Electrochemical Reduction of N2O with a Molecular Copper Catalyst

Deoxygenation of nitrous oxide (N2O) has significant environmental implications, as it is not only a potent greenhouse gas but is also the main substance responsible for the depletion of ozone in the stratosphere. This has spurred significant interest in molecular complexes that mediate N2O deoxygenation. Natural N2O reduction occurs via a Cu cofactor, but there is a notable dearth of synthetic molecular Cu catalysts for this process. In this work, we report a selective molecular Cu catalyst for the electrochemical reduction of N2O to N2 using H2O as the proton source. Cyclic voltammograms show that increasing the H2O concentration facilitates the deoxygenation of N2O, and control experiments with a Zn(II) analogue verify an essential role for Cu. Theory and spectroscopy support metal–ligand cooperative catalysis between Cu(I) and a reduced tetraimidazolyl-substituted radical pyridine ligand (MeIm4P2Py = 2,6-(bis(bis-2-N-methylimidazolyl)phosphino)pyridine), which can be observed by Electron Paramagnetic Resonance (EPR) spectroscopy. Comparison with biological processes suggests a common theme of supporting electron transfer moieties in enabling Cu-mediated N2O reduction.


General Considerations
All manipulations were performed under a dry nitrogen atmosphere using either standard Schlenk techniques or in an M. Braun Unilab Pro glovebox unless otherwise stated.Glassware was dried at 180 °C for a minimum of two hours and cooled under vacuum prior to use.All chemicals were obtained from commercial sources and used as received unless otherwise stated.Solvents were dried on a solvent purification system from Pure Process Technologies before storing over 4 Å molecular sieves under N2.Diethyl ether (Et2O) was stirred over NaK alloy and passed through a column of activated alumina prior to storing over 4 Å sieves under N2.Ultrapure water was obtained from a Milli-Q system.2,6-(Me3Si)2Py was prepared according to a previously reported procedure. 1EPR spectra were recorded on a Bruker Elexsys E500 spectrometer with an Oxford ESR 900 X-band cryostat and a Bruker Cold-Edge Stinger and data was analyzed using the EasySpin Matlab suite. 2 UV-vis spectra were recorded on a Thermo Scientific Evolution 300 spectrometer with the VISIONpro software suite.A standard 1 cm quartz cuvette with an airtight screw cap with a puncturable Teflon seal was used for all measurements.NMR spectra were recorded on either Bruker DRX-400 or AVANCE-500 spectrometers.Single crystal X-ray diffraction data was collected at the Advanced Photon Source of Argonne National Laboratory (beamline 15-ID B,C,D) using X-ray radiation with a wavelength of λ=0.41328Å. Combustion analysis was performed by Midwest Microlab.

Electrochemical Measurements
Electrochemical measurements were carried out using a BAS Epsilon potentiostat and using BAS Epsilon software version 1.40.67NT.Cyclic voltammetry measurements were collected in an undivided three-electrode setup using a 3-mm diameter glassy carbon working electrode, a Pt wire counter electrode, and a Ag wire pseudo-reference electrode.Potentials were referenced by adding ferrocene (1 mM) as an internal standard.
Controlled potential electrolysis (CPE) experiments were performed in a custom-made airtight two-chamber H-cell with a pressure equalizing arm separated by a fine glass frit.A conventional three-electrode set-up for CPE was carried out using RVC as the working electrode, a Ag + /Ag (10 mM AgBF4 + 0.1 M N n Bu4PF6 in MeCN) reference electrode and a sacrificial Zn rod as the counter electrode separated by a glass frit containing 0.1 M N n Bu4PF6 in MeCN.The working electrode (Duocel RVC Foam Panel, 100 PPI, 3% relative density) was assembled by piercing the RVC foam (cut to approximately 0.5 x 0.5 x 2 mm 3 ) with a carbon rod.A conductive carbon glue was used as an adhesive between the carbon rod and RVC and the electrode was dried in an oven overnight.Approximately half of the RVC electrode submerged into solution avoid contact the carbon rod during CPE.The solutions in both chambers were sparged with N2O for 1 h and the headspace was purged for an additional 30 min prior to electrolysis to minimize the amount of trace N2 in the system.A flask containing dry acetonitrile was used as a pre-bubbler during sparging to reduce solvent loss.The cell was covered in aluminum foil and stirred at 720 rpm during electrolysis.

Gas Chromatography (GC)
Headspace gas sampling was analyzed using an Agilent 7890B GC equipped with both a flame ionization detector (FID) and a thermal conductivity detector (TCD).TCD chromatograms were collected for the detection of N2 using He as a carrier gas through a CARBOXEN TM 1010 PLOT fused silica capillary column (30m x 0.32 mm).A calibration curve for product quantification was made by sampling known N2/N2O mixtures.The H-cell was allowed to sit while stirring (720 rpm) for 1 h after sparging and after electrolysis to guarantee gas homogeneity in the headspace.Injection volume 100 L; He carrier gas, flow 1 mL/min; Inlet 150 °C; Oven: 35 °C (15 min) then 24 °C /min to 220 °C (hold 8 min); Split ratio 30:1; TCD detector 230 °C.

Calculating Faradaic Efficiency (FE)
Faradaic efficiency (FE) was determined similarly to a previous report where FE was calculated based on the amount of N2 found in the headspace since the amount of N2 in solution was determined to be negligible (6.4 M in acetonitrile). 3 The volume of N2 in the headspace after 1 h of controlled potential electrolysis was determined as follows: The area of a GC measurement from a 100 L sample of the headspace before electrolysis ( 2  ) was subtracted from the area after electrolysis ( 2  ),  2  =  2  −  2  .Using a calibration curve, the amount of N2 in the 100 L sample was determined and thus the % of N2 from measured can be obtained (% 2 =  2 100  × 100% ).The total volume of N2 in the headspace is obtained by:  2 () = (8.8 × % 2 ) ÷ 100 Where 8.8 mL is the total headspace volume.The Faradaic efficiency for N2 is hence calculated using the equation: Where  2 is the moles of N2 produced determined by  2 =  2 () ÷ (22400   −1 ), Q (in ) is the amount of charge passed, and F is Faraday's constant (96500   −1 ).
MeIm 4 P 2 Py (1).Neat 2,6-(Cl2P)2Py (9.00 g, 32.0 mmol) was added at ambient temperature to a round bottom flask containing a stirring pyridine solution (50 mL) of N-methylimidazole (10.5 g, 128 mmol) and triethylamine (13.0 g, 128 mmol).A white solid precipitated and the reaction mixture was allowed to stir for 24 h.Volatiles were removed under reduced pressure and the solid was taken out of the glovebox and dissolved in chloroform (200 mL).The solution was washed with 200 mL of 1 M NaOH and the organic phase was collected.The aqueous phase was washed two more times with chloroform (200 mL) and the organic phases were combined and dried with Na2SO4.The solution was filtered and the solvent was removed under reduced pressure.The solid product was washed with 200 mL of THF to afford 1 as a white powder (9.20 g, 60% Chemical Reductions with Na/Hg N 2 O Reduction.An acetonitrile solution (10 mL) containing 1 mol % 1-Cu (8.2 mg) and 18 L of H2O (100 mmol) was sparged with N2O for 1.5 h inside a 25 mL Schlenk flask.A slight excess of 0.1 % Na/Hg (210 mmol) was prepared inside a separate 50 mL Schlenk flask under an N2 atmosphere.The headspace was then evacuated for 15 min and the flask was purged with N2O for 15 min.The acetonitrile solution was then cannula transferred using N2O gas into the flask containing Na/Hg and the mixture was stirred overnight (16 h).A control experiment in the absence of 1-Cu was also done following this same procedure.Quantification of N2 was determined and measured using TCD GC; no other gaseous products were detected in the headspace.

EPR of reduced compounds.
A solution of 1-Cu (0.5 mM) and 1-Zn (10 mM) were prepared in acetonitrile and added to approximately 100 equivalents of 20 % Na/Hg.Both reaction mixtures were stirred for 1 h to afford a red solution for the reduced Zn species and a brown solution for the doubly reduced Cu complex.
Although the nature of the decomposition product(s) were not extensively investigated, it is worth noting that there were observable changes in the solutions that indicated degradation of the samples.For example, in the Zn reduction reaction the red solution slowly turns colorless after 1 h.In the case of the Cu reduction reaction, precipitates begin to slowly form after 1 h.The area of O2 from air remains constant in the chromatogram before and after electrolysis, suggesting there is no adventitious N2 from air at the end of the experiment.  .Headspace sampling of the reduction of N2O using 0.1 % Na/Hg as a reducing agent and H2O as proton source under the presence of 1 mol % loading of 1-Cu (red) and in the absence of catalyst (black) over the course of 16h.Gray trace in chromatogram is the starting amount of N2 in the system before the start of the experiment.

Figure S36.
Calibration curve for H2 quantification in the headspace after controlled potential electrolysis of 1-Cu (1 mM) under 1 atm of N2 in the presence of 100 mM H2O.The amount of H2 found in the headspace is less than stoichiometric with respect to the catalyst, affording 0.76×10 -6 mol of H2 per 4.0×10 -6 mol of 1-Cu over 1 h.This indicates that 1-Cu is not a catalyst for H2 evolution using H2O as a substrate at −2.3 V (vs Ag + /Ag).S3.Crystal data and structure refinement for 1-Zn.

Empirical formula C25H26F6N10O6P2S2Zn
Formula weight 867.99

Density Functional Theory (DFT)
5][6][7][8] The basis sets of Weigand and Ahlrichs were used; 9-10 copper was given the def2-TZVPP functional, all atoms bound to copper were given def2-TZVP, and all other atoms were given def2-TZVP(-f).Optimized structures were confirmed by the absence of negative eigenvalues of the Hessian.Entropy and Gibb's free entropy were determined at 296 K and 1 M using the quasi-QRRHO model for Grimme and coworkers. 11For EPR calculations on the reduced ligand radical, copper was given def2-QZVP, atoms bound to copper, phosphorous, and the 3-positions of the imidazole arms were given def2-TZVPP.Hyperfine couplings were calculated with isotropic and dipolar contributions on all copper, nitrogen, and phosphorus, and pyridyl hydrogen nuclei and orbital contributions on copper.Table S4.Summary of calculated energies from DFT calculations.The free energy of disproportionation of two Cu(I) species into a Cu(II) species and Cu(I), reduced ligand species is calculated to be 31.4kcal/mol, compared to 31.1 kcal/mol from the electrochemical data.a Energy reference.b At standard conditions with all species at 1 M. c EPR parameters were calculated for this species.

Figure S20 . 2 ]Figure S21 .Figure S22 .Figure S23 .
Figure S20.CVs of various concentrations under 1 atm of N2O of (a) 1-Cu (top) in the presence of 100 mM H2O and (b) various concentrations of H2O in the presence of 1 mM 1-Cu.The logarithmic plots of the current at -2.4 V vs Fc + /Fc with respect to (c) catalyst concentration and (d) H2O concentration.The linear dependence in both plots suggests that the reaction is likely first order in catalyst and in H2O.Glassy carbon working electrode, Ag wire pseudo-reference electrode, Pt wire counter electrode; MeCN + 0.1 M N n Bu4PF6 supporting electrolyte.

2 Figure S31 .
Figure S31.Headspace analysis before (black) and after controlled potential electrolysis (red).The area of O2 from air remains constant in the chromatogram before and after electrolysis, suggesting there is no adventitious N2 from air at the end of the experiment.

2 Figure S32 .
Figure S32.Headspace analysis before (black) and after controlled potential electrolysis (red).Full chromatogram shows that N2 is the only gaseous product in the reaction.
Figure S33.Headspace sampling of the reduction of N2O using 0.1 % Na/Hg as a reducing agent and H2O as proton source under the presence of 1 mol % loading of 1-Cu (red) and in the absence of catalyst (black) over the course of 16h.Gray trace in chromatogram is the starting amount of N2 in the system before the start of the experiment.

Figure S35 . 2
Figure S35.Headspace analysis after controlled potential electrolysis under 1 atm of N2 in the presence of 100 mM H2O. TCD GC Chromatogram shows that H2 evolution in the reaction in the absence of N2O.

Figure S38 .
Figure S38.UV-vis spectrum of the reduction of 1-Cu (0.5 mM) in MeCN in the presence of excess Na/Hg showing the growth of two bands.

Figure S44. 4 -
Figure S44.4-Coordinate (Left) and 3-Coordinate (right) optimized structures for the Cu(I) (top) and Cu(I), reduced ligand species (bottom).Other permutations of two-arms unbound are also likely accessible.XYZ coordinates are available in the online supporting information.

MeIm 4 P 2 Py)Cu(OTf)][OTf] (1-Cu
An acetonitrile solution (5 mL) of Zn(OTf)2 (118 mg, 0.324 mmol) was slowly added to a slurry of MeIm4P2Py (150 mg, 0.324 mmol) in 5 mL of acetonitrile and stirred for 4 h at 50 °C.The solution was allowed to cool to room temperature and was filtered through Celite.Volatiles were removed under reduced pressure to afford 1-Zn as a white solid in quantitative yield.Crystals suitable for X-ray diffraction were grown at room temperature by layering Et2O on a concentrated solution of 1-Zn in acetonitrile (195 mg, 73%).

Table S1 .
Output from calculated A values (MHz) of the EPR of the 2e − reduced (ligand radical) model complex.Euler angles are reported in z-y-z convention.See DFT section below for computational details.

Single Crystal X-ray Diffraction (SXRD)
Solid state structure of 1-Cu and select bond distances and angles in the table below.Ellipsoids are set to 50% probability and hydrogen atoms, triflate counterion and acetonitrile solvent molecule have been omitted for clarity.C shown in gray, N in blue, O in red, F in green, P in orange, S in yellow, and Cu in brown.

Table S2 .
Crystal data and structure refinement for 1-Cu.Solid state structure of 1-Zn and select bond distances and angles in the table below.Ellipsoids are set to 50% probability and hydrogen atoms have been omitted for clarity.C shown in gray, N in blue, O in red, F in green, P in orange, S in yellow, and Zn in silver.