Instrumental mass-independent fractionation accounting for the sensitivity of the double spike proportion effect by MC-ICP-MS: a case study of Zn isotope measurements

Abstract

The double spike (DS) technique has an advantage over other correction models for correcting isotopic fractionation during chemical separation. However, previous studies have revealed a correlation between measurements and DS proportions in some elements and even more sensitivity to DS proportions, showing inconsistency between experimental results and theoretical expectations, whereas the exact reason is unclear. This study takes Zn with a ⁶⁶Zn–⁷⁰Zn DS combination as a representative to explore the cause of the over-sensitive DS proportion effect. The effect of different DS proportions (0.1–0.9) on isotopic analyses was determined. An inverse correlation was observed between δ⁶⁶Zn and DS proportions in a limited range (0.46–0.48), which differs from the theoretical prediction. High-resolution observation and concentration gradient demonstrate that isobaric interference is negligible and not responsible for the sensitive DS proportion effect. The good linear regression of the three isotope-plot for IRMM-3702 measured values shows that the exponential law is valid in this study. The fractionation factor β calculated using the exponential law shows that the β values of the same isotope pairs at different analytical sessions vary considerably, implying that the β values change with every setup with no fixed pattern. And β values of ^66/64Zn, ^67/64Zn, ^68/64Zn, and ^70/64Zn are not equal in the same sessions, indicating the existence of MIF which is responsible for the sensitive DS proportion effect. Since β values have no fixed pattern and vary within measurement error in a period, they can be used to correct the MIF. After correction, the correlation between δ⁶⁶Zn and DS proportions was corrected, and the DS proportions are comparable to the theoretical predictions, ranging from 0.35 to 0.7. Simulations with β values revealed that β values can vary within a certain range without introducing additional errors in one setup. The corrected long-term reproducibility of CAGS-Zn relative to IRMM-3702 is −0.885 ± 0.022‰, and the corrected Zn isotopic compositions of reference materials agreed with previous studies within 2sd. This paper identifies the cause of the sensitive DS proportion effect and proposes a feasible correction scheme, further emphasizing that instrumental MIF is not a surprise and should be valued in the DS technique.