Molecular-level kinetic modeling of the upgrading of residual oil in supercritical water

Abstract

The objective of this thesis is to develop a molecular-level kinetic model for the upgrading of residual oil in an environment of supercritical water. The Kinetic Modeler’s Toolbox software, developed by the Klein research group, is used to make this kinetic model. A rank zero reaction network is created. This is necessary to limit the number of model compounds. The reaction network contains four major chemical pathways: pyrolysis, hydrolysis, aromatization and coking. The entire network consists of 10025 reactions between 2894 components. Besides creating a reaction network, formats for the rate laws are constructed. Because the system is in a pseudo single phase and no catalyst is used, the underlying principle for the rate laws is microkinetics. Supercritical water has three different solvent effects: solvent cage effect, the effect of the dielectric constant and the water-oil phase behavior. These solvent effects are incorporated into the rate laws. Due to the unavailability of product data for supercritical water upgrading of residual oil in literature, tuning of the kinetic parameters is not done in this thesis. Instead, two molecular representations of feed streams are made. One of them is a molecular representation of residual oil. This is done in order to show that the chosen model compounds can model a residual oil feed stream. The other is a molecular representation of VGO. VGO is a subset of residual oil, that contains the more volatile compounds. For VGO a product data set for supercritical water upgrading is available in literature. Kinetic parameters are tuned, such that the simulated properties of the output of the kinetic model are close to those reported in literature. This is done in order to show that the reactions, in the reaction network, and the format of the rate laws represent the physical and chemical phenomena that occur in supercritical water upgrading of oil fractions.