Abstract | Keywords | Export | Availability | Bookmark

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Microplastic pollution is one of the most critical environmental issues in the world right now. Many studies about human exposure to these micro-and nanoplastics (MNP) and the toxic effects they can cause are being conducted at this moment. Quantification and identification of these MNP lie at the bottom of solving this global problem. The most popular identification methods are Fourier-transform infrared spectroscopy and Raman spectroscopy. These techniques are lacking in non-destructive analysis with high sensitivity and low background noise. This thesis aims to develop an optical spectroscopy method to detect and identify microplastics after staining with Nile Red. The latter is a lipophilic solvatochromic dye. Staining will cause the particle to fluoresce, with its emission spectrum depending on its polarity. More polar plastics will have a more red-shifted emission maximum. Via hyperspectral analysis, the MNP can be identified based on this spectrum. Fluorescence lifetime imaging microscopy is also implemented. The fluorescence lifetime is the time that a fluorophore spends in the excited state, after absorbing a photon, before returning to the ground state. The lifetime depends on the fluorophore's micro-environment and therefore is also different for each MNP type. Differentiation is possible based on this lifetime value, but a graphical, fit-free phasor approach is also used. This allows for the mapping of the lifetime distribution in an image. When two different MNP types are present in one sample, they will each have a separate distribution cloud on the phasor plot, allowing for the graphical distinguishment of MNP. The plastics used are PP, PVC, HDPE, PET, and PS, these samples contain a range of shapes and sizes. A bought suspension of pure PS polymer particles of a fixed shape and size of 0,6 μm was also examined. The spectral data were processed in two ways. By calculating a ‘weighted average’ with the intensity values at each wavelength. This resulted in each MNP being attributed a weighted average value. After recording spectra of different particles, a statistical analysis, ANOVA and t-test, was done. This showed that the weighted averages of PP & HDPE and PS (self-made) & PVC do not differ significantly. The second option is through comparing the spectra themselves, using a relative similarity percentage. Because of the low similarity MNP gave when compared with themselves, only MNP with a significantly large similarity percentage could be distinguished. The same statistical tests were done on lifetime data. Measurements done on different particles of the same plastic showed that only HDPE & PET lifetimes did not differ significantly. Multiple measurements done on the same particle showed slightly different lifetimes. It could be concluded that HDPE & PET and PS (self-made and 0,6 μm) & PET cannot be differentiated. Finally, phasor plot analysis of samples containing two different plastics showed a possibility of recognizing MNP types based on the location of their phasor cloud on a phasor plot. When combining these two techniques, making use of the ’weighted average’ value and lifetime analysis, it could be possible to identify each plastic. The combination with similarly weighted averages, PP & HDPE give significantly different fluorescent lifetimes values and are possible to separate on a phasor plot. Still, both techniques have some flaws and improvement is possible.