探討水耕蔬菜吸收新興污染物與其於餘氯光降解程序中所產生的副產物

Abstract

有鑑於水資源的缺乏，越來越多國家開始使用處理過的廢水來灌溉蔬菜或穀物。然而現有的研究僅分析廢水中的微量污染物本身，並無考慮到其副產物與代謝物，這導致人們嚴重低估暴露污染物的風險。為了探討這種低估的情形，本研究選用不同KOW的藥物──乙醯胺酚 (acetaminophen) (log KOW = 0.46)、氯胺酮 (ketamine) (log KOW = 2.18) 與美沙冬 (methadone) (log KOW = 3.93)作為目標污染物，添加至營養液中，栽培臺灣萵苣（A菜）。同時，為了模擬廢水在使用前所經歷的處理程序，目標污染物會預先進行自然光光降解（sunlight photolysis）、餘氯氧化（chlorination），以及餘氯光降解程序(Sunlight/chlorine)──此程序會發生於農夫將處理過的廢水靜置在戶外，以消除水中餘氯的過程。這些處理過的含目標污染物溶液，會與營養元素混合配製成營養液，然後用來澆灌蔬菜。然而，目前關於Sunlight/chlorine 的機制仍不清楚，故本研究將分成兩個部分：（一）探討Sunlight/chlorine的反應機制；（二）探討使用處理過的廢水澆灌蔬菜後，目標污染物、降解副產物與代謝物在蔬菜中的累積。

在第一個部分， ketamine被作為研究sunlight/chlorine的代表污染物 (50 μg/L)，於模擬自然光源與低濃度的餘氯條件（1 mg Cl2/L）下反應，以驗證在不同反應階段，反應物質的生成與降解貢獻。根據實驗結果，氫氧自由基（•OH）主導sunlight/chlorine的初期反應。當系統中的餘氯消耗完畢，ketamine的降解反應仍持續進行(kobs = 7.6±0.50×10-3 min-1)，且顯著地高於直接光降解的速率( kobs = 2.9±0.15×10-4 min-1)。此時的反應，主要是由sunlight/chlorine系統所生成的臭氧所主導。無論在sunlight/HOCl (pH 5)與sunlight/ClO- (pH 10)，臭氧皆具有12.5±0.5%與10±1%的高生成率。而sunlight/chlorine所生成的µM濃度的臭氧，被證實可以有效地氧化ketamine，並且在自然光下，降解速度可以進一步提升。為了更進一步確認反應物質在不同時間的濃度，本研究使用模擬軟體Kintecus建立sunlight/chlorine的反應動力學模型。依照反應動力學模型的模擬結果，在餘氯濃度為1 mg/L的sunlight/chlorine系統中，累積的臭氧生成量達0.91 μM，並且降解31%的ketamine。而臭氧也被證實能與一級胺與二級胺的官能基反應，因此阿替洛爾 (atenolol) 與美托洛爾 (metoprolol)亦會被臭氧所降解。此外，本研究的實驗與模擬結果指出，ClO2-與ClO2在系統中的濃度低於nM，因此sunlight/ClO2-與sunlight/ClO2僅對ketamine的降解具有微小的貢獻。

在本研究的第二個部分，為了探討微量污染物、降解副產物與代謝物在蔬菜中的累積，三種目標污染物被添加至營養液中，來栽培蔬菜。根據實驗結果，acetaminophen、ketamine與methadone的總生物濃縮係數（total bioconcentration factors，total BCF；包含轉化為代謝物的目標污染物）分別為0、120±7.76與176±16.0 L/kg。而當三種目標污染物經過三種不同處理程序（sunlight photolysis、chlorination與sunlight/chlorine）則會產生不同的副產物，舉例來說，ketamine會轉化成norketamine（最高至6.0%），而methadone則會轉化成2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine (EDDP) （最高至16%）。其中norketamine的BCF為162±22.6 L/kg，高於ketamine。除了主要的降解副產物norketamine與EDDP外，本研究亦發現其它3種微量以及13種未在營養液中測得之副產物被蔬菜所吸收。此外，ketamine與methadone亦會在蔬菜中分別代謝成norketamine與EDDP，且當營養液中的ketamine濃度從1000降至50 μg/L，ketamine代謝為norketamine的比率從22±7.0%提升至45±0.062。

藉由兩個部分的探討，此研究觀察到微量污染物與其在sunlight photolysis、chlorination與sunlight/chlorine下產生的副產物於蔬菜中的累積情形。研究結果揭示了兩項可能被低估的風險：（一） 若僅分析處理過的廢水中的目標污染物濃度，可能會忽略微量或儀器難以偵測的副產物，這些副產物仍可能累積於蔬菜中；（二） 若僅分析蔬菜中的目標污染物，則可能忽視部分污染物在蔬菜內的代謝，進而低估代謝物所造成的風險。鑒於上述成果，本研究預期可作為未來相關研究的基礎，用於評估處理後廢水灌溉蔬菜對人體健康的影響。

Due to the scarcity of water resources, treated wastewater is increasingly being used to irrigate crops and vegetables. However, the risk of micropollutant exposure from vegetables cultivated with treated wastewater has been largely underestimated. To address this gap, a hydroponic system was established for lettuce (Lactuca sativa var. sativa) cultivation using a nutrient solution spiked with three pharmaceuticals— acetaminophen (log KOW = 0.46), ketamine (log KOW = 2.18) and methadone (log KOW = 3.93)—which have varying log KOW values. To simulate the exposure of treatments in a conventional wastewater facility, these pharmaceuticals were pre-treated with sunlight photolysis, chlorination, and reaction with sunlight-induced chlorine (sunlight/chlorine) before being used in lettuce cultivation. The sunlight/chlorine process simulates conditions where farmers leave water undisturbed for several days to allow residual chlorine to dissipate. This study investigated two aspects: (1) the mechanism of the sunlight/chlorine process under simulated sunlight, particularly the formation of reactive species, and (2) the accumulation of micropollutants with different KOW as well as their degradation byproducts and metabolites in vegetables cultivated with treated water.

In the first part, ketamine, an environmentally persistent pharmaceutical, was used as a model compound to elucidate the mechanism of the sunlight/chlorine process. The degradation of ketamine under sunlight irradiation in the presence of a low chlorine concentration (1 mg/L as Cl2), was examined to elucidate the evolution of reactive species and their contributions to ketamine removal. Hydroxyl radicals (•OH) dominates the initial stage of the sunlight/chlorine process. Upon chlorine depletion, O3 became the primary reactant responsible for ketamine degradation. High O3 yields were observed in both sunlight/HOCl (12.5±0.5% at pH 5) and sunlight/ClO- (10±1% at pH 10) systems. At sub-µM levels, O3 resulted in significant ketamine removal, with even faster rates observed in the presence of sunlight. A kinetic model was developed to simulate time-dependent concentration changes during the sunlight/chlorine process. The simulation indicated that the cumulative O3 concentration could reach 0.91 μM, contributing 31% of ketamine removal. Primary and secondary amine functional groups were identified as the reaction sites for O3. Similar degradation phenomena were observed for other pharmaceuticals, such as atenolol and metoprolol. Experimental and model results further suggested that sunlight/ClO2- or ClO2 played a minor role in ketamine degradation, with trace amounts (below nanomolar levels) of ClO2- and ClO2 predicted by the simulation.

In the second part, all three pharmaceuticals were selected to investigate the accumulation of micropollutants, their degradation byproducts, and their metabolites in vegetables cultivated with treated water. The total bioconcentration factors (BCFs) (including the transformation of metabolites) of acetaminophen, ketamine and methadone were found to be 0, 120±7.76 and 176±16.0 L/kg, respectively. During the photolysis, chlorination and sunlight/chlorine processes, ketamine and methadone were transformed into norketamine (up to 6.0%) and 2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine (EDDP) (up to 16%), respectively; the BCF of norketamine (162±22.6 L/kg) was even greater than that of ketamine. Additionally, other degradation byproducts (including three trace and 13 undetected byproducts in the nutrient solutions) were taken up by the lettuce. Furthermore, ketamine and methadone underwent metabolism in lettuce, with the conversion ratio of ketamine to norketamine increasing from 22±7.0% to 45±0.062 as the ketamine concentration decreased from 1000 to 50 μg/L.

The results highlight that it is not sufficient to simply measure the parent pharmaceuticals present in treated wastewater since trace and even undetectable degradation byproducts accumulate in vegetables. The overall exposure can be underestimated when only the parent pharmaceuticals are analyzed in vegetables since a significant amount of the parent pollutants were found to be metabolized into other compounds. Moreover, the results can guide future research in assessing the health risks associated with cultivating vegetables with treated wastewater.