O₂-dependent Aliphatic Carbon-carbon Bond Cleavage Reactivity in a Ni(II) Enolate Complex Having a Hydrogen Bond Donor Microenvironment; Comparison with a Hydrophobic Analog

Abstract

A mononuclear Ni(II) complex having an acireductone type ligand, and supported by the bnpapa (N,N-bis((6-neopentylamino-2-pyridyl)methyl-N-((2-pyridyl)methyl)amine ligand, [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14), has been prepared and characterized by elemental analysis, ¹H NMR, FTIR, and UV-vis. To gain insight into the ¹H NMR features of 14, the air stable analog complexes [(bnpapa)Ni(CH₃C(O)CHC(O)CH₃)]ClO₄ (16) and [(bnpapa)Ni(ONHC(O)CH₃)]ClO₄ (17) were prepared and characterized by X-ray crystallography, ¹H NMR, FTIR, UV-vis, mass spectrometry, and solution conductivity measurements. Compounds 16 and 17 are 1:1 electrolyte species in CH₃CN. ¹H and ²H NMR studies of 14, 16, and 17 and deuterated analogs revealed that the complexes having six-membered chelate rings for the exogenous ligand (14 and 16) do not have a plane of symmetry within the solvated cation and thus exhibit more complicated ¹H NMR spectra. Compound 17, as well as other simple Ni(II) complexes of the bnpapa ligand (e.g. [(bnpapa)Ni(ClO₄)(CH₃CN)]ClO₄ (18) and [(bnpapaNi)₂(μ-Cl)₂](ClO₄)₂ (19)), exhibit ¹H NMR spectra consistent with the presence of a plane of symmetry within the cation. Treatment of [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14) with O₂ results in aliphatic carbon-carbon bond cleavage within the acireductone-type ligand and the formation of [(bnpapa)Ni(O₂CPh)]ClO₄ (9), benzoic acid, benzil, and CO. Use of ¹⁸O₂ in the reaction gives high levels of incorporation (>80%) of one labeled oxygen atom into 9 and benzoic acid. The product mixture and level of ¹⁸O incorporation in this reaction is different than that exhibited by the analog supported the hydrophobic 6-Ph₂TPA ligand, [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2). We propose that this difference is due to variations in the reactivity of bnpapa- and 6-Ph₂TPA-ligated Ni(II) complexes with triketone and/or peroxide species produced in the reaction pathway.

Introduction

Interest in how the secondary environment of a metal center influences the reactivity of metal-bound ligands stems from the possible roles of secondary interactions in metal-catalyzed reactions in biological systems.¹ In the acireductone dioxygenases Ni(II)-ARD and Fe(II)-ARD, which contain the same protein component, it is proposed that differences in the secondary environment surrounding the divalent metal center influence the coordination mode of the acireductone substrate and the regioselectivity of the oxidative carbon-carbon bond cleavage reaction.¹ Specifically, in Ni(II)-ARD the positioning of a tryptophan side chain (W162) is suggested to promote the formation of a six-membered chelate ring in the enzyme substrate adduct (Scheme 1(top)).² Reaction of this adduct with O₂ results in oxidative cleavage of the C(1)-C(2) and C(2)-C(3) bonds and the formation of CO and carboxylate products. For Fe(II)-ARD, a more open active site is suggested to enable the formation of a five-membered chelate enediolate structure, within which only C(1)-C(2) bond cleavage occurs upon reaction with O₂ (Scheme 1(bottom)). For both Ni(II)-ARD and Fe(II)-ARD secondary hydrogen bonding interactions involving an arginine residue are suggested to stabilize enolate/enediolate coordination.

Scheme 1.

Interest in reactions of relevance to Ni(II)-ARD also stems from the fact that carbon monoxide is a molecule of current interest in biological systems. In humans, CO is primarily produced via the oxidative breakdown of heme catalyzed by heme oxygenases.³ While carbon monoxide is typically viewed as a toxic gas, recent studies have shown that this small molecule can have beneficial health effects. The signaling mechanism of CO is similar to that of NO in that it activates soluble guanylyl cyclase to produce cyclic guanylyl monophosphate (cGMP).³ This compound is involved in pathways pertinent to smooth muscle relaxation, platelet inhibition, and cell growth and differentiation. As these factors are related to several diseases, recent studies have focused on efforts toward the development of CO-releasing compounds for use as therapeutic agents.⁴ The majority of the chemical compounds tested so far have been transition metal carbonyl species⁵, although a main group CO-releasing compound (Na₂[H₃BCO₂]) has been recently reported.⁴ We are interested in the CO release properties of acireductones and flavonolates, both of which possess three consecutive carbon centers having an oxygen substitutent. In fungal and bacterial systems, quercetin dioxygenases catalyze CO release from a metal-coordinated flavonolate in a 2,4-dioxygenolytic ring cleavage reaction.^6–14 In a broad sense, we are interested in understanding in detail the chemical factors that modulate the CO-release reactivity of metal-coordinated acireductone and flavonolate species, with a long-term goal of producing new types of CO-releasing compounds containing these novel structural motifs.

Our laboratory has reported the only examples to date of synthetic complexes of relevance to substrate- and product carboxylate-bound forms of Ni(II)-ARD.^15–18 We recently reported that use of a chelate ligand having a mixed hydrophobic/hydrogen bond donor secondary environment produced the first example of a Ni(II) enediolate complex of an acireductone-type ligand (1, Figure 1(top)).¹⁸ Somewhat surprisingly, this complex was originally generated under similar conditions to those used to produce the mononuclear Ni(II) enolate complex [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2, Figure 1(bottom)). These combined results indicate that the presence of a hydrogen bond donor in the secondary environment influences the coordination chemistry of an acireductone-type ligand.

Figure 1.

Supporting chelate ligands used to isolate Ni(II) enediolate (top) and enolate (bottom) complexes of relevance to Ni(II)-ARD.

In terms of carboxylate chemistry, we discovered that the secondary environment of the chelate ligand affects the coordination chemistry of Ni(II)-carboxylate species that are generated in O₂-dependent ARD-type reactions. As shown in Scheme 2(top), a Ni(II)-carboxylate ligand in a hydrophobic microenvironment undergoes a shift from bidentate to monodentate in the presence of water (conversion of compounds 3–5 to 6–8).¹⁷ No evidence was found for this type of carboxylate shift chemistry in bnpapa-ligated Ni(II) carboxylate complexes (9–11, Scheme 2 (bottom)). In these analogs, the hydrogen bond donor appendages interact with one oxygen atom of the bidentate carboxylate. In a Ni(II) benzoate complex supported by a chelate ligand containing both hydrophobic phenyl appendages and a hydrogen bond donor, [(6-NA-6-Ph₂TPA)Ni(O₂CPh)(H₂O)]ClO₄ (12), a bidentate coordinated carboxylate ligand was also identified.¹⁸

Scheme 2.

From our initial studies, it is clear that the ligand environment of a Ni(II) center influences the chemistry of an acireductone-type ligand. In the studies outlined herein, we have further evaluated the influence of the chelate ligand environment by preparing and characterizing the enolate complex [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14) as an analog to the 6-Ph₂TPA-supported 2 (Figure 1). Complex 14 has been characterized by elemental analysis, ¹H NMR, FTIR, and UV-vis. Comparative O₂ reactivity studies for 2 and 14 have been performed under conditions wherein the inorganic and organic byproducts were identified. Interestingly, these reactions differ in the specific mixture of products that is generated and in the level of ¹⁸O incorporation in the benzoate/benzoic acid products. We propose that these differences are due to variations in the reactivity of bnpapa- and 6-Ph₂TPA ligated Ni(II) complexes with triketone and peroxide/hydroperoxide species produced in the reaction pathway.

In the course of this research, as a means of gaining additional insight in the spectroscopic and solution properties of [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄(14), we have also prepared and characterized mononuclear bnpapa-supported Ni(II) acetoacetonato and hydroxamato complexes, as well as perchlorate and chloro derivatives. Preparation of this family of complexes has enabled a structural comparison with a similar family of 6-Ph₂TPA-supported mononuclear Ni(II) complexes. This comparison has provided insight into the structural factors that may be responsible for the differences in O₂ reactivity between [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄(14) and [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2) noted above.

Experimental

General Methods

All reagents and solvents were obtained from commercial sources and were used as received unless otherwise noted. Solvents were dried according to published procedures and were distilled under N₂ prior to use.¹⁹ Air sensitive reactions were performed in the MBraun Unilab glovebox or a Vacuum Atmospheres MO-20 glovebox under a N₂ atmosphere. The ligand bnpapa and the Ni(II) complex [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2) were prepared according to literature procedures.^15,17

Physical Methods

¹H NMR spectra of Ni(II) complexes were obtained using a Bruker ARX-400 spectrometer as previously described.²⁰ Chemical shifts (in ppm) are referenced to the residual solvent peak(s) in CHD₂CN (¹H, 1.94 (quintet) ppm). UV-vis spectra were collected on a HP8453A spectrometer at ambient temperature. FTIR spectra were recorded on a Shimadzu FTIR-8400 spectrometer as KBr pellets. Conductance measurements were made at 22(1) °C using an YSI model 31A conductivity bridge with a cell having a cell constant of 1.0 cm⁻¹ and using Me₄NClO₄ as a 1:1 electrolyte standard. Preparation of the acetonitrile and standard solutions for conductance measurements and subsequent data analysis were performed as previously described.²¹ Mass spectrometry data was obtained at the Mass Spectrometry Facility, Department of Chemistry, University of California, Riverside. Elemental analyses were performed by Atlantic Microlabs, Inc., Norcross, Georgia, or Canadian Microanalytical Service, Inc., British Columbia.

Caution! Perchlorate salts of metal complexes with organic ligands are potentially explosive. Only small amounts of material should be prepared, and these should be handled with great care.²²

[(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14)

Under a N₂ atmosphere, equimolar amounts of bnpapa (0.273 mmol) and Ni(ClO₄)₂·6H₂O (0.273 mmol) were mixed in acetonitrile and stirred until the ligand had fully dissolved. Me₄NOH·5H₂O (0.301 mmol) was then added to the mixture, which was stirred for additional 15 minutes, resulting in a dark green solution. The solution was added to solid 2-hydroxy-1,3-diphenylpropan-1,3-dione²³ (0.301 mmol) and the mixture was stirred for 4 h, yielding a very dark orange/red solution. The solvent was then removed under reduced pressure. The solid was dissolved in CH₂Cl₂ and the solution was filtered through a celite plug. The filtrate was concentrated and brought to dryness under vacuum, leaving a dark orange/red solid, which was washed with Et₂O and dried under vacuum. Yield: 83%. Anal. Calcd for C₄₃H₅₁N₆O₇ClNi: C, 60.19; H, 5.99; N, 9.79. Found: C, 60.52; H, 5.88; N, 9.57. UV-vis [CH₃CN, nm (ε, M⁻¹cm⁻¹)] 393(10,000); FTIR: (KBr, cm⁻¹) 3420 (br s, ν_NH), 1094 (ν_ClO4), 621 (ν_ClO4).

[(bnpapa)Ni(CH₃C(O)CHC(O)CH₃)]ClO₄ (16)

A solution of Ni(ClO₄)₂·6H₂O (0.15 mmol) in CH₃CN (~1 mL) was added to solid bnpapa (0.15 mmol). The resulting mixture was stirred for 20 min at room temperature. The solution was then transferred to a vial containing Me₄NOH·5H₂O (0.15 mmol). To this mixture 2,4-pentanedione (0.15 mmol) was added and the resulting solution was stirred overnight. The solvent was then removed under reduced pressure. The remaining solid was dissolved in CH₂Cl₂ (~5 mL) and the solution was filtered through a celite/glass wool plug. Pentane diffusion into CH₂Cl₂ yielded purple crystals suitable for X-ray crystallography. The crystals were crushed and dried under vacuum prior to elemental analysis. Yield: 46%. Anal. Calcd for C₃₃H₄₇ClN₆NiO₆: C, 55.21; H, 6.60; N, 11.71. Found: C, 55.28; H, 6.55; N, 11.70. FTIR (KBr, cm⁻¹) ~3300 (s, ν_NH), 1089 (ν_ClO4), 622 (ν_ClO4). UV-vis [CH₃CN, nm (ε, M⁻¹cm⁻¹)] 244(31600), 305(17800), 543(sh), 966(19); FAB-MS [m/z (relative intensity)]: 617 ([M – ClO₄]⁺, 100%).

[(bnpapa)Ni(ONHC(O)CH₃)]ClO₄ (17)

A solution of Ni(ClO₄)₂·6H₂O (0.11 mmol) in CH₃CN (~1 mL) was added to solid bnpapa (0.11 mmol). The resulting mixture was stirred for 5 min at the room temperature. The solution was then transferred to a vial containing Me₄NOH·5H₂O (0.12 mmol) and acetohydroxamic acid (0.12 mmol), and the resulting solution was stirred for 1 h. The solvent was then removed under reduced pressure. The remaining solid was dissolved in CH₂Cl₂ (~5 mL) and the solution was filtered through a celite/glass wool plug. Diffusion of diethyl ether into a CH₂Cl₂/MeOH/iPrOH (1:0.5:1) solution of the compound yielded purple crystals suitable for single crystal X-ray crystallography. These crystals were crushed and dried under vacuum prior to elemental analysis. Yield: 60%. Anal. Calcd for C₃₀H₄₄N₇NiO₆Cl·0.25AHA·0.35CH₂Cl₂: C, 51.93; H, 6.34; N, 13.12. Found: C, 51.55; H, 6.41; N, 13.63. The presence of acetohydroxamic acid (AHA) in the crystalline sample was confirmed by X-ray crystallography. The presence of CH₂Cl₂ in the elemental analysis sample was confirmed by ¹H NMR. FTIR (KBr, cm⁻¹) ~3288 (s, ν_NH), 1103 (ν_ClO4), 621 (ν_ClO4); UV-vis [CH₃CN, nm (ε, M⁻¹cm⁻¹)] 247(25800), 325(10400), 548(sh), 977(32); ESI/APCI-MS [m/z (relative intensity)]: 592.2914 ([M – ClO₄]⁺, 100%).

[(bnpapa)Ni(ClO₄)(CH₃CN)]ClO₄ (18)

A solution of Ni(ClO₄)₂·6H₂O (0.07 mmol) in CH₃CN (~1 mL) was added to solid bnpapa (0.07 mmol). The resulting mixture was stirred for 20 min at room temperature. The solvent was then removed under reduced pressure and the residue was dissolved in CH₂Cl₂. Pentane diffusion yielded purple crystals suitable for X-ray crystallography. The crystals were crushed and dried under vacuum prior to elemental analysis. Yield: 54%. Anal. Calcd for C₃₀H₄₃Cl₂N₇NiO₈·1/2CH₂Cl₂: C, 45.69; H, 5.53; N, 12.23. Found: C, 45.32; H, 5.60; N, 12.09. UV-vis [CH₃CN, nm (ε, M⁻¹cm⁻¹)] 318(10300), 553(15), 931(17); FTIR (KBr, cm⁻¹) ~3400 (s, ν_NH), 1103 (ν_ClO4), 624 (ν_ClO4); FAB-MS [m/z (relative intensity)]: 617 ([M – ClO₄ – CH₃CN]⁺, 100%).

[(bnpapaNi)₂(μ-Cl)₂](ClO₄)₂ (19)

A solution of Ni(ClO₄)₂·6H₂O (0.14 mmol) in MeOH (~1 mL) was added to a solution of bnpapa (0.14 mmol) in MeOH (~1 mL). The resulting mixture was stirred for 20 min at room temperature. The solution was then transferred to a vial containing Me₄NCl·5H₂O (0.14 mmol) and the resulting mixture was stirred for 1 h. The solvent was then removed under reduced pressure. The remaining solid was dissolved in CH₂Cl₂ (~5 mL), and the solution was filtered through a celite/glass wool plug. Diffusion of diethyl ether into a CH₂Cl₂ solution of the compound yielded green crystals suitable for X-ray crystallography. The crystals were crushed and dried under vacuum prior to elemental analysis. Yield: 33%. Anal. Calcd for C₂₈H₄₀Cl₂N₆NiO₄: C, 51.40; H, 6.16; N, 12.84. Found: C, 51.15; H, 6.29; N, 12.88. FTIR (KBr, cm⁻¹) ~3340 (s, ν_NH), 1095 (ν_ClO4), 621 (ν_ClO4); UV-vis [CH₃CN, nm (ε, M⁻¹cm⁻¹)] 325(15900), 600(31), 1032(33); ESI/APCI-MS [m/z (relative intensity)]: 553.2348 ([M – 2ClO₄]²⁺, 100%).

O₂ Reactivity of 14: Product Isolation and ¹⁸O Labeling Studies

Complex 14 (0.101 mmol) was dissolved in 10 mL of acetonitrile and O₂ was bubbled through the solution for ~1 min. The reaction was then left stirring overnight at ambient temperature. Over the course of this time, the color changed from deep orange/red to pale yellow. CO production was verified by the PdCl₂ method.²⁴ The solvent was removed under reduced pressure and the resulting solid was dissolved in hexanes:ethyl acetate (4:1) and the solution was passed through a silica plug. This produced a yellow filtrate containing the organic products from the reaction. Ni(II) complexes that adhered to the silica gel were then eluted using acetonitrile, followed by methanol. As determined by TLC, ¹H NMR, and GC-MS, the organic fraction contained primarily benzoic acid (11 mg overall yield – 91% based on production of one equivalent of free benzoic acid per enolate ligand; sample is 88–90% benzoic acid based on ¹H NMR), with a small amount of benzil (PhC(O)C(O)Ph; 10–12%) and a trace amount of the ester PhC(O)OCH₂C(O)Ph present. The ester was identified via independent synthesis.²⁵ This compound is an isomer of 2-hydroxy-1,3-diphenylpropan-1,3-dione²³, and the reaction leading to ester formation will be discussed in detail elsewhere. In each reaction, two Ni(II) complexes were eluted from the column, [(bnpapa)Ni(O₂CPh)]ClO₄ (9) and [(bnpapa)Ni(PhC(O)C(O)CHC(O)Ph)]ClO₄ (15), the latter of which was present in a very low amount (<5%) and was identified via mass spectrometry (ESI-MS [m/z (relative intensity)]: 769.3376 (M – ClO₄)⁺, 100%)). The formation of 15 is related to the trace ester formation in the reaction and will be discussed in detail elsewhere. The yield of the metal complexes was ~88% by mass following the column (based on stoichiometric formation of [(bnpapa)Ni(O₂CPh)]ClO₄ (9)). Thus, from this data, the formation of [(bnpapa)Ni(O₂CPh)]ClO₄ (9) was judged as quantitative. Use of ¹⁸O₂ in the reaction of 14 with O₂ produces benzoic acid with 81% incorporation of one labeled oxygen atom. No ¹⁸O incorporation was found in benzil or the ester PhC(O)OCH₂C(O)Ph, as determined by GC-MS. Mass spectral analysis of the metal complexes indicated 87% ¹⁸O incorporation into one oxygen atom of the benzoate ligand of 9. No ¹⁸O incorporation was found in 15.

Reanalysis of O₂ reactivity of [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2)

Compound 2 was treated with an excess of O₂ in a reaction very similar to that outlined above using 14. An identical work-up procedure was also performed. The organic products generated were benzoic acid and benzil in a reproducible ~75:25 ratio. A trace amount of the ester PhC(O)OCH₂C(O)Ph is also generated, as determined by GC-MS and ¹H NMR. The total yield of organic products was ~96% based on the proposed formation of one equivalent of free benzoic acid per enolate ligand of 2. As previously reported, the Ni(II) complex, [(6-Ph₂TPA)Ni(O₂Ph)]ClO₄ (3) is produced in this reaction.¹⁶ A trace amount of the enolate complex [(6-Ph₂TPA)Ni(PhC(O)C(O)CHC(O)Ph)]ClO₄¹⁶ is also generated. The reaction leading to the formation this complex will be discussed in detail elsewhere. The total mass of complexes isolated suggests a ~70% yield based on the stoichiometric formation of 3. However, a control experiment using [(6-Ph₂TPA)Ni(O₂Ph)]ClO₄ (3) indicated that ~30% of the material is lost in the column purification process used to isolate the metal complexes from the reaction mixture. Thus, the formation of 3 is nearly quantitative in the O₂ reaction of 2. Use of ¹⁸O₂ in the reaction produced free benzoic acid with 59% ¹⁸O incorporation in one oxygen atom position, and the benzoate complex 3 having 64% incorporation in one oxygen atom. The level of isotope incorporation into 3 is slightly higher than that previously reported (~50%) based on multiple ¹⁸O reactions.¹⁶

¹⁸O₂ reactivity of [Me₄N][PhC(O)C(OH)C(O)Ph]

We have previously reported that treatment of the salt [Me₄N][PhC(O)C(OH)C(O)Ph] with O₂ results in the formation of tetramethylammonium benzoate, benzoic acid and CO.¹⁶ We have repeated this reaction using ¹⁸O₂ and have found by mass spectrometry that the benzoate/benzoic acid contains ~65% ¹⁸O incorporation when generated from Me₄NOH·5H₂O and PhC(O)C(OH)C(O)Ph in dry CH₃CN. Pre-drying of the Me₄NOH·5H₂O under vacuum for 24 h prior to use in the reaction produced a slightly higher level of ¹⁸O incorporation (72%) in the benzoate/benzoic acid product.

X-ray Crystallography

A single crystal of each compound 16, 17·1/4AHA (AHA = acetohydroxamic acid), 18·CH₂Cl₂, and 19·2CH₂Cl₂ was mounted on a glass fiber with traces of viscous oil and then transferred to a Nonius KappaCCD diffractometer equipped with Mo Kα radiation (λ = 0.71073 Å) for data collection. For unit cell determination, ten frames of data were collected at 103(1) K or 150(1) K with an oscillation range of 1 deg/frame and an exposure time of 20 sec/frame. Final cell constants were determined from a set of strong reflections from the actual data collection. All reflections were indexed, integrated, and corrected for Lorentz polarization and absorption effects using DENZO-SMN and SCALEPAC.²⁶ The structures were solved by a combination of direct and heavy-atom methods using SIR 97. All of the non-hydrogen atoms were refined with anisotropic displacement coefficients. Unless otherwise stated, all hydrogen atoms were assigned isotropic displacement coefficients U(H) = 1.2U(C) or 1.5U(C_methyl), and their coordinates were allowed to ride on their respective carbons using SHELXL97.²⁷ Complex 16 crystallizes in the space group P-1, with two carbon atoms of one neopentyl group exhibiting disorder over two positions. Complex 17·1/4AHA crystallizes in the space group C2/c with three oxygen atoms of the perchlorate anion and the carbon atoms of one nepentyl group exhibiting disorder. Complex 18·CH₂Cl₂ crystallizes in the space group P2₁/c. The CH₂Cl₂ solvate molecule exhibits disorder. Complex 19·2CH₂Cl₂ crystallizes in the space group P2₁/n, with one CH₂Cl₂ exhibiting disorder in the position of one chlorine atom.

Results

Preparation and Characterization of [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14)

Treatment of the bnpapa chelate ligand with equimolar amounts of Ni(ClO₄)₂·6H₂O, Me₄NOH·5H₂O, and 2-hydroxy-1,3-diphenylpropan-1,3-dione in CH₃CN, followed by work up using CH₂Cl₂, yielded a deep orange/red solid. Unfortunately, repeated attempts to produce crystals of this complex suitable for single crystal X-ray crystallography were unsuccessful. Instead, the orange/red solid was characterized by elemental analysis, UV-vis, FTIR, and ¹H NMR. The elemental analysis data is consistent with the proposed formulation. The π→π* transition of the coordinated enolate ligand of 14 is found at 393 nm (ε ~ 10,000 M⁻¹cm⁻¹). The wavelength and intensity of this absorption feature is generally similar to that reported for [(6-Ph₂TPA)Ni((PhC(O)C(OH)C(O)Ph)]ClO₄ (2; λ_max = 399 nm (ε ~ 6,800 M⁻¹cm⁻¹)).¹⁶ A broad absorption feature at ~3420 cm⁻¹ in the infrared spectrum of 14 is consistent with hydrogen bonding interactions involving the neopentylamine groups of the bnpapa chelate ligand. Based on structural studies of air stable analogs (vida infra), we propose that this hydrogen bonding involves an oxygen atom of the chelating enolate ligand that is positioned trans to the tertiary amine nitrogen donor of the chelate ligand.

Air Stable Analogs

To aid in the interpretation of the ¹H NMR spectroscopic features of 14, air stable analog complexes containing a bidentate acetoacetonato ([(bnpapa)Ni(CH₃C(O)CHC(O)CH₃)]ClO₄ (16)) or hydroxamato ([(bnpapa)Ni(CH₃C(O)NHO)]ClO₄ (17)) ligand were prepared and comprehensively characterized. For solution conductivity and spectroscopic studies, the perchlorate complex [(bnpapa)Ni(ClO₄)(CH₃CN)]ClO₄ (18) and chloride complex [(bnpapaNi)₂(μ-Cl)₂](ClO₄)₂ (19) were also prepared and characterized. Drawings of the cationic portions of these complexes are given in Scheme 3.

Scheme 3.

X-ray Crystallography

The cationic portions of 16 and 17 are shown in Figure 2, and those of 18 and 19 are shown in Figure 3. Details of the data collection and refinement are given in Table 1. Selected bond distances and angles are given in Table 2. The mononuclear 16 and 17 each contain a pseudo-octahedral Ni(II) center with the oxygen donor atoms of the chelate ligand positioned trans to N(3) and N(6). The respective Ni-O/N bonds of the primary coordination sphere are similar in these complexes, with the most noticeable difference in bond lengths being that the Ni-N interactions involving the neopentylamine-appended pyridyl donors are ~0.02 Å longer in 16 than those found in the hydroxamato complex (17). In terms of bond angles, a more acute O(1)-Ni(1)-O(2) angle in 17 (81.86(10)°) than that found in 16 (94.98(3)°) is due to the difference in chelate ring size of the bidentate O,O-donor ligands. In both complexes, an oxygen atom of the bidentate chelate ligand accepts two hydrogen bonds from the neopentyl amine groups of the supporting chelate ligand. These interactions are characterized by N(H)....O distances of 2.8–2.9 Å.

Figure 2.

Thermal ellipsoid drawings of the cationic portions of 16 and 17. All ellipsoids are drawn at the 50% probability level. Hydrogen atoms, other than the neopentyl amine N-H hydrogen atoms, and the hydroxyamato N-H hydrogen in 17, have been omitted for clarity.

Figure 3.

Thermal ellipsoid drawings of the cationic portions of 18 and 19. All ellipsoids are drawn at the 50% probability level. Hydrogen atoms other than the neopentyl amine N-H hydrogen atoms have been omitted for clarity.

Table 1.

Summary of X-ray data collection and refinement^a

	16	17·1/4AHA	18·CH2Cl2	19·2CH2Cl2
Empirical formula	C33H47ClN6NiO6	C30.50H45.25ClN7.25NiO6.5	C31H45Cl4N7NiO8	C58H84Cl8N12Ni2O8
Mr	717.93	711.25	844.25	1478.39
Crystal system	triclinic	monoclinic	monoclinic	monoclinic
Space group	P-1	C2/c	P21/c	P21/n
a/Å	8.9810(2)	32.3396(8)	8.8221(2)	10.9123(2)
b/Å	10.9647(2)	11.4649(3)	20.0037(4)	18.9857(3)
c/Å	18.2657(4)	19.4692(3)	21.7170(4)	16.9677(3)
α/°	102.6548(12)	90	90	90
β/°	96.3498(11)	93.1236(14)	91.7754(11)	96.6495(10)
γ/°	93.1764(13)	90	90	90
V/Å3	1738.34(6)	7207.9(3)	3830.66(14)	3491.68(10)
Z	2	8	4	2
Dc/Mg m−3	1.372	1.312	1.464	1.406
T/K	150(1)	103(1)	150(1)	150(1)
Color	purple	purple	purple	green
Crystal size/mm	0.38 × 0.30 × 0.28	0.20 × 0.20 × 0.15	0.38 × 0.38 × 0.33	0.38 × 0.30 × 0.25
μ/(mm−1)	0.687	0.644	0.842	0.904
2θ max(°)	55.00	55.04	54.96	54.96
Completeness to θ(%)	97.9	99.5	99.2	99.8
Reflections collected	12170	15128	14723	15105
Independent reflections	7825	8259	8717	7984
Rint	0.0219	0.0401	0.0255	0.0200
Variable parameters	577	486	469	422
R1/wR2b	0.0491/0.1289	0.0620/0.1452	0.0636/0.1571	0.0432/0.1038
Goodness-of-fit (F2)	1.043	1.040	1.042	1.081
Δρmax/min/e Å−3	0.758/−0.778	0.790/−0.626	0.783/−1.250	0.499/−0.530

^aDiffractometer: Nonius KappaCCD; Radiation used: Mo Kα (λ = 0.71073 Å).

^bR1 = Σ||F_o| − |F_c||/Σ ||F_o|; wR2 = [Σ[w(F_o²−F_c²)2]/[Σ(F_o²)²]]^1/2, where w = 1/[σ²(F_o²) + (aP)² + bP].

Table 2.

Selected Bond Distances (Å) and Angles (°)

	16	17·1/4AHA	18·CH2Cl2	19·2CH2Cl2
Ni(1)-O(1)	2.0162(18)	2.018(2)	2.177(2)
Ni(1)-O(2)	2.0183(17)	2.058(2)
Ni(1)-N(2)	2.1509(19)	2.137(3)	2.151(3)	2.1998(19)
Ni(1)-N(3)	2.0762(19)	2.076(3)	2.055(3)	2.0648(19)
Ni(1)-N(4)	2.154(2)	2.134(3)	2.164(3)	2.2267(19)
Ni(1)-N(6)	2.061(2)	2.048(3)	2.042(3)	2.038(2)
Ni(1)-N(7)			2.049(3)
Ni(1)-Cl(1)				2.4284(6)
Ni(1)-Cl(1)#1				2.4485(6)
O(1)-Ni(1)-O(2)	90.94(8)	81.86(10)
O(1)-Ni(1)-N(6)	172.86(8)	94.53(10)	90.23(12)
O(2)-Ni(1)-N(6)	95.76(8)	174.68(10)
O(1)-Ni(1)-N(3)	89.89(8)	178.78(10)	176.14(10)
O(2)-Ni(1)-N(3)	178.63(7)	99.35(11)
N(6)-Ni(1)-N(3)	83.46(8)	84.26(11)	85.97(13)	85.13(8)
O(1)-Ni(1)-N(2)	91.27(7)	99.44(11)	99.80(10)
O(2)-Ni(1)-N(2)	99.02(7)	89.69(10)
N(6)-Ni(1)-N(2)	90.13(7)	87.04(11)	88.72(11)	88.00(7)
N(3)-Ni(1)-N(2)	79.86(7)	80.67(11)	80.84(11)	79.27(7)
O(1)-Ni(1)-N(4)	88.86(8)	100.25(12)	98.42(10)
O(2)-Ni(1)-N(4)	99.99(8)	91.27(10)
N(6)-Ni(1)-N(4)	87.55(8)	93.24(11)	89.12(11)	90.16(7)
N(3)-Ni(1)-N(4)	81.12(8)	79.68(11)	80.84(11)	79.58(7)
N(2)-Ni(1)-N(4)	160.98(8)	160.22(11)	161.66(11)
N(6)-Ni(1)-N(7)			179.59(13)
N(7)-Ni(1)-N(3)			94.09(12)
N(7)-Ni(1)-N(2)			90.89(11)
N(7)-Ni(1)-N(4)			91.29(11)
N(7)-Ni(1)-O(1)			89.71(11)
N(6)-Ni(1)-Cl(1)				97.34(6)
N(3)-Ni(1)-Cl(1)				177.53(6)
N(2)-Ni(1)-Cl(1)				100.65(5)
N(4)-Ni(1)-Cl(1)				100.48(5)
N(6)-Ni(1)-Cl(1)#1				177.63(6)
N(3)-Ni(1)-Cl(1)#1				93.76(6)
N(2)-Ni(1)-Cl(1)#1				89.74(5)
N(4)-Ni(1)-Cl(1)#1				91.70(5)
Cl(1)-Ni(1)-Cl(1)#1				83.77(2)
Ni(1)-Cl(1)-Ni(1)#1				96.23(2)

Solution Conductivity Properties

The acetoacetonato complex 16 and the hydroxamato complex 17 both are 1:1 electrolytes in CH₃CN, as determined by variable concentration conductance measurements.²¹ Onsager plots for these complexes (Figure 4) exhibit slopes similar to the 1:1 standard Me₄NClO₄, and a notably different slope than that exhibited by [(bnpapa)Ni(ClO₄)(CH₃CN)](ClO₄)₂ (18 (Figure 3(top)) or [(bnpapaNi)₂(μ-Cl)₂](ClO₄)₂ (19, Figure 3(bottom)), both of which are 1:2 electrolyte species in CH₃CN.

Figure 4.

Onsager plots of the solution conductivity properties of 9, 16–19, and the 1:1 standard Me₄NClO₄ in CH₃CN at 22(1) °C.

¹H NMR Spectra

The ¹H NMR features of analytically pure [(bnpapa)Ni(CH₃C(O)CHC(O)CH₃)]ClO₄ (16) and [(bnpapa)Ni(ONHC(O)CH₃)]ClO₄ (17) in CD₃CN at 22(1) °C in the region of 20–80 ppm are shown in Figure 5(b) and (c). These are compared to the ¹H NMR spectral features of the O₂-sensitive enolate complex [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14, Figure 5(a)). The resonances in the 20–80 ppm region include those of pyridyl β-H ring protons and a portion of the benzylic proton resonances.²⁰

Figure 5.

¹H NMR spectral features of (a) 14, (b) 16, and (c) 17 in the region of 20–80 ppm. Spectra obtained in CD₃CN at 22(1) °C.

The ¹H NMR spectra of 14 and 16 (Figure 5(a) and (b)) are similar, both having a more complicated appearance than that of the hydroxamato compound 17. Based on the conductivity data presented above for the air stable 16 and 17, which are both 1:1 electrolyte species in acetonitrile, the greater complexity of the ¹H NMR spectrum of 16, relative to the spectral features of 17, is not due to the formation of multinuclear species in solution. ²H NMR studies of analogs of 14, 16, and 17 supported by a version of the bnpapa chelate ligand (Figure S1) having deuterium atoms at the benzylic positions, as well as partial deuteration of the β-H positions of the neopentylamine-appended pyridyl rings, indicated the presence of three unique benzylic proton resonances for 14 and 16 in the region of 165–175 ppm. This is consistent with no plane of symmetry within the cation.²⁰ However, 17 exhibits only one benzylic proton resonance in the same region, which is consistent with the presence of a plane of symmetry containing the unsubstituted pyridyl donor and the hydroxamato ligand. The factors that govern the subtle differences in solution structure of the cationic portions of 14, 16, and 17 are not entirely clear, but appear to relate to the chelate ring size of the O,O-donor ligand coordinated to the Ni(II) center. Specifically, the compounds having a six-membered chelate O,O donor lack a plane of symmetry in the solution form of the cation. We note that the 1:2 electrolyte compounds [(bnpapa)Ni(ClO₄)(CH₃CN)](ClO₄)₂ (18) or [(bnpapaNi)₂(μ-Cl)₂](ClO₄)₂ (19) both exhibit ¹H NMR spectra similar to 17 wherein an effective plane of symmetry renders the two neopentylamine-appended pyridyl donors equivalent in solution (Figure S2). Ni(II) carboxylate complexes of the general formula [(bnpapa)Ni(O₂CR)]ClO₄ (R = Ph, (CH₂)₂SCH₃, and H) also exhibit ¹H NMR spectra that are generally similar to the spectrum found for the hydroxamato complex 17.¹⁷ In terms of comparing chelate ligand environments, the mononuclear Ni(II) complexes [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2) and [(6-Ph₂TPA)Ni(RC(O)CHC(O)R)]ClO₄ (R = Ph, CH₃) exhibit spectra which indicate the presence of a plane of symmetry within the cation.^15,16 Thus, the bnpapa-ligated Ni(II) six-membered ring enolate complexes 14 and 16 exhibit unique solution ¹H NMR properties relative to their hydrophobic analogs.

Based on the structural and solution features of the air stable acetoacetonateo complex 16, and the similarity of the ¹H NMR spectrum of this complex to the spectrum of [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14), we propose that the O₂-reactive enolate complex 14 has a structure that is similar to the analog supported by 6-Ph₂TPA (Figure 6), but is distorted such that no plane of symmetry relates the neopentylamine pyridyl donors in the solution form of the cation.

Figure 6.

(a) Proposed structure of [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14). (b) Structure of [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄(2) as determined by X-ray crystallography.¹⁵

Reaction of 14 with O₂

Exposure of a CH₃CN solution of 14 to O₂ results in an immediate color change from deep orange/red to yellow. CO production in this reaction was identified qualitatively using the PdCl₂ method.²⁴ The inorganic and organic products of the reaction were separated by column chromatography and identified (Scheme 4). One equivalent of benzoic acid is generated along with a small amount of benzil (PhC(O)C(O)Ph; ratio of benzoic acid to benzil is ~90:10) and a trace amount of the ester PhC(O)OCH₂C(O)Ph, the latter of which was identified via independent synthesis.²⁵ The reaction leading to ester formation will be described elsewhere. The monobenzoate complex [(bnpapa)Ni(O₂CPh)]ClO₄ (9) is produced in nearly quantitative yield in the reaction, with a trace amount of a second Ni(II) complex, [(bnpapa)Ni(PhC(O)C(O)CHC(O)Ph]ClO₄ (15), being identified by mass spectrometry. The production of 15 is related to the presence of the ester PhC(O)OCH₂C(O)Ph in the reaction and will be discussed elsewhere. Use of ¹⁸O₂ in the reaction with 14 produces benzoic acid with 81% ¹⁸O incorporation in one oxygen atom position. No ¹⁸O incorporation was found in benzil or the ester PhC(O)OCH_sC(O)Ph, as determined by GC-MS. Mass spectral analysis of the Ni(II) products indicated that the benzoate complex 9 had 87% ¹⁸O incorporation in one oxygen atom of the benzoate ligand. No ¹⁸O incorporation was found in 15. We note that the level of ¹⁸O incorporation for the reaction involving 14 is slightly higher than that reported for the Ni(II)-ARD enzymatic reaction (77.5%).²⁸

Scheme 4.

Major products identified in reaction of 14 and 2 with O₂; results of reactions performed using ¹⁸O₂.

The work-up procedure used to isolate the organic and inorganic products of the reaction of 14 with O₂ is slightly different in terms of solvents used from the procedure that we previously reported for identification of the products of the reaction of [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2) with O₂.¹⁶ The new conditions are more amenable to the isolation of free benzoic acid in the reaction mixture. Therefore, the reaction involving 2 was repeated under conditions identical with those employed for [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14), with the results shown in Scheme 4. CO production was again identified qualitatively using the PdCl₂ method.²⁴ In this reaction a ~75:25 mixture of benzoic acid and benzil is produced, along with a trace amount of the ester PhC(O)OCH₂C(O)Ph. The Ni(II)-containing products are [(6-Ph₂TPA)Ni(O₂CPh)]ClO₄ (3) and a trace amount of [(6-Ph₂TPA)Ni(PhC(O)C(O)CHC(O)Ph)]ClO₄ (20).^16,17 Use of ¹⁸O₂ in the reaction with 2 yielded benzoic acid with 59% ¹⁸O incorporation in one oxygen atom position, and the benzoate complex 3 having 64% incorporation in one oxygen atom. We note that the level of isotope incorporation in 3 is slightly higher that than previously reported (~50%), but is similar to that found for the O₂ reaction of the salt [Me₄N][PhC(O)C(OH)C(O)Ph], which decomposes upon reaction with ¹⁸O₂ to give [Me₄N][O₂CPh] and benzoic acid (~65% incorporation of one ¹⁸O into carboxylate group) and CO. Thus, the reaction involving 2 gives a mixture of organic products (benzoic acid and benzil), but with a similar level of ¹⁸O incorporation to that found for the O₂ reaction of [Me₄N][PhC(O)C(OH)C(O)Ph].

Overall, the studies outlined above revealed two interesting differences in the O₂ reactivity of [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14) and [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2). First, the O₂ reaction involving the bnpapa-ligated complex 14 produces less of the side product benzil, Second, the reaction involving 14 has a higher level of ¹⁸O incorporation in the benzoic acid/benzoate products. These combined results provide evidence that the chelate ligand environment is influencing the aliphatic carbon-carbon bond cleavage reaction pathway.

Comparison of the Secondary Environments of bnpapa- and 6-Ph₂TPA-ligated Ni(II) Complexes

To date, we have structurally characterized four bnpapa- and thirteen 6-Ph₂TPA-ligated mononuclear Ni(II) complexes. Shown in Tables 3 and 4 are bond distances and angles involving the neopentyl-appended pyridyl donors in the bnpapa complexes and the phenyl-appended pyridyl donors in the 6-Ph₂TPA derivatives. Several conclusions can be drawn from comparison of the microenvironments in these two families of complexes. First, the neopentylamine amine-appended pyridyl donors in the bnpapa ligand, which have a substituent that is more electron donating and less sterically demanding relative to the phenyl substitutents of the 6-Ph₂TPA ligand, have slightly shorter Ni-N distances (overall average of bnpapa complexes 2.14 Å; 6-Ph₂TPA complexes 2.22 Å). There is also a more linear N-Ni-N bond angle (overall average of bnpapa complexes 161°; 6-Ph₂TPA complexes 154°) between the substituted pyridyl groups. Second, the phenyl-appended pyridyl donors of 6-Ph₂TPA exhibit more variation in terms of Ni-N distances (~0.12 Å) in response to the presence of different exogenous ligands. This is not surprising in that aryl-appended pyridyl donors of chelate ligands have been shown to fully dissociate from a metal center in response to exogenous ligand coordination.²⁹ Third, the microenvironment of bnpapa-ligated complexes is more compact, as determined by comparison of the distance between the two neopentylamine nitrogen atoms (~5.6 Å) versus that of the ipso carbons of the phenyl appendages of 6-Ph₂TPA (~6.4 Å).

Table 3.

Structural properties of bnpapa-ligated mononuclear Ni(II) complexes

Complex	N(2)-Ni-N(4) (°)a	Ni-N(2), Ni-N(4)a (Å)	N(H)…N(H) (Å)	Reference
[(bnpapa)Ni(ClO4)(CH3CN)]ClO4	161.66(11)	2.151(3), 2.164(3)	5.7	this work
[(bnpapa)Ni(ONHC(O)CH3)]ClO4	160.22(11)	2.137(3), 2.134(3)	5.5	this work
[(bnpapa)Ni(CH3C(O)CHC(O)CH3)]ClO4	160.98(8)	2.1509(19), 2.154(2)	5.6	this work
[(bnpapa)Ni(O2Ph)]ClO4a	161.78(12)	2.125(3), 2.136(3)	5.6	17

^aN(2) and N(4) are the neopentylamine-appended pyridyl donors of the bnpapa ligand.

^bData reported for one of two crystalline forms of the complex that were previously reported.

Table 4.

Structural properties of 6-Ph₂TPA-ligated mononuclear Ni(II) complexes

Complex	N(3)-Ni-N(4) (°)a	Ni-N(3), Ni-N(4) (Å)a	C(Ph)..C(Ph)b (Å)	Reference
[(6-Ph2TPA)Ni(CH3CN)(CH3OH)](ClO4)2	158.59(16)	2.218(5), 2.200(5)	6.3	20
[(6-Ph2TPA)Ni(ONHC(O)CH3)]ClO4	147.19(7)	2.229(2), 2.263(2)	6.4	20
[(6-Ph2TPA)Ni(CH3C(O)CHC(O)CH3)]ClO4	150.41(8)	2.217(2), 2.221(2)	6.4	16
[(6-Ph2TPA)Ni(PhC(O)C(O)CHC(O)Ph)]ClO4	158.40(11)	2.209(3), 2.333(3)	6.5	16
[(6-Ph2TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO4	157.04(8)	2.271(2), 2.323(2)	6.6	15
[(6-Ph2TPA)Ni-Cl(CH3CN)]ClO4	157.67(10)	2.081(3), 2.213(2)	6.4	20
[(6-Ph2TPA)Ni(O2CPh)]ClO4	149.83(6)	2.1458(16), 2.234(16)	6.2	16
[(6-Ph2TPA)Ni(O2CPh)(H2O)]ClO4	158.28(10)	2.216(2), 2.241(2)	6.4	17
[(6-Ph2TPA)Ni(O2C(CH2)2SCH3)]ClO4	141.57(10)	2.233(3), 2.193(3)	6.3	17
[6-Ph2TPA)Ni(O2(CH2)2SCH3)(H2O)]ClO4	159.33(12)	2.240(3), 2.186(3)	6.3	17
[(6-Ph2TPA)Ni(O2CCH2SCH3)]ClO4	148.73(8)	2.198(2), 2.165(2)	6.3	17
[(6-Ph2TPA)Ni(O2CCH2SCH3)(H2O)]ClO4	150.97(7)	2.1884(17), 2.2050(17)	6.4	17
[(6-Ph2TPA)Ni(O2CH)(H2O)]ClO4	158.42(11)	2.194(3), 2.218(3)	6.3	17

^aN(3) and N(4) are the phenyl-appended pyridyl donors of the 6-Ph₂TPA ligand.

^bIpso carbon atoms of phenyl appendages.

Discussion

As shown in Scheme 1, a key issue in the chemistry of acireductone dioxygenases is how the coordination mode of the acireductone substrate may influence the reaction with O₂. Pochapsky and coworkers have suggested, on the basis of NMR experiments, that differences in the secondary environment in Ni(II)- versus Fe(II)-containing ARD enzyme may be responsible for inducing different coordination modes of the acireductone substrate.¹ Specifically, in Ni(II)-ARD the side chain of the tryptophan residue is positioned such that the binding pocket for the acireductone is smaller than that found in Fe(II)-ARD. This is proposed to induce the formation of a six-membered chelate ring structure, which leads to 1,3-dioxygenolytic bond cleavage and CO production. We note that in both Ni(II)- and Fe(II)-containing acireductone dioxygenases an arginine residue is positioned within the active site, likely within hydrogen-bonding distance od the coordinated acireductone.¹

As an approach toward evaluating the influence of the secondary environment on the chemistry of Ni(II) complexes having an acireductone-type ligand, we have prepared and characterized [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2), which has a hydrophobic microenvironment, and [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14), which contains a hydrogen-bond donor secondary environment. Based on the results of analytical and spectroscopic studies, we propose that the coordination mode of the acireductone-type ligand in these complexes is similar, with each having a six-membered chelate ring. However, the solutions structures of 2 and 14 differ in terms of the presence of a plane of symmetry that relates the modified pyridyl donors, with the hydrogen bond donor pocket lacking such symmetry. Treatment of each complex with O₂ results in CO production and aliphatic carbon-carbon bond cleavage, with the major products in each reaction being a mononuclear Ni(II) benzoate complex, benzoic acid, and CO. In both reactions, the formation of the side product benzil was also identified. The yield of this byproduct is higher in the reaction involving [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2). Notably, this reaction exhibits a level of ¹⁸O incorporation in the benzoate/benzoic acid products that is ~20% lower than the reaction involving 14. A possible rationale for the differences in product distribution and ¹⁸O incorporation between the reactions involving 2 and 14 could be that differing pathways for carbon-carbon bond cleavage may be operative as outlined in Scheme 5. Pathway A involves a cyclic peroxide resulting from two-electron oxidation of the coordinated acireductone analog. Carbon-carbon bond cleavage in this type of structure would produce two equivalents of benzoate/benzoic acid, with each product having one labeled oxygen atom if the reaction were performed using ¹⁸O₂. The production of benzoic acid in high yield in the reaction involving [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14), and the observed high level of ¹⁸O incorporation, suggests that this pathway could possibly be operative for this reaction. However, the production of a small amount of benzil in the reaction involving 14 suggests that another type of reactivity is also occurring. For both complexes, we propose that benzil is generated via a pathway B type reaction sequence (Scheme 5). Specifically, two electron oxidation of the coordinated acireductone type ligand could produce 1,3-diphenyltriketone and a solvated Ni(II) complex of the chelate ligand. It has been previously shown that 1,3-diphenyltriketone will undergo phenyl or benzoyl migration in the presence of a Lewis acids such as AlCl₃ to give benzil and carbon monoxide.³⁰ If this type of reaction pathway is occurring with the complexes described herein, the secondary environment of the [(bnpapa)Ni(sol)₂]²⁺ and [(6-Ph₂TPA)Ni(sol)₂]²⁺ cations (sol = solvent; e.g. CH₃CN) could influence interactions between the Ni(II) center and the 1,3-diphenyltriketone. For example, the larger pocket for the 6-Ph₂TPA-ligated Ni(II) center could be a key factor in promoting Lewis acid activation of a carbonyl moiety of the triketone, which is necessary for the migration reaction leading to benzil production.³⁰ This is a possible rationale for the higher amount of benzil produced in the reaction involving 2. In terms of the lower level of ¹⁸O incorporation found this reaction, triketones are known to hydrate easily at the central carbonyl, and this may be a pathway for the introduction of unlabeled oxygen atoms into the benzoate/benzoic acid products.³¹ The observation that the salt [Me₄N][PhC(O)CH(OH)C(O)Ph] undergoes reaction with ¹⁸O₂ to produce a similar level of ¹⁸O incorporation (~65%) in the benzoic acid product as is found for the reaction involving 2 suggests that both may proceed via a similar reaction pathway. ¹⁸O incorporation into the carbon-carbon bond cleavage products could occur via reaction of the free triketone with hydroperoxide to form a five-membered cyclic peroxide species from which carbon-carbon bond cleavage could occur. Pochapsky has previously shown that treatment of 2,3,4-pentanetrione with H₂O₂ results in the formation of CO and two equivalents acetic acid.³² No isotope labeling studies were reported for this reaction.

Scheme 5.

Possible reaction pathways for oxidative carbon-carbon bond cleavage in the Ni(II) enolate complexes 2 and 14. The supporting chelate ligands have been omitted for clarity.

The higher level of ¹⁸O incorporation in the benzoate/benzoic acid products of the reaction involving 14 is interesting. Based on the results of our current experiments, it is unclear whether exclusively pathway B chemistry is occurring in the O₂ reactions of 14 and 2, or whether a portion of the labeled product is generated via pathway A type reactivity. If exclusively pathway B chemistry is operative, an important distinction between the two supporting chelate ligand systems is that the bnpapa-ligated Ni(II) center may coordinate the hydroperoxide anion within the hydrogen bond donor pocket. The bnpapa chelate ligand has been previously used to stabilize mononuclear Zn(II) and Cu(II) hydroperoxide complexes.^33,34 To date, similar complexes have not been reported using the 6-Ph₂TPA ligand. Thus, differences in the anion coordination properties between the bnpapa and 6-Ph₂TPA-supported complexes could contribute to the observed differences in reactivity between 2 and 14.

The results of this investigation have enabled the formulation of hypotheses regarding the reaction pathways of 2 and 14 with O₂ (Scheme 5) that can now be tested in individual control reactions. For example, we are now performing experiments to examine the reactivity of [(6-Ph₂TPA)Ni(CH₃CN)(CH₃OH)](ClO₄)₂³⁵ and [(bnpapa)Ni(ClO₄)(CH₃CN)]ClO₄ (18) with 1,3-diphenyltriketone and H₂O₂ to evaluate the efficacy of pathway B chemistry leading to the observed products. Additionally, we are pursuing kinetic studies of the reactions of complexes 2 and 14 with O₂, as well as examining the reactivity of [Me₄N][PhC(O)C(OH)C(O)Ph] with O₂ via computational methods to assess its reactivity relative to that found in the native C(1)-H acireductone substrate processed by the acireductone dioxygenases.³⁶

Conclusions

Using the hydrogen-bond donor ligand bnpapa, a mononuclear Ni(II) complex of the formula [(bnpapa)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (14) has been isolated and characterized. Based on comparison of the spectroscopic features of this complex to those of air stable analogs, the structure of 14 is proposed to be generally similar to that previously reported for [(6-Ph₂TPA)Ni(PhC(O)C(OH)C(O)Ph)]ClO₄ (2). Notably, complexes 2 and 14 exhibit differences in their O₂-dependent aliphatic carbon-carbon bond cleavage reactivity in terms of the product mixture generated and the level of ¹⁸O incorporation in the benzoate/benzoic acid products. Two possible reaction pathways leading to ¹⁸O incorporation into the benzoate/benzoic acid products are proposed, with one pathway involving the formation of a triketone and hydroperoxide anion. Comparison of the chemical features of several bnpapa- and 6-Ph₂TPA-ligated complexes suggests that differences in the pocket size and anion binding properties for these two ligand types may be key factors in producing the observed differences in reaction products.