A multi-isotope deconvolution approach to correct for hydride and xenon interferences during double spike barium isotope analyses with MC-ICP-MS

Precise and accurate stable barium isotope (δ 138/134 Ba relative to the NIST3104a standard) measurements are crucial for geochemical studies, as natural variations can be small. Double spike Ba isotope analysis by MC-ICPMS is typically the preferred method for generating the most reliable Ba isotope data as it effectively corrects for instrumental mass fractionation and isotope fractionation during chemical separation, while also providing precise concentrations via isotope dilution. However, measurement of 50 ng or less of Ba using high sensitivity interface cones can be challenging due to the effects of isobaric interferences from trace Xe in the Ar plasma. Here we demonstrate that interferences from Ba-hydrides (BaH + ) also have significant effects on double spike corrected δ 138/134 Ba-values and, therefore, must be assessed and corrected. We present a multi-isotope deconvolution strategy utilizing a 135 Ba 136 Ba double spike and three independent inversions using different un-spiked isotope pairs ( 134 Ba 137 Ba, 134 Ba 138 Ba, and 137 Ba 138 Ba) to solve the double spike equation. Each inversion independently calculates δ 138/134 Ba that is differently affected by both background Xe and hydride interferences during the measurement. We show that the Xe correction is smallest for the 137 Ba 138 Ba inversion, which allows accurate Ba isotope data to be obtained for this inversion even when <5 ng of Ba is consumed during triplicate sample measurements. However, the 137 Ba 138 Ba inversion also carries the largest effect from hydride formation, which we correct for using a combined 134 Ba- 138 Ba/ 137 Ba- 138 Ba inversion method. This approach produces long-term external reproducibility for rock reference materials of ±0.025 ‰ (2SD) when consuming less than 50 ng of Ba, which represents an improvement over previously reported Ba double-spike MC-ICP-MS methods at similar Ba consumption. We conclude that the double spike multi-isotope deconvolution approach can accurately assess and correct for elemental and hydride interferences for isotope systems with 5 or more stable isotopes suggesting that isotope systems such as Zr, Mo, Cd, Sn and Te might benefit from this strategy.