INTRODUCTION

Fluazifop-P-butyl is a systemic, post-emergence herbicide that selectively controls both annual and perennial grass weeds in non-graminaceous crops. It is rapidly hydrol­ysed by esterases and the main metabolite is its acid de­rivative fluazifop. Conjugates may also contribute to the overall residues, so methods have to capture all the parent and major metabolites. Most of the published methods for the determination of fluazifop residues have used either gas-liquid chromatography (GLC) or liquid chromatogra­phy (HPLC). GLC procedures require extensive clean-up and derivatisation, e.g. pentafluorobenzylation [1, 2], al- kylation using diazomethane [3, 6] or various chlorofor- mates [7]. For GLC detection, electron capture [1, 5, 7], mass selective [2-4, 7] or nitrogen selective detectors [6] have been used. HPLC methods [8-11] are also labour intensive and lack selectivity. Both UV [8-10] and am- perometric [11] detection have been used with HPLC.

LC-MS-MS determination of acidic herbicides [12] has been previously published for water using negative ion electrospray ionisation (ESI). In this paper a robust, highly specific and sensitive LC-MS-MS method is de­scribed for crops using positive ion atmospheric pressure chemical ionisation (APcI).

To facilitate the speed of pro­cedure and minimise possible contamination, disposable labware and solid phase extraction cartridges were used. The method is also suitable for rapid screening of residues using ultaviolet detection with a compromise in sensitiv­ity.

 

EXPERIMENTAL

CHEMICALS AND REAGENTS

a)        Solvents and reagents: acetone, acetonitrile, dichloromethane, ethyl acetate, methanol, hydrochloric acid, glass distilled/HPLC grade solvents and analytical grade reagents.

b)       Standard solutions: fluazifop-P (R)-2-[4-(5-trifluoromethyl-2- pyridyloxy) phenoxy] propionic acid (IUPAC) analytical standard > 98% (Zeneca Agrochemicals) dissolved in acetonitrile.

c)        Solid phase extraction cartridges (standard polypropylene sy­ringe format): C₂ (EC) – end capped 500 mg adsorbent in 10 mL XL cartridge (IST International Sorbent Technology/Jones Chro­matography, UK). Si – high activity silica (500 mg adsorbent in 3 mL cartridge) declared moisture content 7-8% (Varian).

APPARATUS

1.        Ultra Turrax homogeniser – for homogenisation at 10000-­15000 rpm.

2.        Thermostatic heating block – at 60 °C for hydrolysis.

3.        Solid phase extraction manifold.

4.        HPLC column – Kromasil KR-100 C₁₈ 5 |am, 100 A, 50 mm x 3.2 mm for LC-MS-MS determination or 150 mm x 3.2 mm for LC-UV determination.

5.        LC-MS-MS system. – Hewlett Packard 1050 Quaternary HPLC system coupled to a Perkin Elmer Sciex API III triple quadrupole mass spectrometer.

SAMPLE PREPARATION, EXTRACTION AND HYDROLYSIS

Samples were chopped and homogenized using an internationally accepted method [12]. For crops with a low water content, (dry crops e.g.

seeds/oilseeds) a sub-sample (10 g) was soaked over­night (or for a minimum of 2 h) in 1 M hydrochloric acid (50 mL minus water content of 10 g sample). Acetonitrile (50 mL) was added and then sample was extracted using an Ultra Turrax homogeniser at > 10000 rpm for 5 min. Crops with a high water content were extracted by homogenising a sub-sample (20 g) directly with freshly prepared acetonitrile: hydrochloric acid mixture 98: 2 v/v (100 mL minus the average water content of 20 g sample) for 5 min at > 10000 rpm. With all crop types, the homogenate was separated from the extract by centrifugation at 3500 rpm for 2 min. An aliquot of the supernatant equivalent to 0.5 g sample (2.5 mL) was transferred into a 7 mL vial. The organic solvent was removed by evaporation with a stream of dry air at < 60 °C until an aqueous solution remained. An extra aliquot was taken from an untreated control sample extract in order to prepare a standard in the pres­ence of crop matrix for LC-MS-MS determination. The volume of the extract was measured using a pipette and an equal volume of hydrochloric acid (12 M) was added to set the overall acid concen­tration to approximately 6 M. The vial was tightly sealed and hy­drolysed at 60 ±2 °C for 1 h using a thermostatted heating block. After hydrolysis, the hydrolysate was diluted to six times its origi­nal volume using ultra-pure water. This dilution was taken as a pre­ventative measure to minimise acid hydrolysis of the modified sil­ica phase in the clean-up below.

CLEAN-UP

A C₂ (EC) solid-phase extraction column was conditioned with methanol (2.5 mL) followed by water (2.5 mL). The diluted hy­drolysate was applied onto the column at a speed of 3-5 mL/min. The column was washed with ultra-pure water (3 x 1 mL) then dried for 3-5 min by drawing air through the cartridge. Any re­maining water droplets inside the reservoir were wiped off with a clean piece of tissue paper. The column was washed with hexane (2 x 1 mL) then dried as above for 15 min. (It was found vital that no residual water was transferred to the silica clean up to ensure robust elution characteristics). While the C₂(EC) cartridges were drying, the silica cartridges were conditioned with dichloro- methane (2.5 mL). A 2-3 mm layer of dichloromethane was left on the top of the Si adsorbent by stopping the flow of solvent before the C₂(EC) cartridge was coupled to the Si cartridge, in tandem, using a suitable adapter. The sample was quantitatively eluted from the C₂ (EC) column directly onto the silica with a 95: 5 (v/v) mixture of dichloromethane and acetone (2 x 1.5 mL). The upper C₂ column was detached and the Si column was washed with di- chloromethane (2 x 1 mL), followed by ethyl acetate (2 x 1.5 mL). The washes were discarded. The analyte was eluted from the Si cartridge with methanol (2 x 1 mL) into an HPLC vial. The sam­ple was evaporated to dryness with a stream of dry air at < 60 °C and reconstituted in HPLC mobile phase (1.0 mL of 10:90 v/v acetonitrile: 0.4% v/v HCOOH in water) ready for HPLC analysis. A short ultra-sonication was used to assist re-dissolution of the residuum. If precipitation or floating particulates were present prior to HPLC analysis, the sample was centrifuged at 3500 rpm for 1 min and the clear supernatant transferred into a new vial.

PREPARATION OF CALIBRATION STANDARDS

Due to matrix related response enhancement observed with LC-MS-MS analysis, matrix matched standards were prepared in HPLC mobile phase containing the appropriate crop matrix at the relevant sample concentration. Matrix matched standard were pre­pared by adding the appropriate amount of fluazifop-P standard to the extra aliquot of the control matrix at the methanol eluate stage of the silica clean-up. The eluate is evaporated to dryness and re­dissolved in 1.0 mL of HPLC mobile phase in the same fashion as used with the sample.

LC-UV detection is invariable of sample matrix, therefore cali­bration standards were prepared in HPLC mobile phase corre­sponding to the initial composition of the gradient.

With both detection strategies, quantification was based on comparison of peak areas of a sample with that of the relevant standard.

Instrumental analysis

LIQUID CHROMATOGRAPHY WITH TANDEM MASS SPECTROMETRY

Gradient HPLC was found satisfactory for APcl/LC-MS-MS. Mobile phase flow rate was 1 mL/min, column temperature 40 °C, injection volume 50 pL. Mobile phase A was water: acetoni­trile: formic acid (95: 5 : 0.4 v/v); mobile phase B was acetonitrile: water: formic acid (95 : 5 : 0.4 v/v). The initial composition was 20% B held from 0-0.75 min, followed by a linear gradient to 80% B from 0.75-2 min. This is held at 80% B from 2-3 min, ris­ing to 100% B from 3-3.2 min. 100% B is held from 3.2-5 min, then the system is re-equilibrate at initial conditions (20% B) from 5.2-7 min.

Atmospheric pressure chemical ionisation operating in the pos­itive ion mode was found to be the most satisfactory ionisation technique. With the Perkin Elmer API III mass spectrometer, the system was optimised using a 5.5 kV corona discharge voltage, a 40 V orifice voltage, 20 V collision energy and a collision gas thickness of 280 (using 90 : 10 argon: nitrogen as the collision gas). The APcI probe temperature was set at 480 °C. Protonated molec­ular ions generated in the ion source (m/z 328) were selected and subjected to further fragmentation by collisional activation. The most abundant ion, (m/z 282, identified as carboxyl loss), in the re­ re­sulting daughter ion spectrum was then monitored and used for quantitative analysis. The mass spectrometer scanning method was set up with a dwell time of 250 ms and a pause time of 0.02 ms leading to a 0.5 s scan speed. Parent and daughter ion spectra are shown in Figs. 1.

Fig:1 Parent ion[ M+H]+ mass spectrum of fluazifop-P

 

LIQUID CHROMATOGRAPHY WITH ULTRAVIOLET DETECTION

Ultraviolet detection was also used when higher residues (gener­ally > 0.1 mg/kg) were expected. Mobile phase flow rate on a longer column was 1 mL/min, column temperature 40 °C, injection volume 200 pL and detector wavelength 270 nm. Mobile phase A was water containing 0.4% formic acid, mobile phase B was acetonitrile. A linear gradient was employed from 10% to 90% B within 0-15 min, then a second step from 90% to 95% B within 15-20 min, then re-equilibrated at initial 10% B from 20-25 min.

 

Fig. 3 LC-MS-MS chromatogram of fluazifop-P standard (0.1 ^g/ mL), untreated control sunflower seed, fortified control (recovery at 0.1 mg/kg – 105% recovery) and a field treated sample of sunflower seed (residue 0.065 mg/kg). Crop to solvent ratio – 0.5 g/mL.

RESULT AND DISCUSSION

For fast LC-MS-MS analysis, isocratic elution was ini­tially attempted but occasionally false results were ob­tained. It was found that late eluting matrix components from previous sample injections could interfere with the APcI ionisation processes, consequently affecting the de­tector response and the quantification. This effect was overcome with gradient elution, where high percentages of organic solvent were used to remove all matrix compo­nents from the system prior to subsequent injections. This procedure was necessary to ensure robust operation and quantification.

It was found that for a variety of samples, the presence of matrix could lead to signal enhancement (ranging from 0 to 35% dependent on matrix)

when performing LC-MS- MS analysis. This led to inaccurate quantification of fluazifop concentration in some samples. No correlation be­tween matrix type and enhancement was found, hence for consistency and accurate quantification, matrix matched standards were used in all LC-MS-MS analyses. (This ef­fect was not seen when LC-UV detection was used.) For method validation untreated control and fortified crop samples in the range of 0.01-2 mg/kg were analysed by both detection techniques. Similar recoveries were obtained for both detection regimes and summaries of the resultant data are shown in Table 1. Examples of representative chromatograms for LC-MS-MS and LC-UV are given in Figs. 3 and 4, respectively. The LC-MS-MS chromatograms clearly demonstrate a better sensitivity and selectivity compared to that by LC-UV.

Fig. 4 LC-UV chromatograms at 270 nm of untreated cabbage A, fortified cabbage at 0.1 mg/kg B and fluazifop-P standard (0.5 ^g/ mL) C.

Conclusions

A sensitive and specific LC-MS-MS method using atmos­pheric pressure chemical ionisation has been developed to quantify fluazifop residues in

crops. Disposable labware and adsorbents were used for convenience and to min­imise contamination. The method has been validated for a variety of crop types covering most FAO/WHO Codex classification of crop categories [13]. The clean-up meth­od is also suitable for monitoring purposes using less so­phisticated instrumentation such as LC-UV.

TABLE .1: RECOVERIES AND LOQS BY LC-MS-MS AND LC-UV DETECTION

 

Crop	AverageRecover (%)	RSD(%)	n	Range (%)	LOQ (mg/kg)	Average recovery (%)	RSD(%)	n	Range (%)	LOQ (mg/kg)
Potatoes	92	10	14	73-107	0.01	99	7.9	14	83-108	0.05
Spinach	99	7.4	14	90-121	0.05	96	6.2	14	79-101	0.05
Cabbage	96	8.3	20	84-119	0.01	101	5.2	20	95-116	0.01
Peas (mange tout)	93	4.5	16	86- 99	0.05	94	6.7	16	83-103	0.05
Tomato	84	6.7	14	74- 95	0.05	84	6.2	14	75- 91	0.05
Orange	96	7.1	14	83-110	0.01	90	11	10	69-103	0.05
Apples	92	7.6	14	70-101	0.01	91	13	10	72-103	0.05
Soya beans	92	8.8	14	80-104	0.05	107	6.4	6	100-116	0.2
All crops	93	8.5	120	70-121	0.01-0.05	95	9.8	104	69-116	0.01-0.2

 

REFERENCES

1.        Liu W, Chen Z, Lu Q, Shi Y (1991) Analyst 116: 273-276.

2.        HebererT, Butz S, Stan H-J (1995) J Environ Anal Chem 58: 43-53.

3.        Patumi M, Marucchini C, Businelli M, Vischetti C (1987) Pes- tic Sci 21: 193-201.

4.        Wigfield YY, Lanouette M (1993) J Agric Food Chem 41: 84-88.

5.        Kulshrestha G, Singh SB, Gautam KC (1995) Bull Environ Contam Toxicol 55: 276-282.

6.        Clegg BS (1987) J Agric Food Chem 35: 269-273.

7.        Butz S, Stan H-J (1993) J Chromatogr 643:227-238.

8.        Davy GS, Atreya NC, Duffin MR, Wollerton C (1993) Com­prehensive Analytical Profiles of Important Pesticides. CRC Press

Boca Raton Ann Arbor London Tokyo, Chapter 7, pp 149-171.

9.        Balinova A, Nikolova G (1992) Weed Res 32: 143-147.

10.    Zanco M, Pfister G, Kettrup A (1992) Fresenius’ J Anal Chem 344: 39-41.

11.      Worobey BL, Shields JB (1989) J Assoc off Anal Chem 72: 368-371.

12.      Koppen B, Spliid NH (1998) J Chromatogr 803:157-168.

13.      FAO/WHO 1898: Guide to Codex Recommendations Concern­ing Pesticide Residues, Part 4, Codex Classification of Foods and

Animal Feeds, Second Edition, Joint FAO/WHO Food Standards Programme, Codex Alimentarius Programme, CAC/ PR 4-1989 Comity Guidelines.

How to cite this article

Xingguo Chen.; Hauitao Liug.; Huiwen Wang.; Lijun Dong.; Hongping Xue.; Zhide Hug (2015): Pesticide residue analysis in crops by LC-MS/MS method. International Journal of Agricultural and Life sciences 1 (1), pp. 1–4. DOI 10.9379/sfjal. 122057-001-0081-x.

CONFLICTS OF INTEREST

“The authors declare no conflict of interest”.

 

© 2015 by the authors; licensee SKY FOX Publishing Group, Tamilnadu, India. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).