Vanadium isotope fractionation during early planetary evolution: Insights from achondrite analyses

Abstract

Heavy vanadium (V) isotope compositions of bulk silicate Earth (BSE) and Mars (BSM) relative to chondrites have been suggested to result from high pressure-high temperature core segregation processes on terrestrial planets. However, an alternative possibility is that these heavy V isotope signatures could reflect inheritance from their differentiated planetary building blocks if, for instance, early formed planetesimals underwent V isotope fractionation during differentiation and/or magma ocean evaporation. To test this hypothesis, we report the first V isotope compositions of 40 achondrites (eucrites and diogenites, angrites, ureilites, and acapulcoites-lodranites) originating from four distinct parent bodies. We find that the bulk silicate portions of (4) Vesta and the Angrite Parent Body (APB) exhibit heavy V isotope signatures relative to chondrites, comparable to (or greater than) BSM, but lighter than BSE. On the contrary, the Ureilite Parent Body (UPB) and the Acapulcoite/Lodranite Parent Body (ALPB) are indistinguishable from the chondritic value. To investigate the origin of these V isotope variations, we combine V isotope data with the systematics of other elements. First, we show that differentiated planetary bodies do not exhibit clear V–Sr isotope covariations similar to those recently observed for calcium-aluminum-rich inclusions (CAIs). Only the heavy V isotope signature of (4) Vesta could have potentially resulted from accretionary and/or magma ocean volatilization processes (reflecting ∼0.6 % V and Sr loss). To account for the heavy V but chondritic Sr isotope compositions of the APB, BSM, and Moon, we suggest either (i) extremely large isotope fractionation of V during core formation, or (ii) significantly higher metal-silicate partition coefficients (i.e., more siderophile V) than expected based on experimental data and planetary body conditions. Alternatively, conditions of nebular evaporation recorded in CAIs may not apply to planetary evolution, or the volatilities of V and Sr may differ significantly under early planetary conditions. Similarly, we observe no strong correlation between V and evaporation-sensitive elements like K, likely due to K's much higher volatility compared to V. Potential correlations between V and elements such as Mg (R² = 0.98), Si (R² = 0.81), and Fe (R² = 0.46) suggest that these elements – in particular Mg – may exhibit volatilities closer to V than to K or Sr during planetary evaporation. Explaining bulk δ⁵¹V variations through vapor–melt fractionation – rather than V partitioning into the core – would alleviate potential conflicts between our data and previous modeling based on available experimental results regarding V metal-silicate partitioning. In any case, the lack of a V isotope anomaly for the UPB and ALPB indicates that the processes responsible for the heavy V isotope compositions of (4) Vesta, the APB, and the terrestrial planets did not occur on these two bodies. This result is consistent with a lack of global magma ocean formation on these parent bodies, as independently suggested by the preservation of mass independent oxygen isotope heterogeneities throughout the mantles of both parent bodies.