National Technical University of Athens (NTUA)
Development of a group - contribution equation of state for the thermodynamic modelling of associating mixtures
Abstract
dc:descriptionThe purpose of the present thesis is the development of a group – contribution model able to accurately describe the phase equilibria of mixtures comprising of associating compounds. Modelling of phase equilibrium in mixtures involving hydrocarbons, such as those comprising natural gas, water and alcohols or glycols is a very challenging task for the gas industry. The accurate modelling of such mixtures is vital for the safe and optimum operation and design of natural gas transportation pipelines and unit operations. Classic thermodynamic models, such as cubic equations of state (EoS), usually used in industrial practice, fail to accurately describe the phase behavior of strongly polar or hydrogen bonding mixtures. As such, the development of an appropriate thermodynamic framework to describe the phase behavior of such mixtures is deemed necessary.In the first part of the thesis, new experimental vapor – liquid equilibrium (VLE) measurements of the CO2/methane binary are conducted close to the critical temperature of CO2, in order to extend the applicability of the models in a wider temperature range. The obtained data are of high accuracy (3 mK in temperature, 0.02 bar in pressure and 0.0008 in composition) and are successfully validated against literature values. Emphasis is given in the determination of the critical point of the mixture through a scaling law based on statistical mechanics. Moreover, the obtained data are correlated with different EoS of various complexity. Actually, two cubic EoS are used, Peng – Robinson (PR) and Soave – Redlich – Kwong (SRK) and the obtained results are compared with those of two more advanced models; the UMR-PRU which belongs to the EoS/GE class and combines the PR EoS with the UNIFAC activity coefficient model through advanced mixing rules and the Perturbed – Chain Statistical Associating Fluid Theory (PC-SAFT) EoS which is one of the most successful versions of SAFT. All models correlated satisfactorily the experimental data. UMR-PRU yields overall the lowest deviations both away from the critical region, as well as in the prediction of the critical point.In the second part, the UMR-PRU model is extended to mixtures of natural gas components and strongly polar compounds. Very satisfactory correlation of the VLE and liquid – liquid equilibrium (LLE) is obtained for the examined systems. The performance of the model is also evaluated against that of another model belonging to the same class, TST/NRTL, which is suggested by the commercial Aspen HYSYS© (Hysys) process simulator for use in natural gas dehydration applications with glycols. It is shown that in most cases UMR-PRU yields superior results, despite its group – contribution nature, while TST/NRTL systematically fails to predict well the solubility of heavier hydrocarbons (HCs) in the aqueous phase. Both models yield good results in the prediction of multicomponent mixtures containing methane, aromatic hydrocarbons, triethylene glycol (TEG) and water, at conditions similar with those used in natural gas dehydration units, which suggests that both models can satisfactorily predict the aromatic components distribution. After the evaluation in the phase equilibrium of binary and multicomponent mixtures, UMR-PRU is implemented in commercial process simulators, Hysys and Honeywell UniSim© (Unisim), through the CAPE–Open protocol. As expected, identical results are obtained in both simulators. Two cases are examined and the results are compared with those of TST/NRTL; Case I represents a simple absorber, while Case II a simplified offshore dehydration process. In Case I, UMR-PRU yields the same level of dehydration as TST/NRTL, while in Case II it is shown that UMR-PRU and TST/NRTL calculate similar stripping gas rate, TEG purity and circulation rate for achieving the same dry gas water content. Nevertheless, the two models differ in the calculated TEG loss, where UMR-PRU is considered to be closer to the real case than TST/NRTL. UMR-PRU also predicts lower duties in the reboiler, condenser and heat exchangers compared to TST/NRTL, which is considered more realistic due to better prediction of the aqueous TEG heat capacities. Furthermore, a sensitivity analysis of several operating parameters, such as stripping gas rate, operating temperature and pressure of the columns, is conducted. The qualitative comparisons for several properties predicted in Case II are consistent with what is theoretically expected.Based on the observations of the sensitivity analysis, an optimization of the process parameters in terms of operating cost is performed with the UMR-PRU model, along with a preliminary economic evaluation for the base and optimized conditions through the Aspen ICARUS® (Icarus) software and with literature correlations. A relatively low plant capacity (360 SMcm/y) is considered where it is shown that the optimized process conditions do not actually affect the fixed capital cost. The calculated capital cost is in relative good agreement with literature values. For the operating cost, a 6 % decrease is obtained, which corresponds to a cost reduction of about 37 €/SMcm of treated wet gas.In the third part of the thesis, the effect of the mixing rules used in an EoS and the direct accounting for the hydrogen bonding forces, by application of Wertheim’s first order perturbation theory on the phase equilibria of natural gas mixtures with associating compounds is examined. To this end, three versions of the Peng – Robinson equation of state are evaluated in the prediction of the VLE of ternary and multicomponent natural gas mixtures that contain water and methanol or MEG. These are, the cubic Peng – Robinson with conventional van der Waals one fluid (vdW1f) mixing rules, the UMR-PRU model, which combines PR with UNIFAC through the Universal Mixing Rules, and the Cubic Plus Association Peng – Robinson (CPA-PR), which incorporates the Wertheim’s term of statistical mechanics in the PR EoS. It is shown that both UMR-PRU and CPA-PR provide significant improvement over PR EoS in the prediction of multicomponent phase equilibrium. Moreover, UMR-PRU and CPA-PR are able to predict satisfactorily the dew points of natural gas mixtures containing water and methanol, yielding much better results compared to PR. CPA-PR yields the best overall results suggesting that the incorporation of the associating term is necessary to obtain better prediction of hydrogen bonding mixtures. On the other hand, the selection of advanced mixing rules for the attractive and co-volume parameters, also improves the prediction of phase equilibria of complex mixtures containing natural gas components, water and hydrogen bonding compounds such as methanol and MEG, yielding satisfactory results. Based on this observation, a new model, the so-called UMR-CPAPRU, is proposed which combines the advantages of EoS/GE models with the increased accuracy of the perturbation theory. It is first validated in the prediction of two ternary mixtures consisting of methane, water and a hydrate inhibitor (methanol or MEG) and it yields improved results compared to those of UMR-PRU and CPA-PR. This is more pronounced for the methane solubility in liquid phase, where the associating effects are more severe. Very satisfactory correlation results are obtained in the VLE of binary mixtures, where the model improves the UMR-PRU performance both qualitatively and quantitatively. For example, UMR-CPAPRU correlates very well the azeotrope formation in the VLE of hydrocarbons (HC) with methanol yielding substantial improvement over the UMR-PRU predictions. Also, the new model yields better results in the LLE of HC/water and HC/methanol mixtures than UMR-PRU. Then, the model is evaluated in the prediction of the LLE of ternary mixtures, where it yields better qualitative and quantitative results compared to UMR-PRU. Finally, the UMR-CPAPRU improves the performance of UMR-PRU in the dew point prediction of synthetic natural gas mixtures with water and methanol.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2018
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Petropoulou, Eirini
- Πετροπούλου, Ειρήνη
Subjects
dc:subject × 15Rights
dc:rights- Statement dc:rights
-
- BY_NC
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/44926