A Systematic Study of the Effects of Complex Structure on Aryl Iodide Oxidative Addition at Bipyridyl‐Ligated Gold(I) Centers

Abstract A combined theoretical and experimental approach has been used to study the unusual mechanism of oxidative addition of aryl iodides to [bipyAu(C2H4)]+ complexes. The modular nature of this system allowed a systematic assessment of the effects of complex structure. Computational comparisons between cationic gold and the isolobal (neutral) Pd0 and Pt0 complexes revealed similar mechanistic features, but with oxidative addition being significantly favored for the group 10 metals. Further differences between Au and Pd were seen in experimental studies: studying reaction rates as a function of electronic and steric properties showed that ligands bearing more electron‐poor functionality increase the rate of oxidative addition; in a complementary way, electron‐rich aryl iodides give faster rates. This divergence in mechanism compared to Pd suggests that Ar−X oxidative addition with Au can underpin a broad range of new or complementary transformations.


Materials
All starting materials were obtained from commercial supplied and, except where otherwise state, used as purchased. Anhydrous solvents (CH2Cl2, MeCN, hexane and Et2O) were dried using an Anhydrous Engineering Grubbs-type system (alumina) [1] and stored over 4 Å molecular sieves. Mesitylene and TMEDA were distilled over CaH. DMF, MeOH and acetone were used as purchased. CD2Cl2 was distilled over calcium hydride for NMR spectroscopy.
CDCl3 was used directly as purchased.

Methods
For all reactions performed, unless stated otherwise, inert conditions were employed using standard Schlenk line and/or glove box techniques under an atmosphere of dinitrogen or argon using oven-dried glassware. Room temperature (rt) typically fluctuated between 18 -25 C depending on the season and time of day. In instances where reactions were monitored, they were followed by NMR spectroscopy or analytical thin-layer chromatography (TLC). Merck TLC silica gel 60 F254 plates were used for TLC and were visualised using UV light and/or with a potassium permanganate solution and exposure to heat. Normal phase flash chromatography was carried out using 60 Å silica with solvent systems specified below. 165.8 (C=O), 156.6 (CAr), 150.3 (CAr), 138.8 (CAr), 123.4 (CAr), 120.7 (CAr), 52.9 (CH3). The spectroscopic properties of this compound were consistent with literature data. [6]
The reaction was then heated at 170 C for 17 hours after which was allowed to cool to room temperature and deoxygenated water (30 mL) was added. The resultant slurry was filtered and S7 the collected solid was washed with pentane, air-dried and purified by normal phase flash chromatography (30% to 50% ethyl acetate in hexane) to give 1g (584 mg, 45%) as a colorless solid; 1 H NMR (400 MHz, CDCl3)  8.88 (dd, J = 5.0, 0.9 Hz, 2H, CAr-H), 8.72 (dd, J = 1.6, 0.9 Hz, 2H, CAr-H), 7.60 (dd, J = 5.0, 1.6 Hz, 2H, CAr-H); 13 C{ 1 H} NMR (101 MHz, CDCl3)  155.6 (CN), 150.5 (CAr), 126.0 (CAr), 123.3 (CAr), 122.0 (CAr), 116.5 (CAr). The spectroscopic properies of this compound were consistent with literature data. [8] 6-Methyl-2,2-bipyridyl (1k) Following a modified literature procedure, [9] 2,2-bipyridine (8.00 g, 51.2 mmol) was dissolved in diethyl ether (170 mL) and cooled to 0 C. Methyl lithium (32 mL of a 1.63 M solution in diethyl ether) was added dropwise over 30 minutes. Once addition was complete, the reaction mixture was stirred at room temperature for one hour. After heating at 50 C for a further four hours, water (120 mL) was added once the reaction was at room temperature. The organic portion was separated and the aqueous phase was washed with diethyl ether (3 × 100 mL). The organic extracts were combined, dried (Na2SO4), filtered and evaporated in vacuo to give an orange oil. This was dissolved in a sat. solution of potassium permanganate in acetone (200 mL) and stirred for 16 hours. The reaction mixture was filtered and the filtrate was evaporated in vacuo. The crude material was purified by normal phase flash chromatography (10% to 20% ethyl acetate in hexane) to give 1k (3. properties of this compound were consistent with literature data. [9] S8

2,2-Bipyridyl Au(I) ethylene complex synthesis 2.3.1. General Procedure A -Crystallization of complex
Following a modified literature procedure, [5] in a glovebox, a J. Young's tube was charged with gold(I) chloride (1.0 eq.) and silver(I) triflimide (1.0 eq.) in the absence of light. Outside the glove box, the tube was placed under an atmosphere of ethylene (1 bar) and dichloromethane was added to give a 0.02 M solution. After three hours stirring at room temperature, the reaction mixture was filtered through a pad of Celite into a flask containing the 2,2-bipyridyl ligand (1 eq.) whilst maintaining darkness. After stirring for one hour, the solution was evaporated in vacuo to approx. 2 mL and filtered through a pad of Celite. The filtrate was layered with diethyl ether and stored for 24 hours at −18 C to give the desired Au(I) complex.

General Procedure B -Precipitation of complex
Complexes were prepared in a manner identical to that given in General Procedure A, however the Au(I) complexes were precipitated by adding an excess of diethyl ether to the dichloromethane solution. The mother liquor was removed and the complex was washed thrice with diethyl ether and dried in vacuo.

Bona fide oxidative addition complex synthesis
To verify the presence of the oxidative addition complexes in solution for the kinetic studies, bona fide samples were synthesised according to the General Procedure below. Literature spectroscopic data was used for complexes 4a•NTf2, 4d•NTf2, 5b•NTf2 and 5f•NTf2. [5]

General Procedure C -Oxidative addition complex synthesis
Following a modified literature procedure, [5] gold(I) complex (1.0 eq.) was dissolved in dichloromethane (to give a 0.01 M solution) and the aryl iodide (20 eq.) was added. In the cases where the aryl iodides were solids, they were added at the same time as the gold(I) complex.

S15
The reaction mixture was subjected to three freeze-pump-thaw cycles and heated to 50 C under static vacuum. After one hour, the reaction mixture was again subjected to three freeze-pumpthaw cycles and returned to heat under static vacuum. This process was continued until the reaction was complete by 19 F NMR spectroscopy (typically 2 h). The reaction mixture was filtered (using a Millipore Millex-HV 0.45 m PVDF membrane syringe filter) and evaporated in vacuo to a minimum volume (approx. 0.5 mL). Hexane was added to precipitate the gold(III) oxidative addition complexes which were subsequently washed with hexane and dried under vacuum.

Figure S2
: Thermal ellipsoid plot of the grown structure of complex 2e•NTf2 determined by single-crystal X-ray diffraction. Thermal ellipsoids are shown at the 50% probability level with counterion and hydrogen atoms omitted for clarity. Selected bond lengths (Å) and bond angles (°); Au1-N1 2.1793(17), N1-Au1-N1 74.581, Au1-C1 2.092(2), Au1-C1=C1(centroid) 1.970, C1-C1 1.408(5).   subsequently placed under an atmosphere of dinitrogen. The necessary amount of solvent (CH2Cl2) was measured using 0.1 mL or gas-tight L-syringes. In instances where reagents were liquids, the volume required was measured using a gas-tight L syringe. All Au(I) complex stock solutions were freshly prepared as above for immediate use. Amounts of stock solution required were measured using a gas-tight L syringe where volumes were ≤ 0.1 mL.
Volumes above this were measured using a 0.1 mL syringe. Arrays of 19 F NMR spectra were collected at 50 ± 0.1 C at 470 MHz on a Varian VNMRS 500 MHz spectrometer and were implemented using the standard Varian software. A stock solution of trifluorotoluene was used as an internal standard. Each acquisition had 8 scans per spectrum and a 15 s delay between the end of one spectrum and the start of the next. NMR array data was processed using MestReNova 11.0. Spectrometer probe temperature was periodically calibrated using the ethylene glycol/methanol-thermometer methods. [16]

Representative experiment
Au(I) complex (7.6 mol), trifluorotoluene (7.6 mol) and aryl iodide (0.153 mmol) were added as stock solutions in dichloromethane at −78 C into a J. Young's NMR tube under an atmosphere of dinitrogen. Once the tube had been sealed, and the contents mixed by shaking, it was manually loaded into the NMR spectrometer which had been pre-heated, pre-tuned and pre-shimmed on a dummy sample. The kinetics experiment was started immediately without tuning or shimming. The time between removing the NMR tube from the cold bath and the middle of the first acquisition (i.e., 4 th scan, typically <90 s) and the time between collected arrays (typically <30 s) were measured using a stopwatch.

Kinetic data
In all NMR kinetics experiments, initial rates (i.e., before the system reaches equilibrium) were used to determine the rate of oxidative addition as shown by a representative example in Figure   S5. Initial rates are quoted in Table S3 for the ligands and aryl iodides with errors given as 95% confidence intervals. Figure S5: Representative determination of initial rate from raw 19 F NMR spectroscopic concentration data. Example is for the oxidative addition of parent bipy complex 2a•NTf2 and 4-fluoroiodobenzene (3a). Table S3: Initial oxidative addition rates for ligand and aryl iodide with different substituents

van't Hoff analysis
Using a similar 19 Equilibrium constants at the specified temperatures are given in Table S4 and added to the van't Hoff plot given in Figure S6.  lost in the forward reaction and the determination of its concentration is non-trivial, the reaction equation may be simplified to bipyAu-C2H4 ⇌ bipyAu-ArI for the equilibrium constant approximation. Also, given the reaction is under pseudo-first-order conditions (20 eq.), the concentration of ArI is assumed to be constant.

S30
The enthalpy of reaction (H) may be approximated by extracting the gradient of the above by rearrangement of the van't Hoff equation (Equation 2) which gives rxnH  16 kcal mol −1 .
Where K is the equilibrium constant, H is the enthalpy of reaction, R = 1.987 × 10 −3 kcal K −1 mol −1 is the gas constant, T is the temperature, S is the entropy of reaction.

Aryl iodide kinetic isotope effect
Absolute initial oxidative addition rates with both iodobenzene (3b) and iodobenzene-d5 (3j) were measured in a manner similar to above and the kinetic isotope effect was calculated according to Equation 3. Figure S7: 19 F NMR spectroscopic raw data for oxidative addition of iodobenzene (3b) and iodobenzene-d5 (3j) with bipy complex 2d•NTf2. Dashed lines are data fitted to a 6 th order polynomial arbitrary function.

Counterion and solvent kinetic data
For these experiments related to changing of the Au(I) complex counterion and solvent, initial rates were measured in a manner identical to those described above. In all cases with Au(I) complexes 2d•NTf2, 2d•SbF6 and 2d•BF4, 4-fluoroiodobenzene (3a) was used as the substrate. Initial rate data is given in Table S5. Errors are given as 95% confidence intervals. CH2Cl2

General considerations
All calculations were perfomed using Gaussian 09, Revision D.01. [18] The B97-XD functional [19] with an ultrafine integration grid was used throughout with Ahlrichs' def2-TZVP basis set on Pd, Pt and Au; def2-SVP on C, N, I and def2-SV on all other atoms. [20] The 60electron def2 pseudopotentials were used for Pt and Au; 28-electron def2 pseudopotentials were used for Pd and I. [21] Solvation (dichloromethane) was modelled using the SMD model. [

Comparison with experimental rate data
The oxidative addition barriers (G ‡ ) may be calculated from the observed rates (kobs) given in

Calculated potential energy surface
The calculated potential energy surface given in Figure S10 below broadly follows the same profile as the previously calculated examples. [5] There is a clear preference for the formation of