Techniques for measuring carbon and oxygen isotope compositions of atmospheric CO <sub>2</sub> via isotope ratio mass spectrometry

Savio Manaj, Sang‐Tae Kim

Measuring the stable isotope compositions of atmospheric CO2 is common in earth and atmospheric sciences, and various analytical methods have been developed utilizing continuous-flow (CF) or dual-inlet (DI) isotope ratio mass spectrometry (IRMS). Air is typically collected via passive, manual, or automated collection methods and the volume of the air sample ranges from 10 to 300 mL for CF-IRMS to >1 L for DI-IRMS to yield a measurable amount of atmospheric CO2 gas. It has been determined that the integrity of vials and flasks for air sample storage can be compromised after 3 days of air collection for δ13C values and within 10 hours for δ18O values. Air samples must be purified after collection to remove constituents of air, such as Ar, O2, N2, N2O, and water vapor, to avoid isobaric interferences during mass spectrometric measurement. Purification is generally undertaken by utilizing commercial or custom-made preconcentration devices, the blanking method for CF-IRMS, or an offline/online cryogenic separation using a vacuum line for DI-IRMS. Ambient N2O is a component of air that may affect analytical results and thus must either be corrected for or be removed using a gas chromatographic column. In some cases, water is removed during air collection by using a common chemical desiccant, magnesium perchlorate (Mg(ClO4)2), or by a dry ice/alcohol mixture (−78°C). Lastly, a linearity issue for IRMS due to the low amount of purified CO2 from a typical ambient air sample must be considered. In general, analytical precisions of 0.02–0.21‰ and 0.04–0.34‰ for CF-IRMS and 0.01–0.02‰ and 0.01–0.02‰ for DI-IRMS are expected for δ13C and δ18O measurements, respectively.
Savio Manaj and Sang‐Tae Kim. 2021. Techniques for measuring carbon and oxygen isotope compositions of atmospheric CO 2 via isotope ratio mass spectrometry. Rapid Communications in Mass Spectrometry, Volume 35, Issue 4, 35(4).
Copy Citation: