Thermodynamic modeling and simulation of CO2 and H2S removal processes from natural gas and flue gases

Plakia, Anthoula S.; Πλακιά, Ανθούλα Σ.

dc.contributor.author	Plakia, Anthoula S.	en
dc.contributor.author	Πλακιά, Ανθούλα Σ.	el
dc.date.accessioned	2021-04-01T10:24:01Z
dc.date.available	2021-04-01T10:24:01Z
dc.date.issued	2021-04-01
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/53250
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.20948
dc.rights	Default License
dc.subject	Acid gases	en
dc.subject	Alkanolamines	en
dc.subject	Ionic liquids	en
dc.subject	Thermodynamic modeling	en
dc.subject	Process simulation	en
dc.title	Thermodynamic modeling and simulation of CO2 and H2S removal processes from natural gas and flue gases	en
dc.title.alternative	Θερμοδυναμική μοντελοποίηση και προσομοίωση διεργασιών απομάκρυνσης CO2 και H2S από φυσικό αέριο και απαέρια	el
dc.contributor.department	Εργαστήριο Θερμοδυναμικής και Φαινομένων Μεταφοράς	el
heal.type	doctoralThesis
heal.classification	Chemical Engineering	en
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2021-03-18
heal.abstract	The subject of this thesis is thermodynamic modeling and simulation of acid gases removal processes from natural gas and flue gases. Carbon dioxide (CO2) is an acidic gaseous component, which exists naturally in atmosphere, due to the carbon cycle. However, human activities have increased its concentration, a fact that has as a result, significant environmental problems, such as the global warming. The combustion of fossil fuels is the main contributor of CO2 emissions to atmosphere. Therefore, in this thesis, two origin sources of acid gases are examined: flue gases, emitted by fossil fuel combustion and natural gas, which is the cleanest fossil fuel. Thus, CO2 should be removed from natural gas and flue gases and the removal processes are examined in this work. Hydrogen sulfide (H2S) is also present in hydrocarbon resources, such as the natural gas, and it can cause serious problems to piping and other equipment, whereas it reduces its heating value. Therefore, it should also be removed from natural gas. The dominant process used in industry for the removal of acid gases is chemical absorption in aqueous alkanolamine solutions, even though last decades different and environmentally friendly solvents, such as the ionic liquids, have been proposed. The correct description of phase and chemical equilibrium of acid gases-water-alkanolamines mixtures by a thermodynamic model is essential for the proper design and optimization of the process. Classic thermodynamic models, such as equations of state, are not capable of an accurate description of electrolyte mixtures. Therefore, the development of an appropriate thermodynamic framework, in order to describe such mixtures, is necessary. To this purpose, the problem of phase and chemical equilibrium, related to acid gases-water-alkanolamines mixtures, is presented. More specifically, the electrolyte species dissociate in the liquid phase, where chemical reactions take place and ions are formed. Ions remain in liquid phase, whereas molecular species exist both in liquid and vapor phases. In order to solve this complex problem, in which phase and chemical equilibrium coexist, an approach, which describes this problem as analogous to bubble point calculation, is adopted. Therefore, the liquid phase composition is found by the solution of chemical equilibrium, whereas the pressure and vapor phase composition are found by solving the vapor-liquid equilibrium by a thermodynamic model. The thermodynamic model selected is an EoS/GE model, UMR-PRU, which combines Peng-Robinson EoS with UNIFAC via UMR mixing rules, because it provides a consistent description of the mixtures, due to the use of the same equations for both phases, and it is more suitable for electrolyte mixtures than classic equations of state. However, UMR-PRU model needs modification in order to be extended to acid gases-water-alkanolamine mixtures. More specifically, Debye-Hückel term is incorporated, in order to account for the long-range electrostatic forces and the resulting model is called eUMR-PRU. Two alkanolamines have been examined in this thesis: Monoethanolamine (MEA) and Methyl-diethanolamine (MDEA). Thus, the eUMR-PRU is developed by fitting the binary interaction parameters to experimental vapor-liquid equilibrium data of acid gases, CO2 and H2S, in mixtures with H2O, MEA and MDEA. The enormous number of intercorrelated parameters, needed to be estimated, implies the complexity of the problem. At this point, it should be also noted, that UMR-PRU model is extended to H2S-gases (N2, CH4, C2H6) and H2S-hydrocarbon mixtures as well, as there are no available parameters for these mixtures. In the case of CO2, the results of eUMR-PRU yield an average absolute relative deviation in partial pressure, greater than 30 % in all cases examined and the comparison with the results of electrolyte-NRTL, which is a commonly used model for such systems, implies that eUMR-PRU leads to similar or even better results and its extension to these mixtures has been successful. Furthermore, eUMR-PRU model can successfully describe the effect of methane on CO2 solubility in aqueous alkanolamines, i.e. in CO2-H2O-MDEA mixtures, which is of great interest for the natural gas industry. For H2S, the results of the model have been compared to the ones of electrolyte-NRTL and Kent-Eisenberg model, proving once more that the performance of eUMR-PRU is similar or in some cases even more accurate than the other models. Recently, the need of “green”, energy-effective and less volatile solvents has resulted to an increasing interest of the research community to ionic liquids and more specifically to acid gases solubility in ionic liquids, in order to use them as solvents in a physical absorption process for acid gas removal. Therefore, the accurate description of the vapor-liquid equilibrium of acid gases-ionic liquid mixtures by a thermodynamic model is an essential task, in order this model to be used in such a physical absorption process design. Therefore, UMR-PRU model is further extended to CO2-ionic liquid mixtures. The ionic liquids examined are twelve imidazolium-based with anions tetrafluoroborate, hexafluorophosphate and bis(trifluoromethylsulfonyl)imide in order to study the effect of anion and cation on CO2 solubility. Firstly, the correct description of pure ionic liquids is necessary, as their critical properties cannot be measured and their extremely low vapor pressure should be correctly reproduced by a thermodynamic model in order to avoid any solvent loss in a process simulation. Therefore, the critical properties, Soave or Mathias-Copeman parameters of ionic liquids are fitted to density and vapor pressure data. After the definition of UNIFAC groups of ionic liquids used in this work, UMR-PRU binary interaction parameters have been fitted to experimental vapor-liquid equilibrium data of CO2-ionic liquid mixtures. The results are compared to the ones of Peng-Robinson coupled with van der Waals one fluid mixing rules and using kij and lij interaction parameters, expressed as correlations of temperature and carbon number of the cation. It is concluded, that UMR-PRU leads to more accurate results. Process design is maybe the most important part of a chemical engineer’s job. The application of the developed thermodynamic framework in process simulations is the next step needed. Therefore, eUMR-PRU model is incorporated in process simulators (ASPEN HYSYS V8.6 and UNISIM R451) through a CAPE-OPEN protocol and the implementation is found to be successful. At this point, it should be mentioned that the CO2 removal from natural gas and not from flue gases has been examined. To this purpose, three different processes are simulated: CO2 chemical absorption in a 30% w/w aqueous MEA solution, CO2 physical absorption in methanol and CO2 physical absorption in an ionic liquid. For the first process, acid gas thermodynamic package in HYSYS V8.6 is used, whereas for the next two, UMR-PRU and Peng-Robinson EoS are used. Sensitivity analysis has been performed in all cases, in order to define the important parameters of each process and their effect to the results. UMR-PRU is compared to the Peng-Robinson EoS simulation results and the three processes are compared to each other in order to find the most energy-effective process, which is concluded to be the one with ionic liquid as solvent. However, its use is not applicable due to its high viscosity and thus some suggestions of using mixtures of ionic liquid-methanol are made.	en
heal.sponsor	Υποτροφία από τον ΕΛΚΕ ΕΜΠ για το διάστημα 2018-2021	el
heal.advisorName	Βουτσάς, Επαμεινώνδας	el
heal.committeeMemberName	Βουτσάς, Επαμεινώνδας	el
heal.committeeMemberName	Μαγουλάς, Κωνσταντίνος	el
heal.committeeMemberName	Καρώνης, Δημήτριος	el
heal.committeeMemberName	Τασιός, Δημήτριος	el
heal.committeeMemberName	Κροκίδα, Μαγδαληνή	el
heal.committeeMemberName	Σταματάκη, Σοφία	el
heal.committeeMemberName	Βουγιούκα, Σταματίνα	el
heal.academicPublisher	Σχολή Χημικών Μηχανικών	el
heal.academicPublisherID	ntua
heal.fullTextAvailability	false