臺灣水田土壤與稻米中稀土元素之含量分佈與人類潛在健康風險評估

Abstract

稀土元素 (rare earth elements, REEs) 是由鑭系元素及釔所組成的一群元素之統稱。依有效離子半徑，可再區分為輕稀土元素 (light rare earth elements, LREEs) 及重稀土元素 (heavy rare earth elements, HREEs) 兩個亞族。REEs 被廣泛應用於高科技、能源、農業與工業領域，並逐漸成為全球重要的戰略性資源。隨著 REEs 的使用日益增加，可能透過排放與累積流佈於環境中，自土壤轉移至植物，再經由食物鏈對人類健康造成潛在影響。然而，目前針對 REEs 在環境中的分佈及其潛在的暴露及風險，仍舊欠缺系統性的研究。稻米 (Oryza sativa L.) 為全球主要糧食來源之一，是人類暴露 REEs 的重要途徑。本研究藉由探討 REEs 在水田及其生長的稻米中 REEs 之含量及其生物有效性，瞭解 REEs 在土壤與稻米系統中的累積與轉移趨勢。進一步地，針對攝食稻米之途徑，假設每日食用米的情況下，評估不同年齡與性別藉由攝食米對 REEs 潛在的暴露劑量及其非致癌風險。此外，由於 REEs 的分化特徵可反映區域性的地球化學背景，本研究透過比較臺灣不同母質所生長的稻米中 REEs 之組成與分化，初步評估 REEs 作為地理溯源工具的可行性。本研究採集臺灣本島各縣市主要類型的水田土壤和稻米樣品 (n = 85)，分析土壤中的 REEs 近似全量濃度，並以三個單一試劑有效性萃取法，包括 0.05 M EDTA、0.10 M HCl 和 0.01 M CaCl2，萃取土壤中潛在可被利用的 REEs，以及分析所有稻米樣品中的 REEs 濃度。針對健康風險評估，本研究採集臺灣各縣市 265 個稻米樣品，以蒙地卡羅模擬法，評估不同年齡層與性別族群透過稻米攝食所暴露的 REEs 劑量與非致癌風險。而藉由不同母質所採集的 265 個稻米樣品，透過 REEs 的分化指標、常態化特徵圖形及主成分分析等，探討 REEs 作為地理溯源工具之應用性。結果顯示，臺灣水田土壤和稻米中總 REEs 的含量分別為 16.2－137 mg kg-1 和 2.25－13.6 μg kg-1，皆以 LREEs 的濃度較 HREEs 高。三個單一試劑萃取的結果顯示，0.05 M EDTA 與 0.10 M HCl 萃取法，可能作為評估稻米中 REEs 的潛在參考指標。其中，HREEs 因具有較小的離子半徑和較高的電荷密度，使其在酸性環境中可能具有較高的生物有效性，而 REEs 的生物有效性主要受到土壤性質如 pH、有機碳、游離鐵等影響。暴露及風險評估結果顯示，臺灣男性藉由攝食米所暴露的 REEs 較女性高，且 0–3 歲的兒童，其暴露量及非致癌風險為其他年齡層的兩倍之高。敏感性分析結果指出，控制暴露量和風險的因素主要為攝食率、體重以及稻米中特定 REEs (La, Ce, Eu 和 Y) 的濃度。此外，利用 REEs 及其分化指標、以上部大陸地殼進行常態化後的特徵圖形與多變量統計分析，結果顯示，片岩沖積土壤所生長稻米中的 REEs 組成和分化特徵明顯異於砂頁岩、黏板岩沖積土和第四紀洪積母質的土壤所生長稻米中的 REEs。綜上所述，本研究釐清新興污染物——REEs 在土壤與稻米系統中的環境行為，並首次量化 REEs 在主要飲食途徑的潛在暴露風險，亦展現 REEs 在農產品地理來源判別上的應用性。本研究成果針對未來面臨環境與食品中的暴露及風險溝通和管理奠定基礎，同時彰顯 REEs 作為農產品安全與溯源工具的前瞻性價值。

Rare earth elements (REEs), comprising the lanthanides and yttrium, represent a group of technology-critical elements with growing strategic importance worldwide. Based on their ionic radii, REEs are commonly divided into light REEs (LREEs; La–Eu) and heavy REEs (HREEs; Gd–Lu and Y). As their use continues to increase, REEs may be released into and accumulated within the environment, transferred from soils to plants, and subsequently enter the human body through the food chain. However, systematic investigations of their environmental distribution, dietary exposure, and potential health risks remain limited. Rice (Oryza sativa L.), a major staple food worldwide, is considered a major dietary source of REE exposure for humans. This study investigated REE concentrations and bioavailability in paddy soils, their accumulation and transfer to rice grains, and the associated non-carcinogenic health risks through rice consumption. In addition, given that REE fractionation patterns can reflect regional geochemical signatures, this study further evaluated the feasibility of using REEs as tracers for geographical origin authentication of rice. Soil and rice samples were collected from 85 paddy fields across Taiwan. Total REE concentrations in soils were determined using aqua regia digestion, while potentially available REE pools were assessed using three single-extractant methods: 0.05 M EDTA, 0.10 M HCl, and 0.01 M CaCl2. REEs in all rice samples were also quantified. For exposure and risk assessment, 265 rice samples from major rice-producing counties in Taiwan were analyzed. A probabilistic framework based on Monte Carlo simulations was applied to estimate daily exposure doses and non-carcinogenic risks across different age and gender groups under daily rice consumption scenarios. Total REE concentrations ranged from 16.2–137 mg kg-1 in soils and 2.25–13.6 μg kg-1 in rice grains, with LREEs dominating over HREEs. Results from single extractions suggested that 0.10 M HCl and 0.05 M EDTA may serve as suitable indicators for predicting rice REE uptake. HREEs exhibited relatively higher bioavailability, attributable to their smaller ionic radii and higher charge density, while soil properties, such as pH, organic carbon, and iron oxides, were key factors controlling REE bioavailability. Dietary exposure assessment revealed that males generally exhibited higher exposure doses than females, and children aged 0–3 years had approximately twice the exposure and non-carcinogenic risk compared with other age groups. Sensitivity analysis identified rice intake rate, body weight, and rice concentrations of specific REEs (La, Ce, Eu, and Y) as the dominant contributors to total exposure and risk. Multivariate analysis further showed that rice grown on schist-derived alluvial soils exhibited REE compositions and fractionation patterns distinctly different from those grown on soils derived from sand shale, clay slate, or Quaternary deposit, supporting the potential of REEs and their fractionation indices as provenance indicators. Overall, this study clarifies the environmental behavior of REEs within the soil–rice system, provides the first quantitative assessment of dietary REE exposure and associated risks in Taiwan, and demonstrates the applicability of REEs in geographical origin authentication of rice. These findings offer a scientific basis for future risk management and communication regarding emerging contaminants in food systems, and highlight the forward-looking value of REEs for food safety and provenance verification.