innovative semi-aerobic landfill management in tropical
INNOVATIVE SEMI-AEROBIC LANDFILL MANAGEMENT IN TROPICAL COUNTRIES V. GROSSULE* AND M.C. LAVAGNOLO** * ICEA, Department of Civil, Architectural and Environmental Engineering, University of Padova. Via Marzolo 9, 35131 Padova, Italy ** DII, Department of Industrial Engineering, University of Padova. Via Marzolo 9, 35131 Padova, Italy
SUMMARY: Under tropical climate conditions, the rainfall seasonality represents often a limiting factor in proper functioning of semi-aerobic landfill system: lack of moisture during the dry season tends to slow down the biological activity, and the heavy rainfalls during the wet season result in high contaminants load in leachate (Rafizul and Alamgir, 2012; Tränkler et al., 2005). An innovative management strategy of semi aerobic landfills under tropical climates is proposed to improve the stabilization performances, by exploiting the dry season for simulating composting process, recirculating the leachate stored during the rainy season, and the wet season to obtain a flushing effect. The experimental study was carried out during six months, simulating the dry and wet season, on eight landfill simulating bioreactors: four semi-aerobic and four anaerobic. Two types of waste were used: low and high organic content waste. The synergetic effect of semi aerobic conditions and the sequencing first composting phase and second flushing phase was evaluated. Samples of waste were taken and analyzed at the begging and at the end of the experiment and leachate and gas composition were monitored during the entire experiment. The results show the effectiveness of the innovative management under semi aerobic conditions leading to much higher stabilization efficiency, compared to the anaerobic ones, regardless of the kind of waste, the system capability in acting as heavy metals sink and in realizing the Final Storage Quality (FSQ) during the one year simulated period.
1. INTRODUCTION In recent years, the need for the implementation of sustainable landfill concept is increasing more and more in both developing and industrialized countries, requiring the achievement of the FSQ within one generation time, in order to minimize both the short and long term impacts of the landfill system to the environment and to the public (Cossu, 2012; Hrad et al., 2013; Heimovaara, 2014). In industrialized countries, some measures toward the sustainable landfilling have been implemented: in northern Italy, first legislation on sustainable landfill introduced in 2014 Final Storage Quality (FSQ) limits (D.G.R. 2461/14). In developing countries, open dumps are still the most prevalent type of disposal facilities due to financial constraints and limited technical knowhow (Oman and Manandhar, 2011). Thus, implementation of more economic and technological landfill design, able to fulfill the sustainability, is crucial for both developing and industrialized countries. In this context, the semi-aerobic landfill system may represent a proper solution able to simultaneously fulfill the sustainability concept and the need for low cost disposal alternative (Hirata et al., 2012). The semi aerobic landfill, developed at the Proceedings Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium/ 2 - 6 October 2017 S. Margherita di Pula, Cagliari, Italy / © 2017 by CISA Publisher, Italy
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Fukuoka University since more than 20 years ago, aims to reproduce an aerobic environment within the waste mass with a proper engineering designs, in which the ambient air flows into the waste body naturally through the leachate collection pipes, moved by the temperature gradient between inside and outside the landfill (Theng et al., 2005). Many researches have already been managed with the aim of evaluating the advantages and mechanism of the semi aerobic landfill (SALF), demonstrating its capability to achieve the aerobic process performance without any specific technical equipment: § Improvement of carbon degradation rate, higher than in an anaerobic system (Ahmadifar et al., 2016; Cossu et al., 2003; He et al., 2012); § Enhancement of nitrogen removal (Cossu et al., 2003; He et al., 2011, 2012; Shao et al., 2008; Yang et al., 2012; Zeng et al., 2006); § Reduction of methane generation (Ahmadifar et al., 2016; Cossu et al., 2003; Enough and Analysis, 2006; HUANG et al., 2008; Sutthasil et al., 2014; Wu et al., 2017; Yang et al., 2012; IPPC, 2006); § Increased carbon gasification (Shimaoka et al., 2000). Despite that, the rainfall seasonality of tropical climates represents often a limiting factor in SALF proper functioning: lack of moisture during the dry season tends to slow down the biological activity, and the heavy rainfalls during the wet season result in high contaminants load in leachate (Rafizul and Alamgir, 2012; Tränkler et al., 2005). According to the Köpper-Geiger climate classification, the tropical Aw climate (Tropical wet and dry Savanna climate) is the second most diffused climate worldwide (11.5%), characterized by a dry season, 5-6 month long, during which low or no rainfalls occur, while precipitation between 800-1600mm are concentrated during the wet season (Chen and Chen, 2013; Kottek et al., 2006; Peel et al., 2007; Essenwanger, 2001). The focus of this research was to develop an innovative management of semi aerobic landfill under the specific tropical climates in order to improve the SALF stabilization performances, by exploiting the climate seasonality: § During the dry season, the waste tipping in thin layers and the recirculation of leachate, stored during the rainy season, ensure oxygen intrusion and proper moisture content for realizing the composting process. § During the wet season, the leachate generated by the flushing effect, infiltrates trough the composted layer below, in which the humic and fulvic substances may provide a natural insitu physical-chemical treatment by absorbing heavy metals (Qu et al., 2008) and the formed microbial population during the dry season may enhance the in-situ biological treatment. The synergetic effect of semi aerobic conditions together with the innovative management has been evaluated on two different waste typologies, characterized by different organic fraction content: residual waste after separate collection, characterized by low organic content (“R” waste), and rich in putrescible fraction waste (“O” waste). The waste stabilization performances of the innovative management under semi aerobic conditions has been analyzed comparing the effectiveness with the two waste typologies and with the anaerobic system, considering the ammonia removal, the gasification enhancement, the methane generation reduction, the FSQ limits achievement and the capability in acting as final sink for heavy metals. Moreover, the role of the first composting phase on the improvement of leachate quality during the wet phase has been investigated, although further researches are required to clarified mechanisms and real contribute.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
2. MATERIALS AND METHODS 2.1 Waste samples The waste used for the experiment was provided by a public waste management company operating in northern Italy. The urban waste was taken after a mechanical treatment of shredding and a 6 cm sieving, without any biological pretreatment. Starting from the initial waste sample, two type of waste were used to fill the columns: unchanged residual waste (“R” waste) resulting in 4% kitchen residues content, and high in organic fraction waste (“O” waste) obtained modifying initial waste, increasing the putrescible fraction up to 50%, taken kitchen waste from municipal putrescible separate collection. The final waste characterization is provided in Table 1. The “R” (“Residual”) waste simulates the typical waste composition in industrialized countries after the separate collection; the “O” (“Organic”) waste was reproduced in order to simulate the typical waste composition in low-medium income countries (The World Bank, 2012).
Table 1. Waste fractions of “Residual” and ”Organic” waste. Waste categories Kitchen residues Green waste and wooden materials Paper and paperboard Textiles Plastics Metals Glass Inert Under-sieve
"Residual" % 4 6 20 6 24 4 10 10 16
"Organic" % 50 3 11 3 13 2 5 5 8
2.2 Equipment The experiment was carried out using eight Plexiglass® cylindrical lysimeters (1m height, inner diameter of 24 cm) to simulate anaerobic and semi aerobic conditions. At the bottom of each column 10cm thick gravel layer (Ø 20–30 mm) was placed before filling to facilitate the draining out of leachate. In order to simulate the passive aeration in semi aerobic lysimeters, a slotted aeration pipe (4 cm diameter) was placed at the bottom of semi aerobic columns, open to the air, and the top of the semi-aerobic columns was left completely open. 1 cm diameter plastic pipe were inserted in the waste body to measure gas quality. Anaerobic columns were completely sealed. Two valves at upper end of each anaerobic reactor were used for gas extraction by Tedlar sampling bags and water irrigation. Leachate extraction was performed trough a collection port located at the bottom of each reactor and temperature monitoring was ensured using the Thermo Systems TS100 temperature probes installed inside each reactor (Figure 1). Both Anaerobic and Semi aerobic columns were wrapped by thermo insulated material and located in a container equipped to control the environmental temperature.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Figure 1. Semi aerobic (a) and anaerobic (b) reactors set up.
2.3 Methodology The experimental study was carried out during six months, simulating the dry (0-96th day) and wet season (97th-192th day), on eight landfill simulating bioreactors, operated in duplicate under four different process conditions (Table 2): § An-R reactors: anaerobic conditions filled with “Residual” waste; § An-O reactors: anaerobic conditions filled with “Organic” waste; § S-R reactors: semi-aerobic conditions filled with “Residual” waste; § S-O reactors: semi-aerobic conditions filled with “Organic” waste. In semi aerobic columns the aeration was given by natural heat convection according to semiaerobic concept. At the beginning of both wet and dry phases, 9 kg of “Residual” and “Organic” waste were placed in AN-R, S-R columns and in An-O, S-O respectively, achieving a density of about 0.5 kg/L. Table 2. Operational conditions and waste type applied for the different experimental reactors. Waste typologies
Anaerobic
Semi aerobic
“O” waste (50% putrescible content)
An-O
S-O
“R” waste (4% putrescible content)
An-R
S-R
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
The experiment follows two main stages: during the first phase only first half of the bioreactors was filled, the daily water irrigation was initially defined according with PAF model (Cossu et al., 2003) and then adjust from 0.22 L/d to 0.05 L/d to guarantee composting process and thus a good final stabilization at the end of the third month. In the second phase the second halves of the columns were filled with the same amount and kind of waste of the first stratum and daily irrigation of 0.9 L/d simulated the weather condition of wet tropical season, considering a total precipitation of 1400 mm. Daily leachate extraction was performed throughout the entire test and samples were collected for analysis. Environmental temperatures were maintained with a cyclic half day temperature between 18-30°C. Both temperatures and precipitations were selected within the ranges suggested by the Whittaker biome classification for Savanna climate (Whittaker, 1975; Kottek et al., 2006). The stabilization performances of the semi aerobic system under the innovative management have been assessed considering the capability in achieving the Final Storage Quality (FSQ) according to the Italian regional legislation (D.G.R. 2461/14). Samples of waste were taken and analyzed for TOC, IR4, TS, VS and heavy metals, at the begging and at the end of the experiment. Leachate was monitored in terms of pH, VFA, alkalinity, COD, TOC, BOD5, TKN, N-NH4+, N-NO3-, N-NO2- and Cl- and biogas concentrations were analyzed in terms of CO2, CH4 and O2. The results are reported as mean values for each type of columns management.
3. RESULTS AND DISCUSSION 3.1 Leachate quality 3.1.1 pH, Volatile fatty acids, Alkalinity. The behavior of pH, Volatile fatty acids (VFA) and alkalinity, monitored throughout the entire test, are reported in figure 2. Excepted in S-O columns during the first days after fills, the pH values were maintained above 7 in all SALF reactors during the experiment. In An-O columns, acidogenic phase occurred though out the entire test and, maintaining pH values below 6, methanogenesis was inhibited (Cossu et al., 2016). Differently, in An-R reactors, a positive pH trend was observed after day 40, once the methanogenic phase started and then maintained during the second phase, with pH values above 6: from that time CH4 has been observed in AnRs biogas (Figure 6). Consistently with pH variation, VFA steeply decreased in all semi aerobic columns at the beginning of both phases, showing a very similar behavior both in case of “O” and “R” waste. In anaerobic reactors clear difference was observed since 40th day till the end of the dry phase according to the waste typology: in AN-O, VFA values were almost constant around 3400 mgCH3COOH/L, in AN-R, the contribute of methanogens led to a further decrease in fatty acids up to 2000 mgCH3COOH/L at the end of the dry phase. During the second phase no big differences in decreasing trend were detectable between the two columns: the flushing effect dominated the declining behavior of VFA. According to the pH and VFA behavior, alkalinity increased in An-R with the pH at the beginning of methanogens activity. During the wet phase it decreased due to the flushing effect. In all other columns the alkalinity values decreased during the entire test. For all the parameters much lower values were observed in wet phase due to the dilution effect. 3.1.2 TOC and COD Similar behavior to VFA was observed for TOC and COD parameters, reported in Figure 2. High removal performances were evident in all semi aerobic columns, with a steeply decrease
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
in TOC and COD values regardless of the waste typology, justified by the higher kinetics ensured under aerobic conditions (Ritzkowski et al., 2007). In anaerobic columns, the waste typology strongly influence stabilization performances: during the first phase, better results in AN-R were justified by the contribution of methanogens activity starting from 40th day; during the second phase the flushing effect dominated in anaerobic columns, according to the positive trend of TOC/Chloride ratio (Figure 4), and the low organic content in AN-R ensured lower TOC concentrations respect to An-O. Much lower values were observed for both the investigated parameters during the second phase due to the dilution effect.
Figure 2. pH, VFA, alkalinity, TOC and COD variation throughout the entire test.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
3.1.3 TKN, Ammonia, Nitrite, Nitrate The concentrations of nitrogen compounds in all the reactors are presented in Figure 3. The behavior in TKN, ammonia, nitrites and nitrates was very similar among all semi aerobic columns and (with the exception of NO2- and NO3-) between the anaerobic ones. Nitrification process was clearly evident in all semi aerobic reactors. Fast decrease of TKN values in S-R and S-O columns started from 50th day: the nitrification process started and nitrites were observed in leachate for the first time. Consistently, ammonia concentrations showed a steeply decrease since 50th day. Similar behavior was observed for SALF reactors during the wet phase, characterized by much lower concentrations due to the dilution effect and higher removal kinetics. For the anaerobic columns, similar trend of TKN and ammonia occurred for AN-O and AN-R. Much higher removal rate occurred during the wet phase, since the flushing effect. Nitrite and nitrate were not observed in anaerobic columns, since nitrification cannot occur. They were detected in semi aerobic columns once the O2 become available for the nitrification process: in correspondence of day 50 and 120, the TOC consumption was almost completed in all S-O and S-R columns (Figure 2).
Figure 3. Nitrogen compounds behavior during the experiment.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
3.1.4 Ammonia and TOC output loads Ammonia and TOC output loads were calculated in order to analyze the effect of sequencing the two phases. The dilution effect during the second phase hid the real potential impact of the system during heavy rainfall: although the concentrations of TOC and ammonia during second phase were much lower respect to the first phase, the TOC output load was 2 times that of the first phase in all anaerobic columns (Figure 4) and the NH4+ output load was 2.5 and 2.8 times in AN-O and AN-R respectively (Figure 4). The positive synergetic effect of semi aerobic conditions and the innovative management was clear in semi aerobic columns, particular in S-R columns: the mean TOC and NH4+ loads were 1.08 and 0.6 times those of the first phase respectively. The higher the degradation rate, the lower the carbon and nitrogen contaminants release in leachate (Cossu et al., 2003).
Figure 4. Variation of the daily ammonia and TOC output load based on a weekly mean and cumulative ammonia and TOC load during the experiment.
3.1.5 Ammonia and TOC over Chlorides ratios NH4+/Cl- and TOC/Cl- ratios were calculated during the second phase to evaluate in which measure the biodegradation and the flushing effect contribute to the ammonia and TOC removal (Hrad and Huber-Humer, 2016). In particular, since Chloride can be only removed by washing out from waste (Fellner et al., 2009), a constant ratio over the time indicates the only effect of flushing in contaminants removal. The ratios behavior during the wet phase are reported in Figure 5. In anaerobic columns, although the ammonia and TOC concentrations decreased during the second phase, the ammonia and TOC over Chloride ratios increased: NH4+ and TOC decrease much slower than chloride, suggesting that ammonification of organic nitrogen and solubilization of complex TOC exceed the flushing effect. In all semi aerobic reactors both ratios presented similar trend: an initial positive trend, justified by ammonification and solubilization effect, was followed by a steeply decrease of NH4+/Cl- ratio between 120th-134th days and of TOC/Cl- ratio between 115th-120th days: biodegradation process occurred simultaneously to the flushing effect.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Figure 5. Variation of Ammonia/Chlorides and log(TOC/Chlorides) ratios during the second phase. 3.2 Biogas composition In all the semi-aerobic lysimeters biogas composition had the same trend in both dry and wet phase: after initial CO2 peaks after the filling of the columns, the exit gas composition was the same as air composition, demonstrating the effectiveness of the semi aerobic system. In anaerobic columns, no methane production was observed in An-O all along the entire test, since acidogenic phase occurred till the end of the second phase and biogas was mainly composed by CO2. In An-R methane production started around day 40, consistently with the pH increase. The methanogenic phase was maintained till the end of the second phase, achieving 17.2% of CH4. Gas percentages in biogas of SALF reactors and in the anaerobic ones throughout the entire test are presented in figure 6.
Figure 6. Gas composition in reactors throughout the entire test.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
3.3 Waste characterization The results of waste characterization at the beginning and at the end of the experiment are reported in Table 3. Considering the two waste typologies used to fill the columns, the potential impact related to the “O” waste was evident from the IR4 index, much higher respect to the “R” waste, but compensated by the lower quantity of TS in “O” waste. Moreover it was characterized by lower VS and TOC content. Considering the stabilization performances among the different columns, much better results occurred in semi aerobic reactors respect to the anaerobic ones. In particular, greater percentages decrease in VS (63%) and TOC (63%) parameters were observed in S-O columns, since the higher putrescible fraction, achieving better final values. Higher IR4 values in S-O columns, influenced by the different putrescible fraction content in “O” waste, were compensated by the low TS content. Considering the final heavy metals concentrations in the first waste strata, higher values were observed in semi aerobic reactors: anaerobic conditions, characterized by lower pH values, were favorable for Heavy metals mobilization (Sinan Bilgili et al., 2007), while semi aerobic reactors acted as final sink, since the fast stabilization of organic matter into humic substances and alkaline conditions (Qu et al., 2008). Table 3. Initial and final waste characterization. Input waste O
R
TS (kg) (%*)
6.40
VS (kg) (%*) TOC (kg) (%*)
End second phase An - O
An - R
S-O
S-R
11.73 4.72(26)
7.56(36)
3.78(41)
6.95(41)
4.41
5.82
3.24(26)
3.91(33)
1.62(63)
2.62(55)
2.56
2.82
1.65(36)
2.32(18)
0.96(63)
1.50(47)
IR4 (mgO2/gTS)
85.3
55
48.1
38.9
8.9
4.8
Cd (mg/kgTS)
1.8
4.5
0.8
0.7
0.7
0.6
Cr (mg/kgTS)
13.2
15
14
12
17
13.4
Cu (mg/kgTS)
343
735
100
113
211
314
Fe (mg/kgTS)
3208
4319
4065
4581
6363
7667
Mn (mg/kgTS)
72.3
98.2
117
210
245
305
Ni (mg/kgTS)
9
9
16
14
13
27
Pb (mg/kgTS)
36
41
40
51
101
141
Zn (mg/kgTS)
220
125
238
209
316
546
3.4 Carbon mass balance In order to evaluate the fate of carbon under anaerobic and semi aerobic conditions, the carbon mass transfer from solid to liquid and gas phase was evaluated by measuring the TOC in waste samples before filling and at the end of the test (TOCsolid), as well as in the leachate extracted throughout the entire test (TOCleachate). The carbon gasification was calculated as difference (TOCgas) (Figure 7). As confirmed by Shimaoka et al., (2000) and Cossu et al. (2003), the highest gasification occurred under semi aerobic conditions, in which from 50 to 60% of input carbon is supposed to be transferred to biogas, while only 0.6-1.5% was released throughout leachate emissions. The carbon gasification was limited in anaerobic columns, in which TOC was mainly accumulated in the lysimeters (64.2% in AN-O, 82.3% in AN-R) or transferred to the liquid phase (13.4% in AN-O, 9.2% in AN-R). As expected, higher gasification
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
was observed in “O” waste, since the high content in putrescible waste.
Figure 7. Percentages of carbon accumulation in solid phase (TOCsolid), and transfer into liquid (TOCleachate) and gas phase (TOCgas) during the experiment. 3.5 Sustainability achievement The stabilization performances of the semi aerobic system under the innovative management have been assessed and compared to the anaerobic columns, considering the capability in achieving the Final Storage Quality (FSQ) according to the regional legislation (D.G.R. 2461/14) (Table 4). According to the obtained results, all the semi aerobic columns achieved final values of BOD5, COD, BOD5/COD and ammonia below the FSQ limits at the end of the second phase, in particular with much better results in case of “R” waste. Higher Heavy Metals (HM) release occurred in anaerobic columns, exceeding the limit value fixed by the legislation for Iron and Manganese. According to Sinan Bilgili et al. (2007), acidic conditions improved the HM released in anaerobic columns, while fast stabilization of organic matter into humic substances and alkaline conditions in semi aerobic reactors increased sorptive capacity of waste mass and reduced HM mobilization (Qu et al., 2008). Both anaerobic and semi aerobic columns didn’t achieved IR4 standard limit, however much better stabilization conditions have been achieved under semi aerobic conditions, in particular with “R” waste, achieving final values very closed to the IR4 threshold.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Table 4. Final Storage Quality values purposed by Lombardia Region (Italy) (D.G.R. 2461/14), and final values achieved in the reactors at the end of the experiment.
Sample
Leachate
Solid
Parameter
FSQ values
AN-O
AN-R
S-O
SR
COD mg/L
1500
6800
4580
370
305
BOD5/COD
0.1
0.33
0.34
0.06
0.02
Ammonia (mg/L)
50
216
181
11
6
Cd (µg/L)
20
SUMMARY: Under tropical climate conditions, the rainfall seasonality represents often a limiting factor in proper functioning of semi-aerobic landfill system: lack of moisture during the dry season tends to slow down the biological activity, and the heavy rainfalls during the wet season result in high contaminants load in leachate (Rafizul and Alamgir, 2012; Tränkler et al., 2005). An innovative management strategy of semi aerobic landfills under tropical climates is proposed to improve the stabilization performances, by exploiting the dry season for simulating composting process, recirculating the leachate stored during the rainy season, and the wet season to obtain a flushing effect. The experimental study was carried out during six months, simulating the dry and wet season, on eight landfill simulating bioreactors: four semi-aerobic and four anaerobic. Two types of waste were used: low and high organic content waste. The synergetic effect of semi aerobic conditions and the sequencing first composting phase and second flushing phase was evaluated. Samples of waste were taken and analyzed at the begging and at the end of the experiment and leachate and gas composition were monitored during the entire experiment. The results show the effectiveness of the innovative management under semi aerobic conditions leading to much higher stabilization efficiency, compared to the anaerobic ones, regardless of the kind of waste, the system capability in acting as heavy metals sink and in realizing the Final Storage Quality (FSQ) during the one year simulated period.
1. INTRODUCTION In recent years, the need for the implementation of sustainable landfill concept is increasing more and more in both developing and industrialized countries, requiring the achievement of the FSQ within one generation time, in order to minimize both the short and long term impacts of the landfill system to the environment and to the public (Cossu, 2012; Hrad et al., 2013; Heimovaara, 2014). In industrialized countries, some measures toward the sustainable landfilling have been implemented: in northern Italy, first legislation on sustainable landfill introduced in 2014 Final Storage Quality (FSQ) limits (D.G.R. 2461/14). In developing countries, open dumps are still the most prevalent type of disposal facilities due to financial constraints and limited technical knowhow (Oman and Manandhar, 2011). Thus, implementation of more economic and technological landfill design, able to fulfill the sustainability, is crucial for both developing and industrialized countries. In this context, the semi-aerobic landfill system may represent a proper solution able to simultaneously fulfill the sustainability concept and the need for low cost disposal alternative (Hirata et al., 2012). The semi aerobic landfill, developed at the Proceedings Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium/ 2 - 6 October 2017 S. Margherita di Pula, Cagliari, Italy / © 2017 by CISA Publisher, Italy
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Fukuoka University since more than 20 years ago, aims to reproduce an aerobic environment within the waste mass with a proper engineering designs, in which the ambient air flows into the waste body naturally through the leachate collection pipes, moved by the temperature gradient between inside and outside the landfill (Theng et al., 2005). Many researches have already been managed with the aim of evaluating the advantages and mechanism of the semi aerobic landfill (SALF), demonstrating its capability to achieve the aerobic process performance without any specific technical equipment: § Improvement of carbon degradation rate, higher than in an anaerobic system (Ahmadifar et al., 2016; Cossu et al., 2003; He et al., 2012); § Enhancement of nitrogen removal (Cossu et al., 2003; He et al., 2011, 2012; Shao et al., 2008; Yang et al., 2012; Zeng et al., 2006); § Reduction of methane generation (Ahmadifar et al., 2016; Cossu et al., 2003; Enough and Analysis, 2006; HUANG et al., 2008; Sutthasil et al., 2014; Wu et al., 2017; Yang et al., 2012; IPPC, 2006); § Increased carbon gasification (Shimaoka et al., 2000). Despite that, the rainfall seasonality of tropical climates represents often a limiting factor in SALF proper functioning: lack of moisture during the dry season tends to slow down the biological activity, and the heavy rainfalls during the wet season result in high contaminants load in leachate (Rafizul and Alamgir, 2012; Tränkler et al., 2005). According to the Köpper-Geiger climate classification, the tropical Aw climate (Tropical wet and dry Savanna climate) is the second most diffused climate worldwide (11.5%), characterized by a dry season, 5-6 month long, during which low or no rainfalls occur, while precipitation between 800-1600mm are concentrated during the wet season (Chen and Chen, 2013; Kottek et al., 2006; Peel et al., 2007; Essenwanger, 2001). The focus of this research was to develop an innovative management of semi aerobic landfill under the specific tropical climates in order to improve the SALF stabilization performances, by exploiting the climate seasonality: § During the dry season, the waste tipping in thin layers and the recirculation of leachate, stored during the rainy season, ensure oxygen intrusion and proper moisture content for realizing the composting process. § During the wet season, the leachate generated by the flushing effect, infiltrates trough the composted layer below, in which the humic and fulvic substances may provide a natural insitu physical-chemical treatment by absorbing heavy metals (Qu et al., 2008) and the formed microbial population during the dry season may enhance the in-situ biological treatment. The synergetic effect of semi aerobic conditions together with the innovative management has been evaluated on two different waste typologies, characterized by different organic fraction content: residual waste after separate collection, characterized by low organic content (“R” waste), and rich in putrescible fraction waste (“O” waste). The waste stabilization performances of the innovative management under semi aerobic conditions has been analyzed comparing the effectiveness with the two waste typologies and with the anaerobic system, considering the ammonia removal, the gasification enhancement, the methane generation reduction, the FSQ limits achievement and the capability in acting as final sink for heavy metals. Moreover, the role of the first composting phase on the improvement of leachate quality during the wet phase has been investigated, although further researches are required to clarified mechanisms and real contribute.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
2. MATERIALS AND METHODS 2.1 Waste samples The waste used for the experiment was provided by a public waste management company operating in northern Italy. The urban waste was taken after a mechanical treatment of shredding and a 6 cm sieving, without any biological pretreatment. Starting from the initial waste sample, two type of waste were used to fill the columns: unchanged residual waste (“R” waste) resulting in 4% kitchen residues content, and high in organic fraction waste (“O” waste) obtained modifying initial waste, increasing the putrescible fraction up to 50%, taken kitchen waste from municipal putrescible separate collection. The final waste characterization is provided in Table 1. The “R” (“Residual”) waste simulates the typical waste composition in industrialized countries after the separate collection; the “O” (“Organic”) waste was reproduced in order to simulate the typical waste composition in low-medium income countries (The World Bank, 2012).
Table 1. Waste fractions of “Residual” and ”Organic” waste. Waste categories Kitchen residues Green waste and wooden materials Paper and paperboard Textiles Plastics Metals Glass Inert Under-sieve
"Residual" % 4 6 20 6 24 4 10 10 16
"Organic" % 50 3 11 3 13 2 5 5 8
2.2 Equipment The experiment was carried out using eight Plexiglass® cylindrical lysimeters (1m height, inner diameter of 24 cm) to simulate anaerobic and semi aerobic conditions. At the bottom of each column 10cm thick gravel layer (Ø 20–30 mm) was placed before filling to facilitate the draining out of leachate. In order to simulate the passive aeration in semi aerobic lysimeters, a slotted aeration pipe (4 cm diameter) was placed at the bottom of semi aerobic columns, open to the air, and the top of the semi-aerobic columns was left completely open. 1 cm diameter plastic pipe were inserted in the waste body to measure gas quality. Anaerobic columns were completely sealed. Two valves at upper end of each anaerobic reactor were used for gas extraction by Tedlar sampling bags and water irrigation. Leachate extraction was performed trough a collection port located at the bottom of each reactor and temperature monitoring was ensured using the Thermo Systems TS100 temperature probes installed inside each reactor (Figure 1). Both Anaerobic and Semi aerobic columns were wrapped by thermo insulated material and located in a container equipped to control the environmental temperature.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Figure 1. Semi aerobic (a) and anaerobic (b) reactors set up.
2.3 Methodology The experimental study was carried out during six months, simulating the dry (0-96th day) and wet season (97th-192th day), on eight landfill simulating bioreactors, operated in duplicate under four different process conditions (Table 2): § An-R reactors: anaerobic conditions filled with “Residual” waste; § An-O reactors: anaerobic conditions filled with “Organic” waste; § S-R reactors: semi-aerobic conditions filled with “Residual” waste; § S-O reactors: semi-aerobic conditions filled with “Organic” waste. In semi aerobic columns the aeration was given by natural heat convection according to semiaerobic concept. At the beginning of both wet and dry phases, 9 kg of “Residual” and “Organic” waste were placed in AN-R, S-R columns and in An-O, S-O respectively, achieving a density of about 0.5 kg/L. Table 2. Operational conditions and waste type applied for the different experimental reactors. Waste typologies
Anaerobic
Semi aerobic
“O” waste (50% putrescible content)
An-O
S-O
“R” waste (4% putrescible content)
An-R
S-R
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
The experiment follows two main stages: during the first phase only first half of the bioreactors was filled, the daily water irrigation was initially defined according with PAF model (Cossu et al., 2003) and then adjust from 0.22 L/d to 0.05 L/d to guarantee composting process and thus a good final stabilization at the end of the third month. In the second phase the second halves of the columns were filled with the same amount and kind of waste of the first stratum and daily irrigation of 0.9 L/d simulated the weather condition of wet tropical season, considering a total precipitation of 1400 mm. Daily leachate extraction was performed throughout the entire test and samples were collected for analysis. Environmental temperatures were maintained with a cyclic half day temperature between 18-30°C. Both temperatures and precipitations were selected within the ranges suggested by the Whittaker biome classification for Savanna climate (Whittaker, 1975; Kottek et al., 2006). The stabilization performances of the semi aerobic system under the innovative management have been assessed considering the capability in achieving the Final Storage Quality (FSQ) according to the Italian regional legislation (D.G.R. 2461/14). Samples of waste were taken and analyzed for TOC, IR4, TS, VS and heavy metals, at the begging and at the end of the experiment. Leachate was monitored in terms of pH, VFA, alkalinity, COD, TOC, BOD5, TKN, N-NH4+, N-NO3-, N-NO2- and Cl- and biogas concentrations were analyzed in terms of CO2, CH4 and O2. The results are reported as mean values for each type of columns management.
3. RESULTS AND DISCUSSION 3.1 Leachate quality 3.1.1 pH, Volatile fatty acids, Alkalinity. The behavior of pH, Volatile fatty acids (VFA) and alkalinity, monitored throughout the entire test, are reported in figure 2. Excepted in S-O columns during the first days after fills, the pH values were maintained above 7 in all SALF reactors during the experiment. In An-O columns, acidogenic phase occurred though out the entire test and, maintaining pH values below 6, methanogenesis was inhibited (Cossu et al., 2016). Differently, in An-R reactors, a positive pH trend was observed after day 40, once the methanogenic phase started and then maintained during the second phase, with pH values above 6: from that time CH4 has been observed in AnRs biogas (Figure 6). Consistently with pH variation, VFA steeply decreased in all semi aerobic columns at the beginning of both phases, showing a very similar behavior both in case of “O” and “R” waste. In anaerobic reactors clear difference was observed since 40th day till the end of the dry phase according to the waste typology: in AN-O, VFA values were almost constant around 3400 mgCH3COOH/L, in AN-R, the contribute of methanogens led to a further decrease in fatty acids up to 2000 mgCH3COOH/L at the end of the dry phase. During the second phase no big differences in decreasing trend were detectable between the two columns: the flushing effect dominated the declining behavior of VFA. According to the pH and VFA behavior, alkalinity increased in An-R with the pH at the beginning of methanogens activity. During the wet phase it decreased due to the flushing effect. In all other columns the alkalinity values decreased during the entire test. For all the parameters much lower values were observed in wet phase due to the dilution effect. 3.1.2 TOC and COD Similar behavior to VFA was observed for TOC and COD parameters, reported in Figure 2. High removal performances were evident in all semi aerobic columns, with a steeply decrease
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
in TOC and COD values regardless of the waste typology, justified by the higher kinetics ensured under aerobic conditions (Ritzkowski et al., 2007). In anaerobic columns, the waste typology strongly influence stabilization performances: during the first phase, better results in AN-R were justified by the contribution of methanogens activity starting from 40th day; during the second phase the flushing effect dominated in anaerobic columns, according to the positive trend of TOC/Chloride ratio (Figure 4), and the low organic content in AN-R ensured lower TOC concentrations respect to An-O. Much lower values were observed for both the investigated parameters during the second phase due to the dilution effect.
Figure 2. pH, VFA, alkalinity, TOC and COD variation throughout the entire test.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
3.1.3 TKN, Ammonia, Nitrite, Nitrate The concentrations of nitrogen compounds in all the reactors are presented in Figure 3. The behavior in TKN, ammonia, nitrites and nitrates was very similar among all semi aerobic columns and (with the exception of NO2- and NO3-) between the anaerobic ones. Nitrification process was clearly evident in all semi aerobic reactors. Fast decrease of TKN values in S-R and S-O columns started from 50th day: the nitrification process started and nitrites were observed in leachate for the first time. Consistently, ammonia concentrations showed a steeply decrease since 50th day. Similar behavior was observed for SALF reactors during the wet phase, characterized by much lower concentrations due to the dilution effect and higher removal kinetics. For the anaerobic columns, similar trend of TKN and ammonia occurred for AN-O and AN-R. Much higher removal rate occurred during the wet phase, since the flushing effect. Nitrite and nitrate were not observed in anaerobic columns, since nitrification cannot occur. They were detected in semi aerobic columns once the O2 become available for the nitrification process: in correspondence of day 50 and 120, the TOC consumption was almost completed in all S-O and S-R columns (Figure 2).
Figure 3. Nitrogen compounds behavior during the experiment.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
3.1.4 Ammonia and TOC output loads Ammonia and TOC output loads were calculated in order to analyze the effect of sequencing the two phases. The dilution effect during the second phase hid the real potential impact of the system during heavy rainfall: although the concentrations of TOC and ammonia during second phase were much lower respect to the first phase, the TOC output load was 2 times that of the first phase in all anaerobic columns (Figure 4) and the NH4+ output load was 2.5 and 2.8 times in AN-O and AN-R respectively (Figure 4). The positive synergetic effect of semi aerobic conditions and the innovative management was clear in semi aerobic columns, particular in S-R columns: the mean TOC and NH4+ loads were 1.08 and 0.6 times those of the first phase respectively. The higher the degradation rate, the lower the carbon and nitrogen contaminants release in leachate (Cossu et al., 2003).
Figure 4. Variation of the daily ammonia and TOC output load based on a weekly mean and cumulative ammonia and TOC load during the experiment.
3.1.5 Ammonia and TOC over Chlorides ratios NH4+/Cl- and TOC/Cl- ratios were calculated during the second phase to evaluate in which measure the biodegradation and the flushing effect contribute to the ammonia and TOC removal (Hrad and Huber-Humer, 2016). In particular, since Chloride can be only removed by washing out from waste (Fellner et al., 2009), a constant ratio over the time indicates the only effect of flushing in contaminants removal. The ratios behavior during the wet phase are reported in Figure 5. In anaerobic columns, although the ammonia and TOC concentrations decreased during the second phase, the ammonia and TOC over Chloride ratios increased: NH4+ and TOC decrease much slower than chloride, suggesting that ammonification of organic nitrogen and solubilization of complex TOC exceed the flushing effect. In all semi aerobic reactors both ratios presented similar trend: an initial positive trend, justified by ammonification and solubilization effect, was followed by a steeply decrease of NH4+/Cl- ratio between 120th-134th days and of TOC/Cl- ratio between 115th-120th days: biodegradation process occurred simultaneously to the flushing effect.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Figure 5. Variation of Ammonia/Chlorides and log(TOC/Chlorides) ratios during the second phase. 3.2 Biogas composition In all the semi-aerobic lysimeters biogas composition had the same trend in both dry and wet phase: after initial CO2 peaks after the filling of the columns, the exit gas composition was the same as air composition, demonstrating the effectiveness of the semi aerobic system. In anaerobic columns, no methane production was observed in An-O all along the entire test, since acidogenic phase occurred till the end of the second phase and biogas was mainly composed by CO2. In An-R methane production started around day 40, consistently with the pH increase. The methanogenic phase was maintained till the end of the second phase, achieving 17.2% of CH4. Gas percentages in biogas of SALF reactors and in the anaerobic ones throughout the entire test are presented in figure 6.
Figure 6. Gas composition in reactors throughout the entire test.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
3.3 Waste characterization The results of waste characterization at the beginning and at the end of the experiment are reported in Table 3. Considering the two waste typologies used to fill the columns, the potential impact related to the “O” waste was evident from the IR4 index, much higher respect to the “R” waste, but compensated by the lower quantity of TS in “O” waste. Moreover it was characterized by lower VS and TOC content. Considering the stabilization performances among the different columns, much better results occurred in semi aerobic reactors respect to the anaerobic ones. In particular, greater percentages decrease in VS (63%) and TOC (63%) parameters were observed in S-O columns, since the higher putrescible fraction, achieving better final values. Higher IR4 values in S-O columns, influenced by the different putrescible fraction content in “O” waste, were compensated by the low TS content. Considering the final heavy metals concentrations in the first waste strata, higher values were observed in semi aerobic reactors: anaerobic conditions, characterized by lower pH values, were favorable for Heavy metals mobilization (Sinan Bilgili et al., 2007), while semi aerobic reactors acted as final sink, since the fast stabilization of organic matter into humic substances and alkaline conditions (Qu et al., 2008). Table 3. Initial and final waste characterization. Input waste O
R
TS (kg) (%*)
6.40
VS (kg) (%*) TOC (kg) (%*)
End second phase An - O
An - R
S-O
S-R
11.73 4.72(26)
7.56(36)
3.78(41)
6.95(41)
4.41
5.82
3.24(26)
3.91(33)
1.62(63)
2.62(55)
2.56
2.82
1.65(36)
2.32(18)
0.96(63)
1.50(47)
IR4 (mgO2/gTS)
85.3
55
48.1
38.9
8.9
4.8
Cd (mg/kgTS)
1.8
4.5
0.8
0.7
0.7
0.6
Cr (mg/kgTS)
13.2
15
14
12
17
13.4
Cu (mg/kgTS)
343
735
100
113
211
314
Fe (mg/kgTS)
3208
4319
4065
4581
6363
7667
Mn (mg/kgTS)
72.3
98.2
117
210
245
305
Ni (mg/kgTS)
9
9
16
14
13
27
Pb (mg/kgTS)
36
41
40
51
101
141
Zn (mg/kgTS)
220
125
238
209
316
546
3.4 Carbon mass balance In order to evaluate the fate of carbon under anaerobic and semi aerobic conditions, the carbon mass transfer from solid to liquid and gas phase was evaluated by measuring the TOC in waste samples before filling and at the end of the test (TOCsolid), as well as in the leachate extracted throughout the entire test (TOCleachate). The carbon gasification was calculated as difference (TOCgas) (Figure 7). As confirmed by Shimaoka et al., (2000) and Cossu et al. (2003), the highest gasification occurred under semi aerobic conditions, in which from 50 to 60% of input carbon is supposed to be transferred to biogas, while only 0.6-1.5% was released throughout leachate emissions. The carbon gasification was limited in anaerobic columns, in which TOC was mainly accumulated in the lysimeters (64.2% in AN-O, 82.3% in AN-R) or transferred to the liquid phase (13.4% in AN-O, 9.2% in AN-R). As expected, higher gasification
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
was observed in “O” waste, since the high content in putrescible waste.
Figure 7. Percentages of carbon accumulation in solid phase (TOCsolid), and transfer into liquid (TOCleachate) and gas phase (TOCgas) during the experiment. 3.5 Sustainability achievement The stabilization performances of the semi aerobic system under the innovative management have been assessed and compared to the anaerobic columns, considering the capability in achieving the Final Storage Quality (FSQ) according to the regional legislation (D.G.R. 2461/14) (Table 4). According to the obtained results, all the semi aerobic columns achieved final values of BOD5, COD, BOD5/COD and ammonia below the FSQ limits at the end of the second phase, in particular with much better results in case of “R” waste. Higher Heavy Metals (HM) release occurred in anaerobic columns, exceeding the limit value fixed by the legislation for Iron and Manganese. According to Sinan Bilgili et al. (2007), acidic conditions improved the HM released in anaerobic columns, while fast stabilization of organic matter into humic substances and alkaline conditions in semi aerobic reactors increased sorptive capacity of waste mass and reduced HM mobilization (Qu et al., 2008). Both anaerobic and semi aerobic columns didn’t achieved IR4 standard limit, however much better stabilization conditions have been achieved under semi aerobic conditions, in particular with “R” waste, achieving final values very closed to the IR4 threshold.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Table 4. Final Storage Quality values purposed by Lombardia Region (Italy) (D.G.R. 2461/14), and final values achieved in the reactors at the end of the experiment.
Sample
Leachate
Solid
Parameter
FSQ values
AN-O
AN-R
S-O
SR
COD mg/L
1500
6800
4580
370
305
BOD5/COD
0.1
0.33
0.34
0.06
0.02
Ammonia (mg/L)
50
216
181
11
6
Cd (µg/L)
20
