Trace element analyses of carbonates using portable and micro-X-ray fluorescence: Performance and optimization of measurement parameters and strategies.

Variations in elemental abundances in carbonate archives offer a wealth of information that can be used as proxy for palaeoenvironment and diagenetic history. State of the art portable handheld X-ray Fluorescence (pXRF) and laboratory micro X-ray Fluorescence (µXRF) instruments provide a relatively inexpensive, fast and non-destructive way of acquiring these trace element composition data. However, there are well-known issues and limitations regarding the method of spectrum acquisition and the conversion of XRF spectra into quantitative elemental mass fractions. This study offers a guideline for the appropriate use of these XRF techniques for the study of carbonates. Using certified calcium carbonate and dolomite standards, accuracy and reproducibility of a pXRF (Bruker AXS Tracer IV) and a µXRF (Bruker M4 Tornado) device are tested under various measurement conditions. The experimental set-up allowed for the variation of several parameters, including measurement area, integration time, quantification method and measurement strategy. The effects on the accuracy and reproducibility of the quantified elemental abundance results are examined to assess for both devices the optimal performance conditions for the determination of trace element abundances in natural carbonates. The limits of detection and quantification are evaluated for both instruments for a range of trace elements commonly used as palaeoenvironmental proxies (e.g. Sr, Mn and Fe). The quality of the XRF spectra is evaluated using spectral processing software. As a result, two new methods for the determination of optimized parameter combinations are proposed for a range of commonly used elements. The Time of Stable Reproducibility (TSR) is based on optimizing measurement reproducibility by examining the change of the relative standard deviation per time unit and proposing an integration time threshold for reproducible measurement. The Time of Stable Accuracy (TSA) is based on optimizing measurement accuracy by studying changes in accuracy in function of increasing integration time and defining an integration time threshold for accurate measurement. An overview table including minimum integration times by which a reliable measurement is reached is provided for all analyzed elements and experimental set-ups for this study. However, the methodological approach that is developed here is applicable on other (carbonate) materials as well. A comparison between the two X-ray Fluorescence instruments allows the evaluation of their respective advantages and disadvantages. Finally, we recommend optimal measurement strategies and techniques for specific research questions.