A key component of the emerging hydrogen energy economy is the distribution system which facilitates the movement of hydrogen energy between production and end-use. There is interest in valorization of existing gas systems, which today are dedicated to natural gas movement, to facilitate the co-transportation of hydrogen via blending of hydrogen gas into these gas systems at concentrations up to 20% H₂ by volume [1]. Fuel cell quality hydrogen, at a purity of 99.97%> H₂, is required for the most valuable energy end-uses of hydrogen gas and thus the ability to efficiently extract H₂ gas at such high purity is critical to the adoption of natural gas infrastructure as a means of co-transportation and distribution.

Electrochemical hydrogen pump (EHP) utilizes a proton conducting membrane to separate hydrogen from mixtures by driving the electrochemical process of oxidation of hydrogen at an anode and subsequent evolution of hydrogen gas at a cathode, while other gaseous impurities are ideally unable to permeate through the membrane. HT-PEM EHP based on polybenzimidazole (PBI) membranes doped with phosphoric acid exhibit high proton conductivities at temperatures in the range of 160C Celsius while also exhibiting characteristics that are complimentary to their application in gas separation processes. These characteristics include a limited tolerance to common sources of platinum catalyst inhibition such as CO [1] and the presence of an aqueous phosphoric acid phase inhibits gas cross-over, providing better product purity although at the cost of performance in the form of catalyst inhibition.

A two-dimensional model based on an HT-PEM system employing a phosphoric acid doped PBI membrane and free phosphoric acid as the proton conducting phase of the catalyst layer is developed to better understand the underlying processes governing the performance of the EHP. The model is validated with experimental measurements with mixtures as low as 2% H₂ by volume in methane. In-operando micro-CT imaging of an HT-PEM EHP cell taken at the LBNL Advanced Light Source (ALS) is used to further validate physical parameters and assumptions of the model.

The influence of differential pressure, relative humidity of the feed, and concentration of feed gases on separation performance are investigated. Losses due to back-permeation of hydrogen under pressure are accounted for. Power-loss voltage breakdown analysis indicates appreciable losses due to proton transport in the relatively thick catalyst layer, and the dependence of these losses on the volume fraction of acid present (Figure 1). The influence of CO₂ as a catalyst inhibiting contaminant as was previously observed in low temperature EHP [3] is examined. The model and experiments show that a specific energy of separation of 5.1 kWh/kg H₂ at a hydrogen recovery factor (HRF) of 50% can be achieved in a single stage with the EHP, producing 99.995% > H₂ from a 2% H₂ in CH₄ blend, while pressurizing the product H₂ at a product pressure ratio of 1.3 relative to feed pressure.

Bibliography

[1]	C. J. Quarton and S. Samsatli, "Power-to-gas for injection into the gas grid: What can we learn from real-life projects, economic assessments and systems modelling?," Renewable and Sustainable Energy Reviews, vol. 98, pp. 302-316, 2018.
[2]	K. Perry and B. B. Eisman G.A., "Electrochemical hydrogen pumping using a high-temperature polybenzimidazole (PBI) membrane," Journal of Power Sources, pp. 478-484, 2008.
[3]	N. e. al., "Effect of CO2 on the performance of an electrochemical hydrogen compressor," Chemical Engineering Journal, vol. 329, 2020.