Energy Procedia 00 (2008) 000–000
Energy Procedia
www.elsevier.com/locate/XXX
GHGT-9
Mass-transfer efficiency of a spray column for CO2 capture by MEA
Jeffery Kuntz, Adisorn Aroonwilas *
Faculty of Engineering, University of Regina, Regina, Saskatchewan, Canada S4S 0A2 Elsevier use only:Received date here; revised date here; accepted date here
Abstract
A parametric study of carbon dioxide (CO2) absorption performance into an aqueous solution of monoethanolamine (MEA) in the spray column was carried out experimentally over wide ranges of process conditions. The performance of the spray was interpreted in terms of the overall mass transfer coefficient (KGae) and was found to vary with process parameters, including gas flow rate, liquid flow rate, CO2partial pressure, MEA concentration, CO2loading, and size of spray nozzle. The performance of the spray column was compared to that of a packed column and showed a promise for CO2capture application.
© 2008 Elsevier Ltd. All rights reserved
Keywords:CO2capture; CO2absorption; spray column; mass transfer coefficient; mass transfer area; alkanolamine
* Corresponding author. Tel: 1-306-337-2469. Fax: 1-306-585-4855.
E-mail address: [email protected].
1. Introduction
Gas absorption into an aqueous alkanolamine solution is the most well-established of the technologies available for carbon dioxide (CO2) capture [1]. However, its cost is still prohibitively high for the environmental application.
The reduction of cost can be attained through the use of proper gas-liquid contactors. For a number of years, the CO2
absorption using packed and tray columns has produced a large amount of published data with many types of solvents and column internals being tested [2-9]. whereas the CO2absorption using spray column has been studied by few and to date there has been little data published in the area of CO2capture [10]. Specifically in case of the CO2absorption using alkanoalmine such as monoethanolamine (MEA), there has been no data on spray absorber published. This work therefore examined the feasibility of using spray column for CO2 capture using MEA. The performance of spray column was evaluated experimentally under various conditions to reveal effects of process parameters, including CO2partial pressure in gas phase, gas flow rate, liquid flow rate, concentration of MEA, CO2 c
2009 Elsevier Ltd. All rights reserved.
Energy Procedia 1 (2009) 205–209
www.elsevier.com/locate/procedia
doi:10.1016/j.egypro.2009.01.029
loading of absorption solution, and size of spray nozzle. A set of absorption experiments using a packed column was also conducted to determine the performance of conventional column for comparison purposes.
2. Absorption Experiments
The gas absorption setup used in this work is shown in Figure 1. The experiments were conducted in two separate columns,i.e. spray and packed column. The spray column was constructed of acrylic plastic that was 0.55 m high and had an inside diameter (ID) of 0.10 m. The spray column was operated with one of three 316 stainless steel nozzles manufactured by BETE Industrial Spray Nozzles (models P-20, P-28, and P-40). The packed column was also constructed of acrylic plastic with an ID of 0.10 m but had a height of 0.80 m. The packed column was fitted with Mellapak 500Y structured packing provided by Sulzer Brothers Limited, Winterhur, Switzerland. The packing was installed with each layer rotated by 90° with respect to the previous one. The CO2 concentrations were measured using an infrared gas analyzer (Model 302WP, Nova Analytical Systems Inc.). The reading range of the analyzer was 0.0 to 20.0 % of CO2by volume with the accuracy of ±2% of the full-scale reading.
Figure 1 Schematic diagram of CO2absorption apparatus
The experiments began by introducing a mixture of air and CO2at a desired flow rate and CO2concentration to the bottom of the column. The gas mixture once in the column passed through a dispersion outlet to disperse uniformly across the column. At the same time, the prepared liquid solution was pumped to the top of the column where it entered the column through either a spray nozzle (for the spray column) or the liquid distributor (for the packed column). This brought the gas and liquid into contact counter-currently, and the CO2in the gas phase was absorbed. The gas then carried out through the top of the column while the CO2rich solution was collected out of the bottom of the column in the liquid receiver. Once the system reached steady state, the CO2concentration of the exit gas was measured and a liquid sample was collected out of the bottom of the column. The liquid sample was analyzed for CO2loading and total concentration of MEA. Details of experimental setup and procedure can be found in Kuntz (2006) [11].
3. Results and Discussion
More than 400 runs were carried for both spray and packed columns over ranges of conditions as listed in Table 1. The mass transfer performance was determined in terms of the volumetric overall mass transfer coefficient (KGae) by using the following equation: [6]
¸¸¹
·
¨¨
©
§
¸¸
¹
·
¨¨
©
§
dZ
dY y
y P a G
K CO,G
*CO G , CO e I
G 2
2 2
(1)
whereGIis inert gas flow rate in kmol/m2-hr,P is total pressure on the system in kPa,Zis column height in m.,
G
y
CO,2 and *
CO2
y
are mole fraction of CO2in gas stream and equilibrium mole fraction of CO2, andY
CO,G2 is mole ratio of CO2in gas stream. Parametric results are given in Figure 2.
A ir C O2 G a s flo w m e te r s F e e d r e s e r v o ir
L iq u id r e c e iv e r
C O2 a n a ly z e r P a c k e d
c o lu m n
O ff g a s
S p r a y c o lu m n
Table 1 Test conditions of spray and packed columns
Figure 2Effects of parameters on spray performance(a)Effect of CO2partial pressure onKGae. (Nozzle = P-20, Liquid flow rate = 1.53 m3/m2-h, Gas flow rate = 382 – 764 m3/m2-h),(b)Effect of gas flow rate onKGaeat different liquid flow rates (Nozzle = P-28, PCO2=15 kPa, [MEA] =5M),(c)Effect of liquid flow rate onKGaeP-28 nozzle (PCO2=15 kPa, [MEA] =5M, Gas flow rate=382 m3/m2-h),(d)Effect of the concentration of the MEA onKGaeat CO2loadings of 0.25 mole/mole (Nozzle=P-20, PCO2=15 kPa),(e)Effect of CO2loading on KGaefor different nozzles (for P-20 nozzle, [MEA] =3M, Gas flow =76 m3/m2-h, Liquid flow = 1.53 m3/m2-h; for P-28 nozzle, [MEA]
=5M, Gas flow =382 m3/m2-h, Liquid flow = 4.59 m3/m2-h; and for P-40 nozzle, [MEA] =5M, Gas flow =382 m3/m2-h, Liquid flow
=10.32 m3/m2-h), and(f)Effect of the nozzle size onKGae(Gas flow rate=382 m3/m2-h, PCO2=15kPa, [MEA] =5 kmol/m3)
Parameter Condition
Absorption solvent Monoethanolamine (MEA) Gas Phase
Gas flow rate
Feed CO2partial pressure
Up to 764 m3/m2-h 5, 10, 15 kPa Liquid Phase
Liquid flow rate MEA concentration CO2loading Temperature
Up to 10.3 m3/m2-h 3, 5, 7 kmol/m3
0, 0.15, 0.25, 0.35, 0.45 mol/mol 25°C
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0 5 10 15 20
CO2partial pressure (kPa) KGae(kmol/m3*h*kPa)
0.0 mole/mole 0.15 mole/mole 0.25 mole/mole 0.45 mole/mole CO2 loading
0.15 0.25 0.35 0.45
0 200 400 600 800
Gas flow rate (m 3/m2-h) 3 kmol/m3 5 kmol/m3 7 kmol/m3 MEA Concentration KGae(kmol/m3*h*kPa)
0.00 0.40 0.80 1.20 1.60 2.00 2.40
1.9 2.4 2.9 3.4 3.9 4.4 4.9
Liquid flow rate (m3/m2-h) KGae(kmol/m3*h*kPa)
0 mole/mole 0.15 mole/mole 0.25 mole/mole 0.35 mole/mole 0.45 mole/mole CO2Loading 0.00
0.50 1.00 1.50 2.00 2.50
0 200 400 600 800 1000
Gas flow rate (m3/m2-h) KGae(kmol/m3*h*kPa)
1.91 m3/m2-h 4.59 m3/m2-h Liquid flow rate
0 1 2 3 4 5 6 7
0 0.1 0.2 0.3 0.4 0.5
CO2loading (mole CO2/mole MEA)
KGae(kmol/m3*h*kPa) P-20P-28
P-40 Nozzle size
0.00 0.80 1.60 2.40 3.20 4.00 4.80 5.60 6.40
1.9 3.9 5.9 7.9 9.9
Liquid flow rate (m3/m2-h) KGae(kmol/m3*h*kPa)
P-28 0 mole/mole P-28 0.15 mole/mole P-28 0.25 mole/mole P-28 0.35 mole/mole P-28 0.45 mole/mole P-40 0 mole/mole P-40 0.15 mole/mole P-40 0.25 mole/mole P-40 0.35 mole/mole P-40 0.45 mole/mole CO2Loading
(a)
(b)
(c)
(d)
(e)
(f)
Effect of CO2 partial pressure. KGae decreases with CO2 partial pressure. By considering mass flux of CO2
absorption (NCO2), an increase in CO2partial pressure leads to an increasing amount of CO2transferred into liquid phase. However, the increasing mass flux occurs in a lower extent compared to the change in partial pressure, causingKGaeto reduce as partial pressure increases. Note that the effect of CO2partial pressure becomes less at the CO2loading of more than 0.25 mole/mole and the partial pressure of more than 10 kPa. This may be caused by the restricted diffusion and amount of reactive MEA in the liquid phase. The mass transfer may be mainly controlled by CO2 reaction in the liquid, thus resulting in only a small change in the amount of CO2 absorbed as the partial pressure increases.
Effect of gas flow rate. KGae increases with gas flow rate to a certain point and then remains constant. This suggests the gas-phase controlled mass transfer takes place at low gas flow rates (below 300 m3/m2-h) and the liquid-phase controlled mass transfer takes over at high gas flow rate. In general, as the gas flow increases the amount of CO2molecules available for the absorption increases. This would lead to a higher mass transfer flux.
However, the overall rate of gas absorption is not only dependent upon the gas flow rate, but also the liquid flow rate and availability of reactive MEA in the liquid which as seen in this case controls the rate of mass transfer after the gas flow rate reaches the point of 300 m3/m2-h.
Effect of liquid flow rate. KGae increases with liquid flow rate. This is because increasing the liquid flow increases effective interfacial area (ae), between liquid and gas. Note that KGaeincreases more rapidly at low flow rates compared to at high flow rates. The rapid increase was caused by 1) a reduction in size of spray droplets from larger diameter to smaller diameter, thus resulting in an increase in droplet surface area per unit volume of dispersed liquid and 2) an increase in number of droplets produced by the nozzle and also the surface area available for mass transfer. At the high liquid flow rate, the reduction in droplet size by the increasing liquid flow is insignificant, leaving the increasing number of spray droplets to be the primary factor that defined the lower increase in mass transfer performance.
Effect of MEA concentration.KGaeincreases with MEA concentration (up to 7 kmol/m3). This is due to the fact that the increasing MEA concentration yields a higher amount of the active MEA available to diffuse towards the gas-liquid interface and react with CO2. This finding differs from the behavior observed in the packed column in that the KGae of packed column decreases by 5% for every molarity of MEA increasing beyond 3 kmol/m3.[9] Such decrease in KGae is caused by an increase in solution viscosity. This shows that the solution viscosity is more influential on the effective area in the spray column than in the packed column.
Effect of CO2loading.KGaedecreases with CO2loading. This is due to the fact that as the CO2loading increases the amount of active MEA decreases, causing theKGaeto decrease.
Effect of nozzle size. Three nozzles with different orifice sizes were tested,i.e.P-20, P-28, and P-40. It was found thatKGaeof a larger nozzle is lower that of a smaller nozzle at the low end of liquid flow rate. This is because the spray of the lager nozzle is not fully developed, resulting in a lower effective area (ae). As the liquid flow rate increases, the spray is more fully developed with the smaller liquid droplets that offer higherae, causing theKGaeto increase accordingly.
Performance comparison (spray versus packed column). The comparison was made for the spray column fitted with P-40 nozzle and the packed column fitted with Mellapak 500Y. It was found in Figure 3 that the spray column provides a much higher KGae. This is because the spray nozzle offers a much higher gas-liquid interfacial area than the packing does.
Figure 3Mass transfer performance comparison between packed and spray columns
4. Conclusions
The CO2absorption performance of the spray column is affected by process parameters.KGaedecreases with CO2
partial pressure and CO2loading, but increases with liquid flow rate and MEA concentration.KGaeincreases with gas flow rate and then remains constant when the gas flow rate exceeds 300 m3/m2-h. Different nozzle sizes offer different ranges of KGae due to the difference in hydrodynamic capacity of each spray nozzle. The spray column offers a superior CO2absorption performance to the packed column due to a greater interfacial area provided.
Acknowledgment
Authors gratefully acknowledge the financial support received from the Natural Sciences and Engineering Research Council of Canada (NSERC) and the structured packing donated by the Sulzer Chemtech, Winterthur, Switzerland.
Reference
1. J. Davison, P. Freund and A. Smith, IEA Greenhouse Gas R&D Programme. (2001), ISBN 1 898373 28 0.
2. R.E. Treybal, Mass-Transfer Operations: 3rdEdition. McGraw-Hill, Singapore, 1980.
3. A. Aroonwilas, High Efficiency Structured Packing for CO2Absorption Using 2-Amino-2-methyl-1- propanol (AMP): M.A.Sc. Thesis, University of Regina, Regina, Saskatchewan, Canada, 1996.
4. A. Aroonwilas, A. Veawab and P. Tontiwachwuthikul, Industrial Engineering Chemical Research. 38(1999) 2044-2050.
5. A. Aroonwilas, P. Tontiwachwuthikul and A. Chakma, Separation and Purification Technology. 24(2001) 403-411.
6. A. Aroonwilas and A. Veawab, Industrial and Engineering Chemistry Research. 43(9) (2004) 2228-2237.
7. J.T. Yeh, H.W. Pennline and K.P. Resnik, Energy Fuels. 15(2) (2001) 274 -278.
8. D. DeMontigny, P. Tontiwachwuthikul and A. Chakma, Industrial and Engineering Chemistry Research.
44(15) (2005) 5726-5732.
9. R.F. Strigle, Random Packings and Packed Towers: Design and Applications, Gulf Publishing Company, Houston, Texas, 1987.
10. N.K. Yeh and G.T. Rochelle, AIChE Journal. 49(9) (2003) 2363-2373.
11. J. Kuntz, “Absorption of Carbon Dioxide in a Spray Column” Master of Applied Science Thesis, University of Regina, 2006.
0.00 1.00 2.00 3.00 4.00 5.00
4.5 5.5 6.5 7.5 8.5 9.5
Liquid flow rate (m3/m2-h) KGae(kmol/m3*h*kPa)
0.15 mole/mole 0.45 mole/mole Packed 0.15 mole/mole Packed 0.45 mole/mole CO2Loading