Quantitative determination of perchlorate in bottled water and tea with online solid phase extraction high-performance liquid chromatography coupled to tandem mass spectrometry

Abstract

Due to the similarity in ionic radius, perchlorate has been reported to inhibit the iodide intake in the thyroid gland, which may lead to low heart rate, weight gain, and fatigue. In recent years, the presence of perchlorate in drinking water, surface water, soil, and food supplies in the United States has raised a great concern on establishing the maximum residue limit (MRL) for perchlorate to reduce its possible adverse influence on human health. US EPA currently puts perchlorate on the final third Contamination Candidate List (CCL3) and suggests a health reference level at 4.9μgL−1. The MRL of perchlorate was therefore set at 5.0μgL−1 by the authors for method validation. In this study, large volume injection (up to 1-mL) and online solid phase extraction (SPE) were utilized for pre-concentrating perchlorate ions and removing unretained matrix components prior to reversed-phase HPLC analysis using ESI-tandem MS under the negative mode. After eluting perchlorate from online SPE, 0.1% formic acid solution was utilized for isocratic HPLC analysis without any organic solvent. Multiple reaction monitoring (MRM) and the internal standard, Cl18O4−, were utilized for quantitatively determining perchlorate in bottled water and bottled tea samples. Two linear ranges, 0.05–0.50μgL−1 and 0.50–10.00μgL−1, were established to better estimate the residual amounts of perchlorate in bottled water samples with a method detection limit (MDL, signal-to-noise ratio of 3) of 0.01μgL−1. The linear range was 1.50–10.00μgL−1 for bottled tea samples with a MDL of 0.5μgL−1. In addition, the proposed method was further validated based on the EU Commission Decision 2002/657/EC, including within-laboratory reproducibility, decision limit (CCα), and detection capability (CCβ) for bottled water and bottled tea samples. The intra-day/inter-day precision and accuracy as well as within-laboratory reproducibility were determined by calculating the relative standard deviation (RSD) at three spiked levels (0.5 MRL, 1 MRL, 1.5 MRL). The within-laboratory reproducibility (n=18) for both bottled water and bottled tea samples, spiked at MRL (5.0μgL−1) of ClO4−, was less than 10%. The values of CCα/CCβ were reported as 5.43/5.74μgL−1 and 5.03/5.75μgL−1 for bottled water and bottled tea samples, respectively.