AbstractsBiology & Animal Science

Modeling of complex molecules adsorbed on copper surfaces

by Daniel S. Wei

Institution: Georgia Tech
Department: Chemical and Biomolecular Engineering
Degree: PhD
Year: 2014
Keywords: Density functional theory; Temperature programmed desorption; Physisorption; Enantioselectivity; Methanol synthesis; Heteroatom
Record ID: 2042954
Full text PDF: http://hdl.handle.net/1853/53084


There has been growing demands towards the efficient production of enantiopure compounds through either asymmetric synthesis or separation from racemic mixtures. Recent studies have examined numerous different methods that may address this challenge. One of these methods involved the interaction of chiral molecules on achiral metal surfaces such as copper to create chiral templates while another method utilizes the interaction of chiral molecules on intrinsically chiral surfaces. Earlier studies using nonhybrid Density Functional Theory (DFT) functional has provided some insights into the geometric structures and relative energies of some of these interactions, but it failed to achieve quantitative agreement with experimental studies. Using dispersion corrected DFT functionals, this thesis present a study of chemisorbed dense adlayers of glycine and alanine on Cu(110) and Cu(3,1,17), physisorbed R-3-methycyclohexanone (R-3MCHO) on Cu(100), Cu(110), Cu(111), Cu(221), and Cu(643)R, and the hydrogenation of formaldehyde and methoxide on Zn or Zr heteroatoms promoted Cu surfaces. In the dense glycine and alanine adlayer study, we have resolved a disagreement between experimental observation made on LEED, STM, and XPD, and we showed that heterochiral and homochiral glycine adlayer coexist on Cu(110). Our model failed to show the minute enantiospecificity for dense alanine adlayer on Cu(3,1,17) which indicated a numeric limitation for computational modeling of surface adsorption. In the physisorbed system, the dispersion corrected methods calculated adsorption energies were in better quantitative agreement with the experimentally observed values than the nonhybrid functionals, but it also created a significant overestimation of total adsorption energies. On the other hand, our model had indicated a previously unexpected adsorbate-induced surface reconstruction on Cu(110). This is promising news in term of computational modeling's capability in examining surface-adsorbate interaction on an atomic scale. As for the hydrogenation of formaldehyde and methoxide on copper surfaces, the model showed that the increased binding strength between the reaction intermediates and the heteroatom promoted copper surfaces to be the primary contributor of the increased reaction rates. Furthermore, our model had also indicated that while clustered heteroatoms are relatively rare, a significant portion of reaction takes place near these clustered structures. It is our hope that the results and techniques presented in this thesis can be used to better understand and predict the interaction of more complex surface-adsorbate interactions.