A dicoordinate gold(i)–ethylene complex

The use of the exceptionally bulky tris-2-(4,4′-di-tert-butylbiphenylyl)phosphine ligand allows the isolation and complete characterization of the first dicoordinate gold(i)–ethylene adduct, filling a missing fundamental piece on the organometallic chemistry of gold. Besides, the bonding situation of this species has been investigated by means of state-of-the-art Density Functional Theory (DFT) calculations indicating that π-backdonation plays a minor role compared with tricoordinate analogues.

Complex 2 (53 mg, 0.05 mmol) was transferred into a small glass vial and dissolved in dichloromethane (2 mL). The vial solution was loaded into a plastic syringe equipped with a stainless steel needle. Outside the glovebox, the Schlenk flask was cooled down to -30ºC. At this temperature the solution of complex 2 was added to the AgSbF6 suspension while bubbling ethylene. The mixture was allowed to slowly warm up to room temperature, filtered through a short pad of Celite to remove the silver salts, and the solvent was removed under vacuum affording complex 4 (53 mg, 83%). Crystals suitable for X-ray diffraction were grown by slow diffusion of pentane in a concentrated dichloromethane solution. Anal. Calcd. for C62H79AuF6PSb: C, 57.82, H, 6.18.
The solvent was then removed under vacuum. The crude solid was dissolved in CH2Cl2 (10 mL) and filtered through a short pad of Celite.
Complex 2 (53 mg, 0.05 mmol) was transferred into a small glass vial and dissolved in dichloromethane (2 mL). The vial solution was loaded into a plastic syringe equipped with stainless steel needle. Outside the glovebox, the Schlenk flask was cooled down to -30 ºC. At this temperature the solution of complex 2 was added to the AgSbF6 suspension while bubbling CO. The mixture was allowed to slowly warm up to room temperature. Then the mixture was filtered through a short pad of Celite to remove the silver salts affording complex 6 in quantitative NMR yield. Crystals suitable for X-ray diffraction were grown by slow diffusion of pentane into a concentrated dichloromethane solution. 1  (C3), 35.4 (C7), 31.7 (C(CH3)3 ( t Bu7)), 31.6 (C(CH3)3 ( t Bu3)). 31 P{ 1 H} NMR (162 MHz, CD2Cl2, 25 ºC) :               Figure S17. 13  were performed as indicated: In a Young NMR tube complex 2 (10 mg, 0.01 mmol) and AgSbF6 (7 mg, 0.02 mmol) were dissolved in CD2Cl2 (0.5 mL). The tube was charged with 1 bar of a mixture of ethylene (0.5 bar) and CO (0.5 bar) and intensely shaken for half an hour. The reaction was monitored by 1 H and 31 P NMR spectroscopy until equilibrium conditions were reached ( Figure S25).
Similarly, a solution of complex 5 (19 mg, 0.01 mmol) in CD2Cl2 (0.5 mL) was charged either with 1 bar of ethylene or 1 bar of CO and intensely shaken for half an hour. The reactions were both monitored by 1 H and 31 P NMR spectroscopy until equilibrium conditions were reached ( Figure S25).
The equilibrium constants calculated accordingly: were collected by means of ω and φ scans using monochromatic radiation λ(Mo Kα1) = 0.71073 Å. The diffraction images collected were processed and scaled using APEX-II or APEX-III software, respectively.
The structures were solved with SHELXT and was refined against F 2 on all data by full-matrix least squares with SHELXL. 4 All non-hydrogen atoms were refined anisotropically. Hydrogen atoms were included in the model at geometrically calculated positions and refined using a riding model, unless otherwise noted.
In compounds 4 and 4' the hydrogens associated to the bound ethylene ligands were calculated using the AFIX 23 command. The isotropic displacement parameters of all hydrogen atoms were fixed to 1.2 times the U value of the atoms to which they are linked (1.5 times for methyl groups). The nature of the phosphine ligand, containing six tert-butyl groups, lead to several features that are common to most measured structured. First, the highly disordered tert-butyl groups along with the higher libration of their terminal methyl groups led us to model the disorder for the worse behaved fragments, while restraints were applied to the corresponding ADPs. Besides, the presence of these groups generates voids during the packing that is occupied by solvent molecules. In four of the six reported structures we used the program SQUEEZE to compensate for the contribution of disordered solvents, which account for 14 pentane molecules (3), 4 pentane molecules (4), 6 pentane molecules (5) and 4 pentane molecules (6), in the unit cell. In compound 3 Compound 3 was refined as an inversion twin. We could not obtain good quality crystals for compound 5 despite many attempts. However, we could grow crystals of enough quality for X-ray diffraction studies using NaBArF (BArF -= [B(C6H2-3,5-(CF3)2)4] -) instead of AgSbF6 as chloride abstractor. Thus, the reported crystal structure contains the latter anion.
The full numbering scheme of all the reported structures can be found in the full details of the X-ray structure determination (CIF), which is included as Supporting Information.   The molecular structure of complexes 5 ( Figure S30) and 6 ( Figure 2) was further confirmed by X-ray diffraction studies. The two complexes present similar molecular structures with the gold centre adopting a linear geometry with P-Au-N and P-Au-C angles of 178.96(10)° and 176.6(2)°, respectively. The C-O bond length of 1.075(9) Å in complex 6, shortened with respect to free CO (dCO = 1.13 Å), is in agreement with those described in the literature for the very few related gold(I)-carbonyl compounds. Similarly to 4, complex 6 co-crystalized with a silver cation, which linked two gold(I)-CO adducts by coordinating to two ortho-aryl groups of two different phosphine ligands to form a 1D-polymeric structure. Once more, the presence of silver in the bulk sample of 6 is minimal: the reported solid-state structure derives from the considerably improved capacity of silver to provide crystals of good quality. In fact, we have measured other silver-free structures of 6 of noticeably poorer crystalline quality that exhibit virtually identical geometrical parameters for the gold-carbonyl fragment.

4'
S25 Figure S30. ORTEP diagram of compound 5. Counteranion, solvent molecules and hydrogen atoms are excluded for clarity, while tert-butyl groups and one biaryl fragment are represented in wireframe format.
Thermal ellipsoids are set at 50% probability.

Buried volume analysis
The steric description of per cent buried volume (%Vbur) has been shown to be a valid measure of the steric properties of monodentate ligands such as phosphines and NHCs. Comparison of related highly bulky ligands IPr** 5 and phosphine 1 is shown in Figure S28

Computational details
Calculations were performed at the DFT level with the Gaussian 09 (Revision D.01) program. 7 The hybrid functional PBE0 8 was used throughout the computational study, and dispersion effects were accounted for by using Grimme's D3 parameter set with Becke−Johnson (BJ) damping at the optimisation stage. 9 Geometry optimisations were carried out without geometry constraints, using the 6-31G(d,p) 10 basis set to represent the C, H and P atoms and the Stuttgart/Dresden Effective Core Potential and its associated basis set (SDD) 11 to describe the Au atoms. Bulk solvent effects (dichloromethane) were included at the optimization stage with the SMD continuum model 12  311+G(2d,p) basis set and free energies were corrected (ΔGqh) to account for errors associated with the harmonic oscillator approximation with the Goodvibes code. 13 Thus, according to Truhlars's quasiharmonic approximation for vibrational entropy, all vibrational frequencies below 100 cm -1 were set to this value. 14 The NMR shieldings were calculated with the Gauge-Independent Atomic Orbital (GIAO) 15 method at the PBE0/6-311+G(2d,p)//PBE0/6-31G(d,p) level. The CYLview20 visualization software has been used to create some of the figures. 16

Calculation of 1 H NMR chemical shifts
The average calculated 1   Phosphine 1 has been represented as P for clarity.

Energy Decomposition Analysis Calculations
To enable a direct comparison with the available bonding analyses on related gold(I) complexes (see main text), the geometry of complex 4 (as well as complexes A and B) was re-optimized at the dispersion corrected BP86 17 -D3 18 /def2-SVP 19 level. The interaction between the transition metal fragment and ethylene has been investigated with the EDA-NOCV method, 20 which combines the energy decomposition analysis (EDA) 21 with the natural orbitals for chemical valence (NOCV) 22 methods. Within this approach, the interaction energy can be decomposed into the following physically meaningful terms: The term ΔEelstat corresponds to the classical electrostatic interaction between the unperturbed charge distributions of the deformed reactants and is usually attractive. The Pauli repulsion ΔEPauli comprises the destabilizing interactions between occupied orbitals and is responsible for any steric repulsion. The orbital interaction ΔEorb accounts for charge transfer (interaction between occupied orbitals on one moiety with unoccupied orbitals on the other, including HOMO-LUMO interactions) and polarization (empty-occupied orbital mixing on one fragment due to the presence of another fragment). Finally, the ∆Edisp term takes into account the interactions which are due to dispersion forces.
The EDA-NOCV method makes it possible to further partition the total orbital interactions into pairwise contributions of the orbital interactions. Details of the method can be found in the literature. 23 The EDA-NOCV calculations were carried out using the BP86-D3/def2-SVP optimized geometry with the program package ADF 2020 24 using the same functional (BP86-D3) in conjunction with a triple-ζ-quality basis set using uncontracted Slater-type orbitals (STOs) augmented by two sets of polarization function with a frozen-core approximation for the core electrons. 25 An auxiliary set of s, p, d, f, and g STOs were used to fit the molecular densities and to represent the Coulomb and exchange potentials accurately in each SCF cycle. 26 Scalar relativistic effects were incorporated by applying the zeroth-order regular approximation (ZORA). 27 This level of theory is denoted BP86-D3/TZ2P//BP86-D3/def2-SVP.