Pressure baseline correction and high‐precision CO 2 clumped‐isotope (∆ 47 ) measurements in bellows and micro‐volume modes

RATIONALE: CO(2) 'clumped-isotope' measurements (tracking enrichment of (16)O(13)C(18)O, reported as ∆(47) values, on CO(2) derived from carbonate minerals or the atmosphere) are becoming central to a wide range of geochemical investigations. We present a novel approach to address problems with instrument stability, external precision, and the analysis of small samples that have hampered the advancement of Δ(47) measurements. METHODS: We measured Δ(47) values on CO(2) gases introduced via dual inlet to an isotope ratio mass spectrometer. We developed a method for determining the 'pressure baseline' and integrating a correction to ion beam intensity measurements during analysis. We then tested this approach for both bellows and micro-volume modes of sample introduction. Heated gas and equilibrated gas lines (Δ(47) vs. δ(47)) established the effectiveness of this correction. RESULTS: We have determined that drift in instrument calibration that compromises Δ(47) measurements results from a shift in the baseline signal on sensitive collectors (m/z 47, 48, and 49) that occurs when gas is admitted to the ion source. Applying a 'pressure baseline' (PBL) correction significantly stabilizes ∆(47) measurements and reduces the dependence of ∆(47) values on δ(47) values by up to an order of magnitude. CONCLUSIONS: PBL-corrected heated gas and equilibrated gas calibrations in bellows and micro-volume modes are nearly identical and stable through time. Introduction of the PBL correction, a revision to the absolute reference frame approach to determining Δ(47) values, dramatically improves the external precision of Δ(47) measurements to near instrumental analytical uncertainty (6-8 ppm (1σ) in bellows mode; 10-12 ppm in micro-volume mode).