This document presents a model for predicting phase equilibria of oil, gas, and water/brine mixtures using a cubic equation of state to model the oil and gas phases and Henry's law to estimate gas solubility in the aqueous phase. Henry's law constants for various compounds are correlated with pressure and temperature. The scaled particle theory is used to account for salt in the aqueous phase. Experimental data is presented showing the model provides an adequate representation of Henry's law constants.
This document presents a model for predicting phase equilibria of oil, gas, and water/brine mixtures using a cubic equation of state to model the oil and gas phases and Henry's law to estimate gas solubility in the aqueous phase. Henry's law constants for various compounds are correlated with pressure and temperature. The scaled particle theory is used to account for salt in the aqueous phase. Experimental data is presented showing the model provides an adequate representation of Henry's law constants.
Original Description:
Article
Original Title
Phase Oil Equilibria of Oil-water-brine_Yaun Kun Li
This document presents a model for predicting phase equilibria of oil, gas, and water/brine mixtures using a cubic equation of state to model the oil and gas phases and Henry's law to estimate gas solubility in the aqueous phase. Henry's law constants for various compounds are correlated with pressure and temperature. The scaled particle theory is used to account for salt in the aqueous phase. Experimental data is presented showing the model provides an adequate representation of Henry's law constants.
This document presents a model for predicting phase equilibria of oil, gas, and water/brine mixtures using a cubic equation of state to model the oil and gas phases and Henry's law to estimate gas solubility in the aqueous phase. Henry's law constants for various compounds are correlated with pressure and temperature. The scaled particle theory is used to account for salt in the aqueous phase. Experimental data is presented showing the model provides an adequate representation of Henry's law constants.
Phase Equilibria of Oil, Gas and Water/Brine Mixtures
from a Cubic Equation of State and Henrys Law
YAU-KUN LI and LONG X . NGHIEM Computer Modelling Group, 3512-33 Street N . W. , Calgary, Alberta T2L 2A6 This paper presents the predictions and calculations of the phase equilibria of oil, gas and water/brine mixtures where the oil and gas are modeled by a cubic equation of state and where the gas solubility in the aqueous phase is estimated from Henrys law. Henrys law constants for a variety of compounds of interest to the petroleum industry are correlated against pressure and temperature. The scaled-particle theory is used to take into account the presence of salt in the aqueous phase. An extensive match of experimental data is presented, showing the adequacy of the proposed model for Henrys law constant. On donne dans ce travail les pridictions et les calculs dkquilibre de phase des mClanges pktrole, gaz et eau-saumure, 00 le pktrole et les gaz sont reprksentis par une Cquation ditat cubique et ou la solubilite des gaz dans la phase aqueuse est estimie a Iaide de la loi dHenry. Les constantes de Henry sont corrClCes pour une large gamme de composis dintiret pour Iindustrie pktroliere, en fonction de la pression et de la tempkrature. On utilise la thiorie des particules ajusties pour tenir compte de la prisence de sel dans la phase aqueuse. On prksente une vaste comparaison de donnies expkrimentales qui montre la validiti du modkle propost5 pour les constantes de la loi dHenry. ater exists abundantly in hydrocarbon reservoirs, either W as a fluid originally in place or as an injected fluid from an enhanced oil recovery process. The exclusion of water from phase equilibrium predictions may lead to results which are inconsistent with field observations since large amount of light hydrocarbons and carbon dioxide can be dissolved in the aqueous phase. In addition, the presence of salt in the aqueous phase must also be taken into account because of its strong effect on the gas solubility. Cubic equations of state (EOS) have been used ex- tensively to model the oil and gas phases in phase equi- librium computations. Several attempts have been made to model the aqueous phase with the same EOS (Heidemann, 1974; Peng and Robinson, 1976b; Evelein et al., 1976). However, these authors found that accurate predictions of the aqueous-phase compositions are difficult to achieve. To overcome these difficulties, special modifications to the EOS parameters are required for the aqueous phase (Erbar et al., 1980; Peng and Robinson, 1980). Furthermore, the inclusion of the salt effect in an EOS is not obvious. Another approach, which was used by Luks et al. (1976), Heidemann and Prausnitz (1977), Mehra et al. (1982) and Nghiem and Heidemann (1982), consists of using Henrys law to model the gas solubility in the aqueous phase, and an EOS for the other phase(s). Luks et al. (1976) performed calculations for nitrogen-oxygen- water and air-water mixtures. Heidemann and Prausnitz ( 1977) calculated the compositions of combustion gases consisting of nitrogen, carbon dioxide and water. Mehra et al. 11982) estimated the gas solubility in the aqueous phase from Cysewski and Prausnitzs (1976) correlation. Since there did not appear to be any safely generalized correlations for Henrys law constants at this time (OConnell, 1980), Nghiem and Heidemann (1982) carried out flash calculations for mix- tures of crude oil and water with Henrys law constants determined from experimental data. However, only carbon dioxide was assumed soluble in the aqueous phase, and the presence of water in the hydrocarbon-rich phases was ne- glected. All the above references did not consider the effect of salt on the gas solubility. In this paper, correlations for Henrys law constants with respect to pressure and temperature for a variety of com- pounds of interest to the petroleum industry are developed using published experimental data. The effect of salt on the gas solubility in the aqueous phase is handled by using the scaled-particle theory (SPT) to modify Henrys law constants derived for pure water. A robust algorithm for three-phase flash calculations involving an aqueous phase is also discussed. Phase equilibria EQUILIBRIUM EQUATIONS For a liquid-gas- water system with n, components, the (1) necessary conditions for phase equilibrium are: lnfI v - I nf,[ =0 lnfiw - I nfi e =0 i =I , . . ,n, . . . . . . . . . . . . . and i =1,. . ,n, . . . . . . . . . . . . . ( 2 ) wheref,, (m =e , v, w ) is the fugacity of component i in phase m. The subscripts e , v and w denote respectively the liquid, vapor and aqueous phase. In this paper, the term liquid phase refers to the dense phase which is not rich in water. The latter is referred to as the aqueous phase. In phase equilibrium computations, it is convenient to define two sets of K-values (equilibrium ratios) as follows: (3) K,, =y,,/y,, i =I , . . ,n, . . . . . . . . . . . . . . . . . and (4) i =I , . . ,n, . . . . . . . . . . . . . . . . . where y,, denotes the mole fraction of component i in phase m. K,,, =y I w/ y , e 486 THE CANADI AN J OURNAL OF CHEMI CAL ENGI NEERI NG, VOLUME 64, J UNE 1986 The material balance on component i requires that Z , =F, y, , +F, y, , +F,V,,, i =I , . . . , n, . . . . ( 5 ) where z , is the feed composition (global mole fraction), and where F,,, is the mole fraction of phase m. The mole fraction y,,,, are related to z, and F,,, through the following relationship ~, , 11 =~ , , , , z , / ~ Kl yFq i =I , . . . . n, . . . . . . . . . (6) Y m =e , V , M J 4 =e , V, M' where, . . . . . . . . . . . . . . K, , I i = I 11, (7) The thermodynamic equilibrium equations (Equations I In K, , +In - In =0 i =I , . . . , n , ... (8) InK, , , +l n+,,, - I n+,, =0 i =1 . . . . . n, . . (9) where +,,,, denotes the fugacity coefficient of component i in phase m. and 2) can also be rewritten as follows: By definition, the mole fractions satisfy 11, C - I =o m =e , v , w . . . . . . . . . . . . . . (10) C F,,, - I =o m =e , V , ~ . . . . . . . . . . . . . ( 1 1 ) By using Equations (6) and (lo), the following equations 1'1 m can bederived: " I ( K, " - 112, n,. ' ( ' I v - '") =I = ' I F1 +F, K, , +F,,,K,,. =0. . . (12) i = I A three-phase flash calculation corresponds to solving the 2n,. +3 nonlinear equations (Equations 8, 9, 1 I - 13) for the 2n,. +3 primary unknowns K, , , K,,, F1 , F, and F,. The mole fractions y,," can betreated as dependent variables from Equation (6). For the components j which are assumed not soluble in the aqueous phase, take K,,, =0 and remove the corresponding Equation (9) from the equation set. The system of equations can be solved by either Newton's method or the quasi-Newton successive substitution method discussed in Nghiem and Li (1984) and Mehra et al. (1984). Since the aqueous phase is not modeled by an EOS, some modifications to the calculation procedure of Nghiem and Li (1984) are required. These are described in Appendix A. Note that the procedure does not require good initial guesses of the primary variables, nor a prior knowledge of the number of phases in equilibrium. THERMODYNAMIC MODELS As discussed earlier, the liquid and vapor phases are modeled by a cubic EOS, and the solubility in the aqueous phase is handled by Henry's law. Henry's law for a component sparingly soluble in the aqueous phase gives i f w ....................... (14) where the bold subscript w denotes the water component, and the regular subscript w denotes the aqueous phase. Hi is f i w =yI I YHi the Henry's law constant of component i in the aqueous phase. The variation of Henry's law constant with respect to pressure and temperature follows the differential equation (Smith and Van Ness, 1975) V,? h , , - hP +- RT? dT . . . . . . . . . . . d(ln Hi) =-dp RT where vIm is the partial molar volume of component i in the aqueous phase at infinite dilution, h,, is the enthalpy of component i in the gas phase, and hp is the enthalpy of component i in the aqueous phase at infinite dilution. The term h, , - hp depends strongly on temperature, and no general correlation describing its change with respect to temperature is available. It has also been found that vlx is usually not very sensitive to pressure. Therefore, for a given temperature T , integration of Equation (15) from p" to p gives ........... (16) where HP is the Henry's law constant at the reference pressure pp. Equation (16) can also be written as . . . . . . . . . . . . . . . . . . (17) In HI =In HY +vP(p - pY) / ( RT) In H, =In H,*+v, "p/(RT) In H,* =In HP - vl Spy/ (RT) . . . . . . . . . . . . . . . . . (18) where In this study, Equation (17) is used to correlate Henry's law constants from solubility data. H,* is referred to as the reference Henry's law constant. The molar volume at infinite dilution v;" is computed from the correlation of Lyckman et al. (1965) reported by Heidemann and Prausnitz (1977) in the form: (2) Pr i Vi =0.095 +2.35($) TP .............. RTci where C is the cohesive energy density of water given by: C =(h, - h i +p",: - RT) / v: . . . . . . . . . . . . . (20) p i is the saturated vapor pressure of water at the tem- perature T, v: is the molar volume of water at p: and T, and h i - hz is the enthalpy departure of liquid water at p i and T. The fugacity of the component water in the aqueous phase can readily be obtained from the fugacities of the solutes using the Gibbs-Duhem equation as in Prausnitz (1969): 0 p v, f Ww =yWw+i pi exp [IpL ~ d p ] ............... where +: is the fugacity coefficient of pure water at p i , and vi is the molar volume of pure water. In this paper, the vapor pressure of water p i and ( h: - h i ) are calculated respectively from the Frost- Kalkwarf- Thodos and the Yen- Alexander equation reported in Reid et a]. (1977). The molar volume vi is estimated from a correlation due to Chou reported in Rowe and Chou (1970). Alternatively, these values can also be obtained from steam tables. The fugacity coefficient is obtained from the following correlation which was found to match reasonably well the data of Canjar and Manning (1967): +i =0.9958 +9.68330 X 10-5T' - 6.1750 X IO-'T'* - 3.08333 X 10-'"Tf3; T' >90F ........................ (22a) (22b) =I ; T' 5 90F ........................ THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 64, JUNE 1986 487 106 Z z 8 103 W a 1 I I I I 0 50 100 150 200 250 102 I TEMPERATURE, C Figure 1 - Variation of the reference Henrys Law constant, HF, with temperature. 7 =1.8T - 459.67 (22c) where T is the temperature in K, and 7 the temperature in OF. . .. . . .. . . . . . , .. . . .. .. . Henrys law constants for gas solubility in water HENRYS LAW CONSTANTS FROM SOLUBILITY DATA Henrys law constants are obtained by matching the pre- dicted results to experimental binary solubility data. The gas phase is modeled with the Peng-Robinson EOS (Peng and Robinson, 1976a). The compositions of the aqueous and gas phases were matched by regressing on the EOS interaction coefficient di j between water and the solute, and the refer- ence Henrys law constant H:given in Equation (17). Since the Henrys law constants will beused to predict gas solu- bilities at reservoir conditions, more emphasis was given to matching the experimental data at pressures higher than 6.9 MPa whenever these are available. Hence the values of H: obtained do not correspond to Henrys law constants at zero pressure (i.e., p o =0 in Equations 16 and 18) discussed in the literature. Figure 1 shows the variation of H? with respect to tem- perature for a number of compounds. The dashed curves in Figure 1 represent extrapolated values of H?. As depicted in this figure, H: generally exhibits a maximum when plotted versus temperature. Jonah (1983) has found that In (Hi/fs) Y I METHANE-WATER SYSTEM - - I MOLE PERCENT OF METHANE IN VAPOR PHASE 0 Figure 2 - Vapor-phase composition of methane- water system. varies linearly with the inverse of the absolute temperature for the systems he investigated. Here, fs denotes the pure- solvent fugacity. For the binary systems considered in this paper, it has been found that In (H:/fi) deviates slightly from the suggested linear behavior, and follows more accu- rately a second order polynomial in l/T. The coefficients of these polynomials as well as the Peng-Robinson EOS in- teraction coefficients obtained by regression are given in Table 1. The sources of the experimental data used in the regression are also reported. The fugacityf; is obtained by multiplying given in Equation (22) by the water satur- ation pressure pa. Because of space limitation, the match is presented only for a few binary systems in the following. Figures 2 and 3 depict the vapor and aqueous phase compositions of the methane- water binary system. Figures 4 to 6 compare the experimental and calculated composition of the propane which exhibits a three-phase separation. Figures 7 and 8 show the experimental and calculated results for the carbon dioxide-water system. For this system it was found that a temperature-dependent interaction coefficient is required to achieve a good match of the experimental data. The results show that the use of the Peng-Robinson EOS with Henrys law provides adequate predictions of the behavior of binary systems involving water. Although not shown here, similar predictions were also obtained for the other systems shown in Table 1. It has also been found that Henrys law is applicable to the systems considered up to a temperature of about 200C and a pressure around 100 MPa. MULTICOMWNENT SYSTEMS For multicomponent systems, it is assumed that the 488 THE CANADI AN J OURNAL OF CHEMICAL ENGINEERING, VOLUME 64, J UNE 1986 TABLE I Reference Henry's Law Constants and Water Interaction Coefficients T: temperature, K fi: fugacity of saturated water, atm H?: reference Henrv's constant, atm Coefficients Binary interaction Comoonent A B C coefficient Data used in regression Methane Ethane Propane n-Butane n-Pentane n-Octane Carbon dioxide Carbon monoxide Nitrogen 10.9554 13.9485 14.6331 13.4248 16.0045 3 I .943 1 I 1.3021 10.7069 10.7090 11.3569 13.8254 14.4872 13.8865 16.2281 28.6725 10.6030 11.1313 11.4793 1.17105 1.66544 1.78068 1.71879 2.13123 4.37707 1 .20696 I .08920 I .I6549 0.4907 0.491 I 0.5469 0.5080 0.5000 0.4500 0.200 0.2750 Culberson & McKetta (195 I ); Olds et al. (1942) Culberson & McKetta (1950); Reamer et al. (1943) Kobayashi & Katz (1953) Reamer et al. (1944, 1952) Gillespie & Wilson ( I 982) Brady et al. (1982) Wiebe & Gaddy (1939); Todheide & Franck (1963); Gillespie & Wilson (1980) Saddington & Krase (1934); Rigby & Prausnitz (1968) Takenouchi & Kennedy (1964) '0.2, T 5 373 K; 0.49852-0.008 ( TI, T >373 K. 80 METHANE - WATER SYSTEM MOLE PERCENT OF METHANE IN AQUEOUS PHASE Figure 3 - Aqueous-phasecomposition of methane- water system. Henry's law constants developed for binary systems still apply. This is equivalent to assuming that the interaction between the solutes is negligible, which is true at infinite dilution. The phase behavior of the methane-n-butane- water system was predicted with HT given in Table 1 and the Peng-Robinson EOS. The predictions match very well the experimental data of McKetta and Katz (1948) (Figure 9). A typical application of the calculation method to a reservoir fluid is given in Appendix B. Modeling of gas solubility in brines OUTLINE In this section the scaled-particle theory (SPT) is used to modify the Henry's law constants developed previously to take into account the presence of salt in the aqueous phase. HENRY'S LAW CONSTANTS FROM SCALED-PARTICLE THEORY Only the necessary equations concerning the SPT of fluids are reported in this section. More details on the subject can be found in Reiss et al. (1960), Lebowitz et al. (1965) and Pierotti (1963, 1967, 1976). The SPT can be used to obtain an expression for Henry's law constant (23) Hivs - ( : ; ) (Goi) ln-- - + RT .................... RT where Gbi is the Gibbs energy (work) required to create a cavity in the solvent of suitable size to accommodate a solute molecule, GGi is the molar Gibbs energy resulting from the interaction of the solute molecule with the surrounding solvent molecules, and where v, is the molar volume of pure solvent. Lebowitz et al. (1965) have derived an expression for Gbi for gas solubility in mixed solvents. When the mixed solvent is brine, this expression takes the form: 0, =; qw c pj(u;)" n =1,2,3 . . . . . . . . . . (25) i THE CANADIAN J OURNAL OF CHEMICAL ENGINEERING, VOLUME 64, J UNE 1986 489 220 200 180 PROPANE- WATER SYSTEM - - Kobayashi 8 Kat z (19531 V 71.1DC A137.8OC 0 104.4C 160 - 0 87.8C 0 154.4C MOLE PERCENT OF PROPANE IN VAPOR PHASE Figure 4 - Vapor-phase composition of propane- water system. 220 PROPANE -WATER SYSTEM 200 180 160 140 a L ; 120 3 m 100 m 80 a 60 Koboyashi 8 Kat z (1953) 0 15.6OC A 137.8OC V 7 1 . VC 0 4 5 4 4 0 ~ 40 0 104 4 O C 20 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.C MOLE PERCENT OF PROPANE IN AQUEOUS PHASE Figure 5 - Aqueous-phase composition of propane - water system. . . . . . . . . . . . . . a; =up/u, k =i , j (26) r), =Nu: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (27) u, and a, are respectively the molecular diameter of solute i and of species j in the solvent, a, is the molecular diameter of pure water; N is the Avogadros number; p, is the molar density of species j in the solvent. The summation in Equation (25) is over all species in the solvent, e.g. for gas solubility in NaCl aqueous solutibn, the summation involves the solutes, pure water, Na ions and C1- ions. Masterton and Lee (1970) showed that the solutes may be excluded from the summation, and their approach is used in this study. (The term solutes in this paper refers to the com- pounds soluble in the aqueous phase which are also present in the vapor and/or liquid phases). Shoor and Gubbins (1969) proposed the following expression for the interaction Gibbs energy Gu,: 5 A PROPANE -WATER 4 d =, 3 - W 3 v) Vapor a - v) phase w 2 - a a. 0 Kaboyar hi 0 Kat z (19531 Cal cul at i ons MOLE PERCENT OF PROPANE I CRI TI CAL POINT Aqueous phase C3 - r i ch l i qui d phase Aqueous phase Vapor phase J -uulJ 40 60 80 100 TEMPERATURE,OC Figure 6 - Thrce-phase composition and temperature of propanc- water system. TABLE 2 Scaled-Particle-Theory Parameters u, nm e l k , K IOa, cm/molecule Water 0.275 85.3 I ,590 Carbon dioxide 0.332 300.0 2.590 Methane 0.413 160.0 2.700 Nitrogen 0.373 95.0 I .730 Na 0.190 147.4 0.210 C1- 0.362 225.5 3.020 NOTE: N =6.02283 X 10; p w =1.84 X lo- (erg-cm) cm; I nm = m. ............................... El , = (29) a; =F(a, * +0, 3/ 2 . . . . . . . . . . . . . . . . . . . . . . . (30) where p, is the dipole moment of water, a, is the polariza- bility of solute i , k is Boltzmanns constant, E, is a parameter which characterizes the strength of the dispersion force, and where F is unity. Schulze and Prausnitz (1981) found that better predictions could be obtained by making F temperature dependent, i.e., . . . . . . (31) F(T) =16280/T2 - 141.75/T +1.2978 with T in K. Schulze and Prausnitz (1981) reported SPT parameters for the calculation of Henrys law constant for nine gases in water. Masterton and Lee (1970) tabulated the parameters for nine ions for the prediction of gas solubility i n aqueous salt solutions. Values of those parameters which are used in the present study are listed in Table 2. The values a, of the ions shown were calculated from the Mavroyannis- Stephen theory as described in Shoor and Gubbins (1969). Parame- ters for other compounds can befound in Hirschfelder et al. (1964) and Weast and Astle (1979). Equations (25) and (28) require the molar volume of water and the ions. For aqueous NaCl solutions, these are estimated as follows p y =(1 - G) f i s / Mu (32) PNa+ =pel- =6 f i r / h f N & (33) ....................... .................... 490 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 64. JUNE 1986 TABLE 3 Solubility of Nitrogen in Aqueous Sodium-Chloride Solution Mole % nitrogen in aqueous phase H20 1 rn NaCl 4 m NaCl T, P C MPa Measured Calculated Measured Calculated Measured Calculated 51.1 102.5 125.0 10.13 20.27 30.40 40.53 50.66 60.80 10.23 20.37 30.60 40.83 50.97 61.20 10.44 20.67 30.90 41.04 51.37 61.61 0.0799 0.1454 0.2017 0.2490 0.2920 0.3350 0.0777 0.1447 0.2005 0.2520 0.2980 0.3370 0.0808 0.1492 0.2047 0.2570 0.3060 0.3510 0.0822 0. I479 0.2027 0.2499 0.2914 0.3282 0.0812 0. I501 0.2106 0.2641 0.31 15 0.3548 0.0903 0.1682 0.2368 0.2975 0.3533 0.4035 0.0593 0.1076 0. I497 0.1860 0.2216 0.2530 0.0603 0.1113 0.1538 0.1920 0.2252 0.2600 0.0632 0. I102 0.1533 0.1883 0.223 I 0.2550 0.0627 0.1130 0.1551 0.1916 0.2239 0.2528 0.0628 0.1162 0.1633 0.2052 0.2426 0.2769 0.0694 0. I298 0. I833 0.2308 0.2749 0.3146 - 0.0500 0.0700 0.0878 0.1034 0.1 179 0.0523 0.073 I 0.0899 0.1047 0.1205 0.0567 0.0740 0.0921 0.1041 0.1227 - - 0.0296 0.0534 0.0736 0.0912 0.1070 0.1214 0.0306 0.0569 0.0803 0.1012 0.1200 0.1373 0.0336 0.0632 0.0896 0.1133 0.1353 0. I552 Average error (%) 6.77 9.47 12.94 Measurements are from OSullivan and Smith (1970). TABLE 4 Solubility of Methane in Aqueous Sodium-Chloride Solution Mole % methane in aqueous phase Hz0 1 m NaCl 4 m NaCl T, P C MPa Measured Calculated Measured Calculated Measured Calculated 51.1 10.13 0.1427 20.27 0.2279 30.40 0.2870 40.53 0.3340 50.66 0.3730 60.80 0.4090 102.5 10.23 0.1355 20.37 0.2205 30.60 0.2870 40.83 0.3330 50.97 0.3850 61.20 0.4190 125.0 10.44 0.1434 20.67 0.2321 30.90 0.2960 41.04 0.3430 51.37 0.3960 61.61 0.4300 Average error (%) 0.1416 0.2253 0.2822 0.3257 0.361 1 0.3907 0.1269 0.2149 0.2816 0.3348 0.3787 0.4165 0.1389 0.2399 0.3180 0.3808 0.4350 0.481 I 4.10 0.1076 0.1695 0.2138 0.2500 0.2790 0.3070 0.1693 0.2219 0.2570 0.2890 0.3200 0.1058 0.1752 0.2223 0.2600 0.2940 0.3250 - 0.1116 0.1774 0.2224 0.2567 0.2847 0.3084 0.1006 0.1705 0.2235 0.2660 0.301 1 0.3315 0.1097 0.1898 0.2519 0.3021 0.3455 0.3825 6.30 - 0.0805 0.0997 0.1 154 0. I303 0.1444 0.0826 0.1079 0.121 I 0.1319 0.1433 0.0825 0.1005 0. I164 0. I322 0.1438 - - 0.0576 0.0914 0.1144 0.1320 0.1464 0. I587 0.0528 0.0895 0. I174 0. I396 0. I580 0.1739 0.0572 0.0990 0.1316 0.1579 0. I805 0.1997 20.033 Measurements are from OSullivan and Smith (1970). where M, and MNaCl are respectively the molecular weight of water and NaCl, fi I is the mass density of the aqueous NaCl solution and where 6 is the weight fraction of NaCl in water. As pointed out by Pierotti (1967), the use of the SPT does Some predictions of gas solubilities in water and/or not require explicit ionization information about the salt brines with the SPT are presented in Schulze and Prausnitz since the latter is included in the solvent density fis. Thus, (1981), Shoor and Gubbins (1969) and Masterton and Lee an accurate value fi, is required. In this study, bs is estimated (1970). However, because of the uncertainties in the from the correlation of Rowe and Chou (1970). SPT parameters (Schulze and Prausnitz, 1981) and because FITTING SCALE-PARTICLE THEORY HENRYS LAW CONSTANTS TO EXPERIMENTAL DATA THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 64, JUNE 1986 49 I CARBON DIOXIDE- WATER SYSTEM 25' 50' 75' 100' 150' 200' 250' C 500 Witbe ( 1942) - CARBON DIOXIDE- WATER SYSTEM - - 50' 1000 450' 200' 250' C o m Wi tbe B Gaddy (19391 0 x 0 A 0 v T6dheidt B Fr anck(1963) A v Taktnouchi 8 Kt nnt dy(l 96 I - Caiculatianr Two phase regi on McKet t a, Jr. 6 Katz ( 1948) .2 Hydrocarbon ri ch liquid phasm Vapor pharm MOLE PERCENTOF WATER IN VAPOR PHASE Figure 7 - Vapor-phase composition of carbon-dioxide- water system. o A o v TSdheide B Franch ( 1963) A v Taktnouchi B Kt nnt dy(1964) t I I I I I 1 1 2 3 4 5 6 MOLE PERCENT OF CARBON DIOXIDE IN AQUEOUS PHASE Figure 8 - Aqueous-phase composition of carbon-dioxde - water system. of several approximations made in the theory (Masterton and Lee, 1970) there are usually disagreements between predicted and measured values. In this paper, to reduce these uncertainties, Henry's law constants from the STP are first fitted to the values calcu- lated from Equation (17) for gas solubility in pure water. The adjustable parameter is the molecular diameter ui of the solute which has a strong influence on the SPT predictions (Schulze and Prausnitz, 1981). Subsequently, the SPT parameters of the ions and the density of the aqueous salt solution are used in Equations (24), (25) and (28) to calculate the Henry's law constants in the presence of salt. For a given pressure and temperature, the determination 40 30 METHANE - n-BUTANE -WATER SYSTEM Two DhaSe n ai o n - Cal cui at l ons CALCULATED CRITICAL LOCUS \ Three phase regi on I 0.1 1 10 MOLE PERCENT OF WATER Figure 9 - Composition of water in vapor and hydrocarbon-rich liquid phase: methane-n-butane- Water system. CARBON DIOXIDE I N AQUEOUS NoCl SOLUTION Tokenouchi 8 Kennedy ( 1965) "' , zo; c 4.28m 1. 09m O m --- l - Cal cul at i ons 4 . 2 8 m 1. 09m Om 1. 09m Om 0 I I I I I I I 1 2 3 4 5 6 7 MOLE PERCENT OF CARBON DIOXIDE IN AQUEOUS PHASE Figure 10 - Solubility of carbon-dioxide and aqueous NaCl solution. of cri for each solute involves the solution of a nonlinear equation corresponding to the equality of STP Henry's law constant with the Henry's law constant obtained from Equation (17). Since no salt is present, the equation is rela- tively simple and is solved using Newton's method with 492 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 64, JUNE 1986 TABLE 5 Compositions of Reservoir and Recombined Fluids at Various Water Cuts Composition" (mol %) Reservoir Reservoir Water cut (%) aqueous-phase oil-phase Component Oil G 67 89 (mol %) (mol %) compositionb compositionb N2 c02 CI c2 C, nC4 nCs n C6 C,--C16 C17-Cxi H20 Molecular weight Bubble point (MPa) c:, at 65.5" C 0.27 0.26 30.58 4.61 6.06 4.71 4.33 3.18 28.26 12.10 5.66 - 132.2 12.040 0.29 0.38 31.88 4.58 5.94 4.60 4.23 3.10 27.63 I I .83 5.53 - 129.7 12.743 0.34 0.69 35.70 4.49 5.61 4.31 3.95 2.90 25.80 11.05 5.17 - 122.3 14.939 0.0015 0.0088 0.1310 0.0050 0.0017 o.oO05 o.oO01 O.oo00 O.oo00 O.oo00 O.oo00 99.8514 0.27 0.26 30.27 4.57 6.00 4.66 4.28 3. I4 27.98 I I .98 5.60 0.99 131.1 I 1.970 'Dry basis. hWet basis. UNSTABLE STABLE W- aqueous phase L = liquid phoso V = vopor phase ST= stabi l i ty ?*st / ' fl ash calculation - =i n equiilbrlum FigureI I - Flow diagram for liquid-vapor-water flash calcu- lations. analytical derivative. The initial guess for ui corresponds to the value reported in the literature (Table 2). Note that ui is treated as an adjustable parameter and therefore the resulting value may not have any physical meaning. Results The measurements of O'Sullivan and Smith (1970) on the solubility of nitrogen and methane in aqueous NaCl solu- tions are matched with the procedure just described. The results are shown in Tables 3 and 4. Both the solubilities of nitrogen and methane decrease as the salt concentration increases (salting out phenomena). The error in the pre- dictions increases with salt concentration and temperature. The solubility of carbon dioxide in aqueous NaCl solu- tions is also calculated and compared to the measurements of Takenouchi and Kennedy (1965). Adequate predictions were obtained as shown in Figure 10. Again, the error increases with salt concentration and temperature. 100 15 Figure 12 - Effect of water cut on producing GOR and bubble- point pressures of recombined fluids. The values of ui for the solute obtained from matching pure-water Henry's law constants are not far from the initial values reported in the literature. For nitrogen, typical values of ui are 0.37321 nm at 10.132 MPa and 51.5"C, and 0.36627 nm at 61.606 MPa and 125C compared to a value of 0.373 nm shown in Table 2. For methane, the fitted ui is 0.34314 nmat 10.132 MPa and 51.5"C, and 0.34703 nmat 61.606 MPa and 125C compared to a reported value of 0.413 nm. For carbon dioxide, the values are 0.33076 nm at 1 MPa and lOO"C, and 0.2733 nmat 20 MPa and 250C compared to a reported value of 0.332 nm. DISCUSSIONS The SPT provides a simple and adequate method for correlating Henry's law constants for gas solubility in brines. Note that in the calculations presented above, only the ui of the solutes are treated as adjustable parameters to match the Henry's law constants predicted by Equation (17). THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 64, JUNE 1986 493 All other STP parameters are given in Table 2. The procedure can be applied to any aqueous salt solution provided the STP parameters for all ions are available, and the mass density of the solvent is known. In the present study, the density of brine is estimated from Rowe and Chous ( 1970) correlation which is developed for pressures up to 34.3 MPa. However, the pressures in the calculations went beyond 125 MPa, and some of the inaccuracies in the predictions may partly be attributed to an inaccurate solvent density value. Again, the ui of the ions could also be adjusted to give a better fit of the experimental data. However, this was not carried out in this study. Conclusions This paper presents a method for predicting the phase behavior of oil, gas and water or brines where the oil and gas are modeled with the Peng-Robinson equation of state and where the gas solubility in the aqueous phase is estimated from Henrys law constants. The method is simple and gives adequate predictions for pressure and temperature conditions encountered in reservoir engineering studies. Correlations for Henrys law constants and Peng- Robinson equation of state interaction coefficients are de- termined for a variety of compounds of interest to the oil industry. For practical purposes, any hydrocarbon heavier than pentane may be assumed insoluble in the aqueous phase. Satisfactory predictions of gas solubility in pure water up to 200C and 100 MPa have been obtained. The scaled-particle theory was used to predict the change of Henrys law constant due to the presence of salt in the aqueous phase. By adjusting the solute molecular diameter, adequate predictions of solubility in aqueous NaCl solution up to a 4 m concentration have been obtained. Although the reference Henrys law constants in Table 1 were determined with the Peng-Robinson equation of state for the vapor or liquid phase, they could be used with any other equation of state. However, the interaction coeffi- cients in Table 1 are only for the Peng-Robinson equation of state. Appendix A Three-phase flash calculations for liquid-vapor- water systems Procedures for three-phase flash calculations where all phases are modeled with an EOS can befound in Heidemann ( 1974). Peng and Robinson (176b), Fussell (1979), Mehraet al. (1982), Michel- sen (1982b). Risnes and Dalen (1984) and Nghiemand Li (1984). The algorithmdescribed in the following is a modification of the method of Nghiemand Li (1984) to handle three-phase liquid- vapor- water (LVW) flash calculations where theaqueous phase is not modeled by an EOS. The procedure does not require good initial guesses of K-values since the latter are obtained fromsta- bility tests using thetangent plane criteripn. Detailed discussions on this subject are presented in Nghiemand Li (1984). Michelsen (1982a) and Baker et al. (1982). Figure 11 shows a flow diagramof a stagewise procedure for LVW flash calculations. Notethat thesymbols L and V in the flow chart could bepermuted since the first two-phase flash calculation could either bea LW calculation or a VW calculation. When the cubic EOS yields morethan oneroot in the compressibility factor, the one corresponding to the lowest Gibbs energy is selected (Evelein et al., 1976). The stability test of a phase with composition y consists of solving d ; ( u ) =In ui +In & ( 6) - In y , - In $, (y) =0 .... (A.1) i, =u,/$ UL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A.2) The +, in Equation (A. I ) are the EOS fugacity coefficients for u, . The phase with composition y is unstable if, at convergence, i U , >1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A.3) In the stability test of a VW or LW system, it has been found necessary to prevent convergence to a set of u , corresponding to an aqueous phase predicted by an EOS, since the latter has been modeled by Henrys law. For this purpose the following equation is solved for u instead of Equation (A. I ) i =I , . . . , n , . . . . (A.4) where y,,,. is thecomposition of water in theaqueous phase from a two-phase VW or LW calculation. The algorithmin Figure 1 1 requires the initial guesses K:::,. The following equations provide a reasonable estimate for K:::? ! , = I I = I d : ( u ) =d i ( u ) / l u , - yw,,.I =0 K::, =p/H, i f w . . . . . (A.4) ~j : ! =K , * ~/ H, i z w . . . . . . . . . . . . . . . . . . . . . . . ( AS) if a VW systemis expected, or if a LW systemis expected. K,* is Wilsons K-value given by K,* =( p<. / / p) exp [5.42(1 - T<, / T) ( I +w,)] . . . . . . . (A.6) Even if VW Kj is selected instead of LW K rl for LW flash calculations or vice-versa, convergence to thecorrect solution will still beachieved. However, the procedure may take longer to reach the solution. For the water component, a value for Kc,:, is found such that i Kj:!z, >1 ................................ (A.7a) , = I and $ z,/K::! >1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A.7b) I = I This could easily be achieved by setting Kr,:, equal to a large number (say >IOOO). Equation (A.7) will guarantee a split of the feed into two phases when the material balance for two-phase flash calculations is solved (Nghiemet at., 1983; Rachford and Rice. 1952). Appendix B Effect of water on the producing gas-oil ratio An example illustrating the effect of gas solubility in the aqueous phase on the producing gas-oil ratio (GOR) froma producing well is given in the following. Water is usually produced together with oil, and thegas obtaincd at surface conditions comes fromboth the oil and the aqueous phase. Consider a typical reservoir oil (called Oil G) and com- position shown in the second column of Table 5 . The oil is in equilibrium with the aqueous phase at reservoir conditions of 14MPa and 65.6C. It is assumed that the solubility of hydro- carbons heavier than nCs is negligible. The compositions of the recombined reservoir fluids obtained by blending the oil and gas at surface conditions (0.101 MPaand 15.5C) at 67% and 89% water cut (water cut =flow rate of water/(flow rate of water +oil)) are also shown in Table 5. The molecular weights and bubble-point pressures for the various fluids, as well as the aqueous phase composition and the water composition in the oil phase are also reported. Figure 12 illustrates the change in the producing GOR with the water cut for thesame system. The computed bubble-point pressure of the recombined fluid is also depicted. The results show 494 THE CANADI AN J OURNAL OF CHEMI CAL ENGI NEERI NG, VOLUME 64, J UNE 1986 that the effect of gas solubility in the aqueous phase on the producing GOR cannot be ignored at high water cut. Acknowledgement This work was supported by thc Department of Energy, Mines and Resources of Canada under Contract 01 SG-23294-4-07s I . Nomenclature cohesive energy density defined in Equation (20) defined in Equation (A. I ) defined in Equation (A.4) interaction coefficient fugacity of component i in phase m temperaturc-dependent factor defined in Equation (3 1 ) mole fraction of phase m molar Gibbs energy of interaction between the solute i and the solvent molar Gibbs energy for cavity creation molar enthalpy of component i in vapor phase molar enthalpy of component i at infinite dilution molar enthalpy of water at the given temperature and zero pressure molar enthalpy of water at saturation conditions Henrys law constant of component i reference Henrys law constant of component i Henrys law constant of component at a given pressurep; Boltzmanns constant Wilsons K-value defined in Equation (A.6) K-value of component i in phase m molality molecular weight number of components Avogadro number (= 6.02283 X lo) pressure pressure where HY is evaluated universal gas constant temperature temperature in degree F variable in stability-test equation composition corresponding to u, molar volume molar volume of component i at infinite dilution mole fraction of component i in phase m feed (global) composition of component i polarization of solute i parameter which characterizes the dispersion strength of component i geometric mean of E, and E, defined in Equation (27) defined in Equation (25) dipole moment of water molar density mass density molecular diameter of compound i dimensionless molecular diameter of compound i defined in Equation (26) defined in Equation (30) fugacity coefficient of component i in phase m acentric factor of component i weight fraction of salt in water Superscripts s =saturated (0) =initial guess Subscripts i, j. k =indices for component e =liquid phase m, q =indices for phase s =solvent v =vapor phase w =aqueous phase w =water component References Baker, L. E., A. C. Pierce and K. D. Luks, Gibbs Energy Anal- ysis of Phase Equilibra. Soc. Petrol. Eng. J . 22, 731-742 (1982). Brady, C. J ., J . R. Cunningham and G. M. Wilson, Water- Hydrocarbon Liquid-Liquid- Vapor EquilibriumMeasurements to 530F. Research Report RR-62, Gas Processors Association, Tulsa, OK, September, 1982. Canjar, L. N. and F. S. Manning, Thermodynamic Properties and Reduced Correlations for Gases, Gulf Publishing Co., Houston, Texas, 1967. Culberson, 0. L. and J . J . McKetta, J r., Phase Equilibria in Hydrocarbon - Water Systems 11-The Solubility of Ethane in Water at Pressures to 10 OOO psi, Petrol. Trans. AlME 189, 319-322 (1950). Culberson, 0. L. and J . J . McKetta, J r., Phase Equilibria in Hydrocarbon - Water Systems Ill-The Solubility of Methane in Water at Pressures to 10 OOO psia, Petrol. Trans. AIME 192, 233-266 (1951). Cysewski, G. R. and J . M. Prausnitz, Estimation of Gas Solubilites in Polar and Nonpolar Solvents, Ind. Eng. Chem. Fundam. 15, 304-309 (1976). Erbar, J . H., A. K. J agota, S. Muthswamy and M. Moshfeghian, Predicting Synthetic Gas and Natural Gas Thermodynamic Properties Using a Modified Soave-Redlich-Kwong Equation of State, Research Report RR-42, Gas Processors Association, Tulsa, OK, August, 1980. Evelein, K. A., R. G. Moore and R. A. Heidemann, Correlation of the Phase Behavior in the Systems Hydrogen Sulfide - Water and Carbon Dioxide - Water, lnd. Eng. Chem., Process Des. Dev. 15, 423-428 (1976). Fussell, L. T., A Technique for Calculating Multiphase Equilibria, Soc. Petrol. Eng. J . 19 203-210 (1979). Gillespie, P. C. and G. M. Wilson, Vapor-Liquid Equilibrium Data on Water-Substitute Gas Components: N2-H20, H2-H20, CO-H20, H2-CO-H20, and H2S-H20. Re- search Report RR-41, Gas Processors Association, Tulsa, OK, April, 1980. Gillespie, P. C. and G. M. Wilson, Vapour - Liquid and Liquid - Liquid Equilibria: H20-CH4, H20-COz, H20-H2S, H20-nCs, H20-CH4-nCs, Research Report RR-48, Gas Processors Association, Tulsa, OK, April, 1982. Heidemann, R. A., Three-Phase Equilibria Using Equations of State, AlChE J . 20, 847-855 (1974). Heidemann, R. A. and J . M. Prausnitz, Equilibrium Data for Wet-Air Oxidation, Water Content and Thermodynamic Proper- ties of Saturated Combustion Gases, Ind. Eng. Chem., Process Des. Dev. 16, 375-381 (1977). Hirschfelder, J . O., C. F. Curtiss and R. B. Bird, Molecular Theory of Gases and Liquids, J ohn Wiley & Sons, N.Y ., 1964. J onah, D. A., A New Correlation for Henrys Constant for the Solubility of Gases and Liquids, Fluid Phase Equilibria 15, 173-187 (1983). Kobayashi, R. and D. L. Katz, Vapor-Liquid Equilibra for Binary Hydrocarbon-Water Systems, Ind. & Eng. Chem. 45, Lebowitz, J . L., E. Helfand and E. Praestgaard, Scaled Particle Theory of Fluid Mixtures, J . Chem. Phys. 43, 774-779 (1965). Luks, K. D., P. D. Fitzgibbon and J . T. Banchero, Correlations of the Equilibrium Moisture Content of Air and of Mixtures of Oxygen and Nitrogen for Temperatures in the Range of 230 to 600 K at Pressures up to I 00 Atm., Ind. Eng. Chem. Process 440-451 (1953). THE CANADIAN J OURNAL OF CHEMI CAL ENGINEERING, VOLUME 64, J UNE 1986 495 Des. Dev. 15, 326-332 (1976). Lyckman, E. W., C. A. Eckert and J . M. Prausnitz, Generalized Reference Fugacities for Phase Equilibrium Thermodynamics. Chem. Eng. Sci. 20, 685-691 (1965). Masterton, W. L. and T. P. Lee, Salting Coefficients from Scale Particle Theory, J . Phys. Chem. 74, 1776- 1782 (1970). McKetta, J . J . J r. and D. L. Katz, Methan-n-Butane-Water Systems in Two- and Three-Phase Regions, Ind. & Eng. Chem. 40, 853-863 (1948). Mehra, R. K., R. A. Heidemann and K. Aziz, Computation of Multiphase Equilibrium for Compositional Simulation, SOC. Petrol. Eng. J . 22, 61-68 (1982). Mehra, R. K., R. A. Heidemann and K. Aziz, An Accelerated Successive Substitution Algorithm, Can. J . Chem. Eng. 61, 590-596 (1984). Michelsen, M. L., The Isothermal Flash Problem. I . Stability, Fluid Phase Equilibria 9, I - 19 (1982a). Michelsen, M. L., The Isothermal Flash Problem. 11. Phase-Split Calculation, Fluid Phase Equilibria 9, 21 -40 (1982b). Nghiem, L. X. and R. A. Heidemann, General Acceleration Procedure for Multiphase Flash Calculation with Application to Oil-Gas- Water Systems, Proceedings of the Second European Symposium on Enhanced Oil Recovery, November 8- loth, 1982, Paris, Editions Technip, Paris, 303-316. Nghiem, L. X. , K. Aziz and Y . K. Li, A Robust Iterative Method for Flash Calculations Using the Soave-Redlich-Kwong or the Peng-Robinson Equation of State, SOC. Pet. Eng. J . 23, 521 -530 (1983). Nghiem, L. X. and Y. K. Li, Computation of Multiphase Equi- librium Phenomena with an Equation of State, Fluid Phase Equilibria 17, 77-95 (1984). OConnell, J . P., Thermodynamics of Gas Solubility, Proc. Second Int. Conf. on Phase Equilibria and Fluid Properties in the Chemical Industry, Berlin (West), March 17-21. 1980, Dechema, Frankfurt/Main, West Germany, 445-455. Olds, R. H., B. H. Sage and W. N. Lacey, Phase Equilibria in Hydrocarbon Systems-Composition of the Dew-Point Gas of the Methane-Water Systems, lnd. Eng. Chem. 34, OSullivan, T. D. and N. 0. Smith, The Solubility and Partial Molar Volume of Nitrogen and Methane in Water and in Aque- ous Sodium Chloride from 50 to 125C and 100 to 600 Atm., J . Phys. Chem. 74, 1460-1466 (1970). Peng, D. Y. and D. B. Robinson, A New Two-Constant Equation of State, Ind. Eng. Chem. Fundam. 15, 59-64 (1976a). Peng, D. Y. and D. B. Robinson, Two- and Three-Phase Equi- librium Calculations for Systems Containing Water, Can. J . Chem. Eng. 54, 595-599 (1976b). Peng, D. Y . and D. B. Robinson, Two- and Three-Phase Equi- librium Calculations for Coal Gasification and Related Processes, in Thermodynamics of Aqueous Systems with Indus- trial Applications, Newman, S. A., Ed., Amer. Chem. SOC. Symp. Ser. 133, 393-414 (1980). Pierotti, R. A., The Solubility of Gases in Liquids, J . Phys. Chem. 67, 1840- I845 (1963). Pierotti, R. A., On the Scaled-Particle Theory of Dilute Aqueous Solutions, J . Phys. Chem. 71, 2366-2367 (1967). Pierotti, R. A., A Scaled-Particle Theory of Aqueous and Nonaqueous Solutions, Chem. Rev. 76, 717-726 (1976). Prausnitz, J . M., Molecular Thermodynamics of Fluid- Phase Equilibria, Prentice-Hall, Inc., Englewood Cliffs, N.J ., 1969, 170. Rachford, H. H., J r. and J . D. Rice, Procedure for Use of 1223-1227 (1942). Electronic Digital Computers in Calculating Flash Vaporization Hydrocarbon Equilibrium, Trans. SOC. Petrol. Eng. 195, 327-328 (1952). Reamer, H. H., R. H. Olds, B. H. Sage and W. N. Lacey, Composition of Dew-Point Gas in Ethane- Water System, Ind. Eng. Chem. 35, 790-793 (1943). Reamer, H. H., R. H. Olds, B. H. Sage and W. N. Lacey, Composition of Coexisting Phases of n-Butane- Water in the Three-Phase Region Ind. Eng. Chem. 36, 381-383 (1944). Reamer, H. H., B. H. Sage and W. N. Lacey, Phase Equilibria in Hydrocarbon Systems: n-Butane- Water System in the Two- Phase Region, Ind. & Eng. Chem. 44, 609-614 (1952). Reid, R. C., J . M. Prausnitz and T. K. Sherwood, The Properties of Gases and Liquids, 3rd Ed., McGraw-Hill Book Co., New York, N.Y., 1977, pp. 133-115, 188. Reiss, H., H. L. Frisch, E. Helfand and J . L. Lebowitz, Aspects of the Statical Thermodynamics of Real Fluids, J . Chem. Phys. Rigby, M. and J . M. Prausnitz, Solubility of Water in Com- pressed Nitrogen, Argon and Methane, J . Phys. Chem. 72, I (1968). Risnes, R. and V. Dalen, Equilibrium Calculations for Coexisting Liquid Phases, SOC. Petrol. Eng. J . 24, 87-96 (1984). Rowe, A. M. and J . C. S. Chou, Pressure-Volume-Tern- perature-Concentration Relation of Aqueous NaCl Solutions, J . Chem. Eng. Data 15, 61-66 (1970). Saddington, A. W. and N. W. Krase, Vapor-Liquid Equilibria in the System Nitrogen-Water, J . Am. Chem. SOC. 56, 353-361 (1934). Water at High Temperatures, Ind. Eng. Chem. Fundam. 20, 175-177 (1981). Selleck, F. T., L. T. Carmichael and B. H. Sage, Phase Behavior in the Hydrogen Sulfide - Water System, Ind. & Eng. Chem. Shoor, S. K. and K. E. Gubbins, Solubility of Nonpolar Gases in Concentrated Electrolyte Solutions, J . Phys. Chem. 73, Smith, J . M. and H. C. Van Ness, introduction to Chemical Thermodynamics, Third Edition, McGraw-Hill, New York, N.Y., 1975, pp. 241-244, 614. Takenouchi, S. and G. C. Kennedy, The Binary System H20-C02 at High Temperatures and Pressures, Am. J our. Sci. 262, 1055- I074 (1964). Takenouchi, S. and G. C. Kennedy, The Solubility of Carbon Dioxide in NaCl Solutions at High Temperatures and Pressures, Am. J our. Sci. 263, 445-454 (1965). Todheide, K. and E. U. Franck, Das Zweiphasengebiet und die Kritische Kurve imSystem Kohlendioxid-Wasser bis au Drucken von 3500 bar, Z. Physik Chem. (Frankfurt) 37, 387-401 (1963). Weast, R. C. and M. J . Astle, Eds., CRC Handbook of Chemistry and Physics, 60th Ed.. CRC Press, 1979. Wiebe, R., The Binary System Carbon Dioxide - Water Under Pressure, Chem. Rev. 29, 475-481 (1942). Wiebe, R. and V. L. Gaddy, The Solubility in Water of Carbon Dioxide at 50, 75 and IOOC, at Pressures to 700 Atmospheres, J . Am. Chem. SOC. 61, 315-318 (1939). 32, 119-124 (1960). Schulze, G. and J . M. Prausnitz, Solub 44, 2219-2226 (1952). 498 - 505 ( 1969). Manuscript received J une 7, 1985; revised manuscript received December 5, 1985; accepted for publication J anuary 3, 1986. 496 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 64, JUNE 1986