Cu4S Cluster in “0-Hole” and “1-Hole” States: Geometric and Electronic Structure Variations for the Active CuZ* Site of N2O Reductase

The active site of nitrous oxide reductase (N2OR), a key enzyme in denitrification, features a unique μ4-sulfido-bridged tetranuclear Cu cluster (the so-called CuZ or CuZ* site). Details of the catalytic mechanism have remained under debate and, to date, synthetic model complexes of the CuZ*/CuZ sites are extremely rare due to the difficulty in building the unique {Cu4(μ4-S)} core structure. Herein, we report the synthesis and characterization of [Cu4(μ4-S)]n+ (n = 2, 2; n = 3, 3) clusters, supported by a macrocyclic {py2NHC4} ligand (py = pyridine, NHC = N-heterocyclic carbene), in both their 0-hole (2) and 1-hole (3) states, thus mimicking the two active states of the CuZ* site during enzymatic N2O reduction. Structural and electronic properties of these {Cu4(μ4-S)} clusters are elucidated by employing multiple methods, including X-ray diffraction (XRD), nuclear magnetic resonance (NMR), UV/vis, electron paramagnetic resonance (EPR), Cu/S K-edge X-ray emission spectroscopy (XES), and Cu K-edge X-ray absorption spectroscopy (XAS) in combination with time-dependent density functional theory (TD-DFT) calculations. A significant geometry change of the {Cu4(μ4-S)} core occurs upon oxidation from 2 (τ4(S) = 0.46, seesaw) to 3 (τ4(S) = 0.03, square planar), which has not been observed so far for the biological CuZ(*) site and is unprecedented for known model complexes. The single electron of the 1-hole species 3 is predominantly delocalized over two opposite Cu ions via the central S atom, mediated by a π/π superexchange pathway. Cu K-edge XAS and Cu/S K-edge XES corroborate a mixed Cu/S-based oxidation event in which the lowest unoccupied molecular orbital (LUMO) has a significant S-character. Furthermore, preliminary reactivity studies evidence a nucleophilic character of the central μ4-S in the fully reduced 0-hole state.


Experimental Procedures General Considerations
All manipulations of air-and moisture-sensitive materials were performed under an atmosphere of dry dinitrogen with the rigorous exclusion of air and moisture using standard Schlenk techniques, or in a N2-filled glovebox. Solvents were dried with sodium (Et2O) in the presence of benzophenone or CaH2 (MeCN) and were freshly distilled and degassed prior to use. Toluene, acetone, d6-acetone, d3-MeCN and d6-dimethyl sulfoxide (d6-DMSO) were dried with 3 Å molecular sieves.
[LCu2](PF6)2 1 and [Cp*2Fe]PF6 (Cp* = pentamethylcyclopentadienyl) 2 were synthesized according to literature procedures. [ n Bu4N]PF6 was dried in vacuo at 80 °C for 2 d, then transferred to the glovebox prior to use. All other chemicals were purchased from Sigma-Aldrich (Germany) and used as received. 1 H, 13 C, 31 P and 19 F NMR spectra were recorded with Bruker 300 MHz or 400 MHz spectrometer. All chemical shifts are reported in units of ppm and referenced to the residual 1 H NMR signals of the deuterated solvents for proton chemical shifts, the 13 C NMR signal of deuterated solvents for carbon chemical shifts, the 31 P NMR signal of 85% phosphorous acid (external standard) for phosphorous chemical shifts, and the 19 F NMR signal of CFCl3 (external standard) for fluorine chemical shifts. ESI mass spectra were recorded on a Bruker HCT ultra spectrometer. Elemental analyses were performed by the analytical laboratory of the Institute of Inorganic Chemistry at the University of Göttingen using an Elementar Vario EL III instrument. UV-vis-NIR spectra were recorded with an Agilent Cary 60 equipped with an Unisoku Cryostat (CoolSpek) and magnetic stirrer using quartz cuvettes with an attached tube and a screw cap with a septum. IR spectra were recorded inside a glovebox on a Cary 630 FTIR spectrometer equipped with ZnSe ATR module and analyzed by FTIR MicroLab software.

Synthesis of 2 and 3
Scheme S1. Synthetic route for 2 and 3.

Synthesis of [L2Cu4S](PF6)2 (2).
At room temperature, to the MeCN (7 mL) solution of [LCu2](PF6)2 (1, 55 mg, 0.06 mmol) was added Na2S (14 mg, 0.18 mmol). The suspension was stirred at room temperature for 3 days, and during this time the color of the suspension changed from orange to yellow. After centrifugation, the upper yellow solution was collected and left for crystallization through Et2O diffusion at room temperature. Yellow crystals of 2 were obtained in 80% yield (38 mg). 1  See Figures S4-S6 for a more detailed assignment of these peaks. 13 Figure S7 for a more detailed assignment of these peaks. 19  shortly under the glovebox environment using a cooling aluminum block. 1  and NaBF4 (11.0 mg, 0.1 mmol) was dissolved in acetone (5 mL) and was stirred at room temperature for 3 h.
The resulting suspension was put in the glovebox freezer (−35 o C) for 30 min and then was poured into the precooled acetone (1 mL) solution of [Cp*2Fe]PF6 (5.0 mg, 0.11 mmol). The color of the suspension changed from S5 yellow to purple immediately. The reaction mixture was stored in the glovebox freezer (−35 o C) for 3 hours,   during which course the vial was shaken violently every 30 min, to form a large amount of purple precipitate.   After decanting the upper yellowish solution, the resulting purple precipitate was washed with Et2O (3 mL × 3) until the washing liquid becomes almost colorless. The purple solid was put in the glovebox freezer (−35 o C) for 30 mins and then extracted with cold MeCN (2 ml) and fast filtered into a glass tube surrounded by cold Et2O in a big glass vial. Purple crystals of 3-BF4 (8.7 mg, 56%) were obtained after crystallization by layering Et2O (8 mL Figure S15) and ATR-IR (powder; Figure S21) spectra are similar to those of complex 3.
Crystallographic Data X-Ray Structure Determination. Crystal data and details of the data collections are given in Tables S1, selected bond lengths and angles in Tables S2, molecular structures are shown in Figures S1-S3, S36 and S43. X-ray data were collected on a STOE IPDS II or a BRUKER D8-QUEST diffractometer (monochromated Mo-Kα radiation, λ = 0.71073 Å) by use of  or  and  scans at low temperature. The structures were solved with SHELXT and refined on F 2 using all reflections with SHELXL. 3,4 Most non-hydrogen atoms were refined anisotropically. Hydrogen atoms were placed in calculated positions and assigned to an isotropic displacement parameter of 1.2 or 1.5 Ueq(C). Face-indexed absorption corrections were performed numerically with the program X-RED 5 or by the multi-scan method with SADABS. 6 In 2, PF6 -was found to be disordered about two positions (occupancy factors: 0.884(2) / 0.116(2)) and Et2O was found to be disordered about a center of inversion (refined at ½ occupancy). EADP constraints and SAME restraints were applied in case of PF6 -and DFIX (d(C-O) = 1.43 Å, d(C-C) = 1.51 Å), SADI (d(H3C···O)), RIGU, SIMU, DELU restraints in case of Et2O.
In 3-BF4, two BF4 -(occupancy factors: 0.695(4) / 0.305(4) & 0.750(4) / 0.250(4)) and one C5H3N group (occupancy factors: 0.718(7) / 0.282 (7)) were found to be disordered. SAME restraints and EADP constraints were applied in case of one BF4 -for the other SADI (d(B-F), d(F···F)) restraints and EADP constraints were applied, for the disordered C5H3N group SADI (d(H2C-C ar )), SAME, SIMU and FLAT restraints. The unit cell of 3-BF4 contains highly disordered solvent molecules (MeCN) for which no satisfactory model for a disorder could be found. The solvent contribution to the structure factors was calculated with PLATON SQUEEZE 7 and the resulting .fab file was processed with SHELXL using the ABIN instruction. The empirical formula and derived values are in accordance with the calculated cell content.

Magnetic Measurements
Temperature-dependent magnetic susceptibility measurements for 2 and 3 were carried out with a Quantum-Design MPMS-3 SQUID magnetometer equipped with a 7.0 T magnet in the range from 300 K to 2 K on a polycrystalline powdered sample under an applied magnetic field of 0.5 T. The crystalline solid sample was contained in a polycarbonate capsule and fixed in a nonmagnetic sample holder. The raw data file for the measured magnetic moment was corrected for the diamagnetic contribution of the capsule according to M dia = χg × m × H, with experimentally obtained gram susceptibilities of the capsule. The molar susceptibility data of the compounds were corrected for the diamagnetic contribution. Experimental data were modelled with the julX

EPR Spectroscopy
Continuous-wave X-band (~9.63 GHz) EPR spectra of 3 were measured on a Bruker Elexsys-E500 spectrometer equipped with an Oxford liquid helium flow cryostat. Spectra were collected in a dual-mode X-band resonator, operated in perpendicular mode (TE102). All spectra were collected with 100 kHz field modulation at 6 G amplitude. All CW-EPR were simulated in Matlab 2022b with the EasySpin (v 6.0.0-dev49) package. 9 Q-band (34.0 GHz) Two-pulse (Hahn) echo detected EPR spectra were collected with a τ/2 -π -τ -echo pulse sequence on a Bruker Elexsys-580 using a home-built up/down Q-band pulse conversion accessory and home-built TE011 microwave resonator. Cryogenic temperatures were maintained with an Oxford C-935 liquid helium cryostat.

DFT Calculations
Computational Details. All calculations were performed by using the ORCA quantum chemical program package (version 5.0.3). 10,11 Geometry optimizations (OPT) were performed with the B3LYP functional using the def2-TZVP basis set in combination with the auxiliary basis set def2/J, [12][13][14][15][16] and the CPCM (for TD-DFT) method has been applied to include solvent effects in the calculations. 17,18 The RIJCOSX approximations was used to accelerate the calculations. 19,20 The noncovalent interactions were considered via atom-pairwise dispersion corrections with Becke-Johnson (D3BJ) damping. 21 Coordinates from X-ray structural analyses (2 and 3-BF4) were used as starting coordinates and the optimized coordinates of 2 and 3 is given in Table S9. TD-DFT calculations (80 roots) were carried out with the B3LYP functional using the same basis set combination.     Table S4 and Figure  S31.  Figure S31. Plot of the TD-DFT difference densities for the electronic excitations (Table S4) calculated for the cation of 3-BF4 (isosurface value = 0.006 au). Purple: negative densities; Orange: positive densities. Figure S32. Comparison of experimental (in acetone, red) and TD-DFT calculated (in acetone, black) UV-vis-NIR spectra for 3-BF4. The calculated spectrum was convoluted using a Gaussian line shape function with a half-width of 70 nm and a shift of +110 nm was applied. The principal contributions involved in the electronic transitions are described in Table S5 and Figure S33.  Figure S33. Plot of the TD-DFT difference densities for the electronic excitations (Table S5) Table S6. For the energy calibration we converted energies of the tabulated peaks to Bragg angles and fit positions of the picked peaks with a tangential function.  15 on all atoms and the AutoAux basis option for ORCA. 26 The calculations employed the resolution of identity (RI-J) algorithm for the computation of the Coulomb terms and the 'chain of spheres exchange' (COSX) algorithm for the calculation of the exchange terms 20 and a tight self-consistent field (SCF) convergence threshold was chosen via the "TightSCF" keyword. Defgrid2 was used during the SCF iterations and for the final energy evaluation after SCF convergence. The conductor-like polarizable continuum model (CPCM) was used for charge compensation in all calculations of complexes carrying a net positive/negative charge. 17 Cu Kβ and S Kβ VtC XES were calculated using the features described above, with the exception that a scalar relativistic basis set (ZORA-def2-TZVP) 15 was employed. In the XES block, "CoreOrb" dictates the orbitals that will be the electron acceptors. All virtual orbitals S31 were chosen as potential acceptor orbitals. For transition metal XES spectra, the metal 1s orbital is usually orbital 0, and is thus selected. OrbOp defines the operators of the electrons (α = 0, β = 1) that will be calculated, both of which were selected. CoreOrbSOC defines which core orbitals (α = 0, β = 1) are treated with spin-orbit coupling

S36
Note: Signals due to complexes 5 and 1 are expected to be observed in the 1 H NMR spectrum of the reaction of 2 with 1 equiv [Me3O]BF4. However, only one set of signals assigned to the macrocyclic ligand was observed, and the chemical shift of each peak is in-between the corresponding peaks of pure samples of 5 and 1 ( Figure S43).
In addition, integration shows the ratio of the macrocycle and µ-SMe is roughly 2 : 1 ( Figure S42). The ESI mass spectrum of the reaction mixture (2 with 1 equiv [Me3O]BF4, 10 min) shows that both major peaks due to 5 and 1 were observed ( Figure S44). Thus, we speculate that there is an interconversion between 1 and 5 via the exchange of µ-SMe (Scheme S2), which is fast on the NMR timescale. The 1 H NMR spectrum of a 1 : 1 mixture sample of pure 5 and 1 shows only one set of signals due to the macrocyclic ligand with chemical shifts similar to those of the reaction mixture of 2 with 1 equiv [Me3O]BF4 ( Figure S43).