Beruflich Dokumente
Kultur Dokumente
Waste Management
journal homepage: www.elsevier.com/locate/wasman
Department of Environmental Engineering, Technical University of Denmark, Miljvej Building 113, 2800 Kongens Lyngby, Denmark
Chalmers University of Technology/FluxSense AB, SE-412 96 Gteborg, Sweden
c
AV Milj, DK-2650 Hvidovre, Denmark
b
a r t i c l e
i n f o
Article history:
Received 31 May 2010
Accepted 23 October 2010
Available online 24 December 2010
a b s t r a c t
AV Milj is a modern waste disposal site receiving non-combustible waste with a low-organic content.
The objective of the current project was to determine the gas generation, composition, emission, and oxidation in top covers on selected waste cells as well as the total methane (CH4) emission from the disposal
site. The investigations focused particularly on three waste disposal cells containing shredder waste (cell
1.5.1), mixed industrial waste (cell 2.2.2), and mixed combustible waste (cell 1.3). Laboratory waste incubation experiments as well as gas modeling showed that signicant gas generation was occurring in all
three cells. Field analysis showed that the gas generated in the cell with mixed combustible waste consisted of mainly CH4 (70%) and carbon dioxide (CO2) (29%) whereas the gas generated within the shredder
waste, primarily consisted of CH4 (27%) and nitrogen (N2) (71%), containing no CO2. The results indicated
that the gas composition in the shredder waste was governed by chemical reactions as well as microbial
reactions. CH4 mass balances from three individual waste cells showed that a signicant part (between
15% and 67%) of the CH4 generated in cell 1.3 and 2.2.2 was emitted through leachate collection wells,
as a result of the relatively impermeable covers in place at these two cells preventing vertical migration
of the gas. At cell 1.5.1, which is un-covered, the CH4 emission through the leachate system was low due
to the high gas permeability of the shredder waste. Instead the gas was emitted through the waste resulting in some hotspot observations on the shredder surface with higher emission rates. The remaining gas
that was not emitted through surfaces or the leachate collection system could potentially be oxidized as
the measured oxidation capacity exceeded the potential emission rate. The whole CH4 emission from the
disposal site was found to be 820 202 kg CH4 d1. The total emission rate through the leachate collection system at AV Milj was found to be 211 kg CH4 d1. This showed that approximately 1=4 of the emitted gas was emitted through the leachate collections system making the leachate collection system an
important source controlling the overall gas migration from the site. The emission pathway for the
remaining part of the gas was more uncertain, but emission from open cells where waste is being disposed of or being excavated for incineration, or from horizontal leachate drainage pipes placed in permeable gravel layers in the bottom of empty cells was likely.
2010 Published by Elsevier Ltd.
1. Introduction
During the last 1015 years, the waste hierarchy has been the
main principle in waste management in Denmark as well as
Europe. The waste hierarchy aims to encourage the management
of waste materials in order to reduce the amount of waste materials produced, and to recover maximum value from the wastes
that are produced. The waste hierarchy is applied as a guide; it
encourages the prevention of waste, followed by the reuse and
Corresponding author. Tel.: +45 45 25 16 07; fax: +45 45 93 28 50.
E-mail addresses: chs@env.dtu.dk (C. Scheutz), amf@env.dtu.dk (A.M. Fredenslund), jne@amfor.dk (J. Nedenskov), jerker.samuelsson@uxsense.se (J. Samuelsson), pk@env.dtu.dk (P. Kjeldsen).
0956-053X/$ - see front matter 2010 Published by Elsevier Ltd.
doi:10.1016/j.wasman.2010.10.021
947
Table 1
Description of the waste disposal cells 1.3, 1.5.1, and 2.2.2 at AV Milj.
Cell
Cell area
(m2)
Waste amount
(tons)
Active period
1.3
24,000
94,000
19972001
Industrial waste
1.5.1
7200
47,000
19902000
2.2.2
10,500
41,000
Combustible waste
Construction waste
Shredder waste
30 cm bottom layer of ash from
sludge incineration
Industrial waste
Sand from sand traps
No cover
948
1.3 and 2.2.2) or in the top of the waste (cell 1.5.1). The gas probes
consisted of steel tubes, which were closed at the bottom and provided with slits over the lower 3 cm. The steel probes were hammered into the ground at different depths. In general, samples of
the main components (CH4, CO2, O2, and N2) were taken at 10,
20, 30, 40, 60, 80, and 100-cm depth. Samples of 5 mL were
withdrawn with a syringe and stored in vacuumed glass bottles
(Exetainer Vial 819W, 6 mL, Labco Ltd., UK). The probes were
purged before sampling by redrawing three times the volume of
the probes using a 250 mL syringe. Between sampling the probes
were capped. Two sets of gas probes were installed on each cell.
Gas samples from both deep and shallow probes were collected
14 times approximately once a month between February 2005 and
June 2006.
The CH4 oxidation capacity of the cover soils and of the shredder waste was studied by incubation of cover soil and shredder
material sampled at 1020 cm intervals from the surface to
100 cm depth. Material was sampled in 10 cm intervals from the
surface to 60 cm depth below the surface at the cell with shredder
residue. The texture of the shredder waste prevented sampling at
deeper depths than 5060 cm. Samples were collected at two different locations at each cell; an area with CH4 emissions below
detection and a hotspot area, where higher CH4 emissions previously had been measured. A xed amount of soil or shredder material (20 g moist material) was added to a 117 mL batch container
with an initial mixture of CH4 (15% v/v), O2 (35% v/v) and N2
(50% v/v). In order to check if any disappearance could be due to
non-microbial processes (abiotic degradation, sorption and volatilization) control batches with sterilized soil or shredder residues
were conducted. All batch experiments were carried out in duplicates. The experimental set-up as well as the data analysis is described in detail in Scheutz et al. (2004).
2.6. Surface screening of CH4 and ux chamber measurements
An initial surface screening of CH4 at the waste cell was performed, since high spatial variations in surface gas emissions from
landlls are often observed (Scheutz et al., 2003, 2008, 2009). To
provide basis for a qualitative analysis on the spatial variability
of CH4 emission through the soil cover, a grid of locations for measurements of CH4 concentrations were planned. The number of
grid points and the average grid spacing was chosen adopting a
strategy described in Guidance for Monitoring Landll Gas Surface
Emissions (UK Environment Agency, 2004). In total 82 measuring
points were planned, placed in a grid with an average spacing of
1015 m. All grid points were agged and surveyed using GPS
(Trimble 5700 RTK GPS with TSC1 controller, Trimble Navigation
Limited, Sunnyvale, CA, USA). The cell was rst visually inspected
to observe ssures, cracks or other irregularities, which could have
higher gas emissions. Measurements were taken in all grid points
and along the fence and borders of the cells. A small funnel was attached to the inlet tube of the analyzer and placed directly on the
surface during analysis. In total three surface screenings were conducted at each cell.
Surface emission rates of CH4 were determined using static ux
chambers. The ux chambers were constructed of the top of a
drum, which was pushed 3.5 cm into the ground. The removable
top of the chamber was equipped with several sampling ports
and a fan securing that the air inside the chamber was totally
mixed during sampling. The bases of the ux chambers were
sealed with bentonite. When installed, the ux chamber had a
height of approximately 20 cm and an inner diameter of 57 cm
(giving a total chamber volume was 51,035 cm3 over an enclosed
949
Table 2
Average gas concentrations (% v/v) of selected LFG components in three different
waste cells (1.3, 1.5.1, and 2.2.2) based on 14 samples sampled approximately once a
month between February 2005 and June 2006.
LFG components
Methane (CH4)
Carbon dioxide (CO2)
Oxygen (O2)
Nitrogen (N2)
1.5.1
2.2.2
70 4
29 5
11
33
27 6
00
11
71 5
64 13
84
34
24 13
27% CH4, 71% N2, and only traces of CO2. During anaerobic fermentation of organic material CO2 is always formed together with CH4.
The lack of CO2 in cell 1.5.1 and 2.2.2 in comparison with cell 1.3
could be due to dissolution of CO2 in inltrating water, precipitation of different carbonates or microbial methanogenesis by
hydrogenophilic bacteria in combination with a low gas generation
rate. Laboratory experiments and gas modeling showed that waste
types deposited in cells 1.5.1 and 2.2.2 produced less LFG than
waste from cell 1.3 (see Sections 3.2 and 3.3). The shredder waste
disposed in cell 1.5.1 had a metal content of 3% (determined by
hand sorting of the waste), which might corrode under anaerobic
conditions resulting in hydrogen production (Reardon, 1995). The
hydrogen produced can be used for microbial methanogenesis by
hydrogenophilic bacteria (Belay and Daniels, 1990; Lorowitz
et al., 1992).
3.2. Gas production based on waste incubation experiments
Fig. 1 shows accumulated CH4 productions in the three waste
types. In all waste samples a lag phase of approximately ten days
was observed. After the initial lag phase of low or no CH4 production, relatively constant production rates were seen. Due to the initial lag phase of CH4 production, all production rates were
calculated using data measured after 45 days. The CH4 production
rates relative to waste mass shown in Fig. 2 were seen to vary between samples taken at different locations at the same cells. This
was likely due to heterogeneity of the waste deposited in the three
cells with regards to content of degradable organic matter. The
highest CH4 production rate was found for the cell with mixed
combustible waste (2.02 mg CH4 kg waste1 d1) where much
lower rates were obtained for the cells with shredder waste and
mixed industrial waste (0.75 and 0.52 mg CH4 kg waste1 d1,
respectively) (Table 3). Compared to landlled MSW, of which production rates between 0.7 and 9.5 mg CH4 kg waste1 d1 can be
expected (El-Fadel et al., 1997), the production rates measured
were in the lower range. As expected, the more inorganic waste
types deposited in cells 1.5.1 and 2.2.2 produced less CH4 than
waste from cell 1.3. The total average CH4 production rate for the
three cells were estimated by multiplying average CH4 production
rates with the total waste amount deposited known from records
of waste deposited at the landll (Table 3).
950
1.3
1.5.1
2.2.2
2.02 0.24
0.75 0.59
0.52 0.53
190
35
21
Fig. 3. Model simulated production of LFG at cells 1.3, 1.5.1 and 2.2.2 at AV Milj
Landll.
Table 4
LFG production (m3 year1) and CH4 generation (kg d1) in 2006 determined by
modeling. The Afvalzorg values are average values.
LFG production m3 LFG year1
Fig. 2. Observed CH4 production rates related to sampled waste mass.
for 2006 was converted to kg CH4 produced per day. Results are
listed in Table 3.
Cell 1.3
Cell 1.5.1
Cell 2.2.2
Afvalzorg
GasSim
Afvalzorg
GasSim
112,000
27,000
67,000
88,000
85,000
65,000
109
27
65
86
84
63
951
Table 5
Estimation of the total amount of CH4 emitted through cell surfaces.
Cell
1.3
1.5.1
2.2.2
Total
area
28,420
8748
12,227
m2
Maximum
measured
surface
emission rate
g CH4 m2 d1
Total amount of
CH4 emitted
through cell
surfaces
kg CH4 d1
284
437
122
0.01
23.0
0.04
0.003
10.060
0.005
Estimated
hotspot
area
Estimated
hotspot
area
%
1
5
1
Fig. 5. CH4 oxidation rates vs. sampling depth below surface at two locations with
different gas emissions: A represents an area with CH4 emissions below detection
whereas B represents a hot spot with higher CH4 emissions.
952
coefcients often higher than 0.90. In general, very good reproducibility was obtained and results from duplicate batches were
almost identical.
The shredder waste placed in cell 1.5.1 showed a relatively high
capacity for CH4 oxidation with maximum oxidation rates up to
17 lg CH4 g1 h1, which are comparable to rates obtained for
landll cover soils (Scheutz et al., 2009). Fig. 4A shows the CH4,
O2 and CO2 concentrations measured in headspace vs. time, in a
batch experiment containing shredder waste. CH4 oxidation followed zero-order kinetics, indicating that the oxidation was not
CH4-limited. The oxidation was microbially mediated as seen from
comparison with the sterilized control batch (Fig. 4B). Maximal
oxidation activity occurred at 40 cm below the surface (Fig. 5A),
which corresponds to the depth with CH4 oxidation indicated by
gas proles (Fig. 6). The CH4 oxidation activity was signicantly
lower in the upper 20 cm of the prole and at 50 cm depth
(Fig. 5A). Incubation of shredder samples taken at the hot spot
showed a different oxidation potential prole as the maximum oxidation capacity was found in the upper 30 cm of the prole and decreased signicantly at 40 cm depth (Fig. 5B). However, this agrees
well with the measured gas concentrations proles showing a very
shallow oxidation zone between 10 and 20 cm depth (results not
shown).
The cover soil placed at cell 1.3 also showed a high capacity
for CH4 oxidation with maximum oxidation rates up to 13 lg
CH4 g1 h1 (Fig. 5A). The zone with the highest oxidation capacity
was found between 80 and 100 cm.b.s., which is lower than the
zone identied in the shredder prole. This is most likely a result
of the low ux of CH4 through the cover due to its low gas permeability. The small ux of CH4 that is entering the lower part of the
soil cover is relatively rapidly oxidized due to the aerobic conditions also in the deeper part of the cover (Fig. 6). Thus the lower
CH4 oxidation activity is a result of CH4 limitation and thereby a
low number of CH4 oxidizing bacteria. The oxidation prole taken
close to a CH4 leaking area shows a higher oxidation capacity in the
upper prole as a result of a higher CH4 ux at this location
(Fig. 5B). Gas concentrations proles taken at this location showed
elevated CH4 concentrations and reduced O2 concentrations in the
lower part of the cover around 80 cm depth indicating CH4 oxidation (results not shown).
The cover placed at cell 2.2.2 showed a relatively low capacity
for CH4 oxidation throughout the whole prole (Fig. 5A). This is
coherent with the gas proles and the ux measurements either
indicating very low or no CH4 emission through the cover
(Fig. 6). Even the soil prole taken at a leaking area close to the border of cell 2.2.2 showed no CH4 oxidation capacity (Fig. 5B).
The depth integrated CH4 oxidation rate, K0 (g m2 d1), was
calculated using the formula:
K 0 k0 da qb
where k0 is the CH4 oxidation rate (lg g1 d1) obtained from the
incubation experiments showed in Fig. 5A, da is the oxidation zone
(m) and qb is the soil bulk density (metric tons m3). Table 6 lists
the calculated CH4 oxidation rates. The highest CH4 oxidation
capacity was found for cell 1.3 and 1.5.1, which gave a depth integrated CH4 oxidation capacity of 108 and 101 g m2 d1, respectively. In comparison the capacity for cell 2.2.2 was much lower;
10 g m2 d1. The soil covers at AV Milj were not designed with
the purpose to optimize methane oxidation and it is therefore likely
that higher oxidation efciencies could be obtained if more permeable materials like composts or sandy soils were used as cover
materials (Scheutz et al., 2009). At Danish landlls low permeable
materials like clayey soils are often used in order to minimize leachate generation.
953
Fig. 6. Vertical gas proles of main components vs. depth below surface at cell 1.3, 2.2.2 and 1.5.1.
Table 6
Depth integrated CH4 oxidation capacity at cell 1.3, 1.5.1, and 2.2.2 base on CH4 incubation experiments conducted with material sampled at an area with CH4 emissions below
detection (Fig. 5A).
1.3
Cover depth 100 cm
1.5.1
Depth of oxidative zone 50 cm
2.2.2
Cover depth 100 cm
k0
lg g1 d1
K0
g m2 d1
k0
lg g1 d1
K0
g m2 d1
k0
lg g1 d1
K0
g m2 d1
015
1530
3050
5070
7090
90100
Total
24
11
11
11
218
144
6
3
3
4
70
23
108
015
1525
2535
3545
4555
42
38
223
284
24
10
6
36
46
4
015
1530
3050
5070
7090
90100
Total
6
2
1
0
18
15
1
0
0
0
6
2
10
Total
101
Table 7
Total CH4 mass balances from each individual cell. Last column lists the depth integrated CH4 oxidation rate based on batch experiments.
Cell
1.3
1.5.1
2.2.2
Production
Emitted
Potentially oxidized
Through wells
kg CH4 d1
Through cover
kg CH4 d1
190
35
21
28
0
15
0
10
0
163
25
7
108
101
10
6
3
1
954
Table 8
Whole site CH4 emission measured by dynamic plume measurements combined with tracer release.
Campaign
Emission
kg h
October 2006
October 2007
Average 2006 and 2007
1
33.1 7.8
35.2 3.2
34.2 8.4
Distribution (%)
kg d
1
794 187
845 77
820 202
Part A
Part B
Part C
Part A
Part B
Part C
20
22
21
44
48
46
36
30
33
7
8
7
3
2
2
9
3
6
(820 kg CH4 d1) for 2006/2007. The difference between the modeled gas production and the measured gas emission can be due to
CH4 oxidation and/or uncertainties in the modeled gas production.
As most of the cells at the landll are un-covered it is believed that
the overall CH4 oxidation of the whole site is limited. Most of the difference between the modeled gas production and the measured gas
emission is most likely rater a result of uncertainties associated with
the gas production modeling indicating that the model is not
describing the waste types disposed at the landll. However, further
investigations are needed to conclude on this issue.
4. Conclusion and perspectives
Overall the gas analyses of the main components indicated that
LFG containing CH4 was produced in all three cells investigated in
this study. However, the gas generated in the investigated waste
cells showed very different compositions, as the gas generated in
the cell with mixed combustible waste consisted of mainly CH4
(70%) and CO2 (29%) in opposition to the composition of the gas
generated within the shredder waste, which primarily consisted
of CH4 (27%) and N2 (71%), containing no CO2. The results indicated
that the gas composition in the shredder waste was governed by
chemical reactions as well as microbial reactions.
CH4 production rates determined experimentally using waste
sampled at the three cells showed signicant gas production rates
(0.522.02 kg CH4 kg waste1 d1). The modeled CH4 production
rates using the GasSim model and the Afvalzorg Multiphase Model
were 86109 kg CH4 d1 (cell 1.3), 2783 kg CH4 d1 (cell 1.5.1)
and 6365 kg CH4 d1 (cell 2.2.2). For the shredder cell (1.5.1)
and the cell with mixed industrial waste (cell 2.2.2), the gas production models seemed to overestimate the production rates,
which was especially the case using the GasSim Model.
The surface CH4 emission from all three cells was very low and
limited to a few hot spots on slopes, edges, and around leachate
collection wells. The low emission was due to placement of a low
permeable clay cover at two of the cells (cell 1.3 and cell 2.2.2).
At the cell with shredder waste, the waste showed a very high
CH4 oxidation capacity reducing the surface emission from the cell.
Shallow gas proles from the cell with shredder waste showed an
oxidation zone of 40 cm below the surface with signicant CH4 oxi-
dation. This was supported by incubation experiments with shredder material, which showed a high potential of CH4 oxidation in
the upper 40 cm of the shredder waste. Shallow gas proles from
cell 1.3 and 2.2.2 showed a similar gas composition to atmospheric
air indicating that LFG generated in the two cells was not emitted
through the cover. CH4 oxidation in the lower part of the covers
were possible as CH4 incubation experiments showed a tendency
to increasing oxidation rates with increasing cover depth.
A signicant amount of the gas was emitted through the leachate collection system. Mass balances for the two covered cells 1.3
and 2.2.2 indicated that up to 67% of the generated gas was emitted
through leachate collection wells. At the un-covered shredder
cell the emission through the leachate collection wells were
insignicant.
The whole CH4 emission was from the disposal site was found
to be 820 202 kg CH4 d1. The total emission rate through the
leachate collection system at AV Milj was found to be
211 kg CH4 d1. These results indicate that approximately 1=4 of
the emitted gas was emitted through the leachate collections system making the leachate collection system an important source
controlling the overall gas migration from the site. The emission
pathway for the remaining part of the gas was more uncertain,
but emission from open cells where waste was being disposed of
or being excavated for incineration, or from horizontal leachate
drainage pipes placed in permeable gravel layers in the bottom
of empty cells was likely. The modeled gas production for the
whole site (23002900 kg CH4 d1) was about three times higher
than the measured emission again indicating that the model overestimates the gas production and emission as it is believed that the
CH4 oxidation for the whole site is limited.
Overall this study showed that CH4 production and emissions
from landlls receiving non-combustible waste with low organic
carbon content are not negligible. The study also showed that
gas productions models like GasSim seemed to overestimate the
production rates especially for cells with low-organic waste like
shredder waste. However, more work is needed to validate existing
models especially when used on landlls containing waste fractions with a low-organic content. In addition, the emission patterns may be very complicated due to the presence of a leachate
collection system acting as a major gas escaping route and the
use of clay cover soils leading to a very heterogeneous surface
emission pattern. In the future using the leachate collection system
for gas collection followed by treatment in biolters optimized for
CH4 may be a potential technology to reduce the CH4 emission
from the site.
Acknowledgements
This research was funded by AV Milj, Denmark. We thank the
technical staff of AV Milj for their co-operation, comments on
technical issues, eld assistance and arrangements making the
eld campaigns possible.
References
Belay, N., Daniels, L., 1990. Elemental metals as electron sources for biological
methane formation from CO2. Antonie van Leeuwenhoek 57, 17.
955