Levels, Potential Sources and Risk Assessment of Organochlorine Pesticides in Atmospheric Particulate Matter at Regional Background Site

Received: June 25, 2019
Revised: August 9, 2019
Accepted: August 10, 2019

Download Citation: RIS|BibTeX|https://doi.org/10.4209/aaqr.2019.06.0321  

Cite this article:

Liu, J., Wang, H.Y., Song, S.J., Ma, H.C., Sun, W.T., Wang, L., Wang, Y., Yi, X.L., Guo, L.Q. and Li, P.H. (2019). Levels, Potential Sources and Risk Assessment of Organochlorine Pesticides in Atmospheric Particulate Matter at Regional Background Site. Aerosol Air Qual. Res. 19: 2008-2016. https://doi.org/10.4209/aaqr.2019.06.0321

Highlights

• PM_2.5-bound OCPs levels were analyzed in regional background atmosphere.
• Regional sources and multiple sources contributed to the pollutants.
• Estimated non-carcinogenic health risk was below the U.S. EPA’s limit.

Keywords: PM2.5; Organochlorine pesticides (OCPs); Concentration; Long-range transport; Health risk.

INTRODUCTION

Organochlorine pesticides (OCPs), a group of persistent organic pollutants (POPs), were extensively used in agriculture from the 1950s to 1970s all over the world due to their low cost and the effective performance in controlling insects (Jin et al., 2013). However, due to their high toxicity, carcinogenic property and resistance to degradation, OCPs are of special concern in previous studies (Bräuner et al., 2011; Jiao et al., 2018). The application of OCPs have been restricted in most of the developed and developing countries since 1980s. However, they were still widely detected in different regions because of the previous residues (Bozlaker et al., 2009; Zhou et al., 2019). In atmosphere, OCPs mainly came from agricultural soil suspension and synthesis of chlorinated product (Chakraborty et al., 2013), and can be transported over long distances from their sources and had been found in remote locations where they have never been used or produced (Kirchner et al., 2016).

China was a major producer and consumer of OCPs in the world. Before they were prohibited in 1983, the total production of hexachlorocyclohexane (HCHs) and dichlorodiphenyltrichloroethane (DDT) in China was about 4.9 and 0.4 million tons, accounting for 33% and 20% of the total world production, respectively (Xu et al., 2011). Studies concerning OCPs compounds in atmospheric environment have been performed in different regions of China (Han et al., 2016; Wang et al., 2008). In Beijing, the levels were found to be 212 pg m⁻³ for HCHs and 522 pg m⁻³ for DDTs in 2005, and their concentrations were decreased to 101 pg m⁻³ and 146 pg m⁻³, respectively in 2009 (Zhang et al., 2011). However, through the global distillation and grasshopper effect, OCPs were detected in many background sites such as Antarctica (Dickhut et al., 2005), and the Arctic (Becker et al., 2012) where had never used OCPs. Understand the long-range transport of OCPs in atmosphere can give important information for the global-scale existence and fates of these pollutants. In the present study, fine particle (PM_2.5) samples were collected at the summit of Mount Tai, which serves as a regionally background site and a unique environment to investigate the long-range transport of pollutants in troposphere (Liu et al., 2017). OCP compounds were analyzed in collected PM_2.5 samples, and the objectives of the present study were to (1) survey the regional pollution levels of OCPs in PM_2.5, (2) identify possible sources and long-range transport of OCPs from contaminated regions, and (3) provide comprehensive fundamental data for risk assessment of OCPs in atmospheric environment. 

EXPERIMENT AND METHOD

Sampling

The sampling was carried out in the meteorological monitoring station on Mount Tai (117°06′E, 36°16′N, 1532.7 m A.S.L), located in free troposphere, is one of the highest peaks in northern China. A total of 48 PM_2.5 samples were collected from June 15 to August 11, 2015, using a high-volume air sampler (HI-Q 7386, Inc. San Diego, manufactured by Environmental Products Company, CA, USA) at a flow rate of 1000 L min^–1 with 24-h sampling intervals. Before sampling, the quartz fiber filter (QFF; 203 × 254 mm, Munktell, Sweden) was heated at 600°C for 2 h to eliminate organic substances, and equilibrated at controlled environment condition (relative humidity of 35% ± 1% and temperature of 22 ± 2°C) for 48 h (U.S. EPA, 2001). Then they were weighed using a microbalance (balance sensitivity ± 0.001 mg, Mettler Toledo XP6). After sampling, the quartz filters were wrapped by aluminum foil, packaged in polyethylene bags, transported to laboratory as soon as possible. After weighed in controlled conditions described above, samples were stored at –20°C until analysis. 

Sample Analysis

Filters were extracted by accelerated solvent extraction (ASE 300, Dionex, Sunnyvale, US). The extraction procedures were repeated twice, which was operated with a mixture solvent of n-hexane/acetone (1:1, v/v) at 100°C and 1500 psi. The extract was concentrated by a rotary evaporator and then reduced to 1 mL under a gentle stream of nitrogen, transferred into a glass vial for analysis.

The OCPs were analyzed using a gas chromatography-mass spectrometer (GC-MS) (7890B-5977A, Agilent Technologies, USA) equipped with a fused silica capillary DB-5MS column (30 m × 0.25 mm × 0.25 µm) using selective ion monitoring mode. The inlet, transfer line, and ion source temperature were 250°C, 260°C, and 230°C, respectively. 2 µL of each sample was injected in the splitless mode. 

Quality Assurance and Quality Control (QA/QC)

Strict QA/QC procedures were performed during analysis processes. Laboratory and field blanks were prepared and analyzed in the same manner as samples, and OCPs were not detected in these blanks. The method detection limit (MDL) were as follows: 1 pg m^–3 for Hexachlorobenzene (HCB) and α-HCH; 1.5 pg m^–3 for β-HCH, γ-HCH and δ-HCH; 2.0 pg m^–3 for heptachlor, heptachlor epoxide and aldrin; 25 pg m^–3 for p,p’-DDT and mirex. Concentrations below the method detection limit were regarded as the lowest values. Samples were analyzed in triplicate with relative standard deviations (RSDs) less than 13%. The recoveries were 61% ± 23% in the PM_2.5 samples.

RESULTS AND DISCUSSION

OCPs Concentration

Ten OCPs (α-HCH, β-HCH, γ-HCH, δ-HCH, p,p’-DDT, HCB, heptachlor, heptachlor epoxide, mirex and aldrin) were detected in this present study (Table 1). The concentration of ΣOCPs in PM_2.5 ranged from 20.5 to 331 pg m^–3, with an average of 60.3 ± 53.3 pg m^–3.

HCHs used to be a dominant commercial insecticide. In this study, HCHs (ΣHCH, sum of α-HCH, β-HCH, γ-HCH, and δ-HCH) exhibited the highest concentration, with an average of 29.0 ± 28.1 pg m^–3.The concentrations were comparable to those observed in air samples from background sites such as the Atlantic Ocean (3 pg m^–3 for α-HCH and 22 pg m^–3 for γ-HCH) (Jaward et al., 2004) and Singapore (1.38 ± 0.86 pg m^–3 for α-HCH, and 1.36 ± 1.84 pg m^–3 for γ-HCH) (He et al., 2009) (Table 2). They were lower than some regions, such as the South China Sea (67 ± 33 pg m^–3 for α-HCH, and 771 ± 310 pg m^–3 for γ-HCH ) (Lin et al., 2012; Zhang et al., 2007), Taihu Lake (74 ± 45 pg m^–3 for α-HCH, and 46 ± 24 pg m^–3 for γ-HCH) (Qiu et al., 2004), and Korea background sites (71.3 pg m^–3 for HCHs) (Jin et al., 2013).

The concentrations of p,p’-DDT (average: 5.80 ± 12.1 pg m^–3) in this study were comparable to those in the China marginal seas (5.1 ± 5.1 pg m^–3 for p,p’-DDT ) (Lin et al., 2012), Korea background sites (3.5 pg m^–3 for DDTs) (Jin et al., 2013) and the Indian Ocean (5.9 pg m^–3 for p,p’-DDT) (Wurl et al., 2006). But they were significantly lower than results analyzed in particle samples in Beijing, China (273 ± 695 pg m^–3) (Wang et al., 2008), remote area of Turkey (74 ± 163 pg m^–3) (Ozcan and Aydin, 2010), Durban, South Africa (15.0 ± 11.7 pg m^–3) (Batterman et al., 2008), and gas samples in Tai Lake (124 ± 91 pg m^–3 for p,p’-DDT) (Qiu et al., 2004) and the South China Sea (58 ± 65 pg m^–3 for p,p’-DDT) (Zhang et al., 2007)).

HCB was mainly used for producing pentachlorophenol sodium, wood preservation and pesticide for schistosome (Wang et al., 2008). In this study, the average concentration of HCBs (3.65 ± 3.49 pg m^–3) were comparable to those measured at the points in Durban, South Africa (4.7 ± 2.7 pg m^–3) (Batterman et al., 2008), but much lower than those observed over Korea background sites (15–256 pg m^–3) (Jin et al., 2013) and Beijing (53.7 ± 28.1 pg m^–3) (Wang et al., 2008).

Heptachlor is an organochlorine cyclodiene insecticide isolated from technical chlordane in 1946. It was used primarily in termite, ant and soil insect control in seed grains and on crops, as well as in the home (Wang et al., 2008). It was banned in Japan since 1970s, and had not been registered for use in Mongolia, Russia and China (excluding Taiwan, which banned in 1989) (UNEP, 2003). In this study, the concentration of heptachlor ranged from ND to 30.3 pg m^–3 (average 3.72 ± 4.90 pg m^–3) were comparable to the results detected in remote area, Turkey (3 ± 9 pg m⁻³) (Ozcan and Aydin, 2010); higher than Korea background sites (0.41 pg m^–3) (Jin et al., 2013); but significantly lower than those regions, such as Beijing (99.9 ± 190 pg m^–3) (Wang et al., 2008).

Heptachlor epoxide was found in 41% samples in this study, ranged from ND-14.6 pg m^–3 and averaged 3.50 ± 2.88 pg m^–3. This non-systemic stomach and contact insecticide was used primarily against soil insects and termites, as well as cotton insects, grasshoppers, some crop pests and mosquitoes (Batterman et al., 2008). Karlsson et al. (2000) reported heptachlor epoxide levels of 0.82 ± 1.9 pg m^–3 at Lake Malawi, and Batterman et al. (2008) reported the average concentration lower than the detection limit in Durban, South Africa, which were slightly lower than our measurements.

It was worth noticing that aldrin had never been used in large amounts in China, but it was detected in most of samples on Mount Tai (range: ND-34.4 pg m^–3, average 4.48 ± 5.46 pg m^–3). This concentration was higher than that in Korea background sites (ND-0.65 pg m^–3, 0.08 pg m^–3) (Jin et al., 2013), but significantly lower than those in Beijing (376 ± 542 pg m^–3) (Wang et al., 2008). Aldrin was also detected in sediments and surface water from Guanting Reservoir and in most samples of water from Qiantang River (Zhou et al., 2006). Due to its persistence and semi-volatility, aldrin can enter atmosphere by volatilization from sediments and surface water, resulting in a large-scale pollution through “the Grasshopper Effect” or long-range transport (Wang et al., 2008; Wania et al., 2010).

Mirex is a stomach insecticide that has been used to control fire ants in the southeastern US, leaf cutters in South America, harvester termites in South Africa, western harvester ants in the US and mealybugs in Hawaii. It also has been used as a fire retardant in plastics, rubber, paint, paper and electrical products (Batterman et al., 2008). It has not been produced or used in Korea, Mongolia, Russia and Japan, but it was detected in some of these areas such as the study on Korea background sites (Jin et al., 2013). In China, it was used as termiticide in some regions (UNEP, 2003). The levels of mirex in this study ranged from ND to 49.3 pg m^–3, with an average of 10.1 ± 10.9 pg m^–3. The concentration was higher than that in Korea background sites (range: ND-0.78 pg m^–3, average: 0.04 pg m^–3) (Jin et al., 2013). 

Source Identification of OCPs

The concentrations and ratios of some marker OCP compounds can give the indication of the impact from different emission sources. For example, technical HCH and lindane was considered to be the major contribution of HCHs in environment. Technical HCH consisted of α-HCH (55%–80%), β-HCH (5%–14%), γ-HCH (8%–15%) and δ-HCH (2%–16%), while lindane was made up of more than 95% γ-HCH (Rodríguez-Garrido et al., 2010). The sources of HCHs can be estimated by the ratio of α-HCH/γ-HCH. If this ratio is lower than 3, HCHs was from the recent use of lindane. If the ratio is between 3 and 7, the source of HCH comes from fresh input of technical HCH. The ratio higher than 7 represents history use of technical HCH (Ozcan and Aydin, 2010). In the present study, the ratio of α-HCH/γ-HCH is between 0.03 and 1.62, indicating HCHs was from recent use of lindane after technical HCH was banned (Huang et al., 2014; Ji et al., 2015).

To further explore the potential sources of OCPs components, principal component analysis (PCA) was applied by using SPSS version 21. The principle of PCA was to describe the covariance relationship of several variances based on some common components, which was expected to explain most of the variance in analysis results. (Cheng et al., 2017). During the analysis, several factors can be subsequently divided by a certain factor interrelation, and the higher loading in factor represented more contribution to specific variation in data. In this present study, factor loadings higher than 0.6 was supposed to they can be suitable to identify source categories, and the score plot of eigenvalues >1 were considered to be able to explain the variance in output data (Ji et al., 2015; Widiana et al., 2017). Finally, the emission source represented by a component could be determined by the source profiles of pollutants contained in this component (Papaoikonomou et al., 2018). As is shown in Fig. 1, two factors accounted for 69.7% of the total data factors. The first principal component (PC1) explained 39.0% of total variance, showing high loading for p,p’-DDT, α-HCH, β-HCH, mirex and heptachlor. p,p’-DDT was banned as a pesticide in China in 1983. α-HCH and β-HCH were the main components of technical HCHs, which was identified as the residues of old use (α-HCH/γ-HCH ratio lower than 3). Heptachlor was the matrix of heptachlor epoxide, can be identified as previous usage before heptachlor. Thus, PC1 is assumed to be relatively old use of pesticides. The second principal component (PC2) with 30.7% of the variability occurred the high loading on γ-HCH, δ-HCH, HCB and heptachlor epoxide. The presence of γ-HCH could point to the contribution of lindane of recently use. HCB was widely used fungicides in China and emitted as byproduct from chlorinated industry, such as polyvinyl chloride plastics, synthetic rubber, metal production, dye and various combustion processes (Huang et al., 2014). Heptachlor epoxide was derived by heptachlor in environment. Thus, PC2 could reflect the presence of OCPs from recently use.

Fig. 1. Principal component analysis for OCPs.

Long-range Atmospheric Transport

To identify the transport patterns of OCPs and the influence of air mass transport on individual component, 24-hour air parcel back trajectories and potential source contribution function (PSCF) analysis were conducted by the Hybrid-Single Particle Integrated Trajectories (HYSPLIT 4.8) and TrajStat, respectively. The back trajectories were clustered by cluster algorithm using HYSPLIT service (Yeh et al., 2017). The PSCF analysis, combined with back trajectory results, was employed to identify the regional sources of organochlorine components. The study area was divided into some grid cells with equal size (i × j), and the PSCF value in cell (i, j) was calculated by m_ij/n_ij multiple W_ij. Where m_ij and n_ij represented the number of concentration points loading in cell (i, j) and exceeding the threshold value in this cell, respectively. W_ij was an empirical weight function to reduce the uncertainty of results. In the present study, the grid cells with high values exceeding 0.9 were selected as potential regional sources (Li et al., 2018; Tong et al., 2018).

As shown in Figs. 2 and 3, all the air mass back trajectories classified into four types by source direction, namely, SE-air masses, SW-air masses, NW-air masses and NE-air masses, respectively (represented the air from southeast, southwest, northwest and northeast of Mount Tai). The SE-air masses represented travelled across Jiangsu and the Yellow Sea. The SW-air masses mainly from Henan and the western of Shandong province. The NW-air masses that most originated from continental areas in northern China crossed Inner Mongolia, Shanxi and Hebei province and reached the station. The NE-air masses were from Bohai strait and northeast of Shandong.

Fig. 2. 24-h back trajectories of air masses during sampling period.

Fig. 3. Regional sources analysis of (a) α-HCH, (b) p,p’-DDT, (c) HCB and (d) Aldrin using PSCF.

The largest proportion belonged to SE-air masses (31.3%), with the highest OCPs concentration (average 95.3 pg m^–3), indicating that air masses from eastern coastal area was an important potential source region of OCPs on Mount Tai, which was coincident with the previous researches (Niu et al., 2014; Ji et al., 2015). OCPs applied as pesticides still have an impact on environment, and high concentrations for DDTs, HCHs, heptachlor had been reported in this region (Zhou et al., 2013; Wang et al., 2016; Sun et al., 2017).

NW-air masses and NE-air masses accounted for 18.8% and 20.3% of air masses, and average OCPs concentrations were 54.1 pg m^–3 and 36.2 pg m^–3, respectively. Some important OCP compounds had been influenced by NW-air masses and NE-air masses. α-HCH, p,p’-DDT, HCB and aldrin were associated with the air from the northwest of Mount Tai, which was consistent with the results from previous studies (Lammel et al., 2007; Liu, 2014). Agricultural activities in these regions can still consume a lot of pesticides in past decades, and make contribution of OCPs in the environment (Niu et al., 2014).

Toxicity Assessment

For atmospheric particles, exposure through inhalation and dermal contact could lead to adverse health effects. Health risk assessments for both inhalational and dermal exposures of OCPs were calculated in this study. The model was based on the Risk Assessment Guidance for Superfund, Vol. I: Human Health Evaluation Manual (U.S. EPA, 1989). Following two equations were used average daily dose (ADD) (mg kg^–1 day^–1) to estimate human health risks associated with inhalation (ADDinh) and dermal contact (ADDdern) (Wang et al., 2018).

where C is the concentration of OCPs (mg kg^–1). InhR is the inhalation rate (m³ day^–1). PEF is representative of the particle emission factor (m³ kg^–1). EF is the exposure frequency (days year^–1). ED represents the exposure duration (years). BW is the body weight (kg). AT is the averaging time (days). SA is the dermal exposure area (cm²). AF is the dermal adherence factor (mg cm^–2 day^–1). ABS is the dermal adsorption fraction. These values of parameters were referenced from the literature, showing in Table 3.

After the ADD was calculated, health hazard quotient (HQ) based on non-cancer toxic risk on humans can then be estimated by dividing the average daily dose to a specific reference dose (RfD).

The reference dose (RfD) (mg kg^–1 day^–1) values of α-HCH, β-HCH, γ-HCH, p,p’-DDT, heptachlor, aldrin and heptachlor epoxide is 0.8, 0.8, 0.3, 0.5, 0.5, 0.03 and 1.3 × 10^–2, respectively (Witczak and Abdelgawad, 2014). RfD is an estimation of maximum permissible risk on human populations through daily exposure. The threshold of RfD value can be used to indicate whether there is adverse health effect during a lifetime. HQ > 1 indicated likely adverse health effects, whereas HQ ≤ 1 indicated unlikely adverse health effects (Witczak and Abdelgawad, 2014). The estimated HQ values of OCPs detected at Mount Tai were much lower than the acceptable level (HQ ≤ 1) (Table 4), indicating exposure to PM_2.5-bound OCPs lead to an acceptable health effects in this region. However, the health effect of PM-bound OCPs had been highlighted in some regions in Northern China (Ding et al., 2015), much research work should be conducted in the future.

CONCLUSIONS

Despite most OCPs have been restricted in China, many OCP compounds are still detected in PM_2.5 at Mount Tai, a regional background site in Northern China owing to their persistence and transport in the environment. The concentration of total OCPs on Mount Tai ranged from 20.5 pg m^–3 to 331 pg m^–3. HCHs were the most abundant OCP compounds, with an average concentration of 29.0 pg m^–3. The results of PCA, back trajectory and PSCF analysis signaled the dominant source of OCPs were from the southwest agricultural areas and southeast regions of Mount Tai. The HQ values of non-cancer toxic risk on humans were estimated to be in the range of 9.28 × 10^–12 to 6.56 × 10^–3. All of values were far less than 1, suggesting that exposure to PM_2.5-bound OCPs would not lead to significant adverse health effects for general population.

ACKNOWLEDGMENTS

Financial supports were provided by National Natural Science Foundation of China (41601548, 81602827, 21806158), Tianjin Municipal Education Commission Research Project (2017KJ244), Fundamental Research Funds of National Institute of Metrology China (AKY1720), Tianjin 131 innovative talent training project, Natural Science Foundation of Chongqing (cstc2018jcyjAX0776), Research Start-up Funding Project of Chongqing Technology and Business University (1756003) and Tianjin Special Commissioner Project (17JCTPJC53100).