Advances in Automated Sample Preparation and Analysis of Polyfluoroalkyl Substances in Environmental Water Samples

Characterization of Polyfluoroalkyl Substances in the Environment
Oral Presentation

Prepared by J. Pyke1, T. Anumol2, H. Zhao2
1 - Agilent Technologies, Inc., 5301 Stevens Creek Blvd, Santa Clara, California, 95051, United States
2 - Agilent Technologies, Inc., 2850 Centerville Road, Wilmington, DE, 19808, United States


Contact Information: james_pyke@agilent.com; +14085534321


ABSTRACT

Per/Polyfluoroalkyl substances (PFAS), are substances that have uniquely desirable properties for use in various industries. However, their wide-ranging use leads to emission into the environment, and as PFAS are persistent and bioaccumulate in the environment and wildlife, they are contaminants of concern. The United States Environmental Protection Agency (US EPA) has issued drinking water health advisories for perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) at 70 ng/L combined.
PFAS have a few or all of the hydrogen atoms in an alkyl chain replaced by fluorine. The alkyl chain is caped with various classes of chemical moieties, carboxylic acids and sulfonates being the most regularly monitored to date. Analytical techniques make use of the alky chain length to chromatographically resolve PFAS with reverse phase chromatography and to achieve Minimum Reporting Limits (MRL), sample enrichment steps, making use of various chemistries, are often required for analysis.
Sample preparation and analysis is not only costly and burdensome, but due to the prevalence of PFAS in materials common in laboratories, consumables and solvents, there is an increased risk of reporting false positive results. To mitigate costs and risks, we describe an automated analytical method for the analysis of PFAS in environmental water samples. The developed method was demonstrated to have linear calibration curves for all compounds listed in USEPA 537 and be robust across multiple injections of matrix with no significant change in response in a 200+ batch of analysis and proficiency test samples confers with published results.