What water electrolysis is, and the four numbers that describe any electrolyzer

Water electrolysis passes direct current through water and splits it into hydrogen and oxygen. That is the whole reaction, and it has been understood since the beginning of the nineteenth century. What has changed since then is not the chemistry but the cell it happens in.

This is useful to a buyer for one reason. Because the reaction is fixed, several of the numbers on an electrolyzer datasheet are not the supplier's to choose. They are thermodynamic and they are the same for everyone. Knowing which numbers those are tells you where a comparison between two quotations can be real, and where it cannot.

No supplier competes on the energy, charge or water that the reaction itself requires. They compete on the distance between those floors and what the machine actually consumes.

The reaction, and where the two gases come from

Two half reactions run on opposite sides of the cell. On the cathode, water takes electrons and becomes hydrogen. On the anode, water gives up electrons and becomes oxygen. Between them sits a separator that lets the charge carrier through and keeps the two gases apart.

The oxygen half is the slow one. It carries most of the kinetic loss in every technology, which is why an anode is where a materials team spends its time and why the anode is usually the first component a supplier will talk about.

Which ion carries the charge across the middle is the single choice that produces the different technologies. A liquid alkaline cell and an anion exchange membrane cell move hydroxide ions from cathode to anode. A PEM cell moves protons the other way. A solid oxide cell moves oxide ions at high temperature. Everything else about the four technologies follows from that one decision.

The four fixed numbers

For one kilogram of hydrogen, four quantities are set by physics rather than by engineering. They are worth memorising, because they are the only benchmark against which a quoted specification can be judged.

The energy figure is the one most often misread. An electrolyzer quoted at 39.4 kWh per kilogram would be a perfect machine, so any real number is higher, and the size of the gap is the efficiency. Quoting that gap against the lower heating value instead makes the same machine look about 18 percent better, which is why the basis has to travel with the number.

The charge figure explains why stacks are built as they are. Current, not voltage, sets the production rate, and a cell can only carry so much current per unit of area. More hydrogen therefore means more area, more cells in series, or a higher current density with the extra losses that come with it.

The water figure is worth keeping separate from the water a system actually consumes. Our HXB-V1 2 kW system draws about 0.55 litres an hour of feed water at rated output, and the electrolyte circulating through the stack is a much larger flow that is not consumed at all.

Voltage is where the technologies differ

A cell has two thermodynamic marks on it. The reversible voltage at room temperature is 1.23 V, below which the reaction does not proceed. The thermoneutral voltage is 1.48 V, at which the reaction supplies exactly its own heat. Neither can be moved by a catalyst, a membrane or a supplier.

Every real cell runs above 1.48 V, and the excess is what the industry competes on. It divides into the overpotentials of the two half reactions and the ohmic losses of the membrane, the contacts and the electrolyte.

Two of our own measurements make the point concrete. Our HXS-2 stack reaches 88 percent efficiency on a higher heating value basis at beginning of life, at 0.3 M KOH and 60 °C, which corresponds to about 45 kWh per kilogram at the stack boundary. The complete HXB-V1 system around that stack consumes about 54 kWh per kilogram, because the rectifier, the pump, the separators and the dryer all draw power that never reaches the cell.

Those two figures come from the same machine on the same day. Neither is wrong. They answer different questions, and a comparison that mixes the two is not a comparison at all.

The four technologies in one paragraph each

Those five rows explain most of what separates the four. The rest of the difference is maturity, and it runs in the opposite direction to novelty: alkaline has been industrial for over a century, PEM is widely deployed, AEM became commercial in the 2020s, and solid oxide is at the demonstration and early commercial stage.

What to ask, before any of this matters

• The efficiency, with its basis and its boundary. Higher or lower heating value, and cell, stack or system.
• The current density and temperature it was measured at. Any cell looks better run softer.
• The outlet pressure and purity from the stack itself, before a compressor or a dryer is added.
• The turndown floor. Not the ramp rate, but the load below which the machine has to stop rather than reduce.
• The durability, with the load and the voltage the run ended at. Hours alone describe nothing.

For our own products those answers are published rather than available on request. The HXS-2 has a continuous run of 2,281 hours at 50 A galvanostatic, 0.3 M KOH, 40 °C and ambient pressure, degrading at 80.2 µV/h per cell on a linear fit across the full run.