Simultaneous and Rapid Analysis of 500 Pesticide Multiresidues in Agricultural Products Using GC-MS/MS and LC-MS/MS

Abstract

A multiresidue method for the simultaneous and rapid analysis of 500 pesticides in representative agricultural produce (brown rice, orange, spinach, and potato) was developed using a modified QuEChERS procedure combined with gas and liquid chromatography–tandem mass spectrometry (GC-MS/MS and LC-MS/MS). Multiple reaction monitoring parameters (e.g., collision energy, precursor and product ions) in MS/MS were optimized to achieve the best selectivity and sensitivity for a wide range of GC or LC-amenable pesticides. For the GC analysis of 360 pesticides, a short (20 m) microbore (0.18 mm i.d.) column resulted in better signal-to-noise (S/N) ratio with reduced analysis time than a conventional narrowbore column. The use of pulsed pressure injection was also effective to increase the peak response and S/N ratio. After changing a new liner, the priming injection, which caused by masking effect was suggested in order to consistent peak sensitivity. In LC-MS/MS analysis of 332 pesticides, the optimal mobile phase and injection volume was evaluated to acquire high sensitivity and reliable results. The limit of quantitation was <0.01 mg/kg, and the correlation coefficients (r2) of matrix-matched standards were >0.99 within the range of 0.0025–0.1 mg/kg. Acetonitrile with 0.1% formic acid without additional buffer salts was used for pesticide extraction, whereas only primary–secondary amine was used for dispersive solid phase extraction cleanup, to achieve good recoveries for most of the target analytes. The method was validated according to the European Union SANTE guidline. The recoveries ranged from 70 to 120% with relative standard deviations of ≤20% at 0.01 and 0.05 mg/kg spiking levels (n = 6) in all samples, indicating acceptable accuracy and precision of the method. The results of matrix effects were indicated that mainly signal enhancements were observed in GC-MS/MS but in the LC-MS/MS, the evenly spread across the ranges and little difference within the samples. The optimized method was successfully applied to the analysis of pesticide residues in real samples.