A team of researchers led by Professor Dongyan Huang has demonstrated that Earth's iron core may contain a colossal reservoir of hydrogen — equivalent to between 9 and 45 Earth oceans if this element were to react with oxygen. The results, obtained through extreme laboratory experiments, substantially revise long-held views on the formation of the planet and the source of its water resources. The study was published in early September 2026 and has already received extensive coverage in international scientific media, including Live Science.
How to peer into the center of the planet without a drilling tool
Direct sampling from Earth's core at a depth of more than 5,000 kilometers is physically impossible with any existing technology. For decades, geologists and geophysicists have relied on seismic waves reflected from the planet's internal boundaries. As early as the 1960s, analysis of these waves showed that the core's density is lower than that of pure iron and nickel, meaning that lighter elements must necessarily be part of its composition. However, precisely determining their nature and quantity remained an open question: among the candidates were silicon, oxygen, sulfur, and hydrogen — the smallest and most "elusive" of all.
Diamond anvil cells and atomic tomography
Professor Huang's team applied a two-stage methodology. In the first stage, samples of iron and hydrogen-rich silicate glass were clamped between diamond anvil cells and heated by lasers to 4,800 °C at a pressure of 111 gigapascals — conditions corresponding to the core–mantle boundary. In the second stage, after controlled cooling, the researchers performed atomic tomography: atoms were fixed one by one, which allowed them to build a three-dimensional map of elemental distribution at the nanoscale. It was precisely this approach that made it possible to quantitatively assess the hydrogen content in the modeled core medium.
From molar ratio to "oceans" in the core
Analysis of the tomographic data showed that the molar ratio of silicon to hydrogen in the samples was approximately 1:1. Since the mass fraction of silicon in the core is already fairly well established in the geophysical literature, the scientists converted this proportion into a mass fraction of hydrogen: from 0.07 to 0.36 percent. Comparing this figure with the total mass of the outer and inner core (on the order of 3.5 × 10²⁴ kg), the researchers obtained a range: the center of the planet contains hydrogen equivalent to 9–45 Earth oceans. For comparison — all of the world's water, including oceans, glaciers, and groundwater, amounts to roughly 1.4 × 10²¹ kg, meaning that even the lower bound of the estimate exceeds the mass of all surface water by several times.
Revising the "cometary" hypothesis and the primordial nature of water
The discovery has direct consequences for two competing models of the origin of water on Earth. The first is "primordial water": if hydrogen sank into the core during the early stages of accretion, then it was present in the protoplanetary disk from the very beginning, and water could have formed already during differentiation. The second is the "cometary" hypothesis, according to which the bulk of the planet's water was delivered by icy comets and asteroids during the period of the late heavy bombardment. Huang's team points out that, in their model, "cometary" hydrogen should have remained in the upper layers of the mantle and the Earth's crust, whereas the hydrogen detected in the core points to an earlier, "built-in" origin. This does not entirely rule out the cometary contribution, but it significantly narrows its share in the planet's overall water balance.
Contradictory data
The range of estimates — from 9 to 45 oceans — is itself a source of misinterpretation in the media: some outlets emphasize the upper bound ("45 oceans"), while others cite the full interval. Moreover, the authors of the study stress that all conclusions remain a mathematical extrapolation: the experiments were conducted on millimeter-scale samples, and the results were then scaled up to a core more than 3,500 kilometers wide. The laboratory conditions (111 GPa, 4,800 °C) only approximately reproduce the actual parameters of the core–mantle boundary, where pressure reaches 135–140 GPa and temperature is around 5,500–6,000 °C. Thus, the lower bound of the estimate (9 oceans) may be closer to reality than the upper one, but a final verdict will require additional experiments at higher pressures and independent replication of the results by other laboratories.
What's next: the path to confirmation
The scientists note that at least three lines of work are needed for the conclusions to be definitively confirmed: repeating the experiments on other diamond-anvil setups, modeling at pressures that more accurately match the real conditions of the core, and cross-checking against data from seismic tomography and isotopic analysis of meteoritic material. If subsequent studies confirm even the lower bound of the estimate, this will become one of the largest revisions of the internal model of Earth in the past half-century and will open a new layer of questions about how exactly the water envelope that made the planet habitable was formed.
