Date

2014

Document Type

Dissertation

Degree

Doctor of Philosophy

Department

Chemistry

First Adviser

Moore, David T.

Other advisers/committee members

Ferguson, Gregory S.; Vicic, David A.; Kiely, Christopher J.

Abstract

This dissertation is focused on the calculations of potential energy surfaces for carbon monoxide (CO) oxidation mechanisms with oxygen allotropes (O2 and ozone, O3) on monoatomic and diatomic metals. Density functional theory (DFT) calculations have been completed on the interactions of CO and oxygen allotropes with monoatomic and diatomic copper and gold species with charges of +1, 0, and -1 at the B3LYP-SDD/6-311+G(3df) level of theory. Major conclusions, structures and pathways were verified with double-hybrid DFT methods. Trends were also investigated for different DFT functionals and basis sets and the results were in general agreement with the B3LYP result. The computational work is an effort to build a comprehensive set of complexes for aforementioned charge and size, calculated at the same level of theory.Calculations for monoatomic Au showed cooperative binding in the interactions of [O2AuCO]q and [O3AuCO]q (q=+1, 0, -1) along with identification of new minima and transition states along the CO-oxidation pathway(s). The reaction pathways were verified with intrinsic reaction coordinate (IRC) calculations to verify the transitions states and minima connected along the potential energy surface. Ab initio molecular dynamics (AIMD) calculations were completed for the [O2+Au+CO]q pathways to determine if the reaction pathways would be (approximately) followed. Both O2 and O3 systems show similarities in the minima and transition states of pathways across charge and spin states; there are also similarities between structures in the two oxygen allotrope systems, where a number of the O3 structures look very similar to the O2 structures with an additional O atom. The ozone pathways show an ozone-concerted mechanism as favorable to singlet and doublet spin states, while triplet states show preference for stepwise Au oxidation by O3 and subsequent AuO oxidation of CO. Calculations of [CuCO]q, [CuO2]q, [O2CuCO]q, [O3CuCO]q and [CuO3]q complexes and reaction pathways have shown very similar complexes and reaction pathways to the atomic Au systems. However, there are some differences with respect to the energetically favorable pathways and minima between the two elements. The complexes of O2 (O3), CO and diatomic Auq has also been calculated to further understand the affect of size and charge on the interactions of CO and O2 (O3) and to look for trends in the structures and energies of the calculated complexes. Similar complexes were observed in the atom and dimer systems; the [O3+Au2+CO] reaction pathway also shows a pathway that is reminiscent of the Au atomic pathway. Observation of like complexes and pathway for mono- and diatomic gold suggests that trends in the CO oxidation mechanism as a function of size exist. Anti-cooperativity of binding was calculated for all [O2Au2CO]q complexes, where it is not energetically favorable to bind both CO and O2 to Au2q; this anti-cooperativity may be important in the catalytic cycle. The computational work is an effort to support matrix-isolation experiments, probed with infrared spectroscopy (FT-IR). Matrix-isolation experiments have been utilized to test interactions of copper anions with O3 (and CO) in an overall neutralized Ar matrix (with Ar+). Results have shown very small peaks related to CuO3 and copper carbonyls, Cu(CO)- and Cu(CO)3- and peaks associated with O3- (ozonide) and ionic O4 complexes (O4- and O4+) are also observed; however, the very low intensity of the copper complex peaks and the (much) higher intensities of the "parent" peaks (O3 and/or CO) lead to questions about what is happening to the Cu ions. The observation of O4 ionic complexes indicates that ozone is decomposing. For O3 only (and O3:CO) experiments, it is probable that O3 is oxidizing Cu- to form copper oxides, which are not observable with our current IR set-up. The formation of copper oxides would also explain the observation of very small (to no) copper carbonyl peaks in the IR spectra with O3:CO mixes.

Included in

Chemistry Commons

Share

COinS