N₃-ligated Nickel(II) Diketonate Complexes: Synthesis, Characterization and Evaluation of O₂ Reactivity

Abstract

Interest in O₂-dependent aliphatic carbon-carbon (C-C) bond cleavage reactions of first row divalent metal diketonate complexes stems from the desire to further understand the reaction pathways of enzymes such as DKE1 and to extract information to develop applications in organic synthesis. A recent report of O₂-dependent aliphatic C-C bond cleavage at ambient temperature in Ni(II) diketonate complexes supported by a tridentate nitrogen donor ligand [(MBBP)Ni(PhC(O)CHC(O)Ph)]Cl (7-Cl; MBBP = 2,6-bis(1-methylbenzoimidazol-2-yl)pyridine) in the presence of NEt₃ spurred our interest in further examining the chemistry of such complexes. A series of new TERPY-ligated Ni(II) diketonate complexes of the general formula [(TERPY)Ni(R₂-1,3-diketonate)]ClO₄ (1: R = CH₃; 2: R = C(CH₃)₃; 3: R = Ph) was prepared under air and characterized using single crystal X-ray crystallography, elemental analysis, ¹H NMR, ESI-MS, FTIR, and UV-vis. Analysis of the reaction mixtures in which these complexes were generated using ¹H NMR and ESI-MS revealed the presence of both the desired diketonate complex and the bis-TERPY derivative [(TERPY)₂Ni](ClO₄)₂ (4). Through selective crystallization 1-3 were isolated in analytically pure form. Analysis of reaction mixtures leading to the formation of the MBBP analogs [(MBBP)Ni(R₂-diketonate)]X (X = ClO₄: 5: R = -CH₃; 6: R = - C(CH₃)₃; 7-ClO₄: R = Ph; X = Cl: 7-Cl: R = Ph) using ¹H NMR and ESI-MS revealed the presence of [(MBBP)₂Ni](ClO₄)₂ (8). Analysis of aerobic acetonitrile solutions of analytically pure 1-3, 5 and 6 containing NEt₃ and in some cases H₂O using ¹H NMR and UV-vis revealed evidence for the formation of additional bis-ligand complexes (4 and 8) but suggested no oxidative diketonate cleavage reactivity. Analysis of the organic products generated from 3, 7-ClO₄ and 7-Cl revealed unaltered dibenzoylmethane. Our results therefore indicate that N₃-ligated Ni(II) complexes of unsubstituted diketonate ligands do not exhibit O₂-dependent aliphatic C-C bond clevage at room temperature, including in the presence of NEt₃ and/or H₂O.

Graphical Abstract

N₃-ligated Ni(II) diketonate complexes are air stable and do not exhibit oxidative cleavage within the diketonate ligand under aerobic conditions.

Introduction

Oxidative aliphatic carbon-carbon (C-C) bond cleavage reactions mediated by first row metals using O₂ as the terminal oxidant are of significant current interest, particularly for applications in organic synthesis.^1–7 To date, examination of the mechanistic pathways of these reactions has received minimal attention. This is due in part to the limited number of systems wherein a well-characterized metal/substrate type structure can be evaluated in terms of its O₂ reactivity. A few laboratories have developed model systems for enzymes that mediate O₂-dependent aliphatic C-C bond cleavage. DKE1 is an enzyme found in Acinetobacter johnsonii which catalyzes the oxidative cleavage of diketonate substrates, including acetylacetone, using O₂ as the terminal oxidant.⁸ The active site of DKE1 contains a non-heme Fe(II) center ligated in a facial array by three histidine donors (Scheme 1(a)).^9–11 Siewart et al. used a trispyrazolylborato (Tp) ligand to mimic the 3His facial triad in a Fe(II) diethylphenylmalonate complex (Scheme 1(b)).¹² This complex undergoes O₂-dependent aliphatic C-C cleavage within the diketonate unit to give ethyl carbonate (which decomposes to give CO₂ and ethoxide anion) and ethyl benzoylformate as products of a dioxygenase-type reaction. Park et al. reported a similar set of complexes using two different ^R2Tp ligands (R = -Me and -Ph; Scheme 1(c)) and a variety of diketonate ligands.¹³ Their studies showed that O₂-dependent cleavage only occurred for ^Ph2Tp complexes containing a diethyl malonate or diethylphenylmalonate diketonate ligand. Complexes that didn’t undergo oxidative diketonate cleavage exhibited either no reaction or formed a green peroxo-bridged intermediate that decayed without any oxidative C-C bond cleavage. Recently, Banerjee and Paine reported that an Fe(II) dibenzoylmethane complex supported by the monoanionic tris(2-pyridylthio)methanido ligand undergoes reaction with O₂ at ambient temperature to give a small amount of benzoic acid (11–13%).¹⁴ The major product of this reaction is the Fe(III) diketonate complex.

Scheme 1.

a) Reaction catalyzed and active site iron center of DKE1. b) Diethylphenylmalonate-containing model complex for the active site of DKE1. c) Complexes with diketonates e-f undergo oxidative C-C bond cleavage whereas those with diketonates a-d do not exhibit this reactivity.

To examine how a change in the metal center affects O₂ activation, Hoof et al. studied the O₂ reactivity of a ^Me2Tp-ligated Ni(II) diethylphenylmalonate complex.¹⁵ Their studies revealed O₂-dependent reactivity leading to hydroxylation and aliphatic C-C bond cleavage products (Scheme 2(a)). This reaction required an open coordination position at the Ni(II) center as using the more sterically hindered ^Ph2Tp supporting ligand system did not produce the same type of reactivity. A hydroxylated diphenyl diketonate ligand in a model system for Ni(II)-containing acireductone dioxygenase (Scheme 2(b)) undergoes two-electron oxidation upon reaction with O₂ to give a triketone and hydroperoxide anion as intermediates.^16,17 Subsequent reaction between these intermediates in the presence of the remaining Ni(II) complex gives benzoate/benzoic acid and CO as products of the aliphatic C-C bond cleavage reaction. An Fe(II) analog produces similar products as well as an α-keto acid product, which results from iron-promoted hydration of the triketone intermediate followed by C-C cleavage.¹⁸ Recently, Raje, et al. reported similar results using a Ni(II) complex supported by a tetradentate pyridinophane (L₄) ligand.¹⁹ This complex was treated with the hydroxylated dibenzoylmethane (PhC(O)C(OH)C(O)Ph) in the presence of base in CH₃CN. Exposure of this solution to O₂ followed by ESI-MS analysis revealed the presence of the [L₄Ni(OC(O)Ph)]⁺ cation. Organic recovery experiments also revealed the formation of benzoic acid. These results mirror those previously reported by Szajna et al.¹⁶

Scheme 2.

a) Reaction of a ^Me2TpNi(II) diethylphenyl-malonate complex with O₂. b) Reactivity of Ni(II) hydroxydiketonate complex that mimics the reactivity of Ni(II)-containing acireductone dioxygenase (ARD). c) UV-light induced aliphatic C-C bond cleavage of methoxy-substituted chlorodiketonate complex in the presence of O₂.

Notably, a structurally similar Ni(II) dibenzoylmethane complex that does not contain the hydroxyl substituent on the central carbon of the diketonate ligand is air stable.²⁰ However, a Ni(II) chlorodiketonate complex having methoxy-substituted phenyl appendages (Scheme 2(c)) undergoes O₂-dependent aliphatic C-C bond cleavage reactivity within the diketonate moiety to give a carboxylate product upon illumination with UV light.²¹ This reactivity is similar to thermal oxidative C-C bond cleavage identified for structurally-similar copper chlorodiketonate complexes.^22,23

Based on the literature precedent described above, which shows that only electron rich diketonates, or those containing a central carbon substituent that can undergo reaction or act as a leaving group, exhibit significant O₂ reactivity, a recent publication drew our attention.²⁴ Ramasubramanian, et al. reported that Ni(II) diketonate complexes (Scheme 3(a)) supported by the tridentate 2,6-bis(1-methylbenzoimidazol-2-yl)pyridine (MBBP) ligand undergo O₂-dependent oxidative cleavage in the presence of one equivalent of triethylamine (NEt₃) and in the absence or presence of water (10% in CH₃CN) (Scheme 3(b)) to give carboxylic acid products (52%, R = Ph) along with CO₂. These reactions are reported to occur under mild conditions, as a 6.00×10⁻⁴ M solution of each complex undergoes complete reaction in <1 h at 25 °C as determined by UV-vis. ¹⁸O₂ studies showed 43% isotope incorporation in the benzoic acid product. A reaction pathway was proposed for the R = Ph derivative (Scheme 3(b)) wherein the diketonate acts as a two-electron reductant to O₂, leading to the formation of an organoperoxo species that is stabilized via interaction with Ni(II). Intramolecular attack of the terminal peroxo oxygen at a diketonate carbonyl moiety followed by aliphatic C-C bond cleavage gives the benzoic acid product and phenylgloyoxal. Another report in the literature suggested that O₂-dependent oxidative cleavage of an acetylacetonate ligand can occur in Ni(acac)₂(H₂O)₂ the presence of base (ethylenediamine) (Scheme 3(c)).²⁵ However, this reaction was reported to proceed very slowly at room temperature (18 d under air).

Scheme 3.

a) MBBP-ligated Ni(II) diketonate complexes reported by Ramasubramamanin, et al.²⁴ b) Proposed pathway of O₂ reaction of MBBP-ligated Ni(II) phenyl diketonate complex. c) Oxidative cleavage of an aceytylacetonate ligand.²⁵

The reactions outlined in Scheme 3 are interesting in that aliphatic C-C cleavage is reported to occur within an unsubstituted diketonate under mild conditions, which is rare to date in synthetic systems. Based on these results we became interested in further examining such systems. Herein we report synthetic, characterization and reactivity studies of a series of new TERPY-ligated Ni(II) diketonate complexes, [(TERPY)Ni(diketonate)]ClO₄ (1-3). While such complexes could be isolated by selective crystallization, the reaction mixtures leading to their formation always also contained [(TERPY)₂Ni](ClO₄)₂ (4). Unlike the MBBP Ni(II) diketonate complexes reported by Ramasubramamanin, et al.²⁴, 1-3 are stable with respect to O₂ over 24 h, showing no aliphatic C-C bond cleavage reactivity upon exposure to air, even in the presence of base and water. Based on these results, we re-examined the chemistry of the MBBP-ligated diketonate complexes. Using ¹H NMR under paramagnetic conditions we found that the formation bis-MBBP Ni(II) derivatives is prevalent under the reaction conditions.^24,26 Ramasubramanian et al. briefly mention these complexes, but did not report the chemistry associated with such species. Using an organic recovery experiment, we found no evidence for oxidative aliphatic C-C bond cleavage upon exposure of [(MBBP)Ni(PhC(O)CHC(O)Ph)]X (7-X; X = Cl or ClO₄) to O₂ in the presence of NEt₃ or NEt₃/H₂O after 24 h.

Experimental

General Methods

The following chemicals were purchased and used as received: 1,3-diphenylpropane-1,3-dione (98%, Sigma-Aldrich), 2,4-pentanedione (99%, Acros Organics), 2,2,6,6-tetramethyl-3,5-heptanedione (97%, Combi-Blocks) and 2,2’;6’,2”-terpyridine (98%, Sigma-Aldrich). Solvents used in air were used without further purification unless otherwise stated. Solvents used in the glovebox were dried following a previously published procedure prior to use.²⁷ The CD₃CN used in ¹H NMR experiments was dried using molecular sieves. 2,6-bis(2’-benzimidazolyl)pyridine (BBP) was prepared following the literature procedure.²⁸ Manipulations performed under an inert N₂ atmosphere were conducted in a MBraun Unilab glovebox.

Physical Methods

¹H NMR spectra were collected on a Brüker Advance III HD Ascend-500 spectrometer. Chemical shifts (ppm) are reported relative to the residual solvent peak in CD₂HCN (1.94 ppm, quintet) or CHCl₃ (7.26 ppm, singlet). ¹H NMR spectra for high-spin paramagnetic Ni(II) complexes were performed using previously published parameters: 32k data points, 300 scans, a 90° pulse (9.75 μs) at 298 K, and a 250 ms relaxation delay.²⁶ During processing, an exponential weighting function (lb = 7 Hz) and a manual baseline correction were performed using MNova (Version 12.0.4). FTIR spectra were collected as KBr pellets using a Shimadzu FTIR-8400 spectrometer. UV-vis data was collected using a Hewlett-Packard 8453A diode array spectrometer at ambient temperature. ESI mass spectral data was collected using a Shimadzu LCMS-2020. Elemental analyses were performed by Robertson Microlit Laboratories (Ledgewood, NJ) or Atlantic Microlab (Norcross, GA).

Caution! Perchlorate compounds containing organic ligands are potentially explosive. These materials should be handled with care and in small quantities.²⁹

Synthesis and Characterization.

Unless stated otherwise all synthetic reactions were performed in air.

[(TERPY)Ni(CH₃C(O)CHC(O)CH₃)(CH₃CN)]ClO₄ (1).

Ni(ClO₄)₂·6H₂O (49.4 mg 0.135 mmol) was dissolved in CH₃CN (2 mL) under air. An excess of 2,4-pentanedione (~1 mL) and one equivalent of triethylamine (19 μL, 0.135 mmol) were subsequently added. The mixture was then stirred for approximately 30 min during which time it became blue. This solution was added to a vial containing TERPY (31.6 mg, 0.135 mmol) dissolved in CH₃CN (2 mL). The resulting solution became faint yellow. After stirring at ambient temperature for 2 h, Et₂O (8 mL) was added and the solution was placed in the freezer (−15 °C) for 12 h. This produced grayish-green X-ray quality crystals. We note that the ratio of Et₂O to CH₃CN (~1.4:1) is important to selectively crystallize 1. The solution was decanted from the crystals, which were then washed with Et₂O and dried under vacuum (49 mg, 74% yield). ESI-MS: m/z calc. for C₂₀H₁₈N₃NiO₂·CH₃CN, 431.1 [M-ClO₄]⁺; found 431.0; calc. 390.1 [M-CH₃CN-ClO₄]⁺; found 390.0. Anal. Calc. for C₂₀H₁₈ClN₃NiO₆·CH₃CN: C, 49.71; H, 3.98; N, 10.54. Found: C, 49.34; H, 3.92; N, 10.35. UV-Vis λ_max, nm (ε, M⁻¹ cm⁻¹): 233 (23532), 272 (17307), 279 (19884), 305 (18457), 319 (17046), 329 (13074). FT-IR (KBr, cm⁻¹): 3420 (ν_O-H), 3100 (ν_C-H), 2320 (ν_C≡N), 2020 (ν_C-H), 1600 (ν_C=C), 1520 (ν_C=N), 1450 (ν_C=O), 1400 (ν_C=O), 1090 (ν_ClO4), 780(_νC-H), 620(ν_ClO4).

[(TERPY)Ni((CH₃)₃CC(O)CHC(O)C(CH₃)₃)(H₂O)]ClO₄ (2).

Ni(ClO₄)₂·6H₂O (49.4 mg 0.135 mmol) was dissolved in CH₃CN (2 mL) under air. An excess of 2,2,6,6-tetramethyl-3,5-heptanedione (141 μL, 0.675 mmol) was added, along with triethylamine (19 μL, 0.135 mmol). The resulting solution was stirred for 30 min, during which time it became blue. This solution was then added to a vial containing TERPY (31.6 mg, 0.135 mmol) dissolved in CH₃CN (2 mL). The resulting yellow-green mixture was stirred for 2 h. Et₂O (8 mL) was added and the solution was placed in the freezer (−15 °C) overnight which resulted in the deposition of a fine tan precipitate (4 by ¹H NMR; vida infra). The yellow-green solution was brought to dryness under reduced pressure. Crystalline material was obtained by dissolving the precipitate in CH₂Cl₂, adding a few drops of CH₃CN, followed by slow diffusion of Et₂O (41 mg, 53% yield). ESI-MS: m/z calc for C₂₆H₃₀N₃NiO₂·CH₃CN, 515.2 [M-ClO₄]⁺; found 515.2; calc. 474.2 [M-CH₃CN-ClO₄]⁺; found 474.1. Anal. Calc. for C₂₆H₃₀ClN₃NiO₆·H₂O·0.9 CH₂Cl₂: C, 48.28; H, 5.09; N, 6.28. Found: C, 48.48; H, 4.89; N, 6.54. UV-Vis λ_max, nm (ε, M⁻¹ cm⁻¹): 233 (27182), 270 (18653), 278 (22176), 306 (21052), 316 (23235), 328 (14948). FT-IR (KBr, cm⁻¹): 3530 (ν_O-H), 3430 (ν_O-H), 2960 (ν_C-H), 2020 (ν_C-H), 1590 (ν_C=C), 1410 (ν_C=O), 1110 (ν_ClO4), 780(_νC-H), 630(ν_ClO4).

[(TERPY)Ni(Ph(O)CHC(O)Ph)(H₂O)]ClO₄ (3).

Ni(ClO₄)₂·6H₂O (49.4 mg 0.135 mmol) was dissolved in CH₃CN (2 mL) under air. Dibenzoylmethane (30.1 mg, 0.135 mmol) was added to the solution along with triethylamine (19 μL, 0.135 mmol). This mixture was stirred for 30 minutes and then added to a vial containing TERPY (31.6, 0.135 mmol) dissolved in CH₃CN (2 mL). The resulting solution was stirred for 2 h during which time it became green. Excess Et₂O (8 mL) was added and the solution was placed in the freezer (−15 °C) overnight. The yellow-green solution was decanted away from a tan solid (4 by ¹H NMR; vida infra) and was brought to dryness under reduced pressure. The resulting thick green oil was dissolved in CH₂Cl₂ (4 mL) and the solution was passed through a celite plug. Slow diffusion of Et₂O into the CH₂Cl₂ filtrate produced X-ray diffraction quality crystals (35 mg, 42% yield). ESI-MS: m/z calc. for C₃₀H₂₂N₃NiO₂·CH₃CN, 555.1 [M-ClO₄]⁺; found 555.2; calc. 514.1 [M-CH₃CN-ClO₄]⁺; found 514.1. Anal. Calc. for C₃₀H₂₂ClN₃NiO₆·H₂O: C, 56.95; H, 3.82; N, 6.64. Found: C, 57.08; H, 3.75; N, 6.70. UV-Vis λ_max, nm (ε, M⁻¹ cm⁻¹): 236 (20752), 253 (17386), 270 (17553), 278 (18609), 317 (12050), 329 (13187), 364 (10790) (d-d transitions). FT-IR (KBr, cm⁻¹): 3440 (ν_O-H), 3160 (ν_C-H), 2010 (ν_C-H), 1600 (ν_C=C), 1510 (ν_C=N), 1450 (ν_C=O), 1410 (ν_C=O), 1090 (ν_ClO4), 780(_νC-H), 630(ν_ClO4).

[(TERPY)₂Ni](ClO₄)₂ (4).

This compound has been previously reported but a new synthesis is shown here.^30,31 A mixture of Ni(ClO₄)₂·6H₂O (24.7 mg 0.0675 mmol) and TERPY (31.5 mg, 0.135 mmol) in CH₃CN (5 mL) was stirred for 12 h under air at ambient temperature. Slow diffusion of Et₂O into the CH₃CN solution produced orange-brown plates. (31 mg, 63% yield) ESI-MS: m/z calc. for [C₃₀H₂₂N₆Ni·ClO₄]⁺, 623.1; found, 623.1; calc. 262.1 [M-(2·ClO₄)]²⁺; found 262.1. Anal. Calc. for C₃₀H₂₂Cl₂N₆NiO₈·H₂O: C, 48.55; H, 3.26; N, 11.32. Found: C, 48.38; H, 2.95; N, 11.30. UV-Vis λ_max, nm (ε, M⁻¹ cm⁻¹): 235 (31872), 269 (29970), 310 (18235), 320 (27374), 334 (25642). FT-IR (KBr, cm⁻¹): 3610 (ν_O-H), 3410 (ν_O-H), 3120 (ν_C-H), 2360 (ν_C-N), 2010 (ν_C-H), 1610 (ν_C=C), 1450 (ν_C=O), 1090 (ν_ClO4), 770(_νC-H), 620(ν_ClO4).

Analysis of crystallization filtrates by ¹H NMR.

Crystallization of 1-3 produced materials with purity as determined by elemental analysis. Evaluation of the remaining filtrates by ¹H NMR revealed the presence of 4 in each reaction mixture.

Addition of TERPY (1 eq) to 1 in CD₃CN.

Complex 1 (4.30×10⁻³ mmol) was dissolved in CD₃CN (600 μL), and a ¹H NMR spectrum was collected under paramagnetic conditions. TERPY (4.30×10⁻³ mmol) was added, the solution was briefly mixed and another ¹H NMR spectrum was collected. Additional spectra were collected after 3 h and 24 h. The NMR spectra indicated significant formation of 4.

Probing for aliphatic C-C bond cleavage reactivity involving 3.

Complex 3 (10 mg, 0.02 mmol) was dissolved in CH₃CN (2.7 ml) in a glass scintillation vial with a stir bar. Following the addition of H₂O (300 μL) and NEt₃ (279 μL, 2.00 mmol) and purging with O₂ with for 1 min, the mixture was stirred for 24 h. The solvent was then removed under reduced pressure. CH₂Cl₂ (8 ml) and HCl (8 ml, 1 M) were added, followed by stirring for 1 h and subsequently the organic layer was separated. The organic layer was dried over Na₂SO₄ and brought to dryness under vacuum to give a white solid (3.2 mg, 71% based on dibenzoylmethane). The ¹H NMR (CDCl₃) spectrum of the solid showed only dibenzoylmethane and no aliphatic C-C bond cleavage products.

In a separate experiment, a similar reaction was performed with the only difference being 3 (10 mg, 0.02 mmol) was dissolved in dry CH₃CN (3 mL) and one equivalent of NEt₃ (~3 μL, 0.02 mmol) was added. An aliquot was removed before O₂ purge and after stirring for an additional 24 h. ESI-MS of both aliquots only contained patterns for 3 and 4. After workup, a white solid was collected (4.1 mg, 91%). The ¹H NMR of this solid indicated recovering of dibenzoylmethane.

2,6-bis(1-methylbenzoimidazol-2-yl)pyridine (MBBP).

This procedure includes slight modifications from the previously published synthesis.²⁸ 2,6-bis(2’-benzimidazolyl)pyridine (1.00 g, 3.21 mmol) was added to a round bottom flask containing acetone (30 mL) and powdered potassium hydroxide (0.90 g, 16 mmol). This mixture was stirred at room temperature for 30 min during which time it became brown-red. Methyl iodide (0.60 ml, 9.6 mmol) was added and the reaction stirred overnight. Following the addition of water, a light tan precipitate was produced. This solid was removed via filtration, dissolved in dichloromethane, and the solution was dried over sodium sulfate. Following filtration to remove the drying agent, the filtrate was collected and brought to dryness under reduced pressure to give a tan solid (0.50 g, 46% yield) ¹H NMR (CDCl₃) δ (ppm): 8.40 (d, 2H, 8.5 Hz), 8.04 (t, 1H, 7.9 Hz), 7.87 (d, 2H, 7.9), 7.46 (d, 2H 7.37 Hz), 7.36 (m, 4H), 4.24 (s, 6H). These ¹H NMR features match those previously reported.²⁸

Synthesis of [(MBBP)Ni(CH₃C(O)CHC(O)CH₃)]ClO₄ (5) and [(MBBP)Ni((CH₃)₃CC(O)CHC(O)C(CH₃)₃)]ClO₄ (6).

These complexes were prepared under air. Ni(ClO₄)₂·6H₂O (24.7 mg, 0.0675 mmol) was dissolved in CH₃CN (2 mL). The diketone, 2,4 pentanedione (7.0 μL, 0.0675 mmol) or 2,2,6,6-tetramethyl-3,5-heptanedione (13.8 μL, 0.0675 mmol), was added along with triethylamine (9.4 μL, 0.0675 mmol). The resulting blue solution was stirred for 30 min at ambient temperature. MBBP (22.9 mg, 0.0675 mmol) dissolved in CH₃CN (2 mL) was added, which resulted in the precipitation of a green solid. This solid was carefully collected and dried by gravity filtration using a Buchner funnel and filter paper (5: 19 mg, 46%; 6: 18 mg, 40%). ¹H NMR analysis of each solid indicated the presence of a single compound. 5: ESI-MS: m/z calc. for C₂₆H₂₄N₅NiO₂·CH₃CN, 537.2 [M-ClO₄]⁺; found 537.1; 496.1 [M-CH₃CN-ClO₄]⁺; found 496.3. Anal. Calc. for C₂₆H₂₄ClN₅NiO₆: C, 52.34; H, 4.05; N, 11.74. Found: C, 52.28; H, 4.11; N, 11.66. UV-Vis λ_max, nm (ε, M⁻¹ cm⁻¹): 249 (13720), 304 (40254), 350 (22999), 366 (16041). FT-IR (KBr, cm⁻¹): 3440 (ν_O-H), 3120 (ν_C-H), 2360 (ν_C-N), 2030 (ν_C-H), 1930 (ν_C-H), 1600 (ν_C=C), 1400 (ν_C=O), 1110 (ν_ClO4), 750(_νC-H), 620(ν_ClO4). 6: ESI-MS: m/z calc. for C₃₂H₃₆N₅NiO₂·CH₃CN, 621.3 [M-ClO₄]⁺; found 621.2; calc. 580.2 [M-CH₃CN-ClO₄]⁺; found 580.2. Anal. Calc. for C₃₂H₃₆ClN₅NiO₆·0.4 H₂O: C, 55.61; H, 5.34; N, 10.12: Found: C, 55.86; H, 5.39; N, 10.18. UV-Vis λ_max, nm (ε, M⁻¹ cm⁻¹): 249 (11864), 305 (29644), 348 (17675), 366 (10896). FT-IR (KBr, cm⁻¹): 3410 (ν_O-H), 3150 (ν_C-H), 2980 (ν_C-H), 2020 (ν_C-H), 1590(ν_C=C), 1410 (ν_C=O), 1120 (ν_ClO4), 750(_νC-H), 630(ν_ClO4).

Analysis of filtrates.

Analysis of the filtrates remaining following removal of the green solids of 5 and 6 using ¹H NMR and ESI-MS revealed the presence of the diketonate complexes (5 and 6) as well as [(MBBP)₂Ni](ClO₄)₂ (8), which was independently synthesized and characterized (vida infra).

Synthesis of 6 following previously published procedures.

Under N₂, Ni(ClO₄)₂·6H₂O (24.7 mg, 0.0675 mmol) was dissolved in degassed CH₃CN (2 mL). Degassed 2,2,6,6-tetramethyl-3,5-heptanedione (13.8 μL, 0.0675 mmol) and degassed triethylamine (9.4 μL, 0.0675 mmol) was then added to the vial and stirred for 30 minutes. MBBP (22.9 mg, 0.0675 mmol) dissolved in degassed CH₃CN (2 mL) was added, which resulted in the precipitation of a green solid. This solution was stirred for 2 h, then degassed Et₂O (8 mL) was added to the mixture. This mixture was then placed in the freezer overnight (−12 °C). The green precipitate collected on the bottom, so the solution was decanted off and the precipitate washed with more Et₂O and dried. The ¹H NMR under paramagnetic conditions and the ESI-MS of the solid indicated that 6 is present along with 8.

Attempted Synthesis of [MBBP)Ni(PhC(O)CHC(O)Ph)]ClO₄ (7-ClO₄).

The previously reported synthesis for this complex was used in an attempt to make this complex.²⁴ Ni(ClO₄)₂·6H₂O (24.7 mg, 0.0675 mmol) was dissolved in CH₃CN (~2 mL). This solution was combined with dibenzoylmethane (15.2 mg, 0.0675 mmol) and triethylamine (9.4 μL, 0.0675 mmol). The resulting mixture was stirred for 30 min at which point it was transferred to vial containing MBBP (22.9 mg, 0.0675 mmol). This resulted in the formation of a clear green solution. After stirring for 2 h, Et₂O (8 mL) was added, which resulted in the deposition of a green-yellow precipitate. The solid was collected and dried under vacuum. ESI-MS (CH₃CN) and ¹H NMR (CD₃CN) under paramagnetic conditions showed that both [MBBP)Ni(PhC(O)CHC(O)Ph)]ClO₄ (7-ClO₄) and 8 were present. We were unable to selectively precipitate 7-ClO₄ as had been possible with 5 and 6.

Attempted Synthesis of [(MBBP)Ni(PhC(O)CHC(O)Ph)]Cl (7-Cl).

To a solution of NiCl₂ (16.0 mg, 0.0675 mmol) in CH₃CN (2 mL) was added dibenzoylmethane (15.2 mg, 0.0675 mmol) and triethylamine (9.4 μL, 0.0675 mmol). The mixture was stirred for about 30 min at which point it was added to a vial containing MBBP (22.9 mg, 0.0675 mmol). A green precipitate formed which was separated by filtration (11.3 mg). Analysis of this solid by ¹H NMR (d₆-DMSO) showed the presence of a mixture of 7-Cl and 8. Analysis of the filtrate by ¹H NMR (d₆-DMSO) also showed the presence of both complexes. We were unable to selectively crystallize 7-Cl as was possible with 5 and 6. We note that while 7-Cl is insoluble in CD₃CN, complex 8 is soluble.

Synthesis of [(MBBP)₂Ni](ClO₄)₂ (8).

Ni(ClO₄)₂·6H₂O (12.4 mg, 0.0338 mmol) was combined with MBBP (22.9, 0.0675) in CH₃CN (3 mL). This solution was stirred for 12 h at room temperature. Slow diffusion of Et₂O into the CH₃CN solution produced small brown crystals suitable for single crystal X-ray crystallographic characterization. (26 mg, 83% yield). ESI-MS: m/z calc. for C₄₂H₃₄N₁₀Ni·ClO₄, 835.1 [M-ClO₄]⁺, found 835.2, calc. 368.1 [M-(2ClO₄)]²⁺; found 368.5 [M-(2ClO₄)]²⁺. Anal. Calc. for C₄₂H₃₄Cl2N₁₀NiO₈·0.4 H₂O: C, 53.46; H, 3.71; N, 14.84. Found: C, 53.08; H, 3.77; N, 14.66. UV-Vis λ_max, nm (ε, M⁻¹ cm⁻¹): 249 (12825), 304 (33046), 357 (26261), 366 (20572). FT-IR (KBr, cm⁻¹): 3420 (ν_O-H), 3130 (ν_C-H), 2020 (ν_C-H), 1600 (ν_C=C), 1480 (ν_C=O), 1060 (ν_ClO4), 750 (_νC-H), 620 (ν_ClO4).

Probing for aliphatic C-C bond cleavage reactivity involving 7-ClO₄.

Ni(ClO₄)₂·6H₂O (7.3 mg, 0.02 mmol) was dissolved in CH₃CN (2.7 ml). This solution was added to a mixture of dibenzoylmethane (4.5 mg, 0.02 mmol) and NEt₃ (2.8 μL, 0.02 mmol). The resulting mixtures was stirred for 30 min at which point it was added to MBBP (6.8 mg, 0.02 mmol). This clear green solution was stirred for 30 min. To this solution was added H₂O (300 μL) and NEt₃ (279 μL, 2.00 mmol) followed by purging with O₂ and stirring for 24 h at ambient temperature. CH₂Cl₂ (8 ml) and HCl (8 ml, 1 M) were then added, followed by stirring for 1 h and subsequently the organic layer was separated. The organic layer was dried over Na₂SO₄ and brought to dryness under vacuum to give a white solid (3.3 mg, 73% based on dibenzoylmethane). The ¹H NMR (CDCl₃) spectrum of the solid showed only dibenzoylmethane and no aliphatic C-C bond cleavage products.

In a separate experiment, a similar reaction was performed with the only difference being after stirring the mixture of 7-ClO₄ for 30 minutes in dry CH₃CN, one equivalent of NEt₃ (~3 μL, 0.02 mmol) was added. An aliquot was taken before O₂ purging and then again after stirring for 24 h at room temperature. ESI-MS of both aliquots only contained patterns of 7 and 8. After workup using HCl and CH₂Cl₂ extraction as described above, a white solid was isolated (4.1 mg, 91%). The ¹H NMR of this solid indicated only dibenzoylmethane was isolated.

Probing for aliphatic C-C bond cleavage reactivity involving 7-Cl.

The same conditions as 7-ClO₄ were used, except NiCl₂·6H₂O (4.8 mg, 0.02 mmol) was used instead of Ni(ClO₄)₂·6H₂O in dry CH₃CN. After adding one equivalent of NEt₃ (~3 μL, 0.02 mmol), the reaction was stirred for 10 min. An aliquot was taken before O₂ purging and then again after stirring for 24 h at room temperature. ESI-MS of both aliquots only contained patterns of 7 and 8. After workup using HCl and CH₂Cl₂ extraction as described above, a white solid was isolated (4.0 mg, 88%). The ¹H NMR of this solid indicated only dibenzoylmethane was isolated.

Evaluation of the reactivity of 1–7 in the presence of D₂O and NEt₃ using ¹H NMR.

Each isolated complex (4.09×10⁻³ mmol) was evaluated by ¹H NMR over 24 h under aerobic conditions in dry CD₃CN only (600 μL), dry CD₃CN (600 μL) containing NEt₃ (5 eq), dry CD₃CN (540 μL) containing 10% added D₂O (60 μL), or dry CD₃CN (540 μL) containing 10% added D₂O (60 μL) and NEt₃ (5 eq). Evaluations for complexes 5 and 6 were also done under previously published conditions (i.e. dried solvent, 1 eq NEt₃).²⁴ Each isolated complex was evaluated by UV-vis over 24 h in CH₃CN:H₂O (9:1) in the presence of 5 equivalents of NEt₃.

Addition of MBBP (1 eq) to 5 in CD₃CN.

Complex 5 (2.0 mg, 3.40×10⁻³ mmol) was dissolved in CD₃CN (600 μL) and a ¹H NMR spectrum was collected under paramagnetic conditions. MBBP (1.1 mg, 3.40×10⁻³ mmol) was added, the solution was briefly mixed and another ¹H NMR spectrum was collected. Additional spectra were collected after 3 h and 24 h. The NMR spectra indicated significant formation of 8.

X-Ray Crystallography.

Single crystals of 1-3 and 8 were used for X-ray crystallography studies. Structures of 1, 3 and 8 were determined at the University of Montana (UM) where a crystal of each was mounted on a glass fiber using viscous oil and transferred to a Bruker D8 Venture using Mo Kα (λ = 0.71073 Å) radiation for data collection at 100 K. The structure of 2 was determined at Utah State University. Single crystals of 2 were mounted on a glass fiber using viscous oil and transferred to a Rigaku XtaLAB Mini II Diffractometer using Mo Kα (λ = 0.71073 Å) radiation for data collection at 100 K.

1, 3 and 8. Collected data were corrected for absorption using the SADABS.³² Using Olex2³³, the structure of each was solved with the SHELXT³⁴ structure solution program using direct methods and refined with the SHELXL³⁵ refinement package using least squares minimization. All non-hydrogen atoms were refined with anisotropic thermal parameters. For 1 the hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. For 3 the hydrogen atoms attached to the heteroatoms were located in the residual electron density maps, placed, and refined with isotropic thermal parameters. All other hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. For 8 all the hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. Isotropic thermal parameters of riding hydrogen atoms were fixed to 1.2 times the U value of the atoms they are linked to (1.5 times for methyl groups). Calculations and refinement of structures were carried out using APEX2³⁶, SHELXTL³⁷, and Olex2³³ software.

The structure of 8 was found to contain indistinguishable solvent molecules within voids in the lattice. Attempts at modelling this solvent were not able to produce a suitable model. The SQUEEZE³⁸ routine within PLATON³⁹ was utilized to account for the residual, diffuse electron density and the model is refined against these data. A total of 27 electrons per unit cell were corrected.

2: Data were corrected for absorption using Gaussian grid (numerical integration) and a 0.5 mm 1D horizontal Gaussian beam correction for the graphite monochromator. Using Olex2³³, the structure was solved with the SHELXT³⁴ structure solution program using intrinsic phasing and refined with the SHELXL³⁵ refinement package using least squares minimization. All non-hydrogen atoms were refined with anisotropic thermal parameters. Hydrogen atoms for 2 attached to heteroatoms were found in the residual electron density maps, placed, and refined with isotropic thermal parameters. All other hydrogen atoms for 2 were placed in geometrically calculated positions and refined using a riding model. Isotropic thermal parameters of the placed hydrogen atoms were fixed to 1.2 times the U value of the atoms they are linked to (1.5 for methyl groups). Calculations and refinement of structures were carried out using CrysAlisPro⁴⁰, SHELXL³⁵ and Olex2³³ software.

Compounds 1 and 3 crystallize in the monoclinic crystal system, while 2 and 8 are triclinic. The space groups for each complex were the following: 1: 1C_2/c1, 2: P-1, 3: P12_1/c1, 8: P1. In 8, three perchlorate oxygen atoms were modelled as disordered over two positions (O30/O31 0.42:0.58, O17/O19 0.67:0.33, O7/O8 0.66:0.34).

Results and Discussion

Synthesis and characterization of TERPY-ligated Ni(II) diketonate complexes.

To evaluate the solution and O₂ reactivity properties of Ni(II) diketonate complexes supported by a tridentate chelate ligand, three new TERPY-ligated complexes were prepared and characterized. Complexes 1–3 were prepared as outlined in Scheme 4 under air, with the synthesis of 1 and 2 being performed using excess diketone. Isolation of the desired diketonate complex required crystallization or precipitation of the product to separate it from [(TERPY)₂Ni](ClO₄)₂ (4), which was also formed in the reaction mixture (vida infra). Complexes 1-4 were characterized by X-ray crystallography, elemental analysis, ¹H NMR, ESI mass spectrometry, UV-vis, and FTIR (Figures S1-S19).

Scheme 4.

The synthetic pathway for the preparation of 1-3. Compound 4 is a contaminating species in each reaction mixture.

X-ray structures of 1-3.

Details of the X-ray data collection and refinement for 1-3 is provided in Table S1. Selected bond distances and angles are given in Tables S2 and S3. Representations of the cationic portions of the X-ray structures of 1-3 are shown in Figure 1. Each cation contains a pseudo-octahedral Ni(II) center with the meridional coordinated TERPY ligand and one diketonate oxygen donor occupying coordination positions in the equatorial plane. The other diketonate oxygen donor occupies an axial position trans to a coordinated solvent molecule (CH₃CN (1) or H₂O (2 and 3)). The Ni-N bond lengths for the TERPY ligand are in the range of 2.00–2.14 Å. The Ni-O(diketonate) distances are in the range of 1.98–2.05 Å. The Ni-N(CH₃CN) distance in 1 (2.09 Å) is similar to the Ni-O(H₂O) distances in 2 and 3 (2.14 and 2.09 Å, respectively). Comparison of the bond distances to those reported by Ramasubramanian et al. for [(MBBP)Ni((CH₃)₃CC(O)CHC(O)C(CH₃)₃)]ClO₄ (5) and [(MBBP)Ni(PhC(O)CHC(O)Ph)]Cl (8) revealed similar Ni-O distances and Ni-N distances.²⁴

Figure 1.

Cationic portions of a) 1, b) 2 and c) 3. Hydrogen atoms except for those of coordinated water molecules have been omitted for clarity. Ellipsoids are plotted at the 50% probability level.

Solution characterization of 1–3.

Based on prior results from our laboratory²⁶, we hypothesized that the pseudo-octahedral TERPY-ligated Ni(II) diketonate complexes would be amenable to characterization by ¹H NMR under paramagnetic conditions. We also hypothesized that ¹H NMR would be a useful way to monitor the solution reactivity of these diketonate complexes. We therefore pursued studies of the ¹H NMR features of crystalline samples 1-3 in CD₃CN. As shown in Figure 2, each complex exhibits signals over a range of −15 to +75 ppm. An additional broad peak is found at 132 ppm (Figures S1-S3). Assignment of resonances was made on the basis of integrated intensity and chemical shift considering a contact shift model.²⁶ The signal at 132 ppm is tentatively assigned to the pyridyl C-H atoms adjacent to the nitrogen donors. Due to the broad nature of this signal, the integrated intensity could not be determined. The signals b and c for each complex (Figure 2) integrate to two hydrogen atoms and are assigned to the meta C-H positions on the terminal pyridyl donors of the TERPY ligand. A resonance at 38–39 ppm (d), which integrates to one hydrogen, is assigned to the meta C-H on the central pyridyl ring. Resonances at 16 (e, 1H) and 12 (f, 2H) ppm, respectively, are assigned to the para C-H positions of the central and terminal pyridyl rings. A signal integrating to one hydrogen at −15 ppm is assigned as the central C-H moiety of each β-diketonate ligand. The spectrum for 1 also contains a singlet at 5.3 ppm, with an integration of six hydrogens, which is assigned to the methyl hydrogens of the acetoacetonate diketonate ligand. Complex 2 has a similar singlet with an integration of 18 hydrogens for the trimethylacetyacetonate ligand. The ¹H NMR spectrum of 3 contains three signals in the range of 7.7–10.7 ppm which integrate to 10 hydrogens for the phenyl hydrogens of the diketonate ligand.

Figure 2.

¹H-NMR spectra of TERPY-ligated a) 1, b) 2, c) 3 and d) 4 in CD₃CN.

Careful analysis of the ¹H NMR features of analytically pure 1 (Figure 2(a)) revealed the presence of a minor impurity. The signals for this impurity match those of the bis-TERPY complex, [(TERPY)₂Ni](ClO₄)₂ (4, Figure S4)).^29,30 Notably, evaluation of the reaction mixture filtrates (Figures S5-S7) remaining following isolation of the precipitated/crystalline 1-3 using ¹H NMR and ESI-MS indicated the presence of the diketonate complexes 1-3 as well as 4. Analysis of crystalline samples of 1-3 in CH₃CN by ESI-MS revealed molecular ions for the [(TERPY)Ni(diketonate)]⁺ and [(TERPY)Ni(diketonate)(CH₃CN)]⁺ cations (Figure S8-S10) as well as an isotope cluster for the [(TERPY)₂Ni(ClO₄)]²⁺ (4, Figure S11) cation. Overall, it is clear that 4 is formed in the reaction mixtures leading to the formation of 1-3.

Synthesis and solution characterization of MBBP-ligated Ni diketonate complexes.

Based on our experience in preparing and characterizing the [(TERPY)Ni(diketonate)]ClO₄ complexes (1-3), we examined the previously reported MBBP-ligated diketonate complexes 5-7 to evaluate: (a) Is [(MBBPNi)₂](ClO₄)₂ (8) similarly present in the reaction mixtures of the [(MBBP)Ni(diketonate)]ClO₄ complexes? (b) Can ¹H NMR be used to characterize the solution properties of the MBBP-ligated diketonate complexes? Our interest in the first question stems in part from color descriptions of these previously reported compounds.²⁴ Specifically, the diketonate derivatives [(MBBP)Ni(CH₃C(O)CHC(O)CH₃)]ClO₄ (5), [(MBBP)Ni((CH₃)₃CC(O)CHC(O)C(CH₃)₃)]ClO₄ (6), and [(MBBP)Ni(PhC(O)CHC(O)Ph)]ClO₄ (7-ClO₄) were reported as being orange-brown solids whereas the CIF files for 6 and 7-ClO₄ list green crystals.²⁴ Our experience with the [(TERPY)Ni(diketonate)]ClO₄ complexes suggested that the desired diketonate complexes should be green whereas the a bis-MBBP complex [(MBBP)₂Ni](ClO₄)₂ (8) may be tan to orange-brown. In their paper, Ramasubramanian, et al. did briefly mentioned that the [(MBBP)₂Ni](ClO₄)₂ (8) complex is detected by ESI-MS following oxygenation of the diketonate complexes. We hypothesized that it also could be present in the reaction mixtures leading to the isolation of 5-7.

Using the synthetic procedures under air used for the preparation of 1-3, we attempted the synthesis of 5-7 (Scheme 5).²⁴ We identified the following differences from the prior report by Ramasubramanian et al., noting that their reactions were run under N₂.²⁴ In our hands, in each reaction mixture, a green precipitate is produced prior to Et₂O addition. As shown via synthetic investigations for 6, this occurs whether the reaction was run under air or N₂. Isolation and evaluation of the green solid as well as the filtrate by ¹H NMR (Figure 3; Figures S20 (R = -CH_3; solid) and S21 (R = -CH_3; filtrate); Figures S22 (R = -C(CH₃)₃; solid) and S23 (R = -C(CH₃)₃; filtrate)) and ESI-MS (Figures S24 and S25) indicated that the green solids were the desired [(MBBP)Ni(diketonate)]ClO₄ complexes (5 and 6). ESI-MS analysis of solutions of 5 and 6 showed the presence of the [(MBBP)₂Ni]²⁺ cation (Figure S24 and S25). The presence of this ion suggested that a trace amount may be present in the samples or that ligand exchange was occurring under the conditions of the mass spectral experiment. To fully characterize [(MBBP)₂Ni](ClO₄)₂ (8) we independently prepared the complex which was then characterized using X-ray crystallography (Figure 4), ¹H NMR (Figures S26 (CD₃CN) and S27 (d₆-DMSO)), ESI-MS (Figure S28), UV-vis (Figure S29), and FTIR (Figure S30). The structural features of 8 are similar to those of [(TERPY)₂Ni](ClO₄)₂ with Ni-N distances of 2.03–2.12 Å.³⁰

Scheme 5.

Products identified in the syntheses of 5–7. Complex 8 was identified in each reaction mixture using ¹H NMR based on an independently generated sample.

Figure 3.

¹H NMR spectra of MBBP-ligated a) 5, b) 6 c) 7-ClO₄, d) 7-Cl and e) 8 in CD₃CN.

Figure 4.

Thermal ellipsoid representation of a cationic portion of the X-ray structure of 8. The complex crystallizes with two cations per asymmetric unit. Hydrogen atoms have been omitted for clarity. Ellipsoids are plotted at the 50% probability level.

The ¹H NMR features of 5 and 6 (Figure 3a) and b)) are like those of the TERPY diketonate complexes. Resonances for the MBBP derivatives can be assigned based on integrated intensity and consideration of a primary contact shift contribution to the chemical shifts.²⁶ A signal at 65 ppm integrating to two hydrogens is assigned to the aryl C-H closest to each benzimidazole nitrogen donor. A sharp resonance integrating to two hydrogens at ~26 ppm is assigned to the adjacent aryl C-H moiety. Overlapping signals integrating to three hydrogens at ~15 ppm correspond to the meta- and para hydrogens of the central pyridyl donor of the MBBP ligand. The remaining benzimidazole aryl hydrogens as well as the diketonate substituent hydrogens (except the central C-H) are found in the region of 0–10 ppm. The diketonate central C-H resonance is found at −15 ppm. We note that assignment of the ¹H NMR spectral features of 6 enabled us to compare the results of synthetic reactions performed for this complex under air and N₂. For the latter, as is shown in Figures S31 and S32, the ¹H NMR and ESI-MS features of the compound isolated under N₂ indicate that some 8 is present. The presence of the bis-MBBP complex is due to a different work up procedure being employed under N₂ that didn’t enable physical separation of 6 and 8.

Our attempts to generate the MBBP-ligated Ni(II) dibenzoylmethane derivative 7-ClO₄ always resulted in the isolation of mixtures of 7-ClO₄ with 8. Representative ¹H NMR spectra of the filtrate showed neither complex (Figure S33), but isolated precipitate (Figure 3(c); Figure S34) suggested the presence of both complexes. ESI-MS data shows the filtrate contains trace amounts of 7-ClO₄, while the precipitate is consistent with a mixture of 7-ClO₄ and 8 (Figure S35). Attempted syntheses of the chloro derivative [(MBBP)Ni(PhC(O)CHC(O)Ph)]Cl (7-Cl) produced similar results. By ¹H NMR, the precipitated solid contains a mixture of 7-Cl and 8 (Figure 3(d); Figure S36). The remaining filtrate also contains trace amounts of both compounds (Figure S37). ESI-MS of the isolated solid and filtrate also indicates the presence of both compounds (Figure S38).

Exposure of complexes to H₂O/NEt₃ under aerobic conditions.

Ramasubramanian et al. reported that exposure of 5-7 to O₂ in CH₃CN in the presence of NEt₃ (1 eq) led to oxidative aliphatic carbon-carbon bond cleavage within the diketonate unit (52% yield). They additionally reported that the presence of 10% water produced a slightly lower yield (41%) of the aliphatic C-C bond cleavage product (benzoic acid). As we prepared and isolated analytically pure 1-6 under aerobic conditions, it was evident that the complexes were not reactive with O₂. We therefore next examined whether the addition of NEt₃ and/or water would lead to diketonate cleavage reactivity. ¹H NMR samples were prepared under the following conditions and were evaluated after being stored for 24 h at ambient temperature under air: a) dry CD₃CN only; b) dry CD₃CN + 10% D₂O; c) dry CD₃CN + 1 eq NEt₃ (6 and 7 only); d) dry CD₃CN + 5 eq NEt₃; e) dry CD₃CN + 10% D₂O + 5 eq NEt₃. Spectra associated with these experiments are provided in the Supporting Information (1: Figures S39-S42; 2: Figures S43-S46; 3: Figures S47-S50; 5: Figures S51-S55; 6: Figures S56-S60). As shown for the TERPY-ligated Ni(II) acetylacetonate complex 1 in Figures S39-S42, in the presence of D₂O (10%) or NEt₃ (5 eq) a small amount of the bis-TERPY complex 4 appears to form over 24 h relative to that seen in CD₃CN alone. The amount of 4 present appears to be influenced more by the presence of NEt₃ than water. Notably, a solution containing both D₂O and NEt₃ shows the most significant amount of bis-TERPY complex (4) formation after 24 h (Figure 5, Figure S42). The diketonate complexes 2 and 3 exhibit similar reactivity albeit resulting in the presence of an overall smaller amount of 4 relative to the starting material. By ¹H NMR, 4 is stable with respect to the presence of D₂O (10%) and NEt₃ (5 eq) over 24 h (Figure S61). Analysis of the reactivity of 1 and 4 in CH₃CN in the presence of D₂O (10%) and NEt₃ (5 eq) at ambient temperature using UV-visible spectroscopy (Figure S63) shows that 4 is stable whereas 1 exhibits minor changes, possibly suggesting the formation of 4.

Figure 5.

¹H-NMR spectra of 1 in dry CD₃CN with 10% D₂O (v:v) and excess NEt₃ (5 eq) over time, starting with a preaddition spectrum (NEt₃ and D₂O not added) (a), a spectrum after addition of all reagents (b), 3 hours later (c) and 24 hours later (d). The blue arrow highlights the growth of 4, and the red arrow represents the disappearance of the diketonate C-H proton.

The MBBP-ligated complexes 5 and 6 were exposed to the same conditions (CD₃CN, 10% D₂O, NEt₃ (5 eq) for 24 h. As was the case for the TERPY complexes, the acetoacetonate derivative 5 showed an increase in the amount of 8 present (Figure 6). The observed reactivity was complete with 3 h but was monitored over 24 h. The formation of 8 is also observed in solutions of 6, although the reaction proceeds more slowly than for 5. By ¹H NMR, 8 shows no change under the previously described conditions (as was observed for 4; Figure S62). As was done for 1 and 4, the reactivity was also analysed by UV-visible spectroscopy for 5 and 8 under the same conditions (CD₃CN, 10% D₂O, NEt₃ (5 eq; Figure S64). While 8 displayed no change over 24 hours, 5 exhibited a small band shift after 30 minutes, with no change being observed afterward (both at 3 hr and 24 hr).

Figure 6.

¹H-NMR stacked spectra of 5 in CD₃CN with 10% D₂O and excess NEt₃ over time, starting with a pre-addition spectrum (NEt₃ and D₂O not added) (a), a spectrum after addition of all reagents (b), 3 hours later (c) and 24 hours later (d). The blue arrow highlights the growth of 8, and the red arrow represents the disappearance of the α proton.

Notably, addition of TERPY (1 eq) to a CD₃CN solution of 1 showed significant conversion to 4 (Figure S65). This indicates that free TERPY appears to be able to displace the diketonate ligand. In a similar experiment involving addition of MBBP (1 eq) to a CD₃CN solution of 5, growth of signals for 8 can also be observed (Figure S66).

Evaluation of aliphatic C-C bond cleavage reactivity.

To probe for aliphatic C-C bond cleavage, we stirred an CH₃CN solution of 3 containing 10% H₂O (v:v) and NEt₃ (100 eq) for 24 h at ambient temperature after purging with O₂. Exposure of the solution to 1M HCl followed by extraction with CH₂Cl₂ resulted in the isolation of unaltered dibenzoylmethane (71% recovery, Figure S67). A similar experiment performed using an in situ-generated sample of [(MBBP)Ni(PhC(O)CHC(O)Ph)]ClO₄ (7-ClO₄) also resulted only in the isolation of unaltered dibenzoylmethane (73% recovery, Figure S68). Notably, an experiment performed using 3, 7-ClO₄, or 7-Cl in the presence of 1 eq NEt₃ as reported by Ramasubramanian et al.²⁴ produced only unreacted dibenzoylmethane (71%−91%, Figures S69-S74) after 24 h stirring and acidic workup.

Stability of Ni(II) diketonate complexes under CH₃CN/ aqueous basic conditions.

Both sets of tridentate Ni(II) diketonate complexes are stable when exposed to O₂ both as solids and in CH₃CN solution. This is evident as: 1) these complexes can be synthesized and isolated in air; and 2) no oxidative products were identified or isolated from solutions under aerobic conditions. These results are consistent with literature precedent, which suggests that only Ni(II) complexes containing an electron rich diketonate ligand exhibit oxidative C-C bond cleavage reactivity with O₂.^15–23 Some evidence was found for potential ligand exchange reactivity wherein the tridentate Ni(II) diketonate complexes were converted to bis MBBP or bis TERPY Ni(II) complexes in solution. The exchange reactivity is enhanced for diketonates with less steric hinderance.

Conclusions

Oxidative aliphatic C-C bond cleavage reactions of diketones mediated by first-row divalent metals and involving O₂ are of current interest.^9–23 The highest level of oxidative cleavage reactivity has been found for electron rich Ni(II) diketonates. The recent reactivity reported for a MBBP-ligated Ni(II) aryl diketonate complex drew our interest due to the oxidative cleavage of a non-substituted diketone upon reaction with O₂.²⁴ Typically Ni(II) complexes of an unsubstituted aryl diketonate are air stable.²⁰ We have been unable to reproduce these results with either MBBP- or TERPY-supported Ni(II) diketonate complexes. Instead, these complexes are stable in air, and in aerobic CH₃CN/aqueous basic conditions or CD₃CN/NEt₃ only exhibit some possible ligand exchange reactivity.