Ecotoxicology and Environmental Safety 121 (2015) 100–104

Contents lists available at ScienceDirect

Ecotoxicology and Environmental Safety

journal homepage: www.elsevier.com/locate/ecoenv

Kinetic study on the effect of temperature on biogas production using

a lab scale batch reactor
B. Deepanraj a, V. Sivasubramanian b,n, S. Jayaraj a
a
Department of Mechanical Engineering, National Institute of Technology, Calicut, Kerala, India
b
Department of Chemical Engineering, National Institute of Technology, Calicut, Kerala, India

art ic l e i nf o a b s t r a c t

Article history: In the present study, biogas production from food waste through anaerobic digestion was carried out in a
Received 27 October 2014 2 l laboratory-scale batch reactor operating at different temperatures with a hydraulic retention time of
Received in revised form 30 days. The reactors were operated with a solid concentration of 7.5% of total solids and pH 7. The food
24 April 2015
wastes used in this experiment were subjected to characterization studies before and after digestion.
Accepted 28 April 2015
Modified Gompertz model and Logistic model were used for kinetic study of biogas production. The
Available online 8 May 2015
kinetic parameters, biogas yield potential of the substrate (B), the maximum biogas production rate (Rb)
Keywords: and the duration of lag phase (λ), coefficient of determination (R2) and root mean square error (RMSE)
Anaerobic digestion were estimated in each case. The effect of temperature on biogas production was evaluated experi-
Biogas
mentally and compared with the results of kinetic study. The results demonstrated that the reactor with
Food waste
operating temperature of 50 °C achieved maximum cumulative biogas production of 7556 ml with better
Temperature
Modified Gompertz model biodegradation efficiency.
Logistic model & 2015 Elsevier Inc. All rights reserved.

1. Introduction bacteria breaks down polymers to monomers; acidogen-

Biomass has been considered as a major source of renewable
energy to replace the fast declining fossil fuel resources (Demirbas,
2007; Ozcimen and Karaosmanoglu, 2004; Jefferson, 2006). Bio-
mass appears to be an attractive feedstock for three main reasons.
First, it is a renewable resource that could be sustainably devel-
oped in the future. Second, reportedly it has positive environ-
mental properties resulting in no net release of carbon dioxide
(CO2) and very low sulfur content. Third, it also has significant
economic potential provided that, fossil fuel prices continuously
increase in the future (Cadenas and Cabezudo, 1998). Production of
biofuels from biomass can slow down the climate change by
substantially reducing the greenhouse gas emissions (Meena et al.,
2011). Among various renewable technologies, anaerobic digestion
is a commercially proven and widely employed technology for
treating biomass, especially like agricultural and forest residues,
municipal solid waste, etc. (Meena et al., 2011).
Anaerobic digestion is a natural process by which complex or-
ganic materials are broken down into simpler compounds in the
absence of air by several micro-organism communities (Samir
et al., 2010; Cioabla et al., 2012). Anaerobic digestion comprises of
a series of four biochemical reaction steps: hydrolysis – hydrolytic
n
Corresponding author. Fax: þ91 4952287250 .
E-mail address: babudeepan@gmail.com (B. Deepanraj).
esis – acidogenic bacteria converts monomers to short chained
carboxylic acids, carbon dioxide, hydrogen and alcohol; acet-
ogenesis – the products of the previous steps are converted to
acetic acid; methanogenesis – methane is produced from the
acetic acid (Weiland, 2010; Kumar et al., 2010; Decker and Menrad,
2007; Barik et al., 2013).
The major advantage of the anaerobic digestion process is the
production of biogas, a renewable energy source, which can be
used as fuel for automobiles, for direct heating and for power
generation (Appels et al., 2011). The production of biogas, de-
pending on the biomass feedstock used, helps in the reduction of
fossil fuel usage and enables the lowering of carbon dioxide levels.
Apart from biogas yield, anaerobic digestion liberates solid and
liquid by-products which can be used as fertilizer or soil
amendment.
The biogas produced during anaerobic digestion process is a
blend of methane (CH4: 55–65% by volume), carbon dioxide (CO2:
30–40% by volume), and traces of hydrogen sulphide (H2S), hy-
drogen (H2) (Deepanraj et al., 2014). Due to the intricacy of the
bioconversion processes, various factors like solid concentration,
pH, temperature, mixing/agitation, C/N ratio, etc. are affecting the
performance of an anaerobic digestion process. Among those fac-
tors, temperature is one of the important factor and lot of re-
searchers investigated the effects of temperature on biogas pro-
duction using various feedstock and reported their findings (El-
Mashad et al., 2004; Ji-Shi et al., 2006; Vanegas and Bartlett, 2013;

http://dx.doi.org/10.1016/j.ecoenv.2015.04.051
0147-6513/& 2015 Elsevier Inc. All rights reserved.
 B. Deepanraj et al. / Ecotoxicology and Environmental Safety 121 (2015) 100–104 101

Ghatak and Mahanta, 2014). The present study aims to investigate Table 1
the effect of temperature on biogas production from food waste Characteristics of substrate and digestate.
experimentally in an anaerobic batch digester. The experimental
Characteristics Substrate Digestate
kinetics was also analyzed for the goodness of fit with modified
Gompertz model. 30 °C 40 °C 50 °C 60 °C

TS (g/l) 75 37.92 36.28 34.23 35.17

VS (g/l) 71.34 35.02 34.00 31.02 32.74
2. Materials and methods FS (g/l) 3.66 2.90 2.28 1.4 0.53
TDS (g/l) 0.732 1.78 1.90 2.12 2.08
2.1. Feedstock COD (g/l) 69.92 42.70 40.17 37.95 38.88

Food waste used in this experimental study was collected from

a hostel mess of National Institute of Technology Calicut, Kerala.
Food waste is a highly desirable substrate for anaerobic digestion
with regards to its higher biodegradability and biogas/methane
yield. This contains a substantially large amount of organic matter,
which can be digested anaerobically to produce biogas. Also, the
nutrient content analysis showed that the food waste contained
well balanced nutrients for anaerobic microorganisms ([Zhang
et al., 2007,Zhang et al., 2012]). The elemental composition of the
food waste was determined using elemental analyzer. The carbon,
hydrogen, oxygen, nitrogen and sulfur compositions present in the
feedstock are 49.96%, 10.35%, 1.13%, 38.28% and 0.28% respectively.
The food wastes obtained were shredded, mixed and stored at 5 °C
until it is introduced into the anaerobic digester. The solid con-
centration and pH values were already optimized in the previous
experiments (Deepanraj et al., 2014; Jayaraj et al., 2014). Water
was added to obtain the desired total solid concentration (7.5% of Fig. 1. Cumulative biogas production.

total solids) and 1 N sodium bicarbonate solution was used to

maintain the pH value as 7. The characteristics of the substrate
used were determined before and after digestion. For all the ex-
periments, cow dung was used as inoculum (10% inoculum to feed
ratio).
2.2. Experimental setup
Laboratory-scale anaerobic batch digesters made of glass with a
total volume of 2 l and working volume 1.6 l were used in all the
experiments. Biogas production from the digesters was measured
daily by the water displacement method. The reaction mixtures
were stirred twice a day. Each reactors were maintained with
different temperatures (30, 40, 50 and 60 °C) using water bath.
modified Gompertz model and Logistic model. Using these mod-
els, the biogas production potential of the substrate, maximum
biogas production rate and the lag phase of the reaction can be
determined with available experimental results. The kinetic data
obtained from all digesters were checked for the fitness of mod-
ified Gompertz model and Logistic model. The modified Gompertz
model and Logistic model describes cumulative biogas production
from batch digesters, assuming that biogas production is a func-
tion of bacterial growth (Nopharatana et al., 2006; Pommier et al.,
2007). The equations are given by
⎧ ⎡ Rb e ⎤⎫
C = B exp ⎨ − exp ⎢ (λ − t ) + 1⎥ ⎬
⎩ ⎣ B ⎦⎭ (1)

2.3. Analytical methods B

C=
1 + exp ⎡⎣4R b + 2⎤⎦
λ−t
B (2)
Characterization of feedstock is one of the most significant
steps in the biogas production process. Determining the general where ‘C' is cumulative biogas production at digestion time
composition of the substrate (input feed) is essential for calculat- ‘t’ days; ‘B’ is biogas yield potential of the substrate; ‘Rb’ is max-
ing the quantity and composition of the biogas generated. The imum biogas production rate; ‘e’ is exp (1) ¼2.718; ‘λ’ is the
total solids, volatile solids, fixed solids and chemical oxygen de- duration of lag phase.
mand of the substrate and digestate were determined as per the The kinetic parameters of each of the reactor were estimated
standard method (APHA Standard methods for examination of using nonlinear least-square regression analysis with the help of
water and waste water, 1989). Total dissolved solids were de-
termined using TDS meter (Model-161, Deep Vision Instruments, Table 2
Results of experimental study.
India) and pH of the substrate and digestate was determined using
pH meter (pH-201, Lutron Electronic Enterprise, Taiwan). The Parameter 30 °C 40 °C 50 °C 60 °C
elemental composition was determined using elemental analyzer
(Elementar Vario EL III, ELEMENTAR Analysensysteme, Germany). TS removal efficiency (%) 49.44 51.62 54.36 53.10
TS removed (g/l) 37.08 38.72 40.77 39.83
The methane and carbon dioxide composition in the biogas were
VS removal efficiency (%) 50.91 52.34 56.51 54.16
measured using infrared gas analyzers (PIR-89, Technovation VS removed (g/l) 36.32 37.34 40.32 38.70
Analytical Instruments, India). COD removal efficiency (%) 38.93 42.54 45.72 44.39
COD removed (g/l) 27.22 27.22 29.75 31.27
2.4. Kinetic study Cumulative biogas produced – experimental 5673 6449 7556 7081
(ml)
CH4 (%) 60.8 60.5 61.2 59.1
Many dynamic models are available to give a detailed de- CO2 (%) 36.3 38.0 37.1 38.5
scription of bioconversion mechanism. In this study, we used
102 B. Deepanraj et al. / Ecotoxicology and Environmental Safety 121 (2015) 100–104

Fig. 2. (a) Cumulative biogas production – experimental and modified Gompertz model. (b) Comparison of experimental and predicted (modified Gompertz model) cu-
mulative biogas production.

Table 3
Results of kinetic study – modified Gompertz model and Logistic model.

Parameter Modified Gompertz Model Logistic Model

30 °C 40 °C 50 °C 60 °C 30 °C 40 °C 50 °C 60 °C

Cumulative biogas produced – experimental (ml) 5673 6449 7556 7081 5673 6449 7556 7081
Cumulative biogas produced – predicted (ml) 5588.5 6417.2 7517.4 7003 5476.9 6289 7360.6 6859.9
Max biogas potential from Gompertz model (ml) 5628.6 6463.3 7583.2 7062.3 5482.7 6295.2 7369.9 6868.7
Rb (ml/day) 432.4 504.6 581.8 537 430.6 508.0 587.4 538
Lag phase (λ) 1.5 1.9 2.4 2.0 1.8 2.4 2.8 2.4
R2 0.9988 0.9996 0.9998 0.9994 0.9922 0.9962 0.9961 0.9941
RMSE 2.80 1.91 1.71 3.17 11.55 9.74 13.24 13.92

Fig. 3. (a) Cumulative biogas production – experimental & Logistic model. (b) Comparison of experimental and predicted (Logistic model) cumulative biogas production.

s
MATLAB (R2012a). The statistical indicators R2 (coefficient of determina- The cumulative biogas production with respect to retention time
tion) and RMSE were calculated for this analysis (Kafle et al., 2013). for all the reactors is shown in Fig. 1. The cumulative biogas yield
obtained by the reactor with 30, 40, 50 and 60 °C are 5673, 6449,
⎛ m ⎛dj ⎞2⎞
1/2
7556 and 7081 ml respectively. According to the results, the cu-
⎜1
RMSE = ⎜
⎜m
∑ ⎜⎜ ⎟⎟ ⎟⎟⎟ mulative gas production was high at 50 °C, followed by 60, 40 and
Y
j=1 ⎝ j ⎠ ⎠
⎝ (3) 30 °C. The total cumulative gas production increases with increase
in temperature upto 50 °C and then decreases. The cumulative
where ‘RMSE’ is Root Mean Square Error; ‘m’ is number of data
biogas produced by reactor with temperature 50 °C is 24.92%,
pairs; ‘d’ is the difference between experimental and predicted
14.65% and 6.28% higher than the yield of reactors with 30, 40 and
methane yield; and ‘Y’ is the measured biogas yield.
60 °C. Similar trend was observed by Kim et al., 2006 who studied
the effect of temperature on biogas production from anaerobic
digestion of food waste. The amount of biogas produced from the
3. Results and discussions reactors at 50 °C and 60 °C is more than that of 30 °C and 40 °C.
This result suggests that the activity of the methanogenic bacteria
3.1. Experimental study used in this study depends on to operation temperature. There-
fore, a thermophilic temperature 50–60 °C is more effective for
The characteristics of substrate and digestate before and after biogas production than a mesophilic temperature. The specific
digestion process are given in Table 1. The temperature of the biogas yield of 138.94, 144.51, 161.09 and 152.08 ml/g of COD re-
reactors was maintained at 30, 40, 50 and 60 °C using water bath. moval were obtained at 30, 40, 50 °C and 60 °C respectively.
 B. Deepanraj et al. / Ecotoxicology and Environmental Safety 121 (2015) 100–104
Table 2 shows the results of experimental study carried out at
different operating temperatures. The variation of total solids (TS),
volatile solids (VS) and chemical oxygen demand (COD) degrada-
tion efficiencies are shown in Fig. 3. The TS degradation efficiency
of the reactors with operating temperature 30, 40, 50 and 60 °C
are 49.44%, 51.62%, 54.36% and 53.1% respectively. The reactor with
operating temperature of 50 °C produced better TS degradation
efficiency (40.77 g/l), which is 9.05, 5.04 and 2.31% higher than the
reactors of 30, 40 and 60 °C respectively. The VS degradation ef-
ficiency of the reactors with operating temperature 30, 40, 50 and
60 °C are 50.91%, 52.34%, 56.51% and 54.16% respectively. The COD
degradation efficiency of the reactors with operating temperature
30, 40, 50 and 60 °C are 38.93%, 42.54%, 45.72% and 44.39% re-
spectively. Similar to that of TS degradation efficiency, VS and COD
degradation efficiency of the reactor with operating temperature
of 50 °C is higher compared to other three reactors.
3.2. Kinetic study
3.2.1. Using modified Gompertz model
The modified Gompertz equation was used to fit cumulative
biogas production data obtained from anaerobic digestion process.
The estimated kinetic parameters based on modified Gompertz
equation are given in Table 3. This indicates that the modified
Gompertz equation can be used to estimate biogas yield potential,
maximum biogas production rate and duration of lag phase. To
evaluate the soundness of the model results from the modified
Gompertz model, the predicted cumulative biogas values were
plotted against the measured values, as shown in Fig. 2 a.
Biogas yield potential of the substrate (B) with operating
temperature 30, 40, 50 and 60 °C are found to be 5628.6, 6463.3,
7583.2 and 7062.3 ml respectively. This shows that the reactor
with operating temperature 50 °C have a maximum biogas yield
potential followed by 60, 40 and 30 °C. Calculated lag time was
found to be in between 1.5 and 2.4 days, because of the inoculum
added and the ready biodegradation component available in the
substrate. The coefficient of determination (R2) values for the re-
actors with operating temperature 30, 40, 50 and 60 °C are 0.9988,
0.9996, 0.9998 and 0.9994. This shows that the predicted values
were best fitted with experimental values. The root mean square
errors values are in between 1.71 and 3.17. Similar type of results
was reported by researchers who used modified Gompertz equa-
tion for their experiments to evaluate the kinetic study of cumu-
lative gas production. Budiyono and Syaichurrozi (2014) studied
the kinetics of anaerobic digestion process from vinasse with re-
spect to total solids and obtained R2 value in between 0.993 and
0.999 and lag phase in between 0.432 and 2.547 days. Cheong and
Hansen (2006) experimented the cattle manure sludge for anae-
robic hydrogen production and reported that modified Gompertz
equation was best fitted for the kinetic study. They obtained the R2
value in between 0.997 and 1.0 and RMSE in between 0.16 and
7.06.
Fig. 2b shows the comparison of experimental and predicted
cumulative biogas production for all the reactors. Using modified
Gompertz equation, the overall root mean square error (RMSE)
and R2 were obtained as 2.68% and 0.9994% respectively. The
overall percentage deviation for all the reactors with different
temperatures is 717.5%. This shows that modified Gompertz
equation was best fitted for kinetic study of biogas production and
to determine the lag phase of the reaction and maximum biogas
potential.
3.2.2. Using logistic model
Table 3 shows the estimated kinetic parameters values based
on modified Logistic model. To evaluate the soundness of the
model results from the Logistic model, the predicted cumulative
103
biogas production values were plotted against the measured va-
lues, as shown in Fig. 3 a. With operating temperature 30, 40, 50
and 60 °C, the biogas yield potential of the substrate (B) are found
to be 5482.7, 6295.2, 7369.9 and 6868.7 ml respectively. Calculated
lag time was found to be in between 1.8 and 2.8 days, because of
the inoculum added and the ready biodegradation component
available in the substrate. The coefficient of determination (R2)
values for the reactors with operating temperature 30, 40, 50 and
60 °C are 0.9922, 0.9962, 0.9961 and 0.9941 respectively. The
coefficient of determination (R2) values shows that the predicted
values were best fitted with experimental values. The root mean
square errors values are in between 9.74 and 13.92.
Fig. 3b shows the comparison of experimental and predicted
cumulative biogas production for all the reactors. Using Logistic
model, the overall root mean square error (RMSE) and R2 were
obtained as 5.31% and 0.9960% respectively. The overall percentage
deviation for all the reactors with different temperatures is 725%.
This shows that Logistic model was well fitted for kinetic study of
biogas production and to determine the lag phase of the reaction
and maximum biogas potential. But compared to the modified
Gompertz model, the mean square error (RMSE), coefficient of
determination value (R2) and the overall percentage deviation of
the Logistic model are higher. This shows that modified Gompertz
model suits well for the kinetic study of biogas production com-
pared to Logistic model.
4. Conclusion
The effect of temperature on biogas yield from food waste was
investigated experimentally in 2 l batch anaerobic reactors. Kinetic
model of biogas production was obtained using modified Gom-
pertz model and Logistic model. The experimental results were
compared with the predicted values. The best performance was
observed in digester operating at a temperature of 50 °C. However,
this result does not mean the higher temperature is more optimal,
due to the larger energy requirement at higher digesting tem-
peratures. Therefore, careful consideration of the net energy bal-
ance between the increased heating energy demands and im-
proved additional methane production at higher operating tem-
peratures must be simultaneously taken into account when de-
ciding the economical digesting temperature. The modified Gom-
pertz model and Logistic model were used to evaluate the per-
formance parameters of anaerobic digestion. By this study, the
biogas yield potential of the substrate, lag phase of the reaction
and the maximum biogas production rate were determined. The
statistical analysis showed that the modified Gompertz model was
better consisted with the experimental data than the Logistic
model. This was proved by comparing the R2 values and root mean
square errors of the two models.
References
APHA, 1989. Standard Methods for Examination of Water and Waste Water, 17th ed.
American Public Health Association.
Appels, L., Lauwers, J., Degreve, J., Helsen, L., Lievens, B., Willems, K., Impe, J.V.,
Dewil, R., 2011. Anaerobic digestion in global bio-energy production – potential
and research challenges. Renew. Sustain. Energy Rev. 15, 4295–4301.
Barik, D., Sah, S., Murugan, S., 2013. Biogas production and storage for fueling in-
ternal combustion engines. Int. J. Emerg. Technol. Adv. Eng. 3, 193–202.
Budiyono, B., Syaichurrozi, I., Sumardiono, S., 2014. Effect of total solid content to
biogas production rate from vinasse. IJE Trans. B: Appl. 27 (2), 177–184.
Cadenas, A., Cabezudo, S., 1998. Biofuels as sustainable technologies: perspectives
for less developed countries. Technol. Forecast. Soc. Change 58, 83–103.
Cheong., D.Y., Hansen, C.L., 2006. Bacterial stress enrichment enhances anaerobic
hydrogen production in cattle manure sludge. Appl. Microbiol. Biotechnol. 72,
635–643.
Cioabla, A.E., Ionel, I., Dumitrel, G.A., Popescu, F., 2012. Comparative study on factors
104 B. Deepanraj et al. / Ecotoxicology and Environmental Safety 121 (2015) 100–104
affecting anaerobic digestion of agricultural vegetal residues. Biotechnol. Bio-
fuels 5, 1–9.
Decker, T., Menrad, K., 2007. Regulation and innovation in biogas technology in
selected European countries. Int. J. Public Policy 2, 89.
Deepanraj, B., Sivasubramanian, V., Jayaraj, S., 2014. Biogas generation through
anaerobic digestion process – an overview. Res. J. Chem. Environ. 18, 80–93.
Deepanraj, B., Sivasubramanian, V., Jayaraj, S., 2014. Solid concentration influence
on biogas yield from food waste in an anaerobic batch digester. In: Interna-
tional Conference & Utility Exhibition on Green Energy for Sustainable Devel-
opment (ICUE 2014), Thailand, March 19–21.
Demirbas, A., 2007. Progress and recent trends in biofuels. Progr. Energy Combust.
Sci. 33, 1–18.
El-Mashad, H.M., Zeeman, G., Van Loon, W.K.P., Bot, G.P.A., Lettinga, G., 2004. Effect
of temperature and temperature fluctuation on thermophilic anaerobic diges-
tion of cattle manure. Bioresour. Technol. 95, 191–201.
Ghatak, M.D., Mahanta, P., 2014. Effect of temperature on anaerobic co-digestion of
cattle dung with lignocellulosic biomass. J. Adv. Eng. Res. 1, 1–7.
Jayaraj, S., Deepanraj, B., Sivasubramanian, V., 2014. Study on the effect of pH on
biogas production from food waste by anaerobic digestion. In: Proceedings of
the 9th International Green Energy Conference, Tianjin, China, May 25–26,
pp.799–805.
Jefferson, M., 2006. Sustainable energy development: performance and prospects.
Renew. Energy 31, 571–582.
Ji-Shi, Z., Ki-Wei, S., Chang, W.M., Lei, Z., 2006. Influence of temperature on per-
formance of anaerobic digestion of municipal solid waste. J. Environ. Sci. 18,
810–815.
Kafle, G.K., Kim, S.H., Sung, K.I., 2013. Ensiling of fish industry waste for biogas
production: a lab scale evaluation of biochemical methane potential (BMP) and
kinetics. Bioresour. Technol. 127, 326–336.
Kim, J.K., Oh, B.R., Chun, Y.N., Kim, S.W., 2006. Effects of temperature and hydraulic
retention time on anaerobic digestion of food waste. J. Biosci. Bioeng. 102,
328–332.
Kumar, S., Mukherjee, S., Devotta, S., 2010. Anaerobic digestion of vegetable market
waste in India. World Rev. Sci. Technol. Sustain. Dev. 7, 217.
Meena, K., Kumar, V., Vijay, v.K., 2011. Anaerobic technology harnessed fully by
using different techniques: Review. In: IEEE First Conference on Clean Energy
and Technology (CET 2011), Kuala Lumpur, Malaysia, 27–29 June.
Nopharatana, A., Pullammanappallil, P.C., Clarke, W.P., 2006. Kinetics and dynamic
modelling of batch anaerobic digestion of municipal solid waste in a stirred
reactor. Waste Manag. 27, 595–603.
Ozcimen, D., Karaosmanoglu, F., 2004. Production and characterization of bio-oil
and biochar from rapeseed cake. Renew. Energy 29, 779–787.
Pommier, S., Chenu, D., Quintard, M., Lefebvre, X., 2007. A logistic model for the
prediction of the influence of water on the solid waste methanization in
landfills. Biotechnol. Bioeng. 97, 473–47482.
Samir, K.K., Rao, Y.S., Tian, C.Z., Buddhi, P.L., Tyagi, R.D., Kao, C.M., 2010. Bioenergy
and Biofuel from Biowastes and Biomass. vol. 44. American Society of Civil
Engineers, Reston.
Vanegas, C., Bartlett, J., 2013. Anaerobic digestion of laminaria digitata: the effect of
temperature on biogas production and composition. Waste Biomass Valor 4,
509–515.
Weiland, P., 2010. Biogas production: current state and perspectives. Appl. Micro-
biol. Biotechnol. 85, 849–860.
Zhang, Lei, Ouyang, Wenxiang, Li, Aimin, 2012. Essential role of trace elements in
continuous anaerobicdigestion of food waste. Procedia Environ. Sci. 16,
102–111.
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.