3-單氯丙二醇酯(3-monochloro-1, 2-propanediol esters, 3-MCPDEs)為油脂精煉過程中所產生之物質，其已證實經人體-脂解酶作用後產生3-單氯丙二醇(3-Monochloropropane-1,2-diol, 3-MCPD)，而其已被列為國際癌症研究所(International Agency for Research on Cancer, IARC)列為2B級疑似致癌物，過去研究中指出環氧丙醇(glycidol)存在亦導致高估3-MCPDEs含量，故本研究利用降低環氧丙醇或其酯化態之產物的3-MCPDEs分析方法，針對台灣市售87件食用油進行背景值調查。結果顯示，利用硫酸/甲醇處理30分鐘並經過後續水解及衍生化後，可大幅度減少環氧丙醇形成3-MCPD，僅1.2%環氧丙醇轉換成3-MCPD，且經確校實驗結果皆符合食品化學檢驗方法之確效規範，而本方法之定量極限為100μg/kg，於背景值調查中，依3-單氯丙二醇當量(3-MCPD equivalent)由高至低依序為棕櫚油（9,647μg/kg)、葡萄好油（4,224μg/kg)、米糠油（3,695μg/kg)、橄欖粕油（2,994μg/kg）、葵花油（1,088μg/kg)、精製橄欖油(977μg/kg)、芝麻油（841μg/kg)、大豆油（413μg/kg)、芥花油（439μg/kg）、苦茶油（167μg/kg)及初榨橄欖油（未檢出）。綜合以上結果，所使用之樣品製備方法可有效避免3-MCPDEs含量高估之情形，亦針對食用油脂進行調查，此結果可提供未來政府在風險評估或訂定殘留限制之依據。

3-monochloro-1,2-propanediol esters (3-MCPDEs) are substances produced during the oil refining process. 3-MCPDEs could be hydrolyzed by human intestinal lipase to produce 3-monochloro-1, 2-propanediol (3-MCPD), which has been listed as the Class 2B suspected carcinogen by International Agency for Research on Cancer (IARC). Previous studies have shown that the presence of glycidol can overestimate the amount of 3- MCPDEs content. This study established a 3-MCPDEs analytical method by reducing glycidol or its esterification products. This method was used to assess the amount of 3-MCPDEs in 87 commercial edible oils in Taiwan market. The results showed that after treatment with sulfuric acid/methanol for 30 minutes, the 3-MCPD produced from glycidol was significantly reduced. An average of 1.2% glycidol was converted to 3-MCPD in this study. The method was validated and was consistent with the validation regulation of food chemistry. The quantitative limit for this method is 100 μg/kg. According to 3-MCPD equivalent level detected in oil samples, palm oil had the highest concentration (9,647 μg/kg), followed by grape seed oil (4,224 μg/kg), rice bran oil (3,695 μg/kg), olive oil (2,994 μg/kg), sunflower oil (1,088 μg/kg), refined olive oil (977 μg/kg), sesame oil (841 μg/kg), soybean oil (413 μg/kg), canola oil (439 μg/kg), bitter camellia oil (167 μg/kg) and extra virgin olive oil (not detected). In conclusion, we established a method for MCPDEs determination by modifying sample preparation procedure to avoid overestimation of 3-MCPDEs. We also conducted a 3-MCPD investigation to oil sample in the market. These data can be provided as basis on risk assessment and residue limits management for government authority.