The methane production potential of kitchen waste (KW) obtained from different sources was compared through mesophilic and thermophilic anaerobic digestion. The methane yields (MYs) obtained with the same KW sample under different temperatures were similar, whereas the MYs obtained with different samples differed significantly. The highest MY obtained in S7 was 54%–60% higher than the lowest MY in S3. The modified Gompertz model was utilized to simulate the methane production process. The maximum production rate of methane under thermophilic conditions was 2%–86% higher than that under mesophilic conditions. The characteristics of different KW samples were studied. In the distribution of total chemical oxygen demand, the diversity of organic compounds of KW was the most dominant factor that affected the potential MYs of KW. The effect of the C/N and C/P ratios or the concentration of metal ions was insignificant. Two typical methods to calculate the theoretical MY (TMY) were compared, the organic composition method can simulate methane production more precisely than the elemental analysis method. Significant linear correlations were found between TMYorg and MYs under mesophilic and thermophilic conditions. The organic composition method can thus be utilized as a fast technique to predict the methane production potential of KW.

With worldwide economic development and population growth, the amount of kitchen waste (KW) produced by hotels, restaurants, families, canteens, and companies increases. KW is one of the largest waste streams. The amount of KW in China reached 90 million tons in 2010 (Walker et al. 2011). Current KW in China usually contains a large amount of volatile solids (VS) with 70%–90% moisture content (Zhang et al. 2007) and usually accounts for 40%–50% (W/W) of municipal solid waste (Dai et al. 2013). Improper treatment of KW causes serious environmental problems, such as bad odor, leachate production, and groundwater contamination.

The main means of KW disposal in China are incineration and use of landfills. The use of landfills as KW disposal sites has been banned in many countries. Meanwhile, incineration is energy intensive because of the high moisture content of KW, and often results in air pollution (Zhang et al. 2014). Among all the means to deal with KW, anaerobic digestion (AD), which is utilized for industrial or domestic purposes to manage waste and produce fuel, is the most effective technology for the treatment of organic waste. Considering the high concentration of organic matter in KW, KW should be reutilized as a valuable biomass resource (Jiang et al. 2013). The main advantage of this process is that the product can be used as vehicle fuel or for co-generation of electricity and heat; thus, greenhouse gas emissions can be reduced (Ye et al. 2013).

Different methane yields (MYs) have been obtained from KW. Liao et al. (2014) found that a system can be maintained stable with a MY of 369–466 mL/g VS. Wang et al. (2014) obtained a MY of 493.38–652.71 mL/g VS from KW under different hydraulic retention times. Zhou et al. (2014) showed that the biomethane yield of KW was 507 mL/g VS. The difference among these MYs can be attributed to the different operation parameters and characteristics of KW, including different organic compositions, C/N/P ratios, and metal ion contents (Chen et al. 2008). Similar to other organic waste, KW is mainly composed of carbohydrates, proteins, and lipids. Lipids have the highest theoretical MY (TMY), but a high lipid content may lead to long-chain fatty acid accumulation and could further inhibit methane generation (Li et al. 2013c). Carbohydrates are easier to degrade than proteins and lipids (Lay et al. 2003). Thus, different compositions of KW may lead to different MYs. Nitrogen and phosphate are two of the most essential nutrients for efficient methane production because of their nutrient value and buffering capacity. Certain amounts of light (Na, K, Mg, Ca, Al, etc.) and heavy (Cr, Co, Cu, Zn, Ni, etc.) metal ions are also required by anaerobic bacteria because these cations play an important role in enzyme synthesis and maintenance of enzyme activities (Schattauer et al. 2011; Facchin et al. 2013). Kim et al. (2002) found that Ca2+ and Fe2+ can activate protease. Given all the factors that affect methane production ability, several expressions and definitions have been presented to calculate the TMY; consequently, comparison of biodegradability data from previous studies is difficult (Kaparaju et al. 2009; Triolo et al. 2011).

With these considerations, the objectives of this study are to: (1) compare the methane production potential, biodegradability, and degradation performance of KW from different sources under mesophilic and thermophilic conditions; (2) use the modified Gompertz model to simulate the methane generation process; and (3) compare different TMY calculations with experimental data and identify the factor that influences methane production ability the most.

Feedstock and inoculum

The seven KW samples used as substrates were collected from different areas, which were a mixture from lots of restaurants and cafeterias from different districts in Beijing. S1 and S2 were collected from Xicheng District, S3 and S4 were collected from Chaoyang District, S5 was obtained from Haidian District, S6 was obtained from Dongcheng District, those districts make up the central area of Beijing. S7 was gathered from the school cafeterias of Beijing University of Chemical Technology. The inoculum used in this study was digested sludge from a mesophilic anaerobic reactor in the Xiaohongmen municipal wastewater treatment plant in Beijing, China.

Experimental design and procedure

After being sampled, the inorganic fractions (plastic bags, bones, toothpicks and so on) were removed from the KW samples manually, and then the samples were shredded and mixed with a grinder (Waste King SS3300, Anaheim manufacturing Company, USA). The pretreated KWs were then stored at 4 °C before usage to prevent biological decomposition in advance. Before utilization, the inoculum was acclimated at 37 °C for about three weeks in order for degassing (Li et al. 2013a).

To evaluate the ultimate methane potential and biodegradability under mesophilic (37 °C) and thermophilic (55 °C) conditions, biochemical methane potential (BMP) assays of different KWs were carried out in triplicate, 500 mL identical bottles with a working volume of 300 mL were used as the reactor for mesophilic and thermophilic AD. The initial VS concentration of KWs was 2 g/L and the corresponding S/I ratio was 0.6 according to Zhou et al. (2011). After adding and mixing the needed amounts of substrates and inoculum in the reactors, tap water was added to fill up the working volumes. No additional nutrient solution was added into the reactors, and the headspace of each digester was purged with nitrogen gas for 3 min. All the digesters were performed in triplicate. All digesters were tightly closed with rubber stoppers and screw caps. Capped reactors were placed in two air bath shakers (YIHENG, China) set at temperatures of 37°C and 55°C, respectively, and rotated at 120 rpm. Three blank digesters that contained the same amount of inoculum and water were also operated. Biogas production and composition were taken every day, and the characteristics of effluent were measured after the BMP experiments when biogas productions were all ceased.

Analytical methods

After being centrifuged at 10,000 rpm for 10 min, the supernatant of samples were immediately analyzed as a soluble fraction. The pH value was determined by a pH meter (METTLER, Switzerland). Total solids (TS), VS, chemical oxygen demand (COD), total nitrogen (TN), total phosphorus (TP) and NH3-N were measured according to Standard Methods (APHA 1998). Lowry–Folin method was used to determine protein concentration with bovine serum albumin as standard (Lowry et al. 1951). Carbohydrate concentration was measured by the phenol sulfuric acid method, with glucose as standard (Dubois et al. 1956).

Total lipids content was determined by Soxhlet extraction using diethyl ether as solvent. Elemental compositions (C, H, N, S) of substrates were measured by an organic elemental analyzer (Vario EL cube, Germany), and element ‘O’ was estimated by assuming C + H + N + O = 99.5% on a VS basis (Rincon et al. 2012).

All metal concentrations were analyzed with an Inductively-Coupled Plasma-Mass Spectrophotometer (ThermoFisher Scientific).

Biogas (CH4 and CO2) volume was determined according to the pressure in the headspace, and the pressure was measured by a 3151 WAL-BMP-Test system pressure gauge (WAL Mess-und Regelsysteme GmbH, Germany). After the pressure measured (referred as P1, mbar), the biogas in the digester was released under water to prevent gas exchange, and the pressure was then measured again (used as P2 for the next day, mbar). The volume of biogas produced was calculated as in the following equation:
formula
1
where Vbiogas means daily biogas volume (L), Vhead represents the volume of the head space (L), M refers to the molar volume (22.41 L/mol), R stands for the universal gas constant (83.14 L mbar/K/mol), and T is absolute temperature (K).

Biogas compositions were measured by a gas chromatograph (Agilent, 7890B) equipped with a thermal conductivity detector and analytical column of Agilent Hayesep Q. The operational temperature at the column oven and the detector were 60 and 220 °C, respectively. Helium was used as a carrier gas at a pressure of 5 psi.

For the determination of ethanol and volatile fatty acid (VFA) (acetate, propionate, iso-butyrate, n-butyrate, iso-valerate, n-valerate) concentrations, the soluble fraction of samples was filtered through a 0.45 μm microfiber filter and the filtrates were determined by gas chromatograph (Agilent, 7890A) equipped with a flame ionization detector and DB-wax capillary column (30 m × 530 μm × 1.0 μm). The temperature of the injector and the detector were 200 and 250 °C, respectively. Nitrogen was the carrier gas, with a flow rate of 10 mL/min. The GC oven was programmed to begin at 55 °C and remain there for 1 min, then increase at a rate of 30 °C/min to 110 °C, and hold at 110 °C for an additional 1 min, and then increase at a rate of 30 °C/min to 220 °C, and hold at 220 C for an additional 1 min. The sample injection volume was 1.0 μL.

Data analysis

The BMP of substrates was evaluated based on their specific MY (SMY, the calculated methane production per gram VS, mL/g-VSadded), and the gas volume reported in this study was calibrated to a standard temperature (273 K) and pressure (1 atm).

An analysis of variance (ANOVA) was used to test the significance of the results, and p < 0.05 was considered to be statistically significant.

Biogas production

The methane production results are shown in Figures 1 and 2. The modified Gompertz model was utilized to simulate the methane production process (Equation (1)), and the results are summarized in Tables 1 and 2. In Equation (1), H represents the measured methane production rate (mL CH4/g VSadded), p is the maximum MY (mL CH4/g VSadded), R is the maximum production rate of methane in AD (mL CH4/VS*d), and a refers to the lag phase time (day).
formula
2
Figure 1

Cumulative MY under mesophilic AD.

Figure 1

Cumulative MY under mesophilic AD.

Figure 2

Cumulative MY under thermophilic AD.

Figure 2

Cumulative MY under thermophilic AD.

Table 1

Parameters of the modified Gompertz equation obtained with mesophilic methane fermentation

SamplesH (mL CH4/g VS)Modified Gompertz model
p (mL CH4/g VS)a (d)R (mL CH4/(g VS-d))R2
S1 491.1 488 ± 2.3 0.64 ± 0.14 46.22 ± 1.14 0.997 
S2 436.1 426 ± 2.1 0.62 ± 0.14 40.87 ± 1.02 0.997 
S3 353.2 350 ± 1.6 1.50 ± 0.13 39.44 ± 4.73 0.996 
S4 482.4 476 ± 1.3 0.57 ± 0.08 37.73 ± 1.50 0.999 
S5 439.7 436 ± 0.9 0.82 ± 0.06 50.08 ± 0.79 0.999 
S6 538.4 543 ± 4.0 0.87 ± 0.20 42.08 ± 1.25 0.995 
S7 565.2 557 ± 4.3 0.00 ± 0.23 44.93 ± 1.57 0.993 
SamplesH (mL CH4/g VS)Modified Gompertz model
p (mL CH4/g VS)a (d)R (mL CH4/(g VS-d))R2
S1 491.1 488 ± 2.3 0.64 ± 0.14 46.22 ± 1.14 0.997 
S2 436.1 426 ± 2.1 0.62 ± 0.14 40.87 ± 1.02 0.997 
S3 353.2 350 ± 1.6 1.50 ± 0.13 39.44 ± 4.73 0.996 
S4 482.4 476 ± 1.3 0.57 ± 0.08 37.73 ± 1.50 0.999 
S5 439.7 436 ± 0.9 0.82 ± 0.06 50.08 ± 0.79 0.999 
S6 538.4 543 ± 4.0 0.87 ± 0.20 42.08 ± 1.25 0.995 
S7 565.2 557 ± 4.3 0.00 ± 0.23 44.93 ± 1.57 0.993 
Table 2

Parameters of the modified Gompertz equation obtained with thermophilic methane fermentation

SamplesH (mL CH4/g VS)Modified Gompertz model
p (mL CH4/g VS)a (d)R (mL CH4/(g VS-d))R2
S1 497.1 485 ± 5.0 4.29 ± 0.26 48.98 ± 2.43 0.991 
S2 430.8 416 ± 4.0 4.74 ± 0.23 52.82 ± 2.96 0.991 
S3 373.9 364 ± 2.4 3.23 ± 0.20 40.05 ± 1.12 0.994 
S4 512.7 496 ± 3.4 3.47 ± 0.17 56.36 ± 0.75 0.995 
S5 444.5 420 ± 4.2 5.03 ± 0.23 68.41 ± 4.84 0.989 
S6 551.2 536 ± 4.2 5.39 ± 0.19 78.17 ± 3.99 0.994 
S7 574.7 560 ± 5.9 5.02 ± 0.26 64.18 ± 3.59 0.991 
SamplesH (mL CH4/g VS)Modified Gompertz model
p (mL CH4/g VS)a (d)R (mL CH4/(g VS-d))R2
S1 497.1 485 ± 5.0 4.29 ± 0.26 48.98 ± 2.43 0.991 
S2 430.8 416 ± 4.0 4.74 ± 0.23 52.82 ± 2.96 0.991 
S3 373.9 364 ± 2.4 3.23 ± 0.20 40.05 ± 1.12 0.994 
S4 512.7 496 ± 3.4 3.47 ± 0.17 56.36 ± 0.75 0.995 
S5 444.5 420 ± 4.2 5.03 ± 0.23 68.41 ± 4.84 0.989 
S6 551.2 536 ± 4.2 5.39 ± 0.19 78.17 ± 3.99 0.994 
S7 574.7 560 ± 5.9 5.02 ± 0.26 64.18 ± 3.59 0.991 

As shown in Figure 1, under mesophilic conditions, the reactors began to produce methane on the first day. The cumulated MY increased rapidly in the first 10 d and gradually reached a plateau. The cumulative MYs of S1 to S7 were 491.1, 436.1, 353.2, 482.4, 439.7, 538.4, and 565.2 mL, respectively. Although the initial VS of KW in each reactor was adjusted to 2 g/L and the operation parameters were unchanged, the cumulative MYs differed and followed the sequence of S7 > S6 > S1 > S4 > S5 > S2 > S3. An ANOVA was used to test the significance of results. The difference between some samples was indeed not significant, for example, MY of S7 was not significantly higher than S6 (P > 0.05). While, the highest MY of S7 was 60% higher than the lowest MY of S3 (P < 0.05). Table 1 shows that all the R2 values are above 0.99. This result suggests that the modified Gompertz model can effectively describe biogas generation from KW. The value of experimental H is similar to the value of methane production potential P; this similarity indicates the full recovery of methane generation.

Thermophilic methane production is illustrated in Figure 2. After 5 d of accommodation (the inoculum was collected from a mesophilic reactor), the MYs of S1 to S7 began to sharply increase and finally reached 497.1, 430.8, 373.9, 512.7, 444.5, 551.2, and 574.7 mL, respectively. The MYs differed significantly and followed the sequence of S7 > S6 > S4 > S1 > S5 > S2 > S3.

From Tables 1 and 2, the methane production performances could be explained well by the kinetic model since the determination coefficient (R2) ranged from 0.989 to 0.999. With the same KW sample, the MY obtained under thermophilic conditions is similar (only slightly higher) to the MY obtained under mesophilic conditions. It could also be seen that the lag phase time ‘a’ under thermophilic conditions (Table 2) was about 3 to 5 days, whereas the values of a are almost less than 1 d under mesophilic conditions. Although the start of methane generation under thermophilic conditions consumed more time than that under mesophilic conditions, the maximum methane production rate in the former was higher than in the latter. With the same KW sample, the values of R in Table 2 are 2%–86% higher than those in Table 1. This result can be attributed to the thermodynamic advantage. The values of experimental H in Table 2 are similar to the values of methane production potential p. This similarity indicates that the highest methane production potential was obtained under thermophilic conditions. This result proves that mesophilic inoculum can adapt and have activities under thermophilic conditions. Recent studies showed that different inoculum had a difference in starting time, and inoculum with higher enzyme activities and suitable nutrient content achieved higher biogas production (Gu et al. 2014). So thermophilic inoculum should be considered in the future thermophilic tests.

Characteristics of the substrates

The characteristics of different KW samples are summarized in Table 3. Significant differences were observed among these characteristics.

Table 3

Characteristics of different KW samples as substrates

 S1S2S3S4S5S6S7
pH 4.47 4.74 4.50 4.07 4.85 4.63 3.94 
TS (%) 22.16 21.72 16.28 24.30 17.78 21.29 13.95 
VS (%) 20.72 19.38 14.00 21.44 14.56 17.88 11.47 
TCOD (g/g TS) 0.88 1.07 1.11 1.21 1.17 1.04 1.19 
SCOD (g/g TS) 0.34 0.43 0.53 0.57 0.62 0.59 0.90 
Total lipid (mg/g TS) 306.48 254.70 188.08 296.13 195.15 359.73 321.98 
Total carbohydrate (mg/g TS) 64.87 75.05 97.21 102.88 77.90 172.38 216.31 
Total protein (mg/g TS) 307.76 303.87 238.94 379.01 264.90 312.35 320.43 
TP (mg/g TS) 14.37 4.28 10.53 10.12 8.77 7.16 7.67 
TN (mg/g TS) 18.05 15.88 39.31 22.43 33.46 35.23 66.31 
NH3-N (mg/g TS) 1.04 0.87 1.41 1.98 1.29 0.77 0.77 
VFA/ethanol (mg/g TS) 
 Ethanol 8.38 5.84 12.27 16.38 11.95 12.61 8.35 
 Acetate 22.82 34.55 31.45 16.17 30.88 38.56 84.79 
 Propionate 0.30 0.42 0.34 0.52 0.31 0.48 0.70 
 iso-Butyrate 1.13 3.64 0.88 1.05 1.31 2.94 8.52 
 n-Butyrate 0.51 0.45 0.67 0.64 0.58 0.69 0.51 
 iso-Valerate 0.47 0.88 0.40 0.18 0.62 1.14 3.46 
 n-Valerate 0.23 0.41 0.29 0.23 0.29 0.38 1.09 
Metal ion content (μg/g TS) 
 Ca 1,830 1,656 3,080 1,802 2,789 1,140 546.7 
 Cr 0.54 1.38 4.05 0.41 0.84 1.27 0.43 
 Cu 0.86 0.55 1.41 0.62 1.35 0.80 0.65 
 K 738.6 377.5 1,141 355.3 799.8 335.8 548.5 
 Mg 175.6 118.2 290.9 111.7 214.1 107.8 150. 5 
 Mn 2.66 2.03 6.08 1.89 2.87 1.27 1.94 
 Na 216.3 288.8 234.6 182.8 321.5 169.4 487.2 
 Zn 6.50 3.41 5.28 3.09 3.99 2.16 2.94 
Element content (%TS) 
 N 4.12 4.13 3.00 4.03 3.45 3.33 2.39 
 C 47.33 44.81 43.36 49.11 46.98 48.62 53.01 
 H 6.99 6.54 6.09 7.28 6.93 7.16 7.84 
 S 0.28 0.29 0.35 0.31 0.30 0.22 0.22 
 S1S2S3S4S5S6S7
pH 4.47 4.74 4.50 4.07 4.85 4.63 3.94 
TS (%) 22.16 21.72 16.28 24.30 17.78 21.29 13.95 
VS (%) 20.72 19.38 14.00 21.44 14.56 17.88 11.47 
TCOD (g/g TS) 0.88 1.07 1.11 1.21 1.17 1.04 1.19 
SCOD (g/g TS) 0.34 0.43 0.53 0.57 0.62 0.59 0.90 
Total lipid (mg/g TS) 306.48 254.70 188.08 296.13 195.15 359.73 321.98 
Total carbohydrate (mg/g TS) 64.87 75.05 97.21 102.88 77.90 172.38 216.31 
Total protein (mg/g TS) 307.76 303.87 238.94 379.01 264.90 312.35 320.43 
TP (mg/g TS) 14.37 4.28 10.53 10.12 8.77 7.16 7.67 
TN (mg/g TS) 18.05 15.88 39.31 22.43 33.46 35.23 66.31 
NH3-N (mg/g TS) 1.04 0.87 1.41 1.98 1.29 0.77 0.77 
VFA/ethanol (mg/g TS) 
 Ethanol 8.38 5.84 12.27 16.38 11.95 12.61 8.35 
 Acetate 22.82 34.55 31.45 16.17 30.88 38.56 84.79 
 Propionate 0.30 0.42 0.34 0.52 0.31 0.48 0.70 
 iso-Butyrate 1.13 3.64 0.88 1.05 1.31 2.94 8.52 
 n-Butyrate 0.51 0.45 0.67 0.64 0.58 0.69 0.51 
 iso-Valerate 0.47 0.88 0.40 0.18 0.62 1.14 3.46 
 n-Valerate 0.23 0.41 0.29 0.23 0.29 0.38 1.09 
Metal ion content (μg/g TS) 
 Ca 1,830 1,656 3,080 1,802 2,789 1,140 546.7 
 Cr 0.54 1.38 4.05 0.41 0.84 1.27 0.43 
 Cu 0.86 0.55 1.41 0.62 1.35 0.80 0.65 
 K 738.6 377.5 1,141 355.3 799.8 335.8 548.5 
 Mg 175.6 118.2 290.9 111.7 214.1 107.8 150. 5 
 Mn 2.66 2.03 6.08 1.89 2.87 1.27 1.94 
 Na 216.3 288.8 234.6 182.8 321.5 169.4 487.2 
 Zn 6.50 3.41 5.28 3.09 3.99 2.16 2.94 
Element content (%TS) 
 N 4.12 4.13 3.00 4.03 3.45 3.33 2.39 
 C 47.33 44.81 43.36 49.11 46.98 48.62 53.01 
 H 6.99 6.54 6.09 7.28 6.93 7.16 7.84 
 S 0.28 0.29 0.35 0.31 0.30 0.22 0.22 

Although the MYs differed considerably, the total COD (TCOD) of the different KW samples was similar. TCOD is composed of proteins, carbohydrates, lipids, VFA, and other organic compounds. In Table 3, the concentrations of total protein, total lipid, total carbohydrate, and VFA differ considerably in each KW sample. Lipids have the highest theoretical methane production, but a high lipid content may lead to long-chain fatty acid (LCFA) accumulation and could further inhibit methane generation. Early literature suggests that LCFA produced during lipid hydrolysis exerts a permanent toxic effect and even a bactericidal effect on methanogens (Angelidaki & Ahring 1992; Rinzema et al. 1994). Even at low concentrations, LCFA inhibits anaerobic microbial activities (Shin et al. 2003). The proportions of the sum of total protein, total lipid, total carbohydrate, and VFA in TCOD for S1 to S7 were 81.02%, 63.53%, 51.40%, 67.21%, 49.91%, 86.66%, and 81.19%, respectively. These high proportions are in good agreement with high MY since undetected TCOD, such as LCFAs and lignin, could be unfavorable for methane production. Lignocellulosic compounds usually demonstrate low biogas production and biodegradability because of the complex structures of lignin and other cell wall polysaccharides; thus, biodegradation of lignocellulosic waste is difficult (Li et al. 2013c). Furthermore, lignin is not biodegradable in an anaerobic environment (Triolo et al. 2011).

Nitrogen and phosphate are two of the most essential nutrients for efficient methane production because of their nutrient value and buffering capacity. Hassan et al. (2016) found that a C/N ratio of 20–35 is beneficial to the growth of methane-producing bacteria, and thus, to methane production. Recent studies have pointed out that digestion proceeds well at low C/N ratios of 15–20 (Zhang et al. 2014). An increase in phosphate concentration results in an increase in methane production capacity (Li et al. 2016). However, Table 4 and comparison of the MYs in the current study indicate that no significant relationship exists between C/N and C/P ratios and MYs. This result indicates that although methane production performance may be improved by adjusting the C/N and C/P ratios, these ratios are not the most important factor that influences MYs.

Table 4

C/N and C/P ratios of different KW samples

 S1S2S3S4S5S6S7
C/N ratio (element analysis) 11.5 10.9 14.5 12.2 13.6 14.6 22.2 
C/N ratio (TCOD/TN) 48.8 67.4 28.2 54.0 35.0 29.5 18.0 
C/P ratio (TCOD/TP) 61.2 250.0 105.4 119.6 133.4 145.3 155.2 
 S1S2S3S4S5S6S7
C/N ratio (element analysis) 11.5 10.9 14.5 12.2 13.6 14.6 22.2 
C/N ratio (TCOD/TN) 48.8 67.4 28.2 54.0 35.0 29.5 18.0 
C/P ratio (TCOD/TP) 61.2 250.0 105.4 119.6 133.4 145.3 155.2 

Anaerobic bacteria also require metal ions as an important nutrient substance, since these cations play an important role in enzyme synthesis and maintaining enzyme activities (Schattauer et al. 2011; Facchin et al. 2013). Table 3 shows that the concentrations of K and Na are the highest among the concentrations of all metal ions because of the Chinese diet structure. The concentrations of the other ions are comparatively low. With a low concentration of sodium salt, methane production was progressively promoted possibly because of the function of sodium salt in the oxidation of NADH and formation of ATP (Dimroth & Thomer 1989). A high concentration of salt dehydrates cells, causes cell death because of the action of osmosis, and further inhibits methane production (Mendez et al. 1995; Yerkes et al. 1997). Chen et al. (2008) reported that the optimum concentration of sodium for mesophilic methanogens is 350 mg/L. At a concentration of less than 400 mg/L, potassium could enhance the performance of both thermophilic and mesophilic AD. In this study, the concentrations of K in all KW samples were less than 400 mg/L, and the concentrations of Na were less than 350 mg/L. Evidently, the concentrations of metal ions do not have negative effects on the methane production process.

From these results and discussions, we conclude that the distribution of TCOD, the diversity of organic compounds in KW is the most dominant factor that affects the potential MYs of KW, while adjusting C/N and C/P ratios or the concentration of metal ions for MY improvement may have less effect.

Comparison of TMYs

Two typical methods to calculate TMY were applied to determine the difference in their ability to estimate the methane production potential and biodegradability of KW.

TMY was obtained with an elemental analysis method based on the elemental compositions of organic substrates (expressed as TMYele) and using Buswell–Mueller Equations (3) and (4) or based on the organic composition method calculated by Equation (5) (expressed as TMYorg) (Kaparaju et al. 2009), with VFA (as C2H4O2), lipids (as C57H104O6), protein (as C5H7NO2), and carbohydrates (as C6H10O5) as % of VS (Angelidaki et al. 2004).

Biodegradability was also calculated. BDele was calculated with TMYele values (Equation (6)), and BDorg was calculated with TMYorg values (Equation (7)). The calculation results are summarized in Table 5.
formula
3
formula
4
formula
5
formula
6
formula
7
Table 5

Methane production potential and biodegradability of different KW samples

 S1S2S3S4S5S6S7
Mesophilic MY (mL/g VSadded491.1 436.1 353.2 482.4 439.7 538.4 565.2 
Thermophilic MY (mL/g VSadded497.1 430.8 373.9 512.7 444.5 551.2 574.7 
TMYele (mL/g VSadded620.1 607.2 628.0 677.3 706.6 719.4 828.1 
TMYorg (mL/g VSadded570.3 481.4 395.5 576.6 459.3 645.7 678.0 
BDele (%) 
 Mesophilic 0.79 0.72 0.56 0.71 0.62 0.75 0.68 
 Thermophilic 0.80 0.71 0.60 0.76 0.63 0.77 0.69 
BDorg (%) 
 Mesophilic 0.86 0.91 0.89 0.84 0.96 0.83 0.83 
 Thermophilic 0.87 0.89 0.95 0.89 0.97 0.85 0.85 
 S1S2S3S4S5S6S7
Mesophilic MY (mL/g VSadded491.1 436.1 353.2 482.4 439.7 538.4 565.2 
Thermophilic MY (mL/g VSadded497.1 430.8 373.9 512.7 444.5 551.2 574.7 
TMYele (mL/g VSadded620.1 607.2 628.0 677.3 706.6 719.4 828.1 
TMYorg (mL/g VSadded570.3 481.4 395.5 576.6 459.3 645.7 678.0 
BDele (%) 
 Mesophilic 0.79 0.72 0.56 0.71 0.62 0.75 0.68 
 Thermophilic 0.80 0.71 0.60 0.76 0.63 0.77 0.69 
BDorg (%) 
 Mesophilic 0.86 0.91 0.89 0.84 0.96 0.83 0.83 
 Thermophilic 0.87 0.89 0.95 0.89 0.97 0.85 0.85 

As shown in Table 5, both methods were utilized to calculate the TMY of the KW samples. The TMYorg values calculated with the organic composition method were more comparable with the experimental MYs. Thus, the BDorg values were high and within the range of 83%–95%, indicating comparatively good estimation results. These high BDorg values show that the cumulative methane production had reached the potential production. Meanwhile, the elemental analysis method had a higher TMY than the organic composition method. The reason could be the values of carbon, nitrogen or other elements from elemental analysis contained not only easily degradable contents but also contents that were difficult to degrade. The TMYele values were much higher than the experimental MYs, and the calculated BDele values were low (ranging from 56%–80%). Although a previous study concluded that both methods can be reasonably used to calculate the TMY of substrates (Li et al. 2013b), in the current study, the organic composition method could simulate methane production more precisely than the elemental analysis method. The reason could be attributed to the fact that the difficult to degrade contents (such as lignin), as a part of the VS, were not available for biodegradation with the element analysis.

Linear regression was utilized to compare the TMYs with the experimental MYs. To further explore the theoretical estimations obtained from the two different methods, the results are illustrated in Figures 3 and 4. Elemental analysis method was proven to be an unsuitable method for MY estimation because no notable linear correlation was observed between the TMYele values and the MYs. On the contrary, significant linear correlations were found between TMYorg and MYs under mesophilic and thermophilic conditions, with R2 higher than 0.95. Thus, the MYs could be simulated well based on the organic composition of KW without a BMP test or other experiments.
Figure 3

Linear relationship between mesophilic and TMYs.

Figure 3

Linear relationship between mesophilic and TMYs.

Figure 4

Linear relationship between thermophilic and TMYs.

Figure 4

Linear relationship between thermophilic and TMYs.

Degradation performance

The VS/TS ratios of the influent and effluent under different temperatures are summarized in Table 6. The VS/TS ratios of the influent are all higher than 80%, indicating that most components of KW are volatile and may be converted into methane. However, no clear correlation was observed between the VS/TS ratio and MY. S1 has the highest VS/TS ratio of 93.5%, but the MYs are not more than 500 mL/g VSadded. S7 has a comparatively low VS/TS ratio of 82.2%, and its MYs are more than 565 mL/g VSadded.

Table 6

Degradation performance of each reactor

 Influent VS/TS(%)Effluent of mesophilic ADEffluent of thermophilic AD
VS/TS (%)VS/TS (%)
S1 93.50 51.51 ± 0.023 50.20 ± 0.019 
S2 89.23 51.61 ± 0.007 51.14 ± 0.025 
S3 86.00 50.94 ± 0.027 50.72 ± 0.009 
S4 88.23 53.99 ± 0.009 50.48 ± 0 
S5 81.89 51.96 ± 0.047 50.01 ± 0.010 
S6 83.98 50.31 ± 0.015 49.40 ± 0.021 
S7 82.22 51.51 ± 0.023 49.55 ± 0.001 
 Influent VS/TS(%)Effluent of mesophilic ADEffluent of thermophilic AD
VS/TS (%)VS/TS (%)
S1 93.50 51.51 ± 0.023 50.20 ± 0.019 
S2 89.23 51.61 ± 0.007 51.14 ± 0.025 
S3 86.00 50.94 ± 0.027 50.72 ± 0.009 
S4 88.23 53.99 ± 0.009 50.48 ± 0 
S5 81.89 51.96 ± 0.047 50.01 ± 0.010 
S6 83.98 50.31 ± 0.015 49.40 ± 0.021 
S7 82.22 51.51 ± 0.023 49.55 ± 0.001 

After the BMP experiments, the organic compound was maximally converted into methane, and the VS/TS ratios all decreased to approximately 50%. With the same KW source, the VS/TS ratio of the effluent after thermophilic AD was slightly lower than that of the effluent after mesophilic AD. This result can be attributed to the high MY and high degradation efficiency under thermophilic conditions.

The methane production potential of KW obtained from different sources was compared through mesophilic and thermophilic AD. With the same KW sample, the MYs obtained under different temperatures were similar, whereas the MYs of different samples differed significantly. The highest MY obtained in S7 was 54%–60% higher than the lowest MY in S3. The modified Gompertz model was adopted to simulate the methane production process. Although the lag phase time under thermophilic conditions was about 3–5 d longer than under mesophilic conditions, the maximum production rate of methane under thermophilic conditions was 2%–86% higher than that under mesophilic conditions. The characteristics of different KW samples were studied. The results indicated that in the distribution of TCOD, the diversity of organic compounds in KW is the most dominant factor that affects the potential MY of KW. The influence of C/N and C/P ratios or the concentration of metal ions is insignificant. Two typical methods to calculate TMY were compared. The elemental analysis method was proven to be unsuitable for MY estimation, while the organic composition method can simulate methane production more precisely than the elemental analysis method. Significant linear correlations were found between TMYorg and MYs under mesophilic and thermophilic conditions, with R2 higher than 0.95. The organic composition method can be used to quickly predict the methane production potential of KW and could provide a reference for defining a standard protocol to evaluate the methane production potential.

This study was funded by National Science Foundation of China (51508015), Science and Technology Commission of Beijing Municipality (Z151100001115006), and Fundamental Research Funds for the Central Universities (buctrc201505, ZY1510).

Angelidaki
I.
Ahring
B. K.
1992
Effects of free long-chain fatty acids on thermophilic anaerobic digestion
.
Bioresource Technology
37
(
6
),
808
812
.
Angelidaki
I.
Torang
L.
Waul
C. M.
Schmidt
J. E.
2004
Anaerobic bioprocessing of sewage sludge, focusing on degradation of linear alkylbenzene sulfonates (LAS)
.
Water Science and Technology
49
(
10
),
115
122
.
APHA
1998
Standard Methods for the Examination of Water and Wastewater
. 20th edn.
American Public Health Association
,
Washington, DC
,
USA.
Chen
Y.
Cheng
J. J.
Creamer
K. S.
2008
Inhibition of anaerobic digestion process: a review
.
Bioresource Technology
99
(
10
),
4044
4064
.
Dubois
M.
Gilles
K. A.
Hamilton
J. K.
Rebers
P.
Smith
F.
1956
Colorimetric method for determination of sugars and related substances
.
Analytical Chemistry
28
(
3
),
350
356
.
Facchin
V.
Cavinato
C.
Fatone
F.
Pavan
P.
Cecchi
F.
Bolzonella
D.
2013
Effect of trace element supplementation on the mesophilic anaerobic digestion of foodwaste in batch trials: the influence of inoculum origin
.
Biochemical Engineering Journal
70
(
2
),
71
77
.
Gu
Y.
Chen
X. H.
Liu
Z. G.
Zhou
X. F.
Zhang
Y. L.
2014
Effect of inoculum sources on the anaerobic digestion of rice straw
.
Bioresource Technology
158
(
4
),
149
155
.
Hassan
M.
Ding
W. M.
Bi
J. H.
Mehryar
E.
Talha
Z. A. A.
Huang
H. Y.
2016
Methane enhancement through oxidative cleavage and alkali solubilization pre-treatments for corn stover with anaerobic activated sludge
.
Bioresource Technology
200
(
3
),
405
412
.
Jiang
X.
Hayashi
J.
Sun
Z. Y.
Yang
L.
Tang
Y. Q.
Oshibe
H.
Osaka
N.
Kida
K.
2013
Improving biogas production from protein-rich distillery wastewater by decreasing ammonia inhibition
.
Process Biochemistry
48
(
11
),
1778
1784
.
Kaparaju
P.
Serrano
M.
Thomsen
A. B.
Kongjan
P.
Angelidaki
I.
2009
Bioethanol, biohydrogen and biogas production from wheat straw in a biorefinery concept
.
Bioresource Technology
100
(
9
),
2562
2568
.
Lay
J. J.
Fan
K. S.
Chang
J.
Ku
C. H.
2003
Influence of chemical nature of organic wastes on their conversion to hydrogen by heat-shock digested sludge
.
International Journal of Hydrogen Energy
28
(
12
),
1361
1367
.
Li
Y. Q.
Feng
L.
Zhang
R. H.
He
Y. F.
Liu
X. Y.
Xiao
X.
Ma
X. X.
Chen
C.
Liu
G. Q.
2013a
Influence of inoculum source and pre-incubation on bio-methane potential of chicken manure and corn stover
.
Applied Biochemistry and Biotechnology
171
(
1
),
117
127
.
Li
Y. Q.
Zhang
R. H.
Liu
G. Q.
Chen
C.
He
Y. F.
Liu
X. Y.
2013b
Comparison of methane production potential, biodegradability, and kinetics of different organic substrates
.
Bioresource Technology
149
(
12
),
565
569
.
Li
Y. Q.
Zhang
R. H.
Liu
X. Y.
Chen
C.
Xiao
X.
Feng
L.
He
Y. F.
Liu
G. Q.
2013c
Evaluating methane production from anaerobic mono- and co-digestion of kitchen waste, corn stover, and chicken manure
.
Energy & Fuels
27
(
4
),
2085
2091
.
Liao
X. F.
Zhu
S. Y.
Zhong
D. L.
Zhu
J. P.
Liao
L.
2014
Anaerobic co-digestion of food waste and landfill leachate in single-phase batch reactors
.
Waste Management
34
(
11
),
2278
2284
.
Lowry
O.
Rosebrough
J.
Farr
A.
Randall
R.
1951
Protein measurement with the folin reagent
.
Journal of Biological Chemistry
193
(
1
),
265
275
.
Rincon
B.
Heaven
S.
Banks
C. J.
Zhang
Y.
2012
Anaerobic digestion of whole-crop winter wheat silage for renewable energy production
.
Energy & Fuels
26
(
4
),
2357
2364
.
Rinzema
A.
Boone
M.
Van Knippenberg
K.
Lettinga
G.
1994
Bactericidal effect of long chain fatty acids in anaerobic digestion
.
Water Environment Research
66
(
1
),
40
49
.
Schattauer
A.
Abdoun
E.
Weiland
P.
Plochl
M.
Heiermann
M.
2011
Abundance of trace elements in demonstration biogas plants
.
Biosystems Engineering
108
(
1
),
57
65
.
Shin
H.
Kim
S. H.
Lee
C. Y.
Nam
S. Y.
2003
Inhibitory effects of long chain fatty acids on VFA degradation and β-oxidation
.
Water Science and Technology
47
(
10
),
139
146
.
Wang
M.
Sun
X. L.
Li
P. F.
Yin
L. L.
Liu
D.
Zhang
Y. W.
Li
W. Z.
Zheng
G. X.
2014
A novel alternate feeding mode for semi-continuous anaerobic co-digestion of food waste with chicken manure
.
Bioresource Technology
164
(
6
),
309
314
.
Ye
J. Q.
Li
D.
Sun
Y. M.
Wang
G. H.
Yuan
Z. H.
Zhen
F.
Wang
Y.
2013
Improved biogas production from rice straw by co-digestion with kitchen waste and pig manure
.
Waste Management
33
(
12
),
2653
2658
.
Yerkes
D.
Boonyakitsombut
S.
Speece
R.
1997
Antagonism of sodium toxicity by the compatible solute betaine in anaerobic methanogenic systems
.
. Water Science and Technology
36
(
6–7
),
15
24
.
Zhang
R. H.
El-Mashad
H. M.
Hartman
K.
Wang
F. Y.
Liu
G. Q.
Choate
C.
Gamble
P.
2007
Characterization of food waste as feedstock for anaerobic digestion
.
Bioresource Technology
98
(
4
),
929
935
.
Zhang
C. S.
Su
H. J.
Baeyens
J.
Tan
T. W.
2014
Reviewing the anaerobic digestion of food waste for biogas production
.
Renewable & Sustainable Energy Reviews
38
(
5
),
383
392
.
Zhou
Y. L.
Zhang
Z. Y.
Nakamoto
T.
Li
Y.
Yang
Y. N.
Utsumi
M.
Sugiura
N.
2011
Influence of substrate-to-inoculum ratio on the batch anaerobic digestion of bean curd refuse-okara under mesophilic conditions
.
Biomass & Bioenergy
35
(
7
),
3251
3256
.
Zhou
Q.
Shen
F.
Yuan
H. R.
Zou
D. X.
Liu
Y. P.
Zhu
B. N.
Jaffu
M.
Chufo
A.
Li
X. J.
2014
Minimizing asynchronism to improve the performances of anaerobic co-digestion of food waste and corn stover
.
Bioresource Technology
166
(
166
),
31
36
.