Back to Table of Contents

45941-AC3
Harnessing Metal-Ligand Cooperative Redox Reactivity for Bond Functionalization Strategies

Alan F. Heyduk, University of California (Irvine)

           The primary focus of this project is to introduce redox reactivity to electrophilic early transition metal complexes through the use of redox-active ligands. Our previous work has focused on bidentate ortho-amidophenolate and ortho-phenylenediamide ligands coordinated to Group IV metals. In short, our resulted showed that the catechol/semi-quinone/quinone properties of these ligand platforms did in fact enable redox-based bond-activation reactions like oxidative addition and reductive elimination to occur at d⁰ metal centers; however, complications always arose due to dissociation of the redox-active ligands from the metal coordination sphere. To overcome this ligand dissociation problem, in the work reported here we have turned to a tridentate amido bis(phenolate) ligand, [ONO^red]H₃, prepared according to Scheme 1. The three oxidation states of the [ONO] ligand are shown in Scheme 2.

Scheme 1.

Scheme 2.

           We have successfully metallated the [ONO^red]H₃ ligand platform with zirconium and have prepared both zirconium alkyl and zirconium halide complexes that will be used in future reactivity studies. The most straight-forward approach towards the synthesis of a low-coordinate zirconium complex was to treat [ONO^red]H₃ with tetra(alkyl) zirconium(IV) synthons. According to Scheme 3, the dimeric complex, {[ONO^red]Zr(CH₂SiMe₃)]₂ (1), was prepared in modest yields in this manner. The structure of 1 was confirmed by NMR spectroscopy and X-ray crystallography. The surprising feature of the structure is the bridging coordination mode for the chelating [ONO^red]^3– ligands. The Zr–N bond distances are long at 2.31 and 2.50 �, consistent with each amide acting as a donor to two different zirconium centers. The preparation and characterization of 1 as a dimer was an important result because it highlights the difficulty associated with targeting low-coordinate zirconium complexes with the [ONO^red]^3– ligand. As such we began to target complexes that were five- or six-coordinate, but with weakly coordinating solvent molecules that could be removed under mild conditions.

Scheme 3.

           Both mono- and dihalide complexes of zirconium could be prepared through simple methesis reactions. According to Scheme 4, various bases may be used to prepare the dichloride, [ON(H)O]ZrCl₂(solv) (solv = Et₂O, THF, py) (2). Complex 2 is a stable intermediate which can be isolated and stored for long periods of time. The stability of this species derives from the protonation state of the redox-active ligand. The amide nitrogen remains protonated, in effect killing the redox-activity of the ligand; however, it the redox-activity is readily unmasked by addition of another equivalent of strong base such as nBuLi. Alternatively, we have recently been able to prepare directly mono-halide complexes of zirconium by the direct deprotonation of [ONO^red]H₃ with three equivalents of nBuLi and reaction with the appropriate zirconium halide salt in the presence of THF (Scheme 5). Figure 1 shows a the results of an X-ray diffraction experiment on [ONO^red]ZrBr(THF)₂ (3). As expected, the [ONO^red]^3– ligand of 3 occupies three meridional coordination sites on the octahedral complex with the bromide ligand trans to the amide donor. Overall the complex has approximate C_2v symmetry.

Scheme 4.

Scheme 5.

Figure 1. Molecular structure of [ONO^red]ZrBr(THF)₂ (3).            The zirconium monohalide complexes are suitable precursors for the preparation of zirconium alkyl species. As shown in Scheme 6, the reaction of [ONO^red]ZrCl(THF)₂ with Grignard reagents RMgCl (R = CH₂Si(CH₃)₃, CH₂Ph) afforded the expected alkyl derivatives. While the preparation of the THF-solvated alkylzirconium complexes was straight-forward, the purification of these species was hampered by high solubilities. Conversion to the pyridine adducts allowed for purification and characterization by crystallography. The surprising result, shown in Figure 2, is that [ONO^red]Zr(CH₂SiMe₃)(py)₃ (4) adopts a seven-coordinate geometry thanks to the coordination of three pyridine molecules.

Scheme 6.

Figure 2. Molecular structure of [ONO^red]Zr(CH₂SiMe₃)(py)₃ (4).            With the successful preparation of monohalide and monoalkyl zirconium complexes of the [ONO^red]^3– ligand, we have begun examining small-molecule redox reactivity. The monochloride complex, [ONO^red]ZrCl(THF)₂ catalyzes the disproportionation of diphenylhydrazine into aniline and azobenzene (Scheme 7). Control reactions with Lewis acid complexes do not lead to disproportionation indicating the importance of the redox-active ligand in this reaction. Current efforts are aimed at elucidating the mechanism of this reaction, which we believe proceeds through a zirconium imido intermediate, [ONO^q]ZrCl(=NPh). Such an intermediate is only possible if the [ONO] ligand is acting as a two-electron reservoir during the reaction. Further efforts are aimed at diverting this putative imido intermediate for catalytic nitrene transfer reactions.

Scheme 7.