Direct determination of mercury at picomole L-1 levels in bottled water by isotope dilution cold-vapor generation inductively coupled plasma mass spectrometry

The mercury concentration in 17 commercially available bottled waters (artesian, distilled, carbonated, and spring) from 7 different countries determined by isotope dilution cold-vapor inductively coupled plasma mass spectrometry (ID-CV-ICP-MS) ranged from the blank limited detection limit of <0.10 ng L-1 to 2.32 ng L-1 (ppt). Highly enriched Hg-201 isotopic spike is added to approximately 20 mL water and thoroughly mixed. The Hg+2 in the sample is reduced on line with tin(II) chloride and the elemental Hg vapor is separated in a liquid-matrix'' separator and introduced directly into a quadrupole ICP-MS where the Hg isotope ratios (Hg-201/Hg-202) are measured in time-resolved analysis mode. The primary advantages of this method are (1) high sensitivity, the instrument detection limit is less than 0.05 ng L-1, (3σ), (2) very low chemical blank, the average blank (n = 3) is 0.17 ng L-1 +/- 0.03 ng L-1, and (3) high accuracy of isotope dilution-the accuracy is limited by blank and counting statistics. All waters tested, including the major sellers in Europe and the United States, were approximately 1000 times lower than both the U.S. Food and Drug Administration (FDA) limit of 2 μg L-1 Hg (ppb) and the international World Health Organization (WHO) limit of 1 μg L-1 Hg.