Harnessing machine learning to improve identification and quantification of micro- and nanoplastics (MLIdent)
The analysis of micro- and nanoplastics (MNPs) is challenging. This is due to the fact that MNPs are present at low concentrations in complex matrices, as well as the broad range of synthetic polymers and size ranges. It is important to develop a standardized and robust approach to accurately quantify and characterize MNPs.
Goal
The researchers want to develop knowledge about the chemical fingerprints obtained after pyrolysis. Additionally, they want to know how MNPs' characteristics and polymer mixtures affect the chemical fingerprints in the presence of matrices such as human blood. The project aims to provide a robust statistical framework that will improve both the identification and quantification of MNPs in complex samples. Finally, MLIdent will also develop a high-resolution mass spectrometry (HRMS) method and evaluate its added value with respect to conventional low-resolution MS methods.
Approach/method
In this project pyrolysis coupled to gas chromatography and mass spectrometry (Py-GC-MS) will be used. Through the collection of a large dataset, combined with advanced data analysis approaches, this project will provide a robust statistical framework to identify and quantify MNPs.
Collaboration partners
This project is carried out by prof. dr. M.M. Lamoree, dr. F.M. Béen and colleagues from Vrije Universiteit Amsterdam. They collaborate with Utrecht University.
(Expected) results
This project will provide knowledge about chemical fingerprints obtained after pyrolysis. Additionally, a statistical framework will be developed, together with an HRMS method, in order to improve the identification and quantification of MNPs in human and environmental samples.
Continuation
The experimental data and data analysis approaches will be transferable to other laboratories that analyse MNPs in human and environmental samples. This will provide fundamental solutions that can be implemented by the whole scientific community involved in the analysis of MNPs. The project will also provide a better understanding of the effect of polymer characteristics and mixtures on chemical fingerprints relevant to MNPs analysis and plastic recycling.