Quantifying Long‐Term Reproducibility of Zircon Reference Materials by U‐Pb LA‐ICP‐MS Dating. Issue 3 (14th July 2022)

• Record Type: Journal Article
• Authors: Sliwinski, Jakub T.; Guillong, Marcel; Horstwood, Matthew S. A.; Bachmann, Olivier
• Is Part Of: Geostandards and geoanalytical research. Volume 46:Issue 3(2022)
• Page Start: 401
• Page End: 409
• Publication Date: 2022-07-14
• Subjects: LA‐ICP‐MS -- zircon -- reproducibility -- excess variance -- reference materials; Analytical geochemistry -- Periodicals; Géochimie analytique -- Périodiques; 551.9
• DOI: 10.1111/ggr.12442
• Languages: English
• ISSNs: 1639-4488

Abstract

Abstract : Inter‐session excess variance of U/Pb and Pb/Pb ratios in LA‐ICP‐MS zircon dating is the largest contributor to systematic errors, which in turn limit the accuracy of age determinations. Quantifying long‐term excess variance of reference materials allows for the estimation of excess variance in samples, but such compilations are not available in the literature. Here, we present the results of over 100 measurement sessions over 6 years for seven common zircon reference materials and calculate a characteristic excess variance for each. For 206 Pb/ 238 U ages, these values (2 s ) are as follows: AusZ7‐1 = 1.7% (115 sessions), AusZ7‐5 = 0% (74), OD‐3 = 0.6% (19), Temora2 = 1.2% (86), Plešovice = 1.1% (100), 91500 = 1.0% (146) and Mud Tank = 3.0% (12). For 207 Pb/ 206 Pb ages, smaller excess variances are observed: Temora2 = 0.3%, Plešovice = 0.4%, Mud Tank = 1.7% and 91500 = 0.4%. These values are well‐constrained estimates of inter‐session excess variance for the ETH Zürich LA‐ICP‐MS laboratory using a sector‐field ICP‐MS, and may provide either a first‐order estimate for other laboratories or a value to compare against their own reference material compilations. Abstract : Key Points Long‐term reproducibility/intermediate measurement precision (> 100 sessions) of zircon reference materials is presented. Inter‐session excess variance (ε') limits the precision of LA‐ICP‐MS U‐Pb dating. We present a framework for other laboratories to perform long‐term meta‐analysis of data reproducibility.