A sensitive and accurate method for the determination of perfluoroalkyl and polyfluoroalkyl substances in human serum using a high performance liquid chromatography-online solid phase extraction-tandem mass spectrometry

A selective, sensitive, and accurate analytical method for the measurement of perfluoroalkyl and polyfluoroalkyl substances (PFASs) in human serum, utilizing LC-MS/MS (liquid chromatography-tandem mass spectrometry), was developed and validated according to the Centers for Disease Control and Prevention (CDC) guidelines for biological sample analysis. Tests were conducted to determine the optimal analytical column, mobile phase composition and pH, gradient program, and cleaning procedure. The final analytical column selected for analysis was an extra densely bonded silica-packed reverse-phase column (Agilent XDB-C₈, 3.0×100mm, 3.5μm). Mobile phase A was an aqueous buffer solution containing 10mM ammonium acetate (pH=4.3). Mobile phase B was a mixture of methanol and acetonitrile (1:1, v/v). The gradient program was programmed by initiating a fast elution (%B, from 40 to 65%) between 1.0 and 1.5min, followed by a slow elution (%B: 65-80%) in the period of 1.5-7.5min. The cleanup procedures were augmented by cleaning with (1) various solvents (isopropyl alcohol, methanol, acetonitrile, and reverse osmosis-purified water); (2) extensive washing steps for the autosampler and solid phase extraction (SPE) cartridge; and (3) a post-analysis cleaning step for the whole system. Under the above conditions, the resolution and sensitivity were significantly improved. Twelve target PFASs were baseline-separated (2.5-7.0min) within a 10-min of acquisition time. The limits of detection (LODs) were 0.01ng/mL or lower for all of the target compounds, making this method 5 times more sensitive than previously published methods. The newly developed method was validated in the linear range of 0.01-50ng/mL, and the accuracy (recovery between 80 and 120%) and precision (RSD<20%) were acceptable at three spiked levels (0.25, 2.5, and 25ng/mL). The method development and validation results demonstrated that this method was precise, accurate, and robust, with high-throughput (∼10min per sample); thus suitable for large-scale epidemiological studies.