Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion

Bioresource Technology xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

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

Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion Chaoran Li a, Christoph Mörtelmaier a, Josef Winter a,⇑, Claudia Gallert a,b,1 a b

Institute of Biology for Engineers and Biotechnology of Wastewater, Karlsruhe Institute of Technology KIT, Am Fasanengarten, D-76128 Karlsruhe, Germany University of Applied Science, Hochschule Emden Leer, Faculty of Technology, Microbiology – Biotechnology, Constantia Platz 4, D-26723 Emden, Germany

h i g h l i g h t s  Dry anaerobic digestion needs P75% moisture.  Methanosarcinales dominate, no Methanosaeta spec.  Biogas/methane rates and amounts are equal at 37 and 55 °C.

a r t i c l e

i n f o

Article history: Received 19 December 2013 Received in revised form 23 February 2014 Accepted 25 February 2014 Available online xxxx Keywords: Dry anaerobic digestion Water activity Biogas production Volatile fatty acid degradation Microbial population

a b s t r a c t Methane production from biowaste with 20–30% dry matter (DM) by box-type dry anaerobic digestion and contributing bacteria were determined for incubation at 20, 37 and 55 °C. The same digestion efficiency as for wet anaerobic digestion of biowaste was obtained for dry anaerobic digestion with 20% DM content at 20, 37 and 55 °C and with 25% DM content at 37 and 55 °C. No or only little methane was produced in dry anaerobic reactors with 30% DM at 20, 37 or 55 °C. Population densities in the 20–30% DM-containing biowaste reactors were similar although in mesophilic and thermophilic biowaste reactors with 30% DM content significantly less but phylogenetically more diverse archaea existed. Biogas production in the 20% and 25% DM assays was catalyzed by Methanosarcinales and Methanomicrobiales. In all assays Pelotomaculum and Syntrophobacter species were dominant propionate degraders. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Biowaste is the moist organic fraction of municipal garbage with a high percentage of organic matter and is collected separately in many municipalities of Germany. After separation of non-digestible material it can be treated by wet or dry anaerobic digestion (e.g., Lissens et al., 2001; Luning et al., 2003; de Baere, 2000; Nayono et al., 2009). Reports on dry anaerobic digestion (DAD) were dealing with stirred tank reactors and biowaste fractions that contained up to 25% dry matter (DM) (e.g., Cecchi et al., 1991; Mata-Alvarez et al., 1993; Pavan et al., 2000; Bolzonella et al., 2006). Only Abbassi-Guendouz et al. (2012) reported successful DAD of artificial biowaste (card boards) with a DM content of 30% in 2 of 4 parallel reactors. Model equations for batch fermentation revealed that mass transfer was strongly limited in ⇑ Corresponding author. Tel.: +49 721 60842297. E-mail addresses: [email protected] (C. Li), [email protected] (C. Mörtelmaier), [email protected] (J. Winter), Claudia.gallert@hs-emden-leer. de, [email protected] (C. Gallert). 1 Tel.: +49 4921 8071586.

the DAD reactors with P30% DM content, leading to local acidification and reduced methanogenic activity (Abbassi-Guendouz et al., 2012). Even if an almost complete hydrolysis of biowaste solids could be obtained by thermochemical or biological pretreatment to increase the soluble COD (Fernandez-Güelfo et al., 2011) the synergistic action of fermentative bacteria with methanogens and that of acetogenic bacteria with acetate cleaving and hydrogenolytic methanogenic bacteria was still required for stable methane production (Gallert and Winter, 2005). Although in praxi DAD often is established in low-tech box fermenters (‘‘garage fermenters’’) and in high-tech stirred tank vertical or horizontal reactors the focus of most literature reports was layed on stirred tank reactors with much better mass transfer properties. Almost no information about methanogenesis in box fermenters is available. Box fermenters for DAD are considered inexpensive and meet the requirements for the ‘‘technology bonus’’ (2 Euro cent per kWh electricity) offered by the German Renewable Energy Law. No process water is added to DAD during reactor feeding, but in some systems process water or leachate is sprayed on or mixed into the digesting material to improve biogas

http://dx.doi.org/10.1016/j.biortech.2014.02.118 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Li, C., et al. Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.02.118

2

C. Li et al. / Bioresource Technology xxx (2014) xxx–xxx

production and digestion efficiency. Anaerobic microbial consortia for biogas production from organic matter require an aequous environment with a water activity of >0.91 (e.g., Rockland and Beuchal, 1987) for high-rate hydrolysis of polymers, acidogenesis of monomers, acetogenesis of fatty acids and methanogenesis of acetate and of CO2/H2. At the high dry matter (DM) content of nonmoistened municipal biowaste (P30%) there may not be enough bioavailable water (aW 6 0.91) for an optimal, non water limited multistep DAD. As waste treatment by DAD began only about 20 years ago, not many data on the behavior of DAD during startup and during long-term fermentation are available. AbbassiGuendouz et al. (2012) reported a slightly decreasing methane production from card boards for increasing DM content from 10% to 25%. At 30% DM content methanogenic activity was no longer stable and at 35% DM content methanogenesis failed completely. The authors concluded that 30% DM content was the threshold concentration for the solids content. Simulation with the anaerobic digestion model No.1 revealed mass transfer limitations at increasing DM content and in particular a limited hydrolysis rate at high solids content. In another report mesophilic and thermophilic DAD of the organic fraction of municipal solid waste (OFMSW with 20% DM) was compared by modeling organic matter conversion and biogas production (Fernàndez-Rodriguez et al., 2013). Specific growth rates were 27–60% higher during thermophilic than during mesophilic methanogenesis, which could have been due to an increased water activity. More water of the moisture was apparently bioavailable at 55 °C than at 37 °C. As a consequence thermophilic reactor operation would require a shorter retention time and presumably less investment costs for the digester. The start-up of reactors for wet anaerobic digestion (WAD) or DAD depends on the activity of the inoculum, which ideally should come from digestion of the same or of a similar substrate. A major problem during the start-up phase may be the accumulation of volatile fatty acids, especially of propionate by the fast-growing heterotrophs (Gallert and Winter, 2008; Felchner-Zwirello et al., 2012, 2013), which lead to acidification of the reactor content and, if no counteractions are taken, to failure. Monitoring of fatty acids for flexible biowaste addition may shorten the start-up time and is considered helpful for a successful start-up of bioreactors, e.g., after revision (Gallert et al., 2003; Gallert and Winter, 2008). High-rate anaerobic digestion by WAD or DAD depends on syntrophic interaction of fatty acid degrading acetogens with acetate and H2/CO2 utilizing methanogens to avoid volatile fatty acid accumulation (Gallert and Winter, 2005). If fatty acids have been formed due to process imbalances, propionic acid is the most critical organic acid, since its degradation may depend on the established degradation pathway (Felchner-Zwirello et al., 2012), the hydrogen partial pressure and on acetate levels. High degradation rates require close interspecies distances between propionate degraders, hydrogenolytic and aceticlastic methanogens (Felchner-Zwirello et al., 2013) that prevail during DAD, but also a nonlimiting water activity as in WAD. Little is known about the community structure in DAD reactors. Within the methanogens a dominance of Methanosarcinales, either of Methanosaeta spec. (Chu et al., 2010; Montero et al., 2009) or Methanosarcina spec. (Cho et al., 2013) was reported, whereas propionate degraders, with one exception (Zahedi et al., 2013), were only investigated in WAD systems (e.g., Ariesyady et al., 2007; Felchner-Zwirello et al., 2012; Narihiro et al., 2012; Chu et al., 2010). In this contribution we address three aspects of DAD: the optimal water content, the influence of a temperature change from ambient (20 °C) to mesophilic (37 °C) and thermophilic (55 °C) temperatures as well as identification and enumeration of dominant methanogens and propionic acid degraders with respective gene probes.

2. Methods 2.1. Substrates: source of fresh biowaste and digester residues for inoculation Source-sorted biowaste was collected with rotating drum trucks by City authorities of Karlsruhe, Germany for large-scale wet anaerobic digestion (WAD) (Gallert et al., 2003; Nayono et al., 2009). For lab-scale dry anaerobic digestion (DAD) experiments woody material, ornamental plant soils as well as paper, plastic foils, broken glass and metals were manually sorted out from the collected biowaste fraction before shredding in a cutter (ZG Raiffeisen, Karlsruhe) to 1 cm length. The dry matter (DM) content of the sorted biowaste (triplicate analyses) was 30.9 ± 0.6% (first batch for start of experiments) and 30.3 ± 0.6% (second batch for the re-feeding experiments). In both batches the organic dry matter content (ODM, triplicate analyses) varied only little between 65% and 67% of the DM content (calculated by ODM/DM 100, Table 1). As a source of microorganisms solid residues of digested biowaste suspensions were taken from the extrusion pipe of the sludge centrifuge at the WAD plant of the City of Karlsruhe. This inoculum contained 33.7 ± 0.6% DM of which 62% were organic material (bacteria and undigested/undigestible biowaste particles). Portions of 10 kg of shredded fresh biowaste and of 10 kg solid residues of digested biowaste suspension were mixed thoroughly. Little water was added to obtain a calculated DM content of 30%, which was controlled in each fermenter by respective analyses. To obtain biowaste fractions with 25% or 20% DM content (Table 1), the above mixture was accordingly diluted with tap water. Percent amounts were calculated from the mean of triplicate analyses (variation ±0.6%). After 300 days all DAD reactors were re-fed (Table 2). 2.2. Digester set-up, feeding and incubation conditions Parallel DAD experiments with 2 kg of the above prepared biowaste fractions that contained 30%, 25% or 20% DM were started in 3 L glass reactors. One reactor was only fed with 2 kg digester residues that contained 25% DM and was incubated as a control. Reactors were initially flushed with nitrogen, closed with a rubber stopper and incubated at room temperature (20 ± 1 °C). Incubation time, changes of the incubation temperature, e.g., after re-feeding, pH corrections with 5 M NaOH are mentioned at the respective experiments. Biogas from the 3 L reactors was measured with a Ritter model MGC-1 V30 mini gas counter, analyzed with a Blue sense model BACCom 12 CB methane/CO2 gas detector and registered by a computer (System Blue Sense Gas GmbH, D-45099 Herten, Germany). Since no gas was produced after 3 months of incubation in the control reactor, it was stopped. Little leachate water that accumulated at the bottom of the reactors after 3–4 days was regularly remixed into the solid fraction by shaking the reactors. For measuring the pH and for analysis of volatile fatty acids 1 ml of leachate was withdrawn through a bottom valve. Initial incubation of all reactors was at room temperature (20 ± 1 °C). The pH was adjusted after 5, 10 and 30 days to 8 (as good as this is possible at the high DM content) and then stabilized itself at around 8 (Figs. 1b, 2b, 3b and c). Later on the incubation temperature was raised to 37 ± 0.5 or 55 ± 0.5 °C as indicated in Figs. 1–3. Re-fed reactors were incubated at 37 ± 0.5 or 55 ± 0.5 °C. 2.3. Analyses To determine the DM content of biowaste three 1-kg portions were dried to constant weight at 105 °C. The organic DM (ODM) content was obtained from triplicate 20 g-portions of the united

Please cite this article in press as: Li, C., et al. Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.02.118

3

C. Li et al. / Bioresource Technology xxx (2014) xxx–xxx Table 1 Mass data of reactors for dry anaerobic digestion of biowaste with 30, 25, 20% dry matter content at start and after 300 days of digestion. Time

t0 (start)

tend (after 300 days)

Biowaste

DMtotal (kg)

DM (%)

ODM (%)

DMtotal (kg)

DM (%)

ODM (%)

DM DM DM DM DM DM

5.1 5.04 5.54 5.72 6.14 6.02

30.9 30.9 25.2 25.2 21.2 21.2

20.1 20.1 16.3 16.3 13.8 13.8

5.0 4.9 5.3 5.5 5.9 5.8

22.7 22.9 18.4 19.3 14.2 14.7

11.7 12.1 9.0 9.8 7.1 7.2

30% 30% 25% 25% 20% 20%

R1 R2 R1 R2 R1 R2

DM = dry matter; ODM = organic dry matter; R1, R2 = reactor 1 and reactor 2, initially at the same and later on at a different temperature regime as depicted in Figs. 1–3.

Table 2 Re-feeding of dry anaerobic reactors for mesophilic and thermophilic methanogenesis. Reactor designation after re-feeding DM DM DM DM DM DM a

30%,R1⁄ 30%, R2⁄ 25%, R1⁄ 25%,R2⁄ 20%, R1⁄ 20%, R2⁄

Content 1 kg 1 kg 1 kg 1 kg 1 kg 1 kg

residue residue residue residue residue residue

DM = actual dry matter content of each reactor at start;

⁄

DM 30% R1 + 1 kg fresh biowaste DM30% R2 + 1 kg fresh biowaste DM 25% R1 + 1 kg fresh biowaste DM 25% R2 + 1 kg fresh biowaste DM 20% R1 + 1 kg fresh biowaste DM 20% R2 + 1 kg fresh biowaste

Incubation Temperature (°C)

Water addition

DMa (%)

37 55 37 55 37 55

n.a. n.a. n.m. n.m. 250 ml 250 ml

28.6 28.6 24.9 24.8 20.3 20.0

re-fed reactors; n.a. = no addition; n.m. = not measured.

and powdered dried material by heating the dried samples to 550 °C for 2 h and subtracting the ash residues (mineral content) from the 20 g dried biowaste initially weighed in for analysis (APHA, AWWA, WEF, 2005). Ammonia was quantified with a test kit of Dr. Lange (Berlin, Germany). Acetate, propionate and n-buytrate in Figs. 1–4 (i-butyrate, i- and n-valerate were not detected) were determined by gaschromatography with FID detection. Hydrogen, methane and carbon dioxide were analyzed every third sampling using a gas chromatograph with thermoconductivity detector (Gallert and Winter, 2008). All values are the mean of at least duplicate analyses. Gas amount was analyzed with Ritter mini gas counters (BlueSens Gas Sensor GmbH Herten, Germany).

without filters specific for DAPI staining and the fluorescence dyes of the respective gene probes. Images were analyzed with Axiovision 3.1 or DAIME software (Daims et al., 2006) and visually corrected for fluorescent particles other than the bacteria. Image corrections were made manually by taking into account criteria such as particle size and form, color and intensity of fluorescence, and by comparing DAPI, phase contrast and FISH images of the same view field. The lowest detection limit was 1.58  106 cells/ ml. This number was calculated taken the case of 1 FISH positive cell in 10 view fields. Taxa, which were not found in 10 microscopic view fields, are not at all present or only present at lower numbers than 1.58  106 cells/ml.

2.4. Characterization of the biowaste population by fluorescence in situ hybridization (FISH)

3. Results and discussion

Biowaste samples were taken from the reactors that digested biowaste with 20%, 25% and 30% DM and processed according to Felchner-Zwirello et al. (2013). Portions of 0.1 g of each sample were mixed with 0.3 ml of 4% para-formaldehyde solution (Amann et al., 1990). The mixture was incubated at 4 °C for 3 h and then centrifuged at 15,000 rpm for 5 min in a Microfuge (Eppendorf, Hamburg, Germany). The pellet was washed in phosphate buffered saline solution (PBS). Samples were frozen at 20 °C in 1 ml 50% ethanol-PBS-solution before further analysis. For FISH (Amann et al., 1990) 5 ll of sample were transferred on a Teflon coated slide (8 mm diameter), air dried and completely dehydrated by ascending ethanol concentrations of 50%, 80% and 99% (exposure to each concentration for 1 min.). For FISH analyses 10 ll hybridization buffer (Amann et al., 1990) and 2 ll of the respective gen probes (1:20 diluted with distilled water, Table 3) were mixed and incubated 1.5 h at 46 °C, then washed with washing buffer (Amann et al., 1990) and incubated for 200 at 48 °C. All samples were counter stained with 0.1 lM 40 ,6-diamidino-2-phenylindol (DAPI). CitiFluor™ was used as embedding agent. A Zeiss Axioskop A50 equipped with a mercury HBO 50 UV lamp and an Axiocam camera served for microscopy and photography, respectively. From each sample that was specifically labeled with respective fluorescing gene probes for domain, genus or species detection and counterstained with DAPI 10 randomly chosen microscopic view fields were photographed under the phase contrast microscope with and

3.1. Influence of moisture content of biowaste on start-up and final efficiency of biogas production Assays for dry anaerobic digestion (DAD) of source-sorted municipal biowaste at its original dry matter (DM) content of 30% (w/v) and at 25% and 20% DM content were incubated at room temperature 20 ± 1 °C, 37 ± 0.5 °C and 55 ± 0.5 °C. Biogas production, fatty acids levels and pH were monitored for almost 1 year (Figs. 1–3). In the DAD reactors that were moistened to a DM content of 20% methanogenesis at room temperature started after a lag phase of 10–20 days (Fig. 1a). The bulk amount of biogas was generated in the first 150 days and only little more biogas was produced upon further incubation at 37 °C (reactor 1) or 55 °C (reactor 2). Initially the pH of all reactors decreased below 6.7 and was adjusted to 8 with sodium hydroxide after 5, 10 and 25 days (Fig. 1b). Later on no pH correction was necessary, even though volatile fatty acids (VFA) such as acetate, propionate and n-butyrate were still increasing. Whereas the n-butyrate concentration reached about 10 g L1 and n-butyrate was completely degraded after 70 days, degradation of up to 20 g L1 acetate to the low final steady state level required 150 days (Fig. 1b). Propionate concentrations in the reactors increased to 10–18 g L1 during n-butyrate and acetate degradation and reached their low steady state level of 0.5 ± 0.25 g L1 only after 180 days. These VFA concentrations were in the same range as reported by Zahedi et al. (2013) for thermophilic DAD in

Please cite this article in press as: Li, C., et al. Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.02.118

4

C. Li et al. / Bioresource Technology xxx (2014) xxx–xxx

(a)

0.5

Methane (m3*kg-1ODM)

0.4

0.3

RT

37°C

RT

55°C

0.2

0.1

0 0

50

100

150

200

250

300

Time (days) DM 20%, R1

DM 20%, R1

(b) 80

9

70

8

60

7 6

50

5

40

4

30

3

20

2

10

1

0

pH

VFA (g*L-1)

DM 20%, R2

0

50 Acetate

100 Propionate

150 Time (days) n-Butyrate

200

250 total VFA

300

0

pH

Fig. 1. Biogas production in reactor R1 an R2 (a) at room temperature (RT), pH and volatile fatty acid (VFA) concentrations for R1 ((b) Profiles for R2 were similar) during digestion of biowaste with 20% DM content. Raising the temperature to 37 °C or 55 °C after 220 h (a) did not cause significant more biogas generation or VFA production.

a stirred tank reactor. When the gas production had ceased after 210 days an increase of the reactor temperature to 37 ± 0.5 °C (R1) or 55 ± 0.5 °C (R2, Fig. 1a) did not lead to a significant further degradation and only little more biogas was produced in both reactors. Methane production in both reactors differed by 6% (Fig. 1a), which was acceptable for parallel incubations of complex biowaste and the solid residues of WAD as a source of microorganisms. In the DAD reactors that were moistened to a DM content of 25% methanogenesis at 20 ± 1 °C started only after 140 days and proceeded for about 100 days. The total methane amount reached only about 50% of the methane amount that was obtained in reactors with biowaste that contained 20% DM or biowaste with 25% DM content after raising the temperature to 37 °C (Fig. 2a). At 37 °C methane was produced much faster than at 20 ± 1 °C. When methane production ended in the reactor that was incubated at 37 °C the temperature was further increased to 55 °C. About 15% more methane were produced at the higher incubation temperature (Fig. 2a). As in the reactors with 20% DM content VFA were accumulating initially and the pH was adjusted to 8 three times after 5, 10 and 25 days (Fig. 2b). VFA levels in total or distinguished as acetate, propionate and n-butyrate were similar in both reactors. Acetate, propionate and n-butyrate were accumulating in

the first 60–70 days and then were slowly degraded. n-Butyrate and acetate degradation apparently caused an increase of the propionate concentration, either by inhibition of propionate degradation or by propionate formation during acetate or n-butyrate degradation (Fig. 2b). Methanogenesis in the DAD reactors with biowaste that contained 30% DM did not start within 210 days of incubation at 20 ± 1 °C. Even when the temperature in one of the reactors was raised to 37 °C at day 90 (R2 in Fig. 3a) methanogenesis did not start within the next 40 days. Only when the temperature was raised from 20 ± 1 °C to 55 °C (R1, Fig. 3a) or from 37 °C to 55 °C (R2, Fig. 3a) in both reactors biogas production began. However, only about half of the methane amount (0.18–0.22 m3 kg1 ODM) was finally obtained as compared to the biowaste reactors with 20% DM content (0.35–0.38 m3 kg1 ODM; Figs. 1a, 3a), although more digestible substrate (30% DM instead of 20% DM) was available. The maximum methane production rates in the biowaste reactors with 30% DM content (Fig. 3a) were similar, no matter whether 55 °C were approached from 20 ± 1 °C (R1) in one step or via 40 days at 37 °C (R2). VFA levels were high in the reactor that was incubated at 20 ± 1 °C or at 20 ± 1 °C followed by 37 ± 0.5 °C and only decreased after biogas production began at 55 ± 0.5 °C (Fig. 3b and c).

Please cite this article in press as: Li, C., et al. Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.02.118

5

C. Li et al. / Bioresource Technology xxx (2014) xxx–xxx

(a)

0.5

Methane (m3*kg-1ODM)

0.4

RT 0.3

RT

55°C

37°C

0.2

0.1

0 0

50

100

150

200

250

300

Time (days) DM 25%, R1

DM 25%, R1

80

9

70

8

60

7 6

50

5

40

4

30

3

20

2

10

1

0

pH

VFA (g*L-1)

(b)

DM 25%, R2

0

50 Acetate

100 Propionate

150 200 Time (days) n-Butyrate

250 total VFA

300

0

pH

Fig. 2. Biogas production in reactor R1 (300 d at room temperature) and R2 (room temperature ? 37 °C ? 55 °C, (a)), pH and volatile fatty acid (VFA) concentrations in R1 ((b) Profiles for R2 were similar) during digestion of biowaste with 25% DM content.

3.2. Re-feeding of the reactors After biogas production ceased in the parallel biowaste reactors with 20%, 25% and 30% DM content, 1 kg digestion residue of each reactor was mixed with 1 kg fresh biowaste. The moisture content was re-adjusted to 20%, 25% and 30%. Then one of the 2 biowaste reactors with 20%, 25% and 30% DM-content was incubated at 37 °C and the other at 55 °C, respectively. The pH was adjusted twice after 3 and 12 days, when VFAs accumulated and the pH had dropped far below 7 (Fig. 4b–g). The DAD reactors with undiluted biowaste (30% DM content) produced only about 0.03 m3 methane per kg ODMadded at 37 °C and 0.075 m3 methane per kg ODMadded at 55 °C within 5–30 days of incubation, respectively (Fig. 4a). No more methane was produced after day 30 in both reactors. After 60 days propionate dominated the reactor incubated at 37 °C (Fig. 4f) and acetate the reactor incubated at 55 °C (Fig. 4g). In the DAD reactors that contained biowaste with 25% DM significantly less methane (ca. 60%) was produced at 37 °C than at 55 °C and more propionate remained un-degraded (Fig. 4d and e). In both, mesophilic and thermophilic DAD reactors with biowaste that contained 20% DM and in the thermophilic DAD reactor with biowaste that contained 25% DM 0.35 m3 methane/kg ODMadded

were released and volatile fatty acids almost completely degraded after 60 days (Fig. 4a,b and e). Whereas biogas production during the start-up phase at 20 ± 1 °C in the assays with 20% DM containing biowaste started only after 20 days and continued until day 150 (Fig. 1), gas production after re-feeding and incubation at either 37 °C or 55 °C started almost immediately and ended after only 30–40 days. In both re-feeding assays with 20% DM (Fig. 4b and c) and in the thermophilic assay with 25% DM (Fig. 4e) VFAs were almost completely degraded after 60 days, whereas in both assays with biowaste, that contained 30% DM and in the mesophilic assay with 25% DM content, 10–20 g/L of either acetate (Fig. 4g) or propionate (Fig. 4d and f) accumulated and both acids were not degraded. 3.3. Moisture content, volatile fatty acids and methane production rates The moisture content of biowaste for DAD and the incubation temperature are important factors for anaerobic digestion to proceed at all and for the final efficiency of digestion. Maximal volatile fatty acid (VFA) levels in all DAD biowaste reactors were obtained in the first 50–100 days at 20%, 25% and 30% DM content (Figs. 1b,

Please cite this article in press as: Li, C., et al. Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.02.118

6

C. Li et al. / Bioresource Technology xxx (2014) xxx–xxx

0.5

(a)

Methane (m3*kg-1ODM)

0.4

RT RT

0.3

55°C 55°C

37°C

0.2

0.1

0 0

50

100

150

200

250

300

Time (days) DM 30%, R1

DM 30%, R2

9

80

70

8

70

8

60

7

60

7

50

5

40

4

30

3

20

pH

6 VFA (g*L-1)

VFA (g*L-1)

(c)

DM 30%, R1

80

9

6

50

5

40

4

30

3

2

20

2

10

1

10

1

0

0

0

0

100 Time (days) 200 Acetate total VFA

Propionate pH

300 n-Butyrate

pH

(b)

DM 30%, R2

0

100 Time (days) 200 Acetate total VFA

Propionate pH

300

0

n-Butyrate

Fig. 3. Biogas production in reactor R1 (room temperature ? 55 °C) and R2 (room temperature ? 37 °C ? 55 °C, (a)), pH and volatile fatty acid (VFA) concentrations of R1 (b) and R2 (c) during digestion of biowaste with 30% dry matter (DM) content.

2b, 3b and 4b–g). In biowaste DAD reactors with 20% DM content there was apparently enough bioavailable water for non water limited biogas formation by the established microflora at 20, 37 and 55 °C. In biowaste that contained 25% DM content the bioavailable water at 20 °C was apparently still enough to allow rapid hydrolysis and acidification (Fig. 2b), but the methane production by syntrophic interaction of acetogens and methanogens was significantly delayed and did not proceed to completion (Fig. 2a). Since aW is temperature dependent (Starzak and Mathlouthi, 2006) a temperature shift from 20 °C to 37 °C may have increased the amount of bioavailable water so that methanogenesis in the biowaste reactor with 25% DM content could proceed to completion (Fig. 2a). The water activity in biowaste with 30% DM content apparently still allowed acidification but no longer biogas production. Even the increased water activities at 37 or 55 °C seemed to be not high enough for non water limited methanogenesis as in the 20% DM assays. The dependence of biogas production on bioavailability of water was more clearly apparent from the re-feeding experiment at incubation temperatures of 37 and 55 °C. Whereas at a DM content of the biowaste of 30% only very little methane

was generated at 37 and 55 °C, in the assays with 25% DM content at 37 °C incubation temperature by far not as much methane was produced as in the parallel assay at 55 °C. The increase of the water activity by increasing the incubation temperature might be counteracted by a decrease of the metabolic activity of the bacteria due to the unfavorable high temperatures for the prevalent acetogens and methanogens in the DAD reactor. However it turned out that the inoculum from full-scale WAD, pregrown at 33 °C and digesting biowaste from the same source apparently contained phylogenetically diverse hydrolytic, acidogenic, acetogenic and methanogenic bacteria for growth temperatures ranging from 20 °C to 55 °C. Mesophilic and moderately thermophilic bacteria apparently stayed alive during the long incubation period at 20 °C, where those bacteria were enriched that could grow best at this temperature. Their metabolism could be re-activated at higher water activity by e.g., raising the temperature to 37 or 55 °C. Methane production rates of 0.9–3 L kg1 d1 from biowaste with 20–30% DM (13–19.5% ODM) were calculated for the initial batch DAD assays (chapter 3.1) at 20, 37 and 55 °C (Table 4,

Please cite this article in press as: Li, C., et al. Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.02.118

7

C. Li et al. / Bioresource Technology xxx (2014) xxx–xxx

(a) 0.4 0.35

Methane (m3*kg-1ODM)

0.3

DM20%,37°C

0.25

DM20%,55°C

0.2

DM25%,37°C

0.15

DM25%,55°C DM30%,37°C

0.1

DM30%,55°C

0.05 0

2 0

20

40

60

DM 25%, 37°C

4

20

0

20

40

60

DM 30%, 37°C

80

0

Acetate total VFA

Time (days) Propionate pH

60

20

40

60

DM 25%, 55°C

0

10

6 4 2 0

20

40

60

DM 30%, 55°C

n-Butyrate

0

10 8 6

40

4

20 0

0

8

(g)80

VFA (g*L-1)

pH 2

40

0

60

4

20

2

40

0

6

0

4

60

20

8

20

6

20

2

10

40

10 8

(e)80 VFA (g*L-1)

6

40

DM 20%, 55°C

40

0

8

60 VFA (g*L-1)

0

10

60

0

VFA (g*L-1)

4

60

60 pH

6

20

0

(c)80

8

40

50

pH

10

40

pH

DM 20%, 37°C

(d) 80 VFA (g*L-1)

30 Time (days)

60

0

(f)

20

pH

VFA (g*L-1)

(b) 80

10

pH

0

2

0

20

40

60

0

Time (days) Acetate total VFA

Propionate pH

n-Butyrate

Fig. 4. Biogas production in biowaste reactors with 20%, 25% and 30% DM content (a), pH and volatile fatty acid (VFA) concentrations after re-feeding the reactors at mesophilic (37 °C; b,d and f) or thermophilic (55 °C; c,e and g) temperatures.

lines 1–3). After re-feeding methane production rates were much higher, 5.8 L kg1 d1 at 37 °C and 55 °C in the DAD reactors with biowaste that contained 20% DM and 5.8 L kg1 d1 at 55 °C or 4.3 L kg1 d1 at 37 °C in the reactors with biowaste that contained 25% DM content. At 37 °C the methane production rate in the DAD reactor with 25% DM content was lower (4.3 instead of 5.8 L kg1 d1), presumably due to a too low aW value. Very little methane was produced in any DAD reactor with 30% DM content, although the water activity at 55 °C should be higher

than at 20 or 37 °C. At this high DM content most of the moisture apparently was tightly bound to particles and increasing the temperature could not increase bioavailability far enough for acetogenic and methanogenic bacteria. The temperature shift apparently had a more severe effect on the activity of acetogenic and methanogenic bacteria than on acidogenic activity. The reason for this may be a broader range of still growth allowing temperatures of the biowaste hydrolyzing and VFA-producing bacteria.

Please cite this article in press as: Li, C., et al. Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.02.118

8

C. Li et al. / Bioresource Technology xxx (2014) xxx–xxx

Table 3 Oligonucleotide probes with fluorescent marker (50 -FAM or 50 Cy3). Probe

Target

Formamide (%)

Sequence 50 –30

References

Eub388 Arc915 Mg1200 Mb310 MsMx860 Mx825 Glh821m Synbac824 SmiSR354 SmiLR150

Most bacteria Archaea Methanomicrobiales Methanobacteriales Methanosarcinales Genus Methanosaeta Genus Pelotomaculum Genus Syntrophobacter Syntrophus group incl. Smithella propionica Smithella sp. long rod (LR)

30 30 30 30 30 30 10 10 10 10

GCT GCC TCC CGT AGG AGT GTG CTC CCC CGC CAA TTC CT CGG ATA ATT CGG GGC ATG CTG CTT GTC TCA GGT TCC ATC TCC G GGC TCG CTT CAC GGC TTC CCT TCG CAC CGT GGC CGA CAC CTA GC ACCTCCTACACCTAGCACCC GTA CCC GCT ACA CCT AGT CGC AAT ATT CCT CAC TGC CCT TTC GGC ACG TTA TTC

Amann et al. (1990) Stahl and Amann (1991) Raskin et al. (1994) Raskin et al. (1994) Raskin et al. (1994) Raskin et al. (1994) Narihiro et al. (2012) Ariesyady et al. (2007) Ariesyady et al. (2007) Ariesyady et al. (2007)

Table 4 Methane production rates during dry anaerobic digestion of biowaste at 20, 37 and 55 °C. 20 ± 1 °C

37 ± 0.5 °C

55 ± 0.5 °C

DM%

CH4 L kg1 ODM d1

DM%

CH4 L kg1 ODM d1

DM%

CH4 L kg1 ODM d1

First feeding 21.2 25.2 30.9

1.7 0.9 no gas

21.2 25.2 30.9

n.d. 3.0 no gas

21.2 25.2 30.9

n.d. n.d. 1.7

Re-feeding 20.2 24.9 28.6

n.d. n.d. n.d.

20.2 24.9 28.6

5.8 4.3 neg.

20.2 24.9 28.6

5.8 5.8 neg.

DM = dry matter, ODM = organic dry matter, n.d. = not determined, neg. = negligible. Rates were estimated from Figs. 1a, 2a, 3a and 4a for logarithmic/linear CH4 production phases.

Methane production rates and, similarly important, total biogas yields are the main criteria for either WAD or DAD of organic wastes (de Baere, 2000). A comparison of the maximal biogas productivity of the biowaste fraction of the City of Karlsruhe during WAD and DAD revealed that during WAD of biowaste with 5–6% DM content the same amount of biogas per gram (0.59 m3 kg1 VS; Nayono et al., 2009) at a hydraulic retention time (HRT) of 7 d was generated than during batch DAD of biowaste with 20–25% DM content during an almost ten times longer time span (60 d; 0.53– 0.59 m3 kg1 ODM, this paper). This shows that the final biodegradation efficiency of municipal biowaste with 20% or maximally 25% DM content in box fermenters for DAD may be as good as in completely mixed reactors for WAD with 5–6% DM content. The average methane content in the biogas from WAD was 62–70% (Gallert et al., 2003; Nayono et al., 2009) as compared to DAD, where it was 70–75%, due to a higher pH. The space loading for stable WAD in laboratory and in full-scale was 15 kg m3 d1 for a HRT of 6 days (Gallert et al., 2003) and thus was in the same order as in all the references for DAD mentioned by Zahedi et al. (2013). Even in their own work total volatile solids (=ODM) accumulation began at a HRT of 6.6 days, equivalent to a space loading of 13 kg m3 d1, although methane productivity apparently was still stable. 3.4. Microbial population in the dry anaerobic digesters In parallel mesophilic and thermophilic DAD reactors the total populations (DAPI-stained) and the sub-populations of Eubacteria (Eub388-labeled) and Archaea (Arc915-labeled) were in the same range for biowaste of the same DM content. No unidirectional changes towards higher or lower cell numbers were observed in single reactors within 52 days. Mean values of total bacterial counts (DAPI-stained), eubacterial counts (Eub388-labeled) and archaeal counts (Arc915-labeled) in the DAD reactors containing biowaste with 20%, 25% and 30% DM at 37 °C were 4.1  109, 1.4  109 and 0.5  109 cells/ml at 37 °C and at 55 °C 3.9  109, 1.2  109 and 0.6  109 cells/ml (each ± 0.2), respectively. From the test with the Arc915 gene probe for Archaea it can clearly be

seen that the population of Archaea was about twice as high in the mesophilic and thermophilic DAD reactors with biowaste that contained 20% and 25% DM as compared to the DAD reactors that contained 30% DM (Data not shown). This corresponded well with the data on biogas production (e.g., Fig. 4). Comparison of the Eub388 cell counts in the DAD reactors with those of other studies was difficult, due to different digester regimes. In publications on microbial community analyses by FISH in biowaste digesters, substrates were fed semi-continuously into continuously stirred tank reactors, whereas we used batch ‘‘box-type fermenters’’ without wash-out of cells. The numbers for the Eub388 and Arc915 cell counts in thermophilic DAD reactors, as reported by Zahedi et al. (2013), were ranging from 5.1 to 9.9  109 and 1–2  109 cells/ ml, respectively, for different organic space loading rates and were on average 3.6- to 7-fold (Eub388) or 1.4- to 2.8-fold, respectively, higher than in our batch DAD reactor with 20% DM at 55 °C. Since our DAD reactors were not permanently stirred reduced mass transfer may have been this reason for reduced cell growth. This is the ‘‘reality’’ in all DAD reactors without process water recycling. The used probes covered the archaeal population (Arc915 cell counts) in the DAD biowaste reactors which was further differentiated by probes for all known groups of methanogens except for the Methanococcales (Raskin et al., 1994). Species of Methanosarcinales were the dominating group in all reactors, representing up to 96% of all tested methanogens at 37 or 55 °C (Fig. 5). No organisms reacting with the Mx825 gene probe, specific for the genus Methanosaeta, were found at a detection limit of 1.58  106 cells ml1 of the test system. Up to 10% Methanomicrobiales (gene probe Mg1200/Arc915) were found in biowaste samples of the DAD reactors with 20% or 25% DM content, but almost 20% in biowaste samples of the reactors with 30% DM at 37 or 55 °C (Fig. 5), where much less biogas was produced. Methanobacteriales (Mb310 positive cells) only seemed to be present in the DAD biowaste reactors with 30% DM content at both temperatures, 37 or 55 °C. Except for Methanosarcinales, which were detected by probe MsMx860 (Fig. 5), this probe covers various genera of methane bacteria, belonging to the Methanococcoides, Methanolobus,

Please cite this article in press as: Li, C., et al. Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.02.118

9

C. Li et al. / Bioresource Technology xxx (2014) xxx–xxx

160.0%

Archaea (Arc915) =100%

140.0% 120.0% 100.0% 80.0% 60.0% 40.0% 20.0% 0.0%

DM30% t

DM30% m

Methanomicrobiales

DM25% t

DM25% m

Methanobacteriales

DM20% t

DM20% m

Methanosarcinales

Fig. 5. Proportion of major methanogenic taxa (% Methanosarcinales, Methanobacteriales and Methanomicrobiales) in DAD reactors with biowaste that contained 30%, 25%, and 20% DM at mesophilic (m, 37 °C) and thermophilic (t, 55 °C) temperature after 21 days.

Methanohalophilus as well (Raskin et al., 1994). As we did not get a positive reaction with probe Mx825, specific for Methanosaeta spec., we concluded that Methanosaeta spec. do not play a significant role in the biowaste reactors. This is corroborated by high acetate levels at the sampling time in conjunction with reports that Methanosarcina spec. are the dominant acetate degrading methanogens in systems with high acetate levels (Hori et al., 2006). The absence of detectable numbers of Methanosaeta spec. may be the reason for the high acetate levels in biowaste DAD reactors. In recent reports on DAD of food waste, where a dominance of Methanosaeta spec. was found, acetate concentrations were much lower (Chu et al., 2010; Montero et al., 2009). Chu et al. (2010) for instance compared mesophilic and thermophilic digestion of food waste and found that 72% of the mesophilic methanogens belonged to Methanosaeta concilii and 98% of the thermophilic methanogens belonged to Methanosaeta thermophila. Montero et al. (2009) also observed high percentages of thermophilic Methanosaeta spec. at high OLRs. For mesophilic DAD of food waste Cho et al. (2013) observed a decrease of the phylogenetic diversity of acetate utilizing methanogens to finally 96.4–99.1% of M. termophila. In one recent publication a dominance of Methanosarcina spec., identified by q-PCR, at low acetate levels in DAD reactors was reported (Abbassi-Guendouz et al., 2013). High

acetate levels above 8 g L1 in WAD may exert a strong inhibitory effect on both, Methanosarcina spec. and Methanosaeta spec. (McMahon et al., 2004). Although acetate levels above 8 g L1 were reached in our DAD reactors after only 21 d and should have prevented methanogenesis, inhomogeneous spaces with lower concentrations of the inhibiting acetic acid and a higher pH might have been responsible for the slowly increasing methanogenic activity, as argued similarly by Abbassi-Guendouz et al. (2013). Hydrogenotrophic methanogens usually were dominant after start of DAD at high VFA levels (Montero et al., 2009). A dominance of Methanobacteriales was reported in reactors with both, a high VFA content and a low pH (Blume et al., 2010), whereas in reactors with lower VFA levels Methanomicrobiales apparently dominated (Garrity and Holt, 2001). This was in agreement with our results. Among the propionate oxidizing bacteria (POB) in the re-feeded mesophilic assays members of the genus Syntrophobacter (gene probe Synbac824) and Pelotomaculum (gene probe Glh821m) played the major role, whereas members of the Smithella group, including the genus Smithella propionica (Gene probes SmiSR354 and SmiLR150) could not be found (Fig. 6). POP are essentially required members of the anaerobic food chain for syntrophic degradation of fatty acids with a carbon chain longer than C2 (e.g., Gallert and Winter, 2005; Felchner-Zwirello et al., 2013). In

Eubacteria (Eub388)=100%

9.0% 8.0% 7.0% 6.0% 5.0% 4.0% 3.0% 2.0% 1.0% 0.0%

DM30%

DM25%

DM20%

Fig. 6. Percentage of POB Taxa that were detected with gene probes GIh821m (Genus Pelotomaculum) and Synbac824 (Genus Syntrophobacter) in the mesophilic DAD reactors with 30%, 25% and 20% DM during incubation (% POB in relation to the total population).

Please cite this article in press as: Li, C., et al. Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.02.118

10

C. Li et al. / Bioresource Technology xxx (2014) xxx–xxx

the reactor with 30% DM containing biowaste less than 0.8% (exception Syntrophobacter at day 21) of the Eubacteria were propionate degraders. In the reactor containing biowaste with 25% DM the percentage of Pelotomaculum spec. was decreasing with time whereas the percentage of Syntrophobacter spec. increased as long as propionate was available, and then decreased. Only in the reactor containing biowaste with 20% DM a slightly increasing population of Syntrophobacter spec. and a significantly increasing population of Pelotomaculum spec. with time was seen (Fig. 6). Generally little is known about anaerobic propionate oxidizing bacteria (POB) in DAD reactors, although propionate is a major intermediate in anaerobic digestion and can accumulate easily in unbalanced fermenters, e.g., during start up (Gallert and Winter, 2008). In a recent study Zahedi et al. (2013) found 15% and 6% POB at an increasing OLR in samples from thermophilic DAD, by applying FISH and the Synbac824 probe, respectively. No thermophilic species of POB, identified with gene probe Synbac824 were described as yet (Li et al., 2012), but may exist (Zahedi et al., 2013) and would represent new species. Although in our approach for POB identification all available gene probes for known POB were used, the overall proportion of POBs in the reactors was small. This may have been the reason why the propionate concentrations in the DAD reactors were higher than 0.06–1.3 g L1 as reported by Zahedi et al. (2013). Propionate concentrations also correlated well with the absolute numbers of POB in a negative linear relationship (data not shown), indicating that the higher the number of POBs the lower the propionate concentration was. Thus POB are essential to prevent acidification by high VFA contents. 4. Conclusions Dry anaerobic digestion at 20–55 °C was possible with 20% DMcontaining biowaste. With 25% DM-containing biowaste biogas production in the digestion assay at 37 °C was restricted and incomplete, whereas it proceeded to completion in the 55 °C assay. The water activity was apparently too low at 37 °C in biowaste with 25% DM content. No functioning DAD of biowaste with 30% DM content was obtained. Methanosarcinales were the dominant acetate-degrading, Methanomicrobiales the dominant H2/CO2 converting methanogens, Pelotomaculum and Syntrophobacter species the dominant propionate oxidizing bacteria. No members of the Smithella group were detected. Acknowledgement We thank the City Authorities of Karlsruhe, Amt für Abfallwirtschaft for providing separately collected biowaste and inocula from the full-scale digester. This work was financially supported by a Grant of Deutsche Forschungsgemeinschaft (DFG), Grant No. Ga-546/4-2. References Abbassi-Guendouz, A., Brockmann, D., Trably, E., Dumas, C., Delegenés, J.-P., Steyer, J.-P., Escudié, R., 2012. Total solids content drives high solid anaerobic digestion via mass transfer limitation. Bioresour. Technol. 111, 55–61. Abbassi-Guendouz, A., Trably, E., Hamelin, J., Dumas, C., Steyer, J., Delegenés, J., Escudié, R., 2013. Microbial community signature of high-solid content methanogenic ecosystems. Bioresour. Technol. 133, 256–262. Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R., Stahl, D.A., 1990. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56 (3), 1919–1925. APHA, AWWA, WEF, 2005. Standard Methods for the Examination of Water and Wastewater, 21st ed. Washington, DC, New York, USA. Ariesyady, H., Ito, T., Yoshiguchi, K., Okabe, S., 2007. Phylogenetic and functional diversity of propionate-oxidizing bacteria in an anaerobic digester sludge. Appl. Microbiol. Biotechnol. 75, 673–683.

Blume, F., Bergmann, I., Nettmann, E., Schelle, H., Rehde, G., Mundt, K., Klocke, M., 2010. Methanogenic population dynamics during semi-continuous biogas fermentation and acidification by overloading. J. Appl. Microbiol. 109, 441–450. Bolzonella, D., Pavan, P., Mace, S., Cecchi, F., 2006. Dry anaerobic digestion of differently sorted organic municipal waste: full-scale experience. Water Sci. Technol. 53 (8), 23–33. Cecchi, F., Pavan, P., Mata-Alvarez, J., Bassetti, A., Cozzolino, C., 1991. Anaerobic digestion of municipal solid waste: thermophilic vs mesophilic performance at high solids. Waste Management Res. 9, 305–315. Cho, S., Im, W., Kim, D., Kim, M., Shin, H., Oh, S., 2013. Dry anaerobic digestion of food waste under mesophilic conditions: performance and methanogenic community analysis. Bioresour. Technol. 131, 210–217. Chu, C., Ebie, Y., Xu, K., Li, Y., Inamori, Y., 2010. Characterization of microbial community in the two-stage process for hydrogen and methane production from food waste. Int. J. Hydrogen Energy 35, 8253–8261. Daims, H., Lücker, S., Wagner, M., 2006. Daime, a novel image analysis programe for microbial ecology and biofilm research. Environ. Microbiol. 8, 200–213. De Baere, L., 2000. Anaerobic digestion of solid waste: state of the art. Water Sci. Technol. 41 (3), 283–290. Felchner-Zwirello, M., Winter, J., Gallert, C., 2012. Mass spectrometric identification of 13C-labeled metabolites during anaerobic propionic acid oxidation. Chem. Biodivers. 9 (2), 376–384. Felchner-Zwirello, M., Winter, J., Gallert, C., 2013. Interspecies distances between propionic acid degraders and methanogens in syntrophic consortia for optimal hydrogen transfer. Appl. Microbiol. Biotechnol. 97, 9193–9205. Fernandez-Güelfo, L.A., Álvarez-Gallego, C., Sales, D., Romero, L.I., 2011. The use of thermochemical and biological pretreatments to enhance organic matter hydrolysis and solubilization from organic fraction of municipal solid waste (OFMSW). Chem. Eng. J 168, 249–254. Fernàndez-Rodriguez, 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. Gallert, C., Winter, J., 2005. Bacterial metabolism in wastewater treatment systems. In: Jördening, H.J., Winter, J. (Eds.), Environmental Biotechnology – Concepts and Application. Wiley-VCH Weinheim, pp. 1–48. Gallert, C., Winter, J., 2008. Propionic acid accumulation and degradation during start-up of anaerobic biowaste digestion. Bioresour. Technol. 99, 170–178. Gallert, C., Henning, A., Winter, J., 2003. Scale-up of anaerobic digestion of the organic fraction from domestic wastes. Water Res. 37 (1433–1441), 2003. Garrity, G.M., Holt, J.G., 2001. Euryarchaeota. In: Boone, D.R., Castenholz, R.W. (Eds.), Bergeys Manual of Systematic Bacteriology, vol. 1. Springer, New York, pp. 211– 354. 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 a thermophilic anaerobic digester. Appl. Environ. Microbiol. 72, 1623–1630. Li, J., Ban, Q., Zhang, L., Jha, A., 2012. Syntrophic propionate degradation in anaerobic digestion: a review. Int. J. Agric. Biol. 14, 843–850. Lissens, G., Vandervivere, P., De Baere, L., Bley, E.M., Verstraete, W., 2001. Solid waste digestors: process performance and practice for municipal solid waste digestion. Water Sci. Technol. 44 (8), 91–102. Luning, L., van Zundert, E.H.M., Brinkann, A.J.F., 2003. Comparison of dry and wet digestion for solid waste. Water Sci. Technol. 48 (4), 15–20. Mata-Alvarez, J., Cecchi, F., Pavan, P., Bassetti, A., 1993. Semi-dry thermophilic anaerobic digestion of fresh and pre-composted organic fraction of municipal solid wastes: digester performance. Water Sci. Technol. 27, 87–96. McMahon, K., Theng, D., Stams, A., Mackie, R., Raskin, L., 2004. Microbial population dynamics during start up and overload conditions of anaerobic digesters treating municipal solid waste and sewage sludge. Biotechnol. Bioeng. 87, 823– 834. Montero, B., García-Morales, J., Sales, D., Solera, R., 2009. Analysis of methanogenic activity in a thermophilic-dry anaerobic reactor: use of fluorescent in situ hybridization. Waste Manage. 29, 1144–1151. Narihiro, T., Terada, T., Ohashi, A., Kamagata, Y., Nakamura, K., Sekiguchi, Y., 2012. Quantitative detection of previously characterized syntrophic bacteria in anaerobic wastewater treatment systems by sequence-specific rRNA cleavage method. Water Res. 46, 2167–2175. Nayono, S.E., Gallert, C., Winter, J., 2009. Foodwaste as a co-substrate in a fed-batch anaerobic biowaste digester for constant biogas supply. Water Sci. Tech. 59 (6), 1169–1178. Pavan, P., Battistoni, P., Mata-Alvarez, J., Cecchi, F., 2000. Performance of thermophilic semi-dry anaerobic digestion process changing the feed biodegradability. Water Sci. Technol. 41, 75–82. Raskin, L., Stromley, J., Rittmann, B., Stahl, D., 1994. Group-specific 16S rRNA hybridization probes to describe natural communities of methanogens. Appl. Environ. Microbiol. 601, 232–240. Rockland, L.B., Beuchal, L.R., 1987. Water Activity: Theory and Applications to Food, 2nd ed. Marcell Dekker, New York. Stahl, D., Amann, R., 1991. Development and application of nucleic acid probes. In: Stackebrandt, E., Goodfellow, M. (Eds.), Nucleic Acid Techniques in Bacterial Systematics. John Wiley & Sons Inc, NewYork, N.Y., pp. 205–248. Starzak, M., Mathlouthi, M., 2006. Temperature dependence of water activity in aqueous solutions of sucrose. Food Chem. 9 (3), 346–370. Zahedi, S., Sales, D., Romero, L., Solera, R., 2013. Optimization of single-phase dry-thermophilic anaerobic digestion under high organic loading rates of industrial municipal solid waste: population dynamics. Bioresour. Technol. 146, 109–117.

Please cite this article in press as: Li, C., et al. Effect of moisture of municipal biowaste on start-up and efficiency of mesophilic and thermophilic dry anaerobic digestion. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.02.118