Precise determination of lead isotope ratios by MC-ICP-MS without matrix separation exemplified by unique samples of diverse origin and history

Abstract

This study evaluates the feasibility of using the internal standard (IS) method and optimized regression model (ORM) to determine lead isotope ratios without separating Pb from many different types of matrices. The effect of typical interferents at increasing concentrations was investigated using model samples. We have demonstrated that Pb isotope ratio measurements without matrix separation give equally accurate and precise results as those obtained with matrix separation. Finally we used a new protocol to measure the isotope ratio of Pb in environmental, geological and archaeological samples with very limited availability (3 mg) and low lead content (30 mg kg⁻¹). Such samples have so far been problematic for MC-ICP-MS isotopic analysis due to the difficulties with separating such small amounts of lead using ion exchange resins. Lead isotope ratios were measured without separation of lead from the sample using two protocols. In the first, the internal standard method was used as a means of calibration. In the second, the optimized regression model was used. In this case there were no statistically significant differences between the protocols compared. Both approaches gave statistically similar results. The protocol using IS is recommended as it is significantly simpler and less laborious and consumes lower volumes of sample solution. The protocol we developed can be applied to a wide range of samples (following some recommendations). To demonstrate the usefulness of the new procedure, we applied it to different types of samples containing lead at different concentrations. In particular, we focused on historical samples because their availability is very limited and collecting a sufficiently large sample is always a major problem. Where possible, we compared our results with literature values. The new protocol, by eliminating the sample preparation step, allowed us to measure about four times more samples per day (about 12–15) compared to the classical method (with Pb separation using ion exchange resins). We also present, for the first time, a new method for the non-destructive sampling of manuscripts for isotopic analysis, involving the extraction of lead from the object into indicator papers soaked in 4,7-diphenyl-1,10-phenanthroline.