the liquid part could be applied on the nearby lands for grass growing. Para-grass (Branchiria mutica) is a tropical weed that widely grows on wet soils in various tropical countries, which is now a good source of cow and horse feedstock. Over-growing of this grass in many areas can, however, be a burden for fire hazard control. There is an enormous potential for it as feedstock for co-digestion with swine waste in existing

List of abbreviations: AD, anaerobic digestion; ALK, alkalinity; CHONS, carbon, hydrogen, oxygen, nitrogen and sulfur; COD, chemical oxygen demand; CSTR, continuously stirred tank reactor; C:N, carbon to nitrogen ratio; DGGE, denaturing gradient gel electrophoresis; HAc, acetic acid; HPr, propionic acid; HRT, hydraulic retention time; HBu, butyric acid; HVa, valeric acid; i-Hbu, iso-butyric acid; i-HVa, isovaleric acid; MS, mesophilic single stage reactor; M2, mesophilic 2nd stage reactor in TPAD; OLR, organic loading rate; PG, para-grass; PM, pig manure; SVFA, short chain volatile fatty acids; TPAD, temperature-phased anaerobic digestion; TS, total solids; TVFA, total volatile fatty acid; T1, thermophilic 1st stage reactor in TPAD; VFA, volatile fatty acid; VS, volatile solids; Ya, acidification yield; Yh, hydrolysis yield ⁎ Corresponding author at: Department of Civil Engineering, Faculty of Engineering, Prince of Songkla University, Hat Yai Campus, Hat Yai, Songkhla 90110, Thailand. E-mail address: [email protected] (S. Chaiprapat). https://doi.org/10.1016/j.bcab.2018.06.021 Received 28 May 2018; Received in revised form 25 June 2018; Accepted 25 June 2018 Available online 28 June 2018 1878-8181/ © 2018 Elsevier Ltd. All rights reserved.

Biocatalysis and Agricultural Biotechnology 15 (2018) 254–263

K. Saritpongteeraka et al.

and temperature staging.

anaerobic digesters on farm. Previous investigations reported that significant increases in volumetric biogas production can be achieved by adding carbon rich agricultural residues to the AD treating animal manure (Wu et al., 2010), preferably at carbon to nitrogen ratio of 25–30 to 1 (Ward et al., 2008). Lansing et al. (2010) showed that codigesting carbon-rich cooking grease waste with swine manure in lowcost digesters could double the amount of energy production. Adding the crop residue such as sugar beet tops, grass and oat straw at 30% by weight (based on volatile solid) to a CSTR digester receiving cow manure increased methane production by 16–65% (Lehtomäki et al., 2007) while the mixture of 1:3 ratio on dry weight basis of cattle manure to pretreated corn stover gave the highest biogas yield (Li et al., 2009). Optimized C:N of the total feedstock suits metabolic activities in AD and indirectly decrease ammonia inhibition (Xie et al., 2011). Characteristics of the co-substrate in both physical and chemical composition play a vital role on how it can be mixed and to what limit for a particular reactor configuration. Among factors important to successful AD operation such as pH, mixing, pretreatments, and organic loading, temperature is proven to have a great impact on solid digestion. These well-established figures of 35 and 55 °C are typical and proven optimal temperatures for major groups of microorganisms; mesophiles and thermophiles, in anaerobic digestion (Li and Khanal, 2016). Higher temperature increases substrate degradation, and biogas production rate could rise substantially (Yu et al., 2002). However, operating AD at thermophilic temperature is challenging since the system becomes quite sensitive at high temperatures, not to mention a greater energy requirement (Speece, 2008). For conventional anaerobic digestion, acid forming and methane forming organisms reside together in a single reactor. Both groups differ in physiology, growth kinetic, and nutritional and environmental condition requirements (Elbeshbishy et al., 2012). Two-stage reactor configuration is believed to offer a physical separation of acid and methane forming, and optimum environmental condition of each group. Temperature-phased anaerobic digestion (TPAD) is a two-stage AD process that aims to enhance hydrolysis and pathogen destruction in the first stage at thermophilic (55 °C) condition while the second mesophilic stage (35 °C) efficiently polishes a more soluble and digestible pretreated substrate. These combined with its unique microbial diversity could give a remarkable process stability (Montañés Alonso et al., 2016; Riau et al., 2010; Wu et al., 2015). Nonetheless operating two reactors, especially one at a thermophilic temperature poses more difficulty beside the energy expenditure. Simple operation and configuration of the single stage digester system may come in to play when the performance and benefit of the two-stage are not sufficient to overcome the sacrifice it will take. This depends largely on the substrate variety and operating regimen. Organic loading rate (OLR), representing the amount of waste treatable per unit reactor volume, is one critical operational parameter regulated either by feedstock concentration and hydraulic retention time (HRT). While increasing OLR could increase the volumetric methane production, a risk of system failure due to the fast evolution of VFAs or microbial washout of AD if hydraulic retention time is shortened (Xie et al., 2012). Increasing feed concentration seems to be a better way to raise the OLR. It is, in retrospect, up to the limitation of each type of system configuration and operations to handle the more viscose feedstock. Therefore, the objective of the study was to evaluate the effects of substrate concentration which was used as a means to increase OLR on co-digestion of para-grass in piggery wastewater. Comparison between the single stage mesophilic anaerobic digester and the two-stage TPAD system was carried out in continuous operation over 330 days in order to determine the biomethanation performance, which eventually helped identify the necessity and the operating regime that the temperature phased two stage system could be more effective than the single stage one. Microbial analysis was also conducted to explain the shift of community and dominance of different bacteria and archaea in both systems regarding the feed alteration to lignocellulosic substrate

2. Materials and methods 2.1. Inoculum The inoculum used in this work was from anaerobic digester treating piggery wastewater from the unit of finishing (fattening) barn in Songkhla Province, Thailand. The inoculant sludge, which was dispersive, was sieved to remove large particulate impurity. It was then measured for total solids (TS) and volatile solid (VS) concentration. Its specific methanogenic activity (SMA) was measured to ensure active microbial inoculant for the system startup. The methanogenic activity found was 33.0 ± 0.2 mLCH4/gVS which showed high concentration of active methanogenic organisms. The sludge was then inoculated to the reactors within 72 h after field collection. 2.2. Substrates Pig manure (PM) was obtained from excretions in a finishing unit of a pig farm in Pattalung Province, Thailand. It was dried at 60 °C and ground in mortar. Fresh green para-grass (PG), Branchiria mutica, was randomly harvested from the field receiving liquid digestate from the pig farm where PM was collected. It was chopped with a cutting machine to approximately 2 cm, shredded to size of less than 6 mm and then dried at 60 °C until constant weight. Both substrates were kept at 4 °C in plastic bags until use. Each substrate was homogenized and characterized for total solid (TS), volatile solid (VS), moisture, fiber contents and elemental composition (CHONS). Analytical methods of samples are described in Section 2.4. The prepared PM was taken to mix with tap water at concentration around 2.5 gTS/L or 2.0 gVS/L to imitate the piggery wastewater that contained a COD of approximately 3000–4000 mg/L. The introduction of PG as co-substrate to the prepared liquid feed were tested at 4 levels 0%, 2%, 4% and 8%TS (w/w) designated as operating conditions 0% PG, 2%PG, 4%PG, and 8%PG, in order. The substrates were homogenized to slurry before feeding to the reactors. Characteristics of PM and PG used in the experiment are shown in Table 1. 2.3. Reactor systems and operations There were two anaerobic digestion systems evaluated; single-stage mesophilic anaerobic digester (MS) system, and two-stage temperaturephased anaerobic digestion (TPAD) system which consisted of first stage thermophilic reactor (T1) followed by second stage mesophilic digester (M2). Both mesophilic reactor vessels were made of cylindrical glass at 5 L volume and 182 mm in diameter, while the thermophilic glass reactor has a total volume of 1 L. Each system possessed a total effective volume of 4 L, that TPAD system is divided into 0.4 L in T1 and 3.6 L in M2. The reactors were inoculated with an active sludge at 30% of the effective volume yielding an initial sludge concentration of 2.5 Table 1 Chemical and elemental compositions of pig manure and para-grass.

255

Composition

Unit

Pig manure (PM)

Para-grass (PG)

Total solids, TS Volatile solids, VS Cellulose Hemi-cellulose Lignin Carbon, C Hydrogen, H Oxygen, O Nitrogen, N Sulfur, S C:N ratio

g/kg wet wt g/kg dry wt % dry wt % dry wt % dry wt % dry wt % dry wt % dry wt % dry wt % dry wt

248.1 ± 4.5 806.3 ± 3.0 12.7 ± 0.2 25.7 ± 3.9 9.0 ± 5.0 38.1 5.4 22.3 3.0 0.3 12.7

190.6 ± 6.4 914.9 ± 6.2 38.8 ± 2.4 29.5 ± 1.5 8.0 ± 2.0 41.6 5.3 27.3 1.3 0.3 32.2

Biocatalysis and Agricultural Biotechnology 15 (2018) 254–263

K. Saritpongteeraka et al.

defined in terms of hydrolysis yield (Yh), the degree of organic matter solubilization to hydrophilic soluble compounds by hydrolytic bacteria (Eq. (1)). In the next step, acidogenesis efficiency is defined with the same analogy as acidification yield (Ya) that represents the portion of the initial substrate concentration (TCODi) converted to VFAs (Eq. (2)) (El-Mashad et al., 2004). The quantity of each VFA species was presented as COD unit. The COD equivalents of each VFA are 1.066, 1.512, 1.816, 2.036 and 2.204 for acetic, propionic, butyric, valeric and caproic, respectively (Demirel and Yenigun, 2004). COD equivalence of CH4 is added to the nominator since it is considered the reactions’ product.

Table 2 Operating parameters of digester systems in the study. Operational parameters

Reactor effective volume (L) Temperature (oC) HRT (days) VS loading rate (gVS/L.d)

MS*

OLR1 OLR2 OLR3 OLR4

4 35 20 0.10 1.02 1.93 3.76

TPAD T1*

M2*

Overall

0.8 55 4 0.50 5.08 9.65 18.80

3.2 35 16 ** ** ** **

4 20 0.10 1.02 1.93 3.76

Yh =

*MS = mesophilic single stage reactor. *T1 = thermophilic 1st stage reactor. *M2 = mesophilic 2nd stage reactor. **Depends on solid degradation in T1.

Ya =

gTS/L or 2.02 gVS/L, then filled up with the prepared piggery wastewater to an effective volume. The reactors were then capped with a rubber seal having one port to allow biogas collection in a gas bag. Three ports on the sidewall were assigned to influent feeding, effluent withdrawal, and sludge sampling/drainage. The reactors were rested for 24 h prior to the scheduled feeding. All were kept in either 35 °C or 55 °C temperature control water bath. The reactor content was mixed thoroughly twice a day each for approximately 2 min. MS and T1 were fed with the slurry of PM wastewater and PG mixture according to plan until they entered stable operation, i.e. variation of biogas production under 10% with stable pH. The mixed liquor from T1 was withdrawn immediately after mixing and used to feed M2. In MS and M2, effluent withdrawal at mid-depth of the operating level took place after the reactors were well mixed. The organic loading rate (OLR) of the systems was increased stepwise from 0.10 to 1.02, 1.93, and 3.76 gVS/L.d, in order which corresponded to PG mixing of 0%, 2%, 4%, and 8%PG in feed based on TS concentration. Operating parameters of the systems are listed in Table 2.

(CODCH 4 + SCODeffluent ) CODinfluent

× 100

(CODCH 4 + SCODeffluent VFA) CODinfluent

(1)

× 100

(2)

where Yh is hydrolysis yield (%), Ya is acidification yield (%), CODCH4 is COD equivalent of methane at standard temperature and pressure (0 °C, 1 atm) = 0.35 gCOD/LCH4 (g), SCODeffluent is soluble COD of effluent, CODinfluent is theoretical total COD (g) of the substrate, CODeffluent VFA is the summation of COD equivalence from all VFA species. The initial COD of the mixed substrate was calculated from the stoichiometric equations using C, H, O, N, S at different solid loadings (0%, 2%, 4%, and 8%PG) according to Dechrugsa et al. (2013). The COD equivalents of the PG mixture feed stock in the prepared piggery wastewater were calculated based in the equilibrium stoichiometric equations which are 1.69, 1.64, 1.63, and 1.63 gCOD/g substrate dry wt. at 0%, 2%, 4%, and 8%, respectively. 2.6. Microbial community analysis by denaturing gradient gel electrophoresis (DGGE) In order to identify the biological impact of lignocellulosic substrate introduction to existing anaerobic digesters, microbial communities under 0% PG and 4% PG reactors were investigated by collecting sludge samples from the reactors at stable condition of the systems. A slightly modified standard bacterial genomic DNA isolation method as described by Hniman et al. (2011) was adopted to extract the samples for obtaining total genomic DNA. The bacterial 16S rRNA genes were amplified by the first polymerase chain reaction (PCR) with universal primer 1492r and 27f (Table 3). To amplify the fragment of V3 region of 16S rRNA products from the first PCR, the primers K517r and L340f with GC clamp were also used. Sterilized blades were used to excise the DNA bands from the gel for re-amplification with primer 357f (without a GC clamp) and the reverse primer. The PCR products were purified and then sequencing with primer 518r for bacteria using the Macrogen sequencing facility (Macrogen Inc., Seoul, Korea). Identification of closest matches for partial 16S rRNA gene sequences was analyzed by ribosomal database project with SeqMatch program and basic local alignment search tool (BLAST) with nucleotide database in National

2.4. Analytical methods Inoculum and substrates (PM+PG) were analyzed for TS and VS according to the Standard Methods for the Examination of Water and Wastewater (APHA, 2012). The chemical compositions of the samples were analyzed using CHNS-O Analyzer, CE Instruments Flash EA 1112 Series, Thermo Quest, Italy with Dynamic Flash Combustion Technique. Performance of the digesters was evaluated by the determination of pH, COD, and volatile solid (VS) of the influent and effluent following Standard Methods for the Examination of Water and Wastewater (APHA, 2012) while available alkalinity and VFA were determined by direct titration method (DiLallo and Albertson, 1961). For VFA species, samples were filtered through a 0.22 µm Teflon paper and the filtrate was then analyzed by gas chromatography (GC 7820A Agilent Technologies) equipped with a flame ionization detector (FID). A capillary column Agilent 19091N-133 HP- Inowax 30 m × 0.25 mm Internal diameter was used with helium as the carrier gas. The biogas produced was stored in a gas bag and the volume was measured using a multi-chamber rotor wet gas meter (Ritter), Model TG0/5. Biogas composition was analyzed about twice a week by gas chromatography (GC 7820A Agilent technologies) equipped with thermal conductivity detectors (TCD) with a stainless steel packed column SS Hayesep Q80/100 (6 m × 1/8 in.). Helium was used as carrier gas to determine the methane composition. The standard calibration curve was made with pure CH4, CO2, and N2 gases, and verified with a standard gas mixture of 5% N2, 60% CH4, and 35% CO2.

Table 3 Primers used for PCR-DGGE in this study. Group

Primer name

Nucleotide sequences (5′- 3′)

Bacteria

1492r 27f K517r L340f L340f-GC PRA46F PREA1100R PARCH340F PARCH519R GC clamp

GAAAGGAGGTGATCCAGCC GAGTTTGATCCTTGGCTCAG ATTACCGCGCTGCTGG CCTACGGGAGGCAGCAG GC clamp-CCTACGGGAGGCAGCAG C/TTAAGCCATGCG/AAGT T/CGGGTCTCGCTCGTTG/ACC CCCTACGGGGC/TGCAG/CCAG TTACCGCGGCG/TGCTG CGCCCGCCGCGCCCCGCGCCCGTCCCG CCGCCCCCGCCCG

Achaea

2.5. Hydrolysis yield and acidogenesis yield evaluation Effectiveness of the first step in anaerobic digestion partway is 256

Biocatalysis and Agricultural Biotechnology 15 (2018) 254–263

K. Saritpongteeraka et al.

solid loading would not have as severe an impact on the AD acidity balance. On the other hand, fast degradable biomass such as food waste (Bong et al., 2018; Bouallagui et al., 2005) and banana peel (Odedina et al., 2017) could acidify the digester quickly from fast hydrolysis and acidification.

Center for Biotechnology Information (NCBI). On the other hand, the forward primer PRA46F and the reverse primer PREA1100R (Table 3) were used for the PCR amplification targeting archaea to generate a product of 1072 bp. Afterwards, these PCR products were then used as templates for the PCR amplification of 179 bp using the forward primer PARCH340F containing a GC clamp and the reverse primer PARCH519R. For purification PCR products and sequencing, primers PARCH340F and PARCH519R were used by following the methods of Ovreås et al. (1997).

3.2. Organic acid composition Composition of volatile fatty acids in these reactors varied greatly due to the temperature and OLR staging. Distribution of the short chain volatile fatty acids (SVFA, C2-C5) was detected from the effluents of each reactor over time, as presented in Table 4. SVFAs were not detected at low OLR of 0.10 gVS/L.d in MS as they were completely converted to biogas as an end product. At OLR 1.02 and 1.93 gVS/L.d, the dominant SVFA species in MS were the valeric acid (i-HVa+HVa) accounting for approximately 35.5% and 49.9% of total SVFA, respectively. The lesser species in order were iso-butyric and butyric acid (iHbu+HBu), acetic acid (HAc), and propionic acid (HPr) at all tested OLR’s. In opposite direction, less SVFAs were produced when OLR was raised to 3.76 gVS/L.d indicating the decreased hydrolysis and acidification at this state. This observation will be discussed in terms of hydrolysis and acidification yields in Section 3.3. In the thermophilic condition at short HRT (T1 reactor), the obvious dominant VFA species was HAc followed by i-Hbu+HBu, i-HVa+HVa, and HPr in all OLRs operated. HAc (C2) composed of approximately 54–62% of total SVFA and greatly increased at higher OLR. This high total SVFA in T1 led to the lowering pH and subsequently a sharp drop in methane production. It was interesting to observe no SVFA products in M2 at 0% and 2%PG. Only at 4% and 8%PG feed could induce some SVFA remain in M2, but still lower than in MS reactor. M2 was evidently quite effective in utilizing SVFAs from the pre-hydrolyzed substrate of T1. This implies very active acetoclastic methanogens in M2 reactor.

2.7. Statistical analysis The values reported were calculated from data at stable conditions of each OLR. Mean with standard deviation were calculated and compared to determine the effect of key parameters in the performed experiment. One-way analysis of variance (ANOVA) and Scheffe’s multiple-range test were employed to differentiate the means using SPSS software version 11.0. 3. Results and discussion 3.1. pH and organic acids formation The level of pH in the anaerobic reactor is largely resulted from the balance between TVFA (total VFA) and alkalinity. These parameters are interconnected and important in anaerobic digester as they dictate the buffer balance in the system. pH’s of influent and effluent from the single stage mesophilic reactor (MS) and TPAD reactors across all OLR tested are shown in Fig. 1. The average pH of slurry influent was 6.8, 6.5, 5.8 and 5.5 at 0%, 2%, 4% and 8%PG mixture feed, respectively. This lowering pH is a result of acid accumulated from starch and sugar degradation in the grass biomass that could have developed during 2–3 days between field harvest to complete drying in the laboratory. This kind of inherited organic acidity would be biodegraded in the digester. In the first stage thermophilic (T1) of TPAD, pH decreased with increasing OLR (Fig. 1a), which corresponded to the increased TVFA production (Fig. 1b). The average pH values of T1 digester, represented by that of the effluent, at stable condition were 6.3, 5.8, 5.6 and 5.4 at 0%, 2%, 4% and 8%PG mixtures, respectively. These pH shifted toward the optimum range for anaerobic hydrolysis reaction (pH 4–6) (Orozco et al., 2013). Although alkalinity of T1 increased at the elevated OLR probably due to the release of protein degradation to ammonia (Deublein and Steinhauser, 2008), TVFA also increased, pushing TVFA/ alkalinity to 1.0–1.1. This range is not suitable for methanogenic activity, evidenced by the minimal methane production, which will be discussed in Section 3.4. In mesophilic reactors (MS and M2), the pH values varied only slightly between 6.5 and 6.8. The solution 0.1 N NaOH was twice applied to the mesophilic reactors only during early stage of 0%PG feed when pH dropped below 6.5. With such pH adjustment, an obvious drop in VFA concentration was clearly observed around day 85, signaling a successful acclimatization of the anaerobic sludge. At higher PG loadings, the alkalinity increased (data not shown), making the ratio of TVFA/ALK stay in a range of 0.15–0.18 despite more acidity by the higher solid in feed slurry. This result shows that both systems had high buffer capacity (TVFA/ALK below 0.4) (Song et al., 2004). High protein content of the para-grass for as high as 4.3–5.1% played an important role in buffering the acidity in the reactor (Hare et al., 1999). Release of bicarbonate alkalinity (HCO3-) from protein degradation taking place in the methanogenic stage of anaerobic condition is beneficial. Although at OLR at 3.76 gVS/L.d, the system was still operational for MS (20 day HRT), the digester content of MS was so thick that mixing became difficult at times, indicating a physical mass transfer limit of this AD operation. This viscosity was to some extent less in T1 and M2 at this loading. It is interesting to note that for the slowly degradable cellulosic biomass such as grass, high

3.3. Degree of hydrolysis and acidification in single and two-stage systems 3.3.1. Hydrolysis yield The levels of hydrolysis and acidogenesis presented in percentage of the initial COD of substrate were calculated based on Eqs. (1) and (2). The efficiency of these two crucial steps in anaerobic digester was affected by temperature phase separation and solid loading. Theoretically, hydrolysis represents the break down from particulate to soluble form in reactors. The hydrolysis yield (Yh) in T1 reactor increased by 2.4% from 8.3% to 10.7% when PG mix in feed changed from 0% to 4% (Table 5) and stayed at the same level until 8%PG mix. This coincided with the expanding layer of scum in the reactor at 8%PG which separated floating substrate from the microbes in liquid. The overall Yh of MS and TPAD were quite comparable at each of the OLR tested (Table 5). It was at the last OLR of 8%PG that Yh dropped dramatically in both systems resulted from the inability of mesophilic digestion to effectively hydrolyze in comparison to the incoming substrate. Yh obtained from this study is also comparable to that from a two-stage anaerobic digestion of sunflower oil cake at OLRs 7 gVS/L.d (Yh 20.5–30.1%) (De La Rubia et al., 2009), but higher than grass silage in leaching bed reactors, which ferment in a relatively dry environment (Yh 16.5–22.4%) (Xie et al., 2012). Access of enzyme to the biomass is key to the degree of hydrolysis which is influenced by operation mode of AD. Nature of feedstock tissues can also contribute to the hydrolysis as the soft and highly biodegradable fruit and vegetable wastes posed Yh of 81% in wet AD (Bouallagui et al., 2004). Some pretreatment methods such as thermal, chemical, thermo-chemical, and ultrasonic can improve the poor liquefaction of dry substrates. However, it can only make sense when the costs and complexity of such pretreatment worth the benefit gained from the increased biogas production over a simple AD system operation. Careful and thorough evaluations of operation and maintenance must be taken to 257

Biocatalysis and Agricultural Biotechnology 15 (2018) 254–263

K. Saritpongteeraka et al.

Fig. 1. (a) pH and (b) total organic acids accumulation in mesophilic single stage (MS) and TPAD systems digesting piggery waste and grass at different feed concentrations.

et al., 2012). Although TVFA is higher in T1 (Fig. 1b), its Ya is small compared to the other mesophilic reactors in this study due to the low value of nominator in Eq. (2) attributing to the end products being void of CH4. It also indicates that acidogenesis was more active in M2 in the presence of methanogenesis which helped lower the acid products, enabling acidogenesis reaction to proceed. Syntrophic bacteria (also known as acetogens) are as important to the AD culture where large fatty acids must be converted to acetic acid for acetoclastic methanogens. In an effective AD system, presence of these three groups; namely acidogens, acetogens, and methanogens, must exist and work hand-inhand to propel the biomethanation of the organics. Apparent lack of methanogens, T1 could still benefit the TPAD system by mainly hydrolysis by which the accumulated organic acids aided such reactions.

consideration. 3.3.2. Acidification yield Acidogenesis yield (Ya) measures the degree of success in fermentation as the solubilized substrate is converted to organic acids. Ya of MS and M2 reactors were higher when feed concentration increased until OLR 1.931 gVS/L.d at 4%PG as shown in Table 5. When hydrolysis is elevated, the solubilized substrate becomes available at a greater rate to the acidogenic bacteria which typically possess higher reaction kinetic than the other groups in AD. Especially for solid substrate where hydrolysis is a rate limiting step, acidogenesis was then limited by the preceding hydrolysis reaction, hence mathematically, Ya is always less than Yh (Eqs. (1) and (2)). The best Ya in MS and overall TPAD system were 26.6% and 27.4%, respectively, which were achieved at OLR 1.93 gVS/L.d (4%PG). Such Ya values are lower than those of grass silage fermentation in a dry and high acidity environment of LBRs with the highest Ya of 57–60% (Xie 258

Biocatalysis and Agricultural Biotechnology 15 (2018) 254–263

K. Saritpongteeraka et al.

scum forming caused by the higher grass float to the liquid surface in the reactor, particularly MS, giving poorer mass transfer in the reactor. Propensity of fibrous crops to float and form indigestible crust layer has been well documented (Thamsiriroj and Murphy, 2010) depending on substrate characteristics as well as reactor configuration and operation. Koch et al. (2009) suggested that the feed concentration limit as high as 12%TS could still operational in order to obtain satisfying biogas yield from grass silage by special designed loop AD. Nevertheless, scum layer could form even at lower feed concentration where mixing is not adequately provided. It is noteworthy that the scum layer in our two stage TPAD system was obviously less than MS system. The intense hydrolysis and temperature in T1 could have softened the fiber and helped gain liquid access to some gas-filled spaces in the grass biomass structure. A better mixing scheme is so essential in silage digestion to counter their tendency to float.

Table 4 Short chain volatile fatty acids (SVFA) composition of mesophilic single stage (MS) and TPAD systems receiving piggery wastewater and grass at different feed concentrations. Reactor

OLR (gVS/ L.d)

%PG

HAc (mg/L)

HPr (mg/L)

i-Hbu +HBu (mg/L)

i-Hva +Hva (mg/L)

Total SVFA (mg/L)

MS

0.10 1.02 1.93 3.76 0.10 1.02 1.93 3.76 0.10 1.02 1.93 3.76

0% 2% 4% 8% 0% 2% 4% 8% 0% 2% 4% 8%

nd. 17.37 8.84 14.83 27.51 139.72 239.14 234.29 nd. nd. 7.75 10.40

nd. 13.58 12.90 12.45 13.30 20.01 26.14 20.28 nd. nd. 12.78 13.61

nd. 23.59 19.73 20.29 24.96 51.57 77.60 110.77 nd. nd. 18.58 18.94

nd. 30.00 41.27 nd. 29.94 44.34 46.78 44.52 nd. nd. nd. nd.

nd. 84.54 82.73 47.57 95.73 255.63 389.66 409.87 nd. nd. 39.11 42.94

T1

M2

3.4.2. Methane formation Methane content of all reactors fluctuated during start up (Fig. 2) due to the adaptation of methanogenic organisms in the systems. It varied between 43% and 55% in MS and M2 reactors while there was always lower than 30% in T1. Acclimatization of the methanogens in inoculant from ambient temperature digester to the thermophilic temperature would take longer time and is always sensitive to operational parameters (Gebreeyessus and Jenicek, 2016). Because T1 was aimed to serve mainly as a hydrolysis tank, it was decided that OLR be increased as stable operation was reached based mainly on the M2 performance. Methane in both mesophilic reactors varied from 40% to 50% at 0%PG. Increasing OLR with grass mix was also found to promote higher methane content of the biogas in the two mesophilic reactors to reach the stable level of 53.3% and 54.7% at 2%PG for MS and M2 reactors, respectively. A more balance nutrients to the carbon in feed helped enhance the methanogenic activity while probably abating an adverse effect of nitrogen inhibition. T1 reactor obviously offered lower methane content in its biogas compared to the mesophilic ones. It took approximately 100 days to reach stable methane content around 25% on average and stayed in such level during overall system OLR 1.02 gVS/L.d (2%PG), which

Note nd. = not detectable. T1 = thermophilic 1st stage reactor in TPAD. M2 = mesophilic 2nd stage reactor in TPAD.

3.4. Effects of reactor staging on biomethanation 3.4.1. Biogas production One of the most important parameters in successful anaerobic digestion is the quantity and CH4 content of the produced biogas. Biogas production and methane content from MS and TPAD system at different OLR plotted over time are depicted in Fig. 2. Biogas production derived at stable each OLR were 103.0, 1003.7, 2396.6 and 2371.1 mL/d for MS, and 119.7, 1024.1, 2324.9 and 2682.4 mL for TPAD at 0%, 2%, 4%, and 8%PG, respectively. It was obvious that when switching to 8%PG loading, the biogas production rose sharply at day 272 followed by dramatic drop by almost 80% within the next 10 days in MS and never seemed to get to stable operation thereafter. This fluctuation was dampened in the TPAD system as its biogas production returned to stability around day 295 (13 days after OLR jump). There were frequent

Table 5 Summary of performance at stable condition of mesophilic single stage (MS) and TPAD systems receiving piggery wastewater and grass at different feed concentrations. Organic loading rate (gVS/L.d)

Hydrolysis yields, Yh (%) MS TPAD (overall) –T1 –M2 Acidification yields, Ya (%) MS TPAD (overall) –T1 –M2 Theoretical methane yields (mLCH4/gVS) Specific methane yields (mLCH4/gVSadded) MS TPAD (overall) –T1 –M2 Biodegradability (%) MS TPAD (overall) –T1 –M2

0.101 (0%PG)

1.016 (2%PG)

1.931 (4%PG)

25.2 ± 2.6 25.9 ± 1.7 8.3 ± 4.5 17.6 ± 3.5

28.3 ± 4.6 26.6 ± 5.3 9.3 ± 1.3 17.3 ± 4.1

30.0 29.8 10.7 18.4

0.8 1.2 1.2 0.4

16.6 ± 1.1 18.6 ± 1.3 10.7 ± 0.9 7.9 ± 0.4

18.9 ± 1.1 18.4 ± 1.5 6.4 ± 0.4 12.0 ± 1.3 525

23.4 ± 3.3 23.3 ± 4.0 1.4 ± 0.1 21.9 ± 3.9 541

26.6 ± 1.6 27.4 ± 1.1 0.4 ± 0.02 27.0 ± 1.1 542

15.3 ± 3.7 16.4 ± 0.1 0.5 ± 0.1 15.8 ± 0.2 542

119.7 ± 3.3b 117.1 ± 9.4a,b 6.9 ± 3.8c 110.2 ± 5.9a,b

138.2 ± 23.8b 142.5 ± 24.0b,c 1.8 ± 0.1a,b 140.7 ± 24.1b,c

158.3 ± 5.8c 158.6 ± 3.3c 2.7 ± 0.1a,b 155.9 ± 3.2c

83.3 ± 4.7a 94.1 ± 7.8a 0.5 ± 0.4a 93.6 ± 8.3a

22.8 ± 0.6 22.3 ± 1.8 1.3 ± 0.7 21.0 ± 1.1

25.6 ± 4.4 26.3 ± 4.4 0.3 ± 0.02 26.0 ± 4.5

29.2 ± 1.1 29.3 ± 0.6 0.5 ± 0.03 28.8 ± 0.6

15.4 ± 0.9 17.4 ± 1.4 0.1 ± 0.1 17.3 ± 1.5

± ± ± ±

3.761 (8%PG)

Note: Means in each row followed by a different letter are significantly different using Scheffe’s multiple-range test (p < 0.05). MS = mesophilic single stage reactor, TPAD = temperature phased anaerobic digestion, T1 = thermophilic 1st stage reactor, M2 = mesophilic 2nd stage reactor. 259

Biocatalysis and Agricultural Biotechnology 15 (2018) 254–263

K. Saritpongteeraka et al.

Fig. 2. (a) biogas production and (b) methane content from mesophilic single stage and TPAD systems receiving piggery wastewater and grass at different feed concentrations.

concentration shifted from 4% to 8%PG (Fig. 2a). This, in turn, severely dropped the biogas yield down even though the methane content (%) was quite stable (Fig. 2b). The best biogas yield per fresh mass was found at 4%PG at 51.8 and 50.6 m3/tonfresh added for MS and TPAD systems, respectively, with almost identical methane yield of 27.6 and 27.7 m3/tonfresh add accordingly. These calculations are based on the grass TS of 190.6 g/kgfresh (Table 1), CH4 53.3% for MS system, and CH4 54.7% for TPAD system which came solely on M2 since methane yield from T1 is neglectable. These values are useful for financial analysis and project design as the harvest and preparation costs of grass must be well-adjusted with its benefit from the incremental CH4 received. Nevertheless, it must be noted that these values are based on the moisture content of around 81% that can be skewed heavily in technical communication. For instance, the biogas yield in terms of m3/tonfresh can be doubled if one claims the grass moisture content was reduced by field drying to 40–45%. Appropriate pricing of grass feedstock to the biogas plant must be set against dry weight basis and should be the standard means of communication. Thus, biogas yield of grass digestion is 271.7 m3/tondry for MS system and 264.0 m3/tondry for TPAD system at methane content 54% can be used as a reference yield for grass in piggery wastewater co-digestion. In this aspect, mesophilic single stage

corresponds to T1 OLR of 5.08 gVS/L.d (Table 2). Methane content in T1 had declined to the level below 20% at system OLR 1.93 gVS/L.d (4%PG) and later dropped to around 2% at OLR 3.76 gVS/L.d (8%PG), which exhibited T1 OLR of 18.80 gVS/L.d. At 8%PG stage, more biomass was pre hydrolyzed by microbial enzymes and higher soluble organics derived were subsequently acidified, creating VFAs at inhibitory level to methanogens (Table 4). The activities of methanogenic archaea ceased to almost zero by this toxicity caused by the high organic input. This T1 reactor was transformed into fully hydrolysis tank that produced insignificant biogas composed of low methane content. In theory, hydrogen (H2), an intermediate product from acidogenesis, should start to accumulate in T1 but its composition was not measured due to the small overall contribution of energy derived from such a small volume of gas produced. As for MS, since methane content was rather stable at 8%PG feed, the limitation was deemed mostly from mass transfer and increasing heterogeneity within the digester volume, not a biochemical limitation. This explanation is supported by the stability in acidity and pH. 3.4.3. Biogas yield assessment Only slightly more biogas was produced when the feed 260

Biocatalysis and Agricultural Biotechnology 15 (2018) 254–263

K. Saritpongteeraka et al.

Fig. 3. Bacterial community profiles at feed concentration of 0% and 4%PG from mesophilic single stage reactor (MS), thermophilic first stage reactor (T1) of TPAD, and mesophilic second stage reactor (M2) of TPAD.

(0% PG). Detection of Clostridium sp. MF18_Ns and Clostridium stercorarium DSM 8532 in T1 reactor with 4% PG feed suggested that they might have affected on the higher methane yield; particularly the latter organism is an acetate-producing thermophilic bacterium to degrade polysaccharides in plant tissue and uniquely possess a multitude of genes for hydrolytic enzymes production (Poehlein et al., 2013). These syntrophs are also thermodynamically critical in bridging the gap between acidogens and methanogens by cleaving large VFA molecules to acetate and hydrogen, which enables both the exergonic acetoclastic and hydrogenotrophic methanogenesis reactions. In general, acetoclastic methanogenic archaea use acetic acid for producing methane. As in thermophilic reactors at both 0% and 4% PG feed observed only hydrogenotrophic methanogenic archaea (Fig. 4); this led to accumulate acetate in the reactor with 4% PG feed (Table 4). The shift to higher diversity of bacteria when applying a more complex

digester is deemed suitable for grass digestion with piggery wastewater at 4%PG mixture with less energy intensive and simpler operation of a single tank.

3.5. Microbial communities DGGE technique was used to investigate diversity of microbial communities in every reactors at 0% and 4% PG (OLR 0.10 and 1.93 gVS/L.d); and diversities of bacteria and archaea are shown in Figs. 3 and 4. The microbial community found the differences of both of bacteria and archaea. As all reactors observed clostridia, this was due to they are normally bacteria found in anaerobic system as syntrophic acetate producers (Lin et al., 2013; Pampillón-González et al., 2017). In TPAD system, supplementation with grass to the feed (4% PG) produced higher diversity of bacterial species compared to only pig manure

Fig. 4. Archaeal community profile at feed concentration of 0% and 4%PG from mesophilic single stage reactor (MS), thermophilic first stage reactor (T1) of TPAD, and mesophilic second stage reactor (M2) of TPAD. 261

Biocatalysis and Agricultural Biotechnology 15 (2018) 254–263

K. Saritpongteeraka et al.

second stage in the staged configuration and the versatility in the one tank model as they majorly consume the intermediate compounds (acetate, H2 and CO2). Although the biodiversity of the thermophilic reactor is lower with the temperature increase acting as a selective pressure was applied (Gagliano et al., 2015), the overall combined diversity of the dominant specialized microorganisms in whole TPAD system inherited from the two distinct temperature zones, is higher in comparison to the single stage reactor. The emergence of some evolving hydrolytic bacteria within the TPAD system would help achieve greater AD system stability when high solid substrate is introduced as evidenced in the performance of the reactor systems at 8%PG.

substrate (4% PG mixed feed) was also found since it is even more necessary synergistically for the consortia to work together to degrade lignocellulosic substrate. Due to only one uncultured bacterium observed in pure PM feed; while the separated bands of Thermoanaerobacterium sp. enrichment culture clone D5 and Thermophilic anaerobic bacterium K1L1 including four uncultured bacterium groups emerged in 4% PG feed (Fig. 3). In T1 with the existence of diverse species showed a significant role in the hydrolysis step and subsequent production of VFAs; this led to allow a more stable operating of the following M2 to produce biogas. These phenomena did not cause a superior performance yet over the single stage mesophilic digester under 4% PG mixed feed condition, but clearly expressed at 8% PG graphically in Fig. 2, and numerically 83.3 vs 94.1 mLCH4/gVSadd at 4% and 8% PG, respectively, in Table 5. The dominant species found in the MS reactor were much different between 0% and 4% PG feed. For example, dominant genera at 0% PG were Rhodobacteraceae sp., Acinetobacter sp., Halanaerobium sp. and Ruminococcaceae sp. In contrast, at 4% PG feed condition were dominant by Pseudomonas sp., Roseburia sp., and Lachnospiraceae bacterium. For 0% PG had Halanaerobium sp.; while 4% PG feed had Roseburia sp. as the good candidate for producing acetate and butyrate, respectively (Insam et al., 2010). Several researchers reported that some members of Pseudomonas sp. are able to produce endoglucanase, exoglucanase and β-glucosidase, and xylanase, which were very effective in hydrolyzing agricultural wastes such as bagasse (Cheng and Chang, 2011; Singh et al., 2015). One of normal flora found in rumen is Lachnospiraceae bacterium and its metabolites such as acetate, lactate, H2 and CO2 (Madigan et al., 1997; Pampillón-González et al., 2017) are substrates for methane formation. With the changes of substrate clearly showed the shift of dominance in the cultures; however, the presence of many groups with lesser number as seen in light bands still found in the gels in Fig. 3. As the common substrate of pure pig manure; thereby, provided identical microbial community makeup between the MS and M2 at 0% PG feed. Consequently, the bacteria exposed to liquid compositions were rather similar as evidence of VFAs concentration in Table 4. As no detection of each VFA species pointed out that the processes had high efficiency to consume VFAs at this relatively low organic loading (0–4% PG). However, members of bacterial community in M2 experienced a drastic shift from 0% to 4% PG (Fig. 3). Addition of grass found the following bacteria; Flavobacterium sp. 01WB03.1–18, Pseudomonas sp. R-45822, SRB bacterium enrichment culture clone SRtB-otu1-52 and Microbacterium aerolatum KUDC1073. Hence, the different bacterial community as supplementation of 4% PG caused changes to the characteristics of fluid (Table 4). The results demonstrated that those mentioned dominant bacteria were able to hydrolyze the lignocellulose and produce the higher organic acid concentrations. There were four groups of archaea found in mesophilic reactors in this study. However, in each reactor members of the archaea was not much different between 0% and 4% PG. As the predominant methanogens in thermophilic reactors were Methanobacterium formicicum and Methanocella conradii. Both genera are the hydrogenotrophic methanogens that consume hydrogen to form methane and able to withstand the low pH condition. High acetate concentration at this stage in the thermophilic reactor (Table 4) with the low biogas and methane production (Fig. 4) was observed; this was due to the absence of acetoclastic methanogens. T1 reactor was entirely producing methane from H2 and CO2, in contrary to a single tank AD as normally uses acetoclastic pathway for producing 72% methane the rest using hydrogenotrophic pathway (Khanal, 2008; Walter et al., 2016; Zhang et al., 2016). Instead, dominant archaea found in mesophilic reactors at both 0% and 4% PG for the acetoclastic methanogens were Methanomethylovorans sp., uncultured Methanosaeta sp. and uncultured Methanolinea sp.; while for the hydrogenotrophic was Methanobacterium formicicum. This gives explanation the roles of methanogenic archaeal community composition on the effective consumption of the intermediates in the

4. Conclusions Equivalent performance of para-grass digestion in piggery wastewater was achieved between single-stage mesophilic (MS) and TPAD systems under mixture of 4% grass on dry weight basis. Methane production improved with increasing para-grass mixture until 4%PG. At high solid loading, i.e. 8%PG mix in this experiment, TPAD system was superior to MS. The optimum biogas yields of 271.7 and 264.0 m3/ tondry for MS and TPAD, respectively were achieved at 4%PG, beyond which floating scum started to disturb the mass transfer causing a yield drop, more noticeably in the single stage reactor. Para-grass addition shifted domination of bacterial species in the mesophilic reactors while the achaeal species were quite steady in both systems. Benefit of higher diversity of microorganisms found in TPAD system emerged at high solid loading. If existing MS system capacity is to be expanded, a smaller front-end thermophilic tank (instead of a larger second MS tank) should be considered, which should enable the system to operate at an overall higher solid loading. Acknowledgements This research was supported by The National Center of Excellence for Environmental and Hazardous Waste Management, Chulalongkorn University (ENG560080S), Bangkok, Thailand and the Postdoctoral fellowship from the Prince of Songkla University (PSU). The authors would like to thank Faculty of Environmental Management, Biogas and Biorefinery Research Laboratory, Faculty of Engineering of PSU for their generous supports of laboratory facilities. Finally, we would like to thank the Pig Business, Chareon Pokaphand Foods Public Company Limited Group, Thailand for supporting the scholarship. References APHA, 2012. Standard Methods for the Examination of Water and Wastewater. American Public Health Association, Washington, DC. Bong, C.P.C., Lim, L.Y., Lee, C.T., Klemeš, J.J., Ho, C.S., Ho, W.S., 2018. The characterisation and treatment of food waste for improvement of biogas production during anaerobic digestion – a review. J. Clean. Prod. 172, 1545–1558. Bouallagui, H., Torrijos, M., Godon, J.J., Moletta, R., Ben Cheikh, R., Touhami, Y., Delgenes, J.P., Hamdi, M., 2004. Two-phases anaerobic digestion of fruit and vegetable wastes: bioreactors performance. Biochem. Eng. J. 21, 193–197. Bouallagui, H., Touhami, Y., Ben Cheikh, R., Hamdi, M., 2005. Bioreactor performance in anaerobic digestion of fruit and vegetable wastes. Process Biochem. 40, 989–995. Cheng, C.-L., Chang, J.-S., 2011. Hydrolysis of lignocellulosic feedstock by novel cellulases originating from Pseudomonas sp. CL3 for fermentative hydrogen production. Bioresour. Technol. 102, 8628–8634. De La Rubia, M.A., Raposo, F., Rincón, B., Borja, R., 2009. Evaluation of the hydrolytic–acidogenic step of a two-stage mesophilic anaerobic digestion process of sunflower oil cake. Bioresour. Technol. 100, 4133–4138. Dechrugsa, S., Kantachote, D., Chaiprapat, S., 2013. Effects of inoculum to substrate ratio, substrate mix ratio and inoculum source on batch co-digestion of grass and pig manure. Bioresour. Technol. 146, 101–108. Demirel, B., Yenigun, O., 2004. Anaerobic acidogenesis of dairy wastewater: the effects of variations in hydraulic retention time with no pH control. J. Chem. Technol. Biotechnol. 79, 755–760. Deublein, D., Steinhauser, A., 2008. Biogas from Waste and Renewable Resources: An Introduction. WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany. DiLallo, R., Albertson, O.E., 1961. Volatile acids by direct titration. J. WPCP 33, 356–365. DLD, 2017. Data of livestock of Thailand, Department of Livestock Development (DLD). Avaliable in : 〈http://www.dld.go.th/th/〉.

262

Biocatalysis and Agricultural Biotechnology 15 (2018) 254–263

K. Saritpongteeraka et al.

hydrolysis of grass silage in a two-phase anaerobic digestion system. Bioresour. Technol. 143, 117–125. Ovreås, L., Forney, L., Daae, F.L., Torsvik, V., 1997. Distribution of bacterioplankton in meromictic Lake Saelenvannet, as determined by denaturing gradient gel electrophoresis of PCR-amplified gene fragments coding for 16S rRNA. Apply Environ. Microbiol. 63, 3367–3373. Pampillón-González, L., Ortiz-Cornejo, N.L., Luna-Guido, M., Dendooven, L., NavarroNoya, Y.E., 2017. Archaeal and bacterial community structure in an anaerobic digestion reactor (Lagoon Type) used for biogas production at a pig farm. J. Mol. Microbiol. Biotechnol. 27, 306–317. Poehlein, A., Zverlov, V.V., Daniel, R., Schwarz, W.H., Liebl, W., 2013. Complete genome sequence of Clostridium stercorarium subsp. stercorarium Strain DSM 8532, a thermophilic degrader of plant cell wall fibers. Genome Announc. 1 (e00073-13). Riau, V., De la Rubia, M.Á., Pérez, M., 2010. Temperature-phased anaerobic digestion (TPAD) to obtain class A biosolids: a semi-continuous study. Bioresour. Technol. 101, 2706–2712. Singh, S., Thavamani, P., Megharaj, M., Naidu, R., 2015. Multifarious activities of cellulose degrading bacteria from Koala (Phascolarctos cinereus) faeces. J. Anim. Sci. Technol. 57, 1. Song, Y.-C., Kwon, S.-J., Woo, J.-H., 2004. Mesophilic and thermophilic temperature cophase anaerobic digestion compared with single-stage mesophilic- and thermophilic digestion of sewage sludge. Water Res. 38, 1653–1662. Speece, R.E., 2008. Anaerobic Biotechnology and Ordor/Corrosion Control for Municipletilities and Industries. Biogas Utilization. Archea Press, United States of America, pp. 95–167. Thamsiriroj, T., Murphy, J.D., 2010. Difficulties associated with monodigestion of grass as exemplified by commissioning a pilot-scale digester. Energy Fuels 24, 4459–4469. Walter, A., Silberberger, S., Juárez, M.F.-D., Insam, H., Franke-Whittle, I.H., 2016. Biomethane potential of industrial paper wastes and investigation of the methanogenic communities involved. Biotechnol. Biofuels 9, 21. Ward, A.J., Hobbs, P.J., Holliman, P.J., Jones, D.L., 2008. Optimisation of the anaerobic digestion of agricultural resources. Bioresour. Technol. 99, 7928–7940. Wu, L.-J., Kobayashi, T., Li, Y.-Y., Xu, K.-Q., 2015. Comparison of single-stage and temperature-phased two-stage anaerobic digestion of oily food waste. Energy Convers. Manag. 106, 1174–1182. Wu, X., Yao, W., Zhu, J., 2010. Effect of pH on continuous biohydrogen production from liquid swine manure with glucose supplement using an anaerobic sequencing batch reactor. Int. J. Hydrog. Energy 35, 6592–6599. Xie, S., Lawlor, P.G., Frost, J.P., Hu, Z., Zhan, X., 2011. Effect of pig manure to grass silage ratio on methane production in batch anaerobic co-digestion of concentrated pig manure and grass silage. Bioresour. Technol. 102, 5728–5733. Xie, S., Lawlor, P.G., Frost, J.P., Wu, G., Zhan, X., 2012. Hydrolysis and acidification of grass silage in leaching bed reactors. Bioresour. Technol. 114, 406–413. Yu, H.-Q., Fang, H.H.P., Gu, G.-W., 2002. Comparative performance of mesophilic and thermophilic acidogenic upflow reactors. Process Biochem. 38, 447–454. Zhang, B., Zhao, H., Yu, H., Chen, D., Li, X., Wang, W., Piao, R., Cui, Z., 2016. Evaluation of biogas production performance and archaeal microbial dynamics of corn straw during anaerobic co-digestion with cattle manure liquid. J. Microbiol. Biotechnol. 26, 739–747.

El-Mashad, H.M., Zeeman, G., van Loon, W.K.P., Bot, G.P.A., Lettinga, G., 2004. Effect of temperature and temperature fluctuation on thermophilic anaerobic digestion of cattle manure. Bioresour. Technol. 95, 191–201. Elbeshbishy, E., Nakhla, G., Hafez, H., 2012. Biochemical methane potential (BMP) of food waste and primary sludge: influence of inoculum pre-incubation and inoculum source. Bioresour. Technol. 110, 18–25. Gagliano, M., Braguglia, C., Gallipoli, A., Gianico, A., Rossetti, S., 2015. Microbial diversity in innovative mesophilic/thermophilic temperature-phased anaerobic digestion of sludge. Environ. Sci. Pollut. Res. 22, 7339–7348. Gebreeyessus, G.D., Jenicek, P., 2016. Thermophilic versus mesophilic anaerobic digestion of sewage sludge: a comparative review. Bioengineering 3, 15. Hare, M., Booncharern, P., Tatsapong, P., Wongpichet, K., Kaewkunya, C., Thummasaeng, K., 1999. Performance of para grass (Brachiaria mutica) and Ubon paspalum (Paspalum atratum) on seasonally wet soils in Thailand. Trop. Grassl. 33, 75–81. Hniman, A., O-Thong, S., Prasertsan, P., 2011. Developing a thermophilic hydrogenproducing microbial consortia from geothermal spring for efficient utilization of xylose and glucose mixed substrates and oil palm trunk hydrolysate. Int. J. Hydrog. Energy 36, 8785–8793. Insam, H., Franke-Whittle, I., Goberna, M., 2010. Microbes in aerobic and anaerobic waste treatment. In: Microbes at Work from Wastes to Resources. Springer, Heidelberg. Khanal, S.K., 2008. Anaerobic Biotechnology for Bioenergy Production: Principles and Applications. Wiley-Blackwell, Singapore. Koch, K., Wichern, M., Lübken, M., Horn, H., 2009. Mono fermentation of grass silage by means of loop reactors. Bioresour. Technol. 100, 5934–5940. Lansing, S., Martin, J.F., Botero, Rl.B., da Silva, T.N., da Silva, E.D., 2010. Methane production in low-cost, unheated, plug-flow digesters treating swine manure and used cooking grease. Bioresour. Technol. 101, 4362–4370. Lehtomäki, A., Huttunen, S., Rintala, J.A., 2007. Laboratory investigations on co-digestion of energy crops and crop residues with cow manure for methane production: effect of crop to manure ratio. Resour. Conserv. Recycl. 51, 591–609. Li, X., Li, L., Zheng, M., Fu, G., Lar, J.S., 2009. Anaerobic co-digestion of cattle manure with corn stover pretreated by sodium hydroxide for efficient biogas production. Energy Fuels 23, 4635–4639. Li, Y., Khanal, S.K., 2016. Bioenergy: Principles and Applications. Jonh Wiley & Sons, Inc., Hoboken, New Jersey, USA. Lin, L., Wan, C., Liu, X., Lee, D.-J., Lei, Z., Zhang, Y., Tay, J.H., 2013. Effect of initial pH on mesophilic hydrolysis and acidification of swine manure. Bioresour. Technol. 136, 302–308. Madigan, M.T., Martinko, J.M., Parker, J., 1997. Brock's Biology of Microorganisms. Prentice Hall International Inc, New Jersey, USA. Montañés Alonso, R., Solera del Río, R., Pérez García, M., 2016. Thermophilic and mesophilic temperature phase anaerobic co-digestion (TPAcD) compared with singlestage co-digestion of sewage sludge and sugar beet pulp lixiviation. Biomass. Bioenergy 93, 107–115. Odedina, M.J., Charnnok, B., Saritpongteeraka, K., Chaiprapat, S., 2017. Effects of size and thermophilic pre-hydrolysis of banana peel during anaerobic digestion, and biomethanation potential of key tropical fruit wastes. Waste Manag. 68, 128–138. Orozco, A.M., Nizami, A.S., Murphy, J.D., Groom, E., 2013. Optimizing the thermophilic

263