Iron isotopic fractionation between silicate mantle and metallic core at high pressure

Abstract

The +0.1‰ elevated ⁵⁶Fe/⁵⁴Fe ratio of terrestrial basalts relative to chondrites was proposed to be a fingerprint of core-mantle segregation. However, the extent of iron isotopic fractionation between molten metal and silicate under high pressure-temperature conditions is poorly known. Here we show that iron forms chemical bonds of similar strengths in basaltic glasses and iron-rich alloys, even at high pressure. From the measured mean force constants of iron bonds, we calculate an equilibrium iron isotope fractionation between silicate and iron under core formation conditions in Earth of ∼0-0.02‰, which is small relative to the +0.1‰ shift of terrestrial basalts. This result is unaffected by small amounts of nickel and candidate core-forming light elements, as the isotopic shifts associated with such alloying are small. This study suggests that the variability in iron isotopic composition in planetary objects cannot be due to core formation.

Figure 1. Force constants of iron bonds in basaltic glass and Fe and iron-rich alloys as a function of pressure.

Black and blue solid squares, black solid diamonds, and black and magenta solid circles: basaltic glass (a), Fe (b), Fe₈₅Si₁₅ (c), Fe_86.8Ni_8.6Si_4.6 (c) and Fe₃S (d), respectively, from this study (Supplementary Table 2); half-filled diamonds, circles and squares: values for Fe₉₂Ni₈, Fe₈₅Si₁₅ and Fe₃S evaluated based on the NRIXS data from Lin et al. using the SciPhon software; open square and diamonds: bridgmanite and Fe reported by Shahar et al.; dashed line: bridgmanite reported by Rustad and Yin. The error bars are 95% confidence intervals. Each high-pressure data point was measured at least 20 times and as many as 47 times. Solid lines: linear fits to the data for the basaltic glass, Fe and iron-rich alloys. Note the large shift in value of the basaltic glass (a) at ∼30 GPa, which corresponds to structural changes in the glass.

Figure 2. Comparisons of β-factors of iron-rich alloys at high pressure and temperature.

Difference in the ⁵⁶Fe/⁵⁴Fe β-factors of iron-rich alloys with respect to pure iron at 3,000 K (a) and 4,000 K (b) as a function of pressure. Magenta, blue, grey and olive solid lines: Fe_86.8Ni_8.6Si_4.6, Fe₈₅Si₁₅, Fe₉₂Ni₈ and Fe₃S alloys, respectively, from this study. Purple, grey and olive dashed lines: FeO, Fe₃C and FeH_x, respectively, from Shahar et al. Δ⁵⁶Fe_Alloy-Fe=δ⁵⁶Fe_Alloy−δ⁵⁶Fe_Fe=1,000 × (ln β_Alloy^56/54Fe−ln β_Fe^56/54Fe). Black dash-dotted lines represent no iron isotopic fractionation between Fe-rich alloys and pure iron.

Figure 3. Equilibrium ⁵⁶Fe/⁵⁴Fe isotope fractionation between basaltic glass and iron-rich alloys at high pressure and temperature.

a. Pressure at 40 GPa. b. Pressure at 60 GPa. Black, magenta, blue, grey and olive solid lines: Fe, Fe_86.8Ni_8.6Si_4.6, Fe₈₅Si₁₅, Fe₉₂Ni₈, and Fe₃S alloys, respectively, from this study. Purple, grey and olive dashed lines: FeO, Fe₃C and FeH_x, respectively, from Shahar et al. Δ⁵⁶Fe_Basalt-Alloy=δ⁵⁶Fe_Basalt−δ⁵⁶Fe_Alloy. Blue dash-dotted lines represent no iron isotopic fractionation between basaltic glass and Fe-rich alloys.

Figure 4. Predicted shifts in the iron isotopic composition of the silicate mantle due to core formation.

Schematics for δ⁵⁶Fe isotope signatures in the bulk silicate Earth (primitive mantle) with regard to varying compositional models of Earth's core. The segregating metal was assumed to equilibrate with the silicate mantle at the base of the magma ocean at ∼50 GPa and 3,500 K based on refs , , .