Available online at www.sciencedirect.

com

Bioresource Technology 99 (2008) 54185426

Comparison of aerobic and anaerobic degradation of municipal

solid waste in bioreactor landlls
A. Suna Erses, Turgut T. Onay, Orhan Yenigun *
Bogazici University, Institute of Environmental Sciences, Bebek, 34342 Istanbul, Turkey

Received 25 August 2007; received in revised form 2 November 2007; accepted 6 November 2007
Available online 21 December 2007

Abstract

Two landll bioreactors were operated under aerobic and anaerobic conditions in a thermo-insulated room at a constant temperature
of 32 C. Reactors were lled with 19.5 kg of shredded synthetic solid waste prepared according to the average municipal solid waste
compositions determined in Istanbul and operated under wet-tomb management strategy by using leachate recirculation. Aerobic con-
ditions in the reactor were developed by using an air compressor. The results of experiments indicated that aerobic reactor had higher
organic, nitrogen, phosphorus and alkali metal removal eciencies than the anaerobic one. Furthermore, stabilization time considerably
decreased when using aerobic processes with leachate recirculation compared to the anaerobic system with the same recirculation scheme.
2007 Elsevier Ltd. All rights reserved.

Keywords: Aerobic; Anaerobic; Bioreactor; Landll; Leachate recirculation

1. Introduction tal impacts. In order to achieve these goals, the Landll

Directive adopted by the EU requires that the amount of
Municipal solid waste (MSW) landlls are generally biodegradable MSW going to the landll should be
operated by conventional landlling techniques, where reduced by 25% until 2002, 50% by 2005 and 75% by
anaerobic conditions are created within the landll waste. 2010 (Council of the European Union, 1999). In spite of
However, decomposition of solid waste may cause environ- extensive laws and regulations to reduce the waste quan-
mental problems if emission of landll gas is not controlled tity, landlling is expected to be the most commonly
and landll leachate seeps down to groundwaters. Under employed disposal method worldwide since it is a compar-
anaerobic conditions within a landll site, stabilization atively simple and economic way for solid waste disposal.
process is slow and may increase the potential risks to Moreover, landlling is the ultimate disposal method for
human health and the environment (Hudgins and March, wastes that cannot be recovered.
1998). According to Kruempelbeck and Ehrig (1999), the Conventional sanitary landlls consist of cells and lifts
long term environmental impact caused by municipal solid with liners, drains, gas vents, leak detection systems with
waste (MSW) landlls may last for centuries. As a result of intermediate and nal covers. On the other hand, due to
this, more stringent regulations for monitoring landlls, in inadequacies of conventional waste management practices,
addition to their design and operation, are needed. Recent extensive lab-scale and full-scale investigations were carried
European regulations on waste management are primarily out to answer the essential questions of low-emission land-
intended to reduce the use of landlling, mainly because of lling. Increasing attention is being given to the enhance-
problems generated by the above-mentioned environmen- ment of waste stabilization by leachate recirculation to
reduce the time required for waste degradation, improve
*
Corresponding author. Tel.: +90 212 3596946; fax: +90 212 2575033. leachate quality and enhance the rate of gas produc-
E-mail address: [email protected] (O. Yenigun). tion. Bioreactor landll systems as a modication of

0960-8524/$ - see front matter 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2007.11.008
 A.S. Erses et al. / Bioresource Technology 99 (2008) 54185426 5419

conventional landll with the addition of leachate recircu- Precipitation was simulated by feeding the reactors with dis-
lation were developed to minimize environmental impacts tilled water periodically. Daily gas production was mea-
through optimizing waste degradation. Here, leachate is sured by using wet gasmeters (Shinagawa Corporation).
collected, stored and reinjected back into the landll to
promote in situ anaerobic biological degradation. The 2.2. Reactor loadings
advantages of leachate recirculation have been demon-
strated by many researchers that performed numerous Simulated landll reactors were loaded with 19.5 kg of
lysimeters and eld tests (Pohland, 1980; Tittlebaum, shredded and compacted solid waste mixture. The syn-
1982; Kinman et al., 1987; Doedens and Cord-Landwehr, thetic solid waste mixture representing typical solid waste
1989; Otieno, 1994; Townsend et al., 1996; Chugh composition of Istanbul consisted of 45% organic material
et al., 1998; El-Fadel, 1999; Pohland and Kim, 1999, (food + garden), 14.5% paper, 9.5% plastic, 5.6% textile,
2000; Onay and Pohland, 1998; San and Onay, 2001; Erses 3.8% glass, 2.2% metal, 4.4% ceramic, 15% other materials
and Onay, 2003). (dust, wood, brick, miscellaneous) by weight.
On the other hand, as a new perspective on landlling
solid waste, aerobic landll technology has been evalu- 2.3. Reactor operations
ated over the last few years to rapidly stabilize and
detoxify the waste, reduce methane gas, volatile organic In this study, one reactor was operated as aerobic to bet-
compounds and odor emissions as well as eliminate o ter understand the eect of aeration on solid waste degra-
site leachate treatment needs (Purcell, 2000a,b; Read dation. The other one was operated as an anaerobic
et al., 2001; Jacobs et al., 2003; Kim, 2005; Cossu bioreactor which represented conventional sanitary landll
et al., 2003). The results of these studies are summarized conditions. The reactors were constructed under wet-tomb
in Table 1. In spite of the recent practices on aerobic management strategy. Throughout the study period, 1 L of
landlling, the eect of in situ aeration has not been collected leachate in storage bottles was recycled to reac-
understood from the view point of the attenuation mech- tors once per week. Moreover, 500 mL/week distilled
anism of contaminants during the stabilization period. water, corresponding to an equivalent of 20 cm/year rain-
Until today all eld research has taken place using exist- fall, was added to the reactors. In the aerobic reactor, the
ing landll sites, so that many of the sites that are run as air inlet at the bottom of the reactor was connected to an
full-scale aerobic landlls have used waste that has been air pump operating at 2.2 L/min for 5 h in a day to main-
anaerobic for some time. There are questions yet to be tain aerobic conditions from start to end.
answered about the behaviour of waste that has never
been anaerobic (Read et al., 2001; Rich et al., 2007). 2.4. Experimental analysis
The purpose of this study was to compare the opera-
tional parameters and attenuation mechanisms of carbon, Leachate and gas samples were monitored on a regular
nitrogen, phosphorus and alkali metal and their removal basis to understand and compare the degree of waste stabil-
in conventional anaerobic and aerobic bioreactor landlls ization in each reactor. Leachate samples were analyzed for
monitored under laboratory conditions. chemical oxygen demand (COD), total organic carbon
(TOC), biochemical oxygen demand (BOD), pH, alkalin-
2. Materials and methods ity, conductivity, total kjhedal nitrogen (TKN), ammo-
nia-nitrogen, orthophosphate, chloride and alkali metals
2.1. Reactor congurations (Na, K, Ca, Mg). All these analyses were performed
according to Standard Methods (APHA AWWA, 1992).
The process of landll stabilization is strongly depended Gas production and its composition were measured daily.
on temperature. Two laboratory scale PVC reactors were The gas composition in the headspace (O2, N2, CO2 and
constructed in a thermo-insulated room kept at a constant CH4) was analyzed using an HP 6850 gas chromatograph
mesophilic temperature of 32 C to establish a favorable (Carboxen 1010 plot column 30 m  0.53 mm) equipped
growth environment for the microorganisms in the reactors. with a thermal conductivity detector (TCD).
Each reactor had a diameter of 0.35 m and a length of 1 m
(about 96 L). The reactors were equipped with three ports; 3. Results and discussion
one port was used for drainage and sampling while the
other two ports were used to collect gas samples and add 3.1. The impact on bioreactor landll emissions: leachate and
liquid. Leachate was stored in 18 L plexi-glass bottles at gas
the bottom of reactor for leachate recirculation. Aerobic
conditions in the reactor were maintained by using an air 3.1.1. pH and alkalinity
compressor. Leachate recycle was used to provide transport The change in leachate pH in both the reactors is given
mechanism for microorganisms and stabilization products in Fig. 1. Initially, pH values of both reactors were the
for further conversion. To determine daily temperature same (pH 5.83). Both aerobic and anaerobic reactors
changes in the bioreactors, thermocouples were installed. started o in acidic conditions in the beginning of the
5420 A.S. Erses et al. / Bioresource Technology 99 (2008) 54185426

Operation day
12 10000

21 months
11

365 days
120 days

Alkalinity (mg/L as CaCO3)

10
9 1000
8

pH
50%, reduced leachate organic strength by up to 70% and leachate volumes by greater than 80%. Waste
settlement over the trial period of 918 months was also noted to increase by an additional 910%
7
6 100
5
4
More than 90% of COD, BOD and TOC concentration decreased within 100 days 3 10
0 100 200 300 400 500 600 700
Time (Days)
Anaerobic Bioreactor pH Aerobic Bioreactor pH
Anaerobic Bioreactor Alkalinity Aerobic Bioreactor Alkalinity

Fig. 1. Leachate pH values and alkalinity concentrations for the

anaerobic and aerobic bioreactors.
Rapid and marked oxidation of organics and nitrogen

experiment. However, the acidic pH in aerobic reactor

45% VS destruction, 2-year degradation timescales

increased to neutral values in a few days and was measured

between 7.5 and 8.0 after day 35 until the very end. Two
Odour have been reported to be less pungent

phases (acidic and methanogenic or alkaline) of pH in

the anaerobic reactor were observed during the test period.
Initial acidic pH values in the anaerobic bioreactor
15% biodegradation over 2 years

increased sharply and reached to 7.0 on day 446 after the

onset of methanogenic activity and stayed between 7.0
and 7.5 during this phase. As it can be seen from this gure,
the pH of the aerobic reactor was more alkaline than the
pH from anaerobic reactor due to CO2 stripping by air; a
decrease in CO2 leads to a decrease in carbonic acid
(H2CO3) and bicarbonate ion concentrations (HCO3 ) con-
suming H+ ions (Kim, 2005). This observation is also con-
Results

rmed with other studies in literature. The range of pH of

aerobic reactors has been reported as 7.09.0 (Stessel and
Murphy, 1992; OKeefe and Chynoweth, 2000; Agdag
Converted form anaerobic

and Sponza, 2004; Kim, 2005).

100 nL/h continuously

Adequate alkalinity, or buer capacity, is necessary to

maintain a stable pH in the digester for optimal biological
70120 mL/min

activity. An alkalinity level ranging from 1000 to 5000 mg/

to aerobic

L as CaCO3 was recommended by Tchobanoglous and

Air ow

Burton (1979). The measured leachate alkalinity for the

aerobic and anaerobic reactors are also provided in
Fig. 1. Initial alkalinity of aerobic and anaerobic reactors
Bioreactor, 12.8 kg waste
Bioreactor, 8.14 kg waste

were 3219 and 5768 mg/L as CaCO3. In the aerobic reac-

The results of related studies in literature

tor, alkalinity rst increased to its highest value of

(theoretical site) 4 ha
Landll area 6.5 ha

4500 mg/L as CaCO3 and then began to decrease on day

77 and reached to 660 mg/L as CaCO3 at the end of exper-
Characteristic
Landll area

Landll area

iment. Theoretical calculations revealed that there was no

requirement for excess alkalinity. On the other hand, in
the anaerobic reactor, leachate alkalinity changed between
5500 and 7500 mg/L as CaCO3 and remained relatively
constant during the acid formation phase. After transition
to methanogenic conditions, pH values increased and total
Jacobs et al.
(2000a,b)

Cossu et al.
Kim (2005)
Read et al.
Reference

alkalinity concentrations tended to decrease because meth-

(2001)

(2003)

(2003)
Table 1

Purcell

anogens utilized the available VFA as substrate. Total

alkalinity of the anaerobic reactor decreased to 3800 mg/
 A.S. Erses et al. / Bioresource Technology 99 (2008) 54185426 5421

L as CaCO3 at end of the experiment. This was also con- tant carbon source for methanogenesis in landlls, however
rmed by theoretical alkalinity computations. it is not an easily biodegradable material under anaerobic
conditions. In fact, cellulose and hemicellulose, whose
3.1.2. COD, TOC and BOD5 half-lives are about 15 years, contribute to 90% of total
Chemical oxygen demand (COD) and total organic car- methane produced (Swarbrick, 2001). Hydrolysis of these
bon (TOC) were measured as indicators of leachate organic complex organics in the presence of oxygen enhance bio-
strength. Leachate COD and TOC concentrations for aer- degradation rate and convert to organic carbon in the
obic and anaerobic reactors are presented in Fig. 2. The ini- waste mass to mostly carbon dioxide and water, with stabi-
tial leachate COD concentrations of aerobic and anaerobic lized humic material remaining. Waste decomposition
reactors were 17,900 and 38,000 mg/L, respectively. The takes place within a few years under aerobic and controlled
COD concentrations in the aerobic reactor increased to conditions instead of several decades (Hudgins and Harper,
19,237 mg/L and decreased rapidly after pH was stabilized. 1999). The results of the present study conrmed those of
The COD concentrations on days 77 and 374 were deter- previous studies and indicated that approximately 90% of
mined as 1596 and 678 mg/L, respectively. On the other COD removal was complete by days 72 and 462 for aerobic
hand, COD concentration in the anaerobic reactor and anaerobic reactors, respectively.
increased to 61,973 mg/L from 38,022 mg/L due to rapid It is important to note that aeration enables signicantly
release and hydrolysis of complex organics from solid faster biodegradation of organic matter. The results of the
waste to the leachate and then remained at the same high present study are similar to the ndings of Cossu et al.
concentrations due to the accumulation organic acids until (2003), who reported high COD values in the anaerobic
day 290. After the onset of methanogenic conditions that reactor (20,000 mg/L) compared to the aerobic reactor
was conrmed by gas composition, rst, COD concentra- (800 mg/L) after 120 days of operation. Carbon conversion
tions began to decrease slowly to 30,400 mg/L until day as a means of measuring the success of the aeration of the
433 and then declined rapidly to below 900 mg/L at the waste mass has been investigated by Ritzkowski et al.
end of experiment. COD and TOC concentrations showed (2006). They observed a considerable reduction of leachate
similarity (Fig. 2). Initial TOC concentrations for the COD and TOC concentrations after approximately 20 days
reactors are given as follows; aerobic reactor, 1438 mg/L; of aeration.
anaerobic reactor, 17,990 mg/L. TOC concentrations BOD5 indicated similar trend with COD and TOC
decreased to 218 mg/L in the aerobic reactor on day 374 removal (Fig. 3). The BOD5 concentrations of the aerobic
and to 290 mg/L in the anaerobic reactor on day 630. reactor decreased rapidly while BOD5 concentrations of
TOC concentration in the anaerobic reactor declined in the anaerobic reactor decreased slowly and then followed
accordance with the progression of microbially mediated rapid reduction after onset of methanogenic conditions.
stabilization processes also conrmed the increase in gas BOD5 to COD ratio is often used to assess the biodegrad-
production and high methane content of the biogas. ability of the organic matter in leachate, and thus to under-
Landlled wastes are dominated by organic material stand the degree of landll stabilization (Alvarez-Vazquez
comprising (typically) 50% cellulose, 15% lignin, 10% hemi- et al., 2004). In old stabilized landlls, the BOD5/COD
cellulose, 5% protein as well as starch, pectin, and other ratio is below 0.10 (Kjeldsen et al., 2002). A low BOD5/
soluble sugars (Barlaz, 1992). Cellulose is the most impor- COD suggests that leachate is low in biodegradable organic
carbon and relatively high in hard-to-biodegrade organic
compounds such as humic compounds. In this research,
100000 100000 initial high BOD5/COD ratio decreased from 0.57 to 0.03

100000
10000 10000
TOC (mg/L)
COD (mg/L)

10000
BOD5 (mg/L)

1000
1000 1000

100

100 100 10
0 100 200 300 400 500 600 700 Anaerobic Bioreactor
Time (Days) Aerobic Bioreactor
1
Anaerobic Bioreactor COD Aerobic Bioreactor COD 0 100 200 300 400 500 600 700
Anaerobic Bioreactor TOC Aerobic Bioreactor TOC Time (Days)

Fig. 2. Leachate COD and TOC concentrations for the anaerobic and Fig. 3. Leachate BOD5 concentrations for the anaerobic and aerobic
aerobic bioreactors. bioreactors.
5422 A.S. Erses et al. / Bioresource Technology 99 (2008) 54185426

at the end of experiment in the aerobic reactor (Fig. 4). On 25000 1600
the other hand, relatively high BOD5/COD ratios were
obtained from the anaerobic reactors. Initial BOD5/COD

Cumulative Gas Production (L)

20000
ratio was about 0.7, and release of organics increased this

Daily Gas Production (mL)

1200
ratio to 0.80.9 after day 200 after which it decreased to
0.45 at around day 460, nally reaching 0.05 at the end 15000
of experiments. The values proposed by Kjeldsen et al.
(2002) were 0.58 and 0.06 for the acid phase and the meth- 800

anogenic phase, respectively. Borglin et al. (2004) found 10000

that BOD5/COD ratio of aerobic and anaerobic reactors
were 0.03 and 0.45, respectively, after 365 days. 400
5000

3.1.3. Biogas
Cumulative and daily gas production of the anaerobic 0 0
0 100 200 300 400 500 600 700
reactor have been given in Fig. 5. The changes in gas con- Time (Days)
centrations of the anaerobic reactors is presented in Fig. 6.
Fig. 5. Daily and cumulative gas production from the anaerobic
In the start-up period, organic matter in the reactors are bioreactor.
hydrolyzed and the produced organic acids cease gas pro-
duction. During the intermediate anaerobic degradation
stage, methanogenic bacteria slowly start to appear and
gas production rate increases (Murphy et al., 1995). In this 100
O2
study, the anaerobic reactor exhibited some retardation in 90
N2
terms of gas production, because the acidogenesis period 80 CO 2
Gas Composition (%)

lasted a long time, conrmed by high COD concentrations 70 CH4

and low pH values in the reactor. Increase in gas pro- 60

duction was observed around day 330 and reached to 50
20 L/day on day 445 which was the highest value of gas 40
production. Also methane content of biogas was about 30
48% during these days (Fig. 6). Methane and carbon diox- 20
ide are the principal gases produced during the decomposi- 10
tion of organic fraction of waste under anaerobic 0
0 100 200 300 400 500 600 700
conditions. In the anaerobic reactor, methane, at rst, Time (Days)
was not observed due to the acidogenic conditions until
day 165 (Fig. 6). After this day, methane began to appear Fig. 6. Gas composition of the anaerobic bioreactor.
and increased to 50% by composition. Gas composition
of the anaerobic reactor during methanogenic phase was
49% CH4, 43% CO2, 0.5% O2 and 7.5% N2, respectively. by summing the observed daily gas production values
Maximum 9 L of methane was generated on day of highest within the experimental period. The cumulative gas pro-
gas production while 80% of COD had been removed. This duction in the reactor was recorded as 1497 L. Cumulative
result is in good agreement with the ndings of San and methane and carbon dioxide productions were 690 and
Onay (2001). Cumulative gas production was calculated 609 L, respectively. Methane yield was determined as
158 L/kg dry solid waste. Other studies have found that
1 the methane yield from landlls is in the range 60170
Anaerobic Bioreactor
Aerobic Bioreactor
L/kg of dry refuse (El-Fadel et al., 1996). Barlaz and
0.8 Ham (1989) calculated theoretical maximum yield of meth-
ane as 373 L CH4/kg wet solid waste.
BOD5 /COD

0.6 As can be seen in Fig. 5, after conversion of easily

degradable organic substances, gas production decreased
0.4 and gas utilization was no longer possible. In conventional
landlls, gas production inevitably slows down until its use
0.2 as a source of energy is no longer economically viable, but
after this point it is still produced in large enough quanti-
0 ties to require treating before it can be emitted to the atmo-
0 100 200 300 400 500 600 700 sphere (Rich et al., 2007). The uncontrolled CH4 emissions
Time (Days)
from landlls contribute to the greenhouse eect. By short-
Fig. 4. BOD5/COD ratio for the anaerobic and aerobic bioreactors. ening the ensuing unprotable period, aeration of a landll
 A.S. Erses et al. / Bioresource Technology 99 (2008) 54185426 5423

can save on gas treatment costs. The introduction of aero- NH3-N is a signicant long-term pollution problem in
bic conditions throughout the waste mass reduced methane landlls (Burton and Watson-Craik, 1998; Onay and Poh-
level rapidly and kept it at a minimum (Read et al., 2001; land, 1998; Price et al., 2003; He et al., 2007). The increased
Ritzkowski et al., 2006). In one other related work, Leikam ammonia concentrations intensify the toxicity of the leach-
et al. (1997) indicated that methane concentrations in land- ate. Pohland et al. (1987) stated that concentrations
ll gas could be reduced from 60% to 1015% in 710 days between 200 and 1500 mg/L have shown to have no
by air injection. adverse eects on anaerobic process, concentrations rang-
Under aerobic conditions some of the organic carbon is ing from 1500 to 3000 mg/L were shown to have inhibitory
the waste mass is emitted as carbon dioxide and some eects at higher pH levels, and concentrations above
would be stabilized as carbonates. The intensity of the car- 3000 mg/L were very toxic. However Calli et al. (2005b)
bon conversion depends greatly on ambient conditions have shown that NH3-N levels up to 6000 mg/L could be
around the microorganisms (e.g. oxygen concentration, tolerated in anaerobic bioreactors. In this study, NH3-N
pH, temperature, water content) and the availability of bio- concentrations in the anaerobic reactor do not reach inhib-
logically degradable organic substances in the landlled itory or toxic levels (Fig. 7). On the other hand, under aer-
waste. In this study, gas composition of the aerobic reactor obic conditions, the ammonia-nitrogen is oxidized to nitrite
was 4.95% CO2, 14.41% O2 and 80.64% N2, at the end of and nitrate by nitrifying bacteria (nitrobacter and nitroso-
the experiment. monas). TKN and ammonia-nitrogen concentrations in the
leachate of the aerobic reactor indicated the same decreas-
ing trend. Initial TKN and ammonia-nitrogen concentra-
3.1.4. Nitrogen and phosphorus
tions were measured as 620 and 399 mg/L, respectively.
Nitrogen which has potential to pollute water and soil is
TKN concentration decreased to 140 mg/L on day 108
another major constituent in the leachate. The TKN and
and 42 mg/L on day 178 and stayed constant throughout
ammonia-nitrogen concentrations for aerobic and anaero-
the study. NH3-N concentrations, on the other hand,
bic reactors are given in Fig. 7. Initial concentrations of
decreased to 132 mg/L on day 116 and 14 mg/L on day
TKN and ammonia-nitrogen in both reactors were found
175 and continued to decline slightly until reaching to
to be similar, indicating uniformity in waste composition
5 mg/L at the end. These results agree with the ndings
in the reactors. As a result of decomposition of organic
of Cossu et al. (2003), in which the ammonium concentra-
matter containing nitrogen, initial NH3-N concentration
tion declined to 56 NH3-N mg/L after 120 days of opera-
increased from 390 mg/L to a maximum of 1064 mg/L
tion in the aerobic reactor.
NH3-N for the anaerobic reactor. TKN concentrations
The initial concentrations of orthophosphate in the
were also observed between 840 and 1590 mg/L. Protein
aerobic and anaerobic reactors (Fig. 8) were 250 and
fraction of biodegradable wastes release ammonia-nitrogen
168 mg/L, respectively. High initial orthophosphate con-
and they may accumulate in the leachate to concentrations
centration in the aerobic reactor decreased to 50 mg/L on
of 10006000 mg/L in anaerobic systems (Burton and Wat-
day 33. This decrease continued until day 200 and then,
son-Craik, 1998, 2001; Calli et al., 2005a). The recircula-
orthophosphate concentrations stayed at a constant value
tion practice in the anaerobic reactor reintroduces
of about 12 mg/L. On the other hand, orthophosphate con-
ammonia to the system, keeping its value almost constant
centration of the anaerobic reactor increased to 410 mg/L
throughout experiment. Several researches indicated that
as a result of the hydrolysis of organic phosphorus com-
pounds and then began to decrease to 236 mg/L on day
10000 10000 113, 82 mg/L on day 216, 40 mg/L on day 446 and reached
to 25 mg/L at the end of the experiment. The decline in

1000 1000
1000
TKN (mg/L)

NH3 (mg/L)

Anaerobic Bioreactor Aerobic Bioreactor

100 100

100
PO4 (mg/L)

10 10

10
1 1
0 100 200 300 400 500 600 700
Time (Days)
1
Anaerobic Bioreactor TKN AerobicBioreactor TKN 0 100 200 300 400 500 600 700
Anaerobic Bioreactor NH3 AerobicBioreactor NH3 Time (Days)

Fig. 7. Leachate TKN and ammonia-nitrogen concentrations for the Fig. 8. Leachate orthophosphate concentrations for the anaerobic and
anaerobic and aerobic bioreactors. aerobic bioreactors.
5424 A.S. Erses et al. / Bioresource Technology 99 (2008) 54185426

orthophosphate concentration may have been the results of decreased to 56 mg/L at the end. In the anaerobic reactor it
orthophosphate assimilation by microorganisms and declined to 28 mg/L.
dilution. There are numerous studies in which researchers have
reported metal concentrations from full-scale landlls, test
cells and laboratory studies under anaerobic conditions
3.1.5. Chloride, metals and conductivity (Yenigun et al., 1996; Flyhammar et al., 1998; Kruempel-
Chloride was monitored as a conservative tracer in beck and Ehrig, 1999; Revans et al., 1999; Erses and Onay,
order to estimate the dilution and washout eects. The 2003). The release of metal into leachate is generally low in
chloride concentrations of aerobic and anaerobic reactor anaerobic landlls in spite of acidogenesis. This is because
indicated similar decreasing trend because the same metals tend to form hydroxides or undergo suldation in
amount rain water was added and same recirculation ratio the anaerobic phase and the majority of these compounds
was applied to the reactors. Decrease in chloride concentra- are not readily soluble (Rich et al., 2007). Once aerobic
tions caused by the dilution eect of water used for rainfall conditions are re-established in the nal aerobic phase met-
simulation. Initial chloride concentrations were 2584 and als are still immobile.
2536 for the aereobic and anaerobic reactors, respectively. The conductivity of a leachate reects its total concen-
These concentrations decreased to 1340 and 538 mg/L, tration of ionic solutes and is a measure of the solutions
respectively, at the end of the study. Theoretical chloride ability to convey an electric current. In both the reactors,
calculations also exhibited the same decreasing trend in the change in leachate conductivity with time followed a
time. similar trend. Along with degradation, initial conductivity
The change in alkali metals (Na, K, Ca, Mg) in the aer- values (11.78 mS/cm) in the aerobic reactor and
obic and anaerobic reactors was also observed. Both (16.39 mS/cm) in the anaerobic reactor increased to 16.71
sodium and potassium concentrations of the reactors and 23.70 mS/cm as their highest values respectively and
showed the same behavior. Initial Na concentrations of declined slowly to 8.28 and 9.28 mS/cm throughout the
the aerobic and anaerobic reactors were 1344 and experiment due to precipitation and the washout of easily
1548 mg/L, respectively. At the end, Na concentrations mobilized ions such as metals, chloride and sulfate because
decreased to 879 mg/L in the aerobic reactor and of rainfall simulation and sampling. For aerobic reactor
1050 mg/L in the anaerobic reactor. K concentrations of theoretical dilution of conductivity highly agreed with
aerobic and anaerobic reactors declined from 1532 to experimental observations. For the anaerobic reactor theo-
1305 mg/L and from 1354 to 441 mg/L, respectively. Sig- retical conductivity dilution gradient was smaller than that
nicant precipitation for Na and K has not been observed of experimental, this is because of the mobilization of ions
and they do not participate in any considerable way in under acidogenic conditions which was not considered in
complexation reactions and they exist in only one oxida- theoretical computations.
tion state (+1) in the landll systems (Pohland et al.,
1993) and thus, their behavior was expected to be that of
a conservative tracer throughout the study. Both sodium 3.2. The impact on bioreactor landll stabilization
and potassium showed evidence of dilution eects that
was conrmed by chloride concentrations in the reactors. Table 2 shows the characteristics of waste in the bioreac-
On the other hand, Ca and Mg concentrations of the reac- tors at the end of experiment. Settlement or subsidence of
tors indicated variability because they are only slightly MSW is a key parameter that distinguishes between aero-
more chemically complicated than Na and K, with the pri- bic and anaerobic bioreactor landlls. As it is shown in
mary dierence being that both have available and poten- Table 2, aerobic bioreactor technology enhances the rate
tial precipitants. In the case of calcium, the potential and extent of subsidence. Higher degree of settlement was
precipitant is carbonate (CaCO3) while magnesium is sub- achieved in the aerobic bioreactor where settlement was
ject to precipitation as the hydroxide (Mg(OH)2). Ca con- about 37% at 374 days. Anaerobic bioreactor exhibited set-
centrations were measured as 66 mg/L for the aerobic tlement about 5% after 630 days. Several previous studies
reactor and 3190 mg/L for the anaerobic reactor, at the have measured subsidence. The Mountain View Landll
beginning of study. Ca concentration of the aerobic reactor Project in California measured 1315% settlement with
increased slightly and stayed constant during the experi- leachate recirculation over a period of 4 years as compared
ment and then reached to 109 mg/L at the end. On the with 812% settlement in the control reactor (Buivid et al.,
other hand, the high initial Ca concentration of anaerobic
reactor decreased to 1069 mg/L on day 412. After the Table 2
establishment of methanogenic conditions and thus with Waste characteristics from aerobic and anaerobic bioreactors
increase in pH values, Ca concentrations decreased rapidly Anaerobic bioreactor Aerobic bioreactor
to 126 mg/L and stayed constant throughout experiment. Settlement (%) 5 37
Initial Mg concentrations were 52 and 340 mg/L for aero- Volatile solids, VS (%) 47 38
bic and anaerobic reactors, respectively. In aerobic reactor, Wet weight (kg) 15.73 15.31
Dry weight (kg) 5.66 5.19
like Ca concentrations, Mg increased to 143 mg/L and later
 A.S. Erses et al. / Bioresource Technology 99 (2008) 54185426 5425

1981). Stessel and Murphy demonstrated 1830% settling  Concentrations of Ca and Mg in the anaerobic reactor
for aerobic bioreactor in laboratory over a 65-day experi- were signicantly greater than those of the aerobic
ment period. Settlement during actual landll operations reactor.
will impact the performance of the nal surface grade, sur-  The waste settlement in the aerobic reactors occurred
face drainage, roads, gas collection piping system and faster than the waste settlement in the anaerobic reac-
leachate and air distribution piping system. Settlement tors. Higher degree of settlement was achieved in the
impacts can be readily accommodated by the project aerobic bioreactor where settlement was about 37% at
design. Since settlement in aerobic landlls will be largely 374 days. Anaerobic bioreactor exhibited settlement
complete soon after landll closure, long-term maintenance about 5% after 630 days.
costs and the potential for fugitive emissions will be  Comparing the anaerobic and aerobic reactors, the aer-
avoided. obic reactor was more eective in terms of COD reduc-
For comparison, the MSW was weighed as it was tion and waste stabilization. While aerobic reactor was
removed from the reactors. The average weight of wet completed within 374 days, anaerobic reactor was oper-
waste in the anaerobic bioreactor was 15.73 kg, corre- ated for 630 days.
sponding to a dry weight of 5.66 kg, giving a 19% loss of
MSW. For the aerobic bioreactor, the average wet weight Overall, the results of this study showed that aeration in
was 15.31 kg, corresponding to a dry weight of 5.19 kg, municipal solid waste landlls is a viable management
which is a 21% loss in MSW mass. option for the accelerated stabilization of a waste matrix.
The amount of initial VS in the anaerobic and aerobic These results can be used to compare aerobic and anaero-
bioreactors on dry weight basis were 82% and 84%, respec- bic landlls for life-cycle assessment. Aerobic landlls are
tively. Volatile solid was analyzed by loss on ignition at expected to reduce the cost for monitoring and leachate
550 C. Reduction of organic substance in the aerobic bio- treatment due to rapid waste decomposition and low
reactor was 46% while in the anaerobic bioreactor this was organic carbon. Further studies should be extended to
35%. These values correlate well with waste weight and set- investigate aeration rates and enhancement of solid waste
tlement observations. stabilization to render practical solutions in sustainable
landll management.

4. Conclusions Acknowledgements

It may not prove surprising that waste can rapidly The authors acknowledge the nancial support provided
decompose in aerobic conditions. It could be also antici- by Bogazici University Research Fund by Research Project
pated that methane concentrations decreased as a result No. 04 Y102.
of air addition. However, in addition to these common
consequences, the results of this study, which could possi- References
bly occur in large scale landll, may be helpful for landll
operators and engineers when choosing aerobic and anaer- Agdag, O.N., Sponza, D.T., 2004. Eect of aeration on the performance
obic systems. of a simulated landlling reactor stabilizing municipal solid wastes.
Based upon the experimental results obtained during the Journal of Environmental Science and Health Part A-Toxic/Hazard-
experimental study, the following conclusions are ous Substances & Environmental Engineering 39, 29552972.
Alvarez-Vazquez, H., Jeerson, B., Judd, S.J., 2004. Membrane bioreac-
provided: tors vs. conventional biological treatment of landll leachate: a brief
review. Journal of Chemical Technology and Biotechnology 79, 1043
 The anaerobic bioreactor landll shows the highest lev- 1049.
els of all pollutional parameters, with high concentra- APHA, AWWA, WPCF, 1992. Standard Methods for the Examination of
tions of COD, TOC, BOD, ammonia, phosphorus and Water and Wastewater, 18th ed. APHA, Washington, DC.
Barlaz, M.A., 1992. Microbial, chemical and methane production char-
alkali metals in the leachate. acteristics of anaerobically decomposed refuse with and without
 Aeration of the waste mass produces a rapid oxidation leachate recycling. Waste Management and Research 10, 257267.
of organics and nitrogen when compared to the anaero- Barlaz, M.A., Ham, R.K., 1989. Mass balance analysis of anaerobically
bic bioreactor landll. More than 90% of COD was decomposed refuse. Journal of Environmental Engineering ASCE 115,
decomposed within 70 days in the aerobic bioreactor 10881102.
Borglin, S.E., Hazen, T.C., Oldenburg, C.M., Zawislanski, P.T., 2004.
compared to 462 days in the anaerobic bioreactor. Comparison of aerobic and anaerobic biotreatment of municipal solid
 While ammonia concentrations were increasing through- waste. Journal of Air and Waste Management Association 54, 815
out the methanogenic phase of the anaerobic bioreactor, 822.
their concentrations in the aerobic reactor decreased Buivid, M.G., Wise, D.L., Blanchet, M.J., 1981. Fuel gas enhancement by
rapidly. Ammonia concentrations in aerobic reactor controlled landll of municipal solid waste. Resources and Conserva-
tion 6, 320.
decreased to 5 mg/L at the end of experiments, their con- Burton, S.A.Q., Watson-Craik, I.A., 1998. Ammonia and nitrogen uxes
centrations in anaerobic reactor reached to about in landll sites: applicability to sustainable landlling. Waste Man-
600 mg/L. agement and Research 16, 4153.
5426 A.S. Erses et al. / Bioresource Technology 99 (2008) 54185426

Burton, S.A.Q., Watson-Craik, I.A., 2001. Ammonia uxes in landlls: International Waste Management and Landll Symposium, Cagliari,
current research on the acceleration of the decomposition of landlled Italy.
refuse. In: Proceedings Sardinia 2001, 18th International Waste Murphy, R.J., Jones, D.E., Stessel, R.I., 1995. Relationship of microbial
Management and Landll Symposium, Cagliari, Italy. mass and activity in biodegradation of solid waste. Waste Manage-
Calli, B., Mertoglu, B., Inanc, B., Yenigun, O., 2005a. Eects of high free ment and Reseach 13, 485497.
ammonia concentrations on the performances of anaerobic bioreac- OKeefe, D.M., Chynoweth, D.P., 2000. Inuence of phase separation,
tors. Process Biochemistry 40, 12851292. leachate recycle and aeration on treatment of municipal solid waste in
Calli, B., Mertoglu, B., Inanc, B., Yenigun, O., 2005b. Methanogenic simulated landll cells. Bioresource Technology 72, 5566.
diversity in anaerobic bioreactors under extremely high ammonia Onay, T.T., Pohland, F.G., 1998. In situ nitrogen management in
levels. Enzyme and Microbial Technology 37, 448455. controlled landlls. Water Research 32, 13831392.
Chugh, S., Clarke, W., Pullammanappallil, P., Rudolph, V., 1998. Eect Otieno, F.A.O., 1994. Stabilization of solid waste through leachate
of recirculated leachate volume on MSW degradation. Waste Man- recycling. Waste Management and Research 12, 93100.
agement Research 16 (6), 564573. Pohland, F.G., 1980. Leachate recycle as landll management option.
Cossu, R., Raga, R., Rossetti, D., 2003. The PAF model: an integrated Journal of the Environmental Engineering Division 106, EE6.
approach for landll sustainability. Waste Management 23, Pohland, F.G., Kim, J.C., 1999. In-situ anaerobic treatment of leachate in
3744. landll bioreactors. Water Science and Technology 40, 203210.
Council of the European Union, 1999. Council Directive 1999/31/EC of Pohland, F.G., Kim, J.C., 2000. Microbially mediated attenuation
26 of April 1999 on the landll of solid waste. Ocial Journal of the potential of landll bioreactor systems. Water Science and Technology
European Communities. 41, 247254.
Doedens, H., Cord-Landwehr, K., 1989. Leachate recirculation. In: Pohland, F.G., Gould, J.P., Esteves, W.R., Spiller, B.J., 1987. Fate of
Christensen, T.H., Cossu, R., Stegmann, R. (Eds.), Sanitary Landll- heavy metals during landll stabilization of solid waste materials with
ing: Process, Technology and Environmental Impact. Academic Press, leachate recycle. US EPA Project No. R-806498, Cincinnati, OH.
pp. 231249. Pohland, F.G., Cross, W.H., Gould, J.P., Reinhart, D.R., 1993. The
El-Fadel, M., 1999. Leachate recirculation eects on settlement and behavior and assimilation of organic and inorganic priority pollutants
biodegradation rates in MSW landlls. Environmental Technology 20, codisposed with municipal refuse. Research Paper, University of
121133. Pittsburgh, USA.
El-Fadel, M., Findikakis, A., Leckie, J., 1996. Estimating and enhancing Price, G.A., Barlaz, M.A., Hater, G.R., 2003. Nitrogen management in
methane yield from municipal solid waste. Hazardous Waste & bioreactor landlls. Waste Management 23, 675688.
Hazardous Materials 13, 309331. Purcell, B., 2000a. Aerox landlling: the operational implications. Waste
Erses, S., Onay, T.T., 2003. In situ heavy metal attenuation in landlls Management, 2728.
under methanogenic conditions. Journal of Hazardous Materials 99, Purcell, B., 2000b. Aerox landlling: a change of approach. Waste
159175. Management, 2527.
Flyhammar, P., Tamaddon, F., Bengtsson, L., 1998. Heavy metals in a Read, A.D., Hudgins, M., Harper, S., Phillips, P., Morris, J., 2001. The
municipal solid waste decomposition cell. Waste Management and successful demonstration of aerobic landlling: the potential for a
Research 16, 403410. more sustainable solid waste management approach? Resources,
He, R., Liu, X., Zhang, Z., Shen, D., 2007. Characteristics of the Conservation and Recycling 32, 115146.
bioreactor landll system using an anaerobicaerobic process for Revans, A., Ross, D., Gregory, B., Meadows, M., Harries, C., Gronow, J.,
nitrogen removal. Bioresource Technology 98, 25262532. 1999. Long term fate of metals in landll. In: Proceedings Sardinia 99,
Hudgins, M., Harper, S., 1999. Operational characteristics of two aerobic the Seventh International Waste Management and Landll Sympo-
landll systems. In: Proceedings Sardinia 99, the Seventh International sium, Cagliari, Italy.
Waste Management and Landll Symposium, Cagliari, Italy. Rich, C., Gronow, J., Voulvoulis, N., 2007. The potential for aeration of
Hudgins, M., March, J., 1998. In situ Municipal Solid Waste Composting MSW landlls to accelerate completion. Waste Management,
Using an Aerobic Landll System. American Technologies Inc., pp. doi:10.1016/j.wasman.2007.03.022.
144157. Ritzkowski, M., Heyer, K.U., Stegmann, R., 2006. Fundamental pro-
Jacobs, J., Schar, H., Van Arkel, F., de Gier, C.W., 2003. Odour cesses and implications during in situ aeration of old landlls. Waste
reduction by aeration of landlls: experience, operation and costs. In: Management 26, 356372.
Proceedings Sardinia 2003, the Ninth International Waste Manage- San, I., Onay, T.T., 2001. Impact of various leachate recirculation regimes
ment and Landll Symposium, Cagliari, Italy. on municipal solid waste degradation. Journal of Hazardous Materials
Kim, H., 2005. Comparative Studies of Aerobic and Anaerobic Landlls 87, 259271.
Using Simulated Landll Lysimeters. Ph.D. Thesis, University of Stessel, R.I., Murphy, R.J., 1992. A lysimeter study of the aerobic landll
Florida, USA. concept. Waste Management and Research 10, 485503.
Kinman, R.N., Nutini, D.L., Walsh, J.J., Vogt, W.G., Stamm, J., Swarbrick, G.E., 2001. Microbiology of landll. In: Smith, D.W., Fityus,
Rickabaugh, J., 1987. Gas enhancement techniques in landll simu- S.G., Allman, M.A. (Eds.), GeoEnvironment 2001. Australian Geo-
lators. Waste Management and Research 5, 1325. mechanics Society, Newcastle, Australia, pp. 397408.
Kjeldsen, P., Barlaz, M.A., Rooker, A.P., Baun, A., Ledin, A., Christen- Tchobanoglous, G., Burton, F.L., 1979. Wastewater Engineering.
sen, T.H., 2002. Present and long-term composition of MSW landll McGraw-Hill International Editions.
leachate: a review. Critical Reviews in Environmental Science and Tittlebaum, M.E., 1982. Organic carbon content stabilization through
Technology 32, 297336. landll leachate recirculation. Journal WPCF 54, 428433.
Kruempelbeck, I., Ehrig, H.J., 1999. Long term behaviour of municipal Townsend, T.G., Miller, W.L., Lee, H., Earle, J.K.F., 1996. Acceleration
solid waste landlls in Germany. In: Proceedings Sardinia 99, the of landll stabilization using leachate recycle. Journal of Environmen-
Seventh International Waste Management and Landll Symposium, tal Engineering 122, 263268.
Cagliari, Italy. Yenigun, O., Kizilgun, F., Yilmazer, G., 1996. Inhibition eects of zinc
Leikam, K., Heyer, K.U., Stegmann, R., 1997. In situ stabilization of and copper on volatile acid production during anaerobic digestion.
completed landlls and old sites. In: Proceedings Sardinia 1997, Sixth Environmental Technology 17, 12691274.