ARTICLE IN PRESS

Renewable Energy 31 (2006) 2017–2024

www.elsevier.com/locate/renene

Technical Note

Anaerobic co-digestion of primary sludge and

the fruit and vegetable fraction of the municipal
solid wastes
Conditions for mixing and evaluation of
the organic loading rate
X. Gómez, M.J. Cuetos, J. Cara, A. Morán, A.I. Garcı́a
University of León, Institute of Natural Resources, Avda de Portugal no 41, León 24071, Spain
Received 7 November 2004; accepted 28 September 2005
Available online 21 November 2005

Abstract

This paper presents the results obtained for the digestion of primary sludge (PS) and co-digestion
of this sludge with the fruit and vegetable fraction of municipal solid wastes (FVFMSW) under
mesophilic conditions. This mixture was prepared with a PS content of 22%. The anaerobic digestion
process was evaluated under static conditions and with different mixing conditions, with good results
being found for the digesters with limited mixing, this representing an energy saving. The results for
co-digestion of mixtures of PS+FVFMSW are better than those obtained from digestion of PS on its
own. Biogas production for co-digestion is much greater thanks to the larger volatile-solid (VS)
content of this feedstock. Nevertheless, biogas yield and specific gas production for the two digestion
processes are similar, with values in the range 0.6–0.8 l g
1 VS destroyed for the first parameter and in
the range 0.4–0.6 l g
1 VS fed for the second. The co-digestion process was also evaluated at different
organic loading rates (OLR) under low mixing conditions, with stable performance being obtained
even when the systems were overloaded.
r 2005 Elsevier Ltd. All rights reserved.

Keywords: Anaerobic digestion; Mesophilic; Primary sludge; Organic fraction of municipal solid wastes
(OFMSW); Mixing conditions

Corresponding author. Tel.: +34 987 291841; fax: +34 987 291839.
E-mail address: ana.garcia@unileon.es (A.I. Garcı́a).

0960-1481/$ - see front matter r 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.renene.2005.09.029
 ARTICLE IN PRESS
2018 X. Gómez et al. / Renewable Energy 31 (2006) 2017–2024

Nomenclature

FVFMSW fruit and vegetable fraction of municipal solid wastes

HM high mixing
HRT hydraulic retention time.
LM low mixing
OLR organic loading rate
PS primary sludge
SGP specific gas production
SC static conditions
TS total solids
VS volatile solids

1. Introduction

Wastewater treatment plants (WWTPs) in most of the developing countries have only
primary sedimentation treatments, while in developed countries many plants are applying
strategies to reduce the amount of waste activated sludge. The amount of sludge generated
is dependent on the size of the WWTP and the treatment option used. Previous to any land
disposal, the sludge must be stabilized. Land application of biosolids is the choice which is
most compatible with sustainable development principles, as long as the limitations set by
local regulations are complied with, in order to avoid any harm to soils.
Municipal solid wastes (MSW) are another type of residue that is affronting more
restrictive legislation with respect to landfill disposal of the biodegradable fraction.
Treatment of this organic fraction is currently carried out through aerobic composting or
anaerobic digestion. These technologies can maximize recycling and recovery of waste
components [1]. Due to the CO2 emissions associated with the aerobic treatments and the
restrictions that will probably apply to these emissions in the near future, anaerobic
digestion will become the treatment alternative that is most in line with sustainable
development. However; one of the problems most frequently found during biological
processing of the organic fraction of MSW is the high C:N ratio of these residues. To
circumvent this problem, several authors have proposed co-digestion of the organic
fraction of MSW, either with sewage sludge from WWTPs or residues from livestock farms
[2–4]. The benefits of co-digestion include: dilution of potential toxic compounds,
improved balance of nutrients, synergistic effects of microorganisms, increased load of
biodegradable organic matter and better biogas yield [5].
Co-digestion is of considerable technical interest, since it allows the use of existing
installations and greatly increases biogas production and the energy produced in co-
generation units. An additional advantage of the process is the obtaining of a valuable
sludge which can eventually be used as a soil amendant after minor treatments [6,7].
The purpose of this research was to compare digestion of primary sludge (PS) as against
co-digestion of this waste together with the fruit and vegetable fraction of municipal solid
wastes (FVFMSW), evaluating production of gas, the influence of mixing conditions and
performance of the system under different organic loading rates.
 ARTICLE IN PRESS
X. Gómez et al. / Renewable Energy 31 (2006) 2017–2024 2019

2. Experimental

The inoculum used for starting up the digesters was obtained from the WWTP of the
city of León (Spain). Four reactors (R1, R2, R3 and R4) with a 3 l capacity were used for
the experiments. The reactors denoted R1 and R2 were used to evaluate the digestion of
PS, and reactors R3 and R4 to evaluate performance in the co-digestion process. At the
beginning of the experiment, all the reactors were operating under the same conditions.
The PS used to feed the systems was also obtained from the León WWTP. It was
concentrated to a solid content of 6% and stored at 4 1C for further use. The mixture for
co-digestion was prepared with an average total solid (TS) concentration of 6%, with the
PS supplying 22% of the 6% of dry solid content. The mixture representing FVFMSW had
the following proportions: banana (10%), apple (10%), orange (10%), cabbage (35%),
potatoes (25%), bread (8%) and paper (2%). This mixture was ground until it reached a
particle size of less than 3 mm.
The performance of the digesters was evaluated with high mixing (HM) conditions, low
mixing (LM) conditions and static conditions (SC). The HM condition corresponded to
continuous mixing achieving complete homogenization of the system with the particles
inside the reactor kept in movement. The electrical stirrer used was set at a speed of
200 rpm. The LM condition corresponded to a decreased speed of stirring, limited to
80 rpm. At this speed, stratification of the particles was possible and the particles at the top
of the reactor were kept in suspension but not in movement. During this stage, the mixing
speed of the stirrer was increased to 200 rpm before wasting and after feeding the reactor to
attain a complete homogenization of the system. Under SCs, the content of the reactor was
mixed only before wasting and after feeding at a velocity of the electrical stirrer of 200 rpm
to achieve total homogenization, with the mixing devices turned off thereafter. All the
samples obtained from the reactors were taken after this complete homogenization and
previous feeding of the systems.
The energy dissipated into the liquid for the stirrers has been calculated using Eq. (1),
which is recommended for stirrers with deflecting plates and for systems with Reynolds
numbers greater than 104 [8]:
P ¼ ðK T  n3  D5  rÞ=ðgc  V Þ. (1)
For the system considered, P ¼ the power dissipated per unit of volume; KT ¼ constant,
for a stirrer with two blades (1.7); D ¼ diameter of the blade (0.05 m); and r ¼ density of
the sludge. In this case with a sludge with a total solid concentration of 6%, the density can
be approximated to water [9] (1000 kg m
3); V ¼ volume of the reactor (0.003 m3). With
these values, the power dissipated at 200 rpm is 1600 w m
3 and at 80 rpm. is 105 w m
3.
The reactors are fed with PS under HM conditions. Subsequently, the stirrer speed was
reduced, so as to evaluate performance of the digestion process with LM conditions, and
reactors R3 and R4 are fed with the mixture for co-digestion. During this second stage of
the study, the organic loading rate (OLR) applied was continuously increased. For reactors
R1 and R2, which were fed with PS during the whole experimentation period, the OLR
was increased. For reactors R3 and R4, the feed was changed to the co-digestion mixture
and the OLR was continuously modified. In accordance to the variations applied to the
reactors, five sub-periods were distinguished within this stage. The sub-periods were
identified with capital letters and were characterized by changes in the type of feed and the
amount of the organic load applied. During this stage, the intention was to study the
 ARTICLE IN PRESS
2020 X. Gómez et al. / Renewable Energy 31 (2006) 2017–2024

performance of the co-digestion process with LM conditions and overloading. Finally, for
the SC, all four reactors operated without mixing, their performance under such conditions
then was evaluated.

3. Analytical

The anaerobic digestion process was evaluated by measuring the following parameters:
total solids (TS), volatile solids (VS), pH and daily biogas production. TS and VS were
measured twice a week in accordance with standard methods [10]. pH was measured twice
a week using a pH meter. Daily biogas production was measured using a liquid
displacement system that was connected to the digester. The biogas volume was corrected
at standard temperature and pressure conditions.

4. Results and discussion

The daily gas production (expressed as l d
1) together with the OLR (expressed as
g VS fed d
1) applied is presented in Fig. 1 for reactors R1 and R2, with these same data
for reactor R3 and R4 shown in Fig. 2. In these same figures, the different stages studied
are also represented. Under the HM conditions, the four reactors were fed with PS. The
specific gas production (SGP) for the four reactors was in the range 0.2–0.4 l g
1 VS fed
and the biogas yield measured per unit of VS destroyed was in the range of 0.4–0.8 l g
1 VS
destroyed. These results are presented in Tables 1 and 2, together with the parameters
evaluating the performance of the digesters. The performance of the digestion process
under LM conditions was evaluated in a second stage for all digesters, with the same feed
applied to all four reactors. It was possible to observe stratification of the reactor content
as a consequence of the reduction in the stirrer speed. Nonetheless, there was no
downgrading of the performance of the reactors. Application of analysis of variances with

Fig. 1. Evolution of daily gas production and organic loading rate applied to systems R1 and R2.
 ARTICLE IN PRESS
X. Gómez et al. / Renewable Energy 31 (2006) 2017–2024 2021

Fig. 2. Evolution of daily gas production and organic loading rate applied to systems R3 and R4.

Table 1
Performance parameters for digesters R1 and R2

Parameters HM LM SC

A B

HRT (d) 47 40 37 37
OLR (g VS fed d
1) 2.5 2.5 3.3 3.6
Biogas yield R1 0.4 0.7 0.7 0.5
(l g
1 VS destroyed) R2 0.7 0.8 0.8 0.6
SGP R1 0.2 0.3 0.4 0.3
(l g
1 VS fed) R2 0.3 0.4 0.5 0.4
Average daily gas prod. (l d
1) R1 0.5 0.8 1.3 1.0
R2 0.8 1.0 1.5 1.3
VS reactor (g l
1) R1 17.5 18.1 19.1 20.0
R2 18.6 18.4 19.1 20.3
% SV/ST reactor R1 53.2 54.0 53.2 53.6
R2 49.6 53.1 52.2 52.1

one factor (one-way ANOVA at a ¼ 0.05) showed that only reactor R1 presented a
significant increase in the SGP and biogas yield values. These results are translated into
energy savings, since the performance of reactors is acceptable under such mixing
conditions. The energy dissipated by the systems under HM conditions is 1600 w m
3 while
for LM, the reduction in the energy required for the system is only 105 w m
3. The results
are in accordance with those reported by Stroot and co-workers [2] who evaluated the
anaerobic digestion process under minimal and continuous mixing, finding similar
 ARTICLE IN PRESS
2022 X. Gómez et al. / Renewable Energy 31 (2006) 2017–2024

Table 2
Performance parameters for digesters R3 and R4

Parameters HM LM SC

A B C D E

HRT (d) 47 40 40 37 37 37 37
OLR (g VS fed d
1) 2.5 2.4 3.0 3.8 9.0 4.1 4.3
Biogas yield R3 0.6 0.6 0.6 0.8 0.6 0.5 0.6
(l g
1 VS destroyed) R4 0.8 0.7 0.8 0.9 0.5 0.5 0.7
SGP R3 0.3 0.4 0.5 0.6 0.4 0.3 0.5
(l g
1 VS fed) R4 0.4 0.4 0.5 0.6 0.4 0.3 0.5
Average daily gas prod. (l d
1) R3 0.7 0.7 1.2 2.3 4.4 1.4 1.7
R4 0.9 0.8 1.4 2.4 3.7 1.3 2.0
VS reactor (g l
1) R3 19.4 18.9 15.9 15.0 14.0 17.6 16.7
R4 19.8 17.6 16.5 13.4 14.9 16.9 16.3
% SV/ST reactor R3 51.3 51.8 52.0 53.8 52.2 58.0 58.5
R4 51.5 53.4 51.0 52.3 52.3 57.2 57.6

performance for the two conditions when low OLR are applied. The results confirm the
assumption that microorganisms involved in the anaerobic digestion process prefer to live
in balanced micro-ecosystems [11], so the process can be maintained with good results
under minimal mixing conditions.
The increase in the organic load under LM conditions from 2.5 to 3.3 g VS fed d
1 for
reactors R1 and R2 brought about greater production of biogas. The SGP value for both
reactors also increased, although the difference was not acute. Analysis of variances
applied to the results indicated that the mean values were significantly different at a 5%
confidence level. No detriment was observed when the OLR was increased, demonstrating
the capacity of the digesters to treat a higher load while maintaining a healthy microbial
population. The percentage of VS in the reactor relative to total solids was kept constant;
although the reactor underwent a slight increase in VS concentration which might be
explained by the reduction in the hydraulic retention time (HRT) associated with the OLR
increases.
Under LM conditions, the feed for reactors R3 and R4 was changed from PS to the co-
digestion mixture. This new feed represented an increment in the OLR applied to the
systems, since 92% of the solid contents in this feed wereVS, while for primary sludge is
only 70%. Due to this higher concentration of VS, the organic load applied increased from
2.4 to 3 g VS fed d
1 (see Table 2, periods LMA and LMB) and as a consequence, an
increase in biogas production can be observed in Fig. 2. The average biogas production for
reactor R3 increased from 0.7 to 1.2 l d
1 and for reactor R4 from 0.8 to 1.4 l d
1.
Additionally, the SGP of these reactors increased slightly with the change in substrate,
demonstrating the capacity of these systems to treat the new load. It is important to note
the lower values for the concentration of VS in the reactor for co-digestion under LM
conditions, as compared to those obtained for the PS systems.
A further increase in the OLR to 3.8 g VS fed d
1, accompanied by a decrease in HRT
had, as its result, greater daily biogas production for both systems, along with an increase
in the biogas yield and the SGP (data presented in Table 2). The results are in agreement
with the suggestion of Yang et al. [12] concerning increases achieved in the rate and yield of
 ARTICLE IN PRESS
X. Gómez et al. / Renewable Energy 31 (2006) 2017–2024 2023

methane formation when the HRT of the systems decreases, as long as a sufficient
methanogenic population can be maintained in the system [12]. The SGP is in the range
0.4–0.6 l g
1 VS fed for all four systems when the HRT is 37 days (the OLR is
3.3 g VS fed d
1 for reactors R1 and R2, and 3.8 g VS fed d
1 for reactors R3 and R4).
These results agree with those reported in literature for different substrates ([5], [13], [14]).
Overloading of reactors R3 and R4 for 8 days with an OLR of 9 g VS fed d
1 permitted a
greater amount of substrate to be available for the microorganisms. This generated a steep
increase in daily biogas production of the systems. The pH of the reactors remained
unchanged at a value around 7.0, but the parameters that measure performance of the
digesters presented a lower value for both reactors.
Average biogas production during the overloading sub-period for reactors R3 and R4
was 4.4 and 3.7 l d
1, respectively. Nevertheless, the biogas yield for these reactors was
appreciably smaller, which would mean that not all the substrate fed to the systems was
converted into methane. It is reasonable to assume that part of the particulate matter
hydrolyzed and turned into volatile fatty acids (VFA), but not converted into methane,
remained in the system. If the concentration of these acids is not high enough to surpass
the buffer capacity of the system, the pH remains unchanged. Although the overloading
did not affect the system pH, the biogas yield and the SGP of the reactors remained low,
even after the OLR applied was reduced. For the next sub-period, the OLR applied to the
systems is 4.1 g VS fed d
1 and the SGP for both reactors was 0.3 l g
1 VS fed. This value is
lower than that achieved before overloading was applied.
The final experimental stage corresponded to digesters performance under SCs. The
content of the reactors was homogenized only before and after feeding, so that
accumulation of solids inside the reactor was avoided and the HRT of the systems was
maintained within the desired range of values. All four reactors studied showed a reduction
in the biogas yield and in SGP, when compared to the values achieved during LM
conditions. For reactors R1 and R2, the biogas yield values are compared with those
obtained during the last sub-period under LM conditions. The absence of mixing
represents a reduction in the biogas yield and in SGP values. The analysis of variance
indicated that the means were significantly different at a 5% confidence level.
For reactors R3 and R4, values for biogas yield and SGP obtained under SCs must be
compared with those obtained under LM conditions before overloading was applied, since
this disturbance led to downgrading of the performance of the reactors. The value of the
biogas yield for both reactors was reduced 0.2 units and with respect to the SGP value, a
reduction from 0.6 to 0.5 l g
1 VS fed was observed. Although, the reduction in SGP is
small, ANOVA at 5% confidence level indicated that means were significantly different.
Kim and co-workers [15] studied the anaerobic digestion process under mesophilic and
thermophilic conditions evaluating the performance of the digester for continuous mixing
and for SCs. Their results showed better performance for the non-mixed configuration at
both temperatures evaluated. They reported values for SGP in the range 0.4–0.7 l g
1 VS
fed. In this study, on the contrary, the performance of the digester is better at LM than
what it is at SCs.

5. Conclusions

The anaerobic digestion process can be performed with low mixing conditions. Systems
digesting the mixture of PS+FVFMSW showed stable performance under these
 ARTICLE IN PRESS
2024 X. Gómez et al. / Renewable Energy 31 (2006) 2017–2024

conditions. The co-digestion of the FVFMSW with primary sludge produced more biogas
than did the digestion of primary sludge, due to the higher concentration of volatile solids
contained in this feed. The parameters measuring the performance of the digestion process
(specific gas production—SGP, and biogas yield) are more or less the same for the two
kinds of feed evaluated.
Under low mixing conditions, the systems were able to absorb the disturbance of a shock
load. The application of a sudden increase in the organic load of the co-digestion systems
led to higher gas production, accompanied by a downgrading of the performance of the
digesters, even though the pH of the system was not affected. The SGP value was reduced
from 0.6 to 0.4 l g
1 VS fed. The absence of agitation resulted in a reduction in SGP for all
the reactors since good contact between the substrate and the microorganisms was
impeded.

References

[1] Mata-Alvarez J, Macé S, Llabrés P. Anaerobic digestion of organic solid wastes. An overview of research
achievements and perspective. Bioresource Technol 2000;74:3–16.
[2] Stroot P, McMahon K, Mackie R, Raskin L. Anaerobic codigestion of municipal solid waste and biosolids
under various mixing conditions— Digester performace. Water Res 2001;35(7):1804–16.
[3] Bujoczek G, Oleskiewicz J, Danesh S, Sparling R. Co-processing of organic fraction of municipal solid waste
and primary sludge—stabilization and disinfection. Environ Technol 2002;23:227–41.
[4] Callaghan F, Wase D, Thayanithy K, Forster C. Continuous co-digestion of cattle slurry with fruit and
vegetable wastes and chicken manure. Biomass Bioenerg 2000;27:71–7.
[5] Sosnowski P, Wieczorek A, Ledakowicz S. Anaerobic co-digestion of sewage sludge and organic fraction of
municipal solid wastes. Adv Environ Res 2003;7:609–16.
[6] Converti A, Del Borghi A, Zilli M, Arni S, Del Borghi M. Anaerobic digestion of the vegetable fraction of
municipal refuses: mesophilic versus thermophilic conditions. Bioprocess Eng 1999;21:371–6.
[7] Poggi-Varaldo HM, Trejo-Espino J, Fernández-Villagómez G, Esparza-Garcı́a F, Caffarel-Méndez S,
Rinderknecht-Seijas N. Quality of anaerobic compost from paper mill and municipal solid wastes for soil
amendment. Water Sci Technol 1999;40(11–12):179–86.
[8] Mc Cabe WL, Smith JC, Harriot P. Basic operations in chemical engineer. In: Agitation and blending of
liquids. 4th ed. New York: McGraw Hill; 1991. p. 242–97.
[9] Tchobanoglous G, Burton FL. Ingenierı́a sanitaria. Tratamiento, evacuación y reutilización de aguas
residuales. In: Diseño de instalaciones para el tratamiento y vertido del fango. interamericana de España, SA:
Metcalf & Heddy, INC., McGraw Hill; 1995. p. 865–80.
[10] APHA-AWWA-WPCF. Standard methods for the examination of water and wastewater. New York:
American Public Health Associaton; 1997.
[11] Lettinga G. With anaerobic treatment approach towards a more sustainable and robust environmental
protection. Proceeding 10th world congress AD, vol. 1. 2004. p. 2–12.
[12] Yang K, Yu Y, Hwang S. Selective optimization in thermophilic acidogenesis of cheese-whey wastewater to
acetic and butyric acids: partial acidification and methanation. Water Res 2003;37:2467–77.
[13] Murtor M, Björnsson L, Mattiasson B. Impact of food industrial waste on anaerobic co-digestion of sewage
sludge and pig manure. J Environ Manage 2004;70:101–7.
[14] Bouallagui H, Ben-Cheikh R, Marouani L, Hamdi M. Mesophilic biogas production from fruit and
vegetable waste in a tubular digester. Bioresource Technol 2003;86:85–9.
[15] Kim M, Ahn Y- H, Speece R. Comparative process stability and efficiency of anaerobic digestion; mesophilic
vs. thermophilic. Water Res 2002;36:4369–85.