Open Access Dissertation
Date of Award
The novel alkoxide ligand [OCtBu2Ph], or [OR], was synthesized in a single step as a lithium salt. It was then reacted with a series of first-row transition metal(II) halides, with widely varying results. Upon reaction with chromium, manganese, iron, or cobalt(II) chloride, dimeric complexes of the form M2(OR)4Li2Cl2 were formed, which displayed rare seesaw geometry at the metal. This unusual geometry was confirmed by various spectroscopic and computational studies. Computational studies also indicate that the steric bulk of the ligand, as well as the inclusion of lithium atoms in the molecules, are what lead to the seesaw geometry. Reaction of [OR] with nickel(II) halides generates monomeric species of the form Ni(OR)2XLi(THF)2 (X = Cl, Br), which display distorted trigonal planar geometry at three-coordinate nickel. Dimerization likely does not occur for nickel due to its smaller size. DFT studies support preference for nickel to form the monomer. Reaction of [OR] with copper(II) halides leads to reduction of the copper center by one electron, generating the tetramer Cu4(OR)4. Reduction of copper(II) by an alkoxide is a novel transformation. Spectroscopic studies to probe the mechanism suggest that Cu(OR)2XLi(THF)2 may be an intermediate prior to reduction. Observation by NMR of the ketone Ph(C=O)tBu and ROH suggest that alkoxide reduces the copper to give an alkoxide radical, which then decomposes via β-scission.
To form the desired bis(alkoxide) system, the halide-containing alkoxide complexes were reacted with thallium(I) hexafluorophosphate. For manganese, iron, and copper, complexes of the form M(OR)2(THF)2 were isolated. The bis(alkoxide) complexes display distorted tetrahedral geometry at the metal, with large RO−M−OR angles. Cyclic voltammetry of these species show that the iron bis(alkoxide) is the most easily reduced of the three. Attempts to form the chromium bis(alkoxide) in a similar fashion led to the reduction of thallium(I) to thallium metal. Formation of the nickel bis(alkoxide) complex was also unsuccessful.
Reaction of the iron bis(alkoxide) complex with adamantyl azide led to reductive coupling of the azide moieties to give the bridging hexazene complex (RO)2Fe(μ-η2:η2-AdN6Ad)Fe(OR)2. This complex was confirmed to be stable to explosive decomposition. Computational studies suggest a dimerization mechanism, whereby azide initially coordinates to iron(II), and upon dimerization the iron centers reduce the azides before N−N bond formation occurs.
Stoichiometric reaction of the iron bis(alkoxide) complex with mesityl azide leads to nitrene formation followed by nitrene coupling to give the azoarene MesNNMes. Crystallization from a stoichiometric reaction afforded the azoarene and the iron tris(alkoxide) Fe(OR)3. Catalytic azoarene formation with catalyst loading as low as 1 mol% cleanly generates azoarene at room temperature within a day or two for mesityl azide and 2,6-diethylphenyl azide. Asymmetric azoarene can also be formed by reacting mesityl azide and 2,6-diethylpheynyl azides together with the iron bis(alkoxide). This process is highly selective: azoarene is the only product formed, even if the reaction is performed in the presence of cyclohexadiene or isocyanide. Other azides (i.e. those not possessing groups in positions ortho to the azide on the ring) do not form azoarene upon reaction with the iron bis(alkoxide): instead they form bis(imido) complexes of the form (RO)Fe(THF)(μ-NAr)2Fe(THF)(OR) (Ar = 4-(trifluoro)methyl, 3,5-dimethylphenyl, or phenyl). These are stable molecules that do not react with additional equivalents of aryl azide. A tentative mechanism is proposed, whereby the iron(III) imido radical Fe(OR)2(NAr) comproportionates with another equivalent of iron bis(alkoxide) to give the observed iron(III) tris(alkoxide), and the iron(III) bridging imido dimer. Nitrene coupling to give azoarene can occur from either the iron(III) imido radical complex, or from the bis(imido) dimer. For bulkier aryl azides, the bis(imido) dimer is disfavored to form due to sterics. This likely leads to preferential azoarene formation.
Reaction of the iron bis(alkoxide) with diphenyldiazomethane generates the azine Ph2CNNCPh2, suggesting that a reactive iron carbene complex is formed before reacting with another equivalent of the diazoalkane. A similar reaction with the cobalt bis(alkoxide) leads to isolation of the stable carbene complex, Co(OR)2(CPh2). This molecule is the first example of an isolated, structurally characterized high-valent cobalt carbene complex. EPR and DFT studies confirm the electronics of the complex: it is likely that the cobalt oxidation state lies on a continuum between cobalt(III) and cobalt(IV) but that it displays significant alkylidene character. However, the high-valent cobalt carbene complex is surprisingly unreactive. Combination with both styrene and methyl acrylate at elevated temperatures failed to lead to any significant reactivity of the complex. This is likely due to the steric protection offered by both the bulky alkoxide ligands and the bulky carbene.
Bellow, James, "Design Of First-Row Transition Metal Bis(alkoxide) Complexes And Their Reactivity Toward Nitrene And Carbene Transfer" (2016). Wayne State University Dissertations. 1517.