Muller and Spitzer (1) have electrolyzed aqueous ammonia solutions which contained dissolved sodium hydroxide, and the anode products were found to be mixtures of nitrogen, sodium nitrate, and sodium nitrite in varying amounts.2NH3 ~-6 OH-= Ne + 6H20 ~-6e-NH3 + 9 OH-= NOz-+ 6H20 -t-8e-Anodes of smooth platinum and other metals were tried (1,2), but nitrogen was never produced in quantities which corresponded to complete stoichiometric consumption of ammonia with respect to Eq.[1]; substantial quantities of nitrate and lesser quantities of nitrite ions were always formed.As part of a study on an aqueous fuel cell system using ammonia as an anodic reactant, an effort was made to synthesize nitrogen from aqueous ammonia solutions with near 100% yields. This study was undertaken to find the effect of the highly active electrode catalyst, platinum black, on the course of the over-all anode reaction in potassium hydroxide electrolyte of "maximum conductivity" concentration, 6.9M at 18~ (3).
ExperimentalElectrolysis was carried out at 23~ in the cell Pt (black), NH~ (3M), KOH (6.9M)/Cellophane/KOH (6.9M), NiThe anode consisted of 0.2g of precipitated black platinum (4) evenly distributed and clamped between two 1 x 1 in. sintered nickel Clevite plates, 0.032 in. thick, which were connected to an electrical contact tab. A 1/32 in. diameter capillary leading to the anode face was connected to an Hg-HgO, 6.9M KOH reference cell. Electrolyte was pumped from a 2-liter reservoir through the anode at 6.8 cm/min and then recycled. A gas-liquid separator permitted the complete separation of the anode gas from the flowing electrolyte, and the gas volume was measured by standard gasometric techniques after scrubbing in 10% HC1 solution to remove the ammonia.The cathode chamber was static and connected to a eudiometer for the collection and measurement of the hydrogen volume. During a run, the current was kept constant, the time of the run was recorded, and the gases from the anode and cathode were collected simultaneously and measured. A typical run lasted 3 hr, and about 30 ml of anode gas were collected.Ideal gas behavior was assumed throughout the calculations. Corrections were made for the vapor pressure of water over 6.9M KOH and 10% HC1.The anode gas was analyzed with a gas chromatograph and found to be 100 • 2% nitrogen. No hydrogen was present.
Results and Discussion