Light elements in Earth’s outer core
Which light elements are alloyed with iron in Earth’s outer core?
Summary
Why it is open
Nobody can sample the core, so its composition is estimated indirectly, by matching high-pressure experiments and theoretical calculations with seismic observations [13]. Work by Francis Birch and others in the 1950s showed that the core is mainly iron with some nickel, and that light impurities must be present because it is less dense than iron [1]. The abundances of elements in the Solar System, combined with models of Earth's history, limit the candidates to a few [1]; a 2021 review by Kei Hirose, Bernard Wood and Lidunka Vočadlo names sulphur, silicon, oxygen, carbon and hydrogen [13].
Matching is not unique. Different mixtures fit the outer core's density and sound velocity, and the preferred mixture depends on the core's temperature: in calculations published in 2020, hydrogen came out as a primary light element if the boundary between inner and outer core is at 4,800 to 5,400 kelvin, and oxygen as the most important one at 6,000 kelvin [9]. The seismic targets are not settled either. In 2018 Jessica Irving, Sanne Cottaar and Vedran Lekić wrote that discrepancies between models built from body waves and from normal modes (whole-Earth oscillations) show that the outer core's seismic properties "are not yet fully agreed upon" [7].
The source list asks both which elements are present and how they are distributed. No body sets a bar for a resolution. The 2021 review gives ranges rather than a composition, and says that tighter constraints on core temperature and a better link between the compositions of the solid inner and liquid outer core will help narrow them [13].
What is known
The size of the deficit (2002–2020). In 2002 Dario Alfè, Michael Gillan and Geoffrey Price cited estimates that the outer core is 5–10% less dense than iron, and the inner core 2–3% [1]. In 2020 Yasuhiro Kuwayama and colleagues combined density measurements of liquid iron to 116 gigapascals with shock-wave data and found that the outer core is 7.5–7.6% less dense than liquid iron, with a 3.7–4.4% higher sound velocity [11].
Limits from chemistry (2015–2020). Sulphur dissolves readily in iron but is moderately volatile, and geochemical and cosmochemical mass balances limit it to about 2% by weight; Youjun Zhang and colleagues, summarising earlier work in 2016, described carbon and hydrogen as too volatile to enter the core in large amounts [6]. From metal–silicate experiments, Rebecca Fischer and colleagues found in 2020 that carbon becomes less iron-loving at core-formation pressures and temperatures, so that the core holds at most 0.09 to 0.20% carbon by weight, a negligible share of its density [10]. James Badro and colleagues combined the mantle's depletion in trace elements with seismic constraints and concluded in 2015 that Earth accreted material more oxidised than today's mantle, with 2.7–5% oxygen and 2–3.6% silicon by weight entering the core [4].
Calculations (2002–2024). Using ab initio methods, which compute the behaviour of atoms from quantum mechanics, Alfè and colleagues found that oxygen partitions much more strongly than sulphur or silicon from the solid inner core into the liquid, so that oxygen is needed to explain the 4.5 ± 0.5% density jump at the inner-core boundary; they estimated about 8 mole per cent oxygen in the outer core with a similar amount of sulphur, silicon or both [1]. Badro, Alexander Côté and John Brodholt found in 2014 that matching the core's seismic properties requires oxygen but not silicon, sulphur or carbon, with at most 4.5% silicon and 2.4% sulphur [3]. In 2024 Tao Liu and Zhicheng Jing, combining first-principles simulations with neural-network methods, found that hydrogen partitions preferentially into the outer core and contributes much of that density jump, and that hydrogen and silicon are the preferred light elements, implying relatively reducing conditions during core formation [14].
Experiments (2016–2021). From shock-wave sound velocities of an iron–nickel–silicon alloy at outer-core conditions, combined with earlier studies, Zhang and colleagues estimated about 6% silicon, 2% sulphur and 1–2.5% oxygen by weight [6]. In 2021 Shoh Tagawa and colleagues found that at 30–60 gigapascals hydrogen strongly prefers liquid iron to silicate melt, with a partition coefficient of at least 29, implying 0.3–0.6% hydrogen in the core unless water arrived only at the end of accretion; that amount would explain 30–60% of the density deficit [12].
A 2021 synthesis. Hirose, Wood and Vočadlo give the likely outer-core composition as iron with 5% nickel, 1.7% sulphur, 0–4.0% silicon, 0.8–5.3% oxygen, 0.2% carbon and 0–0.26% hydrogen by weight [13].
Distribution. As the inner core crystallises it excludes light elements, which may rise and collect at the top of the core [2]. In 2010 George Helffrich and Satoshi Kaneshima reported wave speeds that require compositional variation in the top 300 kilometres of the outer core, with up to 5% light-element enrichment in an iron–oxygen–sulphur model [2]. The 2018 model of Irving and colleagues explains travel times of waves through the outer core "without requiring an anomalously slow ~90- to 450-km-thick layer" at its top [7].
Neutrinos. Geoneutrinos, antineutrinos from radioactive decay inside Earth, measure heat-producing elements: in its 2020 analysis the Borexino detector inferred 24.6 terawatts (68% interval +11.1 / −10.4) of radiogenic heat from uranium-238 and thorium-232 in the mantle, while potassium-40 geoneutrinos fall below its 1.8 MeV detection threshold [8]. That is another open problem, Earth's internal heat budget. Neutrinos made in the atmosphere offer a route to composition: their oscillations depend on the electron density along their path, which, compared with the density from seismology, gives the outer core's average ratio of atomic number to atomic weight [5].
Recent claims. In February 2026 Dongyang Huang and colleagues reported in Nature Communications metal–silicate experiments in laser-heated diamond-anvil cells, analysed by atom probe tomography, which maps composition in three dimensions at the nanometre scale [15]. They observed hydrogen in silicon- and oxygen-rich nanostructures within the iron and, from a silicon-to-hydrogen ratio close to one, estimated 0.07–0.36% hydrogen by weight in the core, equivalent to 9–45 oceans of water [15]. They note that earlier estimates spanned 10 to 10,000 parts per million by weight, and list their own uncertainties: residual hydrogen in the instrument, the unknown share of hydrogen left in the iron matrix, and the assumption that Earth accreted enough hydrogen [15].
Main approaches
Oxygen as the main light element. Strongest evidence: the inner-core density jump [1] and the match to seismic properties [3] require oxygen, and the combined geochemical and geophysical model gives 2.7–5% [4]. Main difficulty: the seismic match favours oxygen at high temperatures, while with the inner-core boundary at 4,800–5,400 kelvin hydrogen fits instead [9]; Liu and Jing attribute much of the density jump to hydrogen [14].
Silicon as a major element. Strongest evidence: shock-wave sound velocities favour about 6% silicon [6], and Liu and Jing favour silicon together with hydrogen [14]. Main difficulty: the 2014 seismic-matching calculations cap silicon at 4.5% [3], the 2015 model gives 2–3.6% [4], and the 2021 review allows 0–4.0% [13].
Hydrogen as a major element. Strongest evidence: hydrogen strongly prefers liquid iron in experiments [12], and calculations at lower core temperatures favour it [9, 14]. Main difficulty: hydrogen is hard to measure in high-pressure samples, and earlier high-pressure experiments inferred it from the expansion of the iron lattice that hydrogen causes [15]; the 2021 review allows at most 0.26% [13], less than the 0.3–0.6% inferred by Tagawa and colleagues [12], and Huang and colleagues conclude that their lower estimate requires a mixture of light elements rather than hydrogen alone [15].
Neutrino measurements. Rather than infer composition from density and sound velocity, measure the outer core's electron density directly [5]. Strongest evidence: C. Rott, A. Taketa and D. Bose calculated that 10 megaton-years of data could confirm an iron-like core, and one gigaton-year, 20 years with a 50-megaton detector, could discriminate between established composition models and measure hydrogen to a precision of 0.4% by weight [5]. Main difficulty: the method needs next-generation large-volume detectors such as Hyper-Kamiokande and decades of data [5].
Open sub-questions
- Temperature as the switch. Which inner-core boundary temperature, between the 4,800–5,400 kelvin and 6,000 kelvin cases, do melting experiments on the candidate alloys support, and does it select hydrogen or oxygen [9]?
- Hydrogen left in the iron. In atom-probe samples, what fraction of hydrogen stays in the iron matrix rather than the silicon- and oxygen-rich nanostructures, and how does it change the 0.07–0.36% estimate [15]?
- Two hydrogen estimates. Can one core-formation model with silicon and oxygen reconcile the 0.3–0.6% hydrogen from partitioning experiments with the 0–0.26% range of the 2021 review [12, 13, 15]?
- A layer at the top of the core. With the velocity gradient of the 2018 outer-core model, do travel times still require a compositionally distinct layer, and how much light-element enrichment does it allow [2, 7]?
- Neutrino exposure. What exposure would distinguish 0.26% hydrogen from none in the outer core, given the 0.4% precision projected for one gigaton-year [5, 13]?
- Oxides at the core's surface. If oxides crystallise near the top of the core, as the 2021 review suggests, would they collect at the core–mantle boundary as detectable structure [13]? See chemical heterogeneity at the core–mantle boundary.
Work on this in Papex
If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.
Key references
- Alfè, D., Gillan, M. J. & Price, G. D. (2002). Composition and temperature of the Earth's core constrained by combining ab initio calculations and seismic data. Earth and Planetary Science Letters 195(1–2), 91–98. 10.1016/S0012-821X(01)00568-4 · Checked 15 Oct 2026
- Helffrich, G. & Kaneshima, S. (2010). Outer-core compositional stratification from observed core wave speed profiles. Nature 468(7325), 807–810. 10.1038/nature09636 · Checked 15 Oct 2026
- Badro, J., Côté, A. S. & Brodholt, J. P. (2014). A seismologically consistent compositional model of Earth's core. Proceedings of the National Academy of Sciences 111(21), 7542–7545. 10.1073/pnas.1316708111 · Checked 15 Oct 2026
- Badro, J., Brodholt, J. P., Piet, H., Siebert, J. & Ryerson, F. J. (2015). Core formation and core composition from coupled geochemical and geophysical constraints. Proceedings of the National Academy of Sciences 112(40), 12310–12314. 10.1073/pnas.1505672112 · Checked 15 Oct 2026
- Rott, C., Taketa, A. & Bose, D. (2015). Spectrometry of the Earth using neutrino oscillations. Scientific Reports 5, 15225. 10.1038/srep15225 · Checked 15 Oct 2026
- Zhang, Y., Sekine, T., He, H., Yu, Y., Liu, F. & Zhang, M. (2016). Experimental constraints on light elements in the Earth's outer core. Scientific Reports 6, 22473. 10.1038/srep22473 · Checked 15 Oct 2026
- Irving, J. C. E., Cottaar, S. & Lekić, V. (2018). Seismically determined elastic parameters for Earth's outer core. Science Advances 4(6), eaar2538. 10.1126/sciadv.aar2538 · Checked 15 Oct 2026
- Agostini, M., Altenmüller, K., Appel, S., Atroshchenko, V., Bagdasarian, Z., Basilico, D. et al. (Borexino Collaboration) (2020). Comprehensive geoneutrino analysis with Borexino. Physical Review D 101(1), 012009. 10.1103/PhysRevD.101.012009 · Checked 15 Oct 2026
- Umemoto, K. & Hirose, K. (2020). Chemical compositions of the outer core examined by first principles calculations. Earth and Planetary Science Letters 531, 116009. 10.1016/j.epsl.2019.116009 · Checked 15 Oct 2026
- Fischer, R. A., Cottrell, E., Hauri, E., Lee, K. K. M. & Le Voyer, M. (2020). The carbon content of Earth and its core. Proceedings of the National Academy of Sciences 117(16), 8743–8749. 10.1073/pnas.1919930117 · Checked 15 Oct 2026
- Kuwayama, Y., Morard, G., Nakajima, Y., Hirose, K., Baron, A. Q. R., Kawaguchi, S. I. et al. (2020). Equation of state of liquid iron under extreme conditions. Physical Review Letters 124(16), 165701. 10.1103/PhysRevLett.124.165701 · Checked 15 Oct 2026
- Tagawa, S., Sakamoto, N., Hirose, K., Yokoo, S., Hernlund, J., Ohishi, Y. & Yurimoto, H. (2021). Experimental evidence for hydrogen incorporation into Earth's core. Nature Communications 12, 2588. 10.1038/s41467-021-22035-0 · Checked 15 Oct 2026
- Hirose, K., Wood, B. & Vočadlo, L. (2021). Light elements in the Earth's core. Nature Reviews Earth & Environment 2(9), 645–658. 10.1038/s43017-021-00203-6 · Checked 15 Oct 2026
- Liu, T. & Jing, Z. (2024). Hydrogen and silicon are the preferred light elements in Earth's core. Communications Earth & Environment 5, 282. 10.1038/s43247-024-01450-3 · Checked 15 Oct 2026
- Huang, D., Murakami, M., Gerstl, S. & Liebske, C. (2026). Experimental quantification of hydrogen content in the Earth's core. Nature Communications 17, 1211. 10.1038/s41467-026-68821-6 · Checked 15 Oct 2026
Source
Open, as listed by Wikipedia · Unsolved problems in geoscience. Checked 29 Sep 2026.
The status is the source’s. Papex does not decide whether a problem is solved.
Records on this problem
No public record addresses this problem yet.