Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment

Waste Management xxx (2018) xxx–xxx

Contents lists available at ScienceDirect

Waste Management journal homepage: www.elsevier.com/locate/wasman

Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment Luca Morello a, Roberto Raga a, Paolo Sgarbossa b, Egle Rosson b,⇑, Raffaello Cossu a a b

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

a r t i c l e

i n f o

Article history: Received 27 June 2017 Revised 16 January 2018 Accepted 17 January 2018 Available online xxxx Keywords: Waste characterization Waste compounds speciation Landfill long-term emissions Aftercare completion Final Storage Quality (FSQ) Sustainable landfilling

a b s t r a c t The storage capacity and the potentially residual emissions of a stabilized waste coming from a landfill simulation experiment were evaluated. The evolution in time of the potential emissions and the mobility of some selected elements or compounds were determined, comparing the results of the stabilized waste samples with the values detected in the related fresh waste samples. Analyses were conducted for the total bulk waste and also for each identified category (under-sieve, kitchen residues, green and wooden materials, plastics, cellulosic material and textiles) to highlight the contribution of the different waste fractions in the total emission potential. The waste characterization was performed through analyses on solids and on leaching test eluates; the chemical speciation of carbon, nitrogen, chlorine and sulfur together with the partitioning of heavy metals through a SCE procedure were carried out. Results showed that the under-sieve is the most environmentally relevant fraction, hosting a consistent part of mobile compounds in fresh waste (40.7% of carbon, 44.0% of nitrogen, 47.6% of chloride and 40.0% of sulfur) and the greater part of potentially residual emissions in stabilized waste (88.4% of carbon, 90.9% of nitrogen, 98.4% of chloride and 91.1% of sulfur). Landfilled Municipal Solid Waste (MSW) proved to be an effective sink, finally storing more than 55% of carbon, 53% of nitrogen, 33% of sulfur and 90% of heavy metals (HM) which were initially present in fresh waste samples. A general decrease in leachable fractions from fresh to stabilized waste was observed for each category. Tests showed that solid waste is not a good sink for chlorine, whose residual non-mobile fraction amounts to 12.3% only. Ó 2018 Published by Elsevier Ltd.

1. Introduction Landfills should be designed and managed in order to reach the Final Storage Quality (FSQ) condition within the time span of one generation, enhancing as much as possible the waste biochemical stabilization as consequence (Laner et al., 2012). This need is related to the potential threats posed to the environment in short and long-term by mobile fractions in waste and to the significant role that landfills can play as a sink for closing material cycles (Cossu, 2016; Pivnenko and Astrup, 2016). The potential of different landfill technologies to enhance landfill stabilization as well as the long term storage of specific compounds and elements have been hot topics for a few years now and research projects are currently in progress worldwide (Townsend et al., 2015; Cossu et al., 2016; Christensen et al., 2011; Bolyard and Reinhart, 2016). ⇑ Corresponding author.

A landfill stores significant quantities of substances that can be present as non-mobile or harmless mobile compounds, which are not an issue for the environment or the healthcare safety, or in mobile forms. These latter may include several organic substances, soluble salts, catabolites and ions which can potentially become contaminants in the case of uncontrolled emissions into the environment, especially during a long-term aftercare (Christensen and Kjeldsen, 1989). Life Cycle Assessment (LCA) studies, based on literature data, estimated the magnitude of the storage potential of different technologies, referring to different substances, showing that up to 58% of carbon, 87% of nitrogen, 49% of chlorine and 99% of HM were expected to remain stored in a MSW landfill in a 100-year time simulation (Manfredi et al., 2009). Similar results were obtained by Qu et al. (2008), in terms of HM limited mobility in a full-scale bioreactor landfill. In addition to these substances, commonly found in waste, some refractory organic by-products (Bolyard and Reinhart, 2016) and some new chemicals (such as pharmaceuticals, flame retardants, stabilizers, fillers, inks, antioxi-

E-mail address: [email protected] (E. Rosson). https://doi.org/10.1016/j.wasman.2018.01.026 0956-053X/Ó 2018 Published by Elsevier Ltd.

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

2

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

dants, personal care products and others) (Pivnenko and Astrup, 2016) are detected with increasing frequency in leachate and solid waste and can produce long-term emissions. Moreover, some emissions derived from secondary substances, which are contained in these products, have the tendency to be reactive towards dissolved organic matter (as humic and fulvic acids). This implies they could be transported as DOC complexes and therefore cover much larger distances than expected (Van der Sloot et al., 2017). The stored carbon can be divided into organic and inorganic fractions, the first one can be further divided in biodegradable and hardly biodegradable compounds (such as cellulose, hemicellulose, lignin and synthetic organic polymers), which can contribute to the carbon storage. Investigations on carbon storage potential of landfilled waste (De la Cruz et al., 2013) resulted in the estimation that 35% to 95% of the biogenic carbon is likely to go into long-term storage. The lab-scale test conducted by Bolyard and Reinhart (2016), showed that 45–50% of carbon remained stored in the waste mass, while less than 5% was emitted by leaching, even after a flushing treatment reaching a Liquid-Solid ratio (L/S) of 10 L/kgTS. According to Brandstätter et al. (2015a), more than 10% of the biodegradable mobile organic carbon initially present in mobile forms in the samples was converted into nonmobile forms, probably due to the transformation into humic substances. Humic and fulvic acids are refractory by-products, result of the conversion of biomass during the degradation process, and they are only partially leachable, constituting a relevant contribution to long-term COD emissions in leachate. In Bolyard doctoral dissertation more studies about humic acid extraction from solid waste are discussed, considering the flushing of these substances from three bioreactors at different L/S ratios. The highest mass extraction was achieved in the bioreactor characterized by chemical oxidation and aeration at L/S of 15, in comparison to the other two bioreactors in which only flushing and only chemical oxidation were considered, respectively (Bolyard, 2016). The nitrogen mass balance shows that the majority of the initially present nitrogen remains stored into the waste mass, bound to complex non-degradable polymers and organic matter, whatever treatment is applied (Brandstätter et al., 2015b). The lab scale test conducted by Bolyard and Reinhart (2016) showed that 73– 76% of nitrogen remained trapped in the waste mass, despite aeration and flushing reaching a L/S ratio of 10 L/kgTS. Mobile nitrogen is mainly composed by soluble ammonia ion, derived from ammonification, which is considered the most persistent contaminant in landfill leachate. The nitrification–denitrification process is the easiest method to decrease the long-term emissions of NH+4 (Berge et al., 2005; Ritzkowski et al., 2016), unavoidable step for reaching FSQ conditions in shorter time. The chlorine emissions of landfills are mainly constituted by chloride, being persistent also in long-term emissions. Chloride solubility is generally high and poorly influenced by pH; consequently, the reactor biochemical conditions only slightly affect the chloride leaching (Fellner et al., 2009; Morello et al., 2016). The sulfur compounds chemical conditions are influenced by the environment redox properties and by the pH: if a reducing environment is established, sulfides prevail (in both solid samples and leachate). On the contrary, in presence of oxygen, sulfates are the most common detectable fractions. The heavy metals concentration in landfilled MSW has not a great environmental interest due to their low mobility (Qu et al., 2008). However, HM presence in solid state is significant and their partitioning is strictly dependent on the redox conditions and on the pH: low pH increases the amount of HM in ionic mobile form, while reducing conditions enhance their precipitation. According to Kjeldsen et al. (2002) only 0.02% of the initial heavy metals content in the landfill was removed by leachate within 30 years of anaerobic conditions, while less than 1% considering aerobic pro-

cesses. HM partitioning in solid waste samples can be a useful tool for understanding their bonding state within the MSW matrix and how they could be potentially released in the surrounding environment under particular conditions. There are no standard procedures for performing this analysis in a solid status because of the difficulties to identify precisely the HM behavior, dependent on several physico-chemical parameters. This paper aims to identify the waste fractions and the elements or compounds most affecting the overall emission characteristics of a fresh MSW sample and a stabilized waste sample from a landfill simulation experiment. The evolution in time of potential emissions and the mobility of the selected elements or compounds are estimated through the chemical speciation of carbon, nitrogen, chlorine and sulfur and through the partitioning of HM. The final goal is using the collected data on solid waste samples to highlight and quantify which compounds are expected to be reduced or leached away, which can be mobilized even in long term phases and which can be definitively stored in the landfill body.

2. Materials and methods 2.1. Materials This study is based on the raw data obtained through a chemical characterization of a fresh MSW solid sample taken prior to being landfilled (Fresh F sample) and a stabilized material coming from a long-lasting lysimeter reactor test (Stabilized S sample). The waste of these two samples was taken from the same Mechanical Biological Treatment (MBT) plant, located in northern Italy, before aerobic stabilization. F sample was immediately analyzed after collection, highlighting the initial waste conditions. S sample was characterized by material which underwent a long-lasting aerated landfilling simulation in a lysimeter reactor; it was analyzed after the treatment, representing the final waste conditions. The Fresh sample was collected ensuring that its chemical features were equivalent to the ones of the initial MSW loaded into the reactor. The lysimeter reactor was constituted by a column of a square base with a total volume of 2 m3, built for the simulation of a semi-aerobic landfill: letting the natural air circulation driven by the temperature gradient between the waste body and the external environmental temperature (Fig. 1). The reactor was equipped with a pipe (diameter 300 mm) in the bottom gravel layer for allowing the natural air flow, with a thermal insulation system and with a valve system for sampling leachate as well as monitoring gas emissions. This kind of equipment proved to be useful also for simulating long aftercare periods in which the landfill impermeable top cover prevents rain water infiltration and the oxygen can partially enter through the soil interstitial pores. The landfill simulation test consisted of a sequence of phases at the end of which the properties of the material inside the reactor and of the leachate emitted will ensure no harm for the environment (FSQ concept) (Vettorazzi, 2005; Piovesan, 2007; Morello, 2013). The test lasted for approximately 10 years, subdivided in a first anaerobic phase (180 days with injection of 10 L of water per week to reach a L/S ratio of 0.82 L/kgTS) (Vettorazzi, 2005) and a long semi-aerobic aftercare period in which no water was injected. Flushing tests were performed after 1000 days (60 days with the addition of 28 L of water per week until a L/S ratio of 1.14 L/kgTS) (Piovesan, 2007) and at the end of the whole test (180 days with a total amount of water added of 750 L to finally reach a L/S ratio of 2.76 L/kgTS) (Morello, 2013). During the first phase, the quantity of water injected was calculated through the approximation of the average precipitation in northern Italy on the specific surface of the reactor. During the long aftercare phases, no water was added for simulating a superficial impermeable layer

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

3

Fig. 1. Scheme and picture of the lysimeter equipment with the thermal insulation cover. TR: temperature of the reactor; TA: Ambient temperature; nT: temperature probe; nG: gas monitoring valve.

placed over the landfill. Finally, the flushing tests water injection was decided according to the leachate quality, with the purpose of maximizing the compounds extraction. The waste sample taken from the lysimeter at the end of the treatment (S sample) was considered biochemically stable and with a low content of mobile potentially polluting compounds. In order to validate this hypothesis, lab test results were compared with some proposed FSQ values (D.G.R. 2461/14; Laner et al., 2012). 2.2. Samples preparation A fractional classification was performed on the two waste samples collected (F and S samples), as first operation. Three sieves in series were used (100 mm, 60 mm and 20 mm) in order to increase the classification precision; this allowed to identify four different size fractions. After that, each size fraction was divided into some main fractional categories by means of a manual sorting procedure. The waste macro-categories chosen were under-sieve <20 mm, kitchen residues, green waste and wooden materials, plastics, textiles, cellulosic material, inert, glass and metals, according with the subdivision proposed by other tests (Cossu et al., 2016; Morello et al., 2016; Edjabou et al., 2015). F and S sub-samples were prepared by mixing together the same fractional categories coming from all the sieve dimensions (analyzed separately during the fractional classification): UND for under-sieve, KIT for kitchen residues, GRE for green waste and wooden materials, PLA for plastics, TEX for textiles and CEL for cellulosic material. Inert (i.e. ceramic, stones), glass and metals (i.e.

cans, bottle caps) were not directly examined; their composition was indirectly evaluated from the total bulk waste analysis. The impurities contained in the manual sorted samples were considered representative of the real separation obtained by a conventional waste mechanical sorting equipment. Fresh and Stabilized samples, made by the waste categories (UND, KIT, GRE, PLA, TEX and CEL), were all mixed, shredded (10 mm) by means of a soil mill and prepared with three different procedures, according to the chemical analyses to be performed (Table 1):  Raw solid samples were used for standard soil analyses (TOC, TC, TIC, TS, TVS, TKN RI4, RI7, N-NH+4).  Pulverized solid samples were used for some particular analyses requiring low quantities of very homogeneous materials (organic polymers, TN, total chlorine, total sulfur, total HM). Pulverization consisted in milling to obtain a dust-mud, composed by particles smaller than 1 mm size. A SCE (Sequential Chemical Extraction) was applied for analyzing HM presence only in the solid UND sample, after the above described pulverization procedure (bound HM).  Eluates from international standard leaching tests were used for the evaluation of the liquid emission potential (TOC, TC, TIC, VFA, Biochemical Oxygen Demand in 5 days BOD5, Chemical Oxygen Demand COD, humic and fulvic acids, TKN, N-NH+4, NO2 , NO3 , Cl , sulfate, sulfide, total HM). The leaching test performed respected the standard international procedure UNI EN 12,457-2 (shredding at 4 mm, dilution with distilled water until reaching a L/S ratio equal to 10 L/kgTS, mixing for 24 h and filtering to 0.45 mm).

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

4

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

Table 1 List of parameters and analytical methods considered for each solid and eluate sample obtained from the different waste categories (UND: under-sieve, KIT: kitchen residues, GRE: green waste and wooden materials, PLA: plastics, TEX: textiles, CEL: cellulosic material). For the evaluation of organic polymers, humic acids, total chlorine and HM partitioning the reference method is reported being not an international standard. Contaminants

Sample

Parameters

Analytical method

Carbon compounds

Solid

TC, TOC, TIC TS, TVS RI4, RI7 Organic polymers TC, TOC TIC VFA BOD5 COD Humic and fulvic acids

UNI-EN 13137 (TOC-VCSN Shimadzu Analyzer) IRSA-CNR Q 64/84, vol. 2, n. 2 Respiration Index Standard Wang et al., 2015, Yang et al., 2006 and Li et al., 2004 IRSA-CNR 29/2003, vol. 2, n. 5040 (TOC-VCSN Shimadzu Analyzer) Standard method n. 5560 C, 1989 IRSA-CNR 29/2003, vol. 2, n. 5120 A, B, B2 IRSA-CNR 29/2003, vol. 2, n. 5130 Baddi et al., 2004

Eluate

TN TKN NH+4 TKN NH+4 NO2 NO3

CHN/S elemental analysis IRSA-CNR Q 64/85, vol. 3, n. 6 mod. IRSA-CNR Q 64/85, vol. 3, n. 7 mod. IRSA-CNR 29/2003, vol. 2, n. 5030 IRSA-CNR 29/2003, vol. 2, n. 4030C IRSA-CNR 29/2003, vol. 2, n. 4050 IRSA-CNR 29/2003, vol. 2, n. 4040 A1

Chlorine compounds

Solid Eluate

Total chlorine Chloride

Okada et al., 2007 IRSA-CNR 29/2003, vol. 2, n. 4090B

Sulfur compounds

Solid Eluate

Total sulfur Sulfate Sulfide

CHN/S elemental analysis IRSA-CNR 29/2003, vol. 2, n. 4140B IRSA-CNR 29/2003, vol. 2, n. 4160

Heavy metals

Solid

Total metals Bound metals Total metals

EPA 1996, n. 3050B (ICP analysis) Krishnamurti et al., 2002 (ICP analysis) IRSA-CNR 29/2003, vol. 1, n. 3010 A mod. (ICP analysis)

Eluate

Nitrogen compounds

Solid

Eluate

Legend: TC – Total Carbon, TOC – Total Organic Carbon, TIC – Total Inorganic Carbon, RIx – Respiration Index for x days, VFA – Volatile fatty acids, BOD5 – Biological or Biochemical Oxygen Demand in 5 days, COD – Chemical Oxygen Demand, TN – Total Nitrogen, TKN – Total Kjeldahl Nitrogen.

2.3. Methodology The chemical analyses took into account the most relevant compounds of carbon, nitrogen, chlorine, sulfur and heavy metals, present in a common MSW landfill. These analyses were conducted on solid samples (raw or pulverized) and on eluates extracted through the leaching test, representing quantitatively the total mobile fractions of each category (Table 1). Leachate and gaseous emissions from the reactor tests were not considered in the study; they are accounted as mass loss in the discussion. All the collected chemical data are elaborated in order to estimate which compounds have been emitted, which are expected to remain mobile in long-term phases and which can be definitively stored in the landfill body, with respect to their initial presence into the total bulk of the fresh waste. The Total Carbon (TC) content was evaluated through TC analysis; it is composed by Inorganic Carbon (IC) plus Total Organic Carbon (TOC). The amount of organic polymers was determined, representing the part of hardly degradable TOC that contributes to the carbon sink. The leachable fractions of carbon compounds were extracted through the leaching tests and analyzed (TC, TOC and IC). In addition, Volatile Fatty Acids (VFA) were measured to quantify the readily degradable compounds and humic-fulvic acids were monitored to evaluate the long-term emission potential. Finally, the Respiration Index (RI4) and the BOD5/COD ratio were measured to understand the extent of biochemical stabilization of the samples (Cossu et al, 2012). The Total Nitrogen (TN) in solid samples was determined through CNH/S analysis. The speciation of TN considers the Total Kjeldahl Nitrogen (TKN), including ammonia ion (NH+4) plus organic nitrogen, and the oxidized forms, as nitrates (NO3 ) and nitrites (NO2 ). All these compounds are soluble so their quantification was done also in the eluate. The total chlorine was evaluated in the solid samples and in the related leachable fractions, whose only components are chlorides.

The total sulfur was detected through CNH/S on solid samples. In its speciation, sulfate and sulfide were considered as mobile fractions. The total heavy metals (Cr, Cu, Fe, Mn, Ni, Pb, Zn) content was determined both on solid samples and on eluates to estimate the potential emissions. HM in eluates are mainly in ionic forms or bound to DOC (Dissolved Organic Carbon). On the contrary, in solid waste samples, HM can be found in several forms, sometimes nonmobile and sometimes mobile depending on the boundary environmental conditions. For this reason, a SCE procedure was applied on the pulverized UND solid sample of both the fresh waste and the stabilized material of the lysimeter reactor, being the undersieve the highest source of HM. The aim was to evaluate the partitioning of HM in the solid state and to determine if they could be potentially released under particular conditions. The SCE method used for the HM partitioning is an application calibrated on soils and not on waste samples (Krishnamurti et al., 2002). This choice was made not only considering the physico-chemical features of the samples but also for further comparing the stabilized waste with its final form (the soil indeed) in a ‘‘geological repository”, considering a ‘‘Back to Earth” point of view (Cossu, 2016). The contaminants concentration in the eluates can represent the total emission potential of the samples. Starting from this assumption, the results of the analyses conducted were used to identify the fractions which mostly influence the total expected residual emission of the landfill. Considering the chemical speciation data, the waste expectable characteristics after a long-term aerated treatment and a flushed aftercare were estimated. Finally, the biochemical stability indexes for S sample were compared with the FSQ values suggested by the Lombardia Regional Government (D.G.R. 2461/14) to validate the biochemical stabilization of the material obtained. The situations in which the threshold limit concentration for emissions in a liquid matrix (D.G.R. 2461/14) is respected by the leaching test eluates is considered precautionary.

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

5

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

2.4. Analytical methods The chemical analyses on solid and liquid samples were conducted when possible following the certified Italian standards, officially derived from international certified procedures (CEN, 2002) (Table 1). Moreover, the analyses were always performed at least in duplicate and the related results are presented as the average value, ensuring that the standard deviation remains below 5%. The TOC on solid samples was measured using a TOC-VCSN Shimadzu Analyzer. TC, TN and total sulfur were measured by means of a CHN/S elemental analysis equipment. The HM content in solid samples and eluates was evaluated with Inductively Coupled Plasma (ICP), after a three step acid digestion at 95 °C, with concentrated nitric acid for 120 min, hydrogen peroxide for 120 min and hydrochloric acid for 15 min. Respiration Indexes (RI4 and RI7 mg/gTS of O2) were determined by means of Sapromat apparatus (H + P Labortechnik, Germany). The VFA analysis was performed following the two points titration procedure (Standard n° 5560 C, 1989). When standard methodologies could not be applied, alternative methods available in the literature were considered and results were validated comparing them with similar scientific studies. Experimental non-standard procedures were performed for the evaluation of organic polymers (both synthetic and natural, such as cellulose, hemicellulose and lignin), humic and fulvic acids, total chlorine and for the heavy metals partitioning through a SCE (specified in Table 1 as ‘‘bound metals”). Organic polymers in the solid samples were determined by means of analytical procedures adapted from Wang et al. (2015), Yang et al. (2006) and Li et al. (2004). They were classified as hydrolysable and non-hydrolysable organic polymers, considering their behavior with respect to acid digestions. The hydrolysable fractions include cellulose, hemicellulose and some kinds of synthetic polymers while the non-hydrolysable fractions mainly consist in lignin and other synthetic compounds resistant to acidic hydrolysis. Typically, 0.5 g (dry weight) sample of every waste category had been digested in 30 mL of acetone at 90 °C for 2 h (at reflux) to extract semi-polar substances. At the end of the digestion, the sample was filtered (Whatman Glass Microfiber Filters, 1.6 mm) and dried in oven (105 °C) until constant weight. The amount of compounds extracted was calculated by the difference between the dry weight of the sample before and after the acetone digestion. After this determination, the sample underwent a double-step hydrolysis with 150 mL of 72% w/v and the same amount of 3% w/v sulfuric acid solution. To enhance the reaction at each step, it was digested at 110 °C for 1 h (at reflux), after resting at 8–15 °C for 24 h. After hydrolysis, the filtered solid was washed with distilled water until sulfate ions were no more detected in the liquid phase and dried in oven at 105 °C. The amount of polymers soluble in the acid solution was calculated

as the difference between the dry weight of the sample before and after the double-step digestion. To determine the fraction of non-hydrolysable organic polymers, the residues of the digestion processes were pyrolysed at 550 °C for 2 h in a muffle. The weight loss of the solid sample corresponded to the desired fraction, while the remaining compounds were only characterized by inorganic carbon. The amount of hydrolysable organic polymers was calculated by the difference between the TOC of the initial solid samples and the non-hydrolysable organic polymers for each waste category. The determination of humic and fulvic acids in the eluates was performed following the procedure proposed by Baddi et al. (2004). 30 mL of each eluate sample underwent a double step centrifugation at 7000 rpm for 25 min with distilled water washing in order to remove any particulate matter in solution. The pH of the supernatant was adjusted to 2 with a sulfuric acid solution (2 M) to enhance humic acids precipitation. Indeed, humic acids have a very low solubility under these conditions. The samples were left setting for 24 h at 4 °C for a complete coagulation. Subsequently, a second double step centrifugation with washing was performed and the precipitate was dried under vacuum and weighed. The residue represented the amount of humic acids. The supernatant coming from centrifugation was collected and dialyzed through Spectra/PorÒ Dialysis membranes (3500 Da), after the neutralization of the acid which was added during the previous step. The amount of eluate retained by the membrane was finally dried to have the content of fulvic acids in the sample. Total chlorine in solid samples was determined according to Okada et al. (2007). The amount of solid sample considered was 0.5 g (dry weight); it underwent acid digestion with 50 mL of 10% nitric acid at 100 °C for 2 h (at reflux) and titration with silver nitrate to have the final result. The heavy metals SCE in pulverized UND solid F and S samples was performed through the multistep procedure proposed by Krishnamurti et al. (2002) (Table 2). This method allows the identification of eight different forms under which heavy metals could be found in soils: ionic exchangeable, bound to carbonate (adsorbed), as metal–organic complexes (associated with humic and fulvic acids), as easily reducible metal-oxides, bound to organic matter (other than humic and fulvic acids), in amorphous mineral colloids, as crystalline iron-oxides and bound to aluminum–silicate minerals. Although there are not standard procedures to evaluate the HM bonding state and partitioning within the MSW matrix, the results obtained through this sequential extraction can give some information on how the HM are released under particular conditions which are more aggressive than those of leaching in terms of pH and presence of oxidizing species. The reaction-extraction procedure was performed on 1 g sample with definite reagents and under controlled temperature and pH conditions as reported in Table 2. Every step required a precise mixing

Table 2 Heavy metals sequential extraction procedure taken from Krishnamurti et al. (2002). It is useful to determine if HM could be potentially released or remain fixed within the solid waste matrix. For each extraction step is indicated the corresponded fraction in soils, the necessary reagents and reaction conditions. Extraction

Fraction’s name (Krishnamurti et al., 2002)

Reagent

Reaction conditions

1 2 3 and 4a 5 6

Ionic exchangeable Bound to carbonates Fulvic and humic acid bound Easily reducible metal-oxide bound Organic bound

7 8

Amorphous mineral colloid bound Crystalline Fe-oxide bound

10 mL of 1 M NH4NO3 25 mL of 1 M CH3COONa 30 mL of 0.1 M Na4P2O7 20 mL of 0.1 M NH2OHHCl in 0.01 M HNO3 5 mL of 30% H2O2, 3 mL of 0.02 M HNO3 3 mL of 30% H2O2, 1 mL of 0.02 M HNO3 10 mL of 0.2 M (NH4)2C2O4/0.2 M H2C2O4 25 mL of 0.2 M (NH4)2C2O4/0.2 M H2C2O4 in 0.1 M ascorbic acid

pH 7, 25 °C, 4 h pH 5, 25 °C, 6 h pH 10, 25 °C, 20 h 25 °C, 30 min pH 2, 85 °C, 2 h pH 2, 85 °C, 2 h pH 3, 25 °C, 4 h dark pH 3, 95 °C, 30 min

a 30 mL of 0.1 Na4P2O7 extract was brought to pH 1 with addition of 5 M HCl and the suspension was left overnight for the coagulation of humic acids. The suspension was centrifuged at 10,500 rpm for 20 min. The amount of metals bound with fulvic acids were determined in the supernatant. The residue was solubilized with 0.1 M Na4P2O7 and the amount of metals attached with humic acids were calculated for the solution.

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

6

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

time for the extraction completion and was followed by a double step centrifugation (at 10,500 rpm for 20 min) with distilled water washing. The supernatant was collected and analyzed by ICP under the assumption that the amount of liquid added for each extraction was enough to collect the whole analyte from the solid. The results were converted to percentages having the concentration of each metal (Cd, Cr, Pb, Cu, Ni, Zn, Fe, Mn) in the initial UND solid samples as reference. The heavy metals partitioning through the SCE procedure was applied only for the under-sieve of both F and S samples because it was the waste category with the highest metal content.

3. Results and discussion The solid waste loaded into the reactor (Fresh F sample) amounted to 750 kg, which corresponded to 440 kgTS being the total solids 58% of the mass. At the end of the test, the stabilized waste (S sample) quantity was 510 kg, which corresponded to 398 kgTS being the total solids 78% of the mass. Totally, the residual waste dry mass amounted to 91%, while the remaining 9% was emitted with gas and leaching during the reactor life. The results obtained from the waste fractional characterization and from the main chemical analyses on the total bulk F sample (as TOC, TKN, chlorine) were strictly comparable to those obtained for the fresh waste loaded into the lysimeter at the startup of the test: the analytical difference was always under 6% and anyway under the standard analytical error (Fig. 2). Moreover, the collection system and the pre-treatments in the MBT plant from which waste was sampled remained the same over the time. For these reasons, the comparisons done between F and S samples was considered coherent and significant. 3.1. Fractional composition The chemical and biochemical processes occurred to convert fresh waste into the stabilized material promoted waste biological stabilization, emissions dropdown and size reduction of all the fractions, with consequent increase of the under-sieve from 44% to 77% by mass (Fig. 2). Small fractions increased because bigger

particles (paper, plastics, etc.) were broken into smaller pieces by compaction and frictions occurring inside the reactor. In fresh waste, plastics amounted to 20.5%, cellulosic material to 17.5% and the other categories remained below the 8%. For the stabilized material, the main residual categories were glass (8.96%), plastics (5.87%), inert (4.75%) and metals (1.62%), all the others were below 1%. This analysis showed that the under-sieve fraction was doubled in S sample, collecting materials from kitchen residues, green waste, plastics, textiles and cellulosic material, whose percentages drop consistently. The glass, inert and metals (others fraction) remained virtually unchanged in time (Fig. 2). The granulometric analysis showed no retention of waste in the 100 mm sieve for both Fresh and Stabilized samples. Textiles, plastics and metals were mainly collected in the 60 mm mesh, with traces of green and wooden materials for the stabilized waste. All the other categories were found in size smaller than 60 mm for both the samples. The F sample material was mainly between 20 and 60 mm (50%) and in less extent in mesh 60–100 mm (below 10%). The S sample material was mainly in the under-sieve fraction (77%) and in less extent in mesh 20–60 mm (20%). 3.2. Liquid emission potential The chemical and biochemical processes occurred in the reactor promoted changes in the compounds subcategories and generated liquid and gaseous emissions. As mentioned before, the concentration of the main contaminants in the eluates obtained with the standard leaching test (UNI EN 12,457-2) can give an estimation of the liquid potential emissions of the samples. The Fig. 3 shows the potentially total residual emission of each waste category for carbon, nitrogen, chloride and sulfur compounds, referring to the initial TS presence in the solid sample. For all the compounds, the highest impact on liquid emissions was due to the under-sieve fraction, which contributes for approximately 50% in the F sample and is almost the only residual source of emission for the S sample. In fresh waste, CEL and PLA categories contributed to 30–40%, while KIT, TEX and GRE fractions were secondary impacting categories, always totalizing less than 10% of the estimated emissions. The S sample potential emission was an order of magnitude lower respect to the F sample one, excepting for the

Fig. 2. Fractional composition of the fresh waste and the stabilized material. Waste macro-categories chosen: UND – under-sieve, KIT – kitchen residues, GRE – green and wooden materials, PLA – plastics, TEX – textiles, CEL – cellulosic material. Reference waste: waste loaded into the lysimeter reactor.

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

7

Fig. 3. Carbon, nitrogen, chlorine and sulfur concentrations in eluates from leaching tests made in Fresh and Stabilized samples. Concentrations are subdivided considering the contribution of the main waste fractions separately (UND: under-sieve, KIT: kitchen residues, GRE: green waste and wooden materials, PLA: plastics, TEX: textiles, CEL: cellulosic material).

sulfur compounds for which it was halved (Fig. 3). In particular, the leachable carbon dropped down from 19.9 g/kgTS of C to 0.7 g/kgTS of C, totalizing the 96% of emissions reduction. 3.3. Carbon TC in the under-sieve fraction decreased from 176 g/kgTS of C in F sample to 142 g/kgTS of C in S sample mainly due to the leachable fractions reduction (IC-L, OC-L:VFA, OC-L:Humic and OC-L: Others) that became nearly undetectable (Fig. 4). The most abundant subcategory of carbon was the polymeric carbon (OC-NL: Polymers), which dropped from 132.6 g/kgTS of C to 92.7 g/kgTS of C. At the same time, the inorganic and the organic non-mobile carbon registered together an increase from 20 g/kgTS of C to 40 g/kgTS of C, mainly due to the fractionation effect occurring from F to S samples inside the reactor. This means that bigger particles (in particular coming from paper and plastics) broke into smaller particles, increasing the amount of carbon into the under-sieve as consequence. Considering the whole carbon change from F to S sample, weighed accounting all the fraction contribution, the total leachable fraction extracted was 8.9% with respect to the initial TC, the residual extractable fraction was 0.2% and the residual nonmobile fraction was 55% (Fig. 5). A significant part of carbon compounds was stabilized and immobilized in a carbon sink. The processes occurred in the reactor, enhanced by long-term aeration and flushing, caused a consistent reduction of biodegrad-

able organic matter and leaching of mobile substances contained in all the waste categories. As a consequence, the reduction of soluble fractions can be observed with a simultaneous increase of the nonmobile inorganic carbon. The long-term treatments applied to the S sample made the reduction of the organic polymers possible, representing the hardly degradable part of the residual MSW, characterized by a very slow degradation kinetics (Fig. 5). The content of VFA was smaller in stabilized material than in fresh waste, due to their fast consumption since the first stage of the degradation processes (Cossu et al., 2016). As expected, the final concentration of humic substances was smaller than the input ones (Fig. 4) due to their leaching during the lysimeter tests. The organic compounds biodegradability in the S sample can be evaluated through the BOD5/COD ratio and RI4 which are indexes commonly used for stating the biochemical stability (Cossu et al., 2012, D.G.R. 2461/14), while the COD/TOC ratio can highlight the oxidation level of the organic carbon matter (Table 3). The BOD5/ COD ratio indicates the amount of biodegradable organic compounds over the total amount of oxidizable matter in liquid samples (eluate, leachate, wastewater, etc.). The ratio values comprised between 0.02 and 0.13 suggest waste biochemical stability (Sekman et al., 2011), while values between 0.4 and 0.8 mean high biodegradability (Kjeldsen et al., 2002). The BOD5/COD ratio results were always in the range suggested for the biochemical stability, or even lower, in reference to the FSQ threshold limits (Table 3). The F samples showed RI4 values ranging between 44.03 and 21.91 g/kgTS of O2, while the index in the Stabilized

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

8

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

Fig. 4. Carbon, nitrogen, chlorine and sulfur chemical speciation in fresh and stabilized under-sieve fraction of waste samples. Results are reported in concentration respect to the TS of each sample. OC – Organic Carbon, IC – Inorganic Carbon, L – Leachable, NL – Non Leachable.

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

9

Fig. 5. Carbon, nitrogen, chlorine and sulfur chemical speciation in fresh and stabilized bulk waste samples. Results are reported in percentage (w/w) respect to initial solid waste total solids amount. OC – Organic Carbon, IC – Inorganic Carbon, L – Leachable, NL – Non Leachable.

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

10

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

Table 3 Biochemical stability index (BOD5/COD), organic compounds (COD, COD/TOC), nitrogen compounds (N-ammonia, N-Nitric) concentrations for the stabilized waste sample eluate and respiration index (RI4) for the stabilized solid waste sample are reported in comparison with the FSQ values proposed by D.G.R. 2461/14. The results are reported for each waste fraction and for the total stabilized material. Matrix

Parameter

FSQ value

UND

KIT

GRE

PLA

TEX

Stabilized waste

Eluate sample

COD (mg/L of O2) BOD5/COD COD/TOC N-Ammonia (mg/L of N) N-Nitric (mg/L of N)

1500 0.1 N.A. 50 20

188 0.005 4.71 2.7 58

124 0.008 8.38 7.6 91

430 0.05 3.29 7 155

198 0.005 4.44 6 19

254 0.05 3.69 4.5 61

161 0.005 3.97 2.54 47.6

Solid sample

RI4 (mg/gTS of O2)

2

0.44

0.79

2.2

1.5

1.2

0.45

samples ranged between 2.20 and 0.44 g/kgTS of O2 with a weighted average value of 0.45 g/kgTS of O2 (Table 3). As expected, the oxygen consumption in fresh waste was an order of magnitude higher respect to the stabilized material, meaning a consistent reduction of the potential biochemical activity between the two samples. The sole values exceeding the FSQ threshold limits were the Respiration Indexes (RI4) calculated for the cellulosic material. These surpluses were tolerable considering the dilution effect among all the categories. As a matter a fact, the RI4 of the total bulk S sample respected the FSQ threshold limit (Table 3). The COD/TOC ratio increased from fresh to stabilized waste, remaining unchanged only in green and wooden materials. The S sample was constituted by carbon compounds in a higher oxidized status respect to the F sample, due to the aerobic degradation occurred in the lysimeter. 3.4. Nitrogen The total nitrogen content in the under-sieve fraction decreased from 15.8 g/kgTS of N in F waste to 10.5 g/kgTS of N in S sample (Fig. 4). This reduction was mainly due to the organic nitrogen and the ammonium ion depletion from 2.8 g/kgTS of N to 0.2 g/ kgTS of N, caused by nitrification and leaching. Moreover, nitrates were found always below 0.3 g/kgTS of N in both F and S samples, being consumed by denitrification and leaching as well. Considering the whole nitrogen change from Fresh to Stabilized samples, weighted amounting all the fraction contribution, the total leachable fraction extracted was 13.8% with respect to the initial TN, the residual extractable fraction amounted to 4.0% (mainly constituted by nitrates) and the residual non-mobile fraction was 53% (Fig. 5). According with these data, a significant fraction of nitrogen compounds was stabilized and immobilized in a nitrogen sink. The organic nitrogen initially present in solid waste was partially ammonified to N-NH+4 during the whole reactor life. Concomitantly aerobic nitrification converted ammonia ions produced into nitrite and nitrate ions, which are soluble compounds. After that, denitrification occurred in the anoxic zones of the waste mass, converting nitrates into free nitrogen gas (Morello et al., 2016). As consequence of all these processes, 29.1% of nitrogen was progressively mobilized and emitted as free nitrogen gas or leached away as organic nitrogen, ammonia ion, nitrite or nitrate (Fig. 5). Ammonia is generally recognized as one of the most persistent leachate contaminant in the long-term landfill management (Ritzkowski et al., 2016). Long-term emissions of nitrogen compounds are mainly due to the long-term ammonification process, which is able to convert non soluble nitrogen compounds in soluble compounds even after many years (Berge et al., 2005). In this case, N-NH+4 concentration in the eluates of the S samples was always below the FSQ values for all the waste categories (Table 3), meaning a good nitrification efficiency. On the contrary, nitrates concentration in the same samples exceeded the FSQ limits in all the cases, with the only exception for plastics.

3.5. Chlorine The total chlorine content in the under-sieve F sample was 21.7 g/kgTS of Cl, whose 87% was progressively leached away, remaining only 2.9 g/kgTS of Cl in S sample (Fig. 4). Considering the whole chlorine change from fresh to stabilized sample, the final nonmobile fraction amounted to 12.3%, while the still mobile was 4.9% with respect to the initial total chlorine (Fig. 5). These data highlight that the greater part of chlorine was leached away and that a landfill is not a good sink for this element since residual emissions are still present in long term. The chlorine presence in solid waste is often high and its mobile form (chloride) is considered one of the most persistent compound in long-term landfill emissions since it can be removed only by leaching (Morello et al., 2016). The stabilized material contained mostly non-mobile chlorine compounds; this means that the treatments performed in the reactor were sufficient to remove the great part of the leachable fraction. 3.6. Sulfur The total sulfur under-sieve fraction content dropped from 6.7 g/kgTS of S in F sample to 2.5 g/kgTS of S in S one, totalizing 63% of reduction (Fig. 4). Considering the total sulfur change from fresh to stabilized sample, the residual non-mobile fraction amounted to 33.8%, while the residual leachable fraction was 15.5% with respect to the initial total content, meaning that less than 50% of sulfur was emitted (Fig. 5). Sulfur compounds are significantly influenced by the redox conditions of the reactor. In particular, in aerobic and semi-aerobic condition, oxidizing environments enhance sulfate production while in anaerobic conditions sulfates are used as oxygen source. Sulfide compounds can be present both in slightly mobile and in non-mobile forms (included into non-mobile sulfur). In the stabilized material, they are mainly present in non-mobile forms. Moreover, the S sample was mostly characterized by the presence of non-mobile sulfur compounds and in less extent by sulfates. 3.7. Heavy metals Heavy metals presence in solid state was abundant (Table 4) while their solubility was found to be considerably lower, often negligible in Stabilized samples (Table 5). These test results confirmed that HM liquid emissions are generally not considered to be of environmental concern (Qu et al., 2008). The HM presence in solid state was analyzed in detail in F and S samples to understand the bonding state of each metal and so if it could be potentially released or remain fixed within the waste matrix under particular conditions (Table 2). The HM SCE results (Table 4) were presented for the under-sieve fraction, being the most abundant in both F and S samples and with the highest contribution to the liquid emissions. The trend showed an immobilization of heavy metals in the waste matrix over time, however these results must be carefully considered because the heterogeneity of

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

11

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

Table 4 Heavy metals total mass in Fresh and Stabilized samples of the bulk waste and the under-sieve fraction together with its sequential chemical extraction results. Percentages are calculated considering the ICP values of each extraction process related to the total amount of metal in the initial under-sieve solid fraction. The total amount of fresh waste is 750 kg and the total amount of stabilized waste is 510 kg. Cr

Metals mass in the bulk waste (kg) Metals mass in under-sieve (kg)

Cu

Fe

Ni

Pb

Zn

S

F

S

F

S

F

S

F

S

F

S

F

S

0.04 0.01

0.01 0.01

1.07 0.09

0.47 0.39

16.11 2.46

15.01 10.49

0.11 0.04

1.38 1.35

0.05 0.03

0.05 0.05

0.36 0.07

0.14 0.08

0.77 0.62

3.79 3.65

2.5 2.4 18.9 11.3 3.3 11.2 3.3 1.1 46.1

0.2 0.2 8.9 0.2 6.7 4.4 27.7 11.0 40.7

0.0 0.0 8.9 0.2 1.1 7.2 54.2 9.7 18.6

3.4 13.1 3.6 0.2 12.1 5.2 7.9 4.3 50.1

0.5 3.5 1.2 0.0 74.6 4.7 1.2 0.3 14.0

6.5 8.5 6.1 0.8 20.3 16.2 2.9 1.3 37.5

1.4 4.7 6.6 2.3 8.0 19.0 20.6 5.7 31.7

0.3 5.1 33.5 1.0 26.0 10.1 1.6 0.9 21.5

0.2 2.1 9.0 1.7 14.2 43.1 7.5 1.8 20.5

2.6 11.4 9.1 0.2 1.9 2.0 0.6 0.3 72.0

0.9 19.1 12.8 0.2 6.9 1.6 0.9 0.5 57.1

Chemical partitioning of metals in the under-sieve fraction (%) Fraction 1 0.6 1.8 9.5 Fraction 2 0.6 1.9 2.0 Fraction 3 9.8 30.0 15.6 Fraction 4 4.5 15.6 2.8 Fraction 5 2.0 6.4 8.2 Fraction 6 8.2 23.3 90.8 Fraction 7 11.5 22.5 5.2 Fraction 8 29.7 23.7 2.9 Residual fractiona 33.2 25.2 37.0 a

Mn

F

Calculated as difference between the total amount of metal in untreated under-sieve and the sum of the values detected in the extraction steps.

Table 5 Heavy metals content in the eluate from leaching test of Fresh and Stabilized samples. Eluate concentrations are compared with FSQ values for leachate (D.G.R. 2461/14). Results are reported in mg/L. Total amount: concentration of HM in the F and S samples, evaluated as the pondered average of UND, KIT, GRE, PLA, TEX and CEL fractions.

FSQ D.G.R.2461/14 (mg/L) UND KIT GRE PLA TEX CEL Total amount

Cr

Cu

Fe

Mn

Ni

Pb

2 F 3.19 2.01 2.77 1.41 10.93 6.41 3.26

1 F 2.84 0.32 1.92 1.33 3.10 4.03 2.31

2 F 106.8 36.4 20.2 56.2 47.7 145.8 83.5

2 F 5.27 1.77 3.32 3.88 3.16 5.06 4.09

2 F 2.37 1.15 1.57 1.77 2.33 2.58 1.95

0.2 F 1.30 0.15 0.77 0.31 1.11 2.68 1.13

S 0.04 0.04 0.13 0.04 0.04 0.14 0.03

S 1.52 0.40 0.71 2.54 3.47 6.79 1.35

S 14.0 1.3 27.7 31.1 11.0 43.1 12.9

wastes can analytically interfere even after the pulverization of the sample. The residual fractions after the reaction-extraction procedure were generally the highest detected in both the fresh waste and the stabilized material. In general, a decrease of potentially extractable metal ions was observed from F UND to S UND sample in extraction 1, while an increase was registered in extraction 3 and 4, 6 and 7. Cr was mainly found in extracted fractions 5 and 8, as for Pb, Cu, Ni, Zn and Mn in both the under-sieve samples. Moreover, Cr is generally found to be largely associate with particulate organic matter, being it a recalcitrant element in organic forms (Van der Sloot et al., 2017). The Pb, Cu, Ni, Zn and Mn presence was high in the extracted fraction 2. The residual fractions of the reaction-extraction procedure were evaluated as the fraction of each metal remained after the sequential extraction, calculated as the difference between the total content found in the UND solid samples and the sum of all the extracted contributions. Negative percentages were obtained for Cu in the under-sieve of F sample and for Cr in the under-sieve of S sample, possibly due to the difficulties in the evaluation of HM concentration near the ICP detection limits. Summarizing, the presence of potentially non-mobile compounds in the residual fractions and in the extracted fractions 2, 5 and 8 prevailed and the exchangeable residual free metal ions decreased consistently from the F UND sample to the S UND sample (Table 4). Following the ‘‘Back to Earth” approach (Cossu, 2016), these results can be roughly read by approximating the stabilized waste to a soil. Making this strong assumption, it appears that the biggest part of HM is bound to non-mobile Al-Si minerals, carbonates, metal-oxides and Fe-oxides, while the residual mobile exchangeable ionic and the organic bound HM fractions decrease consistently from fresh to stabilized waste samples. These results are interesting to understand the potential fate of HM in landfills but more data on other MSW samples should be required to have a significant quantification of this chemical partitioning.

S 0.80 0.17 0.19 0.22 0.37 0.47 0.64

S 0.14 0.04 0.14 0.13 0.19 0.80 0.12

Zn S 0.25 0.04 0.25 0.40 0.18 0.94 0.22

3 F 10.86 5.44 2.94 1.51 5.91 8.97 6.73

S 3.65 0.48 2.38 3.67 3.83 9.64 3.07

The heavy metals presence in the stabilized material (S sample) is not matter of concern, as proven by the eluates results comparisons with the FSQ values (Table 5). In general, the residual extractable fraction of metals was less than 1% respect to the initial mass for all the HM considered, with the exception for Cd and Zn. The most important contribution in the total emission came from Zn and Fe. The mobile fraction in F sample for Fe ranged between 145.8 mg/L(eluate) in CEL fraction and 20.2 mg/L(eluate) in GRE fraction, with a pondered average value of 83.5 mg/L(eluate). In the stabilized material, the emission potential was consistently lower and the pondered average emission of Fe was 12.9 mg/L(eluate) (Table 5). The residual non-mobile fraction ranged between 60% and 99%, in most of the cases it was higher than 95%. These data proved that a landfill is a good sink for HM deposited inside. The comparisons of the HM concentration in S sample eluates with FSQ values suggested by the Lombardia Regional Government (D.G.R. 2461/14) can give important information on the heavy metals emission hazard (Tables 5). The law limits are given for leachates while the HM emission potential was evaluated for the eluates extracted through the leaching test, so the comparison has only a general validity. Despite that, the FSQ threshold values for Cr, Mn and Ni were always respected in S sample, while a negligible surplus was monitored for Cu, Pb and Zn. On the contrary, Fe showed still high concentrations in eluate, meaning a further significant residual emission potential.

4. Conclusions The fresh waste and the stabilized material characterization showed that the under-sieve was the most abundant fraction, amounting to 44% in F sample and to 77% in S sample. Moreover, the under-sieve can be considered the most environmentally rele-

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026

12

L. Morello et al. / Waste Management xxx (2018) xxx–xxx

vant fraction, hosting a consistent part of the mobile compounds in F sample (40.7% of carbon, 44.0% of nitrogen, 47.6% of chloride and 40.0% of sulfur) and the greater part of the residual emissions in S sample (88.4% of carbon, 90.9% of nitrogen, 98.4% of chloride and 91.1% of sulfur). These data highlight that the highest contribution in the total potential emission during the aftercare of a MSW landfill is due to the under-sieve fraction. The landfilled MSW proved to be a good sink, finally storing more than 55% of carbon, 53% of nitrogen, 33% of sulfur and 90% of HM which were initially present in fresh waste samples. A general decrease in leachable fractions from fresh to stabilized waste was observed for each category. The tests showed that solid waste is not a good sink for chlorine, whose residual non-mobile fraction amounts to 12.3% only. Chlorides have been leached since the beginning of the landfill simulation test in the lysimeter reactor. However residual emissions are still present in long term. Chlorides, sulfates, nitrates and humic substances were considered as the most persistent elements and compounds since they had the highest residual emission potential. Heavy metals emissions were not of great concern. The HM partitioning showed that the biggest part of them is fixed within the MSW solid matrix, while the residual mobile exchangeable free metal ions decreased consistently from fresh to stabilized waste. The comparisons made with FSQ values, law concentration threshold limits and stabilization indexes (as COD/TOC, BOD5/ COD, RI4) highlighted that the degradation processes and the leaching occurred efficiently in the lysimeter, minimizing the residual potential emissions. The results of the stabilized waste analyses could be considered a useful reference for the estimation of the long-term concentrations achievable treating a MSW in an aerated landfill. References Baddi, G.A., Hafidi, M., Cegarra, J., Alburquerque, J.A., Gonzálvez, J., Gilard, V., Revel, J.C., 2004. Characterization of fulvic acids by elemental and spectroscopic (FTIR and C-NMR) analyses during composting of olive mill wastes plus straw. Bioresour. Technol. 93, 285–290. Berge, N.D., Reinhart, D.R., Townsend, T.G., 2005. The fate of nitrogen in bioreactor landfills. Crit. Rev. Environ. Sci. Technol. 35, 365–399. 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. (Oxford) 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, 399–414. Bolyard, S.C., Reinhart, D.R., 2016. Application of landfill treatment approaches for stabilization of municipal solid waste. Waste Manage. (Oxford) 55, 22–30. Bolyard, S.C., 2016. Application of Landfill Treatment Approaches for the Stabilization of Municipal Solid Waste. Doctoral dissertation (http://stars. library.ucf.edu/etd/4878/). CEN, 2002. NEN-EN 12457-4:2002en Compliance test for leaching of granular waste materials and sludges. Christensen, T.H., Kjeldsen, P., 1989. Basic Biochemical Processes in Landfills. In: Christensen, Cossu, Stegmann (Eds.), Sanitary Landfilling: Processes, Technology and Environmental Impact. Academic Press, London, pp. 29–42. Christensen, T.H., Scharff, H., Hjelmar, O., 2011. Landfilling; Concepts and Challenges. Christensen T.H. 2011, Solid Waste Technology and Management, vol. 2. Blackwell Publishing Ltd. Cossu, R., Lai, T., Sandon, A., 2012. Standardization of BOD5/COD ratio as a biological stability index for MSW. Waste Manage. (Oxford) 32, 1503–1508.

Cossu, R., Morello, L., Raga, R., Cerminara, G., 2016. Biogas production enhancement using semi-aerobic pre-aeration in a hybrid bioreactor landfill. Waste Manage. (Oxford) 55, 83–92. Cossu, 2016. Back to Earth Sites: From ‘‘nasty and unsightly” landfilling to final sink and geological repository. Waste Manage. (Oxford) 55, 1–2. De la Cruz, F.B., Chanton, J.P., Barlaz, M.A., 2013. Measurement of carbon storage in landfills from the biogenic carbon content of excavated waste samples. Waste Manage. (Oxford) 33, 2001–2005. D.G.R. 2461/14, Delibera Giunta Regionale 7 ottobre 2014, n. 2461, Linee guida per la progettazione e gestione sostenibile delle discariche, Bollettino Ufficiale della Regione Lombardia 10 Ottobre 2014. Edjabou, M.E., Jensen, M.B., Gӧtze, R., Pivnenko, K., Petersen, C., Scheutz, C., Astrup, T.F., 2015. Municipal solid waste composition: Sampling methodology, statistical analyses, and case study evaluation. Waste Manage. (Oxford) 36, 12–23. Fellner, J., Doeberl, G., Allgaier, G., Brunner, P.H., 2009. Comparing field investigations with laboratory models to predict landfill leachate emissions. Waste Manage. (Oxford) 29, 1844–1851. Kjeldsen, P., Barlaz, M.A., Rooker, A.P., Baun, A., Ledin, A., Christensen, T.H., 2002. Present and long-term composition of MSW landfill leachate: a review. Crit. Rev. Environ. Sci. Technol. 32, 297–336. Krishnamurti, G.S.R., Naidu, R., 2002. Solid-solution and phytoavailability of copper and zinc in soils. Environ. Sci. Technol. 36, 2645–2651. Laner, D., Crest, M., Scharff, H., Morris, J.W.F., Barlaz, M.A., 2012. A review of approaches for the long-term management of municipal solid waste landfills. Waste Manage. (Oxford) 32, 498–512. Li, S., Xu, S., Liu, S., Yang, C., Lu, Q., 2004. Fast pyrolysis of biomass in free-fall reactor for hydrogen-rich gas. Fuel Process. Technol. 85, 1201–1211. Manfredi, S., Christensen, T.H., 2009. Environmental assessment of solid waste landfilling technologies by means of LCA-modeling. Waste Manage. (Oxford) 29, 32–43. Morello, L., 2013. Experimental evaluation of Final Storage Quality threshold values for a semi-aerobic landfill. Master Thesis 2013, Library of Engineering, ICEA Department, University of Padova. Morello, L., Cossu, R., Raga, R., Pivato, A., Lavagnolo, M.C., 2016. Recirculation of reverse osmosis concentrate in lab-scale anaerobic and aerobic landfill simulation reactors. Waste Manage. (Oxford) 56, 262–270. Okada, T., Tojo, Y., Tanaka, N., Matsuto, T., 2007. Recovery of zinc and lead from fly ash from ash-melting and gasification-melting processes of MSW – Comparison and applicability of chemical leaching methods. Waste Manage. (Oxford) 27, 69–80. Piovesan, E., 2007. Valutazione della qualità finale delle discariche: osservazione nel lungo periodo di un modello di discarica sostenibile. Thesis, IMAGE, Università degli Studi di Padova. Pivnenko, K., Astrup, T.F., 2016. The challenge of chemicals in material lifecycles. Waste Manage. 56, 1–2. Qu, X., He, P.J., Shao, L.M., Lee, D.J., 2008. Heavy metals mobility in full-scale bioreactor landfill: Initial stage. Chemosphere 70, 769–777. Ritzkowski, M., Walker, B., Kuchta, K., Raga, R., Stegmann, R., 2016. Aeration of the Teuftal landfill: Field scale concept and lab scale simulation. Waste Manage. (Oxford) 55, 99–107. Sekman, E., Top, S., Varank, G., Bilgili, M.S., 2011. Pilot-scale investigation of aeration rate effect on leachate characteristics in landfills. Fresenius Environ. Bull. 20. Townsend, T.G., Powell, J., Jain, P., Xu, Q., Tolaymat, T., Reinhart, D., 2015. Sustainable Practices for Landfill Design and Operation. Springer, Series: Waste Management Principles and Practice. Van der Sloot, H.A., Kosson, D.S., van Zomeren, A., 2017. Leaching, geochemical modelling and field verification of a municipal solid waste and a predominantly non-degradable waste landfill. Waste Manage. (Oxford) 63, 74–95. Vettorazzi, G., 2005. Bilancio di massa del carbonio e dell’azoto in alcuni modelli di discarica per la valutazione della sostenibilità. Thesis, IMAGE, Università degli Studi di Padova. Wang, X., De la Cruz, F.B., Ximenes, F., Barlaz, M.A., 2015. Decomposition and carbon storage of selected paper products in laboratory-scale landfills. Sci. Total Environ. 532, 70–79. Yang, H., Yan, R., Chen, H., Zheng, C., Lee, D.H., Liang, D.T., 2006. In-depth investigation of biomass pyrolysis based on three major components: hemicellulose, cellulose and lignin. Energy Fuels 20, 388–393.

Please cite this article in press as: Morello, L., et al. Storage potential and residual emissions from fresh and stabilized waste samples from a landfill simulation experiment. Waste Management (2018), https://doi.org/10.1016/j.wasman.2018.01.026