There is still a vast number of unknown per- and polyfluoroalkyl substances (PFAS) in the environment remains unidentified. Current high-resolution screening methods are still unable to adequately find novel PFAS features without prior structural information. In view of this, a new data-driven algorithm used to calculate the similarity between spectra of PFAS is proposed and applied in generating molecular networks to screen unknown PFAS. Using this approach, 81 PFAS of 12 classes were identified in environmental samples collected around an industrial park; several substituted polyfluorinated ether sulfonate (X-PFESA, X=Cl, H, I) have been identified and 12 iodine substituted PFAS are reported for the first time. Standards of four Iodine-substituted polyfluorinated ether sulfonate (I-PFESA) were synthesized for structure confirmation and quantitative analysis, greatly enhancing the understanding of structure and environmental concentration of I-PFAS. Notably, the concentrations of X-PFESA were abnormally high, accounting for 10.8%-94.0% of the total PFAS concentration. Among which, the median concentration of H-PFESA was 63.0 ng/g, far exceeding the median concentrations of legacy perfluoroalkyl carboxylic acids (37.9 ng/g) and legacy perfluoroalkyl sulfonic acid (4.14 ng/g). Through replicating the industrial synthesis route in lab, the environmental source of I-PFESA has been confirmed as a by-product of the production process of 6:2 Cl-PFESA (F53-B) and the pattern of environmental transformation of X-PFESA has also been clarified. The embryotoxicity of zebrafish showed that 6:2 I-PFESA was more toxic than that of 6:2 Cl-PFESA with the same chain length, highlighting that it is necessary to pay attention to the toxic effects of I-PFESA. According to new PFAS spectra similarity, this study provides new perspectives and understanding for the environmental behavior and transformation of I-PFESA, and the in-depth toxicity mechanism needs to be further explored.

iodinated polyfluoroalkyl ester sulfonic acids, nontarget analysis, molecular network