چكيده به لاتين
Abstract:
Increased emission of greenhouse gases, particulary carbon dioxide (CO2), has necessitated its separation to mitigate global warming effects; Moreover, separation of CO2 from natural gas stream is of high importance. Among various separation methods, with taking note of the flue gas characteristics (high volme and low pressure), CO2 chemical absorption, is still regarded as a mature and highly efficient technology. In this thesis, improved non-equilibrium dynamic rate-based modeling and simulation, has beed carried out for CO2 chemical absorption into aqueous piperazine (PZ: C4H10N2) solutions. For this purpose, initially a number of thermodynamic, kinetics and mass transfer sub-program/models were coded in MATLAB software, and via preparation a comprehensive databank of the CO2 equilibrium solubility data in aqueous PZ solutions, an artificial neural network with high precision, (%AARD=2.39%) was developed to independently predict equilibrium CO2 loading values. The developed network results have been applied to improve convergence of the solution of the non-linear set of equations resulted from the (γ-φ) thermodynamic model, as well as in the developed CO2 mass transfer molar flux correlatin. In the kinetics and mass transfer modeline section, through conducting a number of experiments, and applying the Buckingham’s pi (π) theorem, the dimensionless numbers of the the system were recognized and new correlations for kinetic rate constants and CO2 mass transfer molar flux, were developed that compared to the previously developed ones in the literature, are more accurate (%AARD= 2.93%), and in addition to taking into account all important CO2 involving reactions, the pseudo first order (P.F.O) simplifying assumption has been avoided, as well. Ultimately, the improved dynamic rate-based modeling and simulation has been carried out, where the developed kinetics and mass transfer rate correlations in this thesis, have been employed, and also the simplifying assumption of the plug flow regime in the absorber packed column, was removed and the effect of the axial mixing was taken into account. The developed rate-based model partial differential and algebraic equations (PDAE), were solved using the numerical method of lines (MOL), and the ultimate simulation results were validated against the experimental data. The simulation results, ascertain the aversive effect of axial mixing on the column separation performance, which would be more intensified for industrial absorber packed columns. Validation results of the developed rate-based model at different situations, indicates its high ability and accuracy (%AARD= 6.59%) in prediction of the process behavior.
Keywords: Carbon dioxide, Piperazine, Chemical absorption, Rate-based modeling, Simulation.
