Bioresource Technology 131 (2013) 452459

Contents lists available at SciVerse ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Anaerobic co-digestion of grease sludge and sewage sludge: The effect

of organic loading and grease sludge content
C. Noutsopoulos , D. Mamais, K. Antoniou, C. Avramides, P. Oikonomopoulos, I. Fountoulakis
National Technical University of Athens, School of Civil Engineering, Department of Water Resources and Environmental Engineering, 5 Iroon Polytechniou,
Zografou, Athens 15780, Greece

h i g h l i g h t s

" Grease sludge is a suitable co-substrate for sewage sludge anaerobic digestion.
" Co-digestion of GS and SS increases biogas yield up to 55%.
" Biogas yield of co-digestion systems depends on both GS content and OLR.
" The limit GS organic loading which provides for stable operation is 2.4 g VSGS/L/d.
" Anaerobic co-digestion of GS and SS may increase the energy recovery of an WWTP.

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

Article history: The objective of this study was to assess the feasibility of co-digesting grease sludge (GS) originating from
Received 18 October 2012 domestic wastewater along with sewage sludge (SS) and to assess the effect of organic loading rate (OLR)
Received in revised form 23 December 2012 and GS content on process performance. Three lab-scale semi-continuous fed mesophilic anaerobic
Accepted 26 December 2012
digesters were operated under various OLRs and SSGS mixtures. According to the results, addition of
Available online 11 January 2013
GS up to 60% of the total VS load of feed resulted in a 55% increase of biogas yield (700 vs. 452 m3/
tVSadded) for an OLR of 3.5 kg VS/m3/d. A stable and satisfactory operation of anaerobic co-digestion units
Keywords:
can be achieved for a GS-OLR up to 2.4 kg VSGS/m3/d. For such values biogas yield is linearly proportional
Anaerobic co-digestion
Lipids
to the applied GS-OLR, whereas biogas yield is minimal for GS-OLR higher than this limit and acidication
Grease sludge of the anaerobic digestion units is taking place.
Organic loading 2013 Elsevier Ltd. All rights reserved.
Biogas production

1. Introduction various operational problems like biological bulking and foaming,

Lipids (fats, oils, grease) constitute a major part of domestic
wastewater organic matter which accounts for the 2540% of the
total COD of the raw wastewater (Quemeneur and Marty, 1994;
Chipasa and Medrzycka, 2006). The main sources of lipids in
domestic wastewater are originated from food activities (kitchen
waters) and human feces, which account for the 1436% and
423% of the total lipids content respectively (Quemeneur and
Marty, 1994). The major fraction of lipids found in wastewater
are triglycerides and a smaller portion is present as free long chain
fatty acids (LCFA). The transfer of signicant lipids quantities
through biological treatment units is very often associated with
Abbreviations: GS, grease sludge; SS, sewage sludge; OLR, organic loading rate;
PS, primary sludge; WAS, waste activated sludge.
Corresponding author. Tel.: +30 210 7722900; fax: +30 210 7722899.
E-mail address: cnoutso@central.ntua.gr (C. Noutsopoulos).
oc otation, oxygen mass-transfer difculties, odors, or even in-
crease of efuent concentration of organic matter (Noutsopoulos
et al., 2007).
Lipids (especially those in a suspended form) can readily be re-
moved from wastewater by physical methods. Several methods
(i.e. trapping, interception, use of skimmers, air otation) have
been employed in order to prevent lipids passing through the bio-
logical treatment units, achieving signicant lipids removal (to the
order of 5090%). The remaining fraction of lipids is readily re-
moved in the biological treatment units. The common manage-
ment practices adopted for the disposal of the accumulated lipids
in the form of grease sludge (GS) are landlling and incineration.
Alternatively due to its high free fatty acids content GS is a suitable
source for biodiesel production through the reaction of free fatty
acids with an alcohol (usually methanol) to form an alkyl ester
(Jolis et al., 2010; Montefrio et al., 2010). However appropriate
pretreatment of GS is required enabling an acid catalyzed process

0960-8524/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biortech.2012.12.193
 C. Noutsopoulos et al. / Bioresource Technology 131 (2013) 452459
followed by an alkaline catalyzed transesterication process (Long
et al., 2012). Another GS alternative management option might be
composting in order to decrease methane production trough land-
lling (Lemus et al., 2004; Long et al., 2012). Given that landlling
is prohibited by the continuously stricter environmental legislation
regarding the management of such biodegradable wastes (31/
1991/EU), treatment of GS seems to be an unavoidable practice.
Although effective for achieving high removal efciency (Chi-
pasa and Medrzycka, 2006), the aerobic treatment of lipids does
not seem to be a cost efcient method primarily due to the high
oxygen demand and the corresponding high energy consumption.
However, due to their signicant methane yield (1 m3 CH4/kgVS)
when compared to carbohydrates and proteins, lipids are consid-
ered to be a promising substrate for anaerobic treatment and a po-
tential energy source (Hanaki et al., 1981; Kim et al., 2004; Alves et
al., 2009). Several biochemical pathways are related to lipids deg-
radation under anaerobic conditions. Triglycerides are rstly
hydrolysed into free LCFA and glycerol; a process which is cata-
lyzed by extracellular enzymes called lipases. After hydrolysis
the majority of lipids methane potential (more than 90%) is con-
served in LCFA. Degradation of free LCFAs and glycerol is taking
place intracellularly. Glycerol is further degraded mainly to acetate
by acidogenic bacteria, whereas LCFA are transformed to acetate
(or propionate), hydrogen and CO2 via b-oxidation biochemical
pathway (syntrophic acetogenesis). The last step during anaerobic
digestion is methanogenesis (hydrogenotrophic and acetotrophic).
Besides their high methane potential, lipids and lipids-rich
wastes are not commonly used as a sole substrate in anaerobic
digesters due to their inhibition effect to anaerobic biocenosis,
along with the development of other operational problems like
clogging, foaming and biomass otation (Pereira et al., 2004). The
inhibitory problems of lipids are mainly related to LCFA. The main
mechanisms of LCFA toxicity is through their adsorption onto the
cell wall of microorganisms thus inhibiting transport phenomena
(Hwu et al., 1998; Alves et al., 2001; Cirne et al., 2007) and the
acute toxicity on microbial activity of both aceticlastic and hydro-
genotrophic methanogens (Angelidaki and Ahring, 1992; Rinzema
et al., 1994). Recently it was suggested that inhibition is a revers-
ible phenomenon provided that specic recovery practices will
be applied (Cavaleiro et al., 2008; Palatsi et al., 2009).
Anaerobic digestion (AD) of sewage sludge (SS) is a well known
technology resulting in energy generation through biogas produc-
tion and sludge stabilization and sanitization. Several strategies
have been proposed to improve biogas yield in SS-AD units. Among
these, co-digestion of SS along with other organic substrates which
present high methane potential have also be proposed as an inter-
esting option, due to the improvement of nutrient balance and the
positive synergisms established in the digesters, allowing, beyond
others, the dilution of any inhibitory substances contained in SS.
Besides these merits of co-digestion, a number of issues have been
raised regarding potential operational problems of applying such
an option, like process inhibition due to increase content of LCFA
(Hanaki et al., 1981; Rinzema et al., 1994; Shin et al., 2003), sub-
strate transport limitations (Pereira et al., 2004) and digester foam-
ing (Kabouris et al., 2008), with the latter not being fully evidenced
(Ganidi et al., 2009).
Since the anaerobic digestion with lipids as a sole substrate is
not a feasible practice due to the aforementioned inhibition phe-
nomena, the in situ digestion of SS (thickened primary and surplus
biological sludge) produced in a WWTP along with GS collected in
the same WWTP, seems an interesting approach.
There are several studies suggesting that an increased methane
yield can be achieved through co-digestion of SS with lipid-based
material (especially grease trap sludge). More specically Luostari-
nen et al. (2009) reported a methane yield increase to the order of
60% when grease from a meat processing plant was co-digested
453
with SS. According to Luostarinen et al. (2009) the co-digestion
was feasible for a maximum grease content equal to 46% on a VS
basis, whereas Wan et al. (2011) found an increase in methane
yield equal to 137% for a grease content equal to 64% on a VS basis.
Kabouris et al. (2009b) reported that co-digestion of SS along with
grease trap sludge from restaurants (at a 48% GS content on a VS
basis) produced 2.95 times higher methane yield. Davidsson
et al. (2008) reported high grease trap sludge methane yield in sin-
gle substrate anaerobic digestion batch tests (845928 Nm3 CH4/
tVSadded), but could not reach stable methane production in semi-
continuous ow anaerobic digestion systems. Furthermore the
authors ascertained an increase on methane yield of 927% when
1030% of grease trap sludge (on a VS basis) was digested along
with sewage sludge. Finally Silvestre et al. (2011) found an in-
crease in methane yield to the order of 138% when a GS content
of 23% on a VS basis was added to SS.
Based on these ndings it seems reasonable to expect that more
data are required in order to assess the optimal operating condi-
tions for the co-digestion process.
In view of the above the aim of the present study was to assess
the feasibility of co-digesting lipids originated from domestic
wastewater (in the form of GS) along with SS and to evaluate the
effect of the organic loading and GS content on the performance
of anaerobic co-digestion.
2. Methods
2.1. Inoculum and substrates
All experimental systems were inoculated with mesophilic di-
gested sludge from Psyttalia Wastewater Treatment Plant (PWTP).
PWTP is located in the greater Athens region and has a treatment
capacity of 3,500,000 population equivalent. The average wastewa-
ter ow entering PWTP was equal to 750.000 m3/d. Wastewater
treatment train consists of pretreatment, primary treatment and
biological treatment. Primary sludge (PS) and waste activated
sludge (WAS) after separate thickening (gravity thickening for PS
and mechanical thickening for WAS) are pumped to mesophilic
anaerobic digestion tanks at a feed ratio of 74% PS and 26% WAS
on a VS basis. Average owrates pumped to the anaerobic reactors
of PWTP were equal to 3000 and 1200 m3/d for PS and WAS respec-
tively. Average totals solids concentrations of thickened PS and
WAS were equal to 54.5 and 51.5 g TS/L respectively whereas the
average VS/TS ratios for PS and WAS were equal to 75% and 80%
respectively. The aforementioned mixture of PS and WAS thick-
ened sludge from PWTP was used as sewage sludge (SS) substrate
in the present study. Grease sludge (GS) originating from PWTP
was also used as a co-substrate in the experimental units. GS
was collected from the primary settling tanks of PWTP through
surface skimming. The average quantity of GS collected at PWTP
was equal to 60 g VS/m3 wastewater, whereas the quantities of
PS and WAS produced were equal to 170 g VS/m3 wastewater
and 70 g VS/m3 wastewater respectively. Based on laboratory
analyses GS exhibits a high organic content (VS/TS ratio equal to
90 2%) and adverse rheological characteristics. Total and volatile
solids content of GS was equal to 0.71 0.06 kg TS/kg wet GS and
0.64 0.05 kg VS/kg wet GS respectively. SS and GS were shipped
to the laboratory once a week and upon delivery they were
stored at 4 C after being analyzed for total and volatile solids
concentrations.
2.2. Semi-continuous anaerobic digestion systems
Three lab-scale single stage mesophilic anaerobic digesters
were operated under a constant hydraulic retention time (15 d)
454 C. Noutsopoulos et al. / Bioresource Technology 131 (2013) 452459

for various organic loadings and SSGS mixtures for a total period
of 260 d, divided in 3 phases. The selection of a hydraulic retention
time (HRT) of 15 d was based on the fact that most of the Greek
anaerobic digestion units are currently operated at a HRT of 15
18 d. Experiments were conducted in glass conical asks with an
operating volume of 3 L each. The reactors were constantly mixed
using magnetic stirrers and their temperature was controlled with
water recirculation using a heated water circulating bath. The tem-
perature at all reactors was maintained constant at 35 C. Feeding
of the experimental units and withdrawing was carried out once
per day. Inuent and efuent sludge characteristics were measured
twice a week and biogas production on a daily basis.
At the beginning of Phase 1, all systems were inoculated with
mesophilic digested sludge from PWTP. The digesters were initially
lled with 3 L of seed sludge and from the next day feeding was
taking place once a day. Throughout the experimental period 6
alternative experimental systems were operated. One system (C)
was fed on a daily basis with sewage sludge (SS) from PWTP and
served as the control system, while its operation was compared
with the other experimental units. The characteristics of inoculum
material and the feed mixtures of the six experimental units are
summarized in Table 1. All experimental units, with the exemption
of system C, received mixtures of SS and GS with different GS con-
tent (20% for system E20, 40% for system E40, 60% for systems E60
and E60+ and 90% for system E90, on a VS basis). Average organic
loadings of the six experimental units were equal to 1.92 kg VS/
m3/d for system C, 2 kg VS/m3/d for system E20, 3.1 kg VS/m3/d
for system E40, 3.5 kg VS/m3/d for system E60, 4.1 kg VS/m3/d
for system E60+ and 8.3 kg VS/m3/d for system E90. In order to
keep HRT constant at 15 d in all experimental units the daily feed
volume was xed to 200 mL. Based on the desired organic loading
rate, the volumes of SS and GS in the feed were determined every
day. When the total required daily volume of SS and GS was less
than 200 mL, water was added in order to keep HRT constant at
15 d. The average daily volume of each sludge in the feed of every
experimental unit is presented in Table 1. During Phase 1 which
lasted for a period of 150 d, systems C, E20, E60 and E90 were oper-
ated. In day 1 of Phase 1, three systems C, E60 and E90 commenced
operation. After 16 d of operation and due to the complete failure
in its operation, system E90 was changed to system E20 (through
the alteration of the feed) in order to evaluate co-digestion perfor-
mance at moderate OLR and GS content. During Phase 2, which
lasted for 77 d, operation of systems C and E60 continued and sys-
tem E60+ was set in operation (inoculated with acclimatized bio-
mass from system E60). Finally, during Phase 3, systems C, E40
(inoculated with biomass from system E60) and E60+ were oper-
ated. At the beginning of Phase 3 system E60+ was re-inoculated
with biomass from system E60 and addition of ferric chloride at
a concentration of 100 mg Fe/L was performed on a daily basis in
order to improve the efciency of co-digestion of unit E60+.
2.3. Analytical methods
Data collection for the continuous digestion runs commenced
after three times the operating residence time (with the exception
of system E90; its operation data are presented for comparison
with the other experimental systems) and inuent and efuent
sludge samples were analyzed for total and volatile solids (TS
and VS), total and soluble chemical oxygen demand (COD), volatile
fatty acids (VFA) and alkalinity according to Standard Methods
(APHA WEF AWWA, 1992). The foaming potential (FP) and
the foaming stability (FS) were measured according to the Alka
Seltzer test developed by Ho and Jenkins (1991), but owing to
the fact that Alka Seltzer tablets are commercially unavailable in
Greece, Panadol Extra tablets containing the same active sub-
stances (sodium bicarbonate and citric acid) were used instead.
Biogas was collected in inverted cylinders placed in a compart-
mentalized tank with acidied water and sludge biogas production
rates were measured by the water displacement method. Samples
were abstracted from the headspace of each reactor with a gas-tight
syringe (1 mL) and were placed in 60 mL serum bottles sealed with
Teon-faced butyl rubber stoppers. Methane was measured with a
Perkin-Elmer Autosystem XL gas chromatographer equipped with a
PLOT column GS-GasPro (30 m, 0.32 mm diameter, J&W, Folsom,
CA) and a ame ionization detector (FID), by injecting 50 lL of sam-
ple headspace with a gas-tight syringe (100 lL). Analyses were con-
ducted using an isothermal method (50 C) while injector and
detector were held at 220 and 250 C, respectively. Standards were
prepared by adding a known amount to serum bottles of 160 mL
that had the same ratio headspace/liquid as the reactor (6:10).
Detection limit for methane measurements was 2 lg/L and the rel-
ative standard error of this method is 3%.
3. Results and discussion
3.1. Feed characterization
Six lab scale semi-continuous anaerobic reactors operated at
mesophilic conditions for a total period of 260 d. One system (Sys-
tem C) was fed on a mixture of thickened PS and WAS (SS) and its

Table 1
Characteristics of the inoculum and feed substrates of the experimental systems.

Parameter Inoculum Control E20 E40 E60 E60+ E90

VS (kg/m3) 36 28.7 0.5 29.6 0.6 46.0 0.5% 53.2 0.7% 62 0.9 125
VS/TS (%) 62 75 79 81 85 85 89
Alkalinity (mg/L as CaCO3) 4,114 245 3,500 280 4,350 174 4,080 244 4,080 320 nm
Tot-COD (kg/m3) 53.1 5.1 55.6 4.4 109.7 8.7 118.9 8.3 116.9 4.7 nm
Sol-COD (kg/m3) 4.3 4.3 4.9 4.8 4.6 nm
FPa (mL) 332 320 567 270 535 nm
FSb (s) 572 533 540 142 613 nm
OLR (kgVS/m3/d) 1.9 0.04 2.0 0.05 3.1 0.05 3.5 0.06 4.1 0.08 8.3 0.33
OLR (kg COD/m3/d) 3.5 0.3 3.7 0.3 7.3 0.3 7.9 0.5 7.8 0.6 nm
Daily volume of SS in feed (mL) 128 90 106 81 91 90
Daily volume of GS in feed (mL) 0 3 7 11 13 33
Total daily volume of feed (mL)e 200 200 200 200 200 200
Type of substrate SSc 80% SS20% GSd 60% SS40% GS 40% SS60% GS 40% SS60% GS 10% SS90% GS

nm: not measured.

a
FP: foaming potential.
b
Foaming stability.
c
SS: sewage sludge.
d
Grease sludge.
e
When the total volume of SS and GS in the feed was lower than the total volume of 200 mL required to keep HRT constant at 15 d, water was supplemented to the feed.
 C. Noutsopoulos et al. / Bioresource Technology 131 (2013) 452459 455

operation was compared to the operation of the ve experimental
units which were fed on different mixtures of SS and GS as well as
different organic loadings (systems E20, E40, E60, E60+ and E90).
Feed composition for the six experimental units is presented in Ta-
ble 1. Due to the different GS content, feed characteristics of the
ve experimental units varied signicantly. More specically VS/
TS ratio of the feed to the ve experimental units varied between
75% and 89%, with the higher values corresponding to the systems
receiving the higher GS content. The COD/VS ratio was equal to
1.85 for the control unit (System C) and varied between 1.87 and
2.40 for the experimental units receiving different mixtures of SS
and GS. Furthermore almost 8.2% of the feed COD was in soluble
form for system C whereas the respective values for the experi-
mental units were 7.8%, 4.4%, 4% and 3.9% for E20, E40, E60 and
E60+ respectively. Colloidal COD accounted for almost 0.70.9% of
the total COD in the feed for all experimental systems. An interest-
ing observation is that the addition of high lipids quantities in the
feed of system E60 resulted in signicant lower foaming potential
and foaming stability values with respect to the ones of the other
experimental units.
3.2. Semi-continuous experiments
3.2.1. General performance of experimental units
Performance data of the anaerobic units are presented in Table
2. Based on experimental results it can be stated that the experi-
mental units E20, E40 and E60 exhibited stable operation with sat-
isfactory biogas production and VS and COD reduction, whereas
this was not the case for systems E60+ and E90. The VS reduction
was increased with the addition of GS in the feed of the experimen-
tal units and was equal to 52%, 59%, 55% and 59% for systems C,
E20, E40 and E60 respectively. The increase of VS reduction follow-
ing addition of GS should be attributed to the higher biodegradabil-
ity of GS and is in good agreement with the results of other studies
which were performed in semi-continuous mesophilic anaerobic
co-digestion units receiving different GS content (Davidsson
et al., 2008; Luostarinen et al., 2009; Silvestre et al., 2011; Wan
et al., 2011). Reduction of total COD was equal to 59%, 60%, 55%
and 61% for systems C, E20, E40 and E60 respectively, whereas sol-
uble COD reduction varied between 80% and 90% for both system C
and experimental systems E20, E40 and E60 (soluble COD concen-
trations in all experimental units were lower than 1 g/L). Colloidal
COD accounted for 0.60.7% of the total COD for systems C, E20 and
E40, whereas the corresponding fraction for system E60 was equal
to 0.1%. On the contrary, system E60+ exhibited poor VS reduction
to the order of 24% accompanied with low COD reduction (to the
order of 43%) and low soluble COD reduction (to the order of
33%). NH3-N concentrations were moderate and varied between
580 and 1000 mg/L for systems C, E20, E40, E60 and E60+.
Steady state pH values for the experimental units C, E20, E40
and E60 were almost constant, equal to 7.27.3, whereas the corre-
sponding values for systems E60+ and E90 were equal to 6.9 and
5.9. It should be noted that in order to keep pH values around
6.9 for system E60+, base addition on a weekly basis was required.
In the case of system E90, despite of everyday base addition, pH
could not be controlled at values higher than 6.0.
VFA concentrations were comparably high for all anaerobic
units throughout the experimental period and increased with the
increase of the GS content in the feed of the experimental units.
More specically VFA concentrations for systems C, E20, E40,
E60, E60+ and E90 were equal to 651, 664, 711, 1090, 3170 and
7000 mg/L respectively. With the exception of systems E60+ and
E90 and despite the appreciable VFA concentrations (especially
for system E60), biogas production of systems C, E20, E40 and
E60 was satisfactory. This can be explained by the relatively mod-
erate values of VFA:alkalinity ratio which were recorded through-
out the experimental period (lower than 0.2) for all experimental
units. On the other hand average VFA:alkalinity ratio for system
E60+ was equal to 0.5, an indication of units failure to perform
stable operation. It is well known that the operation of anaerobic
digesters with VFA:alkalinity ratios greater than 0.5, indicates an
unfavorable balance between acidogenic and methanogenic micro-
organisms thus producing an acidifying digester.
Systems E20 and E60 exhibited similar foaming potential and
foaming stability values with those of the control system, but this
was not the case for system E40 and E60+. Furthermore, anaerobic
digestion resulted in a moderate decrease in foaming potential for
systems C, E20, E40 and E60+ to the order of 17%, 27%, 20% and 10%
respectively and a signicant decrease in foaming stability to the
order of 70%, 57%, 72% and 28% respectively. Interestingly enough,
this was not the case for system E60 where an average decrease of
foaming potential to the order of 43% was recorded. Based on these
results it is anticipated that the addition of GS to an anaerobic
digester treating sewage sludge should not, normally, produce
signicant foaming problems to the digester, although admittedly
pilot or full scale studies are required to examine any possible
foaming effect.
Based on the experimental results (Table 2) it was also observed
that VS concentrations of system E20 were practically lower than
those of the control system. By taking into account that the two

Table 2
Experimental results of the anaerobic digestion units.

Parameter Control E20 E40 E60 E60+ E90

VS (kg/m3) 13.9 0.3 12.2 0.2 21.3 0.4 21.8 0.4 47.0 0.9 22.8 0.9
VS/TS (%) 61.5 56.7 61.9 69.0 76.3 80.5
VFA (mg/L) 651 95 664 83 711 50 1,090 95 3,170 250 7,000
Alkalinity (kg/m3 as CaCO3) 6.2 0.4 5.4 0.5 6.3 0.25 5.75 0.3 6.3 0.4 nm
VFA/Alkalinity 0.10 0.12 0.11 0.19 0.50 nm
NH3-N (mg/L) 580 600 610 750 1000 nm
pH 7.3 0.01 7.2 0.01 7.2 0.01 7.2 0.01 6.9 0.05a 5.9 0.02a
FP (mL) 277 232 458 156 482 nm
FS (s) 176 232 155 139 444 nm
VS reduction (%) 52 59 55 59 24 81
Tot-COD reduction (%) 59 60 55 61 43 nm
Biogas yield (m3/tVSadded)b 452 23 511 28 660 33 700 35 50 18 10
Biogas yield (m3/tVSreduced)b 870 45 870 48 1120 55 1120 58 208 70 15
Biogas yield (m3/tCODadded)b 240 18 270 21 290 23 360 25 30 2 nm
CH4 (%) 65 66 70 70 nm nm

nm: not measured.

a
Following base addition.
b
Biogas production is given at NTP.
456 C. Noutsopoulos et al. / Bioresource Technology 131 (2013) 452459

experimental systems were practically operated at the same OLR, it
is anticipated that no additional sludge is produced when operat-
ing an anaerobic co-digestion unit compared to an anaerobic unit
treating SS as the sole substrate.
3.2.2. Effect of OLR and grease sludge content on biogas yield
Fig. 1 presents the evolution of biogas production of the exper-
imental units during Phases 2 and 3, after steady state conditions
were achieved. Details for the evolution of biogas production of
systems C and E20 during Phase 1 are presented elsewhere by
Noutsopoulos et al. (2012). Based on the data presented in Table
2 the average biogas yield of the control unit (system C) was equal
to 452 m3/tVSadded. This value is very close to the average biogas
yield of the anaerobic digesters of PWTP (420 m3/tVSadded) which
5
Biogas production rate
(Lbiogas/Lreactor/d)
were operated at similar conditions (OLR = 2.1 kg VS/m3/d, HRT =
17.7 d). In comparison with the results of other anaerobic co-diges-
tion studies (Table 3), specic biogas yield of the control unit (sys-
tem C) is very similar with the corresponding reported values by
Davidsson et al. (2008) and Luostarinen et al. (2009), whereas
much higher than that of Silvestre et al. (2011) and Wan et al.
(2011), most likely due to the higher applied OLR in the former
case and due to the use of a mixture of PS and WAS versus solely
WAS in the latter case.
Average biogas yield for the experimental units receiving differ-
ent mixtures of GS and SS were equal to 511, 660, 700 m3/tVSadded
for systems E20, E40 and E60 respectively, whereas biogas produc-
tion was practically minimal for systems E60+ and E90. Therefore it
is anticipated that the co-digestion of GS along with SS can lead,
under certain operational conditions, to a satisfactory increase in
biogas production in anaerobic digesters treating sewage sludge
as the sole substrate. More specically the co-digestion of SS and
GS at ratios of 2060% on a VS basis and under organic loading

4
rates up to 3.5 kg VS/m3/d resulted in increased biogas yield to
4 the order of 1355% (lower values correspond to lower GS content
3 and lower OLR and higher values correspond to higher GS content
3 and higher OLR). Furthermore the increase in biogas production of
2 Inoculation from the experimental units E20, E40 and E60 compared to the biogas
E60 & Fe addition production of the control system (system C) was equal to 17%,
2
1 228% and 285% (Fig. 1).
1 At the end of Phase 2 the biomass of system E60 was used to
inoculate both systems E40 and E60+ for Phase 3. As it is presented
0
45 55 65 75 85 95 105 in Fig. 1 changing the GS content from 60% to 40%, along with the
Time (d) decrease of OLR from 3.5 to 3.1 kg VS/m3/d, resulted in an immedi-
ate decrease in biogas production. However biogas yield of system
Control E40 E60 E60+
E40 was kept practically equal to that of systems E60 for a period
Fig. 1. Evolution of biogas production rate for experimental units C, E40, E60 and of an HRT (15 d). Thereafter a gradual decrease in biogas yield was
E60+. recorded and the average biogas yield was 6% lower than that of

Table 3
Comparison of results from anaerobic co-digestion research studies with mixtures of sewage sludge and grease sludge.

Study Feeda HRT (d) OLR (kg VS/m3/d) VS reduction (%) VFA (mg/L) Biogas yield (m3/tVSadded)
Davidsson et al. (2008) 100:0 13 2.5 45 na 417
90:10 13 2.5 55 na 447467
70:30 13 2.4 58 na 499
0:100 13 02.3 -d na -d
Luostarinen et al. (2009) 100:0 16 1.562.09 52 113 441
95:5 16 1.672.23 59 145 567
80:20 16 1.932.45 58 172 689
72:28 16 2.8 52 165 728
62:38 16 3.13 64 240 688
54:46 16 3.46 67 - 747
45:55 16 3.99 72 - 505
29:71 16 4.41 70 430 543
Kabouris et al. (2009b) 52.4:47.6b 12 4.35 45 75 680
52.447.6c 12 4.35 51.2 95 747
Silvestre et al. (2011) 100:0 20 1.5 36 <100 346
96:4 20 1.2 46 <100 397
77:23 20 1.6 52 <100 527
63:37 20 1.7 56 <100 488
Wan et al. (2011) 100:0 15 2.34 40 2317 387
36:64 15 2.34 57 2094 895
25:75 10 3.4 29.4 8394 -
Present study 100:0 15 1.9 52 651 452
80:20 15 2.9 59 664 511
60:40 15 3.1 55 711 660
40:60 15 3.5 59 1090 700
40:60 15 4.1 24 3170 50
10:90 15 8.3 -d 7000 10

na: not available.

a
Feed is calculated on a VS basis (SS:GS).
b
Mesophilic anaerobic reactor.
c
Thermophilic anaerobic reactor.
d
No stable operation was achieved.
 C. Noutsopoulos et al. / Bioresource Technology 131 (2013) 452459 457

systems E60. Based on these observations it is concluded that

Biogas yield (m3/tVSadded)

800 y = 124.25x + 466.83
adaptation of biomass to a lower GS content is a rather gradual 700 R2 = 0.9213
process. This was not the case, however, regarding biomass re-
600
sponse to an increase in OLR. More specically in order to help sys-
500
tem E60+ recovering from acidication (as evidenced by the
400
minimal biogas production, low pH values, high VFA concentra- limit GS-OLR=2.4 kgVSGS/m3/d
tions and VFA:alkalinity ratios), on day 75, the system was inocu- 300
lated with biomass from system E60, while at the same time 200
addition of Fe on a daily basis (100 mg/L) in the form of ferric chlo- 100
ride was performed. The addition of micronutrients is a well 0
known measure to improve biogas production in anaerobic digest- 0 1 2 3 4 5 6 7 8
ers and the stimulating effect of Fe addition in mesophilic anaero- Organic loading of GS (kgVSGS/m3/d)
bic digesters to decrease VFA concentrations and to improve biogas
production has been reported by Speece (1988). Nevertheless this Fig. 2. Biogas yield versus grease sludge organic loading rate (present study).
was not the case for the operation of system E60+ during the 3rd
phase. Thus, despite inoculation with biomass from a system
which was operated in a very satisfactory mode (E60) and Fe addi- 1000 Present study
tion, biogas production kept very close to zero after changing OLR

Biogas yield (m3/tVSadded)

900
from 3.5 to 4.1 kg VS/m3/d. Davidsson et al., (2008)
800
Moreover, following Fe addition, although average VFA concen- Luostarinen et al., (2009)
700
tration dropped from 3170 to 2390 mg/L, system E60+ still oper-
ated as an acidied anaerobic reactor, exhibiting the need for 600 Kabouris et al., (2009)
almost every day base addition to keep pH values to the order of 500
Silvestre et al., (2011)
6.9 and producing less biogas than prior to Fe addition (Fig. 1). 400
The failure of micronutrients addition as a measure to provide 300 Wan et al., (2011)
for stable anaerobic co-digestion at high GS content and to increase
200
biogas production is also reported by Wan et al. (2011). The
authors postulate that self recovery of the experimental unit was 100
achieved after 40 d of operation. However stopping the feeding of 0
0 1 2 3 4 5 6 7 8
an anaerobic digestion unit for such a long period does not seem
to be a practical measure to tackle the problem. Organic loading of GS (kgVSGS/m3/d)
The increase of biogas production through co-digestion of sew-
age sludge with GS is also reported by Davidsson et al. (2008), Fig. 3. Biogas yield versus grease sludge organic loading rate based on several
Luostarinen et al. (2009), Kabouris et al. (2009a,b), Silvestre et al. research studies.
(2011) and Wan et al. (2011). However according to Luostarinen
et al. (2009) an upper limit grease content of 46% in feed VS for a
maximum OLR equal to 3.46 kg VS/m3/d, was expected to achieve
Biogas yield (m3/tVSreduced)

1400
feasible co-digestion. According to the authors at higher GS content
1200
(to the order of 55% on a VS basis) degradation was incomplete and
biogas yield was decreased. On the other hand, Wan et al. (2011) 1000
postulate that co-digestion with a GS content of 64% on a VS basis
800
is feasible. Based on the present study, stable digestion operation
with increased biogas production can be achieved even at a GS 600
content of 60% on a VS basis for a maximum OLR of 3.5 kg VS/ 400
m3/d. By taking into account that the experimental units in the
200
three studies operated at almost identical conditions of tempera-
ture and HRT, these differences should be attributed to the applied 0
OLR. Luostarinen et al. (2009) employed an OLR equal to 4.0 kg VS/ E20 E40 E60
m3/d, compared to the OLR applied by Wan et al. (2011) and in the Measured semi-cont exps Calculated theoretical
present study for system E60 that were equal to 2.34 and 3.5 kg VS/
m3/d respectively. Therefore it is postulated that biogas yield de- Fig. 4. Comparison of biogas yield based on semi-continuous experiments with
pends not only on the GS content of the feed substrate but also theoretical biogas yield.

to the applied organic loading.

Fig. 2 presents the dependence of biogas yield on GS organic
loading rate (GS-OLR) expressed as kg VSGS/m3/d. As can be seen
from Fig. 2 stable and satisfactory operation of anaerobic co-diges-
tion units can be achieved for GS-OLR up to 2.4 kg VSGS/m3/d. For
such GS-OLR values biogas yield is linearly proportional to the ap-
plied GS-OLR, whereas biogas yield is minimal for GS-OLR higher
than this limit and acidication of the anaerobic digestion units
is taking place as evidenced by the low pH values, the high VFA
concentrations and the high VFA:alkalinity ratios. The dependence
of biogas yield on the applied GS-OLR for a series of anaerobic co-
digestion studies along with the results of the present study are
presented in Fig. 3. Based on these data the upper limit value of
2.4 kg VSGS/m3/d is satisfactorily veried, along with the linear
dependence of biogas yield on GS-OLR for values lower than the
limit value.
3.2.3. Comparison of measured and calculated theoretical biogas yield
Fig. 4 illustrates a comparison of the measured biogas yields of
the semi-continuous experiments and the calculated biogas yields
based on theoretical values for three co-digestion scenarios. Theo-
retical calculations were based on a lipids methane potential of
1014 Nm3 CH4/tVSremoved and a methane content in biogas equal
to 70% for lipids (Angelidaki and Sanders, 2004) and a sewage
sludge biogas yield of 900 Nm3 biogas/tVSremoved (Metcalf & Eddy,
458 C. Noutsopoulos et al. / Bioresource Technology 131 (2013) 452459
2003). As can be seen the measured biogas yields are very close to
the calculated theoretical estimations. Evidently measured biogas
yields for systems E20, E40 and E60 correspond to the 86%, 107%
and 96% of the theoretical values calculated for systems E20, E40
and E60 respectively.
3.3. Benets from co-digestion of GS and SS
A primary benet of co-digesting GS along with SS is that, fol-
lowing such a practice, GS is being diverted from any disposal
alternatives (landlling, incineration, composting) and is used for
biogas production by using existing infrastructures and expertise.
At Table 4 the results of two alternative GS management practices
are presented for a WWTP with a treatment capacity of 100,000
population equivalents (PE). The wastewater treatment train con-
sists of preliminary treatment (screening, grit removal), primary
treatment in sedimentation tanks and biological treatment (taking
place in an activated sludge unit) and sludge treatment consisting
of primary sludge gravity thickening, wastage sludge mechanical
thickening, anaerobic digestion of the mixture of PS and WAS
and a digested sludge dewatering unit.
In the context of the rst management practice (Scenario A), GS
is being collected from the surface skimmer of primary sedimenta-
tion tanks and then is being transported for landlling away from
the WWTP. Under Scenario B, the GS is used as a co-substrate,
being fed to the existing anaerobic digestion units along with the
mixture of PS and WAS. In order to compare the two alternative
management practices it is assumed that in both scenarios the pro-
duced biogas is being used in a combined heat and power energy
recovery system (CHP) consisting of internal combustion engines
used for generating electrical power and thermal energy with ef-
ciencies equal to 35% and 45% respectively. All the assumptions
made for the comparison of the two systems are shown below
(Metcalf and Eddy, 2003):
Influent grease concentration 100 mg=L:
Influent specific flow rate 200 L=PE=d:
Influent specific VS load 52:5 g VS=PE=d:
Average FOG removal through surface skimming in the primary
settling tanks = 65% (based on measurements in Greek WWTPs).
Average VS removal in primary settling tanks 60%:
Specific WAS production 40 g TS=PE=d:
MLVSS=MLSS 65%:
Table 4
Comparison of two alternative GS management scenarios (Scenario A: Collection,
accumulation and diversion of GS for landlling, Scenario B: Collection, accumulation
and use of GS as a co-substrate in anaerobic digestion unit along with sewage sludge).
Scenario A
GS (kgVS/d) 1300
PS (kgVS/d) 3150
WAS (kgVS/d) 2600
Inuent to AD (kgVS/d) 5750
Digested sludge (kgVS/d) 2760
Biogas yield (m3/d) 2600
Electrical Power produced (KWh/d) 5821
Thermal power produced (kcal/d) 6.44  106
Annual sludge management cost (euros/year) 120,900
Annual value of produced electrical power (euros/ 191,220
year)
Average VS reduction in anaerobic digestion unit = 52%
(Scenario A based on experimental results) and 59% (Scenario B
based on experimental results).
Average biogas yield = 452 m3/tVS added (Scenario A based on
experimental results) and 511 m3/tVSadded (Scenario B based on
experimental results).
Thermal content of biogas = 5,500 kcal/m3.
Average specic sludge handling cost = 120 euros/tn DS.
Energy cost = 0.09 euros/KWh.
Based on the values presented in Table 4 it is anticipated that
for a GS content of 18.5% on a VS basis the co-digestion option
leads to a 38.5% increase of biogas production, whereas the in-
crease of sludge quantity is marginal (to the order of 4.5%). Assum-
ing that conventional anaerobic digestion can recover around 30
40% of the overall energy requirements of a typical WWTP through
the use of energy recovery systems (WERF, 2010), one can con-
clude that by adopting the co-digestion practice, an increase in bio-
gas production to the order of 38.5% may be expected to lead to an
increase of energy recovery to the order of 4050%, thus satisfying
an appreciable portion of the thermal and energy requirements of
an WWTP. Another important benet of applying such a manage-
ment practice is the reduction of greenhouse gas emissions, due
to the reduced need for alternative fuel source and the avoidance
of GS management costs (landlling). Therefore it is anticipated
that co-digestion of GS along with SS is an environmentally sus-
tainable and economically reasonable GS management option.
The above analysis corresponds to a WWTP performing grease
removal in primary clariers. However the specic grease sludge
production depends on the applied grease removal system. More
specically grease removal from domestic wastewater to the order
of 50%, 65% and more than 90% can be achieved in grit-grease re-
moval units, primary clariers and dissolved air otation units
respectively (Einschlag, 2011). Therefore for an inuent grease
concentration of 100 mg/L and an inuent specic ow rate of
200 L/PE/d, average grease quantities removed are equal to 10,
13 and 18 g VS/PE/d for systems enabling grit grease removal
units, primary clariers and dissolved air otation units respec-
tively. Based on an average sewage sludge production of 50 g VS/
PE/d, the respective GS content in the case of applying co-digestion
for the three alternative grease removal options varies between
17% and 26% on a VS basis. In view of the above it is concluded that
anaerobic co-digestion systems using grease sludge collected on
site in the WWTP are likely to be subjected to a GS content to
the order of 20%. In order to maximize the positive effects of apply-
ing co-digestion with higher GS content other sources of grease
sludge may be used additionally (e.g. grease trap sludge from
restaurants).
4. Conclusions
Based on semi-continuous anaerobic digestion experiments, it
is concluded that co-digestion of GS and SS may lead to a signi-
cant increase of biogas yield. GS-OLR appears to be a critical
parameter in controlling co-digestion efciency. Biogas yield
improvement can be achieved for GS-OLR up to 2.4 kg VSGS/m3/d.
Scenario B For GS-OLR higher than this limit, unstable operation followed by
1300 minimal biogas yield should be expected. In domestic WWTPs, typ-
3150 ical GS content and GS-OLR in anaerobic co-digestion units in the
2600
7050
order of 20% and 0.6 kg VSGS/m3/d should be expected which can
2890 lead to the increase of biogas production in the order of 3540%.
3603
8066
8.91  106
References
126,580
264,970 Alves, M.M., Vieria, J.A.M., Pereira, R.M.A., Pereira, M.A., Mota, M., 2001. Effects of
lipids and oleic acid on biomass development in anaerobic xed-bed reactors.
Part II: Oleic acid toxicity and biodegradability. Water Res. 35 (1), 264270.
 C. Noutsopoulos et al. / Bioresource Technology 131 (2013) 452459
Alves, M.M., Pereira, M.A., Sousa, D.Z., Cavaleiro, A.J., Picavet, M., Smidt, H., Stams,
A.J.M., 2009. Waste lipids to energy: how to optimize methane production from
long chain fatty acids (LCFA) minireview. Microb. Biotechnol. 5, 538550.
Angelidaki, I., Ahring, B.K., 1992. Effects of free long chain fatty acids on
thermophilic anaerobic digestion. Appl. Microbiol. Biotechnol. 37, 808812.
Angelidaki, I., Sanders, W., 2004. Assessment of the anaerobic biodegradability of
macropollutants. Rev. Environ. Sci. Biotechnol. 3, 117129.
APHA WEF AWWA, 1992. Standard Methods for Water and Wastewater
Laboratory Analysis, 18th ed. Washington, D.C.
Cavaleiro, A.J., Pereira, M.A., Alves, M., 2008. Enhacement of methane production
from long chain fatty acid based efuents. Bioresour. Technol. 99, 40864095.
Chipasa, K.B., Medrzycka, K., 2006. Behavior of lipids in biological wastewater
treatment processes. J. Ind. Microbiol. Biotechnol. 33, 635645.
Cirne, D.G., Paloumet, X., Bjrnsson, L., Alves, M.M., Mattiasson, B., 2007. Anaerobic
digestion of lipid-rich waste effects of lipid concentration. Renew. Energy 32,
965975.
Davidsson, A., Lvstedt, C., Jansen, J., Gruvberger, C., Aspegren, H., 2008. Co-
digestion of grease trap sludge and sewage sludge. Waste Manage. 28, 986992.
Einschlag, F.S.G., 2011. Wastewater Evaluation and Management. InTech, Croatia.
Ganidi, N., Tyrrel, S., Cartmell, E., 2009. Anaerobic digestion foaming causes a
review. Bioresour. Technol. 100, 55465554.
Hanaki, K., Matsuo, T., Nagase, M., 1981. Mechanism of inhibition caused by long
chain fatty acids in anaerobic digestion process. Biotechnol. Bioeng. 23, 1591
1610.
Ho, C.F., Jenkins, D., 1991. The effect of surfactants on Nocardia foaming in activated
sludge. Water Sci. Technol. 23, 879888.
Hwu, C.S., van Lier, J.B., Lettinga, G., 1998. Physicochemical and biological
performance of expanded granular sludge bed reactors treating long chain
fatty acids. Process Biochem. 33, 7581.
Jolis, D., Loiacono, J., Kwan, L., Sierra, N., Ving, K., Martis, M., 2010. Co-location of
brown grease to biodiesel production facility at the oceanside wastewater
treatment plant in San Francisco, CA. In: Proceedings of the Water Environment
Federation, pp. 68166829.
Kabouris, J.C., Tezel, U., Pavlostathis, S.G., Engelmann, M., Todd, A.C., Gillette, R.A.,
2008. The anaerobic biodegradability of municipal sludge and fat, oil, and
grease at mesophilic conditions. Water Environ. Res. 80 (3), 212221.
Kabouris, J.C., Tezel, U., Pavlostathis, S.G., Englemann, M., Dulaney, J.A., Todd, A.C.,
Gillette, R.A., 2009a. Mesophilic and thermophilic anaerobic digestion of
municipal sludge and fat, oil, and grease. Water Environ. Res. 81 (5), 476485.
Kabouris, J.C., Tezel, U., Pavlostathis, S.G., Engelmann, M., Dulanay, J., Gillette, R.,
Tood, A.C., 2009b. Methane recovery from the anaerobic co-digestion of
municipal sludge and FOG. Bioresour. Technol. 100, 37013705.
Kim, H.W., Han, S.K., Shin, H.S., 2004. Anaerobic co-digestion of sewage sludge and
food waste using temperature-phased anaerobic digestion process. Water Sci.
Technol. 50 (9), 107114.
459
Lemus, G.R., Lau, A.K., Branion, R.M.R., Lo, K.V., 2004. Bench-scale study of the
biodegradation of grease trap sludge with yard trimmings or synthetic food
waste via composting. J. Environ. Eng. Sci. 3 (6), 485494.
Long, J.H., Aziz, T.N., de los Reyes III, F.L., Ducoste, J.J., 2012. Anaerobic co-digestion
of fat oil and grease (FOG): a review of gas production and process limitations.
Process Saf. Environ. 90, 231245.
Luostarinen, S., Luste, S., Sillanp, M., 2009. Increased biogas production at
wastewater treatment plants through co-digestion of sewage sludge with
grease trap sludge from a meat processing plant. Bioresour. Technol. 100, 79
85.
Metcalf & Eddy, 2003. Wastewater Engineering: Treatment and Reuse, fourth ed.
McGraw-Hill, New York.
Montefrio, M.J., Xinwen, T., Obbard, J.P., 2010. Recovery and pre-treatment of fat,
oils and grease from grease interceptors for biodiesel production. Appl. Energy
87 (10), 31553161.
Noutsopoulos, C., Andreadakis, A., Mamais, D., Gavalakis, E., 2007. Identication of
type and causes of lamentous bulking under Mediterranean conditions.
Environ. Technol. 28, 115122.
Noutsopoulos, C., Mamais, D., Antoniou, K., Avramides, C., 2012. Increase of biogas
production through co-digestion of lipids and sewage sludge. Global Nest J. 14
(2), 133140.
Palatsi, J., Laureni, M., Alves, M.V., Flotats, X., Nielsen, H.B., Angelidaki, I., 2009.
Strategies for recovering inhibition caused by long chain fatty acids on
anaerobic thermophilic biogas reactors. Bioresour. Technol. 100, 45884596.
Pereira, M.A., Sousa, D.Z., Mota, M., Alves, M.M., 2004. Mineralization of LCFA
Associated with anaerobic sludge: kinetics, enhancement of methanogenic
activity, and effect of VFA. Biotechnol. Bioeng. 88 (4), 502511.
Quemeneur, M., Marty, Y., 1994. Fatty acids and sterols in domestic wastewater.
Water Res. 28 (5), 12171226.
Rinzema, A., Boone, M., van Knippenberg, K., Letinga, G., 1994. Bactericidal effect of
long chain fatty acids in anaerobic digestion. Water Environ. Res. 66, 4049.
Silvestre, G., Rodriguez-Abalde, A., Fernadez, B., Flotats, X., Bonmati, A., 2011.
Biomass adaptation over anaerobic co-digestion of sewage sludge and trapped
grease waste. Bioresour. Technol. 102, 68306836.
Shin, H., Kim, S.H., Lee, C.Y., Nam, S.Y., 2003. Inhibitory effects of long-chain fatty
acids on VFA degradation and beta-oxidation. Water Sci. Technol. 47 (10), 139
146.
Speece, R.E., 1988. A survey of municipal anaerobic sludge digesters and diagnostic
activity assays. Water Res. 22 (3), 365372.
Wan, C., Zhou, Q., Fu, G., Li, Y., 2011. Semi-continuous anaerobic co-digestion of
thickened waste activated sludge and fat, oil and grease. Waste Manage. 31,
17521758.
WERF, 2010. Energy Efciency in Wastewater Treatment in North-America: A
Compendium of Best Practices and Case Studies of Novel Approaches. Water
Environment Research Foundation, Alexandria, VA.