JKMT (Kinetic Modeler's Toolbox, JPEC 2011), a molecular-level kinetic modeling technology, was applied for ultra-deep hydrodesulfurization of light gas oil (LGO-HDS) process. This LGO-HDS model was consisted of the molecular structure-composition model and the molecular-level reaction network model which were built by CME (Composition Modeling Editor) and INGen (Interactive Network GENerator), respectively. In addition, kinetic parameters which were required for kinetic analysis in KME were decided for each reaction family by fitting to actual reaction data. To verify if this model was suitable for the representation of LGO-HDS reaction, this model was applied for another LGO feedstock. Through this verification, some deviations between actual values and estimated values were found, particularly in aromatic contents. In this model, equilibrium constants were not introduced and reversible reactions were not considered. Therefore, the deviation of the estimation in aromatic contents might become bigger than that in sulfur contents. In addition, in this model, kinetic parameters were decided just for each reaction family and the effect of molecular structures for reactivity was not considered. For example, to consider the reactivity of refractory sulfur compounds and poly-aromatic compounds, kinetic parameters should be given not only for reaction families but also for groups of molecular structures. And it was difficult to give numerous kinetic parameters for all elementary reactions and it is also important to introduce some estimation methods of kinetic parameters, such as LFER (Linear Free Energy Relationship) model.

Molecular modeling for hydrodesulfurization and hydrogenation of light gas oil

246th National Meeting (2013)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.