Physicochemical characterization and quantification of nanoplastics: applicability, limitations and complementarity of batch and fractionation methods

A comprehensive physicochemical characterization of heterogeneous nanoplastic (NPL) samples remains an analytical challenge requiring a combination of orthogonal measurement techniques to improve the accuracy and robustness of the results. Here, batch methods, including dynamic light scattering (DLS), nanoparticle tracking analysis (NTA), tunable resistive pulse sensing (TRPS), transmission electron microscopy (TEM), and scanning electron microscopy (SEM), as well as separation/fractionation methods such as centrifugal liquid sedimentation (CLS) and field-flow fractionation (FFF)-multi-angle light scattering (MALS) combined with pyrolysis gas chromatography mass spectrometry (pyGC-MS) or Raman microspectroscopy (RM) were evaluated for NPL size, shape, and chemical composition measurements and for quantification. A set of representative/test particles of different chemical natures, including (i) polydisperse polyethylene (PE), (ii) (doped) polystyrene (PS) NPLs, (iii) titanium dioxide, and (iv) iron oxide nanoparticles (spherical and elongated), was used to assess the applicability and limitations of the selected methodologies. Particle sizes and number-based concentrations obtained by orthogonal batch methods (DLS, NTA, TRPS) were comparable for monodisperse spherical samples, while higher deviations were observed for polydisperse, agglomerated samples and for non-spherical particles, especially for light scattering methods. CLS and TRPS offer further insight with increased size resolution, while detailed morphological information can be derived by electron microscopy (EM)-based approaches. Combined techniques such as FFF coupled to MALS and RM can provide complementary information on physical and chemical properties by online measurements, while pyGC-MS analysis of FFF fractions can be used for the identification of polymer particles (vs. inorganic particles) and for their offline (semi)quantification. However, NPL analysis in complex samples will continue to present a serious challenge for the evaluated techniques without significant improvements in sample preparation.