What Became of SaltPower's EIC Accelerator Osmotic Power Plant

Salt water that meets fresh water across a membrane releases energy, and the saltier the brine, the more energy there is to take. Saltkraft ApS, a Sønderborg company that trades as SaltPower, spent a decade trying to turn that physics into a business. In the EIC Accelerator pilot's January 2020 cut-off it won an EU grant of €2,411,059.88 towards a €3,444,371.25 project to scale its osmotic power unit from 20 kW to 100 kW and sell it to salt producers, according to CORDIS. The plant was built. It ran at a salt mine owned by the Dutch salt group Nobian. In January 2026 SaltPower stopped development, dismissed its staff and decommissioned the plant, according to the company's website and its 2025 annual accounts filed with the Danish Business Authority.

The company says it concluded its commercial activities "because the economic feasibility was too far in the future", released its patents and published its findings in a white paper so others can continue the work. The trade newsletter Water Desalination Report covered the decision in two 2026 issues, reporting that SaltPower had halted research on the process and exited the PRO market.

Where SaltPower Stands Now

The company still exists on paper. The Danish register lists Saltkraft ApS as active, owned entirely by Saltpower Holding ApS, with Jørgen Mads Clausen on the board, according to FirmaAPI's copy of the CVR record. No liquidation or bankruptcy notice for the company turned up in searches for this article. But the 2025 accounts, adopted on 8 July 2026, say the company "paused its development in January 2026 and will continue without significant activity going forward", and that "employees were dismissed, supplier obligations were settled, and other necessary measures related to the closure were carried out". This article therefore treats SaltPower as dormant: the company has not been dissolved, but by its own account it has had no significant activity since January 2026.

The white paper, written by the science journalist Henrik Bendix for SaltPower and published in 2026, describes how the assets were wound down. The 100 kW plant at Nobian's brine field was decommissioned "so that only the hall remains. Nobian can use that as a tool shed." The small membrane test rig went to the Technical University of Denmark, and the two large test rigs went to Bochum University of Applied Sciences in Germany. SaltPower withdrew from Hy-SaMOE, a follow-on membrane project with DTU and Danfoss funded by Innovation Fund Denmark, and made its whole patent portfolio freely available.

The Technology: Pressure Retarded Osmosis From Brine

SaltPower used pressure retarded osmosis (PRO). Concentrated brine and fresh water flow on either side of a semi-permeable membrane. Water is drawn into the brine, which is held under pressure; the extra volume of pressurised water can drive a turbine or, as SaltPower later chose, do the work of a pump directly, according to the CORDIS results summary and Nobian. Unlike wind and solar, the output does not depend on the weather.

The idea was not new. Norway's state utility Statkraft opened an osmotic pilot plant on the Oslo fjord in 2009 using seawater and closed it in 2012 after reaching about 1 W per square metre of membrane, far below the 5 W/m² usually cited as the threshold for economic operation, the white paper recounts. SaltPower's bet was salinity. Clausen noticed in 2015 that water from the geothermal well at Sønderborg held 16% salt, 4.5 times as much as seawater, and saturated brine from salt mining holds about 26%. Nobian says the brine it pumps is about eight times saltier than seawater, which makes the economics "much better" than with seawater. Ingeniøren's readers voted the Sønderborg pilot plant Denmark's engineering achievement of 2016, with 36% of 4,830 votes.

Before the EIC grant the company relied on Danish money. The Energy Technology Development and Demonstration Programme (EUDP) funded three projects, the last of them, on control methods and components, with DKK 7.88 million of a DKK 15.28 million budget from 2019 to 2021, with Semco Maritime, Dansk Salt, IDE Technologies and Sønderborg Fjernvarme as partners, according to EUDP. That project's own description records the shift from geothermal plants to salt production that defined the EIC proposal. Not every plan worked: a 2019 demonstration at a gas storage cavern at Lille Torup did not go ahead because the pipe that was to deliver the brine had been removed, the company's then technical director wrote under an Ingeniøren article in October 2019.

What the EIC Project Promised, and What CORDIS Shows

The European Commission's list of the January 2020 EIC Accelerator pilot winners names Saltkraft ApS of Sønderborg as the only company in the SaltPower project, grant only. In that list the website column is shifted: SaltPower's address appears on the row of another Danish winner, Radiobotics. The grant agreement was signed on 17 April 2020, and the project ran from 1 May 2020 to 30 April 2023 under the SME Instrument phase 2 funding scheme. CORDIS lists no other EU project for the company.

The objective on CORDIS was specific. SaltPower would move from a mobile 20 kW unit to a 100 kW unit "as a direct result of this EIC project", produce electricity at €0.021 per kWh, later as low as €0.01, and target a turnover of €41 million by 2025 by selling to salt producers. It spoke of "several thousand plants" in Europe.

The final periodic report, covering May 2021 to April 2023, says the company "has built a commercial approximately 100 kW modularized unit", doubled its organisation, set up an engineering department and filed five patent applications. It says the plant operates at a salt production plant in Mariager. Nobian and the white paper place it elsewhere: at Nobian's brine field at Hvornum near Hobro, after a prototype test at the Mariager salt factory in 2021. The white paper also shows that the plant was not running when the project ended: components arrived between autumn 2022 and spring 2023, the plant was assembled in summer 2023, and it was ready to run osmosis from September 2023. Nobian dates the start of operation to 2023.

1. Scale up from 20 kW to a 100 kW unitDeliveredAbout 90 kW gross output from 100 Toyobo membrane modules, per the white paper; Nobian confirms a 100 kW plant from 2023.
2. On time, within the project periodLateReady to run from September 2023, five months after the project ended in April 2023.
3. Electricity at €0.021 per kWhNot achievedThe white paper puts the demonstration plant at about €50/MWh and says the commercial plant's numbers "do not add up".
4. Sales to salt producers, €41M turnover by 2025Not achievedNo second plant sold; the accounts show a gross loss of DKK 5.38M for 2025.
5. 1 MW modulesNot achievedAnnounced as work in progress in 2023; none built before the company stopped.

The delivery was real. CORDIS's 2023 results summary calls it "the world's first industrial energy production plant based on osmotic energy", and said the company had "already identified many customers, especially in Europe and the United States", quoting its sales director, Jesper Culmsee, on the plant's 24/7 baseload output. The commercial targets in the proposal were not met, and nothing in the record suggests they came close.

What Happened at the Salt Mine

The plant drew about 35 cubic metres of saturated brine per hour from one of Nobian's wells, pushed it to 70 bar and pumped the diluted brine back into the salt dome, replacing part of the work of the electric injection pump, according to the white paper. Nobian described the same purpose: to cut the energy demand of solution mining and, it hoped, make its brine field "energy neutral".

Three problems came out of operation, all described in the company's own account. Pre-treating the fresh water with a large reverse osmosis unit took close to 50 kW, just over half of what the osmosis produced; in early 2025 the plant used about 23 kW less than the pump it replaced. The membranes delivered about 2 W per square metre over 45,000 square metres, below the 5 W/m² that short laboratory runs reach. And the hollow fibre membranes cost about €4,500 per module plus €2,000 per pressure housing, which led SaltPower to conclude that the architecture "cannot be used for PRO with hypersaline water". It found no commercial thin-film membrane that could take the pressure, and said it had "neither the expertise nor the financial resources" to develop one.

The Money

Saltkraft has never had meaningful revenue, and its accounts show the owner paying the bills. In the 2023 accounts the company changed its accounting policy and stopped capitalising development costs, which removed DKK 51.4 million of development projects from the balance sheet and cut the restated 2022 equity from DKK 26.1 million to DKK 5.6 million, according to the filings with the Danish Business Authority. The 2021 and 2022 accounts had already flagged uncertainty over the value of those assets.

From then on the losses were visible: DKK 15.7 million in 2023, DKK 16.5 million in 2024 and DKK 11.5 million in 2025. The group covered them with capital contributions of DKK 16.3 million, DKK 14.4 million and DKK 12.6 million, and every set of accounts since 2023 rests its going-concern assumption on a support letter from the company's ultimate owner. At the end of 2025 equity stood at DKK 5.19 million and cash at DKK 1.52 million.

No outside investor appears in the register or in the company's statements. SaltPower is not in the EIC Fund portfolio, which is consistent with a grant-only award. Besides the EIC grant, the white paper lists Danish public funding through EUDP, Innovation Fund Denmark's Innobooster programme and two Innovation Fund Denmark research projects with DTU and Danfoss; how much of the research grants went to SaltPower itself is not stated.

Why Europe Should Care

Salinity gradient power is part of the EU's ocean energy plans. The Commission's offshore renewable energy strategy of 19 November 2020 aims for at least 1 GW of ocean energy by 2030 and 40 GW by 2050, and the Commission lists salinity gradient among the sources, alongside waves and tides, according to the European Commission and A&L Goodbody. SaltPower also pitched its units for the salt caverns that Europe will need to store hydrogen, where they would generate power while the caverns are washed out, according to the CORDIS report.

SaltPower's record is useful to that policy precisely because it failed honestly. It is one of very few attempts to run PRO at industrial scale, and its white paper sets out the numbers that decided the outcome: power density, pre-treatment energy, membrane prices and payback. The white paper notes that a 110 kW PRO plant opened in Fukuoka, Japan, on 5 August 2025, with no public operating data at the time of writing.

The Verdict

SaltPower has not been asked for comment for this article. It is based entirely on public records, company statements and EU project data, linked throughout.