Multiscale modeling serves as a fundamental tool in advancing our understanding of catalytic reactions, vital for catalyst and reactor design, scaling processes, and optimizing operational efficiencies. This thesis applies multiscale modeling to the development of new technologies for the direct catalytic cracking of crude oil into valuable petrochemicals.

At the laboratory scale, our research enhances the crude oil-to-chemicals conversion process through meticulous optimization of catalyst morphology, hydrodynamic parameters, and thermal conductivity. We synthesized and characterized catalysts with diverse silicon carbide (SiC) sizes and compositions, unveiling an optimized formulation that significantly exceeded current industrial performance benchmarks. Employing computational particle fluid dynamics (CPFD) simulations, we effectively coupled hydrodynamics with heat transfer to gain critical insights into catalyst behavior and thermal dynamics under realistic operational conditions. These simulations informed the selection of optimal operating conditions, which were validated through catalytic cracking experiments, resulting in improved yields of light olefins.

We also explored the synergistic effects of cofeeding water and methanol in the catalytic cracking process. Experimental findings demonstrated that cofeeding significantly influenced conversion rates and selectivity for light olefins while markedly reducing coke formation. The reparameterization of the kinetic model, incorporating insights from water cofeeding, elucidated its effect on the reaction kinetics, notably enhancing the rate constant for propylene production while suppressing coke precursor condensation.

Finally, we underscored the critical need for effective scale-up methodologies for novel processes and catalysts. Leveraging the validated CPFD simulation framework, we conducted comparative analyses of hydrodynamic behaviors within pilot-scale reactors under various feeding configurations. We successfully commissioned a pilot-scale multi-zone fluidized bed reactor with a crude oil processing capacity of 15 L day⁻¹. By integrating our lumped kinetic model into the CPFD simulations for yield predictions, we achieved a commendable agreement with experimental results, affirming the efficacy of our modeling approach. This comprehensive study not only advances the field of catalytic cracking for petrochemical production but also aligns with the objectives of the RDIA initiative, fostering innovation in sustainable energy solutions and materials.