Premium
Optimisation of 186 Os/ 188 Os Measurements by N‐TIMS Using Amplifiers Equipped with 10 13 Ω Resistors
Author(s) -
Reisberg Laurie,
Zimmermann Catherine
Publication year - 2021
Publication title -
geostandards and geoanalytical research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.037
H-Index - 73
eISSN - 1751-908X
pISSN - 1639-4488
DOI - 10.1111/ggr.12371
Subject(s) - isobaric process , calibration , amplifier , analytical chemistry (journal) , noise (video) , resistor , accuracy and precision , physics , isotope , nuclear physics , chemistry , optoelectronics , computer science , cmos , chromatography , quantum mechanics , voltage , artificial intelligence , image (mathematics) , thermodynamics
Analysis of 186 Os/ 188 Os variations in most natural samples is hampered by their low Os mass fractions (usually < 100 pg g −1 ), requiring Os separation from up to a kilogram of powder to obtain high precision measurement results. Recently developed amplifiers equipped with 10 13 Ω feedback resistors improve the signal to noise ratio by an order of magnitude, allowing acquisition of high precision data (2SE < 50 ppm) from 186 Os 16 O 3 − ion beams of ~ 40 fA and thus analysis of less material (~ 10–500 g). Factors controlling the uncertainty of 186 Os/ 188 Os N‐TIMS measurement results, using a mixed array of 10 13 and 10 11 Ω amplifiers, include isobaric interferences, baseline variability, gain calibration, decay time corrections and oxygen isotope composition variations. Isobaric interferences present the greatest challenges, but these can be largely overcome by using double Pt filaments and periodically monitoring interferences during each measurement. The optimised procedures produce 186 Os/ 188 Os ratios with precisions close to the theoretical limits imposed by counting statistics and Johnson–Nyquist electronic noise for ion beams of 10–70 fA. The intermediate measurement precision (85 ppm; 2 s ) approaches theoretically expected values (~ 50–80 ppm). This precision is sufficient to allow 186 Os/ 188 Os analyses of crustal rocks, for which relatively large variations are expected.