Bioresource Technology 146 (2013) 619–627
Contents lists available at ScienceDirect
Bioresource Technology
journal homepage: www.elsevier .com/locate /bior tech
Anaerobic digestion of municipal solid waste composed of food waste,wastepaper, and plastic in a single-stage system: Performance andmicrobial community structure characterization
0960-8524/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.biortech.2013.07.140
⇑ Corresponding author. Tel./fax: +86 592 6190769.E-mail address: [email protected] (W. Luo).
Shungang Wan a, Lei Sun a,b, Yaniv Douieb c, Jian Sun a, Wensui Luo a,⇑a Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, PR Chinab Guangxi University, School of Environment, Nanning 530004, PR Chinac AgroParisTech, 16 rue Claude Bernard, F-75231 Paris, France
h i g h l i g h t s
� Anaerobic co-digestion food waste, wastepaper and plastic were examined.� Stable anaerobic digestion of food waste, wastepaper and plastic was achieved.� The accumulation of ammonium and free ammonia does not inhibit anaerobic process.� Significant microbial shift was observed during the anaerobic process.� Co-digestion food waste, wastepaper and plastic were feasible.
a r t i c l e i n f o
Article history:Received 3 June 2013Received in revised form 25 July 2013Accepted 29 July 2013Available online 6 August 2013
Keywords:Anaerobic co-digestionMunicipal solid wasteMicrobial diversityDGGE
a b s t r a c t
The performance of municipal organic solid waste anaerobic digestion was investigated using a single-stage bioreactor, and the microbial community structures were characterized during the digestion. Theresults showed that the biogas and methane production rates were 592.4 and 370.1 L/kg with volatilesolid added at the ratio of 2:1:1 for food waste, wastepaper, and plastic based on dry weight. The meth-ane volume concentration fluctuated between 44.3% and 75.4% at steady stage. Acetic acid, propionicacid, and butyric acid were the major volatile fatty acids produced during the digestion process. Theanaerobic process was not inhibited by the accumulation of ammonia and free ammonia. The bacterialcommunity was found to consist of at least 21 bands of bacteria and 12 bands of archaea at the steadystate. All of the results indicated that the mixture of food waste, wastepaper, and plastic could be effi-ciently co-digested using the anaerobic digestion system.
� 2013 Elsevier Ltd. All rights reserved.
1. Introduction
In China, 164.0 million tons of municipal solid waste (MSW)were collected and transported nationwide in 2011 (China,2012). However, the MSW in China shows distinct compositionalcharacteristics compared with that in developed countries: foodwaste, instead of paper, accounts for the largest fraction (50%) ofMSW, and the moisture levels are significantly higher (typicallyaround 50%, compared with 20–30% in the United States and Euro-pean countries) (Cheng and Hu, 2010). Meanwhile, the organicfraction of municipal solid waste (OFMSW) is collected and treatedalong with other non-organic fractions of MSW through landfilland incineration, which account for 79% and 18% of treatedMSW, respectively (Liu et al., 2012). Traditional landfill treatment
can cause problems such as the generation of heavily pollutedleachates (Renou et al., 2008) and the emission of volatile organiccompounds and odors (Gonzáleza et al., 2013), which present a sig-nificant threat to public health and the environment. From thestandpoint of energy recovery, the OFMSW is not composed of haz-ardous materials but of organic material for energy production. Apromising alternative to landfill or incineration of the OFMSW isto apply an anaerobic digestion process for simultaneous wastetreatment and renewable energy production.
A number of anaerobic biological systems, such as the single-and two-stage anaerobic bioreactors, have been adapted anddeveloped for the treatment of OFMSW (Bouallagui et al., 2005).The single-stage systems have relatively simple designs and areeasy to build and operate. Thus, 90% of full-scale plants in Europerelies on single-stage systems for the anaerobic digestion of organ-ic waste (Forster-Carneiro et al., 2008). During the anaerobic diges-tion, organic substrates are decomposed in the absence of oxygen
620 S. Wan et al. / Bioresource Technology 146 (2013) 619–627
via enzymatic and bacterial activities, in which various microbialprocesses including hydrolysis, fermentation (or acidogenesis),acetogenesis, and methanogenesis occur simultaneously in a singledigester (van Haandel and van der Lubbe, 2007).
Numerous environmental factors affect the performance of sin-gle-stage anaerobic digesters, such as low pH, ammonia inhibition,and the accumulation of volatile fatty acids (VFAs). The accumula-tion of ammonia, particularly, could inhibit the activity of microor-ganisms during the digestion of high-nitrogen organic waste, evenat loading rate of 2.0 g volatile solid/L/day (Angelidaki and Ahring,1993; Chen et al., 2008; El-Mashad et al., 2008). Previous studysuggested that by adjusting parameters such as substrate concen-tration and initial solid loading rate, it could improve the perfor-mances of anaerobic digestion (Fernández et al., 2008). Mixingfood waste with other organic solid waste such as dairy manure(Li et al., 2010) and green waste (Liu et al., 2009) also improvedthe digestion of organic wastes. However, no study has evaluatedthe biogas production potential of OFMSW composing of mixtureof food waste, wastepaper, and plastic in a single-stage anaerobicreactor.
The objective of this study is to investigate the feasibility andperformance of anaerobic co-digestion of food waste, wastepaper,and plastic in a small-scale and mid-scale single-stage semi-dryanaerobic digester, respectively. The variation of biogas productionand composition, VFA concentration, pH values, free ammonia, andsoluble organic matter concentration were evaluated in detail. Thestructure and diversity of the microbial community of the biogasresidue slurry were investigated by polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) analysis.
2. Methods
2.1. Characterization of feed stocks
The food waste was obtained from a local canteen in XiamenCity, Fujian, China. The wastepaper and plastic garbage bags wereprovided by a waste recycling site, and were crushed to less than50 mm. The sludge was employed as inoculums, and it was col-lected from a local wastewater treatment plant in Xiamen City, Fuj-ian, China. The basic characterization of food waste, wastepaper,plastic, and sludge is shown in Table 1. Total solids (TS) contentand volatile solids (VS) content were 18.9% and 90.1% for foodwaste, 91.0% and 90.7% for wastepaper, 99.3% and 69.7% for plastic,and 23.4% and 89.3% for sludge, respectively. All chemicals usedwere of analytical grade.
2.2. Co-digestion system
Two different scales experimental apparatus with single stageanaerobic digester system were employed, respectively. Thesmall-scale anaerobic reactors system consisted in a 1.5 L serumbottle closed with a thick rubber cap and sealed with silicone glue.Three outlets were perforated on the cap: one outlet was con-
Table 1Basic characterization of feedstock.
Item Type of raw materials
Food waste Wastepaper Plastic Sludge
Total solid (wt%) 18.9 91.0 99.3 23.4Volatile solid (wt%) 90.1 90.7 69.7 89.3Carbon, C (%) 41.1 41.5 62.0 19.8Nitrogen, N (%) 3.4 0.24 0.17 2.2Sulfur, S (%) 0.92 0.18 0.23 2.24C/N ratio 12.0 172.8 364.9 9.0
nected to a water displacement system to measure biogas produc-tion. An erlenmeyer flask was connected between the serum bottleand water displacement system to prevent liquid exchange. Theother two outlets were used for liquid samples collection. Reactorswere kept at thermostatic bath at 37 �C. The horizontal mid-scaleanaerobic reactor system consisted of a main reactor body, a tem-perature control unit, a gas–liquid separator, a gas purificationunit, a wet biogas meter, and a biogas analyzer in situ. The mainbody of the digester was made of opaque rigid polypropylene witha total volume of 500 L. The waste treatment unit consisted of astirrer to mix the feed stocks, a jacket to maintain the temperatureconstant, a discharge port, a feed inlet, a sample port, and two bio-gas exports. The temperature control unit kept the temperature ofbioreactor constant by circulating the water between the waterfeed tank and the jacket. The biogas and condensing water vaporwere separated through the gas–liquid separator, and then the bio-gas flowed to the purification union to remove the hydrogen sul-fide produced. Moreover, a wet biogas meter and a biogascollector were connected in turn to the purification unit. The co-di-gester was operated under mesophilic conditions (T = 38 ± 1 �C) bycontrolling the temperature of the feed tank.
2.3. Experiment design
Small-scale experiments were employed to investigate the fea-sibility of the anaerobic co-digestion of food waste, wastepaper,and plastic at mixing ratio 2:1:1 (bench one) and 1:2:1 (benchtwo), respectively, in small-scale apparatus. The set-up was as fol-lows: 50 g of food waste, 25 g of wastepaper, and 25 g of plastic forbench one; and 25 g of food waste, 50 g of wastepaper, and 25 g ofplastic for bench two on dry basis were placed into the 1.5 L serumbottles as anaerobic reactors, respectively. Then, 300 mL anaerobicdigester leachate as inoculum and 480 mL deionized water weretransferred into the serum bottles to obtain final TS of 10%. ThepH values were adjusted to approximately 7.0 by adding 1 M NaOHsolution, and then calcium carbonate at 0.8% of the total weightwas added as pH buffer. Triplicate preparations were performedfor each serum bottle in the study.
The mid-scale experiment was conducted based on the resultsof the small-scale experiments. The mixture of 15 kg food waste,7.5 kg wastepaper, and 7.5 kg plastic on dry basis were added tothe anaerobic digester (TS: 10%). The activated sludge was selectedas inoculum, and the amount was calculated as 3.5 kg on dry basis.After adding the required amounts of substrate and inoculum, thedigester was further filled up to 300 kg with aeration dechlorina-tion of tap water. The pH value was adjusted to about 7.0, and thencalcium carbonate at 0.8% of total weight was added as pH buffer.Finally, the digester was started after closing the feed inlet andopening the valve of the exhaust pipe.
2.4. Analytical method
During the anaerobic co-digester run, the biogas yield, compo-sition of biogas, COD, ammonia-N, short chain VFA, and pH valueswere measured daily. The biogas yield was determined by using awet gas meter, and the biogas composition was analyzed with aninfrared methane gas in situ analyzer (GASBOARD-3200L, WuhanSifang Instrument Factory, China). The VFA value was determinedvia ion chromatography (Model ICS-900, Dionex, USA). The TS,VS, COD, and ammonium-N of the digested samples were deter-mined according to Standard Methods (APHA, 1998). The pH valueswere measured with a pH meter (PHB-4, Shanghai INESA ScientificInstrument Co., Ltd, China) for biogas slurry without any dilution.
S. Wan et al. / Bioresource Technology 146 (2013) 619–627 621
2.5. Calculation
The removal of TS and VS were calculated according to Eqs. (1)and (2) as follows:
REð%Þ ¼ TSinitial � TSend
TSinitial� 100 ð1Þ
REð%Þ ¼ VSinitial � VSend
VSinitial� 100 ð2Þ
where RE is the removal efficiency of TS, and TSinitial (%) and VSinitial
(%) are the concentration of total solids and volatile solids on the0th day, respectively. TSend (%) and VSend (%) are the concentrationof total solids and volatile solids at the end of the anaerobic diges-tion, respectively.
Ammonia equilibrium in aqueous solution is pH- and tempera-ture-dependent, and free ammonia concentration is expressed inEq. (3) (El-Mashad et al., 2004), as follows:
½NH4 � N� ¼NHþ4 � N� �
1þ 10pKa�pH ð3Þ
where [NH3–N] is the free-ammonia concentration in mg/L, andNHþ4 �N� �
is the ammonium concentration in mg/L. Meanwhile,pKa is a decreasing function of the absolute temperature T (Kelvintemperature) in the range of 273–373 K, and the value can be ex-pressed as a function of temperature T by Eq. (4) (Poggi-Varaldoet al., 1997), as follows:
pKa ¼ 0:1075þ 2725=T ð4Þ
By combining Eqs. (3) and (4), the free ammonia concentrationcan be determined.
2.6. Microbiological analysis
The pellets were collected after centrifugation of the biogas res-idue slurry at 11,000 rpm for 7 min. DNA was extracted from a 0.5 gsample of the pellet samples using the MOBIO UltraClean� soil DNAisolation kit (Catalog No. 12800-50) following the manufacturer’sinstructions. The variable V3 region of 16S rDNA of bacteria wasamplified using reverse primer (519R, 50-CGTATTACCGCGGCTGCTGG-30) and forward primer (27F, 50-AGAGTTTGATCCTGGCTCAG-30) with a GC clamp (50- CGCCCGGGGCGCGCCCCCGGGCGGGGCGGGGGCACGGGG-30) attached to the 50 termini (Luo et al.,2008). PCR amplification was performed using a PCR Thermo Cycler(JC-96, China) at a final volume of 25 lL containing 12.5 lL Master-Mix (Tiangen Biotech Co., Ltd., China), 0.5 lL of each primer, 1 lL ofDNA, and 10.5 lL of ddH2O. The thermal cycle was performed usingan initial denaturizing step of 5 min at 94 �C, followed by 10 cyclesof 30 s at 94 �C, 45 s at 65 �C, and 72 �C for 1 min. This step was fol-lowed by 20 cycles at 94 �C for 30 s, 55 �C for 45 s, and 72 �C for1 min, and a final extension at 72 �C for 7 min. In the PCR amplifica-tion for archaea, the forward primer (21F, 50-TCCGGTTGATCCYGCC-30) and reverse primer (958R, 50-YCCGGCGTTGAMTCCAATT-30)were used as the first PCR amplification, and then the reverse pri-mer (519R, 50-CGTATTACCGCGGCTGCTGG-30) and forward primer(340F, 50-AGAGTTTGATCCTGGCTCAG-30) with a GC clamp (50-CGCCCGGGGCGCGCCCCCGGGCGGGGCGGGGGCACGGGG -30) wereused as the second PCR amplification in the aforementioned PCRcomposition. For the first PCR amplification, the thermal cyclewas performed through an initial denaturation at 94 �C for 5 min,followed by 20 cycles of 30 s at 94 �C, 30 s at 52 �C, and 72 �C for1 min, with an extension at 72 �C for 7 min. For the second PCRamplification, the PCR products were used as a template, and thethermal cycle was performed through initial denaturation at 94 �Cfor 5 min, followed by 35 cycles of 30 s at 94 �C, 30 s at 52 �C, and
72 �C for 1 min, with an extension at 72 �C for 7 min. The PCR prod-ucts were cloned and then further sequenced by the DNA Sequenc-ing Facility, Shanghai Majorbio Bio-Pharm Technology Co., Ltd. Allgene sequences were first aligned, and phylogenetic constructionwas performed using the maximum likelihood method with a Kim-ura 2-parameter model and 1000 bootstrap replications using thefree MEGA 5.0 software. In succession, the sequences were classi-fied using the Ribosomal Database Project classifier online softwareat an 80% confidence threshold.
3. Results and discussion
3.1. Small-scale experiments
The small-scale experiments lasted for 30 days, and the resultsare shown in Fig. 1. The daily biogas production abruptly in-creased from 310 mL/L/d to 1063 mL/L/d for bench experimentone (food waste: wastepaper: plastic = 2:1:1), and from 233 mL/L/d to 860 mL/L/d for bench experiment two (food waste: waste-paper: plastic = 1:2:1) (Fig. 1a). Then, the daily biogas productiongradually decreased to 73 mL/L/d and 177 mL/L/d with increasingdigestion time, respectively. The average biogas production rateswere 303 mL/L/d and 373 mL/L/d for bench one and bench two,respectively. In addition, the methane content exceeded 55% dur-ing the steady-state phase. As shown in Fig. 1b, after 30 days ofanaerobic digestion, the COD concentration decreased from11,771 mg/L to 1077 mg/L for bench one and from 5886 mg/L to1103 mg/L for bench two. During the digestion process, pH levelsremained at optimal levels. Fig. 1c showed that the main com-pounds produced were acetic and propionic acid, and they firstlyincreased and then gradually decreased with increasing of diges-tion time for both mixing ratios. In addition, formic, butyrate andlactic acids were also detected at very low concentrations com-pared with acetic and propionic acid. Fig. 1c also demonstratedthat more VFA were produced for bench one than bench two un-der the same anaerobic conditions. For ammonia, the maximumconcentration of 300 mg/L and 220 mg/L was achieved on day 6for bench one and 11 for bench two, respectively (Fig. 1d), andammonia concentration remained under 0.3 g/L, which is notinhibitory for both benches. All the results demonstrated no neg-ative effects on the anaerobic process were found in small-scalestudy, especially for bench one. It was feasible for anaerobic co-digestion of food waste, wastepaper, and plastic in the small-scaleexperiments. Given that food waste accounts for no less than 50%of the actual MSW in China, 2:1:1 mass ratio for food waste,wastepaper, and plastic was selected in the following mid-scaleexperiments.
3.2. Biogas yield and composition of mid-scale experiment
The temporal evolution of biogas and methane produced perday is presented in Fig. 2a. After feeding the feed stocks, the dailybiogas production increased with an increase in digestion time,and the maximum biogas yield was 809 L/d on the 17th day. Thereason is that the organic matter provided by food waste rapidlydecomposed, and carbon dioxide was produced in the presenceof oxygen. Thus, at the beginning of the 14 days, the hydrolysisand fermentation of feed stocks were dominant, and the amountof biogas produced was extremely low. The biogas production be-gan to gradually decrease from the 37th day, until it reached 56 L/dafter 50-day digestion. Compared with biogas production, themethane production had a similar trend, as shown in Fig. 2a. Themethane production always fluctuated approximately below100 L/d in the first 15 days, which might have been caused bythe accumulation of VFA and the inhibition of microorganism
0
200
400
600
800
1000
1200
0
2000
4000
6000
8000
10000
12000
0
1500
3000
4500
6000
7500
9000
10500
12000
0
2
4
6
8
10
0 5 10 15 20 25 300
5
10
15
20
25
30
0 5 10 15 20 25 300
50
100
150
200
250
300
Food:Paper:Plastic=2:1:1 Food:Paper:Plastic=1:2:1
Cum
ulat
ive
gas
prod
uctio
n (m
L)
Bio
gas
Prod
uctio
n (m
L/L/
day)
(a)Food:Paper:Plastic=2:1:1Food:Paper:Plastic=1:2:1
CC
OD
(mg/
L)
pH
(b)
Acetic acidFood:Paper:Plastic=2:1:1
Propionic acidFood:Paper:Plastic=2:1:1
Acetic acidFood:Paper:Plastic=1:2:1
Propionic acidFood:Paper:Plastic=1:2:1
Con
cent
ratio
n (m
M)
(c)
Food:Paper:Plastic=2:1:1Food:Paper:Plastic=1:2:1
Time (day)Time (day)
Am
mon
ia (m
g/L)
(d)
Fig. 1. Performance of the anaerobic co-digestion of food waste, wastepaper, and plastic in the small-scale experiments: (a) evolution of biogas yield per day and the biogasaccumulation, (b) evolution of COD concentration and pH values, (c) accumulation of VFA concentrations and (d) accumulation of ammonia concentration.
0
200
400
600
800
1000
0
5000
10000
15000
20000
0 5 10 15 20 25 30 35 40 45 500
20
40
60
80
100
Volu
me
(L/d
)
(a) Biogas production rate Methane production rate
Volu
me
(L)
Cumulative biogas production Cumulative methane production
(b)
CO2 CH4
Bio
gas
com
posi
tion
(%) (c)
CH4+CO2
Time (day)
Fig. 2. Evolution of biogas yield per day and the accumulation of biogas during theanaerobic co-digestion in the mid-scale experiment.
622 S. Wan et al. / Bioresource Technology 146 (2013) 619–627
activity. From day 16, the methane production gradually increasedto the maximum methane rate of 556.8 L/d on the 37th day andthen slowly decreased to 39.8 L/d on the 50th day.
As shown in Fig. 2b, the cumulative biogas and methane pro-duction slowly increased from day 0 to day 14, and then almost lin-early increased from day 15 to day 42. In addition, the biogasproduction rates at day 15 to day 42, which were based on theworking effective volume of the digester, changed between0.55 L/L/d and 2.6 L/L/d. Most of the biogas production rates inour study were higher than those reported in the literature, whichwere 1.42–1.48 for single food waste and 2.18–2.56 for the mixtureof food waste and dairy manure at a 1:1 ratio (Li et al., 2010). In ourstudy, the final total biogas and methane yields were respectively592.4 L/Kg and 370.1 L/Kg with volatile solid added, which alsoindicated that the single-stage anaerobic digestion performed wellin the co-digesting the mixture of food waste, wastepaper, andplastic. Meanwhile, the evolution of biogas composition is shownin Fig. 2c, which demonstrates that methane and carbon dioxideare the main components of the biogas. For methane, the concen-tration always gradually increased from 1.36% to 60.2% as co-digestion time increased from day 1 to day 19, and then the con-centration slightly increased further to 75.2% after 44 days of co-digestion. Compared with methane, the carbon dioxide concentra-tion abruptly increased from 74.1% to 87.3% as co-digestion timeincreased from day 1 to day 2. As co-digestion time increased today 30, the concentration gradually decreased to 29.8%, and thenthe concentration fluctuated between 23% and 30% as co-digestionfurther increased.
3.3. Soluble COD and VFA variation of mid-scale experiment
As shown in Fig. 3a, the soluble COD of the solution abruptly in-creased from 11,711 mg/L to 24,342 mg/L in the beginning of day 5
0 5 10 15 20 25 30 35 40 45 500
5000
10000
15000
20000
25000
30000
35000
40000
CO
D (m
g/L)
Time (day)
(b)
0 5 10 15 20 25 30 35 40 45 500
50
100
150
200
250
300 Acetic acidPropionic acidButyric acidPyruvic acid Total VFA
Con
cent
ratio
n (m
M)
Time (day)
(a)
Fig. 3. Evolution of soluble COD concentration (a) and VFA and total VFA (b) of the solution during the anaerobic co-digestion in the mid-scale experiment.
0
2
4
6
8
0 5 10 15 20 25 30 35 40 45 50
0
200
400
600
800
1000
1200
1400
1600 Ammonium Free ammonia pH
Time (day)
Am
mon
ia (m
g/L)
0255075100125150175200225250
Free
am
mon
ium
(mg/
L)
pH
Fig. 4. Variation of ammonia-N, free ammonia concentration, and pH values.
S. Wan et al. / Bioresource Technology 146 (2013) 619–627 623
of the co-digestion period. The data demonstrated that the easilydecayed solid organic matter was rapidly decomposed by themicroorganisms. After 11 days of digestion, soluble COD reachedthe highest value of 36,057 mg/L. Successively, the soluble CODabruptly decreased to 22,057 mg/L as digestion time increased today 13, and then increased again to 32,914 mg/L with the furtherincrease of digestion time to day 17. The reason was that the solidparticulate material was firstly converted into soluble organiccompounds such as acetic acid, propionic acid, and butyric acid,which led to the increase in soluble COD concentration in the solu-tion. In addition, the soluble COD always showed a downwardtrend from day 25, which demonstrated that the hydrolysis andfermentation rates of solid organic matter were lower than the uti-lization rate of microorganisms to produce biogas. At the end of thedigestion, the soluble COD decreased to 2206 mg/L, and the finalremoval efficiency was 81.3%. Meanwhile, the removal efficiencyof TS and VS were 46% and 69%, respectively, and all data indicatedthat the co-digestion of the mixed feed stock was feasible.
The concentrations of major volatile short-chain fatty acids pro-duced are shown in Fig. 3b at different co-digestion times. The dataindicated that acetic acid, propionic acid, and butyric acid were themain compounds produced during the co-digestion. The concen-tration variation showed a similar trend for acetic acid and propi-onic acid on day 17, and the concentration consistently fluctuatedin the 32.6 mM and 125.4 mM range for acetic acid, and in the 5.5and 125.4 range for propionic acid. As co-digestion time increasedfrom day 17 to day 50, the concentrations varied between 62.9 mMand 1.6 mM for acetic acid, and were lower than those for propi-onic acid. Compared with acetic acid and propionic acid, the con-centrations of butyric acid were less than the VFAs of the othertwo. Formic acid, pyruvic acid, and lactic acid were also measured,but their concentrations were so low that the average sum of theconcentration was less than 5% of the total detected VFA. The totalconcentration of VFA gradually decreased to 1.9 mM at the end ofthe anaerobic co-digestion period. These data indicated that the or-ganic matter provided by food waste, wastepaper, and plastic werecompletely consumed as the substance to produce methane bymicroorganisms.
3.4. Ammonia-N and pH variation in the mid-scale experiment
The concentration of NHþ4 �N, free ammonia, and pH values areshown in Fig. 4. For NHþ4 �N, the concentration gradually in-creased from 76.7 mg/L to 1437.1 mg/L as digestion time increasedfrom 0 to 25 days, and then the concentration rapidly decreased to1061.2 mg/L on day 26. From the 27th day, the concentration sta-bilized between 890.1 and 1207.8 mg/L with the further increase indigestion time. Compared with NHþ4 �N, the concentration of free
ammonia showed an almost linear increasing trend, and the max-imum concentration was 117.5 mg/L. A high concentration ofammonium is known to be toxic, which will cause the inhibitionfor biogas production during the anaerobic co-digestion processfor waste treatment (Chen et al., 2008). Koster and Lettinga(1988) reported that acidogenic populations in the granular sludgewere hardly affected, whereas the methanogenic population lost56.5% of its activity as ammonia concentrations were increasedin the range of 4051–5734 mg/L. Liu et al. (2012) reported thatthe ammonia nitrogen inhibition concentration was 1400–1700 mg/L for the anaerobic digestion of kitchen waste; particu-larly, the ammonia nitrogen concentration rose to 2000 mg/L,and the anaerobic methanogenesis efficiency consequently de-creased. However, in this study, the NHþ4 �N concentration waslower than the inhibited concentration reported in the literatures,such as to not inhibit the methanogens during the anaerobicdigestion.
In addition, the activities of methane-synthesizing enzymeswere directly inhibited by free ammonia because hydrophobic freeammonia molecules diffused passively into the cell and were rap-idly converted to ammonium as a result of the intracellular pHconditions. The increase in ammonia improved the acid-base sta-bility of the anaerobic system. Specifically, ammonia increasedthe buffer capacity of the methanogenic medium in the mesophilicanaerobic reactor, thereby increasing the stability of the anaerobicdigestion. Optimal ammonia concentration ensured sufficient buf-fer capacity without inhibiting the process. The variation of pH val-ues during the co-digestion period is shown in Fig. 4. The pH leveldropped rapidly from 7.73 to 5.93 as co-digestion time increasedfrom the first to the second day. When the solution pH was lower
d30 d35d5 d40 54d61d d25d21d10d0 d50
1
2
3
4
56
7
8
9 10
13
1415
1617
18 19
2021
11 12
61d04d d45 d25 d21d5d0 d10 d35 d30d50
(a) 5.70.81 17.7 17.1 10.728.1100 27.8 17.4 16.228.2(%)
d30 d35 d40 54d61dd5 d25d21d10 05d0d
123 46 5
78 9 10 11
12
61d04d d45 d25 d21d5d0 d10d35d30d50
4.121.15 50.6 27.0 25.258.9100 51.1 28.4 26.264.4(%)(b)Fig. 5. DGGE band patterns of the 16S rRNA gene fragments: (a) bacterial profiles and (b) archaeon profiles.
624 S. Wan et al. / Bioresource Technology 146 (2013) 619–627
than 5.25 on the seventh day, the pH value was adjusted to above6.5 with sodium hydroxide solution to avoid the inhibition ofmethanogenesis. The reason was that according to Leitão et al.(2006), the methanogenic activity was more likely to proceedoptimally within a narrow pH value range, i.e., between 6.3 and
7.8 under anaerobic conditions. Non-methanogenic microorgan-isms responsible for hydrolysis and fermentation could adapt tolow pH, whereas methanogens would lose activity at low pH. Spe-cifically, the single-stage anaerobic co-digestion of food waste,wastepaper, and plastic requires a coordinated metabolism of
B5Uncultured bacterium clone(HM107047)activated_sludgeUncultured bacterium clone(GQ247136)B6B3Uncultured bacterium clone MY33(JN245656)B1B21B13B7Uncultured Bacteroidetes bacterium clone(HM104941)93B2Lactobacillus fermentum strain SFCB2-3(DQ399352)Uncultured bacterium clone(JX839736)B10B16B12B11Uncultured bacterium clone(EF686914)B14Uncultured Bacteroidetes bacterium clone(JQ580449)B17Uncultured bacterium clone(JF834129)xianweisuB18Uncultured bacterium clone(JF546299)huluoboB20Uncultured bacterium clone(EF559223)xianweisuB19Uncultured bacterium clone(JQ139489)B8B4Uncultured bacterium clone(JF571101)B9Uncultured bacterium clone(JF547253)Acetobacter orientalis gene(AB608070)Uncultured Acetobacter sp. clone(JN423062)B15
100
100100
100
100
100
100
99
89
8499
76
73
67
65
98
93
62
10096
90
82
100
95
Coriobacteridae
Synergistales
Rhodospirillales
Clostridiales
Bacteroidales
Thermomicrobiales
Rhodospirillales
Rhodospirillales
Bacteroidales
Lactobacillales
Bacteroidales
Lactobacillales
Clostridiales
Bacteroidales
(a)
Clostridiales
Lactobacillales
Lactobacillales
Uncultured archaeon clone (JF980509)A2A1A4Archaeon enrichment culture clone (HM630579)Uncultured archaeon clone (JQ085750)A7A5A10Uncultured archaeon (HM193347)Uncultured archaeon gene(AB744707)Uncultured archaeon (FN547123)Uncultured archaeon clone (AY835818)A6Uncultured archaeon clone (JQ241422)Uncultured archaeon clone(FJ347534)Uncultured archaeon clone (HQ224856)Uncultured archaeon clone (JN562371)Uncultured archaeon clone (JX995487)A9A3A8A12A11
99
89
8373
68
6486
64
62
100
90
89
95
Thermoplasmata
Thermoprotei
Methanomicrobia
Thermoplasmata
Methanomicrobia
Methanomicrobia
Methanomicrobia
Thermoprotei
Thermoplasmata
Thermoplasmata
(b)
Fig. 6. Maximum likelihood tree of the 16S rRNA gene sequences showing the affiliation of all detected bands: (a) bacterial and (b) archaeon.
S. Wan et al. / Bioresource Technology 146 (2013) 619–627 625
bacterial groups involved in VFA production and utilization. After13 days of acidification, the pH level gradually increased from
6.24 to 7.99 without the addition of sodium hydroxide solution.This finding could be attributed to the accumulation of NHþ4 �N
626 S. Wan et al. / Bioresource Technology 146 (2013) 619–627
in equilibrium with free ammonia in the digester. In addition, thesubsequent transfer and consumption of VFA by methanogenesiscould also contribute to the improvement of the pH value of theco-digester system. Thus, the results indicated that the single-stagebioreactor was effective in treating the mixture of food waste,wastepaper, and plastic.
3.5. Microbial diversity analysis of the mid-scale experiment
To obtain crucial qualitative information on the evolution ofmicroorganisms during co-digestion, the microbial communitystructure and diversity were analyzed by sampling biogas residuesat different co-digestion times. To compare the composition ofmicrobial communities during the co-digestion, the distinct DGGEbands patterns with time (see Fig. 5) and the 16S rDNA genesamplified were loaded onto both gels for bacteria (Fig. 5a) and ar-chaea (Fig. 5b), respectively. A total of 21 and 12 discernible DGGEbands were respectively observed in the two DGGE profiles for bac-teria and archaea. As shown in Fig. 5a, the bacterial communitychanges in the biogas residue samples varied significantly at differ-ent anaerobic digestion times during the co-digestion of the mix-ture of food waste, wastepaper, and plastic. Generally, thediversity of the bacterial community first increased and then de-creased with the further increase of digestion time. The diversityof the archaeon community showed a similar trend. The DGGE pro-files also indicated that the similarity between the bacterial andarchaeon diversity gradually reduced compared with day 0 withthe increase in co-digestion time, which further demonstrated thatthe bacterial and archaeon structures changed significantly,depending on the persistence of anaerobic conditions.
The bacterial communities in the anaerobic digester wereresponsible for the conversion of organic solid waste into solubleorganic compounds, such as VFAs, which could further serve assubstrates for methanogens. The species identified from the DGGEanalysis could be grouped into five different orders: Bacteroidales(B2, B10, B11, B12, B16, B17, B18, B19, and B20), Lactobacillales(B3, B4, B5, and B6), Rhodospirillales (B1, B14, B15, and B21), Clos-tridiales (B7, B8, and B9), and Synergistales (B13) for bacteria, asshown in Fig. 6a. The results indicated that the predominant pop-ulation was represented by Bacteroidales, and they were consider-ably closer to uncultured bacterium clones detected in theanaerobic digestion and co-digestion processes. For instance, 99%sequence similarity for B2 with JF546299 (Garcia et al., 2011)and EF559223 (Li et al., 2009) and B10 with HM107047 (Cirneet al., 2012) was detected during the anaerobic digestion of carrotwaste, catalyzing methanization of cellulose and continuous fer-mentation of thermally hydrolyzed waste-activated sludge, respec-tively. The population representing Lactobacillales in bands B3, B4,and B6 was significantly closer with 99% sequence similarity to theLactobacillus fermentum strain DQ399352 (Gao et al., 2008) from areactor fermentation that treated rice straw and uncultured bacte-rium JN245656 (Liu et al., 2012) involved in the anaerobic co-digestion of municipal biomass waste and waste-activated sludge.The population representing Rhodospirillales in band B1 was signif-icantly closer with 99% sequence similarity to uncultured aceto-bacter JN423062 (Chandler et al., 2011). Comparatively, thebacterial communities of Clostridiales B8 were considerably closerwith 95% sequence similarity to uncultured bacterium JF571101(Garcia et al., 2011), which was detected in the anaerobic reactorfor digestion of carrot waste. The results also indicated that theBacteroidales, Lactobacillales, and Rhodospirillales were the mostabundant microorganisms present in the used feedstocks, whichmaintained their predominance during the entire digestion pro-cess. Previous studies also found that members of the Bacteroidales,Lactobacillales, and Clostridiales microorganisms were extremelyabundant during the MSW treatment (Bareither et al., 2013). In
addition, Firmicutes and Bacteroidetes are known for being extre-mely resistant microorganisms, which were also versatile and ableto degrade complex organic compounds such as vegetable and fruitresidues (Garcia-Peñaa et al., 2011) and tetrabromobisphenol A(Peng et al., 2012). These could partly account for the fact thatthe lead members of the abundant groups in this study were suc-cessful in anaerobic digesters.
For the archaeon, the species identified through the DGGE anal-ysis could be grouped into three different classes: Thermoplasmata(A1, A2, A4, A5, A8, A10, and A11), Methanomicrobia (A6, A7, A9,and A12), and Thermoprotei (A3), as shown Fig. 6b. The archaeaare similar to the main archaeon present in the anaerobic digestionreported by previous studies. The population of Thermoplasmatawas significantly closer to the uncultured archaeon reported inthe literature. For instance, bands A2 and A11 were close to theuncultured archaeon JF980509, which was detected during theanaerobic co-digestion of chicken feathers and other animal wastes(Xia et al., 2012), and to uncultured archaeon HM193347 (Jung andRegan, 2010) with 100% and 97% sequence similarities, respec-tively. In addition, the populations representing Methanomicrobiain bands A6, A9, and A12 were significantly closer to unculturedarchaeon AY835818 (Collins et al., 2005) for A6, uncultured archa-eon HQ224856 (Zhang et al., 2011) and FJ347534 (McKeown et al.,2009) for A9, and uncultured archaeon AB744707 (Bandara et al.,2012) for A12 with 99%, 100%, and 97% sequence similarities,respectively. All of the detected microorganisms matched the ar-chaea involved in the anaerobic process. Compared with Thermo-plasmata and Methanomicrobia, the band A3 representingThermoprotei had a 97% sequence similarity with uncultured archa-eon JX995487, a domain archaeon present in the anaerobic treat-ment of ammonium from coal-combustion wastewaters(Vishnivetskaya et al., 2013). Thus, all of the results indicated thatthe transition of the mesophilic microbial community was consis-tent with the performance of the bioreactor during the co-diges-tion of food waste, wastepaper, and plastic.
4. Conclusion
A single-stage anaerobic system effectively co-digested foodwaste, wastepaper, and plastic. The final total biogas productionand average percentage of methane were almost 17.0 m3 andabove 50%, respectively, at steady conditions. The total biogasand methane yields were 592.4 L/Kg and 370.1 L/Kg, respectively,with VS added. Acetic acid, propionic acid, and butyric acid werethe main acids produced during digestion process. Firmicutes andBacteroidetes were the dominant microorganisms from the DGGEanalysis of solid biogas residue. The anaerobic co-digestion of foodwaste, wastepaper, and plastic was found to be feasible and effi-cient using the single-stage semi-dry mid-scale anaerobic reactor.
Acknowledgement
This work was financially supported by grants from Natural Sci-ence Foundation of China (51208492, 41071213, 21177121) andFujian Science Funding (2012Y0067).
References
Angelidaki, I., Ahring, B.K., 1993. Thermophilic anaerobic digestion of livestockwaste: the effect of ammonia. Appl. Microbiol. Biotechnol. 38, 560–564.
APHA, 1998. Standard methods for the examination of water and wastewater,twentieth ed., American Public Health Association, Washington, DC.
Bandara, W.M.K.R.T.W., Kindaichi, T., Satoh, H., Sasakawa, M., Nakahara, Y.,Takahashi, M., Okabe, S., 2012. Anaerobic treatment of municipal wastewaterat ambient temperature: analysis of archaeal community structure andrecovery of dissolved methane. Water Res. 46, 5756–5764.
S. Wan et al. / Bioresource Technology 146 (2013) 619–627 627
Bareither, C.A., Wolfe, G.L., McMahon, K.D., Benson, C.H., 2013. Microbial diversityand dynamics during methane production from municipal solid waste. WasteManage.. http://dx.doi.org/10.1016/j.wasman.2012.12.013.
Bouallagui, H., Touhami, Y., Ben Cheikh, R., Hamdi, M., 2005. Bioreactor performancein anaerobic digestion of fruit and vegetable wastes. Process. Biochem. 40, 989–995.
Chandler, J.A., Lang, J.M., Bhatnagar, S., Eisen, J.A., Kopp, A., 2011. Bacterialcommunities of diverse drosophila species: ecological context of a host–microbe model system. PLoS Gen. 7, 1–18.
Chen, Y., Cheng, J.J., Creamer, K.S., 2008. Inhibition of anaerobic digestion process: areview. Bioresour. Technol. 99, 4044–4064.
Cheng, H., Hu, Y., 2010. Municipal solid waste (MSW) as a renewable source ofenergy: current and future practices in China. Bioresour. Technol. 101, 3816–3824.
China, N.B.o.S.o., 2012. 2012 China Statistical Yearbook. China Statistics Press,Beijing.
Cirne, D.G., Bond, P., Pratt, S., Lant, P., Batstone, D.J., 2012. Microbial communityanalysis during continuous fermentation of thermally hydrolysed wasteactivated sludge. Water Sci. Technol. 65, 7–14.
Collins, G., O’Connor, L., Mahony, T., Gieseke, A., de Beer, D., O’Flaherty, V., 2005.Distribution, localization, and phylogeny of abundant populations ofCrenarchaeota in anaerobic granular sludge. Appl. Environ. Microb. 71, 7523–7527.
El-Mashad, H.M., McGarvey, J.A., Zhang, R., 2008. Performance and microbialanalysis of anaerobic digesters treating food waste and dairy manure. Biol. Eng.1, 233–242.
El-Mashad, H.M., Zeeman, G., van Loon, W.K.P., Bot, G.P.A., Lettinga, G., 2004. Effectof temperature and temperature fluctuation on thermophilic anaerobicdigestion of cattle manure. Bioresour. Technol. 95, 191–201.
Fernández, J., Pérez, M., Romero, L.I., 2008. Effect of substrate concentration on drymesophilic anaerobic digestion of organic fraction of municipal solid waste(OFMSW). Bioresour. Technol. 99, 6075–6080.
Forster-Carneiro, T., Pérez, M., Romero, L.I., 2008. Anaerobic digestion of municipalsolid wastes: dry thermophilic performance. Bioresour. Technol. 99, 8180–8184.
Gao, L., Yang, H., Wang, X., Huang, Z., Ishii, M., Igarashi, Y., Cui, Z., 2008. DQ399352-Rice straw fermentation using lactic acid bacteria. Bioresour. Technol. 99, 2742–2748.
Garcia-Peñaa, E.I., Parameswaran, P., Kang, D.W., Canul-Chan, M., Krajmalnik-Brown, R., 2011. Anaerobic digestion and co-digestion processes of vegetableand fruit residues: process and microbial ecology. Bioresour. Technol. 102,9447–9455.
Garcia, S.L., Jangid, K., Whitman, W.B., Das, K.C., 2011. JF546299-Transition ofmicrobial communities during the adaption to anaerobic digestion of carrotwaste. Bioresour. Technol. 102, 7249–7256.
Gonzáleza, C.R.-N., Björklund, E., Forteza, R., Cerdàa, V., 2013. Volatile organiccompounds in landfill odorant emissions on the island of Mallorca. Int. J.Environ. Anal. Chem. 93, 434–449.
Jung, S., Regan, J.M., 2010. Influence of external resistance on electrogenesis,methanogenesis, and anode prokaryotic communities in microbial fuel cells.Appl. Environ. Microb. 77, 564–571.
Koster, I.W., Lettinga, G., 1988. Anaerobic digestion at extreme ammoniaconcentrations. Biol. Wastes 25, 51–59.
Leitão, R.C., van Haandel, A.C., Zeeman, G., Lettinga, G., 2006. The effects ofoperational and environmental variations on anaerobic wastewater treatmentsystems: a review. Bioresour. Technol. 97, 1105–1118.
Li, R., Chen, S., Li, X., 2010. Biogas production from anaerobic co-digestion of foodwaste with dairy manure in a two-phase digestion system. Appl. Biochem.Biotechnol. 160, 643–654.
Li, T., Mazéas, L., Sghir, A., Leblon, G., Bouchez, T., 2009. Insights into networks offunctional microbes catalysing methanization of cellulose under mesophilicconditions. Environ. Microbiol. 11, 889–904.
Liu, G., Zhang, R., El-Mashad, H.M., Dong, R., 2009. Effect of feed to inoculum ratioson biogas yields of food and green wastes. Bioresour. Technol. 100, 5103–5108.
Liu, X., Wang, W., Shi, Y., Zheng, L., Gao, X., Qiao, W., Zhou, Y., 2012. Pilot-scaleanaerobic co-digestion of municipal biomass waste and waste activated sludgein China: effect of organic loading rate. Waste Manage.. http://dx.doi.org/10.1016/j.wasman.2012.03.003.
Luo, W., D’Angelo, E.M., Coyne, M.S., 2008. Organic carbon effects on aerobicpolychlorinated biphenyl removal and bacterial community composition insoils and sediments. Chemosphere 70, 364–373.
McKeown, R.M., Scully, C., Enright, A.-M., Chinalia, F.A., Lee, C., Mahony, T., Collins,G., O’Flaherty, V., 2009. Psychrophilic methanogenic community developmentduring long-term cultivation of anaerobic granular biofilms. ISME J. 3, 1231–1242.
Peng, X., Zhang, Z., Zhao, Z., Jia, X., 2012. 16S ribosomal DNA clone libraries to revealbacterial diversity in anaerobic reactor-degraded tetrabromobisphenol A.Bioresour. Technol. 112, 75–82.
Poggi-Varaldo, H.M., Rodríguez-Vázquez, R., Fernández-Villagómez, G., Esparza-García, F., 1997. Inhibition of mesophilic solid-substrate anaerobic digestion byammonia nitrogen. Appl. Microbiol. Biotechnol. 47, 284–291.
Renou, S., Givaudan, J.G., Poulain, S., Dirassouyan, F., Moulin, P., 2008. Landfillleachate treatment: review and opportunity. J. Hazard. Mater. 150, 468–493.
van Haandel, A.C., van der Lubbe, J., 2007. Handbook biological waste watertreatment: design and optimisation of activated sludge systems, first ed. Quist,Leidschendam.
Vishnivetskaya, T., Fisher, L.S., Brodie, G., Phelps, T., 2013. Microbial communitiesinvolved in biological ammonium removal from coal combustion wastewaters.Microb. Ecol., 1–11.
Xia, Y., Massé, D.I., McAllister, T.A., Kong, Y., Seviour, R., Beaulieu, C., 2012. Identityand diversity of archaeal communities during anaerobic co-digestion of chickenfeathers and other animal wastes. Bioresour. Technol. 110, 111–119.
Zhang, J., Wei, Y., Xiao, W., Zhou, Z., Yan, X., 2011. Performance and spatialcommunity succession of an anaerobic baffled reactor treating acetone–butanol–ethanol fermentation wastewater. Bioresour. Technol. 102, 7407–7414.
