Assessment and Correction of Non-Linearity Effects for In Situ Rb-Sr Geochronology Applied to High 87Rb/86Sr Mica

Abstract

The large range of Rb and Sr mass fractions in natural micas presents a challenge for LA-ICP-MS/MS Rb-Sr geochronology. Mica with disparately high Rb and low common Sr mass fractions are particularly problematic given the limited linear dynamic range (1–2 × 10⁶ cps) of ICP-MS/MS systems. Here we quantify sources of uncertainty and inaccuracy associated with the in situ Rb-Sr method, including assessing laser-ICP-MS/MS timing issues, evaluating statistical bias at low count rates and quantifying non-linearity of the electron multiplier between pulse and analogue detection. These considerations are tested on natural mica with high ⁸⁷Rb/⁸⁶Sr ratios (> 1000), including ca. 92 Ma biotite (Tombstone pluton, Yukon) and ca. 2620 Ma muscovite (Tanco pegmatite, Manitoba), with Rb mass fractions of ca. 2300 and 30000 μg g^-1, respectively. Our results highlight the importance of: (1) optimising analytical parameters (e.g., laser energy, spot size), (2) potential application of non-linearity corrections for high Rb signals and (3) judicious selection of data reduction methods (e.g., signal integration methods, incorporating all sources of uncertainty, error correlation) for the determination of precise and accurate Rb-Sr ages. With these considerations in mind, in situ Rb-Sr measurements of Tombstone and Tanco mica yielded isochron ages within ∼ 1% of their accepted Rb-Sr TIMS ages.