Determination of Stable Strontium Isotopic Compositions by MC-ICP-MS
Juan Xu,Shouye Yang,Yueheng Yang,Yongsheng Liu,Xiaolei Xie
DOI: https://doi.org/10.46770/as.2020.02.003
2020-01-01
Atomic Spectroscopy
Abstract:The investigation of variations within the non-traditional stable Sr isotope composition in various natural systems has gained a broader interest in recent years. Although significant improvements in the analytical precision for isotopic ratio analysis have been made, the accurate determination of stable Sr isotopic ratios is still a challenge. This study shows the key controlling factors affecting the precise determination of stable Sr isotopic ratios by using Neptune Plus MC-ICP-MS, including matrix effect, internal standard correction and nebulizer fractionation effect. Further, a novel step is presented in the Sr purification by using Sr-spec (R) resin, which increased the recovery rates to > 98.5%. In this work, the mass bias effects between Zr-92/Zr-90 and Sr-88/Sr-86 were investigated and compared, and a 3 similar to 5-fold improvement in the precision of the determination of Sr-88/Sr-86 was obtained compared to that based on only the standard-sample-bracketing technique. Moreover, nebulizer fractionation was illustrated by using an in-house standard solution, Alfa Sr, and the seawater standard, IAPSO. As for the matrix effect, residual Ca in the purified Sr fraction was evaluated and showed insignificant influence. Overall, a feasible, highly-efficient procedure for the stable Sr separation and analysis was developed. The seawater standard, IAPSO, yielded delta Sr-88/86 = 0.374 +/- 0.028 parts per thousand (2SD, n = 23), which is in agreement with reported data. The long-term reproducibility for the rock standard materials, BCR-2 and BHVO-2, yielded delta Sr-88/86 = 0.232 +/- 0.015 parts per thousand (2SD, n = 10), delta Sr-88/86 = 0.249 +/- 0.014940 (2SD, n = 7), respectively, and agrees well with published data. The in-house standard solution, Alfa Sr, was produced to check for the long-term reproducibility of the method and yielded a comparable value of 0.113 +/- 0.039 parts per thousand (2SD, n = 46) over 6 months.