Molecular-level kinetic modeling of hydroprocessing for green diesel production
Date
2016
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Green diesel is a promising fuel that is becoming more dominant among other types of fuel. It lowers greenhouse gas emissions by 40–90%, has higher energy density than petro-diesels, and can be introduced into any diesel engine or infrastructure without many mechanical modifications. The production process of green diesel consists of two reactors in series. Both reactors are followed by a flash drum. The first reactor, removes oxygen by hydrotreating the triglycerides. The result is a paraffin mix with high cetane number and cloudpoint. The second reactor lowers both cetane number and cloudpoint by isomerization and cracking of the paraffins. The latter is a beneficial change, the former is not. Therefore, a trade-off in either good combustion properties vs. good cloud point emerges.
The objective of this thesis is to deduce a kinetic model for both reactors, using Klein research group in-house software. The software is on a molecular level. Both models are fit to experimental data. First, a reaction network is composed. The extent of this reaction network is controlled by specifying reactant for every specific reaction and by adding a max rank restriction on isomerization. The rank of the reaction is equal to the number of reaction steps from the feed molecules. Together, both models consists of 959 reactions. Second, properties of all species are computed. Third, for all reactions, a reaction rate constant is determined. The number of parameters is drastically reduced from 1918 to 39 (98% less) by using Linear Free Energy Relationships (LFER). Because a catalyst is involved, Langmuir-Hinshelwood-Hougen-Watson (LHHW) kinetics are applied.
The parity plot for the product distribution of the 1st reactor has a R2 value of 0.9979. This model does not include temperature dependency. The parity plot for the product distribution of the 2nd reactor has R2 value of 0.8291. Also, it concludes temperature dependency, its R2 value for the temperature dependent data is 0.9885. By increasing temperature, the isoparaffin content increases. At about 365°C it reaches a maxima and starts to decrease. This can be related with an increasing thermal cracking and therefore a reduced isoparaffin content.