ISOTOPIC FRACTIONATION AND THE QUANTIFICATION OF 17O ANOMALIES IN THE OXYGEN THREE-ISOTOPE SYSTEM - AN APPRAISAL AND GEOCHEMICAL SIGNIFICANCE

Abstract

The oxygen three-isotope plot provides valuable insights to geochemical and cosmochemical processes in which the relationship of 17O/16O relative to 18O/16O is of interest. It is generally recognised, however, that the usual format of this diagram, illustrating the variation of δ17O with δ18O, is based on the approximation of a power law function to linear format. In contrast, the relationship between δ17O and δ18O for entity a measured with respect to reference b is accurately represented by the linear function:1000ln1+δ17O1000=λ1000ln1+δ18O1000+1000ln[1+ka,b] where ka,b is a measure of the offset (if any) of a from the mass-dependent fractionation line, of slope λ, on which b lies. The respective ordinate and abscissa scales are essentially unchanged from those of the corresponding δ17O versus δ18O plot yet, unlike the latter, the slope of the fractionation line is invariant to both the magnitude of the δ values and to the isotopic composition of the reference.Application to high precision measurements which encompass a wide range of δ values reveals that slope differences (or indeed similarities) between comparable data sets may be disguised by the limitations of the δ17O versus δ18O diagram, with adverse consequences for the accurate quantification of Δ17O values. This may be addressed by re-defining Δ17O, the offset from a reference fractionation line, as 1000ka,b. In turn, this is very closely approximated by 1000ln[1+ka,b], the intercept on the ordinate axis of a 1000ln[1+(δ17O/1000)] versus 1000ln[1+(δ18O/1000)] plot, for values of 1000ka,b not exceeding ~5. The same analysis shows that the fractionation factor α17/16 for a given system is related to the corresponding α18/16 value by:α17/16=[1+ka,b](α18/16)λThese findings are equally applicable to the reporting of data from other three-isotope systems, such as magnesium.