Return to search

Modified UNIFAC-LLE Group-Interaction Parameters for the Prediction of Gasoline-Ethanol-Water Equilibria

Gasoline spills are sources of groundwater contamination. In the event of a spill, timely remediation efforts can advert most of the potential groundwater contamination due to the immiscibility of gasoline in water. Ethanol functions as a cosolvent that can increase the solubility of gasoline in water. Therefore, the risk of groundwater contamination in the event of a fuel spill increases as the ethanol content in automobile fuels increases. This study examines the effect fuel spill size and ethanol content has on the quantities of toluene, ethylbenzene, m-xylene and o-xylene (TEMO) that dissolve into the aqueous phase at equilibrium. Laboratory experiments were preformed to determine the mass fractions of TEMO in waters that were in contact with various volumes of gasoline and ethanol. UNIFAC is a model capable of predicting the concentrations of TEMO in the aqueous phase of a gasoline-ethanol-water system at equilibrium. In this study, the generalized UNIFAC-LLE method, designed for chemical engineering applications, was used to model the laboratory experiments. New UNIFAC-LLE parameters were developed to improve the model's accuracy in predicting the solubilities of aromatic species in ethanol-water mixtures. The new UNIFAC-LLE parameters were also used to model the laboratory experiments. The modeled results were compared to the analogous laboratory experiments. The UNIFAC-LLE parameters developed in this study improved the model's accuracy in predicting the solubilities of TEMO when the aqueous ethanol mass fraction was between 0.114 and 0.431.

Identiferoai:union.ndltd.org:wpi.edu/oai:digitalcommons.wpi.edu:etd-theses-1460
Date29 April 2008
CreatorsLewandowski, Jason A
ContributorsJohn A. Bergendahl, Advisor, ,
PublisherDigital WPI
Source SetsWorcester Polytechnic Institute
Detected LanguageEnglish
Typetext
Formatapplication/pdf
SourceMasters Theses (All Theses, All Years)

Page generated in 0.0024 seconds