Ammonia-fed reversible protonic ceramic fuel cells with Ru-based catalyst

The intermediate operating temperatures (~400–600 °C) of reversible protonic ceramic fuel cells (RePCFC) permit the potential use of ammonia as a carbon-neutral high energy density fuel and energy storage medium. Here we show fabrication of anode-supported RePCFC with an ultra-dense (~100%) and thin (4 μm) protonic ceramic electrolyte layer. When coupled to a novel Ru-(BaO)2(CaO)(Al2O3) (Ru-B2CA) reversible ammonia catalyst, maximum fuel-cell power generation reaches 877 mW cm−2 at 650 °C under ammonia fuel. We report relatively stable operation at 600 °C for up to 1250 h under ammonia fuel. In fuel production mode, ammonia rates exceed 1.2 × 10−8 NH3 mol cm−2 s−1at ambient pressure with H2 from electrolysis only, and 2.1 × 10−6 mol NH3 cm−2 s−1 at 12.5 bar with H2 from both electrolysis and simulated recycling gas.


Challenges on electrochemical ammonia synthesis
In general, there are two reaction mechanisms regarding electrochemical production of ammonia from nitrogen and hydrogen: the dissociative mechanism and the associative mechanism [59][60][61][62][63][64] . These mechanisms are well reported in literature, but with variable nomenclature. For convenience, the reaction mechanisms are reformatted as following. I) Dissociative mechanism: In the dissociative mechanism, each N2 molecule forms two N atoms directly upon adsorption on the surface site. These then capture three protons and three electrons from the reaction interface or the reaction medium. The dissociation energy in R1 reaches 945.4 kJ.mol -1 . Additionally, the electrochemical driving force does no effect 59,61 , so that N2 dissociation is the highly unfavorable rate-determining step.
II) Associative mechanism: The first hydrogenation of nitrogen (R6) is the rate-determining step in the associative mechanism 60,61 65 . The activation energy is ~ 94.5 kJ.mol -1 60 , one order of magnitude lower than that of the rate-determining step in the dissociative mechanism. This makes the associative mechanism more energetically favorable for electrochemical ammonia production.
Following the dissociative mechanism, Singh et al. proposed a model to predict rate of ammonia and hydrogen production based on Langmuir isotherms 61 . The two most-important reactions: where K's are the bulk-to-near-surface equilibrium constants, k's are rate constants, and ~' cs are bulk concentration. Note the termc− is not meant to indicate a physical concentration but to account for cases where the electron transfer rate is limiting 61 . The rate-determining step of this model is the first hydrogenation step of N2 (R5).
The above two equations indicate the rate of ammonia production is zeroth order in the electron and proton concentrations, while the rate of hydrogen production is first order in both electron and proton concentrations. The two equations suggest the ammonia production rate is nonelectrochemically driven and the rate is nearly independent of electron and proton concentrations.
In order to increase selectivity, the electron and proton concentrations must be minimized in order to minimize the hydrogen evolution reaction (HER). Detailed strategies on improving ammonia selectivity are proposed Singh et al. 61 . By suppressing the protonic current in proton-conducting cells, the H2 production rate is reduced, while the nitrogen-reduction rate or ammonia-production rate are nearly unaffected, thereby achieving a higher selectivity to NH3. This has been observed in a few studies where higher ammonia selectivities were achieved under relatively low current densities 52 . Of course, this approach generally is accompanied by low ammonia production rate due to the low current densities. Simultaneously maintaining high rates of ammonia synthesis and conversion efficiency remains very challenging in practice since the activation barrier of HER is much less than that of NRR. Further, no metal catalyst provides better catalytic selectivity in NRR over HER. Montoya  no catalyst demonstrated a more-negative limiting potential under HER than NRR. Rhenium had the minimum difference between HER and NRR, followed by Ru; both metals are good candidates for high-selectivity ammonia synthesis.
These thermodynamic, kinetic, and catalytic analyses all show that it is fundamentally very challenging to suppress hydrogen-evolution reaction while increasing the nitrogen reduction reaction. Low selectivity to NH3 is more favorable by nature. While better catalysts with higher selectivity may be found for electrochemical production of ammonia, rethinking the integrated electrochemical H2 production -ammonia thermocatalytic synthesis coupling may prove more efficient, where higher pressure operation is technically more feasible and unused hydrogen can be recycled back to the NH3-synthesis reactor.