Role of Dispersion in Metallophilic HgÂ·Â·Â·M Interactions (M = Cu, Ag, Au

Article pubs.acs.org/IC

Role of Dispersion in Metallophilic Hg···M Interactions (M = Cu, Ag, Au) within Coinage Metal Complexes of Bis(6diphenylphosphinoacenaphth-5-yl)mercury Emanuel Hupf,† Ralf Kather,† Matthias Vogt,† Enno Lork,† Stefan Mebs,*,‡ and Jens Beckmann*,† †

Institut für Anorganische Chemie und Kristallographie, Universität Bremen, Leobener Strasse, 28359 Bremen, Germany Institut für Experimentalphysik, Freie Universität Berlin, Arnimallee 14, 14195 Berlin, Germany

‡

S Supporting Information *

ABSTRACT: The previously reported bis(6-diphenylphosphinoacenaphth-5-yl)mercury (1) was used as ligand for the preparation of the copper(I) complexes, 1·CuCl and [1·Cu(NCMe)]BF4, which were characterized by multinuclear NMR spectroscopy and X-ray crystallography. DFT calculations employing topological analysis of the electron and electron pair densities within the AIM and ELI-D space-partitioning schemes revealed significant metallophilic Hg···Cu interactions. Evaluation of noncovalent bonding aspects according to the noncovalent interaction (NCI) index was applied not only for the Cu complexes 1·CuCl and [1·Cu(NCMe)]BF4 but also for the previously reported Ag and Au complexes, namely, [1·MCl] (M = Ag, Au) and [1·M(NCMe)n]+ (M = Ag, n = 2; M = Au, n = 0), and facilitated the assignment of attractive dispersive Hg···M interactions with the Hg···Cu contacts being comparable to the Hg···Ag but weaker than the Hg···Au interactions. The localization of the attractive noncovalent bonding regions increases in the order Cu < Ag < Au.

■

INTRODUCTION Metallophilic interactions comprise structurally directing forces between low-coordinated closed-shell metal ions, which are often associated with interesting physical properties such as photoluminescence.1 From a theoretical point of view, metallophilic interactions are dispersive forces, which are amplified by relativistic effects.2 Since the maximum of relativistic effects is reached for the post-transition elements of the sixth period, 5d10-configurated gold(I) and mercury(II) complexes are the most frequently observed closed-shell metal species entailing metallophilic interactions. Apart from pure aurophilic3 and mercurophilic4 interactions, an increasing number of heteronuclear contacts between closed-shell metal species has been reported;5 among those are three examples with prominent Hg(II)···Cu(I) interactions (Chart 1). The cationic complex [Hg{Fe[Si(OMe) 3 ](CO) 3 (μdppm)}2Cu] (I) reported by Braunstein et al. comprises the shortest observed Hg···Cu contact of 2.689(2) Å.6 The cationic metallamacrocycle [{Hg(C6H4−CHN−CH2−CH2−N CH−C6H4)}2Cu] (II) described by Singh et al. shows longer Hg···Cu distances of 2.919(7) and 2.921(7) Å, respectively.7 Very recently, López-de-Luzuriaga et al. prepared a series of coinage metal complexes with a bis(diphenyl© XXXX American Chemical Society

phosphinobenzene)mercurial ligand. In the complex [(oC6H4PPh2)2Hg·CuBr] (III), the Hg···Cu interatomic distances of 2.8656(5) and 2.8619(5) Å are in the midrange.8 Concomitantly, we published a related bis(6-diphenylphosphinoacenaphth-5-yl)mercury ligand (1), which undergoes complex formation with the closed-shell metal ions M = Hg(II), Ag(I), and Au(I).9 In this work we extend and complete the series of the coinage metals by two Cu(I) complexes (Scheme 1). As in our previous study9 we employ a set of real-space bonding indicators (RSBIs), which are derived from the electron and electron pair densities of the C−H distancecorrected experimental molecular structures. The RSBI analysis comprises topological dissection of the electron density (ED) into atomic basins based on the atoms-in-molecules (AIM)10 space-partitioning scheme and dissection of the electron pair density according to the electron localization indicator (ELID)11 space-partitioning scheme. In this work we also evaluate noncovalent bonding aspects according to the recently introduced noncovalent interactions (NCI) index 12 by determination of the reduced density gradient s(r) = [1/ Received: August 26, 2016

A

DOI: 10.1021/acs.inorgchem.6b02056 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Chart 1. Complexes Containing Metallophilic Hg(II)···Cu(I) Contacts

Scheme 1. Reaction of 1 with Coinage Metal Salts

■

2(3π2)1/3]|∇ρ|/ρ4/3. Analysis of the NCI proved to be a helpful tool in the characterization of metallophilic interactions13 and was therefore applied not only to the Cu complexes 1·CuCl and [1·Cu(NCMe)]+ but also to the heavier analogues containing Hg···Ag (1·AgCl and [1·Ag(NCMe)2]+) and Hg··· Au (1·AuCl and [1·Au]+) contacts.9 Besides atomic basins, AIM provides a bond paths motif, which resembles the molecular structure and uncovers strong to very weak interatomic contacts, e.g., between adjacent atoms in the crystal lattice. ELI-D provides core, bonding, and lone-pair basins and thus complements the atomic picture extracted from AIM. However, as ELI-D is best suited to qualify and quantify covalent (including dative) atom−atom interactions, the interpretation of ionic or metallophilic interactions is not straightforward because only small bonding basins are obtained for the ionic case and no bonding basins are formed for metallophilic interactions. Accordingly, the latter two contact types can better be analyzed by NCI. Notably, ELI-D isosurfaces (localization domain representations of the basins) and NCI isosurfaces show a complementary spatial distribution,14 suggesting spatial separation of covalent and noncovalent bonding aspects, which is a matter of recent debate.15 The assignment of different contact types including steric/ repulsive (λ2 > 0), van-der-Waals-like (λ2 ≈ 0), and attractive (λ2 < 0) interactions is facilitated by mapping the ED times the sign of the second eigenvalue of the Hessian (sign(λ2)ρ) on isosurfaces of s(r), making NCI a valuable RSBI. Notably, the NCI, like AIM but unlike ELI-D, can be extracted from experimentally obtained EDs.

RESULTS AND DISCUSSION Synthetic Aspects and Characterization. The reaction of bis(6-diphenylphosphinoacenaphth-5-yl)mercury (1) with equimolar amounts of CuCl and [Cu(NCMe)4]BF4 in dichloromethane provided the 1:1 complexes 1·CuCl and [1· Cu(NCMe)]BF4 as colorless crystals in 62% and 84% yield, respectively (Scheme 1). The 31P{1H} NMR chemical shifts of the neutral complex 1·CuCl (δ = −3.9 ppm) and the cationic complex [1·Cu(NCMe)]+ (δ = −2.1 ppm) are very similar and only slightly low-field shifted in comparison to the parent compound 1 (δ = −7.0 ppm).9 In addition, these chemical shifts are comparable to other triarylphosphine-stabilized Cu complexes, such as (Ph 3 P) 2 CuCl (δ = −3.6 ppm), 16 [(Ph3P)3Cu(NCMe)]BF4 (δ = 1.4 ppm),17 or the related Hg···Cu complex (o-C6H4PPh2)2Hg·CuBr (III, δ = −2.6 ppm).8 The 31P{1H} NMR chemical shifts of the Cu complexes 1·CuCl and [1·Cu(NCMe)]+ are high-field shifted with respect to the heavier coin metal complexes 1·AgCl (δ = 2.1 ppm), [1· Ag(NCMe)2]+ (δ = 4.3 ppm), 1·AuCl (δ = 34.5 ppm), and [1· Au]+ (δ = 42.3 ppm) (Table 1). The 199Hg{1H} NMR chemical shifts consist of triplets centered at δ = −502.4 (1·CuCl) and −494.2 ppm ([1·Cu(NCMe)]+) with J(199Hg−31P) = 133 and 103 Hz, respectively. The chemical shifts and corresponding J(199Hg−31P) coupling constants are in the same range as those of the heavier coin metal complexes (Table 1). The neutral species generally exhibit larger J(199Hg−31P) coupling constants than their cationic analogs. Overall, the J(199Hg−31P) coupling constants decrease with an increasing atomic number of the coinage metal centers (Table 1). Besides the previously known triclinic modification of 1, which was crystallized from CH2Cl2/ hexane,9 a new monoclinic polymorph of 1 was obtained serendipitously during the attempt to crystallize [1·CuB


Article

Inorganic Chemistry

those in the cationic species [1·Cu(NCMe)]+ (3.1530(4)/ 3.1864(4) Å). The peri distances of both Cu species are shorter than the peri distances of the heavier analogs of Ag (1·AgCl and [1·Ag(NCMe)2]+) and Au (1·AuCl and [1·Au]+; Table 2). All complexes in the series show significant Hg···M metallophilic interactions. The Hg···Cu distance of the neutral complex [1· CuCl] (2.9625(4) Å) is larger than that of the cationic complex [1·Cu(NCMe)] + (2.8099(4) Å), and both values are comparable to those of the neutral complex [(oC6H4PPh2)2Hg·CuBr] (III, 2.8656(5)/2.8619(5) Å)8 and the cationic metallamacrocycle [{Hg(C 6 H 4 −CHN−CH 2 − CH2−NCH−C6H4)}2Cu] (II, 2.919(7)/2.921(7) Å).7 To the best of our knowledge, the only shorter Hg···Cu distance (2.689(2) Å) was observed in the cationic complex [Hg{Fe[Si(OMe)3](CO)3(μ-dppm)}2Cu] (I).6 Most significantly, the Hg···Cu distance in [1·Cu(NCMe)]+ is also the shortest Hg··· M contact observed in the investigated series of coinage metal mercurials [1·MCl] and [1·M(NCMe)n]+ (M = Cu, Ag, Au) (Table 2). Bond Topological and Surface Analysis. The experimental C−H-corrected molecular structures of 1·CuCl and [1· Cu(NCMe)]+ were used as starting geometries for DFT computations and subsequent topological analyses according to the AIM and ELI-D space-partitioning schemes to obtain topological and integrated bond properties as well as atomic charges. In addition, noncovalent bonding aspects were visualized by means of NCI surfaces. In order to obtain a comprehensive picture of these noncovalent bonding aspects for all Hg···Cu/Ag/Au cases, we extended the NCI calculations to all compounds of the coinage metal series, including [1· AgCl], [1·AuCl], [1·Ag(NCMe)2]+, and [1·Au]+.9 Common bond topological properties for the Hg···Cu/Ag/Au bond critical points (bcps) are listed in Table 3. All metallophilic interactions show the typical features of weak Hg···M contacts, including low ED values at the bcp, a Laplacian being positive but close to zero, a considerable kinetic density over ED ratio (G/ρ(r)bcp) and a total energy density over ED ratio (H/ ρ(r)bcp) being close to zero. As already discussed9 the Hg···Ag distances are considerably longer than the Hg···Au distances, resulting in lower values of the ED at the Hg···Ag bcps. Seemingly, the Hg···Ag contacts are weaker than the Hg···Au contacts. Remarkably, the Hg···Cu contacts contradict this trend. Although the Hg···Cu distances are comparable to the Hg···Au distances, the ED on the Hg···Cu bcps is much lower than expected, being closer to the Hg···Ag values. This is accompanied by considerably larger bond ellipticities, which are as large as 0.54−0.59, compared to 0.05−0.28 for the four Hg··· Ag/Au contacts. The short Hg···Cu bond distances can be explained by the smaller size of the Cu atom (1.13 Å) compared to Ag (1.33 Å) and Au (1.25 Å).17 The Hg···Cu contacts thus have to be considered to be as weak as the Hg··· Ag contacts. Atomic and fragmental charges for 1·MCl (M = Cu, Ag,9 Au9), [1·Cu(NCMe)]+, [1·Ag(NCMe)2]+,9 and [1· Au]+ 9 are listed in Table 4. The total integration error for the whole molecules never exceeded 0.08 e (Σ), confirming the quality of the integrations. The atomic charges of the metals Hg and M follow the trend Hg (0.50−0.53 e) > Cu (0.37−0.42 e) > Ag (0.28−0.39 e) > Au (from −0.02 to −0.07 e). The Hg atomic charges vary in the extremely narrow range of 0.50−0.53 e. Evidently, the Hg atoms are little involved in electronic rearrangements via metal or ligand exchange. The loss of one charge (from −0.67 to −0.71 e of which were located at the Cl atoms) by going from 1·MCl to [1·M(NCMe)n]+ (n = 0−2) is

Table 1. 31P{1H} and 199Hg{1H} NMR Data of 1, 1·MCl (M = Cu, Ag, Au), [1·Cu(NCMe)]+, [1·Ag(NCMe)2]+, and [1· Au]+ a compound

δ(31P) [ppm]

δ(199Hg) [ppm]

J(199Hg-31P) [Hz]b

1 1·CuCl [1·Cu(NCMe)]+ 1·AgCl9 [1·Ag(NCMe)2]+ 9 1·AuCl9 [1·Au]+ 9

−7.0 −3.9 −2.1 2.1 4.3 34.5 42.3

−452.7 −502.4 −494.2 −553.9 −565.1 −499.8 −470.0

631 133 103 116 96 102 47c

a

All NMR spectra were recorded at ambient temperature. The solvent was either CDCl3, CD2Cl2, CD3CN, or THF-d8 (for details see Experimental Section and ref 9). bJ(199Hg−31P) coupling constants are averaged as both values (199Hg{1H} and 31P{1H} NMR) agree within 20 Hz. cThe observed broad 199Hg{1H} NMR resonance prevented a precise assignment of the J(199Hg−31P) coupling constant.

(NCMe)]BF4 from THF. The molecular structures of monoclinic 1 and the Cu complexes 1·CuCl and [1· Cu(NCMe)]BF4 are displayed in Figure 1, and selected interatomic distances and angles are summarized in Table 2. Comprehensive geometrical parameters are listed in the Supporting Information (Table S2). The spatial arrangement of the Hg atom in the monoclinic modification 1 is linear with respect to the organic substituents and shows a slightly smaller Hg···P peri distance of 2.9160(5) Å with respect to the previously reported triclinic modification of 1 (3.001(2) Å). The most dominant difference between the two polymorphs is the conformation of the two acenaphthyl (Ace) groups. The P−Hg−P angle in the triclinic modification of 1 is linear and gives rise to a long P···P separation of 6.003(3) Å. In the monoclinic modification of 1 the P−Hg−P angle is nearly rectangular (92.53(3)°), leading to a substantially shorter P···P separation of 4.214(1) Å, which is much closer to the arrangement found in the coinage metal complexes (see below). The first coordination sphere of the Cu atoms in the neutral complex 1·CuCl and the cationic complex [1·Cu(NCMe)]+ is distorted trigonal planar defined by PPCl and PPN donor sets, and the second coordination sphere is distorted tetrahedral with additional coordination to the Hg atoms. The P−Cu bond lengths in 1·CuCl (2.2401(8) and 2.2523(8) Å) and [1· Cu(NCMe)]+ (2.2487(5) and 2.2625(5) Å) compare well with those of other triarylphosphine-stabilized copper halides, such as (Ph 3 P) 2 CuCl (2.2564(8)/2.2676(9) Å) 16 or (oC6H4PPh2)2Hg·CuBr (III, 2.246(2)−2.2682(9) Å).8 The P− M bond lengths in the metal chloride complexes 1·MCl as well as the cationic metal complexes [1·M(NCMe)n]+ decrease in the order Ag > Au > Cu (Table 2) and can be related to the covalent radii of two-coordinated M(I) compounds of Ag+ (1.33 Å), Au+ (1.25 Å), and Cu+ (1.13 Å).18 Theoretical calculations show that the smaller size of the Au atomic radius in comparison to Ag is due to relativistic effects.19 The P−Cu− P angle of 1·CuCl (136.04(3)°) gives rise to a P···P separation of 4.1659(9) Å, which is smaller than the respective values for [1·Cu(NCMe)]+ (144.94(2)°/4.3018(5) Å) but larger than in the unconstrained complex (Ph 2 P) 2 CuCl (125.55(4)°/ 4.0228(9) Å) 1 6 and in the copper bromide (oC 6 H 4 PPh 2 ) 2 Hg·CuBr (III, 132.15(4)°/4.130(2) Å and 133.95(4)°/4.146(2) Å).8 The P···Hg peri distances in 1· CuCl (3.2078(9) and 3.2162(8) Å) are slightly longer than C


Article

Inorganic Chemistry

Figure 1. Molecular structures of monoclinic 1 and the copper complexes 1·CuCl and [1·Cu(NCMe)]+ showing 50% probability ellipsoids as well as the crystallographic numbering schemes.

Table 2. Selected Interatomic Distances [Angstroms] and Angles [degrees] of 1, 1·MCl (M = Cu, Ag, Au), [1·Cu(NCMe)]+, [1· Ag(NCMe)2]+, and [1·Au]+ compound 1 (triclinic)9 1 (monoclinic) 1·CuCl 1·AgCl9 1·AuCl9 [1·Cu(NCMe)]+ [1·Ag(NCMe)2]+ 9 [1·Au]+ 9

Hg···M

peri-Hg···P

2.963(1) 3.011(2) 2.965(1) 2.810(1) 3.051(2) 2.866(2)

2 × 3.001(2) 2 × 2.916(1) 3.208(1), 3.216(1) 3.232(3), 3.329(2) 3.307(4), 3.339(4) 3.153(1), 3.186(1) 3.243(4), 3.265(5) 3.413(3), 3.379(3)

D

P−M

P−M−P

P···P

2.240(1), 2.252(1) 2 × 2.446(2) 2.327(3), 2.313(3) 2.249(1), 2.263(1) 2.489(4), 2.503(3) 2.319(2), 2.322(3)

180.0(1) 92.5(1) 136.0(1) 137.1(1) 145.3(2) 144.9(1) 141.7(2) 160.5(1)

6.003(3) 4.214(1) 4.166(1) 4.553(3) 4.429(4) 4.302(1) 4.715(6) 4.575(4)


Article

Inorganic Chemistry

Table 3. Bond Topological Properties of the Hg···M Contacts (M = Cu, Ag, Au) in 1·MCl (M = Cu, Ag,9 Au9), [1·Cu(NCMe)]+, [1·Ag(NCMe)2]+,9 and [1·Au]+ a,9 1·CuCl 1·AgCl 1·AuCl [1·Cu(NCMe)]+ [1·Ag(NCMe)2]+ [1·Au]+

bond

d [Å]

ρ(r)bcp [e Å−3]

▽2ρ(r)bcp [e Å−5]

ε

G/ρ(r)bcp [he−1]

H/ρ(r)bcp [he−1]

Hg···Cu Hg···Ag Hg···Au Hg···Cu Hg···Ag Hg···Au

2.965 3.012 2.966 2.811 3.051 2.865

0.16 0.18 0.23 0.21 0.17 0.27

1.3 1.6 2.0 1.7 1.5 2.5

0.59 0.19 0.08 0.54 0.28 0.05

0.63 0.69 0.72 0.71 0.67 0.79

−0.09 −0.07 −0.11 −0.14 −0.06 −0.14

For all bonds, ρ(r)bcp is the electron density at the bond critical point, ▽2ρ(r)bcp is the corresponding Laplacian, ε is the bond ellipticity, and G/ ρ(r)bcp and H/ρ(r)bcp are the kinetic and total energy density over ρ(r)bcp ratios. a

Table 4. AIM Atomic and Fragmental Charges of 1·MCl (M = Cu, Ag,9 Au9), [1·Cu(NCMe)]+, [1·Ag(NCMe)2]+,9 and [1·Au]+ 9 1·CuCl 1·AgCl 1·AuCl [1·Cu(NCMe)]+ [1·Ag(NCMe)2]+ [1·Au]+

Hg

M

Cl or NCMe

PPh2

ace

PPh2

ace

Σ

0.52 0.52 0.52 0.50 0.50 0.53

0.37 0.28 −0.02 0.42 0.39 −0.07

−0.67 −0.70 −0.71 0.05 0.03

0.65 0.68 0.75 0.68 0.71 0.93

−0.72 −0.72 −0.74 −0.70 −0.68 −0.65

0.68 0.71 0.80 0.70 0.67 0.89

−0.78 −0.75 −0.68 −0.66 −0.67 −0.62

0.05 0.02 −0.08 0.99 0.95 1.01

Figure 2. (a) AIM topology of 1·CuCl and (b and c) ELI-D localization domain representations (b = total, c = zoom; Y = 1.3) of 1·CuCl, NCI surfaces (s = 0.5) of (d) 1·CuCl, (e) 1·AgCl,9 and (f) 1·AuCl.9 Attractive noncovalent bonding aspects are given in blue and steric repulsion is given in red.

mainly compensated by the coinage metals and the PPh2 and acenaphthyl (Ace) fragments. The coinage metals lose 0.05 e

(Cu) or 0.11 e (Ag) or gain 0.05 e (Au). The PPh2 and Ace fragments lose 0.02−0.12 e (Cu), 0.04−0.08 e (Ag), and 0.06− E


Article

Inorganic Chemistry

Figure 3. NCI surfaces of [1·CuCl] at different isovalues: (a and d) s = 0.2, (b and e) s = 0.4, (c and f) s = 0.6. Attractive noncovalent bonding aspects are given in blue, and steric repulsion is given in red.

Figure 4. NCI surfaces of 1·AgCl (top) and 1·AuCl (bottom) looking at the Hg side applying different isovalues: (a and d) s = 0.2, (b and e) s = 0.4, (c and f) s = 0.6. Attractive noncovalent bonding aspects are given in blue and steric repulsion is given in red.

given in the Supporting Information (Figure S4). The NCI isosurfaces for 1·CuCl, 1·AgCl, and 1·AuCl are displaced in Figure 2d, 2e, and 2f, respectively. In the overall picture NCI surfaces complement ELI-D surfaces. The NCI plots show that the attractive noncovalent bonding aspects (given in blue) are also strong for the M−Cl contacts, becoming stronger in the order Cu < Ag < Au. Considering the M−P bonds, Ag−P shows the strongest noncovalent contributions. NCI also unravels numbers of weak dispersive H···H and H···Cπ contacts (given in green) and steric repulsion within the aromatic ring systems (given in red). In all three 1·MCl compounds (M = Cu, Ag, Au) an attractive noncovalent bonding region connecting the Hg atom with the respective coinage metal is found. Accordingly, these interactions are best investigated by different NCI isosurface values, which are

0.18 e (Au). For relaxed gas-phase molecular geometries, the differences between the two PPh2 or Ace moieties, which are already quite small, are expected to be even smaller. Overall, all fragments are involved in the electronic rearrangements, resulting in minimal electronic changes for each individual fragment. The AIM bond topology of 1·CuCl (excluding ring and cage critical points) is displayed in Figure 2a. The Cl atom is involved in two C−H···Cl hydrogen bonds, which stabilizes its location within the structure. An isosurface (localization domain representation) of the ELI-D as well as a magnification of the Hg···Cu region are given in Figure 2b and 2c. Although no Hg···Cu bonding basin is formed, the ELI-D gives a picture of the d-electron arrangement of the metal atoms, which is found to be square planar for Cu (dx2−y2 dominated) and linear for Hg (dz2 dominated). The results for [1·Cu(NCMe)]+ are F


Article

Inorganic Chemistry

from a SPS800 MBraun solvent system. Bis(6-diphenylphosphinoacenaphth-5-yl)mercury (1)9 and [Cu(NCMe)4]BF420 were prepared according to literature procedures. 1H, 11B{1H}, 13C{1H}, 19F, 31P{1H}, and 199Hg{1H} NMR spectra were recorded at room temperature using a Bruker Avance-360 spectrometer and are referenced to tetramethylsilane (1H, 13C{1H}), boron trifluoride diethyl etherate (11B{1H}), trichlorofluoromethane (19F), phosphoric acid (85% in water) (31P{1H}), and mercury dichloride in DMSO-d6 (199Hg{1H}, δ = −1501.6 ppm). Chemical shifts are reported in parts per million (ppm), and coupling constants (J) are given in Hertz (Hz). Electron impact mass spectroscopy (EIMS) was carried out using a Finnigan MAT 95. The ESI MS spectra were obtained with a Bruker Esquire-LC MS. Dichloromethane/acetonitrile solutions (or otherwise stated, c = 1 × 10−6 mol L−1) were injected directly into the spectrometer at a flow rate of 3 μL min−1. Nitrogen was used both as a drying gas and for nebulization with flow rates of approximately 5 L min−1 and a pressure of 5 psi. Pressure in the mass analyzer region was usually about 1 × 10−5 mbar. Spectra were collected for 1 min and averaged. The nozzle-skimmer voltage was adjusted individually for each measurement. Synthesis of (6-Ph2P-Ace-5-)2Hg·CuCl (1·CuCl). A solution of copper(I) chloride (13 mg, 0.13 mmol) and acetonitrile (50 mL) was added to a solution of 1 (115 mg, 0.13 mmol) and acetonitrile/THF (40 mL/10 mL). The reaction mixture was stirred for 16 h at room temperature. The solvent was removed under reduced pressure. Subsequently, the residue was dissolved in dichloromethane, filtrated, and carefully layered with n-hexane. After 1 week, colorless crystals of 1·CuCl were filtered off the mother liquor and dried in air, giving 1· CuCl in 62% yield (79 mg, 0.08 mmol, mp 156−158 °C). 1 H NMR (360.3 MHz, CDCl3): δ = 7.72−7.32 (m, 15H), 7.25 (d, 3 1 J( H−1H) = 6.2 Hz, 4H), 7.19 (d, 3J(1H−1H) = 6.7 Hz, 4H), 6.98− 6.93 (m, 3H), 6.71 (d, 3J(1H−1H) = 5.7 Hz, 2H, H-4), 3.50−3.42 ppm (m, 8H, H-1,2). 13C{1H} NMR (90.6 MHz, CDCl3): δ = 159.3 (m), 151.1 (s), 146.3 (s), 140.0 (m), 138.2 (s), 136.0 (m), 134.9 (m, Co), 133.9 (d, J(31P−13C) = 17.4 Hz), 130.2 (s, Cp), 129.5 (s), 128.9 (d, J(31P−13C) = 7.8 Hz), 128.7 (m, Cm), 119.9 (s), 118.8 (m), 30.2 (s, C1 or C2), 29.9 ppm (s, C2 or C1). 31P{1H} NMR (145.9 MHz, CDCl3): δ = −3.9 ppm (s, J(199Hg−31P) = 136.3 Hz). 199Hg{1H} NMR (64.5 MHz, CDCl3): δ = −502.4 (t, J(31P−199Hg) = 130.5 Hz). ESI MS (CH2Cl2/MeCN 1:10, positive mode): m/z = 939.3 (C48H36P2HgCu) for [1·CuCl − Cl−]+. Synthesis of [(6-Ph2P-Ace-5-)2Hg·Cu(NCMe)]BF4 ([1·Cu(NCMe)]BF4). Compound 1 (250 mg, 0.29 mmol) was suspended in dichloromethane (8 mL), and [Cu(NCMe)4]BF4 (90 mg, 0.29 mmol) was added to the vigorously stirred suspension. Stirring the mixture for 48 h at room temperature gave a homogeneous pale yellow solution. Subsequently, the solution was filtered through a PTFE syringe filter, and the obtained clear filtrate was layered with n-hexane and kept for 12 h at −30 °C. The mixture was allowed to crystallize at room temperature for 1 week. The formed large colorless prisms were decanted from the mother liquor and washed with n-hexane. The crystals were ground to a colorless powder, which was subsequently dried in a stream of argon, giving [1·Cu(NCMe)]BF4 in 84% yield (256 mg, 0.24 mmol, mp 198 °C (dec.)). 1 H NMR (360.3 MHz, CDCl3): δ = 7.70−7.37 (m, 10H), 7.34 (d, 3 1 J( H−1H) = 6.8 Hz, 4H), 7.27 (d, 3J(1H−1H) = 7.0 Hz, 4H), 7.23− 7.04 (m, 8H), 6.75 (d, 3J(1H−1H) = 6.8 Hz, 2H, H-4), 3.55−3.48 (m, 8H, H-1,2), 1.95 ppm (s, 3H, NC−CH3). 11B{1H} NMR (115.6 MHz, CDCl3): δ = −0.1 ppm (s). 13C{1H} NMR (90.6 MHz, CDCl3): δ = 157.7 (m), 152.3 (s), 147.3 (s), 140.0 (m), 139.8 (m), 138.8 (s), 136.2 (m), 134.2 (m, Co), 131.3 (s, Cp), 129.4 (m, Cm), 122.9 (m), 122.3 (s), 120.4 (s), 119.2 (m), 30.1 (s, C1 or C2), 29.9 (s, C2 or C1), 2.0 ppm (s, MeCN). 19F NMR (339.0 MHz, CDCl3): δ = −154.8 ppm (s). 31P{1H} NMR (145.9 MHz, CDCl3): δ = −2.1 ppm (s, J(199Hg−31P) = 99.2 Hz). 199Hg{1H} NMR (64.5 MHz, CDCl3): δ = −494.2 (t, J(31P−199Hg) = 105.9 Hz). ESI MS (CH2Cl2/MeCN 1:100, positive mode): m/z = 939.4 (C48H36P2HgCu) for [[1· Cu(NCMe)]BF4 − NCMe − (BF4)−]+. Crystallography. Intensity data of 1, 1·CuCl·CH2Cl2, and [1· Cu(NCMe)]BF4·2CH2Cl2 were collected on a Bruker Venture D8

presented in Figure 3 for 1·CuCl and in Figures S5 (1·AgCl) and S6 (1·AuCl) of the Supporting Information. The Hg···Cu contact is somewhat asymmetric in that it shows a more localized NCI distribution for the coinage metal but a very diffuse behavior for the Hg atom. At an NCI isovalue of 0.2, localized Hg···Cu and Hg···P contacts are visible (Figure 3a and 3d). However, increasing the isovalue to 0.4 already fuses the Hg···Cu and Hg···P contacts on the Hg side of the NCI basin in a way that a fuzzy boundary surface is formed (Figure 3b). This aspect can clearly be related to the nature of the Hg atom as the Hg···Cu contact is still localized at the Cu side of the NCI basin (Figure 3e). Further increasing the isovalue to 0.6 enhances the differences as all bonds to the Cu atom are still localized, whereas a fully delocalized NCI surface covers large parts of the Hg atom (Figure 3c and 3f). Comparing the rather diffuse boundary surface of the Hg side in 1·CuCl to the Hg side in 1·AgCl and 1·AuCl shows enhanced localization especially in the latter complex (Figure 4). Consequently, the dispersive Hg···M interactions are stronger than the Coulomb repulsion between the positively charged metals Hg (0.52 e) and Cu (0.37 e) in 1·CuCl as well as Hg (0.52 e) and Ag (0.28 e) in 1·AgCl. In 1·AuCl the Au atom is slightly negatively charged (−0.02 e), leading to Coulomb attraction between Hg (0.52 e) and Au, which strengthens the Hg···Au metallophilic interaction. Localized Hg···Hg and Hg···Au contacts are observed for the reference compounds,9 such as dinaphtho[1,8-bc:1′,8′-fg][1,5]dimercurocin, 1,8-bis(chloromercurio)naphthalene, 1·HgCl2, and [1·Au]+ (Figures S7 and S8). Conclusions. The newly synthesized copper complexes 1· CuCl and [1·Cu(NCMe)]BF4 complete the series of neutral and cationic coinage metal complexes 1·MCl (M = Cu, Ag, Au) and [1·M(NCMe)n]+ (M = Cu, n = 1; M = Ag, n = 2; M = Au, n = 0).9 All of these complexes show prominent metallophilic interactions with Hg···M bond lengths ranging between 2.810(1) and 3.051(2) Å. Apparently, the size differences between Cu and Hg hampers the formation of localized noncovalent Hg···Cu bonding regions, which supports the results of the AIM bond topology (low ED values, large ellipticities). The combination of the AIM bond topology, ELID localization domains, and NCI region analysis facilitates the understanding of similarities and differences in the Hg···M (M = Cu, Ag, Au) contacts in real space, which cannot be provided by sole inspection of the molecular geometries. The observed metallophilic Hg···M interactions have been identified as mainly attractive dispersive forces, with the degree of localization of the noncovalent bonding regions increasing in the order Cu < Ag < Au. Computational analysis suggests the Hg···Au interactions to be much stronger than the Hg···Cu and Hg···Ag interactions, which might be attributed to the pronounced relativistic effects prevailing in the gold species. The complementarity of ELI-D and NCI in detecting covalent and noncovalent bonding regions in space allows monitoring minute electronic rearrangements due to structural changes or chemical reactions.

■

EXPERIMENTAL SECTION

General. CAUTION: Chemicals containing mercury are potentially highly toxic. Handle with care in a well-ventilated f ume hood using personal protective equipment including appropriate gloves. If not indicated otherwise, all manipulations were performed under inert conditions in a MBraun Labmaster glovebox or using standard Schlenk techniques. Reagents were obtained commercially (SigmaAldrich, Germany) and used as received. Dry solvents were collected G


Article

Inorganic Chemistry

briefing on aurophilicity. Chem. Soc. Rev. 2008, 37, 1931−1951. (c) Schmidbaur, H.; Schier, A. Aurophilic interactions as a subject of current research: an up-date. Chem. Soc. Rev. 2012, 41, 370−412. (4) Schmidbaur, H.; Schier, A. Mercurophilic Interactions. Organometallics 2015, 34, 2048−2066. (5) (a) Bardajia, M.; Laguna, A. Heteronuclear Metal-Metal Contacts between Gold(I) and Group-11, −12, and −13 Centers. Eur. J. Inorg. Chem. 2003, 2003, 3069−3079. (b) Laguna, A.; Silvestru, C. GoldHeterometal Interactions and Bonds In Modern Supramolecular Gold Chemistry; Wiley-VCH: Weinheim, Germany, 2008. (c) Sculfort, S.; Braunstein, P. Intramolecular d10-d10 interactions in heterometallic clusters of the transition metals. Chem. Soc. Rev. 2011, 40, 2741−2760. (6) Bodensieck, U.; Braunstein, P.; Knorr, M.; Strampfer, M.; Bénard, M.; Strohmann, C. Conformation Control in Polymetallic Mesocycles by Metal-Metal Bonding: The First Example of an Hg-Cu Interaction. Angew. Chem., Int. Ed. Engl. 1997, 36, 2758−2761. (7) Patel, U.; Singh, H. B.; Wolmershäuser, G. Synthesis of a Metallophilic Metallamacrocycle: A HgII···CuI···HgII···HgII···CuI···HgII Interaction. Angew. Chem., Int. Ed. 2005, 44, 1715−1717. (8) López-de-Luzuriaga, J. M.; Monge, M.; Olmos, M. E.; Pascual, D. Study of the Nature of Closed-Shell HgII···MI (M = Cu, Ag, Au) Interactions. Organometallics 2015, 34, 3029−3038. (9) Hupf, E.; Lork, E.; Mebs, S.; Beckmann, J. 6-Diphenylphosphinoacenaphth-5-yl-mercurials as Ligands for d10 Metals. Observation of Closed-Shell Interactions of the Type Hg(II)···M; M = Hg(II), Ag(I), Au(I). Inorg. Chem. 2015, 54, 1847−1859. (10) Bader, R. W. F. Atoms in Molecules. A Quantum Theory; Cambridge University Press: Oxford, U.K., 1991. (11) (a) Kohout, M. A measure of electron localizability. Int. J. Quantum Chem. 2004, 97, 651−658. (b) Kohout, M.; Wagner, F. R.; Grin, Y. Electron localizability indicator for correlated wavefunctions. III: singlet and triplet pairs. Theor. Chem. Acc. 2008, 119, 413−420. (12) Johnson, E. R.; Keinan, S.; Mori-Sanchez, P.; Contreras-García, J.; Cohen, A. J.; Yang, W. Revealing noncovalent interactions. J. Am. Chem. Soc. 2010, 132, 6498−6506. (13) (a) Wächtler, E.; Privér, S. H.; Wagler, J.; Heine, T.; Zhechkov, L.; Bennett, M. A.; Bhargava, S. K. Metallophilic Contacts in 2C6F4PPh2 Bridged Heterobinuclear Complexes: A Crystallographic and Computational Study. Inorg. Chem. 2015, 54, 6947−6957. (b) Ai, P.; Mauro, M.; Gourlaouen, C.; Carrara, S.; De Cola, L.; Tobon, Y.; Giovanella, U.; Botta, C.; Danopoulos, A. A.; Braunstein, P. Bonding, Luminescence, Metallophilicity in Linear Au3 and Au2Ag Chains Stabilized by Rigid Diphosphanyl NHC Ligands. Inorg. Chem. 2016, 55, 8527−8542. (14) Mebs, S. Complex modes of bonding: NCI/ELI-D vs. DORI surface analyses of hapticities and hydrogen-hydrogen contacts in zincocene related compounds. Chem. Phys. Lett. 2016, 651, 172−177. (15) (a) de Silva, P.; Corminboeuf, C. Simultaneous Visualization of Covalent and Noncovalent Interactions Using Regions of Density Overlap. J. Chem. Theory Comput. 2014, 10, 3745−3756. (b) Gillet, N.; Chaudret, R.; Contreras-García, I.; Yang, W.; Silvi, B.; Piquemal, J.-P. Coupling Quantum Interpretative Techniques: Another Look at Chemical Mechanisms in Organic Reactions. J. Chem. Theory Comput. 2012, 8, 3993−3997. (c) Fang, D.; Chaudret, R.; Piquemal, J.-P.; Cisneros, G. A. Toward a Deeper Understanding of Enzyme Reactions Using the Coupled ELF/NCI Analysis: Application to DNA Repair Enzymes. J. Chem. Theory Comput. 2013, 9, 2156−2160. (16) Kräuter, T.; Neumüller, B. Triphenylphosphane complexes of copper(I): Structural and 31P NMR investigations. Polyhedron 1996, 15, 2851−2857. (17) Green, J.; Sinn, E.; Woodward, S.; Butcher, R. Mechanistic insights into catalytic cyclopropanation by copper(I) phosphine complexes. X-ray crystal structures of [Cu(FBF3) (PCy3)2] (Cy = cyclo-C6H11) and [Cu(MeCN)2{1,2-C6H4CH2NMe2(PPh2)}]BF4. Polyhedron 1993, 12, 991−1001. (18) Bayler, A.; Schier, A.; Bowmaker, G. A.; Schmidbaur, H. Gold Is Smaller than Silver. Crystal Structures of [Bis(trimesitylphosphine)gold(I)] and [Bis(trimesitylphosphine)silver(I)] Tetrafluoroborate. J. Am. Chem. Soc. 1996, 118, 7006−7007.

diffractometer at 100 K with graphite-monochromated Mo Kα (0.7107 Å) radiation. All structures were solved by direct methods and refined based on F2 by use of the SHELX program package as implemented in WinGX.21 All non-hydrogen atoms were refined using anisotropic displacement parameters. Hydrogen atoms attached to carbon atoms were included in geometrically calculated positions using a riding model. Disorder was resolved for solvating CH2Cl2 molecules with split occupancies for Cl3 of 1·CuCl·CH2Cl2 (80:20) and for Cl4 [1· Cu(NCMe)]BF4·2CH2Cl2 (75:25). A second even more heavily disordered CH2Cl2 molecule was accounted for using the SQUEZZE routine. Crystal and refinement data are collected in Table S1. Figures were created using DIAMOND.22 Crystallographic data (excluding structure factors) for structural analyses have been deposited with the Cambridge Crystallographic Data Centre, CCDC nos. 1495698− 1495700. Copies of this information may be obtained free of charge from The Director, CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (Fax: + 44-1223-336033; e-mail: [email protected] or http:// www.ccdc.cam.ac.uk). Computational Methodology. For the solid-state molecular geometries of 1, 1·CuCl, and [1·Cu(NCMe)]+ density functional theory (DFT) calculations were performed at the B3PW91/6311+G(2df,p)23 level of theory applying Gaussian09.24 C−H distances were extended to match listed values for neutron diffraction data.25 For the Cu atoms effective core potentials (ECP28MDF)22 and corresponding cc-pVTZ basis sets22 were utilized. The wave function files were used for topological analysis of the electron density according to the atoms-in-molecules space-partitioning scheme10 using AIM2000,25 whereas DGRID26 was used to generate and analyze the electron-localizability-indicator (ELI-D) related real-space bonding descriptors11 applying a grid step size of 0.05 au. The NCI grids were computed with NCIplot.27 Bond paths are displayed with AIM2000;25 ELI-D and NCI figures are displayed with MolIso.28,29

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.6b02056. NMR spectra of 1·CuCl and [1·Cu(NCMe)]BF4, crystal data and structure refinement data of 1, 1·CuCl, and [1· Cu(NCMe)]BF4, as well as experimental interatomic distances; AIM, ELI-D, and NCI representations (PDF) (CIF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS The Deutsche Forschungsgemeinschaft (DFG) is gratefully acknowledged for financial support. REFERENCES

(1) Pyykkö , P. Strong Closed-Shell Interactions in Inorganic Chemistry. Chem. Rev. 1997, 97, 597−636. (2) (a) Pyykkö, P. Relativistic Effects in Structural Chemistry. Chem. Rev. 1988, 88, 563−594. (b) Pyykkö, P. The Physics behind Chemistry and the Periodic Table. Chem. Rev. 2012, 112, 371−384. (c) Pyykkö, P. Relativistic Effects in Chemistry: More Common Than You Thought. Annu. Rev. Phys. Chem. 2012, 63, 45−64. (3) (a) Schmidbaur, H. The aurophilicity phenomenon: A decade of experimental findings, theoretical concepts and emerging applications. Gold Bulletin 2000, 33, 3−10. (b) Schmidbaur, H.; Schier, A. A H


Article

Inorganic Chemistry (19) (a) Schwerdtfeger, P.; Dolg, M.; Schwarz, W. H. E.; Bowmaker, G. A.; Boyd, P. D.W. Relativistic effects in gold chemistry. I. Diatomic gold compounds. J. Chem. Phys. 1989, 91, 1762−1774. (b) Schwerdtfeger, P. Relativistic effects in gold chemistry. 2. The stability of complex halides of gold(III). J. Am. Chem. Soc. 1989, 111, 7261−7262. (c) Schwerdtfeger, P.; Boyd, P. D. W.; Burrell, A. K.; Robinson, W. T.; Taylor, M. J. Relativistic effects in gold chemistry. 3. Gold(I) complexes. Inorg. Chem. 1990, 29, 3593−3607. (d) Liao, M. S.; Schwarz, W. H. E. Effective radii of the monovalent coin metals. Acta Crystallogr., Sect. B: Struct. Sci. 1994, 50, 9−12. (e) Pyykkö, P. Theoretical Chemistry of Gold. Angew. Chem., Int. Ed. 2004, 43, 4412−4456. (20) Kubas, G. J. Tetrakis(acetonitrile)copper(I) hexafluorophosphate. Inorg. Synth. 1979, 19, 90−92. (21) (a) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64, 112−122. (b) Farrugia, L. J. WinGX suite for small-molecule single-crystal crystallography. J. Appl. Crystallogr. 1999, 32, 837−838. (22) Brandenburg, K. DIAMOND, version 3.2i; Crystal Impact GbR: Bonn Germany, 2012. (23) (a) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson, M. R.; Singh, D. J.; Fiolhais, C. Atoms, molecules, solids, and surfaces: Applications of the generalized gradient approximation for exchange and correlation. Phys. Rev. B: Condens. Matter Mater. Phys. 1992, 46, 6671−6687. (b) Becke, A. D. Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993, 98, 5648−5652. (c) Peterson, K. A.; Figgen, D.; Goll, E.; Stoll, H.; Dolg, M. Systematically convergent basis sets with relativistic pseudopotentials. II. Small-core pseudopotentials and correlation consistent basis sets for the post-d group 16−18 elements. J. Chem. Phys. 2003, 119, 11113−11123. (d) Metz, B.; Stoll, H.; Dolg, M. Small-core multiconfiguration-Dirac−Hartree−Fock-adjusted pseudopotentials for post- d main group elements: Application to PbH and PbO. J. Chem. Phys. 2000, 113, 2563−2569. (e) Peterson, K. A. Systematically convergent basis sets with relativistic pseudopotentials. I. Correlation consistent basis sets for the post-d group 13−15 elements. J. Chem. Phys. 2003, 119, 11099−11112. (24) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian09, revision B.01; Gaussian, Inc.: Wallingford, CT, 2010. (25) Wilson, A. J. C. International Tables of Crystallography; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1992; Vol. C. (26) Biegler-König, F.; Schönbohm, J.; Bayles, D. AIM2000 - A Program to Analyze and Visualize Atoms in Molecules. J. Comput. Chem. 2001, 22, 545−559. (27) Kohout, M. DGRID-4.6; Radebeul, 2015. (28) Contreras-García, J.; Johnson, E.; Keinan, S.; Chaudret, R.; Piquemal, J.-P.; Beratan, D.; Yang, W. NCIPLOT: A Program for Plotting Noncovalent Interaction Regions. J. Chem. Theory Comput. 2011, 7, 625−632. (29) Hübschle, C. B.; Luger, P. MolIso - a program for colourmapped iso-surfaces. J. Appl. Crystallogr. 2006, 39, 901−904.

I


Role of Dispersion in Metallophilic HgÂ·Â·Â·M Interactions (M = Cu, Ag, Au

Recommend Documents