First-principles investigation of equilibrium iron isotope fractionation in Fe1−xSx alloys at Earth's core formation conditions
Introduction
Iron is the major element of solar system planetary cores, in relation with the nucleosynthesis sequence. Iron plays a major role in every stance of planetary formation and differentiation. Geophysical approaches have confirmed the presence of iron for Earth's core but with almost 10% of light elements required to make up its mass (Hirose et al., 2013). Traditionally the effects due to the presence of light elements in the core are being inferred by geophysical studies where the density and sound velocity propagation are studied as function of light element concentration to match those values reported from geophysical models such as PREM. However, recently it has been suggested that the Fe isotopic composition in the mantle could be used to infer the composition of Earth's core (Bourdon et al., 2018; Craddock and Dauphas, 2010; Craddock et al., 2013; Lesher et al., 2020; Shahar et al., 2016; Sossi et al., 2016). This is based on the assumption that the presence of elements such as C, O, Si, S should affect the partitioning of iron isotopes between the mantle and the core. Originally, it has been suggested that core-mantle differentiation will leave imprints on the iron isotope signature of Earth's mantle because of the difference of Fe valence state and coordination in the mantle (Fe2+) and in the core (Fe0 metal) (Polyakov, 2009). That study suggested that the mantle should be enriched in heavy isotopes, product of equilibrium isotope fractionation during differentiation. However, subsequent explanations argue that the bulk silicate Earth is chondritic in its iron isotopic composition and that any difference in isotope composition seen in basalts is due to fractionation during partial melting of the rock from which they have been formed (Craddock et al., 2013). Indeed, isotope compositions in natural rocks are scattered in values, with Fe ∼−0.01‰ for carbonaceous chondrites, Fe ∼−0.1 to +0.1‰ for enstatite chondrites and Fe for basalts (Craddock and Dauphas, 2011; Liu et al., 2017; Poitrasson et al., 2013). These scattered values are therefore interpreted to suggest that the accessible mantle has a Fe isotope composition that is indistinguishable from chondritic composition (i.e. Fe ∼0‰). Apart from expected evidence of the presence of light elements in the core due to difference in chemical bonds, one must also consider the effects of pressure and temperature. It is well known that equilibrium fractionation effects should vanish at high temperatures. However, in the case of lower mantle and core conditions where pressures are also high, these fractionation effects might still be important. Recently, Shahar et al. (2016) have carried out experiments and computational modelling on the effects of pressure and light elements (such as H, C, O) on the equilibrium isotope fractionation of 57Fe/54Fe at the core-mantle boundary. Their work has found a significant imprint on isotope fractionation due to light elements such as C and H (Fe∼0.03-0.05‰) whilst elements such as O leave almost no imprint (Fe∼0.007-0.01‰). Their results support the suggestions by Craddock et al. (2013) that in order to have a mantle with a Fe near 0‰, light elements such as C and H that provide a large imprint in fractionation should not be present in the core at a significant level. Additionally, it favours the presence of O as a light element in the core. However, the effects of other light elements such as S on equilibrium Fe isotope fractionation at core conditions are yet unclear as well as any pressure induced magnetic effects that can be of relevance. There has been recent experimental work by Ni et al. (2020), Shahar et al. (2015) on the effects of S on Fe isotope fractionation between solid metal and liquid metal, or between metal and silicate. In both cases, sulfur enters into the metallic phases and preferentially into the liquid metal phase. Shahar et al. (2015) found that metal-silicate fractionation increases significantly with the sulfur content in the metal, whereas Ni et al. (2020) observed that the solid-liquid fractionation in this system does not depend on the sulfur content of the liquid metal. Obviously, much work needs to be done to understand these effects at temperature and pressure conditions of the terrestrial core, far away from experimental studies performed below 2 GPa. In this context, the difficulty to carry out experiments at high pressure and high temperature leads to the use of crystalline phases such as hcp Iron or I4 Fe3S as proxies of the molten systems leading to large extrapolations and possible errors in the interpretation of final results. This approach is used for instance in Liu et al. (2017) where high-pressure data (up to 206 GPa) suggest a minuscule Fe isotope fractionation between metal and silicate that is one order of magnitude lower than the one found by Shahar et al. (2015) at 1-2 GPa.
In recent years, computational techniques have become useful tools to estimate isotope fractionation factors in mineral systems. This is particularly true for methods based on quantum mechanics that allow to describe the vibrational properties of light and heavy isotopes at any pressure and temperature conditions (Blanchard et al., 2017). However, the weakness of these methods typically lies in the use of the quasi-harmonic approximation that falls when temperature increases or when the systems become fully molten. In addition, in dynamical systems such as liquids and melts, it has been shown that configurational disorder needs to be taken into account in order to obtain meaningful equilibrium fractionation factors (Blanchard et al., 2017; Pinilla et al., 2015).
In this work, we use ab-initio computational methods to perform a systematic study of the 56Fe/54Fe equilibrium isotope fractionation in molten and solid Fe1−xSx alloys at conditions of the core formation. By comparing the fractionation factors obtained from solid and molten systems, we estimate the validity of the experimental approximation consisting in using solid metals as a proxy for molten alloys. Additionally, we comment on the effects of S concentration on Fe isotope fractionation in liquid systems. We discuss our findings on view of latest results on equilibrium Fe isotope fractionation and their relevance for the formation of the Earth's core.
Section snippets
Equilibrium isotope fractionation factor from harmonic vibrational modes
The equilibrium isotope fractionation factor of an element displaying two isotopic forms Y and Y* between two phases a and b is related to the ratio of the isotope concentration ratios: where is the mole fraction of isotopes Y in phase a. The equilibrium fractionation factor between two phases can be related to the reduced partition function ratio of each phase by: . Isotopic reduced partition function ratios are usually
Iron and sulfur kinetic energies
To determine the kinetic energy for Fe and S atoms we calculated the velocity auto-correlation function (VCF) for all the studied systems as shown in Fig. 1. In the case of liquid alloys, the typical behaviour for dense liquids was observed, with an intermediate negative region produced by atomic collision and a subsequent rise before oscillating around a zero correlation (Hansen and McDonald, 2005). The fact that the positive part of the VCF is larger than the negative one describes a system
Conclusions
In this work, we have used computational methods to understand the effect of S on Fe equilibrium isotope fractionation in Fe1−xSx alloys as well as the consequences of using solid metal as proxies of molten alloys when performing experimental studies. We have shown that although configurational disorder could be different between these systems, bond type and local environment are the controlling parameters of Fe isotope fractionation and that β-factors obtained from solid systems may in
CRediT authorship contribution statement
Carlos Pinilla: Conceptualization, Investigation, Methodology, Software, Writing – original draft. Aldemar de Moya: Investigation, Visualization. Segolene Rabin: Investigation, Visualization. Guillaume Morard: Conceptualization, Investigation, Writing – review & editing. Mathieu Roskosz: Conceptualization, Investigation, Writing – review & editing. Marc Blanchard: Conceptualization, Investigation, Methodology, Writing – original draft.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work was performed using Granado-HPC from the Universidad del Norte, SCARF from the STFC of the UK, and the HPC resources from CALMIP (Grant 2020 – P1037). The authors acknowledge funding from MINCIENCIAS (No. 2015-710-51568; Contract No. 023-2016) and ECOSNORD (C17U01, No. FP44842-143-2017) through research grants.
References (48)
- et al.
Composition and temperature of the Earth's core constrained by combining ab-initio calculations and seismic data
Earth Planet. Sci. Lett.
(2002) - et al.
The Nosé-Poincaré method for constant temperature molecular dynamics
J. Comput. Phys.
(1999) - et al.
Isotope tracers of core formation
Earth-Sci. Rev.
(2018) - et al.
Abyssal peridotites reveal the near-chondritic Fe isotopic composition of the Earth
Earth Planet. Sci. Lett.
(2013) - et al.
A general moment NRIKX approach to the determination of equilibrium Fe isotopic fractionation factors: application to goethite and jarosite
Geochim. Cosmochim. Acta
(2012) - et al.
Magma redox and structural controls on iron isotope variations in Earth's mantle and crust
Earth Planet. Sci. Lett.
(2014) - et al.
Equilibrium isotope fractionation in the kaolinite, quartz, water system: prediction from first-principles density functional theory
Geochim. Cosmochim. Acta
(2007) On anharmonic and pressure corrections to the equilibrium isotopic constants for minerals
Geochim. Cosmochim. Acta
(1998)- et al.
Spinel-olivine-pyroxene equilibrium iron isotopic fractionation and applications to natural peridotites
Geochim. Cosmochim. Acta
(2015) - et al.
Sulfur-controlled iron isotope fractionation experiments of core formation in planetary bodies
Geochim. Cosmochim. Acta
(2015)
An assessment of iron isotope fractionation during core formation
Chem. Geol.
Iron isotope systematics in planetary reservoirs
Earth Planet. Sci. Lett.
First principles phonon calculations in material sciences
Scr. Mater.
Iron isotopic fractionation in mineral phases from Earth's lower mantle: did terrestrial magma ocean crystallisation fractionate iron isotopes?
Earth Planet. Sci. Lett.
Equilibrium fractionation of non-traditional isotopes: a molecular modelling perspective
Rev. Mineral. Geochem.
Projector augmented-wave method
Phys. Rev. B
Thermal expansion of forsterite up to the melting point
Geophys. Res. Lett.
Iron isotopic compositions of geological reference materials and chondrites
Geostand. Geoanal. Res.
Iron isotopic composition of geological reference materials and chondrites
Geostand. Geoanal. Res.
Symmetry and random sampling of symmetry independent configurations for the simulation of disordered solids
J. Phys. Condens. Matter
Sciphon: a data analysis software for nuclear resonant inelastic X-ray scattering with applications to Fe, Kr, Sn, Eu and Dy
J. Scynchrotron Rad.
Theory of Simple Liquids
Metrc-tensor flexible-cell algorithm for isothermal-isobaric molecular dynamics simulations
J. Chem. Phys.
Composition and state of the core
Annu. Rev. Earth Planet. Sci.
Cited by (8)
Equilibrium indium isotope fractionation between indium-bearing minerals: Insights from first-principles calculations
2024, Geochimica et Cosmochimica ActaSite-specific isotope effect: Insights from equilibrium magnesium isotope fractionation in mantle minerals
2023, Geochimica et Cosmochimica ActaIron and silicon isotope fractionation in silicate melts using first-principles molecular dynamics
2023, Geochimica et Cosmochimica ActaCitation Excerpt :Such corrections in complex silicate melts would be difficult to achieve since it would require precise knowledge of the molar volume contribution of each species in the melt. Furthermore, Pinilla et al. (2021) showed that these fluidicity effects were negligible in the calculation of equilibrium fractionation factor of iron in liquid Fex-1Sx alloys. Therefore, such effects were neglected here.
Fe isotope fractionation caused by phase transition of FeS and implications for Fe isotope signatures of the mantle and core
2023, Geochimica et Cosmochimica ActaCitation Excerpt :Pinilla et al. (2021) showed that the differences between the iron β-factors of hcp Fe and Fe3S from solid systems and those from molten systems are less than 0.01‰ at high pressure and temperature conditions, and the effect of S on the iron β-factor is no more than 0.02‰ with S concentrations ranging from 0.0 to 25 at% at Earth’s core formation conditions. Moreover, our calculated iron β-factors of hcp Fe agree very well with those of molten Fe calculated by Pinilla et al. (2021) at 60 GPa (see Fig. S9). Consequently, it is feasible and valid to use solid crystals as proxies for molten systems under outer core conditions, and the difference between the iron β-factor of FeS and that of hcp Fe can further reveal the possibility of the existence of S in the Earth's core.
The absence of an effect of nickel on iron isotope fractionation during core formation
2022, Geochimica et Cosmochimica ActaCitation Excerpt :Our results are also in agreement with previous experimental studies on Fe isotopic fractionation (Poitrasson et al., 2009; Hin et al., 2012; Liu et al., 2017) demonstrating the absence of Fe isotope fractionation at metal–silicate equilibrium as well as the absence of a resolvable fractionation with addition of limited amounts of Ni (Poitrasson et al., 2009; Liu et al., 2017). Significant S content in the alloy may induce a positive metal–silicate isotope fractionation with light isotope enrichment in the silicate (Shahar et al., 2015; Liu et al., 2017; Pinilla et al., 2021). A similar trend of isotopic fractionation has been observed for other light elements such as Si and C (Elardo et al., 2019; Shahar and Young, 2020).
Structural, redox and isotopic behaviors of iron in geological silicate glasses: A NRIXS study of Lamb-Mössbauer factors and force constants
2022, Geochimica et Cosmochimica Acta