Energy 118 (2017) 377e386

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Effects of thermal pretreatment on degradation kinetics of organics

during kitchen waste anaerobic digestion
Yangyang Li a, b, Yiying Jin a, b, *, Jinhui Li a, b, Hailong Li c, Zhixin Yu d, Yongfeng Nie a, b
a
School of Environment, Tsinghua University, Beijing 100084, China
b
Key Laboratory for Solid Waste Management and Environment Safety (Tsinghua University), Ministry of Education of China, Tsinghua University, Beijing
100084, China
c
Malardalen University, School of Sustainable Development of Society and Technology, SE-721 23 Va
sters, Sweden
d
Department of Petroleum Engineering, University of Stavanger, N-4036 Stavanger, Norway

a r t i c l e i n f o a b s t r a c t

Article history: The inuence of thermal pretreatment on degradation properties of organics in kitchen waste (KW) was
Received 29 April 2016 investigated. The kinetics results showed that thermal pretreatment could enhance the degradation
Received in revised form efciency of crude protein (CP), fat, oil and grease (FOG), volatile solid (VS) and volatile fatty acids (VFA).
1 November 2016
Thermal pretreatment showed no signicant difference in the nal concentration of protein but could
Accepted 11 December 2016
Available online 22 December 2016
decrease the FOG degradation potential (7e36%), while increased the lag phase for degradation of
protein and FOG respectively by 35e65% and 11e82% compared with untreated KW. Cumulative biogas
yield increased linearly and exponentially with the removal efciency of VS and other organics (CP and
Keywords:
Kitchen waste
FOG) respectively. Additionally, the reduction of CP increased exponentially with FOG removal efciency.
Thermal pretreatment The calculating methods of biogas yield, organics reduction and corresponding appropriate digestion
Anaerobic digestion retention based on FOG and CP reduction amount and pretreatment parameters were suggested.
Degradation kinetics 2016 Elsevier Ltd. All rights reserved.

1. Introduction Previous researchers have mainly focused on optimizing the

A large amount of kitchen waste (KW) is generated each year.
Appropriate and effective treatment of KW has always been
prominent concerns in recent years in order to avoid numerous
problems, such as food safety and environmental quality due to the
gutter oil and et al.
High biodegradability and water content make KW suitable for
anaerobic digestion (AD) with the concomitant benet of clean
energy production (such as biogas) and organic waste reduction.
However, the high content of macromolecular organic matters in
KW, such as fat, oil and grease (FOG) and crude proteins (CP), which
accounted for 30e70% of the total organic matter (dry basis), could
easily lead to relatively longer started-up period when AD is
applied. In addition, the content of FOG is usually higher (1e5% wet
basis) in KW, causing excessive acidication and inhibition of nu-
trients transport, thus unfavorable to a sustained and stable
degradation during the digestion process [1].
* Corresponding author. School of environment, Tsinghua University, Beijing,
China.
E-mail address: jinyy@tsinghua.edu.cn (Y. Jin).
anaerobic treatment conditions of KW (single phase/two-phase/
three stage, continuous/sequencing-batch digestion, temperature
and retention time of digestion, mono-digestion/co-digestion, etc.)
[2e4], factors inuencing and inhibiting the digestion process
(feedstock such as carbon and nitrogen ratio, organic compositions,
ammonia nitrogen, salt content, long-chain fatty acid (LCFA), vol-
atile fatty acid (VFA), organic loads, etc.) and pretreatment methods
to enhance the organic reduction and methane production (e.g.,
thermal, thermochemical, ultrasonic, biological, microwave and
chemical pretreatment) [5,6]. In addition, the long retention time of
the AD process of KW is a major concern and numerous methods
for pretreating KW prior to the AD process have been developed to
accelerate the digestion process and to enhance the biomethane
production [7,8]. Among these pretreatment methods, thermal
treatment is one of the easiest and most studied pretreatment
methods and has already been applied at a full-scale [7], which
could be applied as the sole process or as a pre- or post-process
sterilization phase. It can also effectively promote the solubiliza-
tion and liquefaction of complex macromolecule solid organic
matters, thus reducing the particle size of solid materials,
enhancing the solubilization of organics, obtaining an improve-
ment in the total soluble chemical oxygen demand, hence favoring

http://dx.doi.org/10.1016/j.energy.2016.12.041
0360-5442/ 2016 Elsevier Ltd. All rights reserved.
378 Y. Li et al. / Energy 118 (2017) 377e386

for digestion of anaerobic microbes [1,9e13]. Table 1

Additionally, anaerobic digestion of KW is a complex biochem- Characteristics of the KW.

ical reaction process. The biomethane production and composition Compositions of KWa Characteristics of the KWa
at different conditions can be estimated according to the estab- Parameters Percentageb (%) Parameters KW
lishment of biochemical reaction kinetics model and the numerical
Cooked bone 2.7 0.9 pH 6.5 0.2
simulation of anaerobic digestion process [6,14]. It is benecial for
Cooked eggshell 1.0 0.6 Total solids (%) 18.7 0.4
understanding the process of organic matter degradation and Pasta & rice 29.1 2.1 VS (%, dry basis) 93.2 0.5
transformation controlled by microorganism when performing the Fruit peeling 23.3 1.2 Carbohydrate (%, wet basis) 11.8 0.4
investigation of reaction rate, mass transfer characteristics and the Cooked vegetable 19.5 1.8 CP (%, wet basis) 2.5 0.2
Vegetable peeling 23.0 2.2 FOG (%, wet basis) 3.5 0.1
effect on the internal dynamics in the different stages of AD pro-
Others 1.4 0.2 Carbon (%, dry basis) 46.1 1.7
cess, which can also be served as an effective method for guiding Hydrogen (%, dry basis) 6.9 0.2
the design and operation of AD reaction system [5,15]. The rst- Oxygen (%, dry basis) 37.8 0.1
order kinetics, modied Gompertz kinetics and methods provided Nitrogen (%, dry basis) 3.2 0.3
by Koch and Drewes were common approaches to carrying out the a
Each indicator was measured three times.
kinetics study on the hydrolysis and the biogas production of AD b
Determined by weight, wet basis.
process [16e18].
However, in previous researches and practice, thermal pre-
treatment was mainly used for the modication of sludge, and few
researchers have studied the effect of thermal pretreatment on the pressure vessel with a heating shell. It consisted of an electrical
physical and chemical properties of KW and efciency of subse- heating layer, a surface scrape agitator and a reaction chamber
quent AD process. And the existing researches are mainly on ana- (effective volume 20 L, length 459 mm) with a switchable sealing
lyses of the degradation efciency of organic waste, such as KW, valve for pressure control. The reactor was designed to withstand
sludge, etc and achieving higher degradation efciency by thermal pressures as high as 2.0 MPa and temperatures as high as 220
C.
pretreatment through AD tests, such as CP, FOG and volatile solid The temperature and agitator rotation speed were controlled by the
(VS). As for the degradation kinetics, most researchers centered on Fuzzy-PID controller. A thermocouple was used to monitor the
the sewage sludge [19e21], and few studies focused on the detailed sample temperature, and a blender motor was used to mix the
degradation properties of organic degradation during AD process of samples in the vessel completely during the whole heating process.
KW. As for KW with thermal pretreatment, no previous study has During the pre-treatment, approximately 1 kg of KW was
investigated the variation in properties of anaerobic degradation transferred to the vessel, which had been preheated to a pre-
kinetics. determined temperature. After hydrolysis at this temperature for a
In order to achieve a comprehensive understanding and selected time, the heating process was terminated, and the samples
comparative conclusions on the mechanism analysis of organics were chilled by circulating 10
C water until the vessel cooled to
degradation enhanced by thermal pretreatment during AD process room temperature. After pre-treatment, the samples were stored at
of KW, particularly the aim of demonstrating the relevant role 4
C in a refrigerator to minimize the volatilization of organic
played on the biodegradability performance by thermal pretreat- compounds prior to further analysis.
ment, the objective of this study was to assess the kinetic study of We have evaluated the inuence of thermal pretreatment on
organics degradation efciency by thermal pretreatment, investi- physical and chemical properties of KW and the AD efciency in our
gate the inherent relationship between organics biodegradability previous research [22]. In this study, the KW was pretreated for
and biogas production potential. The inuence of thermal pre- 70 min at 55e90
C and 50 min at 120e160
C. Discussion of the
treatment on the degradation kinetics of organics (including CP, reasons for preferring the thermal pretreatment parameters can be
FOG, VS, and VFA), interactions between two main organics com- found in our previous research [22].
ponents in KW, and the relationship between organics degradation
and biogas production during AD of KW were investigated. Addi-
tionally, the feasible strategies for enhancing the AD of KW by
combined thermal pretreatment and AD system was suggested to 2.3. Anaerobic digestion
optimize the operational parameters and provide a theoretical
basis for engineering application. After removal of the oating oil via thermal pretreatment and
subsequent centrifugation, the liquid and solid phases of KW were
2. Materials and methods blended thoroughly for anaerobic digestion experiments, which
were carried out in 250 mL glass bottles with an effective volume of
2.1. KW 200 mL at 35
C. The seed sludge was obtained from a steady-
operation digester in a waste water treatment plant after two-day
Table 1 shows the basic characteristics of KW collected on a gravity sedimentation prior to inoculation. Furthermore, each
weekly basis from a canteen in Tsinghua University that can serve digester was fed with a mixture of seed sludge and KW with a mass
approximately 10,000 students and staff members per meal. After ratio of 14: 1 (with an F/I ratio of 1: 2 on a volatile solid (VS) basis).
manual sorting in order to remove impurities, such as big bones, The upper space of each reactor was ushed with nitrogen for at
plastics, and metals, the KW was mixed with a kitchen blender to least 3 min to assure the anaerobic conditions and then sealed
ensure uniform and representative experimental materials. It was quickly. In each experimental run, three control digesters were
then transferred to a food crusher and shredded into particles with operated. Two blank digesters which contained inoculums only
an average size of 1e2 mm and stored at 4
C in a refrigerator. were also incubated at the same time to correct for the biogas yield
from the inoculums. All bottles were put in a water bath set at 35
C
2.2. Thermal pretreatment and shook manually twice a day, and the digestion experiments
were run for 21 days. The volume of biogas produced during
Thermal hydrolysis pretreatment was performed in a 20 L anaerobic digestion was calculated by a wet-type gas owmeter
stainless steel hydrolysis reactor, which was constructed as a and then collected in gas bags.
 Y. Li et al. / Energy 118 (2017) 377e386
2.4. Kinetics analyses
Two kinetics study models including rst-order kinetics model
and Gompertz kinetics study were used to evaluate the degradation
rate constants (k).
a Firstorder kinetics : ln M kRT b
where M is the concentration of organics, such as protein and fat, b
is the constant, k is the rst-order hydrolysis constant (d1), and RT
is the retention time (d).
As for the rst-order kinetics study, degradation kinetics
including two-section (listed as 1st section and 2nd section) and
only one single-section were applied.
b Gompertz kinetics : P Aexp BexpCRT
where P is the cumulative concentration of organics, A is the
organic concentration potential, B is the duration of lag phase (d), C
is the maximum degradation rate and RT is the retention time in
the anaerobic digester (d).
2.5. Analytical methods
Samples collected from the thermal pretreatment reactor and
anaerobic digesters were dried and the VS, Crude protein (CP) and
lipid content were tested. VS were determined according to the
Standard methods of the American Public Health Association [23].
Total Kjeldahl nitrogen was determined according to the Kjeldahl
method and CP content was calculated by multiplying 6.25 [24].
Lipid content was performed following Soxhlet extraction methods
described by the Association of Analytical Communities [25]. Vol-
atile fatty acid (VFA) was measured by gas chromatography (GC,
Agilent Technologies, Model 7890A, USA) with a ame ionization
detector (FID)). Prior to GC analysis, 10 mL of digested samples was
centrifuged at 15,000 rpm for about 10 min, and then the super-
natant was ltrated over 0.45 mm lter paper.
3. Results and discussion
3.1. CP degradation properties during KW anaerobic digestion
The concentrations of proteins during AD process for untreated
KW and pretreated KW are shown in Fig. 1. For all samples, the
protein concentrations were reduced gradually, while pretreated
KW presented higher CP content in the liquid phase. These results
could be due to the following two reasons. First, the disintegration
process caused by thermal pretreatment could have accelerated the
solubilization of CP in the form of particulate organics into the
liquid phase. Secondly, the readily available soluble CP in the liquid
phase might have been efciently digested by fermentative
bacteria.
CP degradation was kinetically characterized during the AD
process of untreated KW and KW with thermal pretreatment. With
the increase of thermal pretreatment temperatures and treatment
durations, the protein was hydrolyzed from the solid phase into the
liquid phase by hydrolysis and acidogenesis microorganisms, and
hydrolyzation products contained nitrogen was then used to pro-
duce biogas by methanogens, thus, the concentration of CP
continuously decreased. As for CP degradation kinetics of anaerobic
process (kCp), existing researches mainly used rst-order kinetics
tting, including single-section kinetics model [26] and multi-
section kinetics model (including lag phase, rapid degradation
phase, and slow degradation phase) [27].
The kCp values with different treatment temperatures obtained
379
from the single-section kinetics model (Table 2(b)) varied in a
similar tendency with those in Table 2(a). To evaluate the sound-
ness of the model results, statistical indicators (R2) were also
considered. When multi-section kinetics model was used, it could
be found that both the kCp and R2 values in the rst section were
higher compared with the second section, and the results were
accordance with Yang et al. (2015) using the inocula obtained from
a mesophilic anaerobic digester in a wastewater treatment plant for
anaerobic digestion [27]. The hydrolysis of organics is commonly
considered as rate limiting step during AD of organic wastes, such
as cellulose, municipal solid waste, and sludge, etc. However, as for
KW, the inhibition caused by hydrolysis is not so obvious, so the lag
phase can be omitted.
The kCp values obtained from the single-section kinetics model
(Table 2(b)) fell within the range of 0.24e0.45 d1, similar to those
reported by Garc-A-Heras (2004) for the degradation of CP under
mesophilic AD condition (0.25e0.80 d1) [28]. The protein degra-
dation constant for untreated KW was 0.35 d1, higher than those
reported by Yang et al. (2015) for the degradation of sludge
(0.0018e0.0197 d1) [27] and by Lee et al. (2015) for the degrada-
tion of cattle offal (0.28 d1) [26]. In addition, the CP degradation
rate for KW pretreated for 50 min at 140
C increased by 28%
compared with untreated KW, while 17e33% reduction for other
treatment temperatures. In addition, the R2 values ranged from
0.9509 to 0.9913. Thus, the relatively simple rst order kinetics
model can effectively describe the CP degradation during AD pro-
cess of KW.
Numerous factors would inuence the kCp values, such as
operating conditions (like pH, temperature, inoculation proportion,
organic loading, operation mode, etc.), environmental conditions
and pretreatment (such as thermal pretreatment). In this study, the
main reasons led to various differences between kCp values are due
to the thermal pretreatment. Firstly, there can be signicant vari-
ation in the organics constitutes between different thermal pre-
treatment parameters, such as the composition and content of
amino acid, thus resulting in the different content of CP compo-
nents in substrates used for subsequent AD. A specic type of
protease is responsible for the decomposing of corresponding
specic protein to amino acids during AD of KW, hence the
degradation rate of different types of protein compounds will be
different [27]. Besides, the degradation efciency of the same
protein was also affected by the surface to volume ratio of located
particles during the digestion process [29]. After thermal pre-
treatment and subsequent removal of oating oil, the carbohydrate
and FOG content in the substrates for AD would be different [22],
thus leading to different kCp values due to the following two rea-
sons. First, glucose, the main hydrolysate of carbohydrate can cause
repression of protease formation [30]. Second, the FOG concen-
tration could have a great inuence on the degradation efciency of
protein due to the different inhibition effect of FOG concentration
[31], such as adhering to the surface of organics particles and mi-
crobial cells and long-chain fatty acids (LCFA) inhibition [32].
Additionally, when KW were pretreated for more than 50 min at
160
C, Maillard reaction would occur (i.e., a reaction between
amino acids and sugars during the degradation of complex sub-
stances) and result in the production of melanoidins [33], which are
difcult to degrade under anaerobic conditions. Hence, this might
result in lower protein degradation efciency [9]. Moreover, the
Maillard reaction types differ from various thermal pretreatment
temperatures and durations [10], hence leading to the differences
between protein degradation rates.
Factors like maximum kCp values, degradation potential and
duration of lag phase are important for determining the degrada-
tion efciency of protein during anaerobic digestion (Table 2(b)).
No major variation in the concentration occurred to protein
380 Y. Li et al. / Energy 118 (2017) 377e386

Fig. 1. Degradation kinetics of CP during AD of KW (two sections using rst-order kinetics (a), exponential-order kinetics (b) and Gompertz kinetics(c)).

Table 2(a)
Summarization of degradation kinetics of CP in the liquid phase during AD of KW using two-section kinetics.

Item the 1st sectiona the 2nd sectiona

First-order kinetics R2 First-order kinetics R2

KW lnCp 0.045RT9.4431 0.8262 lnCp 0.004RT9.1506 0.9793

55
C 70 min lnCp 0.0813RT9.6423 0.9944 lnCp 0.0094RT9.2536 0.9542
70
C 70 min lnCp 0.0630RT9.6260 0.9025 lnCp 0.0090RT9.2624 0.8527
90
C 70 min lnCp 0.0640RT9.6211 0.9804 lnCp 0.0051RT9.1624 0.9990
120
C 50 min lnCp 0.065RT9.5524 0.9076 lnCp 0.0016RT9.0842 0.8775
140
C 50 min lnCp 0.0640RT9.5851 0.7838 lnCp 0.0042RT9.146 0.8091
160
C 50 min lnCp 0.0650RT9.6724 0.8862 lnCp 0.0076RT9.2248 0.8553
a
The rst section for KW pretreated at 55e120
C began from the initial 5 days of digestion, compared with the initial 8 days for KW pretreated at 140e160
C and untreated
samples; CP: crude protein contents in KW, dry basis.

concentration for pretreated samples after the whole digestion
process, only 1e4% higher compared with untreated KW. However,
the lag phase was increased by 35e65%, while 7e33% reduction
was achieved for maximum kCp values. Besides, protein is the main
nitrogen compounds in the KW and its hydrolysis would also affect
the C/N ratios during the AD process. Our previous research
described the characteristics of C/N ratios in the dry solid phase of
KW after pretreatment with different temperatures and durations
[34]. Basing on these results, a positive correlation could be
concluded between both the degradation potential and the
maximum degradation rate values of protein and the C/N ratios,
and the Pearson correlation (r) were 0.648 and 0.554, respectively,
compared with negative correlation between lag phase of protein
degradation and C/N ratios. This phenomenon could be due to the
following reasons. Firstly, high C/N ration in the feed could result in
the rapid consumption of nitrogen in protein by methanogens, thus
causing higher degradation of the protein. Secondly, low C/N ratio
would lead to the accumulation of nitrogen in the form of
ammonia, thus inhibiting microorganism growth through
ammonia toxicity. Therefore, it is essential to balance the C/N ratio
in the feed for sustainable anaerobic processes and biogas pro-
duction, such as co-digestion with other biomass waste.
 Y. Li et al. / Energy 118 (2017) 377e386 381

Table 2(b)
Summarization of degradation kinetics of CP in the liquid phase during AD of KW using both Exponential kinetics model and Gompertz kinetics.

Item Exponential kinetics modela Gompertz kineticsa

Fitting equation R2 Fitting equation R2

KW CP 4560exp(-0.3530RT)8822 0.9846 CP 8776exp(0.4205exp(-0.2962RT)) 0.9852

55
C 70 min CP 6793exp(-0.2551RT)8750 0.9875 CP 8686 exp (0.5858 exp (0.2113RT)) 0.9857
70
C 70 min CP 6996 exp (0.2826RT)8935 0.9844 CP 8875 exp (0.5889 exp (0.2342RT)) 0.9799
90
C 70 min CP 7038 exp (0.2372RT)8502 0.9913 CP 8458 exp (0.6156 exp (0.1995RT)) 0.9869
120
C 50 min CP 6267 exp (0.2931RT)8404 0.9520 CP 8395 exp (0.5661 exp (0.2759RT)) 0.9462
140
C 50 min CP 7245 exp (0.4519RT)8876 0.9509 CP 8720 exp (0.6010 exp (0.3206RT)) 0.9636
160
C 50 min CP 7979 exp (0.2741RT)8829 0.9684 CP 8662 exp (0.6595 exp (2061RT)) 0.9766
a
CP: crude protein contents in KW, dry basis.

3.2. FOG degradation properties during KW anaerobic digestion

FOG is an essential component in KW, especially the case in

China, including animal fats and vegetable oils. LCFA, the main
component of FOG, are usually thought to inhibit methane gener-
ation and could cause operational problems when occurring in high
concentrations during AD of KW. LCFA are degraded anaerobically
via the b-oxidation pathway to acetate and H2, then subsequently
converting to methane [35]. Saturated fatty acids (SFA) follow the
traditional b-oxidation pathway, while two possible pathways are
suggested for degradation of unsaturated fatty acids (UFA)
including complete saturation followed by the typical b-oxidation
pathway [36] and b-oxidation before fatty acid (FA) saturation [37].
In addition, different fatty acids also show various degradation
pathways, such as stearic acid (C18:0), palmitic acid (C16:0), oleic
acid (C18:1) and linoleic acid (C18:2) [38,39].
Variation in the FA composition could be caused due to different
thermal hydrolysis parameters, which would affect the degradation
rates of FOG during AD (See Fig. 2). The FOG degradation rate
constant (kFOG) during AD process was presented in Table 3. The
kFOG values of untreated KW were 0.0385 d1, and an increase of
12.5e36.6% was achieved for pretreated samples, which increased
with the increasing of thermal pretreatment temperatures. As for
KW pretreated for 50 min at 140
C, the highest kFOG value was
achieved (0.0526 d1), and the rst-order kinetics model also ob-
tained the same results. In addition, according to the Gompertz
kinetics results in Table 3, the untreated KW presented the highest
maximum degradation potential and lowest lag phase duration,
7e36% lower and 11e82% higher respectively for pretreated sam-
ples compared with untreated KW.
Basing on our previous research about the inuence of thermal
pretreatment on the FOG content in the solid phase of KW [22], a
positive correlation between both the maximum degradation po-
tential and maximum degradation rate values of FOG and the FOG
concentration could be concluded, and the r values were 0.568 and
0.442, respectively. These phenomenons could be explained by
following reasons. First, the release of FA in KW was more pro-
nounced with thermal pretreatment, especially at higher treatment
Fig. 2. Degradation kinetics of FOG during AD of KW (rst-order kinetics (a) and
temperatures and longer treatment durations, thus leading to
Gompertz kinetics(b)).
different effects on the FA components. Hence, the different
degradation efciency, degradation pathway and involved bacteria
of FA components, such as oleate and stearate, could cause the concentration because the hydrophobic LCFAs would wrap or
difference between degradation rates. Second, thermal pretreat- adsorb onto the surface of microorganism and blocks the mass
ment results in different FOG concentration in KW after the transfer process between soluble organic matter and microbial
removal of oating oil, and the amount of FOG used for subsequent cells, thus higher concentration of LCFAs will exhibit toxicity effects
AD process varied for KW pretreated with different temperatures, to microorganisms and cause negative effective on organic degra-
thus presenting different degradation efciency of FOG during AD dation and methane production.
process. Thirdly, the accumulation of LCFAs during the hydrolysis However, after the lag phase, anaerobic bacteria could be able to
stage would inhibit anaerobic microorganisms [40]. Thus, the lag digest the accumulated LCFA and continue the biogas production
phase would be increased while the methane yield may be process although long retention times are required due to little
decreased. In addition, the inhibition mechanism of FOG is based on survival of anaerobic consortium [41].
382 Y. Li et al. / Energy 118 (2017) 377e386

Table 3
Summarization of degradation kinetics of FOG in the solid phase during AD of KW using rst-order kinetics and Gompertz kinetics.

Item FOG- First-order kineticsa FOG- Gompertz kineticsa

Fitting equation R2 Fitting equation R2

KW lnFOG 0.0385RT-1.1998 0.9209 FOG 0.1296 exp (0.9454 exp (0.0950RT)) 0.9934
55
C 70 min lnFOG 0.0433RT-1.2863 0.8722 FOG 0.1085 exp (1.0467 exp (0.0968RT)) 0.9087
70
C 70 min lnFOG 0.0439RT-1.2354 0.9299 FOG 0.0827 exp (1.3221exp(-0.0564RT)) 0.9413
90
C 70 min lnFOG 0.0456RT-1.2592 0.9549 FOG 0.0857exp(1.2786exp(-0.0656RT)) 0.9886
120
C 50 min lnFOG 0.0447RT-1.2992 0.8610 FOG 0.1121exp(1.0401exp(-0.1204RT)) 0.9586
140
C 50 min lnFOG 0.0526RT-1.0601 0.9319 FOG 0.1031exp(1.7207exp(-0.1270RT)) 0.9203
160
C 50 min lnFOG 0.0438RT-1.2334 0.8873 FOG 0.1077exp(1.1012exp(-0.0875RT)) 0.9386
a
FOG: FOG contents in KW, dry basis.

3.3. Degradation properties of other organics during KW anaerobic Table 4(a)

digestion Summarization of degradation kinetics of VS in the solid phase during AD of KW
using rst-order kinetics.

VS degradation was kinetically characterized during the AD Item First-order kinetics

process (Fig. 3(a)) and the results are summarized in Table 4(a). Fitting equationa R2
With the increase of thermal pretreatment temperatures, the
KW DVS 0.0087RT0.1496 0.9592
degradation rate constant of VS increased initially to the highest 55
C 70 min DVS 0.0068RT0.2533 0.8663
(0.0151 d1) for samples pretreated at 90
C and then decreased 70
C 70 min DVS 0.0147RT0.0933 0.9626
continuously to 0.0032 d1 for samples pretreated at 160
C. 90
C 70 min DVS 0.0151RT0.1225 0.9198
Thermal pretreatment could destroy large molecules and 120
C 50 min DVS 0.0119RT0.2160 0.7961
140
C 50 min DVS 0.0062RT0.1405 0.8222
160
C 50 min DVS 0.0032RT0.1343 0.9413
a
DVS: VS reduction.

refractory organic matters and reduce the particle size, especially at

Fig. 3. VS (a) and VFA (b) degradation kinetics during AD of KW.
high temperature and long treatment duration. As to the AD pro-
cess of untreated KW, a relatively longer time was required than
those with thermal pretreatment in which the proportion of
biodegradable ingredients increased and the digestion process
could be completed in less time. Thus, the degradation rate of VS
was greatly enhanced for samples pretreated at 70e120
C,
36.8e73.6% higher compared with untreated KW (0.0087 d1).
However, KW pretreated at higher temperatures (>120
C) turns
brownish due to the Maillard reaction, i.e., the reaction between
amino acids and sugars, in which reaction and the formation of
byproducts (such as Amadori-like compounds and byproducts of
melanoidins) under these pretreatment conditions may produce
inferior improvements in the biomethane production and organic
degradation. Hence a continuous decrease of VS degradation con-
stant (approximately 28.7e63.2%) was obtained for the degradation
rates at 140e160
C compared with untreated KW.
During AD process the hydrolysis of soluble organics will be
transformed to VFA, which can be used by methanogens to produce
biogas. Hence, the rate of hydrolysis and VFA production can be
used to determine the competence of subsequent biodegradability
studies. As for VFA degradation, stepwise degradation properties
were obtained during KW AD process (Fig. 3(b)) and the results are
summarized in Table 4(b).
VFA concentration increased rapidly within the rst 5 days for
untreated KW and KW pretreated at 55e90
C and within the 3
days for KW pretreated at 120e160
C, and then decreased
continuously. In addition, both the increase phase (the 1st section)
and decrease phase (the 2nd section) of VFA degradation followed
the exponential reaction equation with its hydrolysis retention
time (RT) for untreated KW and KW pretreated at 55e90
C,
compared with initial rst-order kinetics in the rst section and
then exponential-kinetics in the second section for KW pretreated
at 120e160
C. Such results could be due to the following two
reasons. First, thermal pretreatment could have promoted the
solubilization or liquefaction of particulate organics with large
molecules, thus increasing the soluble chemical oxygen in the
 Y. Li et al. / Energy 118 (2017) 377e386 383

Table 4(b)
Summarization of degradation kinetics of VFA during AD of KW using rst-order kinetics.

Item the 1st sectiona the 2nd section

Exponential kinetics model R2 Exponential kinetics model R2

KW St 2490exp(-0.8371RT)4266 0.9650 St 9607exp(-0.1912RT)712 0.9707

55
C 70 min St 2377exp(-0.9820RT)4132 0.9999 St 8288exp(-0.1691RT)597 0.9667
70
C 70 min St 2630exp(-0.4424RT)4359 0.9802 St 19233exp(-0.3721RT)1026 0.9681
90
C 70 min St 2771exp(-0.2198RT)4472 0.9340 St 33061exp(-0.5242RT)1090 0.9300
120
C 50 min St 503RT1847 0.9703 St 18848exp(-0.7236RT)1161 0.9034
140
C 50 min St 484RT1944 0.8834 St 30821exp(-0.8838RT)1141 0.8073
160
C 50 min St 675RT2217 0.8240 St 10971exp(-0.4286RT)985 0.9610
a
The rst section for untreated KW and KW pretreated at 55e90
C began from to the initial 5 days, compared with the initial 3 days for KW pretreated at 120e160
C. St:
VFA concentration in the digester, mg/L.

liquid phase of KW. Besides, KW subjected to thermal pretreat-
ment, composed of easily bioaccessible and readily biodegradable
water soluble fractions, are expected to be hydrolyzed in a shorter
time [20]. As a result, during the initial stage of acidication phase,
VFA began to be generated.
The kinetics results of VFA also indicated that the degradability
of various organics in KW, such as carbohydrates, FOG and proteins
varied due to different thermal pretreatment temperatures and
treatment durations, which had a major impact on the AD perfor-
mance of KW. In addition, compared with sewage sludge, KW was
much easier to be hydrolyzed, so during the early stage of AD, VFA
production increased rapidly and was then degraded.
kh values are inversely proportional to the characteristic size of
the particle for different particle shape [19]. Compared with un-
treated KW, a reduction of particle size for substrates subjected to
thermal pretreatment could be achieved, and more refractory
organic matters with large molecules will be solubilized to soluble
and small ones, which are much available to microorganisms and
easier to be degraded. Thus, overall higher kh values could be ob-
tained compared with that found for untreated KW during AD
process [42].
3.4. Relationship between organic degradation and biogas
production
3.4.1. Biogas yield and organic reduction
Both the removal efciency of organic matters (including FOG,
protein, and VS) and biogas yield are two important indicators of
the AD process. The relationship between biogas production and
degradation of organics was examined and the results were pre-
sented in Table 5. Due to varied effects of thermal pretreatment on
the organic degradation of KW, different relationships were
obtained.
As to FOG and CP, the cumulative biogas production increased
exponentially with their removal efciency, while the cumulative
biogas production increased linearly with the removal efciency of
VS. The positive correlation between biogas yield and VS reduction
indicated that VS in KW was converted to the biogas steadily during
the whole digestion process. Basing on results in Table 2 and
Table 3, the maximum degradation rates of CP is about 2e4 times
as much as those of FOG (Table 2 and Table 3). As to the conversion
rate of protein and FOG to biogas, although the theoretical biogas
yield from protein is 30% lower than FOG, relatively more biogas
could be obtained for the same degradation amount of CP
compared with FOG (in Table 5). It could be speculated that FOG
was much easier to be degraded and more FOG would be involved
in the bioconversion to biogas compared with CP during AD process
of KW. The highest conversion rates of FOG and CP were achieved
for KW pretreated at 90 and 120
C respectively. In addition, over
the two pretreatment temperatures mentioned above, as the
thermal pretreatment temperature increased, a relatively higher
conversion rates could be obtained while opposite results were
obtained after them. As for VS reduction, the cumulative biogas
production increased linearly with VS removal efciency which
indicated that VS was converted to the biogas steadily during the
whole digestion process.
These relationships in Table 5 indicated that the degradation of
protein and FOG had a major impact but different patterns on KW
biogas production. Thermal pretreatment could improve the sub-
sequent AD efciency, but the effects of thermal pretreatment on
subsequent AD process differed depending on the treatment pa-
rameters, which could lead to variations compositions of CP and
FOG in the KW placed into anaerobic digesters.
In addition, as shown in Fig. 4 (a) and (b), in the initial reduction
ratio of both FOG and proteins, the biogas production was low in all
samples. After achieving about 50% and 40% degradation respec-
tively for FOG and proteins in KW, biogas production in all samples,
especially for the pretreated KW, was increased sharply. It could be
explained by the following two reasons. First, the microbes might
have completed the acclimatization to the surroundings. In

Table 5
The relationship between organics degradation (FOG, CP, and VS) and biogas production.

Item Biogas yield and FOG reductiona Biogas yield and CP reductionb Biogas yield and VS reductionc

Exponential kinetics model R2 Exponential kinetics model R2 First-order kinetics R2

KW y 0.5916exp(x/0.0932)36.8972 0.9417 y 7.1629exp(x/0.0239)39.2078 0.8994 y 1351x-200 0.9668

55
C 70 min y 9.9986exp(x/0.1874)23.7442 0.8990 y 0.0432exp(x/0.0504)56.2332 0.9794 y 1811x-425 0.8602
70
C 70 min y 6.3388exp(x/0.1560)25.5714 0.9615 y 5.8275E-10exp(x/0.01643)80.3680 0.8999 y 1336x-136 0.8117
90
C 70 min y 0.4523exp(x/0.0913)42.7943 0.9909 y 6.3943E-10exp(x/0.0163)50.7073 0.9833 y 1210x-168 0.8025
120
C 50 min y 0.8384exp(x/0.1044)42.0579 0.9721 y 9.5362E-14exp(x/0.0119)59.7037 0.9151 y 1293x-234 0.8635
140
C 50 min y 3.4039exp(x/0.1452)67.5639 0.9429 y 6.0389E-11exp(x/0.0165)74.4643 0.9355 y 2188x-263 0.8402
160
C 50 min y 2.0923exp(x/0.1178)82.7614 0.7614 y 1.2620E-08exp(x/0.0206)101.4990 0.9550 y 5794x-747 0.9538
a
y: Biogas yield, x: FOG reduction, (%).
b
y: Biogas yield, x: CP reduction (%).
c
y: Biogas yield, x: VS reduction (%).
384 Y. Li et al. / Energy 118 (2017) 377e386

Fig. 4. The relationship between FOG (a), protein (b) and VS (c) degradation and biogas yield.

addition, microorganisms responsible for the degradation of spe- Table 6

cic organic matters might have begun to expand faster and The relationship between FOG reduction and CP reduction during AD of KW using
rst-order kinetics.
dominate.
Item Fitting equationa R2

3.4.2. FOG reduction and CP reduction KW y 0.3570exp(-x/0.3520)0.4312 0.9880

As two main components in the organics in KW, the relationship 55
C 70 min y 0.5513exp(-x/0.3296)0.5180 0.9914
70
C 70 min y 0.725exp(-x/0.2057)0.4757 0.9632
between FOG reduction and CP reduction during AD of KW (Fig. 5)
90
C 70 min y 0.551exp(-x/0.2675)0.5016 0.9759
were studied and the results were shown in Table 6. It could be 120
C 50 min y 0.986exp(-x/0.1582)0.4435 0.9810
140
C 50 min y 0.486exp(-x/0.1192)0.4793 0.9353
160
C 50 min y 0.512exp(-x/0.4340)0.6263 0.9842
a
y: CP reduction (%); x: FOG reduction (%).

concluded that the CP degradation increased exponentially with

Fig. 5. The relationship between protein reduction and FOG reduction.
the removal efciency of FOG. In addition, more FOG could be
degraded compared with CP, and the reduction proportion ranges
were 4e65% and 6e49% respectively for FOG and CP. As for un-
treated KW, the degradation percentage of CP began to exceed that
of pretreated samples when more than 30% of FOG was degraded.
The CP degradation efciency for samples pretreated at 55e120
C
presented the similar tendency with the increase of FOG degrada-
tion efciency when FOG reduction ranged from 30 to 60%, about
7e37% lower compared with the CP reduction for samples pre-
treated at 140
C. However, CP reduction of KW pretreated at 160
C
remained stable within 47e48% during AD process after 40% of FOG
was degraded. It could be concluded that AD of KW is a complicated
chemical and physical process, and all kinds of its factors, such as
organic components, interact, thus creating a special relationship
 Y. Li et al. / Energy 118 (2017) 377e386
between the degradation of protein and FOG; besides, thermal
pretreatment also had a major impact on KW anaerobic digestion
performance.
3.4.3. Possible strategies for improving AD of KW
The FOG content is usually very high in KW and is an essential
component. Adding FOG to a digester could enhance the produc-
tion of methane, while FOG could also cause operational problems
due to inhibition effects, such as impeding the mass transfer and
loss of active biomass [43]. Besides, the proteins in KW (with a
proportion range of 11e28%, dry basis) could also provide a nitro-
gen source for microorganisms beyond inhibition during anaerobic
digestion. Hence, both thermal pretreatment and the protein and
FOG could affect the biogas production, organics removal, and
operation stability during the anaerobic digestion process. Basing
on the thermal pretreatment parameters (including temperatures
and treatment durations prior to subsequent digestion) and two
organic component concentrations during the digestion process,
the theoretical biogas yield, organic reduction, and corresponding
appropriate digestion retention could be calculated and predicted.
In addition, it could also be concluded that continuing to
enhance the solubilization of organics in KW, such as proteins and
FOG, from thermal pretreatment by increasing the treatment
temperatures and durations could not lead to signicantly and
continuously increased biogas yield and organics reduction during
AD process. It has been reported that several pretreatment condi-
tions could lead to the production of refractory compounds or
inhibitory intermediates that could reduce the biodegradability
potential [13,44], for example, reduction of methanogenic inhibi-
tion due to the production of ammonia for wastewater sludge when
pretreated at lower than 150
C [44]. On the other hand, pretreat-
ment at higher treatment temperatures requires more energy
consumption. Hence, as for the pretreatment temperature of KW,
no exceeding 140
C is suggested.
4. Conclusions
In order to gain insight into organics degradation and reduction
during the anaerobic digestion process, kinetics models were
applied in this study to assess the effects of the different pre-
treatment parameters on the biogas yield, protein, FOG, VS and VFA
via data tting. It was shown that organics could be efciently
degraded when thermal pretreatment was applied prior to anaer-
obic digestion. Biogas yield increased exponentially and linearly
with removal efciency of both CP and FOG, organics, respectively.
Biogas yield, organic reduction and corresponding retention dura-
tion based on the CP and FOG removal efciency were proposed.
Acknowledgements
The authors acknowledge the nancial support by JH Green
Energy Co., Ltd, the Key Laboratory for Solid Waste Management
and Environment Safety and BMEI Co., Ltd.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.energy.2016.12.041.
References
[1] Liu Y, Li X, Kang X. Effect of volume ratio on anaerobic co-digestion of thermal
hydrolysis of food waste with activated sludge. Int BIODETER BIODEGR
2015;102(SI):154e8.
[2] Grimberg SJ, Hilderbrandt D, Kinnunen M, Rogers S. Anaerobic digestion of
food waste through the operation of a mesophilic two-phase pilot scale
385
digester - assessment of variable loadings on system performance. Bioresour
Technol 2015;178:226e9.
[3] Kiran EU, Trzcinski AP, Wun JN, Liu Y. Bioconversion of food waste to energy: a
review. FUEL 2014;134:389e99.
[4] Meng Y, Li S, Yuan H, Zou D, Liu Y, Zhu B, et al. Evaluating biomethane pro-
duction from anaerobic mono- and co-digestion of food waste and oatable
oil (FO) skimmed from food waste. Bioresour Technol 2015;185:7e13.
[5] Li K, Liu R, Sun C. Comparison of anaerobic digestion characteristics and ki-
netics of four livestock manures with different substrate concentrations.
Bioresour Technol 2015;198:133e40.
[6] Deepanraj B, Sivasubramanian V, Jayaraj S. Experimental and kinetic study on
anaerobic digestion of food waste: the effect of total solids and pH. J Renew
SUSTAIN ENER 2015:7 (0631046).
[7] Carlsson M, Lagerkvist A, Morgan-Sagastume F. The effects of substrate pre-
treatment on anaerobic digestion systems: a review. Waste manage
2012;32(9):1634e50.
[8] Ravindran R, Jaiswal AK. A comprehensive review on pre-treatment strategy
for lignocellulosic food industry waste: challenges and opportunities. Bio-
resour Technol 2016;199:92e102.
[9] Carrere H, Dumas C, Battimelli A, Batstone DJ, Delgenes JP, Steyer JP, et al.
Pretreatment methods to improve sludge anaerobic degradability: a review.
J Hazard Mater 2010;183(1e3):1e15.
[10] Ma J, Duong TH, Smits M, Verstraete W, Carballa M. Enhanced biomethanation
of kitchen waste by different pre-treatments. Bioresour Technol 2011;102(2):
592e9.
[11] Kavitha S, Yukesh Kannah R, Yeom IT, Do K, Banu JR. Combined thermo-
chemo-sonic disintegration of waste activated sludge for biogas production.
Bioresour Technol 2015;197:383e92.
[12] Kavitha S, Jayashree C, Adish Kumar S, Kaliappan S, Rajesh Banu J. Enhancing
the functional and economical efciency of a novel combined thermo
chemical disperser disintegration of waste activated sludge for biogas pro-
duction. Bioresour Technol 2014;173:32e41.
[13] Kavitha S, Rajesh Banu J, Subitha G, Ushani U, Yeom IT. Impact of thermo-
chemo-sonic pretreatment in solubilizing waste activated sludge for biogas
production: energetic analysis and economic assessment. Bioresour Technol
2016;219:479e86.
[14] Latinwo GK, Agarry SE. Modelling the kinetics of biogas production from
mesophilic anaerobic Co-Digestion of cow dung with plantain peels. Int J
Renew ENERGY DEVELOPMENT-IJRED 2015;4(1):55e63.
[15] Ma J, Zhao Q, Laurens LLM, Jarvis EE, Nagle NJ, Chen S, et al. Mechanism, ki-
netics and microbiology of inhibition caused by long-chain fatty acids in
anaerobic digestion of algal biomass. Biotechnol BIOFUELS 2015;(141):8.
[16] Wang B, Stromberg S, Li C, Nges IA, Nistor M, Deng L, et al. Effects of substrate
concentration on methane potential and degradation kinetics in batch
anaerobic digestion. Bioresour Technol 2015;194:240e6.
[17] Zwietering MH, Jongenburger I, Rombouts FM, Vantriet K. Modeling of the
bacterial-growth curve. Appl Environ Microb 1990;56(6):1875e81.
[18] Koch K, Drewes JE. Alternative approach to estimate the hydrolysis rate
constant of particulate material from batch data. Appl ENERG 2014;120:11e5.
[19] Vavilin VA, Fernandez B, Palatsi J, Flotats X. Hydrolysis kinetics in anaerobic
degradation of particulate organic material: an overview. Waste Manage
2008;28(6):939e51.
[20] Panico A, D'Antonio G, Esposito G, Frunzo L, Iodice P, Pirozzi F. The effect of
substrate-bulk interaction on hydrolysis modeling in anaerobic digestion
process. SUSTAINABILITY-BASEL 2014;6(12):8348e63.
[21] Shana A, Ouki S, Asaadi M, Pearce P, Mancini G. The impact of intermediate
thermal hydrolysis on the degradation kinetics of carbohydrates in sewage
sludge. Bioresour Technol 2013;137:239e44.
[22] Jin Y, Li Y, Li J. Inuence of thermal pretreatment on physical and chemical
properties of kitchen waste and the efciency of anaerobic digestion.
J Environ MANAGE 2016;180:291e300.
[23] Federation WE, American PHA. Standard methods for the examination of
water and wastewater. Washington, DC, USA: American Public Health Asso-
ciation (APHA); 2005.
[24] Methodenbuch, Band III. Die chemische Untersuchung von Futtermitteln.
Erganzungslieferung. Darmstadt: VDLUFA, Verlag; 1993.
[25] AOAC. In: Ofcial methods of analysis. fteenth ed. Washington, DC: Associ-
ation of Ofcial Analytical Chemists; 1990.
[26] Lee J, Koo T, Han G, Shin SG, Hwang S. Anaerobic digestion of cattle offal:
protein and lipid-rich substrate degradation and population dynamics of
acidogens and methanogens. BIOPROC Biosyst Eng 2015;38(12):2349e60.
[27] Yang G, Zhang P, Zhang G, Wang Y, Yang A. Degradation properties of protein
and carbohydrate during sludge anaerobic digestion. Bioresour Technol
2015;192:126e30.
[28] Garc-A-Heras J. Reactor Sizing, Process Kinetics and Modelling of Anaerobic
Digestion of Complex Wastes. J. Mata Alvarez (Ed.), Biomethanization of
Municipal Solid Waste ed.: J. Mata-Alvarez (Ed.), Biomethanization of the
Organic Fraction of Municipal Solid Wastes, IWA Publishing, London, 2004.
[29] Vavilin VA, Rytov SV, Lokshina LY. A description of hydrolysis kinetics in
anaerobic degradation of particulate organic matter. Bioresour Technol
1996;56(2e3):229e37.
[30] Breure AM, Mooijman KA, Vanandel JG. Protein-degradation in anaerobic-
digestion - inuence of volatile fatty-acids and carbohydrates on hydrolysis
and acidogenic fermentation of gelatin. Appl Microbiol Biot 1986;24(5):
426e31.
386 Y. Li et al. / Energy 118 (2017) 377e386
[31] Cirne DG, Paloumet X, Bjornsson L, Alves MM, Mattiasson B. Anaerobic
digestion of lipid-rich waste - effects of lipid concentration. Renew ENERG
2007;32(6):965e75.
[32] Sun Y, Wang D, Yan J, Qiao W, Wang W, Zhu T. Effects of lipid concentration
on anaerobic co-digestion of municipal biomass wastes. Waste Manage
2014;34(6SI):1025e34.
[33] Ariunbaatar J, Panico A, Frunzo L, Esposito G, Lens PNL, Pirozzi F. Enhanced
anaerobic digestion of food waste by thermal and ozonation pretreatment
methods. J Environ MANAGE 2014;146:142e9.
[34] Li Y, Jin Y, Li J, Li H, Yu Z. Effects of thermal pretreatment on the biomethane
yield and hydrolysis rate of kitchen waste. Appl ENERG 2016;172:45e58.
[35] Long JH, Aziz TN, de Los Reyes FLI, Ducoste JJ. Anaerobic co-digestion of fat, oil,
and grease (FOG): a review of gas production and process limitations. PRO-
CESS Saf Environ 2012;90(3SI):231e45.
[36] Novak JT, Carlson DA. The kinetics of anaerobic long chain fatty acid degra-
dation. J Water Pollut Control Fed 1970;42(11):1932e43.
[37] Roy F, Samain E, Dubourguier HC, Albagnac G. Synthrophomonas-sapovorans
sp-nov, a new obligately proton reducing anaerobe oxidizing saturated and
unsaturated long-chain fatty-acids. Arch Microbiol 1986;145(2):142e7.
[38] Lalman JA, Bagley DM. Anaerobic degradation and methanogenic inhibitory
effects of oleic and stearic acids. WATER Res 2001;35(12):2975e83.
[39] Lalman J, Bagley DM. Effects of C18 long chain fatty acids on glucose, butyrate
and hydrogen degradation. WATER Res 2002:36 (PII S0043-1354(02)00014-
313):3307-13.
[40] Salminen E, Rintala J. Anaerobic digestion of organic solid poultry slaughter-
house waste - a review. Bioresour Technol 2002:83 (PII S0960-8524(01)
00199-71):13-26.
[41] Pereira MA, Sousa DZ, Mota M, Alves MM. Mineralization of LCFA associated
with anaerobic sludge: kinetics, enhancement of methanogenic activity, and
effect of VFA. Biotechnol Bioeng 2004;88(4):502e11.
[42] Prorot A, Julien L, Christophe D, Patrick L. Sludge disintegration during heat
treatment at low temperature: a better understanding of involved mecha-
nisms with a multiparametric approach. BIOCHEM Eng J 2011;54(3):178e84.
[43] Rasit N, Idris A, Harun R, Ghani WAWA. Effects of lipid inhibition on biogas
production of anaerobic digestion from oily efuents and sludges: an over-
view. Renew Sust ENERG Rev 2015;45:351e8.
[44] Wilson CA, Novak JT. Hydrolysis of macromolecular components of primary
and secondary wastewater sludge by thermal hydrolytic pretreatment. WA-
TER Res 2009;43(18):4489e98.