Accuracy and Efficiency in Computational Chemistry: The Correlation Consistent Composite Approach

Description:

One of the central concerns of computational chemistry is that of efficiency (i.e. the development of methodologies which will yield increased accuracy of prediction without requiring additional computational resources – RAM, disk space, computing time). Though the equations of quantum mechanics are known, the solutions to these equations often require a great deal of computing power. This dissertation primarily concerns the theme of improved computational efficiency (i.e. the achievement of greater accuracy with reduced computational cost). Improvements in the efficiency of computational chemistry are explored first in terms of the correlation consistent composite approach (ccCA). The ccCA methodology was modified and this enhanced ccCA methodology was tested against the diverse G3/05 set of 454 energetic properties. As computational efficiency improves, molecules of increasing size may be studied and this dissertation explored the issues (differential correlation and size extensivity effects) associated with obtaining chemically accurate (within 1 kcal mol-1) enthalpies of formation for hydrocarbon molecules of escalating size. Two applied projects are also described; these projects concerned the theoretical prediction of a novel rare gas compound, FKrOH, and the mechanism of human glutathione synthetase’s (hGS) negative cooperativity. The final work examined the prospect for the parameterization of the modified embedded atom method (MEAM) potential using first principles calculations of dimer and trimer energies of nickel and carbon systems. This method of parameterization holds promise for increasing the accuracy of simulations for bulk properties within the field of materials science.

Creator(s): Wilson, Brent R.
Creation Date: August 2011
Partner(s):
UNT Libraries
Collection(s):
UNT Theses and Dissertations
Usage:
Total Uses: 241
Past 30 days: 29
Yesterday: 0
Creator (Author):
Publisher Info:
Publisher Name: University of North Texas
Publisher Info: Web: www.unt.edu
Place of Publication: Denton, Texas
Date(s):
  • Creation: August 2011
Description:

One of the central concerns of computational chemistry is that of efficiency (i.e. the development of methodologies which will yield increased accuracy of prediction without requiring additional computational resources – RAM, disk space, computing time). Though the equations of quantum mechanics are known, the solutions to these equations often require a great deal of computing power. This dissertation primarily concerns the theme of improved computational efficiency (i.e. the achievement of greater accuracy with reduced computational cost). Improvements in the efficiency of computational chemistry are explored first in terms of the correlation consistent composite approach (ccCA). The ccCA methodology was modified and this enhanced ccCA methodology was tested against the diverse G3/05 set of 454 energetic properties. As computational efficiency improves, molecules of increasing size may be studied and this dissertation explored the issues (differential correlation and size extensivity effects) associated with obtaining chemically accurate (within 1 kcal mol-1) enthalpies of formation for hydrocarbon molecules of escalating size. Two applied projects are also described; these projects concerned the theoretical prediction of a novel rare gas compound, FKrOH, and the mechanism of human glutathione synthetase’s (hGS) negative cooperativity. The final work examined the prospect for the parameterization of the modified embedded atom method (MEAM) potential using first principles calculations of dimer and trimer energies of nickel and carbon systems. This method of parameterization holds promise for increasing the accuracy of simulations for bulk properties within the field of materials science.

Degree:
Level: Doctoral
PublicationType: Doctoral Dissertation
Language(s):
Subject(s):
Keyword(s): quantum chemistry | hydrocarbons | composite methods
Contributor(s):
Partner:
UNT Libraries
Collection:
UNT Theses and Dissertations
Identifier:
  • ARK: ark:/67531/metadc84300
Resource Type: Thesis or Dissertation
Format: Text
Rights:
Access: Public
Holder: Wilson, Brent R.
License: Copyright
Statement: Copyright is held by the author, unless otherwise noted. All rights reserved.