Graduate Thesis Or Dissertation


Global Sensitivity and Stochastic Pathway Analysis of Chemical Mechanisms Public Deposited

Downloadable Content

Download PDF
  • The use of theoretical kinetic modeling provides an invaluable tool for the study of biofuel blend development and optimization. These models provide a way to simulate complex combustion systems and extract information without the need for costly experimental procedures. Indeed, these models often allow one to treat extreme conditions, such as those found in combustion engines, where in situ experiment is not yet feasible. However, many combustion models have large uncertainties in their rate coefficients and many models have been optimized for just a few specific conditions. Therefore, systematic improvement of biofuel combustion models is often necessary. Since these models are quite large, for example the n-butanol combustion mechanism contains 1446 reactions and 243 species, a brute force improvement all rate constant parameters is practically impossible. My Ph.D. work involved applying a global sensitivity analysis (GSA) method to combustion models of biofuel components in order to identify and improve the rate constants within the mechanisms that had the largest effect on the target simulation result. Specifically, my main focus was the combustion of n-butanol. GSA revealed that a target simulation result, the ignition time delay, was quite sensitive to self-reaction of the hydroperoxy radical HO2+HO2->H2O2+O2. The empirical rate coefficient for this reaction had a large uncertainty; therefore, high level ab initio calculations and transition state theory were used to calculate a more accurate rate coefficient for this reaction. The second part of my Ph.D. study involved developing an efficient method for determining the pathways taken during a chemical process using a stochastic method that followed ¡§single atoms¡¨. This study was motivated with the intent of extending the GSA method to account not only for the sensitivity of simulation results to single reactions, but to entire chemical pathways. The method I developed extracts not only reaction flow within the chemical network, but specifies the probabilities of exact chemical pathways.
Date Issued
  • 2012-01-01
Academic Affiliation
Committee Member
Degree Grantor
Commencement Year
Last Modified
  • 2019-11-13
Resource Type
Rights Statement