Stable Au^III complexes with four N-heterocyclic carbene groups can be prepared in high yield directly from KAuCl₄

Abstract

Gold(III) N-heterocyclic carbene (NHC) complexes of form [Au(NHC)₄Cl₂]Cl were synthesized by reaction of KAuCl₄ with bis- and tetrakis(imidazolium) salts in the presence of a mild base. Treatment of these complexes with KPF₆ afforded four-coordinate Au^III complexes of form [Au(NHC)₄](PF₆)₃. X-Ray crystallography showed the [Au^III(NHC)₄]³⁺ cations in the hexafluorophosphate salts to have a square planar Au(NHC)₄ moiety [Au⋯C_NHC 2.024(4)–2.082(7) Å]. In the [Au^III(NHC)₄Cl₂]⁺ cations in the chloride salts, coordination about Au was tetragonally-distorted octahedral, the axial Au–Cl bonds being substantially longer [Au⋯Cl 3.148(2)–3.693(1) Å] than the equatorial Au–C_NHC bonds [Au⋯C 2.024(4)–2.082(7) Å]. NMR and conductance studies suggested that the structures of the complexes seen in the solid state persisted in DMSO solution, except in one case where a chlorido ligand dissociated from [Au^III(NHC)₄Cl₂]⁺ to form [Au^III(NHC)₄Cl]²⁺. The Au^III(NHC)₄ unit was surprisingly robust. An Au^III complex was found to undergo H/D exchange reactions in D₂O solution at 100 °C with no signs of decomposition detectable by ¹H NMR spectroscopy. ¹H NMR studies showed that various complexes containing Au^III(NHC)₄ moieties underwent little or no decomposition when heated at 120 °C in DMSO-d₆ for extended periods.

---

Introduction

Interest in gold chemistry has been growing, in part due to the promise of applications in fields of medicine, nanotechnology and catalysis.^1–5 For example, many gold compounds have been found to inhibit cancer cell growth,^1–3 and so gold complexes are being investigated as potential replacements for or alternatives to cisplatin in cancer chemotherapy. In this context, there has been increased research activity on the synthesis of new gold complexes involving N-heterocyclic carbene (NHC) complexes. These studies have mostly focused on mononuclear Au^I–NHC complexes, with considerably less attention being paid to Au^III–NHC complexes.^6,7

The majority of Au complexes known to date are complexes of Au^I. Synthesis of Au^III complexes is often problematic because of the highly oxidizing nature of Au^III. Attempts to synthesize Au^III complexes from Au^III salts frequently results in side-reactions during which reduction of Au^III occurs.^8–12 Not surprisingly, the most common synthetic pathway to Au^III–NHC complexes involves oxidative addition of halogens or halogen equivalents to Au^I–NHC precursor complexes.^8,9,13–31 Other procedures, such as carbene transfer between Ag^I and Au^III,⁸ or reaction of imidazolium ions with Au^III in the presence of base,^9,10 are usually complicated by reduction of Au^III to Au^I. However, Lu et al. synthesized the Au^III–tetracarbene complex 1 by treatment of HAuCl₄ with an Ag^I–NHC complex.³² Most Au^III–NHC complexes reported to date have square-planar Au^III centres with one or two NHC ligands, 1 being the only Au(NHC)₄ motif reported to date. While the Au^III centre in 1 is square planar, many complexes of Au^III with other ligands contain square pyramidal and octahedral Au centres (e.g. [Au(phen)(CN)₂Br] and [Au(diars)₂I₂]⁺), albeit strongly distorted, the axial metal–ligand bonds being longer than the equatorial ones.^33,34

In this paper, we report six new complexes containing the Au^III(NHC)₄ motif. The complexes were prepared simply by treatment of KAuCl₄ with imidazolium salts in the presence of acetate. Depending on the ancillary ligands, the new complexes contain four- or six-coordinate Au^III centres, and all have been thoroughly characterised by NMR and X-ray studies.

Results and discussion

Synthesis of Au^III–NHC complexes

Au^III–NHC chlorido complexes [Au(L¹)₂Cl₂]Cl 2, [Au(L²)₂Cl₂]Cl 3, and [AuL³Cl₂]Cl 4 (Scheme 1) were synthesized by heating KAuCl₄ with the appropriate imidazolium salt and LiOAc in DMF, following procedures we used previously to synthesize Au^I–NHC cyclophane complexes.¹ The complexes [Au(L¹)₂Cl₂]Cl 2, [Au(L²)₂Cl₂]Cl 3 and [AuL³Cl₂]Cl 4 precipitated from the mixture and were purified by recrystallisation; yields after recrystallisation were 81, 79, and 48% respectively. These yields are surprisingly high, as previous attempts so synthesize Au^III–NHC complexes by reaction of imidazolium salts with Au^III salts in the presence of base were accompanied by a substantial amount of reduction of Au^III to Au^I,^9,10 the problem being exacerbated by long reaction times.¹⁰ We attribute these high yields in part to remarkable stability of the Au^III–tetrakis(NHC) complexes (see NMR studies, below). Interestingly, the [Au(NHC)₄Cl₂]Cl structure seems to be the preferred product even when the stoichiometry is unfavourable—when we allowed L¹·2HCl and KAuCl₄ to react in a 1 : 1 mole ratio in the presence of excess LiOAc, the only product obtained was [Au(L¹)₂Cl₂]Cl 2 (86% yield based on L¹·2HCl). The structure of the starting imidazolium salt (i.e., whether it contains a mono-, bis-, or tetrakis(imidazolium) ion) is also an important factor in determining the outcome of the reactions. The use bis- or tetrakis(imidazolium) salts as precursors makes formation of the Au^III(NHC)₄ moiety kinetically favourable compared to competing processes in which Au(III) participates in redox reactions. When we attempted syntheses using simple monoimidazolium salts (1,3-dimethylimidazolium iodide and 1,3-diisopropylimidazolium bromide), we did not obtain any Au^III(NHC)₄-type complex, instead obtaining small amounts of [(NHC)AuX₃], Au^I–NHC complexes, and the urea-type product of oxidation of the imidazolium ion at C2. Other researchers have also reported poor results from attempts to synthesize Au^III complexes directly from monoimidazolium ions and Au^III salts in the presence of base.^9,10

Scheme 1 Synthesis of Au^III–NHC complexes.

The PF₆⁻ salts [Au(L¹)₂](PF₆)₃5, [Au(L²)₂](PF₆)₃6, and [AuL³](PF₆)₃7 were obtained from the corresponding chlorides [Au(L¹)₂Cl₂]Cl 2, [Au(L²)₂Cl₂]Cl 3, and [AuL³Cl₂]Cl 4 respectively, by metathesis with KPF₆ in methanol, and were obtained as analytically-pure, white powders, in quantitative yields.

X-ray studies

Results of X-ray diffraction studies of the Au^III complexes are summarized in Tables 1 and 2 and Fig. 1–3. The chlorido complexes [Au(L¹)₂Cl₂]⁺, [Au(L²)₂Cl₂]⁺ and [AuL³Cl₂]⁺ each contain an Au^III centre in a strongly distorted octahedral environment, and are examples of rarely-encountered 20-electron complexes. Removal of the chlorido ligands results in [Au(L¹)₂]³⁺, [Au(L²)₂]³⁺ and [AuL³]³⁺, which might be thought of as structures containing commonly encountered four-coordinate Au^III centres (16-electron complexes), but the picture is complicated by fluorines from PF₆⁻ anions occupying axial coordination sites.

Fig. 1 (a) Crystal structure (50% probability level for the displacement ellipsoids) of the cation of [Au(L¹)₂Cl₂]Cl 2. Selected bond lengths (Å) and angles (°): Au(1)–C(11) 2.040(9), Au(1)–C(41) 2.043(9), Au(1)–C(31) 2.046(9), Au(1)–C(21) 2.050(9), Au(1)–Cl(1) 3.148(2), Au(1)–Cl(2) 3.192(2), C(11)–Au(1)–C(41) 179.3(3), C(41)–Au(1)–C(31) 85.0(4), Cl(1)–Au(1)–Cl(2) 176.78(6). (b) Crystal structure (50% probability level for the displacement ellipsoids) of the cation of [Au(L¹)₂](PF₆)₃5. PF₆ anions have been omitted for clarity. Selected bond lengths (Å) and angles (°): Au(1)–C(21) 2.041(2), Au(1)–C(11) 2.045(2), C(21)–Au(1)–C(11) 83.31(9).

Fig. 2 (a) Crystal structure (50% probability level for the displacement ellipsoids) of the cation of [Au(L²)₂Cl₂]Cl 3. Selected bond lengths (Å) and angles (°): Au(1)–C(11) 2.024(4), Au(1)–C(21) 2.054(4), Au(1)–Cl(1) 3.433(1), Au(1)–Cl(2) 3.693(1), C(11)–Au(1)–C(31) 174.87(17), C(11)–Au(1)–C(21) 89.84(17), Cl(1)–Au(1)–Cl(2) 172.17(3). (b) Crystal structure (50% probability level for the displacement ellipsoids) of the cation of [Au(L²)₂](PF₆)₃6. PF₆ anions have been omitted for clarity. Selected bond lengths (Å) and angles (°): Au(1)–C(11) 2.0402(18), Au(1)–C(21) 2.062(2), C(11)–Au(1)–C(21) 88.22(8).

Fig. 3 (a) Crystal structure (30% probability level for the displacement ellipsoids) of the cation of [AuL³Cl₂]Cl 4. Selected bond lengths (Å) and angles (°): Au(1)–C(11) 2.048(8), Au(1)–C(31) 2.082(7), Au(1)–Cl(3) 3.017(4), C(11)–Au(1)–C(31) 95.5(3). (b) Crystal structure (50% probability level for the displacement ellipsoids) of the cation 1 of [AuL³](PF₆)₃7. (Cation 2 is similar.) Selected bond lengths (Å) and angles (°): Au(1)–C(41) 2.051(2), Au(1)–C(31) 2.052(2), Au(1)–C(11) 2.056(2), Au(1)–C(61) 2.056(2), Au(2)–C(71) 2.050(2), Au(2)–C(91) 2.055(2), Au(1)–F(15) 3.004(5), Au(1)–F(23) 3.089(2), Au(2)–F(33) 3.903(2), C(41)–Au(1)–C(31) 84.77(8), C(41)–Au(1)–C(11) 178.57(8), F(15)–Au(1)–F(23) 179.47(8), C(71)–Au(2)–C(91) 93.94(8).

Table 1 Crystal data of new Au^III compounds

Complex	[AuL^1Cl2]Cl·(MeOH)22	[AuL^2Cl2]Cl·(MeOH) 3	[AuL^3Cl2]Cl·(MeOH)3.54	[AuL^1](PF6)3·(MeCN)25	[AuL^2](PF6)3·(MeCN)26	[AuL^3](PF6)3·(MeCN)27
Empirical formula	C20H32AuCl3N8O2	C21H32AuCl3N8O	C33.50H42AuCl3N8O3.50	C22H30AuF18N10P3	C24H34AuF18N10P3	C34H34AuF18N10P3
Formula weight	719.85	715.86	916.07	1066.43	1094.49	1214.59
Wavelength/Å	0.71073	0.71073	1.54184	0.71073	0.71073	0.71073
Crystal system	Monoclinic	Monoclinic	Monoclinic	Monoclinic	Monoclinic	Triclinic
Space group	P21/c	P21/c	I2/m	C2/c	C2/c	P1̄
a/Å	11.2465(4)	8.4520(3)	11.5286(4)	19.6697(5)	19.5075(5)	9.50080(10)
b/Å	12.5742(5)	21.5282(5)	18.4128(5)	11.0832(2)	11.1838(2)	15.6009(2)
c/Å	18.5100(7)	14.1226(4)	16.9015(6)	17.0857(6)	18.0374(4)	21.6100(3)
α/°						95.646(1)
β/°	100.893(4)	101.964(3)	98.985(3)	113.018(4)	113.442(3)	96.197(1)
γ/°						94.683(1)
V/Å^3	2570.44(17)	2513.88(13)	3543.7(2)	3428.18(19)	3610.39(16)	3155.08(7)
Z	4	4	4	4	4	3
ρ(calc)/Mg m^−3	1.860	1.891	1.717	2.066	2.014	1.918
μ/mm^−1	6.070	6.203	10.284	4.565	4.338	3.734
Crystal size/mm^3	0.638 × 0.230 × 0.069	0.186 × 0.127 × 0.080	0.147 × 0.116 × 0.057	0.48 × 0.31 × 0.22	0.40 × 0.285 × 0.245	0.355 × 0.247 × 0.191
θ range for data collection/°	2.455 to 27.0	2.463 to 32.562	3.574 to 67.329	2.276 to 37.068	2.147 to 37.625	1.708 to 30.066
Reflections collected	24 788	31 630	17 415	31 084	43 319	71 746
Independent reflections	14 559	8553	3272	8533	9243	18 510
R(int)	0.1079	0.0591	0.0482	0.0436	0.0423	0.0379
Max./min. transmission	0.665 and 0.204	0.667 and 0.429	0.630 and 0.369	0.644 and 0.417	0.504 and 0.357	1.000/0.741
Restraints/parameters	0/315	0/313	166/247	0/267	12/415	0/971
Goodness-of-fit on F^2	1.105	1.092	1.117	1.124	1.047	1.060
R 1 [I > 2σ(I)]	0.0628	0.0460	0.0546	0.0302	0.0263	0.0252
wR2 [I > 2σ(I)]	0.1842	0.0765	0.1430	0.0660	0.0531	0.0552
R 1 (all data)	0.0712	0.0681	0.0691	0.0437	0.0475	0.0316
wR2 (all data)	0.1887	0.0819	0.1541	0.0711	0.0604	0.0576
Δρ(max/min)/e Å^−3	4.765 and −2.329	2.330 and −0.797	2.749 and −2.259	2.244 and −1.084	1.443 and −1.440	2.804 and −0.744
CCDC number	1528857	1528858	1528859	1528860	1528861	1528862

---

Table 2 Selected bond lengths (Å) and angles (°) for Au^III cations^a

Complexes	Au–C	Au–Cl or Au–FPF5	C–Au–C^b	C–Au–C^c	Cl–Au–Cl	φ	φ′	θ	θ′
a φ = NNHC–C–NNHC; φ′ = NNHC–C–Carene; θ = AuC4/NHC inter-planar angle; θ′ = AuC4-arene inter-planar angle. b Angle between NHCs directly linked by CH2 or (CH2)2 group. c Angle between NHCs not directly linked by CH2 or (CH2)2 group.
[AuL^1Cl2]Cl 2	2.040(9)	3.148(2)	85.0(4)	179.3(3)	176.78(6)	108.6(7)	—	40.6(3)	—
	2.050(9)	3.192(2)	84.5(4)	178.9(4)		109.0(7)		41.1(3)
	2.043(9)			96.0(3)				39.8(3)
	2.046(9)			94.5(4)				38.7(3)
[AuL^1]3PF65	2.045(2)	3.380(2)	83.31(9)	180	—	108.0(2)	—	43.0(1)	—
	2.041(2)							43.7(1)
[AuL^2Cl2]Cl 3	2.024(4)	3.433(1)	89.84(17)	172.65(18)	172.17(2)	—	—	54.5(2)	—
	2.054(4)	3.693(1)	91.40(18)	174.87(17)				46.7(2)
	2.037(4)			90.52(17)				53.5(2)
	2.061(4)			88.88(17)				46.3(2)
[AuL^2]3PF66	2.0402(18)	3.394(3)	88.22(8)	180	—	—	—	57.6(1)	—
	2.062(2)							48.8(1)
[AuL^3Cl2]Cl 4	2.048(8)	3.016(4)	84.5(3)	180	180	110.6(3)	109.9(7)	36.4(4)	64.3(3)
	2.082(7)			95.5(3)			109.8(7)	36.4(3)
[AuL^3]3PF67	2.056(2)	3.004(5)	84.77(8)	178.57(8)	—	107.9(2)	113.0(2)	34.6(1)	74.1(1)
(Molecule 1)	2.052(2)	3.089(2)	85.63(8)	179.75(8)		107.6(2)	112.1(2)	39.4(1)	63.4(1)
	2.051(2)			94.52(8)		109.3(2)		36.8(1)
	2.056(2)			95.08(8)		107.8(2)		35.7(1)
(Molecule 2)	2.050(2)	2.903(4)	86.06(8)	180	—	108.1(2)	111.2(2)	36.7(1)	63.6(1)
	2.055(2)			93.94(8)		109.92(2)		36.4(1)
1 ^32	2.017(9)	—	86.0(4)	174.5(4)	—	109.6(7)	—	29.8(3)	—
	2.042(10)		86.0(4)	174.5(4)		111.4(8)		45.3(3)
	2.052(9)		94.4(3)					42.7(3)
	2.030(11)		94.3(3)					33.7(4)

---

In each of the six complexes, the NHC units comprise an approximate square planar array around the Au^III centre. Au–C distances are in the range 2.024(4)–2.082(7) Å across the series of complexes, and are similar to the values seen for other complexes of form [Au^III(NHC)₄]³⁺ (1 in Table 2),³² [Au^III(NHC)(halide)₃],^8,35 and [Au^III(NHC)₂(halide)₂]⁺.^8,13,35 In [Au(L¹)₂Cl₂]Cl·(MeOH)₂2·(MeOH)₂, [Au(L²)₂Cl₂]Cl·(MeOH) 3·(MeOH), and [AuL³Cl₂]Cl·(MeOH)_3.54·(MeOH)_3.5, two Cl⁻ ligands occupy additional (axial) coordination sites, Au–Cl_axial being in the range 3.148(2)–3.693(1) Å, so that overall coordination geometry is distorted octahedral. These Au–Cl_axial bond distances are about 0.9 Å longer than Au–Cl_equatorial (∼2.27 Å) bond lengths seen in [Au(HHCT)Cl₂]AuCl₄ (HHCT = 1,8-bis(hydroxyethyl)-1,3,6,8,10,13-hexaazacyclotetradec-ane), [Au(dien)C1₃] (dien = diethylenetriamine), etc.,^36–38 but are much shorter than the sum of the van der Waals radii of the atoms (∼4.1 Å for Au⋯Cl).³⁹ This type of distorted octahedral structural motif has been reported^34,36,38,40 in only a few cases for Au^III (Table S1†), and is presumably a consequence of these complexes being 20 electron complexes. For [AuL³Cl₂]Cl·(MeOH)_3.54·(MeOH)_3.5, the coordination of Au^III is broadly similar to that seen in [Au(L¹)₂Cl₂]Cl·(MeOH)₂2·(MeOH)₂ and [Au(L²)₂Cl₂]Cl·(MeOH) 3·(MeOH), but is complicated by the effects of solvation, the axial sites modelled as being partially occupied by chlorido ligands and solvent (methanol) molecules.

[Au(L¹)₂](PF₆)₃·(MeCN)₂5·(MeCN)₂ and [AuL³](PF₆)₃·(MeCN)₃7·(MeCN)₃ have similar coordination environments around Au to those seen in the chlorido series, but with fluorine atoms from PF₆⁻ groups occupying the axial coordination sites (Au^III–F 2.92(3)–3.380(2) Å). For [Au(L²)₂](PF₆)₃·(MeCN)₂6·(MeCN)₂, the coordination environment about Au^III is more strictly square planar, there being only one close contact outside the square planar array (Au^III–F(PF₅⁻), 3.394(2) Å), that contact not in a position that could be considered “axial”.

In the complexes containing an Au(L)₂ core (Fig. 1 and 2), the two bis(NHC) ligands (L¹ or L²) are “inverted” with respect to each other (i.e., for one ligand the methylene/ethylene linker is above the AuC₄ plane while for the other ligand the linker is below the AuC₄ plane), presumably to avoid unfavourable intramolecular interactions between the N–CH₃ groups of each ligand. A similar arrangement is seen in [AuL³]³⁺ and [AuL³Cl₂]⁺ (Fig. 3), but here the N–CH₃ groups are replaced by o-xylylene moieties that complete the (NHC)₄ macrocycle. Not surprisingly, the bite angle (C(11)–Au(1)–C(21)) in the complexes of L² (NHC moieties linked by CH₂CH₂) is larger than for the complexes of L¹ (NHC moieties linked by CH₂).

In each complex, the imidazolyl rings are tilted with respect to the AuC₄ coordination plane. For all complexes of L¹ and L³, in which pairs of imidazolyl units are linked by a methylene bridge, the AuC₄-imidazolyl interplanar angle is about 40°. For complexes of L², in which the imidazolyl units are linked by a longer (ethylene) bridge, the AuC₄-imidazolyl interplanar angle is higher (about 50°). This steeper inclination of the imidazolyl rings to the AuC₄ plane may be the reason why the axial Au–L bonds are longer in [Au(L²)₂Cl₂]Cl·(MeOH) 3·(MeOH) than in the other chloride salts, and absent in [Au(L²)₂](PF₆)₃·(MeCN)₂6·(MeCN)₂—steric hindrance from the endo hydrogens of the ethylene bridge may inhibit approach of axial ligands to the Au centre.

UV-vis electronic absorption spectroscopy and conductance measurements

The UV-vis absorption spectra of all the Au^III–NHC complexes displayed intense absorption in the region λ = 230–290 nm (Fig. 4 and Table 3). We attribute these high energy absorption bands to π–π* intraligand transitions involving the NHC ligands.³⁵ In the spectra of [Au(L¹)₂Cl₂]Cl 2 and [Au(L²)₂Cl₂]Cl 3 (but not for [AuL³Cl₂]Cl 4) a weaker absorbance is observed in the region 300–350 nm. These weaker absorption bands have energies and molar extinction coefficients (ε on the order of 10³–10⁴ M⁻¹ cm⁻¹) similar to those of ligand-to-metal-charge-transfer (LMCT) bands of the [AuCl₄]⁻ ion, which has been studied in detail by Mason et al.⁴¹ Similar absorptions were also displayed by other NHC adducts of Au^III halides.^17,35 Therefore, we tentatively attribute these low energy absorption bands seen for [Au(L¹)₂Cl₂]Cl 2 and [Au(L²)₂Cl₂]Cl 3 to LMCTs from the chlorido ligands to the Lewis acidic Au^III centres. A distinct LMCT band is not seen for [AuL³Cl₂]Cl 4, but may be overlapped with the high energy absorption band below 300 nm. Similar observations were also reported by Huynh et al. for UV-vis studies of [Au^IIICl₂(iPr₂Bim)₂]BF₄ (Bim = 1,3-benzimidazol-2-ylidene).³⁵

Fig. 4 UV-vis absorption spectra of (a) [Au(L¹)₂Cl₂]Cl 2, [Au(L²)₂Cl₂]Cl 3, [AuL³Cl₂]Cl 4 and (b) [Au(L¹)₂](PF₆)₃5, [Au(L²)₂](PF₆)₃6, [AuL³](PF₆)₃7 in MeCN.

Table 3 UV-vis spectroscopic data and conductivity properties of Au^III–NHC complexes

Complex	λ max/nm (ε/M^−1 cm^−1)^a	Λ M ^b (S cm^2 mol^−1)
a Measured in 0.05 mM MeCN solution at 298 K. b Measured in 1 mM DMSO solution at 298 K.
[Au(L^1)2Cl2]Cl 2	302 (3600), 235 (14 100)	41.6
[Au(L^1)2](PF6)35	250 (14 600)	114.6
[Au(L^2)2Cl2]Cl 3	301 (4100), 245 (15 200)	83.5
[Au(L^2)2](PF6)36	241 (15 000)	118.7
[AuL^3Cl2]Cl 4	275 (8400)	41.8
[AuL^3](PF6)37	256 (11 600)	113.4

---

The molar conductance of solutions of the complexes in DMSO (Table 3) was measured to gain insight into possible dissociation of the chlorido ligands from the Au centres. Solutions of each of the salts [Au(L¹)₂](PF₆)₃5, [Au(L²)₂](PF₆)₃6 and [AuL³](PF₆)₃7, showed conductivity in the range 113–119 S cm² mol⁻¹. These values are in the range expected for 1 : 3 electrolytes in DMSO.^42,43 By contrast, [Au(L¹)₂Cl₂]Cl 2 and [AuL³Cl₂]Cl 4 showed a molar conductivity of 44.5 and 41.8 S cm² mol⁻¹ respectively, in the range expected for a 1 : 1 electrolyte, while [Au(L²)₂Cl₂]Cl 3 showed a molar conductivity of 83.5 S cm² mol⁻¹, which falls in the range expected for 1 : 2 electrolytes. These results suggest that for [Au(L¹)₂Cl₂]Cl 2 and [AuL³Cl₂]Cl 4 in DMSO solution, the complex cation is of the form the [Au(NHC)₄Cl₂]⁺ (the same as seen in the solid state), while for [Au(L²)₂Cl₂]Cl 3, the complex cation is of the form [Au(NHC)₄Cl]²⁺ in DMSO solution. Compared to [Au(L¹)₂Cl₂]⁺ and [AuL³Cl₂]⁺, the higher propensity of [Au(L²)₂Cl₂]⁺ to undergo dissociation of Cl⁻ may be consequence of increased steric congestion around the axial coordination sites, as suggested by the steeper inclination of the imidazolyl rings relative to the AuC₄ plane in [Au(L²)₂Cl₂]⁺ (see above).

NMR spectroscopy

The ¹H and ¹³C NMR spectra for Au^III tetracarbene complexes (Fig. 5 and S1–S12†) are broadly consistent with the structures seen in the solid state, but show affects of dissociation of chlorido in some cases.

Fig. 5 ¹H NMR spectra (600 MHz, DMSO-d₆) for: (a) [Au(L¹)₂Cl₂]Cl 2, [Au(L¹)₂](PF₆)₃5; (b) [Au(L²)₂Cl]Cl₂3, [Au(L²)₂](PF₆)₃6; and (c) [AuL³Cl₂]Cl 4, [AuL³](PF₆)₃7.

The ¹H NMR spectrum of [Au(L¹)₂Cl₂]Cl 2 in DMSO-d₆ solution shows a single signal due to the methyl protons (3.69 ppm), two doublet signals in the range 7.5–8.5 ppm, corresponding to the two non-equivalent protons of each imidazolyl ring, and a pair of doublets (AX pattern) centred at 8.09 and 6.79 ppm, corresponding to two non-equivalent protons in the methylene bridge linking the imidazolyl groups in each ligand. The non-equivalence of the methylene protons indicates that the “puckered” conformation adopted by the two L¹ ligands in the complex in the solid state persists in solution and is rigid on the NMR timescale. The ¹H NMR spectrum of [Au(L¹)₂](PF₆)₃5 in DMSO-d₆ solution shows the same number of signals and splitting patterns, but the signals are at different chemical shifts, markedly so for the signals of the methylene protons. This observation is consistent with the change in coordination of Au^III from Au(NHC)₄Cl₂ for [Au(L¹)₂Cl₂]Cl 2 to Au(NHC)₄ for [Au(L¹)₂](PF₆)₃5. We tentatively assign the doublet at 8.09 ppm to the endo protons of the CH₂ groups, and suggest that its markedly downfield chemical shift is a consequence of an intramolecular H⋯Cl hydrogen bonding interaction with a Cl ligand. Interestingly, in the solid state, the Cl⋯H distances for the endo hydrogens are 2.55 and 2.60 Å, which are within the range of ∼2.2 Å–∼3.0 Å seen for C–H⋯Cl hydrogen bonds.⁴⁴

It is interesting to note that the ¹H NMR chemical shift of the signal for the endo CH₂ protons of [Au(L¹)₂Cl₂]Cl 2 in D₂O solutions is close to that seen for [Au(L¹)₂](PF₆)₃5 in DMSO-d₆, and markedly different from that seen for [Au(L¹)₂Cl₂]Cl 2 in DMSO-d₆. This observation suggests in D₂O solution, the cation [Au(L¹)₂Cl₂]⁺ undergoes solvolysis to give [Au(L¹)₂]³⁺. Consistent with this suggestion, we found that when NaCl was dissolved in solutions prepared by dissolving [Au(L¹)₂Cl₂]Cl 2 in D₂O, the signal for the endo CH₂ protons shifted downfield (i.e., towards the chemical shift seen for the endo protons in [Au(L¹)₂Cl₂]⁺), and moved back upfield (i.e., towards the chemical shift seen for the endo protons in [Au(L¹)₂]³⁺) when Cl⁻ was precipitated from the sample as AgCl (Fig. S13†).

As noted above, conductance measurements showed that dissolution of [Au(L²)₂Cl₂]Cl 3 in DMSO results in dissociation of one Cl⁻ ligand from the Au centre and formation of [Au(L²)₂Cl]²⁺ in solution. The ¹H NMR spectra of solutions prepared by dissolving [Au(L²)₂Cl₂]Cl 3 in DMSO-d₆ can thus be considered to be spectra of the cation [Au(L²)₂Cl]²⁺. These spectra again show a single signal (3.35 ppm) due to the methyl protons and two doublet signals (now in the range 7.7–7.9 ppm) corresponding to the two non-equivalent protons of each imidazole ring, with an apparent AA′XX′ pattern of signals centred at 4.91 and 5.46 ppm, corresponding to the protons in each ethylene bridge. These signals and splitting patterns suggest that the “puckered” conformation adopted by the two L² ligands of [Au(L²)₂Cl₂]⁺ in the solid state persists in solution. In the solid state, there are four non-equivalent environments for the hydrogen atoms in each ethylene bridge, so the observation of an AA′XX′ pattern for the ethylene protons suggests that some twisting within each L² ligand causes exo and endo protons on each carbon in an ethylene bridge to be rendered equivalent on the NMR timescale (Fig. 6).

Fig. 6 Proposed twisting of the ethylene bridge to account for symmetry observed for ¹H NMR signals of [Au(L²)₂Cl₂]Cl 3 in DMSO-d₆ solution.

There are small differences between the ¹H NMR signals seen for [Au(L²)₂Cl]²⁺ and [Au(L²)₂]³⁺ (solutions prepared from [Au(L²)₂](PF₆)₃) 6 in terms chemical shift, but the number of signals and splitting patterns are the same in each case. For [Au(L²)₂Cl]²⁺, we note that the existence of an Au(NHC)₄Cl coordination motif (approximately square pyramidal) must render the two L² ligands inequivalent if the ligands are puckered in the same way as seen in the solid state for [Au(L²)₂Cl₂]Cl 3. Since the number of NMR signals is consistent with the L² ligands being equivalent on the NMR timescale, it may be that the two L² ligands are rendered equivalent on the NMR timescale by rapid dissociation and re-association of the chlorido ligand (Fig. 5(b)).

The ¹H NMR spectra for both [AuL³Cl₂]Cl 4 and [AuL³](PF₆)₃7 in DMSO-d₆ solution are analogous to those seen for [Au(L¹)₂Cl₂]Cl 2 and [Au(L¹)₂](PF₆)₃5. Major differences between the two samples in terms of in chemical shifts of the protons of the methylene linkers, and to a lesser extent those of the benzylic protons, are consistent with a change in coordination of Au^III, from Au(NHC)₄Cl₂ in [AuL³Cl₂]Cl 4 to Au(NHC)₄ (or a solvated form) in [AuL³](PF₆)₃7. The number of signals and splitting patterns are consistent with the ligand L³ adopting a “puckered” conformation that does not invert on the NMR timescale.

In the ¹³C NMR spectra of the new Au^III–NHC complexes having an Au(NHC)₄ motif, the signal for the carbene carbon occurs near 146 ppm, which is close to that reported for the related [Au(NHC)₄]⁺ complex 1³² but approximately 35 ppm upfield of that for the carbene carbons in Au^I–NHC complexes.⁸ For comparison, imidazolium-derived Au^III complexes of form [Au(R₂Im)Cl₃] (R₂Im = 1,3-dialkylimidazol-2-ylidene) show a carbene signal in the range 134–141 ppm,^14,45 and [Au(R₂Im)₂Cl₂]⁺ and [Au₂(L¹)₂Cl₄]²⁺ exhibit carbene signals near 150–155 ppm,^46,47 while [Au(R₂Bim)Cl₃] and [Au(R₂Bim)₂Cl₂]⁺ (R₂Bim = 1,3-dialkylbenzimidazol-2-ylidene) exhibit carbene signals³⁵ about 10 ppm downfield of those of their imidazolium-derived counterparts.

H–D exchange reactions

Imidazolium ions are well-known to undergo H/D exchange reactions.^48–50 We have explored H/D exchange reactions for the imidazolium salt L¹·2HCl and the Au^III complex [Au(L¹)₂]³⁺ (formed from [Au(L¹)₂Cl₂]Cl 2, see above) in D₂O solutions (Fig. 7). For L¹·2HCl, the imidazolium H2 protons exchange quickly (<5 min at room temperature) with deuterium from the solvent. This finding is consistent with previous observations that the ¹H NMR signal for the H2 protons of imidazolium salts generally “underintegrates” in spectra recorded in D₂O and methanol-d₄ solutions at room temperature.^51–53 When the sample was heated at 100 °C, the imidazolyl H4/H5 protons were fully exchanged over the course of about 48 h, while the CH₂ protons were about 40% exchanged over the same period, and the CH₃ protons underwent little or no exchange. The analogous exchange reactions were markedly faster for [Au(L¹)₂]³⁺ under similar conditions: the imidazolyl H4/H5 protons (δ 7.91, 7.61 ppm) were fully exchanged within 17 h at 100 °C, as indicated by disappearance of the H4/H5 signals from the ¹H NMR spectrum, and the CH₂ protons were fully exchanged within 48 h at 100 °C (Fig. 7b). Experiments conducted using t-butanol as an internal standard showed negligible H/D exchange of CH₃ protons in [Au(L¹)₂]³⁺ even after heating for 48 h (Fig. S14†). Not surprisingly, the H/D exchange reactions of [Au(L¹)₂]³⁺ were catalysed by base—in the presence of two equivalents of K₂CO₃, the imidazolyl H4/H5 protons (but not the CH₂ protons) underwent partial exchange over 48 h at room temperature (Fig. S15†), and in the presence of two equivalents of NaOH, both the imidazolyl H4/H5 protons and the CH₂ protons were almost completely exchanged overnight at room temperature (Fig. S16†). The high charge on the [Au(L¹)₂]³⁺ cation is likely a factor contributing to the fast H/D exchange seen in D₂O (formation of a deprotonated intermediate [Au(L¹)₂–H]²⁺ would be favoured by the concomitant decrease in concentration of positive charge on the complex).

Fig. 7 ¹H NMR spectra (600.13 MHz, D₂O) for solutions of (a) the imidazolium salt L¹·2HCl and (b) the Au^III complex [Au(L¹)₂]³⁺ (prepared from [Au(L¹)₂Cl₂]Cl) 2, showing disappearance of signals due to H/D exchange reactions. For L¹·2HCl, no signal for the H2 protons was detected even from freshly-prepared solutions.

Stability studies

The new Au^III(NHC)₄ complexes reported in this work were stable in the solid state at room temperature under air for at least 1 year. The D₂O exchange experiment described above suggests surprisingly high stability for the Au^III(NHC)₄ complexes, in that they can survive extended periods in solution at 100 °C, albeit with H/D exchange occurring. To further explore the stability of the new Au^III(NHC)₄ complexes, we have used ¹H NMR spectroscopy to assessed the longevity of various samples in DMSO-d₆ solutions at elevated temperatures. After a DMSO-d₆ solution containing [Au(L¹)₂](PF₆)₃5 was stored at room temperature for 6 months and then heated at 120 °C for 26 h, there was no significant change to the ¹H NMR spectrum (Fig. S17†). The same applied for a DMSO-d₆ solution containing [Au(L³)](PF₆)₃7 that was maintained at 120 °C for 7 days (Fig. S19†). The chlorido complex [Au(L²)₂Cl₂]Cl 3 was somewhat less stable; after a DMSO-d₆ solution containing this compound was heated at 120 °C for 7 days, integration of the imidazolyl H4/H5 region of the ¹H NMR spectrum indicating that ca. 20% of the initial Au^III complex [Au(L²)₂Cl₂]⁺ had decomposed, the major decomposition product being the known⁵⁴ dinuclear Au^I complex [Au₂(L²)₂]²⁺ (Fig. S18†). The slightly lower stability of [Au(L²)₂Cl₂]⁺ may be a consequence of slightly higher strain in the chelate rings (e.g., in the ethylene link joining the NHC units in [Au(L²)₂Cl₂]⁺, N–C–C angles are 113.4(3), 113.8(4), 115.7(4) and 116.4(4)°; in the methylene linker in [Au(L¹)₂Cl₂]⁺, N–C–N angles are 108.5(7) and 109.1(7)°, closer to the ideal tetrahedral angle, 109.5°). For comparison, salts of the dinuclear Au^III complex [Au₂(L¹)₂Cl₄]²⁺ and its bromido counterpart,^8,47 which contain Au^III(NHC)₂X₂ moieties rather than Au^III(NHC)₄ moieties, were far less stable. At room temperature in the solid state, ca. 50% of the Au^III/Au^III complex [Au₂(L¹)₂Br₄]²⁺ had decomposed to the Au^I/Au^III complex [Au₂(L¹)₂Br₂]²⁺ over 6 months (Fig. S20 and S21†), and no Au^III species could be detected after a DMSO-d₆ solution containing [Au₂(L¹)₂Br₄]²⁺ and [Au₂(L¹)₂Br₂]²⁺ was heated at 120 °C for 2 h (Fig. S21†). Similar results were seen for solutions containing [Au₂(L¹)₂Cl₄]²⁺ (Fig. S22†) and [Au₂(L²)₂Br₄]²⁺ (Fig. S23†).

The high stability of Au^III(NHC)₄ complexes compared to their Au^III(NHC)₂(halide)₂ counterparts may reflect the strong σ-donating ability of the NHC ligand, which is better able to stabilise the Au^III centre than the weaker σ-donating halide ligands. The robustness of the metal–NHC bond and steric protection exerted by the chelating NHC ligands must also contribute to the enhanced stability of the Au^III(NHC)₄ complexes compared to their Au^III(NHC)₂X₂ counterparts.

Cyclovoltammetric studies showed that the Au^III complexes did not undergo reversible redox reactions at room temperature in acetonitrile with Bu₄NClO₄ as electrolyte (see ESI†).

Conclusion

Three new Au^III complexes of form [Au(NHC)₄Cl₂]Cl may be synthesized in good yield directly from bis- and tetrakis(imidazolium) salts and KAuCl₄ in the presence of base. These formally 20 electron complexes have tetragonally distorted octahedral coordination of the Au^III centre, the NHC groups making an approximately square planar array and the chlorido ligands occupying more distant axial positions. The Au^III(NHC)₄ motif is remarkably stable, resisting decomposition in DMSO solution at 120 °C or D₂O at 100 °C for extended periods. The axial (chlorido) ligands are labile, and metathesis of the [Au(NHC)₄Cl₂]Cl salts with KPF₆ in methanol/water results in removal of the chlorido ligands and formation of three new, formally 16 electron complexes of form [Au(NHC)₄](PF₆)₃. The [Au(NHC)₄]³⁺ ion is sufficiently stable to undergo base-catalysed H/D exchange reactions at most sites on the NHC ligands in D₂O at 100 °C without evidence of decomposition.

Experimental section

General procedures

Nuclear magnetic resonance spectra were recorded using Bruker ARX400 (400.13 MHz for ¹H), Bruker ARX500 (500.13 MHz for ¹H and 125.77 MHz for ¹³C), or Bruker ARX 600 (600.13 MHz for ¹H, 150.90 MHz for ¹³C) spectrometers at ambient temperature. ¹H and ¹³C NMR chemical shifts were referenced to signals of the solvent (DMSO-d₆: ¹H 2.50 ppm; ¹³C 39.52). When necessary, assignments were made with the aid of ¹H–¹³C HSQC (heteronuclear single quantum coherence) and ¹H–¹³C HMBC (heteronuclear multiple bond correlation) spectra. Conductance measurements were performed by using a TPS Aqua-CP/A conductivity meter. Cyclic voltammetry experiments were recorded using an eDAQ e-corder 401 system in a three-electrode cell with a glassy carbon (1 mm diameter) working electrode, a platinum (1 mm diameter) counter electrode, and a platinum wire reference electrode. Measurements were taken at room temperature (25 °C) in acetonitrile with 0.1 M Bu₄NClO₄ as a supporting electrolyte. Microanalyses were performed by The School of Chemistry & Molecular Bioscience, University of Queensland, Australia, and the Instrument Center of National Chung Hsing University, Taiwan. High resolution mass spectra were measured using Agilent LCMS 6510 Q-TOF and Waters LCT Premier XE spectrometers, using the ESI method, with MeCN : H₂O (9 : 1) as solvent. All organometallic compounds were prepared under a nitrogen atmosphere. The imidazolium salts L¹·2HCl, L²·2HCl,^55,56 and L³·4HCl⁵⁷ were prepared according to literature procedures, and the Au^III complexes [Au₂(L¹)₂X₄]X₂ (X = Cl, Br) and [Au₂(L²)₂Br₄]Br₂ were prepared in the same way as the corresponding hexafluorophosphate salts.^8,47

Synthesis of Au^III–NHC complexes

Complex 2, [Au(L¹)₂Cl₂]Cl. A solution of LiOAc (80 mg, 1.21 mmol) in DMF (5 mL) was added to a solution of the bis(imidazolium) salt L¹·2HCl (99 mg, 0.40 mmol) and KAuCl₄ (75 mg, 0.20 mmol) in DMF (10 mL) at 80 °C. The mixture was then heated to 100 °C and maintained at that temperature for 5 h. During this time a white precipitate formed. The precipitate was collected by filtration and washed successively with DMF, acetone, and Et₂O to give [Au(L¹)₂Cl₂]Cl 2 as a white powder (106 mg, 81%). Found: C, 32.88; H, 3.66; N, 17.10% AuC₁₈H₂₄N₈Cl₃ requires C, 32.97; H, 3.69; N, 17.09%. ¹H NMR (600.13 MHz, DMSO-d₆): δ 8.09 (d, ²J_H,H 15.3 Hz, 2H, CHH), 8.03 (d, ³J_H,H 2.1 Hz, 4H, imidazolyl H4/H5), 7.77 (d, ³J_H,H 2.1 Hz, 4H, imidazolyl H4/H5), 6.79 (d, ²J_H,H 15.3 Hz, 2H, CHH), 3.69 (s, 12H, 4 × CH₃). ¹³C{¹H} NMR (150.9 MHz, DMSO-d₆): δ 146.82 (NCN), 124.96 (imidazolyl C4/C5), 123.63 (imidazolyl C4/C5), 63.15 (CH₂), 37.78 (CH₃). HRMS (ESI⁺): calcd for AuC₁₈H₂₄N₈Cl₂⁺ ([AuL¹Cl₂]⁺), m/z 619.1166. Found, m/z 619.1160.

Complex 5, [Au(L¹)₂](PF₆)₃. A solution of KPF₆ (106 mg, 0.58 mmol) in water (3 mL) was added to a solution of [Au(L¹)₂Cl₂]Cl 2 (106 mg, 0.17 mmol) in MeOH (10 mL). The resulting precipitate was collected by filtration and washed with water and (3 × 3 mL) with methanol to leave [Au(L¹)₂](PF₆)₃5 as a white powder (93 mg, 59%). Found: C, 21.88; H, 2.57; N, 11.43% AuC₁₈H₂₄N₈P₃F₁₈ requires C, 21.96; H, 2.46; N, 11.38%. ¹H NMR (600.13 MHz, DMSO-d₆): δ 8.01 (d, ³J_H,H 1.8 Hz, 4H, imidazolyl H4/H5), 7.77 (d, ³J_H,H 1.8 Hz, 4H, imidazolyl H4/H5), 6.84 (d, ²J_H,H 13.2 Hz, 2H, CHH), 6.75 (d, ²J_H,H 13.2 Hz, 2H, CHH), 3.50 (s, 12H, 4 × CH₃). ¹³C{¹H} NMR (150.9 MHz, DMSO-d₆): δ 146.68 (NCN), 125.16 (imidazolyl C4/C5), 124.28 (imidazolyl C4/C5), 63.09 (CH₂), 37.83 (CH₃). HRMS (ESI⁺): calcd for AuC₁₈H₂₄N₈P₂F₁₂⁺ ([Au(L¹)₂·2PF₆]⁺), m/z 839.1073. Found: m/z 839.1071.

Complex 3, [Au(L²)₂Cl₂]Cl. A solution of LiOAc (61 mg, 0.92 mmol) in DMF (5 mL) was added to a solution of the bis(imidazolium) salt L²·2HCl (105 mg, 0.40 mmol) and KAuCl₄ (75 mg, 0.20 mmol) in DMF (10 mL) at 80 °C. The mixture was then heated to 110 °C and maintained at that temperature overnight. The white precipitate that formed was collected by filtration and washed successively with DMF, acetone, and Et₂O, to leave [Au(L²)₂Cl₂]Cl 3 as a white powder (108 mg, 79%). Found: C, 35.32; H, 4.33; N, 16.05% AuC₂₀H₂₈N₈Cl₃ requires C, 35.13; H, 4.13; N, 16.39%. ¹H NMR (600.13 MHz, DMSO-d₆): δ 7.86 (d, ³J_H,H 2.1 Hz, 4H, imidazolyl H4/H5), 7.73 (d, ³J_H,H 2.1 Hz, 4H, imidazolyl H4/H5), 5.46 (m, 4H, CH₂CH₂), 4.91 (m, 4H, CH₂CH₂), 3.35 (s, 12H, 4 × CH₃). ¹³C{¹H} NMR (150.9 MHz, DMSO-d₆): δ 144.58 (NCN), 126.47 (imidazolyl C4/C5), 125.13 (imidazolyl C4/C5), 47.64 (CH₂CH₂), 38.01 (CH₃). HRMS (ESI⁺): calcd for AuC₂₀H₂₈N₈Cl₂⁺ ([Au(L²)₂Cl₂]⁺) m/z 647.1479. Found, m/z 647.1464.

Complex 6, [Au(L²)₂](PF₆)₃. This compound was prepared in the same way as [Au(L¹)₂](PF₆)₃5, and was obtained in 69% yield. Found: C, 23.13; H, 3.30; N, 10.80% AuC₂₀H₂₈N₈P₃F₁₈·(H₂O) requires C, 23.31; H, 2.93; N, 10.88%. ¹H NMR (600.13 MHz, DMSO-d₆): δ 7.84 (d, ³J_H,H 1.8 Hz, 4H, imidazolyl H4/H5), 7.71 (d, ³J_H,H 1.8 Hz, 4H, imidazolyl H4/H5), 5.30 (m, 4H, CH₂CH₂), 4.93 (m, 4H, CH₂CH₂), 3.30 (s, 12H, 4 × CH₃). ¹³C{¹H} NMR (150.9 MHz, DMSO-d₆): δ 145.14 (NCN), 127.05 (imidazolyl C4/C5), 125.76 (imidazolyl C4/C5), 47.92 (CH₂CH₂), 38.19 (CH₃). HRMS (ESI⁺): calcd for AuC₂₀H₂₈N₈P₂F₁₂⁺ ([Au(L²)₂·2PF₆]⁺) m/z 867.1386. Found, m/z 867.1344.

Complex 4, [AuL³Cl₂]Cl. A solution of LiOAc (20 mg, 0.30 mmol) in DMF (5 mL) added to a solution of the tetra(imidazolium) salt L³·4HCl (54 mg, 0.083 mmol) and KAuCl₄ (32 mg, 0.083 mmol) in DMF (10 mL) at 80 °C and this temperature was maintained for overnight. The white precipitate that formed was collected by filtration and washed successively with DMF, acetone, and Et₂O to leave [AuL³Cl₂]Cl 4 as a white powder (32 mg, 48%). Found: C, 44.65; H, 3.52; N, 13.83% AuC₃₀H₂₈N₈Cl₃ requires C, 44.82; H, 3.51; N, 13.94%. ¹H NMR (600.13 MHz, DMSO-d₆): δ 7.97 (d, ³J_H,H 2.1 Hz, 4H, imidazolyl H4/H5), 7.68, 7.52 (m, 8H, C₆H₄), 7.48 (d, ²J_H,H 12.6 Hz, 2H, CHH), 7.42 (d, ³J_H,H 2.1 Hz, 4H, imidazolyl H4/H5), 6.78 (d, ²J_H,H 12.6 Hz, 2H, CHH), 5.26 (d, ²J_H,H 15.6 Hz, 4H, benzylic CHH), 5.15 (d, ²J_H,H 15.6 Hz, 4H, benzylic, CHH). ¹³C{¹H} NMR (150.9 MHz, DMSO-d₆): δ 145.92 (NCN), 134.96 (C₆H₄), 132.20 (C₆H₄), 129.70 (C₆H₄), 125.11 (imidazolyl C4/C5), 123.73 (imidazolyl C4/C5), 62.20 (CH₂), 52.64 (benzylic CH₂). HRMS (ESI⁺): calcd for AuC₃₀H₂₈N₈Cl₂⁺ ([AuL³Cl₂]⁺) m/z 767.1479. Found, m/z 767.1469.

Complex 7, [AuL³](PF₆)₃. This compound was prepared in the same way as [Au(L¹)₂](PF₆)₃5, and was obtained in 72% yield. Found: C, 31.90; H, 2.39; N, 9.73% AuC₃₀H₂₈N₈P₃F₁₈ requires C, 31.82; H, 2.49; N, 9.89%. ¹H NMR (600.13 MHz, DMSO-d₆): δ 7.98 (d, ³J_H,H 1.8 Hz, 4H, imidazolyl H4/H5), 7.69, 7.56 (m, 8H, C₆H₄), 7.44 (d, ³J_H,H 1.8 Hz, 4H, imidazolyl H4/H5), 6.82 (d, ²J_H,H 13.2 Hz, 2H, CHH), 6.72 (d, ²J_H,H 13.2 Hz, 2H, CHH), 5.34 (d, ²J_H,H 15.3 Hz, 4H, benzylic CHH), 5.00 (d, ²J_H,H 15.3 Hz, 4H, benzylic CHH). ¹³C{¹H} NMR (150.9 MHz, DMSO-d₆): δ 146.21 (NCN), 134.86 (C₆H₄), 132.20 (C₆H₄), 129.86 (C₆H₄), 125.86 (imidazolyl C4/C5), 124.21 (imidazolyl C4/C5), 62.24 (CH₂), 52.32 (benzylic CH₂). HRMS (ESI⁺): calcd for AuC₃₀H₂₈N₈P₂F₁₂⁺ ([AuL³·2PF₆]⁺) m/z 987.1386. Found, m/z 987.1396.

X-Ray crystal structure determinations

Crystals of [Au(L¹)₂](PF₆)₃·(MeCN)₂5·(MeCN)₂, [Au(L²)₂](PF₆)₃·(MeCN)₂6·(MeCN)₂ and [AuL³](PF₆)₃·(MeCN)₃7·(MeCN)₃ were grown by diffusion of vapours between neat diethyl ether and a solution of the complex in acetonitrile, and crystals of [Au(L¹)₂Cl₂]Cl·(MeOH)₂2·(MeOH)₂, [Au(L²)₂Cl₂]Cl·(MeOH) 3·(MeOH) and [AuL³Cl₂]Cl·(MeOH)_3.54·(MeOH)_3.5 were grown by diffusion of vapours between neat diethyl ether and a solution of the complex in methanol. Crystallographic data were collected at 100(2) K on either an Oxford Diffraction Gemini or an Oxford Diffraction Xcalibur diffractometer using Mo Kα or Cu Kα radiation. Following analytical absorption corrections and solution by direct methods, the structures were refined against F² with full-matrix least-squares using the program SHELXL-2014.⁵⁸ Unless stated differently below, all hydrogen atoms were added at calculated positions and refined by use of riding models with isotropic displacement parameters based on those of the parent atoms. Except for those atoms mentioned below, anisotropic displacement parameters were employed throughout for the non-hydrogen atoms. For the structure of [AuL³Cl₂]Cl·(MeOH)_3.54·(MeOH)_3.5, one chlorido ligand was modelled as being disordered with a molecule of methanol with site occupancies constrained to 0.75 and 0.25 from trial refinement and as required for charge balance. Remaining solvent molecules were modelled as three molecules of methanol situated on a crystallographic mirror plane. Geometries and displacement parameters of the solvent were restrained to reasonable values. Solvent hydrogen atoms were not included in the model. For the structure of [Au(L¹)₂](PF₆)₃·(MeCN)₂5·(MeCN)₂, the solvent acetonitrile molecule was found to be disordered over two sites with the methyl atom common to both components. In [Au(L²)₂](PF₆)₃·(MeCN)₂6·(MeCN)₂, the ligand and one hexafluorophosphate anion are both disordered over two sets of sites with occupancies refined to 0.765(4) and its complement after trial refinement showed no significant differences in the refined values. The atoms of the minor component of the cation were refined with isotropic displacement parameters. The remaining hexafluorophosphate anion was modelled as being rotationally disordered with site occupancies were constrained to 0.5 after trial refinement. The solvent was modelled as acetonitrile, disordered over two sets of sites. For the structure of [AuL³](PF₆)₃·(MeCN)₃7·(MeCN)₃, the fluorine atoms of two hexafluorophosphate anions were modelled as being disordered over two sites; hexafluorophosphate anion (1) where site occupancies were refined to 0.75(2) and its complement and hexafluorophosphate anion (3) where fluorine occupancies were constrained to 0.5 after trial refinement.

Acknowledgements

The authors acknowledge the facilities, and the scientific and technical assistance of the Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy, Characterisation & Analysis, The University of Western Australia. Ahmed H. Mageed thanks The Higher Committee for Education Development in Iraq (HCED) for financial support.

References

1. P. J. Barnard, M. V. Baker, S. J. Berners-Price, B. W. Skelton and A. H. White, Dalton Trans., 2004, 1038–1047.
2. M. P. Rigobello, A. Folda, B. Dani, R. Menabò, G. Scutari and A. Bindoli, Eur. J. Pharmacol., 2008, 582, 26–34.
3. W. Liu and R. Gust, Chem. Soc. Rev., 2013, 42, 755–773.
4. D. Pflästerer and A. S. K. Hashmi, Chem. Soc. Rev., 2016, 45, 1331–1367.
5. H. Schmidbaur and A. Schier, in Comprehensive Organometallic Chemistry III, ed. R. H. Crabtree and D. M. P. Mingos, Elsevier, Oxford UK, 2007, vol. 2, pp. 252–308.
6. T. Zou, C. T. Lum, C.-N. Lok, J.-J. Zhang and C.-M. Che, Chem. Soc. Rev., 2015, 44, 8786–8801.
7. I. J. Lin and C. S. Vasam, Can. J. Chem., 2005, 83, 812–825.
8. M. Baron, C. Tubaro, M. Basato, A. Biffis, M. M. Natile and C. Graiff, Organometallics, 2011, 30, 4607–4615.
9. P. de Frémont, R. Singh, E. D. Stevens, J. L. Petersen and S. P. Nolan, Organometallics, 2007, 26, 1376–1385.
10. S. Zhu, R. Liang and H. Jiang, Tetrahedron, 2012, 68, 7949–7955.
11. I. G. Santos, A. Hagenbach and U. Abram, Dalton Trans., 2004, 677–682.
12. S. D. Khanye, N. B. Báthori, G. S. Smith and K. Chibale, Dalton Trans., 2010, 39, 2697–2700.
13. R. Jothibasu, H. V. Huynh and L. L. Koh, J. Organomet. Chem., 2008, 693, 374–380.
14. S. Gaillard, A. M. Z. Slawin, A. T. Bonura, E. D. Stevens and S. P. Nolan, Organometallics, 2009, 29, 394–402.
15. M. C. Jahnke, T. Pape and F. E. Hahn, Z. Anorg. Allg. Chem., 2010, 636, 2309–2314.
16. M. Pažický, A. Loos, M. J. Ferreira, D. Serra, N. Vinokurov, F. Rominger, C. Jäkel, A. S. K. Hashmi and M. Limbach, Organometallics, 2010, 29, 4448–4458.
17. C. Hirtenlehner, C. Krims, J. Hölbling, M. List, M. Zabel, M. Fleck, R. J. Berger, W. Schoefberger and U. Monkowius, Dalton Trans., 2011, 40, 9899–9910.
18. S. Gaillard, X. Bantreil, A. M. Slawin and S. P. Nolan, Dalton Trans., 2009, 6967–6971.
19. M. Kriechbaum, D. Otte, M. List and U. Monkowius, Dalton Trans., 2014, 43, 8781–8791.
20. A. Collado, J. Bohnenberger, M. J. Oliva-Madrid, P. Nun, D. B. Cordes, A. M. Slawin and S. P. Nolan, Eur. J. Inorg. Chem., 2016, 2016, 4111–4122.
21. J. Gil-Rubio, V. Cámara, D. Bautista and J. Vicente, Inorg. Chem., 2013, 52, 4071–4083.
22. W. Liu, K. Bensdorf, M. Proetto, A. Hagenbach, U. Abram and R. Gust, J. Med. Chem., 2012, 55, 3713–3724.
23. H. Sivaram, J. Tan and H. V. Huynh, Organometallics, 2012, 31, 5875–5883.
24. C. Topf, C. Hirtenlehner, M. Zabel, M. List, M. Fleck and U. Monkowius, Organometallics, 2011, 30, 2755–2764.
25. M. Muuronen, J. E. Perea-Buceta, M. Nieger, M. Patzschke and J. Helaja, Organometallics, 2012, 31, 4320–4330.
26. J. P. Reeds, A. C. Whitwood, M. P. Healy and I. J. Fairlamb, Organometallics, 2013, 32, 3108–3120.
27. C. Topf, C. Hirtenlehner, M. Fleck, M. List and U. Monkowius, Z. Anorg. Allg. Chem., 2011, 637, 2129–2134.
28. C. Hemmert, R. Poteau, M. Laurent and H. Gornitzka, J. Organomet. Chem., 2013, 745, 242–250.
29. F. dit Dominique, H. Gornitzka, A. Sournia-Saquet and C. Hemmert, Dalton Trans., 2009, 340–352.
30. M. Baron, C. Tubaro, M. Basato, A. A. Isse, A. Gennaro, L. Cavallo, C. Graiff, A. Dolmella, L. Falivene and L. Caporaso, Chem. – Eur. J., 2016, 22, 10211–10224.
31. C. Tubaro, M. Baron, M. Costante, M. Basato, A. Biffis, A. Gennaro, A. A. Isse, C. Graiff and G. Accorsi, Dalton Trans., 2013, 42, 10952–10963.
32. Z. Lu, S. A. Cramer and D. M. Jenkins, Chem. Sci., 2012, 3, 3081–3087.
33. G. Marangoni, B. Pitteri, V. Bertolasi, G. Gilli and V. Ferretti, J. Chem. Soc., Dalton Trans., 1986, 1941–1944.
34. V. Duckworth and N. Stephenson, Inorg. Chem., 1969, 8, 1661–1664.
35. H. V. Huynh, S. Guo and W. Wu, Organometallics, 2013, 32, 4591–4600.
36. G. Nardin, L. Randaccio, G. Annibale, G. Natile and B. Pitteri, J. Chem. Soc., Dalton Trans., 1980, 220–223.
37. L. S. Hollis and S. J. Lippard, J. Am. Chem. Soc., 1983, 105, 4293–4299.
38. M. P. Suh, I. S. Kim, B. Y. Shim, D. Hong and T.-S. Yoon, Inorg. Chem., 1996, 35, 3595–3598.
39. S. Alvarez, Dalton Trans., 2013, 42, 8617–8636.
40. R. Elder and J. Watkins, Inorg. Chem., 1986, 25, 223–226.
41. H. Isci and W. R. Mason, Inorg. Chem., 1983, 22, 2266–2272.
42. S. Ramalingam and S. Soundararajan, J. Inorg. Nucl. Chem., 1967, 29, 1763–1768.
43. W. J. Geary, Coord. Chem. Rev., 1971, 7, 81–122.
44. C. B. Aakeröy, T. A. Evans, K. R. Seddon and I. Pálinkó, New J. Chem., 1999, 23, 145–152.
45. B. Jacques, D. Hueber, S. Hameury, P. Braunstein, P. Pale, A. l. Blanc and P. de Frémont, Organometallics, 2014, 33, 2326–2335.
46. J. L. Hickey, PhD thesis, The University of Western Australia, 2008.
47. M. Baron, C. Tubaro, M. Basato, A. Biffis and C. Graiff, J. Organomet. Chem., 2012, 714, 41–46.
48. J. L. Wong and J. H. Keck Jr., J. Org. Chem., 1974, 39, 2398–2403.
49. Y. Takeuchi, H. J. Yeh, K. L. Kirk and L. A. Cohen, J. Org. Chem., 1978, 43, 3565–3570.
50. C. Hardacre, J. D. Holbrey and S. J. McMath, Chem. Commun., 2001, 367–368.
51. M. V. Baker, B. W. Skelton, A. H. White and C. C. Williams, Organometallics, 2002, 21, 2674–2678.
52. M. V. Baker, M. J. Bosnich, D. H. Brown, L. T. Byrne, V. J. Hesler, B. W. Skelton, A. H. White and C. C. Williams, J. Org. Chem., 2004, 69, 7640–7652.
53. M. V. Baker, D. H. Brown, C. H. Heath, B. W. Skelton, A. H. White and C. C. Williams, J. Org. Chem., 2008, 73, 9340–9352.
54. M. Baron, C. Tubaro, A. Biffis, M. Basato, C. Graiff, A. Poater, L. Cavallo, N. Armaroli and G. Accorsi, Inorg. Chem., 2012, 51, 1778–1784.
55. Y. Unger, A. Zeller, M. A. Taige and T. Strassner, Dalton Trans., 2009, 4786–4794.
56. O. Sanchez, S. González, M. Fernández, A. R. Higuera-Padilla, Y. Leon, D. Coll, A. Vidal, P. Taylor, I. Urdanibia and M. C. Goite, Inorg. Chim. Acta, 2015, 437, 143–151.
57. T. Pape and F. E. Hahn, Dalton Trans., 2013, 42, 7330–7337.
58. G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct. Chem., 2015, 71, 3–8.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 1528857–1528862. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c7dt01272a

---