Thermochemical conversion of biomass: models and modeling approaches
Moreno, Brian Matthew
University of Delaware
Fuels produced from lignocellulosic biomass are a promising renewable energy source that will aid in the transition from petroleum-derived liquid transportation fuels while utilizing the existing processing and distribution infrastructures. Lignocellulosic biomass is composed of lignin, cellulose, and hemicellulose, in fractions that vary by plant species. Pyrolysis of these biopolymers results in a biomass pyrolysis oil (BPO) containing aromatics, hydrocarbons, and various oxygenated species. The high oxygen content of BPO must be decreased before use in conventional internal combustion engines. Deoxygenation of BPO results in a mixture of hydrocarbons with properties similar to those of petroleum-derived fuels, which can be co-processed in traditional refineries. The compositional variability of both the petroleum and biomass fractions motivates the need for molecular-level kinetic models to track these novel species and reaction pathways during the production of biofuel-petroleum blends. Molecular-level modeling of the thermochemical conversion of biomass requires the prediction of reaction networks that identify each chemical pathway from the biomass feedstock to the desired fuel products. The Klein research group has developed a software package that is capable of rapidly generating complete reaction networks and associated kinetic models. First, the Interactive Network Generator (INGen) produces a reaction network from the molecular reactants. Next, the Kinetic Modeling Editor (KME) automatically writes the corresponding rate equations and material balances for each species. The resulting molecular-level kinetic model is solved numerically to provide model predictions of the product composition at varying process conditions. In this dissertation, the capabilities of these software tools are expanded to incorporate various biomass chemistries. The biomass-to-fuels models described in this dissertation provide insight into competing kinetic pathways for the production of renewable liquid fuels from lignocellulosic materials. Predictions are made for product composition at varying feedstock composition and processing conditions; experimental data provides validation. Comparison of the activation energy barriers for competing deoxygenation routes will direct catalyst research to modulate between pathways and provide the highest quality products at the mildest process conditions. The ability to rapidly generate and solve molecular-level kinetic models will enable real-time optimization of fuel production and increase the viability of biofuels in the renewable energy market.