Can the Electrospray Orbitrap be a new frontier for the stable isotope analysis of oxyanions in ice cores? A statistical study of its performances

Recently, the Orbitrap Exploris instrument, a mass spectrometer designed for molecular identification and classification, has been diverted from its original purpose and transformed into a new tool for quantifying the stable isotope ratios of water-soluble compounds such as nitrates, sulfates or phosphates. At first glance, this new Orbitrap IRMS system is very attractive because it requires 10 to 100 times less ice than current standard procedures, works directly with liquid solutions, quantifies isotope ratios directly on the molecule of interest, and provides a broader range of isotope ratios by moving from the elemental isotope ratio paradigm to that of molecular isotope ratios and thus challenge the regular isotope ratio mass spectrometers (IRMS or ICPMS). Moreover, by leaving the chemical bonds of the molecules of interest intact contrary to IRMS or ICPMS, the Orbitrap technology provides access to clumped isotopes, i.e. doubly substituted isotopic substances, thereby enriching our knowledge of matter. Extending the range of molecular isotope ratios opens up exciting new prospects, particularly for unravelling the mechanisms by which compounds are formed before being incorporated in ice. Such a radical change in the way isotope ratios are measured inevitably raises the question of the real capabilities of this new instrument. With a deeper look into statistics, can the Orbitrap-IRMS challenge the standard IRMS/ICPMS precision for analyzing soluble species? Does its performance live up to expectations? Preliminary results show that the Allan variance for a large variety of molecular isotope ratios of nitrate and sulfate are in the range of one per mil precision, including some of the clumped isotope ratios. However, the bootstrapping approach aimed at reducing acquisition time and thus the drift associated with the instrument appears to be ineffective in improving the Allan variance, indicating a possible limitation of sampling randomization, probably during the electrospray ionization (ESI) process. Other statistics, tests and performances are still in progress and will also be presented.