Molecular-level kinetic modeling of carbon feedstocks upgrading through hydrotreatment on bifunctional catalysts

Date
2018
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
The primary objective of this thesis is to develop molecular-level kinetic models for hydrotreating processes of conventional and non-conventional hydrocarbon feedstocks on bifunctional catalysts. A comprehensive comparison of the kinetics of the most important industrial hydroprocessing operations is obtained by developing a model for the green diesel and petroleum-derived lube base oil production. An overview is given of the different steps required for constructing the models. These are developed using the in-house Kinetic Modeling Toolbox software by reconstructing the molecular feed composition, building a reaction network and using an optimization algorithm to develop the reaction kinetics. Experimental data is used throughout this process to optimize and evaluate the obtained kinetics. Both hydrotreating processes are modeled assuming the Langmuir-Hinshelwood-Hougen-Watson kinetics formalism and using the Bell-Evans-Polanyi linear free energy relationship to reduce the number of tunable kinetic parameters. ☐ To model the hydroprocessing for green diesel production, a fatty acid methyl ester is used as model compound to represent whole vegetable oil feeds. Experimental data of a coconut oil hydroprocessing operation is subsequently used to verify the accuracy of the obtained kinetics. This model consisted of 150 molecular species and 309 reactions, grouped into 8 reaction families. The hydroprocessing of a pretreated deasphalted oil is more complex and represented by a network containing 1 690 species and 24 754 individual reactions grouped in 11 reaction families. The model results for both processes show good agreement with the experimental results for a range of process conditions. While both models comprise the same reaction types, a significant difference in kinetics is witnessed based on operation conditions, feed properties and catalyst type. The biomass hydrotreatment is performed at mild conditions with a mildly acidic catalyst resulting in fast saturation reactions and removal of heteroatoms. A quantitative relation between the Ni and Mo content of the Ni-Mo/Al2O3 catalysts and the reaction rate constants for the hydrotreatment is developed. On the other hand, the heavier feed for the lube base oil production undergoes primarily hydrocracking and -isomerization reactions favored by the high hydrogen pressure and strong acidic catalyst. ☐ The molecular-level kinetic modeling approach appeared feasible to model all hydrotreating processes and has great potential modeling the co-treatment of a petroleum-derived and biomass feedstock and enabling the accurate prediction of important product properties.
Description
Keywords
Applied sciences, Green diesel, Hydroprocessing, Lubricant, Molecular-level kinetic modeling
Citation