Critical evaluation of internal normalization and standard-sample-bracketing for accurate ⁸⁷Sr/⁸⁶Sr analysis

The strontium isotope ratio ⁸⁷Sr/⁸⁶Sr is a key tracer with wide-ranging applications in geochemistry, hydrology, paleoclimatology and migration research. To make sound interpretations of ⁸⁷Sr/⁸⁶Sr isotope ratios in the context of biogeosciences, researchers need high quality data. In this contribution, we critically evaluate the accuracy of two conceptually different analytical protocols for ⁸⁷Sr/⁸⁶Sr determination on the example of a large dataset (n = 135) comprising biogenic and abiogenic materials.  

Water, soil extracts, and hydroxyapatites (tooth enamel) were prepared according to established procedures and analyzed by solution-based multi-collector inductively coupled plasma mass spectrometry (MC ICP-MS). For the calibration we used two protocols: internal normalization (also termed internal mass bias correction or internal calibration) and standard-sample-bracketing (external calibration). Isotope dilution mass spectrometry was not considered suitable, because this calibration approach becomes very time- and cost-intensive when applying to a large sample set.

Analysis of water, soil extracts and hydroxyapatites showed that the majority of ⁸⁷Sr/⁸⁶Sr isotope ratios which were determined by internal normalization shifted towards higher values in comparison to data determined by standard-sample-bracketing. Extensive evaluations ruled out sample preparation or measurement errors. Instead, internal normalization yielded biased data, because it is based on the assumption that all samples have the same ⁸⁸Sr/⁸⁶Sr isotope ratio, which can be used for normalization. However, in 90% of the investigated samples the ⁸⁸Sr/⁸⁶Sr significantly deviated from the assumed invariant value; in particular, the ⁸⁸Sr/⁸⁶Sr that is conventionally expressed as δ(⁸⁸Sr/⁸⁶Sr)_SRM987 ranged from -1.01‰ to 0.20‰. As a consequence, internally normalized ⁸⁷Sr/⁸⁶Sr data biased by up to 0.00043, which was 2-times larger than previously predicted by theoretical calculations. These results demonstrate that the choice of calibration method has a much higher impact on the accuracy of ⁸⁷Sr/⁸⁶Sr isotope ratios than initially expected. The implication of these findings in biogeoscience applications will be discussed.

This project has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement n°856453 ERC-2019-SyG).