Practical comparison of microplastic identification methods: Manual, semi-auto and automated FTIR and Raman spectroscopy

Title
Practical comparison of microplastic identification methods: Manual, semi-auto and automated FTIR and Raman spectroscopy
Author(s)
Song, Young Kyoung; Eo, Soeun; Hong, Sang Hee; Shim, Won Joon
KIOST Author(s)
Song, Young Kyoung(송영경)Eo, So Eun(어소은)Hong, Sang Hee(홍상희)Shim, Won Joon(심원준)
Alternative Author(s)
송영경; 어소은; 홍상희; 심원준
Publication Year
2021-05-05
Abstract
The vibrational spectroscopy such as Fourier transform infrared (FTIR) and Raman microscope is popular microplastic analytical method for their quality and quantity. The spectroscopy can confirm the polymer composition, however, still it is unavoidable to human bias by manually selecting plastic-like particles for FTIR analysis using microscope. We tried to find best practice for microplastic analysis by reducing time demand, human bias (false negative) and automatic identification bias (false positive and false negative) using spectroscopy (FTIR and Raman microscopy). The manual analysis under transmission mode, semiautomated method using ultrafast mapping and spectrum profiling, and automated method using ultrafast mapping, spectrum profiling and fully automatic identification were compapred. In the automated method, to check false positive rate during identifcation, all spectra were manually duble-checked after automated method have done. The automated method took the shortest time (3.2±0.5 h, which is occupied time by operator) to analyze whole filterpaper (Ø25 mm), but the polymer types were limited to the number of profile spectrums, fiber could not be detected, and the rate of false positive was 80±15%. While, semi-auto analytical method using spectrum profiling was suitable for microplastic analysis in all aspects. It took shorter time than those of manual analysis (manual: 6.1±0.8 h and semi-auto: 4.0±0.6 h), fiber could be distinguished by chemical and mapping image. And 22±12% (fasle negative) more microplastic particles were detected using semi-auto than using maual analysis. Two types of spectroscopic analysis (FTIR and Raman) for microplastics were compared in four different methods, manual identification in ATR and transmission mode and semi-auto method using FTIR, and manual using Raman. The analysis duration was overwhelmingly taken for long time (34 ± 1.5 h) in Raman followed by transmission (5.6 ± 0.3 h), ATR (4.8 ± 0.9 h) and semiauto (3.6 ± 0.5 h). However, the detected number of microplastics using Raman were 1.4, 1.9 and 3 times higher than using semiauto, transmission and ATR mode, respectively. And the range of 0-50 µm was peaked in size distribution of Raman, but the range of 50-100 µm was peaked in size distribution of the others. Depending on the microplastic size range of interest, it could be appropriate identification method.
URI
https://sciwatch.kiost.ac.kr/handle/2020.kiost/41439
Bibliographic Citation
SETAC Europe 31st Annual Meeting, pp.54, 2021
Publisher
SETAC
Type
Conference
Language
English
Publisher
SETAC
Related Researcher
Research Interests

Microplastic pollution,Persistent Organic Pollutants,Oil Pollution,미세플라스틱 오염,잔류성 유기오염물질,유류오염

Files in This Item:
There are no files associated with this item.

qrcode

Items in ScienceWatch@KIOST are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse