Electrochemical nitrogen reduction reaction (NRR) that can be powered by renewable energy is currently broadly investigated as a sustainable alternative method for the industrial ammonia synthesis to replace, at least partially, the Haber-Bosch technology. Inspired by recent theoretical reports advocating several types of materials as active NRR catalysts, our experimental work aimed to assess the validity of some of these predictions.Vanadium and niobium nitrides were suggested by the previous computational work to be catalytically active towards the electrochemical reduction of dinitrogen to ammonia occurring via the Mars-van Krevenlen (MvK) mechanism [1]. The present experimental study thoroughly investigates the electrocatalytic activity of cubic vanadium(III) nitride, niobium(III) nitride and tetragonal Nb4N5 for the nitrogen reduction reaction in aqueous electrolyte solutions of different pH under ambient conditions [2]. VN and Nb4N5 (supported on carbon cloth) were synthesised by annealing of hydrothermally produced hydroxide precursors in NH3 atmosphere at 600-1100 °C; NbN was obtained by solid state reaction between niobium(V) chloride and urea at 1000 °C. Comprehensive testing of the materials under a wide range of aqueous conditions unambiguously demonstrates their inability to catalyse the electrosynthesis of ammonia from dinitrogen, as well as the propensity of VN synthesised at 600 °C and Nb4N5 to release lattice nitride in a non-catalytic process, which produces ammonia under reductive conditions. Thus, polycrystalline nitrides of vanadium and niobium are concluded to be catalytically inactive towards the ammonia electrosynthesis from N2 dissolved in water. When tested in non-aqueous aprotic electrolyte, Nb4N5 sustained slow ammonia electrosynthesis (4 pmol s-1 cm-2) at low faradaic efficiency (20 %). The present work additionally emphasises the compulsory requirement for the implementation of reliable testing and analysis procedures for the assessment of the catalytic properties of materials for the nitrogen reduction reaction.Another type of materials suggested by DFT to be catalytically active for the NRR at low overpotentials is molybdenum disulphide promoted with Mo metal particles (Mo/MoS2), which was suggested to provide energetically favourable binding with N2 molecule [3]. However, MoS2 is a well-known hydrogen evolution catalyst rendering any reasonable selectivity for the N2 reduction with Mo/MoS2 in aqueous medium essentially impossible, as confirmed in our experiments. Therefore, we applied two aprotic ionic liquids providing high N2 solubility [4], viz. [C4mpyr][eFAP] and [P66614][eFAP], as electrolyte media for testing the electrocatalytic properties of molybdenum-based materials. Molybdenum metal and Mo/MoS2 catalysts were synthesised by reducing 2H-MoS2 with H2 at 950 and 850 °C, respectively. In [C4mpyr][eFAP], metal catalyst did not demonstrate significant rates of the NRR, but Mo/MoS2 enabled the electrosynthesis of ammonia at the yield rate of 17 pmol cm-2 s-1 and faradaic efficiency of ca 50% at an applied potential of -0.97 V vs. NHE, when using a mixture of 90% dry N2 and 10% H2O-saturated N2 as a feed gas. However, there no ammonia was detected when Mo/MoS2 was tested in [P66614][eFAP] under the same as well as other potential and humidity conditions. Juxtaposition of cyclic voltammograms of Mo/MoS2 recorded in two ionic liquids reveals very different electrochemical behaviour of the material in [C4mpyr][eFAP] and [P66614][eFAP], which is likely to explain the lack of activity in the latter medium. Further tests in tetrahydrofuran as a solvent and using a conventional lithium triflate electrolyte have further supported the hypothesis on the catalytic activity of Mo/MoS2 for the NRR, although the ammonia yield rate (3 pmol s-1 cm-2) and faradaic efficiency (ca 3%) were very low. However, further control experiments are needed before the final conclusion on the genuine NRR catalytic activity of this material can be achieved.Overall, the present study demonstrates that theoretical predictions on the catalytic activity of different materials for the NRR should be considered with caution, and that robust NRR testing and ammonia analysis protocols must be implemented to unambiguously prove genuine ammonia electrosynthesis from N2 [5].
Read full abstract