Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Date of Award

5-2011

Document Type

Campus Access

Degree Name

Doctor of Philosophy (PhD)

Degree Program

Chemistry

First Advisor

Scott M. Auerbach

Second Advisor

Justin T. Fermann

Third Advisor

Michael D. Barnes

Subject Categories

Inorganic Chemistry | Physical Chemistry

Abstract

We have modeled proton transfer using quantum chemical methods in important catalytic material namely Zeolite and polymeric systems to design anhydrous proton exchange membranes for fuel cells (charge transporting materials). In the H-Y Zeolite proton transfer study, we computed the total mean rate coefficient for proton transfer in bare H-Y Zeolite, for comparison with NMR experiments and previous calculations. The proton transfer energies were calculated using two-layer ONIOM calculations on an 8T-53T cluster, where xT indicates x tetrahedral atoms. Rate coefficients were computed using truncated harmonic semi-classical transition state theory. The zero-point energy (ZPE) corrected proton site energies in H-Y (FAU structure) were found to be O3 (0 kJ mole -1 ), O1 (2.1 kJ mole-1 ), O2 (16.1 kJ mole-1 ) and O4 (17.5 kJ mole-1 ), in quantitative agreement with previous calculations and in qualitative agreement with neutron diffraction occupancies. The ZPE corrected activation energies range from 35 to 123 kJ mole-1 . Total mean rate coeffficients were found to exhibit a strong non-Arrhenius temperature dependence, with apparent activation energies in the range ca. 60-100 kJ mole-1 at high temperature, and ca. 3 kJ mole-1 at low temperature. This low-temperature value reflects thermally assisted tunneling to a site with slightly higher energy. NMR experiments by Sarv et al. and Ernst et al. report apparent activation energies of 61 and 78 kJ mole-1 , respectively, extracted from temperature ranges 298-658 and 610-640 K. Our theoretically computed apparent activation energies for these temperature ranges are 72 and 79 kJ mole-1 , respectively, in quite good agreement with experiment.

In the Grotthuss proton transfer and design criteria for anhydrous proton exchange membrane study, we have modeled structures and energetics of anhydrous proton-conducting wires: tethered hydrogen-bonded chains of the form · · · HX · · · HX · · · HX · · ·, with functional groups HX = imidazole, triazole and formamidine; formic, sulfonic and phosphonic acids. We have applied density functional theory (DFT) to model proton wires up to 19 units long, where each proton carrier is linked to an effective backbone to mimic polymer tethering. This approach allows the direct calculation of hydrogen bond strengths. The proton wires were found to be stabilized by strong hydrogen bonds (up to 50 kJ mole -1 ) whose strength correlates with the proton affnity of HX [related to pK b (HX)], and not to pKa (HX) as is often assumed. Proton translocation energy landscapes for imidazole-based wires are sensitive to the imidazole attachment point (head or feet) and on wire architecture (linear or interdigitated). Linear imidazole wires with head-attachment exhibit low barriers for intrawire proton motion, rivaling proton diffusion in liquid imidazole. Excess charge relaxation from the edge of wires is found to be dominated by long-range Grotthuss shuttling for distances as long as 42 Å, especially for interdigitated wires. For imidazole, we predict that proton translocation is controlled by the energetics of desorption from the proton wire, even for relatively long wires (600 imidazole units). Proton desorption energies show no correlation with functional group properties, suggesting that proton desorption is a collective process in proton wires. In the aim of mimicking water, phenolic systems were studied using LSDA/6-311G(d,p). We find using density functional theory calculations on phenolic dimers that their polymers have low re-orientation barrier 13.7 kJ mole -1 compared to the imidazole/triazole systems in which the whole group has to rotate. This study shows that the dynamical nature of the hydrogen bonds in the system is very important to consider when searching for a proton transferring functional group for anhydrous proton exchange membranes.

Share

COinS