Dichloro-μ₂,μ₂-naphthalene-1,8-diyl-bis(N,N,N′,N′-tetramethylethylenediamino)tetracopper(I)

Abstract

A highly reactive dicuprate/CuCl aggregate, Nap(Cu₄Cl₂)(TMEDA)₂ (2, Nap = naphthalene-1,8-diyl, TMEDA = tetramethylethylenediamine), was synthesized by the reaction of 1,8-dilithionaphthalene(TMEDA)₂ with four equivalents of CuCl. The X-ray crystal structure of this complex shows that the four copper atoms form a bent parallelogram-shaped core, with terminally bonded chlorine atoms. The naphthalene ring is bonded through carbons in the 1 and 8 positions (peri-positions), each bridging two copper atoms.

1. Introduction

Organocopper reagents are among the most important organometallic reagents available to chemists [1,2]. Organocuprates, RCu, are well established as milder alternatives to Grignard reagents. A common method of cuprate synthesis is the reaction of a Grignard reagent or an organolithium compound with an anhydrous copper(I) halide, as illustrated in the following general equations:

R-X + Mg → R-MgX

R-MgX + CuX’ → 1/n (R-Cu)_n + MgXX’

Like other organometallic species, organocuprates tend to form aggregates in the solid state. The aggregation can range from species with two copper atoms to polymers.

Tetrameric species, which are relevant to this work, are quite common among organyl copper clusters. Homoleptic copper aryl complexes typically crystallise as tetramers with a nearly planar Cu₄ core, with four carbon atoms each adopting a μ₂-bridged co-ordination mode. This is observed for various aryl groups, such as pentamethylphenyl [3], 2,4,6-triethylphenyl [4], 2,4,6-triisopropyl phenyl [5], pentaflourophenyl [6], and mesityl (for which both tetramers and pentamers have been reported) [7]. Additionally, a heteroaryl variant (thienylcopper) has been crystallised as both tetrameric and pentameric clusters [8].

Square planar Cu₄ core structures have also been reported for Cu₄L₄ aryl cuprates with adjacent tertiary amino groups, resulting in N-coordinated copper centres in the cases of phenyl [9] and naphthalene [10], but, interestingly, not in ferrocene [11]. Even after replacing some of the organyl groups with alumoxane groups, the square planar geometry of the Cu₄ core has been preserved, as shown in several β-diketiminato complexes reported by Roesky [12].

While (nearly) square planar clusters appear to be the most common motif for tetrameric organocuprates, other geometries have also been observed. For example, the coordination of additional ligands, such as PMe₃ in (MesCu)₄(PMe₃)₂ (Mes = mesityl) [13], or thiophene in (MesCu)₄(thiophene)₂ [14], resulted in elongation along the diagonal to a planar parallelogram-shaped Cu₄ core, with the formation of a transannular Cu-Cu bond. Interestingly, a rigid peri-substituted acenaphthene backbone bearing a PPh₂ group also supported a similar Cu₄L₄ (L = acenaphthene-6-diphenylphosphino-5-yl) tetranuclear geometry with a transannular Cu-Cu bond [15].

A few tetrameric species with a non-planar (bent) parallelogram Cu₄ core have also been reported, such as in the π-coordinated [Cu(C₆F₅)]₄(η²-toluene)₂ aggregate [6], or the related [Cu(4-F₃CC₆F₄)]₄(η²-toluene)₂ [16].

Even more closed structures of the Cu₄ cluster were found in a series of NCN pincer complexes reported by Wang [17]. These feature tetrahedral Cu₄ units in which two faces are capped by μ₃-bridging halide atoms and two edges are μ₂-bonded to the organyl groups.

In our attempts to find a derivative with a modified reactivity based on the well-established synthon 1,8-dilithionaphthalene(TMEDA)₂ (1), we have synthesised and characterised the tetrameric copper complex 2. In this report, we present its synthesis and structure.

2. Results and Discussion

1,8-Dilithionaphthalene(TMEDA)₂ [18,19] (1, Scheme 1) has previously been used as a starting material in the synthesis of peri-substituted compounds. For example, Schmutzler used it to prepare an extensive series of 1,8-bis(phosphino)naphthalenes by reacting it with chlorophosphines [20,21,22,23], and Mizuta showed that its reaction with dichlorophosphines yields either peri-bridged phosphines or diphosphines [24,25]. However, compound 1 has some drawbacks, as it cannot be stored for more than a few days without decomposing and is extremely reactive. It was hypothesised that transmetallation with copper might yield a related dicuprate, which could be a highly useful starting material, potentially a more stable, milder, and more selective nucleophile. The synthesis of NapCu₂ was attempted by reacting 1 with two equivalents of CuCl in thf at −78 °C, followed by gradual warming to room temperature while stirring. After removing volatiles in vacuo, toluene was added to the residue. The resulting suspension was filtered through a sinter with filtration aid (porosity 3, Celite). The ¹H NMR spectrum of the mixture after the reaction indicated the formation of a complex mixture of products. A few crystals were obtained from the toluene solution, and X-ray diffraction showed these to be the cluster species Nap(Cu₄Cl₂)(TMEDA)₂ (2).

A more straightforward synthesis of 2 was then developed. The reaction of 1 with four equivalents of CuCl under the same conditions mentioned above led to the isolation of 2 as a red solid in 23% yield (after recrystallisation from toluene) (see Scheme 1). ¹H and ¹³C{¹H} NMR spectra of crude 2 are consistent with the relatively clean formation of a single product, although the spectra show contamination by other compounds with a similar composition. 2 is extremely sensitive to air and moisture, and, unfortunately, it also turns out to be thermally unstable. It cannot be stored for long periods, even at low temperature (−18 °C). The integration of the ¹H NMR spectra also indicated that some of the solvated TMEDA is removed during drying in vacuo.

More crystals suitable for X-ray crystallography were obtained from a toluene solution of 2 and proved to have the same structure as the one obtained from the reaction with two equivalents of CuCl. The X-ray crystal structure of 2 is shown in Figure 1, and selected crystallographic data are presented in Table S1. Compound 2 can be considered a cluster formed from the aggregation of the originally desired cuprate species, NapCu₂, with two CuCl molecules, with all copper atoms being formally in the +1 oxidation state.

The molecule of 2 in the crystal exhibits approximate C₂ symmetry. It has a partially folded (non-planar) parallelogram-shaped Cu₄ core, with each of the two chloride atoms bonded terminally to one copper atom, positioned on opposite sides of the Cu₄ core (Figure 2c). The naphthalene ring spans the Cu₄ core, with the two peri-carbon atoms acting as μ₂-bridges over two copper atoms each. Additionally, each of the two copper atoms not bound to chloride is coordinated to a molecule of TMEDA, forming five-membered chelate rings. The Cu-Cl bond lengths are very similar [Cu1-Cl1 2.1346(9) Å, Cu9-Cl2 2.1496(7) Å], while the Cu-C bond lengths also fall within a narrow range [1.974(3)-2.014(3) Å]. The wider observed range of Cu-N bond lengths [2.031(2)- 2.199(2) Å] indicates a more flexible coordination of TMEDA.

The Cu-Cu bond lengths within the parallelogram motif differ quite significantly from the two shorter sides, being 2.4109(5) Å (Cu1-Cu11) and 2.3652(5) Å (Cu9-Cu21), and the longer sides measuring 2.7517(6) (Cu11-Cu9) and 2.8376(7) Å (Cu21-Cu1). The diagonal Cu1-Cu9 bond is longer, at 2.8065(7) Å. The parallelogram shape of the Cu₄ core results in two obtuse angles [Cu11-Cu9-Cu21 112.17(2)°, Cu11-Cu1-Cu21 107.94(2)°] and two acute angles [Cu1-Cu11-Cu9 64.475(18)°, Cu1-Cu21-Cu9 64.544(19)°]. The Cu-C-Cu angles are relatively acute at 74.90(10)° (Cu1-C1-Cu11) and 72.48(10)° (Cu9-C9-Cu21), while the N-Cu-N angles are close to a right angle [N11-Cu11-N14 86.67(9)°, N21-Cu21-N24 87.85(9)°]. The C-Cu-Cl angles are somewhat short of linear at 161.65(7)° (Cl1-Cu1-C1) and 171.36(8)° (Cl2-Cu9-C9).

The Cu₄ parallelogram is partially folded, with planes defined by the Cu1-Cu9-Cu11 and Cu1-Cu9-Cu21 atoms forming an angle of 28°. The naphthalene ring is nearly planar, with a C1-C10-C5-C6 dihedral angle of 178.2(2)°, indicating that little strain is imposed on it as it bonds with the copper core.

Only a limited number of Cu₄X₂ copper(I) halide aggregates have been reported in the literature. The indolyl-based NCN pincer ligand-supported clusters recently described by Wang comprise a tetrahedral Cu₄ core, as shown in Figure 2a (CSD identifier DEXLUR) [17]. These clusters were obtained by reacting Cu(I) halides with half an equivalent of the lithiated pincer ligand.

A more open, square planar Cu₄ core, with coplanar μ₂-bridging bromine and μ₂-bridging carbon atoms, is observed in clusters with bulky hydrindacene-based ‘Rind’ groups (CSD identifier LUTLOC, Figure 2b) [26]. These clusters were synthesised by reacting organolithiums (RindLi) with two equivalents of CuBr.

The bent parallelogram Cu₄Cl₂ motif in 2 (Figure 2c) appears to be an intermediate between the two structural types mentioned above, with the parallelogram being more open than the tetrahedral core shown in Figure 2a, but less open than the square planar motif shown in Figure 2b. The structure of 2 is also unique in terms of the terminal coordination mode of the chlorides; the halides in the other two clusters are μ₃ and μ₂ bridging, respectively. We believe the unique structural features of 2 are influenced by the rigid geometry of the naphthalene-1,8-diyl backbone, which dictates the relatively short distance between the two peri-carbon atoms coordinated to the copper core.

3. Materials and Methods

3.1. General Considerations

All synthetic manipulations were performed under an atmosphere of dry nitrogen using standard Schlenk techniques or under an argon atmosphere in a Saffron glove box. All items of glass apparatus were stored in a drying oven (ca. 120 °C) prior to use. Dry solvents were collected from an MBraun Solvent Purification System (Garching Germany) and stored over appropriate molecular sieves. Chemicals were taken from the laboratory inventory and used without further purification.

All NMR spectra were recorded at 25 °C using a Bruker Avance (300 MHz, Billerica, MA, USA) spectrometer. Assignments of ¹H and ¹³C spectra were made in conjunction with H-C HSQC spectra. ¹³C NMR spectra were recorded using the DEPTQ pulse sequence with broadband proton decoupling. For ¹H and ¹³C NMR, tetramethylsilane was used as an external standard. Residual solvent peaks were used for secondary calibration (CD₃CN H 1.96 ppm (CH₃); C 118.26 ppm (C≡N). Chemical shifts (δ) are given in parts per million (ppm). Melting and decomposition points were determined by heating solid samples in sealed glass capillaries using a Stuart SMP30 Melting Point Apparatus (Cole-Parmer, Cambridge, UK).

3.2. Synthesis of 1, NapLi₂∙(TMEDA)₂

A literature procedure [19] was followed with slight modifications. nBuLi (9.6 mL of 2.5 M solution in hexanes, 22.5 mmol) was added to diethyl ether (30 cm³) at −20 °C. 1-Bromonaphthalene (4.14 g, 2.78 mL, 20 mmol) was added to the mixture over 2 min. The temperature was allowed to rise to 10 °C while stirring. Solvent was removed in vacuo to yield 1-lithionaphthalene as a white solid, which was dried in vacuo for 30 min. Hexane (15 mL) was added, followed by nBuLi (10.4 mL of 2.5 M solution in hexanes, 26 mmol) and TMEDA (3.26 g, 4.24 mL, 28 mmol) and the suspension was heated under reflux for 3 h, or until there was no gas evolution. 1 was obtained as a red-brown solid after removing the volatiles in vacuo and was used for further synthesis immediately.

3.3. Synthesis of 2, Nap(Cu₄Cl₂)(TMEDA)₂

1 (3.73 g, 10 mmol) in thf (15 mL) was added to a suspension of CuCl (3.96 g, 40 mmol) in thf (10 mL) at −78 °C over 30 min. The resulting dark red suspension was allowed to warm to room temperature and stirred for 16 h. Volatiles were removed in vacuo and toluene (30 mL) was added. The resulting red solution was filtered through a sinter (porosity 3, with filtration aid celite). The undissolved solid and celite/sinter were washed with toluene (20 mL). Volatiles were removed in vacuo to give crude 2 as a red solid. Recrystallisation from toluene at 2 °C gave 2 as red crystals (1.57 g, 23%). Some of these were suitable for X-ray crystallography.

¹H NMR (300.0 MHz, CD₃CN): δ_H 2.40 (s, 24H, 8 × TMEDA CH₃), 2.49 (s, 8H, 4 × TMEDA CH₂), 7.11–7.30 (m, 2H, Ar CH), 7.47–7.56 (m, Ar CH), 7.86–7.93 (m, Ar CH).

¹³C{¹H} NMR (75.5 MHz, CD₃CN): δ_C 48.3 (s, TMEDA, CH₃), 58.6 (s, TMEDA, CH₂), 125.5 (s, Ar CH), 126.9 (s, Ar CH), 128.7 (s, Ar CH), 133.7 (s, Ar q-C). Several quaternary carbon atoms were not observed with certainty due to low solubility. M.p. 134–136 °C (decomp).

3.4. Crystallographic Details

X-ray diffraction data for compound 2 were collected at 93 K using a Rigaku MM007 High Brilliance RA generator/confocal optics [Mo Kα radiation (λ = 0.71073 Å)] and Mercury CCD system. Intensity data were collected using both ω and φ steps, accumulating area detector images spanning at least a hemisphere of reciprocal space. Data were collected and processed (including correction for Lorentz, polarisation, and absorption) using CrystalClear [27]. The structure was solved by direct methods (SIR2004 [28]) and refined by full-matrix least-squares against F2 (SHELXL-2019/3 [29]). Non-hydrogen atoms were refined anisotropically, and hydrogen atoms were refined using a riding model. The structure showed orientational disorder in part of one of the TMEDA ligands; this was modelled with split atom sites, without restraint. All calculations were performed using the CrystalStructure [30] interface. Selected crystallographic data are presented in Table S1. CCDC 2347905 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures.

4. Conclusions

A tetranuclear organocopper(I) cluster has been synthesised and its crystal structure has been determined. The rigid peri-backbone appears to impose a bent parallelogram structure in the Cu₄ core, with terminally bonded chloride atoms and μ₂-bonded carbon atoms from the naphthalene group. The cluster’s high reactivity towards air/moisture limits its characterisation to some extent.