Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters

Waste Management xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Waste Management journal homepage: www.elsevier.com/locate/wasman

Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters S.D. Yap, S. Astals, P.D. Jensen, D.J. Batstone, S. Tait ⇑ Advanced Water Management Centre, The University of Queensland, St Lucia, QLD 4072, Australia

a r t i c l e

i n f o

Article history: Received 1 September 2016 Revised 30 January 2017 Accepted 28 February 2017 Available online xxxx Keywords: Anaerobic digestion Microbial community Manure Straw Inoculum Biological inhibition

a b s t r a c t Batch solid-phase anaerobic digestion is a technology for sustainable on-farm treatment of solid residues, but is an emerging technology that is yet to be optimised with respect to start-up and inoculation. In the present study, spent bedding from two piggeries (site A and B) were batch digested at total solids (TS) concentration of 5, 10 and 20% at mesophilic (37 °C) and thermophilic (55 °C) temperatures, without adding an external inoculum. The results showed that the indigenous microbial community present in spent bedding was able to recover the full methane potential of the bedding (140 ± 5 and 227 ± 6 L CH4 kgVS
1 fed for site A and B, respectively), but longer treatment times were required than for digestion with an added external inoculum. Nonetheless, at high solid loadings (i.e. TS level > 10%), the digestion performance was affected by chemical inhibition due to ammonia and/or humic acid. Thermophilic temperatures did not influence digestion performance but did increase start-up failure risk. Further, inoculation of residues from the batch digestion to subsequent batch enhanced start-up and achieved full methane potential recovery of the bedding. Inoculation with liquid residue (leachate) was preferred over a solid residue, to preserve treatment capacity for fresh substrate. Overall, the study highlighted that indigenous microbial community in the solid manure residue was capable of recovering full methane potential and that solid-phase digestion was ultimately limited by chemical inhibition rather than lack of suitable microbial community. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction Spent bedding is a lignocellulosic residue that is produced when livestock is housed on straw or similar crop residues (Kruger et al., 2006). Anaerobic digestion (AD) is a suitable treatment option for lignocellulosic residues, with the benefit of renewable energy and the potential for enhanced nutrient recovery (Jha et al., 2011). Spent bedding has unique features for AD, because the manure deposited in the bedding can provide intrinsic bioactivity to pre-ferment the bedding prior to AD (Cui et al., 2011; Tait et al., 2009). For a particulate substrate such as spent bedding, the overall rate of AD is likely dictated by hydrolysis and methanogenesis (Batstone and Jensen, 2011). Leachbeds (also known as percolation or batch solid-phase digesters) operate at relatively high solids content (>20%) (Batstone and Jensen, 2011), making them particularly suitable for residues such as spent bedding (Yap et al., 2016). Typically, ⇑ Corresponding author at: Advanced Water Management Centre, The University of Queensland, Level 4, Gehrmann Laboratories Building (60), Brisbane, QLD 4072, Australia. E-mail address: [email protected] (S. Tait).

external inoculum would be required for a balanced microbial population during leachbed start-up. However, external inocula may not be available for decentralised on-farm digestion in Australia, because many farms are very remote and have biosecurity restrictions limiting the flow of materials between farms. A small number of studies have examined whether the indigenous microbial community in manure-laden spent bedding could provide selfinoculation (Kusch et al., 2008; Tait et al., 2009). This would be of benefit because the addition of an external inoculum would reduce the available AD capacity for fresh substrate. However, results from a recent study suggested that the lack of indigenous microbial presence could limit the recovery of methane potential from spent bedding in a pilot-scale leachbed (Yap et al., 2016). Previous studies have shown low recovery of methane potential from AD of agricultural wastes in a leachbed (Kusch et al., 2008; Lehtomaki et al., 2008). Leachbeds may operate at thermophilic conditions for part of their batch life (Pietsch, 2014). That is, when pre-aeration is applied during start-up, temperature is increased by the heat released from pre-composting, thereby reducing the requirements for external heating (Kusch et al., 2008). It has been demonstrated that AD rate can be higher at thermophilic (55–70 °C) than at

http://dx.doi.org/10.1016/j.wasman.2017.02.031 0956-053X/Ó 2017 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Yap, S.D., et al. Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters. Waste Management (2017), http://dx.doi.org/10.1016/j.wasman.2017.02.031

2

S.D. Yap et al. / Waste Management xxx (2017) xxx–xxx

mesophilic conditions (35–37 °C) (De la Rubia et al., 2012; Fernán dez-Rodríguez et al., 2013). This may be related to increased microbial activity and growth rate at thermophilic temperatures (De la Rubia et al., 2012; Fernández-Rodríguez et al., 2013). Despite enhanced rate kinetics, there have been contradicting reports on the stability of start-up and operation of thermophilic digesters (De la Rubia et al., 2012; Hegde and Pullammanappallil, 2007). Specifically, the start-up of a thermophilic digester can be constrained by a lack of acclimated microbes, leading to process instability and higher risk of failure (De la Rubia et al., 2012). Further, due to the downward shift of the ammonia acid-base pKa with increasing temperature, free ammonia concentration at a given total ammonia concentration increases with increasing temperature. This can also contribute to process instability when treating wastes containing manure (Chen et al., 2008). Nevertheless, there have been limited study of the sensitivity of indigenous microbial community in manure residues towards temperature during digester start-up, even though this may be of general importance for leachbeds. Another key operational parameter of the leachbed process is the initial solids loading. Loading of rapidly degradable substrate often leads to digester failure (Motte et al., 2013) due to rapid volatile fatty acid (VFA) accumulation and can subsequently lead to inhibition of methanogenic activity due to low pH. Also, high solids loading may elevate inhibitor and/or toxicant concentrations within the leachbed system, thus affecting digestion efficiency (Motte et al., 2013). For instance, when treating nitrogen-rich substrates, such as manure, the increased total ammonia-nitrogen (TAN) concentration that coincides with higher solids loading may lead to free-ammonia inhibition (Wilson et al., 2013; Yenigun and Demirel, 2013). Further, a study by Fernandes et al. (2014) suggested that humic and fulvic acid inhibited hydrolysis. It remains unclear whether the indigenous microbial community in solid livestock residues could appropriately inoculate start-up of a leachbed, despite the high solids loading and the presence of inhibitors. The aim of the present study was to evaluate the sufficiency of indigenous microbial activity present in fresh spent bedding for solid phase AD. The study examined the response to operating temperature (mesophilic and thermophilic) to better understand the impact of indigenous microbial community under different start-up scenarios. With a view on solid-phase digestion, the effect of solids concentration (5–10%) was also investigated. Lastly, the solid digestate and leachate residue from a previous digestion batch were recycled to inoculate a subsequent batch, in order to test this as another means for on-going inoculation.

2. Material and methods 2.1. Raw material Spent bedding was collected and prepared for further analysis according to Test Methods for the Examination of Composting and Compost (U.S. Composting Council, 2002) (see Sections 2.5 and 2.6 for analytical methods). Fresh samples (0–2 days old at the time of sampling) were collected from stockpiles at two piggeries (site A and B) in Queensland (Australia). The spent bedding from site A was from sheds housing smaller pigs only (called weaners, 10–24 kg), whereas spent bedding from site B was from sheds housing weaners and larger pigs (called growers, 24–36 kg). Different extents of pig exposure were visibly apparent, because the bedding samples from site A contained less faeces and urine (less soiled) than bedding samples from site B. At both sites the pigs were all reared according to a batch ‘‘all in, all out” mode. The batch time of the weaner-to-grower was 4 weeks at site A and

6 weeks at site B. At site B, the time for grower pigs to grow to slaughter weight was 3 weeks. Usually, a substantial amount of bedding straw is added at the start of the pig growth batch to cover the floor and then intermittently through the pig batch life, depending on humidity, straw durability and soilage extent. All bedding was removed at the end of each batch, usually to be stockpiled for passive composting. The average bedding use at sites A and B were about 0.30 and 0.12 kg pig
1 day
1, respectively. Bedding at site A consisted of mixed barley straw and wheat straw (50% w/w each), and was collected for testing during winter when in-shed temperature was an estimated 20 ± 5 °C. Bedding at site B consisted of wheat straw only, and was collected during summer when in-shed temperature was an estimated 27 ± 8 °C. 2.2. Biochemical methane potential tests Ultimate methane potential (B0) and anaerobic biodegradability were quantified by biochemical methane potential (BMP) tests in 310 mL media bottles based on the method of Angelidaki et al. (2009). External inoculum used in the BMPs was digestate from a completely mixed mesophilic digester in South East Queensland, treating primary and secondary municipal sludge. This inoculum was added at an inoculum-to-substrate ratio (ISR) of 2 on a volatile solids (VS) basis. Bottles were flushed with high-purity nitrogen gas for 1 min at 4 L min
1 and were then promptly sealed with a rubber septa retained by an open top screw cap. Tests were performed in triplicate and background methane production from substrate-free blanks were subtracted. Tests were mixed by inverting once after every sampling event. Biogas volume was measured using the method described by Jensen et al. (2011) and biogas composition was determined by gas chromatography (GC) as described in Section 2.5. 2.3. Batch tests without inoculation Batch experiments were conducted at a similar scale and using a similar method to that described above for the BMPs, except that no external inoculum was added. This was to assess the impact of indigenous microbial activity in the spent bedding samples on AD performance. The test conditions, as outlined in Table 1, included combinations of solids concentrations of 5, 10 and 20% total solid (TS), achieved by diluting the spent bedding with milli-QÒ water, and test temperatures of 37 or 55 °C. Tests were performed in sextuplicate at each condition to compensate for spent bedding heterogeneity. During the course of a test set, soluble fractions were analysed by withdrawing 0.5 mL leachate samples intermittently via an 18 Gauge syringe needle. Soluble fractions were assessed from only three of the six test bottles at each test

Table 1 Test conditions of the batch experiments without external inoculum. Spent bedding sample origin

Temperature (T, °C)

Total solids concentration (TS, %)

Test analysis performed

Site A

35

5 10 20 5 10 20

Gas composition and soluble content Gas composition Gas composition and soluble content Gas composition

5 10 20 5 10 20

Gas composition and soluble content Gas composition Gas composition and soluble content Gas composition

55

Site B

35

55

Please cite this article in press as: Yap, S.D., et al. Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters. Waste Management (2017), http://dx.doi.org/10.1016/j.wasman.2017.02.031

3

S.D. Yap et al. / Waste Management xxx (2017) xxx–xxx

condition, while biogas production and composition were assessed for all six bottles. Samples were analysed using the analytical methods in Section 2.5. Liquid samples could not be collected from batch tests at 20% TS, because these had inadequate free draining liquid. Biogas volume and composition were measured as described in Section 2.2 for the BMPs and with correction for changes in gas headspace volume by liquid sampling. The batch experiments were terminated when net methane production was no longer significant. The residue (leachate residual and solid digestate) from each test bottle was subsequently analysed as described in Section 2.4 below. 2.4. Batch tests with inoculation To evaluate the activity of solid digestate and leachate from a prior batch as an inoculum for a subsequent batch, a separate batch test set was performed in accordance with conditions as outlined in Table 2. The test was conducted at fixed TS of 10% and test temperature of 37 °C. Tests inoculated with solid digestate were diluted with milli-QÒ water to the nominal solids concentration, whereas tests inoculated with leachate were diluted with fullstrength or diluted leachate (50% with milli-QÒ water) to the nominal solids content. The chosen solid digestate to fresh spent bedding mix ratio was sufficient for spent bedding digestion based a study by Kusch et al. (2008). The microbial community of the inoculum was analysed according to the method described in Section 2.6. Background methane production contributed by liquid and solid digestate was measured and substrated. The residual methane from inoculum was negligible in magnitude relative to methane produced by the substrate. 2.5. Analytical methods For total fractions (soluble plus particulate), samples were analysed as collected. For soluble fractions, samples were immediately filtered through 0.45 lm syringe filters (PES membrane) and stored at 4 °C prior to analysis. Volatile fatty acids (VFAs) and alcohols were measured with a Perkin–Elmer GC with flame ionisation detector (GD-FID) and a DB-FFAP column. Total ammoniacalnitrogen (TAN) and phosphate were measured using a Lachat Quik-Chem 8000 Flow Injection Analyser (Lachat Instrument,

Table 2 Test conditions for batch tests with inoculation. Spent bedding origin

Inoculum source

Inoculation method

ISR (VS basis)

Site A

From previous batch tests at 5% TS and 35 °C that had no external inoculum added, and that digested spent bedding from site A

20% solid digestate and 80% fresh spent bedding on a weight basis Full-strength leachate only Diluted leachate (50% with miliQ-water) only

0.07 ± 0.01

20% solid digestate and 80% fresh spent bedding on a weight basis Full-strength leachate only Diluted leachate (50% with miliQ-water) only

0.12 ± 0.1

Site B

Batch tests at 5% TS and 35 °C that had no external inoculation added, and that digested spent bedding from site B

0.08 ± 0.01 0.04 ± 0.01

0.06 ± 0.01 0.03 ± 0.01

VS content of full strength leachate from site A and B is 1.05 and 0.93 wt% (wet basis), respectively.

Milwaukee). TS and VS were measured according to Standard Methods procedures 2540G (Franson et al., 2005) and corrected for VFA and alcohols losses (Peces et al., 2014). Total and soluble Chemical Oxygen Demand (tCOD and sCOD, respectively) were measured using a Merck SpectroquantÒ cell and a SpectroquantÒ Move 100 mobile colorimeter (Merck, Germany). Biogas composition (CH4, CO2 and H2) was determined using a Shimadzu GC-2014 with a HAYESEP Q80/100 as described by Astals et al. (2015). pH was measured using a Hannah Instrument HI8614LN Transmitter and HI2910B/H pH probe. Dissolved organic matter (DOM) was characterised via excitation emission matrix (EEM) fluorescence by a PerkinElmer LS-55 luminescence spectrometer (PerkinElmer, Australia) (Chen et al., 2003). The concentration of humic substances was determined using a Dissolved Organic Carbon Labor LC-OCD Model 8 according to the method of Huber et al. (2011). 2.6. Microbial characterisation methods Genomic DNA was extracted by using a FastSpin for Soil Kit (MP-Biomedicals, Santa Ana, California, USA) according to the manufacturer’s protocol. 300 ng DNA of each sample were provided to the Australian Centre for Ecogenomics (The University of Queensland) for 16S Amplicon sequencing using a Illumina Miseq Platform with a 926F (50 -AAACTYAAAKGAATTGACGG-30 ) and 1392wR (50 -ACGGGCGGTGWGTRC-30 ) primer set (Engelbrektson et al., 2010). Raw paired reads were first trimmed by Trimmomatic (Bolger et al., 2014) to remove short reads (less than 190 bp) and low quality reads (lower than Phred-33 of 20). The trimmed paired reads were then assembled using Pandaseq (Masella et al., 2012) with default parameters. The adapter sequences were removed by FASTQ Clipper of the FASTX-Toolkit (Pearson et al., 1997). The joined high quality sequence was analysed using QIIME v1.8.0 (Caporaso et al., 2010) with an open-reference operational taxonomic units (OTU) selecting strategy by uclust (Edgar, 2010) at 1% phylogenetic distance and assigned taxonomy by uclust against the greengenes database (13_05 release, McDonald et al., 2012; Werner et al., 2012). OTUs with only one read were filtered from the OTUs table by command filter_otus_from_otu_table.py in QIIME. Current methods can result in variable and incomplete extraction of microbial DNA from solid samples (Feinstein et al., 2009), thereby preventing reliable and accurate measurement of the microbial concentrations. Therefore, the microbial community profile was showed instead due to reasonable reproducibility. 2.7. Data analysis A two-step model was setup in Aquasim 2.1d to enable parameter identification from the experimental data. The effect of TS levels was not explicitly incorporated as the complex process interactions between mass transfer and biological inhibition is beyond the scope of current model.

Table 3 Characterisation of both spent bedding samples. Spent bedding origin

Site A

Site B

Total solids TS (wt%, wet basis) Volatile solids VS (wt%, wet basis) VS/TS ratios tCOD (kg COD kg TS
1 dry basis) TAN (g NH4-N kg TS
1 dry basis) Phosphate (g PO4-P kg TS
1 dry basis) VFA (g VFA kg TS
1 dry basis) Methane yield (L CH4 kg VS
1 fed) fd (%)
1 kmeth (d )

39 ± 3 27 ± 2 0.70 ± 0.01 0.67 ± 0.04 2.1 ± 0.2 2.1 ± 0.3 2.3 ± 0.4 140 ± 6 44 ± 4 0.12 ± 0.01

31 ± 2 23 ± 2 0.74 ± 0.01 0.86 ± 0.07 4.6 ± 0.6 1.4 ± 0.2 3.7 ± 0.9 227 ± 7 58 ± 5 0.10 ± 0.01

Please cite this article in press as: Yap, S.D., et al. Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters. Waste Management (2017), http://dx.doi.org/10.1016/j.wasman.2017.02.031

4

S.D. Yap et al. / Waste Management xxx (2017) xxx–xxx

In this model, particulate substrate (XS, g COD g COD
1 fed) was degraded to soluble monomers (S, g sCOD g COD
1 fed) at a rate rx
1 (gCOD g COD
1 ) and subsequently to methane (g CODCH4 fed d
1
1 g CODfed) at a rate rs (gCOD g COD
1 ): fed d

X ! S ! methane

rx

rs

ð1Þ

r x ¼
khyd  X S

ð2Þ

 0; t 6 t delay rs ¼
kmeth  S; t P t delay

ð3Þ

where kj is the first-order kinetic coefficient for process j (d
1), t delay is a lag-phase for methanogenesis during start-up (d), and subscript hyd and meth denote the biological process: hydrolysis and methanogenesis, respectively. The initial condition for degradable particulates (XS,0) was set to the total particulate substrate multiplied by the degradable fraction (XS,0 = Xtot  fd), where fd is substrate degradability. Initial conditions for other state variables were zero. The lag phase was generally negligible (i.e., not significantly or quantitatively different from zero) for solubilisation (rs), but significant

and measured for methanogenic activity. With batches for which soluble fraction was not measured, methane production data (cumulative methane COD) was simply fitted with a simple firstorder kinetic model with a lag phase to determine kinetic rate and degradability parameters. Where measured data are presented below with error bars or error (±), these are average values (n  3) with error bars representing a 95% confidence interval based on a two-tailed t-test (5% significance threshold). Where model parameters errors are presented, these are 95% confidence intervals based on a two-tailed t-test (5% significance threshold) based on standard error determined from the Secant Fisher information matrix as used by Jensen et al. (2011). Where relevant, errors were analytically propagated as described by Batstone (2013).

3. Results 3.1. Spent bedding characteristics Table 3 summarizes key characteristics of the spent bedding samples from sites A and B. Nutrients, such as TAN, phosphate and VFA

Fig. 1. Microbial community composition of fresh spent bedding from site A (left) and B (right). The heat map shows the relative abundance of microbial groups present on fresh spent bedding in triplicate (1, 2 and 3). The taxonomic classification is shown at the phylum level (left-hand side) and class and genus level of taxonomic assignment for bacteria and archaea, respectively, at 97% similarity.

Please cite this article in press as: Yap, S.D., et al. Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters. Waste Management (2017), http://dx.doi.org/10.1016/j.wasman.2017.02.031

S.D. Yap et al. / Waste Management xxx (2017) xxx–xxx

concentrations, were higher in the spent bedding from site B than from site A. This was not surprising considering site B has longer batch time and used less bedding used per pig per day than site A (Section 2.1), thus a greater extent of soilage (also higher microbes to bedding ratio) and pre-fermentation in the pig sheds of site B is expected. Further, higher temperature at site B than A (27 ± 8 and 20 ± 5 °C, respectively) likely to promote fermentation rate (Veeken and Hamelers, 1999), thereby contribute to higher nutrient concentrations in spent bedding from site B than A. B0 values and anaerobic biodegradability were higher for the spent bedding from site B. The B0 values measured for spent bedding from site A
1 (140 L CH4 kg VS
1 fed) and site B (227 L CH4 kg VSfed) were comparable to values reported by Kusch et al. (2008) and Tait et al. (2009). Fig. 1 shows microbial community composition of spent bedding from site A and B. The spent bedding from site A only had 2–3% of the total DNA extracted being identified as archaea, and similarly the spent bedding from site B had 4–6% of DNA extracted being identified as archaea (Fig. 1). These results are further discussed below in Section 4.1. In both spent bedding samples, the identifiable archaea were dominated by Methanobacterium and Methanosphaera, with a lesser contribution by Methanoculleus, Methanosarcina and Methanosaeta. The bacteria community of both spent beddings was dominated by the phyla Bacteroidetes, Actinobacteria and Proteobacteria (Fig. 1). 3.2. Effect of solids concentration and temperature Figs. 2–4 present time series data for the batch experiments without an external inoculum (per Section 2.3). The results sug-

5

gested that hydrolysis followed first-order kinetics without a lag time, while methane production followed first-order kinetics with a significant lag time. Table 4 summarizes parameter values for the kinetic model fits of the data. Hydrolysis commenced immediately (Fig. 2) and at 5 and 10% TS at mesophilic and thermophilic temperatures, it was possible to achieve the expected biodegradability of the substrate (as measured independently by the BMP tests). Hydrolysis rate decreased with increasing TS for both spent bedding samples, as indicated by a lower value for khyd at higher TS (Table 4). At each respective TS, hydrolysis was marginally faster at 55 °C than at 35 °C, as indicated by slightly higher khyd values at 55 °C. Substantial VFA accumulation (4–16 gVFA L
1) occurred at start-up, but was consumed over time as methane was being produced (Figs. 3 and 4). As expected, the extent of VFA accumulation increased with increasing TS and was also slightly higher at 55 °C than at 35 °C (Fig. 3). For both bedding samples, total VFA at 10% TS was nearly double that at 5% TS, most likely due to the higher organic loading at the higher TS. The VFA profiles between the two temperatures were comparable, with the exception of the higher peak VFA concentration at 55 °C than at 35 °C. Acetic acid was dominant, followed by propionic acid which contributed a lesser but still major proportion. Methane production showed a substantial time lag, which was more pronounced for spent bedding from site B than for bedding from site A (Fig. 4). Increasing TS resulted in a longer time lag. The lag-phase in methane production was generally shorter at 55 °C than at 35 °C (Table 4), indicating higher methanogenic activity at the higher temperature. Further, for bedding from site A, the

Fig. 2. The hydrolysis extent of spent bedding from site A (a and b) and B (c and d) during batch experiment without external inoculation.

Please cite this article in press as: Yap, S.D., et al. Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters. Waste Management (2017), http://dx.doi.org/10.1016/j.wasman.2017.02.031

6

S.D. Yap et al. / Waste Management xxx (2017) xxx–xxx

maximum methane yield was achieved within 80 days, whereas for bedding from site B it was achieved within 100 days (Fig. 4). In all cases, the full methane yield was recoverable for bedding from both site A (140 ± 5 L CH4 kg VS
1 fed) and site B (227 ± 6 L CH4 kg VS
1 fed) (Fig. 4 and Table 4). The only exception was the batch tests at 20% TS for the site B sample, which had a low methane production and only 4–5% of the B0 value recovered after 180 days of digestion. The methane production rate also decreased with increasing TS, as indicated by a lower value kmeth at higher TS (Table 4). While digestion performance was generally insensitive to temperature (Table 4), process start-up had a 40–50% chance of failure at 55 °C and at TS of 20%. Start-up failure is shown by minimal or no methane production (Fig. 4). Further analysis of tests that had low methane yield revealed a low pH (6.0 ± 0.5) with substantial VFA accumulation (12–16 gVFA L
1). This observation indicated strong inhibition of methanogenesis by low pH and VFA accumulation. 3.3. Post-digestion analysis After the first set of batch tests were terminated (per Section 2.3), leachate and solid residue were collected and analysed. Table 5 summarizes key characteristics of the collected leachate. Due to cost constraints, the concentration of humic substances could only be measured for leachate from the 5% TS batch tests, and was 0.07–0.09 g L
1 for digested site A samples and 0.30 g L
1 for digested site B samples, respectively. DOM analysis showed that the leachate residue contained 10–20%, 20–40% and 50–60% (on the basis of total dissolved carbon) of soluble microbial products, aromatic proteins and humic substance (including humic

and fulvic-like acid), respectively. With the microbial composition analysis on both the solid digestate and leachate residue, 30–40% of the total DNA extracted was identified as archaea. The identifiable archaea appeared to be dominated by Methanosarcina. The leachate and solid residues were also used as an inoculum for a subsequent batch test set as per Section 2.4. Fig. 5 presents the data from this test set. In all cases, methane production begun near immediately, indicating that both the leachate and solid residue had sufficient methanogenic activity for effective digester start-up. The rate of methane production was statistically the same, regardless of whether the batch tests were inoculated with solid or leachate residue (kmeth = 0.072 ± 0.002 d
1). Dilution of leachate by 50% also did not significantly influence the rate of methane production, indicating that the leachate had excess methanogenic activity/capacity and has no inhibition. Observations were similar for digestion of bedding from site B (data not shown). In all cases with bedding from site A, full recovery of methane yield (B0) was achieved within a treatment time of about 50 days (Fig. 5). However, with bedding from site B, digestion was slower (data not shown). These results are further discussed in Section 4.4 below. 4. Discussion 4.1. Microbial community presence for anaerobic digestion start-up The test results (Section 3.2) showed that raw spent bedding contained adequate microbes for start-up of a solid-phase anaerobic digester. However, performance was variable between sites with higher degradation rates from the site A samples, where the

Fig. 3. VFA profile during batch digestion of spent bedding from site A (a and b) and B (c and d) without external inoculation.

Please cite this article in press as: Yap, S.D., et al. Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters. Waste Management (2017), http://dx.doi.org/10.1016/j.wasman.2017.02.031

7

S.D. Yap et al. / Waste Management xxx (2017) xxx–xxx

Fig. 4. Cumulative methane production profile during batch digestion of spent bedding from site A (a and b) and B (c and d) without external inoculation.

Table 4 Model hydrolysis and methanogenic parameters outputs of batch tests without inoculation. Spent bedding origin

Site A

T (°C)

35

55

Site B

35

55

TS (%)

Hydrolysis

Methanogenesis

khyd (d
1)

fd (%)

lag-phase (d)

kmeth (d
1)

Measured Bo (mL CH4 gVS
1)

5 10 20 5 10 20

0.108 ± 0.004 0.087 ± 0.004 – 0.113 ± 0.004 0.101 ± 0.004 –

39 ± 3 42 ± 4 – 42 ± 3 41 ± 3 –

7.8 ± 0.5 10.5 ± 0.2 12.5 ± 0.4 2.0 ± 0.4 4.5 ± 0.3 5.5 ± 0.5

0.075 ± 0.005 0.068 ± 0.004 0.053 ± 0.005 0.078 ± 0.003 0.071 ± 0.004 0.056 ± 0.005

141 ± 5 140 ± 4 145 ± 8 142 ± 4 140 ± 5 146 ± 12

5 10 20 5 10 20

0.055 ± 0.005 0.042 ± 0.006 – 0.057 ± 0.003 0.035 ± 0.005 –

58 ± 3 57 ± 2 – 58 ± 2 58 ± 3 –

12.0 ± 0.6 16.1 ± 0.9 20 ± 2.0 4.8 ± 0.5 13.2 ± 1.0 23 ± 4.0

0.040 ± 0.002 0.032 ± 0.005 0.005 ± 0.001 0.044 ± 0.003 0.028 ± 0.002 0.003 ± 0.001

231 ± 5 225 ± 9 12 ± 7 230 ± 4 223 ± 4 9±7

storage conditions expected to favour growth of the indigenous population, such as extent of soilage, storage time and temperature were all lower. However, the yields were higher at site B, possibly due to differences in the bedding material, or longer storage acting as a pseudo pre-treatment. Further testing with a larger number of storage conditions would be required to assess the relationships between storage and development of the microbial community. However, differences in performance observed in the current study did not impact the viability of the process. That is, all tests were able to fully recover the B0 without an external inoculum. The microbial analysis of fresh spent beddings (Fig. 1) revealed presence of hydrolytic bacteria groups, such as Bacteriodetes and Proteobacteria, that are known to degrade lignocellulosic material (Samet et al., 2015). The relative abundance of methanogenic archaea, on the other hand, increased with digestion of the spent

bedding (compare Sections 3.1 and 3.3). It is possible that inadequate methanogenic archaea were present in the spent bedding to prevent the VFA accumulation observed at the start-up of the batch tests (Fig. 4). An increased relative abundance of archaea in the residue collected from batch digestion appeared to expedite start-up of the digestion with near-immediate methane production (Fig. 5). Again, the full methane yield was recoverable when residue was used as an inoculum, but with a faster onset of methanogenesis, a shorter overall digestion time was required. 4.2. Effect of solids loading (TS) on indigenous microbial performance The test results suggested that dissolved organic compounds and intermediates partially inhibited the AD at higher TS. Specifically, for ammonia, TAN concentration in the tests at 10 and 20% TS (Table 5)

Please cite this article in press as: Yap, S.D., et al. Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters. Waste Management (2017), http://dx.doi.org/10.1016/j.wasman.2017.02.031

8

S.D. Yap et al. / Waste Management xxx (2017) xxx–xxx

Table 5 Characteristics of leachate residue from batch tests without inoculation. Spent bedding origin

Temperature (T, °C)

Total solids concentration (TS, %)

pH

TAN (gNH4-N L
1)

Humic substance concentration (g total dissolved carbon L
1)

Site A

35

5 10 20 5 10 20

6.9 ± 0.2 7.1 ± 0.3 7.3 ± 0.2 7.0 ± 0.3 7.1 ± 0.2/5.9 ± 0.1a 7.2 ± 0.2/5.8 ± 0.3a

0.3 ± 0.1 0.6 ± 0.1 1.3 ± 0.1 0.4 ± 0.1 0.7 ± 0.1 1.5 ± 0.2

0.09 ± 0.01 n/a

5 10 20 5 10 20

7.7 ± 0.2 7.9 ± 0.3 6.0 ± 0.1 7.8 ± 0.2 7.7 ± 0.2 5.9 ± 0.2

0.7 ± 0.1 1.7 ± 0.2 3.3 ± 0.1 0.6 ± 0.1 1.5 ± 0.2 3.4 ± 0.2

0.31 ± 0.02 n/a

55

Site B

35

55

a

0.07 ± 0.01 n/a

0.29 ± 0.01 n/a

Batch test with start-up failure.

was at levels where both hydrolysis and methanogenesis could have been significantly inhibited (Chen et al., 2008; Wilson et al., 2013). Albeit that at 10% TS no VFA accumulation was detected once methane production had commenced (Fig. 3c and d), which indicated that ammonia was low enough to prevent substantial inhibition. With respect to recalcitrant organics, DOM analysis revealed that the aqueous phase contained a considerable amount of humic/ fulvic acid compounds (see Section 3.3). While the measured concentration of humic substances was at the lower end of reported inhibitory values (Brons et al., 1985; Fernandes et al., 2014; Ghasimi et al., 2016) for batch tests at 5% TS, their concentration would have been notably higher in batch tests at 10% and 20% TS. Humic substances are said to influence hydrolysis but not methanogenesis (Fernandes et al., 2014; Ghasimi et al., 2016). Inhibition by humic acid and ammonia could be a particularly important consideration when leachate is recycled as a secondary inoculum for subsequent digestion batches (Section 3.3), thereby causing progressive accumulation of inhibitors (also see Section 4.4).

4.3. Influence of temperature on microbial activity and digestion performance Digestion performance was mostly comparable at mesophilic and thermophilic conditions (Section 3.2), with similar khyd and kmeth values (Tables 3). B0 and f d were also not significantly different at the two operating temperatures. However, the risk of start-up failure at 10 and 20% TS was higher at the thermophilic condition (55 °C) than at the mesophillic condition (35 °C). This could have been caused by (1) an inconsistent level of suitable

methanogenic archaea (Hori et al., 2006) and/or (2) faster rates of hydrolysis and fermentation (Fig. 2) typically accompanied by a depressed pH (Li et al., 2010). In isolation, these results would suggest that a mesophilic start-up is preferred over a thermophilic start-up, given the higher failure risk at thermophilic conditions. The key issue is sourcing an acclimatised inoculum to ensure quick and successful start-up at thermophilic conditions (De la Rubia et al., 2012), because such an inoculum may not be readily available. Future work can explore digestion performance under thermophilic conditions using an inoculum pre-acclimated to thermophilic conditions. 4.4. Strategies for start-up of solid-phase anaerobic digesters treating livestock residues As highlighted above, the indigenous inoculum in spent bedding would likely be adequate for AD start-up at moderate solids loading (<10% TS) and mesophilic temperatures (Sections 4.1– 4.3). To operate at higher solid loading (10% TS) and higher temperatures, methanogenic activity could be boosted by inoculating a subsequent batch with residue from a previous batch. In this regard, leachate residue would probably be preferred over solid digestate, to preserve the treatment capacity for fresh spent bedding and because leachate could also be more easily handled (Cui et al., 2011; Zhu et al., 2015). Inhibitors could accumulate in leachate with multiple reuses and thus impact on digestion performance (Shahriari et al., 2012; Yap et al., 2016). A leachate inoculum may require intermittent dilution or treatment to prevent inhibitory concentrations of substances such as ammonia and/or recalcitrant dissolved organic matter. For example, ammonia might be removed via biogas stripping (Serna-Maza et al., 2015; Walker et al., 2011), and humic substances through rechargeable ion exchange (Bashir et al., 2015). Previous studies have also showed that the addition of calcium, magnesium and iron as inorganic salts (Azman et al., 2015; Brons et al., 1985) could mitigate inhibition by humic substances, but this is likely to be cost-prohibitive for on-farm applications. 5. Conclusion

Fig. 5. Methane production profile during batch digestion of spent bedding from site A inoculated with solid digestate or leachate residue from previous at 10 TS% and mesophilic condition.

The indigenous microbial community in spent piggery bedding samples from two piggeries were able to effectively start-up anaerobic digestion. Digestion without an external inoculum was able to recover the full methane yield (sample from piggery site A: 140 ± 5 L CH4 kg
1 VSfed; sample from piggery site B: 227 ± 6 L CH4 kg
1 VSfed), but extended treatment times were required as compared to when an external inoculum was added. This was likely due to insufficient active biomass indigenous to

Please cite this article in press as: Yap, S.D., et al. Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters. Waste Management (2017), http://dx.doi.org/10.1016/j.wasman.2017.02.031

S.D. Yap et al. / Waste Management xxx (2017) xxx–xxx

the spent bedding. The digestion performance of spent bedding could have been influenced by humic substances and ammonia at high solids content (corresponding to high organic loading in solid-phase digestion). Thermophilic temperatures did not enhance digestion performance, but increased start-up failure risk with minimal or no methane production from some of the test batches. Residues (leachate and solid) from a previous batch digestion were very effective and consistent inocula for start-up of subsequent batches, albeit that leachate would likely be preferred over solid residue as an inoculum to preserve treatment capacity for fresh substrate. Where leachate is reused as an inoculum, care should be taken to manage the accumulation of inhibitors. Acknowledgments This research was funded by the Australian Government Department of Agriculture and Water Resources, Quantum Power Limited, Australian Egg Corporation Limited and the CRC for High Integrity Australian Pork (Pork CRC). Australian Pork Limited was an in-kind contributor. References Angelidaki, I., Alves, M., Bolzonella, D., Borzacconi, L., Campos, J.L., Guwy, A.J., Kalyuzhnyi, S., Jenicek, P., van Lier, J.B., 2009. Defining the biomethane potential (BMP) of solid organic wastes and energy crops: a proposed protocol for batch assays. Water Sci. Technol. 59, 927–934. Astals, S., Batstone, D.J., Tait, S., Jensen, P.D., 2015. Development and validation of a rapid test for anaerobic inhibition and toxicity. Water Res. 81, 208–215. Azman, S., Khadem, A.F., Zeeman, G., van Lier, J.B., Plugge, C.M., 2015. Mitigation of humic acid inhibition in anaerobic digestion of cellulose by addition of various salts. Bioengineering 2, 54–65. Bashir, M.J.K., Aziz, H.A., Amr, S.S.A., Sethupathi, S., Ng, C.A., Lim, J.W., 2015. The competency of various applied strategies in treating tropical municipal landfill leachate. Desalin. Water Treat. 54, 2382. 2382-239. Batstone, D.J., 2013. Teaching uncertainty propagation as a core component in process engineering statistics. Educ. Chem. Eng. 8, e132–e139. Batstone, D.J., Jensen, P.D., 2011. Anaerobic processes. In: Wilderer, P. (Ed.), Treatise on Water Science. Elsevier, Oxford, pp. 615–639. Bolger, A.M., Lohse, M., Usadel, B., 2014. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120. Brons, H.J., Field, J.A., Lexmond, W.A.C., Lettinga, G., 1985. Influence of humic acids on the hydrolysis of potato protein during anaerobic digestion. Agric. Wastes 13, 105–114. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Fierer, N., Peña, A.G., Goodrich, J.K., Gordon, J.I., Huttley, G.A., Kelley, S.T., Knights, D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D., Pirrung, M., Reeder, J., Sevinsky, J.R., Turnbaugh, P.J., Walters, W.A., Widmann, J., Yatsunenko, T., Zaneveld, J., Knight, R., 2010. QIIME allows analysis of highthroughput community sequencing data. Nat. Methods 7, 335–336. Chen, W., Westerhoff, P., Leenheer, J.A., Booksh, K., 2003. Fluorescence excitation– emission matrix regional integration to quantify spectra for dissolved organic matter. Environ. Sci. Technol. 37, 5701–5710. Chen, Y., Cheng, J.J., Creamer, K.S., 2008. Inhibition of anaerobic digestion process: a review. Bioresour. Technol. 99, 4044–4064. Cui, Z., Shi, J., Li, Y., 2011. Solid-state anaerobic digestion of spent wheat straw from horse stall. Bioresour. Technol. 102, 9432–9437. De la Rubia, M.A., Riau, V., Raposo, F., Borja, R., 2012. Thermophilic anaerobic digestion of sewage sludge: focus on the influence of the start-up. A review. Crit. Rev. Biotechnol. 33, 448–460. Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. Engelbrektson, A., Kunin, V., Wrighton, K.C., Zvenigorodsky, N., Chen, F., Ochman, H., Hugenholtz, P., 2010. Experimental factors affecting PCR-based estimates of microbial species richness and evenness. ISME J. 4, 642–647. Feinstein, L.M., Sul, W.J., Blackwood, C.B., 2009. Assessment of bias associated with incomplete extraction of microbial DNA from soil. Appl. Environ. Microbiol. 75, 5428–5433. Fernandes, T.V., Lier, J.B., Zeeman, G., 2014. Humic acid-like and fulvic acid-like inhibition on the hydrolysis of cellulose and tributyrin. Bioenergy Res. 8, 821– 831. Fernández-Rodríguez, J., Pérez, M., Romero, L.I., 2013. Comparison of mesophilic and thermophilic dry anaerobic digestion of OFMSW: kinetic analysis. Chem. Eng. J. 232, 59–64.

9

Franson, M.A.H., Eaton, A.D., Association, A.P.H., Association, A.W.W., Federation, W. E., 2005. Standard Methods for the Examination of Water & Wastewater. American Public Health Association, Washington, DC. Ghasimi, D.S.M., Aboudi, K., de Kreuk, M., Zandvoort, M.H., van Lier, J.B., 2016. Impact of lignocellulosic-waste intermediates on hydrolysis and methanogenesis under thermophilic and mesophilic conditions. Chem. Eng. J. 295, 181–191. Hegde, G., Pullammanappallil, P., 2007. Comparison of thermophilic and mesophilic one-stage, batch, high-solids anaerobic digestion. Environ. Technol. 28, 361– 369. Hori, T., Haruta, S., Ueno, Y., Ishii, M., Igarashi, Y., 2006. Dynamic transition of a methanogenic population in response to the concentration of volatile fatty acids in thermophilic anaerobic digester. Appl. Environ. Microbiol. 72, 1623–1630. Huber, S.A., Balz, A., Abert, M., Pronk, W., 2011. Characterisation of aquatic humic and non-humic matter with size-exclusion chromatography – organic carbon detection – organic nitrogen detection (LC-OCD-OND). Water Res. 45, 879–885. Jensen, P.D., Ge, H., Batstone, D.J., 2011. Assessing the role of biochemical methane potential tests in determining anaerobic degradability rate and extent. Water Sci. Technol. 64, 880–886. Jha, A.K., Li, J.Z., Nies, L., Zhang, L.G., 2011. Research advances in dry anaerobic digestion process of solid organic wastes. Afr. J. Biotechnol. 10, 14242–14253. Kruger, I., Taylor, G., Rosese, G., Payne, H., 2006. Primefact 68, Deep-litter Housing for Pigs. NSW DPI, Sydney, Australia. Kusch, S., Oechsner, H., Jungbluth, T., 2008. Biogas production with horse dung in solid-phase digestion systems. Bioresour. Technol. 99, 1280–1292. Lehtomaki, A., Huttunen, S., Lehtinen, T., Rintala, J.A., 2008. Anaerobic digestion of grass silage in batch leach bed processes for methane production. Bioresour. Technol. 99, 3267–3278. Li, L., Li, D., Sun, Y., Ma, L., Yuan, Z., Kong, X., 2010. Effect of temperature and solid concentration on anaerobic digestion of rice straw in South China. Int. J. Hydrogen Energy 35, 7261–7266. Masella, A.P., Bartram, A.K., Truszkowski, J.M., Brown, D.G., Neufeld, J.D., 2012. PANDAseq: paired-end assembler for Illumina sequences. BMC Bioinformatics 13, 1–7. McDonald, D., Price, M.N., Goodrich, J., Nawrocki, E.P., DeSantis, T.Z., Probst, A., Andersen, G.L., Knight, R., Hugenholtz, P., 2012. An improved greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. ISME J. 6, 610–618. Motte, J.C., Trably, E., Escudié, R., Hamelin, J., Steyer, J.-P., Bernet, N., Delgenes, J.P., Dumas, C., 2013. Total solids content: a key parameter of metabolic pathways in dry anaerobic digestion. Biotechnol. Biofuels 6, 1–9. Pearson, W.R., Wood, T., Zhang, Z., Miller, W., 1997. Comparison of DNA sequences with protein sequences. Genomics 46, 24–36. Peces, M., Astals, S., Mata-Alvarez, J., 2014. Assessing total and volatile solids in municipal solid waste samples. Environ. Technol. 35, 3041–3046. Pietsch, S., 2014. Anaerobic Digestion Systems and Deep Litter Waste Management for Rural Primary Industry Enterprises. VIC, ISS Institute, Melbourne, Australia. Samet, A., Ahmad, K.F., van Lier, J.B., Zeeman, G., Caroline, P.M., 2015. Presence and role of anaerobic hydrolytic microbes in conversion of lignocellulosic biomass for biogas production. Environ. Sci. Technol. 45, 2523–2564. Serna-Maza, A., Heaven, S., Banks, C.J., 2015. Biogas stripping of ammonia from fresh digestate from a food waste digester. Bioresour. Technol. 190, 66–75. Shahriari, H., Warith, M., Hamoda, M., Kennedy, K.J., 2012. Effect of leachate recirculation on mesophilic anaerobic digestion of food waste. Waste Manage. 32, 400–403. Tait, S., Tamis, J., Edgerton, B., Batstone, D.J., 2009. Anaerobic digestion of spent bedding from deep litter piggery housing. Bioresour. Technol. 100, 2210–2218. U.S. Composting Council, 2002. Test Methods for the Examination of Composting and Compost. Composting Council Research and Education Foundation, Bethesda. Veeken, A., Hamelers, B., 1999. Effect of temperature on hydrolysis rates of selected biowaste components. Bioresour. Technol. 69, 249–254. Walker, M., Iyer, K., Heaven, S., Banks, C.J., 2011. Ammonia removal in anaerobic digestion by biogas stripping: An evaluation of process alternatives using a first order rate model based on experimental findings. Chem. Eng. J. 178, 138–145. Werner, J.J., Koren, O., Hugenholtz, P., DeSantis, T.Z., Walters, W.A., Caporaso, J.G., Angenent, L.T., Knight, R., Ley, R.E., 2012. Impact of training sets on classification of high-throughput bacterial 16s rRNA gene surveys. ISME J. 6, 94–103. Wilson, L.P.H., Loetscher, L.E., Sharvelle, S., De Long, S.K., 2013. Microbial community acclimation enhances waste hydrolysis rates under elevated ammonia and salinity conditions. Bioresour. Technol. 146, 15–22. Yap, S.D., Astals, S., Jensen, P.D., Batstone, D.J., Tait, S., 2016. Pilot-scale testing of a leachbed for anaerobic digestion of livestock residues on-farm. Waste Manage. 50, 300–308. Yenigun, O., Demirel, B., 2013. Ammonia inhibition in anaerobic digestion: a review. Process Biochem. 48, 901–911. Zhu, J., Yang, L., Li, Y., 2015. Comparison of premixing methods for solid-state anaerobic digestion of corn stover. Bioresour. Technol. 175, 430–435.

Please cite this article in press as: Yap, S.D., et al. Indigenous microbial capability in solid manure residues to start-up solid-phase anaerobic digesters. Waste Management (2017), http://dx.doi.org/10.1016/j.wasman.2017.02.031