UC Berkeley

The work herein describes progress towards using metal–organic frameworks as scaffolds for stabilizing metal–oxo and dioxygen species, and their application in hydrocarbon oxidation catalysis and O2/N2 separations. Metal–organic frameworks are a class of highly porous and functionally versatile crystalline solids consisting of inorganic cations or clusters bridged by organic linkers. They are attractive as solid supports for metal–oxo and dioxygen chemistry for many reasons, including the presence of well-defined, site-isolated metal centers with highly tunable local and outer coordination spheres.

Chapter 1 provides an introduction to the electronic structure, reactivity, and biological relevance of metal–oxo and dioxygen species, with a particular emphasis on iron and cobalt. In addition, a brief historical overview of the development of biomimetic iron–oxo, iron–dioxygen, and cobalt–dioxygen chemistry, with selected molecular and heterogenous examples, is provided. The chapter concludes with a summary of the methods currently used to install coordinatively-unsaturated redox-active metal sites into metal–organic frameworks. A perspective on the potential of metal–organic frameworks in metal–oxo and dioxygen chemistry is given.

Chapter 2 describes an initial foray into metal–organic framework-supported iron–oxo chemistry. Specifically, the nitrous oxide activation and hydrocarbon oxidation reactivity of the coordinatively-unsaturated iron(II) sites in the metal-organic frameworks Fe2(dobdc) and Fe0.1Mg1.9(dobdc) (dobdc4– = 2,5-dioxido-1,4-benzenedicarboxylate) is detailed. In the presence of N2O, the latter framework is able to selectively and catalytically convert ethane to ethanol upon mild heating. Structural and spectroscopic characterization of the initial iron(II)–N2O adduct and an iron(III)-hydroxide decay product, reactivity studies, and detailed electronic structure calculations strongly suggest that the active oxidant in this system is a high-spin, S = 2 iron(IV)–oxo. This rare electronic structure is a direct result of the weak ligand field imparted by the dobdc4– ligand.

In addition to primary coordination sphere properties such as a weak ligand field, longer-range pore-environment effects could also become a powerful parameter in metal–organic framework catalyst design. Chapter 3 explores this idea in the context of solution-phase cyclohexane oxidation in the biphenyl and terphenyl expanded Fe2(dobdc) derivatives. A three-fold enhancement of the alcohol:ketone (A:K) ratio and an order of magnitude increase in turnover number is observed by simply altering the framework pore diameter and installing nonpolar, hydrophobic functional groups near the iron center. The increase in A:K selectivity is attributed to an increased affinity of the pore walls for cyclohexane, which may help increase its local concentration near the iron site.

Chapter 4 departs from iron–oxo chemistry and oxidation catalysis, focusing instead on cobalt–dioxygen binding for O2/N2 separation applications. Specifically, this chapter details the reversible O2 binding properties of Co-BTTri (H3BTTri = 1,3,5-tri(1H-1,2,3-triazol-5-yl)benzene), a sodalite-type framework containing coordinatively-unsaturated cobalt(II). It was found that the O2 binding affinity could be tuned by altering the local ligand field. Electronic structure calculations reveal the extent of electron transfer from cobalt to O2 in these systems is highly dependent on the local environment, and can vary between 0.2 to 0.7 electrons.

Chapter 5 combines aspects of both Chapter 1 and 4, focusing on the development of new iron(II)-based frameworks with tunable primary coordination spheres, and the effect of ligand field on reactivity and gas adsorption properties. The synthesis and characterization of two new sodalite-type frameworks, Fe-BTTri and Fe-BTP (H3BTP = 1,3,5-tri(1H-pyrazol-4-yl)benzene), is reported. Interesting O2 and CO gas adsorption properties are displayed by bulk Fe-BTTri, and are briefly described in this chapter. In addition, initial N2O/ethane oxidation studies performed on Fe-BTT suggest defect sites are responsible for the reactivity seen in this material, highlighting the inhomogeneity present even in highly crystalline metal–organic frameworks.