以自動加壓流體萃取搭配極致液相層析/串聯式質譜術分析食品中鄰苯二甲酸酯類及雙酚類化合物

Abstract

鄰苯二甲酸酯是目前使用最廣泛、種類最多、產量最大的塑化劑 (plasticizers)，主要用於塑膠製品中以提高產品的彈性、耐用性，常用來軟化聚氯乙烯 (PVC)，例如：食品包裝(如保鮮膜)、醫療用品(如血袋、點滴管)、塑膠玩具中。雙酚類這一類化學物質被用來製造聚碳酸酯(Polycarbonate, PC)塑膠和環氧樹脂 (epoxy resins)，應用範圍相當廣泛與普遍。聚碳酸酯可用來製造食品接觸材料，例如:餐具；環氧樹脂則被使用於食品容器的內塗層來避免內容物與金屬直接接觸。人體可能透過攝食、吸入及皮膚接觸暴露到鄰苯二甲酸酯及雙酚類，而飲食是最主要的暴露來源。包裝材料與食物接觸後，這兩類化學物質可能轉移到食品和飲料中，在人們食入水或食物後進而暴露到這些化學物質。先前的研究證實鄰苯二甲酸酯及雙酚類為內分泌干擾物，會影響人類的內分泌系統、增加癌症發生率、降低免疫功能並且對生殖造成影響。目前同時分析上述兩類化學物質之方法仍有限，因此本研究欲開發一方法同時檢測各類食物中鄰苯二甲酸酯及雙酚類，評估基質效應及萃取效率並進行準確度 (accuracy) 及精密度 (precision) 之方法卻效。本研究進行食物前處理方法之優化，以提升食物中鄰苯二甲酸酯及雙酚類的萃取效率。1 g之均質食物樣本透過自動加壓流體萃取系統，使用 5 mL 丙酮在 65℃、30 psi 下萃取 5 分鐘，進行兩個循環。萃取完成後，使用分散式固相萃取去除油脂、蛋白質及色素等雜質。將 4 mL 萃取液移至 15 mL 分散式固相萃取淨化管中 (植物性樣本使用 900 mg MgSO4, 150 mg PSA, 45 mg GCB 做為吸附劑；動物性樣本使用 Agilent Bond Elut EMR-Lipid做為吸附劑)，震盪 1 分鐘後以 5000 rpm (2,234 g-force) 離心 10 分鐘。離心後將上清液取出濃縮至 1 mL，使用 0.2 μm PTFE 過濾盤進行過濾後上機分析。本研究使用 Waters I-Class 極致液相層析搭配Waters Xevo TQ-XS串聯式質譜儀，分析食物中 12 種鄰苯二甲酸酯及 6 種雙酚類，層析使用Ascentis Express F5 (30 × 2.1 mm, 2.0 μm)管柱。12 種鄰苯二甲酸酯類及 BADGE 之分析，游離源為 UniSpray 正電，有機動相為甲醇，水性動相為 0.1% 醋酸水溶液混合 5 mM 醋酸胺水溶液，pH 4.19，梯度流析流速為 0.5 mL/min，管柱溫度為 35℃，層析時間共需 8.5 分鐘；5 種雙酚類之分析，游離源為 UniSpray 負電，有機動相為甲醇，水性動相為去離子水，梯度流析流速為 0.5 mL/min，管柱溫度為 40℃，層析時間共需 6.4 分鐘。標準品檢量線之線性範圍在0.5–300 ng/mL (DEHT 為 10–500 ng/mL)，決定係數 r2 皆達到 0.994 以上，儀器偵測極限為0.3–1972 fg，儀器定量極限在 2.4–6575 fg 之間。12 種鄰苯二甲酸酯及 6 種雙酚類於地瓜葉、豬肝、鮭魚的基質效應因子介於 10% 到 666% 之間；而萃取效果介於 N.D. 至 151% 之間。同日和異日之回收率在52.2%–1318%，準確度則在 0.73%–131% 之間。地瓜葉、豬肝、鮭魚之方法偵測極限範圍與定量極限範圍為0.05–9.37 ng/g 及 0.08–48.5 ng/g。實驗結果顯示，F5管柱可同時進行鄰苯二甲酸酯及雙酚類之層析；在食物前處理方法的部分，利用 EDGE 萃取系統使用 5 mL 的丙酮在 65℃ 下萃取可以最有效的萃取食物中的鄰苯二甲酸酯及雙酚類，且跟乙?和甲醇相比不會萃取太多雜質；使用 PRiME HLB cartridge 進行淨化會導致鄰苯二甲酸酯及雙酚類被留在 cartridge 中而無法有效定量。使用 900 mg MgSO4、150 mg PSA、45 mg GCB 做為植物性食品樣本的淨化吸附劑，可以有效去除萃取液中的色素。前處理過程使用之試劑及食物中的鄰苯二甲酸酯背景濃度對準確度 (accuracy) 及精密度 (precision) 評估造成影響而無法進行後續之食物分析，例如：環己烷-1,2-二羧酸二異壬酯 (DINCH)、鄰苯二甲酸二異壬酯 (DINP)、鄰苯二甲酸二(2-乙基己基)酯 (DEHP)、鄰苯二甲酸丁基苯酯 (BBP)、鄰苯二甲酸二丁酯 (DnBP/DiBP) 及鄰苯二甲酸二乙酯 (DEP) 在食物中之濃度。除了食物中內生性的鄰苯二甲酸酯外，萃取溶劑及淨化粉劑中皆有鄰苯二甲酸酯，主要為鄰苯二甲酸二異壬酯 (DINP)、對苯二甲酸二(2-乙基己基)酯 (DEHT)、鄰苯二甲酸二(2-乙基己基)酯 (DEHP)、鄰苯二甲酸二丁酯 (DnBP/DiBP)，而在實驗前使用甲醇及丙酮清洗淨化粉劑可以有效降低淨化粉劑中的鄰苯二甲酸酯。若要將此方法應用於食物之分析，仍需要進一步的方法優化。

Phthalate esters (PAEs) are the most widely used plasticizers with the largest variety and production for improving the elasticity and durability of products, such as food packaging, medical appliance and toys, etc. Bisphenols (BPs) are a family of chemicals used to make polycarbonate plastics and epoxy resins and the applications are quite wide and common. Polycarbonate plastics are used to produce food contact appliances such as dinner plates. Epoxy resins are frequently used in the inner coating for food and beverage cans and they can protect the contents from direct contact with metal. Humans exposed to PAEs and BPs through ingestion, inhalation, and skin contact, and ingestion was the major exposure route. PAEs and BPs migrated into food and beverages when the food packaging was in contact with food, and people exposed to these chemicals through the drinking water or food. Previous studies have confirmed that PAEs and BPs are endocrine disruptors that can interfere with the endocrine system of humans, increase cancer rate, decrease immune functions as well as influence reproduction. The methods for analyzing two types of chemicals simultaneously are still limited. This study developed a method to detect PAEs and BPs in various food simultaneously, and validated the matrix effect, extraction efficiency, and the accuracy and precision method. The preparation methods were optimized to improve the extraction efficiency of PAEs and BPs in food. One gram of homogenized sample was extracted by the EDGE system with 5-mL acetone at 65℃ and 30-35 psi for 5 min per cycle and repeated once. After extraction, the d-SPE clean-up method was used to remove impurities such as oils and proteins, and pigments. The 4-mL of the extract was transferred into a 15-mL d-SPE tube (animal food used EMR-Lipid as clean-up adsorbent, planted food used 900 MgSO4, 150 mg PSA, 45 mg GCB as clean-up adsorbent), and was vortexed one minute. Then, the samples were centrifuged for 10 minutes at 5000 rpm (2,234 g-force) and were concentrated to 1-mL. Finally, the extracts were filtered through the filters and analyzed.In this study, 12 PAEs and 6 BPs were detected and quantified by ultra-high performance liquid chromatography coupled with tandem mass spectrometer (UPLC-MS/MS) in multiple reaction monitoring (MRM) mode were analyzed by Ascentis Express F5 column (30 × 2.1 mm, 2.0 μm). The analysis of PAEs and BADGE performed with 0.1% acetic acid/5-mM ammonium acetate (PH 4.19) and methanol as mobile phase at positive Unispray ionization (UniSpray+). The parameters of chromatography have been optimized so that the flow rate was 0.5 mL/min, the column temperature was 35℃, and the time of analysis was 8.5 minutes. The analysis of BPs performed with Milli-Q water and methanol as mobile phase at negative UniSpray ionization (UniSpray-). The parameters of chromatography have been optimized so that the flow rate was 0.5 mL/min, the column temperature was 40℃, and the time of analysis was 6.4 minutes. Furthermore, the linear ranges of calibration curves were 0.5?300 ng/mL (except for DEHT was 10–500 ng/mL) with the coefficient of determination (r2) greater than 0.994. The instrument detection limits (IDLs) were 0.3–1972 fg, and the instrument quantitation limits (IQLs) were 2.4–6575 fg. The matrix effect factors of 12 PAEs and six BPs in sweet potato leaves, pork liver, and salmon ranged from 10% to 666%; the extraction efficiency ranged from N.D. to 151%. Inter-day and Intra-day recovery ranged from 52.2% to 1318%, and accuracy ranging from 0.73% to 131%. The limit of detections (LODs) and the limit of quantitations (LOQs) in sweet potato leaves, pork liver, and salmon were 0.05–9.37 ng/g and 0.08–48.5 ng/g. The F5 column could be used to analyze PAEs and BPs simultaneously. The use of acetone at 65℃ had a better extraction efficiency of PAEs and BPs in food than that of acetone/methanol, and the impurities were lower than that at acetonitrile and methanol. Cleanup with PRiME HLB cartridge retained PAEs and BPs in the cartridge and thus could not be quantified. Using 900 mg MgSO4, 150 mg PSA, 45 mg GCB for plant food could remove pigments effectively. PAEs from the reagents used in pretreatment process and in the tested food matrixes affected the evaluation on accuracy and precision so that subsequent analysis was unable to proceed, such as DINCH, DINP, DEHP, BBP, DnBP/DiBP, and DEP. PAEs were detected in extraction solvents and cleanup adsorbents. Using methanol and acetone to clean up adsorbents could reduce the concentrations of PAEs effectively. Further method optimization is required, and apply this method for PAEs and BPs in food.