Formatted Title:

High-Resolution Mass Spectral Tool for PFAS Forensic Analysis

Background/Objectives:

Per- and polyfluoroalkyl substances (PFAS) are widely used for many commercial and industrial applications, including their use in AFFF by DoD, oil and gas industry, manufacturing facilities, chrome plating operations, etc. Given PFAS are ubiquitously present in the environment, understanding the distinct chemical signatures of different sources is very important. As the terminal transformation products resulting from different PFAS precursors used for different applications are similar, it is very challenging to delineate different sources using routine analytical techniques. As the number of PFAS-contaminated sites identified globally are on the rise, the need for chemical forensic approach to understand the fate and tracking the PFAS sources has grown in the past decade.

Approach/Activities:

Using a combination of high-resolution mass spectrometry techniques and machine learning tools, Battelle has developed a chemical forensic technique, PFAS Signature^®, for PFAS source differentiation and tracking. The PFAS Signature^® source library was developed by compiling high-resolution mass spectral data of samples collected from different known source scenarios and AFFF materials. All of the samples were analyzed using ultra-performance liquid chromatography coupled to a quadrupole time-of-flight (QTOF) mass spectrometry, in combination with PFAS targeted analysis. A library was developed, which is a structure-based library containing over 600 PFAS compounds. Data generated by QTOF are being investigated using both suspect screening methods. Machine learning tools are applied to correlate the PFAS profiles of the unknown source samples to the respective known sources in the library.

Results/Lessons Learned:

The results have shown that the high-resolution mass spectral data of different sources show unique precursor signatures which help differentiate distinct sources. Applying PFAS Signature^®, we were able to correlate the sources of some of the samples collected from contaminated sites. Machine learning analysis of features filtered using mass defect and the suspect list showed good correlation between both the data. A case study on understanding the influent sources and the fate of PFAS from AFFF in wastewater treatment plant systems will be discussed.

Daniel Longbrake

Affiliation

Track

Refine Results

High-Resolution Mass Spectral Tool for PFAS Forensic Analysis