Kinetics of the Reaction of Carbon Dioxide with ... - ACS Publications

Ind. Eng. Chem. Res. 2007, 46, 385-394

385

Kinetics of the Reaction of Carbon Dioxide with Aqueous Solutions of 2-((2-Aminoethyl)amino)ethanol Sholeh Ma’mun,† Vishwas Y. Dindore,‡ and Hallvard F. Svendsen* Department of Chemical Engineering, Norwegian UniVersity of Science and Technology, N-7491 Trondheim, Norway

In view of increased requirements on absorption rates and loading capacity for carbon dioxide (CO2) absorption solvents, more research is directed toward alkanolamines having more than one amino group. In the present work, AEEA {2-((2-aminoethyl)amino)ethanol}, a diamine containing primary and secondary amino groups, is used to study the CO2 absorption kinetics. The reaction kinetics between CO2 and aqueous solutions of AEEA were measured over a range of temperatures from 32 to 49 °C with the concentrations of AEEA ranging between 1.19 and 3.46 kmol m-3 using a string of discs contactor. All kinetic experiments were interpreted using the single-step-termolecular mechanism approach as proposed by Crooks and Donnellan (J. Chem. Soc., Perkin Trans. 2 1989, 331) and reviewed by da Silva and Svendsen (Ind. Eng. Chem. Res. 2004, 43, 3413). Introduction Chemical reactions between gases and liquids play an important role in many industrial processes within both gas purification and production processes. Gas purification is the term used to describe a process to separate a gas mixture from its impurities such as acid gases (e.g., CO2, H2S, and SO2), organic sulfur compounds (e.g., COS, CS2, mercaptans, and thiophene), and certain other impurities (e.g., H2O, HCN, NH3, and hydrocarbons). Removal of acidic gases, e.g., CO2, from industrial and natural gas streams is an important operation in the process industry. Carbon dioxide present in natural gas reduces the heating value of the gas, and as an acidic component, it has a potential to cause corrosion in pipes and process equipment and causes catalyst poisoning in ammonia synthesis.1 Natural gas pipeline specifications usually limit the CO2 content to 2-5%,2 and in the case of liquefied natural gas (LNG) manufacture, the CO2 content needs to be reduced to the 50-100 ppm level. However, most of all, environmental concerns such as global warming are now focused as one of the most important and challenging environmental issues facing the world community and have motivated intensive research on CO2 capture from low-pressure sources. In addition, CO2 capture from various gas streams is viewed as a potentially economic source of CO2 for enhanced oil recovery (EOR) operations. A wide range of technologies currently exist for separation and capture of CO2 from gas streams, as given by Rao and Rubin.3 Such systems have been used in the chemical industry and in the production of technical gases for industrial and laboratory use.4 According to Shaw and Hughes,5 several process-related factors affect the selection of the appropriate method such as CO2 concentration in the feed stream, nature of other contaminants present in the feed stream (e.g., H2S and water in natural gas), pressure and temperature at which the feed stream is available, product purity, and other factors such * To whom correspondence should be addressed. Phone: +47-73594100. Fax: +47-73594080. E-mail: hallvard.svendsen@ chemeng.ntnu.no. † Present address: Department of Chemical Engineering, Gadjah Mada University, Jl. Grafika 2 Jogjakarta, Indonesia 55281. E-mail: [email protected]. ‡ Present address: Aker Kværner Process Systems a.s, P.O. Box 403, N-1327 Lysaker, Norway.

as geographical location, which can be a significant consideration when treating remote natural gas. Carbon dioxide capture using chemical absorption processes is one of the common industrial technologies today and has, in many cases, been found to be the most viable solution compared with other processes.6 For economic reasons, an absorbent must have low cost, high net cyclic capacity, and high reaction/absorption rate for CO2 and, above all, it must be an energy-efficient solvent. Aqueous solutions of alkanolamines are the most commonly used chemical absorbents for the removal of acidic gases from the process industry. Among them, aqueous monoethanolamine (MEA, H2NCH2CH2OH) solutions have been used extensively for this purpose. However, because of well-known disadvantages of MEA such as limited cyclic capacity, high heat of reaction, and degradation in oxygen rich environments,7-10 an experimental study on selection of new absorbents was performed by Ma’mun et al.11 to select solvent candidates having better performance (e.g., high absorption rate, high absorption capacity, low energy of regeneration, etc.) than existing absorbents for CO2 capture. It has been found that 2-((2-aminoethyl)amino)ethanol (H2N(CH2)2NH(CH2)2OH, AEEA) may be a potentially good absorbent for capturing CO2. AEEA offers a high absorption rate and a high net cyclic capacity combined with low regeneration-energy requirements.12 The quest for new absorbents for CO2 capture is very laborious. A tremendous amount of experimental work has to be done on characterizing the solvents with respect to different properties. The VLE data of the CO2-AEEA-water system was established and reported by Ma’mun et al.13 The objective of the present work is the experimental characterization of the kinetics and mechanisms of the reactions between CO2 and AEEA in an aqueous solution. The study of absorption of CO2 into aqueous solutions of AEEA was carried out using a string of discs contactor over a range of temperatures from 32 to 49 °C with concentrations of AEEA ranging between 1.19 and 3.46 kmol m-3. Short Review on the Kinetics between CO2 and Primary and Secondary Alkanolamine Solutions A large number of publications on kinetic experimental data and kinetic mechanisms between CO2 and alkanolamines can

10.1021/ie060383v CCC: $37.00 © 2007 American Chemical Society Published on Web 12/10/2006

386

Ind. Eng. Chem. Res., Vol. 46, No. 2, 2007

be found in the literature. An overview on the reaction mechanisms between CO2 and alkanolamines in both aqueous and nonaqueous solutions is given by Versteeg et al.14 Danckwerts and McNeil15 proposed a reaction mechanism between CO2 and primary and secondary alkanolamines in a two-step reaction:

Step 1: carbamate formation; rate determining CO2 + R1R2NH h R1R2NCOO- + H+

Figure 1. Structure of the AEEA molecule.

(2) k-1/∑kb[B] . 1; resulting in a more complex reaction rate expression with the possibility to have an overall reaction order of 3.

(1) rCO2 ) -

Step 2: protonated alkanolamine formation; instantaneous reaction R1R2NH + H+ h R1R2NH+ 2

(2)

with a second-order overall reaction as follows, CO2 + 2R1R2NH h R1R2NCOO- + R1R2NH+ 2

(3)

Experiments on the reaction between CO2 and primary and secondary alkanolamines in aqueous solutions have been studied by many authors (see refs 14 and 16). However, only for MEA, as a primary amine, the overall reaction order and the value of the kinetic constant are in accordance with the mechanism of reaction in eq 3. For other primary and secondary alkanolamines, the overall reaction orders were found to be varying between 2 and 3, both for aqueous and nonaqueous solutions. For example, the overall reaction order for DEA, as a secondary alkanolamine, was found to be 3 and by some authors to be between 2 and 3. Therefore, the reaction mechanism between CO2 and alkanolamines needs to be modified. Zwitterion Mechanism. This reaction mechanism was first proposed by Caplow17 and reintroduced by Danckwerts,18 in which the reaction between CO2 and alkanolamines results in the formation of a zwitterion intermediate followed by the removal of a proton by a base B: k2Z

z R1R2NH+COOCO2 + R1R2NH y\ k -1

kb

z R1R2NCOO- + BH + R1R2NH+COO- + B y\ k -b

(4) (5)

By using the pseudo-steady-state assumption for the zwitterion concentration, the overall forward reaction rate can be expressed as18

rCO2 ) -

kZ2 [CO2][R1R2NH] 1+

k -1

(6)

∑ kb[B]

where kZ2 and k-1 are the forward and reverse rate constants of reaction 4 where superscript Z denotes zwitterion and ∑kb[B] is the contribution to the zwitterion deprotonation by all bases present in the solution (e.g., H2O, OH-, and free alkanolamine). Simplification of the reaction rate expression above for the two asymptotic situations can be evaluated from its denominators:14 (1) k-1/∑kb[B] , 1; resulting in a simple second-order reaction and the zwitterion formation being the rate-determining step (e.g., MEA in aqueous solutions).

rCO2 )

-kZ2 [CO2][R1R2NH]

(7)

kZ2

∑kb[B] k-1

[CO2][R1R2NH]

(8)

The overall reaction order for the transition region between the two asymptotic cases changes from 2 to 3. Therefore, the reaction rate expression in eq 6 can cover the shifting reaction orders for the reaction system as previously noted. Single-Step-Termolecular Mechanism. This mechanism was proposed by Crooks and Donnellan,19 i.e., amine bonding to CO2 and proton-transfer taking place simultaneously, and has been reviewed by da Silva and Svendsen.20 The reaction mechanism uses the generally accepted mechanism proposed by Danckwerts18 in which the zwitterion intermediate is formed by making the assumption that the reaction proceeds through a loosely bound encounter complex as the initial product. The forward reaction rate for this mechanism is

rCO2 ) -kobs[CO2]

(9)

where

kobs ) kT2 [R1R2NH] ) {kamine[R1R2NH] + kH2O[H2O] + kOH-[OH-]}[R1R2NH] (10) However, eq 9 is equivalent to the asymptotic limit of the zwitterion mechanism described by eq 8 in which the zwitterion deprotonation is caused mainly by water, the alkanolamine, and the hydroxyl ion in aqueous solutions. This reaction mechanism can also explain the broken-order and higher-order kinetics observed as shown by da Silva and Svendsen.20 The CO2-AEEA-H2O is a reactive system. Since AEEA is a diamine containing one secondary and one primary amine group as shown in Figure 1, the chemistry of the system is very complex. The system gives rise to a large number of possible chemical reactions and formed species. A series of physical and chemical reaction equilibria for the CO2-AEEA-H2O system, described in the previous publication,13 together with a possible reaction mechanism between CO2 and AEEA according to the single-step-termolecular mechanism, are as follows:

Diffusion of carbon dioxide from the gas phase into the liquid phase: CO2 (g) h CO2 (l)

(11)

The considered equilibrium reactions that take place in the liquid phase are as follows:

Dissociation of water: 2H2O h H3O+ + OH-

(12)

Dissociation of carbon dioxide: + CO2 (l) + 2H2O h HCO3 + H 3O

(13)

Ind. Eng. Chem. Res., Vol. 46, No. 2, 2007 387

Dissociation of bicarbonate ion: ) + HCO3 + H2O h CO3 + H3O

(14)

Dissociation of monoprotonated AEEA: AEEAH+ + H2O h AEEA + H3O+

(15)

Dissociation of diprotonated AEEA: +

HAEEAH+ + H2O hAEEAH+ + H3O+

(16)

Formation of carbamates: (i) from primary group to form primary carbamate (AEEACOOp ):

(ii) from secondary group to form secondary carbamate (AEEACOOs ):

Dissociation of protonated carbamates: +

+ HAEEACOOp + H2O h AEEACOOp + H3O

(19)

+

+ HAEEACOOs + H2O h AEEACOOs + H3O

(20)

Formation of dicarbamate: (i) from primary carbamate:

(ii) from secondary carbamate:

where R1 is used to represent -(CH2)2- group, R2 is -R1-OH group, and B is a base molecule, e.g., H2O molecule. At zero initial loading of CO2, as in the experiments conducted, only the primary carbamate formed from the primary group will, according to Ma’mun et al.,13 be present as the main product in the solution. This means that the primary group reacts faster than the secondary group. Hence, from this point of view, it seems that the termolecular mechanism is more suitable for aqueous solutions of AEEA. This will be discussed more later.

Figure 2. Experimental set-up of the disc absorption column.

Experimental Section Sample solutions of AEEA (purity > 99 mass %) were prepared from the received chemical from Acros Organics in mixtures with deionized water with concentrations of 1.19, 1.78, 2.38, 2.94, and 3.46 kmol m-3. The concentrations were determined by titration with accuracy to (0.01 kmol m-3. The AEEA as-received does not seem to contain any active amine impurities that significantly affect the overall reaction rate. The CO2 (purity > 99.9992 mol %) and N2 (purity > 99.999 mol %) gases used were obtained from AGA Gas GmbH and used as the source of CO2 and N2, respectively. Apparatus and Procedures. The kinetic experiments for the CO2-AEEA-H2O system were performed using the string of discs contactor. As shown in Figure 2, the apparatus comprises a discs contactor equipped with a Fisher-Rosemount BINOS 100 NDIR CO2 analyzer (2 channels: 2000 ppm and 1 vol % CO2), a Bronkhorst Hi-Tec mass flow controller, a peristaltic liquid pump (EH Promass 83), a gas blower, and five K-type thermocouples. The discs contactor contains 43 discs with a diameter of 1.5 × 10-2 m and a thickness of 4 × 10-3 m, as shown in Figure 3. The active mass transfer area for this arrangement is ∼2.26 × 10-2 m2. The characteristic active length of the column is ∼6.6 × 10-1 m. The disc absorption column was operated in countercurrent mode with liquid flow from top to bottom and gas flow in the opposite direction. The set-up was equipped with thermocouples at the inlet and outlet of both phases and inside the chamber. The liquid and gas flows can be adjusted independently using a liquid pump and gas blower, respectively. The flow of the blower was controlled by a Siemens Micro Master frequency transmitter and has a maximum flowrate of 9.72 × 10-4 Nm3 s-1. A CO2-unloaded AEEA solution with a certain concentration was passed to the column with a flowrate of ∼7.7 × 10-7 m3 s-1, which was found to be about the minimum flowrate where the absorption flux is independent of the liquid flowrate. This is to ensure that the absorption takes place in the fast reaction regime in which the kinetic parameters can be measured. A CO2-N2 gas mixture containing ∼2 vol % CO2 was then led into the column with a flowrate of ∼1.87 × 10-5 Nm3 s-1. The process was terminated when the temperature had stabilized at the desired level and the analyzer showed a constant value for

388

Ind. Eng. Chem. Res., Vol. 46, No. 2, 2007 Table 1. Correlation Details for the Liquid-Side Mass Transfer Coefficient authors

R

n

Stephens & Morris21 Taylor & Roberts22 Taylor & Roberts22 Xu et al.23 Zhang et al.24

6.53 18.0 0.60 14.1 36.4/Ld

0.7 0.4 1.0 0.92 0.65

comment Γ < 0.064 kg m-1 s-1 Γ > 0.064 kg m-1 s-1 Ld is the mean perimeter of liquid flow

Absorption of CO2 into distilled and deionized water is an ideal system for characterization of the liquid-side mass transfer coefficient in the discs contactor. Many researchers have used this system for correlating the liquid-side mass transfer coefficient with operating and physical parameters. The general form of the correlation by Stephens and Morris21 is given by

kL 4Γ n µ 0.5 )R D µ FD

( )( )

(25)

where µ and F are the viscosity and density of the liquid, respectively, and D is the diffusivity of the solute in the liquid phase. Here, (4Γ/µ) is a modified Reynolds number where Γ is the wetting rate in the apparatus and is defined as

Γ)

Figure 3. Arrangement of the disc absorption column.

the exit CO2 volume percent for ∼20-30 min. The data acquisition was performed using FieldPoint and LabVIEW. The total absorption flux of the solute was calculated by taking the solute balance over the entire system, i.e., the difference between the flow of the solute into the system and that going out of the system to the gas analyzer. This flux calculation method gives a higher accuracy compared to that based on the balance just over the discs contactor. The overall absorption rate in the discs contactor is then given by

rsolute )

out Qin solute - Qsolute

V0m

(23)

where rsolute is the molar absorption rate in kmol s-1 and V0m is the molar volume at normal conditions in m3 kmol-1. The amount of the solute entering into the system can be obtained from the mass flow controller reading. The amount of the solute going out of the system through the gas analyzer is obtained from the following equation in Qout solute ) QN2

(

yout solute

)

psolution 1- yout solute P

(24)

Characterization of the Discs Contactor The major drawback of the discs contactor is the unknown hydrodynamics of both gas and liquid flows. Hence, for determination of physico-chemical parameters of new systems as well as for simulating the industrial equipment with a disc column, it is necessary to establish the mass transfer and pressure-drop characteristics of the disc column.

(liquid flow rate) × (length of string) surface area of string

(26)

The values of R and n obtained by various researchers are given in SI units in Table 1. It can be seen from the table that there is no agreement on the R and n values. Thus, the correlations differ from system to system and depend on the various factors such as disc type, disc morphology, etc. Hence, for close control over the technique and to obtain reliable data, it is recommended that the system should be calibrated for the physical mass transfer coefficient. The gas-side mass transfer coefficient was correlated by Stephens and Morris21 using NH3 absorption into water from dilute mixture with air. The obtained correlation is given by

( ) ( ) ()

k GP VdF ) 0.328Γ0.13 VFd µ

- 0.33

µ FD

- 0.56

P pi

(27)

where µ and F are the average viscosity and density of the gas phase, respectively, Fd is the density of solute gas, V is the gas velocity, d is the equivalent diameter for gas flow, P is the total pressure of the system, pi is the partial pressure of solute gas, and D is the diffusivity of the solute in the gas phase. However, for the reasons mentioned above, it is necessary to characterize the gas-side mass transfer resistance of the system. A system with complete gas-side mass transfer control, like absorption of SO2 into aqueous NaOH solutions or absorption of NH3 into aqueous H2SO4 solutions, should be used. Measurement of Liquid-Film Coefficient. Absorption of CO2 into distilled and deionized water was used as a system to measure the liquid-side mass transfer coefficient. The overall liquid-side mass transfer coefficient can be calculated from the absorption rate as given in eq 28.

kL )

rsolute a∆CLM

(28)

where a is total mass transfer area in the discs contactor and ∆CLM is the logarithmic mean driving force over the discs contactor given by eq 29.

Ind. Eng. Chem. Res., Vol. 46, No. 2, 2007 389

∆CLM )

(mCG,in - CL,out) - (mCG,out - CL,in) mCG,in - CL,out ln mCG,out - CL,in

(

)

(29)

Here, m is the temperature-dependent distribution coefficient of the solute and is defined as the ratio of the concentration of the solute in the solvent to the concentration of the solute in the gas phase at equilibrium conditions. To minimize the gas-side mass transfer resistance, a pure CO2 (purity > 99.9992 mol %) was used in the gas phase. During all measurements conducted at an average temperature of 25 °C, the gas flowrate was kept constant at ∼1.1 × 10-6 m3 s-1 using the mass flow controller, which has been calibrated using a gas-flow meter or soap-film meter. Figure 4 shows the effect of liquid flowrate in terms of modified Reynolds number on the overall liquid-side mass transfer coefficient. As seen from the figure, the liquid-side mass transfer coefficient is a strong function of the liquid flowrate. The correlation similar to eq 25 was obtained and is given in eq 30.

kL 4Γ 1.0 µ 0.5 ) 342 D µ FD

( )( )

(30)

The dependency of the (µ/FD) has been taken from ref 21 because only one liquid system and temperature were used during the tests. The dependency on the modified Reynolds number is the same as that found by Taylor and Roberts.22 Measurement of Gas-Film Coefficient. Absorption of a solute followed by an instantaneous irreversible reaction with the reaction plane forming at the gas-liquid interface follows all criteria for measurement of the gas-side mass transfer resistance. As mentioned before, the most convenient system is SO2 absorption into aqueous NaOH solution, and this was, therefore, used to measure the gas-side mass transfer coefficient. A continuous mode of gas-liquid contacting operation was used during the experiments. All experiments were carried out at an average temperature of 30 °C. Aqueous NaOH solution of 1 kmol m-3 was prepared by dissolving a known amount of NaOH pellets into the deionized water. Pure SO2 (purity > 99.9 mol %) and N2 (purity > 99.999 mol %) were used as sources of SO2 and N2, respectively. The gas flowrate and gas-phase SO2 concentration were adjusted using the mass flow controllers. The SO2 concentration at the outlet of the module was monitored using an SO2 analyzer (range 0-1000 ppm). A separate calibration SO2-N2 (1000 ppm SO2) mixture was used to calibrate the SO2 analyzer. The controlling mode of the gasside film resistance was confirmed by no dependence of the absorption flux on the liquid flowrate. During all the measurements, the liquid flowrate was kept constant at ∼9.2 × 10-7 m3 s-1. Figure 5 shows the measured gas-side mass transfer coefficients for various gas flowrates in terms of Reynolds number. A correlation in terms of dimensionless numbers was obtained to predict the gas-side mass transfer coefficient and is given in eq 31.

Sh ) 0.12Re0.79Sc0.44

Figure 4. Effect of liquid flowrate in terms of modified Reynolds number on the liquid-side mass transfer coefficient.

(31)

The Sc number dependency was also taken from ref 21 because only one gas and temperature were used for the tests. In addition, the density and viscosity of aqueous solutions of AEEA were measured at experimental conditions and are

Figure 5. Effect of gas flowrate in terms of Reynolds number on the gasside mass transfer coefficient.

reported elsewhere.12 The physical solubility of CO2 is an important parameter for the determination of reaction kinetics. The physical solubility of CO2 into aqueous AEEA solutions was determined using the N2O analogy.25 The required data on the physical solubility of N2O into aqueous solutions of AEEA were taken from a separate experimental work and are published elsewhere.12 Kinetic Study Using the String of Discs Contactor In chemical absorption, the rate of a reaction cannot be neglected with respect to the mass transfer. The absorption flux is, therefore, enhanced because of the chemical reaction, and the average absorption flux is given by

NA )

1 (C/ - CA,b) RT A 1 + EAkL HkG

(32)

The enhancement factor, EA, describes the effect of chemical reaction on the mass transfer rate. Generally, the enhancement factor is defined as the ratio of absorption flux in the presence of chemical reaction to the absorption flux in the absence of chemical reaction for identical mass transfer driving force. Several approximate solutions to predict the enhancement factor, based on the different mass transfer models, are available in the literature and are applicable over a wide range of process

390

Ind. Eng. Chem. Res., Vol. 46, No. 2, 2007

Figure 6. Effect of liquid flowrate on the average absorption flux of CO2 in aqueous solutions of AEEA.

conditions, reactions of differing complexity, and chemical solute loadings. In the case of a pseudo-first-order reaction regime, the enhancement factor due to chemical reaction based on the penetration theory is given by

EA ) x1 + Ha

2

rCO2 ) -{kAEEA[AEEA] + kH2O[H2O] + kOH-[OH-]} × (33)

xkT2 DACB0 kL

[AEEA][CO2] (35) Using the approach mentioned above, the reaction rate constants and the order of the reaction can then be obtained, with the enhancement factor for the system given by

where

Ha )

Bouhamra and Alper28 studied the reaction mechanism for the same system as this work. This is the only reference found in the literature. A stopped flow conductimetry was used to measure the kinetic parameters where CO2 was first dissolved in distilled water and then mixed with the alkanolamine solution in the stopped-flow apparatus. Because of the low solubility of CO2 in water, the concentration of alkanolamine used was very low. Hence, a high alkanolamine concentration, as typically used industrially, cannot be reached by this technique. Also, care should be exercised when extrapolating data from low-concentration experiments to high-concentration experiments, since the kinetic parameters strongly depend on the alkanolamine concentration. The low alkanolamine concentration used is, however, one of the limitations of this technique. The presence of basic molecules is necessary for the reaction between CO2 and alkanolamines to proceed to form a stable final product, i.e., carbamate. The base species present in the solution are AEEA, H2O, and other bases such as OH-. If AEEA, H2O, and OH- are the dominating bases, the forward reaction rate for this mechanism is, according to Crooks and Donnellan,19 written in eqs 9 and 10. Rewriting these equations leads to

(34)

Thus, measuring the absorption flux in the pseudo-first-order absorption regimes using the string of discs contactor and comparing it with the approximate analytical solutions allows the determination of the reaction kinetics for the reactive systems. For the slow reaction regime (Ha < 0.3), there is no enhancement due to chemical reaction. In this regime, the absorption flux depends on the physical mass transfer coefficient and, hence, on the liquid flowrate.26 The enhancement due to chemical reaction occurs in the fast reaction regime (2 < Ha , EA∞), where the absorption flux is independent of the physical mass transfer coefficient and, hence, independent of the liquid flowrate. Thus, the measurement of the absorption flux must be performed in this regime to determine the mechanism and reaction kinetics of the CO2 absorption. Figure 6 shows the average absorption flux versus the liquid flowrate. It is shown that, at a liquid flowrate of ∼7.5 × 10-7 m3 s-1, the absorption flux starts to be constant. The reaction mechanism of the CO2-AEEA-H2O system can be evaluated by the single-step-termolecular mechanism proposed by Crooks and Donnellan.19 This mechanism has been reviewed by da Silva and Svendsen20 using ab initio calculations in which the more probable mechanism has been evaluated based on the reaction of carbamate formation from CO2 and alkanolamines. It was concluded that the single-step-termolecular mechanism is the most likely and realistic compared to the zwitterion mechanism. In addition, with a large number of fitting parameters in the zwitterion mechanism, it is not feasible to come up with a unique data set to explain the reaction rate constant and some parameters take on unphysical values as observed by Aboudheir et al.27

EA )

x

1+

kobsDCO2

(36)

kL2

where

kobs ) kT2 [AEEA] ) {kAEEA[AEEA] + kH2O [H2O] + kOH-[OH- ]}[AEEA] (37) Since the reaction occurs in the fast reaction regime and at very low CO2 loadings, the concentration of CO2 in the bulk will, in practice, approach zero. Equation 32 can, therefore, be simplified to

1

NA )

1 kL

x

1+

kobsDCO2

RT + HkG

C/A

(38)

kL2

Results and Discussion The reaction kinetics of the CO2-AEEA-H2O system were studied using the string of discs contactor over a range of temperatures from 32 to 49 °C at various concentrations of aqueous CO2-unloaded AEEA solutions. Reaction between CO2 and AEEA will form primary and secondary carbamates where CO2 can react simultaneously with the primary group of the amine and the secondary group of another amine molecule. The sum of eqs 17 and 18 can be written as follows

2AEEA + 2CO2 (l) + 2H2O h AEEACOOp + + (39) AEEACOOs + 2H3O

Ind. Eng. Chem. Res., Vol. 46, No. 2, 2007 391

Figure 7. Effect of AEEA concentration on the average absorption flux for several temperatures: (O), 32.1 °C; (0), 34.9 °C; (4), 39.6 °C; (]), 44.2 °C; (×), 48.8 °C.

Figure 8. Effect of AEEA concentration on kobs over a range of temperatures from 32 to 49 °C: (O), 32.1 °C; (0), 34.9 °C; (4), 39.6 °C; (]), 44.2 °C; (×), 48.8 °C.

Table 2. Comparison of the Second-Order Rate Constant k2 for Alkanolamines at 25 °C alkanolamine

k2,25°C/(m3 kmol-1 s-1)

∆E/(kJ kmol-1)

source

AEEA MEA DEA PZ

12 100 7 000 1 200 53 700

32 500 17 900 41 800 33 600

this work ref 30 ref 31 ref 32

According to the distribution of species in the aqueous solution of AEEA given in ref 13, it was known that, at low CO2 loadings, at which the experiments were performed (∼zero CO2 loading), the species present are mainly the primary carbamate of AEEA as a product, the monoprotonated AEEA, and the unreacted AEEA. The other products (e.g., the secondary carbamate and the dicarbamate of AEEA) are present in negligible amounts, and the formation of them can, therefore, be disregarded. Thus, eq 39 can be simplified to eq 40 as the main reaction. + AEEA + CO2 (l) + H2O h AEEACOOp + H3O

(40)

In addition, to evaluate the reaction mechanisms of the formations of the secondary carbamate and the dicarbamate of AEEA, more experimental work needs to be conducted. Since the experiments were conducted at high enough concentrations of AEEA, the reaction between CO2 and AEEA can then be expressed by a pseudo-first-order irreversible reaction approach. The observed pseudo-first-order reaction rate constants, kobs, were determined using the string of discs contactor at various concentrations of AEEA ranging from 1.19 to 3.46 kmol m-3 over the given temperature range. The rate constants of OH-, kOH-, in eq 37 were taken from ref 29. From the calculations, it is known that the effect of hydroxyl ion on the kobs is relatively small, i.e., max ∼1%. The approximate order of reaction can be evaluated from the effect of the AEEA concentration on the average absorption flux, as shown in Figure 7. It can be seen that the effect of AEEA concentration on the absorption flux is not very significant for a constant temperature. This means that the concentration of AEEA is approximately constant during the reaction. Hence, the assumption of the pseudo-first-order reaction can be made. By assuming the pseudo-first-order reaction, the second-order rate constant, k2, was extracted by taking the best-fit straight line between kobs/[AEEA] and 1/T. An Arrhenius expression

Figure 9. Variation of {kobs - kOH-[OH-][AEEA]}/[AEEA] with [AEEA] over a range of temperatures from 32 to 49 °C: (O), 32.1 °C; (0), 34.9 °C; (4), 39.6 °C; (]), 44.2 °C; (×), 48.8 °C.

for values of the rate constant obtained in this work is defined as follows

[

k2 ) k2, 25°C exp -

∆E 1 1 R T 298.15

(

)]

(41)

where k2, 25°C ) 1.21 × 104 m3 kmol-1 s-1, ∆E ) 3.25 × 104 kJ kmol-1, and R ) 8.314 kJ kmol-1 K-1. Using eq 41, the predicted value of k2 at 25 °C for AEEA is found to be higher than those of MEA30 and DEA31 but much lower than that of piperazine (PZ),32 as shown in Table 2. From the table, it is shown that the activation energy obtained in this work is also comparable with those of MEA and DEA. From the data obtained, it can be seen that kobs strongly depends on the AEEA concentration, as shown in Figure 8. The values of {kobs - kOH- [OH-][AEEA]} divided by the concentration of AEEA result in a linear equation expressed in eq 42. Plots of this equation in Figure 9 gave good straight lines.

{kobs - kOH-[OH- ][AEEA]} [AEEA]

) kAEEA[AEEA] + kH2O [H2O] (42)

The third-order reaction rate constant of AEEA, kAEEA, and that of water, kH2O, were calculated from the plots, and the results

392

Ind. Eng. Chem. Res., Vol. 46, No. 2, 2007

Figure 10. Relationship between ln k and 1/T: (O), AEEA; (0), H2O.

Figure 11. Comparison of measured and predicted kobs obtained by the termolecular mechanism of CO2-AEEA-H2O system: (O), this work; (0), Bouhamra and Alper.28

Table 3. Reaction Rate Constants of AEEA and Those of Water T/°C

kAEEA/(m6 kmol-2 s-1)

kH2O/(m6 kmol-2 s-1)

32.1 34.9 39.6 44.2 48.8

2 968 3 263 3 775 4 349 4 973

224.7 256.7 315.8 386.6 469.6

Table 4. Reaction Rate Constant of AEEA and That of Water for the CO2-AEEA-H2O System in Comparison to Those of PG and MEA Systems at 25 °C absorbent

k × 10-3/ (m6 kmol-2 s-1)

kH2O/ (m6 kmol-2 s-1)

C/ (kmol m-3)

source

AEEA PG MEA

2.35 2.09 1.71

161 118 73.7

1.19-3.46 0.10-4.00 0.19-5.50

this work ref 33 ref 27

are tabulated in Table 3. During the interpretation of the experimental results, an average value of water concentration was used. From this figure, it can be seen that reaction order is between 1 and 2 toward the concentration of AEEA. In Figure 10, an Arrhenius plot is given. From this figure, the reaction rate constant of AEEA, kAEEA, and that of water, kH2O, can be expressed as follows:

(-3030 T ) -4320 ) 3.19 × 10 exp( T )

kAEEA ) 6.07 × 107 exp kH2O

8

(43) (44)

Applying eqs 43 and 44 at a temperature of 25 °C, the predicted values for both the reaction rate constant of AEEA and that of water obtained in this work were found to be comparable to those of the potassium glycinate (PG) system33 and those of the MEA system,27 as seen in Table 4. Kinetic data for the CO2-AEEA-H2O system available in the literature are very limited. The only reference found to be compared to this work was that of Bouhamra and Alper.28 Figure 11 shows a comparison between the predicted kobs in this work and those of Bouhamra and Alper28 at 25 °C. It can be seen that there is reasonable agreement between kobs reported by ref 28 and by the present work, where the results obtained from this work represent the data very well at higher, typically industrial concentrations of AEEA. However, at lower concentrations of AEEA, values reported by ref 28 are higher than those predicted from the present work. This discrepancy might be caused by uncertainties in their experimental procedure in addition to deviations in the interpretation of the data. As shown

by ref 28, a first-order reaction with respect to the AEEA concentration was obtained within the concentration range of 0.015-0.05 kmol m-3. This indicates that there is a shift in the reaction order when going from high to low concentrations of AEEA. In addition to the discrepancy above, the effect of water was also disregarded for their case. This means that the deprotonation of zwitterion is very fast. The zwitterion mechanism was used in their work to interpret and explain the data. However, the termolecular mechanism as given in eq 37 can explain this behavior equally well as at low amine concentrations; the water term in eq 37 will dominate. On the basis of the results obtained, it was shown that the single-step-termolecular mechanism provides a good representation of the experimental data in the whole concentration range. Conclusions The reaction kinetics between CO2 and aqueous solutions of AEEA were measured over a range of temperatures from 32 to 49 °C with the concentrations of AEEA ranging between 1.19 and 3.46 kmol m-3 using the string of discs contactor. The kinetic mechanism of the CO2-AEEA-H2O system was studied using the string of discs contactor and interpreted by the singlestep-termolecular mechanism proposed by Crooks and Donnellan.19 The results show that this mechanism can be applied to determine the reaction mechanism of the CO2-AEEA-H2O system. In addition to the reaction mechanism between CO2 and AEEA, a complete mechanism of the reactions for the CO2AEEA-H2O system covering all species present has been given. The second-order rate constant for the reaction between the primary amine group of AEEA and CO2 has been determined and found to compare reasonably well with earlier data at very low amine concentrations. In comparison to MEA, AEEA yields a higher reaction rate at the same conditions as shown by higher predicted values of the second-order rate constant. Acknowledgment Financial support from the Research Council of Norway Climatic Program through the SINTEF BIGCO2 Project and by the European Commission through the CASTOR Integrated Project (Contract No. SES6-CT-2004-502856) is greatly appreciated. Nomenclature a ) mass transfer area (m2) B ) base species

Ind. Eng. Chem. Res., Vol. 46, No. 2, 2007 393

C ) concentration (kmol m-3) d ) equivalent diameter, column diameter (m) D ) diffusivity (m2 s-1) E ) activation energy (kJ kmol-1), enhancement factor H ) Henry’s law constant (kPa m3 kmol-1) Ha ) Hatta number ) xk2DACB0/kL k -1 ) reverse reaction rate constant (s-1) k2 ) forward reaction rate constant (m3 kmol-1 s-1) kAEEA ) reaction rate constant contributed by AEEA (m6 kmol-2 s-1) kb ) reaction rate constant contributed by all the bases (m3 kmol-1 s-1) kG ) gas-side mass transfer coefficient (m s-1) kH2O ) reaction rate constant contributed by water (m6 kmol-2 s-1) kL ) liquid-side mass transfer coefficient (m s-1) kobs ) observed reaction rate constant (s-1) kOH- ) reaction rate constant contributed by hydroxyl ion (m6 kmol-2 s-1) m ) distribution coefficient n ) constant in eq 25 N ) flux (kmol m-2 s-1) p ) partial pressure (kPa) P ) total pressure (kPa) Q ) volumetric flowrate (m3 s-1) r ) reaction rate (kmol m-3 s-1), molar absorption rate (kmol s-1) R ) universal gas constant (8.314 kJ kmol-1 K-1) Re ) Reynolds number ) (FVd/µ) Sc ) Schmidt number ) (µ/FD) Sh ) Sherwood number ) (kd/D) T ) temperature (K) V ) velocity (m s-1) Vm0 ) molar volume (m3 kmol-1) y ) gas mole fraction Greek Symbols R ) constant in eq 25 ∆ ) difference Γ ) wetting rate (kg m-1 s-1) µ ) viscosity (Pa s) F ) density (kg m-3) Subscripts 0 ) initial condition A ) component A b ) bulk B ) component B G ) gas phase i ) solute gas i in ) inlet L ) liquid phase LM ) logarithmic mean obs ) observed out ) outlet Superscripts T ) termolecular mechanism Z ) zwitterion mechanism * ) equilibrium Literature Cited (1) Astarita, G.; Savage, D. W.; Bisio, A. Gas Treating with Chemical SolVents; New York: John Wiley & Sons, 1983.

(2) Buckingham, P. A. Fluor Solvent Process Plants: How They are Working. Hydrocarbon Process. 1964, 43 (4), 113. (3) Rao, A. B.; Rubin, E. S. A Technical, Economic, and Environmental Assessment of Amine-Based CO2 Capture Technology for Power Plant Greenhouse Gas Control. EnViron. Sci. Technol. 2002, 36, 4467. (4) Desideri, U.; Corbelli, R. CO2 Capture in Small Size Cogeneration Plants: Technical and Economical Considerations. Energy ConVers. Manage. 1998, 39, 857. (5) Shaw, T. P.; Hughes, P. W. Optimize CO2 Removal. Hydrocarbon Process. 2001, 80 (5), 53. (6) Meisen, A.; Shuai, X. Research and Development Issues in CO2 Capture. Energy ConVers. Manage. 1997, 38, S37. (7) Li, Y.-G.; Mather, A. E. Correlation and Prediction of the Solubility of Carbon Dioxide in a Mixed Alkanolamine Solution. Ind. Eng. Chem. Res. 1994, 33, 2006. (8) Jou, F.-Y.; Mather, A. E.; Otto, F. D. The Solubility of CO2 in a 30 Mass Percent Monoethanolamine Solution. Can. J. Chem. Eng. 1995, 73, 140. (9) Liu, Y.; Zhang, L.; Watanasiri, S. Representing Vapor-Liquid Equilibrium for an Aqueous MEA-CO2 System Using the Electrolyte Nonrandom-Two-Liquid Model. Ind. Eng. Chem. Res. 1999, 38, 2080. (10) Bucklin, R. W. DGAsA Workhorse for Gas Sweetening. Oil Gas J. 1982, 80 (45), 204. (11) Ma’mun, S.; Svendsen, H. F.; Hoff, K. A.; Juliussen, O. Selection of New Absorbents for Carbon Dioxide Capture. Energy ConVers. Manage. 2007, 48, 251. (12) Ma’mun, S. Selection and Characterization of New Absorbents for Carbon Dioxide Capture. Ph.D. Thesis, Norwegian University of Science and Technology, Trondheim, Norway, 2005; Paper 2005:116. (13) Ma’mun, S.; Jakobsen, J. P.; Svendsen, H. F.; Juliussen, O. Experimental and Modeling Study of the Solubility of Carbon Dioxide in Aqueous 30 Mass % 2-((2-Aminoethyl)amino)ethanol Solution. Ind. Eng. Chem. Res. 2006, 45, 2505. (14) Versteeg, G. F.; van Dijck, L. A. J.; van Swaaij, W. P. M. On the Kinetics between CO2 and Alkanolamines both in Aqueous and Nonaqueous Solutions. An Overview. Chem. Eng. Commun. 1996, 144, 113. (15) Danckwerts, P. V.; McNeil, K. M. The Absorption of Carbon Dioxide into Aqueous Amine Solutions and the Effects of Catalysis. Trans. Inst. Chem. Eng. 1967, 45, T32. (16) Blauwhoff, P. M. M.; Versteeg, G. F.; van Swaaij, W. P. M. A Study on the Reaction between CO2 and Alkanolamines in Aqueous Solutions. Chem. Eng. Sci. 1983, 38, 1411. (17) Caplow, M. Kinetics of Carbamate Formation and Breakdown. J. Am. Chem. Soc. 1968, 90, 6795. (18) Danckwerts, P. V. The Reaction of CO2 with Ethanolamines. Chem. Eng. Sci. 1979, 34, 443. (19) Crooks, J. E.; Donnellan, J. P. Kinetics and Mechanism of the Reaction between Carbon Dioxide and Amines in Aqueous Solution. J. Chem. Soc., Perkin Trans. 2, 1989, 331. (20) da Silva, E. F.; Svendsen, H. F. Ab Initio Study of the Reaction of Carbamate Formation from CO2 and Alkanolamines. Ind. Eng. Chem. Res. 2004, 43, 3413. (21) Stephens, E. J.; Morris, G. A. Determination of Liquid-Film Absorption Coefficient: A New Type of Column and Its Application to Problems of Absorption in Presence of Chemical Reaction. Chem. Eng. Prog. 1951, 47, 232. (22) Taylor, R. F.; Roberts, F. The Calibration of Disc Type Laboratory Gas-Absorption Column. Chem. Eng. Sci. 1956, 5, 168. (23) Xu, G. W.; Zhang, C. F.; Qin, S. J.; Wang, Y. W. Kinetics Study on Absorption of Carbon Dioxide into Solutions of Activated Methyldiethanolamine. Ind. Eng. Chem. Res. 1992, 31, 921. (24) Zhang, X.; Zhang, C.-F.; Qin, S.-J.; Zheng, Z.-S. A Kinetics Study on the Absorption of Carbon Dioxide into a Mixed Aqueous Solution of Methyldiethanolamine and Piperazine. Ind. Eng. Chem. Res. 2001, 40, 3785. (25) Laddha, S. S.; Diaz, J. M.; Danckwerts, P. V. The N2O Analogy: The Solubilities of CO2 and N2O in Aqueous Solutions of Organic Compounds. Chem. Eng. Sci. 1981, 36, 228. (26) Dindore, V. Y.; Brilman, D. W. F.; Versteeg, G. F. Hollow Fiber Membrane Contactor as a Gas-Liquid Model Contactor. Chem. Eng. Sci. 2005, 60, 467. (27) Aboudheir, A.; Tontiwachwuthikul, P.; Chakma, A.; Idem, R. Kinetics of the Reactive Absorption of Carbon Dioxide in High CO2-loaded, Concentrated Aqueous Monoethanolamine Solutions. Chem. Eng. Sci. 2003, 58, 5195. (28) Bouhamra, W.; Alper, E. Reaction Kinetics of Carbon Dioxide, Carbonyl Sulfide and Carbon Disulfide with Aqueous 2-(2-Aminoethylamino) Ethanol. Chem. Eng. Technol. 2000, 23, 421.

394

Ind. Eng. Chem. Res., Vol. 46, No. 2, 2007

(29) Pohorecki, R.; Moniuk, W. Kinetics of Reaction between Carbon Dioxide and Hydroxyl Ions in Aqueous Electrolyte Solutions. Chem. Eng. Sci. 1988, 43, 1677. (30) Hikita, H.; Asai, S.; Katsu, Y.; Ikuno, S. Absorption of Carbon Dioxide into Aqueous Monoethanolamine Solutions. AIChE J. 1979, 25, 793. (31) Sada, E.; Kumuzawa, H.; Butt, M. A. Gas Absorption with Consecutive Chemical Reaction: Absorption of Carbon Dioxide into Aqueous Amine Solutions. Can. J. Chem. Eng. 1976, 54, 421. (32) Bishnoi, S.; Rochelle, G. T. Absorption of Carbon Dioxide into Aqueous Piperazine: Reaction Kinetics, Mass Transfer and Solubility. Chem. Eng. Sci. 2000, 55, 5531.

(33) Kumar, P. S.; Hogendoorn, J. A.; Versteeg, G. F.; Feron, P. H. M. Kinetics of the Reaction of CO2 with Aqueous Potassium Salt of Taurine and Glycine. AIChE J. 2003, 49, 203.

ReceiVed for reView March 27, 2006 ReVised manuscript receiVed October 20, 2006 Accepted November 1, 2006 IE060383V