1 Introduction

Synthetic pyrethroids (SPs) show enhanced persistence and toxicity. They are chemical pesticides modified from pyrethrins, compounds naturally present in chrysanthemum flowers (Palmquist et al. 2012). SPs can be classified into two different types: type I if they are non-cyano pyrethroids or type II if they have the presence of the cyano group. SPs can also be classified as insecticides according to the type of organisms they act against, as they are used for insect pest control in agriculture, public, and animal health (Pfeil 2014; Albaseer et al. 2011). Their wide use is due to being considered one of the safest insecticides presently accessible for agricultural and public health purposes (Nicolopoulou-Stamati et al. 2016). The mechanism of action of SPs is based on altering the normal nervous system function of insects and other invertebrates and modifying the kinetics of voltage-sensitive sodium channels, leading to the knockdown of their normal functioning and eventually causing death (Soderlund 2010).

Abundant evidence of pyrethroid effects studied in animal models shows plausible outcomes that can be transposed to humans (Burns and Pastoor 2018). Studies using rats as model organisms suggested that pyrethroids can contribute to the etiology of Parkinson’s disease, due to their environmental inductive component (Nasuti et al. 2007), by stimulating cognitive deficits (Nasuti et al. 2013). Children are particularly sensitive to environmental contamination with pesticides. They are susceptible to the uptake and to the adverse effects of such compounds because of developmental, dietary, and physiological factors. Studies suggest that exposure to pyrethroid pesticides might be associated with the increased risk of brain tumors in childhood (Chen et al. 2016) and also the correlation of pyrethroid metabolites with autism spectrum disorder (Domingues et al. 2016).

More than a dozen registered pyrethroid molecules are used in innumerable products (Burns and Pastoor 2018). The intensive use of SP pesticides raise environmental concerns (Zhang et al. 2011) because their residues are frequently detected in soils (Han et al. 2017; Liu et al. 2008; Gu et al. 2010) and sediments (Weston et al. 2005; Aznar et al. 2017; Wang et al. 2012; Jabeen et al. 2015), in water (Feng et al. 2015), and even in crops (Akoto et al. 2013; Wahid et al. 2017). SPs can be found in worrying concentrations in terms of toxicity in some non-target invertebrates (Weston et al. 2013; Palmquist et al. 2012). Bifenthrin reached a maximum of 32 ng g−1 in California’s Central Valley sediments which is highly toxic compared to an LC50 of 3–10 ng g−1 to common sediment toxicity testing species (Weston et al. 2013). The majority of insecticides reach the soil either by direct application (to kill soil-borne pests) or indirectly when applied to the aerial part of the plant, usually by runoff from the leaves and stems (Farina et al. 2016).

City parks and agricultural soils are often contaminated with metals (Laidlaw et al. 2018; Ponavic et al. 2018; Chen et al. 2018), polycyclic aromatic hydrocarbons (Ke et al. 2017), polychlorinated biphenyls (Lu and Liu 2015), and pesticides (Li et al. 2008). Contaminated soils should be considered an important source of human pyrethroid exposure namely by dermal contact (as children in playgrounds and farms) or via ingestion of crops cultivated in those soils. From the few studies performed of city park soils, in Beijing, China, the organochlorine pesticides hexachlorocyclohexanes (HCHs) and dichlorodiphenyltrichloroethanes (DDTs) were reported at 0.2490–197.0 ng g−1 and 5.942–1039 ng g−1 concentration levels, respectively (Li et al. 2008). For agricultural soils, six pyrethroids (bifenthrin, cypermethrin, cyhalothrin, fenvalerate, fenpropathrin, and deltamethrin) were found in 65.8% of the investigated nut-planted soils from China, at levels from 1.5 ng g−1 to 884.3 ng g−1 (Han et al. 2017).

Several methods for pyrethroid residue extraction from environmental samples are described in the literature such as Soxhlet extraction (Woudneh and Oros 2006), solid-liquid extraction followed by salting-out-assisted liquid-liquid extraction (Pastor-Belda et al. 2018), headspace solid-phase microextraction (Fernandez-Alvarez et al. 2008), microwave-assisted extraction (Esteve-Turrillas et al. 2004), and ultrasound-assisted extraction (Aznar et al. 2017; Albaseer et al. 2010). QuEChERS (Quick Easy Cheap Effective Rugged and Safe) method’s numerous advantages are described over the traditional extraction techniques. These include simplicity, low cost, low solvent, and high efficiency with fewer steps (Vera et al. 2013; Fuhrmann et al. 2014). For this reason, QuEChERS was the implemented method in this work for the pyrethroid determination in environmental matrices.

Concerning the increasing use of pyrethroid pesticides worldwide, the present work intended as a major goal the assessment of pyrethroid pesticide soil contamination in Portuguese soils. As described by other authors, a QuEChERS method with gas chromatography (GC) with an electron-capture detector (ECD) was implemented in this work (Dubey et al. 2018). The optimized method for pyrethroid determination was able to quantify trace amounts of eight pyrethroid pesticides (bifenthrin, tetramethrin, cyhalothrin, permethrin, cypermethrin, cyfluthrin, deltamethrin, and fenvalerate) in soils. The efficiency of the methodology was confirmed with ten playground soil samples collected in Porto city and 18 agricultural soils collected in Cávado-Vouga agriculture region. Both types of soils were evaluated for pyrethroid presence and monitored during two seasons (summer and winter).

2 Materials and Methods

2.1 Chemicals and Reagents

For this study, eight of the most widely used SPs were verified: α-cypermethrin, β-cyfluthrin, ʎ-cyhalothrin, bifenthrin, deltamethrin, fenvalerate, permethrin, and tetramethrin. Pesticide standards (more than 95.4% of purity) were purchased from Chemservice (West Chester, USA) and Sigma-Aldrich Co. Merck was the supplier of the acetonitrile and n-hexane (high-purity solvents). Individual pesticide standard stock solutions (between 100 and 200 mg L−1) were prepared in acetonitrile. Standard working solutions (with the eight SPs) were prepared from individual stock solutions at several concentrations’ levels. Stock and working solutions were stored at 4 °C. Agilent technologies (Bond Elut Sample preparation solutions; Lake Forest, USA) was the supplier of QuEChERS and the dSPE. The QuEChERS used was a buffer-salt mixture with 4 g of magnesium sulfate anhydrous grit (MgSO4), 1 g of sodium chloride (NaCl), 0.5 g of sodium hydrogenocitrate sesquihydrate (C6H8Na2O8), and 1 g of sodium citrate (Na3C6H5O7). The dSPE was composed of 150-mg primary secondary amine (PSA), 150-mg MgSO4, and 50-mg bonded silica (C18). Deionized water was obtained and purified (18.2 MΩ cm) using water purification systems from Millipore (Elix apparatus and a Simplicity 185 system). Helium (purity ≥ 99.999%) and nitrogen (purity ≥ 99.999%) were both obtained from Linde Sogás (Lisbon, Portugal).

2.2 Study Area and Sample Collection

Different topsoil samples were collected from several urban playground parks (P1–P10) and agricultural locations (A1–A18) of the north of Portugal. The upper layer (topsoil, 0–10 cm) was removed with a clean spade from each sampling site. Different sampling points, with similar characteristics, of the playground park and agricultural soil were collected in order to have a representative sample of the area to obtain a composite sample. Samples were sieved (2 mm) and were stored at − 18 °C until analysis.

2.2.1 Playground Soils

The sampling was done in playgrounds of urban areas of the north of Portugal, in Porto city and surrounding regions. The playgrounds selected were the ones set in city parks with a high density of green areas, with several different types of trees (e.g., pines, Metrosideros, palm, and cork trees), grass areas, shrubs, camellia gardens, and rosebushes. Dogs are allowed in these parks. Ten soil samples were taken at various locations in Porto city, on the 5th and 6th of September 2015 (summer) and 20th and 21st of February 2016 (winter) (Fig. 1).

Fig. 1
figure 1

Location map of playgrounds sampling sites (P1–P10)

Full size image

2.2.2 Agricultural Soils

The farming sampling sites were in the agricultural regional delegation of Cávado-Vouga. Eighteen soil samples were collected during the first week of July 2016 (summer) and the first week of January 2017 (winter) at the marked sites (Fig. 2). The samples were all coming from conventional agriculture soils. Five of them were from intensive farming practices (A1, A2, A7, A15, and A18) and the remaining ones were from sustainable agriculture practices.

Fig. 2
figure 2

Map of agricultural soil sampling locations (A1–A18). A, Agriculture regional delegations of the north of Portugal; B, Agriculture regional delegation of Cávado-Vouga

Full size image

2.3 Soil Analysis and Characterization

The physicochemical characterization of soils was performed, namely the pH, the water content, and the total organic carbon (TOC) (Nelson 1996; Hesse 1972), and was determined. For the analysis of TOC in soils, a Shimadzu TOC analyzer (VCSN, Japan) and a Shimadzu solid sample module (SSM-5000A) were used. A Kern moisture analyzer (MLS 50-3IR160, Germany) was used to determine the water content. For the measure of the pH, a suspension of soil in water (1:1) was read with an electronic pH meter (Crison 2002, Spain). All determinations were made in triplicate.

2.3.1 QuEChERS Extraction Procedure

Soils were extracted using an adapted QuEChERS method from the European (EN15662 2008) method (EN15662 2008; Dubey et al. 2018). As described before by others authors, a smaller amount of sample (5 g) was used that was weighed into a Teflon tube (50 mL) to which 10 mL of acetonitrile was added (Correia-Sá et al. 2012; Bragança et al. 2012). The tube was capped, shacked vigorously by vortexing for 1 min, and sonicated (10 min) in a 195-W ultrasonic bath from J.P. Selecta (Spain). The QuEChERS buffer-salt mixture was then added to the suspension resultant from this extraction. Phase separation and pesticide partitioning were induced by shaking vigorously for 1 min in a vortex, followed by placing in an ultrasonic bath for 10 min and later centrifugation at 4000 rpm (10 min). A 1.5-mL aliquot was sampled from the upper layer and moved into a 2-mL cleanup tube, vortexed for 1 min, and centrifuged at 4000 rpm (10 min). A 0.7-mL aliquot of the upper layer was removed into a vial and evaporated with a nitrogen gentle stream until dryness. The dry residue was resuspended in 0.7 mL of n-hexane and vortex-shacked vigorously, and finally the extract was analyzed by GC-ECD chromatograph.

2.3.2 Gas Chromatography

SPs were analyzed using a GC-ECD (Shimadzu GC-2010) with a capillary column ZB-XLB (30 m length, 0.25 mm i.d., 0.25 mm film thickness; Zebron - Phenomenex). The temperature of the oven was set to start at 60 °C (stay for 1 min), increase by 30 °C min−1 till 250 °C (held for 11 min), and then an increase of 10 °C min−1 till 290 °C. The injection port was operated at 250 °C (in splitless mode) and the detector was set at 300 °C. Injection (1 μL) was carried out with an autosampler (HT 300A from hta). Helium was used as the carrier gas (1.5 mL min−1), while nitrogen was the makeup gas (30 mL min−1). The software GC-Solutions Shimadzu was used for sample processing.

In addition, the samples positive to deltamethrin were further analyzed using a Zebron ZB-5MSi column (30 m length, 0.25 mm i.d., 0.25 μm film thickness; Phenomenex) in a Thermo Trace-Ultra gas chromatography-mass spectrometry Thermo Polaris Q ion trap detector operated in the electron-impact ionization (EI) at 70 eV. Injection (1 μL) was carried out with an autosampler (AI3000) in an injector set at 240 °C (splitless mode). Helium was used as the carrier gas (1 mL min−1). The temperature of the oven was programmed to start at 90 °C (held for 1 min), followed by an increase of 15 °C/min to 250 °C (held for 1 min), then an increase of 20 °C/min to 255 °C (held for 5 min), a new increase of 10 °C/min till 270 °C (held for 1 min), and finally an increase of 20 °C/min to 290 °C (stay for 2 min). The transfer line temperature was 270 °C and the ion source temperature was 250 °C. The MS/MS conditions were as follows: maximum excitation energy of 0.225, excitation voltage of 0.5 V, excitation time of 15 ms, isolation mass window width of 1.0, and isolation time of 12 ms. The identification of the selected precursor ion for deltamethrin was 172 as it was the one with the maximum abundance factor and m/z ratio. The software Xcalibur 1.3 was used for data processing.

2.4 Method Validation

For SP analysis, the experimental method validation was performed according to the SANCO guidelines from the European Union, regarding pesticide residue analytical methods (EU 2010, 2013). The matrix effect influence on the ECD signal was evaluated by setting up a calibration curve using soil extract and comparing with the one achieved for the standards in hexane. SPs’ calibration curves and linear range response were evaluated through the analysis of eight standard solutions (10 to 360 ng g−1) in triplicate. The organic carbon content (OC) produces a matrix effect on the QuEChERS method (Correia-Sá et al. 2012). To ascertain this effect, two calibration curves were performed: one for a soil with a low TOC content (0.374%) and another for a soil with high TOC content (OC = 3.70%). The detection and quantification limits (LODs and LOQs) were determined using the standard deviation (SD) of the response (SD of y-intercepts of the regression lines), the slope of the calibration curve, and the signal-to-noise ratio convention, 3:1 for LODs and 10:1 for LOQs (Rambla-Alegre et al. 2012). The linearity of the method was established by setting calibration curves over the concentration range using linear regression analysis. Selectivity was verified by comparing the chromatograms of standards dissolved in n-hexane, standards extracted from the spiked soil, and matrix blanks (non-spiked soil). The precision of the method was evaluated by carrying out the extraction and analysis of 40, 200, and 360 ng g−1 fortified samples using three replicates, being the extract injected in triplicate. The analytical method of accuracy was evaluated through recovery studies (EU 2010, 2013).

3 Results and Discussion

3.1 Soil Sample Characterization

The pH values were similar for both types of soil and for both seasons (Table 1). It was noted that for both types of soils, the water content was lower in summer than in winter with much higher rainfalls, in accordance with the climatic conditions of each season. The determination of TOC was an important part of soil characterization since its presence/absence could affect how chemicals would respond in the soil (Correia-Sá et al. 2012). TOC contents in agricultural soils (1.24–7.02%) were generally higher than those in playgrounds soils (0.03–2.87%).

Table 1 Macro parameters of soil samples. Characterization mean values with standard deviation (SD)
Full size table

3.2 Method Performance and Validation

Since some authors reported an influence of the amount of organic carbon on the matrix effect, soils with different TOC contents were tested (Correia-Sá et al. 2012). Considering different total organic carbon contents, comparing the two matrix calibration curves (low and higher TOC content), no matrix effect was observed in all compounds (data not shown). As such, the calibration curves were constructed with fortified extracts using the soil with low TOC content. The chromatograms obtained for extracts with pyrethroid after the QuEChERS method application are shown in Fig. S1, comparing the non-spiked soil (matrix blank), the standards dissolved in n-hexane, and the spiked soil. The QuEChERS method provided extracts that contained the target analyte, with high recoveries, and free from interferences in terms of the pesticide’s retention time.

Analytical parameters for method performance and validation were calculated for the applied method to each pyrethroid pesticide and are reported in Table 2. Method accuracy was expressed as recovery from the fortified blank soil samples at 40, 200, and 360 ng/g concentration levels, with good recoveries (79.4–100.2%), and with precision values, as determined by the mean of the relative standard deviation (RSD), below 9.5% in all cases. These determinations comply with the requirement guidelines of the European Commission concerning methods of analyzing pesticide residues (EU 2013). Linear calibration curves for all the investigated pesticides were obtained in the range of 10 and 360 ng g−1 with coefficients of determination higher than 0.991. The limits of detection and quantification ranged from 0.33 to 2.72 ng g−1 and 1.10 ng g−1 to 9.06 ng g−1, respectively.

Table 2 Recoveries, detection and quantification limits (LOD, LOQ), and matrix effect for the investigated compounds
Full size table

3.3 Pyrethroid in Soils

Eight pyrethroid pesticides (i.e., bifenthrin, deltamethrin, cyhalothrin, cyfluthrin, cypermethrin, fenvalerate, permethrin, and tetramethrin) were analyzed in playgrounds and agricultural soils. When the validated procedure for the monitoring of pyrethroid pesticide residues was applied to the ten non-spiked soil samples from Porto city playgrounds, no pyrethroid contamination was found (not detected) during both seasons. Considering the potential effect of pyrethroids by dermal contact (Hołyńska-Iwan et al. 2018), these city playgrounds may be considered safe for children. Since the understanding of the environmental quality and the impact of these compounds in daily life is crucial future analyses in a wider territory, a higher number of samples and including more pyrethroid compounds are of major importance as these urban facilities are frequently visited. Also, playground air contamination with pyrethroid residues should be considered.

Table 3 contains the results of the pyrethroid contamination detected in agricultural soils. Deltamethrin was the only pyrethroid pesticide confirmed by GC-MS/MS, out of the eight analyzed. It was detected in three of the 18 agricultural soils only during the summer period, while in the winter, no pyrethroid contamination was found. The occurrence only happening in summer may be related to their usage pattern, since pesticides are usually applied in the spring/summer, as advised by the Ministry of Agriculture Forestry and Rural Development (DGAV n.d.). Many pyrethroids undergo microbial biodegradation by the action of several esterases, justifying that SPs were not detected in winter. When pyrethroids were present in measurable amounts, their concentrations were 15.7, 17.8, and 101.7 ng g−1 for samples A18, A8, and A5, respectively. No relation between the type of agriculture practices and the presence of pyrethroid pesticides was found since A18 was a soil sample from intensive farming and while the other positive soil samples were from sustainable agriculture practices. Comparing with available literature, six SPs (bifenthrin, fenpropathrin, cyhalothrin, cypermethrin, fenvalerate, and deltamethrin) were found in chestnut, walnut, and pine soils in China (29.4 ng g−1 to 884.3 ng g−1, 9.7 ng g−1 to 57.3 ng g−1, and 1.5 ng g−1 to 85.0 ng g−1, respectively) (Han et al. 2017). Creeks sediments collected inside a residential neighborhood near Roseville (USA) presented maximum concentrations of 169, 335, 437, and 736 ng g−1 of cyfluthrin, permethrin, bifenthrin, and cypermethrin, respectively (Weston et al. 2005). Several SPs (bifenthrin, cyhalothrin, cyfluthrin, cypermethrin, deltamethrin, fenpropathrin, fluvalinate, permethrin, esfenvalerate, and resmethrin) were also detected in the Mediterranean paddy fields at a maximum concentration of 57.0 ng g−1 before plow and 62.3 ng g−1 during rice production, being resmethrin and cyfluthrin the compounds detected at higher concentrations (Aznar et al. 2017). Deltamethrin was also found in other reports. For instance, in a chestnut soil in China, deltamethrin varied between n.d and 29.3 ng g−1 (Han et al. 2017). Considering their presence in crops, in Ejura, Ghana, nine pyrethroids (bifenthrin, fenpropathrin, cyhalothrin, permethrin, cyfluthrin, cypermethrin, fenvalerate, allethrin, and deltamethrin) were detected in all cowpea-and maize-measured samples, with average concentrations from 1 to 39 ng g−1 and from 2 to 28 ng g−1, respectively. In particular, with deltamethrin, the detected levels ranged from 3 to 13 ng g−1 (cowpea) and from 2 to 3 ng g−1 (maize) (Akoto et al. 2013).

Table 3 Concentrations (ng g−1) of deltamethrin in agricultural soils from the north of Portugal
Full size table

4 Conclusions

The intensive use of pyrethroid pesticides raises environmental concerns, especially regarding soil contamination. The spatial and seasonal occurrence of pyrethroid pesticide contaminants in the environment was the focus of this work. The methodology implemented for pyrethroid analysis could be extremely relevant for further studies related to pyrethroid exposition and its effects. No evidence was found that, in Portuguese playground soil samples, the pyrethroid pesticides analyzed were present. Nevertheless, 17% of the agricultural soil samples tested had measurable levels for one target pyrethroid (deltamethrin) during the summer season. The results confirm the presence of pyrethroid pesticides in Portuguese soils and the need for monitoring and assessment. In order to circumvent this problem, tests should be done in a wider territory, with a higher number of samples and including more pyrethroid compounds, both in urban and rural facilities. A deeper understanding of the environmental and the ecological potential impacts of pyrethroid pesticides in soil is needed.