Every clean energy source comes with a requirement attached. Solar energy stops when the sun sets. The wind stops when the air stops. Even hydropower depends on what the season decides to send down the river.
The only thing the global network really lacks is a clean source that simply runs all night, during a storm, without asking permission from the weather. Since August 2025, a facility inside a desalination plant on Japan’s southern coast has been doing just that, drawing electricity from the gap between freshwater and seawater continuously, around the clock. It is the first osmotic power plant in Asia, and only the second of its kind operating anywhere in the world, that does not burn a single gram of fuel.
The science behind Japan’s 24/7 power plant that runs on waste streams
The physics behind the plant is the same quiet force that allows a tree to draw water through its roots. Put fresh water on one side of the semi-permeable membrane and salt water on the other, and the fresh water will rush through to dilute the salt because nature cannot tolerate a difference in concentration without resolution. Do this inside a sealed pressure chamber, and the volume will rise on the salty side, increasing the pressure.
Connect that pressure through a turbine and you get electricity made from nothing but the difference between two types of water.The technical name is pressure retarded osmosis, or PRO. A 2024 study in Chemical Engineering Science describes new membrane modifications that advance this process, specifically for generating sustainable energy from salinity gradients, a fundamental engineering challenge that has prevented PRO from expanding commercially for decades.
The standard seawater-to-freshwater setup requires a pressure difference of about 26 bars, roughly equivalent to the pressure at the bottom of a 270-meter column of water. Whatever the plant generates must bear the energy costs of pumping both streams and pushing the water through the membranes.
What comes out on the other side is what remains after those losses.The Fukuoka plant, located at the Uminonakamichi Nata Seawater Desalination Center, was officially commissioned on August 5, 2025.
What makes it more efficient than direct seawater installation is what feeds the brine side, not regular seawater, but concentrated brine, which is the salty waste that a desalination plant typically disposes of after stripping fresh water. On the other hand, treated wastewater from a nearby sewage plant.
Two waste currents that existing infrastructure was already producing pass through a membrane, and the output is energy. The Japanese government points out that using this highly salty brine widens the salinity gradient and extracts more available energy from the process than regular seawater would allow.
What the factory actually produces and why it matters to be honest about the numbers
The expected annual production is about 880 thousand kilowatt-hours per year, which is enough to cover part of the desalination plant’s electricity consumption, in addition to the energy needed for 220 to 300 average Japanese families. That’s a modest number by any network-wide standard, and the people who built it didn’t pretend otherwise.What the output has, which solar and wind cannot buy, is near-perfect reliability. Operators estimate the station’s utilization rate is around 90%, meaning it is fully operational regardless of cloud cover, wind speed, or time of day. A technical and economic analysis of PRO published in Frontiers in Energy Research confirmed that integrating PRO with desalination plants represents one of the most commercially viable configurations of this technology, precisely because the brine waste stream is already produced at no additional cost.
The energy generated goes directly back to drinking water production in Fukuoka, making the desalination process less expensive.Kenji Hirokawa, who heads the seawater desalination center, called it a modest first step rather than a final answer. This framework is accurate, and is the right level of expectation for a technology that is still proving itself on a large scale.
Norway tried it first and shut it down in 2014
The Japanese facility is not the first time anyone has attempted to build a working osmotic power plant.
The concept was first proposed by an American researcher in 1976 in the journal Membrane Science, and it took more than three decades before serious devices appeared. Norwegian company Statkraft opened the world’s first PRO prototype at Tofte on the Oslo Fjord in November 2009. It is designed to have a capacity of 10 kilowatts, and in practice generates between 2 and 4.
This concept worked. And the economy didn’t do that.By January 2014, Statcraft shut down the project, saying it could not make the membranes efficient enough to compete commercially and would leave the work to others. The main problem was power density. Research in this area has established that about 5 watts per square meter of membrane is the approximate threshold at which osmotic capacity begins to make financial sense, a number cited in peer-reviewed analysis, including work published in ACS ES&T Engineering. The Statcraft plant was operating at 1 to 3 watts per square metre.
This gap between what the chemistry promises and what the membrane delivers is what has stranded the technology for a decade.The Japanese approach used a mixture of saltwater and wastewater to expand the salinity differential enough to extract meaningful outputs from available membrane technology, avoiding the need to fully solve the membrane cost problem before building something real. It’s a practical engineering decision: using freely available on-site inputs rather than waiting for membrane penetration that has been coming for fifty years.