Ancient Zircons Reveal Oxygen Was Quietly Oxidizing Earth’s Mantle Billions of Years Ago

Deep inside tiny crystals of zircon, trapped like insects in amber, sit microscopic grains of apatite that have guarded a chemical secret for more than three billion years. When researchers led by Hugo Moreira of the University of Portsmouth interrogated those secrets with one of the brightest X-ray beams on the planet, they uncovered evidence that Earth’s mantle beneath volcanic arcs began to oxidize far earlier than anyone had firmly established — long before the atmosphere itself turned permanently oxygen-rich. The finding, published in Nature Geoscience, suggests that the slow chemical weathering of Earth’s earliest continents, and the recycling of their products into the deep Earth, quietly transformed the redox state of the mantle throughout the Archaean eon.

The central question the team tackled is one of the most stubborn debates in Earth science: why is the mantle beneath modern volcanic arcs so much more oxidized than the rest of the upper mantle? Today, the answer seems straightforward. When oceanic plates descend into subduction zones, they carry water, sediments and oxidized sulfur species that are released into the overlying mantle wedge, fundamentally changing its chemistry. But on the early Earth, the picture was murky. The Archaean atmosphere and oceans contained almost no free oxygen, so surface weathering was dominated by anoxic reactions that produced reduced rather than oxidized minerals. If there was essentially no oxygen at the surface, how could oxidized material have been delivered to the mantle?

To answer that question, the researchers needed a geological recorder that could preserve snapshots of mantle chemistry across vast stretches of time. They found it in sanukitoid magmas, a distinctive family of rocks that rose from metasomatized — chemically altered — mantle sources after about 3.5 billion years ago. Unlike the tonalite–trondhjemite–granodiorite suites that formed by direct melting of altered basaltic slabs, sanukitoids are hybrids, born from deep recycling and crust–mantle mixing in ways that resemble many aspects of modern arc magmatism. Once considered rare and obscure occurrences in ancient continents, these rocks have become central to understanding how the lithospheric mantle and continental crust evolved together, and they now serve as natural monitors of oxygen fugacity in sub-arc domains.

The sample suite was deliberately targeted and geochemically coherent, spanning more than three billion years of Earth history. It included three samples dated at 3.42 billion years and two at 2.9 billion years from the Pilbara Craton in Australia; one 2.7-billion-year-old sample from the Karelian Craton in Finland; two 2.7-billion-year-old samples from the Kaapvaal Craton in South Africa; two 2.13-billion-year-old samples from the São Francisco Craton in Brazil; and six 425-million-year-old samples from the Northern Highland Terrane in Scotland. All share the dual chemical signature typical of sanukitoids — enrichment in crustal incompatible elements such as barium and strontium, alongside relatively high abundances of mantle-forming elements including magnesium, chromium, nickel and vanadium — strongly indicating a crust-metasomatized mantle source. By restricting the comparison to this single magma type, the team minimized the petrogenetic, thermal and geodynamic variables that could otherwise confound comparisons across different geological eons.

The analytical challenge was considerable, because apatite is notoriously prone to alteration. Crystals sitting in a rock matrix may have been chemically re-equilibrated long after the magma cooled, erasing the primary magmatic signal. The team’s solution was elegant: they focused on apatite inclusions sealed inside zircon crystals, where the robust zircon host shields its prisoner from secondary alteration. Using sulfur K-edge micro-X-ray absorption near-edge structure spectroscopy at the European Synchrotron Radiation Facility in Grenoble, France, they measured the relative abundances of sulfur in its three key oxidation states — the fully oxidized S6+, the intermediate S4+ and the reduced S2− — within these armoured inclusions. In total, the team presented 337 analyses in 39 apatite inclusions and 213 analyses in 65 matrix apatite grains.

The logic behind the measurement is well established. Apatite incorporates both reduced and oxidized sulfur into its crystal structure, and laboratory experiments have shown that the ratio of S6+ to total sulfur varies systematically with the oxygen fugacity of the magmatic system in which the crystal grew. Oxygen fugacity, expressed in geology as a deviation from the fayalite–magnetite–quartz buffer, or ΔFMQ, is effectively a measure of how oxidizing the environment was. When the team converted their sulfur measurements into fugacity values using an experimental calibration oxybarometer, a striking pattern emerged. The apatite inclusions yielded ΔFMQ values that increased progressively with the age of their host zircons, with the oldest 3.4-billion-year-old samples showing the lowest values and younger samples showing progressively higher ones.

Crucially, the team then ruled out the alternative explanations one by one. The samples are petrologically similar and crystallized within about 100 degrees Celsius of each other, a variation too small to matter. Pressure corrections, if applied, would actually push the trend toward even higher fugacity values, meaning the documented shift is a minimum. Fractional crystallization was tested using compositionally equivalent 425-million-year-old samples from Scotland that range from 48 to 71 weight percent silica and are genetically linked to a common parental magma; the data showed no demonstrable effect on ΔFMQ. Degassing, which can dramatically change fugacity in volcanic environments, was shown by forward modelling to be minor in these deep plutonic rocks, which crystallized at pressures above 0.4 gigapascals, contributing at most 0.4 log units. With every intrinsic process eliminated, the remaining explanation is metasomatism of the sub-arc upper mantle by increasingly oxidized components arising from the recycling of surface materials.

The comparison with the ambient convecting mantle makes the result even more compelling. Recent estimates suggest the ambient mantle has increased in oxygen fugacity by roughly 1.0 to 1.5 log units over the last 3.5 billion years, a rise some researchers attribute to the sluggish homogenization of a primordial oxidized layer generated during core–mantle differentiation. The sub-arc mantle curve traced by the sanukitoid apatites follows a similar overall pattern but sits displaced by approximately 1.0 to 1.5 ΔFMQ units toward higher values, a distinction that appears to have existed for at least 3.4 billion years. That persistent offset requires a substantial input of oxidative material superimposed on a mantle already undergoing intrinsic secular changes — in other words, an external, surface-driven oxidation process operating on the mantle wedge since at least the Mesoarchaean.

Perhaps the most surprising twist lies in the mathematics of sulfur itself. Because the relationship between oxygen fugacity and the proportion of oxidized sulfur is nonlinear, the same fugacity shift demands dramatically different sulfur chemistry at different times. At 3.5 billion years ago, less than 2 percent of total sulfur needed to be in the oxidized S6+ form to shift the mantle’s fugacity by 1.5 log units, because the ambient mantle was then so reduced. Today, more than 65 percent oxidized sulfur is required to achieve a comparable shift. This means that even tiny, localized reservoirs of marine sulfate on the early Earth would have been disproportionately effective at oxidizing the mantle wedge, since sulfur has multiple oxidation states, high mobility in fluids, and immense oxidizing leverage when the starting point is reduced. A veneer of sulfate-bearing sediments riding on downgoing slabs could have acted as a redox filter, oxidizing slab-derived fluids before they reached the mantle.

The implications ripple outward through Earth history. The secular rise in fugacity recorded in zircon-hosted apatite inclusions is the first direct evidence that oxidative weathering and the recycling of surface materials into the mantle occurred before permanent atmospheric oxygenation, which arrived only with the Great Oxidation Event between 2.3 and 2.2 billion years ago. The sub-arc mantle apparently reached conditions comparable to those of the modern Earth by about 2.1 billion years ago, well after that transition. The team proposes a feedback loop: early subduction involving inorganic oxidation and intermittent ‘whiffs’ of oxygen between 3.4 and 2.7 billion years ago regulated an early rise in sub-arc fugacity, which in turn produced more oxidizing volcanic gases, accelerated the formation of more felsic crust, intensified oxidative weathering at the surface, and ultimately contributed to the long-term build-up of atmospheric oxygen. Rather than a passive spectator of surface evolution, the early mantle may have been an active participant — oxidized from within by the very continents it helped create.

Subject of Research: Secular oxidation of the Archaean sub-arc mantle driven by recycling of oxidatively weathered surface materials

Article Title: Progressive oxidative weathering raised sub-arc mantle fugacity on the Archaean Earth

Article References: Progressive oxidative weathering raised sub-arc mantle fugacity on the Archaean Earth. (n.d.). https://doi.org/10.1038/s41561-026-02119-6

Image Credits: AI Generated

DOI: 10.1038/s41561-026-02119-6

Keywords: oxygen fugacity, Archaean, subduction, sanukitoid, apatite inclusions, zircon, sulfur speciation, XANES spectroscopy, mantle metasomatism, oxidative weathering, Great Oxidation Event, Nature Geoscience