Revisiting apatite SIMS oxygen isotope analysis and Qinghu-AP reference material

Abstract

Oxygen is one of the most abundant elements in the universe and ubiquitously participates in geological and biological processes which play a critical role in shaping and maintaining habitability of the Earth. Its isotope composition is a key parameter to decode these geological and biological processes. For example, oxygen isotope composition of biogenic and abiogenic apatite are ideal proxies to reconstruct sea surface temperature and to tackle elemental cycling respectively, in which accurate and precise apatite oxygen isotope analysis is a basic requirement. Our robust evaluation demonstrates a lack of crystallographic orientation effects for Secondary Ion Mass Spectrometry (SIMS) apatite oxygen isotope analysis (expressed as δ18O). Apatite grains separated from the Qinghu monzonite were extensively analyzed by SIMS in the past decade in our laboratory, and 2041 individual δ18O measurements yield a two-standard-deviation of 0.47‰ (per mil, V-SMOW), or 0.16‰ when these measurements were integrated into 195 sessions. Homogenous oxygen isotope composition of Qinghu apatite (Qinghu-AP) at the micrometer level makes it a suitable reference material with a recommended δ18O value of 5.59 ± 0.19‰, as measured by Gas Source Isotope Ratio Mass Spectrometry with further validation from oxygen isotope thermometry. For seven randomly selected IGG (Institute of Geology and Geophysics)-Durango apatite crystals, our SIMS measurements suggest that they also have homogenous δ18O values (9.70 ± 0.22‰, 2 SD, n = 241, using Qinghu-AP as standard) without the presence of inter-grain and intra-grain heterogeneity. We documented a Y-coordinate related artificial effect, with measured δ18O values gradually increasing by ~1‰ from bottom to top of the mount over a distance of ~1.2 cm. This artifact is most likely caused by the presence of a slope (imperfect sample preparation and loading) or a gradient of accelerating voltage (conductivity of holder and sample) on the sample surface along the Y direction, which leads to an imperfect centering of the trajectory of the secondary ions. This effect can be reduced through applying a DTFA (dynamic transfer field aperture)-X correction. The presence of position-specific oxygen isotope fractionation could cause an artificial bias up to −6.7‰ when analyzing carbonated hydroxyapatite, and can reconcile the offset (e.g., −0.9 to 3.3‰) between δ18O values of conodonts measured by SIMS and IRMS. When working on hydrous systems (e.g., ore deposits) and low temperature environment (e.g., conodonts), where apatite is known to be water-rich and could be carbonated, a rigorous evaluation on this artificial bias is warranted. Since position-specific oxygen isotope fractionation is a common phenomenon observed in a variety of minerals (e.g., goethite, tourmaline and muscovite), attention should be paid in future studies. Our study highlights that SIMS is capable of providing accurate and precise δ18O measurements at the ~0.2‰ level, but high-quality reference materials and rigorous evaluations of potential artificial effects (e.g., topography, position, matrix, orientation, sample holder and position-specific oxygen isotope fractionation) degrading analytical accuracy are prerequisites.