微塑料
化学成像
高光谱成像
拉曼光谱
表面增强拉曼光谱
纳米技术
样品(材料)
采样(信号处理)
计算机科学
环境科学
人工智能
材料科学
化学
环境化学
物理
光学
计算机视觉
拉曼散射
色谱法
滤波器(信号处理)
作者
Cheng Fang,Yunlong Luo,Ravi Naidu
标识
DOI:10.1016/j.trac.2023.117158
摘要
As emerging contaminants, microplastics and nanoplastics pose analytical challenges to the scientific community due to the small size, diverse composition and complex environmental background. The research on nanoplastics is far behind that on microplastics because the shrink size and the weak signal lead to more challenging analysis. Herein we review the recent advancements on their analysis including sampling, sample preparation, test and aftermath data analysis. We begin by clarifying the unique properties of microplastics and nanoplastics that complicate the sampling and the sample preparation processes, which have been rarely reported but should be emphasised for the subsequent analysis. For the analysis, there are various techniques for morphological and chemical characterisations, including microscope, element analysis, mass spectroscopy and molecular spectroscopy. They are compared herein to highlight their advantages and disadvantages. Because the microplastics and nanoplastics can have their own sub-structures, there might be bias if the non-imaging analysis is conducted, such as using a single spectrum analysis (point analysis) conducted at a selective position that is only a partial surface area of the whole structure (surface and bulk). Imaging analysis via micro-IR and micro-Raman spectroscopy particularly Raman imaging shows some advantages to overcome this bias. However, Raman imaging is a time-consuming process with a diffraction-limited resolution problem that is also discussed herein. Moreover, it is difficult to convert the scanning hyperspectral matrix to image. To address this, algorithms of chemometrics and artificial intelligent (AI) can be utilised to decode the hyperspectral matrix that acts as a big data, and re-construct the image towards deconvolution. The current analysis techniques should be either improved or combined for the emerging contaminants’ analysis. Overall, this review summarises the analysis challenges and advances, and also suggests the future research directions.
科研通智能强力驱动
Strongly Powered by AbleSci AI