Dissipation pathways of organic pollutants during the composting of organic wastes

Chemosphere 87 (2012) 137–143

Contents lists available at SciVerse ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Dissipation pathways of organic pollutants during the composting of organic wastes Gwenaëlle Lashermes a,1, Enrique Barriuso a, Sabine Houot a,⇑ a

INRA, UMR1091, Environnement et Grandes Cultures, INRA-AgroParisTech, F-78850 Thiverval-Grignon, France

a r t i c l e

i n f o

Article history: Received 25 July 2011 Received in revised form 30 November 2011 Accepted 1 December 2011 Available online 30 December 2011 Keywords: Composting Fluoranthene Linear alkylbenzene sulfonates (LAS) Nonylphenol (NP) Glyphosate Non-extractable residues

a b s t r a c t The organic pollutants (OPs) present in compostable organic residues can be recovered in the final composts leading to environmental impacts related to their use in agriculture. However, the composting process may contribute to their partial dissipation that is classically evaluated through the concentration decrease in extractable OPs, without identification of the responsible mechanisms as mineralization or stabilization of OP as non-extractable residues (NER) or bound residues. The dissipation of four 14Clabeled OPs (fluoranthene; 4-n-nonylphenol, NP; sodium linear dodecylbenzene sulfonate, LAS; glyphosate) was assessed during composting of sewage sludge and green waste. The dissipation of LAS largely resulted from its mineralization (51% of initial LAS), whereas mineralization was intermediate for NP (29%) and glyphosate (24%), and negligible for fluoranthene. The NER pathway mostly concerned NP and glyphosate, with 45% and 37% of the recovered 14C being found as NER at the end of composting, respectively. In the final composts, the proportions of water soluble residues of OPs considered as readily available were <11% of recovered 14C-OPs. However, most fluoranthene remained solvent extractable (72%) and potentially available, whereas only 18% of glyphosate and less than 7% of both NP and LAS remained solvent extractable in the final compost. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Compost can be contaminated with organic pollutants (OPs) initially present in organic feedstock materials. A broad range of OPs can be found including pesticide residues on green waste (Büyüksönmez et al., 2000) or many other chemicals in wastewater sludge (Harrison et al., 2006) or in biowastes (Brändli et al., 2005). Composting has been recognized to largely decrease OP concentrations in the final composts applied on soil (Amir et al., 2005; Pakou et al., 2009). In several countries, thresholds of maximum concentrations into the compost have been defined for some of OPs such as polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) or polychlorinated dibenzo-p-dioxins/furans (PCCDs/Fs) (Hogg et al., 2002). New regulations including limitation for other emergent OPs such as linear alkylbenzene sulfonates (LAS) or nonylphenol (NP) are under discussion (European Commission, 2000, 2001). During composting, OPs can be incompletely degraded by microorganisms into metabolites or mineralized as CO2, lixiviated or establish sorption interactions with the composted organic matter (Michel et al., 1995; Lashermes et al., 2010) culminating with

⇑ Corresponding author. Tel.: +33 130815401; fax: +33 130815396. E-mail addresses: [email protected] (G. Lashermes), enrique. [email protected] (E. Barriuso), [email protected] (S. Houot). 1 INRA, UMR614, Fractionnement des AgroRessources et Environnement, 2 Esplanade Roland Garros, BP 224, F-51686 REIMS cedex 2, France. 0045-6535/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2011.12.004

the formation of non-extractable residues (NER) that remain in the compost after classical analytical extraction procedures (Barriuso et al., 2008). However, the OP dissipation during composting is classically evaluated through the concentration decrease of solvent extractable OP, without consideration of involved mechanisms: mineralization or stabilization as NER. Mineralization is the only true mechanism of OP elimination while the stabilization as NER reducing their mobility and potential ecotoxicological effect (Kästner et al., 1999) limits contamination risks on the short term. However, NER are susceptible to later remobilization (Gevao et al., 2000) and have to be taken into account in risk assessment over medium and long terms. An increasing number of references have reported the dissipation of OPs during composting. In most cases the dissipation of PAHs has been observed, with median decrease of OP concentrations varying from 10% for chrysene to 81% for acenaphtene (Lazzari et al., 2000; Amir et al., 2005; Oleszczuk 2006, 2007; Brändli et al., 2007). A dissipation higher than 77% of LAS has been recorded (Pakou et al., 2009), 87% of NP (Gibson et al., 2007) and 50% for pesticides (Kupper et al., 2008). However, only very few studies differentiated the dissipation pathways, which requires the use of 14C-labeled OPs. Less than 25% of initial 14C-PAH formed NER during composting (Racke and Frink, 1989; Hartlieb et al., 2003). The formation of NER has been reported as very variable for pesticides, from 20% of initial 14C-2,4-D to 86% for 14C-carbaryl with low mineralization in most cases (Racke and Frink, 1989; Michel et al., 1995; Hartlieb et al., 2003).

138

G. Lashermes et al. / Chemosphere 87 (2012) 137–143

The objectives of this study were to assess the dissipation pathways of OPs with contrasted characteristics during composting and their availability in the final compost. A mixture of aerobically digested sewage sludge with green waste including branches, grass clipping, hedge trimming and leaves was spiked with 14C-OPs and composted over 83 d in triplicate, using small-scale instrumented reactors. Four 14C-labeled OPs were studied: a PAH (fluoranthene), two surfactants (NP and LAS) and a widely-used herbicide (glyphosate) whose behavior during composting has been poorly investigated (Büyüksönmez et al., 2000). The mineralization, volatilization and lixiviation of the OPs during composting were determined. The speciation of OP was also followed during composting: (1) water extraction was used to estimate an easily available fraction for degrading microorganisms, plant assimilation or transfer to ground water; (2) solvent extraction allowed to assess potentially available fraction; and (3) the NER was considered as non directly available fraction (Benoit and Barriuso, 1997). 2. Materials and methods 2.1. Organic pollutants The [3C-ring-14C] fluoranthene (specific activity: 1665 MBq mmol1, 98.3% radiopurity) and the [methyl-14C] N(phosphonomethyl)glycine (glyphosate) (specific activity: 81.4 MBq mmol1, 93.8% radiopurity) were purchased from Sigma Chemicals (St. Louis, USA), the [U-ring-14C] 4-n-nonylphenol (specific activity: 1924 MBq mmol1, 99% radiopurity) from ARC-900 (St. Louis, USA), and the [U-ring-14C] sodium linear dodecylbenzene sulfonate (specific activity: 230.9 MBq mmol1, 92.7% radiopurity) from Izotop (Budapest, Hungary). Non-labeled fluoranthene (99% purity), NP (99.5% purity), and LAS (79.9% purity, containing the C10–C13 homologous) were obtained from Sigma Chemicals (St. Louis, USA), Interchim (Montluçon, France), and Sasol (Marl, Germany), respectively. The water solubilities (Sw) were 25  104, 1  104, 7 and 0.26 mg L1 for LAS, glyphosate, NP and fluoranthene, respectively. The octanol–water partition coefficients (Kow) were 5.16, 1.23, 2.73, 3.4 (as log Kow) for NP, fluoranthene, LAS, and glyphosate, respectively (PHYSPROP database). Solutions of 14C-labeled OPs were prepared in methanol for fluoranthene and NP, in MilliQ water (Millipore, Molsheim, France) for LAS and glyphosate. Isotopic dilution with non-labeled OP was used to reach the final concentrations of 44, 28 919, 29 and 700 mg L1 and 53 407, 47 816, 84 233 and 51 071 MBq L1 for glyphosate, LAS, fluoranthene, and NP, respectively. 2.2. Composting set-up The waste mixture and composting reactors have been described in Lashermes et al. (2012). Briefly, the waste mixture was roughly ground and contained (% of DM): aerobic digested sewage sludge (20%), branches (25%), grass clippings (15%), hedge trimmings (20%), and leaves (20%). The composting system comprised 4-L glass reactors supplied with warm humidified air and surrounded by an external jacket through which water circulated from a thermostatic bath. Composting was performed in triplicate for each OP introducing into each reactor 750 g wet weight of the initial waste mixture which corresponded to different equivalent dry weight in the two sets of composting experiments because the DM of organic waste evolved during storage, 250 g DM in glyphosate and LAS experiments and 400 g DM in fluoranthene and NP. The initial waste mixture was spiked with the OP solution drop by drop and under continuous mixing (either 20 mL of the fluoranthene, NP and glyphosate solutions, or 25 mL of the LAS solution) to reach the following initial OP concentrations (mg kg1

DM): 3, 2860, 1.5 and 34, for glyphosate, LAS, fluoranthene and NP, respectively. Finally, the waste mixtures were humidified with 59 mL Milli-Q water for fluoranthene, NP and glyphosate and 54 mL for LAS experiment. During the first 6 d, the temperature increased by self-heating; then the temperature was modulated to mimic a typical composting temperature profile. After 41 d, composts were moved to 21-L glass cells for maturation and placed in a thermostatic room at 28 ± 1 °C for additional 42 d. The composting mixtures were sampled three times after homogenization: at the end of the thermophilic phase (day 13), at the end of the cooling phase (day 41) and at the end of the maturation phase (day 83). At each sampling date, water was sprayed on the compost to maintain the moisture content at 50–70% of wet weight. The composting mixture was characterized during another set of six composting experiments, with the same initial waste mixture and composting procedure but without addition of 14C-OPs (Lashermes et al., 2012). The evolution of the chemical and biochemical characteristics attested of a satisfactory composting process with a decrease of C:N ratio from 15.7 to 12.2 and the increase of the ratio of lignin to holocellulose from 0.4 to 0.9.

2.3. Organic pollutant mineralization, volatilization, lixiviation and speciation during composting During the first 41 d of composting, the exhaust air of each reactor was passed through a 400-mL methanol plug to trap volatile organic compounds (VOCs) and through two successive CO2 traps, each containing 750 mL 3 M NaOH. The traps of methanol and NaOH were replaced respectively 14 and 18 times and analyzed for 14C-VOC and 14C-CO2 concentrations by liquid scintillation counting (LSC) with a Tri-Carb 2100 TR counter (Perkin Elmer Ins., Courtabeuf, France) using Ultima Gold XR (Packard) as scintillation cocktail. Lixiviates were recovered at the bottom of each reactor and were pumped out after 13 and 41 d of composting. The collected volumes were measured and their 14C content was determined by LSC. During maturation (days 41–83), a vial containing 100 mL 3 M NaOH was placed in each maturation cell to trap the 14CO2 produced. These vials were replaced 14 times and the 14C-CO2 was measured by LSC. Sequential extraction was carried out in triplicate on fresh samples of the initial waste mixture and composts. An average weight of 4.8 ± 1.0 g DM of organic sample was placed in a glass centrifuge tube. The first extraction was performed with 75 mL of MilliQ water. After 24 h of shaking, water extracts were recovered by centrifugation at 2400 g for 20 min. Three successive extractions were then performed following the same procedure but with 75 mL of methanol for fluoranthene, NP and LAS, or 75 mL of 0.54 M ammonium hydroxide solution for glyphosate. The 14C-OP concentration in all extracts was measured by LSC. After extraction, the compost residues were dried at 40 °C, ground at 200 lm and the remaining non-extracted radioactivity (corresponding to the NER) was determined by scintillation counting of the 14C-CO2 evolved after combustion of the solid (Sample Oxidizer 307, Packard, Meriden, CT, USA). All 14C-fractions were expressed in percentage of total recovered 14C (sum of the 14C-activity recovered in all the fractions mineralized, volatilized, leached, extractable and NER) at a specific sampling date. The dissipation of OPs during composting was calculated (1) as the difference between the initial and final concentrations (in mg kg1 DM) of total extractable residues (addition of water and solvent extractable 14C-fractions) as usually measured in composts produced on composting plants and (2) as the difference between the initial and final amount (mg) of total extractable residues per reactor in order to assess the dissipation of OP taking into account the compost mass reduction during the process.

G. Lashermes et al. / Chemosphere 87 (2012) 137–143

2.4. Data analyses Microsoft Excel Solver was used to adjust the cumulative 14CCO2 mineralization kinetics to distinct Gompertz sigmoid equations for the thermophilic and maturation phases:

   l e PðtÞ ¼ Ai  exp  exp mi ðki  tÞ þ 1 Ai

ð1Þ

where P(t) is the total mineralized 14C-CO2 at time t, expressed as a percentage of initial 14C; Ai, the asymptote representing the maximum mineralization attained; lmi, the maximum specific growth rate; ki, the lag time (Zwietering et al., 1990), and i index equaled 1 or 2 for the active (thermophilic + cooling phases, 0–41 d) and maturation phases(41–83 d), respectively. Ninety-five percent confidence intervals were calculated considering the three replicates of composting for mineralization and speciation results. 3. Results and discussion 3.1. Overall balance of the OP behavior during composting The evolution of the 14C distribution for each OP and composting experiment among the different analyzed fractions is shown in Fig. 1. Each OP presented different behavior with a good replication between the three replicated composting experiments and with confidence intervals for all fractions generally lower than 10% of recovered 14C. The average 14C-OP recovery considering all sampling dates was (in % of initially applied 14C): 96, 95, 85, and 83 for fluoranthene, glyphosate, LAS and NP, respectively. The 14C-fluoranthene behavior showed little changes during composting maintaining a high solvent extractable fraction. In the case of glyphosate, the proportion of NER increased but a large part remained extractable with an equal distribution between the water and solvent extracts. LAS behavior was characterized by a rapid mineralization whereas the NP mineralization was balanced by a rapid formation of NER which remained at high level during the entire composting process. The volumes of leachates recovered were on average 450 mL, 820 mL, 620 mL and 520 mL for fluoranthene, glyphosate, LAS and NP experiments, respectively. The highest 14C-lixiviated fractions recovered were found for the glyphosate and LAS, representing 5 ± 3% of recovered 14C for both OPs after 41 d of composting. The 14 C lixiviation was negligible for fluoranthene and NP. Volatilization was only detected once for LAS and accounted for 5% of recovered 14 C. This was unexpected since LAS is not a volatile surfactant. 3.2. OP dissipation through mineralization during the composting process Composting can reduce OP contamination of an initial waste mixture through its complete mineralization as observed for PAHs and pesticides (Michel et al., 1995; Hartlieb et al., 2003). In the present experiment, the OP mineralization at the end of composting reached (in % of applied 14C): 0.3 ± 0.2 (95% confidence interval) for fluoranthene, 24 ± 8 for glyphosate, 29 ± 2 for NP and 51 ± 4 for LAS (Fig. 1). The mineralization of glyphosate, LAS and NP mostly occurred during the most biologically active composting period (days 0–41) (Ai, Table 1). During maturation (days 41–83) much less mineralization occurred (on average, 3%, 6% and 1% of initial 14 C for glyphosate, LAS and NP, respectively). A lag phase was observed in the mineralization of LAS, glyphosate and NP at the beginning of composting, corresponding to the time required for the degrading microorganisms to reach an efficient level (Dörfler et al., 1996). The mineralization kinetics were

139

well described by the Gompertz equation (R2 > 0.98) (Table 1). The first lag phase (k1) was on average 3, 5, 8 d for glyphosate, NP, and LAS, respectively, corresponding to the self-heating period during the thermophilic step. Then mineralization rate of these OPs reached the highest values when the temperature increased until 60 °C on days 8–11. During the maturation phase, after homogenization and rewetting, a second lag phase (k2) was on average 7 and 8 d for LAS and glyphosate, respectively; while the mineralization of NP remained very low and nearly linear. The lag phase for NP mineralization was significantly and negatively correlated with the maximum temperature observed during composting (P > 0.001). The microorganisms responsible for NP mineralization may indeed be less efficient at thermophilic than at mesophilic temperatures (Moeller and Reeh, 2003a). The variability observed in glyphosate mineralization kinetics could be explained by temperature and moisture limitations which occasionally occurred because of aeration and high temperatures in some reactors. Indeed, the asymptotes of mineralization kinetics (A1 of Gompertz equation, Table 1) were negatively correlated with the moisture content of the compost sampled after 13 d (P > 0.001) and 41 d (P > 0.01) and positively correlated with the mean temperature during the first 6 d (P > 0.001). The proportions of 14C-OP mineralized at the end of composting in the present study decreased in the same order as found during incubation with composts sampled at different composting stages (Lashermes et al., 2010): LAS > NP > glyphosate > fluroanthene. However, the extent of 14C-OP mineralization was lower during composting than compost incubations. In particular, NP and fluoranthene showed respectively a high potential of mineralization during incubation with compost sampled during the cooling phase (56% of initial 14C) and with mature compost (21% of initial 14C). For fluoranthene, the differences may be explained by a duration of composting (83 d) not long enough to allow the compost colonization by microorganisms capable of degrading aromatic structures, such as white-rot fungi (Hammel, 1995) or prokaryotes (Haderlein et al., 2006) which could have developed during the additional incubation time (92 d). Moreover, the mineralization of NP could have been limited by the high temperature during the thermophilic phase while the mineralization of NP in incubation took place at a lower temperature (28 °C) probably closer to the optimal conditions for NP degradation. In soils, the potentials of mineralization also decrease in the same order among the four OPs. However, the extent of mineralization during composting was always lower than found in soils: for LAS (Dörfler et al., 1996), NP (Gejlsbjerg et al., 2003) or glyphosate (Mamy et al., 2005), while the mineralization of 14C-fluoranthene can reach 25% of initial 14C in agricultural soil (Vessigaud et al., 2007). The presence of specific microflora, in particular for fluoranthene, and differences in sorption on soil and compost matrix limiting the fraction available for degradation can explain the difference. 3.3. OP stabilization as non extractable residues during composting The NER formation participates in the OP apparent dissipation estimated through classical OP extraction. The NER were detected as soon as NP was applied (35% of 14C recovered at the beginning of the composting process), whereas the NER proportions at the beginning of composting reached only 7%, 4% and 3% for fluoranthene, LAS and glyphosate, respectively (Fig. 1). The NER formation mainly occurred during the early phases of composting, as previously observed for pyrene by Hartlieb et al. (2003). The NER from NP tended to slightly decrease during maturation, with a decrease of 4% of recovered 14C. At the end of composting, NER proportions represented (in % of recovered 14C): 45 ± 2 for NP, 37 ± 3 for glyphosate, 24 ± 7 for fluoranthene, and 14 ± 3 for LAS (Fig. 1).

140

G. Lashermes et al. / Chemosphere 87 (2012) 137–143

Fig. 1. Variation during composting of the distribution (as a % of recovered 14C) of mineralized, volatilized, leached, water-extractable, solvent-extractable and nonextractable residues, from 14C-labeled glyphosate, LAS, fluoranthene and NP applied at the beginning of the three replicated composting experiments GLY-1 to GLY-3, LAS-1 to LAS-3, FLT-1 to FLT-3 and NP-1 to NP-3, respectively. The percentage of recovered 14C corresponded to the 14C-activity in a given fraction compared with the sum of the 14Cactivity recovered in all the fractions at a specific sampling date. Bar errors represent the confidence intervals of analytical replicates (n = 3).

The NER formation during composting has been less studied than in soil but the mechanisms described for NER formation in soils are likely to be transposable. The NER can be formed through physical entrapment in the nanoporosity of the humic compounds

or through chemical stabilization with the establishment of crosslinking reactions with compounds in the organic matrix (Kästner et al., 1999; Barriuso et al., 2008). Non humified organic matter may display great affinity to form NER with some OPs (Barriuso

G. Lashermes et al. / Chemosphere 87 (2012) 137–143 Table 1 Values of Ai, lmi and ki parameters obtained by fitting the 14C-CO2 evolution kinetics to the Gompertz equation on the 0–41 (i = 1) and 41–83 (i = 2) day periods for glyphosate, LAS and NP composting experiments. OP

0–41 d period

GLY LAS NP

% of initial 21 ± 8a 45 ± 10 28 ± 2

14

C

41–83 d period A2 GLY LAS NP a b

% of initial 3±1 6±5 1±0

k1

R2

d 1.4 ± 0.7a 5.8 ± 1.4 4.3 ± 1.7

d 3.1 ± 0.4a 8.0 ± 3.0 4.7 ± 2.7

0.998 ± 0.001b 0.997 ± 0.001 0.998 ± 0.003

lm2

k2

R2

d 8.2 ± 2.1 7.4 ± 4.1 0.0 ± 0

0.996 ± 0.004 0.999 ± 0.001 0.983 ± 0.005

lm1

A1

14

C

1

1

d 0.1 ± 0.0 0.3 ± 0.3 0.1 ± 0.0

Mean ± 95% confidence intervals of three composting replicates. Coefficient of determination of linear regression.

et al., 2008). Initial waste mixtures have been found to have higher sorption capacities than mature compost (Lashermes et al., 2010). The organic matter may be more reactive at the beginning of composting than in mature compost possibly because of higher levels of aliphatic components. Moreover, the formation of NER during the early stages of composting might also be related to intense microbial activity (Benoit and Barriuso, 1997; Vessigaud et al., 2007), thus the OPs and metabolites may be incorporated into growing biomass (Barriuso et al., 2008) or linked to organic matter after oxidation reactions catalyzed by extracellular enzymes (Gevao et al., 2000). The high level of NER formation from NP throughout composting was probably related to its highly reactive phenolic structure (Gevao et al., 2000; Dec et al., 2003). Nevertheless, the immediate formation of NER after NP application could also be partly attributed to a lack of efficiency in the sequential water/methanol extraction technique employed, even if methanol has previously been shown to be an efficient extractive of NP from compost material, reaching 90% recoveries (Pakou et al., 2009). The proportion of NER with glyphosate was also high and could be explained by the chemical reactivity of functional groups such as carboxylic, amino and phosphate groups in its chemical structure (Kästner et al., 1999). The level of NER formation during composting with fluoranthene was relatively low and similar to that found for pyrene (Hartlieb et al., 2003). The fused aromatic ring structure of fluoranthene (without any reactive functional groups) has a low potential for coupling with organic matter. And finally, the low NER with LAS revealed the weak reactivity of the molecule to binding with organic matter. The formation of NER leads to a decrease in the toxicity and bioavailability of OPs (Barriuso et al., 2008). However, the formed NER can be subsequently released through microbial degradation, being then potentially available. The environmental impact of NER and their ecological significance thus depends on the reversibility of their stabilization (Gevao et al., 2000). 3.4. OP availability and estimation of dissipation parameters for risk assessment The OP fraction easily available for the degrading microflora or for leaching is usually assessed by water extraction (Benoit and Barriuso, 1997), whereas the use of other solvents can estimate the overall potentially available residues. The water extractable fractions were related to the water solubility of the different OPs (glyphosate > LAS > NP > fluoranthene). This fraction for all OPs decreased during the more active thermophilic step of composting, remaining at relatively constant levels thereafter. The 14C-fluo-

141

ranthene was characterized by a high solvent extractable fraction corresponding to 72 ± 8% of recovered 14C at the end of composting; thus, fluoranthene could be potentially available in soil after compost application. For glyphosate, 47% and 50% of recovered 14 C were in the water and solvent extractable fractions at the beginning of composting, respectively, and only 11 ± 5 and 18 ± 5% at the end of composting. The proportion of 14C-LAS and 14 C-NP in the water extracts were low throughout composting with 9 ± 1% of recovered 14C for LAS, and 5 ± 1% for NP at the end of composting (Fig. 1). The solvent extractable fractions of LAS and NP were very high at the beginning of composting (respectively 89% and 60% of recovered 14C) and decreased quickly to 6 ± 2% and 7 ± 1% at the end of composting. Concerning fluoranthene, due to its low solubility and high hydrophobicity, its water extractable fraction remained very low during composting that could explain its poor availability for degrading microorganisms (Katayama et al., 2010). At the end of composting, the proportion of water soluble residues considered to be directly available was generally small, suggesting a low risk linked to direct OP assimilation by plants or transfer to soil water (Benoit and Barriuso, 1997). For glyphosate, the water extractable fraction was higher during composting than in soil (Mamy et al., 2005). Thus, glyphosate mineralization during composting was probably not limited by sorption on compost organic matter, but rather related to a less active glyphosate degrading microflora in compost than in soil. On the contrary, although LAS and NP are soluble compounds, the small quantities of water extractable residues may have limited the extent of their mineralization. The dissipation of an OP is classically calculated from the decrease in its concentration analyzed after extraction between two sampling dates. We used here the results of the water and solvent extractions to assess the dissipation kinetics of the different OPs, supposing that the measured 14C in the extracts correspond to the initial non degraded OP. This hypothesis implies an overestimation of OP persistence since probably a part of measured 14C corresponded to metabolites formed during composting. Total dissipation over 83 d of composting was calculated from the difference between the initial and final amount of extractable OP per reactor taking into account the mass loss of wastes during composting. The dissipation results expressed in % of initial amount of OP applied were: 84 ± 4 for LAS, 82 ± 4 for NP, 70 ± 10 for glyphosate and 18 ± 8 for fluoranthene. Based on dissipation results, half-life duration (in days) were approached supposing first-order kinetics: 22 ± 8 for LAS, 15 ± 4 for NP, 31 ± 19 for glyphosate and 286 ± 188 for fluoranthene. These values were in the ranges of half-lives reported in soil for fluoranthene (150–300 d, Van Brummelen et al., 1996) and LAS (7–27 d, Jensen, 1999), in biosolidamended soil for NP (7–19 d, Hseu, 2006) but were higher than the ranges reported in soil for glyphosate (1–4 d, Mamy et al., 2005). It is important to point out that very often in the literature the estimation of OP dissipation does not take into account the mass loss during composting, comparing directly the OP concentration expressed in mg kg1 DM between two sampling dates. That can induce an overestimation of the persistence. In our experiment, dry matter losses related to organic matter mineralization during composting mainly occurred during the thermophilic phase, and reached on average 41%, 43% and 52% of the initial DM after 13, 41, and 83 d of composting, respectively. As a consequence, the fluoranthene concentration (in mg kg1 DM) increased during composting meaning that the compost organic matter was degraded more rapidly than the fluoranthene disappeared. The percentages of OP dissipation found in the present study at the end of composting were within the 25th and 75th percentile ranges of the dissipation reported in the literature for LAS (Moeller

142

G. Lashermes et al. / Chemosphere 87 (2012) 137–143

compost, thus the risks related to the application of compost in agriculture. The dissipation mainly occurred during the thermophilic phase rather than during maturation. Thus, the initial thermophilic phase of composting should be carefully managed for feedstock contaminated with these OPs.

GLY

0 reference

NP

4 references n=7

Acknowledgments

LAS

4 references n = 12

FLT

7 references n = 19

-100

-50

0

50

100

Percentage of dissipation (%) Dissipation (decrease in OP concentration) in the literature Dissipation (decrease in OP concentration) in this study Dissipation (decrease in OP amount) in this study Disapearance (mineralization as CO2) in this study Fig. 2. Distribution of the percentages of dissipation of fluoranthene, LAS, NP and glyphosate at the end of composting reported in the literature (box) and from this study (symbols). The dissipation percentages were calculated from the initial (Ci) and final (Cf) concentrations in extractable OP (mg kg1 DM) as ½ðC i  C f Þ=C i   100 or from the initial (Qi) and final (Qf) amount in extractable OP (mg) per reactor as ½ðQ i  Q f Þ=Q i   100. The percentages of OP mineralization measured in the present study are also shown. The boxes correspond to the 25th and 75th percentiles; the bars correspond to the 10th and 90th percentiles. The vertical continuous line in the box is the median and closed circles are the extreme experimental data reported in the literature.

and Reeh, 2003b; Sanz et al., 2006; Pakou et al., 2009) and NP (Jones and Westmoreland, 1998; Gibson et al., 2007; Das and Xia, 2008; Pakou et al., 2009) during composting whose duration varied from 14 to 60 d for LAS, and from 49 to 70 d for NP (Fig. 2). No study on the evolution of glyphosate concentrations has been found. On the other hand, composting appeared to be inefficient to significantly reduce fluoranthene contamination, even with similar duration of composting in this study than in the literature (from 50 to 112 d) (Lazzari et al., 2000; Amir et al., 2005; Oleszczuk, 2006; Brändli et al., 2007; Oleszczuk, 2007; Hafidi et al., 2008; Hua et al., 2008). 4. Conclusion The dissipation pathways of 14C-labeled fluoranthene, NP, LAS and glyphosate and their availability in final composts were assessed during the composting of sewage sludge and green waste. Dissipation mainly occurred during the early stages of composting during which both mineralization and NER formation mainly occurred while both volatilization and lixiviation were negligible. The dissipation of LAS was largely due to mineralization. For NP and glyphosate, both mineralization and NER formation equally contributed to dissipation. The low dissipation of fluoranthene was only related to NER formation. In all cases, OP availability, estimated by the proportion of water soluble 14C-residues, was low in the final compost, whereas the proportion of solvent extractable OPs considered to be potentially available, was very small for NP and LAS, and intermediate for glyphosate. Composting appeared to be ineffective at degrading fluoranthene and decreasing its availability, with no mineralization and most 14C-fluoranthene remaining solvent extractable in the final compost and therefore potentially available. Such procedure to evaluate the evolution of OP availability during composting could be used to optimize the process and determine the best conditions to reduce OP availability in final

We would like to thank ADEME (French Environment and Energy Management Agency), INRA (French National Institute for Agricultural Research) for the grant, and Veolia Environment, Research and Development, for their financial support for these experiments. We thank Valérie Bergheaud and Valérie Dumeny for their assistance with experiments using 14C-labeled pollutants and Maelenn Le Villio-Poitrenaud for her constructive comments. References Amir, S., Hafidi, M., Merlina, G., Hamdi, H., Revel, J.C., 2005. Fate of polycyclic aromatic hydrocarbons during composting of lagooning sewage sludge. Chemosphere 58, 449–458. Barriuso, E., Benoit, P., Dubus, I.G., 2008. Formation of pesticide nonextractable (bound) residues in soil: magnitude, controlling factors and reversibility. Environ. Sci. Technol. 42, 1845–1854. Benoit, P., Barriuso, E., 1997. Fate of C-14-ring-labeled 2,4-D,2,4-dichlorophenol and 4-chlorophenol during straw composting. Biol. Fert. Soils 25, 53–59. Brändli, R.C., Bucheli, T.D., Kupper, T., Furrer, R., Stadelmann, F.X., Tarradellas, J., 2005. Persistent organic pollutants in source-separated compost and its feedstock materials – a review of field studies. J. Environ. Qual. 34, 735–760. Brändli, R.C., Bucheli, T.D., Kupper, T., Mayer, J., Stadelmann, F.X., Tarradellas, J., 2007. Fate of PCBs, PAHs and their source characteristic ratios during composting and digestion of source-separated organic waste in full-scale plants. Environ. Pollut. 148, 520–528. Büyüksönmez, F., Rynk, R., Hess, T.F., Bechinski, E., 2000. Occurrence, degradation and fate of pesticides during composting part II: occurrence and fate of pesticides in compost and composting systems. Compos. Sci. Util. 8, 61–81. Das, K.C., Xia, K., 2008. Transformation of 4-nonylphenol isomers during biosolids composting. Chemosphere 70, 761–768. Dec, J., Haider, K., Bollag, J.M., 2003. Release of substituents from phenolic compounds during oxidative coupling reactions. Chemosphere 52, 549–556. Dörfler, U., Haala, R., Matthies, M., Scheunert, I., 1996. Mineralization kinetics of chemicals in soils in relation to environmental conditions. Ecotoxicol. Environ. Saf. 34, 216–222. European Commission, 2000. Working Document – Organic Contaminants in Sewage Sludge for Agricultural Use. 3rd Draft, pp. 73. European Commission, 2001. Working Document – Biological Treatment of Biowaste, 2nd Draft, pp. 22. Gejlsbjerg, B., Madsen, T., Andersen, T.T., 2003. Comparison of biodegradation of surfactants in soils and sludge-soil mixtures by use of C-14-labelled compounds and automated respirometry. Chemosphere 50, 321–331. Gevao, B., Semple, K.T., Jones, K.C., 2000. Bound pesticide residues in soils: a review. Environ. Pollut. 108, 3–14. Gibson, R.W., Wang, M.J., Padgett, E., Lopez-Real, J.M., Beck, A.J., 2007. Impact of drying and composting procedures on the concentrations of 4-nonylphenols, di(2-ethylhexyl)phthalate and polychlorinated biphenyls in anaerobically digested sewage sludge. Chemosphere 68, 1352–1358. Haderlein, A., Legros, R., Ramsay, B.A., 2006. Pyrene mineralization capacity increases with compost maturity. Biodegradation 17, 293–302. Hafidi, M., Amir, S., Jouraiphy, A., Winterton, P., El Gharous, M., Merlina, G., Revel, J.C., 2008. Fate of polycyclic aromatic hydrocarbons during composting of activated sewage sludge with green waste. Bioresour. Technol. 99, 8819–8823. Hammel, K.E., 1995. Mechanisms for polycyclic aromatic hydrocarbon degradation by ligninolytic fungi. Environ. Health Perspect. 103, 41–43. Harrison, E.Z., Oakes, S.R., Hysell, M., Hay, A., 2006. Organic chemicals in sewage sludges. Sci. Total Environ. 367, 481–497. Hartlieb, N., Erturic, T., Schaeffer, A., Klein, W., 2003. Mineralization, metabolism and formation of non-extractable residues of C-14-labelled organic contaminants during pilot-scale composting of municipal biowaste. Environ. Pollut. 126, 83–91. Hogg, D., Barth, J., Favoino, E., Centemero, M., Caimi, V., Amlinger, F., Devliegher, W., Brinton, W., Antler, S. 2002. Comparison of Compost Standards within the EU, North America and Australasia. Main Report. The Waste and Resources Action Program (WRAP), Banbury UK. Hseu, Z.Y., 2006. Response of microbial activities in two contrasting soils to 4nonylphenol treated with biosolids. Chemosphere 64, 1769–1776. Hua, L., Wu, W.X., Liu, Y.X., Chen, Y.X., McBride, M.B., 2008. Effect of composting on polycyclic aromatic hydrocarbons removal in sewage sludge. Water Air Soil Pollut. 193, 259–267. Jensen, J., 1999. Fate and effects of linear alkylbenzene sulphonates (LAS) in the terrestrial environment. Sci. Total Environ. 226, 93–111.

G. Lashermes et al. / Chemosphere 87 (2012) 137–143 Jones, F.W., Westmoreland, D.J., 1998. Degradation of nonylphenol ethoxylates during the composting of sludges from wool scour effluents. Environ. Sci. Technol. 32, 2623–2627. Kästner, M., Streibich, S., Beyrer, M., Richnow, H.H., Fritsche, W., 1999. Formation of bound residues during microbial degradation of [C-14] anthracene in soil. Appl. Environ. Microbiol. 65, 1834–1842. Katayama, A., Bhula, R., Burns, G.R., Carazo, E., Felsot, A., Hamilton, D., Harris, C., Kim, Y.-H., Kleter, G., Koerdel, W., Linders, J., Peijnenburg, J.G.M.W., Sabljic, A., Stephenson, R.G., Racke, D.K., Rubin, B., Tanaka, K., Unsworth, J., Wauchope, R.D., 2010. Bioavailability of xenobiotics in the soil environment. Rev. Environ. Contam. Toxicol. 203, 1–86. Kupper, T., Bucheli, T.D., Brändli, R.C., Ortelli, D., Edder, P., 2008. Dissipation of pesticides during composting and anaerobic digestion of source-separated organic waste at full-scale plants. Bioresour. Technol. 99, 7988–7994. Lashermes, G., Houot, S., Barriuso, E., 2010. Sorption and mineralization of organic pollutants during different stages of composting. Chemosphere 79, 455–462. Lashermes, G., Barriuso, E., Le Villio-Poitrenaud, M., Houot, S., 2012. Composting in small laboratory pilots: performance and reproducibility. Waste Manage. 32, 271–277. Lazzari, L., Sperni, L., Bertin, P., Pavoni, B., 2000. Correlation between inorganic (heavy metals) and organic (PCBs and PAHs) micropollutant concentrations during sewage sludge composting processes. Chemosphere 41, 427–435. Mamy, L., Barriuso, E., Gabrielle, B., 2005. Environmental fate of herbicides trifluralin, metazachlor, metamitron and sulcotrione compared with that of glyphosate, a substitute broad spectrum herbicide for different glyphosateresistant crops. Pest Manage. Sci. 61, 905–916. Michel, F.C., Reddy, C.A., Forney, L.J., 1995. Microbial-degradation and humification of the lawn care pesticide 2,4-dichlorophenoxyacetic acid during the composting of yard trimmings. Appl. Environ. Microbiol. 61, 2566–2571.

143

Moeller, J., Reeh, U., 2003a. Degradation of nonylphenol ethoxylates (NPE) in sewage sludge and source separated municipal solid waste under bench-scale composting conditions. Bull. Environ. Contam. Toxicol. 70, 248–254. Moeller, J., Reeh, U., 2003b. Degradation of DEHP, PAHs and LAS in source separated MSW and sewage sludge during composting. Compost Sci. Util. 11, 370–378. Oleszczuk, P., 2006. Influence of different bulking agents on the disappearance of polycyclic aromatic hydrocarbons (PAHs) during sewage sludge composting. Water Air Soil Pollut. 175, 15–32. Oleszczuk, P., 2007. Changes of polycyclic aromatic hydrocarbons during composting of sewage sludges with chosen physico-chemical properties and PAHs content. Chemosphere 67, 582–591. Pakou, C., Kornaros, M., Stamatelatou, K., Lyberatos, G., 2009. On the fate of LAS, NPEOs and DEHP in municipal sewage sludge during composting. Bioresour. Technol. 100, 1634–1642. PHYSPROP database, Syracuse Research Corporation. . Racke, K.D., Frink, C.R., 1989. Fate of organic contaminants during sewage sludge composting. Bull. Environ. Contam. Toxicol. 42, 526–533. Sanz, E., Prats, D., Rodriguez, M., Camacho, A., 2006. Effect of temperature and organic nutrients on the biodegradation of linear alkylbenzene sulfonate (LAS) during the composting of anaerobically digested sludge from a wastewater treatment plant. Waste Manage. 26, 1237–1245. Van Brummelen, T.C., Verweij, R.A., Wedzinga, S.A., Van Gestel, C.A.M., 1996. Enrichment of polycyclic aromatic hydrocarbons in forest soils near a blast furnace plant. Chemosphere 32, 293–314. Vessigaud, S., Perrin-Ganier, C., Belkessam, L., Denys, S., Schiavon, M., 2007. Direct link between fluoranthene biodegradation and the mobility and sequestration of its residues during aging. J. Environ. Qual. 36, 1412–1419. Zwietering, M.H., Dekoos, J.T., Hasenack, B.E., Dewit, J.C., Vantriet, K., 1990. Modeling of bacterial-growth as a function of temperature. Appl. Environ. Microbiol. 57, 1094–1101.