On the 17O correction for CO2 mass spectrometric isotopic analysis

To calculate delta(13)C from raw CO(2) isotope data, the ion beam ratio of m/z 45 to 44 is corrected for the contribution arising from the contribution of (17)O-bearing molecules. First, a review on the current state of (17)O-corrections for CO(2) mass spectrometry is presented. The three correction algorithms that are generally in use, however, do produce biased delta(13)C values, and the bias is actually larger than the precision of modern isotope ratio mass spectrometers. The origin of this bias is twofold: different values for (17)R(VPDB-CO2) as well as different values for lambda are used in the correction algorithms. Despite both values being of high importance, large discrepancies between the absolute values published for (17)R(VPDB-CO2) appear to be the main reason for the delta(13)C biases. Next, the question of how to choose the value of lambda to best be used is considered. Natural (e.g. tropospheric) CO(2) as well as primary reference materials (PDB and NBS-19), having been in isotope exchange with water, are assumed to lie on the fractionation line for waters. On this ground, lambda = 0.5281 +/- 0.0015, as determined for waters (Meijer and Li, Isot. Environ. Health Stud., 1998; 34: 349-369), is suggested to be a base for the (17)O-correction algorithm. Finally, an approach to determine the absolute value for (17)R(VPDB-CO2), based on data of relative isotope measurements on two CO(2) gases having a large (17)O difference, is discussed and algebraic formulas are considered. Experimental data and new numerical values determined for (17)R(VPDB-CO2) and (17)R(VSMOW) are given in a companion paper.