Int. J. Environ. Sci. Technol.

DOI 10.1007/s13762-017-1453-5

ORIGINAL PAPER

Feasibility study of waste (d) potential: co-digestion of organic

wastes, synergistic effect and kinetics of biogas production
K. Paritosh1 • S. Mathur1 • N. Pareek2 • V. Vivekanand1

Received: 6 March 2017 / Revised: 12 July 2017 / Accepted: 15 July 2017

Ó Islamic Azad University (IAU) 2017

Abstract The present study investigated the synergistic Introduction

effect of co-digesting food and green waste from institute
campus for enhanced biogas production in different ratios Food waste (FW; both precooked while processing and
in batch tests (37 ± 1 °C, 90 rpm, 45 days). The results after meal leftovers) is a biodegradable waste generated
showed that blending improved the biogas production from various sources including food processing plants,
significantly, with highest biogas yield (660 ± 24 mL g-1 domestic and commercial kitchens, hostel mess, cafeterias
volatile solids) that was achieved at 75:25 of food and and restaurants. According to food and agricultural orga-
green waste ratio on volatile solids basis. The yield was nization (FAO 2012), around 1.3 billion tonnes of food
1.7- and 1.9-fold higher than the mono-digestion of food including fresh vegetables, fruits and meat, bakery and
and green waste (370 ± 34; 342 ± 36 mL g-1 volatile dairy products are lost as well as underutilized along the
solids), respectively. The increase in biogas production food supply chain (FAO 2011, 2012). Reduction of food
may be attributed to optimum carbon to nitrogen ratio waste should be given the highest priority yet so far it has
resulting in higher yield. The addition of TiO2 nanoparti- been observed to be less feasible technically and socially
cles showed virtually no effect on biogas production. (Vandermeersch et al. 2014). To bring down the food waste
Characterization was carried out to gain an insight of generation, awareness activities related to its potential and
feedstocks. Modified Gompertz and logistics models were impact on the environment may be carried out among the
applied for kinetic study of biogas production where populations. However, this is time captivating and never
modified Gompertz model showed goodness of fit ending, for the same one may reprocess and recycle the
(R2 = 0.9978) with the experimental results. food waste by using it as a valuable resource for different
applications viz. anaerobic digestion, compost formation,
Keywords Food waste  Bioenergy  Anaerobic digestion  landfilling and incineration (Paritosh et al. 2017). The
Green waste  Nanoparticles  Kinetic study carbon footprint of FW is estimated to be around 3.3 billion
tonnes of CO2 equivalent of greenhouse gases (GHG)
released into the atmosphere per year (FAO 2011). Simi-
larly, 1.4 billion hectares of land, i.e., 28% of the world’s
agricultural area, is used yearly to produce food that is lost
Editorial responsibility: Binbin Huang.
or wasted in food supply chain (FAO 2011). FW mainly
& V. Vivekanand consists of carbohydrate, protein, fat, organic acids and
[Link]@[Link] traces of inorganic compound. Cooked food usually has
1
roughly *20% total solids (TS).
Centre for Energy and Environment, Malaviya National
Indian cuisines do represent a rich assortment with high
Institute of Technology Jaipur, Jaipur, Rajasthan 302017,
India organic content, easy cooking with extensive choice of
2 ingredients and off-course with a very good taste. Being
Department of Microbiology, School of Life Sciences,
Central University of Rajasthan, organic, it represents high biodegradability and may
Bandarsindri, Kishangarh, Ajmer, Rajasthan 305801, India decompose easily releasing unpleasant odor along with the

123
 Int. J. Environ. Sci. Technol.

growth of pathogenic microbes that may affect the health feedstocks to provide multiple benefits in the form of
of the population (Agdag 2009). It has high moisture energy and soil conditioner.
content and organics that makes it less favorable for Anaerobic co-digestion has been reported to be benefi-
incineration and landfilling yet most of the time FW is cial for enhanced biogas production as compared to mono-
traditionally incinerated along with municipal solid waste digestion. Co-digestion of swine manure with Chinese
(MSW) or dumped in open area which may cause the cabbage waste silage (Kafle et al. 2014) and Kimchi fac-
release of methane and/or dioxins as well as obnoxious tory waste silage (Kafle et al. 2012) has appeared promis-
odor (Katami et al. 2004). The presence of high moisture ing for improved biogas production. Difference in organic
makes the incineration process energy intensive that is matter content, C/N imbalance and the presence of heavy
unfavorable by energy point of view (Pham et al. 2015). It metals may have complications in the form of instability
may further lead to potential loss of chemical value of the and disruption of the process (Callaghan et al. 2002).
food waste. FW with high moisture content, organics and The batch test is the elementary and effective procedure
nutrients may be utilized as a valuable feedstock for biogas to assess the effect of blending different organic wastes into
production (Wang et al. 2015) to decrease the environ- digester for AD (Koch and Drewes 2014). It has also been
mental impact. reported by Koch et al. (2016) that batch tests are suitable to
Complex organic compounds are liable to be converted assess changes in the kinetics of digestion and the methane
into simpler molecules when they undergo through nature- production. The use of small amount of additives in biore-
driven microbial process termed as anaerobic digestion actor could improve its performance significantly. The
(AD) and it may convert wastes into valuable energy applicability of any kind of additive is strongly dependent on
sources, nutrients and minimize the waste. A series of four the type of substrate undergoing AD in the bioreactor
biochemical reactions viz. hydrolysis—conversion of (Yadvika et al. 2001). Trace elements as micronutrients play
polymers into monomers, acidogenesis—conversion of vital role on the performance and stability of bioreactors,
monomers into organic acids, acetogenesis—conversion of which are operated with the crops having energy value,
previously obtained product into acetic acid and human as well as animal excreta, residues of crops such as
methanogenesis in which archaea converts this to biogas. sweet sorghum, maize, wheat straw and municipal solid
Green waste (GW) mainly consists of dry leaves, wastes having organic fraction or any other type of
grass clippings and twigs etcetera collected from parks biodegradable waste (Demirel and Scherer 2011). Limitation
and gardens and is a source of biomass which may be of the metals as cofactors required by the enzymes for
utilized for energy production along with the FW. It has chemical reactions may disturb the overall AD process as
been reported by Pick et al. (2012) that AD of GW as reported in the literature (Thauer et al. 2001). The positive
feedstock for renewable energy production is economi- impact of adding trace elements has also been noted in
cally less feasible. However, co-digesting FW and GW specific methanogenic activity assays with various sub-
generated from various sources for energy recovery may strates, including methanol (Co and Nil; Feng et al. 2010).
be fruitful in terms of odor reduction and waste man- Among four nanoparticles (TiO2, Al2O3, SiO2 and ZnO)
agement. It is imperative to develop sustainable alter- evaluated for enhancing biogas production from cow dung
native processes for FW and GW utilization in eco- slurry, only nano-ZnO showed inhibitory effect on methane
friendly manner. generation, and the effect of nano-ZnO was depending on
AD appears to be a most suitable process to treat FW dosage (Abdelsalam et al. 2016). Although co-digestion has
and GW. Conventional AD has always been focused on been reported with positive results, the existing knowledge
waste treatment and waste minimization and not to maxi- in the area is not adequate on a variety of feedstocks and still
mize energy output. However, the current global scenario additional efforts are needed to define the balancing features,
in the energy sector and climate change issues have com- kinetics of the process as well as a synergistic effect for FW
pelled the researchers to evolve the innovative AD tech- and GW. The present study has been executed to investigate
nologies for slowly replacing the fossil fuels with the effect of blending two different organic wastes, i.e., FW
renewable energy resources (here biogas). The best and GW on biogas production ability with varying substrate
ascendency of AD is the production of biogas that is a ratio in the anaerobic batch digester. The study carried out at
renewable source of energy which may be utilized for National Institute of Technology, Jaipur, India (26.9124°N,
direct heating, power generation and in automobile-related 75.7873°E) between January and April 2016. Also, the
applications. Yield and quality of biogas directly depend on effect of the addition of nanoparticles of titanium dioxide
the type, quality and amount of the feedstocks and also (TiO2) on co-digestion has been evaluated with varying
help to lower the emission of GHG into the atmosphere concentration. The experimental kinetics was also studied
indirectly (Appels et al. 2011). Deepanraj et al. (2014) have for the goodness of fit with modified Gompertz and logistics
also reported that biogas can be produced from common models.

123
Int. J. Environ. Sci. Technol.

Materials and methods VS amount was same in all bottle digesters although they
varied in their feedstocks. A total of nine groups were
Raw materials designed in which inoculum, cellulose, FW, GW and the
mixture of FW/GW in varying ratios viz. 85:15, 75:25,
FW and GW were used as feedstocks in the experiment and 50:50, 25:75 and 15:85 evaluated for biogas potential. Prior
were collected from different hostels situated on the cam- to incubation, the bottles were flushed with nitrogen to
pus of Malaviya National Institute of Technology, Jaipur have the anaerobic conditions; closed with rubber stoppers
(MNITJ). Preliminary survey conducted at the institute for and aluminum screw caps; transferred to the orbital shaker
4 continuous weeks showed that on an average 468 kg of (REMI CIS 24, India) for incubation (37 °C, 90 rpm,
precooked and leftover food waste was generated daily in 45 days; Vivekanand et al. 2013).
the different mess of hostels. FW mainly contained pre-
cooked and meal leftovers such as lentils, beans, rice, Nanoparticle addition
bread, noodles, fruits and vegetables obtained from the
various hostel mess. The GW consisted of dry leaves, grass A total of five groups of batch bottle digesters of FW/GW
clippings and dry flower trimmings. Feedstocks were (75:25) with different dosage of TiO2 (from 0.5 to 2.5 mg
shredded uniformly using an electrical grinder and stored at L-1) were prepared. Batch digesters were added with 0.5,
4 °C until further use. On-campus data were collected from 1, 1.5, 2 and 2.5 mg L-1 of TiO2 and incubated as above
January to April 2016 for the generated feedstocks. (37 ± 1 °C, 90 rpm, 45 days) to evaluate the impact of
nanoparticles on biogas production. One group of FW/GW
Biogas inoculum (75:25) was used as a control without any nanoparticle, and
inoculum alone was used as control (as above).
Active microbial inoculum utilized for biogas experiments
was collected from a local active biogas plant at Rajasthan Biogas measurements
Gosewa Sangh, Durgapura, Jaipur (26.8°N, 75.7°E), run-
ning large-scale continuous stirred-type bioreactor. The Biogas production was calculated by measuring pressure in
plant was operating at mesophilic temperature using cow the headspace of each batch bottle digester. The pressure
manure as feedstock with the presence of a large array of generated in bottles was measured using digital pressure
the highly active methanogenic community for AD pro- meter (testo 512, Germany). The biogas in the headspace
cess. Fresh inoculum initially had total solid (TS) content, was purged to reduce the pressure close to the atmospheric
7.58%, volatile solid (VS) as percent of TS, 65.85%, pH pressure. After releasing the biogas, the pressure in the
7.6 and conductivity of 3.25 mS. headspace was noted again as an initial condition for the
The inoculum was pre-incubated anaerobically in the next-day measurement. Gas composition was measured by
chamber (37 °C) similar to the typical operating tempera- gas chromatgraph (TRACE1300, ThermoFisher Scientific,
ture of biogas plant up to 10 days in order to reduce the India).
endogenous biogas production by undigested biomass. Gas calculations were performed as described by
Following the storage, the inoculum was diluted with water Vivekanand et al. (2013). Using the headspace volume
and divided into 400 mL aliquots into 610-mL bottles for of the bottles, the ideal gas law was applied for calcu-
batch experiments. Diluted inoculum had TS concentration lating biogas production during the experiments. All
of 1.2%. Batch bottle digesters were prepared and stored experiments were run in triplicate, and the average
accordingly until further use. results are given along with standard deviations. The
reported biogas yields are the values only after sub-
Evaluation of biogas potential of different tracting the endogenous biogas production from the
combinations of raw materials in batch mode inoculum alone (control).
ðP  Vhead  CÞ
AD of feedstocks alone and their blends along with cel- VBiogas ¼ ð1Þ
RT
lulose (Avicel, Sigma, USA) as standard reference material
was performed, in sealed batch bottle digesters. Inoculum where VBiogas = daily biogas volume (L), P = absolute
alone was used as control (without any substrate or cellu- pressure difference (mbar), Vhead = volume of the head
lose). Experiments were performed using substrates cor- space (L), C = molar volume (22.41 L mol-1), R = uni-
responding to a final concentration of 1.5 g VS L-l. Blends versal gas constant (83.14 L mbar K-1 mol-1), T = ab-
of FW/GW were prepared in different ratios where the final solute temperature (K)

123
 Int. J. Environ. Sci. Technol.

Analytical methods analyze biogas yields from batch digesters. The kinetic
data obtained from all the batch digesters have been
Characterizing the feedstock is one of the most important checked for the goodness of fit to the above-mentioned
aspects of biogas potential test. The TS, VS and ash content models. It has been observed by Nopharatana et al. (2006)
were determined for raw materials and inoculum as per the that biogas production is a function of bacterial growth and
standard method (APHA 1989). Ultimate analysis viz. these two models describe cumulative biogas production
nitrogen, carbon and hydrogen present in FW, GW and potential from batch digester assuming the same.
inoculum was performed using Elemental Analyzer The equations are:
(FLASH 2000; Thermo Scientific, USA) as reported   
Rb e ð k  t Þ
(Zirkler et al. 2014). ICP-MS (Thermo Scientific, Ger- CBP ¼ B exp  exp þ1 ð2Þ
B
many) was used for determination of elements as described
(Zirkler et al. 2014). Fourier transform infrared spec- B
CBP ¼  h i ð3Þ
troscopy (PerkinElmer, Germany) was performed to 4Rb ðktÞ
1 þ exp B þ2
examine the functional groups present in the substrates as
well as in inoculum. For this, KBr pellets (250 mg of dry where ‘CBP’ is cumulative biogas production at digestion
KBr and 2 mg of samples) were pressed under vacuum and time ‘t’ days; B is biogas yield potential of the substrate;
spectra were recorded in the range of 4000–400 cm-1 ‘Rb’ is maximum biogas production rate; ‘e’ is exp
(Fakharedine et al. 2006). (1) = 2.718; ‘k’ is duration of lag phase. The kinetic
Crude protein and reducing sugar were estimated as per parameters of each of the digesters were estimated using
Lowry et al. (1951) and Miller (1959), respectively. No nonlinear least square regression analysis.
additional material was mixed to assess the synergistic
impact of co-digestion. Measurement of pH and electrical
conductivity was taken for each sample prior to the start of Results and discussion
the experiment, which ranged from 7.43 to 7.57 and 3.20 to
3.25 mS, respectively. The pH and conductivity were in the Feedstock and inoculum characterization
range of 6.49–6.57 and 3.62–4.17 mS, respectively, at the
end of the experiment. The physical and chemical characteristics of FW, GW and
inoculum are shown in Tables 1 and 2. FW, GW and
Kinetic study inoculum had TS content of 21.63, 82.09 and 7.58%,
respectively (Table 1). The moisture content of FW was
There are many dynamic models available for the detailed higher as Indian cuisines are mostly served with curry,
elucidation of the mechanism associated with biological while the content of GW was largely dried, hence showed a
conversions. Modified Gompertz and logistic models low level of moisture.
appeared as a powerful tool for envisaging the anaerobic Ultimate analysis revealed that FW, GW and inoculum
digestion process performance by taking into account the had C/N ratio of 20.43, 40.22 and 20.19, respectively. The
behavior of anaerobic microorganisms and interactions of nitrogen content in FW was detected to be 2.14 that is
operating parameters, hence selected to evaluate the pro- highest among the three samples analyzed. This may be
cess significance in the present study. Deepanraj et al. attributed to the protein present in the FW in the form of
(2017) have also utilized first-order kinetic model, i.e., region and chickpeas, different kinds of lentils as well as
modified Gompertz and logistic models for comparing kidney beans etcetera. Also, the Indian food has complex
predicted and experimental biogas yields. By applying varieties along with different kind of ingredients. It was
these mathematical models, the biogas production potential observed that C/N of GW was more than FW, which makes
of feedstocks, maximum biogas production rate and the lag GW or FW less likable to be used as feedstock alone for
phase of the reaction may be resolved using available standard AD process where C/N of 25–30 is optimum. The
experimental data. Equations (2) and (3) were carefully blending of these feedstocks may prove to be a suit-
used to fit experimental data of this study in order to able substrate for a stable AD process. VS as percent of TS
determine some important kinetic parameters necessary for content in the FW, GW and inoculum was appeared to be
digester design and optimal operation of large-scale 91.85, 85.55 and 65.85%.
anaerobic plants that would treat FW and GW substrates.
The solutions to Eqs. (2) and (3) were obtained using the Biogas production and blending experiment
optimization tools of the Solver function in Microsoft
Excel 2016 version. One-way analysis of variance Experimental results showed that cumulative biogas pro-
(ANOVA) at 5% significant level was used to statistically duction from FW and GW when digested alone was 370 and

123
Int. J. Environ. Sci. Technol.

Table 1 Physical Samples Moisture content Total solid Volatile solid Fixed solid Protein Sugar
characteristics of MNITJ mess
food waste, campus green waste Food waste 77.8 ± 4 21.6 ± 2 91.8 ± 3 8.8 ± 2 13.8 ± 5 57.4 ± 7
and inoculum (%)
Green waste 17.9 ± 2 82.0 ± 5 85.5 ± 4 14.4 ± 3 NA NA
Inoculum 92.4 ± 4 7.5 ± 2 65.8 ± 2 34.1 ± 2 NA NA
NA no analysis

Table 2 Ultimate analysis of Substrate Total C (%) Total H (%) Total N (%) Total Oa (%) C/N ratio
FW, GW and inoculum
Food waste 43.7 6.7 2.1 47.3 20.4
Green waste 43.1 5.7 1.1 50.1 40.2
Inoculum 35.1 4.3 1.74 58.7 20.1
Compositional data are expressed as percentage of dry matter
a
Oxygen was calculated by subtracting C, N and H values from 100% (Vivekanand et al. 2013)

342 ml g-1VS, respectively (Fig. 1). Biogas production was (Vivekanand et al. 2013). Pick et al. (2012) has reported
observed to be 8% more in FW than the GW. It indicated that that the biogas production from GW is less preferred
FW is readily biodegradable and GW is recalcitrant to because of the economy of the process and overall output
degrade biologically due to the presence of lignin (Momoh of biogas. However, FW and GW may be blended to have
and Nwaogazie 2011). Also, less availability of sugars optimum C/N ratio and thus maximum biogas yield.
contributed to lower production levels from GW. Therefore, a mix of FW and nitrogen-rich GW may pos-
The biogas potential observed in Table 3 was in sibly be advantageous for the biogas production. Blending
agreement with the other reported studies for biogas pro- in different ratios showed a positive impact on biogas
duction from FW (Browne and Murphy 2013; Zamanzadeh production ability of the feedstocks (Table 3). Effective
et al. 2016). Moreover, reported biogas yields from FW blending (FW as an easily degradable feedstock and
may vary due to the heterogeneous nature of the FW and recalcitrant GW) may have improved the presence of
difference in nutrient contents depending on type, origin macro- and micro-elements that would also affect the yield
and processing. Furthermore, operating temperature, positively. Co-digestion experiments showed the positive
organic loading rate, the design of bioreactor and pre- synergistic effect of blending.
treatment methods may significantly affect the biogas yield Biogas yield for the feedstock blends was estimated based
(Gadhe et al. 2014). The average menthane percentage in on the production from FW and GW alone, in addition to the
the biogas was 62%. experimental biogas yield (Table 3). All combinations had
The lignocellulose alone has been reported to degrade enhanced experimental/actual yields of biogas for the blends
slowly to produce biogas due to its high C/N ratio compared to the estimated yields of FW or GW alone. This
implies that there is a synergistic effect while co-digesting
the two different feedstocks. All the co-digestion experi-
ments showed significantly (P \ 0.05) higher biogas pro-
duction than mono-digestion of FW or GW.
Estimated and experimental biogas yield for the sub-
strate’s co-digestion ratios is shown in Table 3. The esti-
mated cumulative biogas yield of FW to GW ratio 85:15,
75:25, 50:50, 25:75, 15:85 is 365, 363, 350, 349 and
346 mL g-1 VS, respectively. Experimentally, it was
observed that cumulative yield of biogas was 515, 660,
456, 401, 467 ml g-1 VS of FW to GW ratio 85:15, 75:25,
50:50, 25:75, 15:85, respectively.
Blending ratio of 75:25 (FW/GW) had maximum biogas
yield compared to all other blends (Table 3); this may be
Fig. 1 Cumulative biogas production from FW, GW and blends due to the optimum C/N ratio which is more encouraging

123
 Int. J. Environ. Sci. Technol.

Table 3 Estimated and Blends ratio C/N ratio Estimated biogas yield Experimental biogas yield %Increase
experimental cumulative biogas
yield of blends (mLg-1VS) FW 20.4 – 370 ± 34 –
GW 40.2 – 342 ± 36 –
FW/GW (85:15) 23.3 365 515 ± 21 41.0
FW/GW (75:25) 25.3 363 660 ± 24 81.1
FW/GW (50:50) 30.3 356 456 ± 29 28.0
FW/GW (25:75) 35.2 349 401 ± 19 14.8
FW/GW (15:85) 37.5 346 467 ± 25 34.9

for biogas production than feedstocks alone. In the batch

bottle digester having 75:25 ratio of FW to GW, C/N ratio
was 25.37 which is very close to the standard C/N ratio for
yielding maximum biogas. This blend had cumulative
biogas yield which was 41.09, 28.08, 14.89 and 34.97%
higher than the yield of other reactors when the ratio of
FW/GW was 85:15, 50:50, 25:75, 15:85, respectively
(Fig. 1).
Comparing with individual digestion of FW and GW,
the ratio 75:25 (FW/GW) produced 78.37 and 92.98%
higher biogas which shows that digesting FW with GW
having higher C/N ratio would be profitable in terms of net
production of biogas.
There has been no pattern observed that increasing the Fig. 2 Effect of nanparticle on biogas production
amount of easily degradable FW along with GW will
enhance the biogas production; rather result indicated that the experiment of the blend (75:25; FW/GW) without TiO2
the optimum C/N ratio is responsible for maximum biogas was taken as control. It was evident from the results (Fig. 2)
generation from blends (Table 4). These results are con- that the addition of TiO2 was having almost no encouraging
sistent with previous report (Koch et al. 2016). impact on biogas production. The blend (75:25) without
TiO2 (as control) showed 690 ± 86 mL g-1 VS of biogas
Effect of nanoparticle on biogas production production, whereas the same blend with TiO2 (1.0 mg L-1)
had shown reduced biogas production (647 ± 27 mL g-1
The effect of the addition of nanoparticle (TiO2) on biogas VS). It seemed that there was no significant effect of the
production potential of the blend (75:25; FW/GW) was addition of nanoparticle in the current process although there
studied which had been observed to generate maximum have been reports that nanoparticles may or may not affect
biogas. For the same, five sets of experiments with varying the biogas production based on the type of nanoparticle used
amount of TiO2 starting from 0.5 to 2.5 mg L-1 were carried (Abdelsalam et al. 2016).
out. TiO2 was added into batch bottle digesters, and one set of
FTIR spectroscopy analysis
Table 4 Elemental analysis of FW, GW and inoculum
FTIR spectroscopy was used for qualitative characteriza-
Parameters Unit FW GW Inoculum tion and differentiation of FW and GW. A notable chemi-
cal difference has been observed in the spectrum of these
Aluminum (Al) mg/kg 21.3 137.1 16.3
a wastes that highlights their chemical nature. Figure 5a–c
Arsenic (As) mg/kg ND ND ND
shows the presence of different spectral patterns in FW,
Mercury (Hg) mg/kg ND 0.2 0.2
GW and inoculum. It may be observed that FW (Fig. 5a) is
Copper (Cu) mg/kg 1.9 9.0 2.9
having broad transmittance band at 2900–1600 cm-1
Lead (Pb) mg/kg 0.2 3.5 0.2
attributed to C–H stretch alkanes, C=O stretch esters
Iron (Fe) mg/kg 10.8 327.4 72.5
(Calderón et al. 2006), saturated aliphatic. The peak
a
ND Not detectable; detection limit (0.1) observed nearby 2900 cm-1 shows the presence of protein
and carbohydrate (Provenzano et al. 2014). The peak

123
Int. J. Environ. Sci. Technol.

centered to 3017 cm-1 is ascribed to alkanes stretch. The presence of Fe stimulates the biogas production (Jackson-
transmittance at 1463, 1160 and 1024 cm-1 is the char- Moss and Duncan 1990); contrary to this, in the present study
acteristic peak for C–H bend alkanes, C–N stretch aliphatic it was observed that biogas production from FW (10) is higher
amines and C–N stretch aliphatic amines, respectively. In than the GW. It may be noted that FW is readily biodegradable
Fig. 5b, GW waste is showing maximum number of peaks and GW is recalcitrant to degrade biologically due to the
at 1700–1036 cm-1 which attributed to C=O stretch esters presence of lignin (Momoh and Nwaogazie 2011).
and may responsible to show recalcitrance in gas produc- At the end of the AD (on 45th day), the pH and electrical
tion (Rodrı́guez et al. 2013), saturated aliphatic, N–O conductivity were in the range of 6.49–6.58 and
symmetric stretch nitro compounds, C–N stretch aromatic 3.67–4.17 mS, respectively. Initially, the conductivity
amines, C–N stretch aliphatic amines and C–N stretch ranged in between 3.23 and 3.25 mS and the marginal
aliphatic amines (Smidt et al. 2002). FTIR spectroscopy increase was observed (3.67–4.17 mS) on completion of
clearly showed that FW and GW both are rich in carbon as the experiment. This showed that there would have been
well as in nitrogen which is the primary requirement for degradation of organics that may have led to the deposition
AD. of salts (Manios and Siminis 1988).
Relatively higher intensity peaks were observed at 1650
and 1613 cm-1 in GW and inoculum, which is of very low Kinetic study
intensity in FW. This peak corresponds to the stretching of
C=C in aromatic group and C=O in ketones and/or quinones. Using modified Gompertz model
These groups may be representing their presence in the lig-
nocellulosic complex of GW and also as non-degraded Modified Gompertz equation was applied to fit cumulative
portion in the inoculum. The FW spectrum is having dif- biogas production data that was obtained experimentally
ferences with GW spectra with an intense peak at 1746 and through AD process. The estimated kinetic parameters based on
1160 cm-1, but it is with low intensity in GW representing modified Gompertz equation are given in Table 6. This stipu-
the presence of esters from carbohydrates in the FW. These lated that the equation may be used to estimate biogas yield
results are based on the appearance of the typical bands as potential, maximum biogas production rate and duration of lag
detected in spectra. The FTIR spectra results are showing phase. The predicted cumulative biogas values were plotted
bands associated with functional groups related to the main against the cumulative biogas production obtained experi-
chemical characteristics of FW and GW (Table 5). mentally to evaluate the soundness of the results obtained from
the modified Gompertz model (Fig. 3). Biogas yield potential
Elemental composition of the substrate (B) with composition ratio 85:15, 75:25, 50:50,
25:75, 15:85 were 1587, 1923, 1399, 1389, 1330 ml. This
The FW, GW and inoculum have been showing the presence showed that the digester having 75:25 (FW/GW) ratio had a
of aluminum in the range of 16.30–137.02 mgkg-1. The maximum biogas yield potential. Lag time was observed to be
presence of Cu, Pb and Fe was also detected, while arsenic was oscillating in between 1.34 and 1.78 days.
absent from all three samples (Table 4) The FW and GW had The coefficient of determination (R2) values for all
Fe content of 10.82 and 327.43 mgkg-1 which is required for digesters were in between 0.9971 and 0.9978, depicting
microbial cell growth. Researchers have reported that the that the predicted values best-fit with experimental values.

Table 5 Interpretation of bands (Fakharedine et al. 2006)

Wavelength (cm-1) Assignment

3500–3300 Hydrogen vibration of the OH group of alcohols, phenols and organic acid, amide hydrogen vibration
3000–2800 Aliphatic C–H stretching
1730 Esters from carbohydrates
1651 C=O vibrations of primary amides
1640–1620 Stretching of C=C in aromatic group, and of C=O in ketones and/or quinones
1540 C=H stretching in secondary amides
1460–1440 Aliphatic C–H deformation
1380 O–H deformation, C=O stretching of phenols, anti-symmetric COO– stretching and aliphatic C–H deformation
1230 Absorbance of ethers and aromatic esters and N–H bond of amides

123
 Int. J. Environ. Sci. Technol.

Table 6 Kinetic study: Parameters 1 2 3 4 5 6 7

modified Gompertz and logistic
model Experimental
CBPa (mL g-1 VS) 370 342 515 660 456 401 467
Modified Gompertz model
CBP (mL g-1 VS) 337 319 529 641 466 463 443
Lag phase (k) 1.27 1.87 1.35 1.28 1.34 1.64 1.78
RMSE (%) 6.88 9.82 12.28 8.75 13.20 7.85 11.10
Logistic model
CBP (mL g-1 VS) 402 327 313 629 483 434 513
Lag phase (k) 1.18 1.85 1.24 1.37 1.44 1.59 1.75
RMSE (%) 0.46 11.72 8.57 2.57 1.29 12.39 7.56
1-FW, 2-GW, 3-FW/GW (85:15), 4-FW/GW (75:25), 5-FW/GW (50:50), 6-FW/GW (25:75), 7-FW/GW
(15:85)
a
CBP Cumulative biogas production

Fig. 4 Soundness of experimental and predicted cumulative biogas

Fig. 3 Soundness of experimental and predicted cumulative biogas production (logistic model)
production (modified Gompertz model)

Researchers have applied modified Gompertz equation average coefficient of determination (R2) values for the
for their experiments reported to evaluate the kinetic study digesters was 0.9868. The root mean square error values
of cumulative gas production using vinasse where the R2 are in between 1.29 and 12.39. Figure 4 shows the com-
value was in the range of 0.993–0.999 and lag phase was in parison of experimental and predicted cumulative biogas
between 0.432 and 2.547 days (Budiyono et al. 2014). In production for all the digesters. This represents that logistic
another study performed by Cheong and Hansen (2006), model also fits well with the kinetic study of biogas pro-
the cattle manure sludge was utilized for anaerobic duction and to determine the lag phase of the reaction and
hydrogen production and it was reported that modified maximum biogas potential. Compared to the modified
Gompertz equation had best-fit for the kinetic study for Gompertz model, the mean square error (RMSE), the
these kinds of experiments with R2 value in between 0.997 coefficient of determination value (R2) and the overall
and 1.0 and RMSE in between 0.16 and 7.06. Similar percentage deviation of the logistic model are less. This
results have also been reported by Deepanraj et al. (2015). shows that modified Gompertz model suits well and have
the better goodness of fit for the kinetic studies of biogas
Using logistics model production as compared to logistic model (Fig. 5).

Estimated kinetic parameters’ values based on modified

logistic model are shown in Table 6. The soundness of the Conclusion
results from the model was evaluated by plotting the pre-
dicted and experimental cumulative biogas production The blend (FW:75::GW:25) was most suitable for nota-
values (Fig. 4) with blending ratio (FW/GW) 85:15, 75:25, bly higher biogas yield (1.7 fold) compared to FW alone
50:50, 25:75, 15:85 were 513, 629, 483, 434, 513 ml. that may be ascribed to optimum C/N ratio where
Estimated lag time was in the range of 1.24–1.75 days. The introduction of GW improved overall production and

123
Int. J. Environ. Sci. Technol.

Compliance with ethical standards

Conflicts of interest The authors report no conflicts of interest. The

authors alone are responsible for the content and writing of the paper.

References

Abdelsalam E, Samer M, Attia YA, Abdel-Hadi MA, Hassan HE,

Badr Y (2016) Comparison of nanoparticles effects on biogas
and methane production from anaerobic digestion of cattle dung
slurry. Renew Energy 87:592–598
Agdag ON (2009) Comparison of old and new municipal solid waste
management systems in Denizli, Turkey. Waste Manag
29:456–464
APHA (1989) Standard methods for examination of water and waste
water, 17th edn. American Public Health Association,
Washington
Appels L, Lauwers J, Degreve J, Helsen L, Lievens B, Willems K,
Impe JV, Dewil R (2011) Anaerobic digestion in global bio-
energy production—potential and research challenges. Renew
Sustain Energy Rev 15:4295–4301
Browne JD, Murphy JD (2013) Assessment of the resource associated
with biomethane from food waste. Appl Energy 104:170–177
Budiyono B, Syaichurrozi I, Sumardiono S (2014) Effect of total solid
content to biogas production rate from vinasse. Int J Eng
27:177–184
Calderón FJ, McCarty GW, Reeves JB (2006) Pyrolisis-MS and FT-
IR analysis of fresh and decomposed dairy manure. J Anal Appl
Pyrol 76:14–23
Callaghan FJ, Wase DAJ, Thayanithy K, Forster CF (2002) Contin-
uous codigestion of cattle slurry with fruit and vegetable wastes
and chicken manure. Biomass Bioenerg 22:71–77
Cheong DY, Hansen CL (2006) Bacterial stress enrichment enhances
anaerobic hydrogen production in cattle manure sludge. Appl
Microbiol Biotechnol 72:635–643
Deepanraj B, Sivasubramanianb V, Jayaraj S (2014) biogas gener-
ation through anaerobic digestion process-an overview. Res J
Chem Environ 18(5):80–93
Deepanraj B, Sivasubramanianb V, Jayaraja S (2015) Kinetic study
on the effect of temperature on biogas production using a lab
scale batch reactor. Ecotoxicol Environ Saf 121:100–104
Deepanraj B, Sivasubramanianb V, Jayaraj S (2017) Experimental
and kinetic study on anaerobic digestion of food waste co-
Fig. 5 FTIR spectra of FW (a), GW (b) and inoculum (c) digested with poultry manure. Desalin Water Treat 59:72–76
Demirel B, Scherer P (2011) Trace element requirements of
agricultural biogas digesters during biological conversion of
renewable biomass to methane. Biomass Bioenerg 35:992–998
indicated significant synergistic effect. FTIR analysis Fakharedine N, El Hajjouji H, Ait Baddi G, Revel JC, Hafidi M
showed differences in the chemical structure of biomass. (2006) Chemical and spectroscopic analysis of organic matter
Kinetic studies depicted that modified Gompertz model transformation during aerobic digestion of olive-mill waste-
waters. Process Biochem 41:398–404
was better fit compared to logistic model with the FAO (2011) Towards the future, we want: End hunger and make the
experimental results. Co-digestion (FW/GW) may prove transition to sustainable agricultural and food systems. Food and
to be robust integrated process for biogas generation and agriculture organization of the United Nations, Rome
waste reduction. It may further be explored for the FAO (2012) Towards the future, we want: End hunger and make the
transition to sustainable agricultural and food systems. Food and
development of multi-fuel co-digestion and specific agriculture organization of the United Nations, Rome
analysis for enhanced biogas yield. Feng XM, Karlsson A, Svensson BH, Bertilsson S (2010) Impact of
trace element addition on biogas production from food industrial
Acknowledgements The authors would like to thank Department of waste-linking process to microflora. FEMS Microbiol Ecol
Biotechnology (Government of India) to support this work financially 74:226–240
(No. BT/RLF/Re-entry/04/2013). We also thank Material Research Gadhe A, Sonawane SS, Varma MN (2014) Ultrasonic pretreatment
Centre at MNIT Jaipur, SIC-IIT Indore and Mr. Ishwar at Rajasthan for an enhancement of biohydrogen production from complex
Gosewa Sangh, Durgapura, Jaipur, for their help in the experiments. food waste. Int J Hydrog Energy 39:7721–7729

123
 Int. J. Environ. Sci. Technol.

Jackson-Moss CA, Duncan JR (1990) The effect of iron on anaerobic Pick D, Dieterich M, Heintschel S (2012) Biogas production potential
digestion. Biotech Lett 12:149–154 from economically usable green waste. Int J Sustain Dev
Kafle GK, Kim SH, Sung KI (2012) Batch anaerobic co-digestion of 4:682–702
Kimchi factory waste silage and swine manure under mesophilic Provenzano MR, Malerba AD, Pezzolla D, Gigliotti G (2014)
conditions. Biores Technol 124:489–494 Chemical and spectroscopic characterization of organic matter
Kafle GK, Sujala B, Kim SH, Chen L (2014) Anaerobic digestion of during the anaerobic digestion and successive composting of pig
Chinese cabbage waste silage with swine manure for biogas slurry. Waste Manag 34:653–660
production: batch and continuous study. Environ Technol Rodrı́guez Á, Gómez X, Blanco D, Cuetos MJ, Fernández B, Flotats
35:2708–2717 X (2013) Study of thermal pre-treatment on anaerobic digestion
Katami T, Yasuhara A, Shibamoto T (2004) Formation of dioxins of slaughterhouse waste by TGA-MS and FTIR spectroscopy.
from incineration of foods found in domestic garbage. Environ Waste Manag Res 3:1195–1202
Sci Technol 38:1062–1065 Smidt E, Lechner P, Schwanninger M, Haberhauer G, Gerzabek MH
Koch K, Drewes JE (2014) Alternative approach to estimate the (2002) Characterization of the waste organic matter by FT-IR
hydrolysis rate constant of particulate material from batch data. spectroscopy: application in waste science. Appl Spectrosc
Appl Energy 120:11–15 56:1170–1175
Koch K, Plabst M, Schmidt A, Helmreich B, Drewes JE (2016) Co- Thauer RK, Kaster AK, Seedorf H, Buckel W, Hedderich R (2001)
digestion of food waste in a municipal wastewater treatment Methanogenic archaea: ecologically relevant differences in
plant: comparison of batch tests and full-scale experiences. energy conservation. Nat Rev Microbiol 6:579–591
Waste Manag 47:28–33 Vandermeersch T, Alvarenga RAF, Ragaert P, Dewulf J (2014)
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein Environmental sustainability assessment of food waste valoriza-
measurement with the Folin phenol reagent. J Biol Chem tion options. Resour Conserv Recycl 87:57–64
193:265–275 Vivekanand V, Olsen EF, Eijsink VGH, Horn JS (2013) Effect of
Manios VI, Siminis HI (1988) Town refuse compost of Heraklio. different steam explosion conditions on methane potential and
Biocycle 29:44–47 enzymatic saccharification of birch. Biores Technol
Miller GL (1959) Use of dinitro salicylic acid reagent for determi- 127:343–349
nation of reducing sugar. Anal Chem 31:426–432 Wang Q, Jiang JG, Zhang YJ, Li KM (2015) Effect of initial total
Momoh OLY, Nwaogazie IL (2011) The effect of waste paper on the solids concentration on volatile fatty acid production from food
kinetics of biogas yield from the co-digestion of cow dung and waste during anaerobic acidification. Environ Technol
water hyacinth. Biomass Bioenerg 35:1345–1351 36:1884–1891
Nopharatana A, Pullammanappallil PC, Clarke WP (2006) Kinetics Yadvika S, Sreekrishnan TR, Kohli S, Rana V (2001) Enhancement
and dynamic modelling of batch anaerobic digestion of munic- of biogas production from solid substrates using different
ipal solid waste in a stirred reactor. Waste Manag 27:595–603 techniques: a review. Biores Technol 95:1–10
Paritosh K, Kushwaha S, Yadav M, Pareek N, Chawade A, Zamanzadeh M, Hagen LH, Svensson K, Linjordet R, Horn SJ (2016)
Vivekanand V (2017) Food waste to energy: an overview on Anaerobic digestion of food waste: effect of recirculation and
sustainable approach for food waste management and nutrient temperature on performance and microbiology. Water Res
recycling, BioMed Res Intl 2017:1–19 96:246–254
Pham TPT, Kaushik R, Parshetti GK, Mahmood R, Balasubramanian Zirkler D, Peters A, Kaupenjohann M (2014) Elemental composition
R (2015) Food-waste-to-energy conversion technologies: current of biogas residues: variability and alteration during anaerobic
status and future directions. Waste Manag 38:399–408 digestio. Biomass Bioenerg 67:89–98

123