Pesticide residue has become one of the dominant obstacles to using traditional Chinese medicine safely, and the published researches are mostly aiming at the detection of pesticides themselves. However, during the preparation and storage of traditional Chinese medicinal materials, these rudimental pesticides would possibly go through degradation process such as breaking down and being metabolized by bacteria, which could result in poisonous degradation products. But there are very few published methods focusing on the simultaneous detection of the poisonous products. Hence, to enhance the safety control of traditional Chinese medicine, it is necessary to establish effective and sensitive methods to detect pesticide related poisonous products as well. The thesis is focusing on analyzing poisonous degradants of pesticides on traditional Chinese medicine by capillary electrophoresis (CE) with high efficiency and sensitivity. The main content of the research is as follows: A micellar electrokinetic chromatography method with ultraviolet detector (MEKCUV) was established to determine five poisonous degradation products of pesticides simultaneously. These five analytes were separated completely within 10 minutes. They were readily detected at 0.03-0.07 mg/L with a precision of 0.7%-4.8%. Good linear relationship between concentration and peak area was achieved in the range of 0.2-25.0 mg/L, with correlation coefficients ranging from 0.9992 to 0.9998. A cleanup method was developed to for sample preparation before MEKC-UV. The average recoveries of spiked Lycium chinense Mill. were between 82.4% and 94.1% with RSD less than 10%. This pretreatment method can also effectively applied to treat Lonicera japonica, Pseudostellariae radix and Fritillaria cirrhosa. A capillary electrophoresis method with laser-induced fluorescence detector including a pre-column derivatization process has been developed to determine three poisonous degradants of pesticides simultaneously. These three analytes were separated completely within 12 minutes. They were readily detected at 0.2-10 µg/L with a precision of 0.4%-8.7%. Good linear relationship between concentration and peak area was achieved in the range of 50-1000 µg/L, with correlation coefficients ranging from 0.9984 to 0.9996. The average recoveries were between 86.3% and 98.5%. The method was applied to analyze real Pseudostellariae radix sample, and none of the three degradants was found.

中藥材中農藥殘留是安全使用中藥的重要障礙，現已有的研究中大多是針 對檢測農藥本身的。然而，農藥使用後在藥材的炮製、儲存等過程中可能會發 生分解、被微生物代謝等其他降解行為，得到的某些有毒產物卻少有檢測方法。 因此為了加強對中藥材的安全控制，需要同時建立高效、靈敏檢測農藥相關有 毒產物的方法。 本論文圍繞建立中藥中農藥有毒降解產物殘留的快速、靈敏、高效的毛細 管電泳檢測方法進行了研究，具體內容如下： 1. 建立了 5 種農藥的苯胺類有毒降解產物的毛細管電動力色譜-紫外 （MEKC-UV）檢測方法。10 分鐘內待測有毒降解產物達到基線分離, 在 0.2~25.0 mg/L 濃度範圍內線性良好, 相關係數為 0.9992~0.9998,檢測限為 0.03~0.07 mg/L，重複性實驗中 RSD（n=6）為 0.7%~4.8%。建立了對枸杞樣品 的提取與淨化方法，5 種苯胺在枸杞中的加標回收率為 82.4%~94.1%且 RSD （n=6）＜10%，該前處理方法也適用于金銀花、太子參和川貝母基質的前處理。 2. 建立了 3 種農藥的有毒降解產物的柱前螢光衍生-毛細管電泳-鐳射誘導 螢光（CE-LIF）檢測方法。 該方法在 50~1000 µg/L 濃度範圍內線性良好, 相關 係數 R 為 0.9984~0.9996，檢測限為 0.2~10 µg/L，12 分鐘內 3 種待測農藥達到 基線分離, 重複性實驗中 RSD（n=5）為 0.4%~8.7%，平均加樣回收率為 86.3%~98.5%。將該方法用於檢測太子參中農藥有毒降解產物，未檢測到這三 種降解產物。