To increase the energy density of hydrogen fuel, a common practice is through the production of high pressure hydrogen (up to 70 MPa).^1-4 Furthermore, to ensure green or renewable hydrogen, water electrolysis from solar or wind resources is an attractive pathway, yet such technology has limitations on efficiency, especially when trying to deliver high-pressure hydrogen. Increasing electrolysis efficiency at the cell and system level requires the use of advanced membranes that are thin and have high proton conductivity. However, such thin membranes are more prone to mechanical failure and increased hydrogen crossover when under a large pressure gradient. An intriguing possibility to solve this is to couple the electrolysis directly with an electrochemical compressor. In this fashion, a thin membrane can be used in electrolysis as a lower pressure differential is required. Moreover, the electrochemical compressor has less issues related to noise and scalability that are inherent with mechanical compression. Thus, a combined, optimized system could be deployed hydrogen refueling stations or at larger installations.

In this work, we will explore the tradeoffs and opportunities for electrolysis and compressor system including detailed analysis of the operating physics of both devices using multiscale mathematical modeling. The two devices are modeled together by utilizing the output conditions from electrolysis submodel as the input for hydrogen compressor submodel. Of importance is examining the possibilities for advancements in the cell components such as strong, conducting membrane materials that inhibit crossover and optimizing for multiphase flow. Such latter phenomena are accounted for using a multiscale modeling approach where the microstructure effect on transport properties is considered. The structural information such as pore-size distribution and tortuosity is obtained from X-ray computed tomography (CT) data, which is also used to validate some of the model predictions under operando conditions. Based on the X-ray data, the microstructure of the porous transport layer has been reconstructed and two-phase fluid simulations have been carried out to predict effective transport properties that are incorporated into the marcoscale model. Lastly, the porous electrode parameter optimization such as ionomer loading, porosity, and thickness, is introduced to identify critical areas for improvement at the cell and system levels for better efficiency and less energy consumption.

Acknowledgements
The authors would like to thank Melissa Bosch, Jonne Konink, Martijn Mulder, and Denise Swanborn for insightful discussions, and Bryan Pivovar and Kelly Meek for their work on the EOD determination. This work was jointly funded by a CRADA with HyET Hydrogen and under the HydroGen Consortium of the U.S. Department of Energy, Fuel Cell Technologies Office under contract number DE-AC02-05CH11231.

1. M. Bernt and H. A. J. J. o. T. E. S. Gasteiger, 2016, 163, F3179-F3189.
2. U. Babic, M. Suermann, F. N. Büchi, L. Gubler and T. J. Schmidt, Journal of The Electrochemical Society, 2017, 164, F387-F399.
3. M. Carmo, D. L. Fritz, J. Mergel and D. Stolten, International Journal of Hydrogen Energy, 2013, 38, 4901-4934.
4. K. E. Ayers, E. B. Anderson, C. Capuano, B. Carter, L. Dalton, G. Hanlon, J. Manco and M. J. E. T. Niedzwiecki, 2010, 33, 3-15.