Systematic analysis of precision, long-term stability and data processing of MC-ICP-MS 230Th/U-dating of secondary carbonates

MC-ICP-MS has become the clearly dominant technology in 230Th/U-dating over the last 20 years. The ongoing increase in measurement performance allows for ε-precision by now (Andersen et al., 2004; Cheng et al., 2013) and enabled the production of a large amount of datasets and insights covering a broad range of fields. Despite this large database, systematic studies on the impact of individual corrections on the raw data, long-term instrumental stability and on data processing routines are still sparse. We present the measurement and data analysis protocols developed and applied at the Institute for Environmental Physics at Heidelberg University and examine the contribution of individual corrections, such as tailing and hydride correction, to the total uncertainties of the atomic ratios 230Th/238U and 234U/238U and of the ages for this procedure. As demonstration examples, three different speleothem samples of different U and Th concentrations and ages were chosen. This puts specific quantitative constraints on general findings as the dominance of tailing correction. Long-term instrumental stability is examined by compiling a multi-year dataset of (230Th/238U) and (234U/238U) for the Harwell-Uraninite 1 (HU-1) reference material of more than thousand measurements which is then used for the recalibration of the inhouse 229Th, 233U and 236U spike. Lastly, we present a Python-based GUI for the combined evaluation of mass-spectrometric data and age determination that offers high flexibility with regard to the variation of input constants, such as the initial (230Th/232Th) used for detritus correction.