Elena Dubinina, Alexander Borisov, Andrey Gurenko, S.A. Kossova, A.S. Avdeenko
{"title":"Evaluation of Compositional Matrix Effects on Oxygen Isotope Measurements in Silicate Minerals and Glasses Using SIMS","authors":"Elena Dubinina, Alexander Borisov, Andrey Gurenko, S.A. Kossova, A.S. Avdeenko","doi":"10.1111/ggr.12612","DOIUrl":null,"url":null,"abstract":"<p>Instrumental mass fractionation (IMF) accompanying oxygen isotope measurement by SIMS was studied for seventeen experimental alkali-rich (0–10 mole % of both K<sub>2</sub>O and Na<sub>2</sub>O) glasses along with a set of reference material glasses and minerals (quartz and olivine). Analyses were undertaken in two measurement sessions using a CAMECA IMS 1270-E7 ion microprobe operated under identical instrumental conditions. All employed experimental glasses and reference materials were re-analysed by laser fluorination gas source-mass spectrometry (LF GS-MS) to estimate their IMF values. The obtained IMF values, which ranged from -2.31 to +5.14‰, were combined with the IMF data for alkaline free glasses to estimate the compositional matrix effect for all major rock-forming oxides. Using this joint data set (including forty-four experimental glasses) a number of linear multiple variable regression models describing the link between IMF and major element chemical composition were established. To generalise the established models for the matrix effect (ME) correction we also evaluated an approach of routine measurement of chemically pure quartz with a known <sup>18</sup>O/<sup>16</sup>O ratio along with the other glass reference materials and use it as the “zero” point for multi-component regression fit. Our choice was justified by the fact that the ME in quartz has to be strictly constant. The last represents an advantage of using quartz instead of olivine or any other reference materials for normalisation. Finally, we tested our best multi-component regression model (standard error as low as 0.4‰) established for silicate glasses, being presumably crystallographic-amorphous substances, in application to different structural groups of minerals such as olivine, clinopyroxene, garnet, feldspars and quartz. The difference between the measured ME values and the model-predicted varied systematically, and the magnitude of the observed deviations increased proportionally with increasing magnitude of the measured ME, pointing towards the role of crystallographic structure in producing a compositionally related matrix effect.</p>","PeriodicalId":12631,"journal":{"name":"Geostandards and Geoanalytical Research","volume":"49 3","pages":"629-644"},"PeriodicalIF":3.4000,"publicationDate":"2025-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Geostandards and Geoanalytical Research","FirstCategoryId":"89","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/ggr.12612","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"GEOCHEMISTRY & GEOPHYSICS","Score":null,"Total":0}
引用次数: 0
Abstract
Instrumental mass fractionation (IMF) accompanying oxygen isotope measurement by SIMS was studied for seventeen experimental alkali-rich (0–10 mole % of both K2O and Na2O) glasses along with a set of reference material glasses and minerals (quartz and olivine). Analyses were undertaken in two measurement sessions using a CAMECA IMS 1270-E7 ion microprobe operated under identical instrumental conditions. All employed experimental glasses and reference materials were re-analysed by laser fluorination gas source-mass spectrometry (LF GS-MS) to estimate their IMF values. The obtained IMF values, which ranged from -2.31 to +5.14‰, were combined with the IMF data for alkaline free glasses to estimate the compositional matrix effect for all major rock-forming oxides. Using this joint data set (including forty-four experimental glasses) a number of linear multiple variable regression models describing the link between IMF and major element chemical composition were established. To generalise the established models for the matrix effect (ME) correction we also evaluated an approach of routine measurement of chemically pure quartz with a known 18O/16O ratio along with the other glass reference materials and use it as the “zero” point for multi-component regression fit. Our choice was justified by the fact that the ME in quartz has to be strictly constant. The last represents an advantage of using quartz instead of olivine or any other reference materials for normalisation. Finally, we tested our best multi-component regression model (standard error as low as 0.4‰) established for silicate glasses, being presumably crystallographic-amorphous substances, in application to different structural groups of minerals such as olivine, clinopyroxene, garnet, feldspars and quartz. The difference between the measured ME values and the model-predicted varied systematically, and the magnitude of the observed deviations increased proportionally with increasing magnitude of the measured ME, pointing towards the role of crystallographic structure in producing a compositionally related matrix effect.
期刊介绍:
Geostandards & Geoanalytical Research is an international journal dedicated to advancing the science of reference materials, analytical techniques and data quality relevant to the chemical analysis of geological and environmental samples. Papers are accepted for publication following peer review.