Department of Ecology and Evolutionary Biology, University of Toronto, St. George Campus, Toronto, Ontario, Canada M5S 3B2.
Thermo Fisher Scientific, Wilmington, Delaware 19810, United States.
Anal Chem. 2021 Dec 7;93(48):15878-15885. doi: 10.1021/acs.analchem.1c02549. Epub 2021 Nov 23.
Raman spectral libraries specific to microplastics demonstrated improved spectral matching results when weathered plastics and a variety of particle colors and morphologies were included. Here, we explore if this is true for Fourier transform infrared (FTIR) spectroscopy as well. We present two novel databases specific to microplastics using attenuated total reflection (μATR-FTIR): (1) an FTIR library of plastic particles (FLOPP), containing 186 spectra from common plastic items, across 14 polymer types and (2) an FTIR library of plastic particles sourced from the environment (FLOPP-e), containing 195 spectra across 15 polymer types. Both libraries include particles from a variety of sources, morphologies, and colors. We demonstrate the applicability of these libraries for microplastics research by comparing spectral match results from two microplastic datasets. For this, we use different combinations of libraries including: commercially available reference libraries, an open-access polymer library, and FLOPP and FLOPP-e. Among tests, the greatest mean HQI result was achieved when the greatest number of libraries was included. This work demonstrates that spectral libraries specific to plastic particles found in the environment improve the accuracy of spectral matching and are best used in combination with commercial libraries containing chemical components that are commonly found within plastics and other anthropogenic particles. Multivariate principal component analyses of FLOPP and FLOPP-e spectra confirmed differences among polymer types and higher variation in principal component scores among weathered particles, but no patterns were observed among particle colors or morphologies. These results demonstrate that ATR-FTIR analyses are sensitive to weathering of plastics but not to particle color and morphology.
特定于微塑料的拉曼光谱库在包含风化塑料和各种颗粒颜色和形态时,显示出改进的光谱匹配结果。在这里,我们也探讨傅里叶变换红外(FTIR)光谱是否也是如此。我们使用衰减全反射(μATR-FTIR)呈现了两个特定于微塑料的新型数据库:(1)包含 14 种聚合物类型的 186 种常见塑料颗粒的 FTIR 库(FLOPP),(2)源自环境的微塑料 FTIR 库(FLOPP-e),包含 15 种聚合物类型的 195 种光谱。这两个库都包含来自各种来源、形态和颜色的颗粒。我们通过比较两个微塑料数据集的光谱匹配结果来证明这些库在微塑料研究中的适用性。为此,我们使用了不同的库组合,包括:商业上可用的参考库、开放获取的聚合物库以及 FLOPP 和 FLOPP-e。在测试中,当包含最多数量的库时,HQI 的平均得分最高。这项工作表明,特定于环境中发现的塑料颗粒的光谱库可以提高光谱匹配的准确性,并且最好与包含常见塑料和其他人为颗粒化学组成的商业库结合使用。对 FLOPP 和 FLOPP-e 光谱的多元主成分分析证实了聚合物类型之间的差异,以及风化颗粒中主成分得分的更高变化,但没有观察到颗粒颜色或形态之间的模式。这些结果表明,ATR-FTIR 分析对塑料的风化很敏感,但对颗粒颜色和形态不敏感。