- Email: [email protected]
Performance assessment of landfill in-situ aeration – A case study
Waste Management 101 (2020) 231–240
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Performance assessment of landfill in-situ aeration – A case study Christian Brandstätter a,⇑, Roman Prantl b, Johann Fellner a a b
Institute for Water Quality, and Resource Management, TU Wien, Karlsplatz 13/226-2, 1040 Vienna, Austria blp GeoServices gmbh, Felberstraße 24/1, 1150 Vienna, Austria
a r t i c l e
i n f o
Article history: Received 5 June 2019 Revised 28 August 2019 Accepted 10 October 2019
Keywords: In-situ aeration Landfill Municipal solid waste Aftercare CO2 CH4
a b s t r a c t The assessment of landfill in-situ aeration eludes standardization as its application highly depends on varying local conditions. The prevailing work tries to assess the procedure performance by using typically available data. In the here presented case study the aeration pipes were applied horizontally. To evaluate the airdistribution and its effect on the landfill solid body, two monitoring fields with 10 10 m were created. From there in total 60 solid and 336 gaseous samples were taken over five years. As the material from the landfill was rather old and characterized by comparatively low reactivity, ‘‘new” material from a mechanical biological treatment (MBT) plant was introduced in the landfill. Additionally online gas data from eight technically separated landfill sections were analyzed during in-situ aeration. In total, about 46 Mm3 (0.27 m3/kg waste) air was introduced into the landfill body. The eight sections showed differences in reactivity (overall C-discharge was 8 g C kg1 dry weight, ranging from 4.5 to 11). With solid sampling we could not show a significant decrease in landfill TOC but for the introduced MBTmaterial. Ammonium in solid samples was significantly reduced (to 14.7% initial) and NO3 significantly increased (2.1% initial). The reduction of the initial TOC (4.58%) was on average 11%. The application of horizontal landfill aeration led to a widespread air-distribution in a rather shallow landfill. Monitoring fields allowed for a screening of the impact of the measures on the solid body with reduced sampling costs. Ó 2019 Elsevier Ltd. All rights reserved.
1. Introduction The degradation of organic matter through microorganisms is a major driving process for landfill emissions. Landfill in-situ aeration targets to enhance this process by providing oxygen for microbial aerobic respiration. This enables microorganisms to exploit previously unavailable resources, like anaerobically hardly degradable woody parts and allows for a faster biological waste stabilization. The anaerobic and aerobic degradation processes lead to a change in carbon quality, as the easier digestible carbohydrates are degraded faster. The in-situ aeration method is by now well established especially in Germany, but also in other countries (Liu et al., 2018; Prantl et al., 2006; Ritzkowski and Stegmann, 2012), where numerous in-situ aeration projects are currently in operation. The method ideally should be applied after potential landfill gas extraction is not economically feasible anymore. Its major benefits are the reduction of leachate pollution, mainly in the form of ammonium (NH4, Ritzkowski et al. (2006)) and chemical oxygen ⇑ Corresponding author. Tel.: +43 1 31304 5528. E-mail address: [email protected] (C. Brandstätter). https://doi.org/10.1016/j.wasman.2019.10.022 0956-053X/Ó 2019 Elsevier Ltd. All rights reserved.
demand (COD), as well as the reduction of the remaining methane (CH4) production potential (Ritzkowski and Stegmann, 2007). Landfill in-situ aeration however, also shows negative sideeffects, like increased CO2 emissions through accelerated microbial activity or the increased release of the potent greenhouse gas N2O (Nag et al., 2016). Despite the positive effects, assessing the success of in-situ aeration projects remains a challenge (Hrad and Huber-Humer, 2017). The evaluation is not only a strictly technical question, but also stretches out into the legal issue of when to release a landfill out of care. This is because the method is targeted to older landfills, possibly close to such a potential release point. Six potential criteria for the determination of the endpoints were presented elsewhere (Ritzkowski and Stegmann, 2013). Half of them require testing of the solid waste material in some way, either for determining the remaining emission potential using landfill simulation reactors (LSR), or with other degradation experiments. But because of the heterogeneous nature of landfills, deriving landfill properties from solid samples is difficult (Brandstätter et al., 2014). Other proposed criteria were landfill settling, temperature measurements or CH4-formation.
232
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
It is not feasible for practitioners, municipalities or scientists, to gather the necessary amount of solid samples before, during and after an in-situ aeration measures, for creating a proper data base for sound assessment of the landfill status. This is why we have to search for more cost-effective means of performance assessment, as we are attempting with the present work. Hence, the aim of the present paper is to combine the often available process data of landfill in-situ aeration projects and extend them with information from a limited amount of costly measurements of solid material. The gathered data stem from an Austrian case study, the Heferlbach Landfill. Here an in-situ aeration was applied with the focus of preventing landfill gas to migrate into neighboring buildings. In particular, we 1. analyzed the online data of a horizontal insitu aeration, technically separated into eight sections, 2. established monitoring fields on-site for solid sampling, so we could observe ongoing trends concerning the development of the landfill solid body during in-situ aeration in a cost-effective way and 3. introduced reactive material from a mechanical biological treatment, as a proxy for assessing the impact of the measures on material of known composition. 2. Materials and methods 2.1. Landfill description The landfill was filled between 1965 and 1974. The deposited material consisted of mixed municipal solid waste (~60% total mass TM), construction and demolition waste and excavated soil. The site is rather shallow with an area of roughly 65,000 m2 and an average depth of 3.7 m. The landfill had a total volume of ~240.000 m3. An overview of the landfill, including sampling points and aeration pipes is given in Fig. 1. 2.2. Sampling In 2001, on behalf of the Austrian Federal Agency of Environment, a site screening was conducted. Thereby over 100 solid samples were taken and analyzed. The impact of leachate emissions was considered insignificant, as a neighboring industry extracted and used the ground water underneath the landfill. For multiple reasons, especially to not disturb close-by residents, it was decided to apply the measure of in-situ aeration for minimizing the risk of potential gas migration into neighboring cellars. Since the waste material was more than 40 years old, and the landfill also contained waste material of low biological reactivity, more reactive regions rich in organic material were actively selected for the solid waste sampling in 2012. Finding these reactive landfill regions was based on the available measurement data (TOC in the solids and CO2/CH4) from 2001. During that campaign, 56 samples of roughly 30 kg were taken with an excavator and shovels. The material was screened with a mesh width of 20 mm. Based on the information available from the site screening 2001 and our own analysis, an overview of the main landfill properties is given in Table 1. In a recent research project, laboratory experiments were conducted (Brandstätter et al., 2015a,b) for getting a good measure of biodegradability of the waste material. In addition to the initial sampling, two monitoring fields (each 10 10 m) were defined in sections known to contain more organic material. From these monitoring fields, every year twelve samples were taken in three depth levels out of four pits dug by an excavator. In the first year also 16 gas sampling probes were erected in varying distances to the aeration and gas extraction pipes as well as in differing depths. Roughly every three months the concentrations of CH4 and CO2 were measured in different dis-
tances from the aeration and extraction pipes as well as in different depths. As a grouping variable the distance to the aeration pipes was calculated using the Pythagorean formula:
distanceair ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 ðDXY Þ2 þ ðDdepthÞ
ð1Þ
with DXY as the horizontal distance between the gas probe and the nearest aeration pipe and Ddepth as the difference between the average filter depth (filter upper edge - filter lower edge) and the average depth of the aeration pipes (2.5 m). All units are m. In 2012, ten samples of MBT-waste (organic residues from mechanical biological treatment), each about 10 kg were buried in the landfill. Since the available landfill material was rather old, the idea was to introduce material of known composition and reactivity, in order to detect potential inequalities in air distribution with higher sensitivity. The material was put into ten separate bags, sewed of fly-screen mesh (~40 60 cm, litter bags). Upon placement, temperature loggers (Voltcraft) were placed inside. The MBT-material was placed in different depths and in varying distance to the aeration pipes. For each pit, either two or three litter bags were placed inside. Subsequently, the litter bags were dug out in 2014 after two years to gain access to the buried temperature loggers. All ten were found, but most temperature loggers stopped working after two weeks (data not shown). Then material from each pit was pooled, mixed, divided and placed again (n = 8). This procedure was necessary, as many of the bags were cut open during the sampling procedure. In 2017 they were recovered for good and five valid samples from four different trenches could be collected. 2.3. In-situ aeration On-site, a low-pressure horizontal in-situ aeration was established, since the landfill was rather shallow and long (see Section 2.1). Before the full-scale application, a technical prototype for gathering information about the full-scale capacity was installed, mainly to determine the aeration rates. The prototype was installed for a whole section and could be used even after the test-phase for the main aeration measure. For the insertion of the aeration and extraction pipes, directional drilling was used. The aeration pipes were introduced in 3–4 m depth and the extraction pipes in 2 m. The pipes (PE 90, thickness: 8.1 mm, single pipe length: 6 m) were welded and held around 100 holes/meter with a hole diameter of 10 mm. Water was used as jetting liquid. The pipes were put in place with approximate a distance of 10 m to each other. For safety reasons, the air extraction is roughly 1.5fold higher than the aeration (Table 3). In addition, southwards at the edge of the landfill a security gas extraction system (side extraction) was established, to prevent any gas migration to neighboring buildings and cellars. The whole aeration project was separated into eight sections, where the gas concentrations, temperature, flow and pressure data were analyzed separately in a centralized unit. The off-gas was directed through a biofilter to reduce odor. The facility allowed for reversing both the extraction and the aeration pipes. For determining the mass flow, a MFT-B device from KURZ Instruments Inc. was installed. For measuring the gas pressure, explosion-proof pressure transducers from WIKA were used. For determining the respective concentrations of CO2, CH4 and O2, a measuring device from Siemens AG (ULTRAMAT 23) was used. The determination of CO2 and CH4 was done via infrared measurement and the oxygen concentration was measured with a para-magnetic pressure determination system. In the summer 2013 the landfill got submerged, as it is situated in high proximity to the river Danube. Thus the aeration had to be stopped for a few weeks.
233
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
Fig. 1. Landfill overview.
Table 1 Landfill Characteristics per Section. Variable
Unit
SE01
SE02
SE03
SE04
SE05
SE06
SE07
SE08
Area Total Volume Total Volume dry MSW Top Soil Excavated Soil C&D-Waste TOC* Solid Samples 2012
[m2] [m3] [m3] [m3] [m3] [m3] [m3] [Mg] [n]
7300 25,000 19,700 10,900 2200 2200 11,900 900 15
8900 27,200 21,400 16,800 2500 8700 1600 980 10
8800 29,400 23,100 20,300 1800 3100 5900 1060 20
7700 30,000 23,700 22,800 1500 3100 4100 1100 –
7500 24,900 19,600 16,100 1900 7100 1700 900 –
4300 11,800 9,300 5680 780 2000 4100 430 –
8100 31,000 24,500 28,600 2400 1800 680 1120 9
12,300 39,000 31,000 23,300 2900 11,600 4400 1400 2
Notes: The initial properties were derived from our own sampling campaign (2012) and missing data were filled from a sample campaign conducted by the Austrian Federal Agency of Environment 2001 for overall site assesment (data not shown). The density of the landfilled MSW was assumed 1. The amount of MSW is a rough estimate and was not considered for further calculations. Numbers were rounded to 2 (or three, if the leading digit was <3) significant digits. *For TOC, only the fine fraction (<20 mm) was considered, with an average value of 4.58 [% DM]. For the water content an average value of 21.2 [% DM] was considered. DM dry mass, MSW municipal solid waste, SE section, TM total mass, TOC total organic carbon.
2.4. Analytics
2.5. Calculations
The applied analytical procedures are briefly summarized in Table 2:
For the field data, the total amount of material, or more precise the total dry weight (DW) was used as reference value. For
234
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
Table 2 Sample measurements. Variable Solids Acid soluble lignin Cellulose Total Carbon Respiration Index after 4 days Water content
Abbreviation
Device/Procedure/Norm
LigninAS – TOC RI4
TAPPI (1999) Agilent 1200 series HPLC equipped with a refractive index detector (RID) and a Shodex sugar SP0810 column MACRO CHNS analyzer (Elementar Analysensysteme GmbH, Hanau, Germany) ON S 2027-1 Austrian Standard Institute (2002)
WC
2.5 kg of material in drying oven until weight remained constant.
Eluted Elution Electrical conductuvity Ammonium
– EC NH4
Nitrate
NO3
Sulfate
SO4
Solid/liquid ratio 1:10, 24 h of shaking Seven excellence S470 (Mettler-Toledo S.A.E, Spain) DIN EN ISO 11732 German Insitute for Standardization (2005), continuous flow analyzer (Skalar Analytical, B.V., Netherlands) DIN EN ISO 13395 German Institute for Standardization (1996), continuous flow analyzer (Skalar Analytical, B.V., Netherlands) ON EN ISO 1030-1, Austrian Standard Institute (2007), Dionex ICS 900 (Thermo Fisher Scientific Inc., USA)
Table 3 Technical characteristics of the in-situ aeration. Variable
Unit
SE01
SE02
SE03
SE04
SE05
SE06
SE07
SE08
Inflow Field Extraction Side Extraction Temperature* Settelement
[m3 kg1 DW] [m3 kg1 DW] [m3 kg1 DW] [°C] [% average height]
0.346 0.466 0.249 16.5 (8.7) 3.9
0.280 0.381 0.261 20.9 (9.4) 2.5
0.274 0.352 0.293 18.9 (9.3) 1.7
0.279 0.351 0.270 16.2 (9.2) 1.8
0.209 0.269 0.429 16.1 (9.8) 5.7
0.420 0.542 0.328 21.3 (11) 3.4
0.306 0.465 0.246 25.9 (8.8) 6.4
0.162 0.254 0.246 20.1 (7.9) 3.8
Notes: Data are derived from a time span over 1886 days. *Average Temperature Field Extraction with standard deviation in brackets.
calculating the C-discharge both the concentrations of CO2 and CH4 were considered, used with the general gas equation per time point:
n¼
pV_ RT
ð2Þ
with n as the amount of substance [mol], R as the universal gas constant [J mol1 K1], T as the temperature [K], p as the atmospheric pressure [Pa] and V_ as the volumetric fraction per time point t [m3 t1]. The underlying field data used in this article was aggregated to daily values from five minute time steps. The measurement of temperature, concentration and volumetric flow were all conducted in
Fig. 2. Specific C-discharge from 1st Feb. 2012 to 31st Mar. 2017. Notes: The C-discharge is calculated per per DW (dry weight). The samples were sieved before processing, so only the fraction <20 mm was considered.
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
235
Fig. 3. Specific C-discharge - Field vs. Laboratory. Notes: DW: dry weight, LSR: landfill simulation reactor.
Fig. 4. Gas Concentration vs. Distance to Aeration Pipes. Notes: Gas concentrations were measured in eight gas wells per monitoring field (16 in total). The distance values were classified to low (n = 9), middle (n = 28) and high (n = 9) with breaks at 3.3 and 6.6 m and a maximum distance of 10 m. The numbers in the plot and horizontal lines denote the mean values averaged over the whole time frame. The thin lines denote the measured concentration averaged per distance class. The higher variation starting from early 2015 is a direct result of aeration optimization. The areas denote the standard errors per time point within the respective group.
the central unit. The amount of substance then was multiplied with 12 [g mol1] to receive [g] carbon. All the data processing including visualization was done with R (version 3.5.1, R Core Team (2018)) and the package ggplot2 (version 2_3.0.0, Wickham (2016)). For statistical group comparison of the concentration data one-way ANOVAs were computed, followed by Tukey-HSD tests.
3. Results and discussion 3.1. Online (Gas) data The average off-gas temperature from the field extraction over the whole time-frame and all the sections was 19.5 °C (Table 3). In total, 46 Mm3 (0.284 m3 kg1 DW) of air were added and
236
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
Fig. 5. Solid Samples taken from the Monitoring Fields. Notes: Different letters denote significant differences between the years, if applicable (a is highest). The distance values were classified to low (n = 9), middle (n = 28) and high (n = 9) with breaks at 3.3 and 6.6 m and a maximum of 10 m. The highlighted samples from the same pit correspond to the litterbag samples. DistAer: Distance to Aeration Pipes, EC: Electrical Conductivity, RI4 : Respiration Index after 4 days.
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
237
Fig. 6. C-discharge vs. MSW-content per landfill Section.
112 Mm3 (0.675 m3 kg1 DW) being removed for the considered time period of five years. The high amount of air extraction (2.4fold higher than air inflow) can be attributed to the additional side-extraction system (63 Mm3, or 0.385 m3 kg1 DW) which was established for safety reasons. The average settlement after five years was about 3.6% of the total height (mean: 13.5 cm ± 2.3 (standard error SE)). The sections showed high variation in their constitution, their area as well as the amount of air added and extracted (Tables 1 and 3). The specific C-discharge per landfilled dry mass (Fig. 2) also showed high variations between the sections ranging from 4.5 to 11 g C kg1 DW over the whole timeframe. The C-discharge also decreased over the time ranging from 2.8 in 2012 to 0.6 in 2016, the last completely assessed year. In comparison to the laboratory experiment, the specific aeration rate in the field was much lower (roughly 1/10, see Fig. 3). The average reduction in TOC in the field was 11% (ranging from 9.7 to 23.2%), while in the laboratory experiment it was 33.7% after two years of aeration. It must be noted, that for the laboratory the material was already collected from organic hot-spots, while in the field the whole waste body was considered, including inorganic material. As was to be expected from an old landfill containing municipal solid as well as construction and demolition waste, the different landfill areas showed high variation in composition and degradation behavior (Fig. 2). The overall C-discharge was 1380 Mg (8 g C kg1 dry weight) is rather high in comparison to other projects (Hrad & Huber-Humer, 2017), which reported values ranging from 3 to 17.2 g C kg1 DW. This is especially the case when considering the rather high waste age (closed 1974, the aeration started 2012), low initial reactivity (average RI4: 3.3 mg O2 kg1 DW) of the here presented landfill Heferlbach. For reference, the here investigated aeration duration was about 63 months with an air/solid ratio of 270 m3 Mg1 DW. We contribute the main reason for this rather good performance to the shallow landfill anatomy in combination with the horizontal aeration. In the prevailing case it also was possible to
investigate eight different subsections and even operate them separately. Therefore we could provide a performance assessment for each section and propose changes in the operational parameters. To not exceed the scope of the paper, we deliberately excluded the details of the optimization procedure. In the laboratory, with an air/solid ratio about double as high as in the field, also the relative C-discharge (referred to the TOC content of the waste) was about double as high (see Fig. 3). However, it needs to be noted, that the reference value for the relative C-discharge was TOC, whose determination relies on solid sampling and is thus highly uncertain.
3.2. Data from the monitoring fields 3.2.1. Gas concentrations The gas concentrations in the landfill measured in the monitoring fields showed an overall decrease in CO2 and CH4 (Fig. 4). Grouping the data from the 16 wells according to their distance to the aeration pipes allowed for assessing the air distribution in the landfill. The distance was classified as <3.3 m: low, 3.3 and <6.6 m: middle, 6.6 m: high. In the gas concentration data from the monitoring fields, a clear impact of a flooding event in 2013 can be noticed. After the flooding, CO2 and CH4 concentrations increased. This can be related to technical changes, as the gas flows pipes were reversed to unblock the aeration pipes. At the end of 2013 the gas concentrations in the field changed to a rather stable level. The impact of water transport strongly affected the gas concentrations locally in the monitoring fields. On a landfill level (online data) the effect was not as pronounced, except for the immediate stop for a few weeks after the flooding event. The influence to the extraction distance was vice versa but less pronounced in comparison to the distance to the aeration pipes (data not shown). The aeration measure was stopped for several weeks until the water was drained and the pipes unblocked. This lead to a short term increase in CH4 and CO2 which was to be expected.
238
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
Fig. 7. Composition of MBT-material (litter bag data). Notes: Different letters denote significant differences between the years, if applicable (a is highest). LigninAS: acidsoluble fraction of lignin, RI4: Respiration Index after 4 days.
The disturbance of the online gas-data of the off-gas concentrations was not that big, except during the aeration stop. From the gas concentrations in dependency of the distance to aeration pipes (see Fig. 4) it was possible to derive an indirect over-
view of the quality of the gas distribution through the aeration: after the flooding a rather stable distribution regime was reached. For this aeration measures, a distance to the aeration pipes of more than 6.6 m lead to increased CH4 and CO2 concentrations. This
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
might be caused by the rather low landfill depth and probably also by heterogeneity in the waste body resulting in preferential flow paths in differing directions, as it was already discussed for water (Fellner et al., 2009). However, in the off-gas concentrations of the aeration there was no increase in CH4 to be noticed. 3.2.2. Solids Every year in spring, twelve samples were taken in the monitoring fields. The analyses of TOC did not show a significant reduction (Fig. 5). The analyses of RI4 allowed the observation of a decrease from 3.2 to 1.45 in 2016, where the median also dropped below 0.15 [mg O2 kg1 DW] with slightly higher values in 2017. The median in 2012 was 2.8, the only year with a significant reduction in comparison to 2012 was 2016 according to Tukey’s HSD test. The eluted parameters behaved differently (Fig. 5). For the electrical conductivity (EC) no decreasing or increasing trend could be observed. Over five years, a significant reduction in NH4-N after elution could be observed. This was accompanied by a significant increase of NO3-N only after five years in 2017. For SO4, a parameter indicating an increase in the redox potential, the trend is significantly increasing as well. An increase in SO4 could be observed already in 2013 and in 2017 compared to the initial values prior to aeration. Determining the success of landfill remediation via solid samples is difficult, as the landfilled waste material is highly heterogeneous and landfill sampling is costly. We applied a cost-effective approach of setting up monitoring fields. Due to the close proximity of the samples to each other, the heterogeneity is somehow limited, but with the costs that the received picture is less representative (in comparison to considering the whole landfill). Although the concept of monitoring fields applied is suitable to assess the progress in degradation during the aeration project. The TOC showed no statistical significant decrease over time (see Fig. 5). This was to be expected, as the carbon quality greatly influences the biodegradability (Brandstätter et al., 2015a). The waste material contained many hardly degradable substances like plastics, wood and newspaper, all contributing to the TOC. During five years a significant reduction in the TOC would have been surprising, especially for material deposited 38 years before aeration start without a sealed surface cover. There was a noticeable impact on RI4, a parameter representing short-term biodegradability, a simple assessment for carbon quality. Even if not overall significant, at least the samples from 2016 significantly differed from the initial year 2012. Also, the overall trend in TOC is decreasing. A main driver for differing C-discharge levels between the sections was the different contents of MSW, the waste fraction richest in organics, in each section (Fig. 6). A visual inspection of the excavated material allowed for a rough material classification which led to a useful prediction of C-discharge (linear regression, r2 = 0.63). Leachate data could have contributed to a more representative performance assessment. For older landfills, the lack of a bottom sealing is rather common. In our case, there was none and also no leachate collection. To somewhat overcome this limitation, elution tests from solid material were analyzed. Ammonium is considered to be one of the most (Berge et al., 2006) problematic landfill leachate compound. Its reduction is a main reason for the application of In-situ aeration (Ritzkowski et al., 2006). In our case, NH4 was greatly reduced in the eluted solid samples. Also the significant increase of NO3 and SO4 indirectly showed, that the solids were getting continuously oxygenated. 3.3. MBT-material The litter bag data (MBT-material) showed significant decreases in TOC, RI4, EC, NH4 and cellulose in comparison to 2012 (Fig. 7).
239
Significant increases in NO3 were detected after five years. Sulfate and lignin also showed a significant increase after five years with an intermediate decrease (significant for lignin) after two years. Old landfilled material contains rather recalcitrant substances. This is less the case for younger or even fresh waste material. The significant change in the composition and reactivity of the MBT material (Fig. 7) indicates that the distributed air reached the material in all the litter bags, which can be considered as indication for the comparatively good and homogeneous distribution of the injected air. 4. Conclusions The results of the herein presented case study lead to the following conclusions: By using available online data of in-situ aeration projects, the C-discharge (and thus the mineralization rate) can be well assessed. Compared to laboratory tests, the here investigated field study showed a good performance indicated by a similar relative C-discharge (referred to initial TOC and air/solid ratio). Compared to other in-situ aeration projects, a rather high specific C-discharge (8 g C kg1 dry weight) was observable especially considering the rather high age of the Heferlbach landfill. Significant changes for the solids were observable for reactivity related parameters only (RI4, elution tests), not for the bulk TOC content. The rather big differences between the sections could be related to their respective MSW content. The introduced concept of using monitoring fields allowed for the first time to observe a significant decrease in the NH4-concentration of field material. The introduction of foreign (reactive) material allowed assessing the air distribution inside the landfill body. All samples showed significant reduction in the TOC as well as in the parameters related to reactivity. All in all the shallow landfill morphology in combination with the horizontal aeration of a rather dry landfill showed good results. Acknowledgments We would like to thank Manuel Hahn, Philipp Aschenbrenner, Stefan Spacek, Ole Peer Mallow and Ernis and Zdravka Saracevic for laboratory analyses. Many thanks to Thomas Ters for lignin and cellulose measurements, Ingeborg Hengl for graphical support (Fig. 1) and sincere thanks also to our project partners WGM (Wiener Gewässer Management). Funding This research was partly funded by the Kommunalkredit Public Consulting with the support of the Austrian Federal Ministry of Agriculture, Forestry, Environment and Water Management and MA48. Appendix A. Supplementary material Supplementary data to this article can be found online at https://doi.org/10.1016/j.wasman.2019.10.022. References Austrian Standard Institute, 2002. ON S 2027-1: Stability Parameters describing the biological reactivity of mechanically-biologically pretreated residual wastes – Part 1. Austrian Standard Institute, 2007. ON EN ISO 10304-1: Water quality Determination of dissolved anions by liquid chromatography of ions - Part 1: Determination of bromide, chloride, fluoride, nitrate, nitrite, phosphate and sulfate (ISO 10304-1:2007)
240
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
Berge, N.D., Reinhart, D.R., Dietz, J., Townsend, T., 2006. In situ ammonia removal in bioreactor landfill leachate. Waste Manage. 26 (4), 334–343. Brandstätter, C., Laner, D., Fellner, J., 2015a. Carbon pools and flows during lab-scale degradation of old landfilled waste under different oxygen and water regimes. Waste Manage. 40, 100–111. Brandstätter, C., Laner, D., Fellner, J., 2015b. Nitrogen pools and flows during labscale degradation of old landfilled waste under different oxygen and water regimes. Biodegradation 26 (5), 399–414. Brandstätter, C., Laner, D., Prantl, R., Fellner, J., 2014. Using multivariate regression modeling for sampling and predicting chemical characteristics of mixed waste in old landfills. Waste Manage. 34 (12), 2537–2547 http:// www.sciencedirect.com/science/article/pii/S0956053X1400364X. Fellner, J., Döberl, G., Allgaier, G., Brunner, P.H., 2009. Comparing field investigations with laboratory models to predict landfill leachate emissions. Waste Manage. 29 (6), 1844–1851. https://doi.org/10.1016/j.wasman.2008.12.022. German Insitute for Standardization, 2005. DIN EN ISO 11732-4: Water quality Determination of ammonium nitrogen Method by flow analysis (CFA and FIA) and spectrometric detection. German Institute for Standardization, 1996. DIN EN ISO 11905-1: Water quality Determination of nitrogen Part 1: Method using oxidative digestion with peroxodisulfate. Hrad, M., Huber-Humer, M., 2017. Performance and completion assessment of an in-situ aerated municipal solid waste landfill – final scientific documentation of an Austrian case study. Waste Manage. 63, 397–409. Liu, L., Ma, J., Xue, Q., Shao, J., Chen, Y., Zeng, G., 2018. The in situ aeration in an old landfill in China: multi-wells optimization method and application. Waste
Manage. 76, 614–620 https://www.sciencedirect.com/science/article/pii/ S0956053X18301077. Nag, M., Shimaoka, T., Nakayama, H., Komiya, T., Xiaoli, C., 2016. Field study of nitrous oxide production with in situ aeration in a closed landfill site. J. Air Waste Manag. Assoc. 66 (3), 280–287 http://www.tandfonline.com/doi/full/10. 1080/10962247.2015.1130664. Prantl, R., Tesar, M., Huber-Humer, M., Lechner, P., 2006. Changes in carbon and nitrogen pool during in-situ aeration of old landfills under varying conditions. Waste Manage. 26 (4), 373–380. Core Team, R., 2018. R: A Language and Environment for Statistical Computing. R Foun-dation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/Ritzkowski, M., Heyer, K. U., Stegmann, R., 2006. Fundamental processes and implications during in situ aeration of old landfills. Waste Manage. 26(4), pp. 356–372. Ritzkowski, M., Stegmann, R., 2007. Controlling greenhouse gas emissions through landfill in situ aeration. Int. J. Greenhouse Gas Control 1 (3), 281–288. Ritzkowski, M., Stegmann, R., 2012. Landfill aeration worldwide: concepts, indications and findings. Waste Manage. 32 (7), 1411–1419. https://doi.org/ 10.1016/j.wasman.2012.02.020. Ritzkowski, M., Stegmann, R., 2013. Landfill aeration within the scope of postclosure care and its completion. Waste Manage. 33 (10), 2074–2082 https:// www.sciencedirect.com/science/article/pii/S0956053X13000718. TAPPI, 1999. TAPPI useful method UM 250, Acid-soluble lignin in wood and pulp. In: Test Methods 1998–1999. TAPPI Press, Atlanta, USA. Wickham, H., 2016. ggplot2: Elegant Graphics for Data Analysis. Springer, New York.
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Performance assessment of landfill in-situ aeration – A case study Christian Brandstätter a,⇑, Roman Prantl b, Johann Fellner a a b
Institute for Water Quality, and Resource Management, TU Wien, Karlsplatz 13/226-2, 1040 Vienna, Austria blp GeoServices gmbh, Felberstraße 24/1, 1150 Vienna, Austria
a r t i c l e
i n f o
Article history: Received 5 June 2019 Revised 28 August 2019 Accepted 10 October 2019
Keywords: In-situ aeration Landfill Municipal solid waste Aftercare CO2 CH4
a b s t r a c t The assessment of landfill in-situ aeration eludes standardization as its application highly depends on varying local conditions. The prevailing work tries to assess the procedure performance by using typically available data. In the here presented case study the aeration pipes were applied horizontally. To evaluate the airdistribution and its effect on the landfill solid body, two monitoring fields with 10 10 m were created. From there in total 60 solid and 336 gaseous samples were taken over five years. As the material from the landfill was rather old and characterized by comparatively low reactivity, ‘‘new” material from a mechanical biological treatment (MBT) plant was introduced in the landfill. Additionally online gas data from eight technically separated landfill sections were analyzed during in-situ aeration. In total, about 46 Mm3 (0.27 m3/kg waste) air was introduced into the landfill body. The eight sections showed differences in reactivity (overall C-discharge was 8 g C kg1 dry weight, ranging from 4.5 to 11). With solid sampling we could not show a significant decrease in landfill TOC but for the introduced MBTmaterial. Ammonium in solid samples was significantly reduced (to 14.7% initial) and NO3 significantly increased (2.1% initial). The reduction of the initial TOC (4.58%) was on average 11%. The application of horizontal landfill aeration led to a widespread air-distribution in a rather shallow landfill. Monitoring fields allowed for a screening of the impact of the measures on the solid body with reduced sampling costs. Ó 2019 Elsevier Ltd. All rights reserved.
1. Introduction The degradation of organic matter through microorganisms is a major driving process for landfill emissions. Landfill in-situ aeration targets to enhance this process by providing oxygen for microbial aerobic respiration. This enables microorganisms to exploit previously unavailable resources, like anaerobically hardly degradable woody parts and allows for a faster biological waste stabilization. The anaerobic and aerobic degradation processes lead to a change in carbon quality, as the easier digestible carbohydrates are degraded faster. The in-situ aeration method is by now well established especially in Germany, but also in other countries (Liu et al., 2018; Prantl et al., 2006; Ritzkowski and Stegmann, 2012), where numerous in-situ aeration projects are currently in operation. The method ideally should be applied after potential landfill gas extraction is not economically feasible anymore. Its major benefits are the reduction of leachate pollution, mainly in the form of ammonium (NH4, Ritzkowski et al. (2006)) and chemical oxygen ⇑ Corresponding author. Tel.: +43 1 31304 5528. E-mail address: [email protected] (C. Brandstätter). https://doi.org/10.1016/j.wasman.2019.10.022 0956-053X/Ó 2019 Elsevier Ltd. All rights reserved.
demand (COD), as well as the reduction of the remaining methane (CH4) production potential (Ritzkowski and Stegmann, 2007). Landfill in-situ aeration however, also shows negative sideeffects, like increased CO2 emissions through accelerated microbial activity or the increased release of the potent greenhouse gas N2O (Nag et al., 2016). Despite the positive effects, assessing the success of in-situ aeration projects remains a challenge (Hrad and Huber-Humer, 2017). The evaluation is not only a strictly technical question, but also stretches out into the legal issue of when to release a landfill out of care. This is because the method is targeted to older landfills, possibly close to such a potential release point. Six potential criteria for the determination of the endpoints were presented elsewhere (Ritzkowski and Stegmann, 2013). Half of them require testing of the solid waste material in some way, either for determining the remaining emission potential using landfill simulation reactors (LSR), or with other degradation experiments. But because of the heterogeneous nature of landfills, deriving landfill properties from solid samples is difficult (Brandstätter et al., 2014). Other proposed criteria were landfill settling, temperature measurements or CH4-formation.
232
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
It is not feasible for practitioners, municipalities or scientists, to gather the necessary amount of solid samples before, during and after an in-situ aeration measures, for creating a proper data base for sound assessment of the landfill status. This is why we have to search for more cost-effective means of performance assessment, as we are attempting with the present work. Hence, the aim of the present paper is to combine the often available process data of landfill in-situ aeration projects and extend them with information from a limited amount of costly measurements of solid material. The gathered data stem from an Austrian case study, the Heferlbach Landfill. Here an in-situ aeration was applied with the focus of preventing landfill gas to migrate into neighboring buildings. In particular, we 1. analyzed the online data of a horizontal insitu aeration, technically separated into eight sections, 2. established monitoring fields on-site for solid sampling, so we could observe ongoing trends concerning the development of the landfill solid body during in-situ aeration in a cost-effective way and 3. introduced reactive material from a mechanical biological treatment, as a proxy for assessing the impact of the measures on material of known composition. 2. Materials and methods 2.1. Landfill description The landfill was filled between 1965 and 1974. The deposited material consisted of mixed municipal solid waste (~60% total mass TM), construction and demolition waste and excavated soil. The site is rather shallow with an area of roughly 65,000 m2 and an average depth of 3.7 m. The landfill had a total volume of ~240.000 m3. An overview of the landfill, including sampling points and aeration pipes is given in Fig. 1. 2.2. Sampling In 2001, on behalf of the Austrian Federal Agency of Environment, a site screening was conducted. Thereby over 100 solid samples were taken and analyzed. The impact of leachate emissions was considered insignificant, as a neighboring industry extracted and used the ground water underneath the landfill. For multiple reasons, especially to not disturb close-by residents, it was decided to apply the measure of in-situ aeration for minimizing the risk of potential gas migration into neighboring cellars. Since the waste material was more than 40 years old, and the landfill also contained waste material of low biological reactivity, more reactive regions rich in organic material were actively selected for the solid waste sampling in 2012. Finding these reactive landfill regions was based on the available measurement data (TOC in the solids and CO2/CH4) from 2001. During that campaign, 56 samples of roughly 30 kg were taken with an excavator and shovels. The material was screened with a mesh width of 20 mm. Based on the information available from the site screening 2001 and our own analysis, an overview of the main landfill properties is given in Table 1. In a recent research project, laboratory experiments were conducted (Brandstätter et al., 2015a,b) for getting a good measure of biodegradability of the waste material. In addition to the initial sampling, two monitoring fields (each 10 10 m) were defined in sections known to contain more organic material. From these monitoring fields, every year twelve samples were taken in three depth levels out of four pits dug by an excavator. In the first year also 16 gas sampling probes were erected in varying distances to the aeration and gas extraction pipes as well as in differing depths. Roughly every three months the concentrations of CH4 and CO2 were measured in different dis-
tances from the aeration and extraction pipes as well as in different depths. As a grouping variable the distance to the aeration pipes was calculated using the Pythagorean formula:
distanceair ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 ðDXY Þ2 þ ðDdepthÞ
ð1Þ
with DXY as the horizontal distance between the gas probe and the nearest aeration pipe and Ddepth as the difference between the average filter depth (filter upper edge - filter lower edge) and the average depth of the aeration pipes (2.5 m). All units are m. In 2012, ten samples of MBT-waste (organic residues from mechanical biological treatment), each about 10 kg were buried in the landfill. Since the available landfill material was rather old, the idea was to introduce material of known composition and reactivity, in order to detect potential inequalities in air distribution with higher sensitivity. The material was put into ten separate bags, sewed of fly-screen mesh (~40 60 cm, litter bags). Upon placement, temperature loggers (Voltcraft) were placed inside. The MBT-material was placed in different depths and in varying distance to the aeration pipes. For each pit, either two or three litter bags were placed inside. Subsequently, the litter bags were dug out in 2014 after two years to gain access to the buried temperature loggers. All ten were found, but most temperature loggers stopped working after two weeks (data not shown). Then material from each pit was pooled, mixed, divided and placed again (n = 8). This procedure was necessary, as many of the bags were cut open during the sampling procedure. In 2017 they were recovered for good and five valid samples from four different trenches could be collected. 2.3. In-situ aeration On-site, a low-pressure horizontal in-situ aeration was established, since the landfill was rather shallow and long (see Section 2.1). Before the full-scale application, a technical prototype for gathering information about the full-scale capacity was installed, mainly to determine the aeration rates. The prototype was installed for a whole section and could be used even after the test-phase for the main aeration measure. For the insertion of the aeration and extraction pipes, directional drilling was used. The aeration pipes were introduced in 3–4 m depth and the extraction pipes in 2 m. The pipes (PE 90, thickness: 8.1 mm, single pipe length: 6 m) were welded and held around 100 holes/meter with a hole diameter of 10 mm. Water was used as jetting liquid. The pipes were put in place with approximate a distance of 10 m to each other. For safety reasons, the air extraction is roughly 1.5fold higher than the aeration (Table 3). In addition, southwards at the edge of the landfill a security gas extraction system (side extraction) was established, to prevent any gas migration to neighboring buildings and cellars. The whole aeration project was separated into eight sections, where the gas concentrations, temperature, flow and pressure data were analyzed separately in a centralized unit. The off-gas was directed through a biofilter to reduce odor. The facility allowed for reversing both the extraction and the aeration pipes. For determining the mass flow, a MFT-B device from KURZ Instruments Inc. was installed. For measuring the gas pressure, explosion-proof pressure transducers from WIKA were used. For determining the respective concentrations of CO2, CH4 and O2, a measuring device from Siemens AG (ULTRAMAT 23) was used. The determination of CO2 and CH4 was done via infrared measurement and the oxygen concentration was measured with a para-magnetic pressure determination system. In the summer 2013 the landfill got submerged, as it is situated in high proximity to the river Danube. Thus the aeration had to be stopped for a few weeks.
233
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
Fig. 1. Landfill overview.
Table 1 Landfill Characteristics per Section. Variable
Unit
SE01
SE02
SE03
SE04
SE05
SE06
SE07
SE08
Area Total Volume Total Volume dry MSW Top Soil Excavated Soil C&D-Waste TOC* Solid Samples 2012
[m2] [m3] [m3] [m3] [m3] [m3] [m3] [Mg] [n]
7300 25,000 19,700 10,900 2200 2200 11,900 900 15
8900 27,200 21,400 16,800 2500 8700 1600 980 10
8800 29,400 23,100 20,300 1800 3100 5900 1060 20
7700 30,000 23,700 22,800 1500 3100 4100 1100 –
7500 24,900 19,600 16,100 1900 7100 1700 900 –
4300 11,800 9,300 5680 780 2000 4100 430 –
8100 31,000 24,500 28,600 2400 1800 680 1120 9
12,300 39,000 31,000 23,300 2900 11,600 4400 1400 2
Notes: The initial properties were derived from our own sampling campaign (2012) and missing data were filled from a sample campaign conducted by the Austrian Federal Agency of Environment 2001 for overall site assesment (data not shown). The density of the landfilled MSW was assumed 1. The amount of MSW is a rough estimate and was not considered for further calculations. Numbers were rounded to 2 (or three, if the leading digit was <3) significant digits. *For TOC, only the fine fraction (<20 mm) was considered, with an average value of 4.58 [% DM]. For the water content an average value of 21.2 [% DM] was considered. DM dry mass, MSW municipal solid waste, SE section, TM total mass, TOC total organic carbon.
2.4. Analytics
2.5. Calculations
The applied analytical procedures are briefly summarized in Table 2:
For the field data, the total amount of material, or more precise the total dry weight (DW) was used as reference value. For
234
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
Table 2 Sample measurements. Variable Solids Acid soluble lignin Cellulose Total Carbon Respiration Index after 4 days Water content
Abbreviation
Device/Procedure/Norm
LigninAS – TOC RI4
TAPPI (1999) Agilent 1200 series HPLC equipped with a refractive index detector (RID) and a Shodex sugar SP0810 column MACRO CHNS analyzer (Elementar Analysensysteme GmbH, Hanau, Germany) ON S 2027-1 Austrian Standard Institute (2002)
WC
2.5 kg of material in drying oven until weight remained constant.
Eluted Elution Electrical conductuvity Ammonium
– EC NH4
Nitrate
NO3
Sulfate
SO4
Solid/liquid ratio 1:10, 24 h of shaking Seven excellence S470 (Mettler-Toledo S.A.E, Spain) DIN EN ISO 11732 German Insitute for Standardization (2005), continuous flow analyzer (Skalar Analytical, B.V., Netherlands) DIN EN ISO 13395 German Institute for Standardization (1996), continuous flow analyzer (Skalar Analytical, B.V., Netherlands) ON EN ISO 1030-1, Austrian Standard Institute (2007), Dionex ICS 900 (Thermo Fisher Scientific Inc., USA)
Table 3 Technical characteristics of the in-situ aeration. Variable
Unit
SE01
SE02
SE03
SE04
SE05
SE06
SE07
SE08
Inflow Field Extraction Side Extraction Temperature* Settelement
[m3 kg1 DW] [m3 kg1 DW] [m3 kg1 DW] [°C] [% average height]
0.346 0.466 0.249 16.5 (8.7) 3.9
0.280 0.381 0.261 20.9 (9.4) 2.5
0.274 0.352 0.293 18.9 (9.3) 1.7
0.279 0.351 0.270 16.2 (9.2) 1.8
0.209 0.269 0.429 16.1 (9.8) 5.7
0.420 0.542 0.328 21.3 (11) 3.4
0.306 0.465 0.246 25.9 (8.8) 6.4
0.162 0.254 0.246 20.1 (7.9) 3.8
Notes: Data are derived from a time span over 1886 days. *Average Temperature Field Extraction with standard deviation in brackets.
calculating the C-discharge both the concentrations of CO2 and CH4 were considered, used with the general gas equation per time point:
n¼
pV_ RT
ð2Þ
with n as the amount of substance [mol], R as the universal gas constant [J mol1 K1], T as the temperature [K], p as the atmospheric pressure [Pa] and V_ as the volumetric fraction per time point t [m3 t1]. The underlying field data used in this article was aggregated to daily values from five minute time steps. The measurement of temperature, concentration and volumetric flow were all conducted in
Fig. 2. Specific C-discharge from 1st Feb. 2012 to 31st Mar. 2017. Notes: The C-discharge is calculated per per DW (dry weight). The samples were sieved before processing, so only the fraction <20 mm was considered.
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
235
Fig. 3. Specific C-discharge - Field vs. Laboratory. Notes: DW: dry weight, LSR: landfill simulation reactor.
Fig. 4. Gas Concentration vs. Distance to Aeration Pipes. Notes: Gas concentrations were measured in eight gas wells per monitoring field (16 in total). The distance values were classified to low (n = 9), middle (n = 28) and high (n = 9) with breaks at 3.3 and 6.6 m and a maximum distance of 10 m. The numbers in the plot and horizontal lines denote the mean values averaged over the whole time frame. The thin lines denote the measured concentration averaged per distance class. The higher variation starting from early 2015 is a direct result of aeration optimization. The areas denote the standard errors per time point within the respective group.
the central unit. The amount of substance then was multiplied with 12 [g mol1] to receive [g] carbon. All the data processing including visualization was done with R (version 3.5.1, R Core Team (2018)) and the package ggplot2 (version 2_3.0.0, Wickham (2016)). For statistical group comparison of the concentration data one-way ANOVAs were computed, followed by Tukey-HSD tests.
3. Results and discussion 3.1. Online (Gas) data The average off-gas temperature from the field extraction over the whole time-frame and all the sections was 19.5 °C (Table 3). In total, 46 Mm3 (0.284 m3 kg1 DW) of air were added and
236
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
Fig. 5. Solid Samples taken from the Monitoring Fields. Notes: Different letters denote significant differences between the years, if applicable (a is highest). The distance values were classified to low (n = 9), middle (n = 28) and high (n = 9) with breaks at 3.3 and 6.6 m and a maximum of 10 m. The highlighted samples from the same pit correspond to the litterbag samples. DistAer: Distance to Aeration Pipes, EC: Electrical Conductivity, RI4 : Respiration Index after 4 days.
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
237
Fig. 6. C-discharge vs. MSW-content per landfill Section.
112 Mm3 (0.675 m3 kg1 DW) being removed for the considered time period of five years. The high amount of air extraction (2.4fold higher than air inflow) can be attributed to the additional side-extraction system (63 Mm3, or 0.385 m3 kg1 DW) which was established for safety reasons. The average settlement after five years was about 3.6% of the total height (mean: 13.5 cm ± 2.3 (standard error SE)). The sections showed high variation in their constitution, their area as well as the amount of air added and extracted (Tables 1 and 3). The specific C-discharge per landfilled dry mass (Fig. 2) also showed high variations between the sections ranging from 4.5 to 11 g C kg1 DW over the whole timeframe. The C-discharge also decreased over the time ranging from 2.8 in 2012 to 0.6 in 2016, the last completely assessed year. In comparison to the laboratory experiment, the specific aeration rate in the field was much lower (roughly 1/10, see Fig. 3). The average reduction in TOC in the field was 11% (ranging from 9.7 to 23.2%), while in the laboratory experiment it was 33.7% after two years of aeration. It must be noted, that for the laboratory the material was already collected from organic hot-spots, while in the field the whole waste body was considered, including inorganic material. As was to be expected from an old landfill containing municipal solid as well as construction and demolition waste, the different landfill areas showed high variation in composition and degradation behavior (Fig. 2). The overall C-discharge was 1380 Mg (8 g C kg1 dry weight) is rather high in comparison to other projects (Hrad & Huber-Humer, 2017), which reported values ranging from 3 to 17.2 g C kg1 DW. This is especially the case when considering the rather high waste age (closed 1974, the aeration started 2012), low initial reactivity (average RI4: 3.3 mg O2 kg1 DW) of the here presented landfill Heferlbach. For reference, the here investigated aeration duration was about 63 months with an air/solid ratio of 270 m3 Mg1 DW. We contribute the main reason for this rather good performance to the shallow landfill anatomy in combination with the horizontal aeration. In the prevailing case it also was possible to
investigate eight different subsections and even operate them separately. Therefore we could provide a performance assessment for each section and propose changes in the operational parameters. To not exceed the scope of the paper, we deliberately excluded the details of the optimization procedure. In the laboratory, with an air/solid ratio about double as high as in the field, also the relative C-discharge (referred to the TOC content of the waste) was about double as high (see Fig. 3). However, it needs to be noted, that the reference value for the relative C-discharge was TOC, whose determination relies on solid sampling and is thus highly uncertain.
3.2. Data from the monitoring fields 3.2.1. Gas concentrations The gas concentrations in the landfill measured in the monitoring fields showed an overall decrease in CO2 and CH4 (Fig. 4). Grouping the data from the 16 wells according to their distance to the aeration pipes allowed for assessing the air distribution in the landfill. The distance was classified as <3.3 m: low, 3.3 and <6.6 m: middle, 6.6 m: high. In the gas concentration data from the monitoring fields, a clear impact of a flooding event in 2013 can be noticed. After the flooding, CO2 and CH4 concentrations increased. This can be related to technical changes, as the gas flows pipes were reversed to unblock the aeration pipes. At the end of 2013 the gas concentrations in the field changed to a rather stable level. The impact of water transport strongly affected the gas concentrations locally in the monitoring fields. On a landfill level (online data) the effect was not as pronounced, except for the immediate stop for a few weeks after the flooding event. The influence to the extraction distance was vice versa but less pronounced in comparison to the distance to the aeration pipes (data not shown). The aeration measure was stopped for several weeks until the water was drained and the pipes unblocked. This lead to a short term increase in CH4 and CO2 which was to be expected.
238
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
Fig. 7. Composition of MBT-material (litter bag data). Notes: Different letters denote significant differences between the years, if applicable (a is highest). LigninAS: acidsoluble fraction of lignin, RI4: Respiration Index after 4 days.
The disturbance of the online gas-data of the off-gas concentrations was not that big, except during the aeration stop. From the gas concentrations in dependency of the distance to aeration pipes (see Fig. 4) it was possible to derive an indirect over-
view of the quality of the gas distribution through the aeration: after the flooding a rather stable distribution regime was reached. For this aeration measures, a distance to the aeration pipes of more than 6.6 m lead to increased CH4 and CO2 concentrations. This
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
might be caused by the rather low landfill depth and probably also by heterogeneity in the waste body resulting in preferential flow paths in differing directions, as it was already discussed for water (Fellner et al., 2009). However, in the off-gas concentrations of the aeration there was no increase in CH4 to be noticed. 3.2.2. Solids Every year in spring, twelve samples were taken in the monitoring fields. The analyses of TOC did not show a significant reduction (Fig. 5). The analyses of RI4 allowed the observation of a decrease from 3.2 to 1.45 in 2016, where the median also dropped below 0.15 [mg O2 kg1 DW] with slightly higher values in 2017. The median in 2012 was 2.8, the only year with a significant reduction in comparison to 2012 was 2016 according to Tukey’s HSD test. The eluted parameters behaved differently (Fig. 5). For the electrical conductivity (EC) no decreasing or increasing trend could be observed. Over five years, a significant reduction in NH4-N after elution could be observed. This was accompanied by a significant increase of NO3-N only after five years in 2017. For SO4, a parameter indicating an increase in the redox potential, the trend is significantly increasing as well. An increase in SO4 could be observed already in 2013 and in 2017 compared to the initial values prior to aeration. Determining the success of landfill remediation via solid samples is difficult, as the landfilled waste material is highly heterogeneous and landfill sampling is costly. We applied a cost-effective approach of setting up monitoring fields. Due to the close proximity of the samples to each other, the heterogeneity is somehow limited, but with the costs that the received picture is less representative (in comparison to considering the whole landfill). Although the concept of monitoring fields applied is suitable to assess the progress in degradation during the aeration project. The TOC showed no statistical significant decrease over time (see Fig. 5). This was to be expected, as the carbon quality greatly influences the biodegradability (Brandstätter et al., 2015a). The waste material contained many hardly degradable substances like plastics, wood and newspaper, all contributing to the TOC. During five years a significant reduction in the TOC would have been surprising, especially for material deposited 38 years before aeration start without a sealed surface cover. There was a noticeable impact on RI4, a parameter representing short-term biodegradability, a simple assessment for carbon quality. Even if not overall significant, at least the samples from 2016 significantly differed from the initial year 2012. Also, the overall trend in TOC is decreasing. A main driver for differing C-discharge levels between the sections was the different contents of MSW, the waste fraction richest in organics, in each section (Fig. 6). A visual inspection of the excavated material allowed for a rough material classification which led to a useful prediction of C-discharge (linear regression, r2 = 0.63). Leachate data could have contributed to a more representative performance assessment. For older landfills, the lack of a bottom sealing is rather common. In our case, there was none and also no leachate collection. To somewhat overcome this limitation, elution tests from solid material were analyzed. Ammonium is considered to be one of the most (Berge et al., 2006) problematic landfill leachate compound. Its reduction is a main reason for the application of In-situ aeration (Ritzkowski et al., 2006). In our case, NH4 was greatly reduced in the eluted solid samples. Also the significant increase of NO3 and SO4 indirectly showed, that the solids were getting continuously oxygenated. 3.3. MBT-material The litter bag data (MBT-material) showed significant decreases in TOC, RI4, EC, NH4 and cellulose in comparison to 2012 (Fig. 7).
239
Significant increases in NO3 were detected after five years. Sulfate and lignin also showed a significant increase after five years with an intermediate decrease (significant for lignin) after two years. Old landfilled material contains rather recalcitrant substances. This is less the case for younger or even fresh waste material. The significant change in the composition and reactivity of the MBT material (Fig. 7) indicates that the distributed air reached the material in all the litter bags, which can be considered as indication for the comparatively good and homogeneous distribution of the injected air. 4. Conclusions The results of the herein presented case study lead to the following conclusions: By using available online data of in-situ aeration projects, the C-discharge (and thus the mineralization rate) can be well assessed. Compared to laboratory tests, the here investigated field study showed a good performance indicated by a similar relative C-discharge (referred to initial TOC and air/solid ratio). Compared to other in-situ aeration projects, a rather high specific C-discharge (8 g C kg1 dry weight) was observable especially considering the rather high age of the Heferlbach landfill. Significant changes for the solids were observable for reactivity related parameters only (RI4, elution tests), not for the bulk TOC content. The rather big differences between the sections could be related to their respective MSW content. The introduced concept of using monitoring fields allowed for the first time to observe a significant decrease in the NH4-concentration of field material. The introduction of foreign (reactive) material allowed assessing the air distribution inside the landfill body. All samples showed significant reduction in the TOC as well as in the parameters related to reactivity. All in all the shallow landfill morphology in combination with the horizontal aeration of a rather dry landfill showed good results. Acknowledgments We would like to thank Manuel Hahn, Philipp Aschenbrenner, Stefan Spacek, Ole Peer Mallow and Ernis and Zdravka Saracevic for laboratory analyses. Many thanks to Thomas Ters for lignin and cellulose measurements, Ingeborg Hengl for graphical support (Fig. 1) and sincere thanks also to our project partners WGM (Wiener Gewässer Management). Funding This research was partly funded by the Kommunalkredit Public Consulting with the support of the Austrian Federal Ministry of Agriculture, Forestry, Environment and Water Management and MA48. Appendix A. Supplementary material Supplementary data to this article can be found online at https://doi.org/10.1016/j.wasman.2019.10.022. References Austrian Standard Institute, 2002. ON S 2027-1: Stability Parameters describing the biological reactivity of mechanically-biologically pretreated residual wastes – Part 1. Austrian Standard Institute, 2007. ON EN ISO 10304-1: Water quality Determination of dissolved anions by liquid chromatography of ions - Part 1: Determination of bromide, chloride, fluoride, nitrate, nitrite, phosphate and sulfate (ISO 10304-1:2007)
240
C. Brandstätter et al. / Waste Management 101 (2020) 231–240
Berge, N.D., Reinhart, D.R., Dietz, J., Townsend, T., 2006. In situ ammonia removal in bioreactor landfill leachate. Waste Manage. 26 (4), 334–343. Brandstätter, C., Laner, D., Fellner, J., 2015a. Carbon pools and flows during lab-scale degradation of old landfilled waste under different oxygen and water regimes. Waste Manage. 40, 100–111. Brandstätter, C., Laner, D., Fellner, J., 2015b. Nitrogen pools and flows during labscale degradation of old landfilled waste under different oxygen and water regimes. Biodegradation 26 (5), 399–414. Brandstätter, C., Laner, D., Prantl, R., Fellner, J., 2014. Using multivariate regression modeling for sampling and predicting chemical characteristics of mixed waste in old landfills. Waste Manage. 34 (12), 2537–2547 http:// www.sciencedirect.com/science/article/pii/S0956053X1400364X. Fellner, J., Döberl, G., Allgaier, G., Brunner, P.H., 2009. Comparing field investigations with laboratory models to predict landfill leachate emissions. Waste Manage. 29 (6), 1844–1851. https://doi.org/10.1016/j.wasman.2008.12.022. German Insitute for Standardization, 2005. DIN EN ISO 11732-4: Water quality Determination of ammonium nitrogen Method by flow analysis (CFA and FIA) and spectrometric detection. German Institute for Standardization, 1996. DIN EN ISO 11905-1: Water quality Determination of nitrogen Part 1: Method using oxidative digestion with peroxodisulfate. Hrad, M., Huber-Humer, M., 2017. Performance and completion assessment of an in-situ aerated municipal solid waste landfill – final scientific documentation of an Austrian case study. Waste Manage. 63, 397–409. Liu, L., Ma, J., Xue, Q., Shao, J., Chen, Y., Zeng, G., 2018. The in situ aeration in an old landfill in China: multi-wells optimization method and application. Waste
Manage. 76, 614–620 https://www.sciencedirect.com/science/article/pii/ S0956053X18301077. Nag, M., Shimaoka, T., Nakayama, H., Komiya, T., Xiaoli, C., 2016. Field study of nitrous oxide production with in situ aeration in a closed landfill site. J. Air Waste Manag. Assoc. 66 (3), 280–287 http://www.tandfonline.com/doi/full/10. 1080/10962247.2015.1130664. Prantl, R., Tesar, M., Huber-Humer, M., Lechner, P., 2006. Changes in carbon and nitrogen pool during in-situ aeration of old landfills under varying conditions. Waste Manage. 26 (4), 373–380. Core Team, R., 2018. R: A Language and Environment for Statistical Computing. R Foun-dation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/Ritzkowski, M., Heyer, K. U., Stegmann, R., 2006. Fundamental processes and implications during in situ aeration of old landfills. Waste Manage. 26(4), pp. 356–372. Ritzkowski, M., Stegmann, R., 2007. Controlling greenhouse gas emissions through landfill in situ aeration. Int. J. Greenhouse Gas Control 1 (3), 281–288. Ritzkowski, M., Stegmann, R., 2012. Landfill aeration worldwide: concepts, indications and findings. Waste Manage. 32 (7), 1411–1419. https://doi.org/ 10.1016/j.wasman.2012.02.020. Ritzkowski, M., Stegmann, R., 2013. Landfill aeration within the scope of postclosure care and its completion. Waste Manage. 33 (10), 2074–2082 https:// www.sciencedirect.com/science/article/pii/S0956053X13000718. TAPPI, 1999. TAPPI useful method UM 250, Acid-soluble lignin in wood and pulp. In: Test Methods 1998–1999. TAPPI Press, Atlanta, USA. Wickham, H., 2016. ggplot2: Elegant Graphics for Data Analysis. Springer, New York.