A scrap of the newborn Earth surfaces off Mayotte, and a subsea cable finds more power by running hotter

The signature is an excess of neodymium-142, one particular form of a rare-earth metal. Its importance comes from where that isotope came from. A short-lived parent, samarium-146, decayed into it and then vanished from the planet entirely, its supply exhausted within roughly Earth’s first half-billion years.[1] Because the parent is long gone, any rock carrying an unusual amount of neodymium-142 must have been separated from the rest of the mantle while that parent was still active, which pins the sorting event to the planet’s earliest chapter.[1]

The team traces that early sorting to a mineral called bridgmanite, a magnesium silicate that is the most abundant material in Earth’s lower mantle. When a deep ocean of magma cooled and began to solidify, bridgmanite was among the first crystals to form, and it took up samarium slightly more readily than neodymium.[1] That small preference is enough to leave behind a reservoir with a distinctive isotopic ratio. To reproduce the lavas off Mayotte, the researchers estimate their mantle source holds on the order of eight to ten per cent of material formed in the Hadean, the eon named for Earth’s fiery infancy.[1]

The volcano supplying the samples is itself barely known. It is called Fani Maoré, a cone about 800 metres tall and two kilometres wide that sits roughly 50 kilometres east of Mayotte on the seafloor.[2] It did not exist in living memory. It built itself between 2018 and 2021 in an eruption that released about six cubic kilometres of lava, the largest effusive eruption anywhere since Iceland’s Laki in 1783 and the largest submarine eruption yet documented.[2][3]

Its discovery was an accident of seismology. In May 2018 a swarm of earthquakes began rattling Mayotte, strong enough to be felt but with no obvious source, and it took a research cruise the following year to find the new mountain on the abyssal plain.[2] The fresh, deep lava it produced turned out to be an unusually clean window into the mantle, which is why the samples were worth measuring at extreme precision in the first place.[2]

What the result implies is that the mantle does not mix as thoroughly as a simple picture would suggest. Convection has been stirring Earth’s interior for billions of years, and the intuition is that it should long ago have blended away any pocket of primordial material.[1] A measurable Hadean fingerprint in lava erupted this decade says instead that some early-formed reservoirs have hidden in the deep mantle and can still be tapped by a volcano at the surface.[1] If the interpretation holds, Fani Maoré offers something geochemists rarely get, a sample of the planet’s opening act delivered fresh to the seabed rather than read out of battered ancient rock.[1]

The physics is straightforward. A cable’s capacity is limited by how hot it is allowed to get, because the current flowing through the metal core heats it and the plastic insulation around that core can take only so much before it ages and fails.[1] Raise the permitted temperature and more current can pass through the same metal without laying anything new. The hard part is proving that the insulation, a cross-linked polymer wrapped tightly around the conductor, will survive decades at the higher heat, which is exactly what a qualification test exists to check.[5]

LS Cable is not alone in chasing that headroom. In March 2026 the Italian maker Prysmian qualified a 525-kilovolt sea cable to run at 90 degrees, which it said raises the power a single link can carry from about two gigawatts to about two and a half, enough to swing the supply of roughly half a million more homes onto one cable.[2] Prysmian stressed that the gain needs no larger converter stations or platforms at either end, so the extra capacity drops into infrastructure already built around the lower rating.[2]

That last point is why the tweak matters. Northern Europe’s offshore grid has deliberately frozen its design around a single specification, a 525-kilovolt direct-current link rated at two gigawatts, so that cables, converters and platforms can be ordered as near-identical repeats rather than one-off engineering projects.[3] The Dutch and German grid operator TenneT built its North Sea programme on exactly that template, and in 2023 it awarded roughly 23 billion euros of contracts to wire offshore wind farms to shore under it.[4] A drop-in uprate that squeezes more power from the frozen design compounds across every one of those repeated links.

The gate any such cable must clear is slow and expensive by design. Extruded direct-current cable systems are qualified against a recommendation known as CIGRE Technical Brochure 496 and the international standard IEC 62895, whose prequalification test puts a full cable assembly through a trial lasting about a year before the system is considered proven.[5] Only a handful of manufacturers worldwide can pass it, so LS Cable’s result is also a supply-chain event, widening the short list of vendors that Europe, North America and Korea’s own grid build-out can draw on.[1]

How far these incremental gains can run before the industry has to take a costlier step is the open question. Moving to an entirely new voltage class means new cables, new converters and new factories, whereas letting the insulation run ten or twenty degrees hotter buys capacity on equipment that already exists.[2] For now the bottleneck on connecting offshore wind is less the cable in the water than the small number of factories and test benches that can certify one, and every extra qualified supplier eases it.[1] The jump to 525 kilovolts from the earlier 320-kilovolt norm, first cleared in late 2018, roughly doubled what a single link could carry; the thermal uprates now under way are the industry wringing the last capacity it can from that leap before the next one.[6]