Software Comparison for Nontargeted Analysis of PFAS in AFFF-Contaminated Soil

Per- and polyfluoroalkyl
substances (PFAS) are an emerging class
of toxic environmental contaminants. Over 7500 PFAS exist, but reference
standards are available for less than 2% of compounds. Nontargeted
analysis using liquid chromatography–high-resolution tandem
mass spectrometry is therefore an essential technique for increasing
the analytical coverage of PFAS present in environmental samples.
However, typical nontargeted data analysis is laborious and has a
steep learning curve. Recently, FluoroMatch, a new open source, vendor
neutral software, was published specifically for automating data processing
for nontargeted analysis of PFAS and generating PFAS libraries. Here,
we analyze soil contaminated with PFAS based aqueous film forming
foam (AFFF) and compare the results produced by data analysis workflows
using FluoroMatch and Compound Discoverer, an established nontargeted
analysis program. High-confidence PFAS annotations were nearly identical
between the methods, with 27 out of 32 compounds found using both
Compound Discoverer and the modular version of FluoroMatch. Twenty-two
high-confidence annotations were found using the comprehensive FluoroMatch
Flow. The FluoroMatch method was faster and required significantly
less manual curation than the Compound Discoverer method. Both platforms
produced high-quality data that were useful for assessing PFAS contamination
in the soil.