Determination of Occurrences, Distribution, Health Impacts of Organochlorine Pesticides in Soils of Central China

Organochlorine pesticides are groups of chemicals applied to prevent pest and insect infestation. This study was aimed at investigating the concentration, potential sources, cancer risk and ecological toxicity of organochlorine pesticides (OCPs) in Huangpi district, Wuhan, China. Eight OCPs in soil samples collected from four land-use types at depths of 0–10 and 10–20 cm were examined. Sample extraction was carried out by solid phase matrix extraction method and analyzed using Agilent gas chromatograph 7890B equipped with electron capture detectors (ECD). The total concentration of OCPs ranged from 0.00–32.7 ng g−1 in the surface and 0.01–100.45 ng g−1 in the subsurface soil layer. Beta hexachlorocyclohexanes (β-HCH) with 2.20 and 7.71 ng g−1 in the surface and subsurface soil layers, respectively, was the dominant compound. The mean concentrations of OCPs in all samples were less than the threshold values for hexachlorocyclohexanes (HCHs) and dichlorodiphenyltrichloroethane (DDTs) in China soil. Concentration of OCPs in the four land-use types were in the order of: paddy field > barren land > farmland > plastic greenhouse. Results of composition analysis revealed recent application of lindane as a major and historical use of new technical HCHs as a minor source of HCHs. On the other hand, application of new technical p,p’-DDT is the main source of DDTs in the study area. The estimated lifetime average daily dose, incremental lifetime cancer risks and hazard quotient values revealed that there is less likelihood of carcinogenic and noncarcinogenic health risks on the local residents.

Three replications with a 250 g wet weight were collected and immediately wrapped in polyethylene ziplock bags [14]. The three replications were mixed and lyophilized using a bench top lab vacuum freeze dryer at a −40 °C for 48 h. All samples were ground and sieved through a 100 mesh (0.149 mm) stainless steel sieve [29] and stored in a refrigerator at a −20 °C until the next extraction steps [30].
Three replications with a 250 g wet weight were collected and immediately wrapped in polyethylene ziplock bags [14]. The three replications were mixed and lyophilized using a bench top lab vacuum freeze dryer at a −40 • C for 48 h. All samples were ground and sieved through a 100 mesh (0.149 mm) stainless steel sieve [29] and stored in a refrigerator at a −20 • C until the next extraction steps [30].

Chemicals and Standards
The OCPs reference standards of 1000 mg L −1 containing α-HCH, β-HCH, γ-HCH, δ-HCH, p,p'-DDE, p,p'-DDD, p,p'-DDT and o,p'-DDT was purchased from AccuStandard Inc. (AccuStandard, New Haven, CT, USA). Working standards of OCPs was prepared by diluting the stock solution in n-hexane. Instrumental calibration was carried out using calibration standard solutions of 0.1, 0.2, 0.5, 1, 2, and 10 µg L −1 . The graphs for the calibration were linear with the average correlation coefficient value (R 2 = 0.999). A Chromatographic grade dichloromethane (Fisher Scientific, Waltham, MA, USA) and acetonitrile (Mallinckrodt Baker, Inc., Phillipsburg, NJ, USA) and n-hexane was purchased from (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China). In addition, analytical grade reagents, namely carbon (C18) (SiliCycle, Inc., Quebec City, QC, Canada), anhydrous sodium sulfate (Na 2 SO 4 ) and copper powder (Cu) (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China), Florisil (Beijing Yizhong Chemical Plant, Beijing, China) and neutral silica gel (Qingdao Haiyang Chemical Co., Qingdao, China), were purchased. Pretreatment and activation of the reagents were carried out to improve the effectiveness of the extraction. Accordingly, the anhydrous sodium sulfate was baked at 450 • C for 4 h before use. The 60-100 mesh Florisil and 100-200 mesh neutral silica gel were activated in a dry oven at 150 • C for 10 h and 180 • C for 4 h, respectively [24]. The neutral silica gel was deactivated with 3% ultra-pure water before use. Unlike other reagents, copper was activated by soaking the copper powder in a 2 N of HCl (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China) for 12 h at room temperature, and washed three times with water and three times with acetone (Mallinckrodt Baker, Inc., Phillipsburg, NJ, USA).

Sample Extraction, Cleanup and Analysis
The target compounds were extracted using a modified matrix solid-phase dispersion extraction method [3]. One gram of soil sample was thoroughly blended and homogenized with three grams of C18 using a glass mortar and pestle for 5 min [29]. One gram of anhydrous sodium sulfate (Na 2 SO 4 ), 1 g of Florisil, and 1 g of neutral silica gel were added to remove any trace of water and form a free-flowing powder [31] and then 1 g of activated copper powder was added to remove elemental sulfur [24,31]. The activated Na 2 SO 4 , Florisil, neutral silica gel, and copper powder were packed into a 10 mL syringe barrel column with 0.22 µm membrane filter from bottom to top. After the blended mixture had completely transferred into the syringe column using a funnel, the column was closed with 0.22 µm membrane filter and compressed with the syringe plunger to remove air from the syringe. Final elution was carried out using 20 mL of dichloromethane by gravity flow; before the extracts have concentrated till dryness under gentle high purity nitrogen (N) stream. Finally, the dried residue was re-dissolved in 100 µL of n-hexane [24].

Instrumental Analysis
An Agilent gas chromatograph 7890B electron capture detectors (ECD) was used to determine the levels of DDTs and HCHs in the soil samples. One microliter of sample extracts was automatically injected into DB-1701 (30 m × 320 µm × 0.25 µm) capillary column with a runtime of 30 min [5]. The gas chromatograph column temperature was programmed to begin at 120 • C (held for 1 min) and ramped to 240 • C at a rate of 7 • C/min and held for 0.5 min, whereas the injector and detector were operated at 270 and 300 • C, respectively. To ensure the experimental quality, all glassware was thoroughly cleaned with distilled water, baked in a muffle furnace (Zhengzhou Protech Furnace Co., Ltd., Zhengzhou, China) at 450 • C for 4 h and rinsed with hexane before use. Interference and cross contamination between samples were estimated by including a procedural blank for each set of ten samples [24]. No OCPs were detected in the blank samples and the average recoveries in this study ranged 75-110%. The detection limits (DL) and quantification limits (QL) for soil samples varied within 0.046-0.050 ng g −1 and 0.159-0.165 ng g −1 , respectively.
In addition to OCPs, three selected soil properties were determined by following the methods used by [32]. The total organic carbon (TOC) in the soil was analyzed with the Shimadzu TOC analyzer (Shimadzu, Hadano Kanagawa, Japan) (SSM-5000A). Soil pH was determined using a pH meter in a ratio of 1:5 soil: water [33,34]. Soil moisture content was determined by drying the known weight of soil in a dry oven at 105 • C for 24 h.

Human Health and Cancer Risk Assessment
Human exposure to DDTs and HCHs can induce serious health effects such as endocrine disruption, cancer, immunologic, and neurological problems [35].
where C soil is the concentration of the OCPs in soil (mg kg −1 ), IngR is the soil ingestion rate (100 mg days −1 ), ED is the exposure duration (70 years for adult and 12 years for children), EF is the assumed exposure frequency (365 days/year), CF is the conversion factor (1 × 10 −6 kg mg −1 ), AT is the upper-bound value of averaging time (70 × 365 = 25,550 days for adults, 4380 days for children), BW is the average body weight (70 kg adults and 27 kg for children), SA is the contact surface area of skin with soil (3300 cm 2 ), AFsoil is the Skin adherence factor for soil (0.2 mg cm 2 ), ABS is the dermal absorption factor% (0.2 for DDTs and 0.1 for HCHs), GIABS is the fraction of contaminant absorbed in gastrointestinal tract (1), AFInh is the absorption factor for the lungs (1), InhR is the inhalation rate (15.8 m 3 days −1 for adults), SForal is the oral slope factor (0.2 mg kg −1 days −1 ), PEF is the particle emission factor (1.36 × 10 −9 m 3 kg −1 ), IUR is the inhalation unit risk (0.057 mg m 3 ), and CSF is the cancer slope factor (0.007 mg kg −1 days −1 ). CR ingest , CR dermal , and CR inhale are cancer risk via ingestion, dermal contact and inhalation of soil, respectively. RfD is the reference dose (2 mg kg −1 days −1 ). All the values of selected parameters adopted from health Canada federal contaminated site risk assessment in Canada and [38].

Data Analysis
All descriptive statistics ranges, mean and standard deviation were calculated using IBM SPSS statistics version 20. One-way analysis of variance (ANOVA) was applied to analyze the statistical differences in the mean concentrations OCPs with soil depth and individual OCPs at a significance of α < 0.05. Pearson's correlation analysis was also used to determine the connection between PAHs and soil properties, using a two-tailed test (α = 0.05 and 0.01).

Concentrations of Organochlorine Pesticides in Huangpi Soils
The range, mean, standard deviation, frequency, DL and QL for HCHs and DDTs investigated in the present study are presented below ( Table 1). The detection frequency of OCPs in this study ranged 39-94% (0-10 cm) and 56-100% (10-20 cm). The levels of ∑OCPs (sum of the HCHs and DDTs) in the subsurface soils ranged 0.01-100.45 ng g −1 and were higher than the ND-32.7 ng g −1 in the surface soils (Table 1). Similarly, a slightly lower concentration of OCPs in the surface layer than the subsurface was reported from ditch wetlands of Chinese estuaries [6]. The ∑OCPs in this study were relatively lower than those of reported from soils of North Pacific Ocean (5.14-676 ng g −1 ) [21] and sediments of Weihe River, China (291.16 ng g −1 ) [14]. It was however relatively higher than those of from surface water of central China (0.004-0.011 ng g −1 ) [15] and a mean concentration of OCPs in Xinghua Bay soil, China (2.75 ng g −1 ) [40]. OCP is organochlorine pesticides, DL is detection limit, QL is quantification limit, HCH is hexachlorocyclohexanes, ND is not detected, Ave is average concentration, Std is standard deviation, Fre is frequency, DDT is dichlorodiphenyltrichloroethane, DDD and DDE are dichlorodiphenyldichloroethylene.
The mean concentration of HCHs in Huangpi soil were in a descending order of β-HCH > δ-HCH > α-HCH > γ-HCH in the surface, and β-HCH > γ-HCH > δ-HCH > α-HCH in the subsurface soil layers. β-HCH was the abundant and evenly distributed at both soil depths. The abundance of β-HCH is due to its less degradation, water solubility, and high affinity to adsorb in soil [41,42] as well as the conversion of γ-HCH isomer to β-HCH by microorganisms and photoisomerization [15,43]. A similar phenomenon is reported in soils from Bohai Sea, China [2] and soils from Lahore city, Pakistan [44]. The average concentration ranges of HCHs in this study (Table 1) were similar to those reported in soils of Xinghua Bay (1.22-7.47 ng g −1 ) [40], soils along Bohai Sea (3.5 ng g −1 ) [2] and from sediments of CauBay River, Vietnam (7.82 ng g −1 ) [39]. Meanwhile, they were noticeably higher than those of Wolong natural reserve soils (0.15-1.35 ng g −1 ) [45]. The one-way analysis of variance (ANOVA) showed that there was no a statistically significant variation in the mean concentration of individual OCPs at the two soil depths (F(1,14) = 0.585, p = 0.457) and between mean values HCHs and DDTs across all sampling points (F(1,34) = 2.732, p = 0.108) at α < 0.05.

Distribution of Organochlorine Pesticides in Huangpi across Land-Use Types
The sum of mean concentrations of OCPs (HCHs and DDTs) from four land-use types across the study site from two depths (0-10 cm and 10-20 cm) are displayed in Table 3 Table 3.

0%
10% 20% 30% 40% 50% 60% 70% 80% 90% 100% S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16 S17 S18 OCP composition (%) Sampling sites    According to the literature, variation in OCPs levels among agricultural soils highly depends on the existing and previous land-use types [6]. As illustrated in Table 3, the sum of mean concentration of OCPs found in the surface layers of the four land-use types were in a sequential order of: PF > BL 0% S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16 S17 S18 S9 S10 S11 S12 S13 S14 S15 S16 S17 S18 OCP composition (%) Sampling site ∑DDTs 10-20 cm ∑HCHs 10-20 cm  According to the literature, variation in OCPs levels among agricultural soils highly depends on the existing and previous land-use types [6]. As illustrated in Table 3, the sum of mean concentration of OCPs found in the surface layers of the four land-use types were in a sequential order of: PF > BL > FL > PGH. However, the level of OCPs in the subsurface layer was higher in PF followed by FL, PGH, and BL, respectively. The pooled arithmetic mean concentrations of OCPs from both soil depths were in the order of: PF (11.42 ng g −1 ) > BL (1.92 ng g −1 ) > FL (1.58 ng g −1 ) > PGH (1.07 ng g −1 ). The relatively higher level of OCPs in PF might be related to the high affinity of OCPs to soil organic matter content [24]. However, the difference in concentration of OCPs between different land-use types (p = 0.069) was not statistically significant.
Samples for this study were collected during the cropping season; thus, the indiscriminate application of lindane and rain washings from surrounding sources might play a big role in escalating the levels OCPs in PF [36]. Likewise, the seasonal application of lindane and rain washings might also have a role in increasing the level of OCPs in FL [53]. A similar phenomenon of OCPs suspension into water bodies and sediments from neighboring sources during summer is reported for East Lake, China [24]. Unlike other land-uses, BL is not expected to obtain OCPs from direct application, however, unsafe storage, disposal, and use of agrochemical might contribute greatly in increasing the OCPs level [5]. Some years ago, the current BL might have been FL, which was expected to receive a considerable amount OCPs. Moreover, BL might receive a considerable quantity OCPs washed from FL, PF and other adjacent point and nonpoint sources.
The concentration of OCPs in PGH was lower than for other land-use types. This is could be due to a limited addition of OCPs from adjacent sources and limited application DDTs and HCHs. The difference in concentration of OCPs among the different land-use types might be linked with the difference in application of pesticides and their degradation [37]. The spreadsheet for all observations acrros the study site from both soil depths is presented in the Appendix A (Table A1). Soil properties can also play a big role in concentration of OCPs in soil. The Pearson's correlation analysis at (α = 0.01) exhibited a strong positive correlation between individual OCPs and a weak (positive and negative) relationship with selected soil properties. TOC showed a strong correlation value of (R = 0.60) with α-HCH and (R = 0.25) with p,p'-DDT. The relationship among OCPs and soil properties are displayed in the Appendix A (Table A2). Similarly, positive correlations (R = 0.49) with DDTs and (R = 0.52) with HCHs are reported in CauBay River [39], while no correlation between TOC and OCPs is reported for sediments from East Lake [24].
The distributions of HCHs and DDTs across the sampling points in the study area are illustrated in Figures 2 and 3. The ∑HCHs recorded in 0-10 cm from S8, S10, S11, S13 and S18 showed variably higher concentrations than ∑DDTs. Conversely, the ∑DDTs in the 0-10 cm soil layers of S1, S2, S3, S4, S5 and S17 were higher than ∑HCHs. No HCH and DDT were detected at a soil depth of 0-10 cm in S4 and S8. S1, S7, S10, S13, S14, S15, S16 and S18 depicted a higher ∑HCHs than DDTs in 10-20 cm. In contrast, samples from S3, S4, S5 and S17 showed a higher ∑DDTs concentration than ∑HCHs. There was no DDT detected from S7 at a depth of 10-20 cm. The reasons for variation in OCPs is associated with the historical use of technical (DDTs and HCHs), users preferences in using preference chemicals [6]. One-way analysis of variance was conducted to verify the statistical variation in the mean concentrations of OCPs among different land-use types and comparing the difference in concentration between HCHs and DDTs. The results showed that there was a significant among the concentration of OCPs at the α < 0.05 for the four land-use types (F(3,14) = 4.79, p = 0.017). However, there were no statistically significant differences in the concentrations of OCPs among sampling sites as determined by one-way ANOVA (F(1,34) = 1.650, p = 0.207).
Although there is lack of information about environmental standard ecological risk of values of OCPs for soil in China [24], the Chinese environmental quality standards for soils classified the levels OCPs in soil as: HCHs ≤ 50 and DDTs = 50 ng g −1 , Grade I; HCHs ≤ 500 and DDT = 500 ng g −1 , Grade II; and HCH ≤ 1000 and DDT = 1000 ng g −1 , Grade III [52]. Based on the above classification, obtained concentration < Grade I, Grade I < obtained concentration < Grade II, Grade II < obtained concentration < Grade III are described as negligible, low and moderate pollution levels respectively; whereas obtained concentration > Grade III is classified as a high pollution level [2,44,52]. Accordingly, the levels of OCPs obtained in the study site were much lower than Grade I values for HCHs and DDTs, implying the soil is currently unpolluted.
In addition, the environmental risks of OCPs were evaluated by comparing the obtained values against the Canadian environmental quality standard guideline used by [39]. All the concentrations in this study were less than the interim soil quality guideline (ISQG; 1.42, 3.54, 1.19, 0.94 ng g −1 ) and probable effect level (PEL: 6.75, 8.51, 4.77, 1.38 ng g −1 ) for DDE, DDD, DDT and γ-HCH. The average level of HCHs and DDTs in all samples were lower than the national ocean and atmospheric administration (NOAA) threshold effect levels (TELs) of DDTs in birds (11 ng g −1 ) and soil biological communities (10 ng g −1 ) used in [6]. Generally, the ecological quality values obtained indicated that soils in the study area are suitable for agricultural production without adopting ecological mitigation measures for OCPs [37].
The estimated ∑ILCR values of HCHs and DDTs for both age groups were grouped into very low and low cancer risk level. The LADD, ILCR and HQ values of HCHs were in ascending order of: α-HCH < δ-HCH < γ-HCH < β-HCH. The carcinogenic and noncarcinogenic health impact values of DDTs showed a decreasing trend of: p,p'-DDT > p,p'-DDE > o,p'-DDT > p,p'-DDD. The LADD, ILCR and HQ values of the HCHs and DDTs investigated in this study are presented in Table 4. The ILCR values of OCPs through inhalation for both age groups were all ≤10 −6 indicating very low likelihood of cancer through soil inhalation [38]. The cancer risk values OCPs for both age groups through ingestion, dermal contact and inhalation were between very low (negligible) and low cancer risk, implying there is uncertainty in the likelihood of cancer. Generally, the computed average ILCR of OCPs (HCHs and DDTs) of 4.5724 × 10 −6 for adults and 1.1866 × 10 −5 for children and the average LADD values of 1.2684 × 10 −6 for adults and 3.2871 × 10 −6 for children revealed a low likelihood of cancer risk [61]. According to the estimated ecological and human health risk assessment values, soil in Huangpi is suitable for agricultural production without taking amendment for OCPs.

Conclusions
This study examined the concentration, distribution, possible source and human health impacts of eight OCPs (HCHs and DDTs) across eighteen sampling sites from four land-use types at the depths of 0-10 cm and 10-20 cm. Results obtained revealed the occurrence of OCPs in the study site. The concentrations of OCPs in soil were higher at 10-20 cm depth than at 0-10 cm soil depth. Concentrations of HCHs in both soil depths were dominated by β-HCH, while p,p'-DDT was the dominant DDTs residue recorded. The average concentrations of OCPs in the four land-use types were higher in PF followed by FL, PGH, and BL, respectively. The concentrations obtained in this study were lower than the threshold concentration of OCPs in soil assigned by Chinese environmental quality standards. The one-way analysis of variance results indicated that there were no statistically significant variations in concentration OCPs among soil depths and sampling points. However, there was a significant difference in the concentration of OCPs in the four land-use types. The Pearson's correlation coefficient value showed weakly negative and positive correlations between OCPs and selected soil properties. However, there was a strong and moderately positive correlation between individual OCPs. The HCHs isomers and DDTs metabolites ratios used to evaluate the potential sources of OCPs revealed a limited addition of new technical HCHs and fresh application of technical p,p'-DDT is the primary sources DDT in soil. HQ, LADD and ILCR through ingestion, dermal contact, and inhalation suggested an acceptable level of human and environmental health impacts.