Synthesis of two copper clusters and their catalysis towards the oxidation of benzene into phenol

Abstract

Reaction of a heptadentate ligand (H₂L^I) and a hexadentate ligand (H₂L^II) with three equivalents of Cu(ClO₄)₂·6H₂O in methanol under basic conditions afforded a hexanuclear cluster [Cu₆(L^I)₂(OH)₄](ClO₄)₄·2DMF·3Et₂O (1) and a nonanuclear cluster [(Cu₉(L^II)₃(OH)₇)](ClO₄)₅·0.25CH₃OH·1.15H₂O (2), respectively, where H₂L^I is 2,2′-(((pyridine-2,6-diylbis(methylene))bis((pyridin-2-ylmethyl)azanediyl))bis(methylene))diphenol and H₂L^II 2,2′-((((5-methyl-1,3-phenylene)bis(methylene))bis((pyridin-2-ylmethyl)azane-diyl))bis(methylene))bis(4-methylphenol). The structures of both clusters in their solid states have been determined by X-ray crystallography. Their magnetic susceptibilities reveal that both clusters are antiferromagnetic due to the coupling between the copper centers in the clusters. NMR spectroscopic analysis, conductivity measurements and ESI-MS analysis suggest that the clusters retain their structural integrities in solution. Both clusters show catalytic activity towards the hydroxylation of benzene into phenol with hydrogen peroxide (H₂O₂) as an oxidant at 80 °C in aqueous acetonitrile. The conversion rate is about 20% and their TON/TOF are 564/188 and 905/302 for clusters 1 and 2, respectively.

---

Introduction

Transition metal clusters are a large category of functional materials assembled through coordination chemistry and intermolecular interactions such as hydrogen bonding between metal ions and organic ligands.^1,2 As a unique category of nanometric materials, the transition metal clusters were used as heterogeneous catalysts,³ nonlinear optical materials,⁴ semi-conducting materials,⁵ as well as mimics of the active sites of metalloproteins.⁶ Among the many transition metal clusters, copper clusters have intensively been investigated in the past decades due to their potential applications in these areas.^7,8 Apart from these potential applications, catalysis of the copper clusters is another functionality found not only in synthetic systems,^9–11 but also in metalloenzymes.^12,13 One such an example is the particulate methane monooxygenase (pMMO), an enzyme of dicopper active centers, which catalyzes the conversion of methane to methanol.^14–16 It is also known that multinuclear copper complexes are good catalysts and/or precatalysts for homogeneous oxidative functionalization of alkanes into alkyl hydroperoxides, alcohols, and ketones under mild conditions.¹⁷ Common features for the catalyst of copper clusters are found to be: (i) possessing N,O-donor set, (ii) having insufficient coordination numbers for Cu centers, (iii) decent stability and solubility, and (iv) the relatively close separation between Cu atoms.¹⁸

Although there have been considerable efforts devoted to the oxidative functionalization of organic substrates using copper complexes including copper clusters as catalysts,^19–32 using these categories of catalysts to promote the direct hydroxylation of benzene into phenol has not been well developed,^33–35 particularly using copper clusters for this catalysis.⁶ Since the existing methodologies have a number of disadvantages, for example, low yield and poor selectivity,^36,37 to explore this catalytic chemistry is of practical significance.^38,39 But one challenge of pushing this chemistry ahead is to synthesize copper clusters with certain nuclearity. It is well known that both phenolate and hydroxyl groups are good bridging linkages in assembling metal clusters.^40–42 Therefore, designing and synthesizing appropriate multidentate ligands are essential. Herein, we report two novel ligands, 2,2′-(((pyridine-2,6-diylbis(methylene))bis((pyridin-2-ylmethyl)azane diyl))bis(methylene))diphenol ligand (H₂L^I) and 2,2′-((((5-methyl-1,3-phenylene)bis(methylene))bis((pyridin-2-ylmethyl)a-zanediyl))bis(methylene))bis(4-methylphenol) (H₂L^II), and two copper clusters assembled from the two ligands. HL^I is a heptadentate ligand and the other a hexadentate ligand. Both ligands possess pyridinyl N, tertiary amine N and phenolate O ligating atoms. Reacting these ligands with Cu(ClO₄)₂·6H₂O under basic condition led to two highly charged clusters [Cu₆(L^I)₂(OH)₄](ClO₄)₄·2DMF·3Et₂O (1) and [(Cu₉(L^II)₃(OH)₇)](ClO₄)₅·0.25CH₃OH·1.15H₂O (2), respectively. Their single crystal structures were determined and magnetic behaviors were analyzed. Catalytic assessment showed that the two clusters do catalyze the oxidation of benzene into phenol with H₂O₂ as an oxidant, and the conversion rate is about 20%.

Experimental

Materials and methods

Reactions were performed under an argon atmosphere using standard Schlenk techniques when necessary. Solvents were freshly distilled by using appropriate drying agents prior to use. ¹H NMR and ¹³C NMR spectra were recorded on an AVANCE DRX 400 (Bruker) spectrometer in CDCl₃. UV-vis spectra were recorded on a Cary 50 COCN (Varian) spectrophotometer at room temperature. Infrared data were recorded on an FTIR spectrometer (Scimitar 2000, Varian). Microanalysis was performed on a Heraeus CHN-O-Rapid analyzer. Catalytic reactions were analyzed by GC (gas chromatography) on an Agilent instrument (model 6890N network GC system) equipped with a DB-WAX capillary column (column length: 30 m; internal diameter: 0.32 mm; film thickness: 0.25 μm). Conductivity was measured on a conductivity analyzer (LeiCi, DDS-307A). ESI-MS were performed on an Agilent 1200/6200 mass spectrometer using MeCN as mobile phase. Magnetic measurements were performed on a magnetic property measurement system (MPMS XL-7, QUANTUM DESIGN, USA). Compounds 2,6-bis(bromomethyl)pyridine,⁴³ 1,3-bis(bromomethyl)-5-methyl-benzene,⁴⁴ 2-hydroxy-5-methylbenzaldehyde,⁴⁵ and 4-methyl-2-(((pyridin-2-ylmethyl)amino)-methyl)phenol⁴⁶ were synthesized using modified literature procedures.

Synthesis of 2,6-bis(bromomethyl)pyridine. A mixture of 2,6-bis(methyl)pyridine (9.3 g, 85 mmol) and N-bromosuccinimide (NBS) (33.0 g, 170 mmol) in CCl₄ (400 mL) was refluxed in the presence of azodiisobutyronitrile (AIBN) (1.5 g, 9 mmol) for 6 h under the irritation of 100 W lamp. After being cooled to room temperature, the mixture was filtered and the filtrate was concentrated in vacuo. Purification was achieved by flash column chromatography using ethyl acetate/petroleum ether = 1/5 as eluent. The second band was collected, combined and evaporated in vacuo to afford a white crystal (6.5 g, 30%). ¹H NMR (δ, ppm, 400 MHz, CDCl₃): 4.55 (s, CH₂), 7.38 (d, Py, J = 7.76 Hz), 7.72 (q, Py).

Synthesis of 1,3-bis(bromomethyl)-5-methylbenzene. A mixture of NBS (22.1 g, 125 mmol), benzoyl peroxide (BPO) (0.3 g, 1.2 mmol) and mesitylene (7.0 mL, 50 mmol) in CCl₄ (150 mL) was refluxed for 5 h. After being cooled to room temperature, the mixture was filtered and the filtrate was concentrated by rotary evaporation. Purification was performed using flash chromatography on a silica gel column with petroleum ether as eluent to give a white solid (3.61 g, 27%). ¹H NMR (δ, ppm, 400 MHz, CDCl₃): 7.36 (d, J = 19.2 Hz, 1H, Ph), 7.19 (t, J = 16 Hz, 2H, Ph), 4.45 (s, 4H, CH₂), 2.36 (d, J = 9.6 Hz, 3H, CH₃).

Synthesis of compound 2-hydroxy-5-methylbenzaldehyde. p-Methyl phenol (5.1 g, 47.3 mmol), paraformaldehyde (10.0 g), Et₃N (60 mL) and MgCl₂ (6.6 g, 69.6 mmol) were mixed in THF (100 mL). After being refluxed for 24 h, the reaction mixture was cooled to room temperature, and its pH was adjusted to 3 with concentrated hydrochloric acid. The resultant aqueous solution was extracted successively with ethyl acetate (50 mL × 3). The organic layers were separated, combined and dried with Na₂SO₄. Removal of the solvent gave a crude product which was purified using flash chromatography on a silica gel column with ethyl acetate/petroleum ether = 1/10 as eluents to give a pale yellow solid (3.7 g, 73%). ¹H NMR (δ, ppm, 400 MHz, CDCl₃): 10.77 (s, 1H, CHO), 9.79 (s, 1H, OH), 7.28 (s, 1H, Ph), 7.27 (s, 1H, Ph), 6.83 (t, J = 4.6, 1H, Ph), 2.27 (s, 3H, CH₃).

Synthesis of 4-methyl-2-(((pyridin-2-ylmethyl)amino)-methyl)phenol. To a solution of 2-hydroxy-5-methyl benzaldehyde (3.3 g, 30.7 mmol) in methanol (40 mL) was added pyridin-2-ylmethanamine (3.1 mL, 30.8 mmol) under argon atmosphere. After being stirred for 3 h at room temperature, the reaction was cooled to ice temperature, and sodium borohydride (1.7 g, 44.7 mmol) was added. The mixture was continually stirred for 3 h at room temperature. After removal of the solvent under reduced pressure, distilled water (30 mL) was added, and the mixture was extracted successively with dichloromethane (30 mL × 3). All the extracts were combined and dried with Na₂SO₄. Removal of the solvents gave a crude product which was further purified using flash chromatography on a silica gel column with ethyl acetate as eluent to give yellow oily liquid (4.7 g, 67%). ¹H NMR (δ, ppm, 400 MHz, CDCl₃): 8.58 (d, J = 4.6 Hz, 1H, Py), 7.65 (m, 1H, Py), 7.20 (m, 2H, Py), 6.97 (m, 1H, Ph), 6.77 (m, 2H, Ph), 3.97 (s, 2H, CH₂), 3.91 (s, 2H, CH₂), 2.24 (s, 3H, CH₃).

Synthesis of ligand H₂L^I. 2,6-Bis(bromomethyl)pyridine (265 mg, 1.0 mmol), 4-methyl-2-(((pyridin-2-ylmethyl)amino)methyl)-phenol (684 mg, 3.0 mmol) and K₂CO₃ (690 mg, 5.0 mmol) were mixed in acetonitrile (80 mL). After being refluxed for 36 h, the reaction was filtered, and the filtrate was concentrated by rotary evaporation to give a crude product. Purification was performed by flash chromatography on a silica gel column with ethyl acetate/petroleum ether = 10 : 1 as eluent to give a white solid (340 mg, 61%). ¹H NMR (δ, ppm, 400 MHz, CDCl₃): 10.71 (s, 2H, OH), 8.57 (d, J = 4.8 Hz, 2H, Py), 7.62 (m, 2H, Py), 7.56 (m, 1H, Py), 7.32 (d, 2H, Py), 7.26 (m, 2H, Py), 7.16 (m, 2H, Py), 6.98 (d, J = 8.1 Hz, 2H, Ph), 6.86 (s, 2H, Ph), 6.79 (d, J = 8.1 Hz, 2H, Ph), 3.86 (s, 8H, CH₂), 3.75 (s, 4H, CH₂), 2.23 (s, 6H, CH₃). ¹³C NMR: (aromatic C: 158.2, 157.8, 155.1, 148.9, 137.4, 136.7, 130.6, 129.4, 127.9, 123.2, 122.4, 122.2, 121.7, 116.2), (methylene C: 59.2, 59.0, 57.1), (methyl C: 20.4). Element analysis for C₃₅H₃₇N₅O₂ (FW = 559.7), calc. (%): C, 75.11; H, 6.66; N, 12.51, found (%): C, 75.58; H, 6.93; N, 12.64.

Synthesis of ligand H₂L^II. 1,3-Bis(bromomethyl)-5-methylbenzene (667 mg, 2.4 mmol), 4-methyl-2-(((pyridin-2-ylmethyl)amino) methyl)phenol (1140 mg, 5.0 mmol) and K₂CO₃ (690 mg, 5.0 mmol) were mixed in dichloromethane (50 mL). After being refluxed for 12 h, the reaction was filtered, and the filtrate was concentrated by rotary evaporation to give a crude product. Purification was performed by flash chromatography on a silica gel column with ethyl acetate/petroleum ether = 1.5 : 1 as eluent to give a white solid (212 mg, 37%). ¹H NMR (δ, ppm, 400 MHz, CDCl₃): 10.66 (s, 2H, OH), 8.59 (d, J = 4.8 Hz, 2H, Py), 7.64 (t, J = 6.9 Hz, 2H, Py), 7.28 (s, 2H, Py), 7.18 (t, J = 6.9 Hz, 2H, Py), 7.10 (s, 1H, Ph), 7.03 (s, 2H, Ph), 6.97 (d, J = 8.1 Hz, 2H, Ph), 6.84 (s, 2H, Ph), 6.78 (d, J = 8.1 Hz, 2H, Ph), 3.77 (s, 4H, CH₂), 3.72 (s, 4H, CH₂), 3.63 (s, 4H, CH₂), 2.32 (s, 3H, CH₃), 2.25 (s, 6H, CH₃). ¹³C NMR (δ, ppm, 400 MHz, CDCl₃): (aromatic C: 157.8, 155.1, 149.0, 138.3, 137.4, 136.7, 130.1, 129.3, 129.2, 128.0, 127.7, 123.4, 122.3, 122.1, 116.0), (methylene C: 58.9, 57.9, 57.1), (methyl C: 21.4, 20.4). Element analysis for C₃₇H₄₀N₄O₂ (FW = 572.7), calc. (%): C, 77.59; H, 7.04; N, 9.78, found (%): C, 77.46; H, 6.73; N, 9.56.

Synthesis of cluster 1. To a solution of ligand H₂L^I (55.9 mg, 0.1 mmol) in methanol (10 mL) was added Cu(ClO₄)₂·6H₂O (74.2 mg, 0.2 mmol). Addition of Et₃N (0.15 mL, 1.0 mmol) into the reaction produced blue precipitate under stirring. The precipitate was collected via filtration and washed with dichloromethane (5 mL × 3). Recrystallisation of the solid in DMF/Et₂O mixture accrued blue crystals suitable for X-ray analysis in four days (63.0 mg, 60%). Element analysis for C₇₀H₇₄Cu₆N₁₀O₂₄Cl₄·10H₂O (FW = 2142.6), calc. (%): C, 39.24; H, 4.42; N, 6.54, found (%): C, 39.58; H, 4.93; N, 6.64. IR (cm⁻¹): 3400 (s), 3011 (w), 2919 (w), 2865 (w), 1607 (s), 1489 (s), 1442 (m), 1275 (s), 1142 (s), 1108 (s), 1087 (s), 625 (m). UV-vis (DMF, λ_max/nm (ε/mol⁻¹ L cm⁻¹)): 228 (76 502), 255 (59 380), 310 (25 852), 449 (2920).

Synthesis of cluster 2. To a solution of ligand H₂L^II (286 mg, 0.50 mmol) in methanol (10 mL) was added a solution of Cu(ClO₄)₂·6H₂O (557 mg, 1.5 mmol) in methanol (6 mL). Addition of Et₃N (0.35 mL, 2.5 mmol) into the above solution afforded deep green precipitate under stirring. The precipitate was collected via filtration and washed with MeOH (6 mL × 3). Recrystallization of the precipitate in MeCN/MeOH/Et₂O mixture produced green crystals (210 mg, 30%) suitable for X-ray single crystallographic analysis in three days at room temperature. Element analysis for C₁₁₁H₁₂₁N₁₂O₃₃Cl₅Cu₉·3MeOH·2CH₃CN (FW = 3078.6), calc. (%): C, 46.04, H, 4.55, N, 6.37, found (%): C, 46.58, H, 5.04, N, 5.98. IR (cm⁻¹): 3420 (s), 3008 (w), 2917 (w), 2856 (w), 1610 (s), 1482 (s), 1443 (m), 1283 (s), 1103 (s), 882 (m), 623 (m), 562 (w). UV-vis (DMF, λ_max/nm (ε/mol⁻¹ L cm⁻¹)): 225 (87 200), 255 (69 300), 299 (25 600), 449 (1900).

Typical procedure for catalytic oxidation of benzene with H₂O₂. To a reaction vessel (25 mL) were added a catalyst, benzene (10 mmol) and CH₃CN (3 mL) under vigorously stirring at the room temperature. To the solution was further added an aqueous solution of hydrogen peroxide (30%) before the reacting mixture was heated to appropriate temperature under continuously stirring for a period of time. After the reaction was cooled to room temperature, the reaction was then filtered, and the filtrate formed two layers. The sample taken from the upper organic layer was analyzed by gas chromatography. The same protocol was used to perform catalytic assessment under acidic condition by adding 0.56 μmol hydrochloric acid.

X-ray crystallography. In the crystallographic data collection of clusters 1 and 2, standard procedures were used for mounting the crystals on a Bruker SMART CCD diffractometer with graphite-monochromated MoKα radiation (λ = 0.71073 Å). The structures were solved by the direct methods routines in the SHELXS program and refined on F² in SHELXL.⁴⁷ Due to the partial evaporation of solvents and the severely disordered perchlorate anions, the crystal qualities of clusters 1 and 2 were poor. In cluster 1, one perchlorate anion exhibited disorder over two positions and the O(9)–O(12) atoms, bound to Cl(2) atom, were split into two fragments O(9)–O(12)/O(9A)–O(12A) with an occupancy factors of 0.46/0.54. In crystal 2, two perchlorate anions exhibited disorder over two positions: the O(18)–O(21) atoms bound to Cl(2) atom were split into two fragments O(18)–O(21)/O(18A)–O(20A) with an occupancy factors of 0.50/0.50; the O(30)–O(31) atoms bound to Cl(5) atom were split into two fragments O(30)–O(31)/O(30A)–O(310A) with an occupancy factors of 0.45/0.55. In cluster 1, the site-occupation factor for one diethyl ether molecule (C43–C46 and O15) were fixed at 0.5. In cluster 2, the site-occupation factor for one methanol molecule (O34, C112) was fixed at 0.25. For the three water molecules, the site-occupation factor for two of them (O35, O36) was fixed at 0.5 and the third one at 0.15. All non-hydrogen atoms were modeled anisotropically. All other hydrogen atoms were placed in idealized positions and treated as riding atoms with exception that the hydrogen atom of the hydroxyl group was added in a difference Fourier map. But the hydrogen atoms of water molecules were omitted in the structural refinement. Details of the data collection and refinement are given in Table S1.†

Results and discussion

Synthesis and characterization

The two ligands were synthesized according to modified literature procedure by using dibromides A (A′) and a tridentate compound (B) as building blocks as shown in Scheme 1.⁴⁸ The dibromides A (A′) were synthesized by reacting 2,6-bis(methyl)pyridine or mesitylene with brominating reagent N-bromosuccinimide (NBS) in the presence of the initiators azodiisobutyronitrile or benzoyl peroxide, respectively. The synthesis of compound B was achieved via three-step reactions. First, p-methyl phenol reacted with paraformaldehyde to give 2-hydroxy-5-methylbenzaldehyde, which was further reacted with pyridin-2-ylmethanamine to give Schiff base 2-(((pyridin-2-ylmethyl)imino)methyl)phenol. And the Schiff base was finally reduced into product B by NaBH₄. Reactions of 2,6-bis(bromomethyl)pyridine (A) or 1,3-bis(bromomethyl)-5-methyl-benzene (A′) with 2-(((pyridin-2-ylmethyl)amino)-methyl)phenol (B) in MeCN achieved the finial ligands H₂L^I and H₂L^II, respectively, Scheme 1.

Scheme 1 The synthetic routes for ligands H₂L^I, H₂L^II and their respective copper clusters, (for clarity, only ligating atoms are shown, and the rest of the ligands are simplified as either arc lines or short wave lines in clusters 1 and 2).

As shown in Scheme 1, both ligands are consisted of three types of ligating atoms, phenolate O, pyridinyl N and tertiary amine N. Reaction of ligands H₂L^I and H₂L^II with three equivalents of Cu(ClO₄)₂·6H₂O with the presence of Et₃N led to blue precipitate (1) and deep green precipitate (2), respectively. The reaction does not proceed without a base. Thus the addition of Et₃N is essential because the added Et₃N not only deprotonates the phenol, but also generates hydroxyl group. Both are critically important in assembling the clusters since they act as a bridge in the assembling of the clusters. Clusters 1 and 2 are insoluble in nonpolar solvents but soluble in polar solvents such as DMF and MeCN. Their good solubility in those solvents is somewhat surprising as both clusters are highly ionic. This property ought to originate from their unique structures in which the metal cores are well shielded with a hydrophobic layer offered by the organic motif of the encapsulating ligands (vide infra).

Structural analysis

Cluster 1 has a hexanuclear cation [Cu₆(L^I)₂(OH)₄]⁴⁺, whose charge is balanced by four ClO₄⁻ anions. Its ORTEP drawing is shown in Fig. 1. Selected bond lengths and angles are tabulated in Table 1. The six-copper ions in the structure are approximately arranged in two parallel rows, which are separated by about 3.0 Å and dispositioned by about 1.6 Å. The two ends of the rows are encapsulated by two ligands, respectively. In each row, three copper atoms are nearly in linear arrangement (∠Cu–Cu–Cu = 168.2°). There is a symmetric center in the structure which is located at the crossing point of connections Cu1–Cu1A and O4–O4A. This hexacopper assembly is not unprecedented, although the encapsulating ligands are quite different.⁴⁹ The coordination atmosphere of these copper cores falls into three categories as shown in Fig. 1, with Cu1 and Cu1A being coordinated by five oxygen atoms in a nearly perfect square-pyramidal geometry (τ = 0.087), Cu2 and Cu2A being coordinated by three nitrogen and two oxygen atoms, and Cu3 and Cu3A being coordinated by two nitrogen and three oxygen atoms. The square-pyramidal geometry of the other two types of copper centres are severely distorted with a τ value of 0.60 and 0.44, respectively.

Fig. 1 Crystal structure of the cation of cluster 1, hydrogen atoms were omitted for clarity, and thermal ellipsoids are drawn at the 30% probability level (left); the core of cluster 1 represented by ball and stick model (right).

Table 1 Selected bond lengths (Å) and angles (°) for cluster 1^a

a Symmetric code: a = −x, −y + 1, −z + 1.
Cu(1)⋯Cu(2) = 3.0245(19)	Cu(1)⋯Cu(3)^a = 2.9798(17)
Cu(1)–O(2) = 1.937(3)	Cu(1)–O(1)^a = 1.960(2)
Cu(1)–O(4)^a = 1.961(3)	Cu(1)–O(3) = 1.990(2)
Cu(1)–O(4) = 2.317(2)	Cu(2)–O(3) = 1.958(3)
Cu(2)–O(2) = 2.015(3)	Cu(3)–O(4) = 1.965(2)
Cu(3)–O(1) = 1.995(3)	Cu(3)–O(3) = 2.267(2)
Cu(2)–N(3) = 2.107(3)	Cu(2)–N(4) = 2.006(3)
Cu(3)–N(1) = 1.992(3)	Cu(2)–N(5) = 2.124(2)
Cu(3)–N(2) = 2.050(3)	Cu(3)^a–Cu(1)–Cu(2) = 168.20(2)
Cu(1)^a–O(1)–Cu(3) = 97.79(10)	Cu(1)–O(2)–Cu(2) = 99.83(10)
Cu(2)–O(3)–Cu(1) = 100.00(9)	Cu(2)–O(3)–Cu(3) = 114.61(11)
Cu(1)–O(3)–Cu(3) = 97.40(8)	Cu(1)^a–O(4)–Cu(3) = 98.75(10)
Cu(1)^a–O(4)–Cu(1) = 97.51(10)	Cu(3)–O(4)–Cu(1) = 96.48(8)

---

The structure of cluster 2 is composed of a discrete nonanuclear copper centers [Cu₉(L^II)₃(OH)₇]⁵⁺ and five ClO₄⁻ anions. Its cationic ORTEP drawing is shown in Fig. 2. Selected bond lengths and angles are tabulated in Table 2. As shown in Scheme 1, except for the aromatic moiety in the middle, the two ligands are completely the same. But their reaction with copper perchlorate led to two clusters with entirely different coordinating modes, nuclearity and spatial arrangements of the Cu(II) atoms, which suggests that the middle linker plays an essential role in the assembling of the clusters. It suggests that the stereo conformation and the availability of binding sits of the ligands are the dictating factors in assembling these clusters.

Fig. 2 Crystal structure of the cation of cluster 2, hydrogen atoms were omitted for clarity, and thermal ellipsoids are drawn at the 30% probability level (left); the core of cluster 2 represented by ball and stick model (right).

Table 2 Selected bond lengths (Å) and angles (°) for cluster 2

Cu(1)⋯Cu(2) = 3.1173(6)	Cu(2)⋯Cu(3) = 3.1695(5)
Cu(2)⋯Cu(8) = 2.9648(6)	Cu(2)⋯Cu(5) = 3.1702(6)
Cu(5)⋯Cu(9) = 2.9470(6)	Cu(3)⋯Cu(7) = 2.9684(6)
Cu(5)⋯Cu(4) = 3.0889(6)	Cu(3)⋯Cu(6) = 3.1024(6)
Cu(1)–O(10) = 1.886(2)	Cu(1)–O(3) = 1.918(2)
Cu(2)–O(8) = 1.905(2)	Cu(2)–O(3) = 1.956(2)
Cu(2)–O(4) = 1.965(2)	Cu(2)–O(9) = 2.055(2)
Cu(2)–O(10) = 2.417(2)	Cu(2)–O(2) = 2.530(4)
Cu(3)–O(1) = 1.952(2)	Cu(3)–O(5) = 1.894(7)
Cu(3)–O(6) = 2.053(4)	Cu(3)–O(4) = 1.973(2)
Cu(3)–O(3) = 2.4547(4)	Cu(4)–O(7) = 1.887(2)
Cu(4)–O(2) = 1.932(2)	Cu(3)–O(13) = 2.4751(4)
Cu(5)–O(2) = 1.932(2)	Cu(5)–O(12) = 1.903(2)
Cu(5)–O(11) = 2.108(2)	Cu(5)–O(4) = 1.996(2)
Cu(5)–O(1) = 2.4458(4)	Cu(5)–O(7) = 2.353(2)
Cu(8)–O(8) = 1.873(2)	Cu(8)–O(9) = 1.935(2)
Cu(9)–O(12) = 1.862(2)	Cu(9)–O(11) = 1.923(3)
Cu(6)–O(13) = 1.879(3)	Cu(6)–O(1) = 1.924(2)
Cu(7)–O(5) = 1.876(2)	Cu(7)–O(6) = 1.941(2)
Cu(1)–N(10) = 1.964(3)	Cu(1)–N(9) = 2.025(3)
Cu(4)–N(6) = 1.963(3)	Cu(4)–N(5) = 2.020(3)
Cu(6)–N(2) = 1.957(3)	Cu(6)–N(1) = 2.038(3)
Cu(7)–N(12) = 1.962(3)	Cu(7)–N(11) = 1.993(3)
Cu(8)–N(8) = 1.955(3)	Cu(8)–N(7) = 1.996(3)
Cu(9)–N(4) = 1.957(3)	Cu(9)–N(3) = 2.003(3)
Cu(6)–O(1)–Cu(3) = 106.34(11)	Cu(4)–O(2)–Cu(5) = 106.17(11)
Cu(1)–O(3)–Cu(2) = 107.13(9)	Cu(2)–O(4)–Cu(3) = 107.20(11)
Cu(2)–O(4)–Cu(5) = 106.34(10)	Cu(3)–O(4)–Cu(5) = 105.30(9)
Cu(7)–O(5)–Cu(3) = 103.88(12)	Cu(7)–O(6)–Cu(3) = 96.01(9)
Cu(4)–O(7)–Cu(5) = 92.88(10)	Cu(8)–O(8)–Cu(2) = 103.41(12)
Cu(8)–O(9)–Cu(2) = 95.91(10)	Cu(1)–O(10)–Cu(2) = 92.01(9)
Cu(9)–O(11)–Cu(5) = 93.86(10)	Cu(9)–O(12)–Cu(5) = 103.03(12)

---

In the cation, [Cu₉(L^II)₃(OH)₇]⁵⁺, a tricopper core is surrounded by two layers. The core is an incomplete cubane with one corner missing, constructed by three hexa-coordinated copper atoms and four μ₃-OH groups. The middle layer contains six tetra-coordinated copper centers which are connected to the core by three μ₂-OH groups and six phenolate oxygen atoms. Three of the six copper atoms establish contact with the core via direct bonding to three of the four μ₃-OH groups. The outermost layer is offered by the organic motif of the three ligands. In general, the structure possesses approximately a C₃ axis passed through the central O4 atom and perpendicular to the plane of Cu2–Cu3–Cu4. The nine copper centers fall into two categories in coordination, an octahedral geometry with an axial elongation due to Jahn–Teller effect (vide infra) for the inner core copper atoms and nearly perfect square plane for the other six copper atoms in the middle layer. The seven hydroxyl oxygen atoms in this structure can be divided into two types: four oxygen atoms located at the corners of the incomplete cubane are acted as a μ₃-bridge and the other three hydroxyl oxygen atoms acted as a μ₂-bridge. In the square geometry of six copper centers, the two Cu–N and two Cu–O bonds are in cis positions, respectively.

In both clusters, the significant hydrogen-bonding interactions between the cation and ClO₄⁻ anions were found, Fig. S2 and S3.† In the two clusters, the average distances between two neighboring copper atoms are about 3.0 Å (1) and 3.066 Å (2). Apparently, weak bonding interaction exists between the copper centers. In cluster 1, the Cu–O bonds vary in the range of 1.937(3)–2.317(2) Å, in which the axial Cu–O bond lengths are longer than the equatorial Cu–O bond lengths because of Jahn–Teller effect widely observed in Cu(II) complexes. The Jahn–Teller effect is also found in the octahedron geometry of the three copper atoms (Cu2, Cu3 and Cu5) in cluster 2, the average bond length of the axial Cu–O (2.4458 Å) is much longer than that of the equatorial one (about 1.95 Å).

As the clusters show catalytic activity towards the hydroxylation of benzene (vide infra), thus, it is important to know whether the clusters retain their integrities in solution. Although modern mass spectrometry may be an approach to clarify this issue, traditional technique such as conductivity measurement is proved to be simple and straightforward. The most significant merit of this approach is that no damage could be caused to a sample during the measurement. The correlation of the conductivity of a charged transition metal complex to the charge number of the complex is approximate due to the diversities in both metal ions and ligands. But it is very informative. To ensure the reliability of the measurement, we measured five complexes, clusters 1 and 2, and another three known complexes [Fe(L^I)]ClO₄, [Ni(HL^I)]BPh₄, [Cu(H₂L^I)](Cl)(PF₆).⁵⁰ The data fall essentially into corresponding categories for various cationic complexes, Table S2.†⁵¹ More importantly, the conductivities of the complexes correlate linearly to the charge number of the complexes, Fig. 3. The observed correlation suggests strongly that the clusters retain their integrities in solution.

Fig. 3 The correlation of the conductivity of charged transition metal complexes to the number of the positive charge of the complexes in MeCN at room temperature (please note that the conductivity of mono-charged complex comes from the average of the conductivities of the Fe(III) and Ni(II) complexes).

In supplementary to the conductivity measurement to support the claim of the integrity of the clusters in solution are ESI-MS (Fig. S4 and S5†) and ¹H NMR (Fig. S6 and S7†) analyses. In MeCN solution, both clusters showed molecular fragments associating to their parent clusters. For cluster 1, a peak at 881.8 is likely the evidence of the presence of the fragment of (1 – 2ClO₄⁻)²⁺ (m/z = 881.9) and its isotopic profile agrees well with that theoretically predicted, Fig. S4.† For cluster 2, the peak at 727.1 may be associated with the fragment of (2 + 4H⁺)⁴⁺ (m/z = 726.8). Its isotopic profile and predicted ones are shown in Fig. S5.† ¹H NMR spectra of the two clusters are shown in Fig. S6 and S7.† Although no signals for most of the proton of the organic moieties of the two clusters are observed due to the paramagnetic property of the clusters, the protons of the methyl groups in the organic skeleton do exhibit broad signals because those methyl groups are far away from the paramagnetic Cu(II) centers. For cluster 1, there are two signals at 1.79 and 1.07 ppm which are assigned to the methyl protons and suggest the existence of two types of methyl groups. This is in agreement with its structure in which the four methyl groups fall approximately into two types as schematically shown in Fig. S8.† In cluster 2, the nine Cu(II) ions consist roughly of a cyclic plane around which is surrounded by the three ligands, L^II2−. In the structure, each ligand adopts approximately the shape of an italic capital H, “H”, with the methylbenzene moiety as the horizontal bar of the italic H, Fig. S9.† In this conformation, the two vertical bars of the “H” are the symmetrical units of the ligand in antiparallel arrangement. Thus, all the methyl groups fall into three types. Both groups above and below the horizontal bar of the “H” are in rather similar chemical environment. Therefore, the two signals at around 1.4 ppm are tentatively assigned to these two types of methyl groups. The third type of methyl group is likely responsible to the signal at about 1.9 ppm.

Magnetic properties

The dc magnetic susceptibility measurements of clusters 1–2 were performed in the 2–300 K range under an applied field of 1000 Oe. The results are plotted as the χ_mT product vs. temperature, Fig. 4. In both clusters, the product of the molar magnetic susceptibility and temperature (χ_mT) decreases steadily upon cooling. At 300 K, cluster 1 exhibits a χ_mT value of 2.27 emu K mol⁻¹. The theoretical value is 2.5 emu K mol⁻¹ for six uncoupled Cu²⁺ ions (S = 1/2 and g = 2.15).^52,53 For cluster 2, the measured value (3.31 emu K mol⁻¹) at this temperature is also slightly smaller than the theoretical value (3.8 emu K mol⁻¹). The deviations indicate the presence of antiferromagnetic coupling.⁵⁴ The antiferromagnetic coupling is further supported by the rapid decrease of χ_mT from 10 K to 2 K for both clusters.

Fig. 4 Temperature dependence of χ_mT of clusters 1 (red) and 2 (black) under an applied dc field of 1000 Oe.

Catalytic oxidation of benzene into phenol

It has been reported that multicopper complexes catalyze the hydroxylation of alkane.¹⁷ Since clusters 1 and 2 have coordinatively unsaturated Cu centers, we are interested in whether they show any activity on the direct conversion of benzene into phenol. The catalytic performance of clusters 1 and 2 was assessed by following the procedure well established in literatures^55,56 in acetonitrile–H₂O medium using H₂O₂ as oxidant. All the results are shown in Table 3. The control results (entry 1, Table 3) confirmed that no phenol was obtained without the addition of either cluster. By using cluster 1 as a catalyst, we optimized the reaction conditions and found that the conditions of entry 4 (Table 3) are optimal. Thus the catalytic performance of cluster 2 was also examined under the conditions. As shown in Table 3, while the clusters are comparable in the yield (ca. 20%) to monocopper complexes [Tp^xCu(NCMe)] (yield 25%),⁵⁷ [Cu(Phen)Cl₂] (yield 22%, TON = 242),⁵⁸ the TON values could be nearly 4-fold higher. This can probably be attributed to the high nuclearity of the clusters and the retention of their integrity in solution. Under the same conditions, the nonacopper cluster 2 exhibited better performance by over 50% in both TON and TOF compared with the hexacopper cluster 1.

Table 3 Catalytic oxidation of benzene to phenol by H₂O₂^a

Entry	Catalysts (μmol)	T (°C)	Reaction time (h)	H2O2 (mmol)	Yield (%)	TON	TOF (h^−1)
a The substrate (benzene) is 10 mmol.b HCl (aq., 0.56 μmol) was added.
1	Non	80	1	20	0	0	0
2	1 (3.3)	80	1	20	12.9	390	390
3	1 (3.3)	80	2	20	16.2	491	246
4	1 (3.3)	80	3	20	18.6	564	188
5	1 (3.3)	80	4	20	17.7	536	134
6	1 (3.3)	60	3	20	10.8	327	109
7	1 (3.3)	80	3	10	8.1	245	82
8	1 (1.7)	80	3	20	9.1	535	178
9	1 (3.3)	80	3	30	9.3	282	94
10	1 (5.0)	80	3	20	14.1	282	94
11^b	1 (3.3)	80	3	20	14.5	439	146
12	2 (2.2)	80	3	20	19.9	905	302
13	Cu(ClO4)2(20)	80	3	20	5.2	26	8.7

---

It has been reported that the addition of acid can improve the catalytic activity of multicopper catalyst in the oxidative transformation of alkanes.⁵⁹ It is believed that the added acid can protonate some of the coordinated ligand(s) to generate new catalytic species. But in our case, the addition of hydrochloric (17% of the catalyst) led to decrease in performance by about 20% (entry 11). This is certainly ascribed to the destruction of the catalyst, at least partially, by protonating the basic groups bound to the metal centers, the pyridinyl, phenolate and the μ-hydroxyl bridge. The resulted species would show poor catalytic activity like the simple copper salt in the benzene oxidation (entry 13). This is, on the other hand, to confirm that the clusters stand as a whole to catalyze the reaction. This implies that only those copper centers at the surface of the clusters which are accessible for the oxidant-binding may be involved in the catalysis. This may explain that cluster 2 shows similar catalytic performance to cluster 1 despite its higher nuclearity. When using H₂O₂ as an oxidant in a reaction catalyzed by a transition metal complex, the active species is hydroxyl radical generated by the coordination of the oxidant to the metal center. Such a radical mechanism was confirmed in our recent analogous work in which hydroxyl radical quencher did decrease the reaction yield.⁶⁰ Thus, it is possible that the conversion reaction adopts the same mechanism. But without further investigations, involvement of high valent copper-oxo in the catalysis can not be ruled out as one of the possibilities.

Conclusions

In this work, we reported two copper clusters (1 and 2). The highly charged clusters show good solubility in most organic solvents due to their hydrophobic shields offered by the ligands H₂L^I or H₂L^II, respectively. Conductivity data, ESI-MS and ¹H NMR analyses indicate that these clusters retain their integrities in solution. The clusters catalyze direct hydroxylation of benzene into phenol by hydrogen peroxide. Although the reaction yield is not superior to those reported in literatures, both TON and TOF can be much higher. Our results suggest that in the catalysis, the clusters act as a whole to perform the catalysis. In the catalysis, radical mechanism may be involved although involvement of high valent copper-oxo species can not be ruled out. More detailed mechanistic investigation needs to be pursued and we are actively working on it. Despite the fact that the work reported here is preliminary, this study suggests that copper clusters may be of great potential in the catalytic oxidation of benzene to produce phenol which is one of the most important chemicals in industry.

Acknowledgements

The authors thank NSF of China (Grant no. 20871064, 21171073), the Government of Zhejiang Province (Qianjiang Professorship, XL) and the Education Department of Jiangxi Province (Grant no. JXJG-14-10-6) for financial support. Professor Minghua Zeng at Guangxi Normal University is greatly acknowledged for the collection of magnetic data.

Notes and references

1. J. Lu, J.-X. Lin, M.-N. Cao and R. Cao, Coord. Chem. Rev., 2013, 257, 1334–1356.
2. C. Romanescu, T. R. Galeev, W.-L. Li, A. I. Boldyrev and L.-S. Wang, Acc. Chem. Res., 2013, 46, 350–358.
3. T. Punniyamurthy, S. Velusamy and J. Iqbal, Chem. Rev., 2005, 105, 2329–2363.
4. Z.-H. Wei, C.-Y. Ni, H.-X. Li, Z.-G. Ren, Z.-R. Sun and J.-P. Lang, Chem. Commun., 2013, 49, 4836–4838.
5. N. Uchida, T. Miyazaki, Y. Matsushita, K. Sameshima and T. Kanayama, Thin Solid Films, 2011, 519, 8456–8460.
6. P. Nagababu, S. Maji, M. P. Kumar, P. P. Y. Chen, S. S. F. Yu and S. I. Chan, Adv. Synth. Catal., 2012, 354, 3275–3282.
7. T. Punniyamurthy and L. Rout, Coord. Chem. Rev., 2008, 252, 134–154.
8. S. Perruchas, N. Desboeufs, S. Maron, X. F. Le Goff, A. Fargues, A. Garcia, T. Gacoin and J.-P. Boilot, Inorg. Chem., 2012, 51, 794–798.
9. S. Loew, J. Becker, C. Wuertele, A. Miska, C. Kleeberg, U. Behrens, O. Walter and S. Schindler, Chem.–Eur. J., 2013, 19, 5342–5351.
10. L. Wang, B. Guo, H. X. Li, Q. Li, H. Y. Li and J. P. Lang, Dalton Trans., 2013, 15570–15580.
11. Q. Liu, P. Wu, Y. H. Yang, Z. Q. Zeng, J. Liu, H. Yi and A. W. Lei, Angew. Chem., Int. Ed., 2012, 51, 4666–4670.
12. E. I. Solomon, R. Sarangi, J. S. Woertink, A. J. Augustine, J. Yoon and S. Ghosh, Acc. Chem. Res., 2007, 40, 581–591.
13. S. I. Chan and S. S. F. Yu, Acc. Chem. Res., 2008, 41, 969–979.
14. R. Balasubramanian, S. M. Smith, S. Rawat, L. A. Yatsunyk, T. L. Stemmler and A. C. Rosenzweig, Nature, 2010, 465, 115–U131.
15. R. A. Himes, K. Barnese and K. D. Karlin, Angew. Chem., Int. Ed., 2010, 49, 6714–6716.
16. M. Merkx, D. A. Kopp, M. H. Sazinsky, J. L. Blazyk, J. Muller and S. J. Lippard, Angew. Chem., Int. Ed., 2001, 40, 2782–2807.
17. A. M. Kirillov, M. V. Kirillova and A. J. L. Pombeiro, Coord. Chem. Rev., 2012, 256, 2741–2759.
18. A. M. Kirillov, M. V. Kirillova, L. S. Shul'pina, P. J. Figiel, K. R. Gruenwald, M. F. C. Guedes da Silva, M. Haukka, A. J. L. Pombeiro and G. B. Shul'pin, J. Mol. Catal. A: Chem., 2011, 350, 26–34.
19. J. M. Hoover, B. L. Ryland and S. S. Stahl, ACS Catal., 2013, 3, 2599–2605.
20. A. Sabbatini, L. M. D. R. S. Martins, K. T. Mahmudov, M. N. Kopylovich, M. G. B. Drew, C. Pettinari and A. J. L. Pombeiro, Catal. Commun., 2014, 48, 69–72.
21. M. S. Dronova, A. N. Bilyachenko, A. I. Yalymov, Y. N. Kozlov, L. S. Shul'pina, A. A. Korlyukov, D. E. Arkhipov, M. M. Levitsky, E. S. Shubina and G. B. Shul'pin, Dalton Trans., 2014, 872–882.
22. S. Zhu, Z. Qiu, T. Ni, X. Zhao, S. Yan, F. Xing, Y. Zhao, Y. Bai and M. Li, Dalton Trans., 2013, 10898–10911.
23. M. M. Hossain, W.-K. Huang, H.-J. Chen, P.-H. Wang and S.-G. Shyu, Green Chem., 2014, 16, 3013–3017.
24. M. S. Eberhart, J. R. Norton, A. Zuzek, W. Sattler and S. Ruccolo, J. Am. Chem. Soc., 2013, 135, 17262–17265.
25. K. V. N. Esguerra, Y. Fall, L. Petitjean and J.-P. Lumb, J. Am. Chem. Soc., 2014, 136, 7662–7668.
26. M. R. Halvagar, P. V. Solntsev, H. Lim, B. Hedman, K. O. Hodgson, E. I. Solomon, C. J. Cramer and W. B. Tolman, J. Am. Chem. Soc., 2014, 136, 7269–7272.
27. J. Y. Lee, R. L. Peterson, K. Ohkubo, I. Garcia-Bosch, R. A. Himes, J. Woertink, C. D. Moore, E. I. Solomon, S. Fukuzumi and K. D. Karlin, J. Am. Chem. Soc., 2014, 136, 9925–9937.
28. B. L. Ryland, S. D. McCann, T. C. Brunold and S. S. Stahl, J. Am. Chem. Soc., 2014, 136, 12166–12173.
29. N. Q. Shixallyev, A. V. Gurbanov, A. M. Maharramov, K. T. Mahmudov, M. N. Kopylovich, L. M. D. R. S. Martins, V. M. Muzalevskiy, V. G. Nenajdenko and A. J. L. Pombeiro, New J. Chem., 2014, 38, 4807–4815.
30. S. Biswas, A. Dutta, M. Dolai, M. Debnath, A. D. Jana and M. Ali, RSC Adv., 2014, 4, 34248–34256.
31. L. Ma, Q. Zhang, L. Cheng, Z. Wu and J. Yang, RSC Adv., 2014, 4, 30558–30565.
32. V. Singh, P. C. Mondal, M. Chhatwal, Y. L. Jeyachandran and M. Zharnikov, RSC Adv., 2014, 4, 23168–23176.
33. H. S. Abbo and S. J. J. Titinchi, Appl. Catal., A, 2012, 435, 148–155.
34. M. C. Esmelindro, E. G. Oestreicher, M. Caovilla, J. A. Lessa, C. Fernandes, C. Dariva, S. M. Egues, A. J. Bortoluzzi and O. A. C. Antunes, J. Braz. Chem. Soc., 2006, 17, 1551–1557.
35. Y. Shibata, R. Hamada, T. Ueda, Y. Ichihashi, S. Nishiyama and S. Tsuruya, Ind. Eng. Chem. Res., 2005, 44, 8765–8772.
36. N. Herron and C. A. Tolman, J. Am. Chem. Soc., 1987, 109, 2837–2839.
37. H. Mimoun, L. Saussine, E. Daire, M. Postel, J. Fischer and R. Weiss, J. Am. Chem. Soc., 1983, 105, 3101–3110.
38. K. Ohkubo, A. Fujimoto and S. Fukuzumi, J. Am. Chem. Soc., 2013, 135, 5368–5371.
39. K. Ohkubo, T. Kobayashi and S. Fukuzumi, Angew. Chem., Int. Ed., 2011, 50, 8652–8655.
40. S. Khanra, T. Weyhermueller and P. Chaudhuri, Dalton Trans., 2009, 3847–3853.
41. Y.-Q. Zheng, D.-Y. Cheng, B.-B. Liu and W.-X. Huang, Dalton Trans., 2011, 277–286.
42. X. Li, D. Cheng, J. Lin, Z. Li and Y. Zheng, Cryst. Growth Des., 2008, 8, 2853–2861.
43. C. X. Zhan, X. F. Wang, Z. H. Wei, D. J. Evans, X. A. Ru, X. H. Zeng and X. M. Liu, Dalton Trans., 2010, 11255–11262.
44. B.-L. Lee, M. D. Karkas, E. V. Johnston, A. K. Inge, L.-H. Tran, Y. Xu, O. Hansson, X. Zou and B. Akermark, Eur. J. Inorg. Chem., 2010, 5462–5470.
45. S. Bhatt and S. K. Nayak, Tetrahedron Lett., 2009, 50, 5823–5826.
46. G. N. Ledesma and S. R. Signorella, Tetrahedron Lett., 2012, 53, 5699–5702.
47. G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008, 64, 112–122.
48. J. Martinez-Sanchez, R. Bastida, A. Macias, H. Adams, D. E. Fenton, P. Perez-Lourido and L. Valencia, Polyhedron, 2010, 29, 2651–2656.
49. V. Pashchenko, B. Brendel, B. Wolf, M. Lang, K. Lyssenko, O. Shchegolikhina, Y. Molodtsova, L. Zherlitsyna, N. Auner, F. Schutz, M. Kollar, P. Kopietz and N. Harrison, Eur. J. Inorg. Chem., 2005, 4617–4625.
50. X.-L. You, Z.-H. Wei, B.-B. Xu and X.-M. Liu, Polyhedron, 2014, 81, 743–748.
51. W. J. Geary, Coord. Chem. Rev., 1971, 7, 81–122.
52. F. A. Mautner, M. S. El Fallah, S. Speed and R. Vicente, Dalton Trans., 2010, 4070–4079.
53. J. W. Zhao, D. Y. Shi, L. J. Chen, X. M. Cai, Z. Q. Wang, P. T. Ma, J. P. Wang and J. Y. Niu, CrystEngComm, 2012, 14, 2797–2806.
54. M. Murugesu, P. King, R. Clerac, C. E. Anson and A. K. Powell, Chem. Commun., 2004, 740–741.
55. R. R. Fernandes, M. V. Kirillova, J. A. L. da Silva, J. J. R. Frausto da Silva and A. J. L. Pombeiro, Appl. Catal., A, 2009, 353, 107–112.
56. D. Bianchi, M. Bertoli, R. Tassinari, M. Ricci and R. Vignola, J. Mol. Catal. A: Chem., 2003, 200, 111–116.
57. A. Conde, M. Mar Diaz-Requejo and P. J. Perez, Chem. Commun., 2011, 47, 8154–8156.
58. C. Detoni, N. M. F. Carvalho, R. O. M. A. de Souza, D. A. G. Aranda and O. A. C. Antunes, Catal. Lett., 2009, 129, 79–84.
59. M. V. Kirillova, Y. N. Kozlov, L. S. Shul'pina, O. Y. Lyakin, A. M. Kirillov, E. P. Talsi, A. J. L. Pombeiro and G. B. Shul'pin, J. Catal., 2009, 268, 26–38.
60. B.-B. Xu, W. Zhong, Z.-H. Wei, H.-L. Wang, J. Liu, L. Wu, Y.-G. Feng and X.-M. Liu, Dalton Trans., 2014, 15337–15345.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 960633 and 960634. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c4ra12832j

---