National Technical University of Athens (NTUA)
Θερμοδυναμική ανάλυση συστημάτων δέσμευσης διοξειδίου του άνθρακα
Abstract
dc:descriptionThe Dissertation deals with Carbon Capture Systems (CCS) focusing on chemical absorption and physical adsorption applications. Utmost objective of this work is a thorough investigation of different parameters influencing the efficiency of these processes in order to optimize the process. The tackling of these technologies is performed through simulations for both systems and experimental investigation for chemical absorption systems. The literature review studies different capture technologies with focus on chemical absorption and physical adsorption. It investigates the current state-of-the-art in CCS field, the main modeling approaches followed by researchers and representative steady state and dynamic models.In following, a rate-based, chemical absorption steady state model, is developed in Matlab® and validated for two alkanolamine based aqueous solvents, Mono-ethanolamine (MEA) and 2-Amino-2-methyl-1-propanol (AMP). The model predictions both for the absorber and the desorber columns are validated against experimental data extracted by literature revealing satisfactory predictions of the various parameters’ profiles. The model validation is followed by a parametric analysis, quantifying the effect of various parameters on the process. The sensitivity analysis conducted shows that the gas flow rate entering the column along with the liquid temperature are the most important factors influencing the CO2 removal efficiency as well as the regeneration energy in the desorber. The comparison between the two solvents under a common reference case validates MEA’s predominance as an absorbent when compared to AMP. An integrated chemical absorption unit was also simulated in Aspen Plus software. The system was simulated considering the two amine solvents, a MEA solvent and a MDEA/PZ mixture. MDEA/PZ mixture captured 99% of the CO2 of the flue gas while MEA only 80% for the conditions of the simulation. For the two systems, overall exergy efficiency was calculated considering the carbon capture unit as an integrated thermodynamic system. The overall exergy efficiency of MEA system was higher when compared with MDEA/PZ case. Evaluating the two main columns of these systems in terms of exergy efficiency, the absorber showed low values of efficiency for both solvents, while the desorber had slightly higher efficiencies. In general, MDEA/PZ system was characterized by higher thermodynamic efficiencies both in macroscopic and individual column level. Furthermore, the analysis for the MDEA/PZ system resulted to a 5% lower reboiler heat duty than in the MEA case. Finally for MEA case, a CO2 liquefaction system was simulated, based on Linde-Hampson method. Another case studied is the simultaneous capture of CO2 and H2S by natural gas streams by MEA and MDEA amine solvents. Parametric simulations were conducted in order to investigate the selectivity of these solvents over these pollutants. Moreover, the dynamic behavior of the absorber of a chemical absorption unit was studied. Mathematical solution in the problem was reached through a built-in code in Matlab®, applying the finite differences method. Various disturbances scenarios as well as the start-up process were simulated in order to draw conclusions on the transient behavior of the system. The independent parameters tested were namely the inlet gas and liquid flow rates, the inlet gas and liquid temperatures, the L/G ratio and the initial amine molarity in the26solution. The effect of these factors on the CO2 final capture level was observed and quantified through a sensitivity analysis. The transient behavior of the system showed that changes in gas flow are reflected immediately while changes when liquid flow is varied, the system needs a stabilization time of two minutes.In following, a model for the adsorption stage of the Pressure Swing Adsorption (PSA) process was developed based on Toth and Langmuir models’ isotherms and a Linear Driving Force (LDF) approximation. The mass and energy balances resulted to partial differential equations solved using the Finite Differences Method (FDM). The developed dynamic model was successfully validated against experimental breakthrough curves from literature for zeolite 13X. In following, parametric tests were conducted determining the effect of inlet gas flow rate and gas temperature on the adsorption process and breakthrough curves. The results showed that the adsorption temperature increase leads to a notable decrease of the breakthrough time due to the deterioration of the adsorption efficiency. Finally, the developed model was tested for four different adsorbents, two Faujasite (FAU) type zeolites, 13X and EMC1, activated carbon and Mg-MOF-74 metal organic framework. The simulation results showed that zeolite 13X has significantly longer breakthrough time than the other adsorbents thus leading to more efficient process. Finally, in order to investigate the chemical absorption phenomenon in real conditions, two low cost prototype absorbers were designed and fabricated. Different experiments were conducted varying the initial solvent (MEA) concentration, the solvent temperature and the gas flow rate and pilot results were compared with simulation data. The model was validated against pilot data and theoretical observations deriving from the simulations were confirmed also experimentally.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2019
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Prentza, Loukia
- Πρέντζα, Λουκία
Subjects
dc:subject × 14- Διοξείδιο του άνθρακα
- Απορρόφηση
- Absorption
- Carbon dioxide
- Επιστήμη Χημικού Μηχανικού
- Επιστήμες Μηχανικού και Τεχνολογία
- Χημεία και Τεχνολογία διεργασιών
- Επιστήμη Μηχανολόγου Μηχανικού
- Θερμοδυναμική
- Chemical Engineering
- Engineering and Technology
- Process Chemistry and Technology
- Mechanical Engineering
- Thermodynamics
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/55740