Contributions in Petroleam Geology and Engineering Series Editor: George V. Chilingar, University of Southern California Volume 1: Geologie Analysis of Naturally Fractured Reservoirs Volume 2: Applied Open-Hole Log Analysis Volume 3: Underground Storage of Natural Gus Volume 4: Gas Production Engineering Volume 5: Properties of Oils and Natural Gases Volume 6: Introduction to Petroleum Reservoir Analysis Volume 7: Hydrocarbon Phase Behavior Volume 8: Well ‘Test Analysis Volume 9: Horizontal Drilling Gulf Publishing Company Houston, London, Paris, Zurich, Tokyo Contributions in Petroleum Geology & Engineering ($°7 DC Phase Behavior aContributions in Petroleum Geology and Engineering Yolume 7 ‘Hydrocarbon Phase Behavior Copyright © 1989 by Gulf Publishing Company, Houston, “Texas. All rights reserved. Printed in the United States of "America. This book, or parts thereof, may not be reproduced in any form without permission of the publisher. Printed on acid-free paper (%) Library of Congress Cataloging-in-Publication Data ‘Aled, Tarek H., 1946. Hydrocarbon phase behavior by Tarek Ahmed 1p. em.— (Contributions in petroleum geology and engineering; v. 7) Tneludes indes. 1, Petroleum—Migration. 2. Phase rule and equilibrium. I. Title. II. Series: Contributions in petroleum geology & engineering; 7 ‘TN870.5.A35 1989 89-1903 622" 1828 —de20 cP ISBN 0-87201-589-0 ISBN 0-87201-066-X (Series) 1 . Properties of Natural Gases ..+...+++ Contents Preface .. Basie Phase Behavior - Single-Component Systems, 1; TWwo-Component Systems, 15; Multi Component Systems, 21; Classification of Reservoirs and Reservoir Fluids, 23. Pure Component Physical Properties and Characterizing Undefined Petroleum Fractions .. eenaceers ae) Generalized Correlations for Estimating Physical Properties of Hj drocarbon Fractions, 36; Other Methods of Estimating Physical Properties of Petroleum Fractions, 42; Critical Compressibility Fac- tors, 46; Characterizing Hydrocarbon Heavy Fractions, 50; Methods Based on the PNA Determination, 51; Other Methods of Charac terizing the Hydrocarbon Heavy Fractions, 60; Determination of the Physical Properties of the Heavy Petroleum Fractions from Graphi- cal Correlations, 67, Behavior of Ideal Gases, 77; Ideal Gas Mixtures, 79; Properties of Ideal Gas Mixtures, 81; Behavior of Real Gases, 85; Elfect of Non- hydrocarbon Components on the Z-Factor, 95; Correetion for Non~ hydrocarbons, 95; Correction for High-Molecular-Weight Gases, 995 Direct Calculation of Compressbility Factors, 101; Compressibility oof Natural Gases, 106; Gas Formation Volume Factor, 111; Gas Vis- cosity, 113; Methods of Calculating the Viscosity of Natural Gases, 114; Engineering Applications of the Natural Gas PVT Properties, 123,4, Phase Behavior of Crude Oils .. 136 Crude Oil Density and Specific Gravity, 136; Methods for Determin- ing Density of Crude Oils of Unknown Composition, 137; Methods for Determining Density of Liquids of Unknown Composition, 151; Isothermal Compressibility Coefficient of Undersaturated Crude Oils, 156; Density of Undersaturated Crude Oils, 165; Gas Solubil- ity, 169; Oil Formation Volume Factor for Undersaturated Oils, 1925 ‘Total Formation Volume Factor, 198; Total System Isothermal Com- pressibility Coefficient, 199; Crude Oil Viscosity, 205; Methods of Caleulating the Viscosity of the Dead Oil, 207; Methods of Calculat ing the Viscosity ofthe Saturated Crude Oil, 209; Methods of Caleu- lating the Viscosity of the Undersaturated Crude Oil, 212; Caleulat- ing Viscosities of Crude Oils from their Composition, 2165 Bubble-point Pressure, 220; Surface Tension, 225; Application of the Crude Oil PVT Properties in Reservoir Engineering, 229; Radial Flow of Crude Oils, 229; The Material Balance Equation for Oil Res- ervoirs, 231 5. Vapor-Liquid Phase Equilibria . = 24 Equilibrium Ratios, 244; Flash Calculations, 247; Equilibrium Ra- tios for Real Solutions, 250; Equilibrium Ratios for the Plus-Frac- tions, 263; Applications of the Equilibrium Ratio in Petroleum Engi- neering, 270. 6. Equations of State ......60.0+ vee = 287 ‘The Generalized Form of Equations of State, 330; Applications of the Equation of State in Petroleum Engineering, 331; Three-Phase Equilibrium Caleulations, 339. 7. Splitting and Lumping Schemes of Petroleum Fractions ... 348 Splitting Schemes, 349; Lumping Schemes, 300. 8, Simulation of Laboratory PVT Data by Equations of State .. 378 406 Appent Index Equilibrium Ratio Curves .. 419 Preface “This book explains the fundamentals of hydrocarbon phase behavior and their practical application in reservoir and production enginee though the book was developed from notes prepared for hydrocarbon phase behavior courses given to senior petroleum students, it should be useful as a reference book to practicing petroleum engineers Chapter J reviews the principles of hydrocarbon phase behavior and il- lustrates the use of phase diagrams in describing the volumetric behavior of single-component, two-component, and multi-component systems. Chapter 2 presents numerous mathematical and graphical correlations for estimating the physical and critical properties of the undefined petroleum fractions; and Chapter 3 deals with evaluation of properties of natural gases and intro- ‘duces their applications in Darey’s equation and the material balance equa- tion, ‘A complete and cohesive independent unit, Chapter 4 focuses on methods ‘of determining the crude oil physical properties. Chapter 5 presents the con- cept and application of vapor-liquid phase equilibria. Chapter 6 reviews de- velopments and advances in the field of empirical “cubic” equations of state and demonstrates their applications in petroleum engineering. Schemes of splitting and lumping petroleum fractions are illustrated in Chapter 7, and Chapter 8 discusses the simulation of laboratory PVT data by equations of state. ‘Much of the material on which this book is based was drawn from the publications of the Society of Petroleum Engineers of the American Institute of Mining, Metallurgical, and Petroleum Engineers; the American Gas As- sociation; the Division of Production of the American Petroleum Institutes and the Gas Processors Suppliers Association. Tribute is due to these organi- zations and to the engineers, scientists, and authors who have made so many fine contributions to the field of hydrocarbon phase behavior. 1am indebted to my students at Montana Tech, whose enthusiasm for the subject has made teaching a pleasure. I would like to express my apprecia- tion to all the people who have helped in the preparation of the book bytechnical comment and discussion and by giving permission to reproduce material. Special thanks to my colleagues: Professor Art Story, Dr. Herbert Warren, Dz. Gil Cody, Dr. Gene Collins, and Dr, Dan Bradley, for their en- couragement and for making valuable suggestions for improvement of the book. I would also like to express my appreciation to the editorial staff of Gulf Publishing Company, especially Julia Starr. Thanks to Shanna for her patience and understanding, and believing that one day things would return to normal, Tarek Ahmed 1 Basic Phase Behavior ‘A “phase” is defined as any homogeneous part of a system that is physi- cally distinct and separated from other parts of the system by definite boundaries. For example, ice, liquid water, and water vapor constitute three separate phases of the pure substance HO because each is homogeneous and physically distinct from the others; moreover, each is clearly defined by the boundaries existing between them. Whether a substance exists in a solid, liq- tuid, or gas phase is determined by the temperature and pressure acting on the substance. It is known that ice (solid phase) can be changed to water {liquid phase) by increasing its temperature and, by further increasing tem- perature, water changes to steam (vapor phase). ‘This change in phases is termed Phase Behavior. "Hydrocarbon systems found in petroleum reservoirs are known to display multi-phase behavior over wide ranges of pressures and temperatures. The ‘most important phases which occur in petroleum reservoirs are: *# Liquid phase, e.g., crude oils or condensates © Gas phase, e-g., natural gases ‘The conditions under which these phases exist is a matter of considerable practical importance. The experimental or the mathematical determina tions of these conditions are conveniently expressed in different types of dia sams, commonly called Phase Diagrams. ‘The objective of this chapter is to review the basic principles of hydrocar- bbon phase behavior and illustrate the use of phase diagrams in deseribing and characterizing the volumetric behavior of single-component, two-com- ponent, and multi-component systems. “The simplest type of hydrocarbon system to consider is that containing one component. The word “component” refers to the number of molecular 12 Hydrocarbon Phase Behavior ‘or atomie species present in the substance. A single-component system is composed entirely of one kind of atom or molecule. We often use the word “pure” to deseribe a single-component system. ‘The qualitative understanding of the relationship between temperature T, pressure p, and volume V of pure components can provide an excellent basis for understanding the phase behavior of complex petroleum mixtures. The foregoing relationship is conveniently introduced in terms of experimental ‘measurements conducted on a pure component as the component is sub- jected to changes in pressure and volume at constant temperature. The ef- fects of making these changes on the behavior of pure components are next discussed. Suppose a fixed quantity of a pure component is placed in a cylinder fitted with a frictionless piston at a fixed temperature T, Furthermore, consider the initial pressure exerted on the system to be low enough that the entire system is in the vapor state. This initial condition is represented by point E, oon the pressure-volume phase diagram (p-V diagram) as shown in Figure 1-1. Consider the following sequential experimental steps taking place on the pure component: Step 1. ‘The pressure is increased isothermally by forcing the piston into the cylinder. Consequently, the gas volume decreases until it reaches point F on the diagram, where the liquid begins to con- dense. The corresponding pressure is known as the dew-point pressure pg, and is defined as the pressure at which the first drop- let of li formed. Step 2. The piston is moved further into the eylinder as more liquid con- denses, This condensation process is characterized by a constant pressure and represented by the horizontal line FG. At point G, traces of gas remain and the corresponding pressure is called the bubble-point pressure pp, and defined as the pressure at which the first sign of gas formation is detected. A characteristic of a single-component system is that at a given temperature, the dew-point pressure and the bubble-point pres- sure are equal. Step 3. As the piston is forced slightly into the cylinder, a sharp increase in the pressure (point H) is noted without an appreciable decrease in the liquid volume. That behavior evidently reflects the low compressibility of the liquid phase. By repeating the above steps at progressively increasing temperatures, a family of curves of equal temperatures (isotherms) is construpted as shown in Figure 1-1, The dashed curve connecting the dew points is called the dew-point curve (line FC) and represents the states of the “saturated gas.” ‘The dashed curve connecting the bubble points is called the bubble-point Basic Phase Behavior 9 Pressure Volume ———+ Figure 1-1. Pressure-volume diagram for a single component system. curve (line GC) and similarly represents the “saturated liquid.” These two ‘curves meet at point C which is known as the critical point. The correspond ing pressure and volume are called the critical pressure p. and critical vol- uume Vc, respectively. Notice that as the temperature increases, the length of the straight line portion of the isotherm decreases until it eventually van- ishes, and the isotherm merely has a horizontal tangent and inflection point at the critical point. This isotherm temperature is called the critical temper- ature Tz of the single-component system. This observation can be expressed mathematically by the following relationship; ap : fz). 0 (at the critical point) (ay Pl cal poi - (ea 0 (at the critical point) (2) Referring to Figure 1-1, the area enclosed by the phase envelope AFCGB is called the two-phase region. Within this defined region, vapor and liquid4 Hydrocarbon Phase Behavior can coexist in equilibrium, Outside the phase envelope, only one phase ean exist ‘The critical point (point C) deseribes the critical state of the pure compo- mnt and represents the limiting state for the existence of two phases, ic, liquid and gas. In other words, for a single-component system, the critical point is defined as the highest value of pressure and temperature at which two phases can coexist. A more generalized definition of the critical point which is applicable to a single or multi-component system is: the critical point is the point at which all intensive properties of the gas and liquid phases are equal. {An intensive property is one that has the same value for any part of a homogeneous system as it does for the whole system, i.e., a property which is independent of the quantity of the system. Pressure, temperature, density, composition, and viscosity are examples of intensive properties. ‘Many characteristic properties of pure substances have been measured and compiled over the years. These properties provide vital information for calculating the thermodynamic properties of pure components as well as their mixtures. Those physical properties that are needed for hydrocarbon phase behavior caleulations are presented in Table 1-1 for a number of hy- drocarbon and non-hydrocarbon components. ‘Another means of presenting the results of this experiment is shown, graphically in Figure 1-2 in which the pressure and temperature of the sys- tem are the independent parameters. Figure 1-2 shows a typical pressure- temperature diagram (p-T diagram) of a single-component system. The re- sulting curve, i., line AC, which terminates at the critical point (point C), can be thought of as being the dividing line between the area where liquid ‘and vapor exists. The curve is commonly called the "vapor-pressure curve” Figure 1-2. Pressure-temper ture diagram for a pure compo- rent system. Pressure ‘Temperature ————> Basic Phase Behavior 5 or the “boiling-point curve.” The corresponding pressure at any point on the curve is called the vapor pressure py. Figure 1-2 shows that at the conditions of pressure and temperature speci- fied by the vapor pressure curve, two phases ean coexist in equilibrium. Sys- tems represented by points located below the vapor-pressure curve are com- posed only of the vapor phase. Similarly, points above the curve represent systems that exist in the liquid phase. These remarks can be conveniently summarized by the following expressions: if p < p,~*system is entirely in the vapor phase p > p.-*system is entirely in the liquid phase P = Py7Vapor and liquid coexist in equilibrium where p is the pressure exerted on the pure substance. It should be pointed out that the above expressions are valid only if the system temperature is below the critical temperature 'T, of the substance. ‘A method which is particularly convenient for plotting the vapor pressure as a function of temperature for pure substances is shown in Figure 1-3. The chart is known as the “Cox Charts.” Notice that the vapor pressure seale is, logarithmic, while the temperature scale is entirely arbitrary. Example 1-1. A pure propane is held in a laboratory cell at 80°F and 200 psia. Determine the “existence state” (Le., as a gas or liquid) of the sub- stance. Solution. From the Cox Charts, the vapor pressure p, = 150 psi, and be- ‘cause the propane is under 200 pi (i., p>), this means that the labora- tory cell contains a liquefied propane. "The vapor pressure chart as presented in Figure 1-3 allows a quick estima- tion of the vapor pressure p, of a pure substance at a specific temperature. For computer applications, however, an equation is more convenient. Lee and Kesler (1975) proposed the following generalized vapor pressure equa- tion: Pv = pe EXP (A+ @B) a3) with, A= 5.92714 - 19648 ~ 1,286 In (T,) + 0.16994(T,)® (4) B= 15.2518 - Bes 13.4721 In (T,) + 0.4357(T,)° (sy7 Basic Phase Behavior Hydrocarbon Phase Behavior 6 ui6U3 SIO o4i UI PEYSIIGNE ‘UolleIS0ssy SB\\ddNg si0sse901g SEH Ayp JO ASB}NOD 10 Kousisisu09 on914D8 0; sanyen pareinaye> 0} peppe 16 Sej2euno5e eneY OU Op aIger SIul Ul SJOQUION ‘LON 4P3 uiual, ‘Yoo wie 6 “sanya jeyvowyedxe jo vore|noyeoa: ‘ugaq oney sy/6ip Biix@ S9SE @wWOS UI OOO; Ul Hed | UeUI ‘9st0"0 [iz soc Soez'0 |e osr-les: ze. ri ia ss Zero 0 @ ooze | s0s6:0 3 ‘ata: 0 is| os| c1s0°0 ‘ oa} | Lasoo t ee os] Sois-0 i | | ciso-0 es 3s] tesor0 Ed ss] soro-0 arise ss Fa] tavo-0 Bere les} ce| seco. —— Fe es] teso. — eg 15] tes0 ig os] oss os er] zes0" ¥sor as| vas0:0 ols | ocs0°0 260s oe] cas0-0 ooze Ss] Zes0°0 Sire | $550°0 once Es] o250°0 $568 ey] tes0-0 you is] s6s0°0 08s 955) es ) BAASBRA ARS Raeeas: RRR 88 RR ietee | sez Geil oteed| stousn lee Ba eeakR BSE ETE EE sg: ana: ® SBAROGT AES is 28989 28s! Be os : 35 : zi SEP Rel? : Ualiesiiistis leans ar} ep] E : 7 @ 3 a 3s ed ele; a] § peelec porters S/:S is] Sate ee ale mane won| F £ Siuauoduieg aang 10} samuedosg (eorehid =: Fb arg,8 Hydrocarbon Phase Behavior Basic Phase Behavior 9 RABARRAAAS digits have been wai/s08 HH a v= anlage, \c4yyaos6 21512080) FE Aujmap eanoied ee 2333 | Torsos_| 73.6600 20088 8 Table 1-1 é BRRSRRGE BSISH seasaees E8882 FPITT_TIPPTTTT_FTFET ¥ smyosedee4| 378: | ~0.00300+| 0.1289 | 0:9925 | © sou a1/"108) i is 28 RSRgSER Courtesy of the Gas Processors Suppliers Association. Published in the GPSA Engineering Data Book, Tenth Ect “16/001 2 4609/4900 Beas SoBeEanT 8 3 (fa 0s 215;0t 3 3 e cvntonsieniel eat lad HES EE His EEG | zeaunnf FF RaRAS RRA GNARMRRANS sSTeESTS aaaas AoRAsSwTs| 510 Hydrocarbon Phase Behavior Basic Phase Behavior 11 ‘The term 7, is called the reduced temperature and is defined as the ratio of the absolute system temperature to the critical temperature of the frac- tion, or f. Temperature, °F rocessors where T, = reduced temperature substance temperature, °R ‘T, = critical temperature of the substance, °R P. = critical pressure of the substance, psia ‘& = acentrie factor’ of the substance. Tenth Edition, 1987, ‘The acentric factor w was introduced by Pitzer (1955) as a correlating pa- rameter to characterize the acentricity or non-sphericity of a molecule, and ined by the following expression: w= - Lo |- 1 i) tart for hydrocarbon components. Courtesy of the Gas Pr where p, = vapor pressure of the substance at T = 0.7 Te, psia Po = critical pressure of the substance, psia ls ‘The acentrie factor is frequently used as a third parameter in correspond: ing states and equation-of-state correlations. Values of the acentric factor for pure substances are tabulated in Table 1-1. ‘od in the GPSA Engineering Data Book, Ss Example 1-2. Calculate the vapor pressure of propane at 80°F by using |, 83 the Lee and Kesler correlation. ge 88 Solution BS © Obtain the critical properties and the acentric factor from Table 1-1 28 "7, = 696.01 °R fe Pe = 616.3 psia 22 co = 0.1522 ° B 2 © Calculate the reduced temperature. ‘j94“aunssaud soden 0.81108 + When Equation Ll employed to caleulatep, itis reommended that Eq Chapter 2 be used to compute w. ation 3-13 from12 Hydrocarbon Phase Behavior * Solve for the parameters A and B by applying Equations 1-4 and 1-5, respectively. ~ 1.27359 Be ~ 1147085 ‘Solve for py by applying Equation 1-1 py = 616.3 EXP ( ~ 1.27359 + 0.1572 ( ~ 1.147045)) = 145 psia ‘The densities of the saturated phases of a pure component, i.e., densities of the coexisting liquid and vapor, may be plotted as a function of tempera- ture, as shown in Figure 1-4. It should be noted that for increasing tempera- ture, the density of the saturated liquid is decreasing, while the density of the saturated vapor increases. At the critical point C, the densities of vapor and liquid converge. At this eritical temperature and pressure, all other properties of the phases become identical. crsical Point Average Density qitatee sees eeae Density Tataraved Gas Figure 1-4. Typical density-temperature diagram for a pure component. Figure 1-4 illustrates a useful observation, known as the Law of the Re linear Diameter, which states that the arithmetic average of the densi the liquid and vapor phases is a linear function of the temperature. ‘The straight line of average density versus temperature makes an easily defined intersection with the curved line of densities. Ths intersection then gives the critical temperature and density. Mathematically, this relationship is ex- presed as follows: Basic Phase Behavior 13 Pet Pll aa bT i 7 + a7) where p, = density of the saturated vapor, Ibift® px = density of the saturated liquid, Ibvft? ‘T = temperature, °R a, b = intercept and slope of the straight line, At the ritical point, Equation 1-7 can be expressed in terms ofthe critical density as follows: pow atbT. (18) where pe = critical density of the substance, Ib/ft? Rackett (1970) proposed a simple generalized equation for predicting the saturated liquid density “py,” of pure compounds. Rackett expressed the cor- relation in the following form: (MW) pe PRT ZA + 1-1)" 5 (19) where p;, = saturated liquid density of the pure substance, Ibift? MW = molecular weight of the pure substance Po = critical pressure of the substance, psia ‘T, = critical temperature of the substance, °R Z, = critical gas compressibility factor R = gas constant, 10.73 ft? psia/lb-mole, °R T,= 2, reduoed temperature T = temperature, °R Spencer and Danner (1973) modified Rackett’s correlation by replacing the critical compressibility factor Z. in Equation 1-9 with the parameter Zn, which is a unique constant for each compound. ‘The authors proposed the following modification of the Rackett equation, (MW) Pe RT. Zpall + (= T)*) (2-10) ‘The values of Zn, are given in Table 1-2 for selected components. If a value is not available, Yamada and Gunn (1973) suggested the following correlation to estimate Zan:14 Hydrocarbon Phase Behavior Table 1-2 Values of Zpa for Selected Pure Components Carton dioxide ore ‘-Pontane 0.2684 Nigen 0.2900 nHexane 0.2635 Hydrogen sulfide 0.2885 ‘rHepiane 0.2606 Methane 0.2892 Octane 0.2686 Ethane 0.2808 n-Octane 02871 Propane 0.2766 ‘Nonane 0.2543 Butane 0.2754 nDecane 0.2507 Butane 0.2730 r-Undacane 0.2498 HPentane o27i7 Za = 0.29056 ~ 0.08775 ay where w is the acentric factor of the compound. Example 1-3. Calculate the saturated liquid density of propane at 160°F by using 1. The Rackett correlation b. The modified Rackett equation, Solution. * From Table 1-1 T. = 666.06 °R pe = 616.0 psia MW = 44.097 Ve = 0.0727 Ibitt? # Calculate Z, by applying Equation 3-15 of Chapter 3. peVeMW. RT. __ (616.0) (0.0727) (44.097) 0.2763 (10.73) (666.06) Z 160 + 460 666.06 93085 ‘The Rackett correlation. Solve for the saturated liquid density by apply- ing the Racketp equation, i.e., Equation 1-9 Basic Phase Behavior 15 pL. 0 Ob Or = 25.05 Ibift? b. The modified Rackett equation, From Table 1-2, Za, = 0.2706 © Applying the modified Rackett equation, i.e., Equation 1-10 (44.097)(616) ~ (OT 666. 06)(0-zTERyaaT ~ 25-01 Wie is "TWO-COMPONENT SYSTEM! ‘A distinguishing feature of the single-component system is that, at a fixed ‘temperature, two phases (vapor and liquid) can exist in equilibrium at only ‘one pressure; this isthe vapor pressure. For a binary system, two phases can exist in equilibrium at various pressures at the same temperature, The fol- lowing discussion concerning the description of the phase behavior of a two- component system involves many concepts that apply to the more complex component mixtures of oils and gases. One of the important characteristics of the binary systems is the variation of their thermodynamic and physical properties with the composition, ‘Therefore, it is necessary to specify the composition of the mixture in terms of mole or weight fractions. It is customary to designate one of the compo- nents as the more volatile component and the other the less volatile compo- nent, depending on their relative vapor pressure at a given temperature. Suppose that the experiments previously described for a pure component are repeated, but this time we introduce into the cylinder a binary mixture of a known overall composition. Consider that the initial pressure py exerted on the system, at a fixed temperature of T,, is low enough that the entire system exists in the vapor state. This initial condition of pressure and tem- perature acting on the mixture is represented by point 1 on the p-V diagram of Figure 1-5. As the pressure is increased isothermally, it reaches point 2, at which an infinitesimal amount of liquid is condensed. The pressure at this point is called the dew-point pressure py of the mixture. It should be noted that at the dew-point pressure, the composition of the vapor phase is equal to the overall composition of the binary mixture. As the volume is decreased (by forcing the piston inside the cylinder), a noticeable increase in the pres- sure is observed as more and more liquid is condensed. This condensation process is continued until the pressure reaches point 3, at which traces of gas remain, At point 3, the corresponding pressure is called the bubble-point pressure py. Because at the bubble point the gas phase is only of infinitesimal volume, the composition of the liquid phase is therefore identical with that16 Hydrocarbon Phase Behavior of the whole system. As the piston is forced further into the cylinder, the pressure rises steeply to point 4 with a corresponding decreasing volume. Repeating the above experiments at progressively increasing tempera- tures, a complete set of isotherms is obtained as shown on the p-V diagram of Figure 1-6, The bubble-point curve, as represented by the line AC, repre sents the locus of the points of pressure and volume at which the first bubble of gas is formed. The dew-point curve (line BC) describes the locus of the points of pressure and volume at which the first droplet of liquid is formed, ‘The two curves meet at the critical point (point C). The critical pressure, temperature, and volume are given by ps, Ts, and Ve, respectively. Any point within the phase envelope (line ACB) represents a system consisting of ‘two phases. Outside the phase envelope, only one phase can exist. If the bubble-point pressure and dew-point pressure for the various isotherms on a p-V diagram are plotted as a function of temperature, a p-T diagram similar to that shown in Figure 1-7 is obtained. Figure 1-7 indicates that the pressure-temperature relationships can no longer be represented by a simple vapor pressure curve as in the case of a single-component system, but take on the form illustrated in Figure 1-7 by the phase envelope ACB. ‘The dashed lines within the phase envelope are called the “quality lines”; they describe the pressure and temperature conditions of equal volumes of liquid, Obviously, the bubble-point curve and the dew-point curve repre- sent 100% and 0% liquid, respectively. Figure 1-8 demonstrates the effect of changing the composition of the bi- nary system on the shape and location of the phase envelope. Two of the lines, as shown in Figure 1-8, represent the vapor pressure curves for meth- Figure 1-5. Pressure-volume di- gram for a binary system. Yoese —j-————> Basle Phase Behavior 7 Figure 1-6. Pressure-volume diagram for a two-component system. Pressure Tesperstare Figure 1-7. Typical pressure-temperature diagram for a binary system.} Basic Phase Behavior 19 ‘ane and ethane, which terminate at the critical point. Ten phase boundary curves (phase envelopes) for various mixtures of methane and ethane are also shown. These curves pass continuously from the vapor-pressure curve of the one pure component to that of the other as the composition is varied, ‘The points labeled 1-10 represent the eritical points of the mixtures as de- fined in the legend of Figure 1-8. The dashed curve illustrates the locus of critical points for the binary sys It should be noted by examining Figure 1-8 that when one of the constitu- cents becomes predominant, the binary mixture tends to exhibit a relatively narrow phase envelope and displays critical properties cose to the predomi nant component. The size of the phase envelope enlarges noticeably as the ‘composition of the mixture becomes evenly distributed between the two components. Figure 1-9 shows the critical loci for a number of common binary systems, Obviously, the critical pressure of mixtures is considerably higher than the citical pressure of the components in the mixtures. The greater the differ- cence in the boiling point of the two substances, the higher the critical pres- sure of the mixture will be. No. 17. 3 a =e ‘The Phase Rule It is appropriate at this stage to introduce and define the concept of the “Phase Rule.” Gibbs (1876) derived a simple relationship between the num- ber of phases in equilibrium, the number of components, and the number of independent variables that must be specified to describe the state of the sys- tem completely. Gibbs proposed the following fundamental statement of the phase rule, Teupenarure zi i « a g 2 $ 3 i g 8 : i & 5 é 3 z : F=C-P+2 (1-13) a: iW where F = number of variables required to determine the state of the sys- TELE: 5 tem at equilibrium, or number of degrees of freedom gl sFrscgers sy 2 C = number of independent components te 32 P = number of phases Bh seg gegeees | 2 g ig 6 Sree ease ea . A phase has been defined previously asa homogeneous sytem of uniform as ec physical and chemical composition’In a system containing ice, liquid water, 22 ccccc0e at is and water vapor in equilibrium there are three phases, ie., P= 3. The saenecel number of independent components in the system is one, i.e., C = 1, since ig the system contains only HO. The degrees of freedom F for a system include : * the temperature, the pressure, and the composition (concentration) of e phases, These independent variables must be specified to define the system Buuetelgue Saset: see cseeecsee\- leet eesl® See Sseeued completely i visa aunss3ud20 Hydrocarbon Phase Behavior Basic Phase Behavior 21 i ‘The phase rule as described by Equation 1-12 is useful in several ways. It indicates the maximum possible number of equilibrium phases that can co- exist and the number of components present. It should be pointed out that the phase rule does not determine the nature, the exact composition, or total ‘quantity of the phases, Furthermore, it applies only to a system in stable ‘equilibrium and does not determine the rate at which this equilibrium is at- tained. ‘The importance and the practical application of the phase rule are illus- trated through the following examples: Example 1-4, For a single-component system, determine the number of degrees of freedom required for the system to exist in the single-phase region. Solution. Applying Equation 1-12, gives F = 1-1 +22. There are two degrees of freedom that must be specified for the system to exist in the single phase region. These must be the pressure p and the temperature T. Example 1-5. What are the degrees of freedom allowed for a two-compo- i ‘ nent system in two phases? 3 a3 Solution, Since C = 2.andP = 2, applying Equation 1-12 yields F ~ 2 — 2 e 2 + 2 = 2. The two degrees of freedom could be the system pressure and the g system temperature, or p and the concentration (mole fraction), or some ie other combination of T, p, and composition. Example 1-6. For a three-component system, determine the number of de- agrees of freedom that must be specified for the system to exist in the one: phase region. Convergence Pressures for Hydrocarbons Solution. Using the phase rule expression gives: F = 3 ~ 1 + 2= 4. There are four independent variables that must be specified to fix the system. The variables could be the pressure, the temperature, and the mole fractions of two of the three components MULTI-COMPONENT SYSTEMS ‘The phase behavior of multi-component hydrocarbon systems in the liq- uuid-vapor region is very similar to that of binary systems. However, as the system becomes more complex with a greater number of different compo- Figure 1-9. Critical loci of binary systems. Courtesy of the Gas Processors Suppliers Association, Published in the GPSA Engineering Data Book, Tenth Edition, 1987. 1 HGES a po guaran nents, the pressure and temperature ranges in which two phases exist in- isd "3UNSS3¥é JONZDUIANOD crease significantly.22 Hydrocarbon Phase Behavior scuratod OL Reservoir Lape 12 Gitscat Point ‘Two-Phase Region Figure 1-10. Typical P-T diagram for a multi-component system. Figure 1-10 shows a typical pressure-temperature diagram of a mul component system with a specific overall composition. Although a different hydrocarbon system would have a different phase diagram, the general con- figuration is similar. ‘These multi-component p-T diagrams are essentially used to * Classify reservoirs * Classify the naturally occurring hydrocarbon systems * Describe the phase behavior of the reservoir fluid ‘To fully understand the significance of the p-T diagrams, itis necessary to identify and define the following key points on the p-T diagram: ‘© Cricondentherm (T..)—The ericondentherm is defined as the maxi- mum temperature above which liquid cannot be formed regardless of pressure (point E). The corresponding pressure is termed the criconden- therm pressure Pa. Basic Phase Behavior 23 Cricondenbar (p.,)—The cricondenbar is the maximum pressure above which no gas can be formed regardless of temperature (point D). The corresponding temperature is called the ericondenbar temperature Ty. Critical point—The eritical point for a multi-component mixture is 1e- ferred to as the state of pressure and temperature at which all intensive properties of the gas and liquid phases are equal (point C). At the criti- cal point, the corresponding pressure and temperature are called the critical pressure p, and critical temperature T of the mixture. © Phase envelope (two-phase region)—The region enclosed by the bub- ble-point curve and the dew-point curve (line BCA), wherein gas and liquid coexist in equilibrium, is identified as the phase envelope of the hydrocarbon system. Quality lines—The dashed lines within the phase diagram are called quality lines. They describe the pressure and temperature conditions for equal volumes of liquids. Note that the quality lines converge at the critical point (point ©) * Bubble-point curve—The bubble-point curve (line BC) is defined as the line separating the liquid phase region from the two-phase region. ‘* Dew-point curve—The dew-point curve (line AC) is defined as the line separating the vapor phase region from the two-phase region. ILASSIFICATION OF RESERVOIRS AND RESERVOIR FLUIDS Proper classification of a reservoir requires the knowledge of the thermo dynamic behavior of the phases present in the reservoir and forces responsi- ble for the production mechanism. In general, reservoirs are conveniently classified on the basis of the location of the point representing the initial res: ervoir pressure p; and temperature T with respect to the p-T diagram of the reservoir fluid. Accordingly, reservoirs can be classified into essentially two types. These are: © Oil reservoirs—If the reservoir temperature TT is less than the critical ‘temperature T, of the reservoir fluid, the reservoir is classified as an oil * Gas reservoirs—If the reservoir temperature is greater than the critical temperature of the hydrocarbon fluid, the reservoir is considered a gas Oil Reservoirs Depending upon initial reservoir pressure p,, oil reservoirs can be subclas- sified into the following categories:24 Hydrocarbon Phase Behavior 1. Undersaturated Oil Reservoir. Ifthe initial reservoir pressure p, (as repre- sented by point 1 on Figure 1-10), is greater than the bubble-point pressure po of the reservoir fluid, the reservoir is labeled an undersaturated oil reser- 2, Saturated Oil Reservoir. When the initial reservoir pressure is equal to the bubble-point pressure of the reservoir fluid, as shown on Figure 1-10 by point 2, the reservoir is called a saturated oil reservoir, 3. Gas-cap Reservoir. If the initial reservoir pressure is below the bubble- point pressure of the reservoir fluid, as indicated by point 3 on Figure 1-10, the reservoir is termed a gas-cap or two-phase reservoir, in which the gas or ‘vapor phase is underlain by an oil phase. The ratio of the gas-cap volume to reservoir oil volume is given by the appropriate quality line Crude oils cover a wide range in physical properties and chemical compo- sitions, and itis often important to be able to group them into broad catego- ries of related oils. In general, crude oils are commonly classified into the following types: Ordinary black oil * Low-shrinkage crude oil © High-shrinkage (volatile) crude oil © Near-eritical crude oil The above classification is essentially based upon the properties exhibited by the crude oil, including: * Physical properties * Composition oil ratio ‘ Appearance ‘ Pressure-temperature phase diagram Note that the reservoir temperature also plays a role in the classification of the crude oil. 1. Ordinary Black Oil. A typical p-T phase diagram for an ordinary black oil is shown in Figure 1-11. The phase diagram is characterized by quality lines that are approximately equally spaced. Following the pressure reduc- tion path as indicated by the vertical line EF on Figure 1-11, the liquid shrinkage curve, as shown in Figure 1-12, is prepared by plotting the liquid volume percent as a function of pressure. The liquid shrinkagé curve ap- proximates a straight line except at very low pressures. When produced, or- dinary black oils usually yield gas-oil ratios betwoen 200-700 scf/STB and Basic Phase Behavior 95 Liguig Gptescad Point toot Teaperacure —————> Figure 1-11. A typical P-T diagram for an ordinary black oil Pet tvrano Residval OLL paae Ligatd Volune 2 Figure 1-12. Liquid shrinkage curve for black ol.26 Hydrocarbon Phase Behavior oil gravities of 15 to 40°API. The stock-tank oil is usually brown to dark green in color. 2, Low-shrinkage Oil. A typical p-TT phase diagram for a low shrinkage ofl {s shown in Figure 1-13. The diagram is characterized by quality lines that are closely spaced near the dew-point curve. The liquid shrinkage curve, as given in Figure 1-14, shows the shrinkage characteristics of this category of crude oils. The other associated properties of this type of crude oil are: Figure 1-13. A typical phase diagram for a low shrinkage oll Figure 1-14. A typical liq- uid shrinkage curve for a low shrinkage oil Ligasd Voluse £ Basic Phase Behavior 27 * Gas-oil ratio less than 200 scf/STB © Oil gravity less than 15°APT Black or deeply colored ‘¢ Substantial liquid recovery at separator conditions as indicated by point G on the 85% quality line of Figure 1-13, 3. Volatile Crude Oil. The phase diagram for a volatile (high-shrinkage) crude oil is given in Figure 1-15. Note that the quality lines are close to- gether near the bubble point and at lower pressures they are more widely spaced. This type of crude oil is commonly characterized by a high liquid shrinkage immediately below the bubble point as shown in Figure 1-16, The other characteristic properties of this oil include: *# Gasol ratios between 2,000-3,500 scf/STB © Oil gravities between 45-55°API # Lower liquid recovery of separator conditions as indicated by point G on Figure 1-15 * Greenish to orange in color 4. Near-critical Crude Oil. If the reservoir temperature T is near the critical temperature’T, of the hydrocarbon system, as shown in Figure 1-17, the hy- drocarbon mixture is identified as a near-critical crude oil. Because all the Liguta cpsetcat Point Figure 1-15. A typical P-T diagram for a volatile crude oil28 Hydrocarbon Phase Behavior ‘quality lines converge at the critical point, an isothermal pressure drop (as shown by the vertical line EF in Figure 1-17) may shrink the rude oil from 100% of the hydrocarbon pore volume at the bubble point to 55% or les at 1 pressure 10 to 50 pi below the bubble point. The shrinkage characteristic behavior of the near-crtical crude oil is shown in Figure 1-18. Ligeid Volume Figure 1-16. A typical liquid shrinkage curve for a volatile crude oil Figure 1-17. A schematic phase diagram for the near-crtical crude oil. Basic Phase Behavior 29 100 50 Liquie Voluse Pressure Figure 1-18. A typical liquid shrinkage curve for the near-critical crude ol. Gas Reservoirs In general, if the reservoir temperature is above the critical temperature of the hydrocarbon system, the reservoir is classified as a natural gas reser- voir. Natural gases can be categorized on the basis of their phase diagram ‘and the prevailing reservoir condition into four categories: ‘ Retrograde gas-condensate ‘* Near-critcal gas-condensate # Wet gas © Dry gas Retrograde Gas-condensate Reservoir. Ifthe reservoir temperature T lies be- ‘tween the critical temperature T. and cricondentherm T.; of the reservoir fluid, the reservoir is classified as a retrograde gas-condensate reservoir. This category of gas reservoir is a unique type of hydrocarbon accumulation in that the special thermodynamic behavior of the reservoir fluid is the control- ling factor in the development and the depletion process of the reservoit. Consider that the initial condition of a retrograde gas reservoir is repre sented by point 1 on the pressure-temperature phase diagram of Figure 1-19, Because the reservoir pressure is above the upper dew-point pressure, the hydrocarbon system exists as a single phase (ic., vapor phase) in the reser- voir, As the reservoir pressure declines isothermally during production from the initial pressure (point 1) to the upper dew-point pressure (point 2), liquid begins to condense. As the pressure is further decreased, instead of expand. ing (if a gas) or vaporizing (if a liquid) as might be expected, the hydrocar-30 Hydrocarbon Phase Behavior See egerim cae Figure 1-19. A typical phase diagram of a retrograde system. 1auid drop-out T cle Ligute vot ° T Figure 1-20. A typical liquid drop-out curve. bon mixture tends to condense. This retrograde condensation process contin- ues with decreasing pressure until the liquid drop-out reaches its maximum at point 3. However, at point 4, the dew-point curve must be crossed again. ‘This means that all the liquid which formed mast vaporize because the sy- tem is essentially all vapor at the lower dew-point. “The lguldsifukage vlumo curv, eommorly called the Kiquid drop-out curve, for a condensate system is shown in Figure 1-20. In most gas-conden- Basie Phase Behavior 31 sate reservoirs, the condensed liquid volume seldom exceeds more than 10% of the pore volume. This liquid saturation is not large enough to allow any liquid flow. It should be recognized, however, that around the wellbore where the pressure drop is high, enough liquid drop-out might accumulate to give two-phase flow of gas and retrograde liquid, The associated physical characteristics of this category are: * Gas-oil ratios between 8,000 to 70,000 scfiSTB. Generally, the gas-oil ratio for a condensate system increases with time due to the liquid drop- ‘out and the loss of heavy components in the liquid. # Condensate gravity above 50°API # Stock-tank liquid is usually water-white or slightly colored. Near-critical Gas-condensate Reservoir. Ifthe reservoir temperature is near the critical temperature, as shown in Figure 1-21, the hydrocarbon mixture is classified as a near-eritical gas-condensate. The volumetric behavior of this category of natural gas is described through the isothermal pressure de- lines as shown by the vertical line 1-3 in Figure 1-21 and also by the corre- sponding liquid drop-out curve of Figure 1-22. Because all the quality converge at the critical point, a rapid liquid build-up immediately below the dew-point will result (Figure 1-22) as the pressure is reduced to point 2. critical poine Pressure Figure 1-21. A typical phase diagram for a near crtical-crtical gas condensate reservoir32 Hydrocarbon Phase Behavior a Figure 1-22. Liquid shrinkage curve for a near-critical gas-condensate system. Ligutd votunes ‘This behavior can be justified by the fact that several quality lines are crossed very rapidly by the isothermal reduction in pressure. At the point ‘where the liquid ceases to build up and begins to shrink again, the reservoir goes from the retrograde region to a normal vaporization region. Wet Gas Reservoir. A typical phase diagram of a wet gas is shown in Figure 1.23, where reservoir temperature is above the cricondentherm of the hy- drocarbon mixture. Because the reservoir temperature exceeds the ericon- dentherm of the hydrocarbon system, the reservoir fluid will always remain in the vapor phase region as the reservoir is depleted isothermally, along the vertical line A-B. However, as the produced gas flows to the surface, the pressure and temperature of the gas will decline. If the gas enters the two- phase region, a liquid phase will condense out of the gas and be produced from the surface separators. ‘Wet gas reservoirs are characterized by the following properties: ‘© Gas-oil ratios between 60,000 to 100,000 sef/STB © Stock-tank oil gravity above 60°APL ‘© Liquid is water-white in color © Separator conditions, i.e., separator pressure and temperature, lie within the two-phase region Dry Gas Reservoir. The hydrocarbon mixture exists as a gas both in the res- ervoir and the surface facilities. The only liquid associated with the gas from a dry gas reservoir is water. A phase diagram of a dry gas reservoir i given in Figure 1-24. Usually a system having a gas-oil ratio greater than 100,000 scf/STB is considered to be a dry gas. Basic Phase Behavior 33 eritical point INITIAL RESERVOTR tiger Temperature > Figure 1-23. Pressure-temperature diagram for a wet gas reservoit Pressure ——» / Toneronure Figure 1-24. A typical pressure-temperature diagram for dry gas reservoir.u Hydrocarbon Phase Behavior ‘PROBLEMS 1. A pure component has the following vapor pressure: TF 104 40 176 a2 Pos Bi 46.09 195.04 345.19 713.75 a. Plot the above data so as to obtain a nearly straight line b. Determine the boiling point at 200 psi c. Vapor pressure at 250°F 2. ‘The critical temperature of a pure component is 260°F, The densities of the liquid and vapor phase at different temperatures are: oF 86. Le 158 212 pu. loin | 40.28 38.16 35.79 30.89 Lp, Wit? 0.886 1g | 2.402 5.054 | Determine the eritical density of the substance. 3. Using the Lee and Kesler vapor correlation, calculate the vapor pres- sure of i-butane at 10°F. Compare the calculated vapor pressure with that obtained from the Cox charts. 4, Calculate the saturated liquid density of n-butane at 200°F by using: a, The Rackett correlation. D. The modified Rackett correlation. 5. What is the maximum number of phases that can be in equilibrium at constant temperature and pressure in one-, two-, and three-compo- nent systems? 6, For a sever-component system, determine the number of degrees of freedom that must be specified for the system to exist in the following regions: a. One-phase region b. Two-phase region ‘7. Figure 1-8 shows the phase diagrams of eight mixtures of methane and ethane along with the vapor pressure curves of the two components. Determine: a. Vapor pressure of methane at ~ 160°F . Vapor pressure of ethane at 60°F . Critical pressure and temperature of mixture 7 d. Cricondenbar and cricondentherm of mixture 7 Basic Phase Behavior 35 €. Upper and lower dew-point pressure of mixture 6 at 20°F £. The bubble-point and dew-point pressures of mixture 8 at 60°F 8, Using Figure 1-8, prepare and identify the different phase regions of the pressure-composition diagram (commonly called the F-X diagram) for the following temperatures: a. — 120°F b, 20°F 9. Using Figure 1-8, prepare the temperature-composition diagram (commonly called the FX diagram) for the following pressures: a, 300 psia b, 700 psi . 800 psia __ REFERENCES Clark, N., “It Pays to Know Your Petroleum,” World Oil, March 1953, Vol. 136, pp. 165-172, Gibbs, J. W., The Collected Works of J. Willard Gibbs, Trans. Conn. Acad. Arts Sci, Vol. 1, Yale University Press, New Haven, reprinted 1948; original text published 1876. Lee, B. I. and Kesler, M. G., “A Generalized Thermodynamics Correla- tion Based on Three-parameter Corresponding States,” AIChE Journal, Vol. 21, No. 3, May 1975, pp. 510-527. |. Pitzer, K. $., “The Volumetric and Thermodynamics Properties of Flu- ids,” J. Amer. Chem. Soc., Vol. 77, No. 13, July 1955, pp. 3427-3433, Rackett, H. G., “Equation of State for Saturated Liquids,” J. Chem. Eng. Data, Vol. 15, No. 4, 1970, pp. 514-517. Spencer, F. F. and Danner, R. P, “Prediction of Bubble-point Density of Mixtures,” J. Chem, Eng. Data, Vol. 18, No. 2, 1973, pp. 230-234. Yamada, T. and Gunn, R., “Saturated Liquid Molar Volumes: The Rack- ‘ett Equation,” J. Chem. Eng, Data, Vol. 18, No. 2, 1973, pp. 234-936.2 Pure Component Physical Properties and Characterizing Undefined Petroleum Fractions Many of the physical properties of pure components have been measured and compiled over the years. These properties provide essential information for studying the volumetric behavior and determining the thermodynamic properties of pure components and their mixtures. The most important of these properties are: © Critical pressure, pe ‘Critical temperature, T. © Critical volume, Ve ‘Critical compressibility factor, Z, * Acentrie factor, o # Molecular weight, MW Petroleum engineers are usually interested in the behavior of hydrocarbon mixtures rather than pure components. However, the above characteristic constants of the pure component can be used with the independent state variables such as pressure, temperature, and composition to characterize and define the physical properties and the phase behavior of mixtures. This chapter's primary objective is review. Several of the well-established physical property correlations are presented to illustrate how they can be lused if no experimental data are available on the petroleum fraction. ‘GENERALIZED CORRELATIONS FOR ESTIMATING __ PHYSICAL PROPERTIES OF HYDROCARBON FRACTIO! ‘There are numerous correlations for estimating the physical properties of petroleum fractions, Most of these correlations use the specifi gravity y and the boiling point T, as correlation parameters. Selecting proper values for 36 Pure Component Properties/Characterizing Undefined Fractions 37 the above parameters is very important because slight changes in these pa- rameters can cause significant variations in the predicted results. Several of these correlations are presented below. A ized Correlations // iazi-Daubert Gener Riazi and Daubert (1980) developed a simple two-parameter equation for predicting the physical properties of pure compounds and undefined hydro- carbon mixtures. The proposed generalized empirical equation is based on the use of the normal boiling point and the specific gravity as correlating parameters. The basic equation is: On aThy @) where any physical property normal boiling point, °R specific gravity correlation constants are given in Table 2-1 for each prop- erty Table 2-1 Correlation Constants for Equation 2-1 Deviation % e b c ‘Average Maximum “4.5679 » 10°* aie 1016s 26 118 Jin 242787 Ossses = 03596 1.3 108 papsia 312281109 29125 2s21 at 93 Veto 75214 x 10"? 0206-07685 23 8a ‘The average error for each property is given in Table 2-1. ‘The prediction accuracy is reasonable over the boiling point range of 100- 850°F, Riazi and Daubert (1987), in their development of new correlations for improving the prediction of physical properties of petroleum fractions, con- sidered various factors. These factors were accuracy, simplicity, generality, availability of input parameters, extrapolatability, and finally, comparabil- ity with similar correlations developed in recent years. ‘The authors proposed the following modification of Equation 2-1, which ‘maintains the simplicity of the previous correlation while significantly im- proving its accuraey:38 Hydrocarbon Phase Behavior @ = 2 0} OL EXP (dO; + € © + FO; Oy] where = any physical property e-f = constants for each property Riazi and Daubert stated that @; and ©, can be any two parameters ca- pable of characterizing molecular forces and molecular size of a compound, ‘They identified (T,, 1) and (MW, 7) as appropriate pairs of input parame- ters in the above equation. The authors finally proposed the following two forms of the generalized correlation: Form 1. In this form, the boiling point T, and the specific gravity of the petroleum fraction are used as correlating parameters. © = a Ty EXP [A Ty +07 + FT 7] ) ‘The constants a-f for each property @ are given in Table 2-2. Table 2-2 Correlation Constants for Equation 2-2 ° 5 © aw 581.96 97476 e5i2ra Toth 10.8463 0.81067 0.53691 Be. psia 6.162 x 108 ~ 0.4946 4.0046 Ve, tf 6.238 x 10-* 0.7506 ~ 1.2028 e e 1 643076 x 10-* =o5sse4 1.11056 x 10-° = 5.1747 x 10-* To5eaas 315996 x 10"* = 4.725 x 10-2 = 4p014 3.1998 x 10-9 = 1.4879 x 10-9 = 0.26008 41.095 x 10"* Form 2. In this form, molecular weight MW and specific gravity -y of the component are used as correlating parameters. © = a (MW) 7 EXP [d (MW) + ey + £ (MW) 7] 23) where the correlation constants are given in Table 2-3. In developing and obtaining the coefficients of the above two correla- tions, Riazi and Daubert used data on the properties of 38 pure hydrocar- bons in the carbon number range 1-20, including paraffins, olefins, Pure Component Properties/Characterizing Undefined Fractions 39 Table 2.3, Correlation Constants for Equation 2-3 b e 0.2988 1.0585 es, psia 4.5203 x 10" 018063 1.6015 Ve 18M 1.206 > 10-7 0.20378 ~ 113096 TOR 67887 o.aor673 = 1158262 i a e t 13478 x 10" ~osieat 00 - = 18078 x 10° = 0.3084 00 = 21857 x 10-2 0.5207 2.6012 x 10-2 8.7409 x 10°> 2.984036 = 425288 x 10-2 the naphthenes, and aromatics in the molecular weight range 70-300 a boiling point range 80-650°F. Lin-Chao Generalized Correlation Lin and Chao (1984) correlated the physical properties of hydrocarbon components with the molecular weight, specific gravity, and normal boiling point. The proposed correlation was developed by using perturbation the- ‘ory; it contains 33 numerical constants for each physical property. The phys- ical properties of C, to Cy n-alkanes (n-paraffins) were correlated with the molecular weight. Properties of other hydrocarbons and derivatives were expressed as perturbations from those of the n-paraffins with the boiling point and the specific gravity as the correlating parameters Lin and Chao expressed the physical properties of C, to Coy n-alkanes by the following generalized equation @q = Cy + Cx(MW) + CMW)? + CMW)? + Cy(MW) (2-4) where O, represents either T., In(pJ, Ve, (# *T,), of Ty of a n-alkane. The coefficients C; to Cs are reported in ‘Table 2-4, ‘The correlation produced an average absolute deviation of 0.15% for To, 1.0% for p, excluding methane), 1.2% for w, 0.11% for Ty, and 0.07% for +7, when compared to the physical properties value, as listed in the American Petroleum Institute project 44 table values. Properties of the general hydro- carbons and derivatives were correlated as perturbations of those of n-al kanes according to the following equation:40 Hydrocarbon Phase Behavior “Table 2-4 Costticients for Ox in Equation 2-4 & SEENICs 7. 0A woe 7056062 iio. 6759446 = darore ot ~ 26.21598 2200516 ro 0.65405 {4st00 x 10-2 wen 240.976 Ssowaee Comnnaai & & = aneii6708 Berea «0; ~a100202 2551406 x 10-4 -a7s776 10"* 350737 rracaee x 10> = 9685056 x 10-* = 128:35094 = S702 x 10-* e2i4i4 x 10°? weaseie = .012761604 19,8405 x 10-4 = 2.029.158 ee : = Gy +A, Ay + Az ATy + Ay (Ay)? + Ag (Ay) (ATA) + As (ATL)? + Ag (Ay)? + Ay (Ay)? (ATs) + As (Ay) (ATL)? + Aa (ATL)? (25) with Ay=7-% ATy = Ty ~ (Tie 4 where +y = specific gravity of the substance of interest 4, Ty = boiling point of the substance of interest, °R_“'* ‘1m (Ti)s = specific gravity and boiling point of a hypothetical n-al- kane with a molecular weight (MW) of the substance of interest (given by Equation 2-4), ‘The coefficients A,~As in Equation 2-5 are given by the following expres- (2-6) where values of the coefficients a, and b, are listed in Table 2-5. ‘The proposed correlation can be best explained through the following ex- ample: Example 2-1. Estimate the critical temperature of a petroleum fraction with a boiling point of 731°R, specific gravity of 0.7515 and molecular ‘weight of 127 by using the Lin and Chao method. Solution. : Step 1. Using Equation 2-6, calculate the critical temperature (T.), spe- cific gravity 7,, and boiling point (T,), of a hypothetical n-alkane Pure Component Properties/Characterizing Undefined Fractions 41 Table 2-5 Coefficients tor Equation 2-6 a 8S onis7e 2.008702 x 10° a = 5.68509 ~ 1.844466667 x 10-* 3.48210 a = 219a62 x 10" © — 86.0975 5.000706 x 10* a 75.0863 o.soseatiti1 = 2.05257 x 10 a ~5.3668056 x 10-* ~ 2.777868860 x 10-* 0.2532090009 a 8.908016 x 10* 1.85927 x10? 7.19722 x 10" a = 1.87999 x 108 ~ 018396277778 5.08888 x 10° os 0.20029 1.99582716 x 10-? ~ 0.3390405556 a = 0.851117204 x 10° 0.6541941015 x 10-% 0.5207160494 x 10-2 be = 2131776 = 75087 x 10-8 3.418698 be 5.77384 x 10-2 4.759989993 x 10°? 2.61662 x 10-7 bs 1.992546 x 10? o.saze54 = 4.014316 x 108 be ~ 0.858450 = 2.p97022222 x 10°? 2.06071 be 4.346186667 x 10-+ 2.430015492 x 10-* ~ 2900589333 x 10-? be ~ 367.841 ~ 1.85490 “7.66070 x 10° b 1.32064 o.75seaa8080 x 10-* 5.75141 be = 126440556 x 10-2 ~ 0,9997808642 x 10°F 4.814816867 x 10°? be 21598461358 x 10-7 0.5753412200 x 10-* 05407067001 x 10-* with a molecular weight of 127. These applications produce the following results (1), = 1,067.8°R = 0.1977 (i), = 759.19°R Step 2. Calculate the coefficients A, by using Equation 2-6. A, = 137.09 Ag = L647 Ay = 8,406,714 Ay= 8.5579 As ~ 0.001508 Ay = ~ 7,610.25 A= 9.72928 Ay = 0.03971 ‘Ay = ~ 000005084 Step 3. Solve for . by applying Equation 25. Ay = 0.7515 - 0.7977 = ~ 0.0462 AT, = 731 ~ 759.19 = ~ 28.19 Te = 1,067.8 + 137.09( ~ 0.0462) + 1.6477( ~ 28.19) + 3,406.714( — 0.0462)? — 8.5579( — 0.0462)( ~ 28.19) + 0.001508 — 29.19)? ~ 7,610.25( ~ 0.0462) + 9.72228 ( ~ 0.0462)? ( ~ 28.19) + 0.03971 ( ~ 0.0462)( - 28.19) ~ 0,00005084{ - 28.19}? = 1,012.2°R42 Hydrocarbon Phase Behavior ‘OTHER METHODS OF ESTIMATING PHYSICAL PROPERTIES OF PETROLEUM FRACTIONS Pure Component Properties/Cheracterizing Undefined Fractions 43 5 1 caveres Correlations _// Cavett (1962) proposed correlations for estimating the critical pressure and temperature of hydrocarbon fractions. The correlations have received a ‘wide acceptance in the petroleum industry due to their reliability in extra- polating at conditions beyond those of the data used in developing the corre- lations. The proposed correlations were expressed analytically as functions of the normal boiling point Ty and API gravity. Cavett proposed the follow- ing expressions for estimating the critical temperature and pressure of petro- Jeu fractions: Critical Temperature Ty = ay + ay Ty + ay TH + ay (APD(TS) + aT)? + as(API)(T)* + as (APL)? (Th)? (27) Critical Pressure Log (p.) = by + by (To) + by (Ta)? + by (API) (Ta) + by (To)? + bs (API) (T.)® + be (API)? (Ix) + br (API? (TL)? (2-8) where T, = critical temperature, °R po = critical pressure, psia ‘Ty = normal boiling point, °F API = API gravity of the fraction ‘The coefficients of Equations 2-7 and 2-8 are tabulated in Table 2-6. Cavett presented these correlations without reference to the type and source(s) of data used for their development. y ns U/ Kesler-Lee Cori Kesler and Lee (1976) proposed a set of equations to estimate the critical temperature, critical pressure, acentric factor, and molecular weight of pe- troleum fractions. The equations, as expressed below, use specific gravity and boiling point in °R as input parameters. Table 2-6 Coetticients of Equations 2-7 and 2-8 1 a b ° 768.07121 2aps0406 1 1.7183693, (094120109 x 10" 2 = 0.0010834003 = 090474749 x 10-* 3 = 0.0089212579 = 020876110 x 10-* 4 0.99890588 x 10-* 0.16184108 x 10" 5 (053094920 x 10-* (0.11047899 x 10-? 6 0.32711600 x 10-7 = 0148271598 x 10-7 7 = (019949619 x 10°" Caitical Pressure aaa (tes) In (p) = 8.3634 — 0.0566/y ~ (0.24944 + 2.9898/+y + 0.11857/4*)10-° Ty, + (1.4685 + 3.648/y + 0.47297/2)10-7 T,? - (0.42019 + 1.6977/y")10-"° TY 9 Critical Temperature Ty = 41.7 + SILL y + (0.444 + 0.1174) Th + (0.4669 - 3.26238) 10°/T,, (2-10), Molecular Weight MW = ~ 12,272.6 + 9,486.4 + (4.6525 ~ 3.32877) Ty + (1 = 0.77084y ~ 0.020584)(1.3437 ~ 720.79/T,)10"7Ts ++ (1 = 0.80882y + 0.022264)(1.8828 ~ 181.98/T,)104/T? (eu) The above equation was obtained by regression analysis using the avail- able data on molecular weights ranging from 60 to 650. ‘Acentrie Factor. Defining the Watson characterization factor K and the re- duced boiling point © by the following relationships x- Y44 Hydrocarbon Phase Behavior where Ty = boiling point, °R Kesler and Lee proposed the following two expressions for calculating the acentric factor For @>0.8: = ~ 7.904 + 0.1352K ~ 0.007465K? + 8.359 8 + (1.408 ~ 0.01063K)/9 (2-12) For 0<0.8: —In(p./14.696) — 5.92714 + 6.09648/0 + 1.28862 In(@) ~ 0.169347 08 15.251 15.6875/0 ~ 13.4721 In(O) + 0. (213) where pe = critical pressure, psia T, = critical temperature, °R Ty = boiling point, °R @ = acentric factor MW = molecular weight +7 = specific gravity Kesler and Lee stated that Equations 2-9 and 2-10 give values for p, and ‘T, that are nearly identical with those from the API Data Book up to a boil- {ng point of 1,200°F. Modifications were introduced to extend the correla tions beyond the boiling point limit of 1,200°F. These extensions (extrapola- tions) were achieved by ensuring that the critical pressure approaches the atmospheric pressure as the boiling point approaches critical temperature. Hf Wixm-Siva-Daubert Correlations _// Sim and Daubert (1980) concluded that the Winn (1957) nomograph is the mast accurate method for characterizing petroleum fractions. For this reason, Sim and Daubert represented the critical pressure, critical tempera- ture, and molecular weight of the Winn nomograph analytically by the fol- lowing equations: 48242 x 10? Ty 25177 44865 Ty = EXP [3,9094718 122" yoo — MW se) 15) “4d50476 x 10-9 TEI"! YS (@16> Pure Component Properties/Characterizing Undefined Fractions 45 where p. = critical pressure, psia ‘T, = critical temperature, °R T,, = boiling point, °R ‘Watansiri-Owens-Starling Correlations Watansiri, et. al. (1985) developed a set of correlations to estimate the critical properties and acentric factor of coal compounds and other hydro- carbons and their derivatives. The proposed correlations express the charac- terization parameters as functions of the normal boiling point, specific grav- ity, and molecular weight. These relationships have the following forms: Critical Temperature In(T,) = ~ 0.0650504 ~ 0.005217 7, + 0.03095 In(MW) + 1.11067 In(Ty) + MW [0.078154 ‘ = 0.061061 +! ~ 0.016943 4} (217) where T, = critical temperature, °R Critical Volume In(V,) = 76.313887 - 129.8038 y + 63.1750 4 ~ 13.175 + 1.10108 In(MW) + 42.1958 In(q) (218) where V, = critical volume, ft?lb-mole Critical Pressure In(p.) = 6.6418853 + 0.01617283 (JV) ~ 8.712(MWIT) ~ 0.08843889(T,/MW) (2-19) critical pressure, psia where pe Acentric Factor wo = [5.12916667 x 10-* Ty, + 0.281826067(TYMW) + 382.904/MW + 0.074691 x 10-%(Ty/y)? ~ 0.12087778 x 10-4(T)(MW) + 0.001261(y)(MW) + 0.1265x10~4 (MW)? + 0.2016x10-# ()(MW)! — 66.29959(T,)!"/MW = 0.00255452 1.29/41 TH(9 MW) (2-20) ‘The proposed correlations produce an average absolute relative deviation of 1.2% for Te, 3.8% for Ve, 5.2% for pe, and 11.8% for a.46 Hydrocarbon Phase Behavior . YL eamisters Corration Edmister (1958) proposed a correlation for estimating the acentric factor of pure fluids and petroleum fractions. The equation, widely used in the petroleum industry, requires boiling point, critical temperature, and eritical pressure. The proposed expression is given by the following relationship: 3[Log (p/14.70) _ ~ oe eat (2-21) where w = acentric factor pe critical pressure, psia Ty = critical temperature, °R ‘Ty = normal boiling point, °R If the acentrie factor is available from another correlation, the Edmister equation can be rearranged to solve for any of the three other properties (providing the other two are known). (CRITICAL COMPRESSIBILITY FACTORS ‘The critical compressibility factor is defined as the component compress- ibility factor calculated at its critical point, This property can be conve- niently computed by the real gas equation-of-state at the critical point, or pe Ve he (2-20 RT. aeeet where R = universal gas constant, 10.73 psia.ft'/Ib-mole.°R Ve = critical volume, ft'lb-mole If the critical volume Ve is given in ft/lb, Equation 2-22 is written as _VeMW ccnawT where MW = molecular weight V.. = critical volume, ft/lb ‘The accuracy of Equation 2-22 depends on the accuracy of the valies of p., TT, and V,. Table 2-7 presents a summary of the critical compressibility est- mation methods. ‘Pure Component Properties/Characterizing Undefined Fractions 47 ‘Table 2-7 Critical Compressibility Estimation Methods Method z Equation No. Haugen 169m 286 + Bat) 223 FRoid-Prausnitz- ‘Sherwood 1977 Z=0.291 0.080 224 Salome ota 1985 Z = 0201 - 0.800 0.018 uF 225 Nath 1985 = 02918 - 0.0928 w 226 ‘Ware oi he aoeniia factor Example 2-3. Estimate the critical properties, molecular weight, and centric factor of a petroleum fraction with a boiling point of 198°F and specific gravity of 0.7365, by using the following methods: Riazi-Daubert (Equation 2-1) Riazi-Daubert (Equation 2-2) Cavett's Kesler-Lee Winn-Sim-Daubert Watansiri-Owens-Starling pusepr Solution. 1, Riazi-Daubert: Equation 2-1 © MW = 4.5673 x 10-* (858)? 2 (0.7365) "101 = 96.4 ©, = 24.2787 (658)°5 (0.7365) = 990.67°R fp. = 3.12281 x 10° (658)-*525 (0.7365) = 466.9 psia 2 V. = 7.5214 x 10-9 (658)°% (0.7365) 97% = 0.06227 ft/lb # Solve for Z_ by applying the above calculated properties in Equation 06227)(96.4) (0.73) (o90.67) —~ 26355 © Solve for w by applying Equation 2-21: = Sillog (466.9714.7)] 7 [(@00.677658) —1) 17° a 2, Riazi-Daubert: Equation 2-2 Applying Equation 2-2 and using the appropriate constants yields:48 Hydrocarbon Phase Behavior * MW = 96.911 eT, = 986.7 °R ep, = 465.83 psia ©V. = 0.06257 feb « Solve for the acentrie factor and the critical compressibility factor by applying Equations 2-21 and 2-22, respectively: ¢u = 0.2877 © Z, = 0.2668 3, Cavett’s Correlation: « Solve for T. by applying Equation 2-7: T, = 978.1 °R © Caleulate p. with Equation 2-8: Pe = 466.1 psia © Solve for the acentric factor by applying the Edmister correlation, Equation 2-21: =1=0.3147 © Compute the critical compressibility by using Equation 2-25; Z, = 0.291 ~ (0.08)(0.3147) ~ 0.016 (0.8147)? = 0.2642 ‘* Estimate V, from Equation 2-22: Z_R Tz _, (0.2642)(10.731)(980) = 5.9495 f/lb-mole Pe 466.1 Assume MW = 96 y, = 5:95 _ 9.06197 feb 4. Kesler-Lee * Calculate p, from Equation 20 pe = 470 psta * Solve for T. by using Equation 2-10: 1. = 990 °R : 5 * Calculate the molecular weight MW by using Equation 2-11 MW = 98.7 Pure Component Properties/Characterizing Undefined Fractions 49 ' * Compute the Watson characterization factor K and the parameter 0: (658)! | ~ ocraes “118 658 0 = 88 ap 70-67 © Solve for acentrie factor by applying Equation 2-13: 0 = 0.306 ‘ Estimate for the critical gas compressibility Z, by using Equation 2-26: Z, = 0.9918 — (0,0928)(0.306) = 0.2634 © Solve for V. by applying Equation 2-22: Ze RT, (0.2634)(10.73)(980) peMW — (470)(08.7) = 0.0597 Ibift? 5. Winn-Sim-Daubert: ‘* Estimate pe from Equation 2-16: Po = 478.6 psi * Solve for T. by applying Equation 2-15: T. = 979.2°R © Calculate MW from Equation 2-16: MW = 95.93 © Solve for the acentric factor from Equation 2-21: @ = 0.3280 * Solve for Z, by applying Equation 2-24: Z, = 0.291 ~ (0.08)(0.3980) = 0.2648 © Calculate the critical volume V. from Equation 2-22: __ (0.2848)(10.731)(979.2) oR EER 7 00080 feb 6, Watansiri-Owens-Star © Because Equations 2-17 through 2-19 require the molecular weight, as- sume MW = 9650 Hydrocarbon Phase Behavior ‘© Calculate T, from Equation 2-17: T, = 980.0 °R ‘¢ Determine the critical volume from Equation 2-18 to give ‘V, = 0.06548 fe°/Ib, «@ Solve for the critical pressure of the fraction by applying Equation 2-19 to produce P, = 426.5 psia ‘© Calculate the acentrie factor from Equation 2-20 to give = 0.2282 ‘© Compute the critical compressibility factor by applying Equation 2-26 Z, = 0.27112 ‘Table 2-8 summarizes the results of Example 2-3, Table 2-8 ‘Summary of the Calculated Results of Example 2-3 v % 6b. Method oA ia__mole SMW z FiaziDaubert No.1 99067 4659 0.06227 964 0.2731 0.26065 RiaziOaubert No.2 986.7 465.83 0.06257 96.911 02877 0.668 Cavett g7e1 466.1 0.06197 ——O3147 0.2642 KeslorLee 980 © 479.«=0.0597 987 0.3060 0.2834 Winn 979.2 478.8 0.06059 96.99 (0.9280 0.2648 Watansi 900° 4265 006548 = — aaa? zTII2 (CHARACTERIZING HYDROCARBON HEAVY FRACTIONS ‘Nearly all naturally occurring hydrocarbon systems contain a quantity of heavy fractions that are not well defined and are not mixtures of discretely {identified components. These heavy fractions are often lumped together and identified as the plus fraction, e.g., Cr. fraction. ‘A proper description of the physical properties of the plus fractions and other undefined petroleum fractions in hydrocarbon mixtures is essential in performing reliable phase behavior calculations and compositional model- ing studies. Frequently a distillation analysis or a chromatographic analysis {s available for this undefined fraction, Other physical properties such as Pure Component Properties/Characterizing Undefined Fractions 51 molecular weight and specific gravity may also be measured for the entire fraction or various cuts of it. ‘To use any of the thermodynamic property-prediction models, e.g, equa- tion of state, to predict the phase and volumetric behavior of complex hy- drocarbon mixtures, one must be able to provide the acentric factor, critical temperature, and critical pressure for both the defined and undefined (heavy) fractions in the mixture. The problem of how to adequately charac- terize these undefined plus fractions in terms of their critical properties and acentric factors has been long recognized in the petroleum industry. Whit- son (1984) presented an excellent documentation on the influence of various heptanes-plus (C;) characterization schemes on predicting the volumetric behavior of hydrocarbon mixtures by equations-of-state Numerous characterization procedures have been proposed over the years. Some of the most widely accepted and used procedures are reviewed below. METHODS BASED. THE “PNA” DETERMINATION ‘The vast number of hydrocarbon compounds making up naturally occur- ring crude oil have been grouped chemically into several series of com- pounds. Each series consists of those compounds similar in their molecular make-up and characteristics. Within a given series, there exist compounds ranging from extremely light, or chemically simple, to heavy, or chemically complex. In general, it is assumed that the heavy (undefined) hydrocarbon fractions are composed of three hydrocarbon groups, namely: «© Paraffins (P) © Naphthenes (N) Aromatics (A) ‘The PNA content of the plus fraction of the undefined hydrocarbon frac- tion can be estimated experimentally from a distillation analysis and/or a chromatographie analysis. Both types of analysis provide information valu- able for use in characterizing the plus fractions. In the distillation process, the hydrocarbon plus fraction is subjected to standardized analytical distillation, first at atmospheric pressure, and then ina vacuum at a pressure of 40 mm Hg. Usually the temperature is taken when the first droplet distills over. Ten fractions (cuts) are then distilled off, the first one at 50°C and each successive one with a boiling range of 25°C. For each distillation cut, the volume, specific gravity, and molecular weight, among other measurements, are determined. Cuts obtained in this manner are identified by the boiling-point ranges in which they were collected,52 Hydrocarbon Phase Behavior Generally, there are five different methods of defining the normal boiling point for petroleum fractions. These are: 1, Volume Average Boiling Foint (VABP), which is defined mathemati- cally by the following expression: VABP = )) vi Tux (2-27) T where T= boiling point of the distillation cut i, °R vi = volume fraction of the distillation cut i 2. Weight Average Boiling Point (WABP), defined by the following ex- pression: WABP = JO w, Tw (2-28) where w, = weight fraction of the distillation cut i 3. Molar Average Bolling Point (MABP), given by the following relation- ship: MABP = J) x; Ty (2-29) 7 where x; = mole fraction of the distillation cut i 4, Cubie Average Boiling Point (CABP) which is defined as Pp CABP =|) x: nerf (2-0) 5. Mean Average Boiling Point (MeABP): MeABP = Mapp ons? (231) As indicated by Edmister and Lee (1984), these five expressions for caleu- lating normal boiling points result in values that do not differ significantly from one another for narrow boiling petroleum fractions. Figure 2-1 shows a typical graphical presentation of the molecular weight, specific gravity, and the true boiling point TBP as a function of the volume fraction of liquid vaporized. It should be pointed out that when a single boiling point is given for a plus fraction, itis given as its Volume Aver- age Boiling Point. ‘Bergman et al. (1977) outlined the chromatographic analysis’ procedure by which distillation cuts are characterized by the density and molecular ‘weight as well as Weight Average Boiling Point, Pure Component Properties/Characterizing Undefined Fractions 53 Figure 2-1. Properties of a crude oil fraction. All three parameters (i.e, molecular weight, specific gravity, and VABP/ WABP) are employed, as discussed below, to estimate the PNA content of the heavy hydrocarbon fraction which in turn is used to predict the critical properties and acentric factor of the fraction. Hopke and Lin (1974), Erbar (1977), Bergman et al. (1977), and Robinson and Peng (1978) have used the PNA concept to characterize the undefined hydrocarbon fractions. As a rep- resentative of this characterization approach, the Robinson-Peng method and the Bergman method are discussed below. Robinson-Peng Method Robinson and Peng (1978) proposed a detailed procedure for characteriz- iiig heavy hydrocarbon fractions. The procedure is summarized in the fol- lowing steps: Step 1. Calculate the PNA content (Xp, Xyy X4) of the undefined fraction by solving the following three rigorously defined equations: D Mel (2-32)54 Hydrocarbon Phase Behavior (MW, Te Xi = (MW)(WABF) (2-33) Ra Dp mw. x)-Mw 34) iA where Xp = mole fraction of the paraffinic group in the undefined frac- tion Xw = mole fraction of the naphthenic group in the undefined fraction Xq = mole fraction of the aromatic group in the undefined frac- tion WABP = weight average boiling point of the undefined fraction, °R MW = molecular weight of the undefined fraction (MW) = average molecular weight of each eut, ie, PNA, (fj); = boiling point of each cut, °R Equations 2-32 through 2-34 can be written in a matrix form as follows: 1 1 1 Xe] a [MW = TyJe (MW + Tyly [MW Tula] |X) = |MW + WABP ) (2-35) (MW [MWk (MW, [Xa] MW Robinson and Peng pointed out that itis possible to obtain negative values for the PNA contents. To prevent these negative values, the authors imposed the following constraints: 0=Xps0.90 Xyz0 X20 ‘To solve Equation 2-35 for the PNA content requires the weight average boiling point and molecular weight for the cut of the undefined hydrocar- bbon fraction, If the experimental values of these cuts are not available, the following correlations proposed by Robinson and Peng can be used: Determination of (Ti (Toss and (Toa. Paraffinie group Ln(Ty) = Ln(1.8) + Y) [a(n - 6)'-!] (2-36) ‘© Naphthenic group Ln(T}) = La(l.8) + J) {a(n - 7)'-4] (237) Pure Component Properties/Characterizing Undefined Fractions 55 6 © Aromatic group Ln(Ts) = Ln(1.8) + Y) [a(a - 7) (2-38) where n= number of carbon atoms in the undefined hydrocarbon fraction 1a; = coefficients of the equations and are given in Table 2-9 Table 2-9 Coefficients a, in Equations 2-36 through 2-38, Parattin ‘Naphthene ~~ Aromatic Cootficlent Pp N a‘ 5.89451690 5.8579802 5.86747600, (0,84909035 x 10°? 0,79805995 x 10-' _0,80496947 x 10 a = 0152695420 x 10-7 - 0.49096101 x 10-2 — 0.47196506 x 10-* a (0.21252008 x 10-2 _0.14783123 x 10"? 0.18299965 x 10°? a = 0444999963 x 10-* — 0.27095216 x 10° — 0.98927290 x 10-* % 0137285965 x 10-7 __0.19907794 x 10. 0132550876 x 10-7 Determination of (MW) (MW)y, and (MW). «© Paraffinie group (MW)p = 14.026 n + 2.016 (2-39) ‘© Naphthenic group (MW)y = 14.026 n ~ 14.026 (2-40) ‘* Aromatic group (MW), = 14.026 n~ 20.074 (2-41) Step 2. Having obtained the PNA content of the undefined hydrocarbon fraction, as outlines in Step 1, calculate the critical pressure of the fraction by applying the following expression: Pe = Xe(Pode + Xn(Pedn + XaPoda (2-42) where P, = critical pressure of the heavy hydrocarbon fraction, psia ‘The critical pressure for each cut of the heavy fraction is calculated ac- cording to the following equations: _ 206.126006n 136 © Paraffinic group (P,)p = aeeeErsEareetttt (2-43) :206,126096n ~ 206.126096 (@.28in- 0.137)" ‘© Naphthenie group (P.)y = (2-44)36 Hydrocarbon Phase Behavior 206, 126096n ~ 295.007504 «Aromatic group (Pa = Sone — 5 aa (45) step 3. Calculate the acentri factor of each cut ofthe undefined fraction by using the following expressions: ¢ Paraffinic group (u)p = 0.0432n + 0.0457 (2-46) ‘© Naphthenie group (a)y = 0.0432n ~ 0.0880 (47) ‘© Aromatic group (1), = 0.0445n ~ 0.0995 (2-48) Step 4, Calculate the critical temperature of the fraction under eonsider- ation by using the following relationship: Te Xp (Top + Xn (Ton + Xa (Toda (2-49) where T, = critical temperature of the fraction, °R ‘The critical temperatures of the various cuts of the undefined fractions are calculated from the following expressions: (P,)p] — 3.501952 5 «Pate pou (Te [1M BE] ey .)~ 3.01850 « phe op =f + RAP 818 SLog{ (P.)4] ~ 3.501952 ‘Aromatic group (Te)a = Si(To)a bh eal | (2-52) where the correction factors $ and S; are defined by the following expres- § = 0.996704 + 0.000431550 8, = 0.99627245 + 0.00043155n Step 5. Calculate the acentric factor of the heavy hydrocarbon fraction by using the Edmister correlation (Equation 2-21) to give 3 tLog Pe14.7)) _ 59 7avT =H Pure Component Properties/Characterizing Undefined Fractions 37 where @ = aventric factor of the heavy fraction P, = critical pressure of the heavy fraction, psia TT, = critical temperature of the heavy fraction, °R Ty = average-weight boiling point, °R Example 2-4. Calculate the critical pressure, the critical temperature, and acentric factor of an undefined hydrocarbon fraction with a measured mo- lecular weight of 94 and a weight average boiling point of 655°R. The num- ber of carbon atoms of the component is 7 Solution. Step 1. Calculate the boiling point of each cut by applying Equations 2- 36 through 238 to give (To)p = 666.58°R, (Ty)x = 630.0°R, (Ty), = 635.85°R Step 2. Compute the molecular weight of various cuts by using Equa- tions 2-39 through 2-41 to yield (MW)p = 100.198, (MW)x = 84.156, (MW), = 78.18 Step 3. Solve Equation 2-35 for Xp, Xy, and Xa, to give Xp = 0.6313, Xy = 0.3262, Xq = 0.0495 Step 4. Calculate the critical pressure of each cut in the undefined frac- tion by applying Equations 2-43 and 2-45. (Pp = 395.7 psia, (Pv = 586.61 psia, (P.)q = 718.46 psia Step 5. Calculate the eritical pressure of the heavy fraction from Equa- tion 2-42 to give P, = 471.7 psia Step 6. Compute the acentric factor for each cut in the fraction by using Equations 2-46 through 2-48 to yield (o)p = 0.3481, (a) = 0.2144, (0) Step 7. Solve for (T.)p, (Tx, and (T.)q by using Equations 2-50 through 9-52 to give (Tp = 969.4°R, (T)y = 947.3°R, (Tea = 1014.9°R Step 8. Solve for (T) of the undefined fraction from Equation 2-49. T. = 964.1°R o.21258 Hydrocarbon Phase Behavior stop 9. Calculate the acentric factor from Equation 259, to give wo = 0.3680 Bergman's Method. Bergman et al. (1977) proposed a detailed procedure for characterizing the undefined hydrocarbon fractions based on calculating the PNA content Sf ithe fraction under consideration. The proposed procedure was originated from analyzing extensive experimental data on lean gases and condensate systems. The authors, in developing the correlation, assured that the paraf- finie, naphthenic, and aromatic groups have the same boiling point. The ‘computational procedure is summarized in the following steps: Step 1. Estimate the weight fraction of the aromatic content in the unde- fined fraction by applying the following expression: owe 847 Ky Fe 04 (rA~ ke) (254) where wa = weight fraction of aromatics K.. = Watson characterization factor, defined mathematically by the following expression: Ku = (Th)y (2-55) where _y = specific gravity of the undefined fraction T, = weight average boiling point, °R Bergman etal. imposed the following constraint on the aromatic content: 0.08. w, 0.35 Step 2. With the estimate of the aromatic content, the weight fractions of the paraffinic and naphthenie euts are calculated by sotving the following system of linear equations: wy te a (256) vey we es) © 7 where W> = weight fraction of the paraffin cut eight fraction of the naphthene cut Pure Component Properties/Characterizing Undefined Fractions 50 -y = specific gravity of the undefined fraction “resvwcia = specific gravity of the three groups at the weight aver- age boiling point of the undefined fraction. These sgravities are calculated from the following relatfon- ships: ‘yp = 0.582486 + 0.00069481 (Ty ~ 460) = 0.7572818(10-9\(T, ~ 460) + 0.3207736(10-8)(T, - 460)? (2-58) ‘yw = 0.694208 + 0.0004909267(T, ~ 460) ~ 0.659746(10-*)(T, ~ 460)* + 0.390966(10-*)(T, ~ 460)* (2-59) ‘yn = 0.916103 ~ 0.000250418(T ~ 460) + 0.357967(10"8)(Tr, - 460)? = 0.166318(10-*)(T,, - 460)? (2-60) ‘A minimum paraifin content of 0.20 was set by Bergman et al. Tb insure that this minimum value is met, the estimated aromatic content that results in negative values of wp is increased in increments of 0.03 up to a maximum of 15 times until the paraffin content exceeds 0.20. They pointed out that this procedure gives reasonable results for fractions up to Cis. Step 3. Calculate the critical temperature, the critical pressure, and acentric factor of each cut from the following expressions: © Paraffins: (op = 275.23 + 1.2061(T, ~ 460) ~ 0,00032984(T, ~ 400)? (2-61) (Pp = 573.011 ~ 1.13707(7, ~ 460) + 0.00131625(T, ~ 460) = 0,85103(10-9)(Ty, ~ 460)* (2-62) (o)p = 0.14 + 0.0009(T, ~ 460) + 0.283(10-9)(T, ~ 460)? (2-63) © Naphthenes: (Tox = 156.8906 + 2.6077(Ty ~ 460) ~ 0.003801(7, ~ 460)? + 0.2544(10-5)(Ty = 460)° (2-64) (P)y = 726.414 ~ 1.9275(T, ~ 460) + 0.9846(10-8)(T,, ~ 460)? = 0.45169(10-9)(T,, ~ 460)? (2-65) (on = (ae ~ 0.075 (2-66)60 Hydrocarbon Phase Behavior Bergman et al. assigned the following special values of the acentric factor to the Cs, Co, and Cp naphthenes. (wy = 0.26 (oy = 0.27 Gyo: (wy = 0.35 © Aromatics: (T)q = 289.535 + 1.7017(T, ~ 460) ~ 0.0015843(T, - 460)? + 0.82358(10"9(T, — 460)? (2-67) (Bq = 1.184.514 - 9.44681(T, ~ 460) + 0.0045512¢7, ~ 460)? = 0.23416(10-5)(T, — 460) (2-68) a= Wp 0.1 (2-69) Step 4. Calculate the critical pressure, the critical temperature, and the ‘centric factor of the undefined fraction from the following relationships: P. = Xe(P.)p + Xw(Paln + XalPeda (2-70) Te = Xpl(Tee + Xn(Tw + XaTea (2-71) = Xp(u)p + Xnle)y + X (2-72) ‘Whitson (1984) suggested that the Peng-Robinson and the Bergman PNA ‘methods are not recommended for characterizing reservoir fluids containing fractions heavier than Crp. ‘OTHER METHODS OF CHARACTERIZING ‘THE HYDROCARBON HEAVY FRACTIONS Rowe's Characterization Method Rowe (1978) proposed a set of correlations for estimating the normal boil- ing point, the critical temperature, and the critical pressure of the heptanes- pplus fraction, i.e., C;,. The prediction of the C;, properties is based on the assumption that the “lumped” fraction behaves as a normal pargifin hydro- carbon, Rowe used the number of carbon atoms n as the only correlating parameter. He proposed the following set of formulas for characterizing the Coy fracti Pure Component Properties/Characterizing Undefined Fractions 61 ‘ Caleulation of the eritical temperature: (Toor, = 1.8 [961 — 10") (273) where (Tor, = critical temperature of C;., °R 4 = coefficient of the equation and is given by the follow- ing expression a = 2.95597 ~ 0.090597n2° where 1 is the number of carbon atoms and is calculated from the mo- lecular weight of the C;, fraction by the following relationship: ni» (MWe, ~ 20)14 (en where MWo,, is the molecular weight of the heptanes-plus fraction, * Calculation of the critical pressure: Pde;. = 10° /(T Joy, (2-75) with Y = ~ 0.0137726826 n + 0.6801481651 where (Poy, is the critical pressure of Cr. in psi. © Calculation ofthe cstioal temperature: x!" Vic ~ (Trop, = 0.0004347(T yo, , + 265 (2-76) where (Ty)o;, is the normal boiling point in °R. Mathews, Roland, and Katz (1942) presented graphical correlations for determining the critical temperature and pressure of the heptanes-plus frac- tion. Standing (1977) expressed these graphical correlations more conve- niently in mathematical forms as follows: (Ter, = 608 + 364 Log [(MW)c, + [2,450 Log (MW)c;, -7112] 8,800] Log (aor. em Poder. 1,188 ~ 431 Log [(MW)o, ~ 61.1] + [2,319 ~ 852 Log [(MW)c;, = 53.7]} ((r)cr, - 0-8) (2-78) where (MW, and (no, ate the moleular weight and specific gravity of the Cr, 562 Hydrocarbon Phase Behavior ‘Example 2-5. If the molecular weight and specifie gravity ofthe heptanes- plus fraction are 216 and 0.8605, respectively, calculate the critical tempera- ture and pressure by using 1. Rowe's Correlations 2. Standing’s Correlations Solution. 1. Solution by using Rowe's Correlation Step 1. Calculate the number of carbon atoms of Cr, from Equation 274 to give n= 15.29 Step 2. Solve for the critical temperature from Equation 2-75 to yield (Te, = 1,279.8°R Step 3. Solve for the eritical pressure from Equation 2-75 to give (Poor, = 280-4 psia 2, Solution by using Standing’s Correlations Step 1. Solve for the critical temperature by using Equation 2-77 to give (Tey, = 1,269.3°R Step 2. Calculate the eritical pressure from Equation 2-78 to yield Poy, = 270 psia Katz-Firoozabadi Method Katz and Firoozabadi (1978) presented a generalized set of physical prop- erties for the petroleum fractions Cy through Cys. The tabulated properties inelude the average boiling point, specific gravity, and molecular weight. ‘The authors’ proposed tabulated properties are based on the analysis of the physical properties of 26 condensates and naturally occurring liquid hydro- carbons. Whitson (1983) found inconsistency in the Katz and Firoozabadi tabu- lated molecular weight data after analyzing and comparing these data with sources from which they were developed. Whitson pointed out that this in- consistency was found in the hydrocarbon fractions Caz through Cys. Pure Component Properties/Characterizing Undefined Fractions 63 Whitson modified the original tabulated physical properties to make their use more consistent, ‘The modification was accomplished by employing the Riazi and Daubert correlation form (Equation 2-1) to extrapolate the molec- ular weight data from Czz to Cis. The coefficients a, b, and ¢ of Equation 21 were recalculated by using a nonlinear regression model to fit the molec- ular weight data of Ce through Cac. The equation was then used to calculate molecular weights of Cas through C,s. The author also calculated the crti- cal properties and acentrie factors of Cy through Cg in terms of their boiling point, specific gravity, and the modified values of the molecular weight. ‘These generalized properties are given in Table 2-10, ‘Ahmed (1985) correlated Katz-Firoozabadi-Whitson tabulated physical properties with the number of carbon atoms of the fraction by using a re- gression model. The generalized equation has the following form: © =a + ayn tag rt + ayn? + (ayn) (279) where © = any physical property ‘n= number of earbon atoms, ie., 6,7, . . 45 a,~as = coefficients of the equations and are given in Table 2-11. Riazi-Daubert Method Riazi and Daubert (1980 and 1987) suggested that their proposed correla- tions for calculating the physical properties of pure petroleum fractions (Equations 2-1 through 2-3) can be used to predict the physical properties of the undefined pertroleum fractions. The authors proposed that the follow- boiling points be used for estimating the physical properties. ing “averag Physical Property ° Average Boiling Point rs Molal Average Boiling Point (MABP) Pe, MW, 7, Ve___Mean Average Boiling Point (MeABP) Cavett’s Method With the appropriate boiling point, Cavett's correlations, as given by Equations 2-7 and 2-8, can be used to predict the critical temperature and the critical pressure of the heavy hydrocarbon fraction, Cavett proposed that the molal average boiling point (MABP) is used in Equation 2-7 to cal- culate T, and the mean average boiling point (MeABP) is used in Equation 2-8 for calculating P,.8 vogmnag ebereny YEH = GY FO 6600 e-O1XeseLiOy _.-01 x vORuNSL x eeieouez «Ox esveees aM *A 2 wo coo 01 x aoeerar'2 x goverir'e 67612980 + § (ee art ‘0% 0586999" 1120008 +0129908'0 ~ 3 0 auto Ot x 1890820'L £9212060 ~ exzszi'05 e086 re E es we 996820 eozzzs2t — suzes S12 FO Seh0 oh xeussz0et ~ ‘65009520 - ascizriy wes si8 3 set avo Susze ear 720610%6'0 ~ suche et — 3 4% 7 . 7 3 soa sawn ym € 61:2 vonenb Jo squ2I>1H809 5 Lee age 8 3 3 waht “any 0 neuig wneanag jo ho0g a fa yma 0 HEL — 8 creoo es oreo #0 So eese00 es ee60 "9 = 89 ese00 12s oreo £9 = 70 ezse00 zs ve60 ®) 2 0 aesao0 203 exe0 "0 2 > is800 ser 1680 °9 & 2 lissoo ver 660 #9 eer) 903000 sur 2260 * Eg 2 — sersco v9 5260 “9 => esra0.0 or 260 %9 = 0680.0 ovr oso 9 9 6ara00 oy uso "9 = oae00 ser siso %9 *) sure suv zis0 = tsa0 ent cos “9 60 es $050 * si60 rs 2050 #5 vea0 sat ero = e980 63 960 rd yao eu 2690 = 6.0 ua sero = exo za $290 "0 e920 eat e009 &% ec es euwo "= aco 002 us0 = ose0 1 9980 "9 zo90 we 180 290 zzz ose0 e190 ore 890 zo Wve oreo oss0 ssz 9ea0 5 sso oz sao 3 vero 992 sig0 2 vs¥0 zoe voro & siro see exo 2 sec ise zaxo 2 ereo cae e270 = zco or ero g ovo esr a0 g osz0 say ___ 0680 8 ° ‘exse) * = i] 6 soniedosd 1eotsfyg paziiesou[ on-z organ,66 Hydrocarbon Phase Behavior Willman-Teja Method Willman and Teja (1987) proposed correlations for determining the eriti- cal pressure and critical temperature ofthe n-alkane homologous series. The suthors used the normal boiling point and the number of carbon atoms of the n-alkane as a correlating parameter. The applicability of the Willman and Teja proposed correlations can be extended to predict the eritical tem- perature and pressure of the undefined petroleum fraction by recalculating the exponeats of the original expressions. These exponents were recalculated by using # nonlinear regression model to best match the critical properties data of Berman et al. (1977) and Whitson (1980). The empirical formulas are given by Te = Ty [1 + (1.25127 + 0.197242n)-08800] (2-80) P, = [399.0416805 + 1,184.157759n][0.873159 “+ 0.54985) mt 281) where n= number of carbon atoms ‘Ty, = average boiling point of the undefined fraction, °R Itshould be noted that the Edmister acentrie factor correlation (Equation 291) and the Kesler and Lee correlations (Equations 2-9 through 2-13) can also be used to characterize the undefined hydrocarbon fraction from their average boiling point and specific gravity. Example 2-6. Calculate the eritical properties and the acentrie factor of C), with a measured molecular weight of 198.71 and specifie gravity of 0.8527. Employ the following methods: 1. Rowe's correlation 2. Standing's correlation 3. Riazi-Daubert correlation Solution. 1. Solution by using Rowe's Correlation. Step 1. Calculate the number of carbon atoms of the fraction n= 14,0507 Step 2. Determine T. from Equation Te = 1,247.06°R 113 to give ‘Pure Component Properties/Characteriaing Undefined Fractions 67 Step 3. Compute P, from Equation 2-75 to yield P, = 245.89 psia Step 4. Determine T, by applying Equation 2-76 to give ‘T, = 941.03°R Step 5. Solve for the acentric factor by applying Equation 2-21 to give w = 0.6123 2. Solution by using Standing’s Correlation. Step 1. Solve for the critical temperature of Gr, by using Equation 2-77 to give (Ter, Step 2. Determine the critical pressure from Equation 2-78 to give oer, = 291.41 psia 1,247.73°R 3. Solution by using Riazi-Daubert Correlation. Step 1. Solve Equation 2-3 for T, to give Te = 1,294.1°R Step 2. Caleulate P, from Equation 2-3 to give P. = 263.67 Step 3. Determine Ty by applying Equation 2-3 Ty = 958.5°R Step 4, Solve for the acentrie factor from Equation 2-21 to give = 0.5346 —_ DETERMINATION OF THE PHYSICAL PROPERTIES OF THE. HEAVY PETROLEUM FRACTION FROM GRAPHICAL CORRELATIONS Several mathematical correlations for determining the physical and eriti- cal properties of petroleum fractions have been presented. These correla- tions are readily adapted to computer applications. However, it s important to present the properties in graphical forms for a better understanding of the behaviors and interrelationships of the properties.68 Hydrocarbon Phase Behavior Boiling Points ‘Numerous graphical correlations have been proposed over the years for determining the physical and critical properties of petroleum fractions. Most of these correlations use the normal boiling point as one of the correla- tion parameters. As stated previously, there are five different methods of de- fining the normal boiling point: ‘a. Volume Average Boiling Point (VABP) . Weight Average Boiling Point (WABP) ¢. Molal Average Boiling Point (MABP) 4d. Cubie Average Boiling Point (CABP) fe. Mean Average Boiling Point (MeABP) Figure 2-2 shows the conversions between the VABP and other boiling points. The following steps summarize the use of Figure 2-2. Step 1. On the basis of ASTM D-86 distillation data, calculate the volumet- ric average boiling point from the following expressions: VABP = (ho + too + too + to + i)/5 (2-82) where t is the temperature in °F and the subseripts 10, 30, 50, 70, and 90 refer to the volume percent recovered during the distillation sgnameentn treet gy ¥ Figure 2-2. Correction to volumetric average boiling points. Courtesy of the Gas. Processors Suppliers Association. Published in the GPSA Engineering Data Book, Tenth Edition, 1987. Pure Component Properties/Characterizing Undefined Fractions 6 5 Step 2. Calculate the 10% to 90% “slope” of the ASTM distillation curv: from the following expression: ? slope = (too ~ t)/80 83 Stop 3. Enter the value of the slope in the graph and travel vertically to the appropriate set for the type of boiling point desired. Step 4. Read from the ordinate a correction factor for the VABP and apply the relationsh Desired Boiling Poin VABP + Correction Factor (284 ‘The use of the graph can best be illustrated by the following examples: Example 2-7. The following ASTM distillation data for a 55°API gravity petroleum fraction is given: Distillation cut % Over jan SRG Reem eee Sree eee eee ee oee 7 co 2 10 3 20 4 20 5 40 6 60 7 6 e 70 9 8% 10 90 Residue eps * initial boiling point *+* Bnd poiat Caloulate: 1. WABP 2, MABP, 3. CABP 4, MeABP Solution, Step 1. Calculate VABP from Equation 2-82: VABP = (178 + 200 + 253 + 318 + 410)/5 = 273°F10 Hydrocarbon Phase Behavior Step 2. Calculate the distillation curve slope from Equation 2-83: slope = (410 - 178)/80 = 2.9 Step 3. Enter the slope value of 2.9 in Figure 2-2 and move down to the appropriate set of boiling point curves. Read the corresponding correction factors from the ordinate to give: ‘© Correction factor for WABP = 6°F ‘© Correction factor for CABP = ~ 7°F ‘© Correction factor for MeABP =~ 18°F © Correction factor for MABP = ~ 33°F Step 4. Calculate the desired boiling point by applying Equ: © WABP = 273 + 6 = 279°F © CABP = 273 ~ 7 = 266°F © MeABP = 273 ~ 18 = 255°F © MABP = 273 ~ 93 = 240°F Molecular Weight Figure 2-3 shows a convenient graphical correlation for determining the molecular weight of petroleum fractions from their mean average boiling points (MeABP) and API gravities. ‘The following example illustrates the practical application of the graphical method. Example 2-8. Calculate the molecular weight of the petroleum fraction with an API gravity and MeABP as given in Example 2-7, Solution. From Example 2-7: API = 55° MeABP = 255°F Enter the above values in Figure 2 to give MW = 118. Critical Temperature ‘The critical temperature of a petroleum fraction can be determined by using the graphical correlation shown in Figure 2-4. ‘The required correla- tion parameters are the API gravity and the molal average boiling point (MABP) of the undefined fraction, Molecular weight Pure Component Properties/Characterizing Undefined Fractions. 71 MW, Boling Points, Graves, Petoleum Fractions Tm ETT RAN aN ERASE ty ARITA HH TE i AO eet He Ht i ae HEEL Es at AL a TEE nal AG PE a En MT ‘Mean average boiling point, Figure 2-3. Relationship between molecular weight, API gravity, and mean aver- ‘age boiling points. Courtesy of the Gas Processors Suppliers Association, Pub- lished in the GPSA Engineering Data Book, Tenth Edition, 1987. Example 2-9. Calculate the critical temperature of the petroleum fraction with physical properties as given in Example 2-7. Solution, From Example 2-7: MABP = 240°F API = 55° Enter the above values in Figure 2-4 to give: T. = 600°F72 Hydrocarbon Phase Behavior Pando Pressures ot hyarcarbont Penud te press, ple Mean average bang pale Figure 2-4. Relationship between critical pressure, API gravity, and mean aver- ‘age boiling points. Courtesy of the Gas Processors Suppliers Association. Pub- lished in the GPSA Engineering Data Book, Tenth Edition, 1987, Critical Pressure Figure 2-5 is a graphical correlation of the critical pressure of the unde- fined petroleum fractions as a function of the mean average boiling point (MeABP) and the API gravity. The following example shows the practical tse of the graphical correlation, Example 2-10, Caleulate the critical pressure of the petroleum fraction from Example 2-7 Pure Component Properties/Charactertzing Undefined Fractions 73. aS ‘Normal bolting point, *F Figure 2-5. Critical temperature as a function of API gravity and boiling points. Courtesy of the Gas Processors Suppliers Association. Published in the GPSA Engineering Data Book, Tenth Edition, 1987. Critical temperature, °F14 Hydrocarbon Phase Behavior Solution. From Example 2-7: APL = 55° MeABP = 240°F Determine the critical pressure of the fraction from Figure 2-5 to give: P, = 460 psia — PROBLEMS: 1, Ifa petroleum fraction has a measured molecular weight of 190 and a specific gravity of 0.8762, characterize this fraction by calculating the boiling point, critical temperature, eritical pressure, and critical vol- ‘ume of the fraction. Use the Riazi and Daubert correlation, 2. Caleulate the acentrie factor and critical compressibility factor of the component in the above problem. 3. Using the Lin and Chao generalized correlation, caleulate the physical and critical properties of a component with a MW, Q, and T of 200, 0.8251, and 500°% respectively 4. A petroleum fraction has the following physical properties: API = 50°, T, = 400°F, MW = 165, Calculate p., Te, Ves , and Z; by using the following correlations: a. Cavett b. Kesler and Lee ec. Winn-Sim-Daubert 4. Watansiri-Owens-Starling 55. An undefined petroleum fraction with ten carbon atoms has a mea sured average boiling point of 791°R and a molecular weight of 134. If the specific gravity of the fraction is 0.78, determine the critical pres- sure, the critical temperature, and acentric factor of the fraction by using: . Robinson-Peng PNA Method . Bergman PNA Method Riazi-Daubert Method |. Cavett’s Correlation : Kesler-Lee Correlation Willman-Teja repege Pure Component Properties/Characterizing Undefined Fractions 15 6. A heptanes-plus fraction is characterized by a molecular weight of 200 and specific gravity of 0.810. Calculate P,, T., Ty, and acentric factor of the plus fraction by using: a. Riazi-Daubert Method b. Rowe's Correlation ¢. Standings Correlation 7, Using the data given in Problem 6, and the boiling point as calculated by the Riazi and Daubert correlation, determine the eritical properties and acentric factor by employing ‘a, Cavett’s Correlation b. Kesler-Lee Correlation ¢, Compare the results with those obtained in Problem 6. EFERENGES 1. Ahmed, T, “Composition Modeling of Tyler and Mission Canyon For- ‘mation Oils with CO, and Lean Gases,” final report submitted to Mon- tanans on a New Track for Science (MONTS) (Montana National Sei- ence Foundation Grant Program), 1985. 2, Bergman, D. F,, Tek, M. R., and Katz, D. L., “Retrograde Condensa- tion in Natural Gas Pipelines,” Project PR 2-29 of Pipelines Research Committee, AGA, Jan. 1977. 3. Cavett, R. H., “Physical Data for Distillation Caleulations—Vapor- Liquid Equilibrium,” Proc. 27th Meeting, API, San Francisco, 1962, pp. 351-366. 4, Edmister, W. C., “Applied Hydrocarbon Thermodynamics, Part 4: Compressbility Factors and Equations of State,” Petroleum Refiner, April 1958, Vol. 37, pp. 173-179. 5, Edmister, W. C. and Lee, B. I., Applied Hydrocarbon Thermodynam- ics, Vol. 1, Gulf Publishing Company: Houston, 1986. 6. Erbar, J. H., “Prediction of Absorber Oil K-Values and Enthalpies, search Report 13, GPA, Tulsa, Oklahoma, 1977. 7. Haugen, O. A., Watson, K. M., and Ragatz, R. A., Chemical Process Principles, 2nd ed., New York: Wiley, 1959, p. 57. 8, Hopke, $. W. and Lin, C. J., “Application of BWRS Equation to Ab- sorber Oil Systems,” proceedings 53rd Annual Convention GPA, Den- vver, Colorado, March 1974, pp. 63-71. 9. Katz, D. L. and Firoozabadi A., “Predicting Phase Behavior of Con- densate/Crude-Oil Systems Using Methane Interaction Coefficients, JPT, Nov. 1978, pp. 1649-1655. Re.16 Hydrocarbon Phase Behavior 410. Kesler, M. G. and Lee, B. I, “Improve Prediction of Enthalpy of Frac- tions,” Hydrocarbon Processing, March 1976, pp. 153-158. (i) Lin, H. M. and Chao, K. C., “Correlation of Critical Properties and ? Acentric Factor of Hydrocarbons and Derivatives,” AIChE Journal, Vol. 20, No. 6, Nov. 1984, pp. 981-983. 12. Nath, J, “Acentric Factor and the Critical Volumes for Normal Fluids,” Ind. Eng. Chem, Fundam, Vol. 21, No. 8, 1985, pp. 325-326, 13. Reid, R,, Prausitz, J. M., and Sherwood, '., The Properties of Gases and Liquids, 3rd ed., McGraw-Hill, 1977, p. 21. 1. Riazi, M. R. and Daubert, T. E., “Simplify Property Predictions,” Hy- drocarbon Processing, March 1980, pp. 115-116. 15. Riazi, M. R. and Daubert, T. E., “Characterization Parameters for Pe- troleum Fractions,” Ind. Eng. Chem. Res., Vol. 28, No. 24, 1987, pp. 155-159. 16, Robinson, D. B. and Peng, D. ¥., “The Characterization of the Hep- tanes and Heavier Fractions,” Research Report 28, GPA, Tulsa (1978). 17. Rowe, A. M., “Internally Consistent Correlations for Predicting Phase Compositions for use in Reservoir Compositional Simulators,” paper SPE 7475 presented at the 53rd Annual Fall Technical Conference and Exhibition. 18, Salerno, S., et al., “Prediction of Vapor Pressures and Saturated Vol- umes,” Fluid Phase Equilibria, Vol. 27, June 10, 1986, pp. 15-34. 19, Sim, W. J. and Daubert, T. E., “Prediction of Vapor-Liguid Equilibria of Undefined Mixtures,” Ind. Eng. Chem. Process Des. Dev., Vol. 19, No. 3, 1980, pp. 380-393. 20, Standing, M. B., Volumetric and Phase Behavior of Oil Field Hydrocar- bon Systems, Society of Petroleam Engineers, Dallas, 1977, p. 124. 21, Watansiri, 5., Owens, V. H., and Starling, K. E., “Correlations for Esti- mating Critical Constants, Acentric Factor, and Dipole Moment for Un- defined Coal-Fluid Fractions,” Ind. Eng. Chem. Process Des. Dev., 1985, Vol. 24, pp. 294-296. 2, Whitson, C. H., “Effect of Physical Properties Estimation on Equation- of-State Predictions,” SPEJ, Dec. 1984, pp, 685-696 23, Willman, B. and Teja, A., “Prediction of Dew Points of Semicontinuous Natural Gas and Petroleum Mixtures,” Ind. Eng. Chem, Res., 1987, Vol. 226, No. 5, pp. 948-952. 24, Winn, F W,, “Simplified Nomographic Presentation, Characterization of Petroleum Fractions,” Petroleum Refiner, Vol. 38, No. 2, 1957, p. 157. 3 Properties of Natural Gases Laws which describe the behavior of gases in terms of pressure p, volume \V, and temperature T have been known for many years, These laws are rela~ tively simple for a hypothetical fluid known as a perfect (ideal) gas. This chapter reviews the perfect gas laws and how they can be modified to de- scribe the behavior of real gases which may deviate significantly from these Jaws under certain conditions of pressure and temperature. ‘A gas is defined as a homogeneous fluid of low density and viscosity, which has no definite volume but expands to completely fill the vessel in which itis placed. Knowledge of pressure-volume-temperature (PVT) rela- tionships and other physical and chemical properties of gases is essential for solving problems in natural gas reservoir engineering. ‘The physical proper- ties of a natural gas may be obtained directly either by laboratory measure- ‘ments or by prediction from the known chemical composition of the gas. In the latter case, the calculations are based on the physical properties of indi- ‘vidual components of the gas and upon physical laws, often referred to as mixing rules, relating the properties of the components to those of the gas mixture. BEHAVIOR OF IDEAL GASES ‘The kinetic theory of gases postulates that the gas is composed of a very large number of particles called molecules. For an ideal gas, the volume of these molecules is insignificant compared with the total volume occupied by the gas. It is also assumed that these molecules have no attractive or repul- sive forces between them, and it is assumed that all collisions of molecules are perfectly elastic. Pure Gases Based on the above kinetic theory of gases, a mathematical equation called Equation-of-State can be derived to express the relationship existing 118 Hydrocarbon Phase Behavior between pressure, volume, and temperature for a given quantity of gas. This relationship for perfect gases is called the Ideal Gas Law, and is ex- pressed mathematically by the following equation: pV = oRT @1 where n is the quantity of gas in moles and is defined by the expression m = “ Mw (3-2) or-|g}m = in which, for the conventional field units used in the Petroleum industry, p= absolute pressure, psia V = volume, ft? T = absolute temperature, °R umber of moles of gas, Ib-mole m = weight of gas, 1b MW = molecular weight, IbyIb-mole R = the universal gas constant which, for the above units, has the value 10.790 psia f?/Ib-mole °R Because the density is defined as the mass per unit volume of the sub- stance, Equation 3-3 ean be solved for the density to yield (3-4) where py = density of the gas, Ib/f, and tb refers to Ibs mass in any subse- uent discussions of density in this text. In dealing with gases at a very low pressure, the ideal gas relationship is a convenient and generally satisfactory tool. For the calculation of the physi- cal properties of natural gases at elevated pressure, the use of the ideal gas equation-of-state may lead to errors as great as 500% , as compared to errors of 2-3% at atmospheric pressure. Example 3-1. Assuming an ideal gas behavior, calculate the density of pro- pane with a constant temperature of 100°F and 20 psi Properties of Natural Gases 79 Solution. Applying Equation 3-4 yields (20) (44.097) Pe" To.ay 100 + 460) ~ O48 DE IDEAL GAS MIXTURES Petroleum engineers are usually interested in the behavior of mixtures and rarely deal with pure component gases. Because natural gas is a mixture of hydrocarbon components, the overall physical and chemical properties can be determined from the physical properties of the individual components in the mixture by using appropriate mixing rules. Conventionally, natural gas compositions are expressed in terms of mole fraction, weight fraction, and volume percent. These are derived as follows: Mole Fraction: The mole fraction of a particular component, component i, {s defined as the number of moles of that component divided by the total number of moles of all the components in the mixture. my 65) where yj = mole fraction of component i in the mixture 1, = number of moles of component { n= total number of moles in the mixture Weight Fraction: The weight fraction of any component is defined as the ‘weight of that component divided by the total weight. wee (36) weight fraction of component i ‘eight of component i in the gas phase jotal weight of the gas mixture Volume Fraction: The volume fraction of a specific component in a mixture is defined as the volume of that compound divided by the total volume of the mixture. MMi yu T (37)80 Hydrocarbon Phase Behavior where vj = volume fraction of component i in the gas phase V; = volume occupied by component i YV = total volume of the mixture It is convenient in many engineering calculations to convert from mole fraction to weight fraction and vice versa. The procedure of converting the composition of the gas phase from mole fraction to weight fraction is sum- marized in the following steps. Step 1. Assume that the total number of moles of the gas phase is one, ie,n= 1. Step 2. From Equation 3-5, it is apparent that a= ys Step 3. Because the number of moles of a component is equal to the ‘weight of the component divided by the molecular weight of the component, as expressed mathematically by Equation 3-2, the ‘weight of the component can be expressed as m= ysMW; and m rm Dn Step 4. By recalling the definition of weight fraction then yi MW, (38) yMW, ced Similarly, one can convert from weight fraction to mole fraction by applying the following relationship: (w/MW)) 09 (wiMW)) oo ‘The above procedure is conveniently illustrated through the following ex- amples. Example 3-2. Determine the composition in weight fraction of the follow- ing gas: Properties of Notural Gases 81 Mole Fraction ‘Component tL G 0.65 Ce 0.10 G 0.10 G 0.10 7 0.05 Solution. Component Mw, s GQ 16.04 10.4260 0.3824 G 0.10 30.07 3.0070 0.1103 G 0.10 44.10 4.4100 0.1618 GQ 0.10 58.12 5.8120 0.2132 cs 0.05 72.15 3.6075 0.1393 m = 27.2625 Example 3-3, Determine the composition in mole fraction of the following gas: Weight Fraction Component ws G 0.40 G 0.10 & 0.20 Gq 0.20 Cs 0.10 Solution. Component = _wi_ MW, wiMW, y= min G 0.40 16.04 0.02494 0.6626 Cs 0.10 30.07 0.00333 0.0885 G 0.20 © 44.10 0.00454 0.1206 G 0.20 58.12 0.00344 0.0914 ro 0.10 72.15 0.00139, 0.0369 n= 0.03764 ‘PROPERTIES OF IDEAL GAS MIXTURES. Usually the petroleum engineer is interested in studying the volumetzic behavior and evaluating the basic properties of natural gas mixtures. It is82 Hydrocarbon Phase Behavior ‘appropriate first to introduce and define the physical properties of ideal gas mixtures. “The basic properties of ideal gases are commonly expressed in terms of the apparent molecular weight, standard volume, density, specifie volume, and specific gravity. These properties are defined as follows: Apparent Molecular Weight ‘One of the properties that is frequently of interest to engineers is the ap- parent molecular weight. Ify zepresents the mole fraction of the ith compo- nent in a gas mixture, the apparent molecular weight is defined mathemati- cally by the following equation Mw. De MW, 10) where MW, = apparent molecular weight of the gas mixture MW, = molecular weight of the ith component in the mixture Standard Volume In many natural gas engineering caleulations, it is convenient to measure the volume occupied by 1 Ib/mole of gas at a reference pressure and temper- ature, These reference conditions are usually 14.7 psia and 60°F, and com- monly referred to as standard conditions. The standard volume is then de- fined as the volume occupied by 1 Ib/mole of an ideal gas at standard conditions. Applying the above conditions to Equation 3-1 and solving for the standard volume yields vpn RTs Pe Substituting for pressure and temperature, produces Va. = 879.4 sefilb-mole @u) where V,. = standard volume, sefilb-mole sef = standard cubie feet . ‘Ty, = standard temperature, °R Pe = standard pressure, psia Properties of Natural Gases 83 Density ‘As defined previously by Equation 3-4, the density of an ideal gas mixture is calculated by applying Equation 3-6 and replacing the molecular weight with the apparent molecular weight of the gas mixture to yield: Mw, a 12) where pg = density of the gas mixture, F/lb ‘MW, = apparent molecular weight Specific Volume “The specific volume is defined as the volume occupied by a unit mass of the gas. For an ideal gas, this property can be calculated by applying Equa tion 3-3: iu 8-13) where y = specific volume, f0¥/Ib py = gas density, Ib/f Specific Gravity ‘The specific gravity is defined as the ratio of the gas density to that of the air, Both densities must be taken at the same temperature and pressure, or: noe G14) Par Assuming the behavior of both the gas mixture and the air is described by the ideal gas equation, the specific gravity can be expressed by the following relationship: (3-15)84 Hydrocarbon Phase Behavior where 7g = gas specific gravity par © density of the air MWy, = apparent molecular weight of the air = 28.96 Example 3-4. A gas has the following composition: Component GQ 0.75 Cy 0.07 Cs 0.05 Gy 0.04 Cs 0.04 Ce 0.03 Cc: 0.02 “Assuming an ideal gas behavior, calculate the following gas properties at 1,000 psia and 100°F: a. Apparent molecular weight b. Specific gravity c. Gas density 4. Specific volume Solution. Component ae Mw. yc MW G 0.75 16.04 12.030 CG 0.07 30.07 2.105 G 0.05 44.10 2,205 Cy 0.04 38.12 2.395 Cs 0.04 72.15 2.886 Cs 0,03 86.18 2.585 Gr 0.02 100.21 2.004 26.14 a. Applying Equation 3-10 yields MW, = 26.14 bb, From Equation 3-15 26.14 _ 0.903, : 96 ” ©. Compute the gas density by using Equation 3-12 Properties of Natural Gases 85 __(1,000)(26.14) Pe*"(10.73)(660) 7 1 d. From Equation 2-13, specific volume = == = 0.93 ft*/Ib Equation 2-13, specific volume = —-3¢ = 0.23 4.35 Ioift ——— BEHAVIOR OF REAL GA! Basically the magnitude of deviations of real gases from the conditions of the ideal gas law increases with increasing pressure and temperature and varies widely with the composition of the gas. Real gases behave differently than ideal gases. The reason for this is that the perfect gas law was derived under the assumption that the volume of molecules is insignificant and no ‘molecular attraction or repulsion exists between them. This is not the ease for real gases. ‘Numerous equations-of-state have been developed in the attempt to cor- relate the pressure-volume-temperature variables for real gases with experi- ‘mental data. More recent equations-of-state are presented and discussed in detail in Chapter 6. In order to express @ more exact relationship between the variables p, V, and TT, a correction factor called the gas compressibility factor, gas deviation factor, or simply the Z-factor, must be introduced into Equation 3-1 to account for the departure of gases from ideality. This is: pV = ZnRT (16) where the compressibility factor Z is a dimensionless quantity and is defined as the ratio of the actual volume of n-moles of gas at T and p to the ideal volume of the same number of moles at the same T and p. aos uy G17) where V, = actual gas volume V, = ideal gas volume For a perfect gas, the gas compressibility factor is equal to one. For a real ‘gas, the Z-factor is greater or less than one depending on the pressure, tem- perature, and the composition of the gas. “The value of Z at any given pressure and temperature can be determined ‘experimentally by measuring the actual volume of some quantity of gas at the specified p and ‘T and solving Equation 3-16 for the compressibility fac- tor Z. A typical curve of the Z-factor for a natural gas is shown in Figure 3:1, where the Z-factor is plotted as a function of pressure for a given con-86 Hydrocarbon Phase Behavior 1 Figure 3-1. A typical Z-p diagram. Pressure stant temperature. For different temperatures, the gas deviation factor ‘curves follow a very definite pattern. ‘Numerous independent experimental studies of pure gases showed a well- established relationship between the compressibility factors and both pres- sure and temperature. Selected charts of this relationship for some pure ‘components (methane, ethane, and propane) are given in Figures 3-2 through 3-4. Such experimental determinations of Z-factors for a specific gas as a function of pressure and temperature represent the most reliable method of obtaining the Z, p, and T relationship. However, with the time and expense involved, it is not necessary to conduct such individual experi- ‘ments because sufficient information is known about the variation of com- pressibility with pressure and temperature to permit a correlation. This cor- relation is based on the theory of “Corresponding States.” The theory proposes that all gases will exhibit the same behavior, e.g,, Z-factor, when viewed in terms of reduced pressure, reduced volume, and reduced tempera- ture, The term “reduced” means that each variable is expressed as a ratio of its critical value. These reduced states can be expressed mathematically by the following relationships: Pre (3-18) sic Fe 19) (3-20) Properties of Natural Gases 87 00 08s Figure 3-2. Methane compressibility factors chart. Courtesy of the Gas Proces- sors Suppliers Association, where p, = reduced pressure Pe = critical pressure T, = reduced temperature ‘T. = critical temperature V, = reduced volume Y. = critical volume Notice that only pure gases possess distinct values of critical properties. Therefore, if the theory of corresponding states can be applied without ap- preciable error, all gases would have the same value for Z at the same re88 Hydrocarbon Phase Behavior Figure 3-3. Ethane compressiblity factors chart. Courtesy of the Gas Proces- sors Suppliers Association. uced temperature and pressure. This can be observed and appreciated by solving the following example. Example 3-5. Calculate the compressibility factors of: a. Methane b. Ethane ¢. Propane . at a reduced pressure and temperature of 2, and 1.6, respectively, Properties of Natural Gases 89 Re Figure 3-4. Propane compressibility factors chart. Courtesy of the Gas Proces- sors Suppliers Association. Solution. 1, Z-factor for methane. Because p, = p/p. and T, T/T, then P= ps Pe = (2)(667.8) = 1,935.6 psia T= T,+T. = (1.4)( ~ 116.63 + 460) = 549.392°R, From Figure 3-2 Z= 0.88 b, Z-factor for ethane.90 Hydrocarbon Phase Behavior p= Pr* Pe = (2)(707.8) = 1,415.6 psia 12 ,-"F, = (600.09 + 460) ~880.100°R From Figure 3-3, Z = 0.882 ¢. Z-factor for propane. P= pr’ Pe = (2)(616.3) = 1,232.6 psia ‘T = T,+ Te = (1.6)(206.01 + 460) = 1,065.616°R From Figure 3-4 Z = 0.886 The above example shows that at equal values of T, and p, the Z-factors for the three substances are very similar, indicating the lear power of the corresponding states principle. The application of the corresponding states principle to mixtures is based on the observation that the compressibility factor is a universal function of reduced pressure and temperature. Thus, the corresponding states principle should be applicable to mixtures if proper values for the critical point properties are used for the mixtures. Kay (1936) introduced the concept of pseucio-critical values whieh can be used in place of the true critical pressure and temperature of hydrocarbon mixtures. Kay proposed the following mixing rules for calculating the pseudo-critical prop- erties of hydrocarbon mixtures. Prem Divs Po (@-21) Te= Dy To 6-22) where Pre = pseudo-critical pressure, psia Tye = pseudo-critical temperature, °R Pa = critical temperature of component i, psia ‘Ty = critical temperature of component i, °R yi = mole fraction of component i in the gas mixture. Equations 3-2] and 3-22 are referred to as “Kay's Mixing Rule.” ‘The reduced states for gas mixtures are called the pseudo-reduced pres- sure and temperature, and they are expressed by the following relationships: 2 29 Pe : paeatr 2 a Properties of Natural Gases. 91 where Py = pseudo-reduced pressure of the gas mixture Tye = pseudo-reduced temperature of the gas mixture Studies of the compressibility factors for natural gases of various compos tions have shown that compressibility factors ean be generalized with eafft clent accuracies for most engineering purposes by introducing the eoncept of seudo-reduced pressures and pseudo-reduced temperatures Standing and Katz (1942) presented a generalized compressibility factor chart, as shown in Figure 3-5. The chart represents compressibility factors of sweet natural gus as a function of p,, and T,,. This chart is generally rel, able for sweet natural gases with minor amounts of non-hydrocarbons, Its cone of the most widely accepted correlations in the oil and gas industry. Equation 3-16 may be written in terms of specific volume or density as follows: Because pV=Z Gar RT then V_ zRr on” p MW, ar 1_pMW, ee ORT G26) where » = specific volume, ft/lb 4 = density, bite? Example 3-6, Using the gas composition, pressure, and temperature given in Example 34, and assuming a real gas behavior, calculate gas density Solution, Component _y MW: _yisMWi Ta"R yi: Tw pa vip Ci 0.75 16.04 12.030 343.5 957.6 679 504.7 CG 0.07 30.07 2.105 550.1 38.5 708496 Cy 0.05 44.10 2.205 666.2 333 G17 «09 Gs Cy 0.04 56.12 2.395 765.6 30.6551 oR. 0.04 72.15 2.886 847.0 33.9485 19.4 Co 0.03 86.18 2.585 9146274434 13.0 C 0.02100.21__2.004 972.819.5307 __ 7.9 MW, = 26.14 Tre = 440.8 Dy = ATS92 Hydrocarbon Phase Behavior "eau Pseudo reduced pressure, P, Figure 3-5. Standing and Katz compressibility factors chart. Courtesy of the Gas Processors Suppliers Association, Published in the GPSA Engineering Data Book, Tenth Edition, 1987. Properties of Natural Gases 93 Pye = P/Ppe = 1,000/647.5 = 1.54 Tye = Ti Tye = (100 + 460)/440.8 = 127 From Figure 35, Z = 0.725 Solve for gas density by applying Equation 3.26. (1,000)(26.14) Pe y-7a5j0.73)5) ~ °° NE ‘The compressibility factor chart shown in Figure 3-5 is applicable to most ‘gases encountered in petroleum reservoirs and provides satisfactory predic- tion for all engineering computations. The calculated volumetric behavior of gases containing only minor amounts of non-hydrocarbon can be aceu- rate within 3%. In cases where the composition of a natural gas is not available, the pseudo-critical properties, i.e., Pye and Tye, can be predicted solely from the specific gravity of the gas. Brown et al. (1948) presented a graphical method for a convenient approximation of the pseudo-critical pressure and pseudo- critical temperature of gases when only the specific gravity of the gas is available. The correlation is presented in Figure 3-6. Standing (1977) ex- pressed this graphical correlation in a mathematical form: Case 1: Natural Gas Systems Tye = 168 + 395 y, - 12.5 of 27) Pye * BTT + 15.0 5 - 37.5 4 (3-28) Case 2: Gas Condensate Systems Tye = 187 + 330 4-71.59 (6-29) Pye = 706 ~ 51.7 yy - IL 8-30) where T,, = pseudo-critical temperature, °R Pye = pseudo-critical pressure, psia "Ye = specific gravity of the gas mixture Example 3-7. Using the data given in Example 3-6, recalculate the gas density by estimating the pseudo-critical properties from Equations 3-27 and 3-98. Solution. © Calculate the specific gravity of the gas 194 Hydrocarbon Phase Behavior iseoaneocs 0seraae wera Limitations: Max 5% Ne 26.00 2% HS rial pressure, Pseudo cial temperature, *F Spectc Gravity ofthe Gas Figure 3-6, Pseudo-critical properties of natural gases. Courtesy of the Gas Pro- ‘cessors Suppligrs Association. Published in the GPSA Engineering Data Book, ‘Tenth Edition, 1987, MW, _ 26.14 29.96 29.96 *# Solve for Tye and pye by applying Equations 3-27 and 3-28. ‘Tye = 187 + 330 (0,903) ~ 71.5 (0.003)* = 427°R Pye = 706 ~ 51.7 (0,903) ~ 11.1 (0.903)? = 650 psia # Calculate pye and Tye = 0.903 6 iP 13 -2- 1,000 =O ost Po = “59 ‘Estimate the compressibility factor from Figure 3-5 Z= 0.748 3 “(2,000)(26.14) Pe Gasja0.7ayeaay 821 Properties of Natural Gases 95 EFFECT OF NON-HYDROCARBON COMPONENTS ‘ON THE Z-FACTOR Natural gases frequently contain materials other than hydrocarbon com- ponents, such as nitrogen, carbon dioxide, and hydrogen sulfide. Hydrocar- bbon gases are classified as sweet or sour depending on the hydrogen sulfide content. Both sweet and sour gases may contain nitrogen, carbon dioxide, or both. A hydrocarbon gas is termed a sour gas if it contains one grain of HS per 100 cubic feet. ‘The common occurrence of small percentages of nitrogen and carbon di- oxide is in part considered in the correlations previously cited. Concentra- tions of up to 5% of these non-hydrocarbon components will not seriously affect accuracy. Errors in compressibility factor calculations as large as 10% ‘may occur in higher concentrations of non-hydrocarbon components in gas mixtures, ‘CORRECTION FOR NON-HYDROCARBONS: Natural gases which contain HyS and/or CO» frequently exhibit different compressibility factor behavior than do &Weét* gases. Wichert and Aziz (1972) developed a simple, easy to use calculation procedure to account for these differences. This method permits the use of the Standing-Katz chart, i.e., Figure 3-5, by using a pseudo-critical temperature adjustment factor, which is a function of the concentration of CO; and HaS in the sour gas. This correction factor is then used to adjust the pseudo-critical temperature and pressure according to the following expressions: ‘Tpe! = 7, Comercio de PI OPE PENS ay) De GASES Lon PREsCnes ve HS ¥ Con (3-32) where Tye = pseudo-critical temperature, °R Pre = pseudo-eritical pressure, psia Tye’ = corrected pseudo-critical temperature, °R Pye’ = corrected pseudo-critical pressure, psia 'B = mole fraction of H,S in the gas mixture «= pseudo-critical temperature adjustment factor and is defined ‘mathematically by the following expression98 Hydrocarbon Phase Behavior = 120 (AS ALS) + 15 (5 ~ BAO) (6.33) where the coefficient A is the sum of the mole fraction of HS and COs in the gas mixture, or A™ Ynys + Yoo: ‘The computational steps of incorporating the adjustment factor ¢ into the Z-factor calculations are summarized below: Step 1. Step 2 Step 3. Step 4. Step 5. Calculate the pseudo-critical properties of the whole gas mixture by applying Equations 3-27 and 3.28 or Equations 3-29 and 3-30, Calculate the adjustment factor from Equation 3-33. ‘Adjust the calculated pye and T,. (as computed in Step 1) by ap- plying Equations 3-31 and 3-32. Calculate the pseudo-reduced properties from Equations 3-23 and 3-24, Read the compressibility factor from Figure 3-5. Example 3-8. A sour natural gas has the following composition: Component CO; 0.10 HS 0.20 Ne 0.05 G F 0.63 G 0.02 Determine the density of the gas mixture at 1,000 psia and 110°F 1. Without making any corrections to account for the presence of the non-hydrocarbon components. 2. Using the Wichert-Aziz correction method. Solution. Component _y,_ MW: CO; HS yes Ts 0.10 44.01 «44011071 107.1 Be 0.20 34.08 6.816 yids ST 54.757 1,306 261.2 672.37 134.474 0.05 28.01 1.401 493 24.65 227.29 11.3645 0.63 16.04 10.105 667.8 420.714 343.06 216.128 0.02 30.07 0.601 707.8 14.156 549.78 11.00 MW, = 23.324 827.82 427.72 Properties of Natural Gases Pye = 827.82 Tye = 427.72 1. Determination of gas density without corrections: ‘* From Figure 35, Z = 0.820 © Calculate the gas density by applying Equation 9-26 (1,000)(23.324) 82) (10.73)(570), 2. Determination of gas density with correction: Be yng = 0.2 A= Yoo + Ynys ~ 0.1 + 0.20 = 0.90 * Calculate the correction factor ¢ from Equation 3-33. 29.86 or = 4.651 Ibife ‘© From Equation 3-31, correct the pseudo-critical temperature. Tye’ = 427.72 ~ 29.86 = 397.86 © Calculate the corrected py. by applying Equation 3-32. (827.82) (897.86) © Pe’ = gerne O20 05) ae” 72707 ese 1,000 Por aor 07 375, oT. = To" sarge ~ 458 © From Figure 3-5, Z = 0.837 . (1,000)(23.324) = 11,000)(28.324) __ 4.56 iby? P= @asno-7aye70) ~ +0 1 798 Hydrocarbon Phase Behavior Carr-Kobayashi-Burrows Correction Method. Carr, Kobayashi, and Burrows (1954) proposed a simplified procedure to adjust the pseudo-critical properties of natural gases when non-hydrocarbon components are present. The method can be used when the composition of the natural gas is not available. The proposed procedure is summarized in the following steps: Step 1. Knowing the specific gravity of the natural gas, calculate the pseudo-critical temperature and pressure from Figure 3-6, or by applying Equations 3-27 and 3-28. Step 2, Adjust the estimated pseudo-critical properties by using the fol- lowing expressions: Tye’ = Tye ~ 80 Yoo + 130 yurys ~ 250 yng 34) Pro! = Pye + 440 Yoo, + 600 yiugs ~ 170 yup (3-95) where Tye’ ~ the adjusted pseudo-critical temperature, °R ‘Tye = the unadjusted pseudo-critical temperature, °R Yoo, = mole fraction of CO yun = mole fraction of HS Yq = mole fraction of nitroge Pye’ = the adjusted pseudo-critical pressure, psia Dye * the unadjusted pseudo-eritical pressure, psia Step 3. Use the adjusted pseudo-critical temperature and pressure to cal- culate the pseudo-reduced properties. Step 4. Calculate the Z-factor from Figure 3-5. Example 9-9, Recalculate the gas density of Example 3-8 from its specific gravity. Solution. «© Because the specific gravity of the gas is 0.8054, calculate Tye and Pye by applying Equations 3-27 and 3-28. ‘Tye = 168 + 325 (0.8054) — 12.5 (0.8054)* = 421.65°R Pye ~ 677 + 15 (0.8054) - 37.5 0.8054)" + 650.9 psia> C47 + Adjust the calculated 7, and py. by using Equations 3-34 and 3.35. Tye’ = 421.65 ~ 80(0.10) + 190(0.20) ~ 250(0.05) = 427.15°R, (650.9 + 15(0.10) + 600(0.20) ~ 170(0.05) = 763.9 psia L Pre! Properties of Natural Cases 99 # Caleulate py, and Ty, Rater 4o> 1.33 ‘* From Figure 3-5, estimate the compressibility factor Z= 0.808 9-~ * Solve for the gas density (1,000)(23.324) Pa” Taney o.rayeray 72 te PARA conteGHe Ter Pe {IGH-MOLECULAR-WEIGHT GASES It should be noted that the Standing and Katz. compressibility factor chart (Figure 3-5) was prepared from data on binary mixtures of methane with propane, ethane, and butane, and on natural gases, thus covering a wide range in composition of hydrocarbon mixtures containing methane. No mix- tures having molecular weights in excess of 40 were included in preparing this plot. =—=5) Sutton (1985) evaluated the accuracy of the Standing-Katz compressibil- ity factor chart using laboratory-measured gas compositions and Z-factors, and found that the chart provides satisfactory accuracy for engineering cal- culations, However, Kay’s mixing rules, i.c., Equations 3-21 and 3-22 (or comparable gravity relationships for calculating pseudo-eritical pressure and temperature), result in unsatisfactory Z-factors for high molecular ‘weight reservoir gases. The author observed that large deviations occur to ’ gases with high heptanes-plus concentrations. He pointed out that Kay's mixing rules should not be used to determine the pseudo-critical pressure © and temperature for reservoir gases with specific gravities greater than about 0.75. + “Button proposed that this deviation ean be minimized by utilizing the ‘mixing rules developed by Stewart etal. (1958), together with empirical ad- justment factors related to the presence of the heptane-plus fraction in the gas mixture. The proposed approach is outlined in the following steps: Step 1. Calculate the parameters J and K from the following relation- ships: abb-Rse-RT oe100 Hydrocarbon Phase Behavior x-E[o(f] -- iti em 5 K = Stewart-Burkhardt-Voo correlating parameter, °R/ y= mole fraction of component { in the gas mixture Step 2. Calculate the adjustment parameters Fj, ¢, and ex from the fol- lowing expressions: 1 (r. 2 r\s} r-2|y (2 2), (2 (3-38) : abel. al) I. ne 6 = 0.081 Fy + 11995 Ff ~ 14.004 Fy yor, + 64.434 Fy ye, (6-39) a & (0.3129 yoy, ~ 4.8156 s8,, + 27.5751 98,,] er (3-40) ., = mole fraction of the heptanes-plus component Yor (Teh, = critical temperature of the Cy. °R (bac; = critical pressure of the Cr, psia Step 3. Adjust the parameters J and K by applying the adjustment factors 6; and e according to the relationships: Jet-4 @-4l) Ki =K-e (6-42) where J, K = calculated from Equations 3-36 and 3-37 ¢ éx = calculated from Equations 3-39 and 3-40 where Step 4. Calculate the adjusted pseudo-critical temperature and pressure from the expressions (3-43) Ppo = ae (G-44) Step 5. Having calculated the adjusted Tye and p,q, the regular pfocedure of caleulating the compressibility factor from the Standing and Katz chart is followed. Properties of Natural Gases 101 Sutton’s proposed mixing rules for calculating the pseudo-critical proper- ties of high-molecular-weight reservoir gasts (ie., >1.25) should signifi- cantly improve the accuracy of the calculated DIRECT CALCULATION OF COMPRESSIBILITY FACTORS Alter four decades of existence, the Standing Katz Z-factor chart is still ‘widely used as a practical source of natural gas compressibility factors. AS a result, there was an apparent need for a simple mathematical description of that chart. Several empirical correlations for calculating Z-factors have been developed over the years. Takacs (1976) reviewed the performance of eight of these correlations. The following five empirical correlations are pre- sented here, fall-Yarborough © Dranchuk-Abu-Kassem * Dranchuk-Purvis-Robinson Hankinson-Thomas-Phillips Papay Method Papay (1968) proposed the following simplified equation for calculating the compressibility factor: Za 1-Be oamrasrs oss | 2) (3-45) ie here pyr = pseudo-reduced pressure Ty = pscudo-reduced temperature ‘The above correlation is convenient for application for hand calculations. “Takacs pointed out that the proposed correlation produces an average error of ~ 4.8%. Example 3-10. Using the data given in Example 3-8, re-evaluate the com- pressibility factor by using the Papay correlation, and calculate the gas den- sity. a ‘© From Example 3.8, Pye = for the Z-factor. .3T5 and Tyr = 1.433. Solve Equation 3-45,102 Hydrocarbon Phase Behavior 1.375 137 . — 0,04188423 | 2575 } | - 0, 0.36748758 1 oo (1305) 0.687 Zal- 1433 © Calculate the gas density. = —(1,000)(23.324) __ P=" Desnqio-rayem ~ 8 b Hall-Yarborough Method Hall and Yarborough (1973) presented an equation-of-state that accu- rately represents the Standing and Katz Z-factor chart. The proposed ex- pression is based on the Starling-Carnahan equation-of-state. The coeffi- cients of the correlation were determined by fitting it to data taken from the ing and Katz Z-factor chart, Hall and Yarborough proposed the fol- ‘mathematical form: zo( 2008 mt |x| 12.00" 46) where P,, = pseudo-reduced pressure 't = reciprocal of the pseudo-reduced temperature, i.e., Tyo/T Y = the reduced density which can be obtained as the solution of the following equation: = = 008185 py XP [= 120 ye EMERY = (14.76t ~ 9.76t? + 4.58t°) Y? + (90.7t — 242.2¢ + 42.40) Y2IS+280) 2g (3-47) ‘This non-linear equation can be conveniently solved for the reduced den- sity ¥ by using the Newton-Raphson iteration technique. The computa- tional procedure of solving Equation 3-47 is summarized in the following steps. Step 1. Make an initial guess of the unknown parameter, Y+, where k is an iteration counter. An appropriate initial guess of Y is given by the following relationship: ‘Y* = 0.06125 py, t EXP [ ~ 1.201 - t] Step 2. Substitute this initial value in Equation 3-47 and’evaluate the non-linear function. Unless the correct value of Y has been ini- tially selected, Equation 3-47 will have a non-zero value of £(¥'). Properties of Natural Gases 103 Step 3. A new, improved estimate of Y, i.., Y*¥, is calculated from the following expression: net oye — A Pele oe (6-48) where f”(¥*) is obtained by evaluating the derivative of Equation 3-47 at Y4, or 1+ 4y + 4¥8—4yo4 ¥ (yy = LEA EE (209.52t — 19.528 2) ees, (29.52 — 19.524 + 9.1GE)Y + (2.18 + 2.824)(90,7¢ ~ 249.28 + 42.20) yls+zen) (8-49) Step 4. Steps 1-3 are repeated n times, until the error (i.e., Y* ~ Y°~!) becomes smaller than a preset tolerance, e.g., 10! Step 5. ‘The correct value of ¥ is then used to evaluate Equation 3-46 for the compressibility factor Z. Hall and Yarborough pointed out that the method is not recommended for application if the pseudo-reduced temperature is less than one. Dranchuk and Abu-Kassem Method Dranchuk and Abu-Kassem (1975) proposed an eleven-constant equa- tion-of-state for calculating the gas compressibility factors. The authors pro- posed the following equation: 2 + Aw (1+ An 08) £- EXP{~ An: pf] +1 (350) ‘where p, = reduced gas density and is defined by the following relationship: 0.27 p, = OST oe 3.5 ZT. cee ‘The constants A; through Ax, were determined by fitting the equation, using non-linear regression models, to 1,500 data points from the Standing and Katz Z-factor chart. The coefficients have the following values:104 Hydrocarbon Phase Behavior 0.3265 Ap= 1.0700 Ay = ~ 0.5330 Ay = 0.01569 = 0,05165 Ag= 0.5475 Ay = 0.7361 Ay = 0.1844 ie = 0.1056 = Ay = 0.6134 = Ay = 0.7210 Equation 3-50 can be rearranged by replacing the left hand side of the equation, ie, Z, with 0.27plp, Tye and solving the resulting equation for px by using the Newton-Raphson iteration technique. An appropriate intial {guess of the solution is given by the following expression: irae (8-52) Tre Having obtained the correct p,, Equation 3-51 can be solved for the com- pressibility factor. ‘The proposed correlation was reported to duplicate compressibility fac- tors from the Standing and Katz chart with an average absolute error of 0.585 percent, and is applicable over the ranges Dranchuk, Purvis, and Robinson (1974) developed a correlation based on the Benedict. Webb-Rubin type of equation-of-state. The eight coefficients of the proposed equations were optimized by fitting the equation to 1,500 data points from the Standing and Katz Z-factor chart. The equation has the following form: - As , As| AS) pp + (AsAg ry (1 + As 7) EXP (~ Ag al (8-53) where p, is defined by Equation 3-51 and the coefficients A, through Ay have the following values: Ay = 0.31506237 ‘As = ~ 0.61232032 Ay= ~ 1.067099 ‘Ag = ~ 0.10488813 Ay = ~ 0.57832729 Ay = 0.68157001 59530771 ‘Ag = 0.68446549 Properties of Natural Gases 105, ‘The solution procedure of Equation 3-59 is similar to that of Dranchuk and Abu-Kassem, ‘The method is valid within the following ranges of pseudo-reduced tem- perature and pressure 1.05 5 Tye < 3.0 Hankinson, Thomas, and Phillips (1969) correlated the compressibility factors for natural gases as a function of the pseudo-reduced temperature ‘and pseudo-reduced pressure by using the Benedict-Webb-Rubin equation- of-state. The proposed equation is expressed in terms of compressibility fac- tor as follows: Pa Ty Ai gPe Aa 1 Best] exo —Ae 1 Bere lactea- Aude (3-54) ‘The accuracy of the data representation is improved considerably by breaking the data into two regions, one region for reduced pressures less than 5.0 and one region for reduced pressures between 5.0 and 15.0. ‘Thus, two sets of coefficients are obtained, one for each pressure range. The two sets of coefficients are presented in Table 3-1 It is suggested that the proposed correlation is used only at reduced tem- peratures above 1.1 Table 3-4 Coefticients for Equation 3-52 Coefficients Pr trom 0.4 106.0 Pr from 6 to 18 a ‘o.001280296 0.0014507682 A 0.38183005 0,37922268 & 0.022199287 0.028181399 A‘ 0.12215481 011812287 A ~ 0.015674794 0.037905063, a 0.027271364 0,19845016 a 0.023894019 0.048911693 x 0.49617780 0.0691425417106 Hydrocarbon Phase Behavior Equation 3-54 can be solved for the compressibility factor by using the Newton-Raphson iterative technique as outlined previously. (COMPRESSIBILITY OF NATURAL GASES ‘A knowledge of the variability of fluid compressibility with pressure and temperature is essential in performing many reservoir engineering calcula- tions. For a liquid phase, the compressibility is small and usually assumed to be constant. For a gas phase, the compressibility is neither small nor con- stant. ‘By definition, the isothermal gas compressibility is the change in volume pper unit volume for a unit change in pressure, or, in equation form: (6-55) where C, « isothermal gas compressibility, I/psi From the real gas equation-of-state aRTZ P v Differentiating the above equation with respect to pressure yields 1|az|_Z cr |} (22 _ b 8 Substituting into Equation 3-55 produces the following generalized relation ship: 1 (az) 265. fe For an ideal gas, Z = 1 and (@Z/dp}r = 0, therefore q-l (357) P ‘Equation 3-57 is useful in determining the expected order of magnitude of the isothermal gas compressibility. Properties of Natural Gases 107 Equation 3-56 can be conveniently expressed in terms of the pseudo-re- duced pressure and temperature by simply replacing p with (Pze Pp)» OF iG; La a Dr Pre Z [5B Bed ‘Multiplying the above equation by pye yields (3-58) ‘The term G, is called the isothermal pseuco-reduced compressibility, and {s defined by the relationship CCD (6-59) where C, = isothermal pseudo-reduced compressibility Gy = isothermal gas compressibility, psi? pseudo-reduced pressure, psi Values of (92/0p,.J1,, can be calculated from the slope of the T,, isotherm on the Standing and Katz Z-factor chart at the Z-factor of interest. Example 3-11. Given the following gas composition, Component) ania Cc 0.75 Ge 0.07 G 0.05 ne 0.04 n-Cs 0.04 Co 0.03 Cc; 0.02, calculate the isothermal gas compressibility at 1,000 psia and 100°F by as- suming: a. An ideal gas behavior b. A real gas behavior Solution. a. Assuming an ideal gas behavior108 Hydrocarbon Phase Behavior «© Applying Equation 3.57 yields 1 6 y= gy 7 1000 10°* at . Assuming a real gas behavior Step 1. Calculate pje and Tye of the gas mixture Dye = OATS psia Tye = 440.8°R Step 2. Compute Pye and Tye 1,000 = 1,000 154 Pee” G48 360. Ty = = 12 = ga9.5 7 77 Step 3. Estimate the compressibility factor Z from Figure 3-5. 2 = 0.795 Step 4. Caleulate the slope [3Z/3Pprhyr-tr from Figure 3-5 am 22.) o.1678 lel, Step 5. Solve for C, by applying Equation 3-58. 1 {2 = _[ 1 \(-0.1678) = 0.580 On Ts fes| Dee aEeE Step 6. From Equation 3-59, solve for Cy C, _ 0.8808 — Ce 9-8808 1,361.1 x 10-® psi-* Dpauitur Sieveseeuscedaeat G ‘Trube (1957) presented graphs from which the isothermal compressibility of natural gases may be obtained. The graphs, as shown in Figures 3-7 and 3.8, give the isothermal pseudo-reduced compressibility as a function of pscudo-reduced pressure and temperature, Example 3-12. Using Trube's generalized charts, rework Example 3-11. Solution. Step 1. From Figure 3-7, find C, c, = 0.90 Propertice of Natural Gases 109 PSEuo0 REDUCED COMPRESSIBILITY, Cy PSEUDO REDUCED PRESSURE, Fs Figure 3-7. Tube's psoudo-reduced compres Sion to pubis bythe Society of Petroleum Engh ity for natural gases. Permis- ‘of AIME. Copyright SPE- Step 2. Solve for C, by applying Equation 3-58. 0.90 6 og! Fig 719390 % 10° psi Mattar, Bras, and Aziz (1075) presented an analytical technique fr caleu- lating the isothermal gas compressibility. The authors expressed C, as a func- tion of [82/Ap,}r,, Father than [82/8pprlr,,- Equation 3-51 is differentiated with respect to pp, to give (ra 8 a |_ 0A en Ae Tre [1+ Bezl0p110 Hydrocarbon Phase Behavior Pseudo Reduced Compressii SINMOESSHES/Sm CRBC ESTES 's Peeudo Reduced Pressure, Py Figure 3-8. Trube's psoudo-reduced compressibility for natural gases. Permis- sion to pubiish by the Society of Petroleum Engineers of AIME. Copyright SPE- AIME. Equation 3-60 may be substituted into Equation 3.58 to express the pseudo-reduced compressibility as oe |__ zap | (61) 1+ 2 @2/0e.9 gel Pe where p, = pseudo-reduced gas density ‘The partial derivative [3Z/0p,}ry, appearing in Equation 3-61 is Gbtained from the Benedict-Webb-Rubin equation-of-state developed by Dranchuk, Purvis, and Robinson (Equation 3-53), to give Properties of Natural Gases 1A fe] -faeteed}ea[ecetloesarscgt (1 + As pi A of) EXP (~ As 6%) (6-62) ‘Values of the coefficients Ay-Ag are given previously in Table 3-1. The computational procedure of estimating C, through the use of Mattar, Brar, and Aziz is shown by the flow diagram in Figure 3-9. Gen Talevlate ye 88 Tye Solve eqation 3653 for ue Evaluate equation 3652 Solve for C, from equation 3-61 Figure 3-9, Flow diagram for calculating C, ‘GAS FORMATION VOLUME FACTOR __ ‘The gas formation volume factor is used to relate the volume of gas, as measured at reservoir conditions, to the volume of the gas as measured at standard conditions, i.e., 60°F and 14.7 psia. This gas property is then de- fined as the actual volume occupied by a certain amount of gas at a specified pressure and temperature, divided by the volume occupied by the same112, Hydrocarbon Phase Behavior ‘amount of gas at standard conditions. In an equation form, the relationship is expressed as Vor Ve By where _B, = gas formation volume factor, f8/scf \V,,t = volume of gas at pressure p and temperature TT, f° Vig = volume of gas at standard conditions, sef Applying the real gas equation-of-state (Equation 3-16) to the above rela- tionship gives Zn RT Seepreore baa ZaRT. Te D Pe where _Z, = Z-factor at standard conditions = 1.0 Puy Te = standard pressure and temperature, ie., 14.7 and 60°F or B, = 0.02827 2 Wors\ (3-63) In other field units, the gas formation volume factor ean be expressed in bblscf, to give sas ta eo B, ‘The reciprocal of the gas formation volume factor is called the Gas Ex- pansion Factor and designated by the symbol E,, oF - ae (3-65) E,= 95.57 Pe, slit? (3-65) In other field units, E,= P (3-66) B,= 198.6 2, sofbbl (6-66) Example 3-13. A gas well is producing at a rate of 15,000 ft¥/day from a gas reservoir at a bottom hole with flowing conditions of 1,000 psia and Properties of Natural Gases 113 100°F. The specific gravity of the gas is 0.903. Calculate the gas flow rate in scflday. Solution. ‘ From Equations 3-27 and 3-28, calculate Tye and pe of the gas phase, Tye = 427°R Pye = 650 psia * Compute Ty. and pp; and solve Z to give: Tyra 13 Pye = 154 Z = 0.148 ‘* Calculate the gas formation volume factor by applying Equation 3-65. 1,000 orsye ‘* Calculate the gas flow rate in sefiday gas flow rate = (15,000)(84.44) = 1,266,592 sof/day By = 95.97 = 84.44 soflfe? (GAS VISCOSITY ‘The viscosity of a fluid is a measure of the internal fluid friction (resis- tance) to flow. If the friction between layers of the fluid is small, i.e., low viscosity, an applied shearing force will result in a large velocity gradient. As the viscosity increases, each fluid layer exerts a larger frictional drag on the adjacent layers and velocity gradient decreases. ‘The viscosity of a fluid is generally defined as the ratio of the shear force ‘per unit area to the local velocity gradient. Viscosities are expressed in terms of poises, centipoises, or micropoises. One poise equals a viscosity of 1 dyne- secjem? and can be converted to other field units by the following relation- ships. 1 poise = 100 centipoises 1 x 10° micropoises = 6.72 x 10°? Ib massft-see = 2.09 x 10-8 Ibf-seclft? ‘The gas viscosity is not commonly measured in the laboratory because it can be estimated precisely from empirical correlations. Like all intensive properties, viscosity of a natural gas is completely deseribed by the following function,114 Hydrocarbon Phase Behavior where jug = the viscosity of the gas phase. The above relationship simply states that the viscosity is « function of pressure, temperature, and composi- tion. Many of the widely used gas viscosity correlations may be viewed as modifications of that expression. METHODS OF CALCULATING THE VISCOSITY OF NATURAL GASES Carr-Kobayashi-Burrows Method Carr, Kobayashi, and Burrows (1954) developed graphical correlations for estimating the viscosity of natural gas as a function of temperature, pres- sure, and gas gravity. The computational procedure of applying the pro- posed correlations is summarized in the following steps. Step 1. Calculate the pseudo-critical pressure, pseudo-critical tempera- ture, and apparent molecular weight from the specific gravity or the composition of the natural gas. Corrections to these pseudo- critical properties for the presence of the non-hydrocarbon gases (COs, No, and H;S) should be made if they are present in concen- tration greater than 5 mole percent, Step 2. Obtain the viscosity of the natural gas at one atmosphere and the temperature of interest from Figure 3-10. This viscosity, as deno- ted by 1), must be corrected for the presence of non-hydrocarbon components by using the inserts of Figure 3-10. The non-hydro- carbon fractions tend to increase the viscosity of the gas phase. ‘The effect of non-hydrocarbon components on the viscosity of the natural gas can be expressed mathematically by the following re- lationship, 11 = (erenoreied + (Apis + (itlcos + (Anns (3-67) where m= “corrected” gas viscosity at one atmo- spheric pressure and reservoir tempera- ture, @p (Au)nz = viscosity corrections due to the presence of N (81)co, = viscosity corrections due to the presence of Co, (An) = viseosity corrections due to the presence of HS (ia)anoneaed = uncorrected gas viscosity, ep ‘eas Geavity (ait 1000) Properties of Natural Gases : Sine 0uco welca0e Gelssne aie ne eanl sien Geen eee a> ome fan Asan Figure 3-10. Carr's atmospheric gas viscosity correlation. Permission to publish by the Society of Petro- eum Engineers of AIME. Copyright SPE-AIME. us116 Hydrocarbon Phase Behavior | SCOTT AA, My, Figure 3-11. Carr's viscosity ratio correlation. Permission to publish by the Soci- ‘ty of Petroleum Engineers of AIME. Copyright SPE-AIME. Step 3. Calculate the pseudo-reduced pressure and temperature. Step 4. From the pseudo-reduced temperature and pressure obtain the viscosity ratio (1g/ai) from Figure 3-11. The term yg represents the viscosity of the gas at the required conditions. Step 5. The gas viscosity, ug, at the pressure and temperature of interest is calculated by multiplying the viscosity at one atmosphere and system temperature, 41, by the viscosity ratio. The following examples illustrate the use of the proposed graphical corre- lations. Example 3-14, Given the following gas composition, Component. rani Ee Gq 0.850 C. 0.055 Cy 0.035 yt icy o.o10 | : caleulate the gas viscosity at 3,000 psia and 150°F Properties of Natural Gases NT Solution. Step 1. From the gas composition, calculate MW, Hg, Tyer aNd Ppe- Component ve MW; MW, G 0.850 16.04 13.634 Ce 0.055 30.07 1.654 G 0.035 44.09 1.543 CG 0.010 58.12) 0.5812 MW, = 17.412 ef aA og Gear "O° ¢ From Equations 3-27 and 3-28 Tye = 358.5 Dye = 672.5 Step 2. Calculate the viscosity of the natural gas at one atmosphere and 150°F from Figure 3-10. = 0.0119 ep Step 3. Calculate py, and Tye 3,000 Pe or25” _ 810 58.5 4.46 T, =1.70 Step 4. Estimate the viscesity ratio from Figure 3-12 on page 195 feel Step 5. Solve for the viscosity of the natural gas 1. t=“ Gx) = (1.7)(0.0119) = 0.0202 ep VS Example 3-15. Given the following gas composition Component i N 0.05 CO, 0.05 HS 0.02 GQ 0.80 C 0.05 G 0.03 calculate the gas viscosity at 200°F and 3,500 pia118 Hydrocarbon Phase Behavior Solution. Component ane MW. yMWi Ne 0.05 38.02 1.4010 CO; 0.05 44.01 2.2005 HS 0.02 34.08 0.6816 GQ 0.80 16.07 12.8560 Ce 04 30.07 1.5035 GQ 0.03 44.09 1.3227 MW, = 19.97 Step 1. Caleulate T,. and pye by applying Equations 3-27 and 3-28. Step 2. Step 3. Step 4, Step 5. Step 6. _, = 19.87 "8" 28.97 Tye = 380.43°R Poo = 665.11 psia = 0.689 Correct the calculated pseudo-critical properties to account for the presence of the non-hydrocarbon components using Equa- tions 3-34 and 3-35. Tye! = 980,49 ~ (60)(0.05) + 130(0.02) ~ 250(0.05) = 366.53°R Pye’ = 665.11 + 440(0.05) + 600(0.02) ~ 170(0.05) = 690.61 psia Estimate the gas viscosity of 1 atm and 200°F from Figure 3-10. m = 0.0123 ep Using the inserts in Figure 3-10, estimate (Spa, (Ax)oog, and nuns: (An)vg = 0.00042 ep (As)oos = 0.00026 ep (4i)uys = 0.00005 cp Calculate the corrected gas viscosity at atmospheric pressure and system temperature by applying Equation 3-67. ‘m, = 0.0123 + 0,00042 + 0.00025 + 0.00005 = ).01303 cp Calculate pp, and Ty, Properties of Natural Gases Tee = 365.59 7 1-50 Step 7. Estimate the viscosity ratio yy/a: from Figure 9-12, 9 Step 8. Calculate the viscosity of the natural gas at 3,500 psia and 200°F. my = (1.52)(0.01303) = 0.01981 ep Standing (1977) proposed a convenient mathematical expression for cal- culating the visoosity of the natural gas at atmospheric pressure and reser- voir temperature, j. Standing also presented equations for describing the effects of Nz, COs, and HyS on »;. The proposed relationships are: 11 (einai + (BH)ng + (Bricon + (Spgs ()ancrentes * [1.709(10-8) ~ 2.062(10-* 44} (T - 460) + 8.188(10"%) ~ 6.15(10-%) log (4) (3-08) (ine = ¥nz 18-48(10-®) log (7g) + 9-59(10-9)] (6-69) (i)co, = Yoo, [9-08(10"8) log (4) + 6.24(10°2)] (3-70) (Aa)ngs = Ynys 18.49(10"9) log (40) + 3.73(10-*)] 7) where mm = viscosity of the gas at atmospheric pressure and reservoir temperature, ep reservoir temperature, °R gas gravity cf Yar Yop Yuis = mole fraction of Nz, COs, and H,S, respectively Dempsey (1965) expressed the viscosity ratio u/s by the following rela- tionship, In fr. tl. 8g + i Pye + 82 Phe + A Phe + Tyr (84 + 5 Poe + Be Phe + 8: Ph) + TR, (ae + ao Pye + aio Phe + an PR) + Tye (@iz + aia Por + 14 Phe + is Pe) (8-72)120 Hydrocarbon Phase Behavior where Ty = pseudo-reduced temperature of the gas mixture Pye = pseudo-reduced pressure of the gas mixture ‘~ajg = Coefficients of the equations are given below ao = - 2.46211820 fag = — 7.93985684(10-) ay = 2.97054714 a = 1,39643306 ag =~ 2.86264054(10-!) ay = — 1,49144995(10-) ‘ay = 8.05420522(10"%) 4 = 4.41015512(10"9) a4 = 2,80860949 ‘yg = 8.39387178(10~*) ag = ~ 3.49603305 ais = = 1.86408848(10-) ag = 8.60373020(10-) ayy = 2.08367881(10~2) ay =~ 1,044324(10-4) ys = — 6,09579263(10-4) Uf 1eo-Gonzaler-Kakin Method _// Lee, Gonzalez, and Eakin (1966) presented a semi-empirical relationship for calculating the viscosity of natural gases. The authors expressed the gas viscosity in terms of the reservoir temperature, gas density, and the molecu: lar weight of the gas. Their proposed equation is given by pe 10-4 K EXP |X [Be 1h. 0/29,Pe | 6-73) 62.4 a | iW.) T? > Eee enae _ QA + 0.02 MW,) T'S ; hes 200 + 19 MW, + T sages ay \ x=95+ 8 +o01mw, | ‘ 6-75) Yu24-02% (76) bg = gas density at reservoir pressure and temperature, lb/ft T = reservoir temperature, °R MW, = apparent molecular weight of the gas mixture ‘The proposed correlation can predict viscosity values with a standard de- viation of 2.7% and a maximum deviation of 8.99% . The authors pointed out that the method cannot be used for sour gases, Example 3-16. Rework Example 3-14 and calculate the gas‘viscosity by using the Lee-Conzales-Eakin method. Properties of Natural Gases 421 Solution. ¢ Using pp: and Ty, of 4.46 and 1.7, respectively, calculate the gas com- pressibility factor from Standing and Katz chart. Z= 0.89 ‘Solve for gas density by applying Equation 3-26, (8,000)(17.412) (610)(10.73)(0.89) '* Calculate the parameters K, X, and Y by using Equatic +74, 3 and 3-76 respectively. Beeudezbsenseussuad! x = [9:4 + 0.02 (17.412)) (610)! Pen 97 White? 208+ 19 (17.412) + 10 ~ 12778 986 5 + 988, - + Fp 7 001 (17412) = 5.291 Y= 24-02 6.201) = 1.342 * Solve for the gas viscosity by applying Equation 3-73. pe=10-* (127.79) EXP [smu bz a4 Dean and Stiel (1965) proposed the following mathematical expressions for calculating the viscosity of natural gases at atmospheric pressure and res- exvoir temperature. eae eer aa 5) Td? 6 5 m= 94 00-9) CH forftys & 1.5 (77) 166.8 (10-*)[0.1338 Ty, — po m= 1680 wes Te 0.0882) for Tye > 15 (3-78) where f is the viscosity parameter of the gas mixture as defined by the fol- lowing equation. fm = 5.4402 (Tod WW) Po? (3-79)122 Hydrocarbon Phase Behavior Dean and Stiel recommended the following relationship for calculating the viscosity of natural gases at the prevailing reservoir condition. py + 10:8(00-5[BXP(L.499 p,) — EXP(—1.111(0)!*)] - (8-80) By ies (3-80) gas viscosity at reservoir pressure and temperature, cp gas viscosity at atmospheric pressure and reservoir tempera- ture, ep pr = reduced gas density as defined by Equation 3-51 Qs ont PJ ate. 2= € (Tors Ose) ‘The use of the proposed equation can be best illustrated by the following example. where Hy m Example 3-17, Rework Example 3-14 by using the Dean and Stiel correla- tion. The gas composition, pressure, and temperature are given here for con- Somponmnt a GQ 0.850 G 0.058 G 0.035 G 0.010 System pressure = 3,000 psia System temperature = 150°F Solution. From Examples 3-14 and 3-16, the following data were ob- tained, MW, = 17.412 Pye = 672.5 psia Th = 358.5°R pe = 4.46 1, = 1.70 z= 080 © Calculate the viscosity parameter, fq, by applying Equation 3-79 = 5.4402 O58.) = 0.4527 fn = 5.02 Ts 6TZ. | # Because Ty, > 1.5, apply Equation 3-78 to calculate Properties of Natural Gases 193 = 268:8(00~5)(0.1388(0.7) ~ 0.09925 1s a a = 0.0121 ep * Calculate the reduced density of the gas mixture from Equation 3-51. _ (0.27)(4.46) _ *™ o.gy¢i.70) ~ 07959 * Solve for the gus vigeosty by using Equation 3-80 to yield ‘APPLICATIONS OF THE NATURAL GAS PVT PROPERTIES ‘To demonstrate the practical applications of the natural gas physical properties in reservoir engineering, the following sections present and briefly review some of these applications. Radial Flow of Gases Henry Darcy, a French engineer, developed a fluid flow equation which hhas since become one of the standard mathematical tools of the petroleum engineer. The proposed equation, which is known as Darcy's equation, states that the rate of flow of a homogeneous fluid through a homogeneous ‘Porous media is proportional to the pressure gradient and the cross-sectional area normal to the direction of the flow; and inversely proportional to the of the fluid. In a differential form, the equation is expressed by the following relationship: L.1z71@xth)k dp 1,000 dr (81) where q = fluid flow rate, bbl/day = radial radius, ft h = formation permeability, md viscosity, ep pressure, psia Darcy's equation can be applied to determine the gas Flow rate in terms of volumetric rate at standard conditions, soffday, to give124 Hydrocarbon Phase Behavior 1.1271(2erh)k dp. 1,000 4B, dr ed where Q, = gas flow rate, sefiday gas viscosity, ep gas formation volume factor, bbl/sef ‘The gas formation volume factor, as expressed mathematically by Equa- tion 3-64 is combined with Equation 3.82 to give 0.703 khr(2) dp TT a Separating variables and integrating the above expression from wellbore conditions to the outer reservoir boundary conditions gives fe dr _0.703 kh g ( p « aj" Ba riauee TERE za 0.703 ka gp y= x8 so > "Tales? pad? cae where 1, = well drainage radius, ft wellbore radius, ft par = bottom-hole flowing pressure, psia De = reservoir pressure, psia T = reservoir temperature, °R ‘The term 2 |” (p12) dp in Equation 3-83 can be expanded to give et PP dp (@-84) ap-2f> Zz The expression 2 s (2p/uZ)dp is commonly called the real gas potential or the real gas pseudo-pressure. It is customarily represented by ni(p) or ¥. Thus Properties of Natural Gases 125 mio) == [2 Bap 5) Using the concept of the real gas potential in Equation 3-83 gives 0.703 kh(ve = bot) = (3-86) T La(tJtw) (66) & where Y_ = extemal real gas potential, psi*/ep Yor = bottom-hole real gas potential, ps%/ep ‘The drop in the real gas potential ie., Ye ~ Ys, can be determined nu- merically by plotting the term (2p/,Z) as a function of pressure and caleu- lating the area under the appropriate part of the resulting curve. Figure 3-12 shows a typical plot of (2p/xZ) versus p relationship. Figure 3-12. A typical plot of the gas pressure function. ‘The area under the curve, i.¢., Ye ~ Yor, ean be numerically determined by employing Weddle's rule, to give Verve = 0.3 Ap [Yj +5¥p+ Vet OY + ¥s+5¥o+ V1 (87) with Ap = PexPet (6-88) 2 [pwr + (= 1) APH (Hg Z)i (3-89)126 Hydrocarbon Phase Behavior where Lyon = gas viscosity and compressibility factor as evaluated at [bw + = 1) ap) ‘Equation 3-83 can be approximated by removing the term 1/(j4,Z) outside the integral and evaluated at the average pressure, to give (ug 2) or03ih gm = 07099 oy Tring Eee)? Iu? oe 0.703 kh(p - psc) (3-90) ThigZ)avg Lin(teltw) The expression (\4,Z)ayg is evaluated at the arithmetic average pressure e+ pal ‘The approximation method represented by Equation 3-90 is commonly known as the pressure-squared approximation method. Example 3-18. A natural gas with the following composition is flowing from a gas well at a bottom-hole flowing pressure of 2,800 psia. Component yi C 0.850 CG 0.055 cs 0.035 cor 0.010 Given the following additional data, Pe = 3,200 psia T= 150°F k= 50 md ha Bit 0.333 ft 666 ft calculate the gas flow rate by using the pressure-squared method. Solutio Step 1. Calculate average pressure Paws = (3,200 + 2,800)/2 = 3,000 psia Step 2. Calculate the apparent molecular weight of the gis, to give MW, = 17412 Properties of Natural Gases 127 Step 3. Calculate T,. and py of the gas, to yield io = 358.5, Pye = 672.5 Step 4. Determine the pseudo-reduced temperature and pressure of the gas, to give Teo 7 Por = 4.46 Step 5. Calculate the gas viscosity and compressibility factor, to yield ig = 0.0202 ep Z=0.89 Step 6. Calculate the gas flow rate from Equation 3-90. Q, = 25.331307 MM sefiday Gas Reservoir Material Balance ‘A material balance on a gas reservoir can effectively be used to determine ‘© The size of the gas reservoir © The ultimate gas recovery © The effect of water encroachment in the reservoir For a volumetric gas reservoir, ic., a reservoir with no water influx, the ‘material balance on the gas reservoir in terms of moles of gas is expressed as n= m1 91) where ny = moles of gas produced 1; = moles of gas initially in the reservoir ‘ny = moles of gas remaining in the reservoir ‘Representing the gas reservoir by an idealized gas container (shown sche- ‘matically in Figure 3-13), the terms in Equation 3-91 are replaced by their equivalents using the real gas law, to give eee ale bb oa EEL128 Hydrocarbon Phase Behavior Figure 3-13, A schematic illustration of an idealized volumetric gas reservoir where Gp = cumulative gas produced, scf Tye, Pe = standard conditions: 520°R and 14.7 psia V = gas pore volume, ft? i= initial reservoir pressure, psia 2% = initial reservoir pressure, psia 1p = current reservoir pressure, psia Z = current gas compressibility factor Since py, T, Zy and V are all constants for a volumetric gas reservoir, Equation 3-92 can be expressed as an equation of straight line, or G,=b-m 2 (3-93) where b = 520 p.Vi(14.7 ZT) m = 520 Vi(I4.7T) Equation 3-93 indicates that for a volumetric gas reservoir the graph of the ratio (p/Z) versus cumulative gas production G, on cartesian coordinate paper isa straight line relationship. ‘The gas formation volume factor B,, as expressed in sef/f* by Equation 3.64, can be incorporated into Equation 3-92 to give G, = ByV ~ BV (3-94) Properties of Natural Gases 129 ‘The term (ByV) represents the gas-initially-in-place as measured in sef, or G, = GUP - BV (3-95) where GIP = gas initially in place, sof, Equation 3-95 shows that the plot of G, versus Bisa straight line with a negative slope of V and with an intercept of GIIP. The areal extent of the reservoir, A, can be calculated from the slope V (gas pore volume) by apply- ing the mathematical definition of V, or V = 43,560 Ah & (1 = Syd) (6-96) where V = gas pore volume, ft? A-= areal extent of the reservoir, acres connate water saturation ‘B = reservoir porosity Example 3-19. A gas reservoir with an areal extent of 1,000 acres has an initial reservoir pressure of 3,000 psia. The reservoir has the following rock and fluid properties: = 0.95 Swe = 0.15; T= 150°F Calculate the cumulative gas production when the reservoir pressure drops to 2,500 psia. Solution. Step 1. Calculate the past pore volume from Equation 3-96 to give V = 92.565 MM f° Step 2. Calculate the gas compressibility factor at 3,000 and 2,500 psia to yield Za = 0.89 Zam = 0.86 Step 3. Solve for G, by applying Equation 3-92 to give Gy = 2,489.67 MM sof ‘The material balance as expressed mathematically by Equation 3-92 ean be modified to describe the volumetric behavior of gas reservoirs under wa- ter drive by incorporating the cumulative water influx We into the expres- sion as follows:130 Hydrocarbon Phase Behavior Pe Gp. BV _ pV - (W.-W) ZRT (v= We+ Wy) @7) where W, = cumulative water influx, ft ‘W, = cumulative water produced, f° The performance of a water drive gas reservoir can be represented sche- ‘matically by an idealized gas container as shown in Figure 3-14. Figure 3-14. A schematic illustration of an idealized water drive gas reservoir. Equation 3-97 can be expressed in terms of the gas formation volume fac- tor as follows G, = By V- B[V - We + Wy) (3-98) Since the term (B,V) represents the gas initially in place, Equation 3-98 can be written as UP ~ B, [V ~ W, + W,] + G99) Properties of Natural Gases 131 [our _ UIP ~ B, Wet W, (8-100) where GIIP = gas-initially-in-place, sof initial gas formation volume factor, scf/ft? current gas formation volume factor, scf/ft PROBLEMS: 1. Assuming an ideal gas behavior, calculate the density of n-butane at 290°F and 50 psia. 2. Show that _ (w/w) > CTD 3 Given the following gas Gy 0.65 G ous G 0.10 nC, 0.06 2 0.04 calculate a. Mole fraction of the gas by. Apparent molecular weight e. Specific gravity €1 Specific volume st 300 pra and 120°F by assuming an ideal gas behavior 4, An ideal gas mixture has a density of 1.92 Ib/f@ at 500 psia and 100°F. Caleulate the apparent molecular weight of the gas mixture. 5. Using the gas composition as given in Problem 3, and assuming real ‘gas behavior, calculate a. Gas density at 2,000 psia and 150°F b. Specific volume at 2,000 psia and 150°F c. Gas formation volume factor in seffft?12 Mydrocarbon Phase Behavior 6. A natural gas with a specific gravity of 0.75 has a gas formation vol- ‘ume factor of 0.00529 ft'sef at the prevailing reservoir pressure and temperature. Calculate the density of the gas. 7. A natural gas has the following composition on 0.75 0.10 0.05 0.04 0.03 0.02 0.01 Reservoir conditions are 8,500 psia and 200° * Calculate a. Isothermal gas compressibility coefficient b. Gas viscosity by using (1) Carr-Kobayashi-Burrows Method (2) Lee-Gonzales-Eakin Method (9) Dean-Stiel Method 8. Given the following gas composition Component a CO; 0.06 Ne 0.03 G 0.75 G 0.07 cy 0.04 ny 0.03 Cy, 0.02 If the reservoir pressure and temperature are 2,500 psia and 75°F respectively, calculate ‘a, Gas density, by accounting for the presence of non-hydrocarbon components by using (1) Wichert-Aziz Method (2) Carr-Kobayashi-Burrows Method b, Isothermal gas compressibility coefficient ©. Gas viscosity by using (Q) Carr-Kobayashi-Burrows Method (2) Lee-Gonzales-Eakin Method (8) Dean-Stiel Method Properties of Natural Gases. 133 9. Forty sef of gas are placed in a 0.5 ft cylinder at a temperature of 80°F, The cylinder is fitted with a pressure gauge. The gauge shows a pressure reading of 1,000 psia. Calculate the gas compressibility fac- tor. What is the gas formation volume factor in f° /scf? 10, A gas well is producing at a rate of 1.22 mm seffday. The specific ‘gravity of the producing gas is 0.74. If the average reservoir pressure and reservoir temperature are 2,000 psia and 195°F, calculate a. Gas flow rate in f/day b. Gas viscosity under reservoir conditions LL, A high-molecular-weight natural gas has the following composition: Component Te Gq 0.73 G 0.10 Gs 0.05 G 0.03 ce 0.03 Co 0.02 G, 0.04 If the molecular weight and specific gravity of C;, are 135 and 0.81, calculate a. Specific gravity of the gas mixture b. Density of the gas at 2,000 psia and 120°F 12, A gas well is producing a hydrocarbon gas at a bottom-hole pressure (of 3,100 psia. The reservoir pressure and temperature are 3,420 psia and 160°F. Given the following additional well and gas properties data, gas gravity = 0.700 k= 29 md Te = 660 ft Tw = 0.25 ft h= 12 ft calculate the gas flow rate by using a. The real gas potential approach b. The pressure-squared method 13. A volumetric gas reservoir has the following gas and rock properties: h= 26 ft = 0.16 bg = 0.72 T= 130°F Calculate cumulative gas produced when reservoir pressure drops to 3,850 psia,134 Hydrocarbon Phase Behavior 4. A volumetric gas reservoir has the following pressure-production his- tory: P G, _psia_ MM sf 3,000 100 2,500 500 ‘The reservoir temperature is 170°F. Ifthe specific gravity of the gas is 0.70, calculate a. Initial gas formation volume factor b. Initial gas density ©. Gas initially-in-place in sef 4. Cas pore volume €. Areal extent of the reservoir given = 0.19 Se = 0.24 b= 12 ft 1, Brown, et al, “Natural Gasoline and the Volatile Hydrocarbons,” Tulsa: NGAA, 1948, 2. Carr, N., Kobayashi, R., and Burrows, D., “Viscosity of Hydrocarbon Gases Under Pressure,” Trans. AIME, 1954, Vol. 201, pp. 270-275. 3. Dean, D. E, and Stiel, L. I, “The Viscosity of Non-polar Gas Mixtures fat Moderate and High Pressure,” AIChE Jour., 1958, Vol. 4, pp. 430- 436, 4, Dempsey, J. R., “Computer Routine ‘Treats Gas Viscosity as a Variable,” Oil and Gas Journal, Aug. 16, 1965, pp. 141-143. 5. Dranchuk, P. M. and Abu-Kassem, J. H., “Calculation of Z-factors for Natural Gases Using Equations-of-State,” JCPT, July-September, 1975, pp. 34-36. 6. Dranchuk, P. M., Purvis, R. A., and Robinson, D. B., “Computer Cal- culations of Natural Gas Compressibility Factors Using the Standing and Katz Correlation,” Inst. of Petroleum Technical Series, No. IP 74- 008, 1974. 7. Hall, K. R. and Yarborough, L., “A New Equation-of-State for Z-factor Calculations,” Oil and Gas Journal, June 18, 1973, pp. 82-92. 8. Hankinson, R, W., Thomas, L. K., and Phillips, K. A., “Predict Natural Gas Properties,” Hydrocarbon Processing, April 1969, pp. 106-108. Properties of Natural Gases 135 9. Kay, W. B., “Density of Hydrocarbon Gases and Vapor,” Industrial and Engineering Chemistry, Vol. 98, 1936, pp. 1014-1019. 10. Lee, A. L., Gonzalez, M. H., and Eakin, B. E., “The Viscosity of Natu- ral Gases,” Journal of Petroleum Technology, August 1966, pp. 997- 1000. 11. Mattar, L. G., Brar, S., and Az “Compressbility of Natural Gases," Journal of Canadian Petroleum Technology, October-Decem- ber, 1975, pp. 77-80. 12, Papay, J., “A Termelestechnologiai Parameterek Valtozasa a Gazlelepk Muvelese Soran,” OGIL MUSZ, Td, Kuzl., Budapest, 1968, pp. 267- 273, 13, Standing, M. B., Volumetric and Phase Behavior of Oil Field Hydrocar- ‘bon Systems, Dallas: Society of Petroleum Engineers, 1977, pp. 125. 1236, 14, Standing, M. B., Volumetric and Phase Behavior of Oil Field Hydrocar- bbon Systems, Dallas: Society of Petroleum Engineers of AIME, 1977, pp. 192. 15, Standing, M. B. and Katz, D. L., “Density of Natural Cases,” Tran AIME, Vol. 146, 1942, pp. 140-149. 16. Stewart, W. F, Burkhard, S. F, and Vo, D., “Predi Critical Parameters for Mixtures,” Paper presented at the AIChE Meet- ing, Kansas City, MO (1959). 17, Sutton, R. P,, “Compressbility Factors for High-Molecular-Weight Res- cervoir Gases,” Paper SPE 14265, Presented at the 60th Annual Technical Conference and Exhibition of the Society of Petroleum Engineers, Las ‘Vegas, September 22-25, 1985. 18. Takacs, G., “Comparisons Made for Computer Z-factor Calculation,” Oil and Gas Journal, Dee. 20, 1976, pp. 64-66. 19. Wichert, E. and Aziz, K., “Calculation of Z's for Sour Gases,” Hydro- carbon Processing, Vol. 51, No. 5, 1972, pp. 119-122. 20, Trube, A. S., “Compressbility of Natural Gases,” Trans. ALME, Vol. 210, 1957, pp. 355-357, 21, Yarborough, L. and Hall, K. R., “How to Solve Equation-of State for ‘Zfactors,” Oil and Gas Journal, February 18, 1974, pp. 86-88.4 Phase Behavior of Crude Oils Petroleum (an equivalent term is “crude oil") is a complex mixture consist- ing predominantly of hydrocarbons, and containing sulfur, nitrogen, oxy- gen, and helium as minor constituents. ‘The physical and chemical proper- ties of crude oils vary considerably and are dependent on the concentration of the various types of hydrocarbons and minor constituents present. ‘An accurate description of physical properties of crude oils is of consider- able importance in the fields of both applied and theoretical science and es- pecially in the solution of petroleum reservoir engineering problems. Physi- ceal properties of primary interest in petroleum engineering studies include: ** Fluid densities * Isothermal compressibility *# Solution gas-oil r ‘* Oil formation volume factor '¢ Fluid viscosities ‘ Bubble-point pressure * Surface tension Data on most of these fluid properties is usually determined by laboratory experiments performed on samples of actual reservoir fluids. In the absen of experimentally measured properties of crude oils, itis necessary for the petroleum engineer to determine the properties from empirically derived correlations. (CRUDE OIL DENSITY AND SPECIFIC GRAVITY Evaluating the volumetric phase behavior of oil reservoirs requires accu- rate knowledge of the physical properties of the crude oil at elevated pres- sure and temperature. Pertinent among the properties of interest are the density and the specific gravity of the crude oil 136 Phase Behavior of Crude Oils. 137 ‘The crude oil density is defined as the mass of a unit volume of the crude at a specified pressure and temperature. Itis usually expressed in pounds per cubic foot. The specific gravity of a crude oil is defined as the ratio of the density of the oil to that of water. Both densities are measured at 60°F and atmospheric pressure. nee (1) where 7 = specific gravity of the oil density of the crude oil, Ibift? jensity of the water, Ibi? Although the density and specifie gravity are used extensively in the pe- troleum industry, the API gravity i the preferred gravity scale, This gravity scale is precisely related to the specific gravity by the following expression: MLS % PAPI = 1315 (42) ‘The API gravities of crude oils usually range from 47°API for the lighter crude oils to 10°API for the heavier asphaltic erude oils. During the last forty years, numerous methods of calculating the density of crude oils have been proposed. There are two approaches available in the literature to calculate liquid density: © The equation-of-state approach (discussed in Chapter 6) ‘© The liquid density-correlation approach. ‘The second approach of determining liquid density is presented as follows: METHODS FOR DETERMINING DENSITY OF CRUDE OILS OF KNOWN COMPOSITION Several reliable methods are available for determining the density of com- plex hydrocarbon mixtures from their composition. The best known and ‘mast widely used methods in the petroleum industry are those of Standing- Katz (1942) and Alani-Kennedy (1960) [howe Standing Katz Method _// Standing and Katz (1942) proposed a graphical correlation for determin- ing the density of hydrocarbon liquid mixtures. Standing and Katz devel-198 Hydrocarbon Phase Behavior oped the correlation from evaluating experimental, compositional, and den- sity data on 15 crude oil samples containing up to 60 mole-percent methane. ‘The proposed method yielded an average error of 1.2% and maximum error (of 4% for the data on these crude oils. The original correlation did not have a procedure for handling significant amounts of non-hydrocarbons. ‘The authors expressed the density of hydrocarbon liquid mixtures as a function of pressure and temperature by the following relationship: Tos = bu + pp ~ Aor] (43) where pp = erude oil density at p and T, Ibift? bx = crude oil density at standard conditions, i.e., 14.7 psia and 60°F, Ibite ‘App = density correction for compressibility of oils, Ibift? Apr = density correction for thermal expansion of oils, Ibift* Standing and Katz correlated graphically the liquid density at standard conditions with: © The density of the propane-plus fraction pcs, ¢ The weight percent of methane in the entire system (mej), ‘¢ The weight percent of ethane in the ethane-plus (me,)cy ‘This graphical correlation is shown in Figure 4-1. The following are the specific steps in the Standing and Katz procedure of calculating the liquid density at a specified pressure and temperature. Step 1. Calculate the total weight and the weight of each component in one Ib-mole of the hydrocarbon mixture by applying the follow- ing relationships: m= x MW, (4) m= Ex MW, 45) where m, = weight of component i in the mixture, Ib/Ib-mole x; = mole fraction of component i in the mixture MW, = molecular weight of component i 1m, = total weight of one Ib-mole of the mixture, Ib/lb- male Step 2. Calculate the weight percent of methane in the entire system and the weight percent of ethane in the ethane-plus from the follow- ing expressions: (made. = [23] 100 ce) Phase Behacior of Crude Oils 139 Density of Propane Plus—Ib per eu ft cates Pre acr Conage Density of System including Methane and Ethane—tb per cu ft, Figure 4-1. Standing’s density correlation. From Volumetric and Phase Behavior (of Oi Field Hydrocarbon Systems, Ninth Edition, Permission to publish by the So- Ciely of Petroleum Engineers of AIME. Copyright SPE-AIME, 1981, and <2} 100 = M244 where (mc,)c,, = Weight % of methane in the entire system ‘me, = weight of methane in one Ib-mole of the mix- t (meg)ee, = (4-7) me,)cy, = Weight ‘% of ethane in ethane-plus ‘mc, = weight of ethane in one Ib-mole of the mix- ture Step 3. Calculate the density of the propane-plus fractions at standard conditions by using the following equations:140 Hydrocarbon Phase Behavior Bes (4-8) ma, = Lm 9) Vo.- EVES (4-10) where pc, = density of the propane and heavier components, Wife ‘mcy, = weight of the propane and heavier fractions, tb/ Ib-mole Veq, = volume of the propane-plus fraction, ft/Ib-mole V, = volume of component i in one Ib-mole of the mixture pa density of component i at standard conditions, Ib/f®. Density values for pure components are tabulated in Table 1-1 in Chapter 1, but the den- sity of the plus fraction must be measured. Step 4. Using Figure 4-1, enter the pc, , value into the left ordinate of the chart and move horizontally to the line representing (mo,)co,» then drop vertically to the line representing (me,)c, ,. The den- sity of the oil at standard condition is read on the right side of the chart, Step 5. Correct the density at standard conditions to the actual pressure by reading the additive pressure correction factor, Ap, from Fi ure 4-2, Step 6, Correct the density at 60°F and pressure to the actual tempera- ture by reading the thermal expansion correction term, Apr, from Figure 4.3. Example 4-1. A crude oil system has the following composition, Component, Ei Gy 0.45 G 0.05 Gs 0.05 G 0.03 Cs 0.01 : Ce 0.01 Cy 04 Phase Behavior of Crude Oils 141 ty at 60°F and 14.7 psi, fo per eu tt Density at pressure minus densi Density at 60°F and 14.7 pela, tb por eu ft Figure 4-2. Density correction for compressibility of crude oll. Courtesy of the Gas Processors Suppliers Association. Published in the GPSA Engineering Data Book, Tenth Edition, 1987. M If the molecular weight and specific gravity of C;, fractions are 215; and Q.87, respectively, calculate the density of the crude oil at 4,00) psia and 160°F by using the Standing and Katz method. i Solution. Component x, MW; G 0.45 16.04 CG 0.05 30.07 C 0.05 44.09 * From Table (ecample £1 continued on next page)142 Hydrocarbon Phase Behavior * 4) (Peoreg: 493% * = a Oa ee ee en eee = = © Es . = cs = « aa mat adPse aac Figure 4-3. Density correction for isothermal expansion of crude oils. Courtesy of the Gas Processors Suppliers Association. Published in the GPSA Engineer- ing Data Book, Tenth Edition, 1987. Phase Behavior of Crude Oils 143, 4 a Example 4-1, Continue. iw ae coo Component MW, m= MW: pa lb/ft’? _Vi= mila CG 0.03 58.12 1.7436) 38.71 0.0888 Cy 0.0L «72:15 0.7215 39.08 0.0185 C001 86.17 0.8617 41.38 0.0208 Cy 040 215.0 _8 54.288" im = 100.253 * ore = (O.8T)(G2.4) = 54.288 Step 1. Calculate the weight % of Cy in the entire system and the weight ‘ of Cz in the ethane-plus fraction from Equations 4-6 and 4-7 a 7.218 | 99. pga pet PETME (me)er. (za 100 = 7.2% > om = |_1:5035__} 109 = 1.616% ranen Att (ca)ens | Tig | 100= 1.616% wnen nel © Step 2. Calculate the density of the propane and heavier by using Equa- tion 4-8. 100.253 ~ 7.218 ~ 1.5035. _ 55 55} 2g 7 BS Ibi? Step 3. Determine the density of the oil at standard conditions from Fig- ure 4-1 Ape = 47.5 Ibife Step 4. Correct for the pressure by using Figure 4-2. Ap, = 1.18 Ibife * Density of the oil at 4,000 psia and 60°F is then calculated by the expression Ppto Pun + Spy = 47.5 + 1.18 = 48.68 IDF Step 5. From Figure 4-3, determine the thermal expansion correction factor Apr = 2.45 Ibift? Step 6. The required density at 4,000 psia and 160°F is Po = 48.68 — 2.45 = 46.93 Ib/f?144 Hydrocarbon Phase Behavior With the apparent need for a mathematical description of the Standing and Katz charts, Standing (1977) developed the following expressions for de- termining pro App, and Spr. ‘The mathematical expression for pa. Pu = Pop, {1 — 0.012 (me,)ey, ~ 0.000158 (me,)é .] + 0.0133 (me,)c, , + 0.00058 (me)& (4-1) with ex, = Peay [1 ~ 0.02386 (mies)es, ~ 0.000082 (Ine), ] + 0.379 (mes)cp, + 0.0042 (me,)ty, (4-12) where pcp, is the density of the ethane-plus fraction. ‘The mathematical expression for App App = [0.167 + (16.181) 10-095] (al (4-13) Ps = 0.0608 oye) |_P_| 0.01 [0.299 + (263) 10 7 (ea ‘The mathematical expression for Ap Apr = (0.0133 + 152.4 (pe + Bp,)~*] (T — 520) ~ {8.1 (10-8) ~ (0.0622) 10-7 4p] (T — 520)? (4-14) where T is the temperature in °R. ‘To account for the effect of the presence of non-hydrocarbon components, (HS, CO,, and N,) on the Standing and Katz crude oil density, Pedersen et _ al. 184) proposed the following correction procedure: Step 1. Calculate the density of (HS + Cy.) fraction at standard condi- tions from the following expression XY & Mwy) (HS + Cy.) = = Ibi? (4-15) rE where the index i refers to HS, Cs, and heavier components. P Step 2. Calculate the weight % of ethane in the (HS + Cz») from the following relationship: Step 3. Step 4 Step 5. Phase Behavior of Crude Oils 145 =| ed (none. ~ [petite where MWe, = molecular weight of ethane Xo, = mole fraction of ethane i = refers to H;S, Co, and heavier components Calculate the density of (HS + Cz.) fraction at standard condi- tions from the relationship (416) Pascas) * Pauses) ~ Aa ~ Ana ~ A 417) where Ay = 0.1971 (mey)iys +02, Ay = = 0.1612 (me)iys cay ‘Az = 0.0081 (moans Co. a1 = 3.3 ~ 0.0801 puigseca,) ‘a = 1+ [0.94098 pyugs+cs,) ~ 6.0] (0.0401 puys+ca,) ~ 215) Based on the additive volume concept and @ CO; density of 51.26 Ibift, the density of (COs + HaS + C2,) fraction is calculated at standard conditions by using the additive volume concept and employing the density of (HS + Cz.) as calculated in the pre- vious step, to give co MWeo, +E x MW, Poogeiiass ces (4-18) x» MW, xo, Won, Poo, PlgsoGae where the index i refers to H,S, Cz, and heavier components Caloulate the weight % of C; + Nz in the total mixture from (orc, ende,3 2 MWe My aunaael 5 MW, 100 (419) where Xo, = mole fraction of methane in the mixture MWe, = molecular weight of methane Xx, = mole fraction of nitrogen in the mixture MWg, = molecular weight of nitrogen Ey. % MW, = total weight of the hydrocarbon system, Ibilb-mole16 Hydrocarbon Phase Behavior Step 6. Calculate the density of the crude oil at standard conditions, in pounds per cubie foot, from the following equation: Pe ™ Poostiss cay ~ Bo ~ Bi by (4-20) where By = 5.507112 ~ 5.95976 by + 0.46195 by ~ 0.37627 by By = 8.86573 - 9.97002 b; + 0.41677 by + 0.07257 by by = — 0.65 + 0.01603 pco,ngs+c2y b= 1-0.) (me snver, + 0.015 (meyens)b,, ~ 0.3 (mey+nader. by = 1 ~ 0.6 (me sno, + 0.075 (me, +n), ~ 0.0025 (me, +ualea , Step 7. ‘The density of the crude oil at standard conditions can be cor- rected for an increase in pressure and temperature by applying Equations 4-13 and 4-14 or by using Figures 4-2 and 4-3. Example #2. Calculate the density of the crude oil with the composition given below at 3,000 psia and 120°R Component a Component —& co; 0.07 @ 0.04 Ne 0.03 G 0.03 HS 0.02 G 0.02 a ol Ce 0.03 G 0.04 Gy 0.31 aMW pa Vie uMWiou 3.0808 51.26/ 0.0601 0.8403 = == 0.6816 49.30 0.0138 6.5764 - - 1.2028 = = 1.7636 31.64 0.0557 1.7436 35.71 0.04883, 1.4430 39.08 0.0369 2.5851 41.36 0.9625 62.00 54.3 1.1418 m = 61.9171 1.41963 Phase Behavior of Crude Oile 147 Step 1. Calculate the density of (HS + Cs.) fraction 70.2160 7 70.2169 51.64 tit Tatoes - 0.0601 ~ °! Step 2, Calculate the weight % of Cs in the (HS + C2.) fraction from Equation 4-16. Puigsca.) = 1.2098 71.A197, / Step 3. Calculate the coefficients Ay, A1, As, a1, and a as illustrated in Step 3 of the procedure. Ay= 0.3311 A,= = 0.2708 Ay = 0.0153 ay = — 83636 ay = 0.56315 Step 4. Calculate the density of (HS + Cz.) fraction by applying Equation 4-17. Pigs cg4) = 51.64 ~ .3311 - ( - 0.9708)( ~ .83636) = (0.0153)(0.56315) = 51.07 Ibift? Step 5. Calculate the density of (COy + HaS + Cy.) fraction by using Equation 4-18, to yield 9 Poogr is scyy = Soa AAT 5,08 bit _ 1.07 Step 6. Determine the weight % of C, and N; in the entire system by applying Equation 4-19, to give (mcysnor, = 9.05% 7 Step 7. Caleulate the parameters by, bs, bs, by, By, and B,. by = 0.16881 by = 0.095 by = ~ 0.4865 by = - 0.1404 Bo = 4.769 B, = 7.7625 Step 8. Calculate the density of the oil mixture of standard conditions by applying Equation 4-20. Pac = 51.08 ~ 4.769 — (7.7625)(0.16881) = 45.0 Ibi/fe Step 9. Determine the pressure correction from Figure 4-2 App = 1.06 Ibift®M8 Hydrocarbon Phase Behavior Step 10. Determine the temperature correction from Figure 4-3. Aor = 18 Ibift? Step 11. Calculate the density of the crude at 3,000 psia and 120°F by applying Equatjon 4-3. po = 45.0 + 1.06 ~ 1.8 = 44.26 Ibif? / ‘The Alani-Kennedy Method ‘Alani and Kennedy (1960) developed an equation to determine the molal liquid volume V., of pure hydrocarbons over a wide range of temperature and pressure. The equation was then adopted to apply to crude oils with the heavy hydrocarbons expressed as heptanes-plus fraction, i.c.. Cr. ‘The Alani-Kennedy equation is similar in form to the Van der Waals equation, which takes the following form: va vz42¥a_ab_o (4-21) P PP where gas constant, 10.73 psia ft°/Ib-mole °R T = temperature, °R p= pressure, psia V,, = molal volume, ft/lb-mole a, b = constants for pure substances ‘Alani and Kennedy considered the constants a and b as functions of tem- perature and proposed the expressions for calculating the two parameters. a-ket | (4-22) bemT+e! (4-23) where K, n, m, and care constants for each pure component. Values ofthese constants are tabulated in Table 4-1. ‘Table 4-1 contains no constants from which the values of the parameters a and b for heptanes-plus can be calculated. Therefore, Alani and Kennedy proposed the following equations for determining a and b of C:, 3.8405985 (10-9)(MW)c,, ~ 9.5638281 (10*) pas or In (acy, 7 + 261.80818 4 7.31 04464 (10-)(MW)é,, + 10.759517 (4-24) Phase Behavior of Crude Oils 149 Table 4-1 Alani-Kennedy Constants ‘Component « a max 10" e ©, 70°-s00°F —_—9,160.6419 ersssezs ——«a.gtena72 0 C.g01e-460"F —'147.47393 —«9,247.4593—~ 14.072897 | 1.8328605 ©; 100°-240%F — 46,709.573 = 404.4608 5.1520981 0.52299654 (G;.250°-460°F —17,496.343 ‘34.163551 2.8201736 0.62900877 cy 20287.757 +190.24420 2.1586448 0,90832519 Hy 32,208.420 13183171 3.3962284 1.101384 nc 3016 212 146.15445 2902187 1t68tad ty 37,086.254 200152630 21954785 14964209 no, 37,048.234 299.62630 2.1954785 14364289 a6, 52,088.008 254.56007 S.e961858 15929406 HeS* 13,200.00 ° 17.800 013945 Ne 4300.00 2203, 4.490 0.3063 Oe" 8,166.00 126.00 tgieo 0.3872 be, = 0.03499274 (MW)c,, ~ 7.275403 (y)o,, + 2.232395 (10-*)T [naw ~ 0.016922572 ba + 6.2256545 (4-25) ere 7 where MWo,, = molecular weight of Cr. ‘yor = Specific gravity of Cr, ‘4cz,» be;, = constants of the heptanes-plus fraction, For hydrocarbon mixtures, the values of a and b are calculated using the following mixing rules Ss am = aX (4-26) bas § bx (4-27) where a, and b, refer to pure hydrocarbons at existing temperature and x; is their mole fraction in the mixture, The values aq and by are then used in Equation 4-21 to solve for the molal volume, Vi. The density of the mixture at pressure and temperature of interest is determined from the following re- lationship150 Hydrocarbon Phase Behavior oan te (438 where _p. = density of the crude oil, Ib/ft? MW, = apparent molecular weight ‘Ve = molal volume, ft/lb-mole ‘The Alani and Kennedy method for calculating the density of liquids is summarized in the following steps: Step 1. Caleulate the constants a and b for each pure component from Equations 4-22 and 4-23, respectively. Step 2. Determine ac,, and bo,, by applying Equations 4-24 and 4-25, Step 3. Solve for ap, atid by, from Equations 4-26 and 4-27. Step 4. Calculate the molal volume V,, by solving Equation 4-21 for the smallest real root. Step 5. Compute the apparent molecular weight, MW,. Step 6. Determine the density of the crude oil by using Equation 4-28 ‘The average absolute deviation is 1.6% with a maximum error of / Example 4-3. A crude oil system has the composition: Component a CO; 0.0008 Ne 0.0164 G 0.2840 G 0.0716 G 0.1048 ie 0.042 ay 0.042 in 0.0191 ns 0.0191 Cy 0.0405 Gy 0.3597 Given the following additional data MWo,, = 252 Yop, = 0.8424 pressure = 1708.7 psia temperature = 591°R Caleulate the density of the crude cil : Phase Behavior of Crude Oils 181 Solution. Step 1. Calculate the parameters ac,, and be,, from Equations 4-24 and 4-25, to give fo, = 229269.9 bey, = 4.165811 Step 2. Calculate the mixture parameters ay, and by from Equations 4-26 and 4-27, &q = 9LLLTL * ba = 2.119383 Step 3. Solve Equation 4-21 for the molal volume Vu = 2.598417 ¥ Step 4. Determine the apparent molecular weight of this mixture. MW, = 113.5102 / Step 5. Compute the density of the oil system by using Equation 4-28 = 113.5102 = Syppar 7 14-896 lbit "METHODS FOR DETERMINING DENSITY OF LIQUIDS OF UNKNOWN COMPOSITION Several empirical correlations for calculating the density of liquids of un- known compositional analysis have been propesed. The correlations employ limited PVT data such as gas gravity, oil gravity, and gas solubility as corre- lating parameters to estimate liquid density at the prevailing reservoir pres- sure and temperature. Katz's Method ‘The density, in general, can be defined as the mass of @ unit volume of ‘material at a specified temperature and pressure. Accordingly, the density of a saturated crude oil at standard conditions can be defined mathematically by the following relationship: weight of stock tank oil + weight of solution gas =” Solame of Rock tank ol + increase in-stock tank volume due to wlation gat152 Hydrocarbon Phase Behavior i, mo +m, Pe Wale (BV, where px: = density of the oil at standard conditions, Ib/ft? (Velue = volume of oil at standard conditions, {0/STB total weight of one stock-tank barrel of oil, Ib/STB ig = Weight of the solution gas, Ib/STB (AVo)qe = increase in stock-tank oil volume due to solution gas, f/ STB 14.7 psia Volume of the Viquified Ryscf oF the solution gas 14.7 psia Ry sef ® 60°F 1513 Figure 4-4. Schematic illustration of the Katz's density model at standard condi- tions ‘The procedure of calculating the density at standard conditions is illus- trated schematically in Figure 4-4. Katz (1942) expressed the density of rude oil at standard conditions with the following relationship: pa = at aN? (4-29) Wet BE NG Pa where pp, as introduced by Katz, is the apparent density of the liquefied dissolved gas at 60°F and 14.7 psia. Katz correlated the apparent gas den- sity, in Ib/ft8, with the specific gravity, the solution gas, and the APT gravity of the stock-tank oil as shown graphically in Figure 4-5. The proposed ‘method does not require the composition ofthe crude oll. The only required properties are the gas gravity, the API gravity, and the gas solubility. ‘To arrive at the final expression for calculating p,., let: R, = gas solubility, sefiSTB Ye = Bas gravity s ‘ie = oll gravity Phase Behavior of Crude Oils 153 Figure 4-5. Kat's apparent density of cisolved gas. Courtesy ofthe Ameri Petroleum Institute, Katz, D., Drilling and Production Practice, 1942, p. 137, a ‘The weights of the solution gas and the stock-tank oil can be determined in terms of the above defined variables by the following relationships R = Fy 3-904). Ib of solution gas/STB rm, = (5.615)(62.4)(y,), Ib of oil/STB Substituting the above terms into Equation 4-29 yields waren «[H-lescoy s.ts-+ [las e6t/0u eae | as0.st07.+ fal — wa} fa (4-30)154 Hydrocarbon Phase Behi “The pressure correction adjustment, 4p,, and the thermal expansion ad- justment, Apr, for the calculated p. can be made by using Figures 4-2 and 4-3, respectively. Standing (1981) showed that the apparent liquid density of the dissolved gas as represented by Katz's chart can be closely approximated by the fol- lowing relationship: (88.52) 10(-oo AMD 4 [94.75 ~ 93.93 Log (API)JLog(y,) (4-31) Example 4-4, A crude oil at its bubble point pressure of 4,000 psia and a temperature of 180°F has an API gravity of 50°, ifthe gas solubility at the bubble-point pressure and system temperature is 650 scf/STB and the spe- cific gravity of solution gas is 0.7. Calculate the oil density at the specified pressure and temperature by using Katz's method, Solution. Step 1. From Figure(4-4, determine the apparent density of dissolved gas. be = 20.5 Ibife Step 2. Calculate the stock-tank liquid gravity from Equation 4-2. 141.5 141.5 = 3 _ 0.7796 "APIs 1315 "50+ 1915 Step 3. Apply Equation 4-30 to calculate the liquid density at standard conditions. (950.376)(0.796) + 200-0) 13-19 42,12 Ibift? (65007) 03.0.5) Step 4. Determine the pressure correction factor from Figure 4-2. pp = 1.55 Ibift Step 5, Adjust the oil density, as calculated at standard conditions, to res- ervoir pressure. Powe = 42.12 + 1.55 = 43.67 Ibift? Step 6. Determine the isothermal adjustment factor from Figure 4%. Apr = 3.25 Ibift? Step 7. Calculate the oil density at 4,000 psia and 180°F. bo = 43.67 ~ 3.95 = 40.42 Ib/f Pe= 5.615 + Phase Behavior of Crude Oils 155 Standing’s Method Standing (1981) proposed an empirical correlation for estimating the oil formation volume factor as a function of the gas solubility R,, the specific Eravity of stock-tank oil, the specific gravity of solution gas 7, and the - system temperature ‘. By coupling the mathematical definition of the oil formation volume factor (as discussed in a later section) with Standing’scor- relation, the density of a crude oil at a specified pressure and temperature ean be calculated from the following expression: 62.4 yo + 0.0136R, 7g NR, (4 ir P= — (3), 0.072 + 0.000147 where T = system temperature, °R “= specific gravity of stock-tank oil 1 Example 4-5. Rework Example 4-4 and solve for the density by using Standing’s Correlation. Solution. From Equation 4-32; 62.4(.7796) + 0.0136(650)(0.7) ray S02 Ibe: ee ol ra Ahmed's Correlation Ahmed (1985) developed a correlation for estimating the erude oil density at standard conditions, The correlation is based on calculating the apparent molecular weight of the oil from the readily available PVT on the hydrocar- bon system. Ahmed expressed the apparent molecular weight of the erude by the following relationship: |e, = 20008 Bag MWe (4-33) where MW, ~ apparent molecular weight of the oil MW. = molecular weight of the stock-tank oil and can be taken as the molecular weight of the heptanes-plus fraction. 7. “to = specific gravity of the stock-tank oil or the C;,, fraction156 Hydrocarbon Phase Behavior “The density of the oil at standard conditions can then be determined from the expression 2. = ee (4-34) : 199.71432| sens m+ os ESE If the molecular weight of the stock-tank oil isnot available, the density of the oil at standard conditions can be estimated from the following equa- tion: 0.0763 Ry 75 + 350.4 . Pee" 9 0087 R, + 2.4893 7, + 9.401 (4-35) Example 4-6. Using the data given in Example 4-4, caleulate the density of the crude oil by using Equation 4-35. Solution Step 1. Calculate the density of the crude oil at standard conditions from Equation 4.35, 0.0763(650)(0.7) + 350.400.7796) __ 40 g ib 710027 (650) + 2.4895(0.7796) + 3. Step 2. Determine Ap, from Figure 4-2. Ap, = LS Ibe Step 3. Determine Apr from Figure 4-3, Apr = 3.6 Ibift? Step 4. Caleulate p, at 4,000 psia and 180°F. po = 42.8 + 1.5 ~ 9.6 = 40.7 bite ISOTHERMAL COMPRESSIBILITY COEFFICIENT ‘OF UNDERSATURATED CRUDE OILS [sothermal compressibility coefficients are required in solving many reser- voir engineering problems, including transient fluid flow problems, and Xulso they are required in the determination of the physical pfoperties of the undersaturated crude oil Phase Behavior of Crude Oils. 157 By definition, the isothermal compressibility of a substance is defined For a crude oil system, the isothermal compressibility coefficient is given by the following equation: + + where C, = isothermal compressibility of the crude oil, psi~! (@V/ap)r = slope of the isothermal pressure-volume curve. ‘According to Equation 4-36, the isothermal compressibility coefficient is defined as the rate of change in volume with pressure increase per unit vol- ‘ume of liquid, all variables other than pressure being constant, Generally, isothermal compressibility coefficients of an undersaturated crude oil are determined from a laboratory PVT study. A sample of the crude oil is placed in a PVT cell at the reservoir temperature and at a pres- sure greater than the bubble-point pressure of the crude oil. At these initial conditions, the reservoir fluid exists as a single-phase liquid. The volume of the oil is allowed to expand as its pressure declines. This volume is recorded and plotted as a function of pressures. A typical pressure-volume relation- ship is shown in Figure 4-6. If the experimental pressure-volume diagram for the oil is available, C, can be calculated by graphically determining the volume V and the slope (@V/dp)r and then applying these values in Equa- tion 4-36. Figure 4-6. A typical p-V diagram fora crude158 Hydrocarbon Phase Behavior todo Gaelting G.__— 1 ‘Trube’s Correlation. Trube (1957) presented a correlation for calewating vie ed mel pecudo reduced compresibility C, of undersaturated crude ote ube correlated this property graphically with the pseudo-reduced Shere and temperature, Py and Tyr as shown in Figure 47. ‘The author wined C. mathematically by the following expresion: C= Pe (37) E cr E sent noma +4 © Pseudo Reduced Compress Figure 4-7. Tubo's pseudo-educed compressbilly of undersaturated ese Figure st esion to publish by the Society of Petroleum Engineers of AIME. Copy- right SPE-AIME. [Additionally Trube presented graphical correlations, as show in Figures oct co, toestimate the proudo-critcal properties of erude oils. The cal ‘ilation procedure ofthe propesed method is summarized in the following steps: Phase Behavior of Crude Oils 159 1 shes URE AT 60°F ‘BUBBLE POINT PRESS! Hert HEEtLLH PSEUDO CRITICAL TEMPERATURE, *R Figure 4-8. Ttube's pseudo-critical temperature correlation, Permission to pub- fish by the Society of Petroleum Engineers of AIME. Copyright SPE-AIME. Step 1. From the bottom-hole pressure measurements and pressure-tra dient data, calculate the specific gravity of the reservoir oil from the following expression: dpidh Grobe = Faas (4-38) where (ya)r = specific gravity at reservoir pressure and te perature T dpldh = pressure gradient as obtained from a pressure buildup test, psifft Step 2, Adjust the calculated specific gravity (Equation 4-38) to its value at 60°F by using the following equation. (rao = (re + 0.00046 (T ~ 520) (4-39) at160 Hydrocarbon Phase Behavior where (y.)a0 = adjusted specific gravity to 60°F T = reservoir temperature, °R Step 3. Determine the bubble-point pressure p, of the crude oil at reser- voir temperature. The following convenient correlation, as pro- posed by Standing (1981), can be used to estimate p, if the experi- ‘mental value is not known. aia a 1s (eso) In where R, = gas solubility, sof/STB ‘ag ~ specific gravity of solution gas T = temperature, °"R API = stock-tank oil API gravity Step 4. Correct the calculated bubble-point pressure py at reservoir tem- perature to its value at 60°F by using the following equation as, proposed by Standing (1962): (4-41) Sect ovo Una Figure 4-9. Trube's pseudo-citical properties correlation. Permission to publish by the Society of Petroleum Engineers of AIME, Copyright SPE-AIME. Phase Behavior of Crude Oils 161 where (ps)oo = bubble-point pressure at 60°F, psi Ps = bubble-point pressure at reservoir temperature, pia ‘T = reservoir temperature, °R Step 5. Enter in Figure 4-8 the values of (ps)eo and (yo) and determine the pseudo-critical temperature, Ty., of the erude. Step 6. Enter the value of Ty. in Figure 4:9 and determine the pseudo critical pressure p,e of the erude. Step 7. Calculate the pseudo-reduced pressure p,, and temperature T, from the following relationships: T te (4-42) Pye B (4-43) Pre Step 8. Determine C, by entering Figure 4-7 with the values of Tye and Pre Step 9. Calculate C, from Equation 4-37, cc Pre ‘Trube did not specify the data used to develop the correlation nor did he allude to their accuracy, although the examples presented in his paper showed an average absolute error of 7.9% between calculated and mea- sured values. ‘Trube's correlation can be best illustrated through the following example Example 4-7. Given the following data, Oil gravity = 45° Gas solubility = 600 scf/STB Solution gas gravity = 0.8 Reservoir temperature = 212°F Reservoir pressure = 2,000 psia Pressure gradient of reservoir liquid at 2,000 psia and 212°F = 32 psilft Find C, at 2,000, 3,000, and 4,000 psia. Solution. Step 1. Determine (yp from Equation 4-38162 Hydrocarbon Phase Behavior step 2. Correct the caloulated ol specific gravity to is value at 60°F by applying Equation 4-39. (vdeo = 0.739 + 0.00046 (152) = 0.8089 Step 3. Calculate the bubble-point pressure from Standings correlation (Equation 4-40). $9 ygpowmicr—8 Step 4. Adjust pp to its value at 60°F by applying Equation 4-41. 1.134 (1,866) _ (ea = “gears ~ 8571 pala Step 5. Eximate the pseudo-critial temperature Ty of the crude oil from Figure 4-8, to yield 1, -GER) 08" Step 6. Estimate the pseudo-critial pressure Pye of the erude oll from Figure £9, to ve pn Goes (420) Step 7. Caloulate the pseudo-reduced temperature from Equation 4-42 2812 gg OTH 0 Step 8. Calculate the pseudo-reduced pressure Py, the pseudo-reduced othermal compressibility coefficient C,, and the isothermal a compressibility coefficient of the oil from Equation 4-43, Figure (_) fvand Equation 47, respectively s{ 2 De ct a 10-* psa? 28 BY 2,000 436 0.0195 000% 25.0 -D1.d> os) 3,000 Qia4 — 0.0080 9» 55 78 43.4 S * 4,000 FRR 0.0065 6-44 13.00 4 2. Vasquer-Beggs’ Correlation, From a total of 4,036 experimental data Po tear roresitn model, Vasquez and Beggs (1080) corel ted the sothermal oil compressibility coefficients with Ry, T, “API, vq» and - They proposed the following expression. Gyo DAMS + SR, + TOT =a) = 1,180 7g + 12.61°APL_ 43) P Phase Behavior of Crude Oils 163 Realizing that the value of the specific gravity of the gas depends on the conditions under which itis separated from the oil, Vasquez. and Beggs pro- ‘posed that the value of the gas specific gravity as obtained from a separator pressure of 100 psig be used in the above equation. This reference pressure Was chosen because it represents the average field separator conditions. The Juthors proposed the following relationship for adjustment of the gas grav- ity 7, to the reference separator pressure. Ye = Ig |1 + 5.912 (108) (46)(Teep ~ 460) Log fe] ws where ‘yp = gas gravity at the reference separator pressure ve = gas gravity at the actual separator conditions of Pay and Tay pag = actual separator pressure, psia ‘Toy = actual separator temperature, °R ‘Ezample 48. Rework Example 4-7 by using the Vasquez and Beggs corre- lation and assuming the average field separators pressure of 114.7 psia. Solution. Step 1. Solve for the gas specific gravity at the reference pressure By ap- plying Equation 4-44 yg = 0.80 [1 + 0] = 0.8 Step 2, Calculate C, from Equation 4-43. ,180(0.8) + 12.61(45) 1p c= 10-9 483685 P P Co, 10°F psa”? 2,000 24.184 3,000 16.123 4,000 12.092 3. Ahmed’s Correlation. Based on 245 experimental data points for the {sothermal compressibility coefficients, Ahmed (1985) developed a mathe- matical expression for estimating C, by using a non-linear regression model, ‘The propesed correlation uses the gas solubility R, and the pressure as the only correlating parameters. It should be noted the other correlating param-164 Hydrocarbon Phase Behavior ters such as yp, 7p and T are implemented in the equation through the gas solubility R,. The correlation has the following simplified form. a EMP (es) (4-45) where ay = 24,841.0822 an = M.07428745 a3 = — 0.00018473 ‘The proposed relationship produced an average absolute error of 3.9% when tested against the experimental data used in developing the equation. ‘The isothermal compressibility coefficient can also be determined from the following expression: fore | aeseseceet uses ——| EXP (a; p) (4-46) l 4 Yo + a BY where a; = 1.026638 a = 0,0001553 ay = ~ 0.0001847272 a = 62,400 a= 136 ‘The above correlation was developed by correlating the isothermal com- pressibility coefficients with the oil density at the bubble-point pressure, as given by Equation 4-31 Example 4-9, The data given in Example 4-7 are listed here again for con- Using Equations 4-45 and 4-46, calculate C, at 2,000, 3,000, and 4,000 psia, Solution. ¢ From Equation 4-45, solve for Ca (=0.00018473 p) 0822 + 14.07428745 (600) = 30.0429(10-% EXP (~0.00018473 p) On Phase Behavior of Crude Oils 165 © Applying Equation 4.46 yields , = 0.00002589 EXP ( ~ 0,0001847272p) © Substituting for pressure yields PG 10° psi 179 1488 12.37 ‘DENSITY OF UNDERSATURATED CRUDE OILS Figure 4-10 shows a typical liquid density-pressure diagram. As the ram. As the pres- sure inoeases, the gas doles inthe crude ol and the density decreas, ¢ oil density will continue to decrease until the bubble-point pressure is The al wubble-point p A further increase in pressure will cause an increase in density due to compression of the crude oil. : versity Figure 4-10. A typical liquid density-pressure diagram.166 Hydrocarbon Phase Behavior ‘To account for the compression ofthe oil above the bubble-point pressure, the density of the crude is first calculated at the bubble-point pressure and reservoir temperature. The calculated density is then adjusted by using the isothermal compressibility coefficient as deseribed below. From the mathematical definition of the density, the volume of m pounds of crude ofl with a density of p, is calculated as v Pe Differentiating this expression with respect to the pressure yields: v| _ =m ap. \p/r Po Op Substituting the above relationship into Equation 4-36 gives: =p0[—m a C= =e | Shel 2|g a or 1 apy c= 1 be bo Op Integrating the above expression yields f° codp= [Se (447) Evaluating C, atthe average presure, § = p + pol2, gives C.e-pyetn le Po = Pop EXP [C, (p - ps)] (4-48) where p, = density of the oil at pressure p, Ibift? ay = density of the oil at the bubble-point pressure, Jb/Ft® C, = isothermal compressibility coefficient at average pressure, psi"! { | i | | Phase Behavior of Crude Oils 167 P= reservoir pressure, psia Pp = bubble-point pressure, psia B = average pressure, psia Vasquez-Beggs’ isothermal compressibility expression (Equation 4-43), or Ahmed’s C, expressions (Equations 4-45 and 4-46), can be incorporated in Equation 4-47 to give: For Vasquez-Beggs’ C, equatio | (-. pa ro|ssa ws where A= 10-5 [ ~ 1493 +5 R + 17.2(7 = 1,180 yp + 12.61°APT] 460) For Ahmed’s ©, expresso : Using Equation 4-45 gives Po = Po EXP [B (EXP (ap) ~ (ep) (4-50) where B= ~ (4.588893 + 0.002599 R,)~! — 0.00018473 Using Equation 4-46 gives, Po = Pep EXP [D (EXP (ep) ~ EXP (epy))] 51) — 4 Yo + As Reg = ~ 0.0001847272 a, = 1.026638 a, = 0.0001553 — 00001847272 ~ 11,526938 ~~ 0.00251229 Example 4-10. Data given in Examples 4-7 through 4-9 are summarized below: API= 45° Prep = 114.7 psia ps = 1,866 psia T = 212°F Yo = 0.802 = 0.8 R, = 600 scf/STB168 Hydrocarbon Phase Behavior Caleulate the oil density 2,000, 3,000, and 4,000 psia by using a, Equation 4-48 . Equation 4-49 ce. Equation 4-50 d, Equation 451 Solution. Using the available data, the oil density at the bubble-point pres- sure is calculated by applying Equation 4-32, to give Pow = 40.788 Ibift® 4, Solution by applying Equation 4-48, From Example 4-7 Tye = 0.80 Pye = 500 psia Py = 1,866 eee = PB Pe Figure 47_10°8, p 2,000 1,033" 3.866 0.013 26.00 3,000 2,433 4.866 0.0115 23.00 4,000 2,933 5.866 0,002 18.4 Tat wverage praau, be D, Solution by using Equation 4-49 * Calculate the coefficient A of Equation 4-49, to give A = 0.088369 ‘© Solve for the density by applying Equation 4-49 P be, Ibite psia from Equation 450 2,000 41.06 3,000 4187 4,000 42.46 . Solution by applying Equation 4-50 © Calculate the coefficient B of Equation 4-50, to give B = 0.162632 Phase Behavior of Crude Oils 169 «© Solve for the density by applying Equation 4-50 P Poy Ibitt? psia : (Equation 4-50) 2,000 40.90 3,000 41.59 4,000 42.35; 4. Solution by using Equation 4-51 © Calculate the coefficient D of Equation 4-51, to give D = 0.1401739 ‘Solve for the density from Equation 4-51. P Poy bite i (Equation 4-51) 2,000 40.98 3,000 41.66 4,000 42.93, GAS SOLUBILITY ‘The gas solubility R, is defined as the number of standard cubic feet of gas which will dissolve in one stock-tank barrel of crude oil at certain pressure and temperature, The solubility of a natural gas in a crude oil is a strong function of the pressure, the temperature, the API gravity, and the gas grav- ity. The effect of these complex variables on gas solubility is shown in Fig- ures 4-10 through 4-13. For a particular gas and crude oil to exist at a constant temperature, the solubility increases with pressure until the saturation pressure is reached. At the saturation pressure (bubble-point pressure) all the available gases are dissolved in the oil and the gas solubility reaches its maximum value. Rather than measuring the amount of gas that will disolve in a given stock-tank crude oil as the pressure is increased, it is customary to determine the amount of gas that will come out of a sample of reservoir crude oil as pres- sure decreases ‘A typical gas solubility curve, as a function of pressure for an undersat- urated crude oil, is shown in Figure 4-14. As the pressure is reduced from the initial reservoir pressure p, to the bubble-point pressure py, no gas evolves from the oil and consequently the gas solubility remains constant at its maxi- mum value of Ry. Below the bubble-point pressure, the solution gas is li berated and the value of R, decreases with pressure470 Hydrocarbon Phase Behavior bas sotubi bas solubinity ‘eaperature ———+ Figure 4-11, Rvs. p and AR, vs. T relationships. ft eunstant py Ts and gas gravity BPI gravity Figure 12. API gravity-R, relationship. ‘ Phase Behavior of Crude Oils 71 ‘At constant py Ty and APL gravity fas solubitity Gas specific gravity —s. Figure 4-13. Gas gravity-gas solubilly relationship. ae Figure 4-14. An idealized R,-p diagram for an undersaturated oll. In determining the PVT relationships (including the gas solubility-pres- sure relationship) in the laboratory, itis necessary to record the volume of oil and volume of liberated gas as the pressure is reduced below saturation pres- sure. The manner in which the solution gas is liberated from the oil significantly affect all the PVT relationships. There are two types of separa- tion (liberation, vaporization) process, namely: ‘Flash liberation © Differential liberation172 Hydrocarbon Phase Behavior Flash ration Inthefl ing a press lash liberation process, the gas which is liberated from the oil dur- re deeline remains in contact with the oil from which it was lib- erated, The process, as shown schematically in Figure 4-15, involves the fol- lowing steps: busble of gas 4 Step 1. Step 2 Step 3. Step 4. Step 5, Step 6. Figure 4-15. A schematic diagram of the flash liberation test. ‘The reservoir fluid sample is charged to a PVT cell which is main- tained at reservoir temperature throughout the experiments. ‘The cell pressure is elevated at a pressure higher than the satura- tion pressure. This can be achieved by injecting mercury (or fore- ing a piston) into the cell. ‘The cell pressure is lowered in small increments by withdrawing. mercury from the cell. The total volume of the hydrocarbon sys- tem is recorded at each pressure. ‘A plot of the cell pressure-total hydrocarbon volume is con- structed as shown in Figure 4-16. ‘When the cell pressure reaches the bubble-point pressure of the hydrocarbon system, a sign of formation of a gas phase is noted. ‘This stage is marked by a sharp change in the pressure-volume slope (Figure 4-16). ths pare la i reduce below the bubble-point presse, the liberated gas is allowed to remain in contact and reach an equilibrium state with the oil phase, This thermodynamic equi- librium is assured by agitating the cell. Phase Behavior of Crude Oils 173 Tora) Wyarocarbon Volume Figure 4-16. Flash liberation p-V diagram. Step 7. The equilibrium pressure level and the corresponding hydrocar- bon total volume is recorded, Step 8. Steps 6 and 7 are repeated until the capacity ofthe cell is reached. ‘The experimental data obtained from the flash liberation test include: a, The bubble-point pressure D. The isothermal compressibility coefficient of the li the bubble-point pressure . Below the bubble point, the two-phase volume is measured as a fune- tion of pressure iid phase above Itshould be noted that during the flash liberation process, the overall sys- tem composition remains constant because no gas is removed from the PVT. cell during pressure reductions. The foregoing process simulates the gas lib- eration sequence, whieh is taking place in the reservoir at pressures immedi- ately below the bubble-point pressure, This can be justified by the fact that the liberated gas remains immobile in the pores and in contact with oil until the critical gas saturation is reached at a certain pressure below py. Dodson et al. (1953) pointed out that the flash liberation process best rep- resents the separator type liberation. When entering the separator, the reser- voir fluids are in equilibrium due to the agitation occurring in the tubing. In the separator, the two phases are brought to equilibrium and the oil and gas, are separated. This behavior follows the flash liberation sequence.174 Hydrocarbon Phase Behavior Differential Liberation In the differential liberation process, the solution gas that is liberated from an oil sample during a decline in pressure is continuously removed from contact with the oil, and before establishing equilibrium with the liq- tuid phase. This type of liberation, as presented schematically in Figure 4-17, is characterized by a varying composition of the total hydrocarbon system, rf ie oe le seeea| sssaat| Figure 4-17, A schematic diagram of the differential liberation test “The experimental procedure of the test is summarized in the following steps ° Step 1. The reservoir fluid sample is placed in a PVT cell at reservoir temperature, Step 2. The cell is pressurized to saturation by injection of mercury Step 3, The volume of the all-liquid sample is recorded. Step 4, The cell pressure is lowered by removing mereury from the cell, Step 5. The liberated gas is removed from the cell through the cell flow valve. During this process, the cell pressure is kept constant by reinjecting mercury in the cell at the same rate as the gas dis- charge rate. Step 6, The volume of the discharged gas is measured at standard condi- tions and the volume of the remaining oil is recorded. Step 7. Steps 5 and 6 are repeated until the cell pressure is lowered to atmospheric pressure. Step 8, The remaining oil at atmospheric pressure is measured and con- verted to.a volume at 60°F, This final volume is referred to as the residual ol Phase Behavior of Crude Oils 175 ‘The experimental data obtained from the test include: ‘a, Amount of gas in solution as a function of pressure b. The shrinkage in the oil volume as a function of pressure ©. Properties of the evolved gas including the composition of the liberated ‘gas, the gas compressibility factor, and the gas specific gravity 4. Density of the remaining oil as a function of pressure “The differential liberation tes is considered to better describe the separa- tion process taking place in the reservoir and is also considered to simulate the flowing behavior of hydrocarbon systems at conditions above the critical gas saturation, As the saturation of the liberated gas reaches the critical gas Saturation, the liberated gas begins to flow, leaving behind the oil that origi- nally contained it. This is attributed to the fact that gases have, in general, higher mobility than oils. Consequently, this behavior follows the differen- tial liberation sequence. ‘A comparison of the two different liberation methods for determining the {gas solubility asa function of pressure is shown in Figure 4-18. This relation- Ship between the two processes may occur as shown or in reverse, depending upon the composition of the hydrocarbon system, Bitterential Frese Figure 4-18, Idealized comparison of flash and differential gas solubilities. Example 4-11, A differential liberation test (discussed in greater detail in Chapter 8) was conducted on a crude oil sample taken from an oil field in Montana, The sample, with a volume of 300ce, was placed in a PVT cell at its bubble-point pressure of 3,000 psia and reservoir temperature of 180°. “The temperature was kept constant and the pressure was reduced to 2,500116 Hydrocarbon Phase Behavior psi by removing mercury’ from the cell. The total volume of the hydrocar- bon system was increased to 346.5cc, The gas was bled off at constant pres- sure (by reinjecting the mercury) and found to occupy a volume of 0.145 sc. The volume of the remaining oil was 290.8ce. ‘The previous process was re- peated at 2,000 psia and the remaining oil was flashed through a series of laboratory separators with the separation stage representing stock-tank con- ditions. The collected experimental data are given below: Volume of Volume of. Pressure Temperature Total Volume Liberated Gas n psi °F ce sof ce 2,000 180 302.3 0.290 981.5 147 60°F 7 0.436 230.8 Calculate the gas solubility at 3,000, 2,500, and 2,000 psia. Solution. * Calculation of R, at 3,000: By recalling the definition of R, as the number of sof of gas in solution at p and T per stock-tank barrel of oil, the total sof of gas in solution at 3,000 psia and 180°F is (ccf), 0+ = 0.145 + 0.290 + 0.436 = 0.871 sef Volume of oil at _ yy) ___230.8__ 52 STB standard conditions” " Gp-agp> sais ~ 0.0052 S 9.871 _ 609 sef/STB Pao = 5 ogT452 © Calculation of R, at 2,500: At this pressure, the number of scf of gas in solution is equal to the total scf of gas minus the scf of free gas (liberated gas at 2,500 psia), or (Gcf)2s00 = 0.871 — 0.145 = 0.726 sof 0.726 0.786 500 seflSTB 0.001452, » (Ry)a.s00 = © Calculation of R, at 2,000: (cef)2.00 = 0.871 ~ 0.145 ~ 0.29 = 0.436 sof (in solution) + therefore, Phase Behavior of Crude Oils 77 ASS = 900 scfiSTB Methods of Calculating R, 1, Beal's Correlation. Beal (1946) presented a graphical correlation, as shown in Figure 4-19, for estimating the gas solubility as a function of the saturation pressure and the API gravity of the stock-tank oil. The proposed correlation was derived from 508 gas solubility observations taken from 164 crude oil samples. Beal's correlation yields an average deviation of 25% . 2. Standing’s Correlation. Standing (1947) proposed a graphical correlation for determining the gas solubility as a function of pressure, yas specific grav- ity, API gravity, and system temperature. The correlation was developed from a total of 105 experimentally determined data points on 22 hydrocar- bon mixtures from California crude oils and natural gases. The proposed correlation has an average error of 4.8%. The correlation is shown graphically in Figure 4-20. Standing (1981) pro- posed the following mathematical expression for this graphical correlation: nappy tonne pe where T = temperature, °R abet 3, Lasater’s Correlation. Lasater (1058) developed a graphical correlation for calculating the gas solubility at the bubble-point pressure. The graphical correlation, as shown in Figure 4-21, was based on 158 experimentally mea- sured bubble-point pressures of 137 independent systems. Vasquez and Beggs (1980) stated that Lasater's correlation is more accurate than Stand- ing's correlation for high-gravity crude oil systems. Standing’s correlation is preferred for applications for crudes of API gravity less than 15. 4. Vasquez-Beggs’ Correlation, Vasquez and Beggs (1980) presented an im- proved empirical correlation for estimating R,. The correlation was ob- tained by regression analysis using 5,008 measured gas solubility data points. Based on oil gravity, the measured data were divided into two ‘groups. This division was made at a value of oil gravity of 30° API. The pro- posed equation has the following form: Ry= Ci yp p EXP |e =] (453)178 Hydrocarbon Phase Behavior PREDICTION OF SOLUBILITY FROM SATURATION PRESSURE AND ‘ORUOE iL GRAVITY T GAS IN SOLUTION SATURATION. PRESSURI Figure 4-19. Beal's correlation for determining R. Permission to publish by the Society of Petroleum Engineers of AIME. Copyright SPE-AIME, Phase Behavior of Crude Oils. 179 Figure 4-20. Standing's gas solubility correlation. Reprinted by permission of Chevron Corporation.180 Hydrocarbon Phase Behavior Phase Behavior of Crude Oils 181 Values for the coefficients are as follows: API = 30 API > 30 0.0362 0.0178 1.0837 1.1870 25.7240 23.931 rage error of the above expression is 0.7%. ‘The gas specific gravity 7 at a reference separator pressure of 100 psig was defined previously by Equation 4-44 's gas solubility correlation. Permission to publish by the Society of Petroleum Engineers of AIME, Copyright SPE-AIME. 5. Glaso’s Correlation, Glaso (1980) proposed a correlation for estimating the gas solubility as a function of the API gravity, the pressure, the tempera- ‘ture, and the gas specific gravity. The correlation was developed from study- ing 45 North Sea crude oil samples. Glaso reported an average error of 1.28% with a standard deviation of 6.98% . The proposed relationship has the following form: App | fia oa oa (4-54) Figure 4-21. Las Ra% where pj is a correlating number and is defined by the following expression: pi = 10289-14813. top (4-55) Sutton and Farashad (1984) presented an excellent review of these corre- lations and documented their associated aecuracy. Sutton and Farashad con- cluded that Glaso's corcation showed the most accuracy and best predicted : However, the accuracy of Glaso'scorre- oll ratios in_excess of 1,400 scf/STB. 6. Marhoun’s Correlation. Marhoun (1988) developed an expression for es- WS ‘The correlation originates from 160 experimental saturation pressure data, SN ‘The proposed correlation can be rearranged and solved for the gas solubility to give: R= [arte TP (4-56) gas specific gravity stock-tank oil gravity T = temperature, °R a-€ = coefficients of the above equation having these values:162 Hydrocarbon Phase Behavior a = 185.843208, b = 1.877840 c+ ~ 3.1437 d= ~ 1.32657 1.398441 Example 4-12. A 38°API crude oil has a bubble-point pressure of 9,811 psia at 180°F. The average gas specific gravity is 0.732. Caleulate the gas solubility by using the following correlations: a. Beal's b. Standings ©. Lasater’s d. Vasquez-Beggs’ e. Claso’s f, Marhoun's Compare the results with the experimental value of 909 scfiSTB Solution. ‘a, Beal's Correlation. From Figure 4-19, determine the gas solubility, to give R, = 1,250 scf/STB b. Standing’s Correlation. Solve for Ry by applying Equation 4-50, to yield R, = 1,094 scf/STB ¢. Lasater’s Correlation, From Figure 4-21, the gas solubility is 900 sefi STB 4. Vasquez Beggs’ Correlation # Assuming separator pressure is 100 psig, from Equation 4-46, ip = 0.732 « Solve for gas solubility by applying Equation 4-53, to give R, sof/STB €, Glaso's Correlation '* Determine the correlating parameter pj, from Equation 4-55, to give: ph = 22.027 « Solve for gas solubility by applying Equation 4-56 to give: R, = 935 scflSTB £. Marhoun's Correlation «© Solve for R, by applying Equation 4-56, to give R, = 921 scflSTB al Phase Behavior of Crude Oils 183 ——— ‘OIL FORMATION VOLUME FACTOR ‘The oil formation volume factor, B,, is defined as the ratio of the volume of oil (plus the gas in solution) at the prevailing reservoir temperature and pressure to the volume of oil at standard conditions. Evidently, B, is always ‘greater than or equal to unity. The oil formation volume factor can be ex- pressed mathematically as ne en where By = oil formation volume factor, bbl/STB (V,)pr = volume of oil under reservoir pressure p and temperature T, bbl ‘volume of oil as measured under standard conditions, STB Figure 4-22, Oil formation volume factor versus pressure diagram. ‘A typical oil formation factor curve, as a function of pressure for an un dersaturated crude oil (p, > po), is shown in Figure 4-22. As the pressure is reduced below the initial reservoir pressure, p, the oil volume increases due to the oil expansion. This behavior results n an increase in the oil formation Volume factor and will continue until the bubble-point pressure is reached, ‘At py, the oil reaches its maximum expansion and consequently attains a ‘maximum value of By forthe oil formation volume factor, As the pressure is reduced below py, volume of the oil and B, are decreased as the solution gas