The document discusses using activity-based kinetics models to represent experimental data on the reaction of carbon dioxide with aqueous amine solutions of MEA and MAPA. It describes conducting absorption rate experiments using two different apparatuses and modeling the results. Using activity coefficients in the rate expressions improved the representation compared to concentration-based models.
The document discusses using activity-based kinetics models to represent experimental data on the reaction of carbon dioxide with aqueous amine solutions of MEA and MAPA. It describes conducting absorption rate experiments using two different apparatuses and modeling the results. Using activity coefficients in the rate expressions improved the representation compared to concentration-based models.
The document discusses using activity-based kinetics models to represent experimental data on the reaction of carbon dioxide with aqueous amine solutions of MEA and MAPA. It describes conducting absorption rate experiments using two different apparatuses and modeling the results. Using activity coefficients in the rate expressions improved the representation compared to concentration-based models.
The document discusses using activity-based kinetics models to represent experimental data on the reaction of carbon dioxide with aqueous amine solutions of MEA and MAPA. It describes conducting absorption rate experiments using two different apparatuses and modeling the results. Using activity coefficients in the rate expressions improved the representation compared to concentration-based models.
Activity-based Kinetics of the Reaction of Carbon Dioxide
with Aqueous Amine Systems. Case studies: MAPA and MEA Monteiro, J.G.M.-S.a, Pinto D.D.D.a, Luo X.a, Knuutila H.a, Hussain S.a, Mba E.a, Hartono A.a, Svendsen H.Fa * a The Norwegian University of Science and Technology (NTNU),Trondheim, Norway, NO-7491
Abstract
The kinetics of CO2 reacting monoethanolamine (MEA) and 3-(methylamino)propylamine (MAPA) solutions are studied by conducting absorption rate experiments in two different apparatuses: a wetted wall column and a string of discs column. It is shown that the apparatuses give comparable results. The results are modeled using the direct kinetic mechanism with activity-based rate expressions, and good representation is obtained.
Keywords: Post-combustion capture; activity-based kinetics; CO2 absorption; MEA; MAPA
1. Introduction
The kinetics of CO2 reacting in amine and caustic solutions have been widely studied by conducting absorption rate experiments. Versteeg et al. [1] and Aboudheir et al. [2] present comprehensive literature reviews concerning alkanolamines as solvents.
Two reaction mechanism models are normally used to describe CO2 reactions with amines: the zwitterion formation and the direct (or termolecular) mechanism [3, 4]. Depending on the assumptions made, both mechanisms will lead to the same rate expression for CO2 absorption, and can be used
Selection and/or peer-review under responsibility of GHGT doi:10.1016/j.egypro.2013.06.069 J.G.M.S. Monteiro et al. / Energy Procedia 37 (2013) 1888 – 1896 1889
indistinguishably. However, the results obtained by applying these models in their original concentration- based form do not seem to be able to catch the variations of the kinetic constants with the concentration of the solution.
Haubrock et al. [5] and Knuutila et al. [6] show that modifying the rate models so that activities of the species are taken into consideration (and not their concentrations) can lead to better representations respectively on caustic and carbonate solutions. This modification is made to take into account the strong non-ideality of the reaction mixtures. Also, Dugas and Rochelle [7] apply direct kinetics with activity- based rate expressions to model mass transfer experiments of CO2 into monoethanolamine (MEA) and piperazine (PZ).
In this work, CO2 reaction rates in monoethanolamine (MEA) and 3-(methylamino)propylamine
(MAPA) are represented using the direct mechanism kinetics with activity-based rate expressions. Experimental data are generated using both a wetted wall column and a string of discs contactor.
It is interesting to notice that activity-based rate models are being successfully applied for modeling reactions other than CO2 absorption, such as the esterification of 1-butanol with acetic acid [8] and the enzymatic production of decyl acetate [9].
2. Experimental Apparatus and Procedure
The reactive absorption studies were conducted in two different apparatuses. MEA absorption experiments were conducted both in a wetted wall column (WWC) and in a string of discs contactor (SDC), while MAPA experiments were done only in the SDC. The WWC was described by Luo et al. [10] while the SDC was described by Ma’mun et al. [11] and Knuutila et al. [12] among others.
In both apparatuses, a gas stream (mixture of N2 and CO2) and an aqueous amine solution flow in countercurrent mode. The apparatuses are instrumented so that it is possible to calculate a mass balance for CO2 over the apparatus and thereby determine the rate of absorption. The main difference between the apparatuses is the mass transfer area, which is 2.19*10-2 m2 in the SDC and 1.69*10-3 m2 in the WWC. The smaller contact area makes the WWC column unfeasible for studying the kinetics of slow systems. As both amines considered in this study are relatively fast (MEA is a primary amine, MAPA has one primary and one secondary group), it is possible to use both apparatuses to evaluate all solutions.
MAPA unloaded solutions with concentrations of 1M, 2M, 3M, 4M and 5M were tested. MEA experiments were conducted with 5M solutions, with loading varying from 0 to 0.5. The reactions were performed within the temperature range of 25ºC to 60ºC.
3. Mass transfer modeling
The CO2 absorption flux (NCO2) is determined experimentally. Using a mass transfer model, as the two- film theory or the penetration theory, the CO2 absorption flux can be correlated with the mass transfer parameters. The correlation obtained by using the two film theory is shown in equation 1. The 1890 J.G.M.S. Monteiro et al. / Energy Procedia 37 (2013) 1888 – 1896
experiments are carried out in the pseudo-first order regime so that the enhancement factor is equal to the Hatta number. pCO2 pCO2 N CO2 a 1 H CO2 1 H CO2 (1) akG Eak L0 akG a kobs DCO2
Here kobs is the observed pseudo-first order kinetic constant. Since all other quantities on the right-hand side of equation 1 can be assessed experimentally, this equation can be solved for kobs. Under the reaction conditions prevailing in our experiments the concentration of OH- is very small, so the effect of hydroxyl ions within the observed pseudo-first order kinetic constant is negligible. The reaction rate of CO2 is commonly described by the simple expression in equation 2. This is called the concentration based model. If the concentrations are replaced by the activities of the components, then we have the activity based model. In equation 3, the activities are represented as the product of the concentration by the activity coefficients. In this equation, kobs and k2 are represented with a superscript , to indicate that they were obtained by using the activity based model.
rCO2 kobs CO2 k2 Am CO2 (2)
rCO2 kobs CO2 CO2 k2 Am Am
A CO2 CO2 (3)
A direct mechanism scheme for carbamate formation is shown in Figure 1. According to this mechanism the forward reaction rate model in its activity form can be described by equation 4.
Figure 1 – Direct mechanism scheme
J.G.M.S. Monteiro et al. / Energy Procedia 37 (2013) 1888 – 1896 1891
rCO2 kB B B Am Am CO2 CO2 (4)
bases
For all the experiments addressed in this work, it is reasonable to consider that water and the amine itself are the only bases present in significant concentrations in the systems. Hence, the rate expression simplifies to the one presented in equation 5. rCO2 Am Am H 2O H 2O Am Am CO2 CO2 (5)
Comparing equation 5 to equation 3, the expressions for the observed pseudo-first order and the second order kinetic constants are obtained as: kobs Am Am H 2O H 2O Am Am CO2 (5a) k2 Am Am H 2O H 2O Am CO2 (5b)
When the reaction rate and the system physical properties are known, and a thermodynamic model (extended UNIQUAC or electrolyte NRTL) is available to evaluate the concentrations and activity coefficients, the kinetic and mass transfer coefficients can be determined.
4. Results and discussion
The modeled second order kinetic constant for the MAPA system is in good agreement with the values experimentally determined, as can be seen in Figure 2. Data for 1M, 2M, 3M and 4M MAPA solutions are well represented by the proposed correlations. However, for 5M MAPA, either the representation is not very good or there are two outliers.
When using a concentration based model, the obtained average absolute relative deviation (AARD) was 23%. However, by introducing the activity-based model, the AARD was reduced to 10%, which is within the experimental uncertainty. This gives an indication that treating the non-idealities of the system leads to improved results.
It is important to stress that the activity coefficients were fitted using the e-NRTL model against available vapor-liquid equilibrium (VLE) and CO2 solubility data. While the VLE data are completely independent from the kinetic results, the CO2 solubility data are used (in the form of Henry’s law coefficient) when applying equation 1. As long as the modeling assumptions are correct and data used consistent, improved representation is expected because the modified model takes into account the non- idealities of the systems.
All the MAPA systems tested within this work were thermodynamically modeled as unloaded systems (regarding the systems properties models). It should be noted that while running the experiments, CO2 reacts with the amine and thus some loading occurs. The effect of the loading on the system properties 1892 J.G.M.S. Monteiro et al. / Energy Procedia 37 (2013) 1888 – 1896
was neglected, since the maximum loaded achieved during the experiments was of the order 0.01 mol CO2/mol MAPA for the experiment with 5M MAPA solution at 60ºC.
Since MAPA is a diamine with one primary and one secondary group, it actually forms two distinct carbamates when reacting with CO2. Hence, the carbamate formation reaction measured in the experiments is actually the lumped sum of the two carbamate reactions taking place in the system. With the current set-up, it is not possible to distinguish between them. However, since there is one primary and one secondary amine group in MAPA, and as the final loadings achieved during the experiments always J.G.M.S. Monteiro et al. / Energy Procedia 37 (2013) 1888 – 1896 1893
Figure 2 – Model and experimental values for the second order kinetic constant of formation of MAPA carbamate, in m3mol-1s-1 1894 J.G.M.S. Monteiro et al. / Energy Procedia 37 (2013) 1888 – 1896
were lower than 0.01 mol CO2/mol MAPA, it is believed that the kinetic results are more representative of the primary amine group reaction.
For the MEA system, the representation is generally good, as can be seen in Figure 3. The results for experiments conducted in the string of discs and in the wetted wall column are in good agreement with each other. Unlike for the MAPA system, the activity-based rate model was found not to have any clear advantage over the concentration-based model. The deviations were an AARD = 27% for the concentration based model and an AARD = 23% for the activity based model. Unfortunately, when using loaded solutions in the experiments, the higher experimental uncertainties lead to scatter in the data that is so significant that the corrections for the non-idealities only gives a marginal improvement. It is still believed that using an activity based model is an advantage.
The concentrations and activity coefficients were obtained from the e-UNIQUAC model presented by [13].
Figure 3 – Parity plot for the observed kinetic constant of formation of MEA carbamate J.G.M.S. Monteiro et al. / Energy Procedia 37 (2013) 1888 – 1896 1895
5. Conclusions
Kinetics of CO2 reacting with monoethanolamine (MEA) and 3-(methylamino)propylamine (MAPA)
solutions were studied by conducting absorption rate experiments in two different apparatuses: a wetted wall column and a string of discs column. It is shown that the apparatuses give comparable results.
The results were modeled using the direct kinetic mechanism with activity-based rate expressions, and good representation is obtained. For the reaction of CO2 into unloaded MAPA a clear improvement was found when converting from concentration based to activity based kinetics. For the test with CO2-loaded MEA the data scatter hides any clear advantage of activity based kinetics.
Acknowledgements
Financial support from the EC 7th Framework Programme through Grant Agreement No : iCap- 241391, is gratefully acknowledged.
References
[1] Versteeg GF, van Dijck LAJ, van Swaaij, WPM. On the kinetics between CO2 and Alkanolamines both in Aqueous and Nonaqueous Solutions. An Overview. Chem. Eng. Commun. 1996, 144: 113-158. [2] Aboudheir A, Tontiwachwuthikul P, Chakma A, Idem R. Kinetics of the reactive absorption of carbon dioxide in high CO2- loaded, concentrated aqueous monoethanolamine solutions. Chemical Engineering Science 2003, 58: 5195-5210 [3] Eirik Falck Da Silva and Hallvard F. Svendsen, Ab Initio study of the reaction of carbamate formation from CO2 and alkanolamines. Ind. Eng. Chem. Res, 2004, 43: 3413-3418 [4] Caplow M. Kinetics of carbamate formation and breakdown. J.Am. Chem. Soc., 1968, 90: 6795-6803 [5] Haubrock J, Hogendoorn JA, Versteeg GF. The applicability of activities in kinetic expressions: A more fundamental approach to represent the kinetics of the system CO2-OH-salt in terms of activities. Chemical Engineering Science 2007, 62: 5753- 69 1896 J.G.M.S. Monteiro et al. / Energy Procedia 37 (2013) 1888 – 1896
[6] Knuutila H, Juliussen O, Svendsen HF. Kinetics of the reaction of carbon dioxide with aqueous sodium and potassium carbonate solutions. Chemical Engineering Science, 2010, 65: 6077-88. [7] Dugas RE, Rochelle GT. Modeling CO2 absorption into concentrated aqueous monoethanolamine and piperazine. Chemical Engineering Science 2011, 66: 5212-8. [8] Grob S, Hasse H. Reaction Kinetics of the Homogeneously Catalyzed Esterification of 1-Butanol with Acetic Acid in a Wide Range of Initial Compositions. Ind. Eng. Chem. Res. 2006, 45: 1869-74. [9] Ribeiro AS, Oliveira MV, Rebocho SF et al. Enzymatic Production of Decyl Acetate: Kinetic Study in n-Hexane and Comparison with Supercritical CO2. Ind. Eng. Chem. Res. 2010, 49: 7168-75 [10] Xiao L, Hartono A, Svendsen HF. Modeling the kinetics between CO2 and concentrated MEA under different loadings. To be submitted. 2012. [11] Ma’mun S, Dindore, VY, Svendsen HF. Kinetics of the Reaction of Carbon Dioxide with Aqueous Solutions of 2-((2- Aminoethyl)amino)ethanol. Ind. Eng. Chem. Res. 2007, 46: 385-94. [12] Knuutila H, Svendsen HF, Juliussen, O. Kinetics of carbonate based CO2 capture systems. Energy Procedia 2009, 1: 1011- 8. [13] Aronu UE, Gondal S, Hessen ET, Haug-Warberg T, Hartono A, Hoff KA, Svendsen HF. Solubility of CO2 in 15, 30, 45 and 60% mass MEA from 40 to 120ºC and model representation using the extended UNIQUAC framework. Chem. Eng. Sci, 2011, 66: 6393-6406
Analysis of Equilibrium Data of CO2 in Aqueous Solutions of Diethanolamine (DEA), Methyldiethanolamine (MDEA) and Their Mixtures Using The Modified Kent Eisenberg Model
Analysis of Equilibrium Data of CO2 in Aqueous Solutions of Diethanolamine (DEA), Methyldiethanolamine (MDEA) and Their Mixtures Using The Modified Kent Eisenberg Model
