Levels and Potential Health Hazards of Chlorinated Pesticides in Surface Water Samples of Charsadda Area of Pakistan Using SPME-GC-ECD Technique

: In the present study, we determined the levels of chlorinated pesticide residues in surface water samples collected from the Charsadda district (KPK, Pakistan). SPME-GC-ECD with COMBI PAL CTC autosampler was used for extraction and analysis of 20 organochlorine pesticides in the collected water samples. For maximum efﬁciency of the SPME procedure, several parameters were studied, including the extraction and desorption time of the ﬁber, solution pH, agitation of samples, and stirring speed, etc. This method showed good liner response, with R 2 values in the range of 0.9887 to 0.9999 for all pesticides. This method also provided good percent recoveries at 1 µ g L − 1 (87.5 to 106.0%) and at 2 µ g L − 1 (88.5 to 109.2%). Lower limits of detection for all 20 chlorinated pesticides were found to be lower than their maximum permissible contamination levels. Approximately 50% of the surface water samples collected from the Charsadda district were found to be contaminated with the pesticides γ -BHC, heptachlor, aldrin and dieldrin, with maximum concentrations of 0.023, 0.108, 0.014 and 0.013 µ g L − 1 , respectively. For adults and children, the cancer risk from water due to contamination by various pesticides ranged from 0 to 33.29 × 10 − 6 . The non-carcinogenic risk from each pollutant in the water samples of the Charsadda district was found to be in the order of heptachlor > aldrin > dieldrin > γ -BHC. However, the pesticides α -BHC, β -BHC, heptachlor epoxide, chlordane, endrin, 4,4 (cid:48) -DDD, endrin ketone, 4,4 (cid:48) -DDT, endosulfan sulfate and methoxychlor were not detected in any of the surface water samples of investigated in the present study.

In the present study, we identify residues of approximately 20 types of OCPs in surface water samples from Charsadda district, through the use of solid phase micro-extraction (SPME), in combination with gas chromatograph (GC) with electron capture detection (ECD). Some of these OCPs are banned and not produced in Pakistan, but are being used unofficially.

Sampling Area
Charsadda is a district of Khyber Pakhtunkhwa (KPK) province in Pakistan. Charsadda lies between 34°03′ and 34°38′ North latitudes to 71°28′ and 71°53′ East longitudes. Charsadda is located in the west of the Khyber Pakhtunkhwa and is bounded by Malakand district to the North, Mardan district to the East, Nowshera and Peshawar districts to the South and the Mohmand Agency to the West. The total area of the district is approximately 996 square kilometers (243,753 acres). The total cultivated area is 210,255 acres (61%) and the irrigated area is 180,339 acres, i.e., 86% of the total cultivated area. Three rivers pass through Charsadda: the River Jindi, the Swat River and the Kabul River. These are the main sources of irrigation for Charsadda. The three rivers then merge and join the Indus River. The population of Charsadda was 1.7 million in the year 2000. Charsadda has a density of 1081 people/sq.km. The main crops in Charsadda are tobacco, sugarcane, sugar beet, wheat and maize, etc. The major vegetables of Charsadda are potato, tomato, cabbage, okra and spinach, while the main fruits are apricot, citrus, plum, strawberry and pears. The Charsadda district is a purely agricultural area and is very fertile. Farmers use extensive amounts of pesticides on crops, fruits and vegetables for the control of various pests. Therefore, in this study to improve public awareness, we selected Charsadda district for the analysis of contamination from 20 organochlorine pesticides in surface water samples. Various surface water sample sites in the Charsadda area given in Figure 1.

Extraction of Pesticides from Water
The pesticide concentrations present in contaminated water samples are very low, and efficient extraction from water samples is needed to determine pesticide contamination in water matrixes. Therefore, for the analysis of pesticides in water samples, a good extraction and concentration method is required [32,33]. Solid-phase micro-extraction (SPME) is a new, convenient and efficient extraction method for use with liquid samples. This method needs no solvent and the use of sophisticated equipment can be avoided. SPME decreases sample analysis expenditures and time required for sample preparation. It includes the immersion of a fused silica rod coated with a stationary phase (called fiber) in the solution [32,34]. The fiber adsorbs pesticides, which are then desorbed from the fiber in the GC hot injector (for gas chromatographic analysis) or by solvent desorption (for HPLC). SPME is an easy sample preparation technique that has received increasing attention since its first development in 1990 [35,36].

Standards and Reagents
The organochlorine pesticide mixture (standards) was obtained from Supelco (Bellefonte, PA, USA) with >99% purity. The methanol and acetonitrile (HPLC grade) were purchased from Merck (Darmstadt, Germany). All the chemical reagents used (Per chloric acid and sodium hydroxide) were high-purity analytical grade reagents purchased from Sigma-Aldrich, St Louis, MO, USA. All aqueous solutions were prepared in ultra-pure water. A Milli-Q water purification system (Millipore, Bedford, MA, USA) was used for ultra-pure water. High purity nitrogen gas (99.99% purity) was used in the present study.

Solutions Preparation
Stock solution (10 mg L −1 ) containing each of the 20 organochlorine pesticides was prepared from 2000 mg L −1 standard solution (Supelco, Bellefonte, PA, USA) in methanol and stored at 4 • C. For calibration, fresh standard working solutions of various concentrations i.e., 0.01, 0.05, 0.1, 0.3, 0.5, 1 and 2 µg L −1 were prepared from stock solution in ultra-pure water containing all 20 pesticides. Calibration plots were used for quantitative analysis of pesticides in water samples.

Samples Collection
The agricultural Charsadda district was selected for the analysis of pesticides in water samples. A total of 30 surface water samples were collected from various sites in Charsadda. All samples were collected in 100 mL new and disposable plastic bottles. After collection, water samples were first filtered with Whatman filter paper to remove dust and particulate matter, and then with 0.45 µm filter paper for further purification. After filtration, water samples were stored in 20 mL glass vials (made by Agilent Technologies, Santa Clara, CA, USA) for pesticide analysis to be performed by gas chromatograph (GC). Glass vials were washed with glass cleaner detergent and a brush and then rinsed properly with running tap water. Vials were then rinsed and well shaken with an ethanol:distal water mixture (20:80) before being rinsed with ultra-pure water (milli Q water). Finally, vials were placed for few hours in an oven at 120 • C for dryness purposes. A DVB-PDMS fiber was used for pesticide extraction from water samples.

Analytical Instrumentation and Operating Conditions for GC-ECD
The solid-phase micro-extraction (SPME) fiber used was DVB-PDMS (Supelco, Bellefonte, PA, USA). The fiber was conditioned before the first application by using a GC hot injector at 250 • C and detector at 300 • C for 1h. The injector, column and detector were cleaned to remove any contamination for 30 min at 260, 280 and 310 • C, respectively. A blank test was made by injecting 1-2 microliters of highly pure ethanol and running on GC, followed by checks for the presence of any possible contamination in the injector, column and detector of the GC-ECD. After clearance, water samples were analyzed on GC for the determination of various pesticides.
All chromatographic analysis of chlorinated pesticides was carried out on an Agilent gas chromatograph (6890 N series, Agilent Technologies, Palo Alto, CA, USA) equipped with ECD (electron capture detector) and a split/split less injector. The injector was used in split less mode. Chromatographic separations were performed using HP-5 fused-silica capillary column, 30 m × 0.32 mm id and 0.25 µm film thickness (5% diphenyl-95% dimethyl siloxane coating) from Agilent. The operating GC conditions for the analysis of organochlorine pesticides in water samples were as follows: initial oven temperature was 80 • C (4 min) and then it was increased at 15 • C min −1 to 215 • C (for 30 sec). Then it was raised again at 2 • C min −1 to 230 • C and remained constant for 3 min at this temperature. After this it was increased at 10 • C min −1 to 260 • C. The detector temperature was 300 • C and the injector temperature was 250 • C. A single GC run was completed in 28 min. N 2 gas (99.999% purity) was used as a make-up and carrier gas at a flow rate of 45 mL min −1 and 1ml min −1 , respectively. For fiber, an extraction time of 45 min, desorption time of 2 min and agitator speed of 250 rpm were selected. When using the CTC autosampler, the SPME fiber is fixed in the holder assembly of the autosampler. The autosampler immersed the SPME fiber in the 20 mL vials for 45 min with agitation. The SPME fiber then extracted the pesticides from water samples. After 45 min, SPME fiber was removed from the vial and directly injected to the GC injector, where hot flow of nitrogen gas at 250 • C desorbed the organics (pesticides) from the SPME fiber to the injector, and then to the GC column for separation. Figure 2 shows a typical chromatogram of 20 standard organochlorine pesticides using the above-mentioned GC conditions and using SPME fiber.
with ECD (electron capture detector) and a split/split less injector. The injector was used in split less mode. Chromatographic separations were performed using HP-5 fused-silica capillary column, 30 m × 0.32 mm id and 0.25 µm film thickness (5% diphenyl-95% dimethyl siloxane coating) from Agilent. The operating GC conditions for the analysis of organochlorine pesticides in water samples were as follows: initial oven temperature was 80 °C (4 min) and then it was increased at 15 °C min −1 to 215 °C (for 30 sec). Then it was raised again at 2 °C min −1 to 230 °C and remained constant for 3 min at this temperature. After this it was increased at 10 °C min −1 to 260 °C. The detector temperature was 300 °C and the injector temperature was 250 °C. A single GC run was completed in 28 min. N2 gas (99.999% purity) was used as a make-up and carrier gas at a flow rate of 45 mL min −1 and 1ml min −1 , respectively. For fiber, an extraction time of 45 min, desorption time of 2 min and agitator speed of 250 rpm were selected. When using the CTC autosampler, the SPME fiber is fixed in the holder assembly of the autosampler. The autosampler immersed the SPME fiber in the 20 mL vials for 45 min with agitation. The SPME fiber then extracted the pesticides from water samples. After 45 min, SPME fiber was removed from the vial and directly injected to the GC injector, where hot flow of nitrogen gas at 250 °C desorbed the organics (pesticides) from the SPME fiber to the injector, and then to the GC column for separation. Figure 2 shows a typical chromatogram of 20 standard organochlorine pesticides using the above-mentioned GC conditions and using SPME fiber.

Health Risk Assessment Model
The health risk assessment model is a method used to estimate the nature and probability of adverse health effects in humans who may be exposed to toxic contamination in the environmental media, presently or in the future. There are several mechanisms of exposure to OCPs contamination, however, the most important one is oral exposure [37,38]. In the present study, the US EPA health risk assessment model is applied for the approximation of carcinogenic and non-carcinogenic risks for adults and children consuming underground water as their drinking water source [39]. To estimate human exposure to contaminants, chronic daily intake (CDI) is used and can be calculated with the following formula [40]. According to the US EPA exposure factors handbook, carcinogenic risk (R) can be calculated as follows [41]; where CDI is the chronic daily intake from the oral exposure route (mg kg −1 per day), SF is the slope factor of the contaminant via the oral exposure route [(mg kg −1 per day) −1 ]. Non-carcinogenic risk estimated by hazard quotient (HQ) is calculated using the following formula; where R f D (mg kg −1 per day) is the reference dose of the contaminant via oral exposure route. The values of slope factor and reference dose (Table 1) for OCPs are obtained from the US EPA integrated risk information System [42].

Optimization of Experimental Conditions
The GC-ECD chromatogram using solid-phase micro-extraction (SPME) for the 20 organochlorine pesticides standard is given in Figure 2. All chlorinated pesticides were eluted in 28 min and were well resolved under the selected GC operating conditions as mentioned in Section 2.6. The extraction of pesticides from water was carried out in 20 mL glass vials by dipping the SPME fiber in the solution using a COMBI PAL CTC autosampler. For maximum efficiency in the SPME procedure, several variables, such as the fiber extraction and desorption steps, needed to be optimized. The various parameters studied here were the fiber extraction and desorption time, solution pH, agitation of samples, stirring speed and splitless time. The pH was adjusted with 0.1M HCl and 0.1M NaOH by measurement with a digital pH meter (Newmet, pH/ISE Meter pH-250L, Korea). After the extraction of pesticides by the SPME fiber, the pesticides were injected to the column through thermal desorption in the injector of the GC. To achieve good results, all investigations were performed in triplicate and average values were obtained.
The parameters selected for SPME fiber were: a fiber extraction time of 45 min, desorption time of 2 min, agitator speed of 250 rpm, and pH of 6.2. These selected parameters were used to develop the analytical procedure and for the chlorinated pesticides analysis in real water samples.

Method Validation
The SPME-GC-ECD technique is an excellent and suitable analytical method to determine a large number of chlorinated pesticides in real water samples. The analytical parameters applied to validate the method were linearity, selectivity, accuracy, precision and detection limits. Calibration plots were used to determine various compounds in the water samples by plotting peak area versus compound concentrations.  Table 2. The LOD for all 20 organochlorine pesticides were calculated in the range of 0.002-0.03 µg L −1 , and were found to be lower than their maximum permissible contamination levels. Various researchers have determined that the LOD for different organochlorine pesticides ranges from 0.001-0.025 µg L −1 [43], 0.008-0.020 µg L −1 [19], and 0.039-0.340 µg L −1 (EPA 525.2 method) [44]. The solid-phase micro-extraction (SPME) technique is a non-exhaustive extraction method. Relative recovery can be calculated as the ratio of peak areas (GC) of spiked water sample extracts to spiked ultra-pure water extracts. The real water samples were spiked with pesticide standards (1 and 2 µg L −1 ) to perform relative recovery tests, and the results are given in Table 3. The percentage recoveries found at 1 µg L −1 and at 2 µg L −1 concentrations of pesticides ranged from 87.5% to 106.0% and 88.5% to 109.2%, respectively. At 1 µg L −1 level concentration, the minimum percent recovery value found was for 4,4 -DDT (87.5%) and maximum percent recovery value found was for β-BHC (106.0%). However, at 2 µg L −1 level concentration, the minimum percent recovery value found was for α-chlordane (88.5%) and maximum percent recovery value found was for endrin ketone (109.2%). The percentage recoveries investigated by other researchers for different organochlorine pesticides were ranged from 44.6% to 121.6% [19], 75% to 101% [43] and 69% to 107% [45]. All these results show that the SPME-GC-ECD technique is an excellent and convenient method for chlorinated pesticides analysis in real water samples. Table 3. Percent recovery values of selected pesticides at two concentration levels by using SPME-GC-ECD technique and comparison with some reported values.

Pesticide Residues in Water Samples
In the present study, 20 organochlorine pesticide residues in surface water samples collected from the Charsadda district of Khyber Pakhtunkhwa, Pakistan were investigated. A typical chromatogram of pesticides contaminations in surface water samples collected from the Charsadda area is shown in Figure 3.
Among the various pesticides analyzed, β-BHC, γ-BHC, heptachlor, aldrin, endosulfan I, dieldrin, 4,4-DDE, endrin aldehyde and endosulfan sulfate were detected in the surface water samples of Charsadda district. The concentrations of pesticides measured by the SPME-GC-ECD method, are given in Table 4. A total of 30 surface water samples were analyzed for 20 organochlorine pesticide residues. Approximately 50% of the surface water samples collected from Charsadda were contaminated with the pesticides γ-BHC, heptachlor, aldrin and dieldrin. The pesticide contaminants found in surface water samples were γ-BHC (16 samples), heptachlor (17 samples

Pesticide Residues in Water Samples
In the present study, 20 organochlorine pesticide residues in surface water samples collected from the Charsadda district of Khyber Pakhtunkhwa, Pakistan were investigated. A typical chromatogram of pesticides contaminations in surface water samples collected from the Charsadda area is shown in Figure 3. Among the various pesticides analyzed, β-BHC, γ-BHC, heptachlor, aldrin, endosulfan I, dieldrin, 4,4-DDE, endrin aldehyde and endosulfan sulfate were detected in the surface water samples of Charsadda district. The concentrations of pesticides measured by the SPME-GC-ECD method, are given in Table 4. A total of 30 surface water samples were analyzed for 20 organochlorine pesticide residues. Approximately 50% of the surface water samples collected from Charsadda were contaminated with the pesticides γ-BHC, heptachlor, aldrin and dieldrin. The pesticide contaminants found in surface water samples were γ-BHC (16 samples), heptachlor (17 samples), aldrin (14 samples) and dieldrin (11   Table 5 shows a comparison of pesticide contamination in the Charsadda area with the reported data of some other countries. The organochlorine pesticide residues in the surface water samples of Ghana have been reported in the range of 0.02-0.08 µg L −1 (γ-BHC), 0.01-0.12 µg L −1 (δ-BHC), and 0.02-0.04 µg L −1 (Heptachlor) [7]. Yang et al. (2014) reported pesticide contamination in the surface water samples of China, with maximum concentrations of 0.003, 0.006, 0.003 and 0.012 µg L −1 for α-BHC, β-BHC, γ-BHC, and heptachlor, respectively [46]. Liu et al. (2013) identified pesticide contamination in surface water samples of Lake Chaohu, China, with maximum concentrations of 0.003 µg L −1 (α-BHC), 0.009 µg L −1 (β-BHC), 0.003 µg L −1 (γ-BHC), 0.001 µg L −1 (heptachlor) and 0.001 µg L −1 (aldrin) [8]. However, Golfinopoulos et al. (2003) reported pesticide contamination in surface water samples of Northern Greece with maximum concentrations of 0.031, 0.096, 0.081, 0.189, 0.020 and 0.101 µg L −1 for α-BHC, β-BHC, γ-BHC, δ-BHC, heptachlor and aldrin, respectively [19]. Lindane (γ-BHC) has been used to treat food crops, forestry products, seeds and soil, and to treat pets and livestock. It was applied largely to cotton, coffee, cocoa, sugar cane and other crops. BHC isomers can undergo long-range transport and can be deposited into aquatic systems, where they can bioaccumulate in the food chain. Heptachlor has been used to kill soil insects and termites, to control grasshoppers, cotton insects and some crop pests, particularly those infecting corn. Aldrin and dieldrin are strongly related to each other. They were counted for a number of cases that produced the mortality of many wild life species, because following the application of aldrin in the fields, it rapidly changes to dieldrin [47].
All these studies show that pesticide contamination is a serious threat to human health and the environment. Therefore, efficient analytical and extraction methods are necessary to determine pesticide contamination in environmental samples. The analytical parameters studied with the SPME-GC-ECD method showed good percent recoveries, low detection limits and acceptable linear correlation, making them suitable for application in the analysis of water samples using the solid-phase micro-extraction (SPME) technique. This method allows a significant decrease in the volume of sample extracted, organic solvent consumption, and sample handling time in comparison with classical liquid-liquid extraction (LLE), or solid phase extraction (SPE) that has been commonly used in the analysis of pesticides in waters.

Health Risk Assessment of OCPs in Water Samples
Underground water i.e., pressure pumped and hand pumped water is mostly used for drinking purposes in the Charsadda district. A health risk assessment model [42] was used to calculate health risks from underground drinking water for adults and children. Some OCPs were present and some were not detected in the underground water samples of Charsadda district. The health risk levels caused due to OCPs were calculated for the maximum values of contaminants present in the underground drinking water for adults and children. Table 1 showed the summary statistics for the health risks that may be associated with OCP contaminants present in the water samples of the Charsadda district. For adults, the cancer risk from surface water ranged from 0 to 0.96 × 10 −6 for γ-BHC, 0 to 15.53 × 10 −6 for heptachlor and 0 to 7.61 × 10 −6 for aldrin. The cancer risk for surface water from 4,4 -DDE and dieldrin ranged from 0 to 0.16 × 10 −6 and 0 to 8.69 × 10 −6 , respectively.
However, the cancer risk from surface water ranged from 0 to 0.47 × 10 −6 for 4,4 -DDE and 0 to 14.25 × 10 −6 for dieldrin. In most of the water samples, the cancer risk from heptachlor, aldrin and dieldrin for adults and children was negligible, however, in a few samples it was higher than the acceptable risk level (1 × 10 −6 ) recommended for carcinogens by the US EPA. The concentrations of these OCPs in the water samples indicates that, in most of the Charsadda district, the water may not pose any serious cancer risk to the local population, especially to children, and that the majority of the water samples in the area were suitable for drinking. However, it is essential to prevent new input sources of these OCPs in the surface and underground water to ensure safe drinking water.
The hazard quotients were also estimated according to the assessment model for individuals based on the maximum contamination value of OCPs in water samples from Charsadda district. When the ratio of HQ increases above 1, then it may have adverse effects on human health [45]. The non-carcinogenic risks from heptachlor, aldrin and dieldrin in the water samples from Charsadda district for children was somewhat higher than 1, which is the value recommended by US EPA, 1986. The non-carcinogenic risk from each pollutant in the water samples of the Charsadda district was in the order of heptachlor > aldrin > dieldrin > γ-BHC. Non-carcinogenic risk from α-BHC, β-BHC, endosulfan I, 4, 4-DDE, endrin, 4, 4 -DDT, endrin aldehyde, endrin ketone, γ-chlordane, α-chlordane, 4,4 -DDE, endosulfan sulfate and methoxychlor was not found in water samples from Charsadda district.

Conclusions
In this study, SPME coupled to the GC-ECD method was successfully developed for the extraction and determination of 20 chlorinated pesticides in surface water samples. The SPME method is fast, effective, simple, more environment friendly and decreases expenditures and time required for sample preparation and analysis. It is a good alternative to conventional sample extraction techniques such as liquid-liquid extraction (LLE) and solid phase extraction (SPE) without using expensive and toxic solvents or expensive extraction equipment. The SPME-GC-ECD method showed high extraction efficiency, good percent recoveries (87.5% to 106.0%) and acceptable LODs (0.002 to 0.03 µg L −1 ) for all 20 chlorinated pesticides, which were much lower than their maximum permissible contamination levels. These results indicated that this system is a sensitive, cheaper and simpler alternative to mass spectrometry for assessment the presence of organochlorine pesticides residues in water samples. About 50% of the surface water samples collected from Charsadda district were found to be contaminated with pesticides such as γ-BHC, heptachlor, aldrin and dieldrin. Contamination from the major pesticides was found at maximum values of 0.023 µg L −1 (γ-BHC), 0.108 µg L −1 (heptachlor), 0.014 µg L −1 (aldrin) and 0.013 µg L −1 (dieldrin) in the surface water samples of the Charsadda area. For adults, the cancer risk from water due to contamination of various pesticides ranged from 0 to 15.53 × 10 −6 µg L −1 . The non-carcinogenic risk found from each pollutant in the water samples was in the order of aldrin > dieldrin > heptachlor > γ-BHC. Similarly, for children, the cancer risk ranged from 0 to 33.29 × 10 −6 µg L −1 and non-carcinogenic risk was found in the order of heptachlor > aldrin > dieldrin > γ-BHC. The SPME coupled to GC-ECD has been found to be an excellent approach for the analysis of chlorinated pesticides in water samples. Funding: This research was funded by project nr T190087MIMV and European Commission, MLTKT19481R "Identifying best available technologies for decentralized wastewater treatment and resource recovery for India, SLTKT20427 "Sewage sludge treatment from heavy metals, emerging pollutants and recovery of metals by fungi and by project KIK 15392 and 15401 by European Commission.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.