Kinetics of carbonate based CO2 capture systems - Core

Available online at www.sciencedirect.com

Energy Procedia

(2009) 1011–1018 Energy Procedia Procedia1 00 (2008) 000–000 www.elsevier.com/locate/procedia www.elsevier.com/locate/XXX

GHGT-9

Kinetics of carbonate based CO2 capture systems Hanna Knuutilaa, Hallvard F. Svendsena*, and Olav Juliussenb a

Department of Chemical Engineering, NTNU, N-7491 Trondheim, Norway 2 SINTEF Materials and Chemistry, N-7465 Trondheim, Norway

Elsevier use only: Received date here; revised date here; accepted date here

Abstract The Henry’s law constants for physical solubility of N2O were measured up to 80 oC for sodium and potassium carbonate solutions and the results were compared with the model of Weisenberger and Schumpe (1996). Overall gas phase mass transfer coefficients were measured with a string of discs apparatus for 5-30 wt-% sodium carbonate solution and for 5-50 wt-% potassium carbonate solution up to 70 oC. The Henry’s law constants measured in this study agree well with literature. The Henry’s law constants increase with temperature and carbonate concentration. The model of Weisenberger and Schumpe agrees well with measured data at 25 oC, but it underestimates the Henry’s law constant at higher temperatures. Overall mass transfer coefficients for potassium and sodium carbonate increase with temperature but decreases with concentration at high concentrations. Potassium carbonate solutions have higher overall mass transfer coefficients than sodium carbonate. Adding amines as promoter gives decade higher absorption rates compared to pure carbonate solutions.

c 2008 CC BY-NC-ND license.

2009 Elsevier © ElsevierLtd. Ltd.Open Allaccess rightsunder reserved

Keywords: potassium carbonate; sodium carbonate; N2O solubility; overall mass transfer coefficient

1. Introduction Chemical absorption is a widely used technology for the removal of undesired components from gas streams and it is one of the most promising technologies for CO2 capture from flue gases. In the beginning of the 20th century, carbonate solutions were used in dry ice plants to separate CO2 from flue gas. After alkanolamines were introduced the use of sodium carbonate solutions rapidly decreased. This was mainly because of the CO2 absorption being faster with alkanolamine solutions and that very high CO2 removal efficiencies could be achieved. However, carbonate systems have been used for special applications for many decades and recently the use of carbonates for post combustion CO2 removal has gained renewed interest because of the potentially low energy requirements of the process (Liang et al. 2004, Cullinane and Rochelle, 2004 and Resnik et al., 2004; Knuutila et al. 2006, 2008).

* Corresponding author. Tel.: +47 735 941 00; fax: +47 735 940 80. E-mail address: [email protected]. doi:10.1016/j.egypro.2009.01.134

2

Hanna Knuutila/ Energy Procedia 00 (2008) 000–000 H. Knuutila et al. / Energy Procedia 1 (2009) 1011–1018

1012

In this paper the kinetics of sodium and potassium carbonate solutions are presented based on measurements with string of disc apparatus. The effect of promoting carbonate solutions with monoethanol amine (MEA) and methyl amino propylamine (MAPA) were tested. To predict physical solubility of CO2 into carbonate solutions, the solubility of N2O into sodium and potassium carbonate solutions was measured. The results are compared with literature data and the model of Weisenberger and Schumpe. 2. N2O Solubility Solubility of N2O into electrolyte solutions is normally modeled with the model of Weisenberger and Schumpe (Schumpe, 1993; Weisenberger and Schumpe, 1996). In this model the N2O solubility can be calculated from the equations H ¬ log G , water hi hG ci H G ®

(1)

hG hG ,o hT T 298.15 K

in which HG,water and HG are gas solubilities in water and in salt solution respectively, ci is the concentration of component i,hi is an ion-specific parameter, hT a gas specific parameter for the temperature effect and hG,o is a gas specific parameter. For N2O the model is valid until 40 oC. In this paper the solubility of N2O into sodium and potassium carbonate are measured and the results are compared to the model of Weisenberger and Schumpe.

2.1. Experimental procedure and calculations The solubility apparatus, shown in figure 1, consisted of a stirred jacketed glass reactor with volume of 791.1 cm3 and a stainless steel gas holding vessel with volume (1.17 10-3) m3. A desired amount of solvent was weighted in and transferred to the reactor. The solution was thereafter degassed at ambient temperature by vacuum until vapourliquid equilibrium was established. After degassing the reactor was heated to the desired temperature. Temperature and pressure in the reactor and gas holding vessel were recorded, and N2O gas was added. After equilibrium was established the temperature and pressure in the reactor and N2O gas vessel were recorded. The same apparatus have been used also by Hartono et al. (2008). At equilibrium the partial pressure of N2O (PN2O) in the reactor is difference between the total pressure (PR) and the water vapor pressure of the solvent ( PSo ) PN2O PR PSo (2) When the total volume of the reactor (VR) and the amount of solvent in the reactor (ms) are known the amount of N2O in the gas phase can be calculated as: PN O VR mS S S

(3) nNG2 O 2 zRTR Here TR is the reactor temperature and z the compressibility factor of N2O at equilibrium temperature and pressure. The compressibility factor was calculated using the Peng-Robinson Equation of State. The pressure and temperature difference in the gas vessel before and after adding N2O to the reactor can be used to calculate the total amount of N2O in the reactor: V P P ¬ V V 1 V 2 (4) nNadded 2O R T z T z ® V1 1

V2 2

Here VV is the volume of gas holding vessel, PV the pressure, TV the temperature, z the compressibility factor and R the universal gas constant. Sub-script 1 is initial condition and 2 is final condition. Now the amount of N2O absorbed in the liquid phase can be calculated as the difference between N2O added and N2O in the gas phase G nNadded O nN 2 O c NL O 2 (5) 2 VL

Author name / Energy Procedia 00 (2008) 000–000 H. Knuutila et al. / Energy Procedia 1 (2009) 1011–1018

1013

3

In equation VL is the liquid volume in the reactor. The solubility can be expressed by Henry’s law constant according to the equation PN2O H N 2O cNL 2O (6)

Figure 1. The experimental set-up for N2O solubility experiments.

2.2. Results

The Solubility of N2O in water was measured first to ensure that the system works correctly. The agreement with the literature was good. The solubility of N2O into sodium carbonate at a temperature of 25 oC was measure by Hikita et al. (1974). The new data agrees well with their data as shown in figure 2. From the figure it can also be seen that the model of Weisenberger and Schumpe (1996) and Danckwerts (1970) agrees well with the experiments. In figures 3 and 4 results with sodium carbonate and potassium carbonate at different temperatures are shown. From those figures it can be seen that the model of Weisenberger and Schumpe underestimates the effect of temperature at higher concentrations. At very low concentrations (1 wt-% Na2CO3) the model predicts the experimental results well, but at higher concentrations the model under-predicts the solubility at higher temperatures. For both potassium and sodium carbonate the model predicts the solubility well at 25 oC. It should be noted that in the original paper of Weisenberger and Schumpe they mention that the model for the solubility of N2O into carbonate solutions should be valid up to 40 oC. Solubility of N2O in Na 2CO3 -solutions at 25 o C 30000 Hikita et al. (1974)

Henry's law constant (kPa m 3/kmol)

Experimental data

25000

Weisenberger and Schumpe 1996 Danckw erts (1970)

20000

15000

10000

5000

0 0

4

8

12

16

20

wt-% Na2CO3

Figure 2. Henry’s law constant for N2O in sodium carbonate solution at temperature of 25 oC.

24

4


1014

Henry's law constant for Na2CO3

3

Henry's law constant [kPa m /kmol]

100000

10000

1 w t-%

5 w t-%

10 w t-%

15 w t-%

20 w t-%

Weisenberger and Schumpe

1000 290

300

310

320

330

340

350

360

Temperature [K]

Figure 3. Henry’s law constant for N2O into sodium carbonate solutions.

Henry's law constant for K2CO3

3

Henry's law constant [kPa m /kmol]

100000.0

10000.0

5 w t-% 10 w t-% 20 w t-% 30 w t-% Weisenberger and Schumpe (1996)

1000.0 290

300

310

320

330

340

350

360

Temperature [K]

Figure 4. Henry’s law constant for N2O into potassium carbonate solutions.

3. Kinetic experiments with string of discs Reaction rate measurements with pure unloaded potassium and sodium carbonate have been performed earlier by Hitchock and Cadot (1935). They measured initial steady state absorption rates at 30 oC using batch absorption apparatus. Their data can not be directly used to compare with the new data, but a qualitative comparison can be made. For loaded carbonate solutions a lot of experimental data is available (Comstock and Dodge, 1937; Harte et al., 1933; Williamson and Mathews, 1924; Roper, 1955). Most of the experiments are done at low carbonate concentrations, low temperatures and with high carbonate conversions to bicarbonate. In this study, kinetics measurements with unloaded sodium and potassium carbonate solutions were performed with a string of discs apparatus shown in figure 5 (Ma’mun, 2007). The string of discs is made of unglazed ceramic material and has 43 discs with a diameter of 0.015 m and thickness of 0.004 m each. The string of discs contactor is operated in a counter-current mode with the liquid entering to the top. The liquid flows through a tube that ends in a jet. The liquid is removed with a small tube from the funnel. The gas is fed from the bottom with long enough distance from the first discs to calm the gas flow. The liquid and gas flows are independently adjusted by using a liquid pump and a gas blower. 3.1. Experimental procedure and calculations

Absorption rates of CO2 into sodium carbonate solutions from 5 to 30 wt-% and potassium carbonate solutions from 5 to 50 wt-% were measured at temperatures from 25 to 70 oC. The solutions were prepared by mixing analytical grade chemicals with deionized water. The CO2 analyzer was calibrated every morning before the


1015

5

experiments with calibration mixtures of CO2 and N2. An unloaded sodium or potassium carbonate solution was passed through the column with a flowrate of ~51 mL/min. For every concentration and temperature it was confirmed that this flowrate was high enough so that the absorption flux is independent on the liquid flowrate. After the column reached desired temperature level a known mixture of CO2 and N2 was feed into the column. When the temperature and CO2-analyzer showed constant values for 10-25 minutes, the process was terminated. All the data was collected to computer, and average values were calculated over the 10-25 minutes to be used for calculations of reaction kinetics. The solute balance over the entire system is used to calculate the total absorption flux. It is given by the flow of the gas into the system minus the amount going out of the system through the bleed. The overall absorption rate in the disc contactor is given by in out (7) N CO N CO N CO 2

where N

out CO2

2

2

in

is the CO2 out from the system (mol/s m2), N CO2 is the CO2 in to the absorption column (mol/s m2) and

NCO2 is absorbed CO2. The amount of the solute entering into the system can be obtained from the mass flow controller reading. The amount of CO2 going out of the system is obtained from out yCO out in 2 N CO2 N N2 (8) pvap out yCO 1 2 P out

Where yCO2 is the CO2 concentration in the gas phase in the bleed. The overall mass transfer coefficient for the whole string of discs can be calculated from equation N CO 2 K ov ,G pCO2 , LM

(9)

Since the pressure is not uniform throughout the column the driving force for absorption is calculated in terms of the logarithmic driving force over the whole string of discs pCO2 , LM

in pCO 2

in pCO 2 ,i

ln

out pCO 2

in pCO 2

in pCO 2 ,i

out pCO 2

out pCO 2 ,i

out pCO 2 ,i

in

(10)

out

In the equation the equilibrium pressures ( pCO2 ,i , pCO2 ,i ) can be considered to be zeros since the solution is unloaded when it enters the column and the amount of absorbed CO2 in the column is small. Heated cabin

P-1 T

Liquid tank

I-2

P-4

P-8 T

Liquid flow Flow meter

E-3

CO2 analyser P-3

V-1

P-7

E-7 P-22

T

P-5

T

V-2

CO2

P-11 I-3 Venturi meter

Liquid tank

Figure 5. Experimental set-up of the string of discs apparatus.

N2

6


1016

3.2. Results and discussion The results for sodium and potassium carbonate solutions are shown in figures 6 and 7. For potassium carbonate it can be seen that the overall mass transfer coefficient first increases and reaches a maximum between 10 and 20 wt-% and then starts to decrease. These findings agree with Comstock and Dodge (1937) and Hithcock and Cadot (1935). For sodium carbonate this behaviour is not seen, but this is most likely due to the fact that the maximum may be reached at lower concentrations than 5 wt-%. The absorption rates for sodium carbonate are lower than with potassium carbonate even when molar concentrations are used instead of mass fractions. This has been illustrated in figure 8 where average mass transfer coefficients for sodium and potassium carbonate are shown at a temperature of 50 oC at different concentrations. The mass transfer coefficients are almost the same at very low concentration, but the absorption rate of sodium carbonate decreases faster than potassium carbonate with increasing concentration. This difference might stem purely from the differences in viscosities. Sodium carbonate has higher viscosity than potassium carbonate with same concentration and the concentration dependency is stronger with sodium carbonate with potassium carbonate. The difference between potassium and sodium carbonate solutions seems also to be increasing with increasing concentration. Experimental data of Hithcock and Cadot (1935) supports the results of this study that the difference in absorption rates between potassium carbonate and sodium carbonate depends on the total concentrations and is not constant as reported by Comstock and Dodge (1927). The absorption rates increase with temperature. Based on the data there seems to be only a small difference between the carbonates. When temperature increases from 50 to 70 oC the overall mass transfer coefficient of sodium carbonate seems to increase with factor 1.9-2 while the increase for potassium carbonate is 1.5-1.8. The increase in mass transfer coefficient with temperature for potassium carbonate seems to be higher at lower temperatures. At 70 oC the absorption rate is close to the maximum absorption rate which according to literature is achieved at a temperature of 75 oC (Comstock and Dodge, 1937; Williamson and Mathews, 1924). 0.001 5 wt-% K2CO3 10 wt-% K2CO3 20 wt-% K2CO3 30 wt-% K2CO3 40 wt-% K2CO3

Kg,ov [m/s]

50 wt-% K2CO3

0.0001 0.0029

0.003

0.0031

0.0032

0.0033

0.0034

Temperature [1/K]

Figure 6. Overall gas phase mass transfer coefficient for Potassium carbonate solutions from 5-50 wt-% at temperatures of 25-70 oC.

Kg,ov [m/s]

0.001

0.0001

30 wt-% Na2CO3 20 wt-% Na2CO3 5 wt-% Na2CO3 10 wt-% Na2CO3

0.00001 0.0029

0.003

0.0031

0.0032

Temperature [1/K]

Figure 7. Overall gas phase mass transfer coefficient for sodium carbonate solutions from 5-30 wt-% at temperatures of 40-70 oC.


1017

7

Kov [m/s]

0.001

0.0001

K2CO3 Na2CO3 0.00001 0

1

2

3

4

5

6

Carbonate concentration (m ol/dm 3)

Figure 8. Overall gas phase mass transfer coefficient for sodium and potassium carbonate solutions at 50 oC.

The possibility to increase the absorption rates by adding amines has been studied for decades (Cullinane and Rochelle, 2004; Leder, 1970; Tseng and Savage, 1988). The effect of promoting carbonate solutions with 10 wt-% MAPA and 10 wt-% MEA are shown in figure 9 and 10. Adding amines increases the absorption heavily giving overall mass transfer coefficients which are a decade higher than in pure carbonate systems. The promoting effect is higher with potassium carbonate solutions than with sodium carbonate solutions (figure 10). Both 20 wt-% (2.3 mol/dm3) and 10 wt-% (1.0 mol/dm3) sodium carbonate solutions give lower mass transfer coefficients than 20 wt% (1.7 mol/dm3) potassium carbonate. The relative absorption rate difference in unpromoted systems seems to be approximately same as for promoted systems. 0.1

MAPA 10 % 10 wt-% MAPA+20 wt-% Na2CO3 1+ wt-% MAPA +20 wt-% Na2CO3

K_ov [m/s]

10 wt-% MAPA+20 wt-%K2CO3

0.01

0.001 0.0029

0.003

0.0031

0.0032

0.0033

0.0034

1/T [1/K]

Figure 9. Overall gas phase mass transfer coefficients for sodium and potassium carbonate solutions promoted with 10 wt-% MAPA. 0.01

MEA 10 wt-% 10wt-% MEA+20 wt-% K2CO3 10 % MEA + 20 wt-% Na2CO3

K_ov [m/s]

10 wt-% MEA+10 wt-% Na2CO3

0.001 0.0029

0.0030

0.0031

0.0032

0.0033

0.0034

1/T [1/K]

Figure 10. Overall gas phase mass transfer coefficient for sodium and potassium carbonate solutions promoted with 10 wt-% MEA.

8

1018


4. Conclusions The Henry’s law constants for physical solubility of N2O were measured up to 80 oC for sodium and potassium carbonate solutions and the results were compared with the model of Weisenberger and Schumpe (1996). Overall gas phase mass transfer coefficients were measured with a string of discs apparatus for 5-30 wt-% sodium carbonate solution and for 5-50 wt-% potassium carbonate solution up to 70 oC. The Henry’s law constants measured in this study agree well with literature. The Henry’s law constants increase with temperature and carbonate concentration. The model of Weisenberger and Schumpe agrees well with measured data at 25 oC, but it underestimates the Henry’s law constant at higher temperatures. Overall mass transfer coefficients for potassium and sodium carbonate increase with temperature but decreases with concentration at high concentrations. Potassium carbonate solutions have higher overall mass transfer coefficients than sodium carbonate. Adding amines as promoter gives decade higher absorption rates compared to pure carbonate solutions.

References Comstock, C., Dodge, B. Rate of Carbon Dioxide Absorption by Carbonate Solutions in a Packed Tower. Industrial and engineering chemistry, 29 (1937), pp. 520-529. Cullinane, J. T. and Rochelle G. T. Carbon dioxide absorption with aqueous potassium carbonate promoted by piperazine. Chemical Engineering Science 59 (2004), pp. 3619 – 3630. Danckwerts, PV. Gas-liquid reactions. McGraw-Hill inc, 1970. Harte, C and Baker, E. Absorption of Carbon Dioxide in Sodium Carbonate-Bicarbonate Solution, II. Rate of Absoprtion. Industrial and engineering chemistry 25 (1933), p 1128-1132 Hartono, Ardi, Juliussen, Olav, and Svendsen, Hallvard F. Solubility of N2O in Aqueous Solution of Diethylenetriamine . J. Chem. Eng. Data (2008), In Press. Hikita, H., Asai, S., Ishikawa, I. and Esaka, N. Solubility of nitrous oxide in sodium carbonate-sodium bicarbonate solutions at 25 oC and 1 atm. Journal of Chemical engineering data 19 (1974), pp. 89-92. Hitchcock L.B., and Cadot, H.M. Rate of absorption of Carbon Dioxide. Industrial and Engineering Chemistry 27 (1935), pp. 728- 732. Knuutila H., Anttila M., Børresen E., Juliussen O., Svendsen H.F., CO capture with sodium carbonate. Eighth International Conference on 2 Greenhouse Gas Technologies, Trondheim, Norway, June 19-22.2006. Knuutila, H., Svendsen, H.F., Anttila, M. CO2 capture from coal-fired power plants based on sodium carbonate slurry; a systems feasibility and sensitivity study, International Journal of Greenhouse Gas Control (2008) In Press, Leder, F. The absorption of CO2 into chemically reacting solutions at high temperatures. Chem. Eng. Sci 26 (1971), pp. 1381-1390. Liang, Y., Harrison, D. P., Gupta, R. P., Green, D.A. and McMichael W.J. Carbon Dioxide Capture Using Dry Sodium-Based Sorbents. Energy and Fuels 18 (2004), pp. 569-575. Ma'mun, S., Dindore, V.Y., and Svendsen, H.F.. Kinetics of the Reaction of Carbon Dioxide with Aqueous Solutions of 2-((2Aminoethyl)amino)ethanol. Ind. Eng. Chem. Res. 46 (2007), pp. 7849 – 7850. Resnik, K. P., Yeh, J. T. and Pennline, H. W. Aqua ammonia process for simultaneous removal of CO2, SO2 and NOx. Int. J. Environmental Technology and Management 4 (2004), pp. 89 – 104. Schumpe, A. Estimation of gas solubilities in salt solutions. Chemical engineering science 48 (1993), pp. 153-158. Tseng, P.C. and Savage D.W. Carbon dioxide absorption into promoted carbonate solutions. AIChe Journal 34 (1988), pp. 922-931. Wesenberger, S. and Schumpe, A. Estimation of gas solutbilites insalt solutions at temperatrues from 273 K to 363 K. AIChE Journal l 42 (1996), pp. 298- 300. Williamson R.V. and Mathews, J.H. Rate of absorption and equilibrium of carbon dioxide in alkaline solutions. Industrial and Engineering Chemistry 16 (1924), pp. 1157-1161