文献类型：期刊文献

中文题名：数据驱动的新污染物单体环境分析化学

英文题名：Data-driven single-entity environmental analytical chemistry for new pollutants

作者：李洪双[1];田思语[1];杨泽楷[1];程梦源[1];汤雯[1];赵娴[1];李浦风[1];罗丹[1];谢雪迎[1];邱恺培[1,2,3]

机构：[1]国家环境保护化工过程环境风险评价与控制重点实验室,华东理工大学资源与环境工程学院,上海200237;[2]上海污染控制与生态安全研究院,上海200092;[3]上海市环境保护化学污染物环境标准与风险管理重点实验室,上海200237

年份：2024

卷号：43

期号：2

起止页码：250

中文期刊名：分析试验室

外文期刊名：Chinese Journal of Analysis Laboratory

收录：CSTPCD;;EI(收录号：20241115720599);北大核心:【北大核心2023】；CSCD:【CSCD_E2023_2024】；

基金：国家自然科学基金(21972041,22006037);上海市自然科学基金(23ZR1416300)资助。

语种：中文

中文关键词：新污染物;纳米孔道电化学技术;算法;有限元模拟

外文关键词：new pollutants(NPs);nanopore electrochemical technology;algorithm;finite element method simulation(FEM)

摘要：新污染物是指未纳入监管、但存在较大环境健康风险的物质,具有种类繁多、环境持久性、生物累积性等特点。基于纳米孔道电化学的单体环境分析是一类新兴的监测技术,其测量原理是利用单个待测物过孔时产生的特征电流来定性,并用特征电信号的出现频率定量。本文聚焦两类重点关注的新污染物—全氟/多氟烷基化合物和微纳塑料,分别开发了基于生物纳米孔的单分子分析算法和基于毛细管固体孔的单颗粒分析算法。针对双三氟甲基苯甲酸异构体,通过单分子算法对原始电信号进行多维特征提取,在不使用标准品的情况下,将分类准确率从一维的74.3%提升至五维的92.6%。针对微纳塑料,利用单颗粒算法,实现了粒径4~6μm的羧基聚苯乙烯微球精准区分,准确率为100.0%。进一步通过有限元模拟,提取了受单因素影响的参数,为新污染物精准监测提供了新思路。
New pollutants(NPs) refer to substances that have not been regulated yet, but may pose great environmental and health risks, due to the characteristics of a wide variety, environmental persistence and bioaccumulation. Precision monitoring is a prerequisite for effective treatment of NPs. Single-Entity environmental analysis based on nanopore electrochemistry is an emerging monitoring method, of which the target can be identified by using the characteristic current generated when single analyte passes rough the nanopore, and be quantified by the frequency of events. In this work, two typical NPs, perfluorinated alkyl compounds and micro plastics are focused, while single-molecule analysis algorithm based on biological nanopores and single-particle analysis algorithm based on solid-state pores are developed, respectively. For the isomer of bistrifluoromethyl benzoic acid(BBA), the multi-dimensional features were extracted from the original signal by running the single-molecule algorithm. The classification accuracy was improved from 74.3% of onedimensional to 92.6% of five-dimensional without using standards. For microplastics, the single-particle algorithm was used to accurately distinguish carbox polystyrene(PS) microspheres with particle size of 4-6 μm, with an accuracy of 100.0%. Finally, the parameters affected by single factors through single-particle finite element method simulation(FEM) were further extracted, which provided a new idea for the accurate monitoring of new pollutants.