Stable isotope ratios are widely used to solve environmental, geological, medical, and forensic problems. The double spike technique is considered to be one of the most robust and efficient methods to correct for instrumental mass bias and isotopic fractionation that may occur during sample preparation. However, various hidden errors can arise from data processing and have been largely overlooked in previous studies. Several of these hidden errors were investigated in this work using measurement and synthetic data. Double spike inversion of chromium isotope raw data from 1116 natural samples demonstrated that averaging raw isotope ratios before double spike inversion can add significant errors to inverted isotope values, and such errors can be 1.5 times larger than the true analytical precision. Synthetic data were used to investigate the errors on inverted Cr isotope data caused by spike:analyte ratio and Fe–Ti–V interferences, and the following threshold values are recommended to minimize such errors: ⁵⁴Cr_spike/⁵²Cr_sample ratio greater than 0.5, ⁵⁶Fe/⁵²Cr less than 0.2, ⁴⁹Ti/⁵²Cr less than 0.04, and ⁵¹V/⁵²Cr less than 1. Sample preparation can potentially lead to large errors in inverted Cr isotope data if preparation-induced isotope fractionation deviates from the exponential law used in the double spike inversion, but such errors can be minimized by achieving >70% Cr yield. Our findings provide important insights for the double spike inversion procedure and assessing the reliability of inverted isotope data for not only the chromium isotope system but also other elements commonly analyzed using the double spike technique.

Xiangli Wang, Thomas M. Johnson