Volkswagen Group to tweet bits from its new sustainability report; role of EVs in China
Opinion: Have Natural Gas Prices Bottomed?

EPFL team develops effective membrane-less electrolysis process for H2 production; potential to outperform conventional designs

Researchers at EPFL in Switzerland have developed a system for producing hydrogen through a simplified membrane-less water electrolysis process. By working with the balance between fluid mechanic forces, the researchers eliminated the expensive membrane that sits between the electrodes in conventional electrolysis systems.

The membrane-less approach demonstrates for the first time an electrolyzer capable of operating robustly and continuously with various catalysts and electrolytes across the pH scale, while at the same time generating hydrogen gas streams the oxygen content of which is well below the safety limit. An open access paper on their discovery is published in the RSC journal Energy and Environmental Science.

Schematic diagram of membrane-less electrolysis. Two parallel plates are coated with hydrogen and oxygen evolution catalysts respectively and are separated by less than few hundreds of micrometers. The electrolyte flows between the catalyst plates and the evolved gases move close to the corresponding catalyst surface due to the Segré–Silberberg effect. Each of the product gas streams is collected in dedicated outlets. Stacks of these planes in horizontal can be used for higher throughput. Source: Hashemi et al. Click to enlarge.

In a conventional system, two electrodes are submerged in water and separated by a polymer membrane. An electric current is sent through one of the electrodes (the cathode) and then travels to the other (the anode). The current, with the help of a catalyst, causes the water molecules to break apart into oxygen and hydrogen. To prevent the two gases from mixing together and making an explosive mixture, polymer membranes are implemented between the catalysts to keep the gases separated.

In both research and industry, membranes used for ionic conductivity are most commonly made of Nafion, due to its great stability and ion conductivity. However, they are expensive, have a limited lifetime and only work in highly acidic solutions, which limits the choice of catalysts.

To rid themselves of these constraints, the scientists placed the electrodes less than a few hundreds micrometres apart in a microfluidic device. When the liquid moves above a certain speed between the electrodes, the gases are pushed in opposite directions—due to the lift forces caused by an effect known as Segré-Silberberg effect—without the need for a membrane to guide them into separate outlets.

Our device has the potential to surpass the performance of a similar water-splitting apparatus that relies on an ion conductive membrane. This is due to the higher ion conduction in liquid electrolytes than in common solid membranes.

—Mohammad Hashemi, first author

The devices presented in the paper split water at current densities above 300 mA cm−2, with more than 42% power conversion efficiency. Crossover of hydrogen gas into the oxidation side was as low as 0.4%, leading to a non-flammable and continuous hydrogen fuel stream.

Additionally, the ability to use buffered electrolytes allows for the incorporation of earth-abundant catalysts that can only operate at moderate to high pH values.

The membrane-less electrolyzer demonstrated here has the ability to produce non-flammable hydrogen streams, continuously and stably across the pH scale. Comparing the device ohmic resistance with that of Nafion-based devices, it is clear that this device has the potential to surpass the performance of a similar water splitting apparatus that rely on ion conductive membranes for separation. Although a single electrode pair, such as the one in this proof of concept study, can only produce a limited amount of fuel, scaling it out can be achieved on multi-stack panels for enhanced throughput49 or the implementation of large area electrodes.

As the only dimension to be kept small is the inter-electrode distance, follow up studies are underway to develop high throughput devices where high surface area planar electrodes are used as the side walls of narrow electrolyte channels. Additionally, it is worth noting that this electrolyzer platform may be used in reverse as a fuel cell with two streams of the electrolyte, each saturated with H2 or O2, allowing for the production of electricity. The design simplicity of this membrane-less electrolyzer can facilitate mass production, especially by employing high resolution 3D printers or injection moulding techniques.

—Hashemi et al.

This design sets the stage for the production of devices that work with all types of liquid electrolytes (containing ions) or catalysts, since there is no longer the risk of damaging the components due to a highly acidic environment. This versatility is not possible in conventional systems, in which only catalysts made of noble metals such as platinum can work with the low pH values imposed by the membrane.

The same team is now working on scaling up the design for higher production rates. As the only dimension that needs to remain small is the inter-electrodes distance, it is possible to implement the same concept using high surface area electrodes as side walls of narrow electrolyte channels.


  • S. Mohammad H. Hashemi, Miguel A. Modestino and Demetri Psaltis (2015) “A membrane-less electrolyzer for hydrogen production across the pH scale” Energy Environ. Sci. doi: 10.1039/C5EE00083A

  • Segre G, Silberberg A (1961) “Radial particle displacements in Poiseuille flow of suspensions”. Nature 189: 209–210 doi: 10.1038/189209a0



This could lead to a combined effective electrolizer-FC to convert excess RE into H2 and back to obtain 24/7 clean energy (if the price is right?)

Some of the H2 produced could be compressed for FCEVs.


42% power conversion efficiency doesn't sound good enough.
Norsk Hydro claim 80% for their alkaline electrolysers, but you never quite know the basis on which the calculations are run and presumably this will improve as it is simply a prototype.


If 'wasted' or 'unused' RE can be converted and stored for extended periods with an efficiency of 40% to 80%, it would be better than not doing it. This free energy would serve to supply clean energy 24/7 with FCs and clean compressed H2 for FCEVs.

The would LAO apply to surplus Hydro during rainy season (when reservoirs are full and a lot of water has to be bypassed) and during low consumption hours.

Bob Wallace

I question whether we will see any appreciable amount of 'wasted' or 'unused' RE in the next 15 to 20 years.

Right now there is a lot of fossil fuel generation that can be turned off, hydro that can be withheld, and some pump-up hydro storage that can soak up any surplus. Additionally, more EVs and PHEVs will be coming on line and they can serve as dispatchable loads.

As the price of storage drops we are likely to see coal and some nuclear plants shut down in the spring and fall when demand is lower. It will be cheaper to use wind and solar direct and fill in with hydro, NG and stored wind/solar.


With variable loads (15% to 95%) it is not always possible to use energy production facilities anywhere close to their name plate capability.

Even NPPs, built to handle base loads, are not fully loaded/used during many hours/year.

Wind and Solar facilities do not always produce when needed and visa versa.

As more and more NGPPs, Solar and Wind are started, more surpluses will happen during off peak demands and will become available for H2 production at much lower cost

The comments to this entry are closed.