Renewable Energy 134 (2019) 241e246

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Start-up of a pilot scale anaerobic reactor for the biogas production

from the pineapple processing industries of Belgium
Tesfayesus Zinare Mamo a, Abhishek Dutta b, S. Anuradha Jabasingh c, *
a
Biochemical Engineering Division, School of Chemical and Bio Engineering, Addis Ababa Institute of Technology, Addis Ababa University, Addis Ababa,
Ethiopia
b
Faculty of Engineering Technology, KU Leuven, Campus Greop T, Leuven, Belgium
c
Process Engineering Division, School of Chemical and Bio Engineering, Addis Ababa Institute of Technology, Addis Ababa University, Addis Ababa, Ethiopia

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

Article history: In this study, a pilot scale continuous stirred-tank reactor (CSTR) was used to investigate the startup of an
Received 5 March 2018 anaerobic system, followed by a bench scale test to analyze the biogas production. Wastewater from
Received in revised form pressing and crushing leaves of the pineapple crop was used as a substrate in the CSTR. During the study
18 October 2018
period, the organic loading rate (OLR) was gradually increased and parameters such as volatile fatty acids
Accepted 14 November 2018
Available online 15 November 2018
(VFA), temperature, pH, bio-gas production and chemical oxygen demand (COD) were carefully moni-
tored. The result of the bench scale test showed the potential of the wastewater for producing the biogas.
The study indicated an OLR of 3.8 g COD/L/d, and a maximum biogas production of 3.3 L/d, with an
Keywords:
Anaerobic digestion
average of 2.8 L/d. The COD removal efficiency reached its highest peak with a value of 64.8%, when the
Biogas OLR was 3.43 g COD/L with the hydraulic retention time (HRT) of 20 days. Based on the measurements
Organic loading rate made throughout the test period, there were no indications of the presence of toxic elements in the
Volatile fatty acid wastewater, which could inhibit the anaerobic degradation process. This study is an attempt to provide a
Pine apple industry vigorous solution for the pineapple industry wastewater scenario, thereby aiming at sustainable energy
Continuous stirred-tank reactor production.
© 2018 Elsevier Ltd. All rights reserved.

1. Introduction demand in the Fyffes, Del Monte Foods, Dole Food Company and
Chiquita. Belgium is an important focal point for the baby pineapple
Pineapple (Ananas comosus) is a tropical edible fruit belonging (for edible purposes and decoration) and conserved pineapple (for
to the Bromeliaceae family. They come in the varieties known as the canning, frozen application and dry fruit preparation) exports [2].
Smooth Cayenne, Sugarloaf, Victoria and the MD2 variety. MD2, Hence, the processing and packaging industries in Belgium process
widely known for its sweetness, vitamin C content and longer shelf the imported pineapples and export them at a larger scale to the EU.
life is produced by the Del Monte as a new hybrid pineapple variety. This approach had led to the creation of huge pineapple process
It also constitutes over 80% of the European imports [1]. Belgium industry wastewaters, causing an environmental nuisance. This
ranks fourth among the countries which exports the highest dollar radically led to the primary establishment of anaerobic wastewater
value worth of pineapples during 2016, next only to Costa Rica, treatment plants in Belgium.
Netherlands, Philippines and this is about $105.4 million. This ac- Start-up and successful operation of an anaerobic wastewater
counts for about 5.3% of the 94.2% of global pineapples exports in treatment is a difficult and an extremely sensitive process, which
2016 by value. Belgium finds their export variety pineapples from requires a regular follow up and accurate monitoring techniques
the developing countries and this accounted for about 95% of im- [3,4]. Fluctuations in pH, temperature, alkalinity, organic loading
ports. Belgium is one of the leading importers and distributors of rates (OLRs), Volatile Fatty Acids (VFA), influent characteristics, and
the fruit in the European Union (EU). These fruits are then re- Hydraulic Retention Time (HRT), leads to instability in the reactors,
exported to the countries in the European Union for their during the start-up phase [4]. Approved design and control of the
parameters are essential to maximize the process efficiency, pre-
vent failure and thereby increase stability [5].
* Corresponding author. Anaerobic digestion (AD) is a biological process through which
E-mail address: [email protected] (S.A. Jabasingh). organic material is broken down in an oxygen depleted

https://doi.org/10.1016/j.renene.2018.11.058
0960-1481/© 2018 Elsevier Ltd. All rights reserved.
242 T. Zinare Mamo et al. / Renewable Energy 134 (2019) 241e246

environment involving a group of microbial consortium respon- The characteristics of the raw wastewater plays a crucial role in
sible for catalyzing a complex series of biochemical reaction, which the selection of reactors. Solid waste and sludge are treated in a
in turn stabilizes the organic matter, thereby producing methane continuous flow reactor (CSTR), a suspended growth system, with
and carbon dioxide as the main component [6]. Anaerobic diges- an intermittent or continuous mixing action [19]. Biomass sus-
tion, responsible for the carbon cycle in the ecosystems occurs in pended with the effluent is removed, so as to equalize the solid
wetlands, fields, animal intestines, and sediments [7]. Overall retention time (SRT) with the hydraulic retention time (HRT).
anaerobic digestion is classified into hydrolysis, acidogenesis, ace- Henceforth, the CSTRs are operated at a high HRT between 10 and
togenesis and methanogenesis. During the startup phase, due to 20 days with an organic loading rate of 1e5 g COD/L.d, encouraging
the imbalance in the microbial population, the overall anaerobic the avoidance of methanogen scrubbing [20]. The Up flow Anaer-
conversion presented seems to be less efficient. This imbalance is obic Sludge Blanket (UASB), a high-rate anaerobic reactor treat the
due to the adaptation of the microorganisms like abrupt temper- soluble organic waste efficiently by removing solids with low total
ature, pH drift, toxic substance accumulation, and excess feed [7]. solids (3e4%) contents [3]. The present study investigates the
The acidification of a reactor through a reactor overload is one of startup of a thermophilic anaerobic CSTR, which is meant for pro-
the causes for the backdrop in the process [8]. The chief reason for cessing wastewater from the pineapple industry, by gradually
this is the build-up of VFA's produced by acidogenic and acetogenic increasing the organic loading rate and monitoring the variables
bacteria. The methanogenic bacteria are less tolerant, when like VFA, temperature, pH, bio-gas production and COD removal. In
compared to the acidogenic bacteria. The methane-formers have a addition, a pilot scale CSTR and bench scale tests were performed to
low growth-rate compared to the acidogenic bacteria. Hence, the analyze the biogas production.
acidogenic bacteria grow rapidly and create acidic conditions in the
reactor. This increase in the acidogenic bacteria reduces the activity 2. Materials and methods
of methanogenic bacteria and eventually cause a complete inhibi-
tion of activity. This imbalance should be addressed, or else will 2.1. CSTR pilot plant set up
lead to a reactor failure, during the initial or start-up phases [9].
Low buffering capacity anaerobic reactors rely on pH, partial alka- The pilot plant CSTR setup was built at the Waterleau R&D
linity and VFA's for process imbalance indication. Highly buffered center in Wesplaar, Belgium. The CSTR, employed in this study was
systems rely only on VFA's for reliable process monitoring [10]. a cylindrical vessel made out of stainless steel, with a total volume
Many studies focus on the regulation of VFA concentrations, as it of 20 L operated under thermophilic (50e55
C) conditions. The
is the initial measurable indication of the stress in an anaerobic reactor was heated by a warm water circulation jacket. The influent
treatment system [3,10,11]. The pH is more easily and quickly was prepared in batches and stored in a container and it is
determined than the alkalinity of the contents in the reactor, constantly mixed with a magnetic stirrer to prevent the sedimen-
nevertheless it is an indication of events that have already occurred tation of the solid particles. The reactor has one side outlet for the
in the reactor, whereas, the alkanity changes in the reactor, indicate effluent removal. During feeding, the effluent leaves the reactor
the present scenario of the events prevailing in the reactor and through the effluent outlet to maintain a constant volume within
provides a few corrective measures which could probably be the the reactor. The influent feed and the effluent removal were per-
solution to the reactor functioning. Alkalinity decreases in the formed by peristaltic pumps (Watson Marlow Sci323, UK). All the
reactor, through the production of volatile acids by the acid formers gas produced during the test is measured by a biogas counter
[12]. The VFA concentration in an AD process, particularly depends (Ritter TGI, Germany 120 L/h, 50 mbar).
on the type of substrate introduced as an influent to the reactor and
this concentration is in the range of 200e2000 mg/L acetic acid [13].
2.2. Bench-scale tests
The fast rapid increase in the VFA concentration is responsible for
the process upset, rather than the absolute concentration [14]. But
The bench scale test measures the ability of the anaerobic sludge
the VFA levels cannot be always used as an indicator of the state of
to digest the substrate and convert it into biogas. The test lasted for
an anaerobic process. This is because different systems have their
20 days, and finally the biodegradability was measured. The dry
own levels of VFA's, which are suitable for those scenarios. The
solids, were determined by weighing the empty aluminium dish
conditions which cause instability in one reactor may be favorable
(Wdish), followed by transferring the homogenous well mixed
for another. Henceforth, the volatile acid to the alkalinity ratio (VA/
sludge solution into the aluminium dish. The total weight of the
Alk) is used as an indicator to test the performance of the reactor.
aluminium dish and the sludge was then measured (Wtotal). The
The range of acceptable volatile acid to alkalinity ratio (VA/Alk) is
aluminum dish along with the sludge are subjected to drying in an
0.1e0.2. This tolerable ratio could be attained by regulating the
oven at 104
C for 24 h to ensure the evaporation of water. After
volatile acid concentration, alkalinity concentration, or both. Vari-
24 h, the aluminum dish with the sludge is placed in a desiccator
ation in the influent to the reactor also facilitates to lower the vol-
and then weighed (W1). The aluminum dish with the sludge is
atile acid-to-alkalinity ratio. If the influent cannot be varied, then
further placed in an oven at 550
C for 24 h to allow the oxidation of
the use of chemicals for alkalinity adjustment is mandatory [15].
the organic compounds into CO2 and H2O. The aluminum dish with
The AD process is carried out at either the mesophilic (25e40
C)
the sludge is again cooled in the desiccator and weighed after
or thermophilic (50e65
C) ranges of temperatures [16,17]. Ther-
cooling down (W2). The dry solids (DS) and the volatile solids (VS)
mophilic reactors handle higher loading rate, yield higher methane
are then calculated.
production and higher substrate degradation. Higher temperature
shortens the growth of microorganisms and decrease the required W1  Wdish
HRT, furthermore, it facilitates the degradation of the organic ma- DS ¼ (1)
Wtotal  Wdish
terial. The mesophilic processes operate with robust microbial
consortia which tolerate greater change in the environment, more
W1  W2
stable and easier to maintain, but the mesophilic processes have VS ¼ (2)
longer HRT and low biogas production rates [18]. Wtotal  Wdish
 T. Zinare Mamo et al. / Renewable Energy 134 (2019) 241e246 243

A 1 L graduated flask was used to perform the test. From the conducted at thermophilic temperatures (50e55
C) and sustained,
calculated VS, the right amount of sludge was weighed to get a during the process (Fig. 1a) [24]. The average temperature of the
concentration of 2gVS/L for a volume of 500 ml. Two buffering influent was maintained at an average value of 11.8
C. The tem-
solutions were added; 1 g NaHCO3/L and 2 g KH2PO4/L, to match the perature of the effluent fluctuated significantly due to the drastic
pH of the flask contents to that of the reactor (pH 7.9). The flask was changes in the temperature of the surrounding, but on observation,
then fed with the influent, closed with a manometric measuring this had no effect on the reactor, where the average temperature
head (Oxitop®-C) on a magnetic stirrer and stored at 35
C for 20 d. was 50
C. Anaerobic processes are extremely vulnerable with
Anaerobic degradation of the wastewater samples caused the for- regards to pH [25]. Inhibition on methanogens occurs, when the pH
mation of biogas (CH4 and CO2), this resulted in an increase in the drops below 6.5 or exceeds 8.2. The results indicate that pH values
pressure in the headspace of the flask. The pressure build-up is for the reactor sludge over the study time were within the range of
recorded by the measuring head and automatically stored 7.5e7.8. The average pH of the influent and effluent was 3.7 and 7.9,
throughout the entire period of the experiment. respectively. The average pH of the effluent was slightly higher than
the average pH of the reactor. The increase in the pH could be due to
2.3. Wastewater characteristics CO2 evaporation and the inverse correlation that exists between the
temperature and pH [26].
The wastewater used as the influent was collected from the
pressing and crushing processes carried out on the leaves of the
pineapple in the pineapple processing industries in Belgium and 3.2. Volatile fatty acids, organic loading rate, hydraulic retention
Costa Rica. It is a liquid waste with a high solids concentration, with time and flow rates
a chemical oxygen demand of 75e87.8 g/L, and total suspended
material (TSS) of 20.2e32.5 g/L. The sludge inoculum for the CSTR VFA is an important indicator for the performance of the
was an anaerobic sludge from a thermophilic reactor in Valmass, anaerobic reactor (Fig. 2a). It indicates a correlation between the
Brazil. The characteristics of the influent are listed in Table 1. COD methanogenic bacteria and the COD break down. In this study, VFA
was determined according to the HACH-Lange test cuvettes and a analysis was carried out every day, in order to ensure the balance
HACH-Lange spectrophotometer for both the influent and effluent between the production of volatile fatty acids in the reactor and the
samples. VFA was determined by centrifuging a portion of the consumption of the same. The higher the VFA concentration, the
effluent sample at 60 rpm for 30 min, by performing titration, ac- more its inhibitory effect on the methanogenic bacteria, hence, the
cording to the KAPP method on the supernatant [21,22]. Organic reactor feed should be cautiously monitored. Based on the results
loading rate (OLR) in kg COD/L. day is the amount of organic ma- obtained, throughout the study, when the OLR was increased in the
terial that the reactor processes daily per litre of the active volume system, it resulted in an increase of VFA, but then after a while, the
of the reactor [23], and it is calculated using the formula, VFA returned back to its initial value, due to the balance of acid
formers and acid consumers [21,27].
CODinfluent  Flow rate The OLR (Fig. 2b) was increased by 10%, when the value of VFA
OLR ¼ (3) was the same for two consecutive days. The initial concentration of
Reactor volume
the residue from the inoculation sludge was high, but this fairly
Where CODinf is the influent COD in mg/L. Hydraulic retention
dropped and increased after 10 days of operation. OLR determines
time, is the time measured in days, during which the influent is
the amount of organic matter added to the reactor and creates
retained in the reactor and is calculated as,
awareness on the performance of the microorganism. A higher OLR
demands more bacteria to act on stabilizing the organic matter and
Reactor volume
HRT ¼ (4) this causes problems in the system, if the system does not adapt.
Flowrate
One problem with increasing the OLR too fast, is the behavior of
acidogenic bacteria, this is because, they act early in the digestion
process, multiplying and producing acids, too rapidly. Meanwhile, a
3. Results and discussion scenario is created, whereby the methanogenic bacteria take long
time to increase their population and remain inefficient to feed on
3.1. CSTR operation the acids produced by acidogenic bacteria, all at the same pace. This
would, in turn, cause the VFA to rise, inhibiting more methanogenic
The CSTR employed in this study was operated on a continuous bacteria and thereby halting the digestion process [28]. When the
mode for 55 days (Table 2). During the experiment, the influent was OLR is increased, the HRT of the substrate has to be taken into
added gradually into the reactor to facilitate the adaptation of the greater account. During the period of this study, the OLR was
sludge, without being stressed and the results were obtained dur- gradually increased from 1 g COD/L.d to 3.8 g COD/L.d. With this
ing the first 8 weeks of the testing period. increase, it was observed that the HRT (Fig. 2c) dropped from 67
Temperature and pH were monitored every 24 h during the days to 20 days. HRT indicates the average length of time, a sub-
reactor operation (Fig. 1a and b). Temperature plays a vital role in strate retains, itself in the reactor.
the degradation of the organic particles in the waste which can OLR was then altered by either decreasing or increasing the flow
significantly affect the conversion, kinetics, stability and biogas per day. Depending on the measured COD of a batch of influent, the
production. The experiments in the anaerobic reactor were flow per day was determined in order to get the required OLR. This

Table 1
Influent characteristics.

Parameters pH tCOD (mg/L) CODfiltrate (mg/L) NHþ

4 (mg/L) SO2
4 (mg/L) TSS (mg/L) VSS (mg/L)

Values 3.6 78780 47240 18.5 652 25200 17500

244 T. Zinare Mamo et al. / Renewable Energy 134 (2019) 241e246

Table 2
Effluent characteristics over a study period of 55 days.

Parameters pH tCOD (mg/L) sCOD (mg/L) COD removal (%) VFA (mg/L) Alkalinity (mg/L) Biogasproduced (mL/d)

Values 7.9 38950 20805 57 992 10523 28856

run on the same organic loading rate for a longer period of time, to
allow for the gradual adaptation of the methanogenic bacteria to
the OLR increase [10]. The data obtained for the COD removal (%)
were tested on the best fit curve (exponential, linear, polynomial,
power and logarithmic) (Fig. 3aee). The results were better fitted,
when expressed by the second order polynomial with R2 as 0.8046.
The trend of the output clearly indicates the increase in the COD
removal until 45 days, after which the COD removal did not sub-
stantially change.
Fig. 1. Effect of time on the (a) measured temperature and (b) measured pH.

3.4. Biogas production

The production of biogas could be attributed to the complicated

interaction between the multiple microorganism or between the
individual members of the microbial consortia. During the opera-
tion, the performance of the reactor was evaluated based on the
amount of biogas produced. When the amount of biogas produced
was less than the previously measured values, there could be
problems associated with either the degradation process or with
the reactor configuration. As observed from the Fig. 4a, the amount
of biogas produced initially, was due to the remaining COD in the
inoculation sludge. The concentration was found to drop quickly
during the initial stages and then, increased after 13 days of reactor
operation. The OLR was gradually increased from 1gCOD/L.d to 3.78
gCOD/L.d, with a gradually increase of 5e10%. This increased, the
biogas production possibly, due to the COD removal and the

Fig. 2. Measured concentration of (a) volatile fatty acids (VFA), (b) organic loading rate
(OLR), (c) hydraulic retention time (HRT) and (d) flow rate, over time (days).

was carried out by measuring the COD of the new batch of the
influent and correlating it with previously measured COD. Hence, if
the COD of the new batch of influent was higher than the previous
one, it was more relevant to note that the OLR increased, while the
flow rate decreased. The measure of the flow rate is shown in the
Fig. 2d. The test was initiated with a flow of 0.3 L/d and was
stopped, when the flow remained at 1 L/d. The HRT was found to
decrease as the flow rate increased. Hence, at the flowrate of 1 L/d,
the HRT reached 20 d, and the flow was confined to 20 d. From
these studies, 20 d was observed as the convenient HRT for the
design of the CSTR [11].

3.3. COD and its removal

COD concentrations measured during tests show that the

average COD for the influent during the testing period was
78780 mg/L. The measured amount fluctuated over the time, due to
the discrepancies caused during the preparation of different
batches. The average COD concentration in the effluent was
observed to be 20805 mg/L. The COD removal (%) linked to the
biogas production reached its highest peak with a value 64.8%. Fig. 3. Best fit curve analysis for the COD removal (%) vs time (days) (a) exponential,
Higher removal efficiencies were expected, when the reactor was (b) linear, (c) polynomial, (d) power, (e) logarithmic.
 T. Zinare Mamo et al. / Renewable Energy 134 (2019) 241e246 245

Fig. 4. (a) Measured and theoretically calculated amount of Biogas (a) based on the
total COD (tCOD), (b) based on the soluble COD (sCOD).

consumption of VFA in the system, by the methanogenic bacteria.

The biogas concentration reached a maximum on 47th day and on
the 49th and 55th day, the biogas production dropped by 60%. This
was due to the clogging of the influent, while feeding. This insists
careful monitoring of the system, as suggested. Nevertheless, the
biogas production stabilized after the 45th day. During the period
Fig. 5. Measured dry solids (DS) and volatile solids (VS) over the time (days).
of study, a total of 3.3 L/d of measured biogas was produced
(Fig. 4b). As observed from the Fig. 4b, the measured biogas and the
theoretically calculated values mimic each other in a certain way. 3.6. Bench scale test
On the other hand, in the Fig. 4a, a slight difference between the
two values is visible, due to the methodology adopted in the The bench scale test provides a means to understand the pro-
calculation of the theoretical biogas production. The formula uses duction of biogas on a small scale. The mechanism of the mea-
the COD of the influent and effluent, daily influent flow and makes a surement is similar to the method adopted during the biogas
correction for the expected amount of methane in the biogas. In production. During the test, the pressure inside the bottle was
this case, 55% methane was used (empirical data from WaterLeau) found to increase, and the measuring head was observed to read
[27]. In a CSTR reactor, sludge and influent, become completely the change. As this sludge was directly taken from the CSTR reactor,
homogeneous and the separation of both is impracticable, due to the test was started directly without letting the sludge to adapt to
the higher solid concentration. If a COD measurement was made on the wastewater. On the first day of the test (before the dotted line in
an effluent sample from the reactor, the obtained result would Fig. 6), the pressure build-up was higher, and this was quite normal,
contain the COD of the bacteria, which would then lead to an due to the concentration of CO2 in the sludge [19]. The pressure
under-estimation of the biogas production (Fig. 4a). To overcome build up dropped fairly quickly and increased after 2 days (after the
this, the sample was centrifuged and filtered to remove the larger dotted line in Fig. 6). From then onwards, the biological degradation
solid particles and the sludge. In doing so, a large fraction of COD began, thereby increasing the pressure of the graduated flask, at a
that was not broken down would be discarded. The biogas calcu- fairly constant rate [22]. The COD removal of the test result, gave an
lation, therefore over estimates the amount of biogas that should average of 67.7%, which roughly correlates with the pilot CSTR,
have been produced. Hence, the real production is shown in the
Fig. 4b, as in this case, the adopted methodology, the separation of
the sludge and the influent provides an accurate measure of the
biogas produced.

3.5. Dry solids and volatile solids

The dry solids (DS) and volatile solids (VS) concentrations in the
reactor are important factors to be considered and studied. The DS
value indicates the amount of organic and inorganic solids within
the reactor, whereas, the VS only accounts for the organic part of
the sludge. These values will, thus indirectly illustrate, if the
wastewater stream poses any chronic forms of toxicity. For
example, if the DS concentration increases, but the VS decreases, it
means that the organic part of the sludge is being replaced by the
inorganic sludge and eventually, the COD removal will cease
(Fig. 5). The reactor was initially started with 4.7% of volatile solids
which roughly corresponded with 47 g of volatile solids. During the
study period, the concentration dropped due to several factors,
which includes, the adaptation of the sludge and the hydraulic
retention time [28]. After 45 days, however, the concentration
stabilized, indicating an increase in the organic fraction, discarding
the possibility of the presence of chronic toxic elements in the
wastewater. Fig. 6. Biogas production based on pressure measurement over the 20 days.
246 T. Zinare Mamo et al. / Renewable Energy 134 (2019) 241e246

which had a COD removal of 64.8%. In addition, the production of [7] B.K. Ahring, M. Sandberg, I. Angelidaki, Volatile fatty-acids as indicators of
process imbalance in anaerobic digesters, Appl. Microbiol. Biotechnol. 43
the biogas was observed even after 20 days. This indicates the
(1995) 559e565.
major role imparted by the short HRT of the pilot CSTR, during the [8] M. Myint, N. Nirmalakhandan, R.E. Speece, Anaerobic fermentation of cattle
breakdown of the residual COD. The results also signify the absence manure: modeling of hydrolysis and acidogenesis, Water Res. 41 (2) (2007)
of any chronic toxic elements in the pineapple industry wastewater 323e332.
[9] T. Mechichi, S. Sayadi, Evaluating process imbalance of anaerobic digestion of
and the potential of same for generating the biogas. olive mill wastewaters, Process Biochem. 40 (1) (2005) 139e145.
[10] I.H. Franke-Whittle, A. Walter, C. Ebner, H. Insam, Investigation into the effect
4. Conclusions of high concentrations of volatile fatty acids in anaerobic digestion on
methanogenic communities, Waste Manag. 34 (11) (2014) 2080e2089.
[11] A. Babaee, J. Shayegan, Effects of organic loading rates (OLR) on production of
The bench scale test proved the efficient digestion of the pine- methane from anaerobic digestion of vegetable waste, in: Proceedings of the
apple industry wastewater. Based on the measurement of this test, World Renewable Energy Congress, Linko €ping, Sweden, 2011, pp. 8e13.
[12] J.R. Ryan, Clostridium, peptostreptococcus, bacteroids, and other anaerobes,
there was no indication that the wastewater contained any toxic in: K.J. Ryan, C.G. Ray (Eds.), Sherris Medical Microbiology: an Introduction to
elements that could inhibit the anaerobic degradation process. The Infectious Diseases, fourth ed., McGraw-Hill, USA, 2004.
results obtained during the start-up period of the pilot scale test, [13] D.P. Ho, P.L.D. Jensen, D.J. Batstone, Methanosarcinaceae and acetate oxidizing
pathways dominate in high-rate thermophilic anaerobic digestion of waste
showed a COD removal efficiency of 64.8%, at an OLR of 3.5 gCOD/ activated sludge, Appl. Microbiol. Biotechnol. 79 (2013) 6491e6500.
L.d. Scheduling a longer lab scale test could let us know the exact [14] C. Vahlberg, E. Nordell, L. Wiberg, A. Schnürer, Method for Correction of VFA
OLR for the CSTR reactor, to achieve a maximum production of the Loss in Determination of Dry Matter in Biomass, SGC Rapport, Gas Technology
Centre, Malmo € , Sweden, 2013.
biogas. The concentration of the dry solids and the volatile solids
[15] P. Worm, N. Müller, C.M. Plugge, A.J.M. Stams, B. Schink, Syntrophy in
were stable as indicated by an increase in the organic fraction. The Methanogenic Degradation in (Endo)symbiotic Methanogenic Archaea,
volatile fatty acid concentration remained around its optimum but Microbiology Monographs, J.H.P.Hackstein, vol. 19, Springer, Berlin, Germany,
increased rapidly, when an OLR was increased. The balance be- 2011, pp. 143e173.
[16] A.M. Ziganshin, J. Liebetrau, J. Pro € ter, S. Kleinsteuber, Microbial community
tween the acid formers and acid consumers in the running reactor structure and dynamics during anaerobic digestion of various agricultural
is critical to achieve the desired results. The biogas followed the waste materials, Appl. Microbiol. Biotechnol. 97 (2013) 5161e5174.
theoretically calculated biogas volume, quite well. To further assess [17] S.A. Jabasingh, S. Varma, P. Garre, Production and purification of cellulase from
Aspergillus nidulans AJSU04 under solid-state fermentationusing coir pith,
the biogas production potential of the pineapple industry waste- Chem. Biochem. Eng. Q. 28 (2014) 143e151.
water, a longer lab-scale test is needed. Longer lab-scale tests [18] D. Karakashev, D.J. Batstone, I. Angelidaki, Influence of environmental con-
provide a clear significant information related to the wastewater ditions on methanogenic compositions in anaerobic biogas reactors, Appl.
Environ. Microbiol. 71 (2005) 331e338.
digestion. [19] M. Klocke, P. Ma €hnert, K. Mundt, K. Souidi, B. Linke, Microbial community
analysis of a biogas-producing completely stirred tank reactor fed continu-
Acknowledgments ously with fodder beet silage as mono-substrate, Syst. Appl. Microbiol. 30
(2007) 139e151.
[20] I. Angelidaki, W. Sanders, Assessment of the anaerobic biodegradability of
The authors would like to acknowledge the encouragement and macropollutants, Rev. Environ. Sci. Biotechnol. 3 (2004) 117e129.
research support provided by the Faculty of Engineering Technol- [21] K. Buchauer, A comparison of two simple titration procedures to determine
volatile fatty acids in influents to waste-water and sludge treatment pro-
ogy at the Katholik University of Leuven. The comments and rec-
cesses, WaterSA 24 (1998) 49e56.
ommendations of the anonymous reviewers and the Editor-in- [22] O. Lahav, B.E. Morgan, Titration methodologies for monitoring of anaerobic
Chief, Soteris Kalogirou are greatly acknowledged. digestion in developing countries, a review, J. Chem. Technol. Biotechnol. 79
(12) (2004) 1331e1341.
[23] A.D. Eaton, L.S. Clesceri, A.E. Greenberg, M.A.H. Franson, Standard Methods for
References the Examination of Water and Wastewater, twentieth ed., American Public
Health Association, Washington DC, 1998.
[1] MOA, L’agriculture Biologique, Ministry of Agriculture, Brussels, February, [24] L. Leven, A.R.B. Eriksson, A. Schnürer, Effect of process temperature on bac-
2001. terial and archaeal communities in two methanogenic bioreactors treating
[2] J. Poirier, The Agri-food Market in Belgium, the Canadian Embassy in Belgium, organic household waste, FEMS Microbiol. Ecol. 59 (2007) 683e693.
August, 1999. [25] P.L. McCarty, Anaerobic waste treatment fundamentals, Publ. Works 95 (9)
[3] M.E. Griffin, K.D. McMahon, R.I. Mackie, L. Raskin, Methanogenic population (1964) 107e112.
dynamics during start-up of anaerobic digesters treating municipal solid [26] K.H. Hansen, I. Angelidaki, B.K. Ahring, Anaerobic digestion of swine manure:
waste and biosolids, Biotechnol. Bioeng. 57 (3) (1998) 342e355. inhibition by ammonia, Water Res. 32 (1998) 5e12.
[4] M.H. Gerardi, The Microbiology of Anaerobic Digesters, Wastewater Micro- [27] A.G. Hashimoto, Effect of inoculum/substrate ratio on methane yield and
biology Series, Wiley-Interscience, Hoboken, New Jersey, 2003. production rate from straw, Biol. Wastes 28 (1989) 247e255.
[5] R.M. Smith, Chemical Process:design and Integration, John Wiley & Sons, [28] V. Moset, A. Cerisuelo, S. Sutaryo, H.B. Møller, Process performance of
Chichester, England, 2005. anaerobic co-digestion of raw and acidified pig slurry, Water Res. 46 (2012)
[6] H. Qiang, D.L. Lang, Y.Y. Li, High-solid mesophilic methane fermentation of 5019e5027.
food waste with an emphasis on iron, cobalt, and nickel requirements, Bio-
resour. Technol. 10 (2012) 21e27.