Comparison of two rapid automated analysis tools for large FTIR microplastic datasets

Area of Science:

• Environmental Science
• Analytical Chemistry
• Polymer Science

Background:

• Standardization is lacking in microplastic (MP) research, hindering comparability and reproducibility.
• Micro-Fourier-transform infrared (µFTIR) imaging is a powerful tool for MP analysis, but large datasets require automated algorithms.
• The reliability and comparability of existing MP data analysis algorithms are not well understood.

Purpose of the Study:

• To compare the performance of two established automated data analysis algorithms for microplastic identification.
• To assess the agreement between algorithms regarding MP abundance, polymer composition, and size distribution.
• To highlight the importance of algorithm comparison for improving microplastic data comparability.

Main Methods:

• Utilized focal plane array (FPA) based micro-FTIR imaging for MP analysis in environmental water samples.
• Applied and compared two distinct automated data analysis pipelines: siMPle/MPAPP and BPF.
• Analyzed microplastics within the size range of 11-500 µm.

Main Results:

• Both algorithms demonstrated overall good agreement in microplastic quantification.
• Discrepancies were observed in the identification of specific polymer types and smaller microplastic size classes (11-500 µm).
• The study identified specific areas where algorithm outputs varied, impacting data consistency.

Conclusions:

• Automated algorithms for microplastic analysis show promise but require careful validation.
• Direct comparison of algorithms is essential to understand their limitations and ensure reliable microplastic data.
• Further harmonization of analytical methods and data processing is needed for robust microplastic research.