Alkaline water electrolysis: a century of industrial use, and the two limits that remain

Alkaline water electrolysis is the oldest of the four technologies and still the one with the largest installed capacity. Cells run in a concentrated liquid alkali, typically potassium hydroxide at somewhere between 20 and 30 percent by weight, at temperatures around 60 to 90 °C. Hydroxide ions carry the charge from the cathode to the anode through the liquid.

It got to industrial scale a century ago and it did so with nickel and steel. There is no iridium in an alkaline cell, no platinum, and no fluorinated membrane. That is not a recent achievement in cost engineering. It is the reason the technology existed at all before those materials were available.

Alkaline electrolysis is the proof that water splitting does not require noble metals. Everything the newer technologies do is an attempt to keep that property while fixing the separator.

The separator is a diaphragm, not a membrane

This is the distinction that explains most of the technology's behaviour, and it is routinely blurred in comparison tables that call every separator a membrane.

An alkaline cell separates the two gases with a porous diaphragm. The liquid electrolyte fills those pores and carries the hydroxide ions through them. The diaphragm is a physical barrier against bubbles, not a selective barrier against dissolved gas.

Unfortunately, the diaphragms do not completely prevent the cross-over of the gases nor that some oxygen arrives to the cathode where it can combine with hydrogen to form water.

A dense polymer membrane does the same job differently. There are no open pores, the ions move through the material, and dissolved gas has a much harder path across. That is what PEM and AEM cells use, and it is why they hold a pressure difference between the two sides while an alkaline cell generally cannot.

Once you know that, the two well-known limits of alkaline electrolysis stop being separate facts and become one fact seen twice.

SourcesMolecules 26(21):6326 (2021), DOI 10.3390/molecules26216326: diaphragm crossover in alkaline electrolysis · Polymers 15(9):2144 (2023): KOH concentration and operating temperature of alkaline electrolyzers

Limit one: the turndown floor

Gas crosses any separator at a rate set mostly by the concentration difference across it. That rate does not care how hard the cell is working. Production, on the other hand, is proportional to current.

Reduce the load and the numerator falls while the leak stays roughly where it was. The impurity fraction therefore rises as the machine is turned down, and at some load it reaches the point where the mixture is no longer acceptable and the plant has to stop rather than reduce further.

This allows the intermixing of the produced gases (hydrogen and oxygen – H2 and O2) that are dissolved in the electrolyte, limiting lower power-operating range

This is why the turndown floor belongs in a technical discussion about renewable-powered projects rather than in a footnote. A grid connection at steady base load never approaches it. A solar or wind profile spends a large share of its hours near it.

The behaviour is not unique to alkaline cells. It is a property of electrolysis, and we measure it in our own AEM stacks too. Our HXS-2 stack held hydrogen in oxygen under 2 percent across 49 measured points from 0 to 6 barg, everywhere from 40 to 100 percent of rated load. What differs between technologies is where the floor sits, not whether one exists.

SourceIRENA, Green Hydrogen Cost Reduction (2020), section 2.2: why gas intermixing limits the low-load range and pressure

Limit two: pressure and dynamics

A liquid electrolyte on both sides and a porous separator between them means the two sides must be held at nearly the same pressure. Let one side lead and the electrolyte moves through the diaphragm. Pressurised alkaline systems exist and work, but they pressurise both sides together, which is a different engineering problem from the differential pressure a dense membrane allows.

The liquid inventory also sets the response time. A large volume of hot circulating electrolyte is thermally slow, and the cell chemistry needs that inventory to be at temperature and in composition before the plant can take load.

• Delivery pressure usually needs downstream compression, which is capital cost and parasitic load.
• Ramping is slower than a membrane cell, which matters when the power source is the thing setting the profile.
• The electrolyte itself is a maintenance item, and a concentrated alkali is a handling and materials question through the whole plant.

None of these limits is a reason to dismiss the technology. They are the reasons the two membrane technologies were developed, and they are also the reasons alkaline electrolysis remains the right choice for the projects those limits do not touch.

What alkaline electrolysis is still the right answer for

We build AEM, so this section is written by an interested party and should be read as one. It is still the honest summary.

Alkaline remains the leading technology, with 60% of the additions and the global installed capacity.

The last row is the one that produced everything else in this series. AEM keeps the alkaline chemistry, and with it the freedom from iridium, while replacing the diaphragm with a dense anion exchange membrane. That single substitution is what changes the pressure and turndown behaviour, and it is also where AEM's own unsolved problem lives, because the membrane has to survive the alkaline environment it was put into.

SourcesIEA Global Hydrogen Review 2025, chapter 3: alkaline share of installed electrolysis capacity · U.S. DOE Technical Targets for Liquid Alkaline Electrolysis: status of commercial alkaline technology