Pilot-scale anaerobic digestion of screenings from wastewater treatment plants
Ronan Le Hyaric1*, Jean-Pierre Canler2, Bruno Barillon3, Pascale Naquin4 and Rémy
1 University of Lyon, INSA of Lyon, Lab. LGCIE, 20 av. A. Einstein, F-69621
Villeurbanne cedex, France
2 CEMAGREF of Lyon, 3 bis Quai Chaveau, CP 220, F-69336 Lyon cedex 09, France
3 CIRSEE – Suez Environnement, 1 rue d’Astorg, F-75008 Paris, France
4 Polden INSAVALOR, CEI, BP 52132, F-69603 Villeurbanne cedex, France
. Bioresource Technology, 101 (23), 9006-9011 (2010)
The anaerobic digestion of screenings from a municipal wastewater treatment plant was
studied in a 90 L pilot-scale digester operated at 35°C under semi-continuous
conditions. In the first four weeks, a dry solids residence time of 28 days was applied,
but the installation of inhibitory conditions was observed. Feeding was therefore
suspended for 4 weeks to allow the digester to recover from inhibition, and then
progressively increased up to a constant load of 6 kg of raw waste per week,
corresponding to an average residence time of about 35 days of dry solids. At this stage,
biogas production stabilized between 513 and 618 Nl/kgVSadded per week, with methane
contents around 61 % v/v. The results of this work thereby supported the feasibility of
(co-)digestion as a potential alternative treatment of screenings from municipal
wastewater treatment plants.
Keywords: Anaerobic digestion, screenings, sievings, wastewater.
Solid wastes such as screenings and greases are generated from the operations of
pretreatment of municipal wastewaters. Due to the relatively low production of
screenings in currently operated wastewater treatment plants as compared to sludge
production, little attention has been paid so far to this type of waste. Yet, increasing
production may be expected from the technological evolutions of municipal wastewater
treatment processes (membrane bioreactors MBR in particular) that require increasingly
fine screening (sieving) pretreatments (Frechen et al., 2006).
The most commonly used methods of disposal for screenings in Europe are currently
landfilling and incineration (Bode and Imhoff, 1996; Clay et al, 1996). Landfilling
however is not favoured by the European waste regulations (Council directive
1999/31/EC on waste landfilling) because of (i) the required reduction of the amounts of
biodegradable organic matter disposed into landfills and (ii) the ban for waste with
water content above 70% w/w to be landfilled, whereas screenings often exceed this
upper limit (Clay et al., 1996; Huber et al, 1995; Le Hyaric et al., 2009; Naud et al.,
2007). Incineration is considered as a good alternative, although the high water content
is also an unfavourable characteristic that may jeopardize the operating conditions of the
incineration plant in terms of combustion temperature and gaseous emissions (Bode and
Imhoff, 1996). Alternative treatments to those currently used may therefore prove to be
more adapted for screenings.
2. Materials and methods
2.1. Preliminary characterization of screenings from different WWTP
As a first step towards the development of an adequate treatment strategy for these
wastes, quantitative and qualitative characteristics of the production of screenings and
sievings (respectively collected from screens with gap sizes between 6 and 60 mm or
below 3 mm) were determined from three municipal wastewater treatment plants
(WWTPs) in the Region Rhône-Alpes (France). A total of 6 sampling campaigns,
conducted in accordance with the recommendations of the European standard NF EN
14899:2005 (Characterization of waste – Framework for the preparation and application
of a sampling plan), were done between May 2007 and November 2008, each with a
duration between 3 and 7 days.
The composition of the screenings sampled from each WWTP and each campaign,
determined according to the French standard NF X30-466: 2004 (Characterization of
municipal solid wastes - Analysis on dry product), was found to be relatively similar
(Le Hyaric et al., 2009). The predominant materials in the screenings were found to be
sanitary textiles which accounted for 54.7% to 72.9% of dry weight, along with the fine
fraction (< 20 mm) which accounted for 17.1% to 21.8% of dry weight (Table 1).
Table 2 shows that the sievings (from screens with a gap size of 3 mm or less) and the
predominant fractions of the screenings (screens with gap size from 6 to 60 mm)
exhibited a high volatile solid content which was found to be highly biodegradable as
shown by the methane potential.
Average BMP values obtained from predominant fractions of screenings (from 232 to
343 Nml CH4/gVS for sanitary textiles and from 241 to 439 Nml CH4/gVS for fine
fraction) and from sievings (from 289 to 377 NmlCH4/gVS) are equal or superior to
BMP values of wastes treated by digestion at the industrial level. Indeed, Owens and
Chynoweth (1993) and Jokela et al. (2005) reported municipal solid waste BMP values
were between 180 and 220 Nml CH4/gVS and typical methane production potential of
sewage sludge is approximately 300–400 Nml CH4/gVS (Davidsson et al., 2008;
Sosnowski et al. 2008).
Based on these observations, the present paper considers the anaerobic digestion of
screenings through the implementation of a pilot-scale study. Parameters such as biogas
production and composition were used to evaluate the performance of the treatment.
2.2. Description of experimental setup
A stainless steel cylindrical digester was used as shown in Fig. 1. The total volume of
the reactor was about 90 litres with a working volume of 50 litres. Mechanical stirring
was provided by the rotation of blades at 10.5 rpm around a horizontal axis powered by
a 250 W electric motor. The motor was automatically activated for 15 minutes every 4
The digester was equipped with an internal temperature sensor. It was heated from its
external surface using an electrical tape in order to maintain a constant temperature of
35 ± 1 °C inside the digester during its operation.
A 50 × 25 gastight window allowed the introduction and sampling of waste from the top
of the reactor. After each load/unloading operation, the digester was flushed with
nitrogen gas to remove oxygen.
The biogas production was measured continuously using a drum-type gas meter (range
of 0.1 l/h to 30.0 l/h). The gas meter was equipped with a pulse generator to transfer the
data to a computer. Biogas was regularly sampled for analysis through a sampling port
sealed with a septum (Fig. 1).
2.3. Analytical methods
Samples of waste were taken weekly from the digester and analyzed for dry solids
content (DS – overnight drying at 105°C until constant weight) and volatile solids
content (VS – open-air calcination at 550°C for 4 hours). Water extractions were done
in triplicates by mixing for 2 hours in 100 mL of deionised water a given mass of raw
waste corresponding to 100 g of dry solids (L/S ratio of 10). The suspension was then
filtered and analyzed for pH and volatile fatty acids (VFA) concentration. Standard
analytical methods were employed when applicable. Volatile fatty acids were analyzed
with an AgilentTM gas chromatograph equipped with a flame-ionization detector (CPG –
FID) and a HP-FFAP column (30 m × 0.25 mm).
Biogas was analyzed with an AgilentTM gas micro-chromatograph equipped with a
thermal conductivity detector (CPG – TCD), a Poraplot U column (8 m × 0.320 mm ID)
for CO2 and H2S separation and a Molsieve 5 Å column (10 m × PPU 3 m) for O2, N2,
and CH4 separation.
All analyses were done in triplicates.
2.4. Waste sampling and preparation
A mass of 300 kg of screenings was sampled from the municipal wastewater treatment
plant of Bourg-en-Bresse (Region Rhône-Alpes, France). The waste was made of a mix
of screenings from 15-mm and 3-mm screens and was compacted in the plant. The
sample was taken to the laboratory were it was homogenized and divided into 20 sub-
samples of 8 kg. The procedure allowed to guarantee a good reproducibility of the sub-
samples composition, which was verified analytically. The samples were stored at -
18°C. A digestate sampled from a municipal solid waste digester was used as an
inoculum. The major characteristics of the waste and the inoculum are shown in Table3.
2.5. Experimental procedure
The pilot-scale experiment comprised 4 stages:
(a) Inoculation (2 weeks): The anaerobic reactor was inoculated with about 44.2
kg of digestate (corresponding to 8.0 kgDS) sampled from a municipal solid
waste digester plant (Table 3). The digester was then incubated for 2 weeks
with no feeding in order to allow the stabilization of the microflora and the
biodegradation of the residual organic matter of the digestate, which was
estimated from the biogas production. At the end of this stage, the digestate
was analysed for dry matter and volatile matter contents, pH and VFAs.
(b) Start-up stage at a constant organic load (4 weeks): Every week, a sub-
sample of 8.0 kg of raw waste (corresponding to 2.0 kgDS) was defrozen and
introduced into the digester. A mass of digestate corresponding to the same
mass of dry solids was firstly removed from the digester in order to maintain a
constant mass of dry solids in the reactor, corresponding to an average
residence time of 4 weeks (based on dry solids). The digestates were analysed
immediately after sampling. Biogas production and temperature in the reactor
were monitored continuously. Gas composition was measured before each
(c) Recovery stage (4 weeks): The previous stage induced a strong inhibition of the
digestion process. It was therefore decided to stop feeding the digester for a
while. No waste was introduced into the digester over 4 weeks in the aim of
allowing the microflora to recover from the state of inhibition. Biogas
production was monitored continuously and its composition was measured
each week. At the end of this stage, the digestate was analysed to verify that its
characteristics were adequate (pH near 7-8 and low concentration of VFA).
(d) Anaerobic digestion stage with adapted organic load (8 weeks): After the
recovery stage, the digestion was restarted with a progressive increase of
organic load (2.0 kg of raw waste introduced the first week, 4.0 kg the second
week and 6.0 kg per week in the last six weeks) while digestate was taken out
to maintain constant the mass of dry solids in the reactor. The digestates were
analysed immediately after sampling. Biogas production and temperature in
the reactor were monitored continuously. Gas composition was measured
before each loading/unloading operation.
3. Results and discussion
3.1. Weekly biogas production in the successive stages of the experiment
Fig. 2 shows the weekly biogas production measured during the four experimental
stages. The following discussion focuses on the 2nd stage (constant organic load, weeks
3 to 6) and the last one (progressive organic load, weeks 11 to 18).
3.2. Operation of the digester at a constant organic load (weeks 3 to 6)
During this phase, 8.0 kg of raw screenings were introduced into the digester every
week, corresponding to a load of 2.0 kg of dry solids and an average residence time of 4
weeks. Fig. 4 shows the biogas production obtained, expressed in Nl per week under
normal pressure (1 atm) and temperature (0 °C) conditions. It can be seen that biogas
production decreased drastically from 488 (w3) to 143 Nl/week (w6), revealing the
installation of inhibitory conditions probably due to the overload of the digester. The
analysis of the waste sampled each week from the reactor confirmed the inhibitory
conditions by revealing a regular increase of the total VFA content from 11.1 (end of
w2) to 67.1 g/kgDS, and a pH drop from 8.4 to 6.4 (Fig. 3). Considering that volatile
fatty acids were predominantly present in the aqueous solution and that the moisture of
the waste in the digester was about 80 % w/w, the measured VFA contents
corresponded to aqueous concentrations ranging from 1.6 to 12.2 g/l, well above the
inhibitory concentration normally considered (Aguilar et al., 1995; Gourdon and
Vermande, 1987) (Table 4). Increasingly VFA concentrations underlined the imbalance
between a rapid production of acids by hydrolysis / acidogenesis and their relatively
slow degradation by acetogenesis / methanogenesis, as reported by many authors
(Rodrigues-Iglesias et al, 1998). VFA were not consumed fast enough by acetogenic
and acetoclastic communities and the overall anaerobic digestion process was inhibited.
Biogas composition (% v/v CH4 and % v/v CO2) for this experimental phase is shown in
Table 4. The concentrations of H2S were below the detection limit of 0.01 % v/v.
According to the results from BMP determinations, CH4/CO2 ratio of 58/42 could be
expected (Table 3). However, Table 4 shows that the CH4/CO2 ratio in the biogas from
the pilot reactor varied from 48/52 to 39/61 during the constant load stage. The
relatively low methane content in the biogas from the digester confirmed the ill-
balanced digestion process in the digester, with active hydrolysis and acidogenesis steps
resulting in the production of VFA and CO2 and inhibited methanogenesis.
3.3. Recovery from inhibition (weeks 7 to 10)
To allow the microbial communities to recover from inhibition, it was decided to stop
feeding the digester until excess VFA were biodegraded. No waste was introduced over
a 4-week period (weeks 7 to 10).
Biogas production and its composition were still measured and analysed however
during this stage. It was observed that biogas production increased progressively after
week 8 (Fig. 2) with a strong increase of methane content (Table 5), indicating a good
recovery of methanogenic activity. This was confirmed by the analysis of the waste at
the completion of the stage (Table 5), which revealed an increase of pH from 6.4 (end of
week 6, Table 4) to 8.0 (end of week 10, Table 5) and a significant reduction of VFA
3.4. Operation of the digester at an adapted organic load (weeks 11 to 18)
Based on the good recovery from inhibition, feeding of the digester was resumed at
week 11. The amount of raw waste introduced into the digester was progressively
increased from 2 kg to 6 kg (Fig. 2). Fig. 5 shows that the biogas production increased
with the load and stabilized between 513 and 653 Nl/kgVSadded per week during weeks
13 to 18 where a constant load of 6 kg of raw waste per week was applied (dry solids
average residence time of 5 weeks).
Fig. 3 shows that the pH was quite stable during this stage, ranging between 8.0 and 8.4,
which is close to the optimal working range (Comino et al., 2009; Macias-Corral et al.,
2008). VFA content decreased from 38.8 to 1.5 g/kgDS, indicating that the acids
produced were efficiently consumed by acetogenic and methanogenic communities and
converted into biogas (Fig. 3).
Biogas composition was also relatively stable over the eight weeks of this phase (Table
6). The CH4/CO2 ratio ranged between 59/41 and 63/37, thereby confirming that the
process of anaerobic digestion was conducted under good operating conditions. It
should be noted however that biogas composition was determined at the end each week
(before loading/unloading operations) and therefore did not reflect the average
composition of the total biogas produced between two feedings. Indeed, the biogas
produced in the first days after feeding, which probably contained a higher proportion of
CO2 generated by hydrolysis and acidogenesis, was continuously vented out of the
digester and therefore not analyzed. Only the biogas produced at the end of each week,
containing more methane, was actually analyzed. This overestimation could partly
explain why the CH4/CO2 ratio in this phase of the pilot-scale study was higher than in
the BMP assays (Table 3).
Laboratory analyses of screenings from different wastewater treatment plants revealed a
high fraction of very biodegradable organic matter. Pilot-scale experiments
demonstrated the feasibility of anaerobic digestion of screenings with a strong methane
potential of about 600 Nl of biogas (at 60-62 % CH4 v/v) per kg VS added, but revealed
the sensitivity of the process to overloading during the start-up phase.
Due to the relatively small production of screenings however, the implementation of a
dedicated digester would probably not be economically sound at the industrial level.
Co-digestion in existing digester plants, for example with sludge and/or or other wastes,
would likely be more realistic provided that the screenings were properly pretreated.
The authors wish to thank the competitiveness cluster on chemistry and the environment
of Lyon – Rhône-Alpes (Axelera) for the implementation of this research program.
They also gratefully acknowledge the Region Rhône-Alpes for financial support.
Aguilar, A., Casas, C., Lema, J.M., 1995. Degradation of volatile fatty acids by
differently enriched methanogenic cultures: kinetics and inhibition. Water Research
29 (2), 505-509.
Bode, H., Imhoff, K. R., 1996. Current and planned disposal of sewage sludge and other
products from the Ruhrverband wastewater treatment. Water Science and
Technology 33, 219-228.
Clay, C., Hodgkinson, A., Upton, J., Green, M., 1996. Developing acceptable sewage
screening practices. Water Science and Technology 33, 229-234.
Comino, E., Rosso, M., Riggio, V., 2009. Development of a pilot scale anaerobic
digester for biogas production from cow manure and whey mix. Bioresource
Technology 100, 5072-5078.
Davidsson, Å., Lövstedt, C., la Cour Jansen, J., Gruvberger, C., Aspergen, H., 2008.
Codigestion of grease trap sludge and sewage sludge. Waste Management 28, 986-
Frechen, F.-B., Schier, W., Wett, M. 2006. Pre-treatment of municipal MBR
applications in Germany – Current status and treatment efficiency. Water Practice &
Gourdon, R., Vermande, P., 1987. Effects of propionic acid concentration on anaerobic
digestion of pig manure. Biomass 13, 1-12.
Huber, H., Tanik, A. B. & Gerçek, M., 1995. Case studies on preliminary treatment
facilities at marine outfalls. Water Science and Technology 32, 265-271.
Jokela, J. P. Y., Vavilin, V. A., Rintala, J. A., 2005. Hydrolysis rates, methane
production and nitrogen solubilisation of grey waste components during anaerobic
degradation. Bioresource Technology 96, 501-508.
Le Hyaric, R., Canler, J-P., Barillon, B., Naquin, P., Gourdon, R., 2009.
Characterization of screenings from three municipal wastewater treatment plants in
the Region Rhône-Alpes. Water Science and Technology 60(2), 525-531.
Macias-Corral, M., Samani, Z., Hanson, A., Smith, G., Funk, P., Yu, H., Longworth, J.,
2008. Anaerobic digestion of municipal solid waste and agricultural waste and
effect of co-digestion with dairy cow manure. Bioresource Technology 99, 8288–
Naud, P.-Y., Perret, J.-M., Canler, J.-P., 2007. Vers une meilleure connaissance des
déchets issus de l’étape des prétraitements (Toward a better knowledge of residual
wastes from pre-treatments). Technique Sciences Méthodes 10, pp. 131.
Owens, J. M., Chynoweth, D. P., 1993. Biochemical methane potential of municipal
solid waste (MSW) components. Water Science and Technology 27(2), 1-14.
Rodriguez-Iglesias, J., Castrillon, L., Marañon, E., Sastre, H., 1998. Solid-state
anaerobic digestion of unsorted municipal solid waste in a pilot-plant scale digester.
Bioresource Technology 63, 29-35.
Sosnowski, P., Klepacz-Smolka, A., Kaczorek, K., Ledakowicz, S., 2008. Kinetic
investigations of methane co-fermentation of sewage sludge and organic fraction of
municipal solid wastes. Bioresource Technology 99, 5731–5737.
Composition of screenings of three domestic wastewater treatment plants (WWTP)
wwtp 1 wwtp 2 wwtp 3
Fine fraction (< 20 mm)
DS: dry solids
PE: population equivalent
Characterization of screenings and sievings from three domestic WWTPs in the Region
Fine fraction (< 20 mm)
% dry solids
% volatile solids
(Nml CH4/g VS)
a BMP: Bio Methane Potential;
b Screenings: Waste from fine screen (6 to 10 mm), middle screen (10 to 20 mm) or coarse screen (20 to 60 mm);
c Sievings: Waste from sieve (gap size of 3 mm and less).
Characterization of screenings and inoculum used for pilot-scale experimentation
from wwtp of Bourg-en-Bresse
25.3 (± 0.9)
92.9 (± 0.1)
18.1 (± 0.7)
52.4 (± 1.6)
57.6 / 42.4
50.6 / 49.4
BMP: Bio Methane Potential