Ind. Eng. Chem. Res.

1998, 37, 1473-1477 1473

Gas-Liquid Equilibrium in a CO2-MDEA-H2O System and the

Effect of Piperazine on It
Guo-Wen Xu,* Cheng-Fang Zhang, Shu-Jun Qin, Wei-Hong Gao, and Hua-Bin Liu
Research Institute of Chemical Technology, East China University of Science and Technology, P.O. Box 274,
130 Meilong Road, Shanghai 200237, People’s Republic of China

Solubility data of CO2 and vapor pressure of water in 3.04-4.28 kmol/m3 aqueous N-
methyldiethanolamine (MDEA) solutions were obtained at temperatures ranging from 40 to
100 °C and CO2 partial pressures ranging from 0.876 to 1013 kPa. A thermodynamic model
was proposed and used for predicting CO2 solubility and water vapor pressure. An enthalpy
change of absorption of CO2 in 4.28 kmol/m3 MDEA solution was estimated. The effect of
piperazine (PZ) concentration on CO2 loading in MDEA solutions was determined at piperazine
concentration ranging from 0 to 0.515 kmol/m3. The results show that piperazine is beneficial
to the CO2 loading. The equilibrium partial pressure of piperazine in the PZ-MDEA-H2O
system was measured in an Ellis Cell. Results show that the PZ-MDEA-H2O system is a
typical negative deviation system, with the strength of deviation decreasing with MDEA solutions.

Introduction The MDEA solution was made to the desired strength

and charged to the cell. The volume of solution inside
Aqueous N-methyldiethanolamine (MDEA) solutions
the cell was about 200 cm3.
are widely used for removal of the acid gas (H2S and
CO2) from natural gas synthesis and refinery gas Gas (N2 and CO2) from cylinders with a purity of 99%
streams. The use of MDEA solutions was first described passed through a high-pressure regulator in order to
by Frazier and Kohl (1950). MDEA has a stable keep the flow rate stable. Gas passed through a
structure and does not degrade readily, so it would not molecular sieve adsorber and was then metered by a
cause problems of corrosion of carbon steel. soap film meter. After passing through a gas-mixture
Jou (1982) reported the solubility of H2S and CO2 in tube, the mixture gas was saturated at room tempera-
1.0, 2.0, and 4.28 kmol/m3 aqueous MDEA solutions ture and then introduced into the experimental cell. The
under temperatures and acid gas partial pressures rang- mixture gases are bubbled by a self-suction-type agita-
ing from 40 to 120 °C and from 0.001 to 6600 kPa, res- tor. The partial pressure of CO2 was controlled by
pectively. A mathematical model, where all nonideali- varying the ratio of CO2 to N2.
ties were included in the equilibrium constant expres- The outlet gas was cooled by nature water at room
sion, was proposed to correlate the CO2 solubility data. temperature, and the condensed water flows back to the
Chakma (1987) determined the solubility of CO2 in cell. The temperature of the outlet gas is usually 0.2
1.69 and 4.28 kmol/m3 aqueous MDEA solutions over a °C higher than room temperature. The moisture loss
range of temperatures from 100 to 200 °C and partial with the carrier gas can be neglected during all experi-
pressures from 172 to 4929 kPa. A mathematical model ments.
similar to that of Jou (1982) for predicting CO2 solubility The total flow rate of mixture gases was about 0.018
in MDEA solutions was presented. m3/h during all experiments. The experimental tem-
Research work about CO2 solubility in MDEA solu- perature must be controlled to within (0.1 °C. We had
tions also has been done by Bhairi (1984), Ho and observed CO2 loading vs time at different pressures,
Eguren (1988), Austgen (1991), MacGregor and Mather temperature and MDEA concentrations. It was found
(1991), Shen and Li (1992), and Jou (1993a,b). that, to ensure equilibrium had been reached, gas had
In this paper, solubility data of CO2 and saturated to be bubbled through the liquid for at least 4 h prior
water vapor pressure over aqueous MDEA solutions, to sampling the liquid phases.
which are suitable to engineering design, were pre- The content of CO2 in the liquid phase was deter-
sented and a thermodynamic model of an ionic solution mined by addition of an excess of sulfuric acid into the
was proposed. The enthalpy change of absorption of solutions and then measurement of the evolved volume
CO2 and the effect of piperazine (PZ) on the equilibrium of CO2 by a graduated buret. The correction of tem-
partial pressure of CO2 in aqueous MDEA solutions perature and pressure was incorporated.
were investigated. The vapor pressure of piperazine in The high-pressure apparatus was a stainless steel
the PZ-MDEA-H2O system was measured. autoclave with a magnetic agitator. The volume of the
autoclave was 500 mL. Aqueous MDEA solutions were
Experimental Section charged into the autoclave, and then CO2 gas with a
Solubility data were measured in an atmospheric- purity of 99.9% was bubbled into the MDEA solutions
pressure apparatus and a high-pressure apparatus. and kept at a constant pressure. The temperature was
The atmospheric-pressure apparatus was a glass cell controlled to within (0.1 °C. To ensure that equilibri-
stirred by a self-suction agitator. The inside diameter um had been reached, the experimental time was kept
of the cell was 70 mm, and the height was 130 mm. at least 6 h prior to sampling of the liquid phase. In
the vapor phase, CO2 was saturated with water vapor.
* Author to whom correspondence is addressed. Telephone: The CO2 content in the liquid phase was performed
021-64252386. Fax: 021-64250884. by adding excess BaCl2 to precipitate the carbonate as
S0888-5885(95)00632-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/03/1998
1474 Ind. Eng. Chem. Res., Vol. 37, No. 4, 1998

Table 1. Solubility Data of CO2 in Aqueous MDEA

Solutions
MA, PCO2,e, PCO2,c, Pw,c, Pw,e,
kmol/m3 y × 102 T, K kPa kPa kPa kPa
4.28 0.2690 313 15.40 19.50 6.520 6.30
4.28 0.3653 313 30.11 36.83 6.453
4.28 0.7045 313 203.0 221.9 6.274
4.28 0.7947 313 393.0 384.4 6.240
4.28 0.8806 313 838.0 772.4 6.213
4.28 0.1250 328 8.917 9.935 14.21 13.4
4.28 0.2480 328 32.12 35.78 13.99
4.28 0.4500 328 133.0 126.4 13.71
4.28 0.6179 328 301.0 297.4 13.52
4.28 0.7390 328 603.0 572.1 13.41
4.28 0.7776 328 855.0 725.3 13.38
4.28 0.8128 328 1013.0 921.5 13.36
4.28 0.0740 343 9.220 8.970 28.36 27.0
4.28 0.1400 343 29.32 28.15 28.11
4.28 0.3509 343 174.0 164.7 27.44
Figure 1. Comparison of the present CO2 solubility with that of 4.28 0.5091 343 389.0 388.7 27.04
Jou (1982). 4.28 0.6405 343 754.0 754.0 26.77
4.28 0.0303 353 3.268 3.321 43.38 41.0
BaCO3. The precipitate was washed and titrated with 4.28 0.1394 353 47.31 50.03 42.74
a standard 0.1 N HCl solution using a Methyl orange- 4.28 0.2992 353 207.8 211.0 41.94
xylene cyanol indicator to a gray-green endpoint. 4.28 0.4735 353 552.8 573.3 41.24
4.28 0.5610 353 867.8 890.4 40.95
Water vapor pressures over aqueous MDEA solutions 4.28 0.0091 373 0.876 1.348 93.14 91.5
were determined by using a modified ebulliometer in a 4.28 0.0367 373 11.87 16.01 92.76
MDEA-H2O system. MDEA solutions were prepared 4.28 0.1421 373 159.0 174.0 91.46
from chemically pure MDEA and deionized water. Pure 4.28 0.2744 373 519.0 590.4 90.04
MDEA are supplied by Research and Design Institute 4.28 0.3514 373 824.0 972.3 89.32
of Fine Chemical and Technology in SiChuan Province 3.46 0.5019 328 115.0 113.1 14.07 14.2
3.46 0.7213 328 389.0 354.4 13.90
(China). 3.46 0.7406 328 401.0 396.4 13.89
The experiments in 3.04-4.28 kmol/m3 aqueous MDEA 3.46 0.8197 328 782.0 663.2 13.84
solutions were carried out over a range of temperatures 3.46 0.8488 328 992.0 832.7 13.82
and partial pressures from 40 to 100 °C and from 0.876 3.46 0.3580 343 173.5 120.9 28.17 27.2
to 1013 kPa, respectively. Experimental results are 3.46 0.5120 343 278.5 278.8 27.88
listed in Table 1. Figure 1 shows the comparison of CO2 3.46 0.6040 343 388.5 443.5 27.73
3.46 0.6930 343 608.5 703.0 27.60
partial pressures of this work with those of Jou (1982). 3.46 0.7530 343 808.5 989.8 27.52
It is found that the value of this work is a little higher 3.46 0.2510 353 169.8 106.6 43.19 42.3
in comparison with that of Jou (1982) in high loading 3.46 0.3660 353 254.8 230.0 42.81
of CO2. A possible reason for the different results is 3.46 0.4270 353 364.8 324.1 42.62
the difference in purity of commercial MDEA. 3.46 0.5490 353 599.8 605.9 42.30
3.46 0.6580 353 794.8 1052.0 42.05
3.46 0.1740 363 147.5 98.67 64.37 63.7
Thermodynamic Model 3.46 0.2360 363 247.5 175.0 64.03
3.46 0.3190 363 357.5 317.9 63.61
The main reactions occurring in the CO2-MDEA- 3.46 0.3950 363 552.5 499.4 63.26
H2O system are as follows: 3.46 0.4730 363 737.5 758.1 62.93
3.04 0.2088 328 10.74 13.98 14.54 14.4
Dissociation of R3 NH+: 3.04 0.2318 328 18.85 17.11 14.52
3.04 0.3469 328 42.57 38.88 14.41
3.04 0.4641 328 85.57 75.51 14.31
R3NH+ ) R3N + H+ (1) 3.04 0.6900 328 200.5 242.6 14.15
3.04 0.7790 328 288.5 409.9 14.10
Dissociation of carbon dioxide: 3.04 0.8290 328 395.5 580.0 14.07
3.04 0.8860 328 595.5 956.0 14.05
3.04 0.9110 328 806.5 1268.0 14.03
CO2 + H2O ) H+ + HCO3- (2) 3.04 0.0692 343 6.152 4.445 29.14 29.0
3.04 0.0979 343 12.33 8.390 29.07
Dissociation of bicarbonate ion: 3.04 0.1489 343 23.79 17.86 28.95
3.04 0.2735 343 70.17 57.01 28.70
HCO3- ) H+ + CO32- (3) 3.04
3.04
0.4840
0.5820
343
343
206.8
281.8
199.6
328.2
28.33
28.19
3.04 0.6590 343 376.8 485.8 28.09
Ionization of water: 3.04 0.7400 343 581.8 759.1 28.00
3.04 0.7910 343 806.8 1042.0 27.95
H2O ) H+ + OH- (4)
aHO3-aH+
Corresponding expressions of equilibrium constants K2 ) (6)
aCO2aw
are
aR3NaH+ aCO32-aH+
K1 ) (5) K3 ) (7)
aR3NH+ aHCO3-
 Ind. Eng. Chem. Res., Vol. 37, No. 4, 1998 1475

aOH-aH+ Table 2. Expressions for Constants Used in the

K4 ) (8) Thermodynamic Model
aw
expressions ref
The following balance equations for the reacting species log k1 ) -14.01 + 0.0184T Barth, 1984
can be formed: ln k2 ) 235.482 - 12092.10/T - 36.7816 ln T Edwards, 1978
ln k3 ) 220.067 - 12431.7/T - 35.4819 ln T Edwards, 1978
Amine balance: ln k4 ) 140.932 - 13445.9/T - 22.4773 ln T Edwards, 1978
Aφ ) -1.306568 + 0.1328238 × 10-1T - Chen, 1979
MA ) (mR3N + mR3NH+) (9) 0.350803 × 10-4T2 - 0.3387968 × 10-7T3
pWs ) 0.133289 exp[18.3036 - Zhou and Xu,
3816.44/(T - 46.13)] 1983
Carbon balance: H0CO2 ) 100 exp(A - B/T) Haimour, 1987
A ) 12.212 - 0.4815MA - 0.0467MA2
MAy ) (mCO2 + mHCO3- + mCO32-) (10) B ) 2627.3 - 158.76MA - 15.612MA2

Electroneutrality: selected. Fortunately, the equilibrium calculations are

not sensitive to this estimation method because the
mR3NH+ + mH+ ) mOH- + mHCO3- + 2mCO32- (11) important parameters are Henry’s constant and the
equilibrium constant, not interaction parameters.
Under general conditions, the vapor pressure of Following a suggestion by Pitzer and Mayorga (1973),
MDEA is very low and we can assume that it is only in βik(0) and βik(1) for ion-ion interactions are empirically
the liquid phase. The gas-liquid equilibrium expres- interrelated by
sions of CO2 and H2O are as follows:
ΦCO2pCO2HCO2 ) γCO2mCO2F (12) βik(1) ) 0.018 + 3.06βik(0) (18)

Φwpw ) apwsΦw (13) Coulombic forces between like charged ions do not
allow the ions to approach one another. Therefore, It
Under commercial operation conditions, the gas-phase is assumed that for like charged ions
fugacity coefficient of CO2 and H2O is close to unity.
βik(0) ) βik(1) ) 0 (19)
ΦCO2 ) Φw ) Φws )1 (14)
In the CO2-MDEA-H2O system, the direct prediction
The activity coefficients take into account interactions of the equilibrium composition is impossible since the
between solute species in the liquid phase. In this work, specific interaction parameters are unavailable. These
the activity coefficient of solute species can be expressed parameters can be obtained by least-squares fitting of
by the equation derived from Pitzer’s theory (Edwards, experimental data.
1978) Eight distinct species exist in this system: R3N, CO2,

[ ]
H2O, R3NH+, H+, HCO3-, OH-, and CO32-. If all
xI 2 interaction parameters were considered, the thermody-
ln γi ) -AΦZi2 + ln(1 + 12xI) + namic model would be very complex.
12
1 + 12xI To simplify the calculations, the following assump-

{ }
tions may be made:
βij(1) (1) It is suggested that the parameter βmj(1) for
∑ (0)
2 mj βij +
j*w 2I
[1 - (1 + 2xI) exp(-2xI)] - molecular-ion interactions and βmm(1) for molecular-
molecular interactions can be expressed as
Zi2
2
∑ ∑ mjmkβij(1){1 - (1 + 2xI + 2I) exp(-2xI)} βmj(1) ) βmm(1) ) 0 (20)
4I j*w k*w
(15) (2) The interaction parameters for the species which
are present in very small concentrations can be ne-
where I is the ionic strength of the solution glected, such as H+ and OH-. At very high CO2 partial
pressure, the concentration of free CO2 is also large, and
∑mjZj2
1
I) (16) the concentration of CO32- is not small enough to
2 neglect. Therefore, the most important interaction
The activity of water is determined from the Gibbs- parameters are βR3N-R3NH+(0), βR3N-HCO3-(0), βR3NH+-HCO3-
(0), β
R3NH+-CO2 , βHCO3--CO2 , βR3N-CO32- , and βR3NH+-CO32-
(0) (0) (0)
Duhem equation

{
(0).

2AΦI3/2 The experimental data of 65 points are used for

ln aw ) Mw - ∑ ∑ mimj[βij(0) + calculating the specific interaction parameters. Table
2 lists the constants used in this model. The estimated
1 + 12xI

}
i*wj*w
values of interaction parameters are shown in Table 3.
Having determined specific interaction parameters,
βij(1) exp(-2xI] - Mw ∑ mi
i*w
(17) we may calculate the vapor-liquid equilibrium of the
CO2-MDEA-H2O system. The predicted data of the
Calculation and Results thermodynamic model are listed in Table 1. Figure 2
shows the comparison between the predictions and the
To evaluate the activity coefficients γi and aw, a experimental data. The agreement is good for nearly
method of estimating interaction parameters must be all experimental points. The mean error between
1476 Ind. Eng. Chem. Res., Vol. 37, No. 4, 1998

Table 3. Specific Interaction Parameters βij(0)

R3N-R3NH+ R3N-HCO3- R3NH+-HCO3- R3NH+-CO2 R3HCO3--CO2 R3N-CO32- R3NH+-CO32-
0.06229 0.07832 0.07079 0.3230 0.1557 1.042 2.870

Table 5. Effect of Piperazine on CO2 Partial Pressure in

4.28 kmol/m3 MDEA Solutions
PZ, pCO2, y, PZ, pCO2, y,
kmol/m3 kPa mol/mol kmol/m3 kPa mol/mol
0 4.263 0.04224 0.103 3.8346 0.05246
14.455 0.08542 13.074 0.08256
35.259 0.1652 35.951 0.1591
55.414 0.1972 54.509 0.2037
76.701 0.2287 76.770 0.2377
0.257 14.058 0.1218 0.515 11.365 0.1524
36.859 0.1966 32.254 0.2047
76.770 0.2536 76.574 0.2897

Table 6. Equlibrium Data of MDEA-PZ-H2O System

MDEA, PZ, p s, xp × 103, yp × 103, py × 103,
kmol/m3 kmol/m3 T, K kPa mol/mol mol/mol kPa ψ
4.28 0.103 377.1 8.722 3.898 10.56 10.42 3.263
4.28 0.308 377.6 9.010 12.22 34.38 33.93 3.245
4.28 0.696 378.1 9.297 26.89 71.59 72.54 3.492
3.46 0.103 376.0 8.196 3.283 7.305 7.401 3.635
3.46 0.154 376.5 8.459 4.647 9.653 9.780 4.019
Figure 2. Comparison of calculated and experimental data. 3.46 0.206 376.8 8.591 4.907 11.10 11.247 3.748
1.72 0.103 374.2 7.148 2.984 3.052 3.092 6.899
Table 4. Enthalpy Change of Absorption of CO2 in 4.28 1.72 0.154 374.5 7.406 4.470 4.591 4.652 7.117
kmol/m3 MDEA Solution 1.72 0.206 374.7 7.537 4.474 5.628 5.702 5.914
y
in the PZ-MDEA-H2O system. The initial MDEA and
0.1 0.2 0.3 0.4 0.5 ref piperazine concentrations ranged from 1.72 to 4.28 and
-∆Hs 61.26 59.78 56.18 55.43 54.04 this work from 0.1 to 0.70 kmol/m3, respectively. In the vapor
-∆Hs 59.8 53.2 Jou (1982) phase of PZ-MDEA-H2O, the content of MDEA is too
experimental and calculated CO2 partial pressures is low to determine, so it can be neglected.
less than 15%. Define Ψ as

Enthalpy Change of Absorption of CO2 Ψ ) xpps/py (22)

The enthalpy of absorption of CO2 in 4.28 kmol/m3
aqueous MDEA solutions is calculated by the Gibbs- where ps is the saturated pressure of pure piperazine,
Helmholtz equation. py is the vapor partial pressure of piperazine, and xp is

|
the liquid-phase molar fraction of piperazine. Ψ is the
∆Hs ∂ ln pCO2 strength of deviation between the theoretical solution
) (21) and the real solution of PZ-MDEA-H2O. If Ψ < 1, the
R ∂(1/T) y
solution is a positive deviation system; otherwise, it is
The plots of ln pCO2 vs 1/T are linear within the a negative deviation system.
accuracy of the data, and this indicates that Hs is Experimental data and calculated values are listed
dependent on conversion of CO2. The values of Hs are in Table 6. It is found that PZ-MDEA-H2O is a typical
presented in Table 4. negative deviation system and the value of Ψ decreases
The enthalpy change of absorption depends on the with MDEA concentration.
loading of CO2, and the value agrees well with the data
reported by Jou (1982). Conclusions
Effect of Piperazine on CO2 Loading Solubility data of CO2 and the vapor pressure of water
in 3.04-4.28 kmol/m3 aqueous MDEA solutions were
It has been proved that piperazine is an effective
obtained at temperatures ranging from 40 to 100 °C and
absorption accelerator in aqueous MDEA solutions used
CO2 partial pressures ranging from 0.876 to 1013 kPa.
for the removal of carbon dioxide (Xu et al., 1992).
A thermodynamic model was proposed and used for
Therefore, the effect of piperazine on the CO2 partial
predicting CO2 solubility and water vapor pressure.
pressure should be investigated.
The enthalpy change of absorption of CO2 in a 4.28
Solubility data were measured in 4.28 kmol/m3 MDEA
kmol/m3 MDEA solution was estimated. It changes
solutions, and the piperazine concentration ranged from
with CO2 loading, and the value agrees well with the
0 to 0.515 kmol/m3. Solubility data have shown that
data reported by Jou (1982).
the CO2 loading increases with the concentration of
The effect of piperazine concentration on CO2 loading
piperazine, and the results are shown in Table 5.
in MDEA solutions was determined at piperazine
concentrations ranging from 0 to 0.515 kmol/m3. The
Partial Pressure of Piperazine in the
results show that piperazine is beneficial to the CO2
PZ-MDEA-H2O System
loading in MDEA solutions.
Experiments were carried out in an Ellis cell to The equilibrium partial pressure of piperazine in the
determine the equilibrium partial pressure of piperazine PZ-MDEA-H2O system was measured in an Ellis cell.
 Ind. Eng. Chem. Res., Vol. 37, No. 4, 1998 1477

The initial MDEA and piperazine concentrations ranged and CO2 Solubility in Aqueous Mixtures of MDEA with MEA
from 1.72 to 4.28 and from 0.1 to 0.7 kmol/m3, respec- or DEA. Ind. Eng. Chem. Res. 1991, 30, 543.
tively. Results show that the PZ-MDEA-H2O system Barth, D.; Tondre, C.; Delpuech, J. J. Kinetics and Mechanisms
is a typical negative deviation system, in which the of The Reactions of Carbon Dioxide with Alkanolamines: A
strength of deviation decreases with MDEA solutions. Discussion Concerning The Cases of MDEA and DEA. Chem.
Eng. Sci. 1984, 39 (12), 1753.
Bhairi, A. M. Experimental Equilibrium Between Acid Gases and
Acknowledgment Ethanolamine Solutions. Ph.D. Dissertation, University of
This work was supported by the Science Foundation Oklahoma, Stillwater, OK, 1984.
of National Education Committee of People’s Republic Chakma, A.; Meisen, A. Solubility of CO2 in Aqueous Methyldi-
of China. ethanolamine and N,N-Bis(hydroxyethyl)piperazine Solutions.
Ind. Eng. Chem. Res. 1987, 26 (12), 2451.
Chen, C. C.; Britt, H. I.; Boston, J. F.; Evans, L. B. Extension and
Nomenclature Application of the Pitzer Equation for Vapor-Liquid Equilib-
a ) activity, mol/kg rium of Aqueous Electrolyte Systems with Molecular Solutions.
Aφ ) constant in the Debye-Huckel equation AIChEJ 1979, 25 (5), 820.
H ) Henry’s constant, kmol/(m3‚kPa) Edwards, T. J.; Maruer, G.; Newman, J.; Prausnitz, J. M. Vapor-
∆Hs ) enthalpy of absorption, kJ/mol Liquid Equilibria in Multicomponent Aqueous Solutions of
I ) ionic strength, mol/kg Volatile Weak Electrolytes. AIChE J. 1978, 24 (6), 966.
K ) equilibrium constant Frazier, H. D.; Kohl, A. L. Selective Absorption of Hydrogen Sulfide
MA ) total amine concentration, kmol/m3 From Gas Stream. Ind. Eng. Chem. 1950, 42 (11), 2288.
mi ) molality, mol/kg of solvent Haimour, N.; Bidarian, A.; Sandall, O. C. Kinetics of The Reaction
Mw ) molecular weight of water, 0.018 kg/mol Between Carbon Dioxide and Methyldiethanolamine. Chem.
p ) pressure, kPa Eng. Sci. 1987, 42 (6), 1393.
pws ) saturation pressure of water, kPa Ho, B. S.; Eguren, R. R. Solubility of Acidic Gases in Aqueous DEA
py ) vapor-phase partial pressure of piperazine and MDEA Solutions. AIChE Spring National Meeting, New
T ) temperature, K Orleans, LA, 1988; Paper 69A.
xp ) liquid-phase molar fraction of piperazine Jou, F. Y.; Mather, A. E.; Otto, F. D. Solubility of H2S and CO2 in
y ) CO2 loading, mol of CO2/mol of MDEA Aqueous Methyldiethanolamine Solutions. Ind. Eng. Chem.
yi ) vapor-phase molar fraction of species i Process. Des. Dev. 1982, 21 (4), 539.
yp ) vapor-phase molar fraction of piperazine Jou, F. Y.; Carroll, J. J.; Mather, A. E.; Otto, F. D. Solubility of
Zi ) ionic charges on species i Mixtures of Hydrogen Sulfide and Carbon Dioxide in Aqueous
N-Methyldiethanolamine Solutions. J. Chem. Eng. Data 1993a,
Greek Symbols 38, 75.
β(0), β(1) ) interaction parameters Jou, F. Y.; Carroll, J. J.; Mather, A. E.; Otto, F. D. The Solubility
γ ) molar activity coefficient of Carbon Dioxide and Hydrogen Sulfide in a 35 Wt % Aqueous
Φ ) vapor-phase fugacity coefficient Solution of Methyldiethanolamine. Can. J. Chem. Eng. 1993b,
Ψ ) strength of deviation, defined by eq 22 71, 264.
F ) density, g/cm3 MacGregor, R. J.; Mather, A. E. Equilibrium Solubility of H2S and
CO2 and Their Mixtures in a Mixed Solvent. Can. J. Chem. Eng.
Superscript 1991, 69, 1357.
s ) saturation Pitzer, K. S.; Mayorga, G. Thermodynamics of Electrolytes, II. J.
Phys. Chem. 1973, 77 (19), 2300.
Subscripts Shen, K. P.; Li, M. H. J. Solubility of Carbon Dioxide in Aqueous
w ) water Mixtures of Monoethanolamine With Methyldiethanolamine. J.
i, j ) species or component Chem. Eng. Data 1992, 37, 96.
c ) calculation Xu, G. W.; Zhang, C. F.; Qin, S. J.; Wang, Y. W. Kinetics Study on
e ) experiment Absorption of Carbon Dioxide into Solutions of Activated
Methyldiethanolamine. Ind. Eng. Chem. Res. 1992, 31 (3), 921.
Amine Abbreviations Zhou, J. J.; Xu, Z. H. Vapor-Liquid Equilibrium Study on Ternary
MDEA ) N-methyldiethanolamine Aqueous Solutions of Volatile Weak ElectrolytessNH3-CO2-
PZ ) piperazine H2O, NH3-H2S-H2O, NH3-SO2-H2O Systems. J. Chem. Ind.
R3N ) N-methyldiethanolamine Eng. (China) 1983, 3, 234.

Literature Cited Received for review October 17, 1995

Revised manuscript received December 11, 1997
Austgen, D. M.; Rochelle, G. T.; Chen, C. C. Model of Vapor-Liquid Accepted January 7, 1998
Equilibrium for Aqueous Acid Gas-Alkanolamine Systems. 2.
Representation of H2S and CO2 Solubility in Aqueous MDEA IE9506328