Metallaboratrane Facilitated E‒H Bond Activation and Hydrogenation Catalysis

The E‒H bond activation chemistry of tris-phosophino-iron and -cobalt metallaboratranes is discussed. The ferraboratrane complex (<b>TPBb>)Fe(Nb>2b>) heterolytically activates H‒H and the C‒H bonds of formaldehyde and arylacetylenes across an Fe‒B bond. In particular, H‒H bond cleavage at (<b>TPBb>)Fe(Nb>2b>) is reversible and affords the iron-hydride-borohydride complex (<b>TPBb>)(μ‒H)Fe(L)(H) (L = Hb>2b>, Nb>2b>). (<b>TPBb>)(μ‒H)Fe(L)(H) and (<b>TPBb>)Fe(Nb>2b>) are competent olefin and arylacetylene hydrogenation catalysts. Stoichiometric studies indicate that the B‒H unit is capable of acting as a hydride shuttle in the hydrogenation of olefin and arylacetylene substrates. The heterolytic cleavage of Hb>2b> by the (<b>TPBb>)Fe system is distinct from the previously reported (<b>TPBb>)Co(Hb>2b>) complex, where Hb>2b> coordinates as a non-classical Hb>2b> adduct based on X-ray, spectroscopic, and reactivity data. The non-classical Hb>2b> ligand in (<b>TPBb>)Co(Hb>2b>) is confirmed in this work by single crystal neutron diffraction, which unequivocally shows an intact H‒H bond of 0.83 Å in the solid state. The neutron structure also shows that the Hb>2b> ligand is localized at two orientations on cobalt trans to the boron. This localization in the solid state contrasts with the results from ENDOR spectroscopy that show that the Hb>2b> ligand freely rotates about the Co‒Hb>2b> axis in frozen solution. Finally, the (<b>TPBb>)Fe system, as well as related tris-phosphino-iron complexes that contain a different apical ligand unit (Si, PhB, C, and N) in place of the boron in (<b>TPBb>)Fe, were studied for COb>2b> hydrogenation chemistry. The (<b>TPBb>)Fe system is not catalytically competent, while the silicon, borate, carbon variants, (<b>SiP^Rb>3b>b>)Fe, (<b>PhBP^iPrb>3b>b>)Fe, and (<b>CP^iPrb>3b>b>)Fe, respectively, are catalysts for the hydrogenation of COb>2b> to formate and methylformate. The hydricity of the COb>2b> reactive species in the silatrane system (<b>SiP^iPrb>3b>b>)Fe(Nb>2b>)(H) has been experimentally estimated.

The Design and Synthesis of Transition Metal Complexes Supported by Non-innocent Ligand Scaffolds for Small Molecule Activation

This dissertation focuses on the incorporation of non-innocent or multifunctional moieties into different ligand scaffolds to support one or multiple metal centers in close proximity. Chapter 2 focuses on the initial efforts to synthesize hetero- or homometallic tri- or dinuclear metal carbonyl complexes supported by para-terphenyl diphosphine ligands. A series of [Mb>2b>M’(CO)b>4b>]-type clusters (M = Ni, Pd; M’ = Fe, Co) could be accessed and used to relate the metal composition to the properties of the complexes. During these studies it was also found that non-innocent behavior was observed in dinuclear Fe complexes that result from changes in oxidation state of the cluster. These studies led to efforts to rationally incorporate central arene moieties capable managing both protons and electrons during small molecule activation.

Chapter 3 discusses the synthesis of metal complexes supported by a novel para-terphenyl diphosphine ligand containing a non-innocent 1,4-hydroquinone moiety as the central arene. A Pd⁰-hydroquinone complex was found to mediate the activation of a variety of small molecules to form the corresponding Pd⁰-quinone complexes in a formal two proton ⁄ two electron transformation. Mechanistic investigations of dioxygen activation revealed a metal-first activation process followed by subsequent proton and electron transfer from the ligand. These studies revealed the capacity of the central arene substituent to serve as a reservoir for a formal equivalent of dihydrogen, although the stability of the M-quinone compounds prevented access to the PdII-quinone oxidation state, thus hindering of small molecule transformations requiring more than two electrons per equivalent of metal complex.

Chapter 4 discusses the synthesis of metal complexes supported by a ligand containing a 3,5-substituted pyridine moiety as the linker separating the phenylene phosphine donors. Nickel and palladium complexes supported by this ligand were found to tolerate a wide variety of pyridine nitrogen-coordinated electrophiles which were found to alter central pyridine electronics, and therefore metal-pyridine π-system interactions, substantially. Furthermore, nickel complexes supported by this ligand were found to activate H-B and H-Si bonds and formally hydroborate and hydrosilylate the central pyridine ring. These systems highlight the potential use of pyridine π-system-coordinated metal complexes to reversibly store reducing equivalents within the ligand framework in a manner akin to the previously discussed 1,4-hydroquinone diphosphine ligand scaffold.

Chapter 5 departs from the phosphine-based chemistry and instead focuses on the incorporation of hydrogen bonding networks into the secondary coordination sphere of [Feb>4b>(μb>4b>-O)]-type clusters supported by various pyrazolate ligands. The aim of this project is to stabilize reactive oxygenic species, such as oxos, to study their spectroscopy and reactivity in the context of complicated multimetallic clusters. Herein is reported this synthesis and electrochemical and Mössbauer characterization of a series of chloride clusters have been synthesized using parent pyrazolate and a 3-aminophenyl substituted pyrazolate ligand. Efforts to rationally access hydroxo and oxo clusters from these chloride precursors represents ongoing work that will continue in the group.

Appendix A discusses attempts to access [Feb>3b>Ni]-type clusters as models of the enzymatic active site of [NiFe] carbon monoxide dehydrogenase. Efforts to construct tetranuclear clusters with an interstitial sulfide proved unsuccessful, although a (μb>3b>-S) ligand could be installed through non-oxidative routes into triiron clusters. While [Feb>3b>Ni(μb>4b>-O)]-type clusters could be assembled, accessing an open heterobimetallic edge site proved challenging, thus prohibiting efforts to study chemical transformations, such as hydroxide attack onto carbon monoxide or carbon dioxide coordination, relevant to the native enzyme. Appendix B discusses the attempts to synthesize models of the full H-cluster of [FeFe]-hydrogenase using a bioinorganic approach. A synthetic peptide containing three cysteine donors was successfully synthesized and found to chelate a preformed synthetic [Feb>4b>Sb>4b>] cluster. However, efforts to incorporate the diiron subsite model complex proved challenging as the planned thioester exchange reaction was found to non-selectively acetylate the peptide backbone, thus preventing the construction of the full six-iron cluster.