Ammonia Detection Methods in Photocatalytic and Electrocatalytic Experiments: How to Improve the Reliability of NH3 Production Rates?

The enzyme nitrogenase inspires the development of novel photocatalytic and electrocatalytic systems that can drive nitrogen reduction with water under similar low-temperature and low-pressure conditions. While photocatalytic and electrocatalytic N2 fixation are emerging as hot new areas of fundamental and applied research, serious concerns exist regarding the accuracy of current methods used for ammonia detection and quantification. In most studies, the ammonia yields are low and little consideration is given to the effect of interferants on NH3 quantification. As a result, NH3 yields reported in many works may be exaggerated and erroneous. Herein, the advantages and limitations of the various methods commonly used for NH3 quantification in solution (Nessler's reagent method, indophenol blue method, and ion chromatography method) are systematically explored, placing particular emphasis on the effect of interferants on each quantification method. Based on the data presented, guidelines are suggested for responsible quantification of ammonia in photocatalysis and electrocatalysis.

Commercial TiO 2 (P25) (300 mg) was placed in an alumina tube furnace and then heated to 500 °C in an atmosphere of flowing NH 3 (30 mL min -1 ) at a heating rate of 5 o C min -1 . After ammonolysis at 500 o C for 3 h, the sample was cooled to room temperature, and the product was collected.

Synthesis of N-doped Carbon:
Zn, Co-ZIF [3] (400 mg) was placed in an alumina tube furnace and then heated to 900 o C in an atmosphere of flowing NH 3 (30 mL min -1 ) at a heating rate of 5 °C min -1 . After ammonolysis at 900 o C for 2 h, the sample was cooled to room temperature and the product was collected.

Synthesis of Ni 3 N and Ni 3 FeN:
Commercial NiO (300 mg) or NiFe-LDH (300 mg) was placed in an alumina tube furnace and then heated to 500 o C in an atmosphere of flowing NH 3 (30 mL min -1 ) at a heating rate of 5 o C min -1 . After ammonolysis at 500 o C for 4 h, samples were cooled to room temperature and the product was collected.

Synthesis of g-C 3 N 4 :
g-C 3 N 4 was obtained using a standard thermal polymerization method. [4] Preparation of ammonia solutions with different pH: Ammonia solutions with different pH were prepared by adding an appropriate amount of 28 wt.%. NH 3 to a specific volume of water, followed by pH adjustment with either aqueous H 2 SO 4 (0.05 mol L -1 ) or aqueous NaOH (1 mol L -1 ). The final concentration of ammonia in all solutions was 1000 μg L -1 .

NH 3 production and control experiments:
A series of experiments were performed under UV-visible light irradiation (200-800 nm) supplied by a 300 W Xe lamp (CEL-HXF 300). Typically, photocatalyst or potential interferant (1 mmol) were dispersed in 100 mL of ultrapure water in a 150 mL quartz reactor.
The reactor was equipped with a circulating water outer jacket in order to maintain at a constant temperature of 25 o C. The photocatalyst suspension or solutions containing the possible interferants were stirred continuously in the dark whilst either high-purity N 2 or high purity Ar was bubbled through the suspension at a flow rate of 60 mL min -1 for 10 min. The reactor was then irradiated using the 300 W Xe lamp under either a N 2 or Ar flow (100 mL min -1 ). At regular intervals, 3 mL aliquots of the reaction solution were collected using a syringe, then immediately centrifuged to remove any solid material. The concentrations of NH 3 was determined by ion chromatography (930 compact IC Flex, Metrohm). Errors in the data were expressed as follows: The relative error: , where X is the measured value and M is the true value.

The absolute error:
, where X is the measured value and M is the true value.
The faradaic efficiency was calculated according to the following equation: [5] The faradaic efficiency = 3×F× where [NH 3 ] is the measured NH 3 concentration; V is the volume of the cathodic reaction solution for NH 3 collection; t is the potential applied time; A is the geometric area; m is the loaded mass of catalyst; F is the Faraday constant; and Q is the quantity of applied electricity.  For a Faradaic efficiency of 1%, the NH 3 yield is estimated to be about 300 μg L -1 . Figure S2 shows that the different methods of ammonia quantification afford different    caused some absorption at 420 nm, especially solutions contained Ag + , Ce 3+ , In 3+ or Zn 2+ for which a precipitate formed, leading to cloudiness and therefore interference detection.
Although solutions of Li + and Cu 2+ ions did not form a precipitate, these metal ions still interfere with ammonia detection.            For NH 3 production via electrocatalysis, the reactor, electrode and catalyst (if used for cycling tests) require repeated cleaning with ultra-pure water or acid to remove residual ammonia before commencing a test run. Then, 14 N 2 should be used to screen catalysts and optimize reaction conditions. If the amount of ammonia denerated during the reaction is sufficient to exceed the detection limit of 1 H NMR, then that method should be used to confirm 14 NH 4 + formation. Finally, 14 N 2 should be replaced with 15 N 2 and the electrocatalytic reaction perfromed under the same reaction conditions, with the products again analyzed by 1 H NMR to confirm that N 2 was the origin of the NH 4 + formed.