Simulation of Thermodynamic Models Using Python
Juhlian V. Santos · International Journal of Engineering Development and Research · 2025
This research developed a computational framework for assessing the phase behavior of ternary systems to determine liquid-liquid equilibrium potential for extraction processes. The framework integrated NRTL and UNIFAC thermodynamic models with comprehensive validation protocols to evaluate the phase behavior characteristics of chemical systems. A modular 14- component architecture was implemented using Python 3.12. The framework was validated using the water-ethanol-acetone system at 298.15 K and 101325.0 Pa with literature-sourced parameters. Results demonstrated successful framework operation for both thermodynamic models. The NRTL model calculated activity coefficients of [1.355, 1.386, 1.107] while UNIFAC predicted [2.512, 0.982, 0.841] for composition [0.4, 0.3, 0.3]. Both models consistently identified single-phase behavior, suggesting no tendency for liquid–liquid phase separation at the given composition. Cross-model validation achieved complete consistency in phase behavior assessment. An average calculation time of 0.0078 seconds demonstrated computational efficiency suitable for industrial screening applications. The framework addresses the need for rapid preliminary assessment of ternary systems prior to detailed extraction process design. Phase behavior assessment capability enables identification of systems unsuitable for liquid-liquid extraction. The validated computational framework provides chemical engineers with reliable tools for phase behavior screening and thermodynamic model simulation in process development applications.