Membrane/electrolyte interplay on ammonia motion inside a flow-cell for electrochemical nitrogen and nitrate reduction

Electrochemical ammonia production from molecular nitrogen and nitrate reduction reactions at ambient conditions has gained a lot of attention in recent years, making this topic more and more appealing. The race towards good and quick results in terms of Faradaic efficiency and productivity is not always focused on the possible source of ammonia contamination. In particular, Nafion membrane is the most commonly used in this field as cell separator, discarding the possible known disadvantages coming from ammonium ions absorption and release. The wettable microporous Celgard membrane has been proposed as a substitute for Nafion membranes, despite the separation mechanism, in this case, is only dimension-driven, so it does not assure ammonium ions retention. This paper reveals that the mechanism of ammonium ions absorption and release by Nafion 117 is strongly related to the cations present in the electrolyte and to a lesser extent by its pH value. On the other hand, Celgard membrane does not show any relevant ammonium ions absorption. Moreover, the different trend of ammonium ions motion from catholyte to anolyte solution inside a flow-cell reactor shows that none of the membranes is able to avoid ammonium ions crossover and that there is a correlation between the applied potential and the motion trend. Electrochemical nitrogen and nitrate reduction tests confirm how Nafion membrane can have a big impact on the final result of ammonium ions production, especially when dealing with low production quantities, leading to mistakes in the real quantity of ammonium ions coming from the reduction reaction.