A sodium-battery factory bets against cooling, a borrowed deadline enforces an award against Argentina, and a dinosaur surfaces at Lake Kariba

Sodium-ion cells swap the lithium in a battery for sodium, the element in table salt, and that swap changes the bill of materials. Sodium does not alloy with aluminium, so a sodium cell can use cheap aluminium foil to carry current at both electrodes, where a lithium cell needs costlier copper on one side.[7] The raw feedstock is cheaper too. Sodium carbonate, the soda ash the industry starts from, trades at roughly 300 dollars a tonne, a small fraction of the price of the lithium carbonate that feeds a lithium plant, and its inputs are far more abundant and widely spread.[7] The catch is energy density. Sodium cells store less energy for their weight, which rules them out of cars but matters little for a battery that sits still on the grid.[7]

The cooling claim is where the pitch concentrates. A lithium battery farm spends power and money keeping its cells within a temperature band, and Peak says a sodium system built to tolerate heat needs none of that hardware.[1] On that basis the company projects lower lifetime cost, high uptime and years of running without scheduled maintenance, and estimates the saved refrigeration could spare California ratepayers on the order of 100 million dollars a year.[1] Those are the company's figures rather than independent measurements, and the plant has not yet produced a cell to test them.[1]

There is demand behind the wager. Peak reports more than 6 GWh of supply agreements with the storage developers Jupiter Power, Energy Vault and RWE Americas, and its chief executive has spoken of more than a billion dollars in booked orders.[1][2] In June the company tied itself to a much larger partner, announcing a partnership with General Motors and an investment from GM Ventures, with the carmaker developing the cells in its Michigan battery labs and aiming to make them from 2028.[3]

What makes the factory a genuine test is the graveyard around it. Peak switched on the first grid-scale sodium-ion system in the country, a 3.5 megawatt-hour installation in Colorado, only in September 2025.[4] In the same stretch the field thinned sharply. Natron Energy, a rival that had opened a sodium-ion line, ceased operations in September 2025.[5] Bedrock Materials, a startup making sodium-ion raw materials, wound down in April 2025 and handed most of its seed money back to investors.[6] A plant betting that a home-grown sodium supply chain can undercut lithium is therefore opening into a market that has already buried companies making the same bet.[5][6]

Sodium's case does not rest on this one factory. Its raw inputs are cheaper than lithium's and far more widely distributed, which is the lasting argument for the chemistry.[7] Whether a plant can turn that advantage into cells cheaper than an established and falling lithium price is the question the next two years will answer, and it is a question about industrial economics rather than any single quarter's market.[6]

The award came from ICSID, the World Bank body that arbitrates disputes between foreign investors and states.[2] The law that lets US courts recognise an ICSID award, section 1650a of the federal code, sets no time limit for bringing the case.[1] American courts have a standard move for a statute that is silent on timing: they borrow the limitations period from the most similar state law.[1] The panel borrowed the District of Columbia's rule for enforcing an ordinary money judgment, which allows twelve years.[1] Argentina had argued for a far shorter clock, the three years that applies to confirming commercial arbitration awards under the Federal Arbitration Act and the New York Convention.[1]

The court's reasoning fixed on what an ICSID award is. Section 1650a says the Federal Arbitration Act does not apply and directs that an ICSID award be given the same full faith and credit as a judgment of a state court.[1] Confirmation is close to automatic, with no fresh look at the merits, because Article 54 of the ICSID Convention requires each member state to treat the award as if it were a final judgment of its own courts.[3] On that logic the closest analogy is enforcing a judgment, not policing an ordinary arbitration, and the judgment's longer clock controls.[1]

The money traces back to an airline. In 2008 Argentina renationalised Aerolíneas Argentinas and Austral Líneas Aéreas, then controlled by the Spanish group Marsans, and the Spanish investors Teinver and two affiliated companies took the expropriation to ICSID.[5] The tribunal ruled for them in 2017, and an ICSID committee left the award standing in 2019.[2] Titan Consortium later acquired the right to collect on it and filed the enforcement petition that reached the appeals court.[3]

Because the deadline is borrowed from state law, it is not the same everywhere. Courts in New York have applied a twenty-year period to the same kind of award, so a creditor's window depends on where the case is filed.[3] The result is that the United States, a leading forum for collecting on awards against sovereigns, still offers no single national clock for doing so, which rewards careful choice of court and gives a reluctant state one more variable to litigate.[4]

The animal, Musango matusadonaensis, belongs to the sauropodomorphs, the lineage that would later produce the giant long-necked sauropods.[1] This early member was modest, around 4.5 metres long and about 222 kilograms, roughly the mass of a fully grown pig, and it walked on two legs.[1][3] The study places it in a family of early sauropodomorphs called the unaysaurids, and it becomes the fifth dinosaur species formally described from Zimbabwe, from rocks of the Pebbly Arkose Formation laid down in the Norian stage of the Late Triassic.[1][2][3]

The bones say a little about the individual as well as the species. Growth rings in a lower leg bone record at least eight years of life, and the team reads the animal as nearly full-grown at death, a reading drawn from the bone rather than a certainty.[1][4] The same bone carries a patch of reactive growth that the researchers interpret as a healed injury or infection, while cautioning that they cannot say what befell it.[4] Its diet is left open too. The describers place it as a herbivore or an omnivore and note they cannot settle which.[1]

The wider claim rests on company as much as on anatomy. Musango shared its rock formation with Musankwa sanyatiensis, a dinosaur named in 2024 that weighed more than twice as much, so two distinct forms lived side by side.[3] Set against that, the authors point out that Zimbabwe's Triassic beds share no dinosaur species with the far richer beds of South Africa's Karoo, not a great distance away, and argue that the southern African fauna was regionally split rather than uniform.[1] That argument leans on absence as much as on the new bones, and the team offers it as an interpretation rather than a closed case.[1]

Whatever becomes of the argument over regional faunas, the skeleton itself is the firmer gain. Africa's Late Triassic is far more thinly sampled than the equivalent rocks of Europe and the Americas, so a single well-described skeleton shifts the picture more here than it would elsewhere, and each new form sets a firmer floor under any claim about how these early dinosaurs were spread across the ancient south.[1]