Energy

Procedia
Energy
EnergyProcedia
Procedia 00 (2008)211–215
1 (2009) 000–000
www.elsevier.com/locate/XXX
www.elsevier.com/locate/procedia

GHGT-9

CO2 absorption into MEA-AMP blend: mass transfer and absorber

height index
Anindo Dey, 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 CO2 absorption into aqueous solutions of monoethanolamine (MEA)-2-amino-2-methyl-1-propanol

(AMP) blends was conducted to obtain a much-needed mass transfer data for process design. It was found that the overall mass
transfer coefficient (KGae) generally increases with liquid flow rate, temperature and total amine concentration (except at high
CO2 loading of above 0.35 mol/mol). It however decreases with CO2 partial pressure and CO2 loading. Higher MEA
concentration in the MEA-AMP blends causes KGae to increase, but this trend reverses at high CO2 loadings. An empirical
correlation of KGae was subsequently developed as a function of process parameters.
©c 2008 ElsevierLtd.
2009 Elsevier Lt Open access under CC BY-NC-ND
d license.

Keywords: CO2 absorption; blended amine; packed column; structure packing; mass transfer

1. Introduction

Carbon dioxide (CO2) contributes approximately 60% of the total greenhouse gas (GHG) emissions on a global
warming potential basis. Capturing and sequestering CO2 were voiced in 1992 at the United Nations Framework
Convention on Climate Change (UNFCCC), held in Rio de Janeiro to stabilize the atmospheric concentrations of
GHGs to the levels that would prevent dangerous alterations to the climate system. Today, the most mature and
reliable technology for the post combustion CO2 capture is gas absorption using chemical solvents.
Monoethanolamine (MEA) is received the most attention, primarily due to its high rate of reaction [1-2]. MEA
however has several drawbacks including high energy requirement for solvent regeneration. The problem of high
energy requirement can be reduced by using MEA-based blends that would provide satisfactory absorption
performance compared to MEA but would require less energy of regeneration, thus resulting in lower operating
costs and a lower energy penalty.
Our work focuses on blends of MEA and 2-amino-2-methyl-1-propanol (AMP). MEA-AMP blends were
reported to have a promise in an aspect of CO2 absorption performance compared to MEA [3]. To date, there have
been no comprehensive literature data for CO2 absorption in MEA-AMP blends in packed beds that could reveal the
benefits of using MEA-AMP blends at the concentration used at an industrial scale. It was therefore our objective to

* Corresponding author. Tel.: +1-306-337-2469; fax: +1-306-585-4855.

E-mail address: aroonwia@uregina.ca.
doi:10.1016/j.egypro.2009.01.030
 212 A. and
Dey Dey,Aroonwilas
A. Aroonwilas / Energy
/ Energy Procedia
Procedia 00 1(2008)
(2009)000–000
211–215

investigate CO2 absorption performance of MEA-AMP blends at various molar mixing ratios using a bench-scale
absorber packed with structured packing.

2. Experiments

Figure 1 shows a schematic diagram of the gas absorption setup used in this study. An absorption column with 20
mm in diameter and 500 mm in height was made from acrylic plastic. The stainless steel DX structured packing
(Sulzer Brothers, Switzerland) has 900 m2/m3 total surface area and packing geometry as follows: 60o crimp angle,
0.775 void fraction, 2.9 mm crimp height and 6.4 mm corrugated base. The IR analyzer (Model No. 3C1D, Nova
Analytical Systems Inc.) for CO2 constant analysis had a range of 0-20% CO2 with an accuracy of +2% of full scale.
Details of the setup can be found in Dey [4].
F
Vent
CO 2 Mass
Flowmeter

G as Line
Sampli ng Line N 2 Mass
Flowmeter
Thermocouple
N2 cylinder
T
Sparger

CO2 IR Analyzer
CO2 cylinder
Sampling Line

FI

Vapor
Condenser
Hot Water Bath Rotameter
Gas Li ne

Amine Holding
Tank
Cold Water
Liquid Line

Bath

Packed Column

S ampling Point
Pump

Figure 1 A schematic diagram of gas absorption apparatus

The absorption experiments began by introducing a CO2-N2 gas mixture, set at desired flow rates and CO2 partial
pressure, to the absorber bottom. Simultaneously, the amine (MEA or MEA-AMP) solution was pumped to the
absorber top. The flow rate of the amine solution was calibrated manually with a stopwatch and a volumetric flask.
The column operation continued until a steady-state condition was reached for both gas- and liquid-sides. Liquid
samples were collected from the bottom of the column and analyzed for their amine concentration and CO2 content.
CO2 concentrations in the gas stream at the top and bottom of the absorber were measured by the IR analyzer. The
CO2 absorption performance of each experimental run was calculated as an overall mass-transfer coefficient (KGae)
in kmol/m3.hr.kPa by using the equation below.

Ginert Yin  Yout

K G ae (1)
y in  y out z
P[ ]
y
ln in
y out
where Ginert is the inert gas molar velocity (kmol/hr.m2), Yin and Yout are the mole ratios of CO2 in feed gas and exit
gas, respectively, P is the system pressure (kPa), z is the packing height (m), and yin and yout are the mole fractions of
CO2 in feed gas and exit gas, respectively.

3. Results and discussion

A total of 175 absorption runs were carried out to investigate the behavior of CO2 absorption into aqueous
solutions of MEA-AMP blends under wide ranges of operating conditions summarized in Table 1. Mass balance
 A.Deay
Dey,and
A. Aroonwilas
Aroonwilas/ /Energy
EnergyProcedia
Procedia1 00
(2009) 211–215
(2008) 000–000 213

error was within +10% based on both gas and liquid sides. All experimental data obtained from experiments can be
found in Dey [4] and parametric results are given in Figure 2.

Table 1 Experimental Conditions

Gas flow rate (kmol/m2.hr) 100

3 2
Liquid flow rate (m /m -hr) 2.5 – 5.0
Inlet CO2 (%) 5 - 15
MEA concentration (kmol/m3) 3-5
Molar ratio of amine Blends (MEA:AMP) 1:0, 2:1, 1:1, 1:2, 1:5
CO2 loading (mol CO2/mol amine) 0.0 – 0.59
Temperature (K) 298 - 323

1.6
0.40
KGae (kmol/m .hr.kPa)

1.2 0.35

CO2 Loading = 0.0 mol/mol K Ga e (km ol/m3 .hr.kPa) 0.30

3

0.8 0.25
= 0.15
0.20
= 0.25
0.4
= 0.35 0.15
= 0.45
= 0.54 0.10
0.0 = 0.59 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
295 300 305 310 315 320 325 3 2
Flow rate (m /m -hr)
Temperature (K)
MEA:AMP=1:2 MEA:AMP=1:1 MEA:AMP=2:1 MEA

1.50 (a) (d)

1.20 1.8 323 K

KGa e (kmol/m .hr.kPa)

1.6 313 K
KGae (kmol/m .hr.kPa)

0.90 303 K
1.4
3

298 K
3

0.60 1.2

1.0
0.30
0.8

0.00 0.6
0.00 0.10 0.20 0.30 0.40 0.50 0.60 2 1 1 0
CO2 Loading (mol/mol) 1 1 2 1
MEA-AMP=1:1 MEA-AMP=1:2
MEA-AMP=2:1 MEA Only AMP-MEA Ratio
(b) (e)
1.4
0.45
1.2
0.40
KGae (kmol/m .hr.kPa)
KGae (kmol/hr.m.kPa)

1.0
0.35
3

0.8
3

0.6 0.30

0.4
0.25
0.2
0.20
0.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
4 5 6 7 8 9 10 11 12 13 14 15 16 3
Concentration (kmol/m )
Partial Pressure CO 2 (kPa)
MEA Only MEA:AMP=2:1
MEA only MEA:AMP=1:1
MEA:AMP=1:1 MEA:AMP=1:2
MEA:AMP=1:2 MEA:AMP=2:1
(c) (f)

Figure 2 Parametric effects on mass-transfer coefficient (a) Effect of temperature (molar ratio of MEA/AMP is 1:1), (b) Effect of CO2 loading on
KGae at 298K, (c) Effect of CO2 partial pressure at 0.35 mol/mol CO2 loading, (d) Effect of liquid flow rate at 0.35 mol/mol CO2
loading, (e) Effect of MEA-AMP ratio at 0.00 mol/mol CO2 loading, and (f) Effect of amine concentration on KGae at 0.35 mol/mol CO2
loading.
 214 A. and
Dey Dey,Aroonwilas
A. Aroonwilas / Energy
/ Energy Procedia
Procedia 00 1(2008)
(2009)000–000
211–215

Effect of solution temperature: KGae increases with solution temperature for all blends at low to medium CO2
loading ( -0.35 mol/mol). This is due to the increase in both reaction kinetics and CO2 diffusivity. At the CO2
loadings above 0.35 mol/mol, the KGae decreases with temperature due to exhaustion of active amine molecules
participating in the reaction and a reduction in CO2 solubility.

Effect of liquid-phase CO2 loading: KGae decreases with the increasing CO2 loading. This is attributed to the
fact that the amount of reactive species of both MEA and AMP in the mixture decreases and consequently the
absorbing capacity decreases. At very high loadings, the diffusion rate of CO2 also decreased as the CO2 loading
value was drawn near the absorption capacity of the amine.

Effect of CO2 partial pressure: KGae decreases with the increasing partial pressure of CO2. Since the KGae
values represent mass transfer rate per unit driving force, an increase in the driving force due to the increase in CO2
partial pressure results in a decrease of KGae values. The absolute CO2 absorption rate (kmol/m3.hr), a product of
KGae and CO2 partial pressure, however increases with CO2 partial pressure.

Effect of liquid flow rate: KGae increases with liquid flow rate. This is because the increase in liquid flow rate
results in the spread of liquid on the packing surface and an increase in the effective interfacial area between liquid
and gas in the packing. Since the structured packing modules used in this study are oriented at 90o to one another, it
is possible there were a higher number of occurrences of quasi-stagnant pools and increased vortex formation in
between the series of structured packing modules. Figure 6 also shows that with the increase in liquid flow rates, the
increase in KGae values in MEA-AMP blends containing greater AMP content is less compared to that of MEA. This
can be explained by considering viscosity of amine. The viscosity of MEA ( PROPRONPROP3 solution) is
2.52 mPa.s [5] whereas AMP ( PROPRONPROP3 solution) is 67.7 mPa.s [6] for the temperature of 298 K.
The viscosity of MEA-AMP blends increases with the AMP content. For low CO2 loading ( PROPRO WKH
viscosity factor does not seem to dominate as the active molecules in amine participate in the CO2 absorption. For
higher CO2 loading, the number of active molecules decreases and the viscosity of the amine increases. At this
stage, hydrodynamics starts to influence the reaction regime.

Effect of blending ratio of MEA-AMP: MEA-AMP offers lower KGae values than MEA. This is because MEA
has higher rate constant (k2) than AMP. Higher MEA content in the MEA-AMP blends results in a higher KGae
value. The MEA-AMP blend with the ratio 2:1 has the highest KGae value while the blend with the ratio of 1:2 has
the lowest KGae compared to other mixing ratios tested. This is because the addition of MEA to AMP results in
higher CO2 absorption rate due to an increase in diffusivity, solubility, enhancement factor and overall rate constant
(kov) [7].

Effect of total MEA-AMP concentration: KGae for fresh MEA-AMP blends (  LQFUHDVHVZLWKWKHWRWDO
amine concentration. This is because more active MEA and AMP molecules increase the absorption of CO2.
However, at high CO2 loading (  VXFKLQFUHDVLQJWUHQGRI.Gae is not seen for 5 kmol/m3 solutions of MEA-
AMP=1:1 and 1:2. This may be because the difference between equilibrium CO2 loading and operational loading
values was gradually decreased for higher loadings even though an increase in concentration increased the amount
of reactive species available for absorption.

4. Empirical Correlation of KGae

An empirical correlation of KGae for MEA-AMP based CO2 absorption was developed using the obtained
experimental data. It is expressed as a function of process parameters as shown below.
F
CxCO2
K a min e e A( AMP / MEA ) e BD e
D
K G ae L1 e EC s e T (2)

where, KGae is the overall mass transfer coefficient (kmol/m3.hr.kPa); AMP/MEA is the molar ratio of the blends; D
is the CO2 loading (mol CO2/mol amine); xCO2 is the mole fraction of CO2; L1 is the liquid flow rate in m3/m2-hr; Cs
 A.Deay
Dey,and
A. Aroonwilas
Aroonwilas/ /Energy
EnergyProcedia
Procedia1 00
(2009) 211–215
(2008) 000–000 215

is the concentration of amine in kmol/m3; T is the temperature in Kelvin; A, B, C, D, E and F are the coefficients for
the respective parameters in the proposed equation. Multiple regression was used to find coefficients (Table 2) in the
equation for each amine at a confidence level of 98%. A composite model was also developed for all the blended
amines.

Table 2 Summary of coefficients for KGae correlation

Description MEA MEA:AMP=2:1 MEA:AMP=1:1 MEA:AMP=1:2 All Blends

2
Coefficient of Determination (R ) 0.9896 0.9760 0.9741 0.9735 0.9670
Standard Error 0.0853 0.1362 0.1502 0.1556 0.1533
Average Absolute Deviation 6.75% 10.05% 11.72% 12.27% 11.97%
K 12.40 275.27 901.67 1985.94 313.43
A 0.2405
B -3.9963 -4.7284 -5.1120 -5.3040 -5.0083
C -10.3685 -9.9526 -8.7948 -7.7273 -8.8458
D 0.9478 0.6120 0.5449 0.4429 0.5303
E 0.1719 0.0582 0.0038 -0.0519 0.0023
F -876.74 -1586.62 -1931.61 -2139.17 -1682.92

5. Conclusion

The CO2 absorption performance of MEA-AMP systems is influenced by many process parameters. KGae was
found to increase with liquid flow rate, temperature and total amine concentration (except at high CO2 loading of
above 0.35 mol/mol). It however decreases with CO2 partial pressure and CO2 loading. The presence of MEA
increases KGae values of MEA-AMP blends.

Acknowledgment

The Natural Sciences and Engineering Research Council of Canada (NSERC) is gratefully acknowledged for
generous financial support.

References

1. H. Hikita, H. Asai, H. Ishikawa and M. Honda, Chemical Engineering Journal, 13 (1977) 7-12.
2. IPCC, IPCC Special Report on Carbon Dioxide Capture and Storage, Cambridge University Press,
Cambridge, United Kingdom and New York, 2005.
3. A. Aroonwilas and A. Veawab, Ind. Eng. Chem. Res., 43 (2004) 2228-2237.
4. J. Kuntz, “Study of CO2 Absorption in Blended Aqueous Monoethanolamine and 2-Amino-2-methyl-1-
propanol” Master of Applied Science Thesis, University of Regina, 2007.
5. R.H. Weiland, J.C. Dingman, D.B. Cronin and G.J. Browning, J. Chem. Eng. Data, 43 (1998) 378-382.
6. A. Henni, J. Hromek, P. Tontiwachwuthikul and A. Chakma, J. Chem. Eng. Data, 48(3) (2003) 551-556.
7. J. Xiao, C. Li and M. Li, Chemical Engineering Science, 55 (2000) 161-175.