An in situ nuclear magnetic resonance (NMR) tube extraction coupled with fluorine-19 (19F) NMR spectroscopy was developed to quantify extracted fluorine. A solid-liquid extraction of Viton A was performed in situ in an NMR tube whereby an organic solvent was added to solid cross-linked Viton A (10 mg) that was positioned below the detection region of the NMR probe. The extracted fluorine was measured while relying on diffusion of fluorine-containing molecules from the fluoropolymer into the solvent. The in situ approach was used to correlate extraction efficiency as a function of polymer swelling. Permeation, calculated from swelling, was 2-fold greater using acetone compared to methanol. Acetone greater extracted fluorine concentration (ng F/g fluoropolymer) after 302 min of extraction. This method was also applied to a Viton A cord, a Viton O-ring, and polytetrafluoroethylene. Analysis by 19F diffusion-ordered spectroscopy NMR indicated that the extracted fluorine observed by 19F NMR was composed of non-cross-linked Viton A liberated from the polymer matrix. For individual PFAS analysis, liquid chromatography-quadrupole time-of-flight mass spectrometry (LC-QTOF) showed increased extraction of perfluorooctanoic acid (PFOA) and perfluorohexanoic acid (PFHxA) as a function of time and swelling. Although it appeared that acetone extracted PFOA and PFHxA with greater efficiency than methanol, concentrations were not statistically significantly different. Thus, swelling with methanol was sufficient for extracting nonpolymeric molecules from Viton A. Therefore, measures of extracted fluorine by 19F NMR or other nondiscriminatory methods should not be interpreted as individual PFAS, without LC-QTOF confirmation. Our in situ approach minimizes the need for ex situ sample preparation and can be applied to other in situ NMR experiments such as solid leaching or solid-liquid partitioning.
Kim-Fu et al. (Sun,) studied this question.