利用高效液相層析儀及氣相層析儀分析176種農藥以確認採收前蔬果安全品質   全文下載

Multiresidue Analysis of 176 Pesticides and Metabolites in Pre-harvested Fruits and Vegetables for Ensuring Food Safety by Gas Chromatography and High Performance Liquid Chromatography

本研究建立高敏感且可信賴之多種農藥殘留同時檢出法，以進行蔬果農藥殘留監測工作。檢測農藥及其代謝物共計176 種，包含農民常用農藥及部分政府禁用農藥，其中主要藥劑類別為：有機磷劑、有機氯劑、合成除蟲菊類、有機氮劑及胺基甲酸鹽類。作物樣品經由丙酮抽出、液- 液相萃取及淨化管作用後，分別藉由不同檢出器定性定量，如71 種農藥以氣相層析儀/ 火焰光度附磷檢出器（GC/FPD）檢測、83 種農藥以氣相層析儀/ 電子捕獲檢出器（GC/ECD）檢測、19 種農藥經由後置反應裝置衍生化後再以高效液相層析儀/ 螢光檢出器（HPLC/FLD）檢測，而貝芬替、腐絕及益達胺則由高效液相層析儀/ 紫外光檢出器（HPLC/UV280）檢測。方法之確效係以標準品混合溶液摻入14 種水果及8 種蔬菜中（0.1-5.0 μg/g），各作物均進行三重覆。本方法之敏感度及再現性均佳，其偵測極限大多介於0.001-0.03 μg/g 且符合容許量之要求。176 種農藥在22 種蔬果中之回收率多數介於60% 至120% 之間，且相對變異係數小於20%。2006 年針對22 種採收前蔬果進行4,305 件樣品分析，依據衛生署公告安全容許量評估其安全性，並探討本分析方法在例行監測、農藥使用、農民教育及食品安全執法上之應用性。"

A broad-range, sensitive and reliable multiresidue analytical method is presented for identifying and inspecting 176 pesticide residues, including organophosphate, organochlorine, synthetic pyrethroid, organonitrogen and carbamate, in pre-harvested fruits and vegetables. Pesticide residues were extracted from samples with acetone, followed by liquid-liquid partition and solid-phase extraction. Seventy one and 83 pesticides were determined, respectively, by gas chromatography with a flame photometric detector and an electron capture detector. Nineteen carbamate pesticides and several metabolites were measured by high-performance liquid chromatography (HPLC) with a post-column derivation system and a fluorescence detector; and three additional pesticides, i.e., carbendazim, thiabendazole, and imidacloprid, were detected by HPLC using an ultraviolet detector at 280 nm. The proposed method was validated for 14 fruits and 8 vegetables using samples spiked with pesticide standards (0.1- 5.0 μg/g) in triplicate. Good sensitivity and repeatability were obtained with detection limits of 0.001- 0.03 μg/g, and all limits of detection were lower than one-fifth of their specific maximum residue levels. The recovery rates for most pesticides in various fruits and vegetables were 60-120% with relative standard deviations < 20%. With more than 4,300 samples collected from the field examined to assess the performance, it is proposed that this method be applied for routine monitoring, legislation implementation and farmer education programs.