The team that built Northvolt's recycling division spent eight years on a process the company never managed to scale. Northvolt filed for bankruptcy in March 2025. Its engineers did not scatter.
In late September they resurfaced as Refinyx, bought back the 134 patents they had written, and signed a first major U.S. customer. Their pitch is unfashionably plain: win on cost.
Refinyx, led by the former Revolt team, bought the 134-patent portfolio and the Västerås pilot lab, with backing from Qarlbo Energy.
The thesis: circular battery materials have to win on cost, quality and reliability — not on premiums the market will not keep paying.
The process outlived the company
Northvolt Revolt was the recycling arm of Europe's best-funded battery maker. Eight years, three successive generations of process design, and one industrial-scale plant built on that design, in Skellefteå, Sweden. When the parent collapsed, the technology stayed.
Refinyx pulls battery-grade nickel, cobalt and lithium out of end-of-life cells, gigafactory scrap, industrial residues and municipal waste streams. The engineering claim is narrower than the mission statement, and more useful.
No organic solvents. No intermediate refining step.
That second point is the business. Most Western recyclers stop at black mass — a mixed metallic powder — and ship it to Asia for the final upgrade into battery-grade material. Refinyx says its platform skips that leg of the journey and produces usable material at the plant.
A large part of the sector has been building its business case on three premiums the market is not going to pay long-term: a green premium, a government-subsidy premium, and a national-security premium. There is no rescue price coming.— Dhruv Malhotra, Chief Business Officer and Co-Founder, Refinyx
Co-founders Mahmood Alemrajabi, the chief executive, and Ramiar Vaziri, the chief technology officer, are the two who wrote the process. They carried it through those three generations and into the first commercial-scale plant, then watched the company around it fail. The relaunch is their second attempt to make the same chemistry pay.
A continent's battery wreckage, repurposed
Refinyx is one node in a larger salvage operation. Northvolt raised billions, built factories across Sweden and Germany, and filed for bankruptcy anyway in March 2025. Its manufacturing assets and remaining intellectual property went to the U.S. firm Lyten, together with operational cell capacity and projects still under construction.
The pattern repeated across the continent. Norway's Morrow Batteries filed for bankruptcy in May 2026, the latest European cell-maker undone by rising capital costs, supply-chain shocks and softening demand.
What survives a collapse like that is rarely the factory. It is the process knowledge — the engineers, the patents, the pilot lines. Refinyx is built almost entirely on that residue: the recovery technology proved at Skellefteå, the pilot lab in Västerås, and the people who ran both.
That is why this relaunch reads as more than one company's story. It is a test of whether Europe's deep tech can turn a failed champion's know-how into a durable business, or whether the region can only watch the technology leave with the founders who wrote it.
How Refinyx scales without owning plants
Refinyx does not intend to own recovery plants. It licenses process technology and engineering to industrial operators who build, own and run the facilities themselves.
That is the opposite of the integrated model that defined Northvolt. It is also leaner: licensing is capital-light, and revenue scales with partners' capacity rather than with Refinyx's balance sheet. The company keeps the R&D and pilot operations in Västerås and sells the blueprint.
The risk sits in the same place as the upside. A licensor is only as good as its partners' willingness to break ground. Refinyx does not control the construction schedule, the financing or the local permits behind any of the plants its process will run in.
Qarlbo's long game
The backer is Qarlbo Energy, which took an active ownership role rather than a passive cheque. Qarlbo Energy is owned by two Swedish industrial holdings: Qarlbo, the vehicle of EQT founder Conni Jonsson, and Thisbe, part of the Wallenberg Foundations' industrial portfolio.
That is patient capital of a specific kind — family-office and foundation money that can sit through a licensing ramp measured in years. Michel Thomas, Qarlbo Energy's chief executive, framed the investment around critical-raw-materials security and a long horizon.
The policy wind is real
The timing is not accidental. In 2026 Washington moved to keep end-of-life battery material inside its borders: a presidential determination gave the Commerce Department authority to restrict exports of black mass and other critical minerals, and the Department of Energy named recycling and refining projects in a $500 million grant round.
Brussels is pushing the same direction through the Critical Raw Materials Act. For a company selling domestic recovery capacity, that tailwind is hard to miss.
The material bottleneck is not new. As we wrote in October about where the storage cost curve goes next, the constraint was never really the cell. It was everything upstream of it.
The cost test
Refinyx is not competing in an empty field. Redwood Materials has spent years industrializing cathode recovery in Nevada. Li-Cycle has fought through its own restructuring to keep its plants running. Peers with built capacity and deeper balance sheets can outspend a startup on tonnes, at least for now.
Refinyx's answer is the sentence that separates it from the sector. It argues the green, subsidy and national-security premiums are temporary, and that only cost, quality and reliability compound over a decade.
If that is right, licensing a proven, solvent-free process beats owning an expensive one. If it is wrong, the company is selling engineering services into a market that wanted a climate story.
The technology is the easy part. The plant in Skellefteå proved it can run. What remains unproven is whether industrial operators — steelmakers, chemical firms, waste handlers — will buy a recovery process when a subsidy-supported rival is one procurement cycle away.
The first real sign
A signed U.S. customer with undisclosed value and undisclosed capacity is a signal, not proof. Refinyx has not named the customer, the contract value or the planned throughput.
The next real test is ground-breaking. Watch for a named partner, a site and a capacity figure in megawatt-hours or tonnes per year. Until those appear, the 134 patents are a story about potential, not production.
Whether Refinyx names its first U.S. customer and publishes contract capacity
Whether Qarlbo Energy's patient capital funds a second pilot or a licensed plant
Whether U.S. black-mass export rules tighten again and lift domestic recovery economics
Whether a subsidy-supported peer undercuts Refinyx on price before its licensing model proves out
Can a recycler win without a rescue price?
Probability: 55% — the technology already ran at industrial scale and the policy direction favors domestic processing, but licensing timelines depend on partners Refinyx does not control.
🟢 Bull case (30%)
Consequence: Refinyx becomes the default process vendor for Western operators building recovery capacity.
🟡 Base case (50%)
Consequence: Refinyx survives as a niche technology supplier, profitable but small.
🔴 Bear case (20%)
Consequence: Refinyx runs out of licensing runway and its patents become an asset sale.
The 134 patents were never the question. Whether anyone pays for a process without a subsidy is.