Formatted Title
Profiling of Target and Nontarget PFAS in Agricultural Soils
Background/Objectives
With the increasing public concern about per- and polyfluoroalkyl substances (PFAS), characterization of such contaminants in different environmental metrics have attracted extensive attention. PFAS can infiltrate into agricultural landscapes via multiple channels: the use of PFAS-contaminated biosolids as fertilizer, irrigation with contaminated water, atmospheric deposition, and undesirable anthropogenic activities, such as the spill of aqueous film fire-fighting foams (AFFFs). Soil matrices can serve as significant reservoirs for PFAS, leading to their entry into the food chain. Our study focuses on the detection and quantification of PFAS in agricultural fields by combining the target analysis by EPA Method 1633 and nontarget analysis by nano-electrospray ionization and high-resolution mass spectrometry (Nano-ESI-HRMS).
Approach/Activities
Seventeen soil samples were collected from tomato fields, corn fields, hemp fields, and deep layer soil on a farm near Newton, New Jersey. These samples were processed following EPA Method 1633. Large rocks were removed from the soil samples using tweezers, and large soil clods were sieved out to obtain finely divided soil samples. Prior to extraction, surrogates were added to the homogenized samples. After ensuring homogeneity, 5 g of these prepared samples were extracted using 0.3% methanolic ammonium hydroxide three times. Subsequently, the extract solution was passed through preconditioned weak anion exchange (WAX) SPE cartridges. Then the adsorbed PFAS were eluted using methanolic ammonium hydroxide. The resulting elution samples were first evaporated and then dissolved in 1 ml of methanol. Following the sample preparation, internal standards were incorporated, and the specimens were subjected to analysis using high-performance liquid chromatography (HPLC) coupled with a triple quadrupole mass spectrometer (LC/MS/MS) to determine the target PFAS compounds. In parallel, the extracts were also analyzed by Nano-ESI-HRMS using a high-resolution Q Exactive hybrid quadrupole−Orbitrap MS for suspect screening and non-target analysis. HRMS MS1 data were employed to identify putative PFAS features present in soil metrics. Furthermore, collision-induced dissociation (CID) was used to investigate the structure of these novel PFAS.
Results/Lessons Learned
Target analysis by LC/MS/MS revealed the detection of 24 PFAS across 17 agricultural samples. PFOS and PFOA were found as the most dominant PFAS since they were consistently identified across all soil samples in the concentration range of 30.1 to 18,220 ng/kg. Shorter chain PFAS, like PFBA and PFHxA, were also present largely in those tomato fields and deep-layer soil samples. Additionally, PFAS precursors, like N-EtFOSA, N-MeFOSAA, and FOSA were detected in seven out of 17 collected soil samples.
Nontarget analysis by Nano-ESI-HRMS revealed several PFAS features, including a dominant series of PFAS-adducts found in tomato field and deep-layer soil samples. These analogue adducts displayed a mass difference of 49.99681 m/z, reflecting variations in their perfluoroalkyl component chain lengths. A detailed fragmentation analysis of these PFAS-adduct ions via CID can provide insights into potential molecular structures of this PFAS-adduct series. Additionally, another unsaturated PFAS feature with a possible formula of C11HF15O2- was found in all six corn field samples. The characterization of these unique PFAS showed variances of contamination across different types of agricultural fields.