Student Seminar Series - July 28, 2006
University of Minnesota
School of Statistics
College of Liberal Arts
Methods
for Verifying Kinetic Models and Estimating Reaction Rates
Aaron
Nienow
Friday, July 28, 2006
10:00 AM, B60
Ford Hall
Minneapolis, East Bank Campus
Refreshments at 9:30 AM
300 Ford Hall
Abstract
Kinetic modeling is an important part of chemistry research. True
kinetic models are often difficult to determine due to the various
reactions taking place during an experiment. Often, models are selected
based on similarities to previous research. This can
lead to inaccurate model selection.
The data used in this analysis was generated from an experiment using
ferrihydrite and goethite particles. It was suspected that the
particles react with hydroquinone following one and two dimensional
diffusion models respectively. The method used to check
the validity of the hypothesized kinetic models was a general linear
model. The
results indicate that one and two dimension diffusion models do not
adequately represent the reactions
taking place. Initial reaction rates were estimated using cumulative
sum charts
to give the researchers a better understanding of the reaction process.
A Box-Cox transformation was performed using the estimated reaction
rates, and it was determined that interactions between experimental
settings are present. However, due to the small sample size, the
specific interactions were not able to be determined.