NMR Spectroscopic Signatures of Cationic Surface Sites from Supported Coinage Metals Interacting with N-Heterocyclic Carbenes

N-heterocyclic carbenes (NHCs) have been extensively studied to modulate the reactivity of molecular catalysts, colloids, and their supported analogues, being isolated sites, clusters, or nanoparticles. While the interaction of NHCs on metal surfaces has been discussed in great detail, showing specific coordination chemistry depending on the type of NHC ligands, much less is known when the metal is dispersed on oxide supports, as in heterogeneous catalysts. Herein, we study the interaction of NHC ligands with Au surface sites dispersed on silica, a nonreducible oxide support. We identify the easy formation of bis-NHC ligated Au(I) surface sites parallel to what is found on metallic Au surfaces. These species display a specific 13C NMR spectroscopic signature that clearly distinguishes them from the mono-NHC Au(I) surface sites or supported imidazoliums. We find that bis-ligated surface species are not unique to supported Au(I) species and are found for the corresponding Ag(I) and Cu(I) species, as well as for the isolobal surface silanols. Furthermore, the interaction of NHC ligand with silica-supported Au nanoparticles also yields bis-NHC ligated Au(I) surface sites, indicating that metal atoms can also be easily extracted from nanoparticles, further illustrating the dynamics of these systems and the overall favorable formation of such bis-ligated species across a range of systems, besides what has been found on crystalline metal facets.

Quantification of the -OH density of the oxide supports was performed through the reaction of [Cu(IMes)2][OTf] were synthesized according to previous reports [5][6][7] .

General Methods
Solution NMR spectra were recorded at room temperature (298 K) on a Bruker 300 or 500 MHz solution state spectrometer equipped with a broadband probe or a broadband cryoprobe ( 13 C, 1 H). 1 H and 13 C chemical shifts are referenced relative to residual solvent peaks 8 .Chemical shifts are reported in parts per million (ppm).Where appropriate, signal multiplicity has been condensed to a single letter format, i.e.: s=singlet, d=doublet, t=triplet, q=quartet, m=multiplet.Solvent signals are denoted accordingly.Unless otherwise specified, 13 C spectra were recorded using 2048 scans, and 1 H spectra were acquired using 64 scans.The Small Molecule Crystallography Center (SMoCC) of ETH Zürich provided the equipment for single crystal structure determination.XRD data was recorded on a Rigaku XtaLAB Synergy-S single-crystal diffractometer.Suitable crystals were selected and tipmounted on a MiTeGen Pin covered with Paratone Oil.Using Olex2 9 , the structure was solved with the SHELXT 10 structure solution program and refined with the SHELXL 11 refinement package using CGLS minimization.The hydrogen atoms were placed in ideal positions and refined as riding atoms.
Elemental Analysis was provided by the in-house Molecular and Biomolecular Analysis Service (MoBiAS) of ETH Zürich and Analytisches Labor Pascher in Remagen, Germany.
Infrared (IR) spectra of the molecular compounds were collected under an inert atmosphere using a Bruker Alpha spectrometer in attenuated total reflectance (ATR) mode, equipped with a diamond ATR module (deuterated triglycine sulfate (DTGS) detector, 2 cm -1 spectral resolution, 4000-400 cm -1 , average of 64 scans).Transmission-IR were recorded on Bruker FT-IR Alpha spectrometer equipped with RockSolid interferometer, DTGS (deuterated triglycine sulfate) detector, SiC globar source; solid samples were mounted on a magnetic pellet holder.A typical experiment consisted of the measurement of transmission in 64 scans in the region from 4000 to 400 cm -1 .Spectra were analyzed using OPUS 8 Software.Spectra of silica-supported samples are normalized to the Si-O-Si overtone peak maximum at 1868 cm - 1 .Diffuse reflectance UV-vis (DRUV) spectra were recorded on an Agilent Cary 5000 UV-Vis-NIR spectrometer equipped with a Praying MantisTM Diffuse Reflection accessory in absolute %R under double beam mode, at a scan rate of 600 nm•min -1 and a resolution of 1 nm.The samples were prepared in a glovebox by grinding the material in an agate mortar.The obtained mixture was layered in an ambient chamber for Praying MantisTM (Harrick), which was sealed under the inert atmosphere of the glovebox.The source changeover was performed at 350 nm and the detector/grating changeover was performed at 860 nm.Reflectance (R) data were converted to Kubelka-Munk (KM) units taking pure PTFE powder as the reference reflector.Transmission electron microscope (TEM) images were recorded on a FEI Talos F200X instrument operated at 200 keV.Powdered samples were mixed in solid form with a Lacey-C 400 mesh Cu grid inside of a glovebox under an atmosphere of Ar before mounted on a vacuum transfer tomography holder from Fischione Instruments (model #2560) inside the glove box which was subsequently transferred to the chamber of the TEM in the absence of air.All given values for size distributions assume a normal distribution.The nanoparticle diameter determination to obtain the particle size distribution (PSD) was done by manual measurement with the standard software ImageJ (version 1.52a).The "±" in the depicted particle size distributions indicates the standard deviation of the mean.

Solid-state NMR experiments
All the solid-state NMR experiments were performed on a Bruker 9.4 T NMR spectrometer, using a 3.2 mm low-temperature MAS probe.For the 13 C{ 1 H} cross polarization magic angle spinning (CPMAS) experiments at room temperature (≈298 K), the MAS rate was 16 kHz, the CP contact time was varied from 2-5 ms as optimized for different samples and the 1 H decoupling power was 120 kHz.For 13

Estimation of J coupling constant by 1D-INADEQUATE experiments
In INADEQUATE experiments, the refocused delay in a spin echo is usually set equal to 1/4J for the creation of DQ coherence, where J is the average spin-spin coupling constant.By analyzing the dependence of signal intensity on the refocused delay in 1D INADEAUQTE, the average coupling constant J can be possibly estimated.In addition, it was shown that the dependence on the refocused delay in the spin echo can be complicated by the chemical shift difference, and expressed as followed 13,14 :
To the suspension was added dropwise a solution of Mes2Mg(THF)2 (77.8 mg, 0.186 mmol) in THF (ca. 3 mL).The mixture was stirred for 20 h before 2 drops of dioxane were added to precipitate [MgCl2(C4H8O2)]n.After filtering the reaction mixture through Celite® solvent was removed under reduced pressure to yield a white product.The crude product was dissolved in toluene (5 mL) and cooled to -35 o C for 24 hours.Colorless cubic crystals were obtained in a yield of 62% (73 mg, 0.115 mmol).XRD quality crystals were obtained from recrystallization in toluene.Elemental Analysis C: 58.51% H: 5.33% N: 4.70% (Calc.: 58.06%/ 5.68%/ 4.51%).
To the suspension was added NaO t Bu (21.5 mg, 0.22 mmol) in THF (ca. 3 mL).The mixture was stirred for 2 h.After filtering the reaction mixture through Celite® the solution was taken to dryness, leaving a pale-orange solid.The crude product was dissolved in toluene, filtered, dried, and finally washed with pentane to obtain a yield of 70% (57 mg, 0.12 mmol).XRD quality crystals were obtained from recrystallization in toluene.Elemental Analysis C: 52.01%H: 7.04% N: 3.75% (Calc.: 51.82% / 6.72% / 3.66%).

1-Au = [(IMes)Au]/SiO2
[Au(IMes)Mes] (65.2 mg, 0.11 mmol) is dissolved in 3 mL benzene.SiO2-700 (280 mg, 0.1 mmolOH) are added to this solution.The mixture is stirred at 120 rpm at room temperature for 6 hours.The solid is then allowed to decant and the supernatant pipetted out.The solid is then washed intermittently 3 times with 5 mL benzene and 5 mL of pentane, dried under vacuum, then high vacuum to afford a white powder.

2-Au = [(IMes)2Au]/SiO2-700
IMes (30 mg, 0.10 mmol) is dissolved in 3 mL benzene.1-Au (280 mg) are added to this solution.The mixture is stirred at 120 rpm at room temperature for 3 hours.The solid is then allowed to decant and the supernatant pipetted out.The solid is then washed intermittently 3 times with 5 mL benzene and 5 mL of pentane, dried under vacuum, then high vacuum to afford a white powder.

1-Ag = [(IMes)Ag]/SiO2
The procedure was adapted from our previous work 5 .[Ag(IMes)Mes] (58.3 mg, 0.11 mmol) is dissolved in 3 mL benzene.SiO2-700 (280 mg, 0.1 mmolOH) are added to this solution.The mixture is stirred at 120 rpm at room temperature for 3 hours.The solid is then allowed to decant and the supernatant pipetted out.The solid is then washed intermittently 3 times with 5 mL benzene and 5 mL of pentane, dried under vacuum, then high vacuum to afford a white powder.

2-Ag = [(IMes)2Ag]/SiO2-700
IMes (30 mg, 0.10 mmol) is dissolved in 3 mL benzene.1-Ag (280 mg) are added to this solution.The mixture is stirred at 120 rpm at room temperature for 3 hours.The solid is then allowed to decant and the supernatant pipetted out.The solid is then washed intermittently 3 times with 5 mL benzene and 5 mL of pentane, dried under vacuum, then high vacuum to afford a white powder.

1-Cu = [(IMes)Cu]/SiO2
The procedure was adapted from our previous work 6 .[Cu(IMes)Mes] (53.5 mg, 0.11 mmol) is dissolved in 3 mL benzene.SiO2-700 (280 mg, 0.1 mmolOH) are added to this solution.The mixture is stirred at 120 rpm at room temperature for 3 hours.The solid is then allowed to decant and the supernatant pipetted out.The solid is then washed intermittently 3 times with 5 mL benzene and 5 mL of pentane, dried under vacuum, then high vacuum to afford a white powder.

2-Cu = [(IMes)2Cu]/SiO2-700
IMes (30 mg, 0.10 mmol) is dissolved in 3 mL benzene.1-Cu (280 mg) are added to this solution.The mixture is stirred at 120 rpm at room temperature for 3 hours.The solid is then allowed to decant and the supernatant pipetted out.The solid is then washed intermittently 3 times with 5 mL benzene and 5 mL of pentane, dried under vacuum, then high vacuum to afford a white powder.

1*-H = IMes*/SiO2-700
To a solution of IMes* (30 mg; 0.1 mmol) in 5 mL benzene is added SiO2-700 (280 mg).The mixture is stirred at 120 rpm at room temperature for 6h.The solid is then allowed to decant and the supernatant pipetted out.The solid is then washed intermittently 3 times with 5 mL benzene and 5 mL of pentane, dried under vacuum, then high vacuum to afford a white powder.

2*-H = [(IMes*)2H]/SiO2-700
IMes (30 mg, 0.10 mmol) is dissolved in 3 mL benzene.1-Cu (280 mg) are added to this solution.The mixture is stirred at 120 rpm at room temperature for 3 hours.The solid is then allowed to decant and the supernatant pipetted out.The solid is then washed intermittently 3 times with 5 mL benzene and 5 mL of pentane, dried under vacuum, then high vacuum to afford a white powder.

IMes*-AuNPs/SiO2
To a solution of IMes* (18 mg; 0.06 mmol) in 5 mL benzene is added Au/SiO2 (150 mg).The mixture is stirred at 120 rpm at room temperature for 24 h.The solid is then allowed to decant and the supernatant pipetted out.The solid is then washed 3 times with 5 mL benzene, 3 time with 5 ml pentane, dried under vacuum, then high vacuum for 30 min to afford a red powder.

XRD Reports
The following pages contain detailed crystallographic information on the newly synthesized molecules.
Figure S1.The studied molecular compounds.

Figure S4 .
Figure S4.Solution state 13 C NMR spectrum of the product resulting from reaction of 1 equiv.[Au(IMes)(OTBOS)]and 1 equiv. of IMes*.A peak at 184 ppm suggests the formation of the bis-NHC Au species.

Figure S6 .
Figure S6.1D C INADEQUATE spectra of 2*-Au collected under different refocused delay times (see legend).The spectra were collected at ~110 K and MAS rates of 10 kHz.

Figure S7 .
Figure S7.(a) The normalized integrated area of the peak at 184 ppm in the 1D-INADEQUATE spectrum of 2*-Au with various refocused delays (circle).The calculated results corresponding to different J-coupling constants (18 -32 Hz) with Δ=310 Hz were indicated by lines.(b) The mean square deviation (MSD) between calculated and experimental results shown in (a) indicating the estimated 13 C-13 C J coupling constant in 2*-Au is 26 ± 3 Hz.

Table of Content 1. Experimental Procedures
Clabeled IMes* ligand instead of IMes or 1*-Au instead of 1-Au.