Abstract
Submicroplastics (SMP) in food, beverages, water reservoirs and oceans are currently under-represented in literature. SMP characterization is very complex since i.SMP vary in composition; ii.SMP are ubiquitous; iii.SMP are heterogeneous, deriving from a multitude of degradation processes, size, shape and morphology are unique; and most importantly iv.SMP change identity in the exposure media, as high surface area and strong binding affinity enable the formation of eco-coronas and bio-coronas on the particles’ surface in different media. Corona formation is spontaneous, inevitable and depends on exposure conditions and matrix composition. To date, it remains largely unexplored how it determines SMP biological effects. More efforts are needed to compose suitable approaches towards SMP in environmental and food samples, and perform general and matrix-specific corona assessments. Many analytical techniques have been employed or proposed for SMP analysis, including imaging, spectroscopy and separation, all aiming to extract SMP and characterize specimen. However, a multitude of datasets is generated from this particle-oriented approach which must be correlated to improve extracted knowledge and reduce complexity. At the same time, multiblock chemometric and machine learning (ML) approaches are surprisingly unexplored. DOMANI aims at achieving two orthogonal objectives. The first is the comprehension of SMP eco-coronas through profiling, fractionation and quali-quantitative characterization of SMP in the exposure media. The second and most ambitious is the design of modelling approaches to dynamically profile and recognize clean and polluted water/food matrices in fast-result/fast-action mode, creating screening tools based on simpler analyses before resorting to time, money and environment-heavy ones. DOMANI puts together interdisciplinary expertise to devise matrix-oriented strategies, exploiting analytical, surface and colloidal chemistry and metrology, and benefiting of chemometrics, multiblock analysis and Machine Learning.
Dpartmental scientific manager
Valentina Marassi (PI)
Partnership
IMM-INRIM – Torino (Italy)
CNR-ISSMC – Faenza (Italy)