Automatic kinetic model generation: a novel modeling approach for liquid-phase processes

Gust Popelier
Dissertation for Master of Science in Chemical Engineering, Ghent U (2022).
Science Published: (Jun/2022)

Current societal trends push for innovation and optimization of existing large-scale chemical processes in the pursuit of a sustainable economy. In the last decade, a substantial amount of progress has been made for predictive first principle-based models for numerous gas-phase processes. This is in contrast to the liquid phase, for which the construction of such detailed models has proven to be more challenging. As a case study, the liquid-phase oxidation of cyclohexane to cyclohexanol and cyclohexanone is investigated since it is an industrially relevant process to produce nylon-6 and nylon- 6,6. A new program is developed to aid liquid-phase model development, called ALKIMO, standing for Automatic Liquid-phase Kinetic Modeller. Its function is to transform a gas-phase model into a liquid- phase model by adding new effects present in the liquid phase. Two effects are considered, the (de)stabilizing effect of dissolution and the diffusional limitations present in the liquid phase. The effect of dissolution is modelled using the Gibbs free energy of solvation. This is calculated using the Volume- Translated Peng-Robinson Equation of State, which can be calculated using available group contribution methods. The predicted solvation energies in cyclohexane are compared with the experimental results of the CompSol database, resulting in a root mean square deviation of 1.23 kJ/mol. Diffusional limitations were also calculated based on group contribution methods. These proved to play a significant factor in the liquid-phase oxidation of cyclohexane as a substancial percentage (~21%) of the reactions are diffusion limited.