Increasing Faradaic Efficiency and Current Density for the Lithium-Mediated Ammonia Synthesis

Ammonia, being one of the most produced chemicals worldwide with 180 Mio. ton per year, plays an important role in our society. It‘s main use is as fertilizer and it is predicted that without the industrial scale production of ammonia via the Haber-Bosch process, we would not be able to feed half of the current global population. With continuous global population growth the role of ammonia will be even more important in the future. Moreover, ammonia is a precursor for many chemicals both in the polymer industry as well as in the pharmaceutical industry. As important and positive the influence of ammonia is, it also leaves a negative impact on the environment. With steam reforming from methane as a main contribution, the Haber-Bosch process is responsible for around 340 Mt CO₂ emissions per year. Ammonia is hereby the chemical with the largest green house gas emission, with ethylene (140 Mt CO₂) in a far second place.
Hence, it is important to mitigate the negative impact of the Haber-Bosch synthesis by more sustainable processes, such as electrochemical nitrogen reduction (NRR) powered by renewable electricity sources. The NRR field is relatively young and therefore scattered with a lot of false positive results, since the measured concentrations are often in the sub-ppm regions. Several protocols have been published how to vigorously prove the ammonia production from nitrogen and since then the field has moved in the right direction by revisiting and even subtracting previously published results. The most promising method so far is the lithium-mediated electrochemical ammonia synthesis (LiMEAS), which has been successfully reproduced by several labs. The success and industrial relevance of an electrochemical reaction is usually quantified by values such as Faradaic efficiency (FE), energy efficiency (EE) and current density. The goal of this thesis is to increase both the Faradaic efficiency and current density to levels where the reaction can be considered for scale-up.
The FE was increased significantly from 25 to around 80 % by the addition of small amounts of oxygen (0.6 - 0.8 mol.%) into the reaction atmosphere at 20 bar total nitrogen pressure. This result was quite unexpected, since oxygen was initially believed to hinder the reaction due to competing oxygen reduction reaction (ORR) and Li₂O formation. However, the data presented here prove the opposite, that oxygen greatly enhanced the FE at  the right concentrations. At higher concentrations however, the FE dropped rapidly to 0 %. The explanation of this unusual effect was by modification of the solid electrolyte interphase (SEI) by oxygen. From literature it is already known that oxygen modifies the SEI layer and previous work from our group predicts that the FE is strongly depending on the diffusion rates of the reactants through the SEI layer. To support this hypothesis, air-free characterization (X-ray photoelectron spectroscopy, XPS, X-ray diffraction, XRD) of the SEI layer was conducted at different concentrations of oxygen. With experimental results and theoretical predictions combined, the modification of the SEI layer through O₂ was proven.
To increase the current density, high surface area copper electrodes were synthesized by electro-deposition at high overpotentials via the hydrogen bubble template (HBT) method. Then the electrolyte had to be optimized for high current application. Namely the concentration of the electrolyte had to be increased from 0.3 M to 2 M to increase the conductivity and therefore minimize iR losses. Furthermore, higher concentrations lead to a thinner double-layer, which increases the electrochemical surface area (ECSA). Lastly, the ECSA was determined by capacitive cycling in the same solvent as LiMEAS. With these electrodes, the current density was increased to -100 mA/cm² geo. Further optimization of the HBT method and electrolyte lead to a current density of -1 A/cm² _geo with FE of 71 ± 3 %. 
The last focus in this thesis is to elucidate the role of ethanol (EtOH) in the system. It was initially predicted to be the proton source, but that does not  seem to be it‘s only role. Experiments where the electrolyte was switched from containing EtOH to non containing EtOH electrolyte showed even higher FE than with EtOH. The conclusion from these experiment was that EtOH has an important role to play in the beginning to built the SEI layer, but in the sequential run does not serve as proton source and ammonia can be formed with tetrahydrofuran (THF) as proton source.