PEM Water Electrolysis - New Approaches Towards Catalyst Separation, Recovery, and Recycling

In the following years, polymer electrolyte membrane (PEM) water electrolysis may well be reaching the gigawatt magnitude of hydrogen production, in order to attend the energy storage demand of intermittent power sources such as wind and solar[1]. At its end-of-life, each electrolysis system will have a redundant stack, expecting tough recycling and re-use requirements for this efficient but rather expensive energy storage system. Practical application of PEM electrolysis already strongly depends on the resolution of these issues.

The optimum strategy will require stack dismantling and separation of the major components. Steel, aluminum, and titanium components shall be processed using the general recycling stream, but the membrane electrode assembly (MEA) and bipolar plates will require a specialized recycling process. The most common approach is to shred the used MEA, dissolve and recover the membrane, burn off the carbon and organic additives, and recycle platinum group metals (PGMs) catalysts using solvent extraction.

Platinum and iridium can be in principle infinitely recycled. In practice, however, recycling of precious electrocatalysts used in PEM water electrolysis is often inefficient or essentially nonexistent because of limits imposed by component design, unavailable recycling technologies, low yield, and the thermodynamics of separation. In any case, the recycling of these precious metals can be promoted if beneficial actions are demonstrated.

In this study, we have developed a novel recycling approach in order to obtain high collection rates of the expensive catalyst material, improve the design for the recycling of the different components including the Nafion electrolyte, and demonstrate the deployment of a modern and efficient recycling methodology for PEM water electrolysis. Our goal was to proof the concept of a closed-loop material system. We have found that recycled catalysts and membranes can be used to fabricate new MEAs without significantly affecting the performance.  There is still room for much improvement, but limitations of many kinds - not all of them technological – shall not impede in the long term the complete closure of the materials cycle.

[1] M. Carmo, D. Fritz, J. Mergel and D. Stolten, International Journal of Hydrogen Energy, 38, 4901-4934 (2013)