Available online at www.sciencedirect.

com

Waste Management 28 (2008) 18011808

www.elsevier.com/locate/wasman

Two-phase anaerobic digestion within a solid

waste/wastewater integrated management system
a b a,* a
G. De Gioannis , L.F. Diaz , A. Muntoni , A. Pisanu
a
DIGITA, Department of Geoengineering and Environmental Technologies, University of Cagliari, Piazza DArmi 09123 Cagliari, Italy
b
CalRecovery, Inc., 2454 Stanwell Drive, Concord, California 94520, USA
Accepted 20 November 2007
Available online 10 January 2008

Abstract

A two-phase, wet anaerobic digestion process was tested at laboratory scale using mechanically pre-treated municipal solid waste (MSW) as
the substrate. The proposed process scheme diers from others due to the integration of the MSW and wastewater treatment cycles, which
makes it possible to avoid the recirculation of process e uent. The results obtained show that the supplying of facultative biomass, drawn from
the wastewater aeration tank, to the solid waste acidogenic reactor allows an improvement of the performance of the first phase of the process
which is positively reflected on the second one. The proposed process performed successfully, adopting mes-ophilic conditions and a relatively
short hydraulic retention time in the methanogenic reactor, as well as high values of organic loading rate. Significant VS removal e ciency and
biogas production were achieved. Moreover, the methanogenic reactor quickly reached opti-mal conditions for a stable methanogenic phase.
Studies conducted elsewhere also confirm the feasibility of integrating the treatment of the organic fraction of MSW with that of wastewater.

2007 Elsevier Ltd. All rights reserved.

1. Introduction achieve. Energy recovery through anaerobic digestion rep-

The biodegradable organic matter contained in munici-
pal solid waste (MSW) probably is one of the most prob-
lematic fractions to deal with: it is dicult to sort if
mixed with the other fractions, it aects the performance
of incinerators in terms of energy recovery and it is the
cause of the long lasting emissions from sanitary landfills
(Cossu et al., 2003; Soyez and Plickert, 2002; Zach et al.,
2000). European legislation requires a drastic reduction
of the amount of biodegradable organic residues to be
landfilled. The European MSW management policy relies
on separate collection in order to allow recycling of most
of the packaging waste and the production of high quality
compost from the biodegradable organic fraction. How-
ever, recycling through separate collection and composting
of the entire MSW biodegradable fraction is dicult to
* Corresponding author. Tel.: +39 70 6755546; fax: +39 70 6755523.
E-mail address: amuntoni@unica.it (A. Muntoni).
resents a complementary option. Mechanical pre-treatment
of residual MSW can be performed in order to produce two
streams: a first one, characterized by a significantly high
heating value, to be utilized as refuse derived fuel (RDF)
and a second one which would need biological stabilization
before it is disposed. As far as this second stream is con-
cerned, a process based on anaerobic digestion followed
by aerobic treatment combines well the need for stabiliza-
tion with the opportunity for energy recovery through bio-
gas production and exploitation (Diaz et al., 1981; De
Baere, 2000).
Currently several companies and researchers propose
their own anaerobic process schemes, which usually are
classified on the basis either of the total solid (TS) content
as wet digestion (TS content: 1015%) and dry digestion
(TS content: 2540%), or of the adopted temperature
regimes as mesophilic (3040 LC) and thermophilic diges-
tion (5060 LC). According to a number of research activi-
ties, good results are achievable by a two-phase, wet
mesophilic digestion process (Bhattacharayya et al., 2007;

0956-053X/$ - see front matter 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.wasman.2007.11.005
1802 G. De Gioannis et al. / Waste Management 28 (2008) 18011808

Bolzonella et al., 2003;
Bouallagui et al., 2005;
Cecchi et al., 2002; Maharaj
and Elefsiniotis, 2001; Mata
Alvarez, 2002b; Pavan et al.,
2000; Schober et al., 1999;
Zhang et al., 2005). The
separation of the hydrolysis-
acetogenesis and
methanogenesis phases
results, in fact, in a higher
pro-cess stability;
furthermore, the dilution of
inhibiting sub-stances
contained in the inflow, the
optimization of the contact
between biomass and
substrate and the possibility
to treat a wider range of
waste (MSW, residues from
food industry, etc.) are
achieved. With respect to the
tempera-ture regimes,
mesophilic digestion has
been widely adopted for
anaerobic treatment due to
the good operational per-
formance, whilst the use of
thermophilic processes has
been limited, mainly due to
poor supernatant quality and
pro-cess instability caused,
for example, by chronically
high propionate
concentration (Kim et al.,
2002).
Although it has been
reported that anaerobic
consortia are able to adapt to
adverse conditions if they
are given adequate time
(Speece, 1983, 1996),
anaerobic digestion
processes are often
characterized by an intrinsic
instability deriving from the
high sensitivity of the
anaerobic biomass to a
number of inhibiting or toxic
substances (Garc`a-Heras,
2002; Salminen and Rintala,
2002; Vandevivere et al.,
2002), either endogenous
(ammonia and organic
acids) or exogenous (heavy
+ sidered the rate-limiting step
metals and cations like K ,
+ ++ ++
Na , Mg , Ca ). It is a
general opinion that the
recircu-
lation of the euent in a
wet anaerobic digestion
processes, often in
significant quantities to
achieve the desired mois-
ture content in the solid
waste, can lead to the
accumula-tion of inhibiting
substances (Gallert et al.,
2003).
A two-phase, mesophilic
wet anaerobic digestion pro-
cess was tested at the
laboratory scale using
mechanically pre-treated
MSW as the substrate.
Recirculation of the euent
was avoided assuming a
possible integration of the
MSW and municipal
wastewater treatment cycles.
In fact, this integrated
management would allow:
the supply of facultative
biomass and liquid phase
to the solid waste
acidogenic reactor
through the utilization of
activated sludge from the
aeration tank of a
wastewater treatment
plant;
the supply of anaerobic
sludge to the solid waste
meth-anogenic reactor
drawing it from the
sewage sludge digester of
a wastewater treatment
plant;
the treatment of the
euent of the solid
waste treatment plant at
the wastewater treatment
plant; and
the use of excess capacity
in conventional sewage
sludge digesters.
The supply of the easy to
obtain activated sludge to
the acidogenic reactor leads
to several advantages: (1)
the first phase of the
process, which includes
hydrolysis that is con-
of the entire solid waste
anaerobic digestion process,
benefits by the considerable
presence and high growth
yields of the facultative
biomass (Kim and Speece,
2002; Schober et al., 1999);
(2) activated sludge is
characterized by a low
content of inert material
acidogenic phase and the
and sand (Kim and Speece,
2002); (3) nitrogen, as well
as other nutrients, which are
not present at sucient
levels in MSW, are provided
(Stroot et al., 2001); and
finally,
(4) the potential risk of
accumulation of inhibiting
sub-stances is reduced due to
the avoidance of euent
recirculation.
The expected improved
performance of the
possible avoidance of
inhibiting eects could
allow the achievement of
significant conversion
eciency of the substrate
and adequate process
stability, despite the
problematic nature of MSW.
Therefore, shorter values of
the methanogenic reactor
hydraulic retention time
(HRT) could be adopted.
On the basis of these
considerations, a research
program was developed
aimed at testing, at the
laboratory scale, the
performance of a two-phase,
mesophilic, wet anaerobic
digestion process, where (a)
the euent recirculation is
replaced as a result of the
integration of the MSW and
municipal wastewater
treatment cycles, and (b)
mechani-cally pre-treated
MSW is used as the sole
substrate.
2. Materials and methods
2.1. Materials
Mechanically pre-treated
MSW, activated sludge from
the aeration tank and
anaerobic sludge from the
sewage sludge digester were
sampled at an integrated
platform where municipal
wastewater treatment is
performed and MSW and
digested wastewater
treatment sludge are incin-
erated. The solid waste was
sampled at a mechanical
treat-ment section where the
plastic bags containing
MSW are opened, the solid
waste is shredded and size
classification is conducted
by means of a rotating sieve
(U = 60 mm). These
operations result in two
streams: the over-size,
which is introduced into the
incinerator and the under-
size which has to be
biologically stabilized prior
to landfilling. Sam-ples of
the mechanically pre-treated
waste were collected during
three dierent weeks of the
summer season to take into
account the possible short-
term variations of waste
composition, and, in turn, to
test the performance of the
system under these
circumstances. Indeed, the
characteris-tics of the
sampled residues allowed
for the adoption of dif-ferent
values of the organic loading
rate during the test runs.

The solid waste, the

activated sludge and the
anaerobic sludge were
analyzed to determine total
solids (TS), volatile solids
(VS), total organic carbon
(TOC), and total nitrogen
(Ntot); the ratio of TOC to
total nitrogen (C/N) was
calcu-lated. All of the
analyses were performed in
triplicate; the
average data are reported in
Table 1.

2.2. Experimental device

and operation

The proposed scheme is

shown in Fig. 1. The
reactors were operated in
semi-continuous mode by
feeding and removing an
equivalent volume on a daily
basis.
 G. De Gioannis et al. / Waste Management 28 (2008) 18011808 1803
Table 1

Characterization of mechanically pre-treated MSW, activated sludge and anaerobic sludge used during the test runs
TS (%) VS (%TS) TOC (%TS) Ntot (%TS) C/N
Waste 1 (Wl) 59.7 1.35 36.2 0.79 18.4 1.25 0.9 0.08 20.5 3.24
Waste 2 (W2) 57.7 1.25 22.0 0.70 18.1 0.95 2.0 0.13 9.1 0.94
Waste 3 (W3) 64.3 1.40 44.3 1.83 30.6 1.48 1.4 0.10 21.9 2.6
Activated sludge 0.5 0.01 0.3 0.04
Anaerobic sludge 4.0 0.1 2.4 0.26 1.2 0.08

the values usually adopted (i.e. 18 days, Kim et al. (2002);

14.2 days, Scherer et al. (2000); 1418 days, Hart-mann and
Ahring (2005); 17 days, Fernandez et al. (2005)) since a good
eciency of the hydrolysis/acidogenic phase was expected.

Start-up of the methanogenic reactors was performed by

Fig. 1. Process scheme of the tested process (mass flow in % w/w).
The acidogenic phase was carried out using completely
mixed reactors having a volume of 500 ml; temperature
control was not performed. Mechanically pre-treated MSW
and activated sludge were mixed in order to obtain 1015%
TS content and the resulting mixture fed to the aci-dogenic
reactor. Acclimation of the aerobic biomass to anaerobic
conditions was allowed by not feeding the acido-genic
reactors during the first 48 h after initial filling. After-wards,
the reactors were operated in a semi-continuous mode in order
to achieve a hydraulic retention time (HRT) of 3 days, similar
to those adopted by other authors (e.g. Viturtia et al., 1995;
Scherer et al., 2000).
Reactors of 800 ml, minimally mixed thoroughly shaken
by hand for 2 min each day, i.e. 1 min before wasting and 1
min after feeding as reported by Stroot et al. (2001), were
used as methanogenic reactors. The temperature was set at the
limit of the mesophilic field (39 LC) and controlled by
circulating warm water through a water jacket. Daily feed-ing
of the methanogenic reactors was performed using a mixture
of 80% w/w of the acidogenic reactor euent and 20% of
anaerobic sludge. The HRT of the methano-genic reactor was
set at 8 days, significantly shorter than
first filling them with a mixture of 75% v/v of demineralised
water and 25% v/v of anaerobic sludge. Feeding of the
methanogenic reactors was not performed during the fol-
lowing 48 h; afterwards, the reactors were operated in a semi-
continuous mode by daily wasting 12.5% of the vol-ume and
feeding with an equivalent volume of the mixture of
acidogenic reactor euent and anaerobic sludge in order to
gradually reach the prefixed OLRs, which was achieved in
about 2 HRTs.
All the tests were carried out in triplicates and according to
two runs characterized by dierent operating conditions
(Table 2).
The first run (28 days) was performed using waste W1
(Table 1). The mixture used for feeding the acidogenic reac-
tors was characterized by a TS content of 9.8% and the
3
organic loading rate (OLR) was set at 19.9 kg VS/m d
(OLR1, Table 2).
The second run lasted 60 days; two dierent wastes (W2
and W3) were used and three values of OLR were adopted
(Table 2: periods 1, 2 and 3 of run 2). Waste W2, charac-
terized by low values of the C/N ratio (9.1) and of the VS
content, was used during the first two periods of the second
run. During period 1 of run 2 (20 days), a mixture of 82.3%
w/w of activated sludge and 17.7% of waste W2,
corresponding to a TS content of 10% and an organic load-ing
3
rate of 13.2 kg VS/m d (OLR2, Table 2), fed the acido-genic
reactors. The mixing ratio was changed during period 2 of run
2 (20 days), according to 78% w/w of activated sludge and
22% of waste W2, corresponding to a solid con-tent of 13.2%
3
and in order to have an organic loading rate of 17.6 kg VS/m
d (OLR3, Table 2). Waste W3, character-

Table 2
Operational parameters adopted during runs 1 and 2
Run Period Waste Acidogenic reactor feeding 3 TS (%) Waste HRT (d)
OLR (kg VS/m
(% w/w) d) C/N
Activated sludge Waste Acidogenic Methanogenic
reactor reactor
1 Wl 84.0 16.0 19.9 (OLR1) 9.8 20.5 3 8
2 2.1 W2 82.3 17.7 13.2 (OLR2) 10.0 9.1 3 8
2.2 W2 78.0 22.0 17.6 (OLR3) 13.2 9.1 3 8
2.3 W3 78.0 22.0 33.0 (OLR4) 14.7 21.9 3 8
1804 G. De Gioannis et al. / Waste Management 28 (2008) 18011808
ized by higher values of C/N ratio and VS content, was used
during period 3 of run 2 (20 days). The aim of this last part of
the research was testing the process when high val-ues of
OLR are adopted; in fact, the same waste/activated sludge
mixing ratio of period 2 was adopted but, due to the di erent
characteristics of the solid waste, the resulting OLR was 33.0
3
kg VS/m d (OLR4, Table 2).
The performance of the reactors was evaluated monitor-
ing, on a daily basis, the volatile fatty acids (VFA) produc-tion
(acidogenic reactors), the VS removal eciency (acidogenic
reactors and methanogenic reactors), the spe-cific gas
production (methanogenic reactors) and the pH values
(methanogenic reactors and, periodically, acido-genic
reactors). The concentration of VFA and the pH are
considered parameters that allow determining the sta-bility of
an anaerobic digestion process (Sans et al., 1995; Schober et
al., 1999).
2.4. Data analysis
Statistical significance analysis of the values of di erent
parameters obtained for consecutive adopted OLRs was
carried out by means of F-test (variance analysis) and Stu-
dents t-test (mean analysis), both at a 5% level of probabil-ity
(p < 0.05).
3. Results and discussion
The data in Table 3 present the performance of the sys-tem
in terms of VFA production, VS removal eciency and
specific gas production. The VS removal eciency is defined
as the VS content in the influent minus the VS of the euent
to be divided by the initial VS content. The average values for
each run/period were calculated consid-ering the triplicate
series of the obtained data; standard deviation also is reported.

2.3. Analytical methods The VFA concentration in the acidogenic reactor

(expressed as mg CH3COOH/l) is shown in Fig. 2. The VFA
The total solid content (TS) was measured by weighing the concentration profile shows a stable performance after an
sample before and after drying at 105 LC for 12 h. The initial transitory decrease during the start-up, assessed also by
volatile solids content (VS) was measured by weigh-ing the other authors (Hartmann and Ahring, 2005). After the end of
sample before and after combustion at 550 LC for 6 h. The the start-up period (about 2 HRTs), VFA concentration
total nitrogen content (Ntot) was assessed by means of the dry evolved in accordance with the dier-ent values of the
combustion method with the elemental analyzer CHN600, adopted OLRs, as also reported in other experiences (Mata
Leco Instr. The organic carbon con-tent (TOC) was Alvarez, 2002a,b; Vandevivere et al.,
determined by dierence between the total carbon content
measured by the CHN600 analyzer and the fixed carbon
content measured by thermal oxida-tion at 950 LC with a
MAC400 instrument.
The biogas production was measured by connecting the
methanogenic reactors to a volumetric measurement device
based on the principle of water displacement. Biogas com-
position was assessed using a gas chromatograph (GC
HP5890) equipped with a thermoconductivity detector (TCD)
and a 2-m stainless steel column packed with Pora-pak T. The
dissolved oxygen concentration (DOC) in the reactors was
periodically measured (data not shown) by means of a digital
oxymeter (AMEL Instr. Mod. 360); the concentration of
volatile fatty acids (VFA) was deter-mined by means of
titration according to the Kapp method reported by Buchauer
Fig. 2. VFA concentration in the acidogenic reactor (horizontal solid line:
(1998). OLR adopted during each period/run).

Table 3
Performance of the reactors in terms of VFA production, VS removal eciency and assessed specific gas production (SGP A)
VFA (mg CH3COOH/I) VS removal eciency (%) SGPA (assessed) Nl/kg MSW
Acidogenic reactor Acidogenic reactor Entire system Methanogenic reactor
Average Std dev Average Std dev Average Std dev Average Std dev
Run l 693.2 21.1 48.8 2.6 71.3 1.0 84.6 9.5
Run 2, 106.1 7.7 9.8 0.8 49.8 7.2 56.6 6.7
period 2.1
Run 2, 445.2 24.4 31.5 2.2 58.2 6.9 63.2 13.5
period 2.2
Run 2, 1644.7 26.2 63.0 5.9 78.0 3.3 99.6 2.3
period 2.3
G. De Gioannis et al. / Waste Management 28 (2008) 18011808 1805

2002). During the first run,
characterized by an OLR of
3 an
average
19.9 kg VS/m d,
693.2 21.1 mg CH3COOH/l was
OLR decreased (13.2 kg
3
VS/m d), due to W2 waste
feed-ing instead of W1, the
overall VFA concentration
dropped quickly to as low as
106.0 7.7 mg/l. Adopting a
3 The acidogenic reactor
higher OLR (17.6 kg VS/m
d) resulted in an increase of
the VFA concentration in the
acidogenic reactor up to
445.2 24.4 mg/l, and when
the highly biodegradable
waste W3 was used, the
VFA concentration reached
the peak value of 1644.7
26.2 mg/l. The dierence
between the VFA
concentration values was
found to be statistically
significant, with p < 0.05 for
all of the periods/runs under
investigation.
Concerning the high VFA
concentration values reached
during period 2.3, some
controversy can be found in
the lit-erature about possible
inhibitory eects (Vavilin et
al., 2007). Veeken et al.
(2000) underscore that the
influence of pH values also
has to be taken into account,
concluding that acidic pH is
the main inhibitor factor and
that no inhi-bition by VFA
or by non-ionized VFA can
be measured at pH values
between 5 and 7. However,
it is dicult to dis-tinguish
and separate the role of VFA
from that of pH, as well as
to assess the influence of pH
itself. In fact, Vavi-lin et al.
(2007) state that gradients of
pH occurring close to the
hydrolizable particles and
able to decrease the hydroly-
sis rate could not be
measurable into the reactor
bulk liquid; this could
explain some contradictory
results found in the
literature. As far as the
present research is
concerned, indeed no
adverse e ect was noticed,
in accordance with the pH
values measured in the
euent of the acidogenic
reactor whose results were
always higher than 5.5 (data
not reported), even when the
highest OLR value was
adopted and the highest VFA
concentration was reached.
VS removal eciency and
the entire system one
(acidogenic + methanogenic
reactor) are reported in Figs.
3 and 4. It can be noticed
that low val-ues of the
adopted OLR resulted in low
VS removal e-ciency in
the acidogenic reactor. The
highest VS removal
eciency (63 5.8%) was
achieved when an OLR of
3 chanawong and Deesopa
33.0 kg VS/m d was
adopted. The performance of
the aci-dogenic reactor can
be considered remarkable
when com-pared with data
provided by other authors.
Perot and
Fig. 3. VS removal eciency in the
acidogenic reactor (horizontal solid
line: OLR adopted during each
period/run).
2, characterized by low
Fig. 4. VS removal eciency for
the entire system (acidogenic +
metha-nogenic reactor) (horizontal
solid line: OLR adopted during
each period/ run).
Amar (1989) report a VS
removal eciency of 20%
related to a two-stage
anaerobic digestion of
sewage sludge; Karn-
(2004) underline VS
removal e-ciency values
of 7.2% and 9.4% for a two-
stage process
applied to MSW digestion
3
m d, respectively); Sung and
VS reduction of 21.831.4%
after the first phase of a
diges-tion process applied to
dairy cattle manure (OLR
variable and mesophilic
conditions); and Scherer et
al. (2000) report, for a two-
stage anaerobic digestion
process applied to
mechanically pre-treated
MSW, a VS removal
eciency from 5% to 20%
related to a OLR increasing
from 10 to 26 g VS/l d (55
LC).
As far as the entire
process is considered
(acidogenic and
methanogenic reactors, Fig.
4), the following VS
removal eciency values
were achieved: 71.3 1.0%
(run 1, OLR1 = 19.9 kg
3
VS/m d), 49.8 7.2% (run
2 period 1, OLR2 = 13.2
3
kg VS/m d), 58.2 6.9%
(run 2 period 2, OLR3 =
3
17.6 kg VS/m d) and 78.8
3.3% (run 2 period 3,
OLR4 = 33.0 kg VS/m d).
3
The stability of the process,
even during period 1 of run
values of the OLR and of the
waste C/N ratio, is
underscored by the trend of
the daily values of the total
VS removal e-ciency
(acidogenic and
methanogenic reactors, Fig.
4). The dierence between
the VS removal eciency
assessed for the acidogenic
phase and that for the entire
process was found to be
statistically significant, with
p < 0.05 for all of the
periods/runs under
investigation.
The trend of the assessed
specific gas production
expressed with respect to the
unit mass of fed waste
(SGPA, Nl/kg MSW) is
shown in Fig. 5. The SGP
profile shows an initial
transitory increase which is
typical for the start-up of
anaerobic processes
(Bolzonella et al., 2003; Kim
et al., 2002). After about 2
HRTs, a good correlation can
be noticed between the SGP
and the organic loading rate,
as noticed also by Viturtia et
al. (1995). The positive
evolution of the process can
be noticed also considering
the trend of the pH values in
the methanogenic reactor
(Fig. 6). The pH value of
6.8, indicative of stable
methanogenic conditions,
was achieved within 16 days
from the beginning of both
runs 1 and 2. This occurred
also when waste W2,
character-ized by a C/N of
9.1, considered unfavourable
for the pro-
1806 G. De Gioannis et al. / Waste Management 28 (2008) 18011808

Fig. 5. Values of the specific gas production assessed from the methano-genic
reactor (SGPA) (horizontal solid line: OLR adopted during each period/run).
Fig. 6. Evolution of the pH in the methanogenic reactors (average values)
during runs 1 and 2.
cess (Hartmann et al., 2002), was treated. Statistical analy-ses
indicated that the dierence between SGP values was
significant, with p < 0.05 for all of the periods/runs under
investigation. The CH4 content in the produced gas always
remained close to 60% v/v.
Table 4 summarizes some data considered for a further
evaluation of the process performance. Theoretical specific
biogas production values (SGPT0 and SGPTm) were evalu-ated
on the basis of the amount of the TOC fed to the entire system
and to the methanogenic reactor, respec-tively. The amount of
organic carbon available for biogas production under the
adopted temperature conditions
was estimated using the following relation, as proposed by
Bingemer and Crutzen (1987):
TOCb=TOC 0:014T 0:28
where TOCb: content of organic carbon available for
biogasification; TOC: total organic carbon content (Table 4)
and T: process temperature in LC.
The theoretical specific gas production was then calcu-
lated assuming that the biogasification of 1 g of organic
carbon leads to the formation of 1.867 Nl of biogas (CH 4 +
CO2) (Cossu et al., 1996; Rao and Singh, 2004).
The expected specific biogas production (SGPE) was cal-
culated on the basis of the amount of organic carbon removed
during the methanogenic stage (TOC assessed in the influent
to the methanogenic reactor minus TOC assessed in the
euent) and assuming, as reported before, that 1.867 Nl of
biogas are produced by the biogasification of 1 g of organic
carbon.
Comparing the theoretical specific gas production for the
entire system (SGPT0) with the assessed specific gas
production from the methanogenic reactor (SGPA), the organic
carbon conversion eciency (SGPA/SGPT0) results for the
entire system are lower than 30%. However, this result should
be evaluated taking into account that the bio-gas production
from the acidogenic reactor was not mea-sured. In fact, if the
performance of the entire system is assessed in terms of
organic carbon removal, it can be derived from data reported
in Table 4 that it accounted for 80% and 73% for run 2.3 and
run 1, respectively, Fur-thermore, the lowest SGP A/SGPT0
value was reached in run 2.3, when the highest VS and TOC
removals in the aci-dogenic reactor were measured.
Comparing the theoretical specific gas production for the
methanogenic reactor (SGPTm) with the specific gas
production (SGPA) assessed from the same reactor, it can be
noticed that organic carbon conversion eciencies
(SGPA/SGPTm) up to 51% and 47.7% were reached in run 1
and run 2.3, respectively. Similar values (49.3%) are reported
by Sosnowski et al. (2003) for an UASB reactor treating a
mixture of sewage sludge and MSW biodegrad-able fractions.
Chae et al. (2008) reported maximum con-

Table 4
Theoretical, expected and assessed specific gas production for the dierent runs
Run TOC TOC to the TOC removed a b c d
SGPT0 SGPTm SGPE SGPA SGPA/ SGPA/ SGPA/
to the methanogenic in the (Nl/kg MSW) (methanogenic (methanogenic (methanogenic SGPT0 SGPTm SGPE
system reactor (g) methanogenic reactor) reactor) reactor) % % %
(g) reactor (g) (Nl/kg MSW) (Nl/kg MSW) (Nl/kg MSW)
Run 1 2.62 1.45 0.73 299.3 165.6 100.9 84.6 28.2 51.1 83.3
Run 2.1 1.74 1.55 0.56 198.8 177.1 77.4 56.6 28.5 32.0 73.0
Run 2.2 2.64 1.75 0.54 301.6 199.9 74.6 63.2 20.9 31.6 84.6
Run 2.3 4.66 1.83 0.33 532.3 209.1 121.7 99.6 18.7 47.7 81.9
a
SGPT0: theoretical specific gas production, calculated on the basis of the TOC b (TOC fraction available for biogasification) fed to the system.
b
SGPTm: theoretical specific gas production, calculated on the basis of the TOCb (TOC fraction available for biogasification) fed to the methanogenic reactor.
c d
SGPE: expected specific gas production, calculated on the basis of the amount of organic carbon removed in the methanogenic reactor. SGPA:
assessed specific gas production from the methanogenic reactor.
 G. De Gioannis et al. / Waste Management 28 (2008) 18011808 1807
Table 5

Comparison between the performance of the proposed process and those of some other systems
System Substrate T (LC) HRT VS removal 3
Methane yield (N m /
(d) eciency (%) kg VS)
Proposed process wet, Mechanically pre-treated MSW 39 11 5079 0.24
two-phase
DRANCO Dry, MSW organic fraction from separate collection 50 20 4070 0.210.30
one-phase
TBW Wet, two-phase MSW organic fraction from separate collection 3555 14 60 N.A.
VALORGA Dry MSW organic fraction from separate collection 3755 20 60 0.210.30
one-phase
WABIO Wet, one-phase Codisposal of sludge and MSW organic fraction from 35 1520 57.5 N.A.
separate collection

version eciency of 60.4% for anaerobic digestion of swine
manure.
The data reported in Table 4 show also that the values of
biogas production assessed for the methanogenic stage
(SGPA) are fairly consistent with the expected gas produc-tion
(SGPE) calculated for the same stage.
In order to underscore the interesting perspectives of the
proposed process, Table 5 provides a comparison with the
performances of other anaerobic digestion systems, most
characterized by higher values of the methanogenic reactor
operative temperature and of the hydraulic retention time, as
well as by the use of selected substrate (Mata Alvarez, 2002b;
Vandevivere et al., 2002).
Similar studies have been conducted at the University of
California at Berkeley which demonstrated the feasibility of
integrating solid waste treatment with wastewater treat-ment
and of utilizing excess methanogenic reactor capacity in
conventional wastewater treatment plants. The studies
concluded that sludge methanogenic reactors could be loaded
with a mixture of 80% (w/w) refuse (the highly digestible
fraction of MSW) and 20% sludge. Organic load-ing rates on
the order of 4.8 g VS/l day achieved a TS reduction of about
78% with a corresponding gas produc-tion of about 0.4 l/ g
VS (Diaz et al., 1981).
4. Conclusions
A two-phase, wet anaerobic digestion process was tested at
laboratory scale using mechanically pre-treated MSW as the
substrate. The proposed process diers from others due to the
integration of the MSW and wastewater cycles, which allows
the avoidance of euent recirculation.
Despite the variable characteristics of the substrate, the
process was performed in a successful way adopting
mesophilic conditions (39 LC) and a relatively short hydraulic
retention time (8 days) for the methanogenic reactor, as well
as high values of the organic loading rate. A maximum of 78%
of volatile solid reduction was obtained with an OLR of 33.0
3 Landfilling of Waste: Biogas. Chapman & Hall Publishers, pp. 237 268
kg VS/m d. More-over, the methanogenic reactors quickly
reached (within 1516 days) the optimal conditions for a
stable methano-genic phase. The assessed SGP ranged from
56 to around 100 Nl/kg MSW.
The results obtained show how a proper integration
between the MSW and wastewater cycles could contribute to
improve the perspectives linked to the bio-energetic val-
orization of the biodegradable fraction of MSW.
Acknowledgements
The authors wish to thank the IGAG CNR (Environ-
mental Geology and Geoengineering Institute of the Na-tional
Research Council, Italy) of Cagliari for the precious
cooperation in the activity of chemicalphysical analysis.
References
Bhattacharayya, J.K., Kumar, S., Devotta, S., 2007. Studies on acidifi-cation
in two-phase biomethanation process of municipal solid waste. Waste
Manage.. doi:10.1016/wasman.2006.11.01.
Bingemer, H.G., Crutzen, P.J., 1987. The production of methane from solid
wastes. J. Geophys. Res. 92 (D2), 21812187.
Bolzonella, D., Innocenti, L., Pavan, P., Traverso, P., Cecchi, F., 2003. Semi-
dry thermophilic anaerobic digestion of the organic fraction of municipal
solid waste: focusing on the start-up phase. Bioresource Technol. 86, 123
129.
Bouallagui, H., Touhami, Y., Ben Cheikh, R., Hamdi, M., 2005. Bioreactor
performance in anaerobic digestion of fruit and vegetable wastes. Process
Biochem. 40 (34), 989995.
Buchauer, K., 1998. A comparison of two simple titration procedures to
determine volatile fatty acids in influents to wastewater and sludge
treatment process. Water SA 24 (1), 4956.
Cecchi, F., Traverso, P., Pavan, P., Bolzonella, D., Innocenti, L., 2002.
Characteristics of the OFMSW and behaviour of the anaerobic digestion
process. In: Mata-Alvarez, J. (Ed.), Biomethanization of the Organic
Fraction of Municipal Solid Wastes. IWA Publishing, Department of
Chemical Engineering, University of Barcelona, Spain, pp. 141178.
Chae, K.J., Jang, A., Yim, S.K., Kim, I.S., 2008. The eects of digestion
temperature and temperature shock on the biogas yield from the
mesophilic anaerobic digestion of swine manure. Bioresource Technol. 99,
16.
Cossu, R., Andreottola, G., Muntoni, A., 1996. Modelling landfill gas
production. In: Christensen, T.H., Cossu, R., Stegmann, R. (Eds.),
(ISBN 0 419 19400 2).
Cossu, R., Raga, R., Rossetti, D., 2003. The PAF model: an integrated
approach for landfill sustainability. Waste Manage. 23, 3744.
De Baere, L., 2000. Anaerobic digestion of solid waste: state-of the art. Water
Sci. Technol. 41 (3), 283290.
1808 G. De Gioannis et al. / Waste Management 28 (2008) 18011808

Diaz, L., Savage, G.M., Trezek,
G.J., Golueke, C.G., 1981.
Biogasification of municipal
solid wastes. Transactions of the
ASME Vol. 103, June 1981.
Fernandez, A., Sanchez, A., Font,
X., 2005. Anaerobic co-
digestion of a simulated organic
fraction of municipal solid
wastes and fats of animal and
vegetable origin. Biochem. Eng.
J. 26 (1), 2228.
Gallert, C., Henning, A., Winter, J.,
2003. Scale-up of anaerobic
digestion of the biowaste
fraction from domestic wastes.
Water Res. 37, 14331441.
Garc`a-Heras, J.L., 2002. Reactor
sizing, process kinetics, and
modeling of anaerobic digestion
of complex wastes.. In: Mata-
Alvarez, J. (Ed.),
Biomethanization of the
Organic Fraction of Municipal
Solid Wastes. IWA Publishing,
Department of Chemical
Engineering, University of
Barcelona, Spain, pp. 158.
Hartmann, H., Angelidaki, I.,
Ahring, B.K., 2002. Co-
digestion of the organic fraction
of municipal waste with other
waste types. In: Mata-Alvarez,
J. (Ed.), Biomethanization of
the Organic Fraction of
Municipal Solid Wastes. IWA
Publishing, Department of
Chemical Engineering,
University of Barcelona, Spain,
pp. 181197.
Hartmann, H., Ahring, B.K., 2005.
Anaerobic digestion of the
organic fraction of municipal
solid waste: influence of co-
digestion with manure. Water
Res. 39, 15431552.
Karnchanawong, S., Deesopa, S.,
2004. Resource recovery from
organic fraction of municipal
solid waste by two-phase
anaerobic digestion. In:
Proceedings of the Second
International Conference on
Environmental Concerns:
Innovative Technologies and
Management Options, Xia-men,
1215 October 2004,
conference CD.
Kim, M., Speece, R.E., 2002.
Aerobic waste activated sludge
(WAS) for start-up seeded of
mesophilic and thermophilic
anaerobic digestion. Water Res.
36, 38603866.
Kim, M., Ahn, Y.H., Speece, R.E.,
2002. Comparative process
stability and eciency of
anaerobic digestion; mesophilic
vs thermophilic. Water Res. 36, Cattle Wastes. Tamk-ang J. Sci.
43694385. Eng. 4 (4), 301310.
relationships for process
Maharaj, I., Elefsiniotis, P., 2001. optimization. Bioresource Vandevivere, P., De Baere, L.,
The role of HRT and low Technol. 95, 173 185. Verstraete, W., 2002. Types of
temperature on the acid-phase anaerobic digester for solid
Salminen, E.A., Rintala, J.A., 2002.
anaerobic digestion of wastes. In: Mata-Alvarez, J.
Semi-continuous anaerobic
municipal and industrial (Ed.), Biomethanization of the
digestion of solid poultry
wastewaters. Bioresource Organic Fraction of Municipal
slaughterhouse waste: eect of
Technol. 76, 191197. Solid Wastes. IWA Publishing,
hydraulic retention time and
Mata Alvarez, J., 2002a. loading. Water Res. 36, 3175 Department of Chemical
Fundamentals of the anaerobic 3182. Engineering, University of
digestion process. In: Mata- Barcelona, Spain, pp. 111137.
Sans, C., Mata-Alvarez, J., Cecchi,
Alvarez, J. (Ed.), F., Pavan, P., Bassetti, A., 1995. Vavilin, V.A., Fernandez, B.,
Biomethanization of the Palatsi, J., Flotats, X., 2007.
Acidogenic fermentation of
Organic Fraction of Municipal organic urban wastes in a plug- Hydrolysis kinetics in anaerobic
Solid Wastes. IWA Publishing, degradation of particulate
flow reactor under thermophilic
Department of Chemical conditions. Bioresource organic material: an overview.
Engineering, University of Waste Manage..
Technol. 54, 105 110.
Barcelona, Spain, pp. 118. doi:10.1016/j.wasman.2007.03.
Mata Alvarez, J., 2002b. Anaerobic 02.
Scherer, P.A., Vollmer, G.R.,
digestion of the organic fraction Fakhouri, T., Martensen, S., Veeken, A., Kalyuzhnyi, S., Schar,
of municipal solid waste a 2000. Development of a H., Hamelers, B., 2000. Eect
perspective. In: Mata-Alvarez, methanogenic process to of pH and volatile fatty acids
J. (Ed.), Biomethanization of concentration on anaerobic
degrade exhaustively the
the Organic Fraction of hydrolysis of organic solid
organic fraction of municipal
Municipal Solid Wastes. IWA waste. J. Environ. Eng. 126,
grey waste under
Publishing, Department of 10761081.
thermophilic and
Chemical Engineering, hyperthermophilic conditions. Viturtia, Mtz. A., Mata-Alvarez, J.,
University of Barcelona, Spain, Water Sci. Technol. 41 (3), 83 Cecchi, F., 1995. Two phase
pp. 91105. 91. continuous anaerobic digestion
Pavan, P., Battistoni, P., Cecchi, F., of fruit and vegetable wastes.
Schober, G., Schafer, J., Schmid-
Mata-Alvarez, J., 2000. Two- Res., Cons. Recycl. 13, 257
Staiger, U., Trosch, W., 1999.
phase anaerobic digestion of 267.
One and two-stage digestion of
source sorted OFMSW (organic solid organic waste. Water Res. Zach, A., Binner, E., Latif, M.,
fraction of municipal solid 33 (3), 854860. 2000. Improvement of
waste): performance and kinetic Sosnowski, P., Wieczorek, A., municipal solid waste quality
study. Water Sci. Technol. 41 Ledakowicz, S., 2003. for landfilling by means of
(3), 111118. Anaerobic co-digestion of mechanical biological
sewage sludge and organic pretreatment. Waste Manage.
Perot, C., Amar, D., 1989.
fraction of municipal solid Res. 18, 2532.
Optimization of sludge
anaerobic digestion by waste. Adv. Env. Res. 7, 609 Zhang, B., Zhang, L.-L., Zhang, S.-
separation of hydrolysis- 616. C., Shi, H.-Z., Cai, W.-M.,
acidification and Soyez, K., Plickert, S., 2002. 2005. The influence of pH
methanogenesis. Environ. Mechanicalbiological pre- hydrolysis and acidogenesis of
Technol. Lett. 10, 633644. treatment of waste: state of the kitchen wastes in two-phase
Rao, M.S., Singh, S.P., 2004. art and potentials of anaerobic digestion. Environ.
Bioenergy conversion studies of biotechnology. Acta Biotech- Technol. 26 (3), 329339.
organic fraction of MSW: nol. 22 (34), 271284.
kinetic studies and gas yield- Speece, R.E., 1983. Anaerobic
organic loading
biotechnology for industrial
wastewater treatment. Environ.
Sci. Technol. 17 (9), 427A
467A.
Speece, R.E., 1996. Anaerobic
Biotechnology for Industrial
Wastewater. Archae Press,
Nashville, Tennessee, p. 394.
Stroot, P.G., McMahon, K.D.,
Mackie, R.I., Raskin, L., 2001.
Anaerobic codigestion of
municipal solid waste and
biosolids under various mixing
conditions I: Digester
performance.. Water Res. 35 (7),
1804 1816.
Sung, S., Santha, H., 2001.
Performance of Temperature-
Phased Anaer-obic Digestion
(TPAD) System Treating Dairy