Characterization of polycyclic aromatic hydrocarbons associated with PM10 emitted from the largest composting facility in the Middle East

Abstract This work reports a characterization of PAHs-PM10, including associated health effects in the largest composting facility in Tehran, Iran. Measured PAHs-PM10 stemmed primarily from petrogenic sources with mean concentrations between 231.19 to 401.25 ng m−3. The distribution pattern of PAHs (in terms of ring number) exhibited the following order when combining all sites: 3 > 4 > 5 > 6 > 2 rings. The average cumulative excess lifetime cancer risk values estimated for PAHs-PM10 surpassed the U.S. EPA limit (1 × 10−6): refining site (3.31 × 10−4) > processing site (1.75 × 10−4) > aeration site (9.81 × 10−5).


Introduction
Rapid urbanization and industrialization and alteration in peoples' habits generate higher amounts of solid wastes, such as municipal and industrial solid waste (MSW and ISW) (Norouzian Baghani et al. 2016, Putthakasem et al. 2018, Ramachandra et al. 2018. As a result of such waste generation, there are downstream impacts on emissions of polycyclic aromatic hydrocarbon species (PAHs), particulate matter (PM), volatile organic compounds (VOCs), and bioaerosols from different sites such as landfill and composting facilities, which discharge these pollutants into the ambient air (Abdel-Shafy and Mansour 2016, Dat andChang 2017, S anchez-Monedero et al. 2018).
For example, Richard (1992) and He et al. (1992) expressed that polycyclic aromatic hydrocarbons (PAHs) can volatilize into the air during the composting process and some can remain in the compost. Although composting facilities can be a suitable and cost-effective waste management option for waste disposal in high volumes, there are still concerns about their adverse impacts for workers (Viegas et al. 2014a). For example, these facilities are recognized as the main sources of particulate matter (PM) that bind with PAHs in industrial and urban areas worldwide (Byeon et al. 2008, Song and Li 2014, Viegas et al. 2014b, Hadei et al. 2017. Br€ andli et al. (2007) stated that composting is an important waste management strategy, while the resulting products can contain significant amounts of organic pollutants such as PAHs (Br€ andli et al. 2007). In addition, compost production processes provide a good environment for the creation of PM in considerable quantities and different sizes during the decomposition of organic matter (Byeon et al. 2008, Hadei et al. 2017. PM emitted from different stages of compost production stemming from diverse wastes harbor dangerous chemical substances such as PAHs (Song andLi 2014, Viegas et al. 2014b).
PAHs are of high toxicity in atmosphere, generally caused by combustion of natural gases, wastes, and landfills (Rushton 2003, Verma et al. 2016, Balogun-Birro 2015. Melnyk et al. (2015) conducted an investigation of concentrations and sources of PAHs in municipal solid waste (MSW) landfills in Gdansk in northern Poland; they found that the total PAH concentration ranged from 892 to 3514 ng m À3 . Furthermore, Zhou et al. analyzed the content of 16 PAHs from 24 soil samples near Sheepcote Valleylandfill in the United Kingdom, and reported that the PAH concentrations varied from 1611 to 66.089 lg kg À1 (Zhou et al. 2014). Additionally, Lou et al. (2016) found that the concentrations of 16 PAHs at a landfill site in Shanghai (China) were between 4103 and 19 ng m À3 .
In particular, exposure to PM 2.5-bound PAHs or dust-bound PAHs can lead to diseases such as jaundice, malformations, cataracts, kidney damage, tumors, interfere with hormone systems, inflammation and redness of the skin, liver damage, DNA damage, testicular lesions, behavioral problem, hormonal imbalance, and cancers such as those of the skin, lung, bladder, and gastrointestinal region (Perera et al. 2011, Ali et al. 2016, Di Vaio et al. 2016, Nazmara et al. 2020.
The concentration ratio between individual PAHs can be used as an identifier tool for emission sources of PAHs (Tobiszewski and Namie snik 2012, Tue et al. 2014, Oliveira et al. 2015, Błaszczyk et al. 2017. Pyrogenic and petrogenic sources are regarded as the two main sources of anthropogenic PAHs emissions in the environment (Qi et al. 2014, Ali et al. 2016, Iwegbue et al. 2018. For instance, for distinguishing between pyrogenic, petrogenic, and other sources of PAHs, Flu/(Flu þ Pyr), IcdP/(IcdP þ BghiP), and Ant/ (Ant þ Phe) ratios are generally applied as an identifier of PAH emission sources (Peng et al. 2011, Kamal et al. 2014, Gope et al. 2018). In addition, the IcdP/ (IcdP þ BghiP) versus Flu/(Flu þ Pyr) ratio and the Ant/ (Ant þ Phe) versus Flu/(Flu þ Pyr) ratio are often used as markers for pyrogenic and petrogenic sources (Peng et al. 2011, Kamal et al. 2014, Bao et al. 2018, Gope et al. 2018. For example, previous works reported that if the ratio of Flu/(Flu þ Pyr) is more than 0.40, less than 0.5, or between 0.40 and 0.50, it is suggestive of petroleum/petrogenic sources, pyrogenic sources, or traffic emission/incomplete combustion of fuels, respectively (Menezes and Cardeal 2012, Tobiszewski and Namie snik 2012, Zhang et al. 2016.
In addition, according to previous studies, identification of low-molecular weight (LMW) and highmolecular weight (HMW) PAHs can be used for indicating whether those pollutants (LMW-PAHs and HMW-PAHs) tend to be associated with the gas phase or the particulate phase (Iwegbue et al. 2018, Nazmara et al. 2020. For instance, previous studies indicated that LMW-PAHs and HMW-PAHs tend to be associated with the gas phase and the particulate phase, respectively (Iwegbue et al. 2018, Nazmara et al. 2020. Arad Kouh Waste Process and Disposal Complex (AKWPDC) is located in southern Tehran, Iran. A daily average of more than 7640 tons of waste enters AKWPDC from the waste of 22 distinct geographic regions of Tehran (Iran), and it is also recognized as the largest composting process production site in the Middle East covering about 22 ha. The composting site has operated with a mean processing capacity of more than 82 tons of wastes per day from 2007. To date, there has been no report on the exposure of workers to PAH species associated with PM 10 in different sampling locations in this composting facility in Tehran, Iran. More broadly, little is known about composting facility emissions regardless of region. Domingo and Nadal (2009) suggested that environmental monitoring of PAHs and VOCs in the composting facilities is critically important with regard to the well-being of workers.
The goal of this work is multi-fold: (1) report on the concentration of PM 10 and PAH species associated with PM 10 (PAHs-PM 10 ) at AKWPDC in Tehran, Iran; (2) identify the percentage of PAH species, develop a spatial distribution map of PAH species, investigate isomeric (diagnostic) ratios for PAHs, study lowmolecular weight (LMW) and high-molecular weight (HMW) PAHs at different sampling points; (3) conduct health risk assessment (HRA) (carcinogenic and noncarcinogenic) for different age groups between 19 and 60 years who are exposed to PAHs-PM 10 and report on the ratio of carcinogenic PAHs to RPAH (RPAH carcinogenic /RPAH); and (4) conduct a statistical analysis of concentration of PM 10 and PAH species associated with PM 10 at different sampling points. This work is novel in that there are no reports to our knowledge of the nature and impacts of PAHs-PM 10 for outdoor environments associated with a compost facility. The results of this work have broad implications for other regions owing to the pervasiveness of composting facilities and PAHs species associated with PM 10 .

Description of sampling sites
The Arad Kouh Waste Process and Disposal Complex (AKWPDC) is in southern Tehran (35.4665 N,51.3404 E, 530 m above sea level), Iran ( Figure 1). More detailed information about description of sampling sites are provided in the supporting information (Supplementary Section S1 and Figure S1).

PM10 sampling
Samples of PM 10 were collected every three days in the winter (February 1 2018 to March 31 2018) during work shifts at the workers' breathing zone (at a height of about 1.5-2 m above the ground). At the three sites, a total of 60 PM 10 samples were collected between February and March. The Omni Ambient Air Sampler (BGI Incorporated, Waltham, MA, USA) was used for sampling, which was used with 47 mm polytetrafluoroethylene (PTFE) substrates having a pore size of 0.5 mm (Whatman, Clifton, NJ, USA). Air sampling was conducted eight hours per day eight hours at a flow rate of 5 L min À1 (Supplementary Figure S2). Before and after sampling, substrates were kept in a desiccator for 24 h. After sampling, substrates were put into Inhibitory Mold Agar (IMA) plates and guarded from light using aluminum foil and kept at 4 C with a portable plastic cooler box. Finally, collected samples were kept at À20 C until they were digested and extracted (Sharma et al. 2016, Goudarzi et al. 2017. Meteorological conditions including temperature ( C), relative humidity (%) and wind speed (ms À1 ) were also simultaneously recorded by a portable instrument (Preservation Equipment Ltd, UK and Campbell Scientific, Inc., USA).

Sample extraction and PAH analysis
PAHs in each sample were extracted based on NIOSH 5515 method (NIOSH 1994, Chuang 1996. For extracting and analyzing PAHs, the substrates were picked up from the freezer and held in the desiccator for 24 h. Then, half of each substrate was cut into very fine pieces and poured into 5 ml vials. Subsequently, 5 ml of methanol and 5 ml of dichloromethane (HPLCgrade) were added to each vial containing sample. The vials were subsequently shaken via ultrasonication for more than 30 min. Afterwards, the samples were filtered through a 0.22 mm syringe filter (PTFE, Whatman Schleicher & Schuell) to separate suspended particles.
Each extracted sample was dried using a stream of a slow purified nitrogen (5-10 psi, steadily rising with vaporization), resulting in approximately 1 ml for analysis. Finally, a 2 ml mixture of dichloromethane/ methanol with a 1:1 ratio was added to each dried samples. (Ahmed et al. 2013, Fabia nska et al. 2016, Jalili et al. 2019. Sixteen priority PAHs were speciated using gas chromatography-mass spectrometry (GC-MS) (GC 7890 N, AGILENT-MS 5975 C, MODE EI.MS) in the mode of selective ion-monitoring (SIM). A fused silica capillary column (DB5/MS, 30 m Â 0.25 mm Â 0.5 lm) was applied for separation of PAHs. The injection volume and injection technique were two lL and splitless mode, respectively. Helium was injected as the carrier gas at a flow rate 1 ml min À1 . The initial temperature of column was programed for 60 C and then raised at a rate of 10 C per minute to 100 C (held for one minute). Subsequently, temperature was increased to 285 C at a rate of 4 C per minute (held for fifteen minutes).

Quality control (QA/QC)
All glassware used for extraction steps were cleaned via ultra-sound, soaked in 20% nitric acid (overnight soaking), and washed with distilled water prior to being heated in an oven (5 h at 180 C) to volatilize and remove any organic contaminants. Standard reference materials (SRM; Urban Dust 1649 b, USA) provided by the National Institute of Standards and Technology (NIST) were employed for PAH analysis. For all validation, quantified concentrations of each PAH species were usually about 25% of the SRM-NIST value, and for the total of PAHs was about 5% of the SRM-NIST values. For QC, 18 background samples were gathered after every five samples from upwind rural areas near of AKWPDC. These background filters were obtained and treated similarly to the true samples for the purpose of characterizing background contamination levels. The concentration of PAHs and PM 10 for each measurement was determined by subtracting the concentration of the background control samples from that in the true samples. It should be noted that determination of limit of detection (LOD) is useful considering that the levels of PAHs in various matrices are present in low concentrations. The LOD for each compound was calculated three times, which was taken as the standard deviation of the lowest level standard. The recovery test was also performed by spiking known amounts of the standard mixture of 16 PAHs onto the filters and performing and using surrogate standards added to the samples undergoing the same analytical process. In addition, the limit of quantification (LOQ) was also quantified as the LOD divided by the sampling volume. Recoveries of PTFE filters spiked with the mixed PAH standard were 80-104%. Furthermore, a five-point calibration curve  ranging from 2 to 150 ng m À3 was applied for quantification of the target components (PAHs) with coefficients of determination (R 2 ) higher than 0.997. In addition, the LOD and LOQ of target components ranged from 0.009 to 0.1 ng g À1 and 0.032 to 0.348 ng g À1 , respectively. Table 1 reports the LOD and LOQ of the target PAHs and chosen parameter values utilized for calculation of carcinogenic equivalents (TEQ) including carcinogenic equivalency factor (TEF), and the carcinogenic classification of PAH components in IARC.

Statistical analysis
Analysis was performed by the statistical program R (version 3.0.1, May 16 2013) (Team 2013). The Fligner-Killeen test was applied to assess for homogeneity of variance. If the p values obtained from the Fligner-Killeen test exceeded 0.05, the ANOVA test was performed for further analysis. If the p values were less than 0.05, the Kruskal-Wallis test was applied for further analysis. Furthermore, if the Kruskal-Wallis test was significant, the Kruskalmac post hoc analysis was carried out to show that levels of the independent variable vary from other levels. Various ratios (IcdP/ (IcdP þ BghiP), Flu/(Flu þ Pyr), Ant/(Ant þ Phe), and LMW-PAHs/HMW-PAHs) were calculated in order to infer emission sources of target components. Furthermore, the Monte Carlo Simulation (MCS) model in the Crystal Ball software (version 11.1. USA, Inc.) (Firestone et al. 1997) was used for HRA calculations of PAHs associated with PM 10 for workers at the study site.

Spatial distributions
The Arcgis 10.4.1 software package was applied for spatial distribution analysis of PAH species associated with PM 10 at the study site.

Health risk assessment (HRA)
A HRA is essential to quantify the level of threat to human health from exposure to hazardous pollutants such as PAHs, which can create cancer or other adverse health effects (Mohajer et al. 2020, Nabizadeh et al. 2020a In this study, it was important to compute the non-carcinogenic and carcinogenic health effects of target components for workers at the compost site of AKWPDC (age range: 19 to 60 years) , Dehghani et al. 2019.  (1)), each PAH concentration was multiplied by a carcinogenic equivalency factor (TEF) ( Table 1) for cancer concerning BaP (Qiao et al. 2006, Iwegbue et al. 2018).

Local meteorological conditions of study area
The wind direction during sampling from February to March 2018 in the study area was toward the west, with slight inclination to the southwest. In addition, the temperature, relative humidity, and wind speed measured during sampling in working hours ranged from 10 to 15 C, 40 to 60%, and 1 to 4 ms À1 , respectively.

Concentration of PM 10 at different sampling sites
The mean ± standard deviation (SD) values of 16 PAH species associated with PM 10 at different sampling sites (processing, aeration, and refining site) are summarized in Supplementary Table S2. The average concentrations of PM 10 at processing, aeration, and refining site were 1291 ± 363, 143 ± 34, and 3557 ± 980 mg m À3 , respectively. The highest average concentration of PM 10 was obtained at the refining site likely due to the height above the rotary screen (at a height of about 4 m above the ground) (Supplementary Figure S2) and low humidity of fine compost/granulated compost (<6 mm) (lowerr than 35%). The lowest average concentration of PM 10 was reported at the aeration site (143 ± 34 mg m À3 ) (Figure 1; Supplementary Figures S1 and S2) since in this step coarse compost (>6 mm) is produced at high humidity (50-60%). At the aeration site, water was added for producing mature compost which limited emissions of particles. The average concentrations of PM 10 in the processing site (1291 ± 363 mg m À3 ) were higher than the aeration site (143 ± 34 mg m À3 ) due to the processing site being at low humidity and in a closed space, and the lack of proper ventilation leads to accumulation of pollutants.
For comparison, Chalvatzaki et al. (2010) reported that the concentration of PM 10 at different stages in Akrotiri landfill site in Crete (Greece) ranged from 42 to 601 mg m À3 (Chalvatzaki et al. 2010). Furthermore, Hryhorczuk et al. (2001) and Ray et al. (2005) reported that the concentration of total suspended particulate matter (TSP) in a suburban yard waste composting facility in Northern Illinois (USA) and in open landfill site in Delhi (India) ranged from 99 to 631 mg m À3 and 559 to 2082 lg m À3 (Hryhorczuk et al. 2001, Ray et al. 2005. In addition, the concentration of PM 10 in this study (143-3557 mg m À3 ) was far higher than those reported for in close to the weighing facility of the Akrotiri Landfill of Crete (Greece) (84-140 mg m À3 ) (Chalvatzaki et al. 2014), in the ambient air of the Greater Athens Area (Koropi), Greece (21.8-94.3 mg m À3 in winter versus 34.1-87.4 mg m À3 in summer) (Vassilakos et al. 2007), and in the ambient air of the Greater Athens Area (Spata), Greece (16.4-109 mg m À3 in winter versus 13.3-24.8 mg m À3 in summer) (Vassilakos et al. 2007).

Levels and profile of PAHs species associated with PM 10 in different sampling locations
The mean ± standard deviation (SD) values and the spatial distribution of 16 PAHs species associated with PM 10 in different sampling sites (processing, aeration   Downard et al. (2015) reported that total concentration of PAHs in coarse particulate matter (PM 10 ) at the University of Iowa air monitoring site near uncontrolled combustion of shredded tires in a landfill was 33.4 ng m À3 . Additionally, Dallarosa et al. (2005) reported that the mean concentrations of PAHs associated with PM 10 in ambient air of Porto Alegre (Brazil) ranged from 0.04 to 2.3 ng m À3 . In addition, Byambaa et al. (2019) described that the concentrations of PAHs associated with total suspended particles (TSP) in ambient air of Ulaanbaatar (Mongolia) in spring, autumn and winter ranged from 22.2 to 530.6, 1.4 to 54.6, and 131 to 773 ng m À3 , respectively. Moreover, Sadej and Namiotko (2010), Berset and Holzer (1995), Milo sev et al. (2007), Niederer et al. (1995), and Barker and Bryson (2002) reported that the primary source of emissions of PAHs to atmospheric air is incomplete combustion of fossil fuels, while PAHs can also originate from other sources, such as waste dumps, composting of wood materials treated with creosote or tarry and sewage sludge blends with municipal solid waste (MSW), which is consistent with the findings of our study.
Municipal solid waste (MSW) separation was not performed properly at the processing site, and as a result, non-biodegradable waste along with compostable/biodegradable waste also entered into the composting facility. On the other hand, there are types of waste that are naturally similar to industrial waste and enter into the composting facility, which can be sources of PAHs.
In addition, other anthropogenic PAHs emissions in the environment that were observed in the processing and aeration sites of this study were wood materials treated with creosote or tarry and sewage sludge. These materials included different PAHs species, especially Ant, Phe, and Pyr, which is in line with the findings of former studies (Abdel-Shafy and Mansour 2016, U.S. EPA 2008). Also, according to Canadian Environmental Protection Act (CEPA), creosote-impregnated waste materials (CIWM) such as railway ties, bridge timbers, pilings, and lumber were the main sources of PAHs species; these species can constitute up to 90% of these waste products (Canadian 1994), which is consistent with the results of our study.
A box plot of PAHs species concentration at the three sampling stations (processing, aeration, and refining site) is shown in Supplementary Figure S3. Accordingly, the highest and lowest concentrations of PAH species at the three stations were fluoranthene and fluorene, respectively. These findings are opposite of previous studies since benzo [b]fluoranthene and benzo[a]pyrene were the most abundant as compared to other PAH species in study of Arfaeinia et al. (2017) in indoor dust of Bushehr (Iran) (Arfaeinia et al. 2017) and Norramit et al. (2005) in airborne particles in the Bangkok metropolitan area (Thailand) (Norramit et al. 2005), respectively.
In addition, the percentage of 16 PAH species associated with PM 10 at the refining site was 1.30-3.05 and 0.63-2.08 times higher than the aeration site and processing site, respectively. The percentage of target pollutants associated with PM 10 in the aeration site was 0.29-1.61 times higher than the processing site. The reasons for why the percentage of 16 PAH species associated with PM 10 in the refining site was higher than other sites could be related to purification and separation of undesirable substances from coarse compost using a rotary drum screen at a height of 4 m for converting into fine compost. Concentrations of PM 10 were increased due to height above the rotary screen and low humidity of fine compost (lowerr than 35%), which resulted in an enhancement of PAH species.

Statistical analysis of concentration of PM 10 in three stations
The output of the Fligner-Killeen test showed that p values for PM 10 concentrations in different sampling stations were higher than 0.05 (p > 0.05). This indicates that the difference between the variances in different processes were not significant (p > 0.05). Therefore, parametric ANOVA test was used for further analysis. The results of the ANOVA test on PM 10 concentrations in different sampling stations were not significant at a level of 0.05.

Statistical analysis of concentration of PAHs in three stations
The output of the Fligner-Killeen test showed that the p values for PAH concentrations in different sampling stations were lower than 0.05. This indicates that the difference between the variances in different processes were significant (p < 0.05). Therefore, analysis of variance was used to compare normally distributed variables for more than two groups (Kruskal-Wallis test). The results of the Kruskal-Wallis test on PAHs concentrations in different sampling stations were significant at a level of 0.05. Consequently, the results of the Kruskalmac post hoc analysis showed that the mean concentrations of PAHs between two different sampling locations of processingaeration and aerationrefining were significantly different (p < 0.05).

Low-Molecular weight (LMW) and High-Molecular weight (HMW) PAHs
In this work, LMW-PAHs include the sum of species with two rings (Nap) and three rings (Acy, Ace, Flo, Phe, and Ant), while HMW-PAHs comprise the sum of species with four rings (Flu, Pyr, BaA, and Chr), five rings (BbF, BkF, BaP, and IcdP) and six rings (DahA and BghiP). The LMW-PAHs and HMW-PAHs are summarized in Supplementary Figure S4 based on percent (%) abundance at different sampling locations and in sum of three sites. The distribution pattern of PAHs in processing and refining site was different from aeration site due to the high presence of phenanthrene (Phe) in processing and refining site (Supplementary Figure  S4). According to Supplementary Figure S4, the concentrations of Phe and Chr were different at the three sites due to these PAHs being produced from several sources: combustion processes (e.g. incinerator and landfill sites) near the composting site; mobile sources such as front-end loaders and vehicles that push solid wastes from one site to other sites; mixing uncombusted petroleum products such as lubricants, petroleum product spills, crude oil and fuels with solid wastes at homes; plants and workshops and wood materials treated with creosote or tarry (Canadian 1994, Chen et al. 2013, Stogiannidis and Laane 2015, Nazmara et al. 2020).
In addition, Chen et al. (2013) and Stogiannidis and Laane (2015) stated that low-molecular weight (LMW) PAHs such as Phe are more inclined to move into the ambient air from solid wastes during composting, while the high-molecular weight (HMW) PAHs such as Chr are slightly inclined to get into the ambient air during composting, which is in agreement with the findings of the present study. Furthermore, previous work reported that with passing time from 1 to 3 months during composting, and from processing site to aeration site, the concentrations of high-molecular weight (HMW) PAHs increased (Sadej and Namiotko 2010), which is in line with the findings of this work.
In addition, the distribution pattern of PAHs (in terms of ring number) exhibited the following order when combining all sites: 3 > 4 > 5 > 6 > 2 rings. For comparison, the distribution pattern of the target compounds in indoor dust in semi-urban, rural and urban areas in Niger Delta (Nigeria) were, respectively, 3 > 6 > 4 > 5 > 2 rings, 5 > 6 > 4 > 3 > 2 rings and 4 > 6 > 5 > 3 > 2 rings (Iwegbue et al. 2018). In this study, HMW-PAHs (36.28-86.5%) were the more abundant PAH compounds across the three sites, while the LMW-PAHs ranged from 15.43 to 91.62%. The LMW PAHs were relatively more abundant than the HMW-PAHs in PM 10 samples from various sampling sites (processing, aeration, and refining site), which illustrates that LMW-PAHs tend to be associated with the gas phase, while HMW-PAHs tend to be associated with the particulate phase, which is in agreement with the results of the past works (Li et al. 2006, Nazmara et al. 2020). On the other hand, the LMW-PAHs constituted 15.43 to 91.62%, while the HMW-PAHs constituted 36.28-86.5% of the total PAHs in PM 10 samples from various sampling sites, that may be explained by how the gas-phase and particulate-phase are important for association of LMW-PAHs and HMW-PAHs with PM 10 , respectively, which is in line with the findings of former studies (Iwegbue et al. 2018). Hence, this work showed that the percentage of LMW-PAHs (15.43 to 91.62%) was higher than HMW-PAHs with percentage of 36.28-86.5% (of the total PAHs) in PM 10 samples from various sampling sites, which indicate that PAHs mostly stemmed from petrogenic sources and a lesser extent from pyrogenic sources.

PAHs source identification (PAHs ratios)
The ratio values of individual concentrations of PAHs can be used as an identifier tool for emission sources of PAHs (Tobiszewski and Namie snik 2012, Tue et al. 2014, Oliveira et al. 2015, Błaszczyk et al. 2017. Generally, pyrogenic and petrogenic sources are regarded as the two main sources of anthropogenic PAHs emissions in the environment (Iwegbue et al. 2018, Nazmara et al. 2020. To distinguish between pyrogenic sources, petrogenic sources, and other sources of PAHs, Flu/(Flu þ Pyr), IcdP/(IcdP þ BghiP), and Ant/(Ant þ Phe) ratios are generally applied as an identifier of PAH emission sources (Peng et al. 2011, Kamal et al. 2014, Gope et al. 2018. More detailed information about PAH source apportionment (PAHs ratios) are provided in the supporting information (Supplementary Section S2 and Tables S4 and S5).
This work showed that the ratio of Ant/(Ant þ Phe), Flu/(Flu þ Pyr) and IcdP/(IcdP þ BghiP) ranged from 0.22 to 0.29, from 0.07 to 0.49 and from 0.36 to 0.80, respectively, suggesting that PAHs mainly stemmed from petrogenic sources and to a lesser extent from pyrogenic sources, especially stationary sources such as incinerator and landfill sites near the composting site, mobile sources such as front-end loaders and vehicles that push solid wastes from one site to other sites and mixing uncombusted petroleum products such as lubricants, crude oil and fuels with solid wastes at homes, plants and workshops (Katsoyiannis and Breivik 2014, Franco et al. 2015, Stogiannidis and Laane 2015, Zhang et al. 2016, Mohseni Bandpi et al. 2017. For comparison, Bao et al. (2018) reported that the Flu/(Flu þ Pyr) and Ant/(Ant þ Phe) in urban soils from Xi'an (China) were found from 0.06 to 0.85 and 0.02 to 0.24, respectively. The same study described that PAHs species associated with urban soils were mainly derived from combustion of coal, biomass and petroleum (Bao et al. 2018). In addition, Gope et al. (2018) and Hussain et al. (2015) based on the ratios of the Ant/(Ant þ Phen) and Fla/(Fla þ Pyr) reported that pyrogenic sources and wood/coal burning were the main sources of PAHs in the street dust of Asansol (India) and in street dust of Guwahati (India), respectively. Diagnostic ratios for identification of PAH pollution sources at the study site are depicted in Supplementary Figure S5.
In addition, the LMW-PAHs/HMW-PAHs ratio can be applied for source identification (Błaszczyk et al. 2017, Rogula-Kozłowska 2015. For instance, Błaszczyk et al. (2017) and Rogula-Kozłowska (2015) reported that ratios exceeding one indicate that petroleum is the main source of target components, while ratios less than one indicate pyrogenic sources are the main sources of PAHs. In this work, LMW-PAHs/HMW-PAHs ratios ranged between 0.47 and 0.65 for three sampling locations (Supplementary Table S2). Hence, these findings indicate that pyrogenic sources were mostly responsible for HMW-PAHs, consistent with works by Franco et al. (2015) and Qi et al. (2014).

1. Lifetime excess cancer risks (LTCRs)
To the best of our knowledge, this work is the first to recognize the potential health risk assessment of age groups between 19 to 60 years old from exposure to PAHs-PM 10 in a composting facility. Supplementary Figure S6 shows a box plot of carcinogenic equivalents (TEQ BaP ) based of different sites of AKWPDC. According to Supplementary Figure S6, the values of TEQ BaP in this work (35.55-86.99 ng m À3 ) were much higher than those introduced for dust in the United States (2.9 ng m À3 ) (Liu et al. 2017) The average (5% and 95%) cumulative excess lifetime cancer risk (CELCR) estimated for PAHs species in refining site was from 3.31 Â 10 À4 (2.29 Â 10 À4 and 4.54 Â 10 À4 ), which surpass suggested values by the U.S. EPA (1 Â 10 À6 ). In addition, the mean (5% and 95%) CELCR calculated for PAHs species in processing site was from 1.75 Â 10 À4 (1.21 Â 10 À4 and 2.41 Â 10 À4 ), which exceed suggested values by the U.S. EPA. Besides, the mean (5% and 95%) CELCR calculated for PAHs species in aeration site was from 9.81 Â 10 À5 (6.80 Â 10 À5 and 1.35 Â 10 À4 ) and in exceedance of the suggested values by the U.S. EPA. Hence, the mean cumulative excess lifetime cancer risk (CELCR) for PAHs species in different sites was as follows: refining site > processing site > aeration site. In this respect, previous study such as Tsai et al. (2001) examined additional cancer risks among carbon factory workers and estimated the value of 4.35 Â 10 À2 that was far much higher than the acceptable level (1 Â 10 À6 ). Furthermore, Huang et al. (2016) similarly conducted a study on cancer risk assessment of residents dwelling around e-waste recycling sites in China and reported that levels of cancer risk assessment for size-fractionated particle-bound heavy metals had reached 1.3 Â 10 À3 and 3.9 Â 10 À3 for adults and children, respectively, which exceeded the suggested value (1 Â 10 À6 ) by the U.S. EPA. Comparing values of additional cancer risks in regions of production with the additional risks from air pollution in urban areas such as 8.87 Â 10 À9 to 1.76 Â 10 À9 for male and females, respectively, in Curitiba (Brazil) (Froehner et al. 2011), 9.3 Â 10 À6 for in the USA habitants (Liu et al. 2017), and less than 1 Â 10 À6 in Brno (Czech Republic) (Bulejko et al. 2016), it can be concluded that the additional cancer risks of PAH compounds in industrial areas are much higher than those in urban ones. Thus, individuals employed in industrial areas are at a greater risk of developing cancer (Jamhari et al. 2014).

Non-Carcinogen or hazard quotient (HQ)
The mean (± SD) HQs of PAHs for the processing site, refining site, and aeration site were 2.57 ± 1.01, 2.43 ± 0.91, and 1.66 ± 0.65 during sampling, suggestive of the need for concern about the non-carcinogenic risk of PAHs in the study area of AKWPDC. Hence, the HQs of PAHs exceeded one, which is "not at an acceptable level."  Figure S7 and Table S2). This ratio provides insight into the potential carcinogenicity of PAHs for workers in composting facilities around world (Supplementary Table S2). As described in Supplementary Table S2, the potential carcinogenicity of PAHs in AKWPDC ranged from 0.32 to 0.40, which is lower than PM 1 -bound PAH in outdoor in Warszawa (Poland) (0.48) (Rogula-Kozłowska et al. 2018) and in PM 2.5 -PAHs of a business office area in Jinan (China) (0.92) (Zhu et al. 2015). In addition, the spatial distribution of the RPAH carcinogenic /RPAH ratio showed that the highest levels of carcinogenic species were observed in aeration site (Supplementary Figure S7). The total concentration of carcinogenic pollutants (RPAH carcinogenic ) relative to the total concentration (RPAH) in the aeration site was higher than other sites. Hence, the non-carcinogenic (hazard quotient (HQ)) and carcinogenic (cumulative excess lifetime cancer risk (CELCR)) health effects of PAH components for workers at aeration site of AKWPDC can be considerable (based on Section 3.7). According to this section, the mean (5% and 95%) CELCR calculated for PAHs species in aeration site was from 9.81 Â 10 À5 (6.80 Â 10 À5 and 1.35 Â 10 À4 ) and in exceedance of the suggested values by the U.S. EPA (1 Â 10 À6 ) (Tsai et al. 2001, Huang et al. 2016). In addition, the mean (± SD) HQs of PAHs for aeration site was 1.66 ± 0.65 during sampling (1 HQ), suggestive of the need for concern about the non-carcinogenic risk of PAHs in the study area of AKWPDC.

Conclusion
This study reports on a characterization of outdoor PAH species associated with PM 10 and their associated health effects in the largest composting facility in the Middle East within Tehran, Iran. The mean concentrations of PM 10 detected at different sites were in the following order: refining site (3557 mg m À3 ) > processing site (1291 mg m À3 ) > aeration site (143 mg m À3 ). The total PAH-PM 10 at the three different sampling locations ranged from 231.19 ± 145.94 to 401.25 ± 210.66 ng m À3 . The percentage of target pollutants associated with PM 10 in the aeration site were 0.29 to 1.61 times higher than the processing site. The results of the Kruskalmac post hoc analysis showed that the mean concentrations of PAHs-PM10 between different sampling locations were significantly different (p < 0.05). The average cumulative excess lifetime cancer risk (CELCR) estimated for PAHs-PM 10 in different sites was as follows: refining site (3.31 Â 10 À4 ) > processing site (1.75 Â 10 À4 ) > aeration site (9.81 Â 10 À5 ), which surpass suggested values by the U.S. EPA (1 Â 10 À6 ). The mean HQs of PAHs-PM 10 in different sites ranged from 1.66 to 2.57, which is "not at an acceptable level." Hence, exposure to PAHs-PM 10 at this composting facility can lead to high risk (HQ ¼ 1.66 to 2.57 and CELCR ¼ 9.81 Â 10 À5 to 3.31 Â 10 À4 ) and potential carcinogenic effects are significant. Additionally, procedures such as adjustment of the factory to enclose the conveyors, designing proper ventilation and air conditioning systems, minimization of PAH-contaminated waste generation (pre-treatment), and using personal protective equipment (PPE) such as respirators and gloves should be considered to decrease pollutants and to keep workers from the non-carcinogenic and carcinogenic effects. The results of this work motivate improved strategies for this and other composting facilities to limit the exposure of workers to emitted toxic species. Furthermore, the results of this work build on the growing evidence that PM 10 particles serve as a carrier of different types of harmful constituents that can partition to the aerosol-phase during transport of particles. This is a relevant issue for the biogeochemical cycling of environmental contaminants.