This information comes from the dataset README covering the NTA data and associated metadata for this dataset. Dust samples were collected from home vacuum bags and sieved (<150 µm). Internal standard (MPFAC-MXA, Wellington Labs) was spiked , dependent on dust mass, to a concentration of 10 ng/g. Native standards (PFAC-MXA in methanol, Wellington Labs) were dosed into each calibration and QC standard. The next day, 5 ml methanol was added to each sample. Samples were sonicated, centrifuged, and cleaned up. Samples were then blown down using a dry nitrogen gas stream not to dryness; samples were reconstituted to 0.5 ml with methanol as needed. A 100 µl aliquot of sample extract was combined with 300 µl mobile phase A (see below).

Sample extracts were analyzed by UHPLC-MS/MS on a Thermo Scientific (Waltham, MA) system consisting of a TriPlus RSH autosampler/injector, Vanquish Horizon UPLC/pump system, and Thermo Orbitrap Fusion tribrid mass spectrometer. Chromatographic separation was performed using a Restek (Bellefonte, PA) Raptor C18 column at 55 °C. A 15-minute reverse-phase gradient was applied consisting of mobile phases A (95:5 v/v deionized water:methanol containing 2.5 mM ammonium acetate) and B (5:95 v/v deionized water:methanol containing 2.5 mM ammonium acetate). Negative-polarity electrospray ionization [ESI (-)] was applied first, with positive-polarity applied afterwards. Mass spectra were collected using a resolving power of 50,000, with preferred-ion data-dependent acquisition (DDA) applied to select molecular features for MS2 fragmentation.

Samples were analyzed in a single batch. The full batch was repeated twice with randomized sample order. Separate batches were run for ESI(+) and ESI(-) analysis, for a total of six sample batches. Targeted methanolic calibration standards at concentrations of 1-1000 ng/g Wellington PFAC-MXA PFAS mixture were run at the start and end of the first batch. Method blanks, QC standards, pooled samples, and solvent-only blanks were run every ten samples across all batches.

After data collection, chromatograms were processed and peak areas integrated in Thermo Scientific Xcalibur Quan Browser 4.3 for targeted quantitation. For nontargeted identification, chromatograms and associated mass spectra were processed in Thermo Scientific Compound Discoverer 3.3. Features were prioritized for expert identification based on a combination of high maximum abundance, strong match to library spectra, negative mass defect (for PFAS), presence as a member of a likely hologous series (for PFAS and surfactants); and/or presence of diagnostic PFAS-related fragments in their MS2 spectra. Overall, 742 features of interest at confidence 1-3 were identified, as well as 7 confidence-5 features meriting inclusion in the final dataset, and excluding hundreds of features representing false positives/adducts/etc. tentatively identified by Compound Discoverer.