Abstract
For several decades, an invariable 238U/235U ratio of 137.88 was assumed throughout the solar system, including Earth, due to uranium‘s heavy mass. However, recent studies have shown surprisingly variable 238U/235U values in terrestrial and extraterrestrial environments, including both low- and high-temperature settings. This has profound implications for both the accuracy of the U-Pb chronometer and understanding isotopic fractionation of the heavy elements. Isotope fractionation between 238U and 235U is controlled by both mass-dependent, and nuclear volume-dependent fractionation mechanisms during geochemical reactions. Mass-dependent effects preferentially remove the lighter 235U isotope from the residue, while volume-dependent fractionation preferentially removes the heavier 238U isotope, leaving the residue isotopically lighter. In this study, 238U/235U was determined with high analytical precision on samples from a range of terrestrial and extraterrestrial environments, utilizing a 233U-236U double spike procedure and multiple collector ICP-MS. A terrestrial 238U/235U reference value for ‘Bulk Silicate Earth‘ of 137.795 ± 0.008 was determined, revoking the previously assumed value of 137.88. Additional uranium isotopic measurements of a wide range of low-temperature environments provide key insights into the mechanisms driving uranium isotopic shifts. To this end, uranium isotopic fractionation controlled by volume-dependent effects of 0.36‰ were observed at the redox interface of an anoxic basin. Most likely, this mechanism is induced by microbially-mediated uranium reduction, which preferentially removes isotopically heavy U(IV) from the water column. This process is in agreement with Rayleigh fractionation in a closed system, corresponding to a fractionation factor of α of 0.9985 between the aqueous (oxidized) and solid (reduced) phase. In contrast, a 238U/235U fractionation of 0.35‰ heavier isotopic compositions was observed in the aqueous phase of an aquifer, following mass-dependent fractionation caused by adsorption processes.
Equally significant results were obtained from extraterrestrial meteorites. Samples of different meteorite groups, namely angrites, eucrites and chondrites, show homogeneous uranium isotope compositions between groups, but variations between single samples within the groups. The investigations of low-temperature samples indicate that these variations are likely caused by redox related fractionation during accretion processes, aqueous alteration, or thermal metamorphism, although no direct evidence of these processes was found.
Fractionation mechanisms aside, the 238U/235U ratios of the angrites analyzed result in revised absolute Pb-Pb ages for this important group of achondrites that are ~1 Myr younger than previously assumed. This reconciles previously reported deviations between absolute Pb-Pb ages and those obtained utilizing different extinct short-lived chronometers.
Revised Pb-Pb ages of the carbonaceous chondrite Allende, including chondrules and a calcium-aluminium rich inclusion extracted from it, combined with other recent data [Connelly et al., 2012] give rise to a CAI formation age of 4567.30 ± 0.30 Myr, representing a revised age for the beginning of the solar system and an extended interval of chondrule formation of ~3 Myr. These new uranium isotopic data also have implications for the 247Cm-235U short-lived chronometer, yielding an initial 247Cm/235U at the beginning of solar system formation of 2.9 x 10-4 - 1.5 x 10-3, using Nd and Th as proxies for the extinct Cm. The resulting time interval of free decay between the last r-process event and solar system formation corresponds to 140-195 Myr, in very good agreement with previously published data.