In the search for life beyond Earth, astronomers have tended to look up first and down second. They scan the skies for rocky planets in the so called habitable zone, measure starlight filtering through alien atmospheres, and hunt for oxygen, methane, or water vapor. But a new study in Nature Astronomy argues that one of the most decisive factors in whether a planet can host life is forged much earlier and much deeper, inside a global ocean of magma, when a young world is still tearing itself apart.
The study, led by Craig R. Walton of ETH Zurich and the University of Cambridge, focuses on two elements that every high school biology student learns are indispensable: phosphorus and nitrogen. Phosphorus forms the backbone of DNA and RNA and sits at the heart of ATP, the molecule that powers cells. Nitrogen is essential for amino acids and nucleic acids. On Earth, both elements have limited biological productivity at various times in the planet’s history. Remove too much of either from the surface environment and life struggles to get started, let alone flourish.
Walton and his colleagues argue that the fate of those elements is largely sealed during a violent episode known as core formation. In the first tens of millions of years after a rocky planet assembles, heat from accretion and radioactive decay melts much of its interior into a global magma ocean. Heavy iron sinks toward the center to form a metallic core. Lighter silicates float upward to form the mantle. In that differentiation process, elements partition between metal and rock according to pressure, temperature, and one crucial parameter: oxygen fugacity, a measure of how oxidizing or reducing the environment is.
In laboratory experiments that simulate conditions of high pressure and high temperature, phosphorus and nitrogen behave in opposite ways as oxygen fugacity changes. Under reducing conditions, phosphorus becomes siderophile. It prefers metal and dives into the core. Nitrogen, in contrast, becomes more lithophile and stays in the silicate mantle. Under oxidizing conditions, the pattern flips. Phosphorus remains in the mantle while nitrogen is more likely to avoid the core and potentially degas into the atmosphere.
That trade off produces what the authors call a chemical Goldilocks zone. Too reducing, and a planet’s mantle ends up stripped of phosphorus, with most of it locked in the core and effectively inaccessible to life. Too oxidizing, and nitrogen may be lost to space after degassing, or sequestered in forms that do not support robust atmospheres. Somewhere in between lies a narrow window in which both phosphorus and nitrogen remain available in biologically meaningful amounts.
The numbers are stark. In their model, a shift in oxygen fugacity of only a few logarithmic units relative to the iron wüstite buffer, written as ΔIW, can change mantle phosphorus concentrations by five orders of magnitude. At ΔIW minus four, strongly reducing conditions, the predicted mantle phosphorus concentration falls below one part per million. Earth’s mantle contains roughly 100 parts per million. At ΔIW plus one, a more oxidizing scenario, mantle phosphorus exceeds 1,000 parts per million while nitrogen drops to tenths of a part per million. For biology, such swings are not academic. Geological reconstructions suggest that changes of only a few fold in crustal phosphorus availability on early Earth may have influenced ocean oxygenation events more than two billion years ago.
The team tested their model against the known compositions of Earth and Mars. Using estimated oxygen fugacities for core formation on both planets, the calculations reproduce Earth’s mantle phosphorus concentration and fall within error bars for Mars. That agreement gives the authors confidence to extend the framework to exoplanets, where direct sampling is impossible but stellar abundances and mass radius measurements provide indirect clues.
To understand how much variation might arise simply from differences in stellar composition, the researchers turned to the Hypatia catalog of nearby stars. They examined main sequence stars with measured phosphorus, nitrogen, and iron abundances. The spread in stellar phosphorus to nitrogen ratios spans about 1.2 orders of magnitude. Low metallicity stars tend to be richer in phosphorus relative to nitrogen, reflecting the different astrophysical sources of those elements. Phosphorus is forged mainly in massive stars that explode as supernovae, enriching the interstellar medium on short timescales. Nitrogen is produced in lower mass stars through the carbon nitrogen oxygen cycle and released over longer timescales.
Despite that galactic scatter, when the team propagated stellar variability through their core formation model, they found that inheritance from the host star is a secondary effect. Oxygen fugacity during core formation dominates. Across plausible exoplanet scenarios, mantle nitrogen concentrations vary by less than an order of magnitude due to redox state alone, but phosphorus can swing by factors of 100,000. In other words, how a planet differentiates internally matters more for phosphorus availability than how much phosphorus its star happened to contain.
The implications ripple outward to some of the most talked about exoplanet classes. Hycean worlds, a term coined for hypothetical hydrogen rich sub Neptunes with deep global oceans, are expected to form under strongly reducing conditions. A hydrogen envelope buffers oxygen fugacity at low values. Under such circumstances, the model predicts severe phosphorus depletion in the mantle. Even if nitrogen remains abundant, life would face a phosphorus famine unless late impacts delivered fresh supplies after core formation.
There are hints in our own Solar System that core formation is decisive. Saturn’s moon Enceladus, which likely did not experience extensive metal silicate differentiation, shows evidence of abundant phosphorus dissolved in its subsurface ocean, according to a 2022 study in Proceedings of the National Academy of Sciences. Without a large iron core to siphon it away, phosphorus remains accessible. That contrast underscores how a planet’s deep interior can set surface chemistry billions of years later.
The study also threads into current debates about the Fermi paradox, the question of why we do not see obvious signs of extraterrestrial civilizations. If the chemical Goldilocks zone defined by oxygen fugacity is indeed narrow, then many otherwise Earth sized planets in the liquid water habitable zone may lack sufficient phosphorus in their crusts to kick start prebiotic chemistry. In one scenario explored by the authors, combining the fraction of planets in the liquid water habitable zone with the fraction that are Earth sized and the fraction with mantle phosphorus at least as high as Earth’s yields sobering odds. Thousands of candidate planets might need to be surveyed before finding one that satisfies all three criteria.
The argument rests on a simplified model of core formation. The team assumes a single stage differentiation rather than the likely multistage process Earth experienced. They estimate oxygen fugacity from the ratio of iron metal in the core to iron oxide in the mantle, using the relation ΔIW equals minus two times the base ten logarithm of the molar ratio of iron metal to iron oxide. They do not explicitly track progressive oxidation driven by silicon partitioning into the core, which experiments suggest may have shifted Earth’s mantle by two to three log units during formation. Nor do they model nitrogen loss to space in detail, a process that may have stripped a significant fraction of Earth’s initial inventory.
Still, the order of magnitude effects are large enough that the central conclusion appears robust. Planets that form too reducing will bury their phosphorus. Planets that form too oxidizing may struggle to retain nitrogen in useful forms. Earth, by this measure, sits in the middle. Its mantle oxygen fugacity during core formation, estimated near ΔIW minus two, yields mantle phosphorus near 100 parts per million and nitrogen near a few parts per million before subsequent atmospheric evolution. Not perfect. Not extreme. Balanced.
For astronomers planning the next generation of space telescopes, the message is subtle but urgent. Detecting oxygen or methane in an exoplanet atmosphere is only part of the story. To interpret those signals as biosignatures, researchers must also consider whether the planet’s interior chemistry ever allowed life to emerge. Bulk elemental ratios inferred from stellar spectra, combined with models of differentiation, may help flag worlds that are chemically primed or chemically starved.
Later this decade, missions such as the European Space Agency’s Ariel, scheduled for launch in 2029, will measure the atmospheres of hundreds of exoplanets. Ground based observatories like the Extremely Large Telescope will probe smaller, rockier worlds around nearby stars. Each spectrum will offer clues about atmospheric composition. Behind every spectrum lies a deeper question: what happened in the magma ocean?
The chemical habitability zone described by Walton and colleagues adds a new axis to the search for life. It is not enough for a planet to orbit at the right distance from its star. It must also have formed under the right redox conditions, with a core that did not hoard too much of life’s critical inventory. The next step is to refine estimates of exoplanet oxygen fugacity during formation, a task that will require better stellar abundance measurements, improved mass radius constraints, and continued high pressure experiments that simulate alien interiors.
Life depends on more than sunlight and water. It depends on chemistry locked in place when a planet is still molten. Earth’s deep past may have given it a rare balance. Whether that balance is common or scarce across the galaxy is now a question that reaches from the bottom of magma oceans to the top of telescope domes.
Source:
Walton, C. R., Rogers, L. K., Bonsor, A., Spaargaren, R., Shorttle, O., & Schönbächler, M. (2026). The chemical habitability of Earth and rocky planets prescribed by core formation. Nature Astronomy. https://doi.org/10.1038/s41550-026-02775-z
