| dc.description.abstract | In biological systems, manganese enzymes have been shown to perform vital proton-coupled electron transfer (PCET) reactions. In manganese lipoxygenase, a mid-valent MnIII-hydroxo moiety is proposed to abstract a hydrogen atom from poly-unsaturated fatty acids to initiate substrate peroxidation. In the oxygen evolving complex (OEC) of photosystem II (PSII), a Mn4Ca cluster performs water oxidation involving the abstraction of 4 protons and 4 electrons and the formation of dioxygen. Reactivity of the OEC has been shown to shut down upon the removal of the Ca2+. However, the exact role of the Ca2+ is not well understood. Understanding how these enzymes achieve their reactivity is of ongoing interest, and relevant model complexes can be used to probe how changing the electronic structure can modulate reactivity. However, the number of MnIII-OH complexes capable of performing PCET are scarce, with only [MnIII(OH)(PY5)]2+, [MnIII(OH)(SMe2N4(tren))]+, and [MnIII(OH)(dpaq)]+ previously reported. These complexes have a wide range of reported PCET rates, but due to the significant variation in ligand design it can be difficult to identify the exact structure-function relationship. To address this, a more systematic approach to complex variability was utilized through modification of the [MnIII(OH)(dpaq)]+ complex. The first of such complexes, [MnIII(OH)(dpaq2Me)](OTf), which contains a methyl group at the 2-position of the quinoline, was synthesized. The goal of this complex was to provide steric bulk near the reaction center. With the substrate TEMPOH, [MnIII(OH)(dpaq2Me)]+ was shown to react at rates more than two orders of magnitude faster than the unsubstituted complex. The methyl-substitution near the reaction center of the 2Me derivative was shown to increase reactivity with small substrates by increasing the reduction center of the manganese center. However, the [MnIII(OH)(dpaq2Me)]+ was shown to react with the C-H bonds of xanthene at rates similar to the initial complex. This damping of reaction rate was determined to be due to steric interactions of the methyl substitution, which hindered reactivity with bulkier substrates. X-ray absorption spectroscopy (XAS) experiments were then performed on the solid and frozen solution samples of both [MnIII(OH)(dpaqR)]+ (R = H, 2Me) complexes and their MnII precursor to gain insight into structural and electronic properties. An interesting result was observed, where [MnIII(OH)(dpaq)]+ showed evidence of a dimeric species in acetonitrile (MeCN), but the expected monomeric species in H2O. This result differed from that of the [MnIII(OH)(dpaq2Me)]+ complex, which showed only the monomeric species in MeCN. The relevance of these results was initially uncertain, as the XAS experiments were performed at cryogenic temperatures (ca. 10 K), while all kinetic experiments were performed above 238 K. To address this, 1H NMR experiments were performed on the [MnIII(OH)(dpaq)]+, and revealed the presence of a mixture of a paramagnetic species, as expected, as well as an antiferromagnetically coupled species as determined by temperature dependent 1H NMR. Combining XAS, density functional theory (DFT), and the 1H NMR results, it was determined that this species was a [MnIIIMnIII(mu-O)(dpaq)2]2+ dimer that existed in equilibrium with the [MnIII(OH)(dpaq)]+ monomer. Meanwhile, the [MnIII(OH)(dpaq2Me)]+ showed no mixture of species in solution, due to the steric bulk of the methyl group disfavoring the formation of such a dimer. Through the addition of a small amount of H2O to [MnIII(OH)(dpaq)]+ in MeCN, the equilibrium could be shifted to favor the monomer in solution, and the kinetics with TEMPOH were reevaluated. The rates obtained for the [MnIII(OH)(dpaq)]+ were now two orders of magnitude faster, and nearly as fast as [MnIII(OH)(dpaq2Me)]+. These studies demonstrated potential hurdles with MnIII-OH model complexes, where their proclivity to form an oxo-bridged MnIIIMnIII dimer in non-hydroxylic solvents could complicate kinetic and spectroscopic results. With the reactivity difference between the two complexes reconciled, additional modifications could be made to the ligand in order to test other effects. The next series of complexes synthesized were [MnIII(OH)(dpaqR)]+ (R = 5OMe, 5Cl, 5NO2), where electron donating and withdrawing groups were appended away from the reactive site. Crystal structures were obtained of the [MnIIIMnIII(u-O)(dpaqR)2]2+ dimers, confirming the results from the above spectroscopic studies. Kinetic studies were performed with the derivatives in the presence of H2O, and these complexes showed only a modest variation in reactivity with the fastest (R = 5NO2) being only 9 times faster than the slowest (R = 5OMe). It was shown with DFT calculations that the substitutions of electron withdrawing groups create a more favorable electron transfer, but disfavors proton transfer. Overall the electron withdrawing group created a more potent PCET oxidant, but the results were somewhat dampened by the increased acidity and minor modulation of reduction potential. With the small variance in reactivity by the ligand modifications, a different approach was used where [MnIII(OH)(dpaq)]+ was modified by the addition of Sc(OTf)3. In the presence of Sc(OTf)3, a new intermediate is formed that is able to abstract hydrogen atoms from substrates with moderately strong C-H bonds such as ethylbenzene. This shows significant enhancement over reactivity of the complex in the absence of Sc(OTf)3, and is approaching the reactivity of the more reactive MnIV-oxo complexes. Spectroscopic techniques identify the intermediate as a MnIII complex that is similar in structure to [MnIII(OH)(dpaq)]+, but the exact mode of interaction between scandium and the manganese complex are still under investigation. Despite the synthetic efforts so far, MnIII-OH complexes cannot access PCET reactivity of the stronger C-H bonds of substrates such as cyclohexane. MnIV-oxo complexes, however are often proposed as intermediates in catalytic C-H bond activation reactions. Synthetic MnIV=O complexes with varying ligand structures have shown a wide range of reactivity, with some complexes reacting only with very weak C-H bonds, while others are capable of abstracting strong C-H bonds from substrates such as cyclohexane. The understanding for this variability in reactivity is currently lacking. Previous studies from Nam and Shaik utilizing DFT have proposed the involvement of an excited state to explain the rapid reactivity of complexes such as [MnIV(O)(N4py)]2+. To further evaluate this computationally, DFT calculations along with a multireference method, complete active-space self-consistent field (CASSCF) followed by N-electron valence perturbation theory to the second order (NEVPT2) are performed on [MnIV(O)(N4py)]2+ with a substrate with experimental activation parameters. The nature of the multistate reactivity is evaluated, and the energetics are compared with the experimental values. | |
