Bioresource Technology 102 (2011) 5048–5059

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Anaerobic co-digestion of food waste and piggery wastewater: Focusingon the role of trace elements

Lei Zhang a, Yong-Woo Lee b, Deokjin Jahng a,⇑a Department of Environmental Engineering and Biotechnology, Myongji University, San 38-2, Namdong, Cheoin-Gu, Yongin, Gyeonggi-Do 449-728, Republic of Koreab Major in Chemistry & Applied Chemistry, College of Science and Technology, Hanyang University, 1271 Sa-3 Dong, Sangnok-Gu, Ansan, Gyeonggi-Do 426-791, Republic of Korea

a r t i c l e i n f o

Article history:Received 19 October 2010Received in revised form 26 January 2011Accepted 28 January 2011Available online 23 February 2011

Keywords:Anaerobic digestionCo-digestionFood wastePiggery wastewaterTrace elements

0960-8524/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.biortech.2011.01.082

⇑ Corresponding author. Tel.: +82 31 330 6690; faxE-mail addresses: wxzyfx@yahoo.com (L. Zhang), d

a b s t r a c t

The objective of this study was to evaluate the feasibility of anaerobic co-digestion of food waste and pig-gery wastewater, and to identify the key factors governing the co-digestion performance. The analyticalresults indicated that the food waste contained higher energy potential and lower concentrations of traceelements than the piggery wastewater. Anaerobic co-digestion showed a significantly improved biogasproductivity and process stability. The results of co-digestion of the food waste with the different frac-tions of the piggery wastewater suggested that trace element might be the reason for enhancing theco-digestion performance. By supplementing the trace elements, a long-term anaerobic digestion ofthe food waste only resulted in a high methane yield of 0.396 m3/kg VSadded and 75.6% of VS destructionwith no significant volatile fatty acid accumulation. These results suggested that the typical Korean foodwaste was deficient with some trace elements required for anaerobic digestion.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction Two-stage anaerobic digestion system and HASL (hybrid anaerobic

Generation of municipal solid waste (MSW) has been increasingover the years worldwide, a considerable fraction of which is foodwaste (Banks et al., 2011; Zhang et al., 2007). In Korea, the foodwaste reached about 14,452 tons per day in 2007, which accountedfor 28.7% of the MSW (MOE, 2010). The untreated food waste isknown to cause many environmental problems, such as contami-nations of soil, water, and air during collection, transportationand storage due to its rapid decomposition (Han and Shin, 2004).Currently, various methods for reutilization and disposal of thefood waste are available, which include landfill, incineration, useof animal feed, aerobic composting and anaerobic digestion. Dueto newly issued environmental regulations, some disposal meth-ods are going to be prohibited and becoming less desirable (Kelleyand Walker, 2000; Oh et al., 2008).

Nonetheless, food waste is a highly desirable substrate foranaerobic digestion because its biodegradability and nutrient con-tents are high. The typical food waste contains 7–31 wt.% of totalsolid (TS), and the biochemical methane potential (BMP) of the foodwaste is estimated to be about 0.44–0.48 m3 CH4/kg of the addedvolatile solid (VSadded) (Han and Shin, 2004; Heo et al., 2003; Zhanget al., 2007). Nowadays, anaerobic digestion of the food waste isattracting strong interest, and many novel anaerobic digestion sys-tems have been developed and applied to treat the food waste.

ll rights reserved.

: +82 31 336 6336.jahng@mju.ac.kr (D. Jahng).

solid–liquid) system are well known examples (Lee et al., 1999;Wang et al., 2005). In practical application, however, about 90% ofthe full scale plants currently in use in Europe rely on continuousone-stage systems (De Baere, 2000). Even so, there are rare reportson the successfully operating single-stage anaerobic digestion ofthe food waste. These results indicated that anaerobic digestion ofthe food waste still remains as a challenge.

Piggery wastewater is another major organic waste. Anaerobicdigestion has long been employed to treat piggery wastewater be-cause through which waste treatment and energy productioncould be achieved simultaneously. The methane yields of animalmanures are generally in the range of 12.0–13.9 m3 CH4/m3 waste(wet basis). According to the economic analysis of the existing bio-gas plants, however, the methane yield should be higher than20 m3 CH4/m3 substrate to meet the economic balance (Angelidakiand Ellegaard, 2003). In addition, ammonia inhibition was oftenobserved in the anaerobic digestion of the pure manure, which re-sulted in a low methane production and a high VFA level in theeffluent (Hansen et al., 1998).

Nowadays, anaerobic co-digestion has attracted more atten-tions (Angelidaki and Ellegaard, 2003; Creamer et al., 2010; Heoet al., 2003). Generally, animal manures like piggery wastewaterare considered to be excellent co-substrates because of its highbuffering capacity, high nitrogen content and the wide range ofnutrients needed by the methanogens (Moral et al., 2008; Weiland,2000). The co-digestion of animal manure with other substrate canbe successful because C/N ratio, concentrations of the macro andmicronutrients, and buffering capacity are adjusted by mixing

L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059 5049

the substrates. Co-digestion of animal manure with a biodegrad-able waste appears as a robust process technology that can in-crease the biogas production by 80–400% in biogas plants (Braunet al., 2003; Weiland, 2000). Moreover, many studies showed thatthe sensitivity of the anaerobic digestion process to the environ-mental changes may be improved by combining several wastestreams (Creamer et al., 2010; Heo et al., 2003; Kayhanian andRich, 1995; Romano and Zhang, 2008; Wu et al., 2010). These prac-tices suggest that anaerobic co-digestion of the food waste and thepiggery wastewater could solve the operational problems and loweconomical feasibility found in anaerobic digestion of food wasteor piggery wastewater alone.

The aim of this study was to evaluate the technical feasibility ofanaerobic co-digestion of the food waste and the piggery wastewa-ter in a mesophilic single stage reactor and to identify the key fac-tors governing the process performance. In order to more clearlyexplain the co-digestion results, the food waste and piggery waste-water were characterized. Special focus was put on trace elements,since many previous reports showed that the trace element sup-plementation enhanced anaerobic digestion of different substrates(Agler et al., 2008; Jarvis et al., 1997; Wilkie et al., 1986). In addi-tion, the literatures reported that piggery wastewaters were rich intrace elements (Creamer et al., 2010; Moral et al., 2008), and thefood wastes contained less trace elements or analytical data wereunavailable (Zhang et al., 2007; Zhu et al., 2008).

2. Methods

2.1. Feedstocks and inoculum

The food waste used in this study was collected from a Koreanfood waste restaurant on the campus of the Myongji University,Yongin, Korea. The obtained food waste was crushed using an elec-trical kitchen blender (HMF-347, Hanil, Korea) and the resultingslurry food waste was sieved (No. 10) to remove coarse particleslarger than 2 mm and kept at �18 �C until use.

The piggery wastewater used in this study was obtained fromstorage tanks on a swine farm located in Yongin, which contained

Table 1Summary of reactor operating conditions.

Experiment Digester HRT (day) OLR (g COD/L day) Comp

Experiment 1 EX1-1 20 4.71 100%EX1-2 20 6.35 100%EX1-3 20 6.35 7% wEX1-4 20 6.35 17% w

Experiment 2 EX2-1 20 6.35 100%EX2-2 20 6.35 83% FEX2-3 20 6.35 83% FEX2-4 20 6.35 83%F

Experiment 3 Period 1 (Day 0–92) 20 6.35 Conti83% F17% P

Period 2 (Day 93–155) 40 3.2 83% F1 and

Period 3 (Day 156–161) 30 5.1 83% F1 and

Period 4 (Day 162–179) 30 4.3 100%Period 5 (Day 180–253)b 30 4.3 100%

elemPeriod 6 (Day 254–367)b 20 6.35 100%

elem

a 7% PW/93% FW represents that 7% of OLR was contributed by the piggery wastewatb During this period, Run 1 served as the control, whereas synthetic trace elements w

the mixture of feces, urine and tap water. After arriving at the lab-oratory, the wastewater was filtered through a 2 mm mesh to re-move coarse particles, and then stored at 4 �C until use.

In order to analyze the distribution of trace elements, the pig-gery wastewater was centrifuged at 14,000g for 30 min (SUPRA21 K, Hanil, Korea). The solid part was resuspended in the samevolume of distilled water (DW) as that of the liquid fraction. Theconcentrations of trace elements in the liquid fraction and the solidsuspension were separately analyzed. The liquid fraction and thesolid suspension were stored at 4 �C until use as substrates forco-digestion experiments.

The seed sludge used in this study was taken from a 20-L benchscale anaerobic reactor treating the piggery wastewater for morethan 2 years. The operating conditions were hydraulic retentiontime (HRT) of 20–30 days and OLR of 1.0–2.3 g VS/L day. The vola-tile suspended solids (VSS) concentration of the seed culture wasapproximately 15 g/L.

2.2. Anaerobic digestion in semi-continuous mode

Semi-continuous anaerobic digestion was carried out in a 500-mL Schott Duran bottle with a 200-mL working volume. Initially,the bottle was filled with 190 mL seed sludge and 10 mL substrate.After sparged with nitrogen gas, the digester was capped with arubber septum, and then inversely incubated in a shaking incuba-tor at 37 �C and 140 rpm, which provided the constant agitationand temperature. Incubation was proceeded in a semi-continuousmode with daily or every other day withdrawing and feeding of thesame of amount, and the HRT was kept at 20–40 days. All the oper-ations were made under nitrogen atmosphere to avoid contactingoxygen.

Three series of semi-continuous experiments were conductedto evaluate the feasibility of anaerobic co-digestion of the foodwaste and the piggery wastewater and to identify the key factorsresponsible for enhancing co-digestion performance. Table 1 sum-marizes the experimental design and operating conditions. InExperiment 1, four digesters were operated by feeding differentmixtures of the piggery wastewater and the food waste. In Exper-iment 2, in order to figure out the location of the effective

osition of feedstock (on COD basis) Objective

whole PW (piggery wastewater) Examine the possibility of anaerobic co-digestion of the food waste and thepiggery wastewater

FW (food waste)hole PW/93% FWa

hole PW/83% FW

FW Identify the stimulatory substances inpiggery wastewater for co-digestion,including the C/N ratio, bufferingcapacity and concentrations of traceelements

W/17% whole PWW/17% PW liquid fraction

W/17% PW solid fraction

nued operation of EX2-2 and EX2-4W/17% whole PW (Run 1) 83% FW/W solid fraction (Run 2)W/17% whole PW (identical for RunRun 2)

W/17% whole PW (identical for RunRun 2)

FW (identical for Run 1 and Run 2)FW (Run 1) 100% FW + Trace

ents (Run 2)Examine the process stability in a long-term operation, and confirm the role oftrace elements on anaerobic digestion ofthe food waste by supplying synthetictrace elements instead of the piggerywastewater

FW (Run 1) 100% FW + Traceents (Run 2)

er and 93% by the food waste on COD basis.ere added in Run 2 (see Fig. 5).

5050 L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059

substances of the piggery wastewater which enhanced the co-digestion performance, the food waste was co-digested separatelywith the whole piggery wastewater, the solid fraction or the liquidfraction of the piggery wastewater. Finally, in Experiment 3, the ef-fects of trace elements on anaerobic digestion of the food wastewere investigated. For this set of experiments (Experiment 3),the stock solution of trace elements containing 200 mg Co2+/L asCoCl2, 500 mg Mo2+/L as MoCl2, 1000 mg Ni2+/L as NiCl2, and10,000 mg Fe3+/L as FeCl3 was prepared in distilled water (DW).For Run 2 (Table 1), 0.67 mL of this stock solution was added tothe digester (200 mL of working volume) once a day for 3 daysfrom Day 180 to Day 182. From Day 194, 0.67 mL of the stock solu-tion was added intermittently with the interval of 6–12 days. Sincethe HRT of the digester was 20 days with a negligible volumechange, the concentrations of each trace elements in the digesterwere assumed to be 2.0 mg Co2+/L, 5.0 mg Mo2+/L, 10 mg Ni2+/Land 100 mg Fe3+/L.

2.3. Analytical methods

TS (total solid) and VS (volatile solid) were measured accordingto the Standard Methods (APHA, 2005). pH value was determinedusing a pH meter (Orion, Model 370). Chemical oxygen demand(COD) measurements were made using COD ampoules (HachChemical) and a spectrophotometer (DR/2010, Hach). SolubleCOD was measured for the supernatant of a sample after centrifuga-tion (Micro 17R centrifuge, Hanil Science Industrial Co., Ltd., Korea)at 14,000g for 10 min. Total Kjeldahl nitrogen (TKN) was analyzedusing a Kjeldahl apparatus (Kjeltec 2100, Foss, Sweden), and totalammonia (free ammonia and ionized ammonia) content was deter-mined by the Kjeldahl method without the destruction step. Proteincontent was estimated by multiplying the organic nitrogen value(TKN subtracted by total ammonia nitrogen) by 6.25 (Ahn et al.,2006). Lipid was gravimetrically measured (Bligh and Dyer, 1959).

Biogas composition (N2, CH4 and CO2) was determined using agas chromatograph (GC) (Hewlett Packard 6890, PA, USA) equippedwith a thermal conductivity detector (TCD) and an HP-PLOT Q(Agilent Technologies, USA) capillary column (30 m � 0.32 mm �20 lm). A gas standard consisting of 60% (v/v) CH4 and 40% (v/v)CO2 was used for calibrating gas chromatographic results. The

Table 2The characteristics of the food waste and the piggery wastewater as compared to the lite

Parameter Food waste

Han and Shin (2004) Zhang et al. (2007) This

pH –a – 6.TS (wt.%) 20.5 30.9 ± 0.1 18.VS (wt.%) 19.5 26.4 ± 0.1 17.VS/TS ratio 0.95 0.85 0.9Total COD (g/L) – – 238.Soluble COD (g/L) – – 106.Carbohydrate (g/L) – – 111.Lipid (g/L) – – 23.Total protein (g/L)b – – 32.Carbon, C (% of TS) 51.4 46.78 ± 1.15 46.6Hydrogen, H (% of TS) 6.1 – 6.3Oxygen, O (% of TS) 38.9 – 36.3Nitrogen, N (% of TS) 3.5 3.16 ± 0.22 3.5Sulfur, S (% of TS) 0.1 0.81 ± 0.03 0.3TKN (g/L) – – 5.4TP (g/L) – – 1.4Ammonia-N (g/L) – – 0.1Alkalinity (g CaCO3/L) – – 0.3C/N ratio 14.7 14.6 13.BMP (mL CH4/g VSadded) – 435 479.

a –, Not available.b Protein content = (TKN-ammonia nitrogen) � 6.25.

measurement of biogas generation was described previously(Zhang and Jahng, 2010). Concentrations of volatile fatty acids(including acetate, propionate, n-butyrate, iso-butyrate, n-valerateand iso-valerate) were determined using another gas chromato-graph (M600D, Younglin, Korea) equipped with a flame ionizationdetector (FID) and an HP-INNOWAX (Agilent Technologies, USA)capillary column (30 m � 0.25 mm � 0.25 lm). The sample prepa-ration procedures and GC operational conditions can be found else-where (Zhang and Jahng, 2010).

Elements were assayed using an element analysis instrument(Flash EA1112, Thermo Electron SPA). Nitrogen, carbon, hydrogen,sulfur and oxygen were the target elements. Metal analysis wasperformed using an ion coupled plasma-atomic emission spectrom-eter (ICP-AES) (OPTIMA 4300DV, Perkin–Elmer, USA) or inductivelycoupled plasma-mass spectrometer (ICP-MS) (ELAN6100, Perkin–Elmer SCIEX, USA).

3. Results and discussion

3.1. Characterization of food waste and piggery wastewater

3.1.1. General featuresThe results of the feedstock characterization are summarized in

Tables 2 and 3. As shown in Table 2, the food waste contained18.1% (w/w) of TS and 17.1% (w/w) of VS, which were nearly threefolds higher than those of the piggery wastewater (5.64% of TS and3.69% of VS). Similarly, the food waste also contained much highertotal COD and soluble COD (238.5 and 106.6 g/L, respectively) thanthe piggery wastewater (92.8 and 53.2 g/L, respectively). The vola-tile fraction of the total solid of the food waste was 0.94, while theVS/TS ratio of the piggery wastewater was only 0.65, indicatingthat the food waste contained more digestible organic mattersthan the piggery wastewater. These results well matched withthe literature reports (Han and Shin, 2004; Zhang et al., 2007) aslisted in Table 2, where the moisture content and VS/TS ratio ofthe food wastes was 69–93% and 0.85–0.95, respectively. By con-trast, the piggery wastewaters contained 6.37–7.62% of TS and0.72 of VS/TS ratio (Ahn et al., 2006). The biodegradability was alsoestimated by measuring BMP. The BMP value of the piggery waste-water was 242.3 mL CH4/g VSadded, which was only 51% of the food

rature reports.

Piggery wastewater

study Ahn et al. (2006) Hansen et al. (1998, 1999) This study

5 ± 0.2 6.37 ± 0.10 7.62 ± 0.02 6.60 ± 0.201 ± 0.6 6.18 ± 0.04 – 5.64 ± 0.341 ± 0.6 4.45 ± 0.02 4.5 ± 0.1 3.69 ± 0.224 ± 0.01 0.72 – 0.65 ± 0.015 ± 3.8 130.8 ± 3.0 – 92.8 ± 1.36 ± 5.3 59.7 ± 0.9 – 53.2 ± 0.97 ± 6.2 – – –3 ± 0.45 20.1 ± 0.1 4.86 4.10 ± 1.759 ± 1.4 15.8 ± 0.9 8.13 14.98 ± 1.167 – – –9 – – –9 – – –4 – – –3 – – –2 ± 0.26 7.3 ± 0.1 6.6 7.31 ± 0.229 ± 0.09 – – 0.50 ± 0.146 ± 0.04 4.8 ± 0.1 5.3 ± 0.1 4.91 ± 0.063 ± 0.06 – – 7.52 ± 0.922 ± 0.2 6.72 – 4.8 ± 0.15 ± 21.3 – 300 ± 20 242.3 ± 7.3

Table 3Comparison of the metal element levels in the food wastes and the piggery wastewaters.

Elements Food waste Piggery wastewater

Zhu et al. (2008) Zhang et al. (2007) This study Moral et al. (2008) Creamer et al. (2010) This study

All values below (mg/L) were adjusted to an 8 wt.% of TS concentration.a

Sodium (Na) 1003.5 –b 1528.8 3171.8 ± 1832.6 248.0 919.2Magnesium (Mg) 87.7 112 ± 8 62.5 – 881.6 1018.4Aluminum (Al) – – 4.31 – – 62.55Potassium (K) 1122.8 720 ± 88 546.7 – 801.6 5995.2Calcium (Ca) 266.7 1728 ± 232 118.2 – – 2689.4Chromium (Cr) – – 0.17 3.88 ± 4.05 – 0.26Manganese (Mn) 0.84 15.53 ± 7.77 0.96 88.11 ± 112.78 72.16 37.78Iron (Fe) 9.47 198.32 ± 104.08 3.17 447.58 ± 563.88 284.16 149.86Cobalt (Co) – – <LDc 0.493 ± 0.564 – 0.18Nickel (Ni) – 0.52 ± 0.26 0.19 3.31 ± 3.31 0.96 0.69Copper (Cu) 1.19 8.03 ± 0.26 3.06 148.02 ± 179.74 21.92 59.36Zinc (Zn) 2.53 19.68 ± 5.70 8.27 606.17 ± 620.26 213.12 234.15Molybdenum (Mo) 0.070 – 0.025 – – 0.634Cadmium (Cd) – <0.26 0.023 0.35 ± 0.32 0.16 0.022Lead (Pb) – 1.04 ± 0.78 0.18 2.29 ± 1.94 – 0.51

a All the original measured values were corrected for the samples containing 8% (w/w) TS in order to quantitatively compare concentrations of metals on the same massbasis of each organic waste.

b –, Not available.c <LD, lower than detection limit (detection limit: Cd, Co > 30 lg/L).

L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059 5051

waste (479.5 mL CH4/g VSadded). The low BMP of the animalmanure was also reported elsewhere (Mackie and Bryant, 1995;Hansen et al. 1998). The high BMP values for food wastes (435–489 mL CH4/g of VSadded) were observed by other researchers aswell (Heo et al., 2003; Zhang et al., 2007). The high organic content(VS and COD) together with the high biodegradability (BMP) sug-gested that the food waste was a highly desirable substrate foranaerobic digestion.

The nitrogenous substances were also measured and shown inTable 2. The carbon to nitrogen (C/N) ratios were 13.2 and 4.8 forthe food waste and the piggery wastewater, respectively, whichwere comparable to the literature report as shown in Table 2.The C/N ratio suggested that the food waste was near the optimalrange (15.5–25) (Wu et al., 2010), while the C/N ratio of the piggerywastewater seemed low, which was due to the high concentrationof ammonia. The food waste and the piggery wastewater used inthis study contained significant concentrations of TKN as well.However, the nitrogen forms were quite different. In the foodwaste, most of nitrogen existed as the organic nitrogen like pro-teins, while the majority of nitrogen was ammonia in the piggerywastewater. The elemental analysis results also indicated thatthe food waste contained balanced concentrations of phosphorusand sulfur.

3.2. Trace elements

Table 3 shows the trace element contents in different feed-stocks. Both of the food waste and the piggery wastewater con-tained high concentrations of macronutrients (Na, K, Ca, Mg)ranged from 100 to 6000 mg/L. It was believed that the concentra-tions of macronutrients were high enough for metabolic activitiesof microorganisms. By contrast, the food waste and the piggerywastewater were found to contain low concentrations of all traceelements and heavy metals as compared to macronutrients. In par-ticular, the levels of most inorganic elements in the piggery waste-water were considerably higher than those in the food waste. Forexample, the concentrations of Fe, Ni and Mo in the food wastewere only about one tenth of those in the piggery wastewater.Moreover, the food waste contained a below-detection level of co-balt. Considering the important roles of these trace elements (Co,Ni, Mo, Fe) for activating and maintaining enzyme activities ofanaerobic microorganisms (Agler et al., 2008; Jarvis et al., 1997;

Murray and van den Berg, 1981; Wilkie et al., 1986; Williamset al., 1986; Zitomer et al., 2008), these trace elements might beinsufficient for stable and efficient anaerobic digestion of the foodwaste alone. Therefore, it was expected that the concentrations ofessential trace elements could be properly adjusted by mixing thefood waste and the piggery wastewater.

3.3. Fractionation of piggery wastewater

The liquid and solid fractions of the piggery wastewater con-tained quite different concentrations of ammonia, macronutrients,micronutrients, and alkalinity. As expected, most alkalinity andammonia were included in the liquid fraction (data not shown).The distribution of metal elements in the liquid and the solid frac-tions is presented in Fig. 1. Except Na+ and K+, most trace elementswere present in the solid fraction. This result was in agreementwith Moral et al. (2008), who also found that 95% of trace elementswere located in the solid fraction of the pig slurry. These two frac-tions of the piggery wastewater were prepared by centrifugationfor subsequent co-digestion experiments with the food waste inorder to investigate which fraction was more effective for improv-ing the performance of anaerobic digestion.

3.4. Anaerobic digestion experiments

3.4.1. Anaerobic co-digestion of food waste and piggery wastewater(Experiment 1)

Fig. 2 shows the time course of anaerobic co-digestion of thefood waste and the piggery wastewater as compared to anaerobicdigestion of the food waste or the piggery wastewater alone. Theorganic loading rate (OLR) was fixed at 6.35 g COD/L day exceptfor anaerobic digestion of the piggery wastewater, the OLR ofwhich was 4.71 g COD/L day due to its energy depleted feature.The OLR in terms of VS were 2.65, 4.79, 4.86 and 4.36 g VS/L dayfor EX1-1, EX1-2, EX1-3 and EX1-4 (Table 1), respectively. For allcases, the HRT was set at 20 days. When the food waste was usedas a sole substrate, the methane production rate decreased pro-gressively from 1.35 L CH4/L day on Day 18 to zero on Day 35,and the steady state was not reached during 47 days of operation.Concurrently, the total concentrations of VFA increased up to18,000 mg/L, by which pH decreased from pH 7.2 on Day 18 topH 4.4 on Day 47. Interestingly, it was found that propionicacid firstly appeared as the main component among the total

Metal elementNa Mg Al K Ca Cr Mn Fe Co Ni Cu Zn Mo Ag Cd

Met

al e

lem

ent d

istr

ibut

ion

(%)

0

10

20

30

40

50

60

70

80

90

100

Liquid fractionSolid fraction

Fig. 1. The distribution of metals in the solid and the liquid fractions of piggery wastewater.

5052 L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059

accumulated VFAs. For example, when total VFA concentration was9 804 mg/L (on Day 32 for anaerobic digestion of food waste onlyin Fig. 2C), the concentrations of acetic acid, propionic acid, n-butyric acid, iso-butyric acid, n-valeric acid and iso-valeric acidwere 548, 5247, 336, 488, 1077 and 2108 mg/L, respectively. Theseresults indicated that the acetogenesis might be a limiting step inanaerobic digestion of food waste only. In addition, the unstableprocess performances were also observed in anaerobic digestionof the food waste alone even at a reduced OLR (4.3 g COD/L day)and a longer HRT (30 days) (data not shown). El-Mashad et al.(2008) also found that the single stage digester treating food wastewas not stable at a OLR (organic loading rate) of 4.0 g VS/L day oreven at a reduced OLR of 2.0 g VS/L day, as indicated by a low bio-gas production rate, high concentrations of volatile fatty acid (VFA)and pH drop. Similarly, Climenhaga and Banks (2008) and Bankset al. (2011) also found that the residual concentrations of VFAswere high (about 15 g/L), indicating that energy loss was signifi-cant. Lee et al. (1999) and Cho et al. (1995) reached a similar con-clusion in that the single-stage anaerobic digestion was verydifficult for the easily degradable Korean food waste.

Although the stable performance was obtained during the oper-ating period, anaerobic digestion of the piggery wastewater aloneshowed a lower methane yield (187 mL CH4/g VSadded) than twoco-digestion trials (358 mL CH4/g VSadded for 7% PW/93% FW,388 mL CH4/g VSadded for 17% PW/83% FW). The low methane yieldfrom piggery wastewater only was similar to that obtained fromthe bench scale anaerobic reactor. It was attributed to the inherentlow BMP of the piggery wastewater (242.3 mL CH4/g VSadded) ascompared to that of the food waste (479.5 mL CH4/g VSadded) (Table2). Another reason might be ascribed to the ammonia inhibition,since a large proportion of ammonium nitrogen existed as freeammonia at high pH values (pH 8.5–8.8).

Compared to the failure of methane production from the foodwaste only after the operating period of 35 days, the methane pro-duction from the co-digestion reactors were much higher and sta-ble. The increasing percentage of the piggery wastewater greatlyimproved the system performance and stability as indicated bythe high methane yield, low effluent VFA level and stable pH valuethroughout the experimental period. By contrast, anaerobic co-digestion with the lower mixing ratio (7% PW and 93% FW basedon COD) showed less stable profile as indicated by the reducedmethane production rate and chronic VFA accumulation in the latestage (Day 30–45).

In summary, the co-digestion results demonstrated that it waspossible to obtain stable and active single-stage anaerobic diges-tion of the food waste by co-digesting with a small amount ofthe piggery wastewater (17% organic strength on COD basis). Onthe side of the piggery wastewater, the food waste in co-digestionprocess greatly increased the volumetric methane productivity ascompared to anaerobic digestion of the piggery wastewater alone.These results suggested that some factors were limiting in the foodwaste, and this limitation was corrected by providing essential fac-tors from the piggery wastewater via co-digestion strategy.

3.4.2. Anaerobic co-digestion of food waste with solid and liquidfractions of piggery wastewater (Experiment 2)

(Fig. 2) clearly showed that the piggery wastewater addition ata moderate ratio improved anaerobic digestion of the food waste.However, it was still not clear what factors of the piggery wastewa-ter were responsible for improving process performance of the co-digestion. As mentioned above, the liquid fraction and the solidfraction of the piggery wastewater obtained by centrifugationshowed quite different characteristics. For example, large parts ofalkalinity, nitrogen sources, and soluble organic materials were in-cluded in the liquid fraction. By contrast, the solid fraction con-tained more trace elements (Fig. 1). In order to identify theeffective factors of the piggery wastewater that increased themethane production rate, the food waste was co-digested withthe different fractions of the piggery wastewater (whole piggerywastewater, liquid fraction and solid fraction). For avoiding the ef-fect of different organic loading rates (OLR), the strength of feedingsubstrates was adjusted to the same level (6.35 g COD/L day), 83%of which was contributed by the food waste and the other by thefractionated piggery wastewater or whole piggery wastewater.

Figs. 3 and 4 show the process performance during semi-continuous anaerobic co-digestion of the food waste with the dif-ferent fractions of the piggery wastewater. As shown in Fig. 2,the anaerobic digestion of the food waste only became unstableafter 40 days, and the co-digestion of the food waste with thewhole piggery wastewater showed a very stable performance until68 days of incubation. When the food waste was co-digested withthe liquid fraction of the piggery wastewater, the process perfor-mance was unstable during 60 days of operation as demonstratedby the declining methane production rate (from 1.5 L CH4/L day tozero), lower pH value (pH 5.2), high residual soluble COD level(46,000 mg/L) and high total VFA level (about 20,000 mg/L) in the

0 5 10 15 20 25 30 35 40 45 50

pH v

alue

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

0 5 10 15 20 25 30 35 40 45 50

CH

4 pro

duct

ion

rate

(L/L

. day

)

0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

100% PW100% FW7% PW + 93% FW17% PW + 83% FW

Operating time (day)0 5 10 15 20 25 30 35 40 45 50

Tota

l VFA

con

cent

ratio

n (m

g/L)

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

A

B

C

Fig. 2. Performance of anaerobic co-digestion of the food waste (FW) and the piggery wastewater (PW) at two mixing ratios (7% PW + 93% FW; 17% PW + 83% FW on CODbasis) as compared to anaerobic digestion of the food waste and the piggery wastewater only. The OLR was adjusted to be 6.35 g COD/L day by adding distilled water (DW)except for anaerobic digestion of piggery wastewater alone, for which the OLR was 4.71 g COD/L day.

L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059 5053

effluent. Interestingly, when the food waste was co-digested withthe solid fraction of the piggery wastewater, a superior perfor-mance was obtained as anaerobic co-digestion with the whole pig-gery wastewater in terms of the high methane production rate(about 1.5 L CH4/L day), low residual soluble COD and no VFA accu-mulation (Figs. 3 and 4). The constant pH (pH 7.4–7.6) suggestedthat extraneously added buffering agents were not necessary

because robust methanogenesis consumed VFA and producedenough alkalinity.

Addition of buffering agents and pH control are often adoptedstrategies to prevent the failure of anaerobic digestion accompa-nied by VFA accumulation. We had also attempted the pH controlby adding lime, NaOH or NaHCO3 for anaerobic digestion of thefood waste alone. But the steady state was not be achieved, as

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70

CH

4 pro

duct

ion

rate

(L/L

. day

)

0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

2.25

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70

CH

4 con

tent

(%)

0

10

20

30

40

50

60

70

100% FW83% FW + 17% Whole PW83% FW + 17% PW liquid fraction83% FW + 17% PW solid fraction

Operating time (day)0 5 10 15 20 25 30 35 40 45 50 55 60 65 70

pH v

alue

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

A

B

C

Fig. 3. Performance of anaerobic co-digestion of the food waste (FW) with different fractions of the piggery wastewater (PW) obtained by centrifugation. The organic loadingrate (OLR) was set at 6.35 g COD/L day. The mixing percentages were based on COD.

5054 L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059

indicated by the methane production rate progressively decreasedafter about two HRTs, even when pH values were remained in theoptimum range (pH 6.8–7.5) (data not shown). Therefore, it wasbelieved that the accumulation of VFA was mainly due to the slowbiomass synthesis and reduced metabolic activities of the metha-nogenic microorganisms. Similarly, Kayhanian and Rich (1995)also found that the addition of alkalinity was ineffective for anaer-

obic digestion of the biodegradable organic fraction of MSW. Crea-mer et al. (2010) also observed that anaerobic digestion ofdissolved air flotation (DAF) sludge under conditions of HRT of10 days and OLR of 5.24 g VS/L day was highly unstable, character-ized by rapid and irreversible pH drop. Attempts were made to buf-fer the digester with lime but stable operation was still difficult toobtain. These results suggested that there must be reasons other

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70

Solu

ble

CO

D (m

g/L)

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70

Tota

l VFA

con

cent

ratio

n (m

g/L)

0

2500

5000

7500

10000

12500

15000

17500

20000

22500

25000

100% FW83% FW + 17% Whole PW83% FW + 17% PW liquid fraction83% FW + 17% PW solid fraction

Operating time (day)0 5 10 15 20 25 30 35 40 45 50 55 60 65 70

Am

mon

ia-N

con

cent

ratio

n (m

g/L)

0

500

1000

1500

2000

2500

3000

3500

4000

A

B

C

Fig. 4. Profiles of soluble COD, total VFA and ammonia nitrogen in anaerobic co-digestion of the food waste (FW) with the different fractions of the piggery wastewater (PW)obtained by centrifugation at the identical organic loading rate (6.35 g COD/L day). The mixing percentages were based on COD.

L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059 5055

than the buffering agents introduced from the piggery wastewaterto increase the process stability of anaerobic digestion of the foodwaste alone.

C/N ratio and ammonia level were another important issues inanaerobic digestion of animal manure and other organic wastes(Hansen et al., 1998, 1999; Romano and Zhang, 2008; Wu et al.,

2010). The high C/N ratio might result in the nitrogen deficiency,and the low C/N ratio could lead to ammonia inhibition. Fig. 4Cshows the ammonia profile during anaerobic co-digestion of thefood waste with the different fractions of the piggery wastewaterand anaerobic digestion of the food waste only. Due to the lowerC/N ratio (4.8) of the piggery wastewater as compared to the food

5056 L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059

waste (13.2), co-digestion of the food waste with the piggerywastewater resulted in a higher ammonia level than anaerobicdigestion of the food waste only. Since most nitrogen was con-tained in the liquid fraction of the piggery wastewater (data notshown), co-digestion of the food waste and the liquid fractionshowed the highest ammonia level at steady state (2000–3000 mg/L ammonia-N). In contrast, co-digestion of food wastewith the solid fraction of piggery wastewater showed a lowerammonia concentration (1000–1500 mg/L ammonia-N), whichwas quite similar to anaerobic digestion of the food waste only.Ammonia level in the anaerobic digestate of the food waste fellin the optimum range for anaerobic digestion (Zhang and Jahng,2010). The C/N value of the food waste (13.2) also suggested thatthe nitrogen content was close to the optimum range (15–30).Since anaerobic co-digestion of the food waste with the solid frac-tion of the piggery wastewater showed superior performance ascompared to the failure of the anaerobic digestion of the foodwaste only, it was likely that the improvement of anaerobic diges-tion of the food waste by co-digestion with the piggery wastewaterwas not due to the introduction of extra nitrogen sources from thepiggery wastewater.

These results indicated that other substances must be containedexclusively in the solid fraction of the piggery wastewater. FromFig. 1, it was noticed that the solid fraction of the piggery wastewa-ter contained much higher concentrations of trace elements thanthe liquid fraction. Furthermore, the food waste contained lessertrace elements (Table 3). Considering previous reports supportingthat the trace element supplementation enhanced anaerobic diges-tion of various substrates, as in the anaerobic digestion of a grass-clover silage by adding cobalt (Co) (Jarvis et al., 1997), Napiergrassby adding nickel (Ni), cobalt (Co), molybdenum (Mo), selenium (Se)and sulfur (S) (Wilkie et al., 1986), and a thin stillage from corngrain ethanol process by adding cobalt (Co) (Agler et al., 2008), itseemed that the trace elements introduced from whole or the solidfraction of the piggery wastewater reversed the trace element defi-ciency of the food waste.

3.4.3. Effects of trace elements supplementation on a long-termanaerobic digestion of the food waste (Experiment 3)

As mentioned above, the trace elements supplied from the pig-gery wastewater seemed to increase the process stability of anaer-obic digestion of the food waste. In order to prove this speculation,a trace element solution instead of the piggery wastewater wasadded to the food waste, and anaerobic digestion was carried outfor a prolonged period (188 days). The trace element solutionwas designed to contain cobalt (Co), molybdenum (Mo), nickel(Ni) and ion (Fe), since these elements were considered to enhancethe performance of anaerobic digestion of different substrates(Agler et al., 2008; Jarvis et al., 1997; Murray and van den Berg,1981; Wilkie et al., 1986; Williams et al., 1986; Zitomer et al.,2008). In addition, these elements were found to be rich in thepiggery wastewater but significantly poor in the food waste (Table 3).

Two runs of co-digestion of the food waste with the whole pig-gery wastewater (EX2-2) and the solid fraction of the piggerywastewater (EX2-4) in Experiment 2 were continued until 92 daysfor Experiment 3, and named as Run 1 and Run 2, respectively.During Day 0–67, the excellent performances were observed inboth digesters, which had already been reported in Figs. 3 and 4.However, due to irregular feeding during Day 68–74, the processupsets occurred as indicated by the VFA accumulation and an in-creased SCOD (soluble chemical oxygen demand) level. In orderto avoid the process failure, the feeding schedule was changed toevery other day on Day 93, and correspondingly the HRT increasedfrom 20 to 40 days. At that time point (on Day 93), the feeding sub-strate for Run 2 was also changed to the mixture of the food wasteand the whole piggery wastewater as Run 1. After two more

months of operation, the two co-digestion processes were stabi-lized with similar performances. On Day 156, the HRT was changedto 30 days by daily feeding and withdrawing. The digesters contin-ued their stable performance after the change of HRT. The results ofco-digestion of the food waste and the piggery wastewater (Day 0–161) indicated that it was possible to achieve stable continuousoperations under varying conditions of HRT and OLR.

From Day 162, the feeding substrate of both digesters (Run 1and 2) was changed to the food waste only. Run 1 was served asthe control without trace element addition during the remainingoperation period. In order to evaluate the effect of adding trace ele-ments on the anaerobic digestion of the food waste only, 0.67 mLof the trace element stock solution was supplemented daily toRun 2 for 3 times during Day 180–182, to obtain the desired traceelement levels in the digestate (2.0 mg/L of Co2+, 5.0 mg/L of Mo2+,10 mg/L of Ni2+, 100 mg/L of Fe3+). In the remaining operation per-iod, 0.67 mL of the trace element stock solution was added every6–12 days. The concentrations of added trace elements in the di-gester were supposed to be 2.0 mg Co2+/L, 5.0 mg Mo2+/L, 10 mgNi2+/L, and 100 mg Fe3+/L, assuming no volume change.

As shown in Fig. 5, after about 20 days of feeding the food wastewithout the piggery wastewater from Day 163, the methane pro-duction rate of Run 1 progressively decreased from 1.3 L CH4/L dayto zero on Day 209. During this period, pH decreased from pH 7.5to 5.2 on Day 220. These results were in agreement with the anaer-obic digestion of the food waste alone in Figs. 2 and 3, and con-firmed that it was infeasible to anaerobically digest the foodwaste in a single stage reactor even at the conditions of a reducedOLR (4.3 g COD/L day) and an increased HRT (30 days). By contrast,Run 2 which was fed with the trace element solution from Day 180showed a very stable methane production throughout the remain-ing experimental period (Day 180–367). The methane contentslightly decreased from 60% to 52%, which might be due to the highcarbohydrate content of the food waste (Table 2). The pH main-tained constant between pH 7.4 and 7.6 without alkali addition.This result again confirmed that the buffering capacity introducedfrom the piggery wastewater was not the contributing factor forenhancing anaerobic digestion of the food waste only.

Fig. 6 shows the organic profiles of Run 1 and Run 2. In Run 1,after 20 days of feeding the food waste only, concomitantly withthe declining methane production rate, the rapid accumulationof soluble COD and total VFA, the increased levels of TS, VS andVS/TS ratio were observed. These results indicated that the metab-olism on organic materials was seriously affected in anaerobicdigestion of the food waste only. By contrast, in Run 2, the supple-mentation of the trace elements resulted in superior performancesin terms of higher TS and VS destruction, lower residual concentra-tions of soluble COD and negligible VFA level. In Run 2, the residualsoluble COD gradually decreased from around 10,000 to about3000 mg/L, and the total VFA level also quickly decreased from2000 to lower than 100 mg/L after starting feeding of the traceelement solution. Although the VFA concentration increased whenthe HRT decreased from 30 to 20 days, it was mainly due to theincreased OLR. After subsequent 30 days of continuous operation,the VFA level was stabilized at 300–400 mg/L. The second VFAaccumulation in Run 2 between Day 317 and 360 was due to thetemperature control problem (temperature unexpectedly jumpedfrom 37 to 55 �C for about 12 h). After correcting the temperatureproblem, the VFA concentration was gradually decreased and sta-bilized at a low level.

The steady state performance data of anaerobic digestion of foodwaste with trace element addition were summarized in Table 4together with two relevant studies (Banks et al., 2011); Zhanget al. 2007). The methane content (52%) in this study was lowerthan others (73.14% and 62.6%), which might be due to theacidification of food waste before feeding. The methane yield was

Fig. 5. Methane production rate, methane yield and pH profile during a long-term operation under varying conditions (HRT, OLR and type of feedstock). On Day 162, thefeeding material for both Run 1 and 2 was changed to the food waste without the piggery wastewater. Arrows (;) indicate the addition time points of the trace elementsolution for Run 2.

L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059 5057

comparable with that of Banks et al. (2011), and high VS destruc-tion rate (75.6%) was achieved. The volumetric methane productiv-ity of this study was much higher than that of Banks et al. (2011)(1.34 vs 1.00 L/L day), indicating that the economic feasibility couldbe improved. Interestingly, the total VFA concentration of the

effluent of this study was significantly lower than that of Bankset al. (2011), suggesting that the trace elements played an essentialrole in VFA metabolism.

In short, the results indicated that the food waste was deficientin some trace elements required for robust and stable anaerobic

Fig. 6. Organic profiles (total solid/volatile solid, soluble COD and total VFA) during a long-term operation under varying conditions (HRT, OLR and type of feedstock). On Day162, the feeding material for both Run 1 and 2 was changed to the food waste without the piggery wastewater. Arrows (;) indicate the addition time points of the traceelement solution for Run 2.

5058 L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059

digestion. Anaerobic digestion of the food waste supplementedwith the trace element-rich piggery wastewater or synthetic traceelements resulted in the significantly improved biogas productionrate and the enhanced process stability.

4. Conclusions

This study showed that anaerobic digestion of the food wastealone in a single stage reactor was not feasible under examined

conditions even with pH control. By contrast, co-digestion ofthe food waste with the piggery wastewater showed a high meth-ane production rate without VFA accumulation. By adding a traceelement solution instead of the piggery wastewater to the foodwaste, a superior performance was reproduced in terms of thehigh methane yield and no VFA accumulation. From these results,it was suggested that trace elements supplemented from the pig-gery wastewater was the key factors enhancing co-digestionperformances.

Table 4The performance of anaerobic digestion of food waste only in this study as comparedwith literature report.

Item Zhang et al. (2007) Banks et al. (2011) This study

Source of food waste Source separatedfood waste

Source segregateddomestic food waste

Restaurant

Type of reactor Batch CSTR CSTRHRT (day) 28 80 20pH value 7.57 8.13 7.37Total VFA

concentration(mg/L)

– 15,000 1279

Methane content (%) 73.14 62.6 52Methane yield

(mL/g VSadded)440 402 396

Methaneproductivity(L/L day)

– 1.00 1.34

VS destruction (%) 80.57 – 75.6

L. Zhang et al. / Bioresource Technology 102 (2011) 5048–5059 5059

Acknowledgements

This work was supported by the KENTEC (2009) and the PriorityResearch Centers Program through the National Research Founda-tion of Korea (NRF) funded by the Ministry of Education, Scienceand Technology (2010-0028300).

References

Agler, M.T., Garcia, M.L., Lee, E.S., Schlicher, M., Angenent, L.T., 2008. Thermophilicanaerobic digestion to increase the net energy balance of corn grain ethanol.Environ. Sci. Technol. 42, 6723–6729.

Ahn, J.-H., Do, T.H., Kim, S.D., Hwang, S., 2006. The effect of calcium on the anaerobicdigestion treating swine wastewater. Biochem. Eng. J. 30, 33–38.

Angelidaki, I., Ellegaard, L., 2003. Codigestion of manure and organic wastes incentralized biogas plants. Appl. Biochem. Biotechnol. 109, 95–105.

APHA. 2005. Standard Methods for the Examination of Water and Wastewater. 21sted. American Public Health Association (APHA), Washington, DC.

Banks, C.J., Chesshire, M., Heaven, S., Arnold, R., 2011. Anaerobic digestion of source-segregated domestic food waste: performance assessment by mass and energybalance. Bioresour. Technol. 102, 612–620.

Bligh, E., Dyer, W., 1959. A rapid method of total lipid extraction and purification.Can. J. Physiol. Pharmacol. 37, 911–917.

Braun, R., Brachtl, E., Grasmug, M., 2003. Codigestion of proteinaceous industrialwaste. Appl. Biochem. Biotechnol. 109, 139–153.

Cho, J.K., Park, S.C., Chang, H.N., 1995. Biochemical methane potential and solidstate anaerobic digestion of Korean food wastes. Bioresour. Technol. 52, 245–253.

Climenhaga, M., Banks, C., 2008. Anaerobic digestion of catering wastes: effect ofmicronutrients and retention time. Water Sci. Technol. 57, 687–692.

Creamer, K.S., Chen, Y., Williams, C.M., Cheng, J.J., 2010. Stable thermophilicanaerobic digestion of dissolved air flotation (DAF) sludge by co-digestion withswine manure. Bioresour. Technol. 101, 3020–3024.

De Baere, L., 2000. Anaerobic digestion of solid waste: state-of-the-art. Water Sci.Technol. 41, 283–290.

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.

Han, S.-K., Shin, H.-S., 2004. Biohydrogen production by anaerobic fermentation offood waste. Int. J. Hydrogen Energy 29, 569–577.

Hansen, K., Angelidaki, I., Ahring, B., 1998. Anaerobic digestion of swine manure:inhibition by ammonia. Water Res. 32, 5–12.

Hansen, K., Angelidaki, I., Ahring, B., 1999. Improving thermophilic anaerobicdigestion of swine manure. Water Res. 33, 1805–1810.

Heo, N., Park, S., Lee, J., Kang, H., Park, D., 2003. Single-stage anaerobic codigestionfor mixture wastes of simulated Korean food waste and waste activated sludge.Appl. Biochem. Biotechnol. 107, 567–579.

Jarvis, Å., Nordberg, Å., Jarlsvik, T., Mathisen, B., Svensson, B.H., 1997. Improvementof a grass-clover silage-fed biogas process by the addition of cobalt. BiomassBioenergy 12, 453–460.

Kayhanian, M., Rich, D., 1995. Pilot-scale high solids thermophilic anaerobicdigestion of municipal solid waste with an emphasis on nutrientrequirements. Biomass Bioenergy 8, 433–444.

Kelley, T.R., Walker, P.M., 2000. Bacterial concentration reduction in swine wasteamended livestock feed using a single-screw dry-extrusion process. Bioresour.Technol. 75, 189–195.

Lee, J., Lee, J., Park, S., 1999. Two-phase methanization of food wastes in pilot scale.Appl. Biochem. Biotechnol. 79, 585–593.

Mackie, R.I., Bryant, M.P., 1995. Anaerobic digestion of cattle waste at mesophilicand thermophilic temperatures. Appl. Microbiol. Biotechnol. 43, 346–350.

Ministry of Environment, Republic of Korea. Environmental Whitebook. http://www.me.go.kr. [accessed 18 October 2010].

Moral, R., Perez-Murcia, M.D., Perez-Espinosa, A., Moreno-Caselles, J., Paredes, C.,Rufete, B., 2008. Salinity, organic content, micronutrients and heavy metals inpig slurries from South-eastern Spain. Waste Manage. 28, 367–371.

Murray, W.D., van den Berg, L., 1981. Effects of nickel, cobalt, and molybdenum onperformance of methanogenic fixed-film reactors. Appl. Environ. Microbiol. 42,502–505.

Oh, G., Zhang, L., Jahng, D., 2008. Osmoprotectants enhance methane productionfrom the anaerobic digestion of food wastes containing a high content of salt. J.Chem. Technol. Biotechnol. 83, 1204–1210.

Romano, R., Zhang, R., 2008. Co-digestion of onion juice and wastewater sludgeusing an anaerobic mixed biofilm reactor. Bioresour. Technol. 99, 631–637.

Wang, J., Zhang, H., Stabnikova, O., Ang, S., Tay, J., 2005. A hybrid anaerobic solid–liquid system for food waste digestion. Water Sci. Technol. 52, 223–228.

Weiland, P., 2000. Anaerobic waste digestion in Germany: status and recentdevelopments. Biodegradation 11, 415–421.

Wilkie, A., Goto, M., Bordeaux, F.M., Smith, P.H., 1986. Enhancement of anaerobicmethanogenesis from Napiergrass by addition of micronutrients. Biomass 11,135–146.

Williams, C.M., Shih, J.C.H., Spears, J.W., 1986. Effect of nickel on biological methanegeneration from a laboratory poultry waste digester. Biotechnol. Bioeng. 28,1608–1610.

Wu, X., Yao, W., Zhu, J., Miller, C., 2010. Biogas and CH4 productivity by co-digestingswine manure with three crop residues as an external carbon source. Bioresour.Technol. 101, 4042–4047.

Zhang, L., Jahng, D., 2010. Enhanced anaerobic digestion of piggery wastewater byammonia stripping: effects of alkali types. J. Hazard. Mater. 182, 536–543.

Zhang, R., El-Mashad, H.M., Hartman, K., Wang, F., Liu, G., Choate, C., Gamble, P.,2007. Characterization of food waste as feedstock for anaerobic digestion.Bioresour. Technol. 98, 929–935.

Zhu, H., Parker, W., Basnar, R., Proracki, A., Falletta, P., Béland, M., Seto, P., 2008.Biohydrogen production by anaerobic co-digestion of municipal food waste andsewage sludges. Int. J. Hydrogen Energy 33, 3651–3659.

Zitomer, D., Johnson, C., Speece, R., 2008. Metal stimulation and municipal digesterthermophilic/mesophilic activity. J. Environ. Eng. 134, 42–47.