J. Chem. Eng.

Data 1989, 34, 355-360 355

solvent mixtures has not been done since it is not available in Reglstry No. CO, 630-08-0; DMF, 68-12-2; EtOH, 64-17-5; BuNH,,
the literature. 109-73-9; cyclohexene, 110-83-8.

Glossary
Llterature Cited
S solubility of CO, m3/m3
h height of the water column in the gas desorption (1) Gillies, M. T. C ,-Based Chemicals from Hvdrogen and Carbon Mon
oxide; Noyes Data Corporation: Park RMge, NJ, 1982.
-
apparatus, m (2) Taqul Khan, M. M.; Halligudi, S. 8.; Abdi, S.H. R. J . Mol. Catal. 1888,
S1 solubility of CO at atmospheric pressure and room 45, 215.
temperature, m3/m3 (3) Botteghl, C.; Ganzerla, R.: Lenarda, M.; Moretti, G. J. Mol. Catal.
1987, 40, 129.
H Henry's coefficient of solubility, kmol/(m3.kPa) (4) Taqui Khan, M. M.; Halligudi, S. B.; Abdi, S. H. R. J. Mol. Catal., in
P barometric pressure, kPa press.
( 5 ) Chaudhary, V. R.; Parande, M. G.; Brahme, P. H. Ind. Eng. Chem.
PW vapor pressure of water at T w ,kPa Fundam. 1982, 21, 472.
pco partial pressure of CO in the autoclave, kPa (6) Voael. A. I. A Textbook of Practical Organic Chemistw: Longman:
New York, 1978; p 332.
T temperature in the autoclave, K
(7) Battino, R.; Clever, H. L. Chem. Rev. 1986, 66, 395.
TW temperature in the gas buret, K (8) Taqui Khan, M. M.; Halligudi, S. B. J. Chem. Eng. Data 1988, 33,
V volume of water displaced by the desorbed CO gas, 276.
(9) Dake, S. 8.; Chaudhari, R . V. J. Chem. Eng. Data 1985, 3 0 , 400.
m3
volume of the liquid sample withdrawn from auto-
clave, m3 Received for review October 31, 1988. Accepted April 12, 1989.

Solubilities of Carbon Dioxide in Water and 1 wt % NaCl Solution at

Pressures up to 10 MPa and Temperatures from 80 to 200 OC

John A. Nighswander, Nicolas Kaiogerakis, and Ani1 K. Mehrotra'

Department of Chemical and Petroleum Engineering, The University of Calgary, Calgary, Alberta, Canada T2N IN4

T a b l e I. PVT D a t a for t h e C 0 2 - W a t e r S y s t e m at T=
Experimental gas solublilty data for the C0,-water and 75-250 "C a n d P = 1-15 MPa
C0,-1 wt % NaCi solution blnary systems are reported.
Measurements were made at pressures up to 10 MPa and
source T, "C P, MPa no. data uta
temperatures from 80 to 200 OC. A thet'modynamlc model Drummond ( I ) 80-250 3-15 34
of these systems Is also presented. The model employs Ellis and Golding (2) 175-250 1.5-9 6
Malinin and Kurovskaya (3) 100-150 5 2
the Peng-Robinson equation of state to represent the Malinin and Savelyeva ( 4 ) 75 5 1
vapor phase and an empirical Henry's law constant this work 80-200 2-10 32
correlation for the liquid phase. I t is shown that the
salting-out effect of the 1 wt YO NaCl solution on CO, greater than 15 MPa. A summary of these data was provided
solubility Is small. Also described is a new experimental by Drummond ( 7 ) .
apparatus consisting of a variable-volume equllibrlum cell The total dissolved solids concentration in a "typical" res-
enclosed in a constant temperature controlled oven and ervoir produced water is approximately 0.7 wt % (5).The
the procedure used in conducting the experiments. solids primarily consist of sodium, potassium, chloride, and
bicarbonate ions. On the basis of this, it was decided to per-
form equilibrium experiments on a synthetic salt solution con-
Introduction taining 1 wt % NaCl for which no data exist in the literature.
Carbon dioxide solubility data in water and dilute salt solutions The data available at higher salt concentrations (3-35 wt %)
were summarized by Drummond ( 7 ). At NaCl concentrations
typical of oil-field-produced waters (approximately 1 wt %) are
greater than 6 wt YO,the logarithm of the Henry's law constant,
required in the modeling of many enhanced oil and bitumen
In ( H J , follows the linear Setschenow relationship with salt
recovery processes. Applications are also found in geochem-
ical and natural gas systems. I n spite of their significance, concentration. However, it has been shown that the relationship
is invalid in the low salt concentration range (5).
limited data on the C0,-water and no data on the C0,-1 wt %
Thermodynamic models utilizing cubic equations of state are
NaCl solution systems are available in the literature.
I n the range of temperature and pressure of interest to this widely used for the modeling of oil and gas phases in petroleum
study, data for the C0,-water system were given by Drummond reservoir simulation. However, attempts at modeling the
aqueous phase with these equations have concluded that ac-
(I), Ellis and Golding (2),Malinin and Kurovskaya (3),and Ma-
curate predictions of gas solubility are difficult to obtain (6-8).
linin and Savelyeva ( 4 ) . A summary of the data is provided in
Furthermore, the effect of salts in the water cannot be repre-
Table I. Both Drummond ( 7 ) and Ellis and Golding (2) have
reported additional data in the 250-350 OC range. Also, Table sented by a cubic equation of state. Instead, empirical Henry's
law constant correlations have been used to model the aqueous
I does not include numerous studies that reported solubility
phase with equations of state used for the oil and vapor phases
measurements at temperatures below 75 OC or at pressure
(9-73).
The effect of the inclusion of NaCl on CO, solubility has been
* Author to whom correspondence should be addressed modeled empirically by Drummond ( 7 ) and semiempirically by

0021-9568/89/1734-0355$01.50/0 C 1989 American Chemical Society

358 Journal of Chemical and Engineering Data, Vol. 34, No. 3, 1989

8 II I
INCONEL 6 2 5

316 STAINLESS

SAMPLING LINES

OVERFLOW

THERMOCOUPLE 1'APS

HYDRAULIC FLUID LINE

PRESSURE TAPS

I U

LEGEND

fl VIEWING WINDOW FLOW CONTROL VALVE

@ THERMOCOUPLE x ON / O F F VALVE

@ PRESSURE GAUGE xi CHECK V A L V E 3/8" TUBING

@ DP CELL QUICK CONNECT V A L V E 1/8" TUBING INSIDE

0 GAS REGULATOR
@ SAMPLING LINE
Figure 1. Schematic diagram of experimental apparatus.
Flgure 2. Equilibrium cell (dimensions in cm).
Li and Nghiem (9)and Barta and Bradley (74). These models
were based on CO, solubility data for more concentrated NaCl
Procedure. Following is a brief description of the experi-
solutions (>6 wt %), and they do not accurately predict COP
mental procedure used for the vapor-liquid equilibrium (VLE)
solubility in the 1 wt % NaCl solution (5).
experiments. Referring to Figure 1, approximately 500 mL of
I n this paper, Henry's law constant correlations for the liquid
liquid was injected from the liquid reservoir into the equilibrium
phase in the C0,-water and the C02-1 wt % NaCl solution
cell, ensuring complete removal of air bubbles. The cell was
systems are used in conjunction with the Peng-Robinson
then completely sealed and heated to the test temperature.
equation of state (PR EOS) (75)for the modeling of the vapor
Once thermal equilibrium was achieved, approximately 500 mL
phase. I t is shown that the proposed model accurately predicts
of CO, at 6 MPa was added from the C02 cylinder. The system
C02 solubility in the range of temperature and pressure con-
pressure was then increased to about 11 MPa by inserting the
sidered.
cell piston.
Experimental Sectlon Subsequently, the equilibrium cell was agitated until equilib-
rium at 10 MPa was achieved. Equilibrium was assumed when
Apparatus. A schematic of the experimental apparatus is no cell pressure variation was observed over a 12-h period.
shown in Figure 1. The apparatus consists of a 9.8 cm inside Once equilibrium was reached, agitation was stopped and a
diameter equilibrium cell which is constructed of Inconel 625 sample of the liquid phase was withdrawn for analysis. The
and fitted with a piston for volume adjustment (Figure 2). The piston was then withdrawn to lower the pressure to approxi-
cell is enclosed in a temperature-controlled oven and is atta- mately 8 MPa.
ched to a rotating cam device to provide gentle agitation. The The above procedure was then repeated, decreasing the
cell is fitted with ports for sampling from both the top of the cell pressure by 2-MPa steps until the last samples at 2 MPa were
and through the cell piston and rod. Cell volume is accurately collected. The above steps were performed at temperatures
adjusted by displacing silicone hydraulic fluid to or from the cell of 80, 120, 160, and 200 O C .
with a Jefri high pressure displacement pump. Sample Withdrawal. An important consideration in the de-
Two iron-constantan type J thermocouples were used to sign of any PVT cell is to devise a method to withdraw samples
monitor the cell temperature, one in the upper equilibrium without disrupting equilibrium. This is especially crucial with
chamber and the other in the lower flange. Calibration of the regard to liquid samples, as the flashing of even a small volume
thermocouples was performed in boiling water, and the agree- of liquid from the cell can result in a significant reduction in the
ment obtained was well within the manufacturer's specified cell pressure. To avoid this problem we used a novel sampling
accuracy (f0.5%). The deviation in temperature measured at technique which is described below.
the top and lower flange of the cell was within the specified Referring to Figure 1, a back pressure equal to the equilib-
error for all the tests, indicating that temperature gradients rium cell pressure was maintained in the sampling line with
within the cell were negligible. nitrogen. To obtain a liquid sample, the sampling valve was
Cell pressure was monitored with a precision gauge (3-0 opened and the piston was very slowly injected, producing a
Instrument Inc.) as well as a Rosemount Alphaline gauge slight increase in the cell pressure. As a result, the sample was
pressure transmitter. Both gauges were calibrated to an ac- forced out against the back pressure and into an approximately
curacy of 35 kPa over the range of operation considered. Both 3-mL sampling cell. When sufficient amount of fluid had been
the pressure transmitter and the thermocouple reading were withdrawn, as verified by visual observation through the viewing
continuously monitored and recorded by use of a SAFE8000 window, the sampling valves were closed and the sample was
front-end device connected to a supervising microcomputer. isolated and removed for analysis. I t should be noted that
 Journal of Chemical and Engineering Data, Vol. 34, No. 3, 7989 357

2.0

8 1

GRADUATED LlPUlD
C Y LlNDER RESERVOIR

0 2 4 6 8 10 12
Pressure (MPa)
Flgure 4. Comparison of C0,-water solubility data with model pre-

n dictions.
PERISTALTIC

PUMP

A A
- 1
SAMPLE
m
E
950
B
A
0
0
A>
n
I,
-
T = 80 O C

-.-
Y
x
c
900 - 0 n
Y Y Y
T = 120 O C

T= 160 O C
Flgure 3. Sample analysis apparatus. VI
0
a 850 - -0
0 -b O
--$

nitrogen was prevented from entering the cell by use of a check 0 T = 200 O C
valve. Furthermore, the solubility of nitrogen in the liquid is very
small (9). The back pressure was maintained constant by -""
providing a continuous bleed from the sample line. Pressure 0 2 4 6 E 10 12
fluctuations were kept to a minimum with the help of a large- Pressure (MPa)
volume damping cell connected to the sampling line. Figure 5. CO, saturated water densities.
Sample Ana&s/s. After a sample cell was removed, it was
dried and weighed to an accuracy of 0.001 g. The sample cell
5.00 I
was then connected to a depressurization chamber (Figure 3)
where the gases evolved from the sample (CO, and H,O)
bubbled through a 1 N sulfuric acid solution. Water vapor
condensed in the solution, and all the COPescaped the acidic
environment and was collected in a volumetric U-tube. The
4.75 I n o

volume of this gas was then determined to an accuracy of 0.5

mL at ambient temperature and barometric pressure. The
empty sample cell was then removed from the depressurization
v
L
c I A
= This Work'
= Drummond (1)
cell, dried, and weighed once again. From the mass and carbon 0 = Ellis and Golding (2)
dioxide content of the sample, the density and the gas solubility = Maiinin et al. (3,4)
values were calculated. -.--
Error Ana&*. I n addiiion to the experimental error involved 50 100 150 200 250
in sample mass, volume CO, evolved, temperature, and pres- Temperature ("C)
sure measurements, the uncertainty in the sample cell volume Flgure 6. Comparison of carbon dioxide solubility data in water. (*:
must be considered in an overall error analysis for the density The data for this work represent the average of 7-10 data points.)
and COPsolubility measurements. The sample cell volume was
calibrated by filling the cell with distilled water and measuring
the mass when it was full and empty. From pure water density thus making it possible to evaluate both the accuracy and re-
producibility of the data from the new experimental apparatus.
data (76), the sample cell volume was determined as 3.302
In this work, experimental density and liquid-phase COPsolubility
cm3 with a standard deviation of 0.007 cm3. Using this value
as the uncertainty in the cell volume and the results of a typical measurements were made at pressures of 10, 8, 6, 4,and 2
sample analysis, the standard error in the density and gas MPa for each of the four temperatures of 80, 120, 160, and
200 OC. Results of our experiments are plotted in Figure 4 for
solubility measurements were calculated to be 3 kg/m3 and
0.02 mol %, respectively. the liquid-phase solubility data and Figure 5 for the C02 satu-
Materials. The carbon dioxide used in all experiments was rated water densities. As shown in Figure 4, the solubility of
Coleman Instrument grade with a minimum purity of 99.99 % . CO, increases with an increase in pressure, but decreases with
Water was purified by using a reverse osmosis system to a an increase in temperature. The range of CO, solubility is from
minimum resistance of 17 MR. The synthetic 1 wt % NaCl approximately 0.2 mol YO at 200 O C and 2 MPa to approxi-
mately 1.7 mol % at 80 OC and 10 MPa. However, the CO,
solution was prepared with Fisher Scientific laboratory grade
NaCl with 99.98% purity. saturated water densities do not indicate any pressure depen-
dence. The density values are slightly lower than those for pure
Results and Discussion water and decrease from 956 kg/m3 at 80 OC to 853 kg/m3
at 200 OC. The maximum standard deviation for the density
,- .
CO Water Vapor -Liquid Equilibrum The first system measurements is 6 kg/m3 or 0.8 % .
chosen for study was the C0,-water VLE. This system was To compare our data with those reported in the literature, the
selected due to the availability of previous experimental data, COPsolubilities have been expressed in terms of Henry's law
358 Journal of Chemical and Engineering Data, Vol. 34, No. 3, 1989

Table 11. Experimental Solubility, Density, and Henry's

Law Data for the C02-Water System
X P m8' P, In (I-I,), 1.5 .
T,"C P,kPa mol % mol/ka 6; ka/m3 kPa 0"
0
80.5 2330 0.47 0.262 0.9280 951 8.967 .I-

80.2 4310 0.85 0.476 0.8698 956 8.911 ; 1.0 .

80.6 4340 0.84 0.470 0.8695 951 8.930 a
L
80.5 6110 1.12 0.629 0.8203 963 8.906 -0
80.5
80.5
7760
7840
1.37
1.38
0.771
0.777
0.7767
0.7746
950
951
8.868 = 0.5
8.868
80.3 10160 1.64 0.925 0.7166 96 1 8.846 -Model Predictions
79.7 10180 1.66 0.937 0.7145 96 1 8.833 -.- ~~

In (If,) = 8.891 f 0.045 0 2 4 6 a 10 12

119.9 2110 0.34 0.189 0.9552 936 9.155 Pressure (MPa)

120.1 2130 0.35 0.195 0.9549 933 9.135 Figure 7.Comparison of C0,-1 wl % NaCl solution solubility data with
120.1 4050 0.65 0.363 0.9155 929 9.141 model predictions.
119.9 4090 0.67 0.374 0.9146 926 9.120
120.0 5960 0.91 0.510 0.8785 933 9.144

I
1000
120.0 6050 0.91 0.510 0.8765 926 9.157 "
A
;-4
119.9 8110 1.19 0.668 0.8393 93 1 9.124 T = 80 "C 4,
120.1
120.1
8160
9960
1.20
1.42
0.674
0.799
0.8387
0.8081
926
930
9.111
9.097 ""
"E
.E 1 T
0
= 120 "C
a
a 0
0
0 0
120.3

159.7
9980

2040
1.45

0.29
0.817 0.8081
= 9.126 f 0.025
In (H,)
0.161 0.9740
930

896
9.078

9.050
-Y

r
A
900 T= 160 "C
u
o
U
R 0

159.9 2060 0.29 0.161 0.9738 893 9.061

159.9 3910 0.54 0.301 0.9458 894 9.214
160.0 3940 0.56 0.313 0.9454 890 9.185
159.7 5940 0.80 0.448 0.9177 890 9.249
159.8 8060 1.10 0.617 0.8908 890 9.203 ---
159.8 8070 1.10 0.617 0.8907 890 9.204 0 2 4 6 8 10 12

In (H,)
= 9.167 f 0.078 Pressure (MPa)
197.8 2450 0.22 0.122 0.9914 851 8.946 Figure 8. CO, saturated 1 wt YO NaCl solution densities.
198.1 4560 0.57 0.318 0.9648 840 9.107
198.0 4600 0.52 0.290 0.9643 860 9.212
197.7
198.0
197.6
198.1
198.0
6240
7970
8020
10200
10210
0.78
1.06
1.09
1.30
1.40
0.436
0.595
0.612
0.731
0.788
0.9471
0.9313
0.9304
0.9126
0.9124
854
853
850
857
856
9.193
9.166
9.147
9.215
9.141
9'50
9.25 I
In (If,) = 9.141 f 0.087

constants and these are plotted in Figure 6. The Henry's law

constant, H, is defined as -Model Predictions (Equation 9)
A = Experimental Data (Water)
0 = Experimental Dota (1 wt% NaCI)
50 100 150 200 250
Temperature ("C)
Figure 9. Comparison of data to proposed correlation.
I n this equation, the partial molar volume of CO, at infinite
dilution, O,", is estimated as ( 7 )
molality (m,). Therefore, absolute measurements of solubility
P," = 1000P,"(at 25 "C)/p (2) should show a minimum in solubility at a temperature higher
than that corresponding to the maximum in the Henry's law
where V,"(at 25 "C) is 0.033 m3/kmol (77) and p is the ex- constant. This explains why the Henry's law constant in Figure
perimental liquid density. Since at each temperature up to 10 6 decreases beyond 160 OC, while the absolute solubility of CO,
solubility measurements were made, a Henry's law constant in water continues to decrease as shown in Figure 4.
was calculated for each point and the average of all these ,-
CO 1 wi % NaCl Wutlon Vapor -Uquld Equlbium The .
values used as the Henry's law constant for that temperature. data obtained for the COPsolubility and NaCl solution densities
The solubility data and the corresponding values of the Henry's are plotted in Figures 7 and 8, respectively. Comparison of the
law constant for the C0,-water system are given in Table 11. CO, solubility in salt solution with that in water (Le., Figures 7
I t should be noted that Drummond ( I ) had used the pure com- and 4) indicates that the solubility is slightly reduced by the
ponent vapor-phase fugacity coefficient and not the fugacity inclusion of the salt. I t is also observed that the difference in
coefficient for CO, in the vapor phase mixture as defined in eq CO, solubility between the water and the salt solution becomes
1. To provide a consistent basis for comparison, our data were smaller with increasing temperature. Also, CO, saturated salt
analyzed using the pure component coefficient and are pres- solution densities are lower than those for the pure salt solution
ented in Figure 6. and show no apparent pressure dependence.
As seen in Figure 6, our data agree with the literature values As with the C0,-water data, the data for the C0,-1 wt %
quite well. The data suggest a broad maximum in the Henry's NaCl solution binary have been reduced in terms of the Henry's
law constant between 150 and 225 "C. I t is noted that this law constant as defined in eq 1. I n these calculations, the
maximum in Henry's law constant does not correspond to a effect of the salt on the vapor pressure of the aqueous phase
minimum in the COPsolubility since with increasing temperature was approximated as 99.4% of the vapor pressure of pure
the partial pressure &,P) of COP fails more rapidly than the water at the same temperature. An analysis of vapor pressure
 Journal of Chemical and Engineering Data, Vol. 34, No. 3, 1989 359
100 . Table 111. Experimental Solubility, Density, and Henry's
Law Data for the C02-l wt YO NaCl Solution System
80 T=200 *C
%' mgt
0 0 = Experimental Data T,"C P,kPa mol % mol/kg $i
r"
+ 60 -Model Predictions 80.3 4040 0.74 0.414 0.8777 9.001
E 80.2
80.2
4130
6020
0.76
1.01
0.425
0.566
0.8750
0.8222
974
986
8.983
8.995
2
- 40
T=160 'C 80.2 6070 1.08 0.606 0.8208 982 8.933
I 80.3 8040 1.35 0.759 0.7691 988 8.904
20 80.5 8050 1.34 0.754 0.7692 981 8.913
80.0 9490 1.41 0.794 0.7320 986 8.967
0-
80.1 9940 1.54 0.868 0.7214 981 8.899
In (H,)
= 8.949 f 0.042
120.0 2110 0.29 0.161 0.9553 955 9.304
120.1 2140 0.31 0.173 0.9547 951 9.252
120.1 4040 0.59 0.329 0.9158 945 9.225
120.1 4060 0.61 0.341 0.9154 942 9.197
120.4 6020 0.97 0.544 0.8778 935 9.076
120.1 6040 0.97 0.544 0.8771 928 9.079
120.1 8150 1.20 0.674 0.8389 934 9.106
119.9 8160 1.18 0.663 0.8384 945 9.124
120.0 9970 1.39 0.782 0.8078 935 9.104
120.2 10030 1.38 0.777 0.8072 945 9.116
In (H,)
= 9.158 f 0.080
160.2 2150 0.28 0.156 0.9724 916 9.128
160.0 4100 0.54 0.301 0.9431 903 9.238
160.1 4120 0.55 0.307 0.9429 911 9.225
159.9 6040 0.84 0.470 0.9165 909 9.185
159.8 6060 0.87 0.487 0.9162 902 9.154
159.9 7950 1.06 0.595 0.8923 912 9.202
160.0 8030 1.03 0.578 0.8914 918 9.241
159.9 9930 1.31 0.737 0.8688 915 9.186
160.1 9970 1.36 0.765 0.8686 914 9.151
In (H,)
= 9.190 * 0.040
200.0 4120 0.47 0.262 0.9708 869 9.077
200.1 4150 0.47 0.262 0.9705 879 9.086
199.8 6010 0.69 0.386 0.9508 878 9.199
199.9 6010 0.76 0.425 0.9509 87 1 9.101
199.9 8120 0.94 0.527 0.9317 873 9.237
200.0 8130 0.98 0.549 0.9317 878 9.196
200.5 9930 1.32 0.742 0.9177 862 9.090
In (H,)
= 9.141 f 0.067

The CO, molality is directly related to mole fraction in the liquid

phase as
m , = [55.494~,/(1 - x,)] (7)
and H," is defined as

(8)

where In (H,), over the temperature range of 353-473 K and

the pressure range from 2 to 10 MPa, is given by the following
empirical quadratic expressions:
for pure water
In (H,) =
1.384196 i-
3.568271 X 10-'(T) - 4.079688 X 10-5(T)2
for 1 wt % NaCl solution
In (H,) =
1.672958 4- 3.487636 X 10-,(T) - 4.037500 X 10-5(T)2
(9)
I t should be emphasized that H, has the units of kilopascals
and is evaluated by using $,, the fugacity coefficient of COP
in the vapor-phase mixture. Since Henry's law is valid for the
solute CO, the Lewis-Randall rule applies for the solvent. The
fugacity of H,O in the aqueous phase is determined by the
fk = H,"m, expression
360 Journal of Chemical and Engineering Data, Vol. 34, No. 3, 1989

X liquid-phase mole fraction

Y vapor-phase mole fraction
Greek Letters
The molar volume of the water or brine phase, v,, was P density, kg/m3
calculated from a correlation by Rowe and Chou (22). The 4 pure component vapor-phase fugacity coefficient
fugacity coefficient 4 wwt was evaluated from the Canjar and 4 vapor-phase fugacity coefficient in gas mixture
Manning correlation (9):
subscripts
4wwt= 0.9958 -I-9.6833 X 10-5(T') - 6.175X lO-'(T')' - 9 CO, component
3.08333 X ~ . O - ' ~ ( T ' ) ~ for T'> 90 OF i ith component
W H,O component
@wMt = 1 for T ' I 90 OF (1 1)
Superscripts
where T' is the temperature in O F . Pwsat for the salt solution
L liquid
was estimated as mentioned previously to be 99.4% of the
value for pure water.
v vapor
sat saturated
The results of the calculations are shown in Figures 4 and
7 as the calculated lines. Clearly, the model correlates the CO, Reglrtry No. CO, 124-38-9; NaCI, 7647-14-5.
solubility data quite well over the range of temperature and
Literature Cited
pressure.
(1) Drummond, S. E., Jr. R.D. Dissertation, Pennsylvania State University,
Conciuslons 1981.
(2) Ellis, A. J.; GoMing. R. M. Am. J . Sci. 1963, 261, 47.
A new PVT apparatus employing a novel technique for liquid (3) Mallnin, S.D.; Kurovskaya, N. A. Geochem. Int. 1975, 12, 199.
(4) Mallnln, S. D.; Savelyeva, N. I. Geochem. I n t . 1972. 5 . 410.
sample withdrawal has been developed and shown to provide (5) Nighswander, J. A.; Kalogerakis, N.; Mehrotra, A. K. P r m s of
accurate and reproducible gas solubility measurements at t e m the 4th UNITARIUNDP International Conference on Hsavy Crude and
Tar Sands, August, 1988; AOSTRA: Edmonton, Canada; p 8:l.
peratures up to 200 O C and pressure up to 10 MPa. I t was (6) Heidemann, R. A. AIChE J. 1974, 20, 847.
found that CO, solubility is slightly lower in a 1 wt % NaCl (7) Peng, D. Y.; Robinson, D. 6. Can. J. Chem. Eng. 1976, 54, 595.
solution than in pure water due to the salting-out effect. (8) Eveiin, K. A.; Moore, R. 0.; Heidemann, R. A. Ind. Eng. Chem. Pro-
cess Des. Dev. 1076, 15, 423.
A model of the C0,-water and CO,-1 wt % NaCl solution (9) Li, Y.; Nghiem, L. X. Can. J. Chem. Eng. 1966. 6 4 , 486.
vapor-liquid equilibria was also presented. The model uses the (IO) Luks. K. D.; Ftzgibbon, P. D.; Bancharo, J. T. Ind. Eng. Chem. Pro-
cess Des. Dev. 1976, 75, 326.
Peng-Robinson equation of state to represent the vapor phase (11) Heidemann, R. A.; Prausnitz, J. M. Ind. Eng. Chem. Process Des.
and a proposed Henry's law correlation for the liquid phase. Dev. 1977, 16, 375.
Predictions are shown to be very good over the range of tem- (12) Mehra, R. K.; Heidemann, R. A.; Aziz, K. SPE J. 1062, 22. 61.
(13) Nghbm, L. X.; Heidemann, R. A. Report CMG R12.02, Computer Mo-
perature and pressure considered. delling Group, Calgary, Canada, 1982.
(14) Barta, L.; Bradley, D. J. Geochlm. C o s m h l m . Acta 1966, 49, 195.
Glossary (15) Peng, D. Y.;Robinson, D. 6. Ind. Eng. Chem. Fundam. 1976, 15,
59.
f fugacity, kPa (16) Hass, J. L. U S . Geol. Survey Bull. 1076, 1421-A, 623.
pressure-corrected Henry's law constant, kPa (17) Reid, R. C.; Prausnitz. J. M.;Poling, B. E. The Properties of Gases and
E;*
k,
Henry's law constant for CO, in water, kPa
binary interaction parameter in the PR EOS
Liquids, 4th ed.;McGraw-Hill: New York, 1987; p 336.
(18) Reiss, H.; Frisch, H. L.; Helfand, E.; Lebowitz, J. L. J. Chem. Phys.
1960, 32, 119.
CO, molality in liquid phase, mol/kg (19) Lebowitz, J. L.; Heifand, E.; Praestgaard, E. J. Chem. Phys. 1965,
m, 43, 774.
P pressure, kPa (20) Pierotti, R. A. Chem. Rev. 1076, 76, 717.
R universal gas constant (8.314 kl'a m3/(kmol K)) (21) Pitzer, K. S.J . Phys. Chem. 1073, 7 7 , 268.
T temperature, K (22) Rowe, A. M.; Chou, J. C. S.J . Chem. Eng. Data 1070, 15, 61.
T' temperature, O F
V molar volume, m3/kmol Received for review September 26, 1988. Accepted Merch 17, 1989. We
thank the Alberta Oil Sands Technology and Research Authority (AOSTRA)
P," partial molar volume of CO, at infinite dilution, m3/ and the Natural Sciences and Engineering Research Council of Canada
kmol (NSERC) for the funding of this research.