Earth’s deep magma ocean never reached sulfide saturation
Affiliations | Corresponding Author | Cite as | Funding information
Published by the European Association of Geochemistry
under Creative Commons License CC BY-NC-ND 4.0
Keywords: Deep magma ocean, sulfur solubility, hadean matte
- Share this article
-
Article views:238Cumulative count of HTML views and PDF downloads.
- Download Citation
- Rights & Permissions
top
Abstract
Figures and Tables
 Figure 1 Backscattered images of two typical recovered runs. (a) Sample SCSS5 synthesised using pyrolytic glass as starting material, whereas (b) was synthetised with basaltic glass (SCSS4). Both runs were equilibrated at 53 GPa and 3800 K. |  Figure 2 Evolution of SCSS as a function of pressure for 80 GPa adiabat (a) and 20 GPa adiabat (b). For models that are composition dependent, we used Primitive Upper Mantle composition by (Palme and O’Neill, 2014). |  Figure 3 Comparison between the evolution of SCSS over the course of accretion from Equation 2, and the amount of sulfur in the magma ocean linked to core-mantle differentiation for both partial and full metal-silicate equilibration from Suer et al. (2017). The evolution of SCSS over the course of accretion is shown at constant fO2 (black solid curve), in an oxidised scenario (purple curve), and in a reduced scenario (blue curve). |  Table 1 Summary of experimental conditions. |
| Figure 1 | Figure 2 | Figure 3 | Table 1 |
top
Introduction
The formation of Earth’s main reservoirs, the metallic core and the surrounding silicate mantle, occurred early in Earth’s history. During the first few tens of million years, the Earth was covered by a deep magma ocean with pressure and temperature conditions exceeding 40 GPa and 3000 K respectively (Li and Agee, 1996
Li, J., Agee, C.B. (1996) Geochemistry of mantle-core differentiation at high pressure. Nature 381, 686–689. https://doi.org/10.1038/381686a0
; Siebert et al., 2012Siebert, J., Badro, J., Antonangeli, D., Ryerson, F.J. (2012) Metal–silicate partitioning of Ni and Co in a deep magma ocean. Earth Planetary Science Letters 321–322, 189–197. https://doi.org/10.1016/j.epsl.2012.01.013
). These conditions led to the separation of siderophile (iron-loving) from lithophile elements (silicate-loving) to form the core and the mantle, respectively. Understanding how the Bulk Silicate Earth (BSE) composition was established is of primary importance, and is directly linked to our comprehension of the Earth’s differentiation as well as the accretion process and the nature of Earth’s building blocks.Sulfur is a key element for understanding Earth’s formation. It has been proposed that the mantle’s signature in Highly Siderophile Elements (HSEs) could be explained by HSE segregation into a sulfide matte exsolved from the magma ocean towards the end of accretion (O’Neill, 1991
O’Neill, H.S.C. (1991) The origin of the Moon and the early history of the Earth—A chemical model. Part 2: The Earth. Geochimica Cosmochimica Acta 55, 1159–1172. https://doi.org/10.1016/0016-7037(91)90169-6
; Rubie et al., 2016Rubie, D.C., Laurenz, V., Jacobson, S.A., Morbidelli, A., Palme, H., Vogel, A.K., Frost, D.J. (2016) Highly siderophile elements were stripped from Earth’s mantle by iron sulfide segregation. Science 353, 1141–1144. https://doi.org/10.1126/science.aaf6919
). For this hypothesis to hold, the magma ocean must have reached sulfur saturation, enabling the segregation of a sulfide matte. The solubility of sulfur at sulfide saturation (SCSS) in molten silicate has been the subject of numerous studies, focusing mainly on the effects of the silicate composition on SCSS (Liu et al., 2007Liu, Y., Samaha, N., Baker, D.R. (2007) Sulfur concentration at sulfide saturation (SCSS) in magmatic silicate melts. Geochimica Cosmochimica Acta 71, 1783–1799. https://doi.org/10.1016/j.gca.2007.01.004
; Fortin et al., 2015Fortin, M., Riddle, J., Desjardins-langlais, Y., Baker, D.R. (2015) The effect of water on the sulfur concentration at sulfide saturation (SCSS) in natural melts. Geochimica Cosmochimica Acta 160, 100–116. https://doi.org/10.1016/j.gca.2015.03.022
; Namur et al., 2016Namur, O., Charlier, B., Holtz, F., Cartier, C., McCammon, C. (2016) Sulfur solubility in reduced mafic silicate melts: Implications for the speciation and distribution of sulfur on Mercury. Earth Planetary Science Letters 448, 102–114. https://doi.org/10.1016/j.epsl.2016.05.024
). Recently, the effects of pressure (P) and temperature (T) have been studied (Laurenz et al., 2016Laurenz, V., Rubie, D.C., Frost, D.J., Vogel, A.K. (2016) The importance of sulfur for the behavior of highly-siderophile elements during Earth’s differentiation. Geochimica Cosmochimica Acta 194, 123–138. https://doi.org/10.1016/j.gca.2016.08.012
; Blanchard et al., 2021Blanchard, I., Abeykoon, S., Frost, D.J., Rubie, D.C. (2021) Sulfur content at sulfide saturation of peridotitic melt at upper mantle conditions. American Mineralogist 106, 1835–1843. https://doi.org/10.2138/am-2021-7649
; Steenstra et al., 2022Steenstra, E.S., Lord, O.T., Vitale, S., Bullock, E.S., Klemme, S., Walter, M. (2022) Sulfur solubility in a deep magma ocean and implications for the deep sulfur cycle. Geochemical Perspective Letters 22, 5–9. https://doi.org/10.7185/geochemlet.2219
), but either at limited P–T conditions (Laurenz et al., 2016Laurenz, V., Rubie, D.C., Frost, D.J., Vogel, A.K. (2016) The importance of sulfur for the behavior of highly-siderophile elements during Earth’s differentiation. Geochimica Cosmochimica Acta 194, 123–138. https://doi.org/10.1016/j.gca.2016.08.012
; Blanchard et al., 2021Blanchard, I., Abeykoon, S., Frost, D.J., Rubie, D.C. (2021) Sulfur content at sulfide saturation of peridotitic melt at upper mantle conditions. American Mineralogist 106, 1835–1843. https://doi.org/10.2138/am-2021-7649
) or using basalt as starting composition, which might not be relevant in the context of the Earth’s formation (Steenstra et al., 2022Steenstra, E.S., Lord, O.T., Vitale, S., Bullock, E.S., Klemme, S., Walter, M. (2022) Sulfur solubility in a deep magma ocean and implications for the deep sulfur cycle. Geochemical Perspective Letters 22, 5–9. https://doi.org/10.7185/geochemlet.2219
).In this work, we have used a laser-heated diamond anvil cell (LH-DAC) to replicate the high P–T conditions of Earth’s differentiation in the laboratory. We have equilibrated a pyrolytic melt with pure FeS and showed that sulfur solubility in silicate melt remains high under these conditions. Additionally, we have also used a basaltic composition to compare our results with those reported by Steenstra et al., (2022)
Steenstra, E.S., Lord, O.T., Vitale, S., Bullock, E.S., Klemme, S., Walter, M. (2022) Sulfur solubility in a deep magma ocean and implications for the deep sulfur cycle. Geochemical Perspective Letters 22, 5–9. https://doi.org/10.7185/geochemlet.2219
. Our experimental results, along with data from the literature, were used to parameterise a model predicting the evolution of sulfur saturation in a deep magma ocean. We show that SCSS remains significantly high throughout the accretion process, with mean values around 3000 ppm.In contrast, sulfur content in the mantle produced from metal-silicate partitioning is always much lower (<600 ppm; Suer et al., 2017
Suer, T.A., Siebert, J., Remusat, L., Menguy, N., Fiquet, G. (2017) A sulfur-poor terrestrial core inferred from metal–silicate partitioning experiments. Earth Planetary Science Letters 469, 84–97. https://doi.org/10.1016/j.epsl.2017.04.016
). Hence, it seems unlikely that sulfur saturation was achieved during Earth’s differentiation. This challenges the Hadean Matte hypothesis and suggests that alternative mechanisms should be considered to explain the budget of HSEs of the mantle.top
Methods
We synthetised five samples between 53 and 72 GPa and 3800 to 4050 K using a laser-heated diamond anvil cell. Synthetic pyrolytic and basaltic glasses were melted and equilibrated with pure FeS. Experimental runs were subsequently recovered using Focused Ion Beam (FIB) and analysed using scanning electron microscopy energy dispersive X-rays. More details of the experimental and analytical procedures are given in the Supplementary Information. Table 1 summarises experimental conditions while Table S-1 gives the full chemical analysis of each run.
Table 1 Summary of experimental conditions.
| Starting Material | Experiment | P (GPa) | T (K) | ΔIW | SCSS (wt. %) |
| Pyrolite + FeS | SCSS1 | 64 ± 5 | 4000 ± 300 | −1.31 | 1.02 ± 0.27 |
| Pyrolite + FeS | SCSS2 | 61 ± 5 | 4000 ± 300 | −1.27 | 1.12 ± 0.36 |
| Pyrolite + FeS | SCSS3 | 72 ± 5 | 4050 ± 300 | −1.64 | 1.15 ± 0.15 |
| Basalt + FeS | SCSS4 | 53 ± 5 | 3800 ± 300 | −0.92 | 0.83 ± 0.31 |
| Pyrolite + FeS | SCSS5 | 53 ± 5 | 3800 ± 300 | −1.42 | 0.93 ± 0.25 |
top
Experimental Results
Each run consisted of a spherical sulfide blob surrounded by a quench silicate melt (see Figs. 1, S-1). The silicate melt of experiments performed using pyrolite as starting glass (SCSS1, 2, 3 and 5) or basaltic glass (SCSS4) fall in the range of terrestrial peridotites or basalts respectively. The amount of sulfur in the silicate melt from SCSS4 is slightly lower than in the other experiments, although the experimental runs look similar (see Fig. 1). The sulfide phase contains between 25 and 29 wt. % sulfur, along with roughly 6 to 10 wt. % oxygen, and 0.58 to 1.17 wt. % Si. This oxygen concentration is similar to that observed in previous metal-silicate partitioning diamond anvil cell experiments containing no sulfur. The concentration of silicon in sulfide is similar to observations from Steenstra et al. (2022)
Steenstra, E.S., Lord, O.T., Vitale, S., Bullock, E.S., Klemme, S., Walter, M. (2022) Sulfur solubility in a deep magma ocean and implications for the deep sulfur cycle. Geochemical Perspective Letters 22, 5–9. https://doi.org/10.7185/geochemlet.2219
, but much lower than in experiments containing no sulfide (e.g., Blanchard et al., 2017Blanchard, I., Siebert, J., Borensztajn, S., Badro, J. (2017) The solubility of heat-producing elements in Earth’s core. Geochemical Perspective Letters 1–5. https://doi.org/10.7185/geochemlet.1737
; Fischer et al., 2020Fischer, R.A., Cottrell, E., Hauri, E., Lee, K.K.M., Le Voyer, M. (2020) The carbon content of Earth and its core. Proceedings National Acadamey of Sciences U.S.A. 117, 8743–8749. https://doi.org/10.1073/pnas.1919930117
; Huang et al., 2021Huang, D., Siebert, J., Badro, J. (2021) High pressure partitioning behavior of Mo and W and late sulfur delivery during Earth’s core formation. Geochimica Cosmochimica Acta 310, 19–31. https://doi.org/10.1016/j.gca.2021.06.031
). Traces of Mg and Al were also detected in the sulfide (see Table S-1). Overall, the compositions of the sulfide phases are similar, regardless of whether the starting material was pyrolytic or basaltic glass. We calculated the redox conditions of our experiments relative to the iron-wüstite buffer (IW; see Supplementary Information), and derived values between −0.9 to −1.41, consistent with previous diamond anvil cell experiments on other systems (e.g., Siebert et al., 2012Siebert, J., Badro, J., Antonangeli, D., Ryerson, F.J. (2012) Metal–silicate partitioning of Ni and Co in a deep magma ocean. Earth Planetary Science Letters 321–322, 189–197. https://doi.org/10.1016/j.epsl.2012.01.013
; Blanchard et al., 2022Blanchard, I., Rubie, D.C., Jennings, E.S., Franchi, I.A., Zhao, X., Petitgirard, S., Miyajima, N., Jacobson, S.A., Morbidelli, A. (2022) The metal–silicate partitioning of carbon during Earth’s accretion and its distribution in the early solar system. Earth Planetary Science Letters 580, 117374. https://doi.org/10.1016/j.epsl.2022.117374
). No chemical or textural differences were observed between samples for which the temperature was maintained for either 10 or 120 seconds, suggesting that equilibrium is likely achieved in a very short time.Figure 1 Backscattered images of two typical recovered runs. (a) Sample SCSS5 synthesised using pyrolytic glass as starting material, whereas (b) was synthetised with basaltic glass (SCSS4). Both runs were equilibrated at 53 GPa and 3800 K.
top
Thermodynamics
The solubility of sulfur in a magma ocean is intrinsically linked to the evolution of SCSS with pressure and temperature. To date, no study has focused on the effects of P and T in a system relevant for Earth’s formation, specifically a pyrolytic magma ocean. Apart from the effects of pressure and temperature, it is well documented that the composition of the silicate itself has an impact on SCSS (Liu et al., 2007
Liu, Y., Samaha, N., Baker, D.R. (2007) Sulfur concentration at sulfide saturation (SCSS) in magmatic silicate melts. Geochimica Cosmochimica Acta 71, 1783–1799. https://doi.org/10.1016/j.gca.2007.01.004
; Fortin et al., 2015Fortin, M., Riddle, J., Desjardins-langlais, Y., Baker, D.R. (2015) The effect of water on the sulfur concentration at sulfide saturation (SCSS) in natural melts. Geochimica Cosmochimica Acta 160, 100–116. https://doi.org/10.1016/j.gca.2015.03.022
; Namur et al., 2016Namur, O., Charlier, B., Holtz, F., Cartier, C., McCammon, C. (2016) Sulfur solubility in reduced mafic silicate melts: Implications for the speciation and distribution of sulfur on Mercury. Earth Planetary Science Letters 448, 102–114. https://doi.org/10.1016/j.epsl.2016.05.024
). Among the most important elements influencing SCSS, experimental works have shown that iron (FeO) and silicon (SiO2) play major roles (O’Neill and Mavrogenes, 2002O’Neill, H.S.C., Mavrogenes, J.A. (2002) The Sulfide Capacity and the Sulfur Content at Sulfide Saturation of Silicate Melts at 1400°C and 1 bar. Journal of Petrology 43, 1049–1087. https://doi.org/10.1093/petrology/43.6.1049
; Wykes et al., 2015Wykes, J.L., O’Neill, H.S.C., Mavrogenes, J.A. (2015) The effect of FeO on the sulfur content at sulfide saturation (SCSS) and the selenium content at selenide saturation of silicate melts. Journal of Petrology 56, 1407–1424. https://doi.org/10.1093/petrology/egv041
; Smythe et al., 2017Smythe, D.J., Wood, B.J., Kiseeva, E.S. (2017) The S content of silicate melts at sulfide saturation: New experiments and a model incorporating the effects of sulfide composition. American Mineralogist 102, 795–803. https://doi.org/10.2138/am-2017-5800CCBY
; Blanchard et al., 2021Blanchard, I., Abeykoon, S., Frost, D.J., Rubie, D.C. (2021) Sulfur content at sulfide saturation of peridotitic melt at upper mantle conditions. American Mineralogist 106, 1835–1843. https://doi.org/10.2138/am-2021-7649
). Recent use of machine learning has confirmed that Si and Fe in silicate are indeed among the most important parameters influencing SCSS (ZhangZhou et al., 2024ZhangZhou, J., Li, Y., Chowdhury, P., Sen, S., Ghosh, U., Xu, Z., Liu, J., Wang, Z., Day, J.M.D. (2024) Predicting sulfide precipitation in magma oceans on Earth, Mars and the Moon using machine learning. Geochimica Cosmochimica Acta 366, 237–249. https://doi.org/10.1016/j.gca.2023.11.029
). While other components might also play a role, we chose not to overload our regression with parameters that are potentially not well constrained. Hence, we modelled SCSS following:
Eq. 1with XFeO and XSiO2 as the mole fraction of FeO and SiO2 respectively. We combined more than 200 data from the literature that cover a wide range of pressure (1 bar to 72 GPa), temperature (1573 to 4300 K) and silicate composition. Details of the selection is given in the Supplementary Information, and the final data set used for our parameterisation is provided in the Supplementary Information.
We obtain the following equation, with uncertainties in brackets:
Eq. 2Equation 2 shows that SCSS decreases with increasing pressure and XSiO2 , and increases with increasing T and XFeO. The effects of pressure and temperature that we derive from this study are coherent with previous ones from Laurenz et al. (2016)
Laurenz, V., Rubie, D.C., Frost, D.J., Vogel, A.K. (2016) The importance of sulfur for the behavior of highly-siderophile elements during Earth’s differentiation. Geochimica Cosmochimica Acta 194, 123–138. https://doi.org/10.1016/j.gca.2016.08.012
and Steenstra et al. (2022)Steenstra, E.S., Lord, O.T., Vitale, S., Bullock, E.S., Klemme, S., Walter, M. (2022) Sulfur solubility in a deep magma ocean and implications for the deep sulfur cycle. Geochemical Perspective Letters 22, 5–9. https://doi.org/10.7185/geochemlet.2219
who did not explicitly propose an effect of silicate composition. Nevertheless, it is not possible to directly compare our coefficients with theirs, since the regression used is not the same.Other models that include compositional effects previously predicted stronger effects of pressure (−190 and −265.8 for Blanchard et al., 2021
Blanchard, I., Abeykoon, S., Frost, D.J., Rubie, D.C. (2021) Sulfur content at sulfide saturation of peridotitic melt at upper mantle conditions. American Mineralogist 106, 1835–1843. https://doi.org/10.2138/am-2021-7649
model 2 and Smythe et al., 2017Smythe, D.J., Wood, B.J., Kiseeva, E.S. (2017) The S content of silicate melts at sulfide saturation: New experiments and a model incorporating the effects of sulfide composition. American Mineralogist 102, 795–803. https://doi.org/10.2138/am-2017-5800CCBY
respectively), and either a stronger temperature term (−14683 in Smythe et al., 2017Smythe, D.J., Wood, B.J., Kiseeva, E.S. (2017) The S content of silicate melts at sulfide saturation: New experiments and a model incorporating the effects of sulfide composition. American Mineralogist 102, 795–803. https://doi.org/10.2138/am-2017-5800CCBY
) or even positive ones (18159 in the model 2 from Blanchard et al., 2021Blanchard, I., Abeykoon, S., Frost, D.J., Rubie, D.C. (2021) Sulfur content at sulfide saturation of peridotitic melt at upper mantle conditions. American Mineralogist 106, 1835–1843. https://doi.org/10.2138/am-2021-7649
).In Figure S-3, we compare the SCSS values measured in the experimental runs with those calculated from Equation 2. This figure highlights the excellent agreement between our model and experimental data, despite the wide P–T compositional range studied here.
We also tested the effect of including only high P–T data to derive the evolution of SCSS. This is discussed in the Supplementary Information.
top
Discussion
The solubility of sulfur at sulfide saturation (SCSS) predicted from this work (Eq. 2) can be used to better understand the behaviour of sulfur during Earth’s formation and differentiation.
The behaviour of sulfur in the Earth’s magma ocean is of great importance, because many elements are chalcophile (sulfur-loving), hence their fate during episodes of melting is directly linked to that of sulfur. We have calculated SCSS as a function of pressure and temperature in a different context. First, we present in Figure 2a and 2b the evolution of SCSS along an adiabatic temperature profile for a magma ocean with a basal pressure of 80 GPa and 20 GPa (Miller et al., 1991
Miller, G.H., Stolper, E.M., Ahrens, T.J. (1991) The equation of state of a molten komatiite 2. Application to komatiite petrogenesis and the Hadean mantle. Journal of Geophysical Research 96, 849–864. https://doi.org/10.1029/91jb01203
) respectively. For models that have a composition dependency, we used the composition of the PUM (Primitive Upper Mantle) from Palme and O’Neill (2014)Palme, H., O’Neill, H.S.C. (2014) Cosmochemical Estimates of Mantle Composition, in: Treatise on Geochemistry 2nd Edition. Elsevier Ltd., pp. 1–39. https://doi.org/10.1016/B978-0-08-095975-7.00201-1
. From this figure, we observe that models of SCSS built from relatively low P–T experiments have a much more dramatic evolution, with higher starting SCSS values evolving toward much lower ones at high pressure. This is especially true for the 80 GPa adiabat, where for instance the Laurenz et al. (2016)Laurenz, V., Rubie, D.C., Frost, D.J., Vogel, A.K. (2016) The importance of sulfur for the behavior of highly-siderophile elements during Earth’s differentiation. Geochimica Cosmochimica Acta 194, 123–138. https://doi.org/10.1016/j.gca.2016.08.012
model starts at values way higher than 1.5 wt. % of sulfur of SCSS to finish at 35 ppm at 80 GPa. Conversely, models built with both low and high P–T experiments (this study; Steenstra et al., 2022Steenstra, E.S., Lord, O.T., Vitale, S., Bullock, E.S., Klemme, S., Walter, M. (2022) Sulfur solubility in a deep magma ocean and implications for the deep sulfur cycle. Geochemical Perspective Letters 22, 5–9. https://doi.org/10.7185/geochemlet.2219
) show a decrease (especially for 80 GPa adiabat), but it is much less intense. In our case, we predict an evolution of SCSS from about 11750 ppm at 0 GPa to about 4060 ppm at 80 GPa adiabat. For the 20 GPa adiabat, our study predicts almost a flat evolution (from about 6400 to about 5200 ppm), coherent with predictions from Steenstra et al. (2022)Steenstra, E.S., Lord, O.T., Vitale, S., Bullock, E.S., Klemme, S., Walter, M. (2022) Sulfur solubility in a deep magma ocean and implications for the deep sulfur cycle. Geochemical Perspective Letters 22, 5–9. https://doi.org/10.7185/geochemlet.2219
. Again, models obtained from large volume press experiments predict a much stronger decrease of SCSS with pressure (Laurenz et al., 2016Laurenz, V., Rubie, D.C., Frost, D.J., Vogel, A.K. (2016) The importance of sulfur for the behavior of highly-siderophile elements during Earth’s differentiation. Geochimica Cosmochimica Acta 194, 123–138. https://doi.org/10.1016/j.gca.2016.08.012
; Smythe et al., 2017Smythe, D.J., Wood, B.J., Kiseeva, E.S. (2017) The S content of silicate melts at sulfide saturation: New experiments and a model incorporating the effects of sulfide composition. American Mineralogist 102, 795–803. https://doi.org/10.2138/am-2017-5800CCBY
; Blanchard et al., 2021Blanchard, I., Abeykoon, S., Frost, D.J., Rubie, D.C. (2021) Sulfur content at sulfide saturation of peridotitic melt at upper mantle conditions. American Mineralogist 106, 1835–1843. https://doi.org/10.2138/am-2021-7649
), with almost an order of magnitude difference for Laurenz et al. (2016)Laurenz, V., Rubie, D.C., Frost, D.J., Vogel, A.K. (2016) The importance of sulfur for the behavior of highly-siderophile elements during Earth’s differentiation. Geochimica Cosmochimica Acta 194, 123–138. https://doi.org/10.1016/j.gca.2016.08.012
. All in all, we predict for both low and high pressure adiabat high values of SCSS (>2500 ppm), in contrast to previous models obtained at lower P–T conditions.Figure 2 Evolution of SCSS as a function of pressure for 80 GPa adiabat (a) and 20 GPa adiabat (b). For models that are composition dependent, we used Primitive Upper Mantle composition by (Palme and O’Neill, 2014
Palme, H., O’Neill, H.S.C. (2014) Cosmochemical Estimates of Mantle Composition, in: Treatise on Geochemistry 2nd Edition. Elsevier Ltd., pp. 1–39. https://doi.org/10.1016/B978-0-08-095975-7.00201-1
).It has been suggested that at the end of accretion, an episode of Hadean Matte occurred, during which ponds of FeS segregated from the magma ocean, due to the saturation of the magma ocean in sulfur (O’Neill, 1991
O’Neill, H.S.C. (1991) The origin of the Moon and the early history of the Earth—A chemical model. Part 2: The Earth. Geochimica Cosmochimica Acta 55, 1159–1172. https://doi.org/10.1016/0016-7037(91)90169-6
). Using our formalism, we can now assess whether the terrestrial magma ocean ever reached sulfur saturation over the course of its existence.The parameterisation derived in Equation 2 was used in a core formation model to assess the evolution of SCSS during the magma ocean phase on Earth. The temperature of the magma ocean was fixed as the arithmetic mean of the mantle liquidus proposed by Fiquet et al. (2010)
Fiquet, G., Auzende, A.L., Siebert, J., Corgne, A., Bureau, H., Ozawa, H., Garbarino, G. (2010) Melting of peridotite to 140 gigapascals. Science 329, 1516–1518. https://doi.org/10.1126/science.1192448
and Andrault et al. (2011)Andrault, D., Bolfan-Casanova, N., Nigro, G.L., Bouhifd, M.A., Garbarino, G., Mezouar, M. (2011) Solidus and liquidus profiles of chondritic mantle: Implication for melting of the Earth across its history. Earth Planetary Science Letters 174, 181–191. https://doi.org/10.1016/j.epsl.2011.02.006
, and the depth of the magma ocean was fixed at 40 % of the mantle’s depth at each stage. The core formation model was discretised into 100 steps, each representing the accretion of 1 % of Earth’s mass. We have also tested the effect of changing fO2 over the course of accretion by changing the amount of FeO, since the redox path of Earth’s formation is a matter of debate (e.g., Siebert et al., 2013Siebert, J., Badro, J., Antonangeli, D., Ryerson, F.J. (2013) Terrestrial accretion under oxidizing conditions. Science 339, 1194–1197. https://doi.org/10.1126/science.1227923
; Rubie et al., 2016Rubie, D.C., Laurenz, V., Jacobson, S.A., Morbidelli, A., Palme, H., Vogel, A.K., Frost, D.J. (2016) Highly siderophile elements were stripped from Earth’s mantle by iron sulfide segregation. Science 353, 1141–1144. https://doi.org/10.1126/science.aaf6919
). Three different scenarios were tested: one where the fO2 is increasing over the course of accretion from IW−4.5 to IW−2.3, one where fO2 decreases from IW−1.3 to IW−2.3, and one constant fO2 of IW−2.3. Figure 3 presents the evolution of SCSS with the accreted fraction of the Earth, following Equation 2, and changing fO2 conditions as detailed previously. The composition of the silicate is fixed to the composition of the Primitive Upper Mantle proposed by Palme and O’Neill (2014)Palme, H., O’Neill, H.S.C. (2014) Cosmochemical Estimates of Mantle Composition, in: Treatise on Geochemistry 2nd Edition. Elsevier Ltd., pp. 1–39. https://doi.org/10.1016/B978-0-08-095975-7.00201-1
in the case of the constant fO2, and the amount of FeO is varied for the two other redox paths. Given the importance of FeO in the evolution of SCSS (see Eq. 2), we additionally varied its abundance on the mantle on the two other redox paths through the course of accretion.Figure 3 Comparison between the evolution of SCSS over the course of accretion from Equation 2, and the amount of sulfur in the magma ocean linked to core-mantle differentiation for both partial and full metal-silicate equilibration from Suer et al. (2017)
Suer, T.A., Siebert, J., Remusat, L., Menguy, N., Fiquet, G. (2017) A sulfur-poor terrestrial core inferred from metal–silicate partitioning experiments. Earth Planetary Science Letters 469, 84–97. https://doi.org/10.1016/j.epsl.2017.04.016
. The evolution of SCSS over the course of accretion is shown at constant fO2 (black solid curve), in an oxidised scenario (purple curve), and in a reduced scenario (blue curve).Rubie et al. (2016)
Rubie, D.C., Laurenz, V., Jacobson, S.A., Morbidelli, A., Palme, H., Vogel, A.K., Frost, D.J. (2016) Highly siderophile elements were stripped from Earth’s mantle by iron sulfide segregation. Science 353, 1141–1144. https://doi.org/10.1126/science.aaf6919
were first to define empirically an effective pressure that describes SCSS and equilibration pressure for the entire magma ocean (Peq−s=kS×PCMB), provided that the corresponding temperature is between the liquidus and solidus of peridotite. It was shown by Steenstra et al. (2022)Steenstra, E.S., Lord, O.T., Vitale, S., Bullock, E.S., Klemme, S., Walter, M. (2022) Sulfur solubility in a deep magma ocean and implications for the deep sulfur cycle. Geochemical Perspective Letters 22, 5–9. https://doi.org/10.7185/geochemlet.2219
that kS can range from 0.3 to 0.9 without impacting the evolution of SCSS, so we did not try to vary this value. Here, we used a constant value of 0.4, as suggested by Rubie et al. (2016)Rubie, D.C., Laurenz, V., Jacobson, S.A., Morbidelli, A., Palme, H., Vogel, A.K., Frost, D.J. (2016) Highly siderophile elements were stripped from Earth’s mantle by iron sulfide segregation. Science 353, 1141–1144. https://doi.org/10.1126/science.aaf6919
. To assess whether the magma ocean ever reached SCSS, we compared the estimation of sulfur concentration at saturation with the residual amount of sulfur in the magma ocean due to metal-silicate partitioning during core formation. Figure 3 shows the amount of sulfur in the mantle predicted from metal-silicate partitioning experiments by Suer et al. (2017)Suer, T.A., Siebert, J., Remusat, L., Menguy, N., Fiquet, G. (2017) A sulfur-poor terrestrial core inferred from metal–silicate partitioning experiments. Earth Planetary Science Letters 469, 84–97. https://doi.org/10.1016/j.epsl.2017.04.016
. We tested two homogeneous accretion scenarios: one where the magma ocean fully equilibrates with the forming core, and another where partial equilibration occurs between the magma ocean and the core (Deguen et al., 2014Deguen, R., Landeau, M., Olson, P. (2014) Turbulent metal–silicate mixing, fragmentation, and equilibration in magma oceans. Earth Planetary Science Letters 391, 274–287. https://doi.org/10.1016/j.epsl.2014.02.007
). In this second scenario, equilibration is complete during the first 80 %, but becomes partial during the later stages of accretion when larger bodies are accreted (Suer et al., 2017Suer, T.A., Siebert, J., Remusat, L., Menguy, N., Fiquet, G. (2017) A sulfur-poor terrestrial core inferred from metal–silicate partitioning experiments. Earth Planetary Science Letters 469, 84–97. https://doi.org/10.1016/j.epsl.2017.04.016
).Regardless of the extent of equilibration, the model suggests that a maximum of about 500 ppm of sulfur can be stored in the magma ocean linked to core-mantle differentiation. This is far below the estimated SCSS, which is shown to be always higher than 2500 ppm, irrespective of the evolution of redox over the course of accretion (see Fig. 3). This is also true using a model derived only from high P–T experiments (see Supplementary Information, and Fig. S-4).
It has been proposed that oxygen can influence the partitioning of sulfur between metal and silicate (e.g., Gendre et al., 2022
Gendre, H., Badro, J., Wehr, N., Borensztajn, S. (2022) Martian core composition from experimental high-pressure metal-silicate phase equilibria. Geochemical Perspective Letters 21, 42–46. https://doi.org/10.7185/geochemlet.2216
). We could not include the effect of oxygen on SCSS due to insufficient data. However, extrapolating data from Gendre et al. (2022)Gendre, H., Badro, J., Wehr, N., Borensztajn, S. (2022) Martian core composition from experimental high-pressure metal-silicate phase equilibria. Geochemical Perspective Letters 21, 42–46. https://doi.org/10.7185/geochemlet.2216
obtained at relatively low P–T conditions (<12 GPa and <2473 K) to core formation conditions would only strengthen our conclusions, as the presence of oxygen in the metallic phase enhances the siderophility of sulfur.As can be seen in Figure 3, we predict a high SCSS during the whole course of accretion, with values above 2500 ppm, but a low concentration of sulfur predicted from metal-silicate partitioning experiments (<500 ppm). Previous experiments and models proposed by Steenstra et al., (2022)
Steenstra, E.S., Lord, O.T., Vitale, S., Bullock, E.S., Klemme, S., Walter, M. (2022) Sulfur solubility in a deep magma ocean and implications for the deep sulfur cycle. Geochemical Perspective Letters 22, 5–9. https://doi.org/10.7185/geochemlet.2219
are comparable to our results, although they predict slightly lower values of SCSS at the end of accretion (about 3500 ppm). In addition to a significant increase of the investigated pressure range, the main difference with our results is that we incorporate new high P–T data points with the effects of FeO and SiO2 present in the silicate on SCSS. These results suggest that saturation of the magma ocean in sulfur was probably never reached over the course of accretion.These estimations differ significantly from those of Rubie et al. (2016)
Rubie, D.C., Laurenz, V., Jacobson, S.A., Morbidelli, A., Palme, H., Vogel, A.K., Frost, D.J. (2016) Highly siderophile elements were stripped from Earth’s mantle by iron sulfide segregation. Science 353, 1141–1144. https://doi.org/10.1126/science.aaf6919
, who proposed up to about 7000 ppm of S in the mantle at the end of accretion. The difference stems from major differences in the accretion models. Rubie et al. (2016)Rubie, D.C., Laurenz, V., Jacobson, S.A., Morbidelli, A., Palme, H., Vogel, A.K., Frost, D.J. (2016) Highly siderophile elements were stripped from Earth’s mantle by iron sulfide segregation. Science 353, 1141–1144. https://doi.org/10.1126/science.aaf6919
considered that a large part of planetesimals that built the Earth were undifferentiated (fully oxidised). The evolution of SCSS they proposed is based on relatively low P–T experiments, that are very rich in HSEs (Laurenz et al., 2016)Laurenz, V., Rubie, D.C., Frost, D.J., Vogel, A.K. (2016) The importance of sulfur for the behavior of highly-siderophile elements during Earth’s differentiation. Geochimica Cosmochimica Acta 194, 123–138. https://doi.org/10.1016/j.gca.2016.08.012
. Interactions of HSEs with sulfur are poorly constrained, but given that HSEs are chalcophile, their presence in the metallic phase of the runs could influence sulfur behaviour. Laurenz et al. (2016)Laurenz, V., Rubie, D.C., Frost, D.J., Vogel, A.K. (2016) The importance of sulfur for the behavior of highly-siderophile elements during Earth’s differentiation. Geochimica Cosmochimica Acta 194, 123–138. https://doi.org/10.1016/j.gca.2016.08.012
propose a strong decrease of SCSS with increasing pressure, leading to sulfur saturation in a deep enough magma ocean. Consequently, Rubie et al. (2016)Rubie, D.C., Laurenz, V., Jacobson, S.A., Morbidelli, A., Palme, H., Vogel, A.K., Frost, D.J. (2016) Highly siderophile elements were stripped from Earth’s mantle by iron sulfide segregation. Science 353, 1141–1144. https://doi.org/10.1126/science.aaf6919
proposed that during the course of accretion, the magma ocean experienced at least one episode of sulfide liquid segregation, which stripped HSEs from Earth’s mantle. They suggested that a subsequent episode of late accretion is mandatory to reproduce supra-chondritic Pd/Ir and Ru/Ir ratios observed in today’s mantle, and remove the excess of HSEs (Becker, 2006Becker, H., Horan, M.F., Walker, R.J., Gao, S., Lorand, J.-P., Rudnick, R.L. (2006) Highly siderophile element composition of the Earth’s primitive upper mantle : Constraints from new data on peridotite massifs and xenoliths. Geochimica Cosmochimica Acta 70, 4528–4550. https://doi.org/10.1016/j.gca.2006.06.004
; Walker et al., 2015Walker, R.J., Bermingham, K., Liu, J., Puchtel, I.S., Touboul, M., Worsham, E.A. (2015) In search of late-stage planetary building blocks. Chemical Geology 411, 125–142. https://doi.org/10.1016/j.chemgeo.2015.06.028
). In light of our new high P–T experiments, we suggest that sulfide saturation was likely never reached during Earth’s accretion (see Fig. 3). Hence, the proposition that HSE’s mantle signatures can be explained by sulfide segregation early in the Earth’s history appears unlikely.Recent work have proposed that a significant decrease in the metal-silicate partitioning of platinum during core formation might explain abundance of this HSE in today’s mantle (Suer et al., 2021
Suer, T.-A., Siebert, J., Remusat, L., Day, J.M.D., Borensztajn, S., Doisneau, B., Fiquet, G. (2021) Reconciling metal–silicate partitioning and late accretion in the Earth. Nature Communications 12. https://doi.org/10.1038/s41467-021-23137-5
), rather than a combined effect of sulfide saturation and a late arrival of this element.top
Acknowledgments
We would like to thank Edgar Steenstra, an anonymous reviewer and the editor of this paper, Ambre Luguet, for their constructive and helpful comments. We thank James Badro for laser heating the samples. The LHDAC lab is supported by the LabEx UnivEarthS, ANR-10-LABX-0023 and ANR-18-IDEX-0001. Parts of this work were supported by IPGP multidisciplinary programme PARI, and by Paris–IdF region SESAME Grant no. 12015908. We acknowledge Anja Schreiber from GFZ who extracted the FIB lamellae, Stephan Borensztajn from IPGP who helped us with SEM analyses. JS acknowledges support from the French National Research Agency (ANR project VolTerre, grant No. ANR-14-CE33-0017-01) and support from the Institut Universitaire de France.
Editor: Ambre Luguet
top
References
Andrault, D., Bolfan-Casanova, N., Nigro, G.L., Bouhifd, M.A., Garbarino, G., Mezouar, M. (2011) Solidus and liquidus profiles of chondritic mantle: Implication for melting of the Earth across its history. Earth Planetary Science Letters 174, 181–191. https://doi.org/10.1016/j.epsl.2011.02.006
Show in context The temperature of the magma ocean was fixed as the arithmetic mean of the mantle liquidus proposed by Fiquet et al. (2010) and Andrault et al. (2011), and the depth of the magma ocean was fixed at 40 % of the mantle’s depth at each stage.
View in article
Becker, H., Horan, M.F., Walker, R.J., Gao, S., Lorand, J.-P., Rudnick, R.L. (2006) Highly siderophile element composition of the Earth’s primitive upper mantle : Constraints from new data on peridotite massifs and xenoliths. Geochimica Cosmochimica Acta 70, 4528–4550. https://doi.org/10.1016/j.gca.2006.06.004
Show in context They suggested that a subsequent episode of late accretion is mandatory to reproduce supra-chondritic Pd/Ir and Ru/Ir ratios observed in today’s mantle, and remove the excess of HSEs (Becker, 2006; Walker et al., 2015).
View in article
Blanchard, I., Siebert, J., Borensztajn, S., Badro, J. (2017) The solubility of heat-producing elements in Earth’s core. Geochemical Perspective Letters 1–5. https://doi.org/10.7185/geochemlet.1737
Show in context The concentration of silicon in sulfide is similar to observations from Steenstra et al. (2022), but much lower than in experiments containing no sulfide (e.g., Blanchard et al., 2017; Fischer et al., 2020; Huang et al., 2021).
View in article
Blanchard, I., Abeykoon, S., Frost, D.J., Rubie, D.C. (2021) Sulfur content at sulfide saturation of peridotitic melt at upper mantle conditions. American Mineralogist 106, 1835–1843. https://doi.org/10.2138/am-2021-7649
Show in context Recently, the effects of pressure (P) and temperature (T) have been studied (Laurenz et al., 2016; Blanchard et al., 2021; Steenstra et al., 2022), but either at limited P–T conditions (Laurenz et al., 2016; Blanchard et al., 2021) or using basalt as starting composition, which might not be relevant in the context of the Earth’s formation (Steenstra et al., 2022).
View in article
Among the most important elements influencing SCSS, experimental works have shown that iron (FeO) and silicon (SiO2) play major roles (O’Neill and Mavrogenes, 2002; Wykes et al., 2015; Smythe et al., 2017; Blanchard et al., 2021).
View in article
Other models that include compositional effects previously predicted stronger effects of pressure (−190 and −265.8 for Blanchard et al., 2021 model 2 and Smythe et al., 2017 respectively), and either a stronger temperature term (−14683 in Smythe et al., 2017) or even positive ones (18159 in the model 2 from Blanchard et al., 2021).
View in article
Again, models obtained from large volume press experiments predict a much stronger decrease of SCSS with pressure (Laurenz et al., 2016; Smythe et al., 2017; Blanchard et al., 2021), with almost an order of magnitude difference for Laurenz et al. (2016).
View in article
Blanchard, I., Rubie, D.C., Jennings, E.S., Franchi, I.A., Zhao, X., Petitgirard, S., Miyajima, N., Jacobson, S.A., Morbidelli, A. (2022) The metal–silicate partitioning of carbon during Earth’s accretion and its distribution in the early solar system. Earth Planetary Science Letters 580, 117374. https://doi.org/10.1016/j.epsl.2022.117374
Show in context We calculated the redox conditions of our experiments relative to the iron-wüstite buffer (IW; see Supplementary Information), and derived values between −0.9 to −1.41, consistent with previous diamond anvil cell experiments on other systems (e.g., Siebert et al., 2012; Blanchard et al., 2022).
View in article
Deguen, R., Landeau, M., Olson, P. (2014) Turbulent metal–silicate mixing, fragmentation, and equilibration in magma oceans. Earth Planetary Science Letters 391, 274–287. https://doi.org/10.1016/j.epsl.2014.02.007
Show in context We tested two homogeneous accretion scenarios: one where the magma ocean fully equilibrates with the forming core, and another where partial equilibration occurs between the magma ocean and the core (Deguen et al., 2014).
View in article
Fiquet, G., Auzende, A.L., Siebert, J., Corgne, A., Bureau, H., Ozawa, H., Garbarino, G. (2010) Melting of peridotite to 140 gigapascals. Science 329, 1516–1518. https://doi.org/10.1126/science.1192448
Show in context The temperature of the magma ocean was fixed as the arithmetic mean of the mantle liquidus proposed by Fiquet et al. (2010) and Andrault et al. (2011), and the depth of the magma ocean was fixed at 40 % of the mantle’s depth at each stage.
View in article
Fischer, R.A., Cottrell, E., Hauri, E., Lee, K.K.M., Le Voyer, M. (2020) The carbon content of Earth and its core. Proceedings National Acadamey of Sciences U.S.A. 117, 8743–8749. https://doi.org/10.1073/pnas.1919930117
Show in context The concentration of silicon in sulfide is similar to observations from Steenstra et al. (2022), but much lower than in experiments containing no sulfide (e.g., Blanchard et al., 2017; Fischer et al., 2020; Huang et al., 2021).
View in article
Fortin, M., Riddle, J., Desjardins-langlais, Y., Baker, D.R. (2015) The effect of water on the sulfur concentration at sulfide saturation (SCSS) in natural melts. Geochimica Cosmochimica Acta 160, 100–116. https://doi.org/10.1016/j.gca.2015.03.022
Show in context The solubility of sulfur at sulfide saturation (SCSS) in molten silicate has been the subject of numerous studies, focusing mainly on the effects of the silicate composition on SCSS (Liu et al., 2007; Fortin et al., 2015; Namur et al., 2016).
View in article
Apart from the effects of pressure and temperature, it is well documented that the composition of the silicate itself has an impact on SCSS (Liu et al., 2007; Fortin et al., 2015; Namur et al., 2016).
View in article
Gendre, H., Badro, J., Wehr, N., Borensztajn, S. (2022) Martian core composition from experimental high-pressure metal-silicate phase equilibria. Geochemical Perspective Letters 21, 42–46. https://doi.org/10.7185/geochemlet.2216
Show in context It has been proposed that oxygen can influence the partitioning of sulfur between metal and silicate (e.g., Gendre et al., 2022).
View in article
We could not include the effect of oxygen on SCSS due to insufficient data. However, extrapolating data from Gendre et al. (2022) obtained at relatively low P–T conditions (<12 GPa and <2473 K) to core formation conditions would only strengthen our conclusions, as the presence of oxygen in the metallic phase enhances the siderophility of sulfur.
View in article
Huang, D., Siebert, J., Badro, J. (2021) High pressure partitioning behavior of Mo and W and late sulfur delivery during Earth’s core formation. Geochimica Cosmochimica Acta 310, 19–31. https://doi.org/10.1016/j.gca.2021.06.031
Show in context The concentration of silicon in sulfide is similar to observations from Steenstra et al. (2022), but much lower than in experiments containing no sulfide (e.g., Blanchard et al., 2017; Fischer et al., 2020; Huang et al., 2021).
View in article
Laurenz, V., Rubie, D.C., Frost, D.J., Vogel, A.K. (2016) The importance of sulfur for the behavior of highly-siderophile elements during Earth’s differentiation. Geochimica Cosmochimica Acta 194, 123–138. https://doi.org/10.1016/j.gca.2016.08.012
Show in context Recently, the effects of pressure (P) and temperature (T) have been studied (Laurenz et al., 2016; Blanchard et al., 2021; Steenstra et al., 2022), but either at limited P–T conditions (Laurenz et al., 2016; Blanchard et al., 2021) or using basalt as starting composition, which might not be relevant in the context of the Earth’s formation (Steenstra et al., 2022).
View in article
The effects of pressure and temperature that we derive from this study are coherent with previous ones from Laurenz et al. (2016) and Steenstra et al. (2022) who did not explicitly propose an effect of silicate composition.
View in article
This is especially true for the 80 GPa adiabat, where for instance the Laurenz et al. (2016) model starts at values way higher than 1.5 wt. % of sulfur of SCSS to finish at 35 ppm at 80 GPa.
View in article
Again, models obtained from large volume press experiments predict a much stronger decrease of SCSS with pressure (Laurenz et al., 2016; Smythe et al., 2017; Blanchard et al., 2021), with almost an order of magnitude difference for Laurenz et al. (2016).
View in article
The evolution of SCSS they proposed is based on relatively low P–T experiments, that are very rich in HSEs (Laurenz et al., 2016).
View in article
Interactions of HSEs with sulfur are poorly constrained, but given that HSEs are chalcophile, their presence in the metallic phase of the runs could influence sulfur behaviour. Laurenz et al. (2016) propose a strong decrease of SCSS with increasing pressure, leading to sulfur saturation in a deep enough magma ocean.
View in article
Li, J., Agee, C.B. (1996) Geochemistry of mantle-core differentiation at high pressure. Nature 381, 686–689. https://doi.org/10.1038/381686a0
Show in context The formation of Earth’s main reservoirs, the metallic core and the surrounding silicate mantle, occurred early in Earth’s history. During the first few tens of million years, the Earth was covered by a deep magma ocean with pressure and temperature conditions exceeding 40 GPa and 3000 K respectively (Li and Agee, 1996; Siebert et al., 2012).
View in article
Liu, Y., Samaha, N., Baker, D.R. (2007) Sulfur concentration at sulfide saturation (SCSS) in magmatic silicate melts. Geochimica Cosmochimica Acta 71, 1783–1799. https://doi.org/10.1016/j.gca.2007.01.004
Show in context The solubility of sulfur at sulfide saturation (SCSS) in molten silicate has been the subject of numerous studies, focusing mainly on the effects of the silicate composition on SCSS (Liu et al., 2007; Fortin et al., 2015; Namur et al., 2016).
View in article
Apart from the effects of pressure and temperature, it is well documented that the composition of the silicate itself has an impact on SCSS (Liu et al., 2007; Fortin et al., 2015; Namur et al., 2016).
View in article
Miller, G.H., Stolper, E.M., Ahrens, T.J. (1991) The equation of state of a molten komatiite 2. Application to komatiite petrogenesis and the Hadean mantle. Journal of Geophysical Research 96, 849–864. https://doi.org/10.1029/91jb01203
Show in context First, we present in Figure 2a and 2b the evolution of SCSS along an adiabatic temperature profile for a magma ocean with a basal pressure of 80 GPa and 20 GPa (Miller et al., 1991) respectively.
View in article
Namur, O., Charlier, B., Holtz, F., Cartier, C., McCammon, C. (2016) Sulfur solubility in reduced mafic silicate melts: Implications for the speciation and distribution of sulfur on Mercury. Earth Planetary Science Letters 448, 102–114. https://doi.org/10.1016/j.epsl.2016.05.024
Show in context The solubility of sulfur at sulfide saturation (SCSS) in molten silicate has been the subject of numerous studies, focusing mainly on the effects of the silicate composition on SCSS (Liu et al., 2007; Fortin et al., 2015; Namur et al., 2016).
View in article
Apart from the effects of pressure and temperature, it is well documented that the composition of the silicate itself has an impact on SCSS (Liu et al., 2007; Fortin et al., 2015; Namur et al., 2016).
View in article
O’Neill, H.S.C. (1991) The origin of the Moon and the early history of the Earth—A chemical model. Part 2: The Earth. Geochimica Cosmochimica Acta 55, 1159–1172. https://doi.org/10.1016/0016-7037(91)90169-6
Show in context It has been proposed that the mantle’s signature in Highly Siderophile Elements (HSEs) could be explained by HSE segregation into a sulfide matte exsolved from the magma ocean towards the end of accretion (O’Neill, 1991; Rubie et al., 2016).
View in article
It has been suggested that at the end of accretion, an episode of Hadean Matte occurred, during which ponds of FeS segregated from the magma ocean, due to the saturation of the magma ocean in sulfur (O’Neill, 1991).
View in article
O’Neill, H.S.C., Mavrogenes, J.A. (2002) The Sulfide Capacity and the Sulfur Content at Sulfide Saturation of Silicate Melts at 1400°C and 1 bar. Journal of Petrology 43, 1049–1087. https://doi.org/10.1093/petrology/43.6.1049
Show in context Among the most important elements influencing SCSS, experimental works have shown that iron (FeO) and silicon (SiO2) play major roles (O’Neill and Mavrogenes, 2002; Wykes et al., 2015; Smythe et al., 2017; Blanchard et al., 2021).
View in article
Palme, H., O’Neill, H.S.C. (2014) Cosmochemical Estimates of Mantle Composition, in: Treatise on Geochemistry 2nd Edition. Elsevier Ltd., pp. 1–39. https://doi.org/10.1016/B978-0-08-095975-7.00201-1
Show in context For models that have a composition dependency, we used the composition of the PUM (Primitive Upper Mantle) from Palme and O’Neill (2014).
View in article
For models that are composition dependent, we used Primitive Upper Mantle composition by (Palme and O’Neill, 2014).
View in article
The composition of the silicate is fixed to the composition of the Primitive Upper Mantle proposed by Palme and O’Neill (2014) in the case of the constant fO2, and the amount of FeO is varied for the two other redox paths.
View in article
Rubie, D.C., Laurenz, V., Jacobson, S.A., Morbidelli, A., Palme, H., Vogel, A.K., Frost, D.J. (2016) Highly siderophile elements were stripped from Earth’s mantle by iron sulfide segregation. Science 353, 1141–1144. https://doi.org/10.1126/science.aaf6919
Show in context It has been proposed that the mantle’s signature in Highly Siderophile Elements (HSEs) could be explained by HSE segregation into a sulfide matte exsolved from the magma ocean towards the end of accretion (O’Neill, 1991; Rubie et al., 2016).
View in article
We have also tested the effect of changing fO2 over the course of accretion by changing the amount of FeO, since the redox path of Earth’s formation is a matter of debate (e.g., Siebert et al., 2013; Rubie et al., 2016).
View in article
Rubie et al. (2016) were first to define empirically an effective pressure that describes SCSS and equilibration pressure for the entire magma ocean (P eq−s=k S×P CMB), provided that the corresponding temperature is between the liquidus and solidus of peridotite.
View in article
Here, we used a constant value of 0.4, as suggested by Rubie et al. (2016).
View in article
These estimations differ significantly from those of Rubie et al. (2016), who proposed up to about 7000 ppm of S in the mantle at the end of accretion.
View in article
Rubie et al. (2016) considered that a large part of planetesimals that built the Earth were undifferentiated (fully oxidised).
View in article
Consequently, Rubie et al. (2016) proposed that during the course of accretion, the magma ocean experienced at least one episode of sulfide liquid segregation, which stripped HSEs from Earth’s mantle.
View in article
Siebert, J., Badro, J., Antonangeli, D., Ryerson, F.J. (2012) Metal–silicate partitioning of Ni and Co in a deep magma ocean. Earth Planetary Science Letters 321–322, 189–197. https://doi.org/10.1016/j.epsl.2012.01.013
Show in context The formation of Earth’s main reservoirs, the metallic core and the surrounding silicate mantle, occurred early in Earth’s history. During the first few tens of million years, the Earth was covered by a deep magma ocean with pressure and temperature conditions exceeding 40 GPa and 3000 K respectively (Li and Agee, 1996; Siebert et al., 2012).
View in article
We calculated the redox conditions of our experiments relative to the iron-wüstite buffer (IW; see Supplementary Information), and derived values between −0.9 to −1.41, consistent with previous diamond anvil cell experiments on other systems (e.g., Siebert et al., 2012; Blanchard et al., 2022).
View in article
Siebert, J., Badro, J., Antonangeli, D., Ryerson, F.J. (2013) Terrestrial accretion under oxidizing conditions. Science 339, 1194–1197. https://doi.org/10.1126/science.1227923
Show in context We have also tested the effect of changing fO2 over the course of accretion by changing the amount of FeO, since the redox path of Earth’s formation is a matter of debate (e.g., Siebert et al., 2013; Rubie et al., 2016).
View in article
Smythe, D.J., Wood, B.J., Kiseeva, E.S. (2017) The S content of silicate melts at sulfide saturation: New experiments and a model incorporating the effects of sulfide composition. American Mineralogist 102, 795–803. https://doi.org/10.2138/am-2017-5800CCBY
Show in context Among the most important elements influencing SCSS, experimental works have shown that iron (FeO) and silicon (SiO2) play major roles (O’Neill and Mavrogenes, 2002; Wykes et al., 2015; Smythe et al., 2017; Blanchard et al., 2021).
View in article
Other models that include compositional effects previously predicted stronger effects of pressure (−190 and −265.8 for Blanchard et al., 2021 model 2 and Smythe et al., 2017 respectively), and either a stronger temperature term (−14683 in Smythe et al., 2017) or even positive ones (18159 in the model 2 from Blanchard et al., 2021).
View in article
Again, models obtained from large volume press experiments predict a much stronger decrease of SCSS with pressure (Laurenz et al., 2016; Smythe et al., 2017; Blanchard et al., 2021), with almost an order of magnitude difference for Laurenz et al. (2016).
View in article
Steenstra, E.S., Lord, O.T., Vitale, S., Bullock, E.S., Klemme, S., Walter, M. (2022) Sulfur solubility in a deep magma ocean and implications for the deep sulfur cycle. Geochemical Perspective Letters 22, 5–9. https://doi.org/10.7185/geochemlet.2219
Show in context Recently, the effects of pressure (P) and temperature (T) have been studied (Laurenz et al., 2016; Blanchard et al., 2021; Steenstra et al., 2022), but either at limited P–T conditions (Laurenz et al., 2016; Blanchard et al., 2021) or using basalt as starting composition, which might not be relevant in the context of the Earth’s formation (Steenstra et al., 2022).
View in article
Additionally, we have also used a basaltic composition to compare our results with those reported by Steenstra et al., (2022).
View in article
The concentration of silicon in sulfide is similar to observations from Steenstra et al. (2022), but much lower than in experiments containing no sulfide (e.g., Blanchard et al., 2017; Fischer et al., 2020; Huang et al., 2021).
View in article
The effects of pressure and temperature that we derive from this study are coherent with previous ones from Laurenz et al. (2016) and Steenstra et al. (2022) who did not explicitly propose an effect of silicate composition.
View in article
Conversely, models built with both low and high P–T experiments (this study; Steenstra et al., 2022) show a decrease (especially for 80 GPa adiabat), but it is much less intense.
View in article
For the 20 GPa adiabat, our study predicts almost a flat evolution (from about 6400 to about 5200 ppm), coherent with predictions from Steenstra et al. (2022).
View in article
It was shown by Steenstra et al. (2022) that k S can range from 0.3 to 0.9 without impacting the evolution of SCSS, so we did not try to vary this value.
View in article
Previous experiments and models proposed by Steenstra et al., (2022) are comparable to our results, although they predict slightly lower values of SCSS at the end of accretion (about 3500 ppm).
View in article
Suer, T.A., Siebert, J., Remusat, L., Menguy, N., Fiquet, G. (2017) A sulfur-poor terrestrial core inferred from metal–silicate partitioning experiments. Earth Planetary Science Letters 469, 84–97. https://doi.org/10.1016/j.epsl.2017.04.016
Show in context In contrast, sulfur content in the mantle produced from metal-silicate partitioning is always much lower (<600 ppm; Suer et al., 2017).
View in article
Comparison between the evolution of SCSS over the course of accretion from Equation 2, and the amount of sulfur in the magma ocean linked to core-mantle differentiation for both partial and full metal-silicate equilibration from Suer et al. (2017).
View in article
Figure 3 shows the amount of sulfur in the mantle predicted from metal-silicate partitioning experiments by Suer et al. (2017).
View in article
In this second scenario, equilibration is complete during the first 80 %, but becomes partial during the later stages of accretion when larger bodies are accreted (Suer et al., 2017).
View in article
Suer, T.-A., Siebert, J., Remusat, L., Day, J.M.D., Borensztajn, S., Doisneau, B., Fiquet, G. (2021) Reconciling metal–silicate partitioning and late accretion in the Earth. Nature Communications 12. https://doi.org/10.1038/s41467-021-23137-5
Show in context Recent work have proposed that a significant decrease in the metal-silicate partitioning of platinum during core formation might explain abundance of this HSE in today’s mantle (Suer et al., 2021), rather than a combined effect of sulfide saturation and a late arrival of this element.
View in article
Walker, R.J., Bermingham, K., Liu, J., Puchtel, I.S., Touboul, M., Worsham, E.A. (2015) In search of late-stage planetary building blocks. Chemical Geology 411, 125–142. https://doi.org/10.1016/j.chemgeo.2015.06.028
Show in context They suggested that a subsequent episode of late accretion is mandatory to reproduce supra-chondritic Pd/Ir and Ru/Ir ratios observed in today’s mantle, and remove the excess of HSEs (Becker, 2006; Walker et al., 2015).
View in article
Wykes, J.L., O’Neill, H.S.C., Mavrogenes, J.A. (2015) The effect of FeO on the sulfur content at sulfide saturation (SCSS) and the selenium content at selenide saturation of silicate melts. Journal of Petrology 56, 1407–1424. https://doi.org/10.1093/petrology/egv041
Show in context Among the most important elements influencing SCSS, experimental works have shown that iron (FeO) and silicon (SiO2) play major roles (O’Neill and Mavrogenes, 2002; Wykes et al., 2015; Smythe et al., 2017; Blanchard et al., 2021).
View in article
ZhangZhou, J., Li, Y., Chowdhury, P., Sen, S., Ghosh, U., Xu, Z., Liu, J., Wang, Z., Day, J.M.D. (2024) Predicting sulfide precipitation in magma oceans on Earth, Mars and the Moon using machine learning. Geochimica Cosmochimica Acta 366, 237–249. https://doi.org/10.1016/j.gca.2023.11.029
Show in context Recent use of machine learning has confirmed that Si and Fe in silicate are indeed among the most important parameters influencing SCSS (ZhangZhou et al., 2024).
View in article
top
Supplementary Information
The Supplementary Information includes:
- Methods and Data Selection
- Tables S-1 and S-2
- Figures S-1 and S-2
- Supplementary Information References
Download the Supplementary Information (PDF)
Download Table S-2 (.xlsx)
