The membrane electrode assembly is the layer stack where water is actually split. It sets the efficiency ceiling of everything built around it, and most of its performance is decided at the interfaces rather than by the ingredients.
Every specification an electrolyzer carries, its efficiency, its current density, its degradation rate, is set first by a component a fraction of a millimetre thick. The membrane electrode assembly, or MEA, is the membrane with a catalyst layer bonded to each face. Everything else in the machine exists to give the MEA current, water and a way out for the gas.
An electrolyzer can be no better than the MEAs inside it, and a stack built around a mediocre MEA cannot be engineered back to a good number.
What each layer does
The membrane conducts ions and keeps the two gases apart. In an AEM cell it carries hydroxide from the cathode to the anode, and it has two competing requirements: thin enough that ionic resistance stays low, tough enough to seal against a pressure difference and survive years of alkaline conditions.
The anode catalyst layer runs the oxygen evolution reaction. This is the slow half of water splitting and the reason most catalyst research in the field is on the oxygen side. In alkaline conditions it can be done with nickel-iron catalysts rather than iridium, which is the single largest cost difference between AEM and PEM. The cathode side runs hydrogen evolution, which is kinetically faster, so it is usually not what limits the cell.
The ionomer is easy to overlook because it is not a layer of its own. It is the polymer inside the catalyst layers that carries hydroxide the last distance to the catalyst particles, and it is a genuine design variable: too little and much of the catalyst is electrochemically unreachable, too much and gas cannot escape the layer.
Two ways to put catalyst where it is needed
| Coated | Grown in place | |
|---|---|---|
| What happens | Catalyst powder is milled into an ink with ionomer and solvent, then coated onto the membrane or the porous transport layer | The catalyst is synthesised directly on the metal support, with no binder between catalyst and substrate |
| What decides the result | The inkdispersion, ionomer ratio, solvent, drying profile | Growth conditions and the substratefewer interfaces to go wrong |
| How it scales | With the coating line | With the size of the electrode |
| Where we sit | Not our route | NiFe-LDH grown on nickel foam |
Neither route is inherently better, and both are used in commercial products. The question to put to a supplier is not which one they use, but how tightly they can repeat it. Two laboratories with the same catalyst powder and different ink recipes will not report the same performance.
What separates a good MEA from a bad one
- Interfaces decide more than ingredients. A strong catalyst on a poor interface performs worse than a modest catalyst on a good one, because the loss is not in the catalyst's activity but in getting electrons, ions and gas to and from it.
- Ionomer loading is a design variable and not a detail. The same catalyst gives measurably different curves at different loadings, which is one reason published catalyst results are hard to compare across groups.
- Reproducibility tells you about the process. The spread across a batch is more informative than the best cell in it.
- Conditions travel with the number. An MEA measured at 80 °C in concentrated KOH will not repeat that figure at 40 °C in 0.3 M KOH.
Ask a supplier for the spread across a batch, not the best cell in it.
Why we make our own
MEA performance is decided at the boundary between the membrane, the catalyst layer and the porous transport layer, and those boundaries only exist once the parts are assembled and compressed together. A membrane that performs well in a half-cell can behave differently under compression against a particular electrode structure.
So we develop the catalysts and electrodes, synthesise the ionomer, and design the cell and stack they go into. That is not a claim that integration is the only way to build an electrolyzer. It is the reason we can change one of those layers without waiting for three suppliers to agree, and the reason we can publish the conditions behind our numbers rather than repeat someone else's.
Three founders from a KAIST hydrogen laboratory started the company in December 2023. More than 60 customers in 12 countries now use our electrodes, stacks and systems, from university laboratories evaluating their own materials to companies producing hydrogen on site.
- What is an electrolyzer stack→Where the MEA sits, and the three mechanical decisions that decide how much of it survives assembly.
- What changes when the membrane is anionic→Why an anion exchange membrane lets the catalyst leave the noble metal market.
- HXP-an and HXP-ca electrodes→The electrodes themselves, with datasheets.
Looking at AEM for a specific application?