High precision determination of silver isotope ratios in commercial products by MC-ICP-MS

Silver isotope ratios in several fortified commercial products were determined by multicollector inductively coupled plasma mass spectrometry (MC-ICP-MS). Mass discrimination and instrument drift were corrected for by a combination of internal normalization with Pd and standard-sample-standard bracketing, without assuming identical mass bias for Pd and Ag. The certified value of 1.07638 for Ag-107/Ag-109 in NIST SRM 978a was used for mass bias correction of Pd-106/Pd-108 in two adjacent solutions of SRM 978a. Their average was then used for mass bias correction to calculate the mass bias corrected Ag isotope ratio in the sample. An approximately 2.5-fold improvement in precision of determination of delta(107)/Ag-109 was obtained with the use of the proposed technique compared to data obtained solely with a standard-sample-standard bracketing approach. Internal and external precisions (2SD) of +/- 0.05 parts per thousand (n = 10) and +/- 0.04 parts per thousand (n = 19) for delta(107/109) Ag were obtained, suitable for the detection of varying Ag isotope abundances in commercial products fortified with Ag. Amongst the commercial products examined, no significant differences were detected in Ag isotopic compositions among NIST 978a standard, Ag high purity metal ( Johnson Matthey Plc), Ag Nanopowder (Sigma-Aldrich) and Silversoft socks. Significant differences in Ag isotope ratios were found among NIST SRM 978a standard, a colloidal Ag dietary supplement, decanethiol functionalized Ag nanoparticles and X-static socks. The X-static socks and decanethiol functionalized Ag nanoparticles have negative delta Ag-107/109 values of -0.08 +/- 0.05 parts per thousand and -0.15 +/- 0.02 parts per thousand (2SD, n = 4), respectively, and the colloidal Ag dietary supplement has +0.83 +/- 0.06 parts per thousand (2SD, n = 4). These observations suggest that such data may provide a useful tool for fingerprinting sources of Ag in the environment.