Food Chemistry 113 (2009) 1297–1300

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Analytical methods

Simplified pesticide multiresidues analysis in fish by low-temperature cleanup

and solid-phase extraction coupled with gas chromatography/mass spectrometry
Shubing Chen *, Xuejun Yu, Xiaoyu He, Donghua Xie, Yuanmu Fan, Jinfeng Peng
Food Safety Testing Subsection, Inspection and Quarantine Technology Center, Ningbo Entry-Exit Inspection and Quarantine Bureau of the People’s Republic of China,
Ningbo 315012, PR China

a r t i c l e i n f o a b s t r a c t

Article history: A simple and fast method for the simultaneous analysis of thiobencarb, deltamethrin and 19 organochlo-
Received 25 April 2008 rine pesticide residues in fish by gas chromatography–mass spectrometry was investigated in this study.
Received in revised form 16 August 2008 Samples are extracted with acetonitrile. Most of lipids in the extract are eliminated by low-temperature
Accepted 18 August 2008
cleanup, prior to solid-phase extraction cleanup. The lipids extracted from the fish samples were easily
removed without any significant losses of the pesticides. Aminopropyl (NH2) cartridge was effective to
eliminate the remaining interference. Spiked experiments were carried out to determine the recovery,
Keywords:
precision and limits of detection (LODs) of the method. The method detection limits ranged from
Thiobencarb
Deltamethrin
0.5 lg kg1 to 20 lg kg1, whilst recoveries of the pesticides were in the range of 81.3–113.7% with rela-
Organochlorine pesticides tive standard deviations 613.5% at a spiked concentration of 0.05 mg kg1, 0.02 mg kg1 and 0.1 mg kg1.
Low-temperature cleanup The newly developed method is demonstrated to give efficient recoveries and LODs for detecting pesticide
Fish multiresidues in fish.
Gas chromatography–mass spectrometry Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction approaches have been developed to eliminate co-extracted lipid

interferences, including liquid-liquid partitioning (Manirakiza,
Although banned for several decades, organochlorine pesticides Covaci, & Schepens, 2001), gel permeation chromatography (Hong,
(OCPs) are still detectable in fish tissue today because of their high Eo, Rhee, Kim, & Kim, 1993), column chromatography (Hong et al.,
lipid solubility and chemical stability (Eaton & Lydy, 2000). In addi- 1993; Ling, Chang, & Huang, 1994; Manirakiza, Covaci,
tion to the long-lived OCPs, current-use pesticides like pyrethroid Nizigiymana, Ntakimazi, & Schepens, 2002; Muir, Ford, Grift, Met-
and herbicide are also a concern because of their extensive use ner, & Lockhart, 1990) and multiple clean-up methods (Smith,
and high toxicity to nontarget species, and its high bioaccumula- Stalling, & Johnson, 1984). However, most of these methods are
tion in fish tissue (Hancock, Yasin, Baugh, Bonwick, & Davies, time consuming and need large quantities of organic solvents or
1997). Large quantities of aquatic products are exported from an expensive instrument or complex operation.
China to Japan, European Union (EU) and other countries every The low-temperature cleanup method was ever introduced to
year, so the monitoring of pesticide residues in aquatic products determine organophosphorus pesticides in olive oil (Lentza-Rizos,
is rather stringent in our laboratory. Therefore, an effective and Avramides, & Cherasco, 2001) and organochlorine pesticides in fish
simple method to analyse pesticide multiresidues simultaneously (Hong, Kim, Kim, Seo, & Kim, 2004). There is a significant difference
in fish is necessary. of melting points between lipids and pesticides. After extraction,
Generally, aquatic products often have high lipid content, and lipids in organic extracts are precipitated as frozen form at sub-
most of pesticides are lipophilic compounds, so their extraction zero temperature in the freezer, whilst pesticides are still dissolved
from aquatic products is accompanied by considerable amounts in the organic solvents. Therefore most of lipid components can be
of fatty material. During instrumental analysis, lipid components easily separated from pesticides. However, the remaining lipids in
tend to adsorb in GC system such as injection port and column, the solvent would still interfere the determination after low-tem-
resulting in poor chromatographic performance. So, for the deter- perature cleanup. So a further cleanup step was needed to remove
mination of lipophilic pesticide residues in fatty extracts, the cru- the remaining interference. Solid-phase extraction (SPE) is being
cial step in the analytical procedure is the separation of the increasingly used in food analysis, mainly for sample clean-up be-
pesticide residues from the bulk of the lipidic material. Several cause it requires small solvent volumes, requires no specialised
equipment, is easy to operate and has a rapid sample throughput.
* Corresponding author. Tel.: +86 574 87022808; fax: 86 574 87123765. A Florisil cartridge was employed to clean the remaining interfer-
E-mail address: [email protected] (S. Chen). ence after the low-temperature cleanup to determine chlorinated

0308-8146/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2008.08.045
1298 S. Chen et al. / Food Chemistry 113 (2009) 1297–1300

pesticides in fish (Hong et al., 2004). A tandem SPE method was rinsed with 2 mL acetonitrile-toluene (3:1) twice, and the wash-
used for clean-up of 10 OCPs in wildlife tissues (Volz & Johnston, ings were also applied to the cartridge. A reservoir was attached
2002). A tandem C18 and Florisil cartridges was applied to deter- to the cartridge and the pesticides were eluted with 25 mL aceto-
mine pyrethroid, organophoshpate and organochlorine pesticides nitrile-toluene (3:1). The eluate was concentrated to ca 0.5 mL by
in fish tissue (You & Lydy, 2003). rotary evaporation at 40 °C and then evaporated to dryness with
The purpose of the current study is to develop a simple and fast a nitrogen stream. The residue was dissolved in 1 mL hexane for
method that would permit the determination of thiobencarb, del- GC–MS analysis.
tamethrin and 19 organochlorine residues in fish by gas chroma-
tography–mass spectrometry. This developed method can be 2.4. GC–MS analysis
applied for the monitoring of pesticide residues in a mass of ex-
ported fish from China. GC–MS was performed with an Agilent 5973 instrument
equipped with EI and a 30 m  0.25 mm capillary column coated
2. Materials and methods with a 0.25 lm film of DB-1701. Column temperature was main-
tained at 40 °C for 1 min, then programmed at 30 °C min1–
2.1. Materials and reagents 130 °C, then at 5 °C min1–250 °C, and finally at 10 °C min1–
300 °C, which was held for 5 min. Helium, purity>99.999%, was
The organochlorine pesticides included a-, b-, c- and d-hexa- used as carrier gas at a flow rate of 1.2 mL min1. The injection port
chlorcyclohexane (HCH), p,p0 -DDD, p,p0 -DDE, o,p0 -DDT, p,p0 -DDT, temperature was 260 °C and 1 lL samples were injected splitless
hexachlorobenzene (HCB), quintozene, dicofol, endosufan I, with the purge on after 1.5 min. The MS ionisation energy was
endosufan II, endosulfan-sulphate, aldrin, heptachlor, heptachlor 70 eV, the ion-source temperature 230 °C, and the GC–MS interface
epoxide, endrin, dieldrin. The pyrethroid analysed in this study is temperature 280 °C. The selected ion monitoring (SIM) was used
deltamethrin, and the herbicide is thiobencarb. Pesticide standards and the dwell time of each ion was set at 100 ms. To improve sen-
were purchased from Sigma (Poole, UK). sitivity, selected ions used in the SIM mode are divided into four-
Individual stock standard solution of pesticide was prepared by teen groups, guiding by the individual pesticide retention times.
dissolving 10 mg of each compound in 10 mL hexane and stored in The selected ion groups in SIM mode were listed in Table 1. All pes-
amber bottles. A mixed standard solution was prepared from the ticides were identified by retention time and specific ions, and
individual stock solutions with a concentration of 100 mg L1. A ser- quantified by the external standard method.
ies of calibration standards were prepared by diluting 100 mg L1 of
the mixed standard solution to produce a final concentration of 0.1, 2.5. Fortified recovery studies
0.2, 0.5, 1.0, 2.0 mg L1 in hexane. Stock and working solutions were
stored at 4 °C and used for no longer than 3 months and 1 week, Untreated aquatic samples were fortified, on average at 0.02,
respectively. 0.05, 0.1 mg kg1, by adding standard solution. Samples were al-
Acetone, n-hexane, methylene chloride, toluene and acetoni- lowed to equilibrate for 30 min prior to extraction, and were pro-
trile, of pesticide grades, were purchased from Dikma (Ontario, cessed according to procedure described above. The recovery
Canada). Anhydrous sodium sulphate and sodium chloride are both assays were replicated five times. All samples were treated and
of analytical reagent grade. Before use, sodium sulphate was analysed using GC–MS–SIM mode described above.
heated at 650 °C for 4 h and kept in a desiccator. Distilled water
was obtained with a Milli-Q system (Millipore, Bedford, MA, 3. Results and discussion
USA). For SPE, aminopropyl (NH2) cartridge was purchased from
Waters. 3.1. Extraction and cleanup

2.2. Sample preparation Both the selection of solvent and extraction method can be crit-
ical in obtaining a satisfactory recovery of the target pesticides from
Ten grams of previously minced fish sample was placed into a
50 mL centrifuge tube and mixed with 3.0 mL water. The mixture
Table 1
was vortexed for 1 min. A 20 mL aliquot of acetonitrile was added
Selected ion groups of the pesticides in GC–MS–SIM analysis
as extraction solvent. The resulting mixture was stirred for 15 min.
Add 5 g sodium chloride to the mixture and vortex for another Compounds Time (min) Quantified ion (m/z) Confirm ion (m/z)
2 min, then centrifuge for 5 min at 4000 rpm. 10 mL of the extrac- HCB 15.05 284 286, 282
tion solution was collected in 100 mL round flask, which was a-HCH 16.9 181 219, 217
stored in the freezer at 24 °C for 20 min to freeze lipids. Most Quintozene 17.2 237 249, 295
d-HCH 18.6 181 219, 217
of the lipids were precipitated as pale yellow, condense lump on Heptachlor 19.4 272 237, 337
the flask surface. Cold extract at 24 °C was immediately filtered Aldrin 20.45 263 265, 293
with filter paper to remove frozen lipids. The precipitated lipid b-HCH 21.6 181 219, 217
on the flask surface was redissolved in 10 mL of acetonitrile to per- Thiobencarb 21.8 100 125, 257
c-HCH 22.5 181 219, 217
form filtration again by same procedure. The filtered extract was
Dicofol 22.6 139 250, 141
concentrated to 1 mL by rotary evaporation to follow SPE Heptachlor epoxide 23.04 353 355, 357
procedure. Endosufan I 24.1 241 265, 339
p,p0 -DDE 24.95 246 318, 316
2.3. Clean-up by NH2 SPE dieldrin 25.4 263 277, 380
endrin 26.2 263 279, 345
o,p0 -DDT 26.55 235 318, 316
Before sample application, anhydrous sodium sulphate (ca p,p0 -DDD 27.8 235 318, 316
1 cm) was placed on top of NH2 cartridge. The cartridge was condi- endosufanII 27.9 195 339, 277
tioned with 10 mL acetonitrile–toluene (3:1). When the condition- p,p0 -DDT 28.4 235 318, 316
Endosulfan-sulphate 30.2 272 387, 389
ing solution reached the top of the sodium sulphate, the
Deltamethrin 37.4 181 253, 172
concentrated extract was added to the cartridge. The flask was
 S. Chen et al. / Food Chemistry 113 (2009) 1297–1300 1299

the sample matrix. In this study, methylene chloride, n-hexane, remaining interference in the extract, which was used to remove
and acetonitrile were used as extraction solvent to extract the target the coextractives such as pigments and lipids (Fillion, Sauve, &
pesticides from the spiked fish samples. No significant difference Selwyn, 2000; Pang et al., 2006). The efficiency of SPE depends
was found amongst these solvents used in extraction, probably on the polarity of the elution solvent and volume (Yi, Hua, &
due to their good dissolving capability for the interesting pesticides. Lu, 2006). It was reported that the combined use of acetonitrile
However, large amounts of lipids were extracted when the methy- and toluene as the eluent could provide a satisfactory recoveries,
lene chloride or n-hexane was used as extraction solvents. When where aminopropyl (NH2) cartridge was also used for cleanup
acetonitrile or acetone was used, the fish tissue was aggregated, (Fillion, Sauve, & Selwyn, 2000; Pang et al., 2006). So, the different
not enable to penetrate fish tissue. Thus, ca. 3 mL water was added volume ratio of the combined acetonitrile-toluene eluent and the
to the fish tissue to homogenise the previously minced fish sample. eluent volume was investigated. As the result, all the compounds
As the result, actonitrile or acetone can easily penetrate the mixture could be eluted from the NH2 cartridge with only 5 mL acetoni-
to perform the extraction. However, acetone is miscible with water, trile–toluene (3:1), except for the following: quintozene (10 mL),
and it is difficult to separate acetone from the water phase. Acetoni- hexachlorobenzene (25 mL).
trile is miscible with water, but it can be separated easily by using
salt. Moreover, acetonitrile has low solubility for lipids. So, acetoni- 3.2. Method application
trile was used as extraction solvent in this study.
When the acetonitrile extract was stored in the freezer at Good linearity was obtained for all the pesticides with a concen-
24 °C for 20 min, most lipid components in the extract solution tration of 0.1, 0.2, 0.5, 1.0, 2.0 mg L1 in hexane. Correlation coeffi-
were precipitated as pale yellow lump on the flask surface, owning cients were higher than 0.998 in all cases. For method validation,
to their low solubility in acetonitrile. On the other hand, the inter- control and fortified samples were extracted, purified and analysed
esting pesticides were soluble even in cold acetonitrile solvent. The under the same conditions. Typical SIM chromatogram obtained
cold extract containing precipitated lipids was promptly filtered from a control sample and a fortified sample extract using low-
with filter paper to prevent melting lipids. To improve the extrac- temperature cleanup and NH2 cartridge is shown in Fig. 1 and
tion yield of the target pesticides, low-temperature cleanup was Fig. 2, respectively. No significant interferences were observed in
repeated once more. SIM chromatogram. Twenty-one pesticides at 0.02 mg kg1 level
A significant amount of lipids was eliminated by low-temper- were successfully detected with excellent sensitivity.
ature cleanup. However, the remaining lipids would still interfere Table 2 shows the recoveries, precision values and the detection
with the determination of the target pesticides. In this study, limits of pesticides obtained in the validation portion of the study.
aminopropyl (NH2) cartridge was also tested to eliminate the As indicated in Table 2, recoveries of pesticides ranged from 78.7%

Fig. 1. SIM chromatogram of control fish extracts purified by low-temperature cleanup and NH2 cartridge and analysed by GC–MS–SIM mode.

Fig. 2. SIM chromatogram of pesticides in spiked fish sample after low-temperature cleanup, NH2 SPE and analysed by GC–MS–SIM mode. Peaks: 1.HCB; 2. a- HCH; 3.
quintozene; 4. d- HCH; 5. Heptachlor; 6. Aldrin; 7. b- HCH; 8. thiobencarb; 9. c- HCH; 10. dicofol; 11. Heptachlor epoxide; 12. endosufan I, endosufanII; 13. p,p0 -DDE; 14.
Dieldrin; 15. Endrin; 16. o,p0 -DDT; 17. p,p0 -DDD + endosufanII; 18. p,p0 -DDT; 19. endosulfan-sulphate; 20. deltamethrin.
1300 S. Chen et al. / Food Chemistry 113 (2009) 1297–1300

Table 2
Recoveries, relative standard deviations (RSD) and detection limits of pesticides at different spiked levels in fish

Compounds Recovery ± RSD (%) LODs (lg kg1)

1 1 1
0.02 (mg kg ) 0.05 (mg kg ) 0.1 (mg kg )
HCB 92.4 ± 8.2 87.2 ± 2.3 93.2 ± 4.3 1.0
a-HCH 85.7 ± 5.8 91.3 ± 3.2 87.7 ± 5.6 2.0
Quintozene 92.4 ± 9.8 87.8 ± 5.6 86.7 ± 7.3 2.5
d-HCH 91.2 ± 8.3 84.1 ± 7.4 94.5 ± 3.4 2.0
Heptachlor 84.4 ± 7.2 90.9 ± 8.4 89.9 ± 2.3 2.5
Aldrin 88.3 ± 4.6 92.5 ± 3.9 89.4 ± 5.3 2.5
b-HCH 91.4 ± 6.7 85.3 ± 6.3 95.3 ± 4.3 2.0
Thiobencarb 101.5 ± 6.9 81.3 ± 6.9 92.3 ± 6.7 0.5
c-HCH 89.4 ± 8.7 82.4 ± 7.6 87.3 ± 4.3 2.0
Dicofol 78.9 ± 9.4 86.9 ± 9.7 85.7 ± 7.4 1.0
Heptachlor epoxide 86.4 ± 6.6 96.8 ± 4.6 91.0 ± 3.4 2.0
Endosufan I 97.1 ± 4.4 88.9 ± 13.5 90.3 ± 5.6 20
p,p0 -DDE 92.7 ± 5.5 88.5 ± 5.5 97.2 ± 6.6 1.0
Dieldrin 93.2 ± 7.2 97.3 ± 10.0 94.5 ± 5.3 10
Endrin 86.7 ± 4.3 83.4 ± 5.5 88.0 ± 10.0 10
o,p0 -DDT 85.5 ± 10.3 89.3 ± 7.6 88.4 ± 2.4 1.0
p,p0 -DDD 82.5 ± 5.6 92.5 ± 11.3 95.1 ± 6.3 1.0
EndosufanII 80.5 ± 11.3 97.4 ± 9.9 92.4 ± 9.0 20
p,p0 -DDT 85.4 ± 9.3 84.7 ± 7.6 90.3 ± 6.3 1.0
Endosulfan-sulphate 97.3 ± 6.3 84.6 ± 4.3 91.2 ± 4.4 5
Deltamethrin 90.2 ± 10.0 113.7 ± 6.4 97.3 ± 6.0 20

to 101.5%, from 81.3% to 113.7%, and from 85.7% to 97.3% for the operated in negative-ion chemical ionization mode. International Journal of
Environmental Analytical Chemistry, 67, 81–95.
control fish spiked at 0.02, 0.05, 0.1 mg kg1. The relative standard
Hong, J., Eo, Y., Rhee, J., Kim, T., & Kim, K. (1993). Simultaneous analysis of 25
deviations (RSD) were less than 13.5% at all the three spiked con- pesticides in crops using gas chromatography and their identification by gas
centrations, indicating the good precision and accuracy of the chromatography–mass spectrometry. Journal of Chromatography, 639, 261–271.
method. The detection limits of some pesticides are around 0.5– Hong, J., Kim, H. Y., Kim, D. G., Seo, J., & Kim, K. J. (2004). Rapid determination of
chlorinated pesticides in fish by freezing-lipid filtration, solid-phase extraction
5 lg kg1 in SIM mode, except for endosufan I and II, deltamethrin, and gas chromatography–mass spectrometry. Journal of Chromatography A,
dieldrin and endrin, whose detection limits were 20 and 10 lg kg1, 1038, 27–35.
respectively, at signal-to-noise ratio of 3. Lentza-Rizos, Ch., Avramides, E. J., & Cherasco, F. (2001). Low-temperature clean-up
method for the determination of organophosphorus insecticides in olive oil.
In view of their recoveries and removal of interference, low- Journal of Chromatography A, 912, 135–142.
temperature cleanup and NH2 SPE is effective for the reliable con- Ling, Y. C., Chang, M. Y., & Huang, I. P. (1994). Matrix solid-phase dispersion
firmation and quantitation analysis of thibenocarb, deltamethrin extraction and gas chromatographic screening of polychlorinated biphenyls in
fish. Journal of Chromatography A, 669, 119–124.
and 19 organochlorine pesticides. Manirakiza, P., Covaci, A., Nizigiymana, L., Ntakimazi, G., & Schepens, P. (2002).
Persistent chlorinated pesticides and polychlorinated biphenyls in selected fish
species from Lake Tanganyika, Burundi, Africa. Environmental Pollution, 117,
4. Conclusions
447–455.
Manirakiza, P., Covaci, A., & Schepens, P. (2001). Comparative study on iotal Lipid
A rapid extraction, low-temperature cleanup and GC–MS mea- determination using soxhlet, roese-gottlieb, bligh & dyer, and modified bligh &
surement method was developed and used to determine thibeno- dyer extraction methods. Journal of Food Composition and Analysis, 14, 93–100.
Muir, D. C. G., Ford, C. A., Grift, N. P., Metner, D. A., & Lockhart, W. L. (1990).
carb, deltamethrin and 19 organochlorine pesticide residues in Geographic variation of chlorinated hydrocarbons in burbot (Lota lota) from
fish samples. The low-temperature cleanup combined with NH2 remote lakes and rivers in Canada. Archives of Environmental Contamination and
cartridge enabled efficient removal of lipids extracted from fish Toxicology, 19, 530–542.
Pang, G. F., Cao, Y. Z., Zhang, J. J., Fan, C. L., Liu, Y. M., Li, X. M., et al. (2006). Validation
sample without any significant loss of the target pesticides. This study on 660 pesticide residues in animal tissues by gel permeation
method shows good sensitivity and recoveries and allows for rapid chromatography cleanup/gas chromatography–mass spectrometry and liquid
sample analysis. It is suited for monitoring pesticide multiresidues chromatography–tandem mass spectrometry. Journal of Chromatography A,
1125, 1–30.
in fish in a regulatory laboratory. Smith, L. M., Stalling, D. L., & Johnson, J. L. (1984). Determination of part-per-trillion
levels of polychlorinated dibenzofurans and dioxins in environmental samples.
References Analytical Chemistry, 56, 1830–1842.
Volz, S. A., & Johnston, J. J. (2002). Solid phase extraction/gas chromatography/
electron capture detector method for the determination of organochlorine
Eaton, H. J., & Lydy, M. J. (2000). Assessment of water quality in wichita, kansas,
pesticides in wildlife and wildlife food sources. Journal of Separation Science, 25,
using an index of biotic integrity and analysis of bed sediment and fish tissue
119–124.
for organochlorine insecticides. Archives of Environmental Contamination and
Yi, X. H., Hua, Q., & Lu, Y. T. (2006). Determination of organophosphorus pesticide
Toxicology, 39, 531–540.
residues in the roots of Platycodon grandiflorum by solid-phase extraction and
Fillion, J., Sauve, F., & Selwyn, J. (2000). Multiresidue method for the determination
gas chromatography with flame photometric detection. Journal of AOAC
of residues of 251 pesticides in fruits and vegetables by gas chromatography/
International, 89, 225–231.
mass spectrometry and liquid chromatography with fluorescence detection.
You, J., & Lydy, M. J. (2003). Simultaneous determination of pyrethroid,
Journal of AOAC International, 83, 698–713.
organophosphate, and organochlorine pesticides in fish tissue using tandem
Hancock, P. M., Yasin, M., Baugh, P. J., Bonwick, G. A., & Davies, D. H. (1997).
solid-phase extraction clean-up. International Journal of Environmental Analytical
Optimization of high quality analytical methods for the monitoring of
Chemistry, 84, 559–571.
pyrethroid mothproofing agents by gas chromatography–mass spectrometry