sign in sign up
<<
Home , Ethane, Pentane

Phase Behavior in the --n- System

By G. W. BILLMAN,· B. H. SAGE,· MEMBER AIME AND W. N. LACEY·

(New York Meeting, March 1947)

ABSTRACT However, these results do not establish THE composition of the coexisting phases in the influence of the and amount of the methane-ethane-n-pentane system was the other substances making up a multi­ determined at 100°F. This was accomplished by component system upon the -liquid withdrawing samples of the coexisting phases equilibrium constant of anyone com­ under isobaric-isothermal conditions. The rela­ tive amount of each of the components present ponent. A preliminary correlation on the in the liquid and gas phases was determined by basis of vaiues for binary systems was conventional analytical fractionation methods. made2 utilizing such experimental informa­ The phase behavior of this system approxi­ tion as was available. However, there is mates that estimated from generalized corre­ need for much additional information lations which take into account the influence regarding behavior in systems containing of the nature and amount of each of the com­ more than two components. ponents present upon the equilibrium. It was found that for particular pressures and tem­ Study of the phase behavior of ternary Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021 peratures the equilibrium constant of each of systems offers a feasible attack upon the the components was influenced significantly by determination of the influence of com­ the composition of the system and this effect position as well as of pressure and tem­ was more pronounced at the lower pressures perature upon the several equilibrium for methane than for the components of greater constants concerned. An investigation of molecular weight. However, at the higher pres­ the phase behavior of the methane­ sures approaching the maximum two-phase -n-pentane system at IOOo, 160°, pressure for the system at this temperature the and 220°F was reported.3,. This work composition markedly influences the equilib­ rium constants of all the components at a served to illustrate the marked influence particular pressure and temperature. The re­ of concentration of the other components sults of this study are presented in graphical upon the equilibrium constants. Variations and tabular form. of as much as 50 pct in the equilibrium constants for methane for fixed tempera­ INTRODUCTION ture and pressure were observed in regions The phase behavior of many of the remote from the critical state. Likewise binary systems containing paraffin hydro­ there was a marked variation in the components from methane through equilibrium constants of propane and n-pentane has been investigated during n-pentane as a result of changes in the recent years. In addition the compositions composition. These results indicated the of the coexisting phases in mixtures of crude desirability of investigating additional and have been studied. I ternary systems. The present paper deals Manuscript received at the office of the with a study of the methane-ethane-n- pen­ Institute Feb. 10, 1947. Issued as TP 2232 in TECHNOLOGY, July 1947. tane system at IOooF, which is similar in • California Institute of Technology, Pasa­ many respects to the work upon the meth­ dena, California. 1 References are at the end of the paper. ane-propane-n-pentane system cited above. 13 14 PHASE BEHAVIOR IN THE METHANE-ETHANE-N-PENTANE SYSTEM

METHOD dioxide. The equipment was essentially The compositions of the phases coexisting the same as that used in earlier studies.a.'.8 under isobaric-isothermal conditions were The fractionating columns employed established by withdrawing samples of the in the determination of the composition of gas and liquid phases which had been gas and liquid phase samples had internal brought to thermodynamic equilibrium diameters of 0.12 and 0.18 in. and lengths by prolonged mechanical agitation. The of 4 and 5 ft, respectively. The analyses sample was confined in a steel vessel over were carried out in conventional fashion mercury, the pressure being controlled by except that in the transition region the addition or withdrawal of the latter between one component and another the fluid. After equilibrium had been reached, was repeated in order to check isobaric-isothermal conditions were main­ the completeness of separation of the com­ tained during the withdrawal of the ponent being withdrawn. Comparison of samples by controlled addition of mercury duplicate samples indicates that the over­ to the equilibrium vessel. In general the all analytical uncertainty in the mol fraction of a component was approximately withdrawals were accomplished with devia­ Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021 tions from isobaric conditions of less than 0.002, while a precision of 0.001 was usually 3 psi. Mechanical agitation was dis­ easily obtainable. continued during the withdrawal process At low pressures it is possible that very and it is believed that the divergences from small amounts of liquid phase on the walls the original equilibrium state as a result of the equilibrium chamber near the sample of slight changes of pressure did not cause outlet were carried out when a sample of significant error In the experimental the gas phase was withdrawn. At higher results. pressures this difficulty was less probable. The temperature of the equilibrium cell After considering these somewhat intangi­ was kept at desired values by immersing ble quantities in addition to the uncer­ it in an agitated oil bath, thermostatically tainties in pressure, temperature, and controlled by use of suitable electronic composition it appears that, when a com­ circuits to give variations of less than ponent was present only in small quantity, 0.3°F with respect both to time and the uncertainty in mol fraction is probably location within the bath. Temperatures less than 0.005. were related to the International Platinum Scale by a mercury-in-glass thermometer MATERIALS which had been compared with the tem­ The methane used in this investigation perature indications of a standardized, was obtained from the Buttonwillow field -free platinum resistance thermom­ in California. As received, the sample eter. It is believed that the temperature was saturate, with and contained of the equilibrium mixture was known approximately 0.003 mol fraction carbon within 0.2°F. dioxide and 0.0004 mol fraction heavi:r The pressure during the approach to . analyses have equilibrium and during the withdrawal of indicated that nitrogen and other non­ the samples was measured by means of a combustible materials are present in this commercial type of piston-and-cylinder gas in negligible amounts. Before use, the pressure balance with an estimated uncer­ methane was dried and purified by passing tainty of 2 psi or less. This instrument it at a pressure of 300 psi or more through was calibrated against a precision pressure layers of calcium chloride, balanceS which in turn had been calibrated hydroxide, activated charcoal, and ascarite. against the known of carbon The partially purified methane was then G. W. BILLMAN, B. H. SAGE AND W. N. LACEY 15

passed through a copper coil immersed of the coexisting phases for the experi~ in a mixture of · and solid carbon mentally studied mixtures have been dioxide. connected by dotted lines. For the pur~ The ethane was procured from the pose of systematic portrayal of the be~ Carbide and Carbon Chemicals Corpora­ havior of the system, solid combining tion and when received it contained sub­ lines are shown for even values of a stantial amounts both of more and less parameter C which is defined by Eq 1.3 volatile components. This crude ethane C _ X 2 was purified by repeated low temperature [Il - (X2 + X 5) fractionation in which the first and last tenths of the overhead product were In this equation, X 2 and Xs represent the discarded. Analytical measurements upon mol fractions of ethane and n-pentane, the purified material indicated that it respectively, in the mixture. did not contain more than 0.002 mol TABLE I-Experimentally Determined Com­ fraction of material other than ethane. positions of Coexisting Phases The n-pentane was obtained from the Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021 Phillips Petroleum Company whose special Pres· I Gas Phase. I Liquid Phase. Meth-IEthan n-Pen- Meth-IEth I n-Pen- analysis upon a similar sample indicated s~~i' ane e I tane ane ane tane the material to contain approximately 0.005 100°F mol fraction and less than 0.001 mol fraction of other hydrocarbon 500 0.904" 0.0377 0.0583 0.154 0. 0284 0.818 0.652 0.297 0.0519 0.115 0.223 0.662 impurities. Care was exercised to a void 0.517 0·431 0.0511 0.0947 0·327 0·578 0.275 0.681 0.0440 0.0550 0.514 0·431 contamination of the sample with air and a 0.000 0.965 0.0349 0.0000 0.736 0.264 further precaution was taken by sub­ 1,000 0.762 0.188 0.0499 0.263 0.211 0.5.6 mitting the n-pentane to prolonged boiling 0.674 0.282 0.0454 0.246 0.314 0.440 0.596 0.360 0.0443 0.237 0.396 0.368 after addition to the equilibrium apparatus. 0.548 0·405 0.0468 0.224 0·444 0.331 0·483 0·473 0.0438 0.214 0.515 0.271 Addition of the material to the equilibrium 0·394 0·562 0.0441 0.198 0.60. 0.200 chamber was accomplished in substantially 0.196b 0·758b 0.0450b 0.196· 0.759· 0.0451b 1,500 0.814 0.125 0.0608 0.420 0.152 0.428 the same way as was described for the 0.689 0.244 0.0670 0.412 0.291 0.297 methane-propane-n-pentane sYstem. 3,4,6 0.588 0·334 0.0779 0.413 0.378 0.209 0.461• 0·443· 0.0953· 0.462• 0·443· 0.0948· 2,000 EXPERIMENTAL RESULTS 0.899 0.0173 0.0833 0.580 0.021 7 0·399 0.801 0.0981 0.101 0·587 0.116 0.297 The equilibrium states were at pressures 0.763 0.121 0.116 0·599 0.140 0.261 of 500, 1000, 1500, and 2000 psi. o The • Compositions are expressed as mol fraction. compositions of the coexisting phases • Single phase present. are reported in terms of mol fraction in As an aid in visualizing the interrelation Table I. At two states the system was of pressure and composition with the homogeneous and so the compositions of phase behavior of the system an equilateral the two samples taken from the upper projection of a pressure-composition figure and lower part of the equilibrium equip­ of the methane-ethane-n-pentane system ment are the same within the uncertainty at 100°F is presented in Fig 5. Values for of measurement. the vapor pressure of pentane were taken The compositions of the coexisting from published data,3.s as was the informa­ phases for each of the experimentally tion for the methane-n-pentane system.s studied pressures are shown in Figs I to 4, These measurements were carried out inclusive. In each figure the compositions primarily by the determination of the volumetric behavior of individual mix­ • Throughout this paper pressure is ex­ pressed in pounds per square inch absolute. tures of methane and n-pentane. Values for I6 PHASE BEHAVIOR IN THE METHANE-ETHANE-N-PENTANE SYSTEM the ethane-n-pentane system were not The behavior of the system at 500 psi available and those given were determined is shown by the boundary curve BMOKE by extrapolating the equilibrium constants and it is apparent that two phases are determined for the ternary system. to obtainable within limited ranges of com- Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021

9" C?a. FIG I-MoLAL COMPOSITION DIAGRAM AT 500 PSI AND Ioo°F.

FIG 2-MoLAL COMPOSITION DiAGRAM AT 1000 PSI AND 100°F. compositions corresponding to varying position in both the methane-pentane mixtures of ethane and n-pentane. The and the ethane-pentane systems, but only data are not believed to be more than a single gaseous phase exists at 100°F qualitatively indicative of the behavior for the methane-ethane system at any of the ethane-n-pentane system. pressure. Combining lines for several G. W. BILLMAN, B. H. SAGE AND W. N. LACEY values of the parameter C are shown in The locus of critical states from N to T this diagram, corresponding to the behavior has been indicated in the figure and the shown in Fig I. Similarly, the behavior maximum critical pressure for this system of this system at 1500 psi is shown by the at 100°F appears to occur in the binary Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021

n-PENTANE 9., ~ 9,.. ~ FIG 3-MoLAL COMPOSITION DIAGRAM AT 1500 PSI AND 100°F.

n-PENTANE 9., 9«1' 9,.. ~ ETHANE FIG 4-MoLAL COMPOSITION DIAGRAM AT 2000 PSI AND 100°F. boundary CHID for which combining methane-n-pentane system. In other words, lines are shown for several values of the there is no maximum in the critical pressure parameter C. Corresponding curves are locus for ternary mixtures lying between shown for 1000 and 2000 psi. the critical state of the methane-n-pentane G. W. BILLMAN, B. H. SAGE AND W. N. LACEY 19 system and that of the ethane-n-pentane be considered to be a function of pressure, system. temperature, and the composition of the By way of illustration, the locus of system. From the data recorded in Table I coexisting phases corresponding to a value the equilibrium constants were computed

~ 2500 d en a: 2000 w IL cO ...J 1500 Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021 ~.... IL

1000

500 1000 1500 2000 PRESSURE LB. PER SQ. IN. FIG 6-EQUILlBRWM CONSTANTS FOR METHANE IN METHANE-ETHANE-N-PENTANE SYSTEM AT 100°F.

of the parameter C of 0.2 has been indi­ and, after graphical smoothing with cated. This curve FGHIJKZ is generated respect to the parameter C, the results by the intersection of combining lines with were plotted in relation to the equilibrium the corresponding bubble-point and dew­ pressure as shown in Figs 6, 7, and 8 point surfaces of Fig 5. for methane, ethane, and n-pentane, respectively. GAS-LIQUID EQUILIBRIUM CONSTANTS In the case of Fig 6 the product of pres­ The ratio of the mol fraction of a com­ sure and the equilibrium constant has been ponent in the gas phase to its mol fraction shown as the dependent variable in order in the coexisting liquid phase has been to increase the graphical precision with called an equilibrium constant. Such which the data may be presented. In each ratios are actually a function of the state of these figures, curves end at low pressure of a heterogeneous system and vary· with at values for the ethane-n-pentane system, the pressure, temperature, and the nature and at higher pressure at the critical states and amount of each of the components for the ternary system. present. Under certain conditions a number It is often of interest to establish the of the form substantially limiting value of the equilibrium constant ideal solutions and under these con­ of one of the more volatile components ditions the equilibrium constant is as its concentration approaches zero. primarily a function of pressure and tem­ For example, if it is desired to determine perature. However, in the present system the limiting value of the equilibrium con­ the deviation from such simplification is stant of methane in the methane-ethane-n­ large and the equilibrium constant must pentane system as the pressure is decreased G. W. BILLMAN, B. H. SAGE AND W. N. LACEY system and that of the ethane-n-pentane be considered to be a function of pressure, system. temperature, and the composition of the By way of illustration, the locus of system. From the data recorded in Table I coexisting phases corresponding to a value the equilibrium constants were computed

~ 2500 d II) a: w 2000 Do cxi ...J 1500 Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021 If:.- Do

1000

500 1000 1500 2000 PRESSURE LB. PER SQ. IN. FIG 6-EQUILlBRIUM CONSTANTS FOR METHANE IN METHANE-ETHANE-N-PENTANE SYSTEM AT 100°F. of the parameter C of 0.2 has been indi­ and, after graphical smoothing with cated. This curve FGHIJKZ is generated respect to the parameter C, the results by the intersection of combining lines with were plotted in relation to the equilibrium the corresponding bubble-point and dew­ pressure as shown in Figs 6, 7, and 8 point surfaces of Fig 5. for methane, ethane, and n-pentane, respectively. GAS-LIQUID EQUILIBRIUM CONSTANTS In the case of Fig 6 the product of pres­ The ratio of the mol fraction of a com­ sure and the equilibrium constant has been ponent in the gas phase to its mol fraction shown as the dependent variable in order in the coexisting liquid phase has been to increase the graphical precision with called an equilibrium constant. Such which the data may be presented. In each ratios are actually a function of the state of these figures, curves end at low pressure of a heterogeneous system and vary. with at values for the ethane-n-pentane system, the pressure, temperature, and the nature and at higher pressure at the critical states and amount of each of the components for the ternary system. present. Under certain conditions a number It is often of interest to establish the of the hydrocarbons form substantially limiting value of the equilibrium constant ideal solutions and under these con­ of one of the more volatile components ditions the equilibrium constant is as its concentration approaches zero. primarily a function of pressure and tem­ For example, if it is desired to determine perature. However, in the present system the limiting value of the equilibrium con­ the deviation from such simplification is stant of methane in the methane-ethane-n­ large and the equilibrium constant must pentane system as the pressure is decreased 20 PHA.SE BEHAVIOR IN THE METHANE-ETHANE-N-PENTANE SYSTEM

1.30

1.25 \

1.20 \

1.15 \

1.10 \\ ~ Z ~ ~ ~ 1.05 CIlz 0 ~~ 0 1.00 cql ~ °1 ~: ~:~~ :J I 1 095 ~ 1 ~ m Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021 :i i "5 I 0 , ) w 0.90 ~ ~~ 1/ J 0.85 J ~ 1/ t'--....-/ 0.80 ~ --

0.75 500 1000 1500 2000 2500 3000

PRESSURE LB. PER SQ •. IN. FIG 7-EQUILIBRIUM CONSTANTS FOR ETHANE IN METHANE-ETHANE-N-PENTANE SYSTEM AT 100°F

~ Z ~ CIl oZ o ~ :J m ":i o5 w

0.06~---+-----r----~----+---~r----1

0.04'---""5~O"O"O-""1""'0~0""0-""I$""'0"""0'--""2~0,-L:0""'0~""2-='50'="0::------'

PRESSURE LB. PER SQ. IN. FIG 8-EQUILIBRIUM CONSTANTS FOR N-PENTANE IN METHANE-ETHANE-N-PENTANE SYSTEM AT 100°F. G. W. BILLMAN, B. H. SAGE AND W. N. LACEY 21

to the two-phase pressure of the ethane-no ment with values obtained by extrapolation pentane system this may be accomplished of the data to the two-phase pressure of the as will be described. ethane-n-pentane system.

TABLE 2-Equilibrium Constants for Components of Methane-ethane-n-pentane System

Equilibrium Constants Composition, Gas Phase Composition, Liquid Phase Pres- I sure, C Psia Methane Ethane n-Pentane Methane Ethane n-Pentane Methane Ethane n-Pentane ------500 0.0 5.87 1.33 0.0718 0.940 0.000 0.0603 0.160 0.0000 0.840 0.' 5·74 1.33 0.078. 0·71' 0"~3 0.0547 0.1'4 0.1752 0·701 0·4 5·40 1.32 0.0895 0.469 0·482 0.0490 0.0868 0.365 0.548 0.6 4·79 1.32 0.108 0.200 0·759 0.0414 0.0417 0·575 0·383

1,000 0.0 3.08 0.882 0.0765 0·9471 0.0000 0.0530 0.308 0.0000 0.693 0.2 2.98 0.887 0.0858 0.8.25 0.128 0.0497 0.276 0.145 0·579 0·4 2·77 0.897 0.102 0. 6834 0.270 0.046 1 0.247 0.301 0·452 0.6 '·39 0·914 0.147 0.5265 0·4·8 0.0459 0.'20 68 0·312 0.8 1.85 0·94' 0.250 0·347' 0.612 0.0406 0.188 o·aO. 50 0.163

1,500 0.0 2.14 0.81I 0.105 0·9412 0.0000 0.0588 0·440 0.0000 0·560

0.2 2.00 0.816 0.130 0.8460 0·0942 0.0601 0.423 0·1I5 0.46 2 Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021 0·4 1.79 0.829 0.181 0.7428 0. I9 0.0636 0.415 0.234 0·351 0.6 1.50 0.86g 0.315 0.6204 0.30a 0.0739 0·414 0.352 0.235 2,000 0.0 1. 59 0·79' 0.189 0·9205 0.0000 0.0796 0·579 0.0000 0·421 0.2 1.44 0.825 0.278 0.8381 0.0690 0.0930 0.582 0.0836 0·334 0·4 1.21 0.878 0·540 0·7347 0.1379 0.1272 0.607 0.157 0.236

The equilibrium constant for methane Equilibrium constants for the com­ is related to the equilibrium constants ponents of the methane-ethane-n-pentane of the other components and the com­ system at even values of the parameter C position of the system in the following have been recorded in Table 2. In addition way:9 the corresponding mol fractions of the Y I I - (Y2 + Y 5) components in the coexisting liquid and Kl = Xl = I - (X2 + Xs) gas phases have been indicated. These I - (K~2 + KsXs) data were obtained from the interpolated I - (X 2 + Xs) values of the equilibrium constant by in which Y 1, Y 2, Y 5 represent the mol methods that are outlined elsewhere. 9 fractions of methane, ethane and n-pentane, The values recorded agree well with those respectively, the gas phase and Xl, X 2, X 5 obtained directly from Figs I, 2, and 3. represent the mol fractions of the same As was indicate'd above, the analytical components in the liquid phase. The techniques were not such as to permit the limiting value of this expression as the mol distribution of n-pentane to be determined fraction of methane approaches zero in with satisfactory accuracy when it was both the gas and liquid phases may be present only in small amounts. For this obtained by differentiating the numerator reason uncertainty exists in regard to and denominator and rearranging. It the limiting behavior of n-pentane as C follows at the limit that approaches unity. As a result of this un­ certainty, values for the equilibrium con­ K I = C[ K 2 + K 5 (I ~ C) ] stant for n-pentane corresponding to a value of C of unity have not been shown C2 [3] + [(:~:) + (:~)(I ~ C)] in Fig 8. Similar uncertainty made it Eq 3 has been applied to the data pre­ desirable to eliminate the corresponding sented in Fig 6 and indicates goo d agree- curve for ethane in Fig 7. 22 PHASE BEHAVIOR IN THE METHANE-ETHANE-N-PENTANE SYSTEM

The marked influence of the nature emphasize the fact that the so-called and amount of the components present "equilibrium constant" is a function of upon values of the equilibrium constant the state of the system rather than a func­ of all the components at pressures in excess tion only of pressure and temperature for of 750 psi is demonstrated in Figs 6, 7, systems in which the components differ and 8. In the case of methane this influence markedly from one another in regard to persists throughout the two-phase re­ pertinent physical properties. gion of the methane-ethane-n-pentane Because of the inherent complexity of system. However, for ethane and n-pentane the state of a heterogeneous multicom­ the equilibrium constants become more ponent mixture, some difficulties are to nearly functions only of pressure and be expected in obtaining general correla­ temperature at pressures below 500 psi. tions of equilibrium constants for the It appears that ideal solution behavior lighter hydrocarbons in such systems. is closely approximated for pressures Since the chemical potential of a com­ below 250 psi in so far as these components ponent in a phase is a function of the are concerned. However, no experimental state only of that phase, rather than that Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021 measurements were made below 500 psi of the system as a whole, its use offers and this statement is based entirely upon somewhat better promise for purposes of graphical correlation of the data. generalization. However, insufficient. in­ The influence of the concentration of formation is as yet available to per­ other components upon the equilibrium mit serious attempt to obtain accurate constant for n-pentane is most pronounced correlations. at relatively high ratios of ethane to ACKNOWLEDGMENT n-pentane. This is shown by the small influence of changes in C upon the equilib­ This investigation was carried out under rium constant for n-pentane in Fig 8 at a fellowship supported by the Standard low values of this parameter and the Oil Company of California. The coopera­ markedly greater influence of high values. tion and financial assistance of this organi­ This behavior is similar to that found for zation is acknowledged. n-pentane in the methane-propane-n-pen­ REFERENCES tane system.3.' I. Katz and Hackmuth: Ind. and Eng. Chem. (1937) 29, 1072. SUMMARY OF RESULTS 2. Sage. Hicks and Lacey: Amer. Petro Inst. Drill. and Prod. Prac., 1938 1939386. This study indicates that at 100°F 3. Carter. Sage and Lacey: Trans. AIME (1941) 142, 170. the critical pressures of 'all ternary mix­ 4. Dourson. Sage and Lacey: Trans. AIME tures of methane, ethane and n-pentane (1943) 151, 206. 5. Sage and Lacey: Trans. AIME (1940) 136, are lower than the critical pressure of the 136. 6. Taylor. Waldo Sage and Lacey: Oil and Gas methane-n-pentane system at this tem­ Jnl. (1939) 38 (10). 46. perature. This behavior is similar to that 7. Sage. Reamer. Olds and Lacey: Ind. and Ene: Chem. (1942) 34, 1I08. •. which has been found for other ternary 8. Sage.\Lacey and Schaafsma: Ind. and Eng. mixtures of hydrocarbons so far investi­ Chem. (1935) 27, 48. 9. Sage and Lacey: Volumetric and Phase gated. The gas-liquid equilibrium con­ Behavior of Hydrocarbons. 86-89. 223- stant of each component is markedly 228. Stanford Univ. Press. 1939. influenced by the nature and amount of DISCUSSION other components present. The influence W. T. LIETZ*-This paper, like all previous persists to pressures well below 500 psi, ones presented by Dr. Lacey and his co- particularly in the case of the lighter • Shell Oil Company. Los Angeles. Cali­ components. This work serves further to fornia. DISCUSSION 23 workers, is an excellent contribution to the given in this paper are essential for accurate general information on phase behavior of prediction of phase equilibria in hydrocarbon ternary systems. The equilibrium constants systems. Because of the practical impossibility of these simplified systems have been applied of measuring all possible mixtures at all tem­ successfully to calculations on·" crude . peratures and pressures of interest, some provided the gravity is not too low. It is only method of correlating and extending these to be regretted that practically no data are data is required. As suggested by the authors, available for pressures over 3000 psi, especially the chemical potential, or fugacity, of each since with the increased drilling depths reser­ component in each phase, is the property pre­ voir pressures over 8000 psi and temperatures ferred for correlation. of 275°F or more are quite common. In our An equation of state;O initially developed computations we have found that the use of for mixtures of methane, ethane, propane and extrapolated equilibrium constants for the , provides a method for predicting these calculation of phase behavior of reservoir chemical potentials, and from them, equilibrium fluids leads to results which differ considerably constants, or k-values. Although there has not from experimental observations. been time to make a careful study of the best values for the constants in the equation of J. E. SHERBORNE*-Dr. Lacey and his co­ state for pentane, preliminary calculations Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021 workers have presented another significant indicate that the results of this paper may be contribution to add to the long list of excellent correlated satisfactorily by means of this equa­ researches they have performed for the tion of state. The maximum deviation of . The data presented equilibrium constants predicted in this way further supplement the somewhat discon­ from observed equilibrium constants is around certing knowledge that the kind and amount 4 pct for methane, 4 pct for ethane and 10 pct of other components in a system have an for pentane. The dependence of equilib­ important bearing on the equilibrium constant. rium constants on composition is correctly It is interestitlg to observe, however, that represented. Dr. Lacey offers us the possibility of another means of correlating complex hydrocarbon G. W. BILLMAN, B. H. SAGE and W. N. behavior in the use of the chemical potential. LACEY (authors' reply)-This laboratory is in Dr. Lacey observes that insufficient data are entire agreement with the statements by W. available concerning the chemical potential Tempelaar Lietz in regard to the difficulty of to permit its use in correlation at present. extrapolating equilibrium constants available It would be of interest to know Dr. Lacey's to the higher pressures encountered in under­ opinion as to, first, whether more information ground reservoirs. At the present'time work relative to the chemical potential could be is in progress at this laboratory upon another made available by further analysis of the ternary system involving one component of a existing experimental data, or if further somewhat higher molecular weight thus yield­ experimental work will be required; and second, ing correspondingly higher two-phase pressures. if further experimental work is required, would One of the primary limitations to the extension it be a relatively small amount supplementary of such studies to the higher pressures lies in to that already performed, or would extensive the difficulty of obtaining PUre materials of experimental research be required? In short, is sufficiently high molecular weight to obtain there a chance that a correlation based on two-phase pressures in excess of 10,000 psi. chemical potentials will be available to the If an impure constituent is employed as the industry reasonably soon? component of high molecular weight significant differences in the distribution of the heavier M. BENEDIcTt-Reliable experimental meas­ components making up this constituent are urements of the composition of coexistent liquid obtained in the liquid and the gas phases. and vapor such as are It is believed that if suitable methods of • Union Oil Company of California. Santa Fe analysis are available for the materials of Springs. California. t Hydrocarbon Research. Inc .• New York. 10 Benedict. Webb and Rubin: Jnl. Chem. New York. Ph,s. (1942) 10, 747. 24 PHASE BEHAVIOR IN THE METHANE-ETHANE-N-PENTANE SYSTEM higher molecular weight investigations can industries. The work presented in the current be carried to the ranges of pressure and tem­ paper needs no technical comment. However, perature of interest in petroleum production. having learned of the fact that Dr. Lacey is to However, the difficulty in obtaining adequate receive the Lucas Award of this year, I would supplies of pure compounds of high molecular like to take this opportunity to express my weight has made such data relatively scarce. sincere feeling that the award is highly merited When volumetric data for systems involving on the basis of his outstanding services to components of high molecular weight are industry in the fields of scientific research and available it will be possible to evaluate the education. constants of the Benedict equation of state for these components. By the application of this S. E. BUCKLEy*-It is always a real pleasure to read or hear a paper prepared by Dr. Lacey equation the corresponding phase behavior may be estimated with an accuracy equivalent and his associates, not solely because each to that with which this equation of state pre­ paper has technical merit and is clearly and concisely prepared in well-chosen words, but dicts the phase behavior of mixtures of the because each paper is another mile post repre­ lighter hydrocarbons. senting substantial, practical progress. During In regard to the discussion by Mr. Sherborne, the years that Dr. Lacey and his many co­ it is believed that a significant contribution Downloaded from http://onepetro.org/trans/article-pdf/174/01/13/2178576/spe-948013-g.pdf by guest on 30 September 2021 workers have labored with hydrocarbon toward the application of chemical potentials systems, they have progressed steadily and to the prediction of the composition of co­ continuously through the lowlands of in­ existing phases of hydrocarbon systems has dividual hydrocarbons, the rising plateau of been made by Manson Benedict.1o In another binary mixtures, and now into the higher discussion 11 of this paper by Manson Benedict altitudes of ternary systems. The highest it is indicated that his equation of state which peaks of multicomponent mixtures are clearly is already available will predict the equilibrium in sight ahead. constants for the components of the methane­ One who reads or listens' to one of these ethane-n-pentane system with uncertainties straightforward and lucid explanations of the varying from 4 to 10 pct over the entire range manner in which hydrocarbons behave is apt of composition. It is believed that the accuracy to get the impression that hydrocarbon mix­ with which chemical potentials may be derived tures are really very simple things to deal with. from this equation of state or by application of One whose primary technical pursuits in the other methods will be improved gradually as petroleum industry do not involve detailed additional data are accumulated. The possi­ calculation of the behavior of individual bility of the gradual improvement of accuracy hydrocarbons or their mixtures may overlook without the need of a complete revision of the the fact that the entire petroleum industry method of correlation is one of the strongest is a hydrocarbon industry, founded on and arguments in favor of the use of chemical consisting in the exploitation, production, potentials in this connection. This advantage processing, transportation, and marketing of also applies to the prediction of volumetric hydrocarbons, that millions of dollars are and thermodynamic data from the partial invested on the basis of processes and designs values of the corresponding properties. directly dependent on an exact quantitative R. J. SCHILTHUIs*-The current paper by knowledge of what hydrocarbons will do in a Billman, Sage, and Lacey represents another known environment. addition to the knowledge of complex hydro­ If, in retrospect, the job looks simple, no carbon phase behavior. Over the years, Dr. greater tribute could be paid to the foresight, Lacey and his various associates have been the careful planning and execution which Dr. outstanding contributors to this field of basic Lacey and his co-workers have demonstrated knowledge which has proved to be of great in giving to the petroleum industry the wealth value to the petroleum branch of the mineral of information which year after year has come from their papers. 11 Lewis: Proc. of Amer. Acad. Arts and Sci. (1907) 43, 273· • Humble Oil and Refining Company, Cor­ • Humble Oil and Refining Company, Hous­ pus Christi, Texas. ton, Texas.