Master of Science, The Ohio State University, 2007, Graduate School

Committee: Not Provided (Other) Subjects:

Doctor of Philosophy, The Ohio State University, 2015, Chemical Physics

Astrochemistry studies chemical reactions that occur in the interstellar medium. Most astrochemical studies focus on the detection of interstellar molecules via rovibrational spectroscopy or the interpretation of their abundance using reaction dynamics. This thesis is divided into two parts, and in each of them we address one of the two aspects from the perspective of theoretical chemistry. In the first part we propose the synthetic route for propene (CH3-CH=CH2), which has a surprisingly high abundance as the most saturated organic molecule in the extremely cold and thin gas-phase medium. Based on the reaction rates obtained from experiments and ab initio calculations for all three reactions in this route, we simulate the time-evolution for the abundance of propene and find that our synthetic route is able to reproduce part of the observed propene abundance. In the second part we discuss the spectroscopy and the dynamics of H5+, aiming to decipher its role as the intermediate of the astrochemically important proton transfer reaction, H3+ + H2 -> H5+ -> H2 + H3+. The large amplitude motions (LAM's) in H5+ allow the protons to permute between H3+ and H2 and result in products that have different nuclear spins and rovibrational states from the reactants. These LAM's introduce challenges to the theoretical studies of H5+ because the conventional harmonic oscillator and rigid rotor approximation is no longer valid. Diffusion Monte Carlo and its extensions are used to capture the couplings between LAM's and other rovibrational modes in H5+. Specifically, we focus on three LAM's: the proton transfer vibration, the H2-H2 torsion, and the internal rotation of H3+ about its C3 symmetry axis. For selected states of H5+, we evaluate energies and wave functions as well as the reaction paths for the above-mentioned proton transfer process. In the spectroscopic studies, we find that the proton transfer vibration has a significant mixing with the dissociation vibration, (open full item for complete abstract)

Committee: Anne McCoy (Advisor); Terry Miller (Committee Member); John Wilkins (Committee Member) Subjects: Physical Chemistry

Doctor of Philosophy, University of Toledo, 2013, Chemistry

Two-dimensional self-assemblies of plasmonic nanoparticles (NPs) could one day become a useful technology for mankind as it has a potential to produce desirable structures with various patterning and ordering that is difficult to achieve by the top-down approach. Furthermore, these patterned and ordered structures of NPs have been known to display interesting optoelectronic properties. While applications using self-assemblies of plasmonic NPs seem promising, the fundamental forces that govern the evolution of these structures are not fully understood yet. Interesting similarities between the interfacial NP self-assembly and epitaxial growth exist despite a number of differences such as diffusion, desorption, coalescence, and ordering. The goal of this dissertation is to determine to what extent established submonolayer epitaxy theories can be applied to modeling interfacial NP self-assembly, and thereby develop new tools for understanding NP-NP interactions and self-assembly. Different sizes of 1-dodecanethiol (DDT) capped Au NPs were used to study the submonolayer growth behavior of NP islands. However, for the study, synthesizing DDT Au NPs > 8 nm by a conventional method was a challenge. To solve this synthesis problem, large NPs were synthesized in water and a phase transfer-based method was developed and used. While solving this synthesis problem, we developed general guidelines for NP phase transfer and determined that successful phase transfer only depended on three key surfactant parameters: bulkiness, length, and ability to get onto the interface. Furthermore, we also shed light on the mechanistic details of NP phase transfer into the organic medium. The experimental results for the submonolayer growth study such as NP island size distribution (ISD), capture zone distribution (CZD), percolation threshold, diffusion, and flux were compared and contrasted with the known epitaxy theories. It was found that islands were compact and circular at low covera (open full item for complete abstract)

Committee: Terry Bigioni PhD (Advisor) Subjects: Chemistry

Doctor of Philosophy, The Ohio State University, 2013, Physics

Ultracold atomic gases may be the ultimate quantum simulator. These isolated systems have the lowest temperatures in the observable universe, and their properties and interactions can be precisely and accurately tuned across a full spectrum of behaviors, from few-body physics to highly-correlated many-body effects. The ability to impose potentials on and tune interactions within ultracold gases to mimic complex systems mean they could become a theorist's playground. One of their great strengths, however, is also one of the largest obstacles to this dream: isolation. This thesis touches on both of these themes. First, methods to characterize phases and quantum critical points, and to construct finite temperature phase diagrams using experimentally accessible observables in the Bose Hubbard model are discussed. Then, the transition from a weakly to a strongly interacting Bose-Fermi mixture in the continuum is analyzed using zero temperature numerical techniques. Real materials can be emulated by ultracold atomic gases loaded into optical lattice potentials. We discuss the characteristics of a single boson species trapped in an optical lattice (described by the Bose Hubbard model) and the hallmarks of the quantum critical region that separates the superfluid and the Mott insulator ground states. We propose a method to map the quantum critical region using the single, experimentally accessible, local quantity R, the ratio of compressibility to local number fluctuations. The procedure to map a phase diagram with R is easily generalized to inhomogeneous systems and generic many-body Hamiltonians. We illustrate it here using quantum Monte Carlo simulations of the 2D Bose Hubbard model. Secondly, we investigate the transition from a degenerate Fermi gas weakly coupled to a Bose Einstein condensate to the strong coupling limit of composite boson-fermion molecules. We propose a variational wave function to investigate the ground state properties of such a Bose-Fermi mixture w (open full item for complete abstract)

Committee: Nandini Trivedi Ph.D. (Advisor); Tin-Lun Ho Ph.D. (Committee Member); Gregory Lafyatis Ph.D. (Committee Member); Richard Furnstahl Ph.D. (Committee Member) Subjects: Condensed Matter Physics

Doctor of Philosophy, The Ohio State University, 2011, Chemistry

Using Diffusion Monte Carlo, vibrational and rotational excited states of CH5+ and its deuterated isotopologues are evaluated and analyzed. A method for evaluating anharmonic corrections to energies along a minimized energy path for the reaction CH3+ + H2 → CH5+ → CH3+ + H2 is also discussed. For the vibrational excited states, the fundamentals in the five modes for CH5+ and CD5+ are calculated. The fundamentals are generated by requiring that the wave functions change sign at specified values of the five Symmetry Adapted Linear Combinations (SALC's) of the CH or CD bond lengths. While the definitions of these modes are based on displacements of the CH or CD bond lengths, the frequencies are found to be low compared to previously calculated CH vibrational frequencies of CH5+. The totally symmetric mode, with A1+ symmetry, has a calculated frequency of 2164 and 1551 cm-1 for CH5+ and CD5+. The frequencies of the four fundamentals that arise from excitation of the four SALC's that transform under G1+ symmetry have frequencies that range from 1039 to 1383 and 628 to 893 cm-1 in CH5+ and CD5+, respectively. The origins of the broken degeneracy are investigated and are found to reflect extensive coupling to the two low-frequency modes that lead to isomerization of CH5+. For the rotational excited states, the J=1, |K|=0,1 rotationally excited states of CH5+ and its deuterated isotopologues are calculated. The calculated J=1, |K|=0,1 rotationally excited state energies are high in energy when compared to the rotational energies calculated from vibrationally averaged rotational constants. The energy of a low-lying inversion mode that corresponds to a low-energy tunneling doublet is also calculated. When the inversion energy is subtracted from that of the J=1, |K|=0,1 rotational energy, the energies are in good agreement with those calculated from the vibrationally averaged rotational constants. The low-lying inversion mode cannot be removed from the calculations because of (open full item for complete abstract)

Committee: Anne McCoy PhD (Advisor); Terry Gustafson PhD (Committee Member); Sherwin Singer PhD (Committee Member) Subjects: Physical Chemistry

Doctor of Philosophy, The Ohio State University, 2005, Materials Science and Engineering

With the tremendous increase in computational power over the last two decades as well as the continuous shrinkage of Si-based Metal Oxide Semiconductor Field Effect Transistors (MOSFET), quantum mechanically based ab initio methods become an indispensable tools in nano-scale device engineering. The following atomistic simulations, including ab initio, nudged elastic band (NEB) and kinetic Monte Carlo methods, have been presented in this work. Using VASP, an ab initio simulation package, we calculated B segregation energy at different atomic sites in perfect and defected Si/SiO2 interfaces and arsenic segregation energy in Si/LaAlO3 structures. With the presence of O vacancies and H in B doped systems, the predicted segregation energy is 0.85 eV for neutral systems and 1.12 eV for negatively charged systems, which is consistent with experimental measurements (0.51 to 1.47 eV). Focussing on the La deficient Si/LaAlO3 interfacial structure, we find that the arsenic prefers energetically not to segregate into LaAlO3 nor does it pile up in front of the interface. In combation of atomic-resolution Z-contrast imaging and electron energy loss spectroscopy (EELS), we theorectically calculated the band structure and EELS of a Ge/SiO2 interface. we actually found a chemically abrupt Ge/SiO2 interface, which has never been reported before and which is quite desirable for applications. Furthermore, we formulated a kinetic Monte Carlo model to simulate the oxidation process of Ge ion-implanted Si, which explained the formation of abrupt Ge/SiO2 interface. Using nudged elastic band (NEB) method, we systematically calculate the vacancy formation, diffusion activation energy and pre-exponential diffusion factor at pure and Cu doped Al grain boundaries. Though grain boundary diffusion is still much faster than that of bulk, adding small amounts of Cu can dramatically improve the electromigration reliability of Al interconnects. A comparison of the vacancy formation energy at Al, Al(C (open full item for complete abstract)

Committee: WOLFGANG WINDL (Advisor) Subjects: Engineering, Materials Science

Doctor of Philosophy, University of Akron, 2006, Polymer Science

A two-bead move algorithm for dynamic Monte Carlo simulation has been designed to reflect the true randomness of local torsion dynamics on a high coordination lattice (2nnd lattice). All possible configurations of two consecutive beads of a chain on this high coordination lattice have been included in a library by this algorithm. The moves were implemented by randomly choosing one of the possible configurations from the library. Thus there is no artificial rule for the moves. The algorithm is capable of introducing new bond vectors to local configurations without going through chain ends. The chain-cross has been eliminated in this two-bead move algorithm. The algorithm is fast enough for simulating polyethylene (PE) melts ranging from C40 to C324 on a regular desktop computer. The results of our simulation confirmed there were finite chain length effects, e.g. chain length dependent friction coefficients and non-Gaussian statistics for short PE chains. A detailed comparison has been made among the experiments, prior simulations by other groups, and the results of our new algorithm. The diffusion coefficients scale with molecular weight (M) to the -1.7 power for short chains and -2.2 for longer chains at 180°C, which coincides very well with experimental results. Due to the finite chain length effect, no pure Rouse scaling in diffusion has been observed. The reptation-like slowdown can be clearly observed when M is above 2400 according to the mean square displacements of middle beads. The slope 0.25 predicted by the reptation theory was missing for the intermediate regime of diffusion; instead a slope close to 0.4 appeared, indicating that additional relaxation mechanisms exist in this transition region. The relaxation times extracted by fitting the autocorrelation function of the end-to-end vectors scale with M to 2.5 and 2.7 power using the reptation model and KWW equation, respectively, for entangled chains. The dynamic Monte Carlo algorithm has also been used fo (open full item for complete abstract)

Committee: Wayne Mattice (Advisor) Subjects:

Doctor of Philosophy, The Ohio State University, 2013, Chemical Physics

Molecular spectroscopy remains a powerful tool for obtaining insights into fundamental chemical processes, studying the properties of reactive intermediates, and investigating the chemistry of the interstellar medium. The spectra of semi-rigid molecules are often well-described using spectroscopic Hamiltonians obtained from the application of perturbation theory to the full rotation-vibration Hamiltonian. However, the presence of large amplitude vibrational motions causes the perturbation theory expansion to become slow to converge or even divergent. As a result, the spectra of fluxional systems become difficult to interpret and a proper theoretical description of the nuclear dynamics challenging to achieve. The majority of the work reported in this thesis involves diffusion Monte Carlo (DMC), an approach that has been shown to successfully describe the ground state properties of highly fluxional systems and, through the fixed-node approximation, vibrationally excited states. In this thesis, we first describe an extension of fixed-node DMC to the study of the rotationally excited states of symmetric and asymmetric top molecules that undergo large amplitude vibrational motions. The nodal surfaces used to impose rotational excitation into the DMC calculations are obtained from the roots of the rigid rotor wave functions. Despite using a zeroth-order description of the nodal surfaces and without placing any constraints in the vibrational coordinates, the fixed-node DMC approach is able to capture the effects of rotation-vibration coupling in highly fluxional molecules for states with J as large as 12. One limitation of DMC is the need to perform the simulations in Cartesian coordinates, requiring the use of a full-dimensional potential energy surface and preventing the possibility of restricting the calculations to the subset of coordinates that are physically relevant to the questions being asked. As a first step towards overcoming this limitation, we (open full item for complete abstract)

Committee: Anne McCoy (Advisor); Terry Miller (Committee Member); Frank De Lucia (Committee Member) Subjects: Physical Chemistry