Rapid and sensitive determination of pyrethroids indoors using active sampling followed by ultrasound-assisted solvent extraction and gas chromatography

1 Rapid and sensitive determination of pyrethroids indoors using 1 active sampling followed by ultrasound-assisted solvent extraction and 2 gas chromatography 3 Ruth Barro, Carmen Garcia-Jares*, Maria Llompart, Maria Herminia Bollain 4 and Rafael Cela 5 Departamento de Quimica Analitica, Nutricion y Bromatologia, Facultad de 6 Quimica, Instituto de Investigacion y Analisis Alimentario, Universidad de Santiago de 7 Compostela, E-15782 Santiago de Compostela, Spain. 8 * Corresponding author: 9 Tel.: +34 981563100; fax: +34 981595012. 10 E-mail address: [email protected] (C. Garcia-Jares). 11 12 Abstract 13 14 A fast and simple method to analyze pyrethroids as well as other components of 15 frequently used domestic insecticide preparations in indoor air is presented. The 16 proposed method, based on sampling with an adsorbent followed by ultrasound-assisted 17 solvent extraction, was developed with the aim to simplify the traditional extraction 18 methodologies applied up to date to determine pesticides in air. The analytes were 19 retained on a very small amount of adsorbent, which allowed using solely 1 mL of 20 solvent for desorption. The quantification was performed by gas chromatography with 21 micro electron-capture detection (GC-µECD) and gas chromatography coupled to mass 22 spectrometry (GC-MS). The influence of main factors involved in the ultrasound23 assisted solvent extraction step (type of adsorbent and type of solvent, solvent volume 24 2 and extraction time) was studied using an experimental design approach to account for 1 possible factor interactions. The sampling step was studied for two adsorbents (Tenax 2 TA and Florisil), finding that 1 m3air could be sampled without losses of analytes. In 3 this way, the analysis of pyrethroids in air by the proposed method could be carried out 4 within a total time shorter than an hour, including sampling. Linearity was demonstrated 5 in a wide concentration range. Efficiency of the total sampling-extraction process was 6 studied at several concentration levels (2, 10, 100 and 1000 ng/m3), obtaining 7 quantitative recoveries for all compounds, with good precision (RSD<10%). Method 8 detection limits were below 1 ng/m3in air when GC-µECD was employed, and about 9 one order of magnitude higher for GC-MS. In addition, the proposed method was 10 applied to real samples collected in contaminated closed rooms, in which some of the 11 target compounds were determined. 12 13 Key words: pyrethroids, ultrasound-assisted solvent extraction, air analysis, 14 factorial design optimization, gas chromatography 15 16 17 3 1. Introduction 1 2 Interest and demand for ambient air analysis have increased as it has the number 3 and diversity of air pollutants of concern. Synthetic pyrethroids are chemicals that have 4 been manufactured since 1950s based upon the structure of natural pyrethrins, which are 5 very unstable in the environment, due to oxidation and UV radiation [1-3]. Since then, 6 pyrethroids are widely applied as insecticides in households and greenhouses, as well as 7 for the protection of crops. Nevertheless, laboratory tests showed that pyrethroids are 8 toxic for fish, aquatic arthropods and honey bees [4,5]. 9 Releases to the air represent the most important emission pathway for 10 pyrethroids. Because of that, inhalation is an important route of exposure for humans, 11 especially just after spraying application in domestic indoors or agricultural close areas. 12 The Occupational Safety and Health Administration (OSHA) has established the 13 occupational exposure limit for an 8-hour workday, 40-hour workweek, at 5 mg of 14 pyrethrins and pyrethroids per cubic meter of workplace air (5 mg/m3) [1]. 15 Among pyrethroids available today, allethrin, phenothrin, tetramethrin and 16 cyphenothrin are mainly used for household insects, and cypermethrin, deltamethrin, 17 permethrin, λ-cyhalothrin and cyfluthrin are usually utilized for agricultural purposes. 18 Pyrethroids are often commercially combined with synergist compounds like piperonyl 19 butoxide, which enhance their insecticidal activity, or mixed with fungicides as 220 phenylphenol or other pesticides like propoxur (a carbamate pesticide). They can be 21 found in indoor air, sometimes in much higher concentrations than pyrethroids; so, their 22 simultaneous determination in air could be of interest. 23 Due to the commonly low concentration of pesticides in air, sampling usually 24 consists on collecting high volumes of contaminated air using sampling cartridges filled 25 4 with one or more adsorbents where the compounds are retained and then, an appropriate 1 solvent is required, usually at high volumes, to quantitatively elute the analytes. This, in 2 turns, leads to time-consuming steps for concentration and clean up of the organic 3 extracts with the risk of analyte losses. In the particular case of pyrethroids, an 4 additional problem could arise from the possible photodecomposition of certain of these 5 compounds, which has been reported in some multi-pesticide studies [6-8], showing that 6 determination of pyrethroids in air might require performing a rapid and careful 7 trapping-extraction process. 8 Polyurethane foam (PUF) [6], Empore disks [7], Silicagel [8], Chromosorb 106 [9], 9 Tenax [9,10] and mixtures of PUF and Tenax [11], are adsorbents used to retain certain 10 pyrethroids together with other pesticides in air. Five pyrethroids were retained on 11 Cambridge filter discs after collecting the cigarette smoke [12]. Other adsorbents used 12 to retain pyrethroids from other matrixes than air are graphitised carbon black (GCB) 13 from oils, lipids and fat [13] or Florisil from water [14]. Tenax TA or Florisil have been 14 employed to trap other pesticides different from pyrethroids from air [15,16]. 15 In addition to classical solvent extraction techniques, other extraction techniques 16 have been used to extract some pyrethroids from different matrices, such as 17 supercritical-fluid extraction (SFE) from wool [17] or honey [18], microwave-assisted 18 extraction from soil [19] or matrix solid-phase dispersion (MSPD) from juice samples 19 [20]. The non-exhaustive extraction technique solid-phase microextraction (SPME) is 20 relatively new and could constitute an alternative to solvent-based extraction 21 methodologies [21]. Recently, SPME has been applied for the determination of several 22 pesticides including bioallethrin in confined atmospheres [22]. In addition, the 23 possibilities of SPME to the analysis of pyrethroids in air have been fully studied by the 24 authors and reported elsewhere [23]. 25 5 Gas chromatography with electron-capture detection (GC-ECD) and GC coupled 1 to mass spectrometry (MS) or tandem MS are the techniques of choice for the analysis 2 of semivolatile pesticides including pyrethroids [6-9,24,25), although the use of liquid 3 chromatography (LC) with fluorescence and MS detection has also been reported [264 28]. 5 The aim of the present study is to develop a fast method for the determination of 6 the components of highly consumed insecticide formulations used for agriculture or 7 indoors, with particular interest in pyrethroid compounds, with enough sensitivity to 8 detect these compounds at their low concentrations in air. For this purpose, 11 9 pyrethroid insecticides, a fungicide (2-phenylphenol), a carbamate pesticide (propoxur) 10 and an insecticide synergist (piperonyl butoxide) have been selected. The method 11 proposed is based on the use of a very low amount of adsorbent to retain the 12 compounds, and the rapid ultrasound-assisted solvent extraction using a very low 13 volume of solvent. No more exhaustive sample preparation was required to 14 quantitatively extract the target analytes from the adsorbent. The optimization of the 15 methodological parameters was carried out using an experimental design approach to 16 study the main factors and possible factor interactions. Limits of detection lower than 1 17 ng/m3were achieved using GC-µECD. In addition, the proposed method was applied to 18 real samples collected in contaminated closed rooms, in which some of the target 19 compounds have been determined. 20 21 2. Experimental 22 23 2.1. Reagents 24 6 Cypermethrin (mixture of isomers) and deltamethrin were supplied by Supelco 1 (Bellefonte, PA, USA). 2-hydroxybiphenyl (2-phenylphenol), cyphenothrin (mixture of 2 cis and trans isomers), allethrin (mixture of stereo isomers), transfluthrin, empenthrin, 3 cyfluthrin (mixture of isomers), piperonyl butoxide, tetramethrin, permethrin (mixture 4 of cis and trans isomers), phenothrin (mixture of isomers), propoxur and λ-cyhalothrin 5 were of pestanal grade and were purchased to Riedel-de-Haën (Seelze, Germany). All 6 organic solvents (acetone, n-hexane and ethyl acetate) were of pesticide grade and were 7 obtained from Merck (Mollet del Vallés, Barcelona, Spain). 8 Individual standard stock solutions of 9 000-11 000 µg/mL were prepared in 9 acetone, and a stock mixture solution of all target analytes at a concentration of 100 10 µg/mL was obtained by appropriate dilution of individual standard solutions in acetone. 11 All working solutions were prepared by convenient dilution of the stock mixture 12 solution in hexane or ethyl acetate. All solutions were stored in amber-colored vials at - 13 20ºC. 14 15 2.2. Air sampling and extraction of analytes 16 The air-sampling device is similar to that previously applied by the authors to 17 determine polychlorinated biphenyls in air [29]. Using a vacuum pump working at 100 18 L/min (Telstar model S-8, Tarrasa, Spain), a known volume of air was pumped through 19 a glass tube containing 25 mg of an adsorbent. In this study, Tenax TA of mesh size 6020 80 (Supelco) and Florisil (activated overnight at 105ºC) of 60-100-mesh size (Aldrich, 21 Steinheim, Germany) have been used as adsorbents. Teflon (PTFE) tubing was used for 22 all connections. The adsorbent was then poured into a 22-mL glass vial and analytes 23 were extracted into an appropriate volume of organic solvent (n-hexane or ethyl acetate) 24 using an ultrasound bath (J.P. Selecta, Barcelona, Spain) for a few minutes. After 25 7 filtered through a 0.45 µm Millex HV filter (13 mm diameter) (Millipore, Bedford, 1 USA), the extract (1to 2-µL) was then injected into the chromatographic injection port. 2 To study the retention and extraction efficiencies of target compounds, a volume 3 of 100 µL of standard mixtures of the analytes in n-hexane were directly spiked on 25 4 mg of the adsorbent. The spike was left to homogenize at room temperature for two 5 hours and then, it was treated as described above in this section. Air sample volume 6 ranged from 1 to 10 m3. 7 To detect a possible breakthrough of the adsorbent, some experiments required 8 the coupling of a second glass tube filled with 25 mg of the adsorbent (blank) to the first 9 spiked one. Both portions of adsorbent were individually extracted. 10 11 2.3. Gas chromatographic analysis 12 Gas chromatography with micro electron-capture detection (GC-µECD) analysis 13 was performed in an Agilent Technologies 6890N Network GC System, operated by 14 GC Chemstation software and equipped with a split/splitless injector. A HP-5 (30 m x 15 0.32 mm I.D.. 0.25 µm film thickness) column was used to separate the pyrethroids. 16 Nitrogen was employed as the carrier and make-up gas with a constant pressure of 12 17 psi (a flow of 2.5 mL/min at 60ºC). The GC oven temperature was: 60ºC hold 2min, 18 first rate 20ºC/min to 230ºC, second rate 5ºC/min to 270ºC hold 5min, third rate 19 5ºC/min to 290ºC with a total acquisition program of 27.5 min. Detector and injector 20 temperatures were set at 285ºC and 270ºC, respectively. Injector operated in the 21 splitless mode and programmed to return to the split mode after 2 min from the 22 beginning of a run. 23 The GC-MS analyses were performed on a Varian 3800 gas chromatograph 24 (Varian, Walnut Creek, CA, USA) equipped with an ion trap mass detector Varian 25 8 Saturn 2000 (Varian). The system was operated by Saturn GC-MS WorkStation v5.4 1 software. Analytes were separated on a 25 m length x 0.25 mm I.D., CP-Sil8 CB Low 2 bleed/MS column coated with a 0.25 µm film. Helium (purity 99,999%) was employed 3 as carrier gas, with a constant column flow of 1.2 mL/min. Injector was operated in the 4 splitless mode (2 min). The GC oven temperature program was similar to that used for 5 GC-ECD. Injector temperature was held constant at 270ºC. Trap, manifold and 6 transfer line temperatures were maintained at 250ºC, 120ºC and 300ºC, respectively. 7 The ion-trap mass spectrometer was operated in the electron ionisation mode (70 eV). 8 The mass range was scanned in the full scan mode from 70 to 270 m/z. Experimental 9 parameters for ionisation were, multiplier voltage, 1750 V; filament emission current, 10 12 A; axial modulation voltage, 4V; ionisation control, automatic mode; 11 filament/multiplier delay, 8 min. From the total ion current chromatograms, one ion or a 12 group of ions were selected for quantification and are presented in Table 1. 13 14 3. Results and discussion 15 16 Most of compounds considered in this study showed analytical response in GC17 µECD. Nevertheless, some of the target compounds (phenothrin, 2-phenylphenol, 18 propoxur and piperonyl butoxide) did not show measurable signals with this detector. 19 Then, methodology was also developed for GC-MS. In Figures 1 and 2, chromatograms 20 obtained using GC-MS in the fullscan mode and GC-µECD for standard mixtures of 21 compounds at 1000 ng/mL and 10 ng/mL, respectively, are shown. It can be noticed that 22 some of the target pyrethroids, such as cyphenothrin, cyfluthrin and cypermethrin, gave 23 isomeric peak clusters. 24 Linearity was evaluated for each chromatographic system. Five concentration 25 9 levels ranging from 1 to 100 ng/mL for the GC-µECD system (10-100 ng/mL for 1 empenthrin), and six concentration levels ranging from 10 to 5000 ng/mL for GC-MS 2 system (50-5000 ng/mL for cyfluthrin, cypermethrin and deltamethrin) have been 3 considered. In both cases, two replicates have been obtained, and the correlation 4 coefficients were higher than 0.999 for all compounds (see table 1). To validate the 5 regression data, an analysis of variance (ANOVA) with a lack-of-fit test was run [30]. 6 This test allows determining whether the selected model is adequate to describe the 7 observed data, or whether a more complicated model should be used. The test is 8 performed by comparing the variability of the proposed model residuals to the 9 variability between observations (chromatographic response) at replicate values of the 10 independent variable (known concentration of compounds in the standard solutions). 11 Table 1 shows the results of F-ratio and P-values obtained in the calibration range 12 considered. As can be seen, P-values were greater than 0.05 for all compounds and thus, 13 linear regression models were adequate for the obtained data at a confidence level of 14 95%. 15 16 3.1. Optimization of the ultrasound-assisted solvent extraction 17 Desorption step determines the efficiency of the final method and then, 18 experimental work was initially focused on the optimization of the ultrasound-assisted 19 extraction process using an experimental design approach. 20 Four factors were selected and studied at two levels: type of adsorbent, type and 21 volume of extracting solvent and ultrasounds application time. The factors selected and 22 their levels are presented in Table 2. Tenax TA and Florisil were the choice for the two 23 levels of factor type of adsorbent. The efficiency of Tenax TA in the retention of some 24 organic pollutants in air, even at such little amounts as 25-mg, was previously reported 25 16 A very simple and sensitive method to analyze pyrethroids as well as other 1 components of domestic insecticide preparations in indoor air was developed. The 2 method is based on the active retention of the target compounds on a very small amount 3 of Tenax TA and the subsequent desorption by application of ultrasound using only 14 mL ethyl acetate, avoiding for the requirements of extract concentration prior to the 5 chromatographic analysis. Optimization of the extraction step was achieved by an 6 experimental design approach studying four factors (type of adsorbent and type of 7 solvent, solvent volume and extraction time) at two levels. Retention efficiency was 8 studied, finding that no breakthrough occurred for any compound sampling 1 m3air. 9 The study of method performance demonstrated its linearity, quantitative recoveries, 10 and good sensitivity, with LODs lower than 1 ng/m3using GC-µECD. The use of GC11 MS allowed the simultaneous determination of pyrethroid and non-pyrethroid 12 compounds. In addition, the proposed method allows high sample throughput since the 13 total sampling-extraction-analysis process is completed within one hour. 14 The analysis of several contaminated air samples demonstrated the validity of 15 the proposed method for the analysis of the target compounds in indoors atmospheres. 16 17 Acknowledgements 18 19 This research was supported by the projects BQU2003-02090 from CICYT, 20 Spanish Commission for Research and Development, (Ministerio de Ciencia y 21 Tecnologia) and PGIDIT04PXIC23701PN and PGIDIT05RAG50302PR from Xunta de 22 Galicia. R. B. would like to acknowledge her Doctoral grant to the Xunta de Galicia. 23 24 25 17 References 1 [1] Toxicological Profile for pyrethrins and pyrethroids, US Department of Health and 2 Human Services, Public Health Service, Agency for Toxic Substances and Disease 3 Registry (ATSDR), Atlanta, GA, 2003. 4 [2] International Labour Organisation and World Health Organization (WHO), United 5 Nations Environment Programme, The International Programme on Chemical Safety, 6 Environmental Health Criteria Nos. 64, 82, 87, 94, 96-99 and 142, Geneva, 1986-1992. 7 [3] Y. L. Chen, J. E. Casida, J. Agr. Food Chem. 17 (1969) 208-215. 8 [4] J. R. Clark, L. R. Goodman, P. W. Borthwick, J. M. Patrick Jr., G. M. Cripe, P. M. 9 Moody, J. C. Moore, E. M. Lores, Environ. Toxicol. Chem. 8 (1989) 393-401. 10 [5] J. R. Coats, D. M. Symonik, S. P. Bradbury, S. D. Dyer, L. K. Timson, G. J. 11 Atchison, Environ. Toxicol. Chem. 8 (1989) 671-679. 12 [6] L. Elflein, E. Berger-Preiss, K. Levsen, G. Wünsch, J. Chromatogr. A 985 (2003) 13 147-157. 14 [7] T. Yoshida, I. Matsunaga, H. Oda, J. Chromatogr. A 1023 (2004) 255-269. 15 [8] T. J. Class, J. Kintrup, Fresenius J. Anal. Chem. 340 (1991) 446-453. 16 [9] F. J. Egea Gonzalez, A. Mena Granero, C. R. Glass, A. Garrido Frenich, J. L. 17 Martinez Vidal, Rapid Commun. Mass Spectrom. 18 (2004) 537-543. 18 [10] J. Siebers, P. Mattusch, Chemosphere 33 (1996) 1597-1607. 19 [11] R. G. Lewis, K. E. MacLeod, Anal. Chem. 54 (1982) 310-315. 20 [12] J. Cai, B. Liu, X. Zhu, Q. Su, J. Chromatogr. A 964 (2002) 205-211. 21 [13] A. Ramesh, M. Balasubramanian, Analyst 123 (1998) 1799-1802. 22 [14] W. R. Barrionuevo, F. M. Lanças, Bull. Environ. Contam. Toxicol. 69 (2002) 12323 128. 24 18 [15] R. Barro, S. Ares, C. Garcia-Jares, M. Llompart, R. Cela, J. Chromatogr. A 1072 1 (2005) 99-106. 2 [16] J. Castro, R. A. Pérez, E. Miguel, C. Sánchez-Brunete, J. L. Tadeo, J. Chromatogr. 3 A 947 (2002) 119-127. 4 [17] A. M. Nguyen, P. J. Marriot, J. Hughes, J. Biochem. Biophys. Methods 43 (2000) 5 411-429. 6 [18] S. R. Rissato, M. S. Galhiane, F. R. N. Knoll, B. M. Apon, J. Chromatogr. A 1048 7 (2004) 153-159. 8 [19] F. A. Esteve-Turrillas, C. S. Aman, A. Pastor, M. de la Guardia, Anal. Chim. Acta 9 522 (2004) 73-78. 10 [20] B. Albero, C. Sánchez-Brunete, A. Donoso, J. L. Tadeo, J. Chromatogr. A 1043 11 (2004) 127-133. 12 [21] J. Pawliszyn, Solid Phase Microextraction, Theory and Practice, Wiley, New York, 13 1997. 14 [22] F. Ferrari, A. Sanusi, M. Millet, M. Montury, Anal. Bioanal. Chem. 379 (2004) 15 476-483. 16 [23] R. Barro, C. Garcia-Jares, M. Llompart, R. Cela, J. Chromatogr. Sci. (2006) (in 17 press). 18 [24] E. Berger-Preiss, A. Preiss, K. Sielaff, M. Raabe, B. Ilgen, K. Levsen, Indoor Air 7 19 (1997) 248-261. 20 [25] G. Leng, E. Berger-Preiss, K. Levsen, U. Ranft, D. Sugiri, W. Hadnagy, H. Idel, 21 Int. J. Hyg. Environ.-Health 208 (2005) 193-199. 22 [26] T. López-López, M. D. Gil-Garcia, J. L. Martínez-Vidal, M. Martínez-Galera, 23 Anal. Chim. Acta 447 (2001) 101-111. 24 19 [27] C. Ferrer, M. J. Gómez, , J. F. García-Reyes, I. Ferrer, E. M. Thurman, A. R. 1 Fernández-Alba, J. Chromatogr. A 1069 (2005) 183-194. 2 [28] I.-H. Wang, R. Moorman, J. Burleson, J. Chromatogr. A 983 (2003) 145-152. 3 [29] R. Barro, S. Ares, C. Garcia-Jares, M. Llompart, R. Cela, Anal. Bioanal. Chem. 4 381 (2005) 255-260. 5 [30] Statgraphics-Plus V. 5.1. Experimental design, Appendix C, Manugistics, 6 Rockville, MD, 2002. 7 [31] J. You, D. P. Weston, M. J. Lydy, Arch. Environ. Contam. Toxicol. 47 (2004) 1418 147. 9 10 20 Figure captions 1 2 Figure 1. Total ion current (TIC) chromatogram of a standard solution of the target 3 analytes at 1 µg/mL. 1: 2-phenylphenol, 2: propoxur, 3: empenthrin, 4: transfluthrin, 5: 4 allethrin, 6: piperonyl butoxide, 7: tetramethrin (2 peaks), 8: phenothrin, 9: λ5 cyhalothrin (2 peaks), 10: cyphenothrin (3 peaks), 11: permethrin, 12: cyfluthrin (4 6 peaks), 13: cypermethrin (4 peaks), 14: deltamethrin (2 peaks). 7 8 Figure 2. µECD chromatogram of a standard solution of the target analytes at 10 ng/mL. 9 1: empenthrin, 2: transfluthrin, 3: allethrin, 4: tetramethrin (2 peaks), 5: λ-cyhalothrin (2 10 peaks), 6: cyphenothrin (3 peaks), 7: permethrin, 8: cyfluthrin (4 peaks), 9: 11 cypermethrin (4 peaks), 10: deltamethrin (2 peaks). 12 13 Figure 3. Main effects plots for two selected pyrethroids: tetramethrin and λ14 cyhalothrin. 15 16 Figure 4. Combined effect of factors type of adsorbent and type of solvent (A) and type 17 of solvent and solvent volume (B), for two selected pyrethroids: 18 tetramethrin and lambda-cyhalothrin. 19 20 Figure 5. Variation of the chromatographic response with the volume of air sampled for 21 the target pyrethroids (GC-ECD), as well as phenothrin and the non-pyrethroid 22 compounds (GC-MS). 23 24 21 Figure 6. GC-µECD chromatogram of a contaminated air sample collected following 1 the proposed method in a closed room sprayed with insecticide aerosols and treated with 2 anti-mosquito electro-evaporators. * Quantified on a diluted sample. 3 4 Figure 7. GC-MS analysis of a contaminated air sample collected following the 5 proposed method in a closed room sprayed with insecticide aerosols and treated with 6 anti-mosquito electro-evaporators. 7 8 Table 1. Study of linearity using two different detection systems. MS detection µECD detection Quantification ions Correlation coefficient (R) Fratio Pvalue Correlation coefficient (R) Fratio Pvalue 2-Phenylphenol 169+170 0.9998 0.22 0.9198 n.d. n.d. n.d. Propoxur 110+152 0.9993 0.29 0.8719 n.d. n.d. n.d. Empenthrin 123 0.9991 0.02 0.9992 0.9992 3.43 0.1609 Transfluthrin 163 0.9990 1.19 0.4030 0.9999 0.03 0.9982 Allethrin 123 0.9999 5.63 0.0642 0.9999 0.32 0.8536 Piperonyl butoxide 176 0.9997 0.37 0.8230 n.d. n.d. n.d. Tetramethrin 164 0.9998 0.04 0.9958 1.0000 1.97 0.2188 Phenothrin 123+183 0.9998 0.39 0.8066 n.d. n.d. n.d. λ-Cyhalothrin 181+197 0.9998 1.49 0.3158 0.9998 4.42 0.0526 Cyphenothrin 123 0.9998 0.49 0.7475 0.9999 0.21 0.9261 Permethrin 183 0.9997 0.03 0.9973 0.9997 2.30 0.1733 Cyfluthrin 163 0.9998 6.61 0.0539 0.9998 1.76 0.2557 Cypermethrin 163 0.9998 5.50 0.0711 0.9997 0.13 0.9646 Deltamethrin 253+181 0.9995 10.5 0.0838 0.9996 0.92 0.5085 Table 2. Factors and levels considered in the experimental design. Factor Code Low level High level Continuous Solvent A Ethyl acetate n-hexane No Solvent volume B 1 mL 3 mL Yes Adsorbent C Tenax Florisil No Extraction time D 10 min 20 min Yes Table 3. Analysis of variance (ANOVA) showing the significance of main effects and their second order interactions. Factors A: Solvent B: Solvent volume C: Adsorbent D: Extraction time F-ratio p-value F-ratio p-value F-ratio p-value F-ratio p-value Empenthrin 10.5 0.023 4.54 0.086 53.9 0.001 1.19 0.326 Transfluthrin 393 0.000 32.9 0.002 129 0.000 11.1 0.021 Allethrin 1239 0.000 47.8 0.001 683 0.000 1.77 0.241 Tetramethrin 918 0.000 0.00 0.980 303 0.000 2.41 0.182 λ-Cyhalothrin 857 0.000 62.7 0.000 309 0.000 2.52 0.173 Cyphenothrin 698 0.000 9.16 0.029 277 0.000 1.57 0.266 Permethrin 443 0.000 0.15 0.713 189 0.000 6.02 0.058 Cyfluthrin 535 0.000 11.2 0.020 196 0.000 3.41 0.124 Cypermethrin 386 0.000 9.57 0.027 137 0.000 3.82 0.108 Deltamethrin 269 0.000 10.7 0.022 123 0.000 3.26 0.131 Interactions AB AC BC F-ratio p-value F-ratio p-value F-ratio p-value Empenthrin 8.18 0.035 47.4 0.001 5.48 0.066 Transfluthrin 77.2 0.000 145.4 0.000 22.1 0.005 Allethrin 35.1 0.002 674 0.000 3.88 0.106 Tetramethrin 3.42 0.124 361 0.000 3.64 0.115 λ-Cyhalothrin 91.1 0.000 161 0.000 6.49 0.051 Cyphenothrin 37.0 0.002 179 0.000 0.75 0.425 Permethrin 32.4 0.002 103 0.000 3.87 0.106 Cyfluthrin 38.4 0.002 65.1 0.000 0.52 0.505 Cypermethrin 24.9 0.004 38.9 0.002 1.90 0.226 Deltamethrin 24.1 0.004 18.8 0.008 0.15 0.712 Table 4. Extraction efficiency of pyrethroids from Tenax at three spike levels. Analyses were performed by GC-ECD. 2 ng (n= 3) 10 ng (n= 3) 100 ng (n= 3) Recovery (%) R.S.D. (%) Recovery (%) R.S.D. (%) Recovery (%) R.S.D. (%) Empenthrin - - 111 12 97 2.9 Transfluthrin 97 5.6 95 2.5 99 2.5 Allethrin 100 4.1 96 2.0 104 3.2 Tetramethrin 114 4.1 103 5.5 99 7.1 λ-Cyhalothrin 95 5.7 103 8.0 105 7.0 Cyphenothrin 99 4.4 114 4.4 106 7.2 Permethrin 96 5.1 109 3.8 106 6.5 Cyfluthrin 101 6.9 108 1.8 97 8.8 Cypermethrin 101 7.6 103 8.0 102 6.6 Deltamethrin 110 7.5 118 6.7 97 6.4 Figure 3 Figure 4 tetramethrin 0 2 4 6 8 response (area cts x 104) ethyl acetate n-hexane Tenax Florisil λ-cyhalothrin 0 4 8 12 16 20 24 response (area cts x 104) ethyl acetate n-hexane Tenax Florisil 7 11 15 19 23 λ-cyhalothrin response (area cts x 104) ethyl acetate n-hexane 1 mL 1 mL 3 mL 3 mL (A) (B) 21 31 41 51 61 71 response (area cts x 103) tetramethrin ethyl acetate n-hexane 1 mL 3 mL 1 mL 3 mL Figure 5 0 20 40 60 80 100 transfluthrin λ-cyhalothrin permethrin cyfluthrin cypermethrin deltamethrin relative response (%) 0 20 40 60 80 100 empenthrin allethrin tetramethrin cyphenothrin relative response (%) 0 20 40 60 80 100 2-phenylphenol propoxur piperonyl butoxide phenothrin relative response (%) 1 m3 2.5 m3 5 m3 10 m3 Figure 6 minutes 13 14 15 16 17 18 19 20 21 Hz 500 1000 1500 2000 2500 3000 allethrin + bioallethrin (3.0 ng/m3) tetramethrin (176 ng/m3) cyphenothrin (85 ng/m3) permethrin (3.3 ng/m3) cyfluthrin (4.3 ng/m3) cypermethrin (22 ng/m3) minutes 13 14 15 16 17 18 19 20 21 Hz 500 1000 1500 2000 2500 3000 allethrin + bioallethrin (3.0 ng/m3) tetramethrin (176 ng/m3) cyphenothrin (85 ng/m3) permethrin (3.3 ng/m3) cyfluthrin (4.3 ng/m3) cypermethrin (22 ng/m3) * Figure 7 10.0 12.5 15.0 17.5 minutes kCounts 0 1 2 3 4 5 6 0 25 50 75 100 0.0 0.5 1.0 1.5 0.0 2.5 5.0 7.5 10.0 12.5 0 25 50 75 100 0.0 2.5 5.0 7.5 0 5 10 15 Ions: 169+170 Ions: 110+152 Ion: 123 Ion: 176 Ion: 164 Ion: 123+183 Ion: 123 2-phenylphenol (84 ng/m3) propoxur (863 ng/m3) allethrin (33 ng/m3) piperonyl butoxide (151 ng/m3) tetramethrin (1651 ng/m3) phenothrin (81 ng/m3) cyphenothrin (603 ng/m3) 10.0 12.5 15.0 17.510.0 12.5 15.0 17.5 minutes kCounts 0 1 2 3 4 5 6 0 25 50 75 100 0.0 0.5 1.0 1.5 0.0 2.5 5.0 7.5 10.0 12.5 0 25 50 75 100 0.0 2.5 5.0 7.5 0 5 10 15 Ions: 169+170 Ions: 110+152 Ion: 123 Ion: 176 Ion: 164 Ion: 123+183 Ion: 123 2-phenylphenol (84 ng/m3) propoxur (863 ng/m3) allethrin (33 ng/m3) piperonyl butoxide (151 ng/m3) tetramethrin (1651 ng/m3) phenothrin (81 ng/m3) cyphenothrin (603 ng/m3)