Gas Fraction in an Alkaline Water Electrolysis Cell

Abstract

Renewable hydrogen from alkaline water electrolysis (AWE) is central to a low-carbon energy system, yet the efficiency of AWE cells is limited by the gas they produce. Bubbles formed at the electrodes raise the gas fraction of the electrolyte, which lowers its ionic conductivity and adds an ohmic loss that grows as cells are driven to higher current densities. This gas is not spread evenly through the cell but is concentrated in a thin layer near the electrode, where a balance between how quickly gas is produced and how quickly it is carried away sets the build-up, and that region is difficult to reach experimentally: intrusive probes and optical imaging cannot resolve it, and the near-wall profile of a classical cell has not been measured.

This thesis measures the steady-state gas fraction in a classical alkaline water electrolyser using X-ray imaging. The cell was purpose-built for the experiment: a wide inter-electrode gap of 114 mm filled with 6 M KOH, with no forced flow, so that the near-wall layer develops unconfined. Time-averaged radiographs are converted through the Beer–Lambert law into a two-dimensional gas-fraction field ε(x, y), at a pixel size of 45 μm, for current densities j spanning from 36.5 A m−2 to 2434 A m−2, and for both the hydrogen and the oxygen electrode.

The measured fields show a thin near-wall layer that decays into an almost gas-free bulk. The peak gas fraction is set by the current density, growing roughly as √j to about 0.2 at the highest current, while the width of the layer is set dominantly by height, broadening steadily at the highest current from about 0.3 mm at the bottom to 1.5 mm at the top of the 87 mm of electrode height captured in the field of view, and changing only weakly with current. It is therefore the current that fixes how much gas the layer holds at its peak, and the vertical length of the electrode that fixes how far that gas spreads horizontally. At equal current density the hydrogen side carries about twice the near-wall gas fraction of the oxygen side, in line with the reaction stoichiometry; matched for gas production, the hydrogen peak is the higher and the oxygen layer the wider, so the two sides hold much the same near-wall gas but arrange it differently. Only the hydrogen side develops a dispersed-gas background, which builds up towards the top of the cell to a gas fraction of about 0.026 at the highest current, whereas the oxygen side stays close to the detection floor.

Interpreted via a one-dimensional production–removal balance, the profiles imply an apparent lateral dispersion coefficient of order 10−7 m2 s−1. It is not a material constant: it grows with both current and height and varies by over an order of magnitude, and is best read as an apparent, lumped quantity. A likely, but unproven, reading is that a crowding of gas bubbles, a solid pressure, caps the near-wall peak while dispersion sets the width of the layer.

The experimental data from this study thus provide an unconfined, wall-resolved reference for the gas fraction in an alkaline electrolyser, useful for reasoning about the physics of the near-wall layer, for validating transport models and for cell design, though transfer to pressurised, forced-flow or zero-gap cells needs further validation.