High-precision and high-accuracy magnesium isotope analysis on multiple-collector inductively coupled plasma mass spectrometry using a critical mixture double spike technique

A new procedure has been developed for high precision and high accuracy Mg isotope analysis on multiple-collector inductively coupled plasma mass spectrometry using a critical mixture double spike technique. Errors arising from improper preparation of the critical double spike solutions are able to be corrected against the regression on a set of over- and under-spiked standards. Accuracy is ensured by correcting mass bias offset based on Monte Carlo calculations. Doping experiments indicate that the double spike method is robust to non-spectrum matrix effects. A long-term precision and accuracy of +/- 0.03 parts per thousand (2SD) was demonstrated for delta Mg-26 by replicate analyses of well-characterized in-house pure Mg solutions and synthetic samples passed through the column chemistry, provided that each sample was measured four times. The robustness of the method was further assessed by replicate analyses of fifteen geological reference materials ranging from peridotite, basalt, granodiorite, carbonatite to seawater. Rock standards yielded consistently higher delta Mg-26 by 0.076 +/- 0.052 parts per thousand (2SD, N = 12) compared to data previously reported by standard-sample-bracketing from the same lab. This discrepancy might result from the difficulty in matching the matrices of natural samples exactly the same to the bracketing standards, even after purification, suggesting a careful evaluation on residual matrix effect for the standard-sampling-bracketing method. Our new data for geological reference materials serve as a reference for quality assessment and inter-laboratory comparison in future studies. Copyright (C) 2022, Guangzhou Institute of Geochemistry. Production and hosting by Elsevier B.V.

Automated summary

A new procedure for high precision and high accuracy Mg isotope analysis has been developed using a critical mixture double spike technique and multiple-collector inductively coupled plasma mass spectrometry. The accuracy is ensured by correcting errors arising from improper preparation of the double spike solutions and correcting mass bias offset based on Monte Carlo calculations. Doping experiments show that the method is robust to non-spectrum matrix effects.