Seas and oceans are brimming with untapped potential for producing energy. Two EPFL research groups are developing nanoscale devices for converting saltwater into electricity.
Rising tides and rolling waves release massive amounts of energy from the seas and oceans every day. Attempts to harness this energy are nothing new: for decades, engineers have been building tidal power plants, marine turbines and wave energy systems in order to use the mechanics of water to produce electricity. But take-up of such technology has been marginal. According to the International Renewable Energy Agency, the installed capacity of marine energy production worldwide is around 500 megawatts. For comparison, the equivalent figure for wind power is 1,100 gigawatts - or 2,200 times higher.
There are also other, less evident ways to produce energy from water: one is the salinity difference between saltwater and freshwater, and another is the evaporation that occurs at the surface of seas and lakes. Such natural phenomena are continuous and can be found everywhere. Researchers have long known how to describe them using physics - but concrete applications are thin on the ground. At EPFL, two research groups, each with its own approach, have taken major steps forward in developing systems for generating renewable marine energy at a nanometric scale.
Sliding ions
Osmotic power, also known as blue energy, is one of the promising options being explored. It can be produced anywhere freshwater comes into contact with saltwater, such as where a river pours into an ocean. In theory, osmotic power has huge potential - over 15,000 TWh per year, or 75% of total global energy use. But in practice, formidable technical constraints mean that only around 600-2,000 TWh per year of power can be harnessed, or between France's annual electricity demand and that of India. The method for producing osmotic power has been around for decades. It entails separating freshwater and saltwater by a selective membrane and then, as ions naturally pass from the solution with a high salt concentration to that with a lower one, taking advantage of the voltage difference. The problem is that, with existing membranes, engineers must choose between speed and precision: either the ions pass rapidly but in a haphazard way, or they pass slowly but with good selectivity.
The research group led by Aleksandra Radenovic at EPFL's Laboratory of Nanoscale Biology (LBEN) has found an elegant way to square this circle. In a study appearing in Nature Energy in February 2026, the scientists developed a device made of hafnium oxide nanopores that have been coated with a lipid biofilm - the same lipid molecules as those in our cell membranes. Water molecules are attracted to this biofilm and form a layer only a few molecules thick. This layer serves as a lubricant, enabling ions to slide through the pores quickly while maintaining a high degree of selectivity. As a result, the device containing 1,000 nanopores has a power density of around 15 watts/m², which is two to three times higher than that achieved by existing polymer systems.
These systems already use evaporation to turn saltwater into freshwater. So we can envisage creating a system that generates both drinking water and electrical power.
The next challenge will be to scale up the technology. "We need to move from chips in a lab to large membranes with a high pore density, precise pore-size controls and a reliable coating - all at a reasonable cost," says Yunfei Teng, a doctoral assistant at LBEN. The team must also demonstrate the device's long-term stability and resistance to biofouling. However, the underlying concept can be applied universally. Such water-based lubrication could also improve the performance of membranes used for desalination and medical devices.
Evaporation as a continuous energy source
The research group led by Giulia Tagliabue at EPFL's Laboratory of Nanoscience for Energy Technologies (LNET) is exploring an even more surprising energy source: evaporation. Their method uses the hydrovoltaic effect, which is based on a simple mechanism: when liquid evaporates at the surface of a nanodevice, that creates a continuous flow of liquid in the nanochannels - similar to the transpiration that causes sap to rise in a tree. If an electric charge is applied to the nanochannel walls, the ions start to move and generate an electric current. In an article published in Device in 2024, the LNET team reported an unexpected discovery: contrary to what scientists had previously thought, this process works with saltwater and not just purified water. That paves the way to concrete applications using evaporation at the ocean surface.
The team made another breakthrough in 2026, in a study published in Nature Communications. They developed a silicon nanodevice composed of three layers: one for evaporation, one for ion transport and one for collecting the electrical charge. The device uses sunlight and heat to not just speed evaporation, but also directly control the movement of ions and electrons. "This natural process has always occurred, but we're the first to harness it," says Tagliabue. Thanks to the device's dual approach, it can generate five times more energy than systems employing evaporation alone.
One big advantage of LNET's technology is that it can run continuously. Unlike the solar radiation needed for solar power, evaporation occurs day and night over a broad range of temperatures. The nanodevice can therefore be employed wherever there's water, heat and light. For example, it can be used to run self-powered environmental sensors. Tarique Anwar, a doctoral assistant at LNET, also points to potential synergies with desalination systems. "These systems already use evaporation to turn saltwater into freshwater. So we can envisage creating a system that generates both drinking water and electrical power."
Being able to produce both of these would bring massive benefits to coastal areas with little access to freshwater and electricity. For now, nanofluidic devices can't compete with major power generators, but their ability to take advantage of continuous, ubiquitous phenomena at a small scale opens up exciting prospects.
