PEM Water Electrolysis at Elevated Temperatures

Global warming and the accelerating depletion of fossil based fuels have catalysed a tremendous surge in the development of alternative and sustainable energy sources e.g. wind-, solar- and hydropower. Common for most of these alternative energy sources is that they at times provide more power than needed and hence it has become acute to be able to store the energy.

Hydrogen has been identified as a suitable energy carrier and water electrolysis is one way to produce it in a sustainable and environmentally friendly way. In this thesis an introduction to the subject (chapter 1) is given followed by a literature review of the field of water electrolysis (chapter 2), with a focus on proton exchange membrane (PEM) electrolysis.

In chapter 3 a short description of the experimental techniques used for synthesis of catalyst and characterisation of the components in the electrolysis cell is given. This is followed in chapter 4 by a description of the electrolysis setups and electrolysis cells used during the work. Two different setups were used, one operating at atmospheric pressure and another that could operate at elevated pressure so that liquid water electrolysis could be performed at temperature above 100 °C.

It was found that the gas diffusion layer on the anode side played an important role for the electrolyser performance. Different thicknesses and types, i.e. a single layer- and double layer type, were tested. Chapter 5 presents a characterisation of the gas diffusion layers, using parameters such as porosity and resistance which were supported by images acquired using scanning electron microscopy (SEM).

In chapters 6 and 7 the results of the steam electrolysis and pressurised water electrolysis, respectively, are presented and discussed. The steam electrolysis was tested at 130 °C and atmospheric pressure, whereas the pressurised water electrolysis was performed at 120 °C and 3 bar. For the steam electrolysis three different electrolytes were used.

Chapter 6 is divided into subchapters in which the results are presented and discussed before a comparison between them is given. First phosphoric acid doped membranes of polybenzimidazole - poly[2,2'(m-phenylene)-5,5'-bibenzimidazole] (PBI) were used as electrolyte. Reasonably good short-term results were achieved using this membrane reaching a current density of 775 mA·cm-2 at a cell voltage of 1.84 V.

The durability of phosphoric acid doped PBI however was quite poor only lasting few hours in the setup used. Afterwards a range of different phosphoric acid doped commercial Nafion® and ternary phosphoric acid doped composite Nafion® membranes were tested. The performance was not as good as for the PBI system; only 310 mA·cm-2 at 1.7V for the best ternary composite membrane.

The durability, on the other hand, was greatly improved and test was run for approximately 70 hours (constant voltage of 1.7 V) with a 0.17 mA·h-1 decline in performance. Finally, a new class of perfluorosulfonic acid (PFSA) membranes were tested. This was Aquivion™, which is a short side chained PFSA membrane.

Aquivion™ was also tested doped with phosphoric acid. It showed better mechanical properties, larger uptake of phosphoric acid and hence improved performance. The best result achieved was 775 mA·cm-2 at 1.8 V. Aquivion™ also showed quite promising durability features running for approximately 760 hours (constant current density of 400 mA·cm-2) with a 0.023-0.04 mV·h-1 decline in performance over the last 660 hours.

For the pressurised water electrolysis the best result obtained was for an Aquivion™ membrane with a current density of 2125 mA·cm-2 at 1.85 V. An attempt was made to quantify the significance of various parameters such as membrane electrode assembly (MEA) technique (chapter 6), binder content in anode (chapter 6), anode catalyst loading (chapter 6), gas diffusion layer (chapter 5 and 6) and flow patterns (chapter 4, 6 and 7).

The different machined flow patterns used are described in chapter 4, and the results with the different patterns are shown and discussed in chapters 6 and 7. Chapter 8 is devoted to a general discussion and the conclusions and outlook are given in chapter 9. It seems obvious that further effort should be put into characterisation and development of a more sophisticated anode electrode structure.

Only when this parameter is optimised and performing at a high level, will it be possible to really quantify the importance of the other parameters under full single cell electrolysis tests.