Premium
High precision δ 17 O isotope measurements of oxygen from silicates and other oxides: method and applications
Author(s) -
Miller M. F.,
Franchi I. A.,
Sexton A. S.,
Pillinger C. T.
Publication year - 1999
Publication title -
rapid communications in mass spectrometry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.528
H-Index - 136
eISSN - 1097-0231
pISSN - 0951-4198
DOI - 10.1002/(sici)1097-0231(19990715)13:13<1211::aid-rcm576>3.0.co;2-m
Subject(s) - chemistry , fractionation , isotopes of oxygen , isotope , analytical chemistry (journal) , oxygen , mass spectrometry , oxide , mineralogy , isotope analysis , stable isotope ratio , isotope fractionation , oxygen isotope ratio cycle , environmental chemistry , geology , nuclear chemistry , chromatography , physics , oceanography , organic chemistry , quantum mechanics
The use of infrared laser‐assisted fluorination to release oxygen from milligram quantities of silicates or other oxide mineral grains is a well‐established technique. However, relatively few studies have reported the optimisation of this procedure for oxygen‐17 isotope measurements. We describe here details of an analytical system using infrared (10 µm) laser‐assisted fluorination, in conjunction with a dual inlet mass spectrometer of high resolving power (∼250) to provide 17 O and 18 O oxygen isotope measurements from 0.5–2 mg of silicates or other oxide mineral grains. Respective precisions (1) of typically 0.08 and 0.04‰ are obtained for the complete analytical procedure. Departures from the mass‐dependent oxygen isotope fractionation line are quantified by Δ 17 O; our precision (1) of such measurements on individual samples is shown to be ±0.024‰. In turn, this permits the offset between parallel, mass‐dependent fractionation lines to be characterised to substantially greater precision than has been possible hitherto. Application of this system to investigate the 17 O versus 18 O relationship for numerous terrestrial whole‐rock and mineral samples, of diverse geological origins and age, indicates that the complete data set may be described by a single, mass‐dependent fractionation line of slope 0.5244± 0.00038 (standard error). Copyright © 1999 John Wiley & Sons, Ltd.