Boosting selective nitrogen reduction to ammonia on electron-deficient copper nanoparticles

Production of ammonia is currently realized by the Haber–Bosch process, while electrochemical N2 fixation under ambient conditions is recognized as a promising green substitution in the near future. A lack of efficient electrocatalysts remains the primary hurdle for the initiation of potential electrocatalytic synthesis of ammonia. For cheaper metals, such as copper, limited progress has been made to date. In this work, we boost the N2 reduction reaction catalytic activity of Cu nanoparticles, which originally exhibited negligible N2 reduction reaction activity, via a local electron depletion effect. The electron-deficient Cu nanoparticles are brought in a Schottky rectifying contact with a polyimide support which retards the hydrogen evolution reaction process in basic electrolytes and facilitates the electrochemical N2 reduction reaction process under ambient aqueous conditions. This strategy of inducing electron deficiency provides new insight into the rational design of inexpensive N2 reduction reaction catalysts with high selectivity and activity.

(UPS) measurements were conducted on a Kratos Axis Ultra DLD spectrometer using a monochromated Al Kα radiation. Inductively coupled plasma (ICP) optical emission measurements were conducted on an iCAP7600 spectrometer. Temperature programmed desorption (TPD) tests were carried on a Micrometritics AutoChem Ⅱ chemisorption analyzer.
Low-pressure gas adsorption measurements were performed on a Nova 2200e surface area & pore size analyzer. Samples were degassed under dynamic vacuum for 12 h at 100 o C prior to each measurement. N2 isotherms were measured using a liquid nitrogen bath (77 K).

Electrochemical measurements
Electrochemical NRR performances were investigated using a CHI 730 C electrochemistry workstation (CH Instruments, Inc., Shanghai) in a 150 mL five-necked flask based on a standard three-electrode system. The three-electrode electrochemical cell was made up of a Cu/PI/carbon cloth electrode, a graphite rod and a saturated calomel electrode (SCE), which were served as working electrode, counter electrode, and reference electrode, respectively. The NRR tests were performed in N2 saturated 0.1 mol L -1 KOH solution (90 mL). The electrolyte was stirred and bubbled with N2 gas for 30 min before the tests. During the tests, the N2 flow was continuously inputted using properly positioned spargers so that the whole cathode was hit by the gas bubbles where N2 and water (H2O) combine with electrons to form hydroxide (OH -) and the N2 reduction product. For comparison, potentiostatic tests in Ar saturated 0.1 mol L -1 KOH solution were also conducted in this work. Linear sweep voltammograms (LSV) measurements were conducted with a scan rate of 10 mV s −1 . All the LSV curves are the steady-state ones after several cycles. The long-term NRR tests were performed using chronoamperometric measurements. An empty glass tube was set at the end of the flask. After NRR reaction, the liquid in glass tube was poured back into the flask for following ammonia determination. All potentials were described versus the reversible hydrogen electrode (RHE) via the following equation: ERHE = ESCE + 0.059 V × pH + 0.241 V. Cyclic voltammogram (CV) measurements were conducted by potential cycling in Ar saturated 0.1 mol L -1 KOH solution to evaluate the corresponding HER performance. The working electrode was the rotating disk electrode (RDE) with a mass loading of Cu/PI-300 at 1 mg cm -2 .
The rotate speed was around 1000 rpm and the curves were iR corrected.
The CV curves were converted as overpotential vs log current (log j) to get Tafel plots. Tafel slope was obtained by fitting the Tafel plots (the linear portion) to the Tafel equation (η = blog(j) +a).
Mott-Schottky plots were performed in a typical three-electrode system in the 0.1 mol L -1 PBS solution on an electrochemical workstation (CHI 660C) by using the Impedance-Potential technique. The Cu/PI/FTO electrodes were prepared using the same method referred to the Cu/PI/ carbon cloth electrode and used as the working electrode. A graphite rod was used as the counter electrode and SCE was used as the reference electrode. All measurements were performed at room temperature.

Colorimetric method
The quantity of NH3 formation was determined via a reported colorimetric method S1 using Nessler's reagent. Calibration curve (Supplementary Figure 13b) was plotted according to the following method: Firstly, a series of reference solutions were prepared by adding suitable volumes of the ammonia working 0.1 M KOH solution in colorimetric tubes. Then, the solution was made up to the mark (10 mL) with 0.1 M KOH solution before adding 1 mL of 0.2 M potassium sodium tartrate (KNaC4H4O6) solution to each of the tubes and mix thoroughly; Next, 1 mL of Nessler's reagent was added to each of the tubes and the solution was standing for 20 minutes for color development; Finally, the absorbance of the solutions was measured at 425 nm using a 10 mm glass cuvette.

Ion chromatography method
After NRR test, 10 mL obtained NH3 solution in 0.1 M KOH was adjusted pH to 3 with 1.01 mL 1 M HCl and filtered through a nylon membrane filter (0.45 mm) before analysis by Ion Chromatography (Dionex 1500i with an AS 4 ASC column; Sunnyvale, USA). The NH4 + peak was observed at 7.05 min. Calibration curve (Supplementary Figure 14b) was built to quantify the ammonium ion in the solution.

N2 isotope labeling experiments
99% 15 N2 was used as the feeding gas to perform the isotopic labeling NRR experiment in order to clarify the source of ammonia. The electrochemical reactor was sealed, degassed and filled with Ar for three times, then refilled with 15 N2 gas. 20 mL 15 N2 was injected to the system every 10 min during the test. After NRR for 6 h, the pH value of obtained solution was adjusted to 7 with 0.5 M H2SO4 and 15 NH 4+ was identified using 1 H NMR spectroscopy on the obtained solutions (Supplementary Figure 15). 14 N2 experiment was also performed in the same condition for comparison.

Faraday efficiency, generation rate and turnover frequency
The Faradaic efficiencies for NRR were defined as the quantity of electric charge used for synthesizing ammonia divided by the total charge passed through the electrodes during the electrolysis. The Faraday efficiency (FE) can be calculated as FE = nN × 96485/(It), where N is the electron number for the NRR (here, N is 3); n represents the mole of NH3; I represents the current (A); and t represents time (s). The generation rate (GR) of ammonia was calculated using the following equation: GR = 17n/(tA), where n represents nNH3; t represents time (h); and A is the effective area of the electrode (cm 2 ). The turnover frequency (TOF) can be calculated by the equation: TOF = n/(tn'), where n represents the mole of NH3; t represents time (h); and n' represents the mole of Cu on the electrode.

Theoretical calculations
All theoretical calculations were performed using density functional theory (DFT), as implemented in the DMol3 program based on Materials Studio 8.0. The electronic exchange-correlation energy was described by the generalized gradient approximation (GGA) method with spin polarized Perdew-Burke-Ernzerhof (PBE) functional. S2 Valence orbitals were described with the double numerical plus polarization (DNP) basis. S3 The Cu (111) surfaces in the experiment were modelled as a Cu model by cutting a 3×3×3 repetitive rhomb Cu cluster (i.e. pristine Cu). S4