This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.
Article pubs.acs.org/IC
Heterobimetallic N‑Heterocyclic Carbene Complexes: A Synthetic, Spectroscopic, and Theoretical Study Thomas P. Pell,† David J. D. Wilson,† Brian W. Skelton,‡ Jason L. Dutton,† and Peter J. Barnard*,† †
Department of Chemistry and Physics, La Trobe Institute of Molecular Science, La Trobe University, Melbourne, Victoria 3086, Australia ‡ Centre for Microscopy, Characterization and Analysis, The University of Western Australia, 35 Stirling Highway, Crawley, Western Australia 6009, Australia S Supporting Information *
ABSTRACT: A new synthetic methodology has been developed for the preparation of heterobimetallic group 11 and group 12 complexes of a symmetrical bis-NHC “pincer” ligand. The synthetic route involved the initial preparation of a mononuclear [Au(NHC)2]+ complex with pendent imidazole moieties on the NHC ligands. Subsequent alkylation of the imidazole groups with Et3OBF4 and metalation with a second metal ion (Ag(I) or Hg(II)) provided two heterobimetallic complexes. Four homobimetallic (Cu(I)2, Ag(I)2, Au(I)2, and Hg(II)2) complexes of the same bis-NHC “pincer” ligand were also prepared. The homobimetallic Cu(I)2, Au(I)2, and Hg(II)2 complexes and heterobimetallic Au(I)−Ag(I) and Au(I)−Hg(II) complexes and the synthetic intermediates for the heterobimetallic complexes were characterized by X-ray crystallography. These X-ray structures show that the bimetallic complexes adopt “twisted” conformations in the solid state, supporting short M···M interactions. Crystalline samples of the homobimetallic Ag(I)2 and Au(I)2 and heterobimetallic Au(I)−Ag(I) and Au(I)−Hg(II) complexes were emissive at room temperature and at 77 K. The geometries of the synthesized complexes were optimized at the M06-L/def2-SVP level of theory, and the electronic nature of the M···M interactions for all synthesized complexes was investigated using natural bond orbital (NBO) calculations.
■
INTRODUCTION The synthesis of heterobimetallic coordination complexes incorporating N-heterocyclic carbene (NHC) ligands is of great interest for applications including the study of M1···M2 bonding interactions1,2 and the development of novel catalysts.3 Heterobimetallic complexes offer the potential for cooperative catalysis, where the combination of two different metal ions into a single molecule produces synergistic catalytic activity.3−12 Recently, examples of heterobimetallic complexes of NHCs have been described that display a range of interesting properties. For example, a mixed Ru(III)−Pd(II) complex (1) exhibited synergistic hydrodefluorination catalytic behavior under mild reaction conditions.3 Previously, heterobimetallic complexes of NHC ligands have been prepared using a stepwise approach, where the chosen metals are bound sequentially to the bidentate proligand. The strategy generally relies on the capacity to selectively deprotonate the precursor diazolium salt (e.g., 1, Figure 1)3,16,17 or by utilizing a metal fragment that has a specific ligand-site binding preference (e.g., 2, Figure 1).12,18,19 Unfortunately, these methods often produce a mixtures of isomers, which limit reaction yields and require chromatographic purification to separate the homo- and heterobimetallic products. Other alternative methods require the use of different azolium groups (e.g., imidazolium and triazolium) incorporated into the proligand, providing different basicities and con© XXXX American Chemical Society
sequently the capacity for selective deprotonation (e.g., 3, Figure 1).13 The use of bidentate ligands with mixed donors, for example the combination of NHC and pyridine groups, facilitates the selective introduction of the metals based on the chosen reaction sequence. This concept has been applied by Catalano and co-workers in the preparation of a series of luminescent heterobimetallic coinage metal complexes (e.g., 4 and 5, Figure 1).14,15,20−23 The remarkable heterotrimetallic Au(I)−Cu(I) complex (5) exhibits a solvochromic shift when exposed to volatile organic compound (VOC) vapors.15 There have been numerous reports of luminescent dinuclear coinage metal complexes of bidentate NHC ligands that display short M···M (metallophilic) interactions.24−27 Previously the luminescent properties of dinuclear Au(I) complexes of “pincer” type NHC ligands have been investigated.28,29 X-ray crystallographic studies show that these cations may adopt either a “twisted” molecular conformation, which supports a short (aurophilic) Au···Au interaction (∼3.3 Å), or an “open” conformation, with a longer Au···Au separation (∼6.5 Å).27,30 As such, “pincer” NHC ligands represent a useful system for probing the influence of metallophilic interactions on luminescence properties of coinage metal complexes.27 We are interested in the synthesis of luminescent complexes of Received: February 2, 2016
A
DOI: 10.1021/acs.inorgchem.6b00222 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Figure 1. Structures of selected heterobimetallic complexes of N-heterocyclic carbene ligands.3,12−15
Scheme 1. Synthesis of Homobimetallic Cationic Complexes 7−10a
(a) 7, Cu2O (0.55 equiv), NH3(aq) (10 equiv), EtOH, 40 °C, 3 h. (b) 8, Ag2O (1.2 equiv), CH3CN, rt, 24 h. (c) 9, (THT)AuCl (1 equiv), NaOAc (4 equiv), DMF, 110 °C, 2 h. (d) 10, Hg(OAc)2, CH3CN, 82 °C, 24 h. a
Figure 2. ORTEP46 representation of the cations for (a) 7, (b) 9, and (c) 10; displacement ellipsoids at 50% probability; hydrogen atoms, anions, and solvent of crystallization have been omitted for clarity.
NHC ligands,31−33 and recently we became interested in the possibility of using “pincer” type NHC ligands for a comparative study of homo- and heterobimetallic complexes. Herein we report, for the first time, a postsynthetic modification and metalation approach for the synthesis of heterobimetallic coinage metal complexes of a symmetric bidentate NHC “pincer” ligand, with homobimetallic complexes of the same ligand prepared in order to conduct a comparative study of the luminescent properties. A complementary theoretical study, including molecular orbital (MO) and natural bond orbital (NBO) calculations, has also been carried out.
imidazolium salt proligand 6 using the synthetic procedures outlined in Scheme 1. Complex 7 was prepared from 6 with the addition of 0.55 equiv of Cu2O and excess ammonia (10 equiv), and after workup the dinuclear complex was isolated as a bright yellow crystalline solid in an 82% yield. The Ag(I) complex 8 was prepared conventionally by the reaction of 6 with Ag2O in acetonitrile, however this compound was highly susceptible to photolytic and thermal decomposition and as a result the reaction was conducted at room temperature (rt) in the absence of light. The homodinuclear Au(I) and Hg(II) complexes, 9 and 10 respectively, were readily prepared utilizing previously described synthetic procedures for dinuclear NHC complexes of these metals. Complex 10 was recrystallized by the diffusion of vapors between a solution of the complex in acetonitrile and diethyl ether, and this resulted in a low yield (27%) of the pure compound.24,27,34−40 The formation of complexes 7−10 was confirmed by NMR spectroscopy, mass spectrometry, and in the case of 7, 9, and
■
RESULTS AND DISCUSSION Synthesis and Characterization. The homobimetallic complexes 7−10 (metals Cu(I), Ag(I), Au(I), and Hg(II) respectively) were prepared from the ethylated pincer B
DOI: 10.1021/acs.inorgchem.6b00222 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry Scheme 2. Synthesis of Heterobimetallic Precursors 11−13, and heterobimetallic complexes 14−16a
a
(a) 14, [Cu(CH3CN)4]BF4 (1 equiv), K2CO3 (2.5 equiv), CH3CN, rt, 16 h. (b) 15, Ag2O (2 equiv), CH3CN, rt, 72 h. (c) 16, Hg(OAc)2 (1.4 equiv), CH3CN, 80 °C, 24 h.
complex 9, unreacted proligand, and other unidentified products. To selectively form a mono-Au(I) complex precursor in good yield, a postmetalation ligand synthesis strategy was developed, where the second imidazolium moiety was generated after the initial metalation step (Scheme 2). To our knowledge a synthetic strategy of this type has not been previously used for the preparation of heterobimetallic NHC complexes. Due to the insolubility of the required monoalkylated imidazolium salt (11, Scheme 2) in THF, 2,6-bis(imidazole-1yl)pyridine could be readily monoethylated with bromoethane in this solvent with high yield and purity (Scheme 2). Subsequent treatment of 11 with (THT)AuCl and NaOAc in DMF at 110 °C produced the cationic Au(I) complex, for which the anion was converted to BF4− via metathesis using KBF4 in water. The subsequent step was initially quite challenging and involved ethylation of the pendent imidazole group of the Au(I) complex (12, Scheme 2). Early attempts to introduce the second ethyl group using an alkyl halide (e.g., bromoethane or iodomethane) were unsuccessful, with 1H NMR analysis showing decomposition of the Au(I) complex, possibly as a result of oxidative addition of the alkyl halide to the metal center. To overcome the problem, the nonoxidizing alkylating agent triethyloxonium tetrafluoroborate (Et3OBF4) was investigated. This reagent proved suitable, and the desired trication (13, Scheme 2) was obtained in excellent yield. The 1H and 13C NMR data for all of the compounds 11−13 are in agreement with their formulation. Upon monoethylation of 2,6-bisimidazolepyridine the 1H NMR spectrum of 11 displayed nine separate aromatic signals consistent with the unsymmetrical compound, with one downfield resonance (10.38 ppm) corresponding to the imidazolium (procarbenic) proton. The 13C NMR spectrum for the Au(I) complex 12 showed a characteristic downfield chemical shift for the carbenic carbon (180.65 ppm) indicating formation of the coordinated NHC unit. The alkylation of 12, yielding 13, was confirmed by the shift in the NCHN proton resonance of the
10 X-ray crystallography. In all cases metalation of the proligand 6 was confirmed by the loss of the procarbenic proton signal and a downfield shift for the carbenic carbon (178.07, 210.23, 180.89, and 170.43 ppm for 7, 8, 9, and 10 respectively). The number of signals observed in the 1H and 13 C NMR spectra is consistent with the ligands being magnetically equivalent. X-ray crystal structures were obtained for the homodinuclear compounds for 7, 9, and 10 (Figure 2). The helical structure of these compounds brings the metal centers into close proximity (Cu(1)···Cu(1*) 2.9552(13) Å, Au(1)···Au(2) 3.5412(1) Å, and Hg(1)···Hg(2) 3.6941(1) Å for 7, 9, and 10, respectively). In the case of 7 there is a short Py−N(5)−Cu(1) interaction of 2.666(5) Å. The Au···Au distance in 9 is shorter than the sum of the van der Waals radii for Au (3.80 Å) and can therefore be classified as an aurophilic interaction.41 In contrast, for 7 and 10 the Cu···Cu and Hg···Hg distances are greater than the sum of the van der Waals radii for each of these metals, respectively.42,43 Complex 7 crystallizes in a Sohncke space group (trigonal, R32), consistent with the complex being chiral in its twisted form due to axial chirality or helicity. In the crystal structure a short interaction of 2.666(5) Å is measured between the pyridine nitrogen atom and the Cu(I) centers. This is greater than the Cu−NPy distances determined when pyridine is the main donor for Cu(I) (e.g., 1.937(4) Å),44 but similar to distances reported for NHC−Cu(I) complexes with ancillary pyridine donors (e.g., 2.50 Å).45 An interesting feature of complex 7 is the inequivalent N-ethyl CH2 protons in the 1H NMR spectrum (resonating at 3.60 and 3.69 ppm), possibly suggesting that the chiral helical structure (as seen in the solid state) may be retained in solution. To synthesize heterobimetallic analogues of complexes 7− 10, initial attempts were made to monometalate the proligand 6 with 0.5 equiv of the Au(I) source (THT)AuCl in the presence of a weak base, either NaOAc or Cs2CO3. These experiments were not successful, instead producing the homobimetallic C
DOI: 10.1021/acs.inorgchem.6b00222 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Figure 3. ORTEP46 representation of the cations for (a) 11, (b) 12, and (c) 13; displacement ellipsoids at 50% probability; hydrogen atoms and anions have been omitted.
inequivalent aromatic protons and two sets of signals for the ethyl groups. A comparison of the 1H NMR spectra for the homobimetallic Au2 complex 9 and the heterobimetallic Ag(I)−Au(I) complex 15 is shown in Figure 4. The 13C NMR spectrum for 15 showed the characteristic downfield shifts for the Au−CNHC and the Ag−CNHC signals at 180.72 and 181.34 ppm, respectively. In addition, the 13C−107Ag and 13 C−109Ag couplings (185.15 and 213.91 Hz respectively) were resolved. These coupling constants are consistent with those
pendant imidazolium moiety downfield from 9.17 (imidazoleNCHN proton) for 12 to 9.40 ppm and the appearance of a second set of ethyl group signals. The solid-state structures of the heterobimetallic complex precursors 11−13 were determined by X-ray crystallography (Figure 3). The proligand 11 has the expected monoethylated structure with one bromide counterion. Complex 12 crystallized in a helical configuration, with an apparent π−π stacking interaction between the two pendant pyridylimidazole groups of the [Au(NHC)2]+ complex, with the shortest distance between the π stacked ring systems being 3.461 Å (Py-C(22)··· Im-C(4)). The helical configuration adopted by 12 results in axial chirality, with both the right- and left-handed helical configurations present in the centrosymmetric crystal structure. The crystal structure of the tricationic complex 13 (formed by ethylation of the pendant imidazole groups of 12) showed a pseudoplanar configuration, with the pendant pyridyl− imidazolium groups oriented on opposite sides relative to the linear geometry about the Au(I) center. It is likely that the π−π stacking interaction seen for 12 is absent in the case of 13 due to electrostatic repulsion associated with the positive charged imidazolium units. The heterobimetallic complexes 14 (Cu(I)−Au(I)), 15 (Ag(I)−Au(I)), and 16 (Hg(II)−Au(I)) were prepared using similar synthetic procedures, by treating the mono-Au(I) precursor compound 13 with either [Cu(CH3CN)4]BF4 and either K2CO3, Ag2O, or Hg(OAc)2 respectively. The Cu(I)− Au(I) complex 14 was obtained as a crystalline yellow solid, however NMR analysis (Figures S5, S6) of this material showed it to be a mixture of the desired heterobimetallic Cu(I)−Au(I) complex 14 (∼70%) and the homobimetallic Au2 complex 9 (∼30%). Despite numerous recrystallization attempts, these compounds could not be separated. In contrast, complexes 15 and 16 were obtained as pure materials from the respective reactions (without further purification) as confirmed by NMR spectroscopy, mass spectrometry, elemental analysis, and X-ray crystallography. In each case, the 1H NMR spectra for 14−16 were consistent with the unsymmetrical nature of these molecules with seven
Figure 4. NMR spectra of (a) homobimetallic Au(I) complex 9 and (b) heterobimetallic Au(I)−Ag(I) complex 15. Regions of the spectrum between 1.5−3.5 ppm and 4.5−7.5 ppm have been omitted for clarity (see Figures S3 and S7 for full spectra). D
DOI: 10.1021/acs.inorgchem.6b00222 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry reported previously for a mononuclear Ag(I) complex of a macrocyclic NHC ligand.47 Similarly, the 13C NMR spectrum for 16 shows two downfield signals corresponding to the Au− CNHC and Hg−CNHC resonances at 180.60 and 169.99 ppm, respectively. X-ray crystal structures were obtained for complexes 15 and 16 (Figure 5), and in both cases the structures confirmed the
Figure 6. (a) Room temperature and (b) 77 K solid state emission spectra for crystalline samples of 6·Br2 (λex = 260 nm), 8·[BF4]2 (λex = 300 nm), 9·[BF4]2 (λex = 366 nm), 15·[BF4]2 (λex = 353 nm), and 16· [BF4]2 (λex = 350 nm).
Theoretical Studies. From a preliminary study of homobimetallic Au(I) NHC complexes, it is evident that M06-L/def2-SVP is an appropriate level of theory for geometry optimizations of such systems. For computational efficiency, methyl R-groups were used in place of ethyl R-groups as a model for DFT and ab initio calculations. The optimized geometries of each of the complexes were found to be consistent with those determined crystallographically for the homobimetallc Au(I) complex (9Me) and the heterobimetallic Ag(I)−Au(I) (15Me) and Au(I)−Hg(II) (16Me) complexes. However, the M···M distances (which are considered an important criterion when assessing the quality of the geometry optimization calculations) were found to deviate from experimental values for the homobimetallic Cu(I) (7Me) and Hg(II) (10Me) complexes by 0.342 Å (shorter) and 0.125 Å (longer), respectively. The deviation may be related to the weaker metallophilic interactions observed for these metals.43,48 To study the electronic nature of the M···M interactions in the bimetallic complexes, natural bond orbital (NBO) calculations were performed at the M06-L/def2-TZVP level of theory (at the M06-L/def2-SVP geometries), with results presented in Table 1. There is only a slight positive charge on all of the group 11 metal centers (NPA average of 0.26−0.40, Table 1), which results in a breaking of the formal 5d10 6s0 electronic structure of the metal centers. Previously it has been shown that metalophilic bonding is produced by hybridization
Figure 5. ORTEP46 representation of the cations for (a) 15 and (b) 16; displacement ellipsoids at 50% probability; hydrogen atoms and anions have been omitted and ethyl group carbon atoms shown as spheres for clarity.
heterobimetallic nature of these compounds. The complexes crystallized in the same helical or twisted conformation as that described previously for the homobimetallic complexes (vide supra), giving rise to short Au···Ag and Au···Hg interactions of 3.2897(4) Å and 3.2135(4) Å for complexes 15 and 16, respectively. Photophysical Properties. Complexes displaying aurophilic interactions (and to a lesser extent cuprophilic, argentophilic, and mercurophilic interactions) are well-known for their intriguing luminescent properties. As the homo- and heterobimetallic complexes prepared here are formed from the same ligand system and adopt matching helical structures in the solid state, they represent an interesting series for probing the influence of the short M···M interactions on emissive properties. Similar UV−visible spectra were obtained for the proligand 6 and the bimetallic complexes 7−10, 15, and 16 (Figure S1). The proligand and the metal complexes are nonemissive in solution, however as crystalline solids, complexes 8, 9, 15, and 16 (Ag(I)−Ag(I), Au(I)−Au(I), Au(I)−Ag(I), and Au(I)− Hg(II), respectively) are photoluminescent at both rt and 77 K (Figure 6, panels a and b respectively). At rt complexes 8, 9, 15, and 16 give low-energy (LE) emission with maxima at 453 nm (λex = 300 nm), 478 nm (λex = 366 nm), 461 nm (λex = 353 nm), and 447 nm (λex = 350 nm) respectively, and these emission bands are red-shifted by ∼2 nm at 77 K. In contrast, a crystalline sample of the proligand 6 (Figure 6 a) shows highenergy emission at 325 nm at rt. The crystal structures of 9, 15, and 16 show that these complexes exhibit short M···M interactions in the solid state, and the solid state emission spectra show a trend where the lowest energy emission is associated with the Au(I)···Au(I) interaction, with blue-shifted emission maxima for Au(I)···Ag(I), Au(I)···Hg(I), and Ag(I)··· Ag(I) respectively. These complexes are not luminescent in solution, and this is likely to be due to the fluxional behavior in fluid medium leading to loss of the metallophilic interaction and efficient nonradiative deactivation.
Table 1. M06-L/def2TZVP Calculated Wiberg Bond Indices, NBO Charges, and Electronic Configurations of Homobimetallic and Heterobimetallic Group 11 and 12 Complexes occupation
E
complex
WBI (M···M)
metal
charge
s
p
d
7Me 8Me 9Me 10Me 15Me
0.0860 0.1367 0.1186 0.0457 0.1480
16Me
0.1524
Cu Ag Au Hg Ag Au Au Hg
0.40 0.38 0.26 0.99 0.38 0.26 0.26 0.94
0.56 0.62 0.94 1.01 0.62 0.94 0.96 0.93
0.31 0.22 0.20 0.18 0.23 0.21 0.17 0.30
9.73 9.78 9.60 9.82 9.78 9.60 9.60 9.83
DOI: 10.1021/acs.inorgchem.6b00222 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry among the filled 5d shell and the empty 6s and 6p orbitals at the metal center.49 That is, the greater the population of the 6p orbital, the greater the degree of hybridization and the stronger the metalophilic interaction. The greatest degree of hybridization is observed for complex 7Me, for which the 6p orbital occupation is calculated to be 0.31 electrons. The experimental X-ray crystal structure of complex 7 does not exhibit a Cu···Cu distance shorter than the sum of the van der Waals radii of copper (2.9551(1) Å), whereas the optimized geometry does. This means that the high degree of hybridization at the Cu(I) center (5d9.73 6s0.56 6p0.31) calculated for complex 7Me may not be entirely representative of the actual molecule. Of the complexes containing group 11 metals, 9Me has the greatest amount of electron density within the 6s orbital, which is most likely a result of the high electronegativity of Au, allowing greater acceptance of electrons from the NHC donor groups. The hybridization at the metal centers of the heterobimetallic complexes is very similar to that for the same metal centers in the respective homobimetallic complexes. The Wiberg bond indices (WBI) for the M···M interaction for all complexes are also presented in Table 1. The bond orders for the complexes 7Me and 10Me are smaller than those observed for the other homobimetallic complexes, but are still significant enough to be described as weak interactions.49 The bond orders observed for the heterobimetallic complexes are slightly larger than those observed for the homobimetallics, which is what might be expected for stronger dipolar interactions between different metals. Molecular orbital (MO) contour plots for all complexes (with methyl R-groups in place of ethyl R-groups to increase computational efficiency) were calculated at the CAM-B3LYP/ def2-TZVP level of theory (M06-L/def2-SVP geometry), and plots containing the HOMO−1 to the LUMO+1 for all complexes can be found in Tables S3 and S4. The HOMOs for all bimetallic complexes are shown in Figure 7. Generally, the
For the complexes that exhibit luminescent properties in the solid state (at rt or 77 K), the HOMO and LUMO energies and HOMO−LUMO gaps have been calculated using CAMB3LYP/def2-TZVP (Table 2). The HOMO−LUMO gap Table 2. CAM-B3LYP/def2-TZVP Calculated HOMO and LUMO Energies and HOMO−LUMO (H−L) Gapsa
a
complex
HOMO
LUMO
H−L gap
Ag2 Au2 Au(I)−Ag(I) Au(I)−Hg(II)
−7.60 −7.81 −7.65 −8.32
−0.56 −0.59 −0.62 −0.87
7.04 7.22 7.03 7.46
Units of eV.
increases in the order Au(I)−Ag(I) < Ag(I)2 < Au(I)2 < Au(I)−Hg(II) for these complexes. This calculated trend is similar to that seen for the experimental emission energies, which increase in the order Au(I)−Au(I) < Au(I)−Ag(I) < Ag(I)−Ag(I) < Au(I)−Hg(II). Only the Au(I) complex breaks the trend, with a calculated HOMO−LUMO gap that is blueshifted compared to the experimentally observed value.
■
CONCLUSION A postsynthetic metalation methodology has been developed for the synthesis of heterobimetallic transition metal complexes of a “pincer” bis-NHC ligand. This synthetic approach is versatile, is expected to be broadly applicable for the synthesis of heterobimetallic NHC complexes, and avoids timeconsuming purification steps that are often required in conventional approaches for the synthesis of complexes of this type. Recently a related study reported the alkylation of metalated imidazolylidene/imidazole ligands, yielding Rh(III) bis-NHC chelate complexes.50 Crystallographic studies show that in all cases the prepared bimetallic complexes support short homo- and heterometallophilic interactions and as such these molecules provide a useful series for investigating the influence of the M···M interaction on the luminescent properties associated of these compounds. The bimetallic complexes are nonluminescent in solution, and this is probably due to these compounds being conformationally fluxional in fluid medium leading to loss of the metallophilic interactions and efficient nonradiative deactivation of the excited state. This is consistent with previous X-ray crystallographic studies of binuclear Au(I)2 cations analogous to those prepared here, where two distinct conformations (with markedly different Au···Au distances) are observed in the solid state.27,30 As crystalline solids the Cu(I) (7) and Hg(II) (10) homobimetallic complexes were nonemissive, however the other homo- and heterobimetallic complexes were emissive. Calculation of the HOMO and LUMO energies and HOMO− LUMO gaps predicts the experimentally observed trend in the emission wavelengths with the exception of the Au(I) complex, where a significantly larger HOMO−LUMO gap is predicted. Wiberg bond indices were undertaken to investigate the magnitude of the metallophilic bond order for the synthesized complexes, and it is evident that the heteronuclear metallophilic interactions are stronger than homonuclear interactions.
Figure 7. CAM-B3LYP/def2-TZVP highest occupied molecular orbital (HOMO) plots for homobimetallic and heterobimetallic group 11 and 12 complexes.
group 11 homobimetallic complexes have very similar HOMOs, which are primarily metal-centered (Figure 7). Complex 10Me is different, with a HOMO that is mostly ligand centered (>99% ligand from Mulliken population analysis) with little contribution from the metal centers. This trend is also observed for the heterobimetallic complexes, where the HOMO for complex 15Me is centered on both the metals (45% Ag and 27% Au), whereas the HOMO for complex 16Me is primarily based on the Au(I) center (68%), with little contribution from Hg(II) (
