Effect of Fractionation Rate on High-Precision Multidynamic TIMS Sr Isotope Analysis and the ⁸⁴Sr Abundance of the Earth

Abstract

The multidynamic acquisition mode in TIMS isotopic analysis can effectively minimize the bias in isotopic ratios due to Faraday cup deterioration [1, 2]. However, a potential problem of this mode is that the normalising and normalised ratios are not measured simultaneously. If the fractionation is fast, this can lead to a bias in the fractionation-corrected isotopic ratios, as demonstrated in the recent studies on highprecision multidynamic Nd isotope analysis [2, 3]. Here we show that this effect can be pronounced in the high-precision TIMS analysis of 84 Sr. We measured 84 Sr/ 86 Sr ratios of the standard SRM987 and a series of terrestrial samples (BCR-2, BHVO-2, BIR-1, and seawater) using a 3line multidynamic method with the center cup sequencially collecting masses 85, 86, and 87. The µ 84 Sr values (defined as 10 6 [(84 Sr/ 86 Srsampl e)/(84 Sr/ 86 SrSRM987)-1]) of the terrestrial samples are positively correlated with fractionation rates, consistent with the expectation from our cup configuration (measuring 84 Sr/ 86 Sr prior to 88 Sr/ 86 Sr). After correcting the fractionation drifting between magnetic settings using a linear interpolation method, this correlation is eliminated, and a better measurement precision is achieved. The drift-corrected multidynamic measurements yield a weighted average µ 84 Sr of-31 ± 8 ppm (2 S.E., N = 33) for the terrestrial samples, which is in good agreement with our static measurement results (-31 ± 7 ppm) and those reported by [4-7]. The uncorrected multidynamic measurents yield a less negative µ 84 Sr value (-17 ± 9 ppm), similar to the results of [7-10]. We suggest that the higher fractionation rates of terrestrial samples relative to SRM987 may have biased the multidynamic µ 84 Sr in some of the previous reports, and that the Earth has a 84 Sr deficit of ~30 ppm relative to SRM987. [1] Thirlwall (1991) CG, 94, 85-104. [2] Garçon et al.