The Solubility of C 0 2 in a 30 Mass Percent Monoethanolamine

Solution
FANGYUAN JOU,ALAN E. MATHER and FREDERICK D. OTTO*

Depahaent of Chemical Engineering, University of Alberta, Edmonton, Alberta T6G 2G6

The solubility of CO, in a 30 mass % monoethanolamine (MEA) solution has been measured at eight temperatures
between 0 and 150°C at partial pressures of CO, ranging from 0.001 to 20,000 kPa. The data have been correlated
using the Deshmukh-Mather (1981) model.

La solubilitk du dioxyde de carbone dans une solution aqueuse contenant 30% en masse de monokthanolamine
a Ctk dktermintk dans un domaine de temp6rature variant entre 0 et 150°C et pour des pressions partielles en CO,
comprises entre 0,001 et 20 OOO Wa. Les donnkes de solubilitk ont ktk correlkes h I'aide du modtle de Deshmukh-
Mather (1981).
Keywords: carbon dioxide, aqueous amine, acid gas, gas treating.

M onoethanolamine (MEA) solutions are widely used for

the removal of C02 from gas streams. MEA was
one of the first alkanolamines to be used for this purpose,
0.61 x 0.91 x 1.22 m. The bath is equipped with a 3000 W
heater, as well as a refrigeration system. The range of
operation of the air bath is -25 to 160°C. On top of the
and in 1990 it comprised 40% of the market (DuPart et a]., cell is a 50 cm3 vapor reservoir. A magnetic pump modi-
1993). Advantages of MEA solutions are the rapid reaction fied from the design of Ruska et al. (1970) is used to
rate, low cost of the solvent, ease of reclamation, and the recirculate the gas phase and bubble it through the liquid
relatively low solubility of hydrocarbons in the solution. phase. The total volume of the apparatus is about 250 cm3.
Also, the molar mass of MEA is low which minimizes the A heated vapor sampling line leads from the reservoir to a
amount of solution which has to be circulated in a gas treating Valco sampling valve mounted outside the air bath. Two
process. The disadvantages include the high enthalpy of liquid sampling lines lead from the cell to a vented compart-
reaction with carbon dioxide and the formation of a stable ment underneath the air bath. All sampling lines are 1.59 mm
carbamate. Other factors include the inability to remove O.D.; other tubing is 3.175 mm O.D. The temperature of
mercaptans and the formation of degradation products with the contents of the cell is measured by a calibrated iron-
COS or oxygen-bearing gas. MEA solutions suffer higher constantan thermocouple and the pressure in the cell is
vaporization losses than other alkanolamines because the measured by a digital Heise (Stratford, CT) gauge (ranges
vapor pressure is higher. As well, MEA solutions are non- 0-10 MPa or 0-35 MPa). The Heise gauges have an accuracy
selective for the removal of H2S in the presence of C02. of 0.1 % of full scale according to a calibration against a dead-
However, the solution is capable of reducing the concentra- weight gauge. The thermocouple has an accuracy of f 0.1 "C
tion of carbon dioxide to less than 50 ppm (vol), a feature by comparison with a platinum resistance thermometer. The
that makes it suitable for processing natural gases which are apparatus was checked by measurement of the critical points
to be liquefied. of propane, carbon dioxide and hydrogen sulfide. Differences
A number of determinations of the solubility of carbon of less than 0.1 "C in temperature and 0.2% in pressure from
dioxide in monoethanolamine solutions have been made. the compilation of Braker and Mossman (1980) were found.
The conditions of the previous solubility determinations are The monoethanolamine (99 + %) was supplied by Aldrich
summarized in Table 1. Many of the previous data were Chemical (Milwaukee, WI),as was the methyl red indicator.
obtained for 15 mass % solutions because in the past, more Diglycolamine (DGA) (98%) was obtained from Pfaltz and
concentrated solutions were found to be corrosive. In recent Bauer (Waterbury, CT). Carbon dioxide and nitrogen with
years the use of corrosion inhibitors has allowed concentrated a purity of 99.5% were supplied by Linde (Edmonton, AB).
MEA solutions to be used. Standard HCI solution, sodium hydroxide and methyl orange
This work was undertaken to rationalize the data available xylene cyanol indicator were supplied by Fisher Scientific
and to produce a body of data over a wide range of tempera- (Nepean, ON).
tures and pressures which could serve as a standard for use The monoethanolamine solution was prepared by weight
in the modelling of vapor-liquid equilibria in MEA solutions. from distilled water and the purchased MEA. The exact
concentration of MEA was determined by titration with 0.1
Experimental apparatus and procedure mol/dm3 HCl solution using methyl red indicator. In the
absence of C02 the color change from yellow to pink is
The equipment and procedures used in this experiment sharp. The gas chromatograph was then used to determine
have changed over the years (Jou et al., 1985, 1987) and the area ratio of H20/MEA. This ratio serves as a standard
the latest version will be described in detail. A schematic to check the MEA concentration in loaded solutions.
diagram of the apparatus is given in Figure 1. The equili- About 100 cm3 of the amine solution was introduced into
brium cell consists of a Jerguson (Peacock Bros., Montreal, the evacuated cell at room temperature. The apparatus was
PQ) liquid level gauge (Model 17-T-30) which has been then purged with nitrogen or carbon dioxide for a short time
strengthened by two steel bars spaced between the ends of to remove oxygen. C02 was then added to the cell through
the cell. It is mounted in an insulated air bath of dimensions the gas inlet valve to an amount monitored by the pressure
gauge. As the gas phase was circulated at 100 mL/min, the
'Author to whom correspondence should be addressed. carbon dioxide was absorbed and the pressure in the system
140 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 73, FEBRUARY, 1995
 TABLE 1
Literature Review of MEA Solubility Data
Partial Pressure
Concentration Temperature of co2
Author (mass %) ("C) (kPa)
Mason and Dodge (1936) 3,12,30,56,74 0,25,50,75 1.3-100
Reed and Wood (1941) 15 100, 120,140 138-1724
Lyudkovskaya and Leibush (1949) 3,12,30 25,50,75 254-4054
Atadan (1954) 15,30,45 3 9 30,50,70 103-3447
Muhlbauer and Monaghan (1957) 15 25,100 < 133
Goldman and Leibush (1959) 6,12,15,30 75,100,120,140 0.3-467
Jones et al. (1959) 15 40,60,80,100,120,140 <931
Murzin and Leites (1971) 3,6,12,15,21 30,40,50,60,70,80 < 93
Lee et al. (1974) 15,30 40,100 1.4-6620
Lee et al. (1976) 6,15,23,30 25,40,60,80,100,120 0.2-6616
Lawson and Garst (1976) 15 40,60,80,100,120,134,140 2.9-2786
30 94 23-453
Nasir and Mather (1977) 15,30 60,80,100 0.001-1.3
Isaacs et al. (1980) 15 80,100 0.007-1.6
Shen and Li (1992) 15,30 40,60,80,100 1.1-2550
Murrieta-Guevara et al. (1993) 15,30 30,50,100 1.5-22 10

aging to allow the BaC03 particles to agglomerate, the

mixture was filtered with a 12.5 cm Whatman #5 paper and
washed with N2-purged distilled water. The washing was
stopped when the filtrate showed a pH of 5.0-5.6 with
Duotest paper. The precipitate and the filter paper were
transferred to a 250 cm3 Erlenmeyer flask and distilled
water added to the 150 cm3 mark. The paper was shredded
with a magnetic stirring bar. The mixture was then titrated,
with stirring, against 0.10 mol/dm3 HCl and bubbled with
N2 to remove C02. Methyl orange xylene cyanol indicator
was used and a grey-green end point was found. The final
volume in the flask was adjusted to the 200 cm3 mark with
distilled water when the pH = 5.3-5.4 which is the pH at
the end point, as determined by a pH meter.
The liquid phase analysis was also done by a chromato-
graphic technique in some cases. A 1.63 m x 3.175 mm
stainless steel packed column of Chromosorb 104 (Johns-
Manville, Denver, CO) was used in conjunction with a Model
Figure 1 - Schematic diagram of the apparatus. 57 10A Hewlett-Packard (Mississauga, ON) chromatograph.
The thermal conductivity detector was at 200°C and the
decreased. Additional amounts of C 0 2 gas were added until injection port was at 300°C. The initial oven temperature
the desired partial pressure was obtained. If necessary, was at 120°C for four minutes and then it was heated at
nitrogen was added to maintain the total pressure well above 32"C/min to 250°C until the amine eluted. The helium
atmospheric pressure. After the apparatus was brought to flowrate was 22 cm3/min. For an injection of 5 pI of
the desired conditions of temperature and pressure, the solution the retention time is 0.44 min for C02, 1.04 min
concentration of the solution was checked and adjusted if for H20, and 8.20 min for MEA. After the water peak
necessary. appeared, 100 pL of C02 was injected for use as a standard
When the vapor phase contained nitrogen, it was analysed or to check the calibration curve. Calibration curves were
to determine the N2/C02 ratio. A 3 m X 3.175 mm O.D. constructed by plotting the number of moles versus the area
Porapak QS (Waters Associates, Milford, MA) column obtained by integration on a logarithmic scale. The amount
operated isothermally at 70°C was used to effect the sepa- of aqueous MEA solution was obtained by weighing the
ration. The response factors were l . 17 for N2 and l .00 for syringe before and after the injection on a Sartorius 2434
c02. (Gottingen, Germany) balance. The tip of the needle must
To analyse the liquid phase, the precipitation-titration be covered during weighing to prevent evaporation loss. In
method was used for many points. The liquid sam le was the case of C02, a volume of between 30 pL and 5 mL was
P
withdrawn from the equilibrium cell into a 50 cm Erlen- used. When the pressure was less than lo00 kPa and ac0.5,
meyer flask or a 40 cm3 stainless steel bomb containing the liquid sample was taken directly from the liquid outlet
1 mol/dm3 NaOH solution. The C 0 2 is converted into of the equilibrium cell for the injection into the chromato-
involatile species and an aliquot was added to a 250 cm3 graph. At pressures greater than lo00 kPa or a>0.5, the
Erlenmeyer flask containing enough 0.1 mol/dm3 BaC12 liquid sample was taken and shaken well into a 25 cm3
solution to precipitate the carbonate as BaC03. When the Erlenmeyer flask which contains about 15 g of 30 mass %
loading was low the liquid sample could be taken into the diglycolamine (DGA) solution. The retention time for DGA
BaC12 solution without dilution. After at least two days was 12.6 min.

THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 73, FEBRUARY, 1995 141
 An evaluation of internal consistency was made by For the volatile solvent, the vapor-liquid equilibrium is
weighing an amount of C02 into a known mass of 30 given by
mass % MEA solution in a 40 mL steel bomb. The loading,
a,was restricted to less than 0.3, which at room tempera- y,&P = ylxlPf &exp[v, (P - P f ) / RT] . . . . . . (10)
ture corresponds to a retention of more than 99.9% C02 by
the liquid phase. Analysis showed that the error in the where vl is the molar volume of the solvent at the system
BaC03 method is f 3 % and in the gas chromatograph temperature and saturation pressure. The pure component
method is f 2 % . vapor pressures of MEA and water are taken from Daubert
MEA does not decompose at temperatures up to 350°C and Danner (1991) and Lide (1991). The fugacity coeffi-
(Daubert et al. 1987). In the present work, the MEA solution cients, r&, were calculated by the Peng and Robinson (1976)
was exposed to temperatures of 100°C or 120°C for up to equation of state.
5 consecutive days, and to temperatures of 150°C for three The equilibrium constants for the molecular electrolytes
days. Chromatographic analysis of the liquid samples did not which dissociate or react in the liquid phase may be expressed
show any decomposition products. as a general product of a polynomial equation:

Results Ki = nj
(mj*'yj)"J......................... (1 1)
The solubility of carbon dioxide in a 30 mass % solution
of monoethanolamine has been determined at temperatures where i refers to the reaction number. The equilibrium
of 0, 25, 40, 60, 80, 100, 120 and 150°C and partial constants, such as the dissociation constants of carbonate and
pressures of C02 ranging from 0.001 to 20,000 kPa. The bicarbonate as well as the dissociation constants of MEA in
data are presented in Table 2. In Table 2, Pf is the vapor water, are obtained from Patterson et al. (1982), Kent and
pressure of the 30 mass % MEA solution which was calcu- Eisenberg (1976), and Sweeton et al. (1974), as well as Perrin
lated by Raoult's law using the vapor pressure of MEA taken ( 1965).
from Daubert et al. (1987). The value of Pf was subtracted The activity coefficients of the species in the liquid phase
from the total pressure to obtain the pressure of the N2/C02 are determined by the extended Debye-Huckel equation.
mixture. From the analysis of the gaseous mixture, the partial Deshmukh and Mather (1981) indicated that this equation
pressure of C02 was calculated. has a precision which is appropriate for engineering appli-
cations. This equation is described as:
Correlation of the data
-AZ: f i
In yi = + 2 CPijmj . . . . . . . . . . . . . . .
The data obtained in this work were correlated using 1 + B&fi J

the model of Deshmukh and Mather (1981). It will be out- The first term of this equation represents the electrostatic
lined briefly. At equilibrium, the fugacities of each volatile forces, and the second term represents the short-range van
component are equal in the vapor and liquid phase: der Waals forces. The constants A and B are related to the
dielectric constant of the aqueous MEA solutions !nd the
p =y: .................................. (1) system temperature and pressure. The quantity a which
depends on the effective diameter of the ion was obtained
The C02 in the liquid phase reacts with other chemical from Barner and Scheuerman (1978). The interaction
components to produce ions according to the following parameters, Pij, represent the mutual effect of the two
reactions: species. These parameters have to be obtained by correlating
solubility data, and are independent of temperature and
C02 + 2RNH2 Y RNH: + RNHCOO- . . . . . . . (2) concentration. The dielectric constant of water and MEA
co2+ RNH2 + H20 o RNH: 4- HCO;. . . . . . (3) solvent were obtained from Lide (1991) and Ikada et al.
(1968). The following mass and electroneutrality balances
COz + H20 o H+ + HCO;. . . . . . . . . . . . . . . . . (4) also apply.
Mass balances:
HCO; Y H+ + COT ....................... (5)
mAM= [RNH,] + [RNHa + [RNHCOO-] . . . . . . (13)
H20 0 Hf + OH- ......................... (6)
RNH2+ H 2 0 Y RNH; + O H - , . . . . . . . . . . . . . (7)
RNHCOO- + H20 o RNH2 + HCO; . . . . . . . . (8) Electroneutrality:

The vapor-liquid equilibria for the system are given by: [ H f l + [RNHa =[OH-] + [HCO;] + 2 [ C O F l +
[RNHCOO-] ........................... (15)
y&P = yimiHiexp(;f(P - P f ) / RT] . . . . . . . . . (9)
Given the COz partial pressure of the gas-liquid equilibrium
Here Hi and ;,'represent the Henry's constant and partial system, the concentration of molecular C02 in the solution
molar volume at infinite dilution for molecular solute i in is obtained using Equation (9). Therefore, there are seven
the solution at the equilibrium temperature and at the vapor species of components and ions (RNHZRNHCOO - , H C O ,
pressure of the solution. The Henry's constants of C 0 2 in CO,; H+,0H-,RNH2) in the liquid phase that need to be
the aqueous MEA solutions are assumed to be equal to the solved by seven independent equations [four equilibrium
Henry's constants of C02 in water. Equations (4)-(7) and three balance Equations (13)-( 15)]

142 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 73, FEBRUARY, 1995
 TABLE2
Solubility of CO, in a 30 mass percent Monoethanolamine Solution
t = 0°C; Pp = 0.54 kPa t = 25°C; Pp = 2.8 kPa

ace, aco,
P(kPa) pco, (@a) GC BaCO, P( kPa) pco, (kPa) GC BaCO,
~ ~ ~~

2oooo 19954 1.324 1.331 2 m 19936 1.247

15000 14967 1.291 I5000 14955 1.209
loo00 9979 1.262 1.320 loo00 9973 1.170
6OOo 5989 1.221 6440 6425 1.109 1.104
35mb 3495 1.140 6410' 6402 1.102 1.100
3470' 3468 1.132 3000 2996 1.039 1.037
1000 999 I .022 1000 996 0.923 0.938
600 599 0.970 600 596 0.844 0.855
300 299 0.902 300 297 0.809 0.815
280 68.4 0.772 100 97.2 0.694 0.699
200 13.2 0.646 300 55.1 0.648 0.678
200 1.76 0.604 210 11.8 0.583 0.588
200 0.253 0.550 200 2.80 0.540 0.532
200 0.0236 0.495 200 0.392 0.494
200 0.00595 0.459 200 0.0600 0.439
200 0.00120 0.400 300 0.00927 0.323
200 0.00213 0.21 1
b Bubble point of C0,-rich liquid phase.
c Dew point of CO, vapor phase.

t = 40°C; Pp = 6.6 kPa t = 60°C; Pp = 18 kPa

ace, ace,
P( kPa) Pco, @Pa) GC BaCO, P(kPa) pco, GC BaCO,
2 m 19914 1.180 1.151 2 m 19893 1.109 1.082
15000 14945 1.132 1.108 I5000 I493 1 1.062 1.055
loo00 9969 1.097 1.073 loo00 9959 1.027 1.001
6000 5986 1.049 1.023 6OOo 5968 0.942 0.948
3000 2992 0.965 0.929 3000 2977 0.880 0.878
lo00 993 0.844 0.808 1000 982 0.749 0.736
600 593 0.794 0.778 600 582 0.701 0.715
300 293 0.709 0.71 1 300 282 0.642 0.694
110 103 0.646 0.638 100 82.0 0.575 0.583
2 10 36.1 0.609 250 34.1 0.565 0.557
230 8.09 0.557 250 11.0 0.504
220 2.57 0.513 200 2.01 0.438
210 0.604 0.461 0.486 300 0.528 0.389
300 0.0677 0.365 220 0.0579 0.206
200 0.00896 0.203 260 0.0193 0.119
200 0.00147 0.0888 220 0.00428 0.0564

besides the activity coefficients. A non-linear optimization where a is the loading of C 0 2 in the mixed solvent, k is the
calculation method is used to solve the non-linear algebraic kth experimental point and n is the total number of
equations. The parameters for the extended Debye-Huckel experimental points.
equations are correlated by the same non-linear optimization The results of the correlation for the equilibrium solubility
method. The objective function used in the data regression of Co2 in a 30 mass % aqueous MEA solution over the
systems is given by temperature range 0 to 150°C are plotted on Figure 2 and
on Figures 3 to 7 where the results of the present work are
compared with previously published solubility data. As
shown on Figure 2, the correlation provides a good represen-
= E -P(;: p;;*)* tation of the experimental data obtained in this study. The
k = I
G; k
goodness of fits is very good for C02 partial pressures
below 4500kPa where the average absolute error of the
regression is 1.48% for the sum of the loading of C02 and
the partial pressure of C 0 2 . The major interaction

( )
2
aexp. - acal. paramaters, Pii, obtained from this work are listed in
+ !.=I (yexp.
. . . . . . . . . . . . . . . . . . . (16) Table 3. The equilibrium constant K 2 , and the carbonate
k stability constant, K, = l/K8, were also obtained from the

THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 73, FEBRUARY, 1995 143
 TABLE2 cont’d
t = 80°C; Pp = 42 kPa t = 100°C; Pp = 91 kPa
QC02 Qco*
P(W4 PCO,(Wa) GC BaC03 P(kPa) pco, (kPa) GC BaC0,
2 m 19862 1.033 2 m 19812 0.941
15000 14902 0.998 1.006 15o00 14842 0.914
loo00 9925 0.953 0.962 loo00 987 1 0.856
6Ooo 5943 0.875 0.901 6000 589 1 0.810
3000 295 1 0.797 0.808 3000 2899 0.705
1000 958 0.680 0.698 1000 909 0.605 0.622
600 558 0.640 600 509 0.571
500 235 0.576 0.581 700 376 0.589
100 56.0 0.517 200 109 0.481
200 16.0 0.460 160 69.0 0.477
300 2.67 0.348 130 39.0 0.422
250 0.278 0.187 110 19.0 0.381
250 0.0992 0.118 300 1.43 0.188
250 0.0465 0.078 1 300 0.136 0.0566
280 0.00484 0.0236 300 0.00724 0.01 17
250 0.00296 0.0174

t = 120°C; Pp = 178 kPa t = 150°C; Pp = 429 kPa

QCO2 Qco*
P(kPa) pco, (kPa) GC BaCO, P(kPa) pco, (kPa) GC BaCO,
18000 17723 0.863 17000 16441 0.755
15000 14741 0.829 12000 11504 0.727
loo00 9770 0.780 0.783 9000 8525 0.651 0.650
6Ooo 5809 0.719 0.722 5o00 4544 0.583
3000 2804 0.644 0.634 2400 570 0.484
lo00 822 0.536 lo00 560 0.388
600 422 0.473 850 420 0.358
400 222 0.444 730 123 0.301
300 122 0.403 0.420 630 34.2 0.153
350 46.8 0.349 600 2.63 0.0509
300 2.29 0.119 600 0.184 0.0134
400 0.0984 0.0247 520 0.0843 0.00992
350 0.0221 0.01 12 650 0.0239 0.00496
250 0.00202 0.00333 600 0.00477 0.00199

correlation results. Values for selected temperatures are given Other comparisons with earlier data are shown in Figure 5.
in Table 4. The high pressure data of Lyudkovskaya and Leibush (1949)
and the low pressure data of Mason and Dodge (1936) are
Discussion in agreement with the present work. The data of Nasir
and Mather (1977) at 100°C is in agreement at a partial
Comparisons with the data of Shen and Li (1992) are pressure of 1 kPa, but the trend with lower partial pressures
shown in Figure 3. At low pressures the results are in rela- is incorrect. Nasir and Mather used a flow through system
tively good agreement, but the high pressure data deviate and it is now realized that such an apparatus is unsuitable
markedly from the present work. In Figure 4 the present data for use with C02 because of the relatively slow rate of
are compared with the work of Lee et al. (1974) which were reaction. The data of Goldman and Leibush (1959) are
obtained in this laboratory. There is a consistent bias, the compared with the present work in Figure 6. At the highest
data of Lee et al. always about 0.04 moles of C02/mole partial pressures the results are in good agreement, but the
MEA less than the present work. The reason for this bias trend with partial pressure of the data of Goldman and
lies in the procedure Lee et al. used to analyse the liquid Leibush appears to be incorrect. As shown on Figure 7, the
phase. The liquid sample was withdrawn from the cell into recent data reported by Murrieta-Guevara et al. (1993) are
a solution of H2S04. The C 0 2 evolved from the solution in poor agreement with the present work.
and was measured volumetrically. The C 0 2 which remained
in the acidic solution was neglected as the amount was
thought to be small; however, the amount of amine in the Conclusions
acidic solution was also small, and the resulting (Y appears
to be about 0.04. The same bias is present in the work of This paper reports experimental data for the solubility of
Lee et al. (1976). C 0 2 in a 30 mass % monoethanolamine solution over

144 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 73, FEBRUARY, 1995
 1 o4

T o3

1 o2
h
m
%
v 10'
0
"
a
1o'

lo-'

1 o-2

1 o-3
I 0-3 10-2 lo-' 100 00 02 0 4 0 6 08 10

a (mol C02/mol MEA) ace, (rnol of CO,/rnol of MEA)

co2

Figure 2 - Solubility of CO, in a 30 mass % MEA solution. Figure 3 - Comparison of the solubility data with those of Shen
and Li (1992). -This work; -------- Shen & Li.
TABLE 3
Interaction Parameters in the Extended Debye-Huckel Equation

Interaction parameter
Binary pair B;; x 10'
RNH, - RNH: 12.493
RNH2 - RNHCOO- 1.1159
RNH2 - HCO; -1.1226
RNH2 - H + -3.8228 h

RNH2 - O H - - 1.8050 m
a
RNH2 - COT
RNH: - RNHCOO-
0.12213
-4.6317 -x
ON
lo2
RNH; - HCO; -5.69 14 a"
RNH: - H' -1.1601
RNH: - O H - 0.23100
RNH: - COT 3.4093
RNHCOO- - HCO; - 1.0084 10'
RNHCOO- - H + 7.5028
RNHCOO- - O H - 6.0378
HCO; - O H - 4.2815
HCO; - COT 7.8160
H' - COT 5.7164 .-
ino
0.1 0.2 0.3 0.4 0.5 0.6 0.70.80.91

a (mol of C02/mol of MEA)

CO,
TABLE4
Equilibrium Constants for Reactions 2 and 8
Figure 4 - Comparison of the solubility data with those of Lee
Temperature et al. (1974). -This work; -------- Lee et al.
K, Kc
"C kg/mol kglmol
a range of temperatures from 0°C to 150°C for partial
0.0 3.93 x 10' 87.91 pressures ranging over seven orders of magnitude. The
25.0 3.70 x 104 3 1.38 Deshmukh and Mather (1981) model provides a good
40.0 1.15 x lo4 22.80 representation of the experimental data.
60.0 2.43 x lo3 15.08
80.0 5.78 x 10' 10.97 Acknowledgement
100.0 2.46 x lo2 14.48
120.0 4.08 x 10' 7.06 This work was supported by the Natural Sciences and
150.0 6.74 5.83 Engineering Research Council under Strategic Grant

THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 73, FEBRUARY, 1995 145
 This work
A Nasir and Mather (1977)
1o3 0 Lyudkovskaya and Leibush (194
0 Mason and Dodge (1936)

1 o2

7
a
10'
3
v

ON
a" 1oo

lo-'
o Murrieta-Guevara
et a l . (1993)
1o-2 50°C

1o -~ 0.01
10-2 lo-' 1 oo 0.1 0.2 0.3 0.4 0 . 5 0.6 0.7 0.8 0.9 1.0

a (mol of COJrnol of MEA) a (mol of C02/mol of MEA)

COZ
COZ

Figure 5 - Comparison of the solubility data with those of Lyudkov-

skaya and Leibush (1949), Mason and Dodge (1936), and Nasir Figure 7 - Comparison of the solubility data with those of Murrieta-
and Mather (1977). -This work; -------- Previous work. Guevara et al. (1993). -This work; -------- Munieta-Guevara
et al.

3 = fugacity of a component in solution, kPa

F = objective function
H = Henry's constant, MPa
I = ionic strength, mollkg
1 o2 K = equilibrium constant
K, = l/K8, stability constant, kg/mol
rn = molality, mol/kg
P = pressure, kPa
h
R = gas constant, 8.314 J/mol.K
m
T = absolute temperature, K
v B 10' V = molar volume m3/mo1
0
N
V" = partial molar volume at infinite dilution, m3/mol
R" X = mole fraction in liquid phase
y = mole fraction in vapor phase
Z = charge on an ion
1 oo
Greek letters
ci = loading, mol/mol MEA
0 = parameter in the extended Debye-Huckel equation,
kglmol
lo-' J, , , , , , , , y = activity coefficient
0.02 0.6 = stoichiometric coefficient
0.1 Y
I#I = fugacity coefficient
a (mol of C02/mol of MEA)
co,
Subscripts
Figure 6 - Comparison of the solubility data with those of Goldman
and Leibush (1993). This work; Goldman & AM = amine
Leibush. - --_---_- ij = species ij
k = no. of points
No. 100777. The authors are grateful to Mr. S. Xu for his co, = coz
assistance with the correlation of the data. 1 = solvent

Nomenclature Superscripts
a = maximum distance between ions, lo-'' m cal. = calculated
A = constant in the extended Debye-Huckel equation, exp. = experimental
kao.5 mol -0.5
G = gas
B = cinstant in the extended Debye-Huckel equation, L = liquid
kg'.' cm/mol'~' 0 = saturation

146 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 73, FEBRUARY, 1995
References Lee, J. I., F. D. Otto and A. E. Mather, “Equilibrium Between
Carbon Dioxide and Aqueous Monoethanolamine Solutions’’, J.
Atadan, E. M., “Absorption of Carbon Dioxide by Aqueous Appl. Chem. Biotechnol. 26, 541-549 (1976).
Monoethanolamine Solutions”, Ph.D. Thesis, University of Lide, D. R., Ed., “CRC Handbook of Chemistry and Physics”,
Tennessee, Knoxville, TN (1954). 72nd ed., CRC Press, Inc., Boca Raton, FL (1991).
Barner, H. E. and R. V. Scheuerman, “Handbook of Thermo- Lyudkovskaya, M. A. and A. G. Leibush, “Solubility of Carbon
chemical Data for Compounds and Aqueous Species”, Wiley , Dioxide in Solutions of Ethanolamines Under Pressure”, Zhur.
New York (1978). Priklad. Khim. 22, 558-567 (1949).
Braker, W. and A. L. Mossman, Eds., “Matheson Gas Data Mason, J. W. and B. F. Dodge, “Equilibrium Absorption of Carbon
Book”, Matheson Co., Secaucus, NJ (1980). Dioxide by Solutions of the Ethanolamines”, Trans. Amer. Inst.
Daubert, T. E and R. P. Danner, “Physical and Thermodynamic Chem. Eng. 32, 27-48 (1936).
Properties of Pure Chemicals: Data Compilation”, Hemisphere Muhlbauer, H. G. and P. R. Monaghan, “New Equilibrium Data
Pub. Co., Washington, D. C. (1991). on Sweetening Natural Gas with Ethanolamine Solutions”, Oil
Daubert, T. E., J. W. Jalowka and V. Goren, “Vapor Pressure Gas J. 55 (17), 139-145 (1957).
of 22 Pure Industrial Chemicals”, AIChE Symp. Ser. 83 (256), Murrieta-Guevara, F., E. Rebolledo-Libreros and A. Trejo, “Gas
128-156 (1987). Solubility of Carbon Dioxide and Hydrogen Sulfide in Mixtures
Deshmukh, R. D. and A. E. Mather, “A Mathematical Model for of Sulfolane with Monoethanolamine”, Fluid Phase Equilibria
Equilibrium Solubility of Hydrogen Sulfide and Carbon Dioxide 86, 225-231 (1993).
in Aqueous Alkanolamine Solutions”, Chem. Eng. Sci. 36, Murzin, V. I. and I. L. Leites, “Partial Pressure of Carbon Dioxide
355-362 (1981). over its Dilute Solutions in Aqueous Aminoethanol”, Zhur. Fiz.
DuPart, M. S., T. R. Bacon and D. J. Edwards, “Understanding Khim. 45, 417-420 (1971).
Corrosion in Alkanolamine Gas Treating plants”, Hydrocarbon Murzin, V. I., I. L. Leites, L. S. Tyurina and Z. I. Perstina,
Process. 72 (3,89-94 (1993). “Partial Pressure of Carbon Dioxide over Slightly Carbonated
Goldman, A. M. and A. G. Leibush, “Solubility of Carbon Dioxide Aqueous Solutions of Monoethanolamine”, Trudy Nauch.-
in Aqueous Solutions of Monoethanolamine in the Temperature Issledov. Proekt. Proekt. Inst. Azot. Prom. Prod. Org. Sin., No.
Range 75”-140°C”, Trudy Gosudarst. Nauch.-Issledov Proekt. 10, 26-37 (1971).
Inst. Azot. Prom. 10, 54-82 (1959). Nasir, P. and A. E. Mather, “The Measurement and Prediction
Ikada, E., Y. Hida, H. Okamoto, J. Hagino and N. Koizumi, of the Solubility of Acid Gases in Monoethanolamine Solutions
“Dielectric properties of Ethanolamines”, Bull. Inst. Chem. Res. at Low Partial Pressures”, Can. J. Chem. Eng. 55, 715-717
Kyoto Univ. 46 (3,239-247 (1968). (1977).
Isaacs, E. E., F. D. Otto and A. E. Mather, “Solubilities of Patterson, C. S., G. H. Slocum, R. H. Bushy and R. E. Mesmer,
Mixtures of H,S and CO, in a Monoethanolamine Solution at “Carbonate Equilibrium in Hydrothermal Systems: First
Low Partial Pressures”, J. Chem. Eng. Data25, 118-120 (1980). Ionization of Carbonic Acid in NaCl Media to 300”C”, Geochim.
Jones, J. H., H. R. Froning and E. E. Claytor, Jr., “Solubility Cosmochim. Acta 46, 1653-1663 (1982).
of Acidic Gases in Aqueous Monoethanolamine”, J. Chem. Eng. Pemn, D. D., “Dissociation Constants of Organic Bases in Aqueous
Data 4, 85-92 (1959). Solutions”, Buttenvorths, London (1965).
Jou, F.-Y., F. D. Otto and A. E. Mather, “Equilibria of H,S and Reed, R. M. and W. R. Wood, “Recent Design Developments in
CO, in Triethanolamine Solutions”, Can. J. Chem. Eng. 63, Amine Gas Purification Plants”, Trans. Amer. Inst. Chem. Eng.
122-125 (1985). 37, 363-384 (1941).
Jou, F.-Y., R. D. Deshmukh, F. D. Otto and A. E. Mather, Ruska, W. E. A., L. J. Hurt and R. Kobayashi, “Circulating Pump
“Vapor-Liquid Equilibria for Acid Gases and Lower Alkanes for High Pressure and -200 to +400°C Application”, Rev. Sci.
in Triethylene Glycol”, Fluid Phase Equilibria 36, 121-140 Inst. 41, 14441446 (1970).
(1987). Shen. K-P. and M.-L. Li, “Solubility of Carbon Dioxide in Aqueous
Kent, R. L. and B. Eisenberg, “Better Data for Amine Treating”, Mixtures of Monoethanolamine with Methyldiethanolamine”, J.
Hydrocarb. Proc. 55 (2), 87-90 (1976). Chem. Eng. Data 37, 96-100 (1992).
Lawson, J. D. and A. W. Garst, “Gas Sweetening Data: Equilibrum Sweeton, F. H., R. E. Mesmer and C. F. Baes, Jr., “Acidity
Solubility of Hydrogen Sulfide and Carbon Dioxide in Aqueous Measurements at Elevated Temperatures. VII. Dissociation of
Monoethanolamine and Aqueous Diethanolamine Solutions”, J. Water”, J. Solution Chem. 3, 191-214 (1974).
Chem. Eng. Data 21, 20-30 (1976).
Lee, J . I., F. D. Otto and A. E. Mather, “The Solubility of H,S
and CO, in Aqueous Monoethanolamine Solutions”, Can. J. Manuscript received August 13, 1993; revised manuscript
Chem. Eng. 52, 803-805 (1974). received October 3,1994; accepted for publication October 12, 1994.

THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 73, FEBRUARY, 1995 147