MONTANUNIVERSITÄT LEOBEN

PETROLEUM ENGINEERING DEPARTMENT
TEXTBOOK SERIES
VOLUME 2
RESERVOIR FLUIDS
by
Zoltán E. HEINEMANN
Professor for Reservoir Engineering
Brigitte E. WEINHARDT
Assosiate Professor for Reservoir Engineering
Leoben, October 2004
© No part of this publication may be reproduced in any form.
Only students of the University of Leoben may copy for studying purposes.
Edited by DI Claudia Scharf - October 2004
Table of Contents
Chapter 1
Review of Thermodynamic Terminology ........................................................ 1
Chapter 2
Phase Behavior ................................................................................................... 3
2.1 Gibbs’ Phase Rule ........................................................................................................3
2.2 Single-Component System ..........................................................................................4
2.2.1 Water .............................................................................................................4
2.2.2 n-Butane .........................................................................................................7
2.3 Critical State and Quantities of Corresponding States ...............................................11
2.4 Binary Systems ..........................................................................................................13
2.5 Multi-Component Systems ........................................................................................20
2.5.1 Ternary Phase Diagrams ..............................................................................20
Chapter 3
Equations of State ............................................................................................ 31
3.1 Change of State at Low Compressibility ...................................................................32
3.2 Equation of State of Perfect and Real Gases .............................................................33
3.3 Cubic Equations of State ............................................................................................35
3.4 Virial Equation of State .............................................................................................47
Chapter 4
Calculation of Phase Equilibria ...................................................................... 49
4.1 Mixtures .....................................................................................................................49
4.1.1 Definitions ...................................................................................................49
4.1.2 K-factors ......................................................................................................50
4.2 Composition of Phases in Equilibrium ......................................................................54
4.2.1 Definitions ...................................................................................................54
4.2.2 Evaluation of K-Factors Using Convergence Pressures .............................62
4.2.3 Evaluation of Convergence Pressure ...........................................................68
4.2.4 Evaluation of by use of Peng-Robinson equation of state ..........................69
Chapter 5
Phase Properties ............................................................................................... 81
5.1 Natural Gases .............................................................................................................81
5.1.1 Volume ........................................................................................................81
ii Table of Contents
5.1.2 Formation Volume Factor ........................................................................... 83
5.1.3 Compressibility ........................................................................................... 84
5.1.4 Correlation of Z-Factor on the Basis of Reduced State Variables .............. 84
5.1.5 Water Content .............................................................................................. 93
5.1.6 Viscosity ...................................................................................................... 94
5.2 Hydrocarbon Liquids ................................................................................................. 98
5.2.1 Volume ........................................................................................................ 98
5.2.2 Formation Volume Factor ......................................................................... 105
5.2.3 Compressibility of Undersaturated Liquids .............................................. 110
5.2.4 Viscosity .................................................................................................... 114
5.3 Brines ....................................................................................................................... 117
5.3.1 Composition of Brines ............................................................................... 118
5.3.2 Formation Volume Factor ......................................................................... 121
5.3.3 Viscosity .................................................................................................... 126
5.3.4 Natural Gas Hydrates ................................................................................ 127
Chapter 6
pVT-Measurements ....................................................................................... 133
6.1 Sampling .................................................................................................................. 133
6.1.1 Objectives .................................................................................................. 133
6.1.2 General Criteria ......................................................................................... 133
6.1.3 Sampling Methods ..................................................................................... 134
6.1.4 Special Problems ....................................................................................... 137
6.2 Experimental Determination of the Volumetric and Phase Behavior ..................... 138
6.2.1 Equipment ................................................................................................. 138
6.2.2 PVT-Cells .................................................................................................. 139
6.2.3 Volumetric Pumps ..................................................................................... 144
6.2.4 Auxiliary Equipment ................................................................................. 144
6.3 Methods ................................................................................................................... 145
6.3.1 Flash Process ............................................................................................. 145
6.3.2 Differential Process ................................................................................... 146
6.3.3 Reverse Differential Process ..................................................................... 146
List of Figures
Figure 2.1: Water system - schematic (not drawn to scale)4
Figure 2.2: Phase equilibrium surface of a pure substance (from Gyulay, 1967)5
Figure 2.3: Vapor pressure diagram of n-butane (from Gyulay, 1967)7
Figure 2.4: Pressure - volume phase diagram of n-butane (from GYULAY, 1967)8
Figure 2.5: Temperature - density phase diagram of n-butane (from GYULAY, 1967)9
Figure 2.6: Critical pressure as a function of number of C-atoms in homologous series (after
Gyulay, 1967)11
Figure 2.7: Critical temperature as a function of numbers of C-atoms in homologous series (after
Gyulay, 1967)12
Figure 2.8: Combined reduced pressure - reduced volume phase diagram of paraffins with low
molecular weight (after Gyulay, 1967)14
Figure 2.9: Phase equilibrium surface of the binary system ethane/n-heptane (from Gyulay, 1967)14
Figure 2.10: Pressure - temperature phase diagram of the binary system ethane/n-heptane (from Kay,
1938)15
Figure 2.11: Pressure - temperature phase diagram of the binary system ethane
(z = 0.9683)/n-heptane17
Figure 2.12: Mole fraction(ethane) - temperature diagram of the binary system ethane/n-heptane
(from Gyulay, 1967)17
Figure 2.13: Mole fraction (ethane) - Pressure diagram of the binary system ethane/n-heptane (from
Gyulay, 1967)18
Figure 2.14: Properties of ternary diagrams20
Figure 2.15: Typical features of a ternary phase diagram21
Figure 2.16: Triangular diagrams for the methane/propane/n-pentane system at 160
o
F(71
o
C) (after
Dourson et al., 1943)23
Figure 2.17: Critical loci of methane/propane/n-pentane systems (from Katz et al., 1959)24
Figure 2.18: Phase diagram of a dry gas (from McCain, 1973)26
Figure 2.19: Phase diagram of a wet gas (from McCain, 1973)26
Figure 2.20: Phase diagram of a retrograde gas condensate (from McCain, 1973)26
Figure 2.21: Phase diagram of a high-shrinkage crude oil (from McCain, 1973)27
Figure 2.22: Phase diagram of a low shrinkage crude oil (from McCain, 1973)27
Figure 2.23: Phase diagram pairs of gas cap and oil zone28
Figure 2.24: Phase equilibrium surface of oil/natural gas systems (from Gyulay, 1967)29
Figure 3.1: The Van der Waals isotherms near the critical point35
Figure 4.1: Fugacity of natural gases (from Brown, 1945)52
Figure 4.2: Ideal and real K-factors of n-butane at 60[
o
C]52
Figure 4.3: Flash and differential vaporization59
Figure 4.4: K-factors for methane-propane at T
c
= 100
o
F (from Sage, Lacey and Schaafsma) (1934)
63
Figure 4.5: Comparison of K-factors at 100
o
F for 1,000 and 5,000-psia convergence pressure (from
NGAA, 1957)64
Figure 4.6: K-factors for methane, 5,000 psia convergence pressure (from NGAA, 1957)66
Figure 4.7: K-factors for hexane, 5,000 psia convergence pressure (from NGAA, 1957)67
Figure 4.8: Convergence pressure data - methane for binary hydrocarbon mixtures (from Winn,
1952)70
Figure 5.1: Z-factor of methane, ethane and propane versus pressure at T = 140oF (from Standing,
1977)83
iv List of Figures
Figure 5.2: Z-factor as a function of reduced pressure for a series of reduced temperatures (from
Sage and Lacey, 1949)85
Figure 5.3: Z-factor for natural gases (from Brown et al., 1948) .................................................... 87
Figure 5.4: Pseudo-critical temperatures and pressures for heptanes and heavier
(from Matthews et al, 1942)88
Figure 5.5: Pseudo-critical properties of Oklahoma City Gases
(from Matthews et al., 1942)88
Figure 5.6: Water content of natural gas in equilibrium with liquid water
(from Katz et al., 1959)92
Figure 5.7: Viscosity of paraffin hydrocarbon gases natural gases at atmospheric pressure (from
Carr et al., 1954)95
Figure 5.8: Viscosity of natural gases at atmospheric pressure (from Carr et al, 1954) ................. 96
Figure 5.9: Correlation of viscosity ratio with pseudo-reduced pressure and temperature (from Carr
et al., 1954)96
Figure 5.10: Variation of apparent density of methane and ethane with density of the system (from
Standing and Katz, 1942)99
Figure 5.11: Pseudo-liquid density of systems containing methane and ethane
(from Standing, 1952)100
Figure 5.12: Density correction for compressibility of liquids (from Standing, 1952)................... 101
Figure 5.13: Density correction for thermal expansion of liquids
(from Standing, 1952)102
Figure 5.14: Apparent liquid density of natural gases in various API gravity oils
(from Katz, 1952)104
Figure 5.15: Typical graph of formation-volume factor of oil against pressure ............................ 106
Figure 5.16: Pseudo-reduced compressibility for undersaturated reservoir fluids
(from Trube, 1957)112
Figure 5.17: Pseudo-critical conditions of undersaturated reservoir liquids
(from Trube, 1957)112
Figure 5.18: Viscosity of subsurface samples of crude oil (from Hocott and Buckley, 1941, after Beal,
1946)114
Figure 5.19: Viscosity of gas-saturated reservoir crude oils at reservoir conditions
(from Chew and Connally, 1959)115
Figure 5.20: Prediction of crude oil viscosity above bubble point pressure
(from Beal, 1946)116
Figure 5.21: Essential feature of the water pattern analysis system (from Stiff, 1951) .................. 120
Figure 5.22: Course of Arbuckle formation through Kansas shown by water patterns (from Stiff,
1951)120
Figure 5.23: Solubility of natural gas in water (from Dodson and Standing, 1944) ....................... 122
Figure 5.24: Typical graph of formation volume factor of water against pressure......................... 122
Figure 5.25: B
w
for pure water (dashed lines) and pure water saturated with natural gas (solid lines)
as a function of pressure and temperature (from Dodson and Standing, 1944)123
Figure 5.26: Density of brine as a function of total dissolved solids
(from McCain, 1973)124
Figure 5.27: The isothermal coefficient of compressibility of pure water, including effects of gas in
solution (from Dodson and Standing, 1944)125
Figure 5.28: The viscosity of water at oil field temperature and pressure
(from van Wingen, 1950)127
Figure 5.29: Hydrate portion of the phase diagram for a typical mixture of water and a light
List of Figures v
hydrocarbon (from McCain, 1973)129
Figure 5.30: Pressure-temperature curves for predicting hydrate formation
(from Katz, 1945)129
Figure 5.31: Depression of hydrate formation temperature by inhibitors
(from Katz et al., 1959)130
Figure 5.32: Permissible expansion of 0.8 gravity gas without hydrate formation
(from Katz, 1945).131
Figure 6.1: Scheme of PVT equipments......................................................................................... 138
Figure 6.2: Blind PVT cell ............................................................................................................. 140
Figure 6.3: PVT cell (after Burnett) ............................................................................................... 140
Figure 6.4: PVT cell (after Dean-Poettman) .................................................................................141
Figure 6.5: Variable volume cell (after Velokivskiy et al.) ...........................................................142
Figure 6.6: PVT cell (after Sloan) .................................................................................................. 142
Figure 6.7: PVT cell (after Wells-Roof)......................................................................................... 143
Figure 6.8: Ruska cell .................................................................................................................... 143
Figure 6.9: Ruska volumetric mercury pump ................................................................................144

1
Chapter 1
Review of Thermodynamic Terminology
When considering hydrocarbon reservoirs, terms such as “oil reservoirs” and “gas
reservoirs” are used both in colloquial speech and technical literature. However, these
terms are insufficient. Changes in the state of aggregation during production should
always be taken into account in consequence of changes of the reservoir pressure and
changes of pressure and temperature in the production system (tubing, pipe lines,
separator, tank).
Thermodynamics has evolved to a science of studying changes in the state of a system
with changes in the conditions, i.e. temperature, pressure, composition. A systematic
presentation of basic thermodynamic tools (charts, tables and equations) for sketching the
state of a hydrocarbon system as a function of the state variables is one of the objectives
of this textbook. Therefore, it may be helpful to refurbish the thermodynamic terminology
at the beginning as far as it is indispensable to the understanding of these tools..
Thermodynamic studies are generally focused on arbitrarily chosen systems while the rest
of the universe is assumed as “surroundings”. The surface of the system - real or
imaginary - is called a “boundary”. A system is called a “closed system” if it does not
exchange matter with the surroundings, in opposite to an “open system” which exchanges
matter with the surroundings. Both systems may exchange energy with the surroundings.
The concept of a closed system is of major interest in applied hydrocarbon
thermodynamics. It is called a “homogeneous” closed system if it contains a single phase,
e.g. a natural gas phase or an oil phase. A “heterogeneous” closed system contains more
than one phase.
A “phase” is defined as a physically homogeneous portion of matter. The phases of a
heterogeneous system are separated by interfaces and are optically distinguishable. It is
not obligatory that a phase is chemically homogeneous. It may consists of several
compounds, e.g. of a large number of various hydrocarbons.
The thermodynamic properties of a system are classified into “intensive” and “extensive”
properties. Intensive properties such as temperature and pressure are independent of the
size of the system, i.e. of the amount of the substance in the system. Extensive properties
depend on the amount of the substances, such as volume, enthalpy, entropy etc. However,
2 Review of Thermodynamic Terminology
the extensive properties per unit mass or mole are intensive properties, e.g. the mole
volume.
“State functions” or “state variables” are those properties for which the change in state
only depends on the respective initial and final state. It is this path-independent
characteristic of the state functions that makes it possible to quantify any change of a
system.
“Equilibrium” has been defined as a “state of rest”. In an equilibrium state, no further
change or - more precisely - no net-flux will take place unless one or more properties of
the system are altered. On the other side, a system changes until it reaches its equilibrium
state.. Any change of a system is called a “thermodynamic interest” in the thermodynamic
study of the system:
• adiabatic (no heat added to or removed from the system),
• isothermal (constant temperature),
• isobaric (constant pressure),
• isochoric (constant volume).
A process is called “reversible” if it proceeds through a series of equilibrium states in such
a way that the work done by forward change along the path is identical to the work
attained from the backward change along the same path. However, all real processes are
“irreversible” with varying degrees of departure from a reversible one.
3
Chapter 2
Phase Behavior
Hydrocarbon reservoirs consist of rock and fluids. Water in brine form and a gaseous
and/or liquid hydrocarbon phase are regarded as reservoir fluids. The phase behavior of
the actual hydrocarbon mixture in the reservoir can be described as a function of the state
of the system.
A system in thermodynamic equilibrium posesses an accurately defined relationship
between the state variables. These are united in the so-called “equation of state”:
. (2.1)
By specification of two variables, the third will be stipulated.
2.1 GIBBS’ Phase Rule
When referring to the number of phases coexisting in the thermodynamical equlibrium,
the phase rule introduced by GIBBS (1928) is applied.
(2.2)
where
• P: number of phases,
• C: number of components,
• F: number of degrees of freedom.
C is defined as the smallest number of constituents by which the coexisting phases can be
completely described. F is defined as the number of quantities such as pressure,
temperature, concentrations which can be varied within finite boundaries without
changing the number of phases in the system.
F p V T , , ( ) 0 =
F C P – 2 + =
4 Phase Behavior
Eq. 2.2 describes the system in a qualitative and very general manner. However, no
reference to the state variables (p,T), to the composition of the particular phases or to the
proportions of the phases are given.
To gain a full understanding, it is best to discuss the phase behavior of pure substances
(single-component systems) first. The circumstances in case of 2- or even
multi-component systems are much more complicated.
2.2 Single-Component System
2.2.1 Water
Figure 2.1: Water system - schematic (not drawn to scale)
Water is one of the most thoroughly studied chemical compounds. Therefore, it is
discussed as a single-component system in this context. The possible phases are ice (solid
state), water (liquid state) and steam (gaseous state). The phase diagram in Figure 2.1
illustrates at which state of the system - charaterized by p and T - two or all three phases
are in equilibrium:
• The sublimation curve OA signifies the equilibrium between the solid and vapor.
• The melting point curve OB combines the states of equilibrium between the solid
22.09
0.01
0.0006
0.0075 100 374
Temperature [°C]
P
r
e
s
s
u
r
e

[
M
P
a
]
A
B
C
0
Ice
Water
Vapor
Phase Behavior 5
and liquid state.
• The vapor pressure curve OC specifies the states of the system at which the liquid
and vapor coexist. On this curve, the "wet" vapor is in equlibrium with the "saturated"
liquid.
• At the triple point O, all three phases are in eqilibrium. In case of water, the
thermodynamical data at this point are p = 610.6 Pa and T = 273.16 K.
• The end point C of the vapor pressure curve is the critical point and signifies the
highest temperature and pressure at which the liqiud and vapor coexist (p
c
(H
2
O) =
22.09 MPa, T
c
(H
2
O)= 647.15 K.
Example 2.1
The degree(s) of freedom in different states of a single-component
system. by use of Eq. 2.2 (GIBBS’ phase rule) and Figure 2.1:
• F = 0 at the tripel point (P = 3).
• F = 1 on the curves describing the 2-phase (P = 2) equlibria.
Either the temperature or the pressure is freely eligible
without counteracting any given phase equlibrium.
• F = 2 in any area of single phase state (P = 1. Both pressure
and temperature (naturally inside finite boundaries) are freely
eligible without transforming the system into a multi-phase
system.
Figure 2.2: Phase equilibrium surface of a pure substance (from GYULAY, 1967)
A
A
A
A
B
B
B
B
C
C
C
C
D
D
D
D
E
E
E
E
v
v
v
p
p
p
T
T
T
6 Phase Behavior
The state variables, p, T, V can only assume positive values. Thus, the graphical
illustration of the state of any system is only situated in the positive section of the p,V,
T-coordinate system An example for an equilibrium surface is giben in Figure 2.2. The
shape of such an equilibrium surface is substance specific.
Assuming that the partial derivatives are steady, it is possible to draw only one single
tangential plane at an optional point of the surface. However, a plane is defined by two
vectors which infers that the differential quotients
are not independent of one another. On the basis of
which describe a change of a system’s state by changing pressure and temperature at
constant specific volume, it is proven that
(2.3)
The three differential quotients describe three essential fundamental properties of the
system: (i) the isothermal compressibility , (ii) the cubic expansion coefficient a, and
(iii) the pressure coefficient :
, (2.4)
(2.5)
(2.6)
Then, according to Eq. 2.3, the following is valid:
(2.7)
Since the specific volume - in contrast to the representation in Figure 2.1 - appears now
as a state variable, the 2-phase state (e.g. water in equilibrium with steam) is characterized
by any area surrounded by two curves which converge at the critical point:
• On the bubble point curve, an infinitesimal small amount of vapor is in equilibrium
with the “saturated” liquid.
• The dew point curve characterizes states in which a negligible small amount of liquid
∂p
∂T
------
¸ ,
¸ _
V
∂V
∂T
------
¸ ,
¸ _
p
∂V
∂p
------
¸ ,
¸ _
T
, ,
∂p
∂T
------
¸ ,
¸ _
V
∂V
∂T
------
¸ ,
¸ _
p
∂V
∂p
------
¸ ,
¸ _
T
---------------
– =
κ
β
κ
1
V
---

¸ ,
¸ _
∂V
∂p
------
¸ ,
¸ _
T
=
α
1
V
---
∂V
∂T
------
¸ ,
¸ _
p
=
β
1
p
---
∂p
∂T
------
¸ ,
¸ _
V
=
α pκβ =
Phase Behavior 7
is in equilibrium with “wet” vapor.
It is common to simplify the complex spatial illustration of the equilibrium surface by
applying normal projections. Figure 2.2 displays that
• the projection into the p, V-plane results in isotherms (T = const),
• the projection into the V, T-plane results in isobares (p = const),
• the projection into the p, T-plane results in isochores (V = const).
When regarding the projection which represents the 2-phase area (liquid-vapor) in the p,
T-plane, the bubble point curve and dew point curve coincide. The resulting single curve
is named vapor pressure curve.
Of course, the vapor pressure curve is not isochoric. However, it is possible to draw
isochores: One upwards into the liquid phase and one downwards into the gas phase. This
aspect will be described in detail by discussing the phase behavior of the simple
hydrocarbon n-butane.
2.2.2 n-Butane
Projections of the equilibrium surface into two planes of the positive section of the p,V,T
coordinate system are displayed in Figure 2.3 and Figure 2.4.
Figure 2.3: Vapor pressure diagram of n-butane (from GYULAY, 1967)
50 100 150 200
Temperature [°C]
P
r
e
s
s
u
r
e

[
M
P
a
]
0
1
2
3
4
5
6
V
=
0
.
5
m
/M
o
l
3
V
=
0
.
0
5
m
/
M
o
l
3
Liquid
F
A
B D
H
C
E
Vapor
G
8 Phase Behavior
Figure 2.3 illustrates the vapor pressure curve of n-butane, including the critical point C.
In addition, the isochores V = 0,05 m
3
/kmoleinside the liquid phase region and V = 0.5
m
3
/kmole inside the vapor region are shown. In case of the state A, butane is an
undersaturated liquid. When moving to the bubble point B by isothermal expansion,
vaporization commences. Then the continuation of this isothermal expansion includes no
further pressure drop in the system until the last molecule has passed over to the gas phase.
From this moment, further expansion will result in further pressure decrease. At the point
E, n-butane is in the state of a “dry” vapor. An isochoric change of state must be
analogously discussed.
By applying the projection of the equilibrium surface into the p, V-plane (see Figure 2.4),
it is possible to comprise the whole 2-phase region. In this region, the isothermal
vaporization or condensation takes place as an isobaric process.
Figure 2.4: Pressure - volume phase diagram of n-butane (from GYULAY, 1967)
Isotherms, which do not intersect the 2-phase region, describe those states of the system
without any phase transformation by changing the pressure. The intersection point of all
other isotherms with the bubble point curve (e.g. A) marks the specific volume of the
saturated liquid which is in phase equilibrium with the specific volume of the wet vapor
(e.g. marked by point B). Considering point D within the 2-phase region of the system
(specific volume of the system in total), the mass ratio of the liquid and vapor phase being
in equilibrium with one another can be calculated by the principle of the lever:
(2.8)
A D B
130
140
150
C
B
P
D
P Liquid
Vapor
Liquid and Vapor
1
5
5
1
6
0
1
7
0

°
C
1
5
2
.
8
0 0.005 0.010 0.015
V [m/kg]
3
p

[
M
P
a
]
2
3
4
5
m
L
m
v
⁄ DB/AD =
Phase Behavior 9
Example 2.2
100 kg n-butane are filled up in a sealed 10 m
3
container. The
volume of the vapor phase at T = 130°C and p = 2.7 MPa can be
evaluated from Figure 2.4 using Eq. 2.8
From Figure 2.4:
,
m
L
+ m
v
= 100 = 0.3125 m
v
+ m
v

m
v
= 76.19 kg.
The specific volume of the vapor phase is marked by point B in
Figure 2.4:
V = 0.0125 m
3
/kg.
The vapor volume of the system, V
v,
can be now calculated by
multiplying V with m
v
:
V
v
= 0.95 m
3
.

Figure 2.5: Temperature - density phase diagram of n-butane (from GYULAY, 1967)
Figure 2.5 demonstrates the T, -diagram of n-butane. The isobare touching the critical
point has an inflection point just as the critical isotherm in Figure 2.4. Inside the 2-phase
region, average values of fluid and vapor density are located on a straight line. With the
help of this rule (CAILLETET-MATHIAS rule (1886)), the critical density can be calculated
m
L
m
v
⁄ DB AD ⁄ 0.3125 = =
Liquid
D
P
B
P
0 200 400
100
150
200
T

[
°
C
]
ρ [kg/m]
3
Vapor
C
Liquid and Vapor
5

M
P
a
ς
10 Phase Behavior
by extrapolation.
Example 2.3
Use the CAILLETET-MATHIAS rule to evaluate the critical density of
methane. The densities of the liquid and the vapor phase being in
equilibrium have been measured at different temperatures (see
table below). The values of averaged densities have already been
calculated.
It is known as the CAILLETET-MATHIAS rule that the averaged densities are situated on a
straight line. The slope of a straight line can be evaluated by regression analysis. On the
basis of the averaged densities given above:
.
To evaluate the critical density, , the line must be
extrapolated to the critical temperature of methane, :
,
.
Table 2.1:
Temperature
T
o
C
Liquid Density
kg m
-3
Vapor Density

kg m
-3
Averaged Density
kg m
-3
- 158.3
- 148.3
- 138.3
- 128.3
- 118.3
- 108.3
4.192 E + 2
4.045 E + 2
3.889 E + 2
3.713 E + 2
3.506 E + 2
3.281 E + 2
2.311 E + 0
2.798 E + 0
7.624 E + 0
1.240 E + 1
1.925 E + 1
2.899 E + 1
2.1076 E + 02
2.0365 E + 02
1.9827 E + 02
1.9185 E + 02
1.8493 E + 02
1.7855 E + 02
ρ
liq eq ,
ρ
v eq ,
ρ
eq
tg α 0 654 , kg m
3
⁄ – =
ρ
c
T
c
82 3 , C
o
– =
ρ
c
178.55 - 26 x 0.654 kg m
3
⁄ =
ρ
c
161.546 kg m
3
⁄ =
Phase Behavior 11
2.3 Critical State and Quantities of Corresponding
States
Figure 2.4 illustrates the inflection point of the critical isotherm at the critical point. At
the point of the inflection, both the first and the second partial derivates of p = p(V) equal
zero that
. (2.9)
The state of the system at this point is characterized by the critical specific volume V
c
, the
critical pressure p
c
, and the critical temperature T
c
.
Figure 2.6: Critical pressure as a function of number of C-atoms in homologous series
(after GYULAY, 1967)
∂p
∂V
------
¸ ,
¸ _
0

2
p
∂V
2
---------
¸ ,

¸ _
= =
Cyclohexane Naphtalene
Toluene
Benzene
iC
4
iC
5
n
o
r
m
a
l
-
P
a
r
a
f
f
i
n
s
m
o
n
o
-
O
l
e
f
i
n
e
s
0 2 4 6 8 10 12 14
C-Atoms per Mole
p

[
M
P
a
]
c
1
2
4
3
5
6
12 Phase Behavior
Figure 2.7: Critical temperature as a function of numbers of C-atoms in homologous
series (after GYULAY, 1967)
Considering the critical data p
c
and T
c
, the homologous series of hydrocarbons show
regularities which can be used for extrapolation. The experimental data in Figure 2.6 and
Figure 2.7 refer to the homologous series of paraffins, C
n
H
2n+2
, and olefines, C
2
H
2n
(n =
1,2,... k)
,
with a margin of error 1 to 2%. Because of thermal decomposition, it is not
possible to obtain experimentally information about the critical data in case of high
molecular weight. However, the critical data of homologous compounds with longer
carbon chains can be extrapolated though an increasing error has to be taken into
consideration.
The “principle of corresponding states” for chemically similar substances - e.g. for
homologous series - results in a close relation between the p, V, T-properties of pure
hydrocarbons if the state variables are substituted by the so called “reduced quantities”
which are
(2.10)
Figure 2.8 shows a p
r
, V
r
phase diagram which is valid for paraffins from methane (CH
4
)
to hexane (C
6
H
14
).
0 2 4
6
8 10 12 14
C-Atoms per Mole
T

[
°
C
]
c
-200
-100
0
100
200
300
400
500
n
o
r
m
a
l
-
P
a
r
a
f
f
i
n
s
m
o
n
o
-
O
l
e
e
f
i
n
s
Toluene
Benzene
Naphtalene
p
r
p
p
c
----- ; V
r
V
V
c
------ T
r
T
T
c
-----
= ; = =
Phase Behavior 13
2.4 Binary Systems
If a systems consists of more than one component, its state is also a function of
composition. In general, the composition is defined by “mole fractions”.
The mole fraction is defined as the ratio between the number of moles of a certain
component and the sum of moles of all components. A system being composed of k
components is defined by the specification of (k - 1) mole fractions because the sum of
the mole fractions always equal 1.Considering a 2-component system, every change in
state is described by the equation of state F(p, V, T, z) = 0. z may be the mole fraction of
one (lighter) component.
The phase behavior of the ethane/n-heptane system is graphically illustrated by the p, T,
z-coordinate system in Figure 2.9. The volume is equivalent to the mole volume.
In the plane z = 1, the vapor pressure curve of ethane appears, whereas in the plane z = 0
the one of n-heptane appears. Covering all other z-planes, an envelope surface encloses
the 2-phase state. This is demonstrated by the example of three additional z-planes.
The upper broken line marks the critical points of all compositions which are possible.
This curve divides the envelope surface into two parts: the bubble point surface and the
dew point surface. The region of an undersaturated liquid state is positioned outside the
bubble point surface (low temperature). Outside of the dew point surface (high
temperature), the state of a dry gas is given.
Analogous to the pure substance, the critical state of binary systems is defined as the state
at which the intensive properties of the phases are no more distinguishable. Just as in case
of 1-component systems, the critical isotherms have an inflection point according to Eq.
2.9.
14 Phase Behavior
Figure 2.8: Combined reduced pressure - reduced volume phase diagram of paraffins
with low molecular weight (after GYULAY, 1967)
Figure 2.9: Phase equilibrium surface of the binary system ethane/n-heptane (from
GYULAY, 1967)
0 1.0 2.0 3.0 4.0
V
r
0.6
0.8
1.0
1.2
1.4
p
r
0.94
0.96
0.98
C
1
.
0
2
1
.0
0
1
.
0
4
1
.
0
6
=
T
r
10
7.5
5.0
2.5
0
0 100 200 300
Temperature [°C]
P
r
e
s
s
u
r
e

[
M
P
a
]
z
1.00
0.75
0.50
0.25
0
C
2
nC
7
Phase Behavior 15
Figure 2.10: Pressure - temperature phase diagram of the binary system
ethane/n-heptane (from KAY, 1938)
Figure 2.10 shows the projection of Figure 2.9 into the p, T-plane. At a given pressure, the
bubble point temperature of the mixture is always higher than that of the pure lighter
component. Physically, it can be explained by the fact that the thermal motion of the
lighter molecules is obstructed by the heavier ones which exhibit more inertia.
On the other side, the dew point temperature of the mixture at a given pressure is always
lower than that of the pure heavier component. This is due to the fact that lighter
molecules partially transfer their higher kinetic energy to the heavier ones by collision.
Consequently, the system maintains the state of a gas phase.
Figure 2.10 also shows that T
c
of a mixture lies between the critical temperatures of the
pure substances. In contrast to this, p
c
of the mixture may be obviously higher than the
one of the pure substances.
If the mixture consists of two homologous compounds with quite different volatility (in
consequence of quite different molecular weights), the critical data curve envelopes a very
extensive temperature and pressure region. For example, the maximum of the critical
pressure of a methane/n-decane system equals 37 MPa. The smaller the difference
between the molecular weights and thus between the volatility, the more flat the envelope
curve will be.
Temperature [°F]
100 200 300 400 500
0
200
400
600
800
1000
1200
1400
P
r
e
s
s
u
r
e

[
p
s
i
a
]
Composition
No. [Wt%] Ethane
1 100.00
2 90.22
3 70.22
4 50.25
5 29.91
6 9.78
7 6.14
8 3.27
9 1.25
10 n-Heptane
10
9
8
7
6
5
1
3
4
2
16 Phase Behavior
Figure 2.11 illustrates the phase behavior of a certain ethane/n-heptane system. Besides
the critical point, the curve enveloping the 2-phase region possesses two additional
characteristic points:
• C’: the point of highest pressure on the curve that is called cricondenbare.
• C”: the point of highest temperature on the curve that is called cricondentherm.
As on Figure 2.11, so called "quality lines" are shown on p,T-diagrams. A quality line
represents a certain mole percentage being liquid or vapor in the state of phase
equilibrium. In Figure 2.11, the quality line "20%" represents the states in which 20% of
the system account for the liquid phase. The bubble point curve and the dew point curve
represent 100% and 0% liquid, respectively. All the quality lines (isochores) converge at
the critical point.
Figure 2.11 also shows an isothermal decrase along the path EF where E defines the
sastem to be a dry gas. If the constant temperature is higher than T
c
but lower than the
cricondentherm - like in case of the path EF -, the path surpasses the dew point line twice.
Consequently, a condensate drops out at the dew point D’. At some point between D’ and
D”, the volume of condensate (liquid) will be at its maximum. This maximum is given by
the intersection point of the path EF with the dotted line connecting C and C”. If the
decrease in pressure will be continued, the condensate will be vaporized again. As soon
as the dew point D” has been reached, the condensated phase has been vaporized in total.
This process is called a “retrograde condensation”.
Similar phenomena occur when the temperature is changed by an isobaric process where
the constant pressure is higher than p
c
but lower than the cricondenbar of the system.
In Figure 2.11, the dotted line connecting point C with point C’ marks the states of the
system which exhibit the highest volume percentage of condensate dropout.
Phase Behavior 17
Figure 2.11: Pressure - temperature phase diagram of the binary system ethane
(z = 0.9683)/n-heptane
It depends on the composition of the system if the cricondenbar is located on the dew point
curve or on the bubble point curve. As far as the system ethane/n-heptane is concerned,
Figure 2.10 elucidates that the cricondenbar is located on the bubble point curve at low
mole fractions of ethane.
Figure 2.12: Mole fraction(ethane) - temperature diagram of the binary system
ethane/n-heptane (from GYULAY, 1967)
7
6
5
4
3
2
1
0
0 50 100 150
Temperature [°C]
P
r
e
s
s
u
r
e

[
M
P
a
]
C
C'
A
B
D'
D''
E
F
C''
Vapor
Liquid
and
Vapor
Liquid
1
0
%
2
0
%
0
0.2
0.4
0.6
0.8
1.0
-50 0 50 100 150 200 250 300
Temperature [°C]
z
α
β
δ
δ'
1.4
2.0
4.2
5.6
7.0
B
P
L
0.7 [MPa]
D
P
L
β'
18 Phase Behavior
In Figure 2.12, the phase behavior of ethane/n-heptane systems is graphically illustrated
in the z,T-plane corresponding to another possible projection of the surface in Figure 2.9.
The mixture achieves bubble point β due to an isobaric (1.4 MPa) heat supply. Point β
symbolizes the composition of the liquid phase which is in equilibrium with an
infinitesimal small vapor phase whose composition is symbolized by the point β’. During
further increase of temperature, dew point state is reached at point . The composition of
the infinitesimal small liquid phase in equilibrium with the vapor phase corresponds with
point ’.
Figure 2.13: Mole fraction (ethane) - Pressure diagram of the binary system
ethane/n-heptane (from GYULAY, 1967)
The design of the corresponding p, z-diagram is also possible (see Figure 2.13). An
example may be the composition at the point A(T = 150
o
C). The composition of the liquid
phase is given by point A’, the one of the vapor phase by point A”. Again the relative
masses of both phases can be determined by applying the principle of the lever (see
Example 2.4).
Example 2.4.
Determining the phase composition.
A sealed container (p = 2.86 MPa), T = 150
o
C) is filled up with
100 kg of a ethane(z = 0.47)/n-heptane mixture. The mole number of
ethane in the liquid phase and in the vapor phase, respectively,
can be evaluated from Figure 2.13 by using the principle of lever.
α
δ
δ
0 0.2 0.4 0.6 0.8 1.0
A A' A''
1
2
3
4
5
6
7
8
9
z
P
r
e
s
s
u
r
e

[
M
P
a
]
B
P
D
P
2
0
0
°
C
1
5
0
°
C
1
0
0
°
C
Phase Behavior 19
At first, the mole weights (M
C2
= 30 kg/mole, M
C7
= 100 kg/mole)
are inserted into
to evaluate the weight of n-heptane in the system, m
C7
. The weight
of ethane, m
C2
, is given by m
C2
= 100 - m
C7
.
Now the total mole number of the system, n = n
C2
+ n
C7
, can be
calculated:
,
,
.
From Figure 2.12
where
n
liq
:total mole number in the liquid phase
n
vap
:total mole number in the vapor phase
Thus the total mole number in the vapor phase results in
The composition of the vapor phase is given by point A” in Figure
2.12:
The mole number of ethane in the vapor phase can now be calculated
by
z
m
C2
M
C2

m
C2
M
C2
m
C7
M
C7
⁄ + ⁄
----------------------------------------------------------
=
z 0.47
100 m
C7
– ( ) 30 ⁄
100 m
C7
– ( ) 30 m
C7
100 ⁄ + ⁄
----------------------------------------------------------------------
= =
m
C7
79.215kg =
m
C2
100 79.215 20.785kg = – =
n
C2
m
C2
M
C2
-----------
20.785
30
----------------
0.693kmole = = =
n
C7
m
C7
M
C7
-----------
79.215
100
----------------
0.792kmole = = =
n 0.693 0.792 1.485kmole = + =
n
l i q
n
vap
-----------
A A″
A A′
------------ 1.458 = =
n
vap
0.604 kmole =
z 0.82 =
20 Phase Behavior
.
The composition of the liquid phase is given by point A’ in Figure
2.12:
.
The total mole number in the liquid phase results in
.
The mole number of ethane in the liquid phase can now be calculated
by
.
Figure 2.14: Properties of ternary diagrams
2.5 Multi-Component Systems
2.5.1 Ternary Phase Diagrams
It is common to illustrate the phase behavior of 3-component systems at constant pressure
and temperature in so called triangular diagrams. Each corner of the triangle represents
one pure component. On the basis of the equilaterality of the triangle, the sum of the
perpendicular distances from any point to each side of the diagram is a constant equal to
length of any of the sides. Thus, the composition - expressed in mole fractions - of a point
n
C2 vap ,
z n
vap
× 0.82 0.672 × 0.495kmole = = =
z 0.23 =
n
liq
n n
vap
1.485 0.604 0.881 kmole = – = – =
n
C2 liq ,
0.23 0.881 0.203kmole = × =
3
3
3
3
2
2
2
2
1
1
1
1
a.
c.
b.
d.
L
2
L
1
L
3
Const. Fraction
Component1
C
o
n
s
t
a
n
t

R
a
t
i
o

o
f
1

t
o

2
A
D
B
Phase Behavior 21
in the interior of the triangle is given by
(2.11)
where
. (2.12)
Several other useful properties of the triangular diagrams are also illustrated by Figure
2.14:
• For mixtures along any line parallel to a side of the diagram, the fraction of the
component of the corner opposite to that side is constant.
• Mixtures lying on any line connecting a corner with the opposite side contain a
constant ratio of the component at the ends of the side.
• Mixtures of any two compositions lie on a straight line connecting the two initial
points on the ternary diagram. The principle of the lever finds application again and
(2.13)
gives the mixing ratio leading to mixture D.
Figure 2.15 shows the 2-phase region for chosen p and T. Any mixture with an overall
composition lying inside the binodal curve will split into a liquid and a vapor phase. The
“tie lines” connect compositions of liquid and vapor phases in equilibrium. Any overall
composition on a certain tie line gives the same liquid and vapor composition being in
euqilibrium. Only the amounts of the phases change as the overall composition changes.
Figure 2.15: Typical features of a ternary phase diagram
z
1
L
1
L
T
------ , = z
2
L
2
L
T
------, = z
3
L
3
L
T
------ , =
L
T
L
1
L
2
L
3
+ + =
n
A
n
B
------
DB
DA
--------
=
Plait
Point
Critical
Region
Liquid
Region
Two Phase
Region
Vapor
Region
Binodal Cu
rv
e
T
i
e

L
i
n
e
3 2
1
22 Phase Behavior
The liquid and vapor portions of the binodal curve meet at the “plait point” which
represents the critical composition. By drawing the tangent in the plait point on the
binodal curve, the single-phase region is splitted into three sections. Mixtures of a
composition being located in the critical region with another one being located in the
liquid or vapor region will, in any case, also result in a single-phase system if the straight
line connecting the two initial compositions does not intersect the 2-phase region.
Figure 2.16 illustrates the influence of pressure on the phase behavior of a certain ternary
system at constant temperature. As pressure increases, the 2-phase region shrinks.
It is useful to comprise the two heavier components of a ternary system and to reduce this
system to a fictitious binary system, on the basis of a hypothetical component. Figure 2.17
illustrates a corresponding application by the respective p,T-diagram of the
methane/propane/n-pentane system. The mole-% of methane are specified along the
outermost envelope curve. All envelope curves are characterized by the portion of
propane in the hypothetical component (propane/n-pentane) which is given by
(2.14)
In accordance to this aspect, the critical state properties, p
c
and T
c
, can be determined for
any mixture of the three components (see Example 2.5).
C
z
3
z
3
z
5
+
----------------
=
Phase Behavior 23
Figure 2.16: Triangular diagrams for the methane/propane/n-pentane system at 160
o
F(71
o
C) (after DOURSON et al., 1943)
Propane
Propane
Propane
n-Pentane
n-Pentane
n-Pentane
Methane
Methane
Methane
0.8
0.8
0.8
0.8
0.8
0.8
0.2
0.2
0.2
0.6
0.6
0.6
0.6
0.6
0.6
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.6
0.6
0.6
0.2
0.2
0.2
0.2
0.2
0.2
0.8
0.8
0.8
p=500 [psia]
p=1000 [psia]
p=1500 [psia]
T=160°F
T=160°F
T=160°F
a.
b.
c.
C
=
1
.
0
C
=
1
.
0
C
=
0
.
0
C
=
0
.
0
C
=
0
.
0
C
=
0
.
2
C
=
0
.
2
C
=
0
.
2
C
=
0
.
4
C
=
0
.
4
C
=
0
.
4
C
=
0
.
6
C
=
0
.
6
C
=
0
.
6
C
=
0
.
8
C
=
0
.
8
1
]
1

¸

+
·
5 3
3
c c
c
x x
x
C
1
]
1

¸

+
·
5 3
3
c c
c
x x
x
C
1
]
1

¸

+
·
5 3
3
c c
c
x x
x
C
24 Phase Behavior
Figure 2.17: Critical loci of methane/propane/n-pentane systems (from KATZ et al.,
1959)
Example 2.5
The hydrocarbon mixture is composed of 8 [kg] methane (M = 16[kg
kmol
-1
], 13,2 [kg] propane (M = 44.1[kg kmol
-1
]) and 32.5 [kg]
n-pentane (M = 72.2[kg kmol
-1
]). The critical data of this mixture
can be evaluated by use of Figure 2.17.
At first, the mole numbers and the respective mole fractions must
be calculated.
,
,
,
,
,
.
The portion of propane in the hypothetical component
propane/n-pentane is given by
0
500
1000
1500
2000
2500
3000
-200 -100 0 100 200 300 400
Temperature [°F]
C
r
i
t
i
c
a
l

P
r
e
s
s
u
r
e

[
p
s
i
a
]
C
1 C
3
C
5
0
10
20
30
40
50
60
70
80
75
M
o
l
e

[
%
]

C
H
4
C
=1
.0
C
=
0
.
6
C
=
0
.8
C
=
0
.
0
C
=
0
.
2
C
=
0
. 4
1
]
1

¸

+
·
5 3
3
nC C
C
C
n
1
8
16
------
0.5kmole = =
n
3
13.2
44.1
----------
0.3kmole = =
n
5
32.5
72.2
----------
0.45kmole = =
z
1
0.5
0.5 0.3 0.45 + +
--------------------------------------
0.4 = =
z
3
0.3
0.5 0.3 0.45 + +
--------------------------------------
0.24 = =
z
5
0.45
0.5 0.3 0.45 + +
--------------------------------------
0.36 = =
Phase Behavior 25
.
From Figure 2.17 at C = 0.4 and 40 mole percent methane:
and
.
The application of the triangular diagram is not solely confined to ternary systems. For
example it is possible to partition the paraffinic hydrocarbons into three
pseudo-components which are
• methane (C
1
) as the light component,
• the lighter pseudo-component including ethane to hexane (C
2
-C
6
),
• the heavier pseudo-component including heptane and higher hydrocarbons (C
7+
).
Anyway, only poor information of complex natural hydrocarbon systems has been
reported until now. Nevertheless, some generalization makes a description of these
complex systems possible - according to known data. The phase behavior of several
complex and natural hydrocarbon systems are demonstrated in Figure 2.18 to Figure 2.22
by p, T-phase diagrams. For the classification of natural hydrocarbon systems, it is
essential to know
• if the critical temperature is lower or higher than the reservoir temperature,
• which state will be achieved at surface conditions (separator).
Not considered in this classification are changes in composition during production.
Figure 2.18 represents a hydrocarbon system whose critical temperature is significantly
lower than the reservoir temperature. In case of an isothermal pressure decrease (full line
from point 1 to 2), which occurs in the reservoir adjacent to the production well the the
course of production, the system remains in the single-phase (gaseous) state. Even in case
of both pressure and temperature decrease (dotted line), no liquid phase will drop out.
Consequently, the considered hydrocarbon mixture is called a “dry gas”. Dry gases
contain mainly methane, small amounts of ethane, possibly propane and some
hydrocarbons of higher molecular weights.
A so called “wet gas” (see Figure 2.19) remains in a single-phase (gaseous) state in the
reservoir during production (line 1-2). Anyway, condensate will drop out under separator
conditions.
C
z
3
z
3
z
5
+
----------------
0.24
0.24 0.36 +
---------------------------
0.4 = = =
T
c
262.5°F 128°C = =
p
c
1344 psia 9.27MPa = =
26 Phase Behavior
In case of the system shown in Figure 2.20, the reservoir temperature is higher than the
critical one but lower than the cricondentherm. The initial conditions given by Point 1
specifies the hydrocarbon mixture as a dry gas. If the pressure will decrease adjacent to
the production well during production, the dew point of the system is reached at point 2.
Consequently, condensate drops out inside the reservoir. The pressure at point 3
corresponds to the state in which the condensed liquid phase reaches the maximum (in
mole%). In the separator, the amount of condensate is larger than in case of wet gases.
Systems as shown in Figure 2.20 are called “gas condensates”.
Figure 2.18: Phase diagram of a dry gas (from MCCAIN, 1973)
Figure 2.19: Phase diagram of a wet gas (from MCCAIN, 1973)
Figure 2.20: Phase diagram of a retrograde gas condensate (from MCCAIN, 1973)
P
r
e
s
s
u
r
e
Temperature
Gas
Liquid Sep.
1
2
Critical
Point
75
50
25
P
r
e
s
s
u
r
e
Temperature
Gas
Liquid
1
2
Critical
Point
Mole % Liq.
75
50
25
5
0
100
Sep.
P
r
e
s
s
u
r
e
Temperature
Gas
Liquid
1
2
Critical
Point
Mole % Liq.
75
50
25
5
0
10
100
Sep.
3
Phase Behavior 27
Figure 2.21: Phase diagram of a high-shrinkage crude oil (from MCCAIN, 1973)
The so called “white oils” - as characterized in Figure 2.21 - are referred to as “high
shrinkage oils”. The reservoir temperature is below the critical temperature. Since the
bubble point curve will be reached by the decrease in pressure due to production, from the
initial pressure (point 1) to the pressure 2, a further pressure drop in the reservoir will lead
to point 3 and thus to an increased development of the vapor phase. At separator
conditions, about 65% of the produced hydrocarbon mixture will exist as liquid phase if
the reservoir is produced at bubble point conditions.
Figure 2.22: Phase diagram of a low shrinkage crude oil (from MCCAIN, 1973)
Figure 2.22 shows a “black oil” or “low shrinkage oil”. The initial state is characterized
by point 1 at which the state of the system can be regarded as “undersaturated” liquid. If
the pressure in the neighbourhood of the production well will decrease during production
to point 2, the bubble point curve is reached and the state of the system is now considered
“saturated”. The separator conditions are near the bubble point curve. Consequently,
about 85 mole% of the produced hydrocarbon mixture is in the liquid phase at separator
conditions. In accordance to this fact, the shrinkage of the oil due to gas liberation is less
pronounced than in case "white oils" (see Figure 2.21).
If the hydrocarbon mixture in the reservoir is a 2-phase state under initial reservoir
conditions, oil and gas phase can be considered apart from one another (see Figure 2.23).
The equilibrium conditions at the initial state of the system are given by the intersection
point of the dew point curve of the gas cap and the bubble point curve of the oil zone.The
P
r
e
s
s
u
r
e
Temperature
Gas
Liquid
1
2
Critical
Point
Mole % Liq.
Sep.
3
75
50
25
100
P
r
e
s
s
u
r
e
Temperature
Gas
Liquid
Critical Point
Mole % Liq.
Sep.
3
75
50
25
0
100
B
u
b
b
l
e
-
P
o
i
n
t

L
i
n
e
D
e
w
-
P
o
i
n
t

L
i
n
e
1 Undersaturated
2 Saturated
28 Phase Behavior
gas cap shows a “retrograde” behavior, if the intersection point is located on the dew point
curve of the gas cap between the critical point and the cricondentherm.
Just as in case of binary systems (see Figure 2.9), the phase behavior of natural
hydrocarbon mixtures can also be illustrated in p, T, z-diagrams.
Figure 2.23: Phase diagram pairs of gas cap and oil zone
In Figure 2.24, composition I represents the separator gas while composition IV
represents the corresponding separator oil of the well stream. Furthermore, the phase
behavior of two representative mixtures of I and IV are given by the compositions II and
III. The system of composition II corresponds to a gas-condensate system, the one of
composition III to a white oil. Inside the 2-phase region of system II and III, isochores of
the liquid phase and - as dotted lines - the locations of maximum retrograde condensation
are drawn. Again an envelope surface comprises the 2-phase region in dependence on the
composition. The spatial curve, which connects the critical points, splits the surface into
two parts which are the dew point surface and the bubble point surface. Outside the
envelope surface, the system is in a single-phase state.
By projecting the phase surface into the p, z-plane, information about the composition of
the system will be obtained. If the state of the system is represented by point 1, the
equilibrium composition of the liquid (x
1
), and the one of the vapor phase, (y
1
), is given
by point 4 in the p, z-plane.
Temperature Temperature
P
r
e
s
s
u
r
e
P
r
e
s
s
u
r
e
C
C
C
C
Oil Zone
Oil Zone
p
i
p
i
Retrograde Condensating Gas Cap Gas Cap
Phase Behavior 29
Figure 2.24: Phase equilibrium surface of oil/natural gas systems (from GYULAY, 1967)

C''
C'
T
1
T
2
T
3
I
II
III
IV
I
II
III
IV
T
1
x
1
y
1
P
d
P
b
T
2
T
3
C
I
4
C
I
C'
C
I
C''
I
C
1
C
II
C
II
C
II
C'
C''
K
II
C
III
C
III
C'
C
III
C
IV
C
IV
C
IV
C''
L
III
IV
G
GL
GL'
C
0
.
0 0
.
2
0
.
4
0
.
6
0
.
8
1
.
0
0
.
0
0
.
2
0
.
4
z
P
30 Phase Behavior
31
Chapter 3
Equations of State
The preliminary chapter included graphical illustrations of equilibrium surfaces and their
normal projections into the (p,T), (p,V), (p,z) etc. planes. The application of such diagrams
enable the determination of the respective volume for certain states defined by the
corresponding pressure p and temperature T. However, a pure graphical application of the
state functions is not very practical and - on top of that - impossible for multi-component
systems. This aspect obviously makes a mathematical consideration of these problems
necessary.
On the basis of Eq. 2.1, the following relation regarding the p, V, T-data is valid in case
of any chemically homogeneous phase:
(3.1)
In this context, V must be defined as the volume of one mole (intensive property). In case
of a 2-phase system:
(3.2)
and therefore
(3.3)
where
V
t
: mole volume of the system
V
liq
: mole volume of the liquid phase
V
vap
: mole volume of the vapor phase
n: number of moles in the system
n
liq
: number of moles in the liquid phase
n
vap
: number of moles in the vapor phase.
If phase equilibrium is given, the phases can be regarded as seperate thermodynamic
systems. If a phase - may be the liquid - phase consists of k components, the corresponding
equation of state may be written as follows:
V V p T , ( ) =
n n
l i q
n
vap
+ =
nV
t
n
liq
V
liq
n
vap
V
vap
+ =
32 Equations of State
(3.4)
or
(3.5)
where x
i
is defined as the mole fraction of the component i in the (liquid) phase.
Considering the mole fractions, the so called “constraint equation” is valid:
. (3.6)
In various cases of even practical interests, a multi-component phase behaves as an ideal
mixture and the volumes are strictly additive. If V
i
is defined as the mole volume of
component i in the phase, the mole volume of the phase will result in
(3.7)
If Eq. 3.7 is valid, the enthalpy of the system must be generally considered additive. This
means that the enthalpy of the system is equal to the sum of the enthalpies of the single
components. In this case, no thermal effect will take place during the mixing procedure.
Put into other words: The mixing energy will be zero.
Regarding Eq. 3.6 in discussing Eq. 3.5, it is obvious that V is a function of 2 + (k - 1) =
k + 1 variables. It is impossible to approximate the equilibrium surface for the entire (p,
V, T, x
k
)-space by one single equation of state. Therefore some procedure bit by bit is
necessary.
3.1 Change of State at Low Compressibility
The expansion of Eq. 3.1 - or just the same of Eq. 3.5 at constant composition - into a
TAYLOR-Series leads to
(3.8)
By assuming that the higher derivations are neglectable, Eq. 3.8 may be truncated to
F p, V, T, x
1
x
2
, …x
p
( ) 0 =
V V = p, T, x
1
x
2
, …x
k
( )
x
i
1 =
i 1 =
k

V x
i
V
i
i 1 =
k

=
V p T , ( ) V p
o
T
o
, ( )
V ∂
p ∂
------
¸ ,
¸ _
T
p p
o
– ( )
1
2
---

2
V
∂p
2
---------
¸ ,

¸ _
T
p p
o
– ( )
2
x
V p T , ( ) V p
o
T
o
, ( )
V ∂
T ∂
------
¸ ,
¸ _
p
T T
o
– ( )
1
2
---

2
V
∂T
2
---------
¸ ,

¸ _
p
T T
o
– ( )
2
...... + + + =
+ + + =
Equations of State 33
(3.9)
where
.
Considering the Eq. 2.4 and Eq. 2.5, Eq. 3.9 may be transformed to
(3.10)
where
: isothermal compressibility
: cubic expansion coefficient
Eq. 3.10 can at best be applied for fluids in a 1-phase state. Experience and practice have
shown that the cubic equations of state are most sufficient and beneficial for calculating
the state of gases and of 2-phase systems. Of course, there may exist equations which
approximate the measured values more accurately. Anyway, the constants included in
these equations are not always given. Therefore, only the cubic equations of state,
particularly the PENG-ROBINSON equation, will be discussed in this textbook. In doing so,
the proceedings of generalization in derivating this equation of state will be elucidated.
3.2 Equation of State of Perfect and Real Gases
A gas is defined as perfect, if the intermolecular (VAN DER WAALS) forces are
neglectable.
Then for a molar system:
(3.11)
where R is defined as “universal gas constant” and
R = 8.31434 J/mole K
The compressibility factor Z is defined as
(3.12)
V p T ( , ) V
o
1
1
V
o
------
∂V
∂p
------
¸ ,
¸ _
T
p p
o
– ( )
1
V
o
------
∂V
∂T
------
¸ ,
¸ _
p
T T
o
– ( ) + + =
V
o
V p
o
T
o
, ( ) =
V p T , ( ) V
o
1 κ p p
o
– ( ) – α T T
o
– ( ) + [ ] =
κ
α
pV RT =
Z
pV
RT
-------
=
34 Equations of State
or
, (3.13)
respectively.
For ideal gases, the factor Z equals 1. For real gases, Z is a state variable and depends on
the pressure, the temperature and the composition. The critical point is defined by the
pressure p
c
, temperature T
c
, and specific volume V
c
and may be determined
experimentally for 1-component systems.
The critical compressibility factor, Z
c
, can be evaluated by substituting the critical data p
c
,
T
c
and V
c
into Eq. 3.12:
. (3.14)
Example 3.1
Evaluation of a Z-factor from laboratory data. A cylinder with
volume of 0.075 m
3
has been filled with a gas under atmospheric
pressure and at a temperature of 90.5
o
C. Then the volume of the
cylinder has been isothermally reduced to 0.00035 m
3
(volume
reduction in consequence of mercury insertion). After the volume
reduction, the pressure has been recorded as 13.79 MPa.
If the gas would show an ideal behavior at 90
o
C and 13.79 MPa, the
specific volume could be calculated by

and so
,
.
By use of Eq. 3.13 and considering the measured volume amounting
to 0.00035 m
3
:
Z
V
actual
V
ideal
-----------------
=
Z
c
p
c
V
c
RT
c
-----------
=
P
1
V
1
P
2
V
2
=
V
2
V
1
p
1
p
2
----- =
V
2
75 10
3 –
×
101
3 –
×10
1379
2
×10
------------------------ 55
5 –
×10 m
3
= =
Z
35
5 –
×10
55
5 –
×10
-------------------- 0.63 = =
Equations of State 35
3.3 Cubic Equations of State
If the pressure of any gaseous system is low, the ideal gas equation remains sufficient to
describe the volumetric behavior. In the year 1873, VAN DER WAALS deducted the first
equation of state which is able - up to a certain degree - to describe the continuity from
gaseous to the liquid state:
. (3.15)
a and b are substance specific constants. b can be interpreted as the inherent volume of the
molecules which is not available for the thermal motion of the molecules. The term a/V
2
regards the pressure reduction in consequence of intermolecular attraction.
Eq. 3.15 may also be written in the following form:
. (3.16)
Figure 3.1 illustrates the VAN DER WAALS isotherms in the vicinity of the critical point.
The dotted section of the isotherms represents the the data which are predicted by using
the VAN DER WAALS equation. Obviously, Eq. 3.16 cannot predict the real behavior of the
system during the vaporization, respectively condensation. The real behavior is shown by
the straight full line BD inside the 2-phase region.
The cubic equations of state, which have been formulated by REDLICH and KWONG
(1949), SOAVE (1972), and PENG and ROBINSON (1976), have achieved much better
results.
Figure 3.1: The VAN DER WAALS isotherms near the critical point
p
a
V
2
------
+
¸ ,
¸ _
V b – ( ) RT =
p
RT
V b – ( )
-----------------
a
V
2
------
– =
Critical Point
Calculated from
Van der Waals
Equation
T>T
2 1
T
=
c
o
n
s
t
.
2
Actual
Path
P
r
e
s
s
u
r
e
Specific Volume
T
=
c
o
n
s
t.
1
B
D
36 Equations of State
The REDLICH-KWONG equation is given as
(3.17)
where b again is a substance specific constant. Anyway. a is now a function of the
temperature. It is useful to write the parameter a as follows:
(3.18)
where is constant. The original REDLICH-KWONG equation included
.
By multiplication of Eq. 3.17 with
and after arrangement of V corresponding to its order of power:
. (3.19)
At the critical point :
. (3.20)
The comparison of Eq. 3.19 and Eq. 3.20 leads to
, (3.21)
, (3.22)
(3.23)
where
. (3.24)
Substitution of Eq. 3.21 and Eq. 3.23 into Eq. 3.22 results in
p
RT
V b – ( )
-----------------
a
V V b + ( )
---------------------
– =
a a′f T ( ) =
a′
f T ( ) T
0.5 –
=
V V b + ( ) V b – ( ) p ⁄
V
3
RT
p
------- V
2

a
p
---
bRT
p
----------
– b
2

¸ ,
¸ _
V
ab
p
------
0 = – +
V V
c
= ( )
V V
c
– ( )
3
V
3
3V
c
V
2
– 3V
2
c
V V
3
c
0 = – + =
3V
3
RT
c
p
c
---------
=
3V
2
c
a
c
p
c
-----
bRT
c
p
c
-------------
– b
2
– =
V
3
c
a
c
b
p
c
--------
=
a
c
a′f T
c
( ) =
Equations of State 37
(3.25)
or after rearrangement, in
, (3.26)
, (3.27)
. (3.28)
Furthermore, Eq. 3.28 and Eq. 3.21 can be combined to
. (3.29)
Inserting Eq. 3.21 and Eq. 3.23 into Eq. 3.29:
(3.30)
The constants and have the following numerical values:
= 0.08664,
= 0.42748.
From Eq. 3.18 and Eq. 3.24:
. (3.31)
It is obvious that , if .
The substitution of into the Eq. 3.19 and the arrangement of Z
corresponding to its order of power results in
. (3.32)
Therefore:
3V
2
c
V
3
c
b
-------
3bV
c
b
2
– – =
b
3
3b
2
V
c
3bV
2
c
V +
3
c
2V
3
c
= + +
b V
c
+ ( )
3
2V
3
c
=
b 2
1 3 ⁄
1 – ( )V
c
=
b
2
1 3 ⁄
1 – RT
c
3p
c
------------------------------

b
RT
c
p
c
--------- = =
a
c
RT
c
( )
2
9 2
1 3 ⁄
1 – ( )p
c
---------------------------------- Ω
b
RT
c
( )
2
p
c
----------------- = =

b

a

b

a
a a
c
f T ( )
f T
c
( )
------------ a
c
α = =
α 1 = T T
c
=
V ZRT p ⁄ =
ZRT
p
-----------
¸ ,
¸ _
3
RT
p
-------
ZRT
p
-----------
¸ ,
¸ _
2

a
p
---
bRT
p
----------
– b
2

¸ ,
¸ _
ZRT
p
-----------
ab
b
------ 0 = – +
38 Equations of State
(3.33)
or
(3.34)
where
(3.35)
and
. (3.36)
The substitution of Eq. 3.29 and Eq. 3.30 into Eq. 3.35 and Eq. 3.36 leads to
(3.37)
and
. (3.38)
In the original REDLICH-KWONG equation:
(3.39)
and so
. (3.40)
Eq. 3.34 includes only two parameters which are p
c
and T
c
. Please note that the ideal gas
equation contains no substance-specific parameters. Since the Redlich-Kwong cubic
equations of state consider these two substance-specific parameters, it has improved the
calculation of PVT-properties in a fundamental way. Anyway, the increasing yield of
experimental data has more and more indicated that the behavior of many liquids with a
non-spherical molecule structure deviates greatly from the predicted one. This made the
introduction of a third factor necessary.
Z
3
Z
2

ap
RT ( )
2
--------------
bp
RT
-------
b
2
p
2
RT ( )
2
--------------
– – Z
ap
RT ( )
2
--------------
bp
RT
------- 0 = – +
Z
3
Z
2
– A B B
2
– – ( )Z AB 0 = – +
A
ap
RT ( )
2
--------------
=
B
bp
RT
-------
=
A 4.2748
p
r
T
2
r
------
¸ ,



¸ _
α =
B 0.8664
p
r
T
r
----- =
f T ( ) T
0.5 –
=
α T
r
0.5 –
=
Equations of State 39
Beginning with the year 1951 (MEISSNER and SEFERIAN), several proposals for a third
parameter have been made. The so called “acentric factor”, , has become the one with
greatest acknowledgement:
(3.41)
where is the reduced boiling point pressure.
The equation of state from SOAVE (1972) only differs from the REDLICH-KWONG equation
with respect to the definition of the factor
. (3.42)
The weakness of all these equations ranging from the original REDLICH-KWONG equation
to all its modifications (including the SOAVE equation) is the fact of an universal
unrealistic Z
c
factor of . Moreover, the prediction of liquid density is combined with
large errors.
Improved approximation has been achieved with the PENG-ROBINSON equation:
(3.43)
where a is given by Eq. 3.18 and b is further a substance specific constant.
Just as in case of the REDLICH-KWONG equation, the following terms and equations can
be obtained:
, (3.44)
, (3.45)
, (3.46)
, (3.47)
, (3.48)
ω
ω lgp
s
r
1 +
¸ ,
¸ _
at T
r
0.7 = – =
p
s
r
p
s
p
c
⁄ =
α
0.5
1 0.48 1.574 + ω 0.176ω
2
– ( ) + = 1 T
r
0.5
– ( )
1 3 ⁄
p
RT
V b –
------------
a
V V b + ( )+ b V b – ( )
------------------------------------------------
– =
V
3 RT
p
-------
b –
¸ ,
¸ _
V
2

a
b
---
2bRT
p
--------------
– 3b
2

¸ ,
¸ _
V b
a
p
---
RT
p
------- b – b
2

¸ ,
¸ _
0 = – +
V
c
0.307
RT
c
p
c
--------- =
b 0.07796
RT
c
p
c
--------- =
a a
c
α =
a
c
0.457235
RT
c
( )
2
p
c
----------------- =
40 Equations of State
(3.49)
, (3.50)
, (3.51)
. (3.52)
Analogies between Eq. 3.44 to Eq. 3.52 on the one side and between Eq. 3.19, Eq. 3.21,
Eq. 3.29 to Eq. 3.31, Eq. 3.34 to Eq. 3.38, and Eq. 3.42 on the other side are obvious. The
universal critical Z-factor of the PENG-ROBINSON equation results in 0.307 which is much
better than but still far away from reality. Anway, the calculated fluid densities are
much more accurate than the ones calculated by the equations of state previously
discussed.
Eq. 3.16, Eq. 3.17, and Eq. 3.43 have been established for pure substances. The extension
for multi-component systems requires the calculation of the respective data of the pure
components and mixing rules in order to get the parameters of the mixture.
The mixing rule for the parameter b, which is included in the equations of
REDLICH-KWONG, SOAVE and PENG-ROBINSON, is universally defined as an arithmetic
average by using
. (3.53)
For the temperature-dependent coefficient a , different mixing rules exist and are
presented below.
REDLICH-KWONG:
. (3.54)
where a
i
can be calculated on the basis of the critical data of each component by using Eq.
3.30 and Eq. 3.40 into Eq. 3.31.
Z
3
1 B – ( )Z
2
– A 2B – 3B
2
– ( )Z AB B
2
B
3
– – – – + 0, =
A 0.457325
p
r
T
2
r
------
¸ ,



¸ _
α ⋅ =
B 0.07796
p
r
T
r
----- =
α
0.5
1 0.3676 1.54226ω 0.26992ω
4
– +
¸ ,
¸ _
1 T
r
1 2 /

¸ ,
¸ _
+ =
1 3 ⁄
b x
i
b
i
i 1 =
k

=
a x
i
a
i
0.5
i 1 =
k

2
=
Equations of State 41
SOAVE:
(3.55)
where a
i
and a
j
, respectively, can be evaluated by Eq. 3.30 and Eq. 3.42.
PENG-ROBINSON:
(3.56)
where a
i
is defined by Eq. 3.48 and Eq. 3.52.
The mixing rules used by SOAVE and PENG-ROBINSON consider the binary interaction
between the molecules of the components i and j. In Eq. 3.55 and Eq. 3.56, the terms
are binary interaction coefficients which are assumed to be independent of pressure and
temperature.
Values of the binary interaction coefficients must be obtained by fitting the equation of
state to gas-liquid equilibria data for each binary mixture. They have different values for
each binary pair and also take on different values for each equation of state.
Obviously, Eq. 3.55 and Eq. 3.56 reduce to the form of Eq. 3.54 if all binary interaction
coefficients are zero.
Another possibility of obtaining this coefficient - if no data are available - is by mean of
matching the phase behavior of multi-component systems.
Example 3.2
Calculation of the pressure by use of the PENG-ROBINSON equation
of state. A laboratory cell at temperature of 100 with volume
of 0.00025 m
3
contains 0.25 mole of gas. The composition of the
system, the critical data and the acentric factors of the
components are tabled below.
a x
i
j 1 =
k

i 1 =
k

x
j
a
i
a
j
( )
0.5
1 k
i j
– ( ) =
a x
i
j 1 =
k

i 1 =
k

x
j
a
i
a
j
( )
0.5
1 k
i j
– ( ) =
k
i j
C
o
42 Equations of State
First, the parameters b and a must be calculated for each component.
For methane from Eq. 3.46:
,
from Eq. 3.48:
,
from Eq. 3.52 by inserting the acentric factor for methane
and the reduced temperature of methane
:
and thus
,
and from Eq. 3.47:
.
Correspondingly to the calculations for methane, the parameters of ethane and n-butane
were evaluated. They are tabled below.
Component Composition
y
i
Critical Data
T
c,i
p
c, i
o
K MPa
Acentric Factor
C
1
0.75 190.6 4.60 0.0115
C
2
0.20 305.4 4.88 0.0908
n - C
4
0.05 425.2 3.80 0.1928
ω
i
b
1
77796 10 ×
6 – 831434 1907 10
6
× ×
463 10
8
×
--------------------------------------------------- 2677 10
5 –
× = =
a
c,1
457235 10
6 –
×
831434 1907 10
6
× × ( )
2
463 10
8
×
---------------------------------------------------------- 0.2494 = =
ω
1
0.0115 =
T
r 1 ,
273.15
190.6
----------------
1.958 = =
α
0.5
1
0.8954 =
α
1
0.8018 =
α
1
0.2494 0.8018 0.1999 = × =
Equations of State 43
The parameter b can be calculated by use of Eq. 3.53:
.
To calculate the coefficient a by use of Eq. 3.56, the interactive
coefficients, K
ij
, must be known. They are given below:
a = 0.75 x 0.75(0.1999 x 0.1999)
0.5
(1 - 0.000000) +
+ 0.75 x 0.20(0.1999 x 0.6045)
0.5
(1 - 0.002648) +
+ 0.75 x 0.05(0.1999 x 1.8260)
0.5
(1 - 0.014640) +
+ 0.20 x 0.75(0.6045 x 0.1999)
0.5
(1 - 0.002648) +
+ 0.20 x 0.20(0.6045 x 0.6045)
0.5
(1 - 0.000000) +
+ 0.20 x 0.05(0.6045 x 1.8260)
0.5
(1 - 0.004904) +
+ 0.05 x 0.75(1.8260 x 0.1999)
0.5
(1 - 0.014640) +
+ 0.05 x 0.20(1.8260 x 0.6045)
0.5
(1 - 0.004904) +
+ 0.05 x 0.05(1.8260 x 1.8260)
0.5
(1 - 0.000000) = 0.3108
Now the pressure can be calculated through Eq. 3.43 and by
inserting
,
,
Component b
i
a
c,i
α
i
a
i
C
1
2.677 E-5 0.2494 0.8018 0.1999
C
2
4.048 E-5 0.6042 1.0010 0.6045
n - C
4
7.245 E-5 1.5050 1.2130 1.8260
C
1
C
2
n - C
4
C
1
0.0000 E+0 0.2648 E-2 0.1464 E-1
C
2
0.2648 E-2 0.0000 E+0 0.4904 E-2
n - C
4
0.1464 E-1 0.4904 E-2 0.0000 E+0
b 0.75 2.67E 5 0.20 4. 048E 5 – 0. + × + – × 05 7. × 245E 5 – =
b 3. 180E 5 – =
V
0.00025
0.25
---------------------
0.001 m
3
mole ⁄ = =
T 373.15 K =
R 8.31434 J K mole × ⁄ =
44 Equations of State
and
.
The calculation procedure results in
Example 3.3
Calculation of the density of methane by use of the PENG-ROBINSON
equation of state. 2.0 kg methane (MC
1
= 16 kg k/mole is hold at
temperature of 305.12 K and at pressure of 9.26 MPa. The critical
data and the acentric factor of methane are known as
,
,
.
First, the parameters a and b have to be calculated.
From Eq. 3.46:
.
From Eq. 3.48:
.
From Eq. 3.52 by inserting the acentric factor and the reduced
temperature of the system
,
and thus
,
and from Eq. 3.47:
.
Now the mole volume of methane at the given state variables has to
be evaluated through Eq. 3.44. To solve this cubic equation of
b 3.18
5 –
×10 =
a 0.3108 =
p 2.328 = MPa
T
c
190.6 K =
p
c
4.63 MPa =
ω 0.0115 =
b 0.077796
8.31434 190.6 ×
4.63 10
6
×
------------------------------------------ 2.662
5 –
×10 = =
a
c
0.457235
8.31434 190.6 × ( )
2
463 10
4
×
------------------------------------------------ 0.248 = =
T
r, 1

305. 12
190. 6
------------------
1. 601 = =
α
0.5
0.8954 =
α 0.8018 =
a 0.248 0.8018 0.1988 = × =
Equations of State 45
state, the CARDAN equation is applied for
where
,
and
.
The calculation of r, s and t result in
,
,
and
.
By substitution of
,
the formula
is reduced to
where
and
.
The calculation of p and q result in
x
3
rx
2
sx t 0 = + + +
r
RT
p
-------
b –
¸ ,
¸ _
– =
s
a
p
---
2bRT
p
--------------
– 3b
2

¸ ,
¸ _
=
t b
a
p
---
RT
p
-------
– b b
2

¸ ,
¸ _
– =
r 0. 2473
3 –
×10 – =
s 0.4755
8 –
×10 =
t 0. 3585
12 –
×10 – =
x y
r
3
---
– =
x
3
rx
2
sx t 0 = + + +
y
3
py q 0 = + +
p
3s r
2

3
----------------
=
q
2r
3
27
--------
rs
3
-----
– t + =
p 1. 563
8 –
×10 – =
46 Equations of State
and
.
The discriminant D is defined as
and thus
.
Based on the relationships
,
and
,
the mole volume can be calculated through
:
,
,
and thus
,
and
.
q 1. 089
12 –
×10 – =
D
p
3
---
¸ ,
¸ _
3
q
2
---
¸ ,
¸ _
2
+ =
D 1.53776
25 –
×10 =
u
q
2
---

3
D + =
v
q
2
---
– 3 D – =
y u v + =
x y
r
3
---
– =
u 9.7800
5 –
×10 =
v 5.3272
5 –
×10 =
y 1.5107
4 –
×10 =
V 2. 3350
4 –
×10 m
3
mole ⁄ =
ρ
m
nV
-------
M
V
----- 68.522 kg m
3
⁄ = = =
Equations of State 47
3.4 Virial Equation of State
All cubic equations of state mentioned above are more or less empirical or at best
semi-theoretical. However, they are obviously qualified for practical application.
In contrast, the virial equation of state has some theoretical background in statistical
mechanics. However, the form of an infinite series concerning the volume is not very
appropriate for practical application.
The pressure is expressed as
. (3.57)
The virial coefficients B, C etc. are solely a function of temperature. Sometimes Eq. 3.57
is applied for the calculation of gaseous states (e.g. natural gas) by considering only the
first two or three terms (depending on the availability of the virial coefficients).
If V is substituted into Eq. 3.57 according to the gas law by Z RT/p, then:
(3.58)
The second virial coefficient B can be shown to be the slope of isotherms on a plot of
. When p approaches zero, the third term in Eq. 3.58 may be
neglected so that
. (3.59)
Since also Z becomes zero as p goes to zero, de l’Hospital’s rule can be applied resulting
at constant T in
. (3.60)
Eq. 3.60 implies that the 2. virial coefficient vanishes at the BOYLE-temperature. The
BOYLE temperature is defined as the temperature above which Z-values become greater
than 1 for all pressures.
p RT
1
V
---
B
V
2
------
C
V
3
------ … + +
¸ ,
¸ _
=
Z 1
Bp
ZRT
-----------
Cp
2
ZRT ( )
2
------------------
… + + + =
Z Z p ( ) at p 0 = =
B ZRT
Z 1 –
p
------------
¸ ,
¸ _
p 0 →
lim =
B RT
∂Z
∂p
------
¸ ,
¸ _
p 0 →
lim
T
=
48 Equations of State
49
Chapter 4
Calculation of Phase Equilibria
4.1 Mixtures
4.1.1 Definitions
In case of thermodynamical phase equilibrium of a multi-component system, all phases
are - in physical terms - homogeneous mixtures. Vaporized compounds are always
mixable in all relations, independent of their chemical characteristics. Therefore, only one
vapor phase exists.
Compounds existing in a liquid phase are only mixable in all relations if they are
chemically similar. Since hydrocarbons - especially whose which belong to the same
homologous series - exhibit such chemical conformity, their mixtures will be physically
homogeneous, independent of the composition of the mixture.
The composition of the system and its phases are specified by mole fractions. The mole
fraction of the component i is given by the ratio between its mole number, n
i
, and the total
mole number in the system (in the respective phase), n. The sum of all mole fractions
equals 1. In the following, the different mole fractions are marked by
z
i
: mole fraction of the component i in the total system
x
i
: mole fraction of the component i in the liquid phase
y
i
: mole fraction of the component i in the vapor phase,
so that in the system in total
, (4.1) z
i
n
i
k
i =1
n
i

----------------------- =
50 Calculation of Phase Equilibria
, (4.2)
. (4.3)
If a multi-component system is in the state of thermodynamic phase equilibrium, the
distribution of the component i among the vapor and the liquid phase is characterized by
its distribution coefficient, the so called “K-factor”:
. (4.4)
The numerical value of K
i
depends on the state of the system. Accordingly, the
composition of the phases will change at any variation in state.
4.1.2 K-factors
Ideal liquid mixtures have the property of being met by RAOULT’s law so that
(4.5)
where
p
i
: partial pressure of the component i
: tension of component i as a single-component system
x
i
: mole fraction of component i in the liquid phase.
The total pressure of the system adds up to
(4.6)
If a vapor phase exhibits perfect behavior, DALTON’s law can be applied and
(4.7)
The combination of Eq. 4.5 and Eq. 4.7 results in
x
i
n
i
k
i=1
n
i

----------------------
¸ ,



¸ _
liq
=
y
i
n
i
k
i=1
n
i

-----------------------
¸ ,



¸ _
vap
=
K
i
y
i
x
i
----
=
x
i
p
i
p
o
i
------
=
p
o
i
p p
i
i= 1
k

x
i
p
o
i
i= 1
k

= =
y
i
p
i
p
----
=
Calculation of Phase Equilibria 51
(4.8)
Anyway, certain limitations are placed on the definition of the K-factor given above. It is
evident that any pure coomponent has a vapor pressure only up to its critical temperature.
In practice, the extrapolation of the vapor pressure curves to higher temperatures than T
c
does not yield satisfactory results in most cases. Since T
c
= -190.37 K for methane, it
becomes evident that some other means must be obtained for the determination of
K-factors if they are to be used for hydrocarbon mixtures.
On the basis of the thermodynamic concept of fugacity, the vapor pressures in Eq. 4.8 can
be corrected and extrapolated to cover a wider range of conditions. A first approach
originated from DODGE and NEWTON (1937) who have shown that at any given p and T
(4.9)
and
(4.10)
respectively, where
f
i(liq)
: fugacity of component i in a liquid mixture
f
i(vap)
: fugacity of component i in a vapor mixture
: fugacity of pure component i as liquid
: fugacity of pure component i as vapor.
If the phases are in equilibrium at given conditions and thus
, (4.11)
the definition of the K-factor is given by
. (4.12)
A generalized fugacity chart for natural gases based on reduced temperatures and
pressures is given in Figure 4.1 in terms of f/p, the so called “activity coefficient”. These
fugacities serve to provide ideal equilibrium constants as defined by Eq. 4.12.
K
i
id ( )
y
i
x
i
----
p
o
i
p
------
= =
f
i liq ( )
x
i
f
o
i liq ( )
=
f
i vap ( )
y
i
f
o
i vap ( )
=
f
o
i liq ( )
f
o
i vap ( )
f
i vap ( )
f
i l i q ( )
=
K
i id ( )

y
i
x
i
----
f
°
i liq ( )
f
°
i vap ( )
------------------
= =
52 Calculation of Phase Equilibria
Figure 4.2 shows that the -values are not sufficiently accurate to describe phase
equilibria at high pressures. Anyway, the extrapolation of the fugacities on the basis of
Eq. 4.10 - though also limited by the critical state conditions - covers a wider range of state
conditions than the vapor pressure relations.
Figure 4.1: Fugacity of natural gases (from BROWN, 1945)
Figure 4.2: Ideal and real K-factors of n-butane at 60[
o
C]
In case of multi-component hydrocarbon systems, attraction forces between the molecules
K
i id ( )

1 0.1
0.01
0.1
1.0
2 2 2 3 3 3 4 4 5 5 10
2
2
2
3
3
3
4
4
4
5
5
Pressure [MPa]
l
o
g

K
Temperature=60 °C
K
K'
id
n-Butane
K
id
Calculation of Phase Equilibria 53
affect the vapor-liquid equilibria at higher p and T. As both RAOULT’s and DALTON’s laws
and the fugacity relations given above (see Eq. 4.9 and Eq. 4.10) result in the calculation
of K-factors without considering the composition, it is necessary to find another solution
for the determination of the K-factors.
K-factors can be evaluated in satisfactory accordance with the measured data by the
application of the “successive solution method”. This method - it will be discussed
through Example 5.5 - consists in an iterative process. It is based on the definition of the
“fugacity coefficient” of the component i (i = 1, 2,...k) in a multi-component phase P - in
analogy of the “activity coefficient”, f/p - as
(4.13)
where p is used for the total pressure of the system.
After introducing not only the PENG-ROBINSON equation of state
in
but also other terms which have been defined by PENG-ROBINSON (see Chap. 3), NGHIEM,
AZIZ and LI (1983) have calculated fugacity coefficients by use of
(4.14)
where a
ij
is defined by
. (4.15)
If the phases are in equilibrium at given conditions and thus
(see Eq. 4.11), the definition of the K-factor is given by
(4.16)
φ
i ,P
f
i ,P
x
i P ,
p
---------------
=
p
RT
V b –
------------
a
V V b + ( ) b V b – ( ) +
-------------------------------------------------
– =
f
p
--- Z 1 Z ln
1
RT
-------
RT
V
-------
p –
¸ ,
¸ _
V d

V

+ – – = ln
φ
i
ln
b
i
b
---- Z 1 – ( ) Z B – ( )
A
2 2B
--------------

2
k
i 1 =

x
i
a
ij
a
--------------------------------------
b
i
b
----

Z 2.414 B +
Z 0.414 B –
-----------------------------
¸ ,
¸ _
ln ln – =
a
i j
1 K
ij
– ( ) a
i
0.5
a
j
0.5
( ) =
f
i vap ( )
f
i l i q ( )
---------------
1 =
K
i
φ
i l i q ( )
φ
i vap ( )
----------------
=
54 Calculation of Phase Equilibria
(see Eq. 4.13),
where
: fugacity coefficient of component i in the liquid phase
: fugacity coefficient of component i in the vapor phase.
4.2 Composition of Phases in Equilibrium
4.2.1 Definitions
If the system at a given pressure, temperature and composition is in a liquid-vapor phase
equilibrium, the following must be valid (on the basis of an overall and particular material
balance, respectively):
(4.17)
(4.18)
where
n: total mole number of the system
: mole number of the liquid phase
: mole number of the vapor phase
x
i :
mole fraction of component i in the liquid phase
y
i
: mole fraction of component i in the vapor phase
z
i
: mole fraction of component i in the system
After substitution of Eq. 4.8 into Eq. 4.18 and after rearrangement, the following is valid:
, (4.19)
(4.20)
If the numerical value for n
liq
and has n
vap
have been chosen correctly, the calculations of
x
i
and y
i
(i = 1,...k) (Eq. 4.19 and Eq. 4.20) result in
φ
i liq ( )
φ
i vap ( )
n n
l i q
n
vap
+ =
z
i
n x
i
n
liq
y
i
n
vap
; i 1, …, k = ( ) + =
n
liq
n
vap
x
i
z
i
n
n
liq
n
vap
K
i
+
--------------------------------
=
y
i
z
i
n
n
liq
K
i
⁄ n
vap
+
------------------------------------
=
Calculation of Phase Equilibria 55
(4.21)
and
. (4.22)
Example 5.1 will demonstrate that it is possible to use the Eq. 4.21 and Eq. 4.22 for
iterative calculation purposes which result in the evaluation of the phase compositions at
given p, T and system composition.
If n = 1, Eq. 4.17 gives
and Eq. 4.21 becomes
. (4.23)
Eq. 4.23 elucidates that
n
liq
must be chosen larger in the following iteration if ,
n
liq
must be chosen smaller in the reverse case of .
Example 4.1
Determination of the gas and liquid phase compositions at p = 14
MPa and T = 92
o
C through an iterative calculation procedure. The
actual composition of the hydrocarbon system is tabled below.
FIRST ITERATION:n
liq
= 0.790
n
vap
= 0.210
x
i
z
i
n
n
l i q
n
vap
K
i
+
--------------------------------
1 =
i 1 =
k

=
i 1 =
k

y
i
z
i
n
n
liq
K
i
⁄ n
vap
+
------------------------------------
1 =
i 1 =
k

=
i 1 =
k

n
vap
1 n
liq
– =
x
i
i 1 =
k

z
i
n
l i q
K
i
1 n
l i q
– ( ) +
--------------------------------------------
i 1 =
k

1 = =
x
i
1 >

x
i
1 <

56 Calculation of Phase Equilibria
n
liq
must be chosen smaller in the following iteration because of
.
SECOND ITERATION: n
liq
= 0.776
n
vap
= 0.224
THIRD ITERATION: n
liq
= 0.778
n
vap
= 0.222
Component System
Composition
z
i
K-Factor
K
i
n
vap
K
i
nl
iq
+n
vap
K
i
C
1
0.4404 2.85 0.598 1.388 0.3173
C
2
0.0432 1.17 0.246 1.036 0.0417
C
3
0.0405 0.68 0.143 0.933 0.0434
C
4
0.0284 0.41 0.086 0.876 0.0324
C
5
0.0174 0.24 0.051 0.841 0.0207
C
6
0.0290 0.15 0.031 0.821 0.0353
C
7+
0.4011 0.01 0.003 0.793 0.5055
Σ 1.0000 0.9963
x
i
z
i
n
liq
n
vap
K
i
+
--------------------------------
=
x
i 1 <

Calculation of Phase Equilibria 57
At the bubble point of the system, the amount of vapor is infinitesimal small and thus
.
Eq. 4.22 then becomes
. (4.24)
Example 4.2 illustrates the iterative process of calculating the bubble point pressure, p
b
,
for a given hydrocarbon composition and for a fixed temperature. Naturally, K
i
= K
i(p)
must be known.
In the vicinity of the bubble point, K
i
x
i
are in a linear relation to pressure. Therefore, linear
interpolation can be applied for acceleration purposes.
Example 4.2
Determination of the bubble point pressure of a crude oil. The
composition of the system is given in Example 4.1.
Iteration II Iteration III Iteration III
Component
C1 0.3114 0.3123 0.8896
C2 0.0416 0.0416 0.0487
C3 0.0436 0.0436 0.0296
C4 0.0327 0.0327 0.0134
C5 0.0210 0.0209 0.0051
C6 0.0358 0.0357 0.0054
C7+ 0.5145 0.5132 0.0082
Σ 1.0006 0.9999 1.0000
x
i
z
i
n
liq
n
vap
K
i
+
--------------------------------
= x
i
z
i
n
liq
n
vap
K
i
+
--------------------------------
= y
i
z
i
n
liq
K
i
+n
vap

---------------------------------
=
n
vap
0, n
l i q
n and x
i
z
i
= = =
z
i
n
n
liq
K
i
n
vap
+ ⁄
--------------------------------------- K
i
z
i
K
i
x
i
i 1 =
k

1. = =
i 1 =
k

=
i 1 =
k

58 Calculation of Phase Equilibria
By interpolation: p
b
= 23 MPa
At the dew point of the system, the amount of liquid phase is infinitesimal small and thus
.
From Equation 4.21:
. (4.25)
In analogy of the iterative process demonstrated by Example 4.2 (evaluation of p
b
), it is
theoretically possible to evaluate the dew point pressure by use of Eq. 4.25 by summing
up the values and by variating the pressure as long as .
However, the evaluation of the K-factor for heavier components is crucial. Therefore, the
calculation of the dew point pressure of hydrocarbon systems by using Eq. 4.25 may be
too inaccurate. This notion is mostly used to check the K-factors which have been
evaluated for certain state conditions.
Component System
Compos.
z
i
1. Assumption
p
b
= 21 MPa
K
i
y
i
= K
i
z
i
2. Assumption
p
b
= 22.5 MPa
K
i
y
i
= K
i
z
i
3. Assumption
p
b
= 23.2 MPa
K
i
y
i
= K
i
z
i
C
1
0.4404 2.1500 0.9469 2.060 0.9072 2.020 0.8896
C
2
0.0432 1.0300 0.0445 1.025 0.0443 1.020 0.0441
C
3
0.0405 0.6720 0.0272 0.678 0.0274 0.680 0.0275
C
4
0.0284 0.4400 0.0125 0.448 0.0127 0.450 0.0128
C
5
0.0174 0.3000 0.0052 0.316 0.0055 0.323 0.0056
C
6
0.0290 0.2150 0.0062 0.230 0.0067 0.239 0.0069
C
7+
0.4701 0.0235 0.0094 0.250 0.0100 0.026 0.0104
Σ 1.0000 1.0519 1.0138 0.9969
n
liq
0 n
vap
n and y
i
z
i
= = , =
z
i
n
n
liq
n
vap
K
i
+
--------------------------------
z
i
K
i
-----
y
i
K
i
----- 1 =
i 1 =
k

=
i 1 =
k

=
i 1 =
k

y
i
K
i
⁄ Σy
i
K
i
⁄ 1 =
Calculation of Phase Equilibria 59

Figure 4.3: Flash and differential vaporization
In evaluating hydrocarbon systems, two characteristic processes are of importance. These
are illustrated schematically in Figure 4.3:
• Mass and composition of the system remain constant (closed system). The portions of
the phases and their compositions in various states of thermodynamic equilibrium (p
and V are variables) are to be evaluated. This process is specified as the “flash
vaporization (condensation)”.
• Preceding every further change of state, mass and composition of the system are
changed in removing one phase - in most cases the vapor phase - as soon as the
previous phase equilibrium is achieved (open system). Such a process is called
“differential vaporization (condensation)”.
In case of a flash process, it is of importance whether the system is signified as liquid at
bubble point pressure or as a vapor at dew point pressure. In both cases, the volume
increases during an isothermal pressure decrease.
If the temperature of a dry gas system ranges between T
c
and the cricondentherm, a
condensate will drop out below the dew point. The amount of the condensate will increase
up to a certain value as a function of pressure decrease (retrograde condensation). Below
this point, further pressure decrease will now effect the vaporization of the condensate.
In case of an undersaturated liquid system, the vapor phase appears below bubble point
pressure. At the same time, the liquid volume diminishes. Correspondingly, the oil phase
shrinks at decreasing pressure.
V
Hg Hg Hg Hg
L
Hg Hg
Hg Hg
V
L
Hg Hg
V
flash vaporization
at the dew point
flash vaporization at
the bubble point
differential vaporization
at the dew point
differential vaporization
at the bubble point
n-times n-times
60 Calculation of Phase Equilibria
The equilibria states appearing at a flash vaporization in series after one another can be
calculated independent of one another by Eq. 4.19 to Eq. 4.21 in regard of their phase
composition. However, the K-factors must always correspond to the given pressure and
temperature.
In case of a differential vaporization, some material balance has to be applied. The volume
remains constant meanwhile every change in state is connected with a change of the mole
number in the system and its composition (open system).
This would correspond to the following reflections: In case of a saturated liquid, the
amount of component i will decrease by y
i
dn
v
resulting from the liberation and production
of an infinitesimal small amount of gas dn
v
, and
. (4.26)
Since y
i
= K
i
x
i
and dn
vap
= dn
liq
, the following is obtained after neglection of the term
dn
vap
dx
i
:
(4.27)
Eq. 4.27 cannot be integrated, because K
i
is a function of p. Therefore, the method of finite
differences is often applied. The integral is then substituted by summation formulae.
Results have been often proved to correspond with the experimental data in a most
accurate way (see Example 4.3).
Example 4.3
Determination of tank oil composition after 3-stage separation on
the basis of the system composition in Example 5.1.
FIRST SEPARATOR: p = 3.33 MPa
T = 48.7
o
C
n
liq
= 0.582
n
vap
= 0.418
yj
i
dn
vap
n
l i q
x
i
n
l i q
dn
vap
– ( ) x
i
dx
i
– ( ) – =
K
i
x
i
dn
l i q
x
i
dn
l i q
n
l i q
dx
i
+ =
Calculation of Phase Equilibria 61
*K
7+
= 0.15 K
C7
SECOND SEPARATOR: p = 0.44 MPa
T = 48.7
o
C
n
liq
= 0.844
n
vap
= 0.156
*K
7+
= 0.15 K
C7
Component Well Stream
Composition
z
i
K-Factor
K
i
Fluids
Liquid Vapor
x
i
y
i
C
1
0.4404 8.100 0.1110 0.8991
C
2
0.0432 1.650 0.0339 0.0560
C
3
0.0405 0.590 0.0489 0.0288
C
4
0.0284 0.230 0.0419 0.0096
C
5
0.0174 0.088 0.0281 0.0025
C
6
0.0290 0.039 0.0485 0.0019
C
7+
0.4011 0.003* 0.6877 0.0021
Σ · 1.0000 1.0000 1.0000
Component Feed
Composition
z
i
K-Factor
K
i
Fluids
Liquid Vapor
x
i
y
i
C
1
0.1110 60.000 0.0109 0.6530
C
2
0.0339 10.600 0.0135 0.1430
C
3
0.0489 3.400 0.0355 0.1207
C
4
0.0419 1.250 0.0403 0.0504
C
5
0.0281 0.415 0.0309 0.0128
C
6
0.0485 0.170 0.0557 0.0095
C
7+
0.6877 0.013* 0.8132 0.0106
Σ · 1.0000 10000 10000
62 Calculation of Phase Equilibria
TANK: p = 0.1 MPa
T = 48.7
o
C
n
liq
= 0.9486
n
vap
= 0.0514
*K
7+
= 0.15K
C7
One mole reservoir crude oil produced amounts to n
liq1
x n
liq2
x n
liq3
= 0.582 x 0.844 x 0.9486 = 0.466 mol tank oil
4.2.2 Evaluation of K-Factors Using Convergence
Pressures
As already mentioned, the K
i
-factors in real systems not only depend on temperature and
pressure, but also on composition. KATZ and HACKMUTH have been the first to provide
for appropriate data of gas/oil-systems. Since then the collection of data has been
continuously enlargend.
Component Feed
Composition
z
i
K-Factor
K
i
Fluids
Liquid Vapor
x
i
y
i
C
1
0.109 265.0 0.0007 0.1965
C
2
0.0135 46.6 0.0038 0.1885
C
3
0.0355 14.3 0.0210 0.3022
C
4
0.0403 5.35 0.0330 0.1775
C
5
0.0309 1.72 0.0299 0.0513
C
6
0.0557 0.70 0.0566 0.0396
C
7+
0.8132 0.052* 0.8550 0.0444
Σ · 1.0000 1.0000 1.0000
Calculation of Phase Equilibria 63
Figure 4.4: K-factors for methane-propane at T
c
= 100
o
F (from SAGE, LACEY and
SCHAAFSMA) (1934)
The most simple relations exist in any 2-component system. Figure 4.4 is a plot of the
K-factors for the methane-propane system at the critical temperature of the mixture T
c
=
100
o
F. At any pressure - at which the vapor and the liquid phase are in equilibrium - K >
1 for the highly volatile component methane. If the pressure increases, its K-factor will
decrease.
On the other side K<1 for the less volatile component propane since its concentration in
the vapor phase is less than in the liquid. At any isothermal change in state, K for the less
volatile component propane will equal 1 at two different pressures. One corresponds - just
as in case of the highly volatile component - with the critical pressure of the mixture and
the other one with the vapor pressure of the pure substance. Above the critical pressure of
the system, K-factors have no longer any physical meaning.
Generally, the K-factors increase with increasing T at constant pressure. However,
methane shows some irregular behavior. Its K-factor increases with temperature up to a
certain limit and decreases beyond that.
According to observations, the two K-isotherms of binary system will also converge to a
certain point at other temperatures than T
c
. This point is also characterized by K = 1(y =
x). However, this convergence is only an “apparent” convergence. In contrast to the
critical isotherms, the K-values have no physical meaning up to the point of apparent
convergence because of the existence of only one phase beyond a certain pressure limit.
100
1.0
10.0
1500 200 300 500
0.3
3.0
0.5
5.0
log p [psia]
l
o
g

K
1000
64 Calculation of Phase Equilibria
Anyway, the isotherms may be extrapolated to K = 1 defining the pressure at this point as
the convergence pressure p
k
of the system at the actual temperature.
Figure 4.5: Comparison of K-factors at 100
o
F for 1,000 and 5,000-psia convergence
pressure (from NGAA, 1957)
It has been proved that multi-component systems have properties very similar to those of
the binary systems with respect to the convergence of all K-isotherms. The (apparent)
convergence pressure p
k
is a function of the composition of the mixture. Figure 4.5
emphasizes the dependence of the K-factors on the convergence pressure being a function
of the system composition. The differences in K-values for the two convergence pressures
at pressures below 0.7 MPa (100 psia) are not significant for the lighter hydrocarbons
ethane through pentane. It then becomes apparent that at low pressures and temperatures,
the K-factors are nearly independent of composition.
Note:
• If the evaluation of the K-factors is based on the convergence pressure of the mixture,
the K-factors depend only on p and T and are independent of composition.
• At a given temperature, the convergence pressure of a mixture coincides with the
critical pressure of another mixing ratio of its components.
In 1957, the NGAA (Natural Gasoline Association of America) published a diagram
series lgK = f(lg p)
T
in its “Equilibrium Ratio Data Book”. The diagrams were based on
average values of already available data from various gas condensates and gas/crude
10 100 1000 2 2 2 3 3 3 4 4 4 5 5 5
Pressure [psia]
0.0001
0.001
0.01
0.1
1
10
100
K
C
1
C
1
C
2
C
3
C
4
C
5
C
6
C
7
C
8
C
9
C
10
Calculation of Phase Equilibria 65
oil-systems. This series includes the paraffinic homologous compounds - methane
through decane - just as ethylene, propylene, nitrogen, carbon dioxide and hydrogen
sulfide within the pressure range from 0.07 to 140 MPa and for temperatures ranging from
- 148 to +260 °C.
Figure 4.6 and Figure 4.7 present the respective charts for methane and hexane,
respectively, in case of the convergence pressure p
k
= 34.5 MPa (5000 psia). The values
for this convergence pressure can be used for a large number of hydrocarbon systems.
Today, these data are already available for data-processing with computers.
If p
k
is known, it is possible to choose from the measured series those K-factors , which
are best suited to describe the phase equilibrium vapor/liquid of the given system.
The second parameter, which has to be evaluated to proceed with the prediction of the
phase and volumetric behavior of complex hydrocarbon systems, is the K-factor of the
heaviest component. In conventional analyses of hydrocarbon fluids, everything heavier
than hexane is grouped together and reported as the pseudo-component “C
7+
”. This
fraction is a mixture of materials of varying volatility. However, the vapor pressure curves
and the critical properties of the hydrocarbons heavier than hexane are fairly close
together. Thus, it is possible to characterize the mixture “C
7+
” by an average set of
K-values.
66 Calculation of Phase Equilibria
Figure 4.6: K-factors for methane, 5,000 psia convergence pressure (from NGAA,
1957)
10
1
2 5 10
2
2 5 10
3
2 5 10
4
Pressure [PSIA]
10
-2
10
-1
10
0
10
1
10
2
10
3
2
2
2
2
2
5
5
5
5
5
K

=

y
/
x
-
1
0
0
-
1
2
0
-
1
4
0
-
1
6
0
-
1
8
0
-
2
0
0
-
2
2
0
-
2
4
0
-
2
6
0
0
5
0
0
1
0
0
2
0
0
T
e
m
p
e
r
a
t
u
r
e

[
°
F
]
5
0
Calculation of Phase Equilibria 67
Figure 4.7: K-factors for hexane, 5,000 psia convergence pressure (from NGAA,
1957)
10
1
2 5 10
2
2 5 10
3
2 5 10
4
Pressure [PSIA]
10
-4
10
-3
10
-2
10
-1
10
0
10
1
2
2
2
2
2
5
5
5
5
5
K

=

y
/
x
-80
-1
0
0
-
1
0
0
-60
-40
-20
0
20
40
60
1
0
0
8
0
1
2
0
1
4
0
1
6
0
1
8
0
2
0
0
2
2
0
2
4
0
2
6
0
2
8
0
3
0
0
4
0
0
4
5
0
5
0
0
T
e
m
p
e
ra
t
u
re


F
]
68 Calculation of Phase Equilibria
4.2.3 Evaluation of Convergence Pressure
Several iterative methods of estimating p
k
have been proposed. One of the most
convenient methods is based on the assumption of a pseudo-binary system and consists of
the following steps:
• A certain convergence pressure is assumed and the K-factors of components are
evaluated from the NGAA-diagrams (examples are given through Figure 4.6 and
Figure 4.7).
• The respective composition of the liquid phase being in equilibrium are evaluated by
a flash process (see Example 4.1).
• With exception of the lightest component methane, all other components are
summarized in a second fictitious component. The pseudo-critical temperature of the
pseudo-component C
2+
in the liquid phase is evaluated in calculating the averaged
mole weights in the liquid phase, W
i
, through
(4.28)
and the pseudo-critical temperature of C
2+
through
. (4.29)
• The critical point of the component C
2+
is plotted on Figure 4.8 which shows the
vapor pressure curves of homologous hydrocarbons.
• The “binary” critical envelope curve of methane with this pseudo-component is
drawn.
• For the given temperature, the convergence pressure p
k
can be read off because it
coincides with the critical pressure of one of the possible mixing ratios in the “binary”
system.
• If this numerical value of the convergence pressure does not coincide with the value
initially assumed, the whole procedure has to be repeated starting with the assumption
of another convergence pressure. The iterative process must go on until
correspondence of the assumed convergence pressure and the evaluated one will be
achieved.
This iterative method will be demonstrated by Example 4.4.
The K-factors obtained through this method exhibit only satisfactory results, if the
operating pressure is considerably lower than p
k
. As the operating pressure approaches p
k
,
the K-factors become very sensitive to the convergence pressure used and care must be
W
i
x
i
M
i
=
T
pc
k
i 2 =

W
i
T
ci
k
i 2 =

W
i
---------------------------------------
=
Calculation of Phase Equilibria 69
taken in the selection of the correct value of p
k
. However, charts exist only for some
distinct convergence pressure (800, 1000, 1500, 2000, 3000, 5000 and 10,000 psia). Thus
the interpolation - inevitable in estimating the K-values - may lead to great errors.
4.2.4 Evaluation of by use of PENG-ROBINSON
equation of state
With help of the “successive solution method”, the K-factors of the components of the
system can be evaluated by an iterative process like the one presented through Example
4.5:
• The composition of the system, the pressure and temperature as well as the critical
data of the components, their acentric factors and the interactive coefficients must be
initially known.
• On the basis of assumed K-factors, the phase compositions are calculated through an
iterative calculation procedure (see Example 4.1).
• On the basis of the calculated phase compositions, the EOS-parameters a, b, A, and B
after PENG-ROBINSON are calculated for both phases.
• The Z-factors both for the liquid and the vapor phase can be evaluated solving the
equation of state (Eq. 3.49) through use of the method of CARDAN (see Example 3.3).
• The fugacity coefficients for both phases are evaluated by use of Eq. 4.14 and Eq.
4.15:
and
.
• New K-values can be evaluated by inserting these values in Eq. 4.16:
• On the basis of the evaluated K-values, the calculation procedure is repeated. The
steps described above will be repeated until the K-factors converge within a certain
limit.
• The last flash iteration results in the evaluation of the mole number in the liquid
φ
i
ln
b
i
b
---- Z 1 – ( ) Z B – ( ) ln –
A
2 2B
--------------
2
k
i 1 =
x
i
a
ij

a
------------------------------
b
i
b
----

Z 2 414B ⋅ +
Z 0.414B –
------------------------------
¸ ,
¸ _
ln – =
a
i j
1 K
i j
– ( ) a
i
0.5
a
j
0.5
( ) =
K
i
φ
i l i q ,
φ
i vap ,
---------------
=
70 Calculation of Phase Equilibria
phase, n
liq
, the mole number in the vapor phase n
vap
and in the evaluation of the
composition of the phases.
• Using Z
liq
and Z
vap
, the volume and the density of the two phases can be calculated,
too.
Figure 4.8: Convergence pressure data - methane for binary hydrocarbon mixtures
(from WINN, 1952)
200 300 400 1000 2000 4000 6000 600
Convergence Pressure [psia]
T
e
m
p
e
r
a
t
u
r
e

[
°
F
]
-100
0
100
200
300
400
500
600
700
800
900
C
3
C
4
C
7
C -C
7 23
C-C
5 24
C
16
C
10
C
8
iC
5
C
5
C
6
iC
4
C
4
C*
C
3
C*
C
2
C*
C
2
C
1
Calculation of Phase Equilibria 71
Example 4.4
Evaluation of the convergence pressure by the iterative NGAA
process described above. The composition of the system is tabled
below, and the temperature and the pressure are fixed as T = 49
o
C
and p = 15.5 MPa.
FIRST ITERATION: p
k
= 26.7 MPa (4000 psia)
n
liq
= 0.5876
n
vap
= 0.4124
*assumption:
On Figure 4.8 T
pc
= 514.06
o
F is close to T
c
of pure n-heptane. For
the given temperature of the system (T = 49
o
C and 120
o
F,
respectively), the convergence pressure p
k
can be read off on the
envelope curve between methane and n-heptane at
p
k
= 3150 psia = 21.72 MPa.
This value does not coincide with the initially assumed value. Thus
a second iteration must be done.
SECOND ITERATION: p
k
= 20.4 MPa (3000 psia)
Component z
i
M
i
kg/kmole
T
c,i
K
K
i
x
i
W
i
=x
i
M
i
W
i
T
ci
C
1
0.6973 - - 1.6500 0.5499 - -
C
2
0.0318 30.07 305.4 0.9000 0.0332 0.998 304.79
C
3
0.0194 44.09 369.8 0.6300 0.0229 1.010 373.50
nC
4
0.0223 58.12 425.2 0.4300 0.0292 1.697 721.56
nC
5
0.0221 72.15 469.7 0.2250 0.0325 2.345 1101.45
nC
6*
0.0309 86.17 507.4 0.1700 0.0470 4.050 2054.97
C
7*
0.1762 114.22 568.8 0.0720 0.2854 32.598 18541.74
Σ 1.0000 42.698 23098.01
C
7+
nC
8

T
pc
k
i 2 =

W
i
T
c i ,
k
i 2 =

W
i
----------------------------------
23098.01
42.698
----------------------
540.96 K 514.06 F
o
= = = =
72 Calculation of Phase Equilibria
n
liq
= 0.5604
n
vap
= 0.4496
*assumption:
From Fig. 4.8:
and thus
.
The agreement with the assumption is now satisfactory. Further
refinement is impossible (K-values only evaluated for 1000 psia
steps in p
k
).
Example 4.5
Evaluation of the K-factors of the three components of a system at
p = 1.38 MPa and T = 338.7 K. The successive solution method is
used on the basis of the composition of the system, the critical
and reduced data as well as the acentric factors which are given
below in Table I. Table II presents the interaction coefficients
K
ij
which must be known, too.
Component K
i
x
i
W
i
= x
i
M
i
W
i
Tc
i
C
1
1.700 0.5332 - -
C
2
0.8800 0.0336 1.010 308.45
C
3
0.6200 0.0233 1.027 379.78
nC
4
0.4200 0.0299 1.738 739.00
nC5 0.2200 0.0336 2.424 1138.55
nC
6*
0.1800 0.0483 4.162 2111.80
C
7*
0.0700 0.2980 34.038 19360.81
Σ 1.0000 44.399 24038.39
C
7+
nC
8

k
i=2
W
i
T
c i ,

k
i=2
W
i

------------------------------------
24038.39
44.399
----------------------
541.42K 514.9 F
o
= = =
pk 3150psia =
p
k
21.72MPa =
Calculation of Phase Equilibria 73
Table 4.1: Composition of the System, Critical Data and Acentric Factors
For the first iteration of a flash process,K-factors as well as
the mole numbers in the phases, n
liq
and n
vap
, are assumed. The gas
and the liquid phase compositions are calculated through an
iterative calculation procedure (see Example 4.1). The assumed
K-factors as well as the evaluated phase compositions are given in
Table 4.3.
Then the flash iteration starts with the evaluation of a and b. At
first the parameters b and a for each component have to be
evaluated through
Component Composition
z
i
Critical Data
T
c
K p
c
MPa
Reduced Data
T
r
p
r
Acentric Factor
ω
i
C
3
0.61 369.8 4.245 0.916 0.325 0.152
n - C
4
0.28 425.2 3.800 0.797 0.363 0.193
n - C
5
0.11 469.6 3.374 0.721 0.409 0.251
Table 4.2: Interaction Coefficients of the Components
C
3
n - C
4
n - C
5
C
3
0.000 0.001 0.003
n - C
4
0.001 0.000 0.001
n - C
5
0.003 0.001 0.000
Table 4.3:
FIRST ITERATION: n
liq
= 0.52
n
vap
= 0.48
Component K-Factor
C
3
1.745 0.440 0.767
n - C
4
0.537 0.369 0.198
n - C
5
0.182 0.191 0.035
Σ 1.000 1.000
x
i
z
i
n
v
K
i
n
L
+
-----------------------
= y
i
z
i
n
L
K
i
⁄ n
v
+
---------------------------
=
74 Calculation of Phase Equilibria
,
(see Eq. 3.46),
,
(see Eq. 3.48),
,
(see Eq. 3.52) and through
.
(see Eq. 3.47).
The results of these calculations are tabled below (see Table 4.4).
By use of Eq. 3.53
and Eq. 3.56
,
a and b for the liquid phase as well as for the vapor phase are
calculated on the basis of phase compositions evaluated through
the first evaluation (see Table 4.3.) The calculation procedure is
presented through Example 3.2.
Then the parameters A and B are calculated after PENG-ROBINSON
using Eq. 3.35
Table 4.4: First Iteration
Component b
i
a
i
C
3
5.635 E-5 1.072
n - C
4
7.247 E-5 1.729
n - C
5
9.003 E-5 2.557
b
i
0.0778
RT
c
p
c
--------- =
a
c i ,
0.45724
R
2
T
2
p
c
------------ =
α
i
0.5
1 0.37646 1.54226ω 1.54226ω
2
– + ( ) 1 T
r
0.5
– ( ) + =
a
i
a
c
α ( )
i
=
b x
i
b
i
i 1 =
k

=
a x
i
x
j
a
i
a
j
( )
0.5
1 K
i j
– ( )
j 1 =
k

i 1 =
k

=
Calculation of Phase Equilibria 75
and Eq. 3.36
.
The evaluated numerical values of the parameters a, b, A and B for
both phases are summarized in Table 4.5.
Now the Z-factors both for the liquid and the vapor phase can be
evaluated solving the equation of state given by Eq. 3.49:
.
To solve this cubic equation of state, the CARDAN equation is
applied for
where
,
and
.
The factors r, s and t have to be evaluated both for the liquid
and the vapor phase.
By substitution of
,
the formula
is reduced to
Table 4.5: First Iteration
a b A B
Liquid Phase 1.551 6.873 E-5 0.2699 3.368 E-2
Vapor Phase 1.232 6.072 E-5 0.2144 2.975 E-2
A
ap
R
2
T
2
------------
=
B
b p
R T
---------
=
Z
3
1 B – ( ) Z
2
– A 2B – 3B
2
– ( )Z AB B
2
– B
3
– ( ) 0 = – +
x
3
rx
2
sx t 0 = + + +
r 1 B – ( ) – =
s A 2B 3B
2
– – ( ) =
t AB B
2
B
3
– – ( ) – =
x y
r
3
---
– =
x
3
rx
2
sx t 0 = + + +
76 Calculation of Phase Equilibria
where
and
.
The discriminant D is defined as
Contrary to Example 3.3, D < 0 both in case of the liquid phase and in case of the vapor
phase. Thus the cubic equation of state has three real solutions, respectively. In defining
and
,
the three solutions for the reduced form
are given by
and
.
Introducing y
1
, y
2
, and y
3
into
,
the Z-factors(=x) of the phases can be evaluated. The solutions
are summarized in Table 4.6.
y
3
py q 0 = + +
p
3s r
2

3
----------------
=
q
2r
3
27
--------
rs
3
-----
– t + =
D
p
3
---
¸ ,
¸ _
3
q
2
---
¸ ,
¸ _
2
+ =
ς
p
3
27
------

2
=
ϕ cos
q

-----
– =
y
3
py q 0 = + +
y
1
2 ς
3
ϕ 3 ⁄ ( ) cos =
y
2
2 ς
3
ϕ 3 2π 3 ⁄ + ⁄ ( ) cos =
y
3
2 ς
3
ϕ 3 4π 3 ⁄ + ⁄ ( ) cos =
x y
r
3
---
– =
Calculation of Phase Equilibria 77
Table 4.6: First Iteration Solution f or the Z-Factors after CARDAN
Obviously, the actual Z-factor of the liquid phase must be
identical with the smaller one of the three solutions for the
liquid phase, meanwhile the actual Z-factor for the vapor phase is
equal to the greatest one of the three solutions for the vapor
phase. Thus the first iteration results in
and
.
The Z-factors are introduced into Equation 4.15
,
respectively, to evaluate the fugacity coefficients (of
propane), , of n-butane) and (of n-pentane) both for the
liquid and the vapor phase. The results of this calculation
procedure are summarized in Table 4.7.
As a result of the first iteration flash, K-factors are evaluated
through Eq. 4.16
and summarized in Table 4.8
Z
1
Z
2
Z
3
Liquid Phase 0.0523 0.2172 0.6968
Vapor Phase 0.0526 0.1325 0.7852
Table 4.7: First Iteration Fugacity Coefficients
φ
3
φ
4
φ
5
Liquid Phase 1.2580 0.4707 0.1831
Vapor Phase 0.8473 0.7519 0.6672
Z
l i q
0.0523 =
Z
vap
0.7852 =
φ
i
ln
b
i
b
---- Z 1 – ( ) Z B – ( ) ln –
A
2 2B
--------------
2 x
i
a
ij
i 1 =
k

a
----------------------
b
i
b
----

Z 2 414B ⋅ +
Z 0.414B –
------------------------------
¸ ,
¸ _
ln – =
φ
3
φ
4
φ
5
K
i
φ
i liq ,
φ
i vap ,
---------------
=
78 Calculation of Phase Equilibria
Table 4.8: First Iteration
These K-factors are the basis of the second iteration which starts
again with the iterative calculation procedure presented through
Example 4.1 (see Table 4.9)
On the basis of the data in Table 4.4 and in Table 4.9 the numerical
values of the parameters a, b, A and B for both phases are
evaluated again and summarized in Table 4.10.
Now the Z-factors both for the liquid and the vapor phase can be
evaluated solving the equation of state given by Eq. 3.49
through the CARDAN equation. Again D < 0 both in case of the liquid
phase and in case of the vapor phase. Correspondingly, the cubic
equation of state has three real solutions. The solutions are
summarized in Table 4.11.
Table 4.9:
SECOND ITERATION: n
liq
= 0.556
n
vap
= 0.444
Components K-Factor
C
3
1.4847
n - C
4
0.6260
n - C
5
0.2744
Component K-Factor
C
3
1.4847 0.5020 0.7453
n - C
4
0.6260 0.3358 0.2102
n - C
5
0.2742 0.1623 0.0445
Σ 1.0001 1.0000
Z
3
1 B – ( )Z
2
– A 2B 3B
2
– – ( )Z AB B
2
B
3
– – ( ) 0 = – +
x
i
z
i
n
v
K
i
n
L
+
-----------------------
= y
i
z
i
n
L
K
i
n
v
+ ⁄
---------------------------
=
Calculation of Phase Equilibria 79
Table 4.10: Second Iteration
Once more again, the actual Z-factor of the liquid phase must be
identical with the smallest one of the three solutions for the
liquid phase, meanwhile the actual Z-factor for the vapor phase is
equal to the greatest one of the three solutions for the vapor
phase. Thus the second iteration results in
Z
liq
= 0.0521
and
Z
vap
= 0.7812.
The Z-factors are introduced into Eq. 4.15
,??????
respectively, to evaluate the fugacity coefficients (of
n-butane)and (of n-pentane) both for the liquid and the vapor
phase. The results of this calculation procedure are summarized in
Table 4.12.
Table 4.12: Second Iteration
Fugacity Coefficients
a b A B
Liquid Phase 1.490 6.725 E-5 0.2592 3.296 E-2
Vapor Phase 1.249 6.116 E-5 0.2173 2.997 E-2
Table 4.11: Second Iteration Solution for the Z-Factors, after CARDAN
Z
1
Z
2
Z
3
Liquid Phase 0.0521 0.1988 0.7161
Vapor Phase 0.0525 0.1364 0.7812
φ
3
φ
4
φ
5
Liquid Phase 1.2520 0.4717 0.1846
Vapor Phase 0.8477 0.7511 0.6655
φ
i
ln
b
i
b
---- Z 1 – ( ) Z B – ( ) ln –
A
2 2B
--------------
2
k
i 1 =
x
i
a
i j

a
-----------------------------------------
b
i
b
----

Z 2 414B ⋅ +
Z 0.414B –
------------------------------
¸ ,
¸ _
ln – =
φ
4
φ
5
80 Calculation of Phase Equilibria
The K-factors are evaluated through Eq. 4.16
and given in Table 4.13
Table 4.13: Second Iteration
K-Factors
The K-factors from the two flash iterations exhibit satisfactory
convergence. On the basis of the K-factors given in Table 4.13, a
final iterative calculation procedure analogous to Example 4.1
results in the mole numbers and in the composition of the phases.
Table 4.14:
Components K-Factor
C
3
1.4769
n - C
4
0.6280
n - C
5
0.2773
FINAL RESULTS: n
liq
= 0.562
n
vap
= 0.438
Component K-Factor
C
3
1.4769 0.5046 0.7452
n - C
4
0.6280 0.3345 0.2101
n - C
5
0.2773 0.1609 0.0446
Σ 1.0000 0.9999
K
i
φ
i liq ,
φ
i vap ,
---------------
=
x
i
z
i
n
v
K
i
n
L
+
-----------------------
= y
i
z
i
n
L
K
i
n
v
+ ⁄
---------------------------
=
81
Chapter 5
Phase Properties
5.1 Natural Gases
5.1.1 Volume
As previously discussed in Chap. 3, the state of ideal gases is defined by the following
equation of state:
(5.1)
where R is the universal gas constant and n is the number of moles. The value R depends
on the system of units. In the SI-system:
R = 8.31432 J / K mole.
In case of ideal gases, no intermolecular forces are acting. At any isothermal change of
state, the decrease in volume is directly proportional to the increase in pressure. It has
been observed that in many cases - e.g. in case of natural gases - the volume becomes less
than half if the pressure is doubled. This behavior of a gas in called a real one.
Below the BOYLE temperature, the real gases are more compressible than an ideal one,
because of intermolecular forces of attraction and repulsion. If the pressure is low (the
forces of attraction can be neglected because of the great distance of molecules) or in case
of high temperatures (dominance of the kinetic energy of the molecules), real gases
behave similarly to perfect gases.
In case of real gases, the change in state can be well described by the equations of state
which are also presented in Chap. 3 (VAN DER WAALS, REDLICH-KWONG,
PENG-ROBINSON). In the petroleum industry, the basic relation of Eq. 3.12
pV = ZnRT
pV nRT =
82 Phase Properties
is commonly used because of its easy applicability.
The Z-factor is the so called “deviation factor” or “compressibility factor”. It is defined as
the ratio of the volume of a real gas to the volume of a perfect one at identical temperature
and pressure.
By introducing
where
m: mass
M: molecular weight,
pV = ZnRT can be transformed into
(5.2)
if m = 1[g] gas
The reciprocal of the specific volume V is the density and so
. (5.3)
The atmospheric density of the gas is often related to the atmospheric density of air at the
same standard temperature. In doing so, the “specific gravity” of the gas is defined by
(5.4)
where the subscript “a” indicates the values for air and the superscript “o” those at
standard conditions (atmospheric pressure).
If Eq. 5.4 is combined with Eq. 5.2, the gas density results in
(5.5)
As expected, the Z-factor depends both on pressure and temperature. Z-factors of different
gases, which can be determined experimentally, vary considerably. Figure 5.1 shows the
Z-factor of methane, ethane and propane as a function of the pressure.
n
m
M
-----
=
pV
ZRT
M
-----------
=
ρ
g
1
V
---
pM
ZRT
-----------
= =
γ
g
γ
g
ρ
g
0
ρ
g a ,
0
-----------
M
M
a
-------
M
28.97
-------------
= = =
ρ
g
1
V
---
M
a
R
-------γ
g
p
ZT
------- 0.003473γ
g
p
ZT
------- = = =
Phase Properties 83
Figure 5.1: Z-factor of methane, ethane and propane versus pressure at T = 140
o
F
(from STANDING, 1977)
5.1.2 Formation Volume Factor
In petroleum industry, the term “formation volume factor” is widely used. It expresses
which volume is taken up by one cubic meter (at standard state conditions, p
o
and V
o
) at
a pressure p and a temperature T.
The formations volume factor for gases, B
g
, is defined by
(5.6)
where the standard conditions p
o
and T
o
are generally defined in
Europe: p
o
: 1.00 bar
T
o
: 273.15 K
and in
USA: p
o
: 14.7 psia (1.01325 bar)
T
o
: 60.0
o
F (15
o
C).
In case of one mole, Eq. 5.3 results for standard conditions in
(5.7)
If Eq. 5.3 and Eq. 5.7 are substituted in Eq. 5.6, then:
Pressure [MPa]
0 4 8 12 16 20
0
0.2
0.4
0.6
0.8
1.0
z
p
V
n
R
T
Methane
P
r
o
p
a
n
e
E
t
h
a
n
e
B
g
B
g
p T , ( )
V
V °
------
= =
p°V ° Z
°
RT
°
=
84 Phase Properties
(5.8)
where
.
5.1.3 Compressibility
On the basis of the gas law, the isothermal gas compressibility results in
. (5.9)
For ideal gases: Z = 1, i.e. .
Therefore, the gas compressibility is only a function of pressure and is inversely
proportional to the pressure.
5.1.4 Correlation of Z-Factor on the Basis of Reduced
State Variables
In general, the components of natural gases are homologous paraffins with similar
molecular structures and the theorem of corresponding states can be applied. Figure 5.2
shows the excellent correspondence of the Z-factors of methane, propane and pentane as
a function of reduced pressure for a series of reduced temperatures.
B
g
p,T ( )
ZT
p
------- C =
C
p
o
Z
o
T
o
--------------
=
c
g
1
V
---
∂V
∂p
------
¸ ,
¸ _
T
1
B
g
------
∂B
g
∂p
---------
¸ ,
¸ _
T
1
p
---
1
Z
---
∂Z
∂p
------
¸ ,
¸ _
– = – = – =
∂Z
∂p
------ 0 =
Phase Properties 85
Figure 5.2: Z-factor as a function of reduced pressure for a series of reduced
temperatures (from SAGE and LACEY, 1949)
The theorem of corresponding states can also be applied to gas mixtures through the
evaluation of “pseudo-critical” data, T
pc
and p
pc
, which are given by
(5.10)
and
(5.11)
where
p
ci
: critical pressure of component i
T
ci
: critical temperature of component i
y
i
: mole fraction of component i in the gas phase.
Then the “pseudo-reduced” data, T
pr
and p
pr
, can be evaluated:
, (5.12)
(5.13)
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
Reduced Pressure, p
r
z
p
V
n
R
T
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
CH
4
C H
3 8
C H
5 12
T=0.9
r
CH
4
C H
3 8 C H
5 12
T=1.0
r CH
4
CH
4
CH
4
CH
4
C H
3 8
T=1.1
r
1
1
3
3
5
5
T=1.2
r
T=1.3
r
T=1.5
r
C H
3 8
C H
5 12
C H
5 12
C H
5 12
C H
3 8
C H
3 8
C H
5 12
T
pc
y
i
T
c,i
i 1 =
k

=
p
pc
y
i
p
ci
i 1 =
k

=
T
pr
T
T
pc
--------
=
p
pr
p
p
pc
--------
=
86 Phase Properties
where
T: operating temperature
p: operating pressure
Note that both T and T
pc
in Eq. 5.12 must be absolute temperatures, either given in K
(Kelvin, SI unit) or in R (Rankin, Field Unit).
The definition of corresponding states on the basis of the pseudo-reduced data results in
a generalized plot of Z-factors for natural gases (see Figure 5.3). This plot can be used to
evaluate the Z-factor of a natural gas on the basis of its composition and the critical data
(see Example 5.1).
Several natural gases contain considerable amounts of heavy components which will be
combined - depending on the feasibility of chemical analysis - to the pseudo components
C
7+
or C
9+
. The compositions of these mixtures cannot exactly be defined. In the
laboratory practice, the C
7+
fraction is obtained as liquid residue from a fractional
destillation and two properties - averaged molecular weight and specific gravity - are
determined. These properties have been correlated with the pseudo-critical data of this
fraction (see Figure 5.4). On the basis of these correlations, the engineer is able to
calculate pseudo-reduced properties of natural gases for which conventional analyses are
available.
In many cases, the data of fractional distillation are not available. Anyway, several
diagrams have been published which correlate the pseudo-critical functions of state with
the gas gravity. By means of these diagrams, the Z-factor can be determined even if the
gas composition is not known. These diagrams are only valid for a given quality or only
for one area, formation, reservoir, basin, etc. The respective diagram for Oklahoma City
gases is given in Figure 5.5 (see also Example 5.2.)
Phase Properties 87
Figure 5.3: Z-factor for natural gases (from BROWN et al., 1948)
7
0.9
1.0
1.1
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
8 9 10 11 12
13 14 15
Pseudo Reduced Pressure
C
o
m
p
r
e
s
s
i
b
i
l
i
t
y

F
a
c
t
o
r


Z
0 1 2 3 4 5 6 7 8
Pseudo Reduced Pressure
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
0.95
1.0
1.1
C
o
m
p
r
e
s
s
i
b
i
l
i
t
y

F
a
c
t
o
r


Z
0.25
3.0
2.8
2.6
2.4
2.2
2.0
1.9
1.8
1.7
1.6
1.5
1.45
1.4
1.35
1.3
1.25
1
.2
1
.
1
5
1
.1
1
.
0
5
Pseudo Reduced
Temperature
88 Phase Properties
Figure 5.4: Pseudo-critical temperatures and pressures for heptanes and heavier
(from MATTHEWS et al, 1942)
Figure 5.5: Pseudo-critical properties of Oklahoma City Gases
(from MATTHEWS et al., 1942)
300
400
500
1.000
1.100
1.200
1.300
1.400
200
120 140 160 180 200 220 100 240
Molecular Weight
P
s
e
u
d
o

C
r
i
t
i
c
a
l

P
r
e
s
s
.

[
p
s
i
a
]
P
s
e
u
d
o

C
r
i
t
i
c
a
l

T
e
m
p
.

[
°
R
]
Sp. gr.
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2
Gas Gravity
P
s
e
u
d
o

C
r
i
t
i
c
a
l

T
e
m
p
e
r
a
t
u
r
e

[
°
R
]
P
s
e
u
d
o

C
r
i
t
i
c
a
l

P
r
e
s
s
u
r
e

[
i
b
/
s
q

i
n
.
]

a
b
s
o
l
u
t
e
700
675
650
625
600
575
550
525
500
475
450
425
400
375
350
325
Phase Properties 89
Example 5.1
Evaluation of the density of a natural gas at T = 105
o
C and at p
= 22.55 MPa. The Z-factor is determined by using Figure 5.3 The
table below summarizes the composition of the gas, the critical
data as well as the calculated reduced data of the components.
assumption:
From Figure 5.3
Z = 0.925
which is introduced in
where
R = 8.31432[J°K
-1
mol
-1
].
Component M
i
kg/kmole
y
i
y
i
M
i
kg/kmole
T
c,i
K
y
i
x T
ci
K
p
ci
MPa
y
i
x p
ci
MPa
C
1
16.0 0.9443 15.109 190.6 179.98 4.60 4.344
C
2
30.1 0.0158 0.476 305.4 4.83 4.88 0.077
C
3
44.1 0.0132 0.582 369.8 4.88 4.25 0.056
iC
4
58.1 0.0051 0.296 408.1 2.08 3.65 0.019
nC
4
58.1 0.0053 0.308 425.2 2.25 3.79 0.020
iC
5
72.2 0.0023 0.166 461.4 1.06 3.38 0.008
nC
5
72.2 0.0021 0.152 469.7 0.99 3.37 0.007
C
6
86.2 0.0033 0.284 507.4 1.67 3.01 0.010
114.23 0.0086 0.982 568.8 4.89 2.48 0.021
Σ 1.000 18.355 202.63 4.562
C
*
7+
C
7+
C
8

T
pr
T
T
pc
--------
105 + 273.15
202.63
-------------------------------
1.87 = = =
p
pr
p
p
pc
--------
22.5
4.562
-------------
4.93 =
¸ ,
¸ _
= =
ρ
g
p
k
i 1 =
y
i
M
i

ZRT
--------------------------------
=
90 Phase Properties
Example 5.2
Evaluation of the density and the formation volume factor of an
Oklahoma City Gas of the specific gravity at
temperature T = 90.5
o
C(655

R)and pressure p = 12.5 MPa(1813 psia).
At first the pseudo critical data are read off from Figure 5.5 as
and
.
Now the pseudo reduced data are calculated as
and
.
On the basis of the pseudo-reduced data,
results from Figure 5.3.
Because
,
where
M
a
= 28.97 E-3 kg/mole,
and
R = 8.31432J/K mole,
the density can be calculated through
so that
ρ
g
225.5 E 5 18.355 E - 3 × +
0.925 8.31432 378.15 × ×
----------------------------------------------------------------- kg m
3
⁄ =
ρ
g
142.32kg m
3
⁄ =
γ
g
0.7018 =
T
pc
390R =
p
pc
670psia =
T
pr
T
T
pc
--------
655
390
--------- 1.68 = = =
p
pr
p
p
pc
--------
1813
670
------------ 2.71 = = =
Z 0.86 =
ρ
g
M
a
R
------- γ
g
p
ZT
------- =
ρ
g
0.02897 0.7018 125 × × E 5 +
8.31432 0.86 363.65 × ×
-------------------------------------------------------------------------- kg m
3
⁄ =
Phase Properties 91
= 97.774 kg/m
3
.
The formation volume factor can be evaluated through Eq. 5.8
where
.
Assuming that the gas is a perfect one at European standard
conditions
and
so that Z = 1.0, its formation volume factor results in
,
.
ρ
g
B
g
p,T ( )
ZT
p
------- C =
C
p
o
Z
o
T
o
------------
=
p
o
1bar =
T
o
273.15 K =
B
g
0.86 (90.5 + 273.15) ×
12.5
------------------------------------------------------
0.1
273.15
---------------- =
B
g
0.00916 =
92 Phase Properties
Figure 5.6: Water content of natural gas in equilibrium with liquid water
(from KATZ et al., 1959)
-70 -50 -30 -10 10 30 50 70 90 110 130 150 170 190 210 230 250 270 290 320 340 360 380400 440 480 520 560 600 640 680 720
10
0
10
1
2
5
10
2
2
5
10
3
2
5
10
4
2
5
2
5
W
a
t
e
r
c
o
n
t
e
n
t
I
b
H
O
p
e
r
m
i
l
l
i
o
n
c
u
f
t
t
o
t
a
l
g
a
s
2
Temperature, °F
Saturation 100% HO, 47,489 Ib per mm cu ft
2
1
4
.
7
p
s
i
a
3
0
5
0
1
0
0
2
0
0
4
0
0
5
0
0
6
0
0
8
0
0
1
0
0
0
1
2
5
0
1
5
0
0
2
0
0
0
2
5
0
0
3
0
0
0
3500
4000
5000
6000
8000
10,000
Phase Properties 93
5.1.5 Water Content
The conventional gas analysis does not indicate any water content. However, the natural
gas in a reservoir is in thermodynamic equilibrium with the interstitial water. Therefore
the gas must be assumed to be saturated with steam. The mole fraction of the steam, y
w
,
can be calculated by the DALTON law (see Eq. 5.7).
(5.14)
where
p
w
: vapor pressure of water at operating temperature
p: operating pressure of the system.
From numerous measurements it could be deduced that the steam content can be well
correlated with pressure and temperature (see Figure 5.6 and Example 5.3). On the other
hand, the influence of the gas composition has been found to be insignificant.
Natural gas and liquid water will combine to form solids resembling wet snow at
temperatures somewhat above the crystallization temperature of pure water. These solids
are called “gas hydrates”. Because these solids can form at temperatures and pressures
normally encountered in natural-gas pipelines, this phenomenon is of particular interest
to the petroleum industry. The conditions for the gas hydrate formation will be
summarized later on.
Example 5.3
The evaluation of the water content of a natural gas as a function
of operating pressure and temperature. This can be done by
correlation through Figure 5.6
FIRST STATE: p = 13.8 MPa(2000 psia)
T = 93.3
o
C (200
o
F)
The water content at this state is evaluated through Figure 5.6 as
.
SECOND STATE: p = 3.45 MPa (500 psia)
T = 60
o
C (140
o
F)
The water content at this state was evaluated through Figure 5.6 as
.
THIRD STATE: p = 2.8 MPa (400 psia)
y
w
p
w
p
------
=
3.8 E 4lb cu ft ⁄ 6.1 E 3kg m
3
⁄ – = –
3.25 E 4 lb cu ft ⁄ 5.2 E 3kg m
3
⁄ – = –
94 Phase Properties
T = 60
o
C (140
o
F)
The water content at this state was evaluated through Figure 5.6 as

5.1.6 Viscosity
Like all intensive physical properties, the viscosity is completely described by
(5.15)
in case of natural gases and
(5.16)
in case of hydrocarbon liquids.
Eq. 5.15 and Eq. 5.16 simply state that the viscosity is a function of pressure, temperature
and composition. LOHRENZ et.al. (1964) presented respective calculations which are
useful for
• determining viscosities of reservoir fluids in compositional material balance
computations and
• predicting the decrease in viscosity which occurs when gases dissolve in reservoir
oils.
The correlations presented below may be viewed as modifications of Eq. 5.15. The
assumptions are practical because the composition is frequently not known. Anyway, the
assumptions are sufficiently valid so that these correlations are frequently used for
reservoir engineering computations.
The gas viscosity µ
g
increases with the temperature. Figure 5.7 shows the dependence of
the viscosity of pure gases on the temperature at atmospheric pressure (0.1 MPa and 14.7
psi, respectively).
Figure 5.8 presents the viscosity of natural gases at atmospheric pressure as a function of
the averaged molecular weight. This diagram elucidates that the viscosity decreases along
with increasing molecular weight. It also contains correction functions for CO
2
, N
2
and
H
2
S.
The theorem of corresponding states has been proved to be valid once more again. The
viscosity at any state of the gaseous system can be predicted through the reduced
viscosity which is defined by
3.8 E 4 lb cu ft ⁄ 6.1 E 3 lb cu ft ⁄ – = –
µ F p T y
1
,y
2
, y
3
…y
k 1 –
, , ( ) =
µ F p T x
1
,x
2
, x
3
…x
k 1 –
, , ( ) =
Phase Properties 95
. (5.17)
The procedure to evaluate µ at certain p and T is the following (see Example 5.4):
• From Figure 5.8, the atmospheric viscosity µ
g,1
has to be evaluated for the given
gravity (or averaged molecular weight) of the gas and the operating temperature.
• According toFigure 4.4 and Figure 5.5, the pseudo-critical pressure, p
pc
, and the
pseudo-critical temperature, T
pc
, are determined.
• After the calculation of the pseudo-reduced state quantities p
pr
and T
pr
, the reduced
viscosity µ
g,r
, which is given by the ratio between the actual viscosity and the one at
atmospheric pressure, is evaluated by use of Figure 5.9.
• The actual gas viscosity can be calculated through Eq. 5.17 where
The gas viscosity determined through this method deviates from laboratory
measurements by 3% at the maximum. This is a satisfactory solution for practical
purposes.
Figure 5.7: Viscosity of paraffin hydrocarbon gases natural gases at atmospheric
pressure (from CARR et al., 1954)
µ
g r ,
p
pr
T
pr
, ( )
µ
g
p T ( , )
µ
g 1 ,
p 1 T , = ( )
------------------------------------
=
µ
g
µ
g r ,
µ
g 1 ,
=
50 100 150 200 250 300 350 400
Temperature [°F]
V
i
s
c
o
s
i
t
y

[
c
P
]
0.004
0.006
0.008
0.010
0.012
0.014
0.016
0.018
0.020
0.022
0.024
H
e
l
i
u
m
A
i
r
N
i
t
r
o
g
e
n
C
a
r
b
o
n

D
i
o
x
i
d
e
H
y
d
r
o
g
e
n

S
u
l
f
i
d
e
M
e
t
h
a
n
e
E
t
h
a
n
e
P
r
o
p
a
n
e
n
-O
c
ta
n
e
n
-D
ec
a
n
e
n
-
N
o
n
a
n
e
i-
B
u
ta
n
e
n
-B
u
ta
n
e
96 Phase Properties
Figure 5.8: Viscosity of natural gases at atmospheric pressure (from CARR et al, 1954)
Figure 5.9: Correlation of viscosity ratio with pseudo-reduced pressure and
temperature (from CARR et al., 1954)
0.0020
0.0010
0.0015
0
0
C
o
r
r
e
c
t
i
o
n

o
d
d
e
d

t
o
V
i
s
c
o
s
i
t
y


[
c
P
]
5 10
Mol e % HS
2
15
H S
2
γg
=
2
.0
1.0
1
.5
0.6
0.0020
0.0010
0.0015
0
0
C
o
r
r
e
c
t
i
o
n

o
d
d
e
d

t
o
V
i
s
c
o
s
i
t
y


[
c
P
]
5 10
Mole % CO
2
CO
2
γ g
=
2
.
0
1
.0
0
.6
1
.
5
0.0020
0.0010
0.0015
0
0
C
o
r
r
e
c
t
i
o
n

o
d
d
e
d

t
o
V
i
s
c
o
s
i
t
y


[
c
P
]
5 10
Mole % N
2
15
N
2 γ g
=
2
.
0
1
.
5
0
.6
1
.
0
15
0
0.004
0.005
0.006
0.007
0.008
0.009
0.010
0.011
0.012
0.013
0.014
0.015
0.016
0.5 1.0 1.5 2.0 2.5 3.0 3.5
10 20 30 40 50 60 70 80 90
Molecular Weight
Gas Gravity (Air = 1.000)
V
i
s
c
o
s
i
t
y

a
t

1

[
a
t
m
]





[
c
P
]
µ
1
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
V
i
s
c
o
s
i
t
y

R
a
t
i
o
,

µ
/
µ
1
0.1 0.2 0.4 0.6 1.0 2 4 3 6 8 10 20 0.3
Pseudo Reduced Pressure, p
r
P
s
e
u
d
o

R
e
d
u
c
e
d

T
e
m
p
e
r
a
t
u
r
e
,

T
r
0
.8
0
.
9
1
.
0
0
1
.
0
5
1
.
1
0
1
.
1
5
1
.
2
0
1
.
3
0
1
.
4
0
1
.
5
0
1
.
7
5
2
.
0
0
2
.
2
5
2
.
5
0
3
.
0
0
Phase Properties 97
Example 5.4
Determination of the gas viscosity at temperature T = 90.5
o
C
(194.9
o
F, 654.7

R) and at pressure p = 12.5 MPa (1800 psia). The
specific gas gravity was found as .
From Figure 5.8, the viscosity at atmospheric pressure and
operating temperature can be read off as
.
Then the pseudo-critical data of the system are evaluated from
Figure 5.5 where
p
pc
= 670 psia
and
T
pc
= 385
o
R.
The calculation of the pseudo-reduced data of the system result in
and
.
The ratio between the actual gas viscosity, µ
g,,
and its viscosity
at atmospheric pressure, µ
g,1
, can now be read off from Figure 5.9:
so that
and
,
respectively.
γ
g
0.7018 =
µ
g 1 ,
0.0122cP =
p
pr
p
p
pc
--------
1800
670
------------ 2.69 = = =
T
T
pc
--------
654.6
385
------------- 1.70 = =
µ
g r ,
1.35 =
µ
g
µ
g r ,
µ
g 1 ,
× 1.35 0.0122 0.0165 = × cP = =
µ
g
0.0165 mPa s =
98 Phase Properties
5.2 Hydrocarbon Liquids
5.2.1 Volume
For practical purposes, it is sufficient to consider hydrocarbon liquids as ideal mixtures
whose volumes can be calculated by adding up the volumes of the components.
The interesting properties (density, viscosity, etc.), which depend on the oil composition,
can be easily determined at the standard (normal) state (atmospheric pressure and T =
60
o
F (15
o
C)). The theorem of corresponding states can be applied again to determine the
properties at any other state of the system (e.g. at reservoir conditions).
However, a crude oil does not remain in the single-phase state at “normal” conditions if
it contains any high mole fraction of light components which will be liberated at
atmospheric pressure and surface temperature. This inconvenience can be avoided by
attributing “apparent densities” or “apparent specific volumes” at standard conditions to
the dissolved and easily volatilized components. Then the respective “pseudo-liquid” data
of any hydrocarbon liquid at standard conditions may be evaluated from the real one at
standard conditions and from the apparent data of the liberated light components.
STANDING and KATZ (1942) examined the properties of many crude oils. They established
that the apparent densities of the two light components methane and ethane depend on the
density of the system (see Figure 5.10). Based on these data and assuming that propane
and heavier components follow the rule of additive volumes, they developed a method for
computing the density of hydrocarbon mixtures. This calculation procedure is presented
through Example 5.5. It is based on Figure 5.11 presenting the correlation of the density
of a system containing methane and ethane with the density of the propanes-plus fraction,
the weight -% ethane in the ethanes-plus and the weight-% methane in the system. Then,
the resulting pseudo-liquid density of the system containing methane and ethane has to be
corrected with regard to the operating temperature and pressure. Figure 5.12 and Figure
5.13 serve this purpose.
The correlation method after STANDING and KATZ is satisfactory for stock-tank liquids
which have low concentrations of methane and ethane.
Phase Properties 99
Figure 5.10: Variation of apparent density of methane and ethane with density of the
system (from STANDING and KATZ, 1942)
Ethane - N-Butane
Ethane - Heptane
Ethane - Crystal Oil
Methane - Cyclo Hexane
Methane - Benzene
Methane - Pentane
Methane - Hexane
Methane - Heptane
Methane - Propane
Methane - Crystal Oil
Methane - Crude Oil
0.3
0.3
0.4
0.4
0.5
0.5
0.6
0.6
0.7 0.8 0.9
Density of System [g/cc], 60°F & atm. Pressure
0.1
0.2
0.3
0.4
A
p
p
a
r
r
e
n
t

D
e
n
s
i
t
y

o
f

M
e
t
h
a
n
e

[
g
/
c
c
]
100 Phase Properties
Figure 5.11: Pseudo-liquid density of systems containing methane and ethane
(from STANDING, 1952)
50 40 30 20 10 0
0
10
20
30
10
20
30
40
50
60
70
30
40
50
60
70
D
e
n
s
i
t
y

o
f

P
r
o
p
a
n
e

p
l
u
s

[
i
b
/
c
u

f
t
]
D
e
n
s
i
t
y

o
f

S
y
s
t
e
m

i
n
c
l
u
d
i
n
g

M
e
t
h
a
n
e

a
n
d

E
t
h
a
n
e

[
i
b
/
c
u

f
t
]
W
t

%

E
t
h
a
n
e

i
n
E
t
h
a
n
e

p
l
u
s

M
a
t
e
r
i
a
l
W
t

%

M
e
t
h
a
n
e
i
n

e
n
t
i
r
e

S
y
s
t
e
m
Phase Properties 101
Figure 5.12: Density correction for compressibility of liquids (from STANDING, 1952)
25 30 35 40 45 50 55 60
65
Density at 60 [°F] and 14.7 [psia], [ib/cu ft]
D
e
n
s
i
t
y

a
t

P
r
e
s
s
u
r
e

m
i
n
u
s

D
e
n
s
i
t
y

a
t

6
0

[
°
F
]

a
n
d

1
4
.
7

[
p
s
i
a
]
,

[
i
b
/
c
u

f
t
]
0
1
2
3
4
5
6
7
8
9
10
1
0
0
0
2
0
0
0
3
0
0
0
4
0
0
0
5
0
0
0
6
0
0
0
8
0
0
0
1
0
0
0
0
1
5
0
0
0
P
r
e
s
s
u
r
e

[
p
s
i
a
]
102 Phase Properties
Figure 5.13: Density correction for thermal expansion of liquids
(from STANDING, 1952)
Example 5.5
Determination of the volume of 1 mole oil at bubble point through
STANDING-KATZ correlations. The bubble point is defined by
and
.
The composition of the system is tabled below.
25 30 35 40 45 50 55 60
65
Density at 60 [°F] and Pressure p, [ib/cu ft]
D
e
n
s
i
t
y

a
t

6
0

[
°
F
]

m
i
n
u
s

D
e
n
s
i
t
y

a
t

T
e
m
p
e
r
a
t
u
r
e
,

[
i
b
/
c
u

f
t
]
0
1
2
3
4
5
6
7
8
9
10
2
4
0
2
2
0
2
0
0
1
8
0
1
6
0
1
4
0
1
2
0
100
60
T
e
m
p
e
r
a
t
u
r
e
[
°
F
]
p
ob
3300psia 22.55MPa ( ) =
T
ob
220
o
F 105°C ( ) =
Phase Properties 103
*) at standard conditions: 0.1 MPa, 15
o
C
The density of the propane-plus can be calculated through
Then the weight -% ethane in the ethane-plus material
and
the weight -% methane in the entire system
are calculated.
From Figure 5.11, the pseudo-liquid density of the hydrocarbon
mixture at standard conditions is read off as 51.5 lb/cu ft.
From Figure 5.12, the pressure correction for p
ob
(3300 psia) is
read off as
Composition Mole
Fraction
z
i
= x
i
Mole
Weight
M
i
kg/kmole
z
i
M
i
Density
kg/m
3
Volume
V
i
*
m
3
C
1
0.4404 16.0 7.046
C
2
0.0432 30.1 1.300
C
3
0.0405 44.1 1.786 507.5 0.0035
C
4
0.0284 58.1 1.650 575.0 0.0029
C
5
0.0174 72.2 1.256 626.0 0.0020
C
6
0.0290 86.2 2.500 662.5 0.0038
C
7+
0.4001 287.0 114.829 907.0 0.1271
Σ 1.0000 130.367 0.1393
ρ
i
k
i 3 =
z
i
M
i

k
i 3 =
V
i

----------------------------
122.021
0.1393
------------------- 876kg m
3
⁄ 54.69 lb cuft ⁄ = = =
z
2
M
2
k
i 2 =
z
i
M
i
---------------------------- 100
1.300
123.321
-------------------
= × 100 1.05 % [ ] = ×
z
1
M
1
k
i 1 =
z
i
M
i

---------------------------- 100
7.046
130.367
-------------------
= × 100 5.4% = ×
104 Phase Properties
0.9 lb/cu ft
so that the liquid density at 3300 psia and 60
o
F results in
51.5 + 0.9 = 52.4 lb/cu ft.
Then the temperature correction for 220
o
F is read off from Figure
6.13 as
3.7 lb/cu ft
so that the oil density at 3300 psia and 220
o
F results in

and
,
respectively.
Figure 5.14: Apparent liquid density of natural gases in various API gravity oils
(from KATZ, 1952)
ρ
ob
52.4 3.7 48.7 lb cu ft ⁄ = – =
ρ
ob
778.5kg m
3
⁄ =
15
20
25
30
35
40
45
0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4
Gas Gravity - Air = 1.0
A
p
p
a
r
r
e
n
t

D
e
n
s
i
t
y

i
n

P
o
u
n
d
s

p
e
r

C
u
b
i
c

F
o
o
t
o
f

d
i
s
s
o
l
v
e
d

G
a
s

a
t

6
0

[
°
F
]

a
n
d

1

A
t
m
o
s
p
h
e
r
e
3
0
°
4
0
°
5
0
°
6
0
°
2
0
°

A
.
P
.
I
.

C
r
u
d
e

G
r
a
v
it
y
Phase Properties 105
Natural gases (dissolved gas) predominantly consist of methane. Therefore, KATZ (1942)
has applied the method of the apparent liquid volume also to crude oil - natural gas
systems. The easily volatilized component is marked by its gravity, , and the scarcely
volatilized fraction is the crude oil with the density and its gravity (in °API) at
standard conditions. The apparent liquid density of the respective natural gas can be
determined according to Figure 5.14.
The amount of the dissolved gas is always indicated as the number of normal cubic meter
per actual liquid volume at standard conditions, nm
3
/m
3
. This ratio is designated as R
s
(solution ratio). In terms of the production, the ratio is also often designated as GOR
(producing gas - oil ratio).
In other words: R
s
expresses how much gas will be dissolved in one cubic meter oil
(volume at standard conditions) at the operating conditions. Therefore, R
s
depends on
pressure and temperature, but also on the oil quality.
5.2.2 Formation Volume Factor
The oil volume inside the tank at surface conditions differs from the volume of the oil
leaving the reservoir. This change in volume has three main sources:
• The most important factor is the gas liberation below bubble point pressure p
ob.
• The difference between reservoir pressure and atmospheric pressure causes slight
expansion of the remaining oil phase.
• The increase in volume because of expansion is more or less compensated by the
shrinkage of the oil resulting from the transfer from reservoir to surface temperature.
γ
g
ρ
o
γ
o
106 Phase Properties
Figure 5.15: Typical graph of formation-volume factor of oil against pressure
The formation volume factor of the oil phase, which must be taken into account, is defined
as
Correspondingly, B
o
can be defined as the volume of one cubic meter tank oil (oil at
standard conditions) at certain operating pressure and temperature.
Figure 5.15 illustrates typical characteristics the relationship between B
o
and p. As long
as the pressure does not fall below the bubble point pressure p
ob
during a pressure
decrease from initial reservoir conditions, change in B
o
will be ruled by the expansion of
the oil. As soon as the reservoir pressure falls below p
ob
, gas will be liberated. This gas
liberation results in a shrinkage of the oil volume. Thus, B
o
decreases at decreasing
pressure as soon as p
ob
has been crossed.
The reciprocal value of B
o
is called “shrinkage factor”.
By multiplying the tank oil volume with the actual B
o
, the volume of the oil inside the
reservoir is obtained. On the other side, the multiplication of the shrinkage factor with the
volume inside the reservoir will result in the stock oil volume.
On the basis of many investigations, STANDING (1947) amplified the correlation to permit
the calculation of bubble point pressures and volume factors, B
o
. The respective
STANDING charts are appended at the end of this textbook:
• Working with Chart 1, it is possible to take the solution ratio, gas gravity, tank oil
gravity and reservoir temperature to determine the pressure at which the given amount
of gas would be in solution. This pressure is the bubble point pressure of the
0
p
b
Reservoir Pressure
B
o,b
1.0
B
o
oil volume and volume of dissolved gas at reservoir conditions
oil volume at standard conditions
---------------------------------------------------------------------------------------------------------------------------------------------------------
=
Phase Properties 107
respective hydrocarbon system, p
ob
.
• Charts 2 permits the evaluation of B
o
at bubble point conditions if GOR represents the
solution ratio.
• The total formation volume factor, B
t
, can be estimated by use of Chart 3. B
t
is
defined by
(5.18)
where
Bo: volume factor of the actual oil phase
Bg: volume factor of the actual gas phase
Rs,i: solution ratio at initial conditions
Rs: solution ratio at operating conditions.
B
o
can be evaluated
• by use of the KATZ correlation (see Figure 5.14) and the STANDING diagrams (see
Figure 5.12 and Figure 5.13)
• or by use of the Charts from STANDING (see appendix).
In most cases, these empirical correlations (see Example 6.6) yield values which are
comparable within 3 or 4% to measured laboratory data.
Example 5.6
Evaluation of the oil volume factor at bubble point conditions from
operation data tabled below both in SI-units and Field-units. The
thermodynamic data of the system were given in Example 5.5. The
calculation is based on one cubic meter tank oil.
B
t
B
o
R
si
R
s
– ( )B
g
+ =
108 Phase Properties
In summing up the producing gas/oil ratios GOR, the initial solution ratio R
si
results in
and
respectively.
The computation of the average specific gas gravity results in
Now the amount of gas can be calculated through
where is the air density:
m
g
= 94 x 1.225 x 0.722 = 83.1 kg
Because the specific gravities are given as
γ
g
= 0.722
and
γ
o
= 30
o
API,
the apparent liquid density of the gas can be evaluated through
Figure 5.14 as
and
,
respectively.
GOR
nm
3
/m
3
]
GOR
cu ft/bbl
kg/m
3 o
API
Separator 1 74 415.5 0.640
Separator 2 16 89.8 0.897
Tank 4 22.5 1.540 876.2 30
Σ 94 527.8
γ
g
ρ
o
γ
o
R
si
94nm
3
m
3
⁄ =
R
si
527.8 cu ft bbl ⁄ =
γ
g
74 0. 640 16 0. 897 4 1. 540 × + × + ×
74 16 4 + +
--------------------------------------------------------------------------------------------
67.9
94
---------- 0.722. = = =
m
g
R
si
ς
a
γ
g
=
ρ
a
ρ
g a ,
24.3 lb/cu ft =
ρ
g a ,
389.2 kg/m
3
=
Phase Properties 109
Now the apparent liquid volume of the liberated gas at standard
conditions (T =15
o
C and atmospheric pressure) can be calculated
through
and so
.
In summarizing 1 m
3
tank oil and 0.214 m
3
apparent liquid volume
of the liberated gas, as well as 876.2 kg tank oil and 83.1kg
liberated gas,the pseudo-liquid density of the given hydrocarbon
system at standard conditions results in
and so
.
From Figure 5.12, the pressure correction of this density for p =
3300 psia results in
.
From Figure 5.13, the final temperature correction for T = 220
o
F
results in
or, in SI-units, respectively, in
.
On the basis of the calculated oil density under the conditions of
the reservoir, the volume of one cubic meter oil under reservoir
conditions results in
and so
.
In other words: 1.288 m
3
saturated reservoir liquid under
reservoir conditions shrinks to 1 m
3
tank oil. The shrinkage
results in
V
g app ,
m
g
ρ
g a ,
-----------
=
V
g app ,
83.1
389.2
------------- 0.214 m
3
= =
ρ
o app ,
876.2 83.1 +
1000 0.214 +
--------------------------------- 790.2kg m
3
⁄ = =
ρ
o app ,
49.3 lb/ cu ft =
ρ
o
49.3 0.9 50.2 lb/ cu ft = + =
ρ
o
50.2 3.8 – 46.4 lb/ cu ft = =
ρ
o
743.7kg m
3
⁄ =
V
ob
959.3
743.7
-------------
1.29m
3
= =
B
ob
1.29 =
110 Phase Properties
5.2.3 Compressibility of Undersaturated Liquids
An undersaturated oil is a compressed liquid in the pressure range above the bubble point
pressure p
b
. Its isothermal compressibility is to be calculated in the same way as for the
gases (see Eq. 5.9) so that
(5.19)
and after integration
. (5.20)
Because
where C is a constant, the density of an undersaturated oil phase can be calculated through
(5.21)
In general, the bubble point pressure, p
b,
is chosen as reference. Eq. 5.20 and Eq. 5.21 will
develop in a Taylor series. Because c
o
is very small, the terms of higher order can be
disregarded so that
(5.22)
and
. (5.23)
Similar formulae are also valid for water.
The compressibility of undersaturated liquids is constant within wide pressure intervals.
Therefore, it can be calculated between any two states of the system, (V
1
, p
1
) and (V
2
, p
2
)
by
1.29 1.000 –
1.29
------------------------------ 100 × 22.48% =
c
o
1
V
---
∂V
∂p
------
¸ ,
¸ _
T
1
B
o
------
∂B
o
∂p
---------
¸ ,
¸ _
T
= – =
B
o
B
ob
e
c
o

p p
b
– ( )
=
ρ
o
C
B
o
------
=
ρ
o
ρ
ob
e
c
o
p p
b
– ( )
=
B
o
~B
ob
1 c
o
p p
b
– ( ) – [ ]
ρ
o
ρ
o b ,
1 c
o
p p
b
– ( ) + [ ] =
Phase Properties 111
(5.24)
where V is averaged volume between p
1
and p
2
.
TRUBE (1957) has reported a method of calculating the compressibility of undersaturated
liquids. This method is based on the determination of the pseudo-critical data of the
system and on the evaluation of the pseudo-reduced compressibility of the undersaturated
liquid, c
r
as a function the pseudo-reduced pressure, p
pr
, and pseudo-reduced temperature,
T
pr
(see Figure 5.16). Then the oil compressibility can be calculated through
(5.25)
If the composition of the undersaturated system is not known, the pseudo-reduced
pressure and temperature can be approximately evaluated through Figure 5.17 on the basis
of
• the calculation of the pseudo-liquid density of the hydrocarbon mixture at standard
conditions (see procedure of Example 5.6),
• the pressure correction of this pseudo-liquid density from Figure 5.12.
Then, the evaluated oil density is related to the density of water at the standard
temperature, .
c
o
V
1
V
2

V p
1
p
2
– ( )
--------------------------
– =
c
o
c
r
p
pc
⁄ =
ρ
w
1000kg m
3
⁄ =
112 Phase Properties
Figure 5.16: Pseudo-reduced compressibility for undersaturated reservoir fluids
(from TRUBE, 1957)
Figure 5.17: Pseudo-critical conditions of undersaturated reservoir liquids
(from TRUBE, 1957)
1 2 3 5 7 10 20 30 50 70100
p
r
c
r
10
-1
10
-2
10
-3
10
-4
0.60
0.50
0.40
1.10
1.00
0.95
0.90
0.80
0.70
Pseudo Reduced
Temperature
0.62 0.64 0.66 0.68 0.70 0.72 0.74 0.76 0.78 0.80 0.82 0.84 0.86 0.88
Specific Gravity of Undersaturated Reservoir Liquid
at Reservoir Pressure. Corrected to 60 [°F]
200
300
400
500
600
700
800
900
1000
1100
1200
1300
P
s
e
u
d
o

C
r
i
t
i
c
a
l
P
r
e
s
s
u
r
e

[
p
s
i
a
]
P
s
e
u
d
o

C
r
i
t
i
c
a
l
T
e
m
p
e
r
a
t
u
r
e

[
°
R
]
P
se
ud
o
C
ritic
a
l T
e
m
p
e
ra
tu
re
Pseudo C
ritic
al P
re
ss
u
r
e
Phase Properties 113
Example 5.7
Evaluation of B
oi
for an under saturated oil at reservoir
temperature T = 105
o
C and under the initial pressure p
i
= 25 MPa
through the reduced compressibility from TRUBE. The composition of
the system was given in Example 5.5 and B
ob
was calculated through
Example 5.6. The critical data are tabled below.
The computation of the pseudo-reduced temperature and pressure
results in
and
.
From Figure 5.16, the reduced compressibility can be read off as
.
The compressibility of the undersaturated oil phase can be
calculated through
Because
(see Example 5.6), B
oi
can be calculated through Eq. 5.23
and results in
B
oi
= 1.29[1 - 0.00292(25 - 22.5)] = 1.281.
T
pr
T
T
pc
--------
105 273.15 +
374.70
------------------------------- 1.01 = = =
p
pr
p
p
pc
--------
25
3.655
------------- 6.84 = = =
c
r
0.02 =
c
o
c
r
p
pc
--------
0.02
6.84
----------
2.92 E 3MPa
1 –
– = = =
B
o b ,
1.29 =
B
oi
B
ob
1 c
o
p
i
p
b
– ( ) – [ ] =
114 Phase Properties
5.2.4 Viscosity
Figure 5.18: Viscosity of subsurface samples of crude oil (from HOCOTT and
BUCKLEY, 1941, after BEAL, 1946)
In general, liquids decrease in viscosity with increasing temperature and increase in
viscosity with increasing pressure. HOCOTT and BUCKLEY (1941) presented
measurements of the viscosity of crude oils containing dissolved gases. Figure 5.18
demonstrates that with increasing pressure, the viscosity decreases up to the bubble point
because of the increasing fraction of light components. The dependence on pressure is
more significant for high viscous oils than for low viscous oils. Anyway, above the bubble
point pressure, the viscosity increases with increasing pressure.
The viscosity of the crude oil can be determined with sufficient accurancy on the basis of
the known viscosity of the tank oil µ
1
at the reservoir temperature T and atmospherical
pressure, the solution ratio R
si
and the undersaturated pressure in the reservoir, ∆p = p
i
-
p
b
. Example 5.8 elucidates the respective correlation procedure.
0 500 1000 1500 2000 2500 3000 3500
Reservoir Pressure [psia]
A
b
s
o
l
u
t
e

V
i
s
c
o
s
i
t
y

[
c
P
]
0
1
2
3
4
5
6
7
8
9
A Permian, West Texas
B Frio Sand, Gulf Coast
C Palo Pinto Lime, North Texas
D Frio Sand, Southwest Texas
B
u
b
b
l
e

P
o
i
n
t
B
u
b
b
l
e
P
o
i
n
t
B
u
b
b
l
e
P
o
i
n
t
Bubble Point
158°F 800ft /
3
bbl GOR 45.7°API G=0.84
1
6
4
°
F

3
8
5
f
t
/
3
b
b
l
G
O
R
2
6
.1
°
A
P
I G
=0
.65
A
B
C
D
100°F Reservoir Temperature
191 ft/bbl Gas in Solution at
Bubble Point Pressure
29.7°API Oil Gravity
1.12 Gas Gravity
3
Legend
Phase Properties 115
Figure 5.19: Viscosity of gas-saturated reservoir crude oils at reservoir conditions
(from CHEW and CONNALLY, 1959)
Example:
Problem: Find the gas-saturated viscosity of
a crude oil having a solution gas/oil ratio of
600 ft/bbl and dead oil viscosity of 1.50 cP
all at the same temperature.
Procedure: Locate 1.5 cP on the dead oil
viscosity scale (abscissa) and go up vertically
to the 600 gas/oil ratio line. Then go left
horizontally to read the answer 0.58 on the
gas-saturated oil viscosity scale (ordinate).
3
0.3
0.3
0.5
0.5
0.7
0.7
1.0
1.0
2
2
3
3
5
5
7
7
10
10
20
20
30
30
50
50
70
70
100
100
Viscosity of Dead Oil [cP]
(at Reservoir Temperature and Atmospheric Pressure)
V
i
s
c
o
s
i
t
y

o
f

G
a
s
-
S
a
t
u
r
a
t
e
d

O
i
l

[
c
P
]
(
a
t

R
e
s
e
r
v
o
i
r

T
e
m
p
e
r
a
t
u
r
e

a
n
d

S
a
t
u
r
a
t
i
o
n

P
r
e
s
s
u
r
e
)
0.1
0.2
Example
1
6
0
0
1
4
0
0
1
2
0
0
1
0
0
0
8
0
0
6
0
0
4
0
0
3
0
0
2
0
0
0
Solution Gas/Oil Ratio
[ft /bbl]
3
1
0
0
5
0
116 Phase Properties
Figure 5.20: Prediction of crude oil viscosity above bubble point pressure
(from BEAL, 1946)
7
6
5
4
3
2
1
Detail
0
0
1
1000
2
2000
3
3000
4
4000
5
5000
6
7
8
9
Pressure [psi]
A
b
s
o
l
u
t
e

V
i
s
c
o
s
i
t
y

[
c
P
]
0
100
10
20
30
40
50
60
70
80
90
A
b
s
o
l
u
t
e

V
i
s
c
o
s
i
t
y

o
f

C
r
u
d
e

O
i
l

a
b
o
v
e
B
u
b
b
l
e

P
o
i
n
t

P
r
e
s
s
u
r
e

[
c
P
]
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Undersaturated Pressure [psi]
(Pressure above Bubble Point less Pressure at Bubble Point)
8
5
8
0
7
5
7
0
6
5
6
0
5
5
5
0
4
5
4
0
3
0
25
20
15
10
5
A
b
s
o
lu
te
V
isc
o
s
ity
o
f G
a
s S
a
tu
ra
te
d
C
ru
de
a
t B
u
b
b
l e
P
o
in
t P
res
s
u
re
[c
P
]
3
5
Crude Oil Prediction above
Bubble Point Pressure
Average Deviation 2.7%
Extrapolated
1
.
0
0
.
9
0
.8
0
.
7
0
.6
0
.5
0
.4
0
.3
0.2
0.1
0 1 2 3 4 5
Pressure 10 [psi]
3
0
10
1
11
2
12
3
13
4
5
6
7
8
9
V
i
s
c
o
s
i
t
y

[
c
P
]
Detail
Phase Properties 117
Exampl e 6. 8
Eval uat i on of t he oi l vi scosi t y at r eser voi r condi t i ons whi ch ar e gi ven
bel ow as wel l as t he i nt er est i ng pr oduct i on dat a.
.
Fr om Fi gur e 5. 19, t he oi l vi scosi t y at bubbl e poi nt condi t i ons can be
r ead of f as
Fr om Fi gur e 5. 20, t he vi scosi t y of t he under sat ur at ed cr ude oi l at p =
3300 psi a can be eval uat ed:
or
5.3 Brines
It has been recognized relatively late that water is - without any exception - one of the
reservoir fluids. GARRISON (1935) and SCHILTHUIS (1938) have been the first who
presented detailed information on the distribution of oil and water in porous rock and on
the origin and occurrence of “connate water” (formation water, capillary water, interstitial
water). Furthermore, they established a relationship between the water saturation and the
rock permeability.
Nowadays, knowledge of the chemical and physiochemical data of the respective
reservoir water is of great importance for the petroleum engineer. Water analyses
• give valuable information about the area of prospecting,
• help to answer questions about the origin of the produced water,
SI-Units Field Units
Reservoir Temperature T
105
o
C
220
o
F
Bubble Point Pressure pb 22.55 MPa 3300 psia
Reservoir Pressure p 25.00 MPa 3600 psia
Initial Solution Ratio Rsi
94 nm
3
/m
3
] 527.8 cu ft/bbl
Standard Oil Viscosity µ
1
0.9 mPas 0.9 cP
µ
ob
0.45cP =
µ
ob
0.45cP =
µ
oi
0.48cP =
µ
oi
0.48 mPas [ ] =
118 Phase Properties
• make sure that injected water is chemically compatible with the reservoir water or
with added water (plugging by precipitations must be avoided),
• are indispensable for feasibility studies in case of chemical EOR processes.
Last but not least, data such as density, compressibility, formation volume factor and
viscosity are required for calculating the displacement process and for establishing a
material balance.
Additionally, this section presents an unusual phenomenon, the so-called “gas hydrate
formation”, during which water and natural gas form a solid phase at temperatures above
the freezing point of water.
5.3.1 Composition of Brines
Reservoir water always contains dissolved inorganic salts, above all sodium chloride,
NaCl. Therefore, these waters are sometimes called “brines” although they cannot be
compared with sea water because both the total salt concentration and the content of
distinctions are different.
The cations usually dissolved in reservoir brine are
Na
+
, K
+
, Ca
2+
, Mg
2+
In certain cases,
Ba
2+
, Li
+
, Fe
2+
,
and sometimes Sn
2+
can be found.
The generally present anions are
and, in some cases, also
.
The elements are present in reservoir water within the following concentration ranges:
Na, Cl in%
Ca, SO
4
in% or ppm
K, Sr in 100 ppm
Cl
-
, SO
2-
4
, HCO
-
3
CO
2-
3
, NO
-
3
, Br
-
, J
-
, S
2-
Phase Properties 119
Al, B, Ba, Fe, Li in 1 - 100 ppm
Cr, Cu, Mn, Ni, Sn, Ti, Zr in ppb (in most cases)
Be, Co, Ga, Ge, Pb, V, W, Zn in ppb (in some cases)
where
% : g substance per 100 g brine
ppm: mg substance per 1 kg brine
ppb: mg substance per 10
3
kg brine
Usually micro-organisms of different species are also present in oil field brines. The
origin of these organisms is not yet totally clear. They contribute to corrosion in the bore
hole and to permeability reduction during water flooding.
There is a wide variability of concentrations and chemical characteristics resulting from
many factors. Such factors may be
• the non-marine origin of a few sediments,
• dilution with ground water,
• enrichment because of vaporization by migrating gas,
• adsorption and cation exchange with clay minerals,
• dissolution of mineral salts in the migrating formation water,
• the chemical reaction with components of adjacent sediments.
Therefore, it is important that water from a certain horizont is characteristically distinct
from all the other brines, even those from the immediate neighbourhood. Thus, water
analyses are mainly used for identifying the produced water with regard to the geological
formation it originates from. In doing so, a simple and easily readable representation of
the great data set from a water sample is desired. The graphical method proposed by STIFF
(1951) may be the simplest graphical method, has maximum utility and, hence, is the most
popular.
This graphical method is shown in Figure 5.21. Horizontal lines extend right and left from
a centrally located vertical line. The positive ions are plotted on the the left of the vertical
line, while the anions are plotted on the right. Characteristic positions are designated for
the ions which are most frequently found in reservoir brines.
120 Phase Properties

Figure 5.21: Essential feature of the water pattern analysis system (from STIFF, 1951)
The analysis data for all ions in question are indicated in ppm (milligram per liter). To
make these values comparable with each other, the contents have to be converted to meq
(milliequivalents) by dividing the respective value with the milliequivalent weight (ionic
weight in mg per valence). If there is a difference between Σ meq of the positive and Σ
meq of the negative ions, the difference is presented as sodium.
Although various scales can be employed, most reservoir waters may be plotted with
sodium and chloride on a scale of 100 meq and a scale of 10 meq for the others. When the
data of the water analysis are plotted on the graph and the adjacent points are connected
by straight lines, a closed “pattern” is formed as a “fingerprint” of the respective brine.
One of many practical applications in correlating producing formations is illustrated by
Figure 5.22. The characteristic pattern of the formations is evident.
Figure 5.22: Course of Arbuckle formation through Kansas shown by water patterns
(from STIFF, 1951)
Scale: meq/liter
Na
100
Ca
10
Mg
10
Fe
10
Cl
100
HCO
10
3
CO
10
3
SO
10
4
Scale:
meq/liter
Na
100
Ca
10
Mg
10
Fe
10
Cl
100
HCO
10
3
CO
10
3
SO
10
4
St. John Field Drach Field
Bloomer Field
Stoltenberg Field
Barton
Stafford
Rice
Ellsworth
Phase Properties 121
5.3.2 Formation Volume Factor
The definition of the formation volume factor of water B
w
corresponds to the definition
of B
o
. Similarly to the hydrocarbon liquids, the change in volume may have three reasons:
• liberation of dissolved gas due to pressure decrease,
• expansion of the brine because of pressure decrease,
• contraction of the brine due to temperature reduction.
In contrast to the solubility of natural gases in oil, their solubility in water is low (see
Figure 5.23a). Additionally, the solubility decrease with increasing salinity of the brine
(see Figure 5.23b). Consequently, the contribution of the gas liberation to B
w
is low.
Since the expansion due to the decrease in pressure and the contraction because of
temperature decrease are also insignificant, B
w
is always numerically small, rarely above
1.06.
Figure 5.24 shows a typical relationship between B
w
and p. In the present case, the initial
reservoir pressure is higher than the bubble point pressure. If the reservoir pressure is
reduced from initial pressure to the bubble point pressure, B
w
increases because of the
water expansion. A further pressure reduction leads to gas liberation. This loss of liquid
volume only balances partly the continuing expansion so that B
w
further increases with
further pressure drop. Along with the reduction of the pressure to atmospheric pressure,
B
w
will reach its maximum value. The difference between this value and 1.0 represents
the thermal contraction resulting from the cooling of the brine from the reservoir
temperature to the standard temperature of 15
o
C. However, B
w
will be smaller than 1.0,
if the increase in volume by dissolved gas is too low to compensate the volume reduction
resulting from high pressure.
122 Phase Properties
Figure 5.23: Solubility of natural gas in water (from DODSON and STANDING, 1944)
Figure 5.24: Typical graph of formation volume factor of water against pressure
The assumption is customary that the bubble point pressure of a gas-saturated brine is
equal to the bubble point pressure of the crude oil inside the corresponding reservoir.
24
22
20
18
16
14
12
10
8
6
4
2
0
60 100 140 180 220 260
Temperature [°F]
a)
S
o
l
u
b
i
l
i
t
y

o
f

N
a
t
u
r
a
l

G
a
s

i
n

W
a
t
e
r


[
f
t
/
b
b
l
]
3
4
0
00
35
0
0
2
0
5
0
200
0
1500
500
1000
4
5
0
0
5
0
0
0
3
00
0
Pressure [psi]
0 10 20 30 40
Total Solid in Brine 10 [ppm]
b)
-3
Correction of Brine Salinity
R
a
t
i
o
:
S
o
l
u
b
i
l
i
t
y

i
n

B
r
i
n
e
S
o
l
u
b
i
l
i
t
y

i
n

W
a
t
e
r
0.8
0.9
1.0
150°F
100°F
250°F
200°F
0
p
b
Reservoir Pressure
B
w,b
1.0
0
Phase Properties 123
Figure 5.25 presents data of B
w
for pure water saturated with natural gas (solid lines).
Naturally, gas saturated pure water has a higher volume factor than pure water at a given
pressure and temperature (dashed lines). B
w
for a reservoir water can be computed with
aid of Figure 5.25 and Figure 5.23 as follows:
• The volume factors for pure water and pure water saturated with natural gas are read
from Figure 5.25 for the given reservoir data.
• The gas solution ratio, R
sw
, for pure water is read from Figure 5.23a. and corrected for
salinity using Figure 5.23b.
• Assuming the effect of gas solubility on B
w
to be linear, the volume factor is
computed by interpolation.
Figure 5.25: B
w
for pure water (dashed lines) and pure water saturated with natural gas
(solid lines) as a function of pressure and temperature (from DODSON and STANDING,
1944)
Example 5.9
Formation volume factor of reservoir water. The reservoir
conditions as well as the salinity of the brine are given below
both in SI- and Field units.
0 1000 2000 3000 4000 5000
Pressure [psia]
W
a
t
e
r

F
o
r
m
a
t
i
o
n

V
o
l
u
m
e

F
a
c
t
o
r

[
b
b
l
/
b
b
l
]
0.98
0.99
1.00
1.01
1.02
1.03
1.04
1.05
1.06
1.07
200°F
150°F
250°F
100°F
124 Phase Properties
The evaluation of B
w
for pure water from Figure 5.25 results in
B
w
= 1.0275
B
w
for pure water saturated with natural brine can be read off
from Figure 5.25 as
B
w
= 1.034.
The gas solution ratio in pure water from Figure 5.23a results in
R
sw
= 15.3 cu ft/bbl
After correction of the gas solubility from Figure 5.23b, R
sw
results in
R
sw
= 15.3 x 0.88 = 13.46 cu ft/bbl
Then the formation volume factor of the reservoir water results in
B
w
= 1.0275 + (1.034 - 1.0275) .
Figure 5.26: Density of brine as a function of total dissolved solids
(from MCCAIN, 1973)
In Figure 5.26, the density of reservoir water at standard conditions is plotted as a function
of the total salt concentration. By dividing the respective density with B
w
, the
SI Units Field Units
Reservoir Temperature T
93.3
o
C 200
o
F
Reservoir Pressure p 20.56 MPa 3000 psia
Salinity 30,000 ppm 30,000 ppm
13.46
15.3
------------- 1.033 =
0 5 10 15 20 25 30
Total dissolved Solids [%]
B
r
i
n
e

D
e
n
s
i
t
y

a
t

1
4
.
7

p
s
i
a

a
n
d

6
0
°
F

[
i
b
/
c
u

f
t
]
76
60
62
64
66
68
70
72
74
Phase Properties 125
corresponding density under reservoir conditions can be received.
From Figure 5.27a the coefficient of the isothermal water compressibility, c
w
, can be
evaluated without considering the gas in solution. At a given pressure and temperature,
the effect of gas in solution is to increase c
w
(see Figure 5.27b).
Figure 5.27: The isothermal coefficient of compressibility of pure water, including
effects of gas in solution (from DODSON and STANDING, 1944)
Example 5.10
Evaluation of the compressibility of reservoir water at the
reservoir data of Example 5.9.
From Figure 5.27a at 3000 psia and 200
o
F, c
w
can be read off as
.
60 100 140 180 220 260
Temperature [°F]
a)
2.4
2.8
3.2
3.6
4.0
2.6
3.0
3.4
3.8
0 5 10 15 20 25
Gas - Water Ratio [ft /bbl]
b)
3
1.0
1.1
1.2
1.3
R
a
t
i
o
:
S
o
l
u
t
i
o
n

C
o
m
p
r
e
s
s
i
b
i
l
i
t
y
W
a
t
e
r

C
o
m
p
r
e
s
s
i
b
i
l
i
t
y
Correction for Gas in Solution
C
o
m
p
r
e
s
s
i
b
i
l
i
t
y

o
f

W
a
t
e
r
c

1
0
[
b
b
l
/
b
b
l

p
s
i
]
w
6
1
0
0
0
[p
s
i
a
]
4
0
0
0
5
0
0
0
6
0
0
0
2
0
0
0
3
0
0
0
T
w
P
V
V
c

,
_

¸
¸

,
_

¸
¸
− ·
1
c
w
3.1 E - 6 psi
1 –
=
126 Phase Properties
From Figure 5.23a, the gas solubility in pure water results in
.
Correction of gas solubility from Figure 5.23b:
.
After evaluating the effect of the dissolved gas from Figure 5.23b,
the compressibility of the reservoir brine results in
,
or
,
respectively.
5.3.3 Viscosity
Regarding the viscosity of oil field brines, only few data are available. Figure 5.28 shows
the water viscosity as a function of the temperature. The viscosity should rise with
pressure and increase significantly with increasing salt concentration. These
dependencies can be observed in Figure 5.28. It is recommended to use this graph for
estimating the viscosity of reservoir waters without conisdering both the pressure and the
salinity of the brine.
The fact that low molecular paraffins are dissolved in water under reservoir conditions
should furthermore lead to a considerable reduction of the brine viscosity. Unfortunately,
no corresponding data have been published until now.
R
sw
15.3 cu ft bbl
1 –
=
R
w
15.3 0.88 × 13.46 cu ft / bbl = =
c
w
1.12 3.1 × E 6 3.47 E - 6 psi
1 –
= – =
c
w
5.03 E - 10 Pa
1 –
=
Phase Properties 127
Figure 5.28: The viscosity of water at oil field temperature and pressure
(from VAN WINGEN, 1950)
5.3.4 Natural Gas Hydrates
Natural gas and liquid water can form solid material which remind of wet snow. This
process takes place at temperatures slightly higher than the freezing point of pure water.
These solid substances are called “gas hydrates”.
The phenomenon is of special interest for petroleum industry. For example, the pressure
decreases in the production line, such as across a choke or in a separator. This can cause
a reduction in the temperature of the gas (adiabatic process) and, therefore, the
condensation of water from the gas. The conditions necessary for hydrate formation could
be established.
Gas hydrates behave rather like dissolutions of gases in ice than like chemical
compounds. The lattice of the hydrate crystal is mainly formed by water molecules while
the hydrocarbon molecules occupy vacancies within the lattice. Therefore, the hydrate
formation is rather of physical than of chemical nature. The water lattice seems to be
similar to ice, because the formation enthalpy is comparable.
Saline Water (60000 [ppm]) at
14.2 [psia]
14.7 [psia]
7100 [psia]
Vapor Pressure
0 50 100 150 200 250 300 350
Temperature [°F]
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
D
y
n
a
m
i
c

V
i
s
c
o
s
i
t
y

[
c
P
]
128 Phase Properties
Although gas hydrates can be compared rather to solid solutions than to chemical
compounds, a certain number of water molecules is associated to each gas molecule. This
may be caused by the crystal structure. The ratio of hydrocarbon to water depends mainly
on the size of gas molecules.
The presence of liquid water is the most important prerequisite for hydrate formation.
However, even if liquid water is present, a meta-stable equilibrium between water and
hydrocarbon gas may occur (under conditions at which generally hydrate formation takes
place). Seed crystals can immediately induce the hydrate formation.
Figure 5.29 shows a part of the phase diagram of a mixture of water and a light
hydrocarbon. The point Q
2
is a “quadruple point” at which four phases are in equilibrium.
The quadruple point Q
1
is at approximately 0
o
C (ice, gas hydrate, water and natural gas
are in equilibrium). The connection Q
1
- Q
2
represents the equilibrium natural
gas-water-hydrate. This line is of special interest.
Figure 5.30 presents the hydrate formation conditions for natural gases of different
gravities. The curves can be used to estimate the conditions of hydrate formation.
Anyway, the results should be applied with caution because of the large discrepancies of
the few published experimental data and the correlation present, especially in case of high
gas gravity and temperature.
Additionally, it has to be taken into account that electrolytes present in the water will lead
to decreasing temperatures at which hydrate formation will occur. Thus, water-soluble
substances are often added to the system to inhibit or suppress the hydrate formation. The
effect of the concentration of various inhibitors on the hydrate-formation temperatures of
a natural gas is plotted in Figure 5.31.
Hydrate portion of the phase diagram for a typical mixture of water and a light
hydrocarbon (from MCCAIN, 1973)
Phase Properties 129
Figure 5.29: Hydrate portion of the phase diagram for a typical mixture of water and a
light hydrocarbon (from MCCAIN, 1973)
Figure 5.30: Pressure-temperature curves for predicting hydrate formation
(from KATZ, 1945)
P
r
e
s
s
u
r
e
Temperature
Q
1
Q
2
C
Hydrate
+
Ice
Hydrate
+
Water
Hydrocarbon Liquid
+
Water
Water
+
Hydrocarbon Gas
Ice
+
Hydrocarbon Gas
30 40 50 60 70 80 90
Temperature [°F]
40
60
80
100
150
200
300
400
600
800
1000
1500
3000
4000
6000
30
500
2000
5000
50
P
r
e
s
s
u
r
e

f
o
r

H
y
d
r
a
t
e

F
o
r
m
a
t
i
o
n

[
p
s
i
a
]
M
e
t
h
a
n
e
0
.
6

G
r
a
v
i
t
y

G
a
s
0
.
7
0
.
8
0
.
9
1
.
0
130 Phase Properties
Figure 5.31: Depression of hydrate formation temperature by inhibitors
(from KATZ et al., 1959)
Temperature calculations have been carried out for typical natural gases, and the results
have been combined with the hydrate formation conditions given in Figure 5.30. This has
been done to evaluate diagrams by which the maximum adiabatic pressure reduction
without gas hydrate formation can be estimated. One of these diagrams is given by Figure
5.32.
Example 5.11
Evaluation of the maximum of adiabatic expansion of a natural gas
without any danger of hydrate formation.
Problem: Up to which pressure can this gas be expanded
0 4 8 12 16 20 24
Weight per cent Inhibitor
0
2
4
6
8
10
12
14
16
18
20
D
e
p
r
e
s
s
i
o
n

o
f

H
y
d
r
a
t
e

F
o
r
m
a
t
i
o
n

T
e
m
p
e
r
a
t
u
r
e

[
°
F
]
1
2
3
4
1 Methanol on 0.63 Gravity
Gas (5-21)
2 Sodium Chloride on 0.68
Gravity Gas (5-43)
3 Calcium Chloride on 0.63
Gravity Gas (5-21)
4 Sodium Chloride on
Methane on 0.555 Gas
Gravity (5-43)
Phase Properties 131
adiabatically without hydrate formation and what will the
temperature be?
Solution steps:
- The ordinate of Figure 5.32 was entered at p
i
= 3000 psia.
- It follows a horizontal movement to the 160
o
F isotherm.
- The next step consists of the vertical movement to the abscissa.
800 psia is the lowest final pressure which will preclude hydrate
formation.
The intersection of the 800 psia line with the dashed line results
in 65
o
F (18.3
o
C) after expansion to 800 psia (5.52 MPa).

Figure 5.32: Permissible expansion of 0.8 gravity gas without hydrate formation
(from KATZ, 1945).
SI-Units Field-Units
Gas Gravity 0.80 0.80
Initial Pressure
p
i 20.68 [MPa] 3000 psia
Initial Temperature
T
i 71[
o
C] 160
o
F
γ
g
100
100
150
150
200
200
300
300
600
600
400
400
1000
1000
2000
2000
3000
3000
Final Pressure [psia]
I
n
i
t
a
i
l

P
r
e
s
s
u
r
e

[
p
s
i
a
]
4000
10000
8000
6000
5000
1500
500
800
40
200
12
0
11
0
10
0
9
0
8
0
70
60
50
197.5
19
5
1
9
0
180
170
160
15
0
14
0
1
3
0
Initial Temp. [°F]
132 Phase Properties
133
Chapter 6
pVT-Measurements
Accurate crude oil data are necessary to control the production of a hydrocarbon reservoir
efficiently. They are of importance for further calculations of the processes which take
place inside a reservoir.
6.1 Sampling
6.1.1 Objectives
The objectives of sampling are to receive samples from a suitable place in the production
wells or surface facilities. The samples should represent the system in the reservoir under
its initial conditions. Only then, the determination of its type, volumetric and phase
behavior, and its composition is assured to supply data for
• geological and reservoir engineering evaluation and forecasting,
• laboratory studies concerning enhanced oil recovery (EOR) methods modifying PVT
properties (and the viscosity) of the system.
6.1.2 General Criteria
Based on the general remarks about the types and behavior of hydrocarbon systems,
conclusions can be drawn about general sampling criteria. It is obvious that - except in
case of dry and wet gases - characteristic samples can be received only in the early life of
the reservoir. Therefore, it can be generally stated that the best time of sampling is at the
beginning of production.
134 pVT-Measurements
To meet all requirements, the careful selection of wells for sampling is very important.
Representative samples cannot be received from wells which are perforated at the
water/oil or gas/oil contact and are producing water and/or non-equilibrium gas. The wells
should be allocated at characteristic sections of the reservoir. Number of samples will
depend on the dip of the structure as well as on the pay thickness. To determine the fluid
properties along with the depth of the pay zone (where monophase system exists), a
minimum of one sample per 30 m can be recommended.
There are restrictions which must always be considered:
• Considering the gravitational effect, it is clear that the sampling in closed wells could
easily lead to misleading data.
• No characteristic fluid can be received within the flowing wells, where a multiphase
system exists or is developing.
• Reliable sampling conditions in the flowing wells and at the wellhead only exist, if
the total composition of the system entering the well and leaving it at the wellhead
remains constant.
Favorable conditions at steady state flow are indicated by stabilization of the pressure and
temperature at the wellhead and in the separator. Moreover, the production gas/liquid
ratio, GOR, should remain constant during a long period of time.
6.1.3 Sampling Methods
There are three methods of sampling. Samples can be taken from the flowing wells
(subsurface sampling) or from the equilibrium phases of the separators (separator
samples). The third possibility is to get representative samples from the wellhead
(wellhead sampling). In every case, steady state conditions must be maintained. The
method used depends on the type of system to be sampled.
There are certain approximations for selecting sampling methods which consider some of
the properties of the produced fluids. Values summarized in Table 6.1 represent average
characteristics which can be used as some rule of thumb. It is clear, however, that the
behavior of any system also depends on reservoir conditions which are missing on the
tabulated data. Therefore, preliminary screening must be used cautiously.
pVT-Measurements 135
On the basis of the general remarks on the phase behavior of different systems in their
initial state and on the changes in their behavior in the depression zone and in the wells,
the following guidelines can be applied in screening sampling methods.
Sampling of equilibrium phases from separators can be used for all systems, but steady
state conditions must exist. Note, that no separator sampling is needed for dry gases and
for saturated and undersaturated oils, if no free water enters the production well and if
there is a considerable length along which a1-phase oil.
For sampling, only those separators can be used where construction ensures highly
efficient separation of the entering phases. In a wide region of flow rates, no liquid
droplets should depart the system with the gas.
Separators must contain pressure gauges and thermometers as well as devices for the
accurate measurement of the gas volumes (liberated and/or delivered), liquid
hydrocarbons, and of water. Each phase can be sampled through sampling valves built in
the separator at that part where pure gas, oil, and water exist.
To determine the phase boundary during the time of sampling, separators often contain
high pressure windows for visual control. When highly paraffinic systems are produced
exhibiting a cloud point in the vicinity of the separator temperature, separators should be
heated. The continuous monitoring of the concentration of some key components in the
gas phase is recommended. This procedure should be considered to control the
development and maintenance of steady-state flow.
Samples of the equilibrium phases should be take using specially designed bottles with
valves at their extreme ends. These bottles are in vertical position during sampling. The
upper valve is connected with the sampling valve of the separator by a flexible tubing
which is resistant to pressure and corrosion by phases sampled.
Sometimes, the bottles and the connecting tube are filled up with an indifferent liquid
(mercury or water from the separator or from other sources) to maintain pressure and to
Table 6.1: Screening sampling methods
Type Density of the
produced liquid
[kg/m
3
]
Gas/liquid ratio
[m
3
/m
3
]
Dry gas -
Wet gas < 740 11000 - 20000
Gas/condensate
system
740 - 780 1000 - 13000
High shrinkage
oils
780 - 800 200 - 1500
Low shrinkage
oils
> 800 < 200
136 pVT-Measurements
exclude air. It is then displaced by the sampled phase (gas or liquid). Whenever the
temperature in the separator is lower than the ambient temperature, the use of mercury (or
brine) is recommended in sampling the liquid phase. In doing so, about 90% of mercury
(or brine) has to be displaced by the separator liquid. After closing the valves into he
container and in the separator and after dismounting the container, additional 10% of
mercury (or brine) should be removed by opening the lower valve. The created gas cap
will prevent uncontrolable pressure increase during shipping. The valves in the contains
must be protected by steel caps during shipping. Data for identification must be delivered
with the sample.
Bottom hole sampling
Bottom hole sampling can be used when only the phase to be sampled is present at the
perforated zone. If water is also produced with the hydrocarbons but remains at the bottom
under steady-state conditions, samples can be taken above the water level. However, the
flow of a 1-phase hydrocarbon fluid must be assured. The method can be controlled by
sampling at different depths and by analyzing the water content of samples. No sampling
within the wells is needed for dry gases or - in general - for systems which exist in a
1-phase state up to the wellhead.
Bottom hole samplers are specially designed devices containing one valve or two valves
at their extreme ends. They are run into the well through a sealed tube mounted to the
christmas tree by a mechanically driven cable. After reaching the desired depth (usually
with open valves to allow the flushing of the sampler by the flowing fluids), the valves
are closed. The valves are operated mechanically or electrically from the surface or within
the sampler at the predetermined depth or time.
At the surface, the sample should immediately be transferred to containers through a
flexible steel tube which - together with the shipping bottle - is filled up with an inert
liquid (e.g. mercury). The sample is displaced by injection of the inert liquid into the
sampler using a volumetric pump or - in a simple manner - by gravitational effects. The
protection of the samples in the shipping bottles and their correct marking for
identification are important as it has already been mentioned.
Sampling at the wellhead
Sampling at the wellhead is a suitable process in case of the production of dry gas or
undersaturated oil at wellhead conditions.
The container is connected with a valve at the wellhead and flushed with the fluid
produced. The previous remarks referring to the protection and identification of the
shipping bottles are also valid for wellhead samples.
pVT-Measurements 137
6.1.4 Special Problems
There are sometimes problems arising from low permeability oil reservoirs (wells being
produced with high depression). As a result, the ratio of the phases entering the wells does
not characterize the initially-in-place fluids (bubble point oil system), especially if water
is also present. In this case, it is only possible to take samples from the equilibrium phases
of the separator and to recombine them under separator conditions corresponding to their
volume ratio. Afterwards, pressure and temperature are increased up to the reservoir
conditions.
The basic concept of this method is that - if the liquid is in equilibrium with other phase(s)
- the composition of the oil phase is not influenced by the contacted volume of the other
phase(s) being in equilibrium with the oil. However, no processes in the reservoir/well
system should take place which may result in a difference between the overall
composition of the reservoir fluid and the produced fluid.
A similar approach can also be used for gas-condensate systems.
Whenever water production cannot be avoided and the volume of the produced water is
high, dissolved gases may be liberated from the water having a composition which could
largely differ from that of the equilibrium hydrocarbon system. In this case, a similar
procedure as above should be used to restore the composition of the hydrocarbon phase.
However, the samples must be preserved in their original state. As for liquids, it is
important
• to prevent any loss - especially in their lighter fractions -,
• to avoid the contact with air because oxidation processes could drastically modify the
composition of hydrocarbon liquids, primarily those of oils.
Specially designed sampling is needed to determine the microbiological contamination of
the produced fluid(s). This procedure is often necessary in case of reservoirs containing
only water from which water will be produced for industrial or public supply. Sterile
samplers and shipping bottles but also anaerobic conditions (in the course of sample
transfer) must be guaranteed.
138 pVT-Measurements
6.2 Experimental Determination of the Volumetric
and Phase Behavior
To determine the volumetric and phase behavior of naturally occurring systems under
reservoir/well/surface conditions, equipments and methods have been developed for the
petroleum industry. A general scheme of the set of equipments used for PVT studies
shown in Figure 6.1.
6.2.1 Equipment
Figure 6.1: Scheme of PVT equipments
Samples to be studied are transferred to the P V T cell. Some known volume of an inert
liquid is injected into the sampling bottle(s), while the same liquid is withdrawn from the
cell initially filled with it.
The selection criteria for the inert liquid are:
• It must exhibit low volatility under (i) ambient conditions and also (ii) under
temperatures and pressures which represent the working conditions.
• It should not react with the fluids contacted.
Control
unit
p
V
T

C
e
l
l
p
Gauge
S
a
m
p
l
e
S
a
m
p
l
e
Hg
Volumetric Pump
Separator
Gasometer
Analysis
V
g
L
G
pVT-Measurements 139
• Its physical and chemical properties must be well known.
Although mercury is widely used, sometimes low melting point alloys are used. No
restriction is needed, however, if the liquid transported by the volumetric pump(s) is
separated from the samples. In this case, the samples will be transferred and measured by
a piston built into the containers and the PVT cell.
Double-acting or two single-acting volumetric pump(s) are needed for this process.
Sometimes only one pump may be used for this purpose, if the inert liquid leaves the cell
through a suitable operated valve at the bottom of the cell. This valve controls the rate and
volume of the out-flowing inert liquid.
The sample being transferred into the cell must be in a 1-phase state. Therefore, liberated
gas or condensate in the shipping bottle has to be dissolved or evaporated in advance.
Referring to separator samples, the proportions of gas and liquid to be injected into the
PVT cell must be equal to those measured into the field separator at the time of sampling.
6.2.2 PVT-Cells
Evaluation of the PVT behavior is carried out on the system transferred into the PVT cell.
The cell has to be a heat- and pressure-resistant vessel which is equipped with valves
made of non-corrosive materials. The cell is mounted into a thermostatic bath which
ensures from detrimental temperature inconstancy in the cell. The electrically heated (or
cooled) medium circulating around the cell could be air or liquid. To obtain the desired
rapid establishment of the equilibrium conditions at any given pressure and/or
temperature, the cell is agitated or stirred inside.
However, the accuracy of measurements depends on the resolution of the measuring
instruments. Nevertheless, the relative error can be decreased by a reasonable increase of
the sample volume studied. Therefore, “micro PVT cells” are not used in practice.
Moreover, considering the change in volume of the system within the pressure range
applied in the PVT studies, larger cell volumes are used to study (i) gases, (ii)
gas-condensate systems, (iii) critical oils, and (iv) water. On the other hand, relatively
smaller ones are used to investigate characteristic oil systems.
Although many different PVT cells are specified, two basic types can be distinguished:
• blind PVT cells (without any possibility of visual observation of the system to be
measured),
• windowed PVT cells, where the volumetric and phase behavior can directly be
observed through one or more pressure- and temperature- resistant windows built into
the cell body. The inside of the cell is lighted from outside by transmitted or reflected
(from the inner wall of the cell) light.
140 pVT-Measurements
Figure 6.2: Blind PVT cell
Blind PVT cells are usually used to study systems in which no phase transition takes place
or phase transition can easily be determined using the isothermal p - V relationship (dry
gases, oil systems, water). A characteristic blind cell is shown in Figure 6.2. Its volume is
about 0.001[m
3
] which is large enough to be used for gas and oil studies as well.
Figure 6.3: PVT cell (after BURNETT)
A special PVT cell for gases was developed by BURNETT (see Figure 6.3). It contains two
or more chambers with different volumes which are separated by valves. The chamber
possessing the smaller volume contains the gas under given pressure. By opening
Valve
Valve
Valve
Hg
Heating Wire
Jacket
Heat Isolator
Bath
Thermocouple
Steel Body
Thermometer
Cell
V
1
V
2
In
Out
Pressure Gauge
pVT-Measurements 141
successively the valves of the evacuate other chambers with known volume, the change
in pressure can be measured. No volumetric pumps are needed in this case.
Windowed P V T cells
Windowed PVT cells are usually used to evaluate systems where the knowledge of the
volumes of coexisting phases is necessary (e.g. wet gases under well or separator
conditions, gas condensate systems). It is also required whenever phase transition cannot
exactly be determined by isothermal p - V relationships (volatile oils or partially miscible
liquid/liquid systems with or without free gas phase).
Figure 6.4: PVT cell (after DEAN-POETTMAN)
A typical one-windowed cell is shown in Figure 6.4. The volumes of the liquid phase can
be determined by the angle of rotating the cell until liquid phases appear in the window.
A PVT cell with variable volume is presented in Figure 6.5. The condensed phase can be
moved to the window by simultaneous lifting (or sinking) the two built-in plungers. The
volume corresponding to that motion can be read on the calibrated stem of the plungers.
Gas
Hg
Liquid
Gas
Hg
Liquid
Windows
142 pVT-Measurements
Figure 6.5: Variable volume cell (after VELOKIVSKIY et al.)
As far as SLOAN’s windowed cell (see Figure 6.6) is concerned, there are three sections.
In the upper part, a plunger is moved by the fluid injected or withdrawn by a volumetric
pump. In the lower part, the condensed liquid is collected and can be lifted up to the
windows which are built into the middle part. This can be done by injecting inert fluid.
Meanwhile, pressure can be kept at the desired constant value by the simultaneous lifting
of the plunger.
Another type contains one capillary glass tube within the cell through which the inner
volume can be lighted and observed using an optical system. It serves also for the
measurement of the volume of phases. The scheme of this cell is shown in Figure 6.7.
Three windows are built into the body of the most generally used cell as shown in Figure
6.8.
Figure 6.6: PVT cell (after SLOAN)
Window
1 Orifice for Introduction of Mercury through
the Upper Cap, for Operation of the Piston
2 Piston with joist Packing
3 Hole and Valve for Introduction of Gasoline
4 Hole and Valve for Introduction and Sampling
of Gas
5 Spiral for stirring Gas. The Liquid is stirred by
a Spiral Coil Wound around the axis
6 Sapphire Viewing Window
7 Viewing Window
8 Lighting Window
9 Liquid Volume Measurement Chamber
10 Magnetic Agitator
1
2
3 4
5
6
7 8
9
10
pVT-Measurements 143
Figure 6.7: PVT cell (after WELLS-ROOF)
Figure 6.8: RUSKA cell
Capillary
(Glass)
Windows
144 pVT-Measurements
Figure 6.9: RUSKA volumetric mercury pump
6.2.3 Volumetric Pumps
Volumetric pumps - the so called positive displacement pumps - are used to measure the
volume of the inert liquid injected to and withdrawn from the cell respectively. The
pressure can be increased or decreased. The transported volume is proportional to the
change in volume of the system studied.
In the generally used types, the inert liquid is displaced by a plunger from a pressure
resistant and thermostatically controlled cylinder. The movement of the connecting
screw-spindle can be read on a calibrated rod and on a dial to increase the accuracy.
In practice, hand- and mechanically operated pumps are used, the latter sometimes with
variable speed. A special type of volumetric pumps is the double-acting one: Two
plungers are working simultaneously, but in opposite directions.
A scheme of a hand-operated volumetric pump is shown in Figure 6.9. Its general
construction does not differ from the pump which is mechanically driven.
6.2.4 Auxiliary Equipment
The PVT cell, the volumetric pump(s), and other parts of the PVT system are
interconnected by pressure-resistant stainless steel tubes. In order to separate single parts
of the equipment, valves are mounted in the network of tubes.
Pressure in the system is measured by gauges containing BOURDON-tubes. However, also
dead weight gauges can be used for this purpose and for the calibration of the
BOURDON-tube-gauges as well. There are also gauges in which an electric signal is
proportional to the pressure measured. These gauges are also built into the tube network.
They are filled with the inert liquid used in the volumetric pumps.
In order to store the inert liquid (mercury), a container is used which is also interconnected
pVT-Measurements 145
through valves with the different parts of the PVT system.
The temperature of the volumetric pump(s), in the PVT cells, and in the thermostate
around the cell is usually measured by thermocouples (may be controlled by
thermometers). Whenever pressure, temperature, and the volume in the pumps are
measured electrically, the process can be automated and computerized.
If the gas phase liberated in the PVT cell will be displaced, its volume must be measured.
For this purpose gasometers are used. However, gas could condense under ambient
conditions or at pre-selected pressures. Therefore, a laboratory separator must be inserted
between the PVT cell and the gasometer.
6.3 Methods
Basically, there are three methods generally used in PVT studies. They are distinguishable
by considering the change and the direction of change in the composition of the system.
6.3.1 Flash Process
In the course of flash processes, pressure is gradually increased or decreased at constant
temperature under equilibrium conditions. At each pressure step, the volumes of the
system and that of the existing phases are measured and the development of phases is
observed. Measurements are usually repeated at other temperatures, too, covering some
range between reservoir and surface temperatures (in the separator or in the stock tanks).
Measurements start at or above the reservoir pressure. The lowest pressure applied is
determined by the system studied. For systems in which no phase transition takes place,
the lowest pressure selected could be the atmospheric one. In this case the flash process
simulates the whole course of production.
Whenever phase transition occurs, measurements should be carried out in a pressure
region which correctly characterizes the phase transition and the 2-phase (or multiphase)
behavior of the system. In this case the flash process - carried out at reservoir temperature
- simulates the behavior of the system far inside the reservoir where its composition is not
influenced by the production. A scheme of the flash process has been presented in Figure
4.3.
146 pVT-Measurements
6.3.2 Differential Process
The differential process describes the change in the volumetric and phase behavior of the
system at reservoir temperature and under equilibrium conditions, where any decrease in
the reservoir pressure results from the stepwise production of one of the equilibrium
phases (usually liberated or free gas).
In contrast to the flash process during which the composition is maintained and the
volume is varied, the differential process is carried out at constant volume and by
variation of the system composition.
In the course of the differential study - also explained in Figure 4.3 -, the volumes of the
system and those of the existing phases are measured. The development or disappearance
of phases is observed. To follow the change in composition of the system and to
characterize the displaced phase (gas), the composition of the latter is also determined. By
use of a laboratory separator in which pressure and temperature correspond to those
during the field separation process, further information on the volume (and composition)
of phases separated can be given.
The differential process is repeatedly carried out, usually down to the lowest possible
reservoir (separator) pressure. As a result of the gradually changing composition, the
values of characteristic parameters will differ from those determined by the flash process.
except the starting point representing the intact system under initial reservoir conditions.
6.3.3 Reverse Differential Process
The reverse differential process is used to study the effect of any gas injection process on
the volumetric and phase behavior of systems. This can be done under initial conditions
(gas cycling) or by starting from a given depleted state (gas injection with pressure
increase). Both are constant volume processes. In the course of gas injection (using a
pre-selected gas), the phase volumes are measured and phase transition(s) are observed at
reservoir temperature. There are two subgroups of the reverse differential process.
If the effect of pressure maintenance by gas injection is being studied after each step of
the differential measurement, p is re-established by injecting a measured volume of the
pre-selected gas. This method is used, when
• in the initial state of a gas condensate system, the liquid phase exists beyond the
restriction of retrograde condensation,
• partial or total evaporation of the immobile liquid condensate is needed.
In depleted (gas condensate or oil) reservoirs, the effect of increasing the pressure by
pre-selected gas can also be studied. In the course of a stepwise process, an isothermal
pressure increase is realized under constant volume conditions. The highest pressure to be
pVT-Measurements 147
reached corresponds to the fracturing pressure of the formation, or - if this information is
not available - it is about 10% higher than the initial reservoir pressure. If some 1-phase
system develops below this pressure, it is obviously no need to continue the process in a
higher pressure region. The effect of gas injection is often the combination of the methods
discussed above.

© No part of this publication may be reproduced in any form. Only students of the University of Leoben may copy for studying purposes. Edited by DI Claudia Scharf - October 2004

Table of Contents
Chapter 1 Review of Thermodynamic Terminology ........................................................ 1 Chapter 2 Phase Behavior ................................................................................................... 3
2.1 2.2 Gibbs’ Phase Rule ........................................................................................................3 Single-Component System ..........................................................................................4 2.2.1 Water .............................................................................................................4 2.2.2 n-Butane .........................................................................................................7 Critical State and Quantities of Corresponding States ...............................................11 Binary Systems ..........................................................................................................13 Multi-Component Systems ........................................................................................20 2.5.1 Ternary Phase Diagrams ..............................................................................20

2.3 2.4 2.5

Chapter 3 Equations of State ............................................................................................ 31
3.1 3.2 3.3 3.4 Change of State at Low Compressibility ...................................................................32 Equation of State of Perfect and Real Gases .............................................................33 Cubic Equations of State ............................................................................................35 Virial Equation of State .............................................................................................47

Chapter 4 Calculation of Phase Equilibria ...................................................................... 49
4.1 Mixtures .....................................................................................................................49 4.1.1 Definitions ...................................................................................................49 4.1.2 K-factors ......................................................................................................50 Composition of Phases in Equilibrium ......................................................................54 4.2.1 Definitions ...................................................................................................54 4.2.2 Evaluation of K-Factors Using Convergence Pressures .............................62 4.2.3 Evaluation of Convergence Pressure ...........................................................68 4.2.4 Evaluation of by use of Peng-Robinson equation of state ..........................69

4.2

Chapter 5 Phase Properties ............................................................................................... 81
5.1 Natural Gases .............................................................................................................81 5.1.1 Volume ........................................................................................................81

............................................. 98 5..........................................................................138 6..............................................................................................................................................2.............................................................................1 Flash Process ......5 Water Content ......3 Sampling Methods .............. 114 Brines ...............4 Auxiliary Equipment ........................... 105 5.......................................................4 Correlation of Z-Factor on the Basis of Reduced State Variables ..................................1.............. 138 6...3 Reverse Differential Process .......2...........2 6......................3 Compressibility .......... 134 6................................................ 146 6....................3.......................1..........................................................................................1.................................... 139 6.....................................................................2...3 Viscosity .............................. 93 5.....1.............3.....2....84 5. 118 5........2 5................... 133 6..................................3....... 117 5..............................................4 Viscosity ........ 84 5..............................3 Volumetric Pumps ...................................1.....................................2..........ii Table of Contents 5.......... 126 5................2 Formation Volume Factor .............................................. 127 Chapter 6 pVT-Measurements ............... 133 6................... 83 5................2 Differential Process .......................4 Special Problems ...............3..........................1....................2.....................................................................................................................................................3 5...........................................................................................2 PVT-Cells ........... 145 6............1.....................................3 ...............................................................................3.................................................. 144 6.........1 Volume .........2................................................................3 Compressibility of Undersaturated Liquids ...............110 5........ 94 Hydrocarbon Liquids ....................................1.................................2 Formation Volume Factor ....................................................................4 Natural Gas Hydrates .................................1.............1 Sampling .........................1 Composition of Brines ............................................................................... 98 5....... 137 Experimental Determination of the Volumetric and Phase Behavior ......................... 121 5.........2 Formation Volume Factor .....................................6 Viscosity ........ 144 Methods ......................3.................................................................................2...............................................................2 General Criteria .........................................................................3..... 133 6................................................... 133 6...1 Equipment ................... 146 6................ 145 6...1 Objectives ...........

temperature phase diagram of the binary system ethane (z = 0.11: Figure 2.1: Figure 4. 1957)67 Convergence pressure data .1: Figure 4.17: Figure 2. 1973)26 Phase diagram of a wet gas (from McCain. 5.3: Figure 2.4: Figure 2. ethane and propane versus pressure at T = 140oF (from Standing.density phase diagram of n-butane (from GYULAY. 1967)9 Critical pressure as a function of number of C-atoms in homologous series (after Gyulay. 1945)52 Ideal and real K-factors of n-butane at 60[oC]52 Flash and differential vaporization59 K-factors for methane-propane at Tc = 100 oF (from Sage.000-psia convergence pressure (from NGAA.List of Figures Figure 2. 1967)11 Critical temperature as a function of numbers of C-atoms in homologous series (after Gyulay. 1967)5 Vapor pressure diagram of n-butane (from Gyulay. 1967)8 Temperature .methane for binary hydrocarbon mixtures (from Winn.2: Figure 4. 1967)14 Phase equilibrium surface of the binary system ethane/n-heptane (from Gyulay. 1973)26 Phase diagram of a high-shrinkage crude oil (from McCain.8: Figure 5. 1967)7 Pressure . 1957)64 K-factors for methane.4: Figure 4..1: Figure 2.000 psia convergence pressure (from NGAA.9683)/n-heptane17 Mole fraction(ethane) .000 and 5.23: Figure 2. 5.18: Figure 2. 1967)12 Combined reduced pressure .22: Figure 2.21: Figure 2.9: Figure 2.19: Figure 2.5: Figure 4. Lacey and Schaafsma) (1934) 63 Comparison of K-factors at 100 oF for 1. 1973)27 Phase diagram of a low shrinkage crude oil (from McCain. 1959)24 Phase diagram of a dry gas (from McCain. 1938)15 Pressure .2: Figure 2.7: Figure 4.5: Figure 2.000 psia convergence pressure (from NGAA.12: Figure 2.16: Figure 2. 1973)27 Phase diagram pairs of gas cap and oil zone28 Phase equilibrium surface of oil/natural gas systems (from Gyulay.schematic (not drawn to scale)4 Phase equilibrium surface of a pure substance (from Gyulay.volume phase diagram of n-butane (from GYULAY. 1977)83 .6: Figure 4.temperature phase diagram of the binary system ethane/n-heptane (from Kay.10: Figure 2.8: Figure 2. 1967)18 Properties of ternary diagrams20 Typical features of a ternary phase diagram21 Triangular diagrams for the methane/propane/n-pentane system at 160 oF(71 oC) (after Dourson et al..3: Figure 4.6: Figure 2. 1952)70 Z-factor of methane.reduced volume phase diagram of paraffins with low molecular weight (after Gyulay.Pressure diagram of the binary system ethane/n-heptane (from Gyulay.7: Figure 2. 1973)26 Phase diagram of a retrograde gas condensate (from McCain.13: Figure 2.temperature diagram of the binary system ethane/n-heptane (from Gyulay.20: Figure 2.14: Figure 2. 1967)17 Mole fraction (ethane) .15: Figure 2. 1943)23 Critical loci of methane/propane/n-pentane systems (from Katz et al. 1967)14 Pressure . 1957)66 K-factors for hexane.24: Figure 3. 1967)29 The Van der Waals isotherms near the critical point35 Fugacity of natural gases (from Brown.1: Water system .

. 101 Figure 5... 1957)112 Figure 5.... 1951)120 Figure 5.. 1952)..................22: Course of Arbuckle formation through Kansas shown by water patterns (from Stiff...............6: Water content of natural gas in equilibrium with liquid water (from Katz et al..8: Viscosity of natural gases at atmospheric pressure (from Carr et al...25: Bw for pure water (dashed lines) and pure water saturated with natural gas (solid lines) as a function of pressure and temperature (from Dodson and Standing.... after Beal. 1950)127 Figure 5........ 1949)85 Figure 5.. 1942)88 Figure 5....11: Pseudo-liquid density of systems containing methane and ethane (from Standing.. 1941.28: The viscosity of water at oil field temperature and pressure (from van Wingen.... 1944)123 Figure 5.....20: Prediction of crude oil viscosity above bubble point pressure (from Beal.7: Viscosity of paraffin hydrocarbon gases natural gases at atmospheric pressure (from Carr et al...21: Essential feature of the water pattern analysis system (from Stiff.15: Typical graph of formation-volume factor of oil against pressure .. 1973)124 Figure 5. 1954)96 Figure 5...18: Viscosity of subsurface samples of crude oil (from Hocott and Buckley.14: Apparent liquid density of natural gases in various API gravity oils (from Katz.....5: Pseudo-critical properties of Oklahoma City Gases (from Matthews et al....... 1942)88 Figure 5....19: Viscosity of gas-saturated reservoir crude oils at reservoir conditions (from Chew and Connally. 106 Figure 5. 1942)99 Figure 5....3: Z-factor for natural gases (from Brown et al.. 1946)114 Figure 5....10: Variation of apparent density of methane and ethane with density of the system (from Standing and Katz...26: Density of brine as a function of total dissolved solids (from McCain..2: Z-factor as a function of reduced pressure for a series of reduced temperatures (from Sage and Lacey... 1954)95 Figure 5... 1954) ..... 120 Figure 5. 96 Figure 5..27: The isothermal coefficient of compressibility of pure water.16: Pseudo-reduced compressibility for undersaturated reservoir fluids (from Trube. 87 Figure 5. 122 Figure 5.......24: Typical graph of formation volume factor of water against pressure..4: Pseudo-critical temperatures and pressures for heptanes and heavier (from Matthews et al... 122 Figure 5........ 1951) .....29: Hydrate portion of the phase diagram for a typical mixture of water and a light ....12: Density correction for compressibility of liquids (from Standing.. 1952)102 Figure 5....9: Correlation of viscosity ratio with pseudo-reduced pressure and temperature (from Carr et al... 1952)100 Figure 5.13: Density correction for thermal expansion of liquids (from Standing.. 1957)112 Figure 5.... 1944)125 Figure 5.. 1948) ....... 1959)92 Figure 5...... 1946)116 Figure 5..........17: Pseudo-critical conditions of undersaturated reservoir liquids (from Trube.....iv List of Figures Figure 5..23: Solubility of natural gas in water (from Dodson and Standing.. 1959)115 Figure 5....... including effects of gas in solution (from Dodson and Standing..... 1952)104 Figure 5......... 1944) ...

.....................................................143 Figure 6............................................138 Figure 6.......................4: PVT cell (after Dean-Poettman) .....30: Pressure-temperature curves for predicting hydrate formation (from Katz......................................................8: Ruska cell .......32: Permissible expansion of 0.............................5: Variable volume cell (after Velokivskiy et al............................................................................142 Figure 6.............................8 gravity gas without hydrate formation (from Katz...................................9: Ruska volumetric mercury pump .................................List of Figures v hydrocarbon (from McCain................... 1945)129 Figure 5............140 Figure 6..31: Depression of hydrate formation temperature by inhibitors (from Katz et al................................................. 1945)..6: PVT cell (after Sloan) ............................................................ 1973)129 Figure 5..........................................144 ........1: Scheme of PVT equipments...... 1959)130 Figure 5..3: PVT cell (after Burnett) .............................7: PVT cell (after Wells-Roof)...................................................................................................................................142 Figure 6..140 Figure 6.....................141 Figure 6.) .....................143 Figure 6.................131 Figure 6..................2: Blind PVT cell .......

.

it may be helpful to refurbish the thermodynamic terminology at the beginning as far as it is indispensable to the understanding of these tools. terms such as “oil reservoirs” and “gas reservoirs” are used both in colloquial speech and technical literature.g. Intensive properties such as temperature and pressure are independent of the size of the system. e. Therefore. However. It may consists of several compounds. separator. of a large number of various hydrocarbons. The concept of a closed system is of major interest in applied hydrocarbon thermodynamics. 1 . It is not obligatory that a phase is chemically homogeneous. entropy etc. such as volume. of the amount of the substance in the system. temperature. i. tank). However. The thermodynamic properties of a system are classified into “intensive” and “extensive” properties.real or imaginary . enthalpy. A “phase” is defined as a physically homogeneous portion of matter.is called a “boundary”.e. tables and equations) for sketching the state of a hydrocarbon system as a function of the state variables is one of the objectives of this textbook. pipe lines. Both systems may exchange energy with the surroundings. The phases of a heterogeneous system are separated by interfaces and are optically distinguishable. The surface of the system .. Thermodynamics has evolved to a science of studying changes in the state of a system with changes in the conditions. A systematic presentation of basic thermodynamic tools (charts.Chapter 1 Review of Thermodynamic Terminology When considering hydrocarbon reservoirs. Changes in the state of aggregation during production should always be taken into account in consequence of changes of the reservoir pressure and changes of pressure and temperature in the production system (tubing.e. these terms are insufficient. It is called a “homogeneous” closed system if it contains a single phase. Thermodynamic studies are generally focused on arbitrarily chosen systems while the rest of the universe is assumed as “surroundings”. e. in opposite to an “open system” which exchanges matter with the surroundings. Extensive properties depend on the amount of the substances. A system is called a “closed system” if it does not exchange matter with the surroundings. i. composition. pressure.g. A “heterogeneous” closed system contains more than one phase. a natural gas phase or an oil phase.

. a system changes until it reaches its equilibrium state. • isothermal (constant temperature). • isochoric (constant volume). “Equilibrium” has been defined as a “state of rest”. • isobaric (constant pressure). “State functions” or “state variables” are those properties for which the change in state only depends on the respective initial and final state. no further change or . all real processes are “irreversible” with varying degrees of departure from a reversible one. the mole volume.more precisely . It is this path-independent characteristic of the state functions that makes it possible to quantify any change of a system.g.2 Review of Thermodynamic Terminology the extensive properties per unit mass or mole are intensive properties. e. On the other side.no net-flux will take place unless one or more properties of the system are altered.. Any change of a system is called a “thermodynamic interest” in the thermodynamic study of the system: • adiabatic (no heat added to or removed from the system). In an equilibrium state. A process is called “reversible” if it proceeds through a series of equilibrium states in such a way that the work done by forward change along the path is identical to the work attained from the backward change along the same path. However.

By specification of two variables. These are united in the so-called “equation of state”: F ( p.1) 2. the phase rule introduced by GIBBS (1928) is applied. F = C–P+2 where • P: number of phases. • F: number of degrees of freedom. temperature. • C: number of components. V. concentrations which can be varied within finite boundaries without changing the number of phases in the system. the third will be stipulated. F is defined as the number of quantities such as pressure. 3 (2. Water in brine form and a gaseous and/or liquid hydrocarbon phase are regarded as reservoir fluids. T ) = 0 . C is defined as the smallest number of constituents by which the coexisting phases can be completely described.1 GIBBS’ Phase Rule When referring to the number of phases coexisting in the thermodynamical equlibrium. (2. The phase behavior of the actual hydrocarbon mixture in the reservoir can be described as a function of the state of the system. A system in thermodynamic equilibrium posesses an accurately defined relationship between the state variables.Chapter 2 Phase Behavior Hydrocarbon reservoirs consist of rock and fluids.2) .

To gain a full understanding. it is discussed as a single-component system in this context. The circumstances in case of 2. Therefore. to the composition of the particular phases or to the proportions of the phases are given. However. 2. 2.1 Water B 22. The phase diagram in Figure 2.2 Single-Component System 2. it is best to discuss the phase behavior of pure substances (single-component systems) first.2 describes the system in a qualitative and very general manner.charaterized by p and T .schematic (not drawn to scale) Water is one of the most thoroughly studied chemical compounds.01 0.T).0075 100 374 Temperature [°C] Figure 2.4 Phase Behavior Eq.or even multi-component systems are much more complicated.0006 0 Vapor A 0. water (liquid state) and steam (gaseous state). no reference to the state variables (p.1 illustrates at which state of the system .09 C Pressure [MPa] Water Ice 0. The possible phases are ice (solid state).1: Water system .two or all three phases are in equilibrium: • The sublimation curve OA signifies the equilibrium between the solid and vapor.2. • The melting point curve OB combines the states of equilibrium between the solid .

• At the triple point O. all three phases are in eqilibrium. 2.2 (GIBBS’ phase rule) and Figure 2.2: Phase equilibrium surface of a pure substance (from GYULAY. 1967) . Either the temperature or the pressure is freely eligible without counteracting any given phase equlibrium. • The vapor pressure curve OC specifies the states of the system at which the liquid and vapor coexist. • F = 2 in any area of single phase state (P = 1.1 The degree(s) of freedom in different states of a single-component system.1: • F = 0 at the tripel point (P = 3).Phase Behavior 5 and liquid state. C D E A B p D A C C p T D E B v E B A p T v T v C A D B E Figure 2. the thermodynamical data at this point are p = 610.16 K. Tc(H2O)= 647. Example 2. the "wet" vapor is in equlibrium with the "saturated" liquid.09 MPa. On this curve. by use of Eq. Both pressure and temperature (naturally inside finite boundaries) are freely eligible without transforming the system into a multi-phase system.15 K.6 Pa and T = 273. • F = 1 on the curves describing the 2-phase (P = 2) equlibria. In case of water. • The end point C of the vapor pressure curve is the critical point and signifies the highest temperature and pressure at which the liqiud and vapor coexist (pc (H2O) = 22.

and (iii) the pressure coefficient β : 1 ∂V κ =  – -.7) (2. (ii) the cubic expansion coefficient a. V can only assume positive values. a plane is defined by two vectors which infers that the differential quotients ∂p  -----. the following is valid: α = pκ β (2. .V.6 Phase Behavior The state variables. On the basis of which describe a change of a system’s state by changing pressure and temperature at constant specific volume.  ------ .3) The three differential quotients describe three essential fundamental properties of the system: (i) the isothermal compressibility κ . Thus. the graphical illustration of the state of any system is only situated in the positive section of the p. the 2-phase state (e.2. • The dew point curve characterizes states in which a negligible small amount of liquid .6) Since the specific volume . according to Eq. p. = – -------------- ∂T  V  ∂V ----- ∂p  T (2. The shape of such an equilibrium surface is substance specific. T.1 .in contrast to the representation in Figure 2. water in equilibrium with steam) is characterized by any area surrounded by two curves which converge at the critical point: • On the bubble point curve. ------ p  ∂T V Then. it is possible to draw only one single tangential plane at an optional point of the surface.  ∂V  ∂T  V  ------ p  ------ T ∂T ∂p are not independent of one another.g. an infinitesimal small amount of vapor is in equilibrium with the “saturated” liquid. T-coordinate system An example for an equilibrium surface is giben in Figure 2.5) (2.4) (2.appears now as a state variable. However. 2.  ∂V . it is proven that  ∂V ----- ∂T p ∂p  -----.3.  V  ∂p T 1 -----α = -. ∂V  V  ∂T  p 1 ∂p β = . Assuming that the partial derivatives are steady.

the vapor pressure curve is not isochoric. 2.V.T coordinate system are displayed in Figure 2. T-plane results in isobares (p = const). Figure 2.05m3/Mol A C 3 3 F Liquid B D m 0 . This aspect will be described in detail by discussing the phase behavior of the simple hydrocarbon n-butane.4.3 and Figure 2. it is possible to draw isochores: One upwards into the liquid phase and one downwards into the gas phase. T-plane results in isochores (V = const).Phase Behavior 7 is in equilibrium with “wet” vapor. the bubble point curve and dew point curve coincide. It is common to simplify the complex spatial illustration of the equilibrium surface by applying normal projections. • the projection into the p. 1967) . However.5 V= l /Mo 2 H E Vapor 1 0 50 100 150 200 Temperature [°C] Figure 2.2. Of course.2 displays that • the projection into the p. When regarding the projection which represents the 2-phase area (liquid-vapor) in the p. T-plane.2 n-Butane Projections of the equilibrium surface into two planes of the positive section of the p.3: Vapor pressure diagram of n-butane (from GYULAY. The resulting single curve is named vapor pressure curve. V-plane results in isotherms (T = const). 6 5 G Pressure [MPa] 4 V=0. • the projection into the V.

V-plane (see Figure 2. describe those states of the system without any phase transformation by changing the pressure. The intersection point of all other isotherms with the bubble point curve (e. including the critical point C. At the point E.4).volume phase diagram of n-butane (from GYULAY. In case of the state A. it is possible to comprise the whole 2-phase region. 5 Vapor 4 C 150 Liquid DP 17 0 °C p [MPa] 16 15 0 15 5 2 . further expansion will result in further pressure decrease.8 140 3 130 BP A D B Liquid and Vapor 2 0 0. A) marks the specific volume of the saturated liquid which is in phase equilibrium with the specific volume of the wet vapor (e. When moving to the bubble point B by isothermal expansion. 1967) Isotherms. vaporization commences. marked by point B). the mass ratio of the liquid and vapor phase being in equilibrium with one another can be calculated by the principle of the lever: mL ⁄ m v = DB/AD (2. Then the continuation of this isothermal expansion includes no further pressure drop in the system until the last molecule has passed over to the gas phase. which do not intersect the 2-phase region.g. An isochoric change of state must be analogously discussed. n-butane is in the state of a “dry” vapor.015 V [m/kg] 3 Figure 2. From this moment.010 0. In addition. the isochores V = 0.8) .005 0.3 illustrates the vapor pressure curve of n-butane. By applying the projection of the equilibrium surface into the p. butane is an undersaturated liquid.4: Pressure .5 m3/kmole inside the vapor region are shown.05 m3/kmoleinside the liquid phase region and V = 0.g. Considering point D within the 2-phase region of the system (specific volume of the system in total). the isothermal vaporization or condensation takes place as an isobaric process.8 Phase Behavior Figure 2. In this region.

19 kg. average values of fluid and vapor density are located on a straight line. The isobare touching the critical point has an inflection point just as the critical isotherm in Figure 2. Vv. 200 Liquid T [°C] 5 M C 150 Vapor Pa DP Liquid and Vapor 200 400 3 BP 100 0 ρ [kg/m ] Figure 2.7 MPa can be evaluated from Figure 2. 2. mL + mv = 100 = 0. The specific volume of the vapor phase is marked by point B in Figure 2.5: Temperature .4.8 From Figure 2.3125 . The vapor volume of the system.density phase diagram of n-butane (from GYULAY. With the help of this rule (C AILLETET-MATHIAS rule (1886)).3125 m v + mv mv = 76. can be now calculated by multiplying V with mv: Vv = 0. Inside the 2-phase region. 1967) Figure 2. ς -diagram of n-butane. the critical density can be calculated . The volume of the vapor phase at T = 130°C and p = 2.2 100 kg n-butane are filled up in a sealed 10 m3 container.95 m3 .4: V = 0.Phase Behavior 9 Example 2.5 demonstrates the T.4: mL ⁄ mv = DB ⁄ AD = 0.0125 m3/kg.4 using Eq.

889 E + 2 3.281 E + 2 2.240 E + 1 1.3 .624 E + 0 1.899 E + 1 2. the line must o be extrapolated to the critical temperature of methane.3 Use the CAILLETET-MATHIAS rule to evaluate the critical density of methane.7855 E + 02 It is known as the C AILLETET-MATHIAS rule that the averaged densities are situated on a straight line. To evaluate the critical density.108.713 E + 2 3.192 E + 2 4.148. The values of averaged densities have already been calculated.8493 E + 02 1. 3 C : ρc = 178.506 E + 2 3.128.925 E + 1 2. eq kg m-3 Averaged Density ρe q kg m-3 .0365 E + 02 1. Table 2.eq kg m-3 Vapor Density ρ v.118.10 Phase Behavior by extrapolation. 3 ρc .1: Temperature T oC Liquid Density ρ liq . On the basis of the averaged densities given above: tg α = –0.3 .654 kg ⁄ m .9185 E + 02 1. ρc = 161.138. The densities of the liquid and the vapor phase being in equilibrium have been measured at different temperatures (see table below).3 .55 .3 .798 E + 0 7.158. T c = – 82.1076 E + 02 2.9827 E + 02 1.26 x 0.546 kg ⁄ m . Example 2. 654 kg ⁄ m .3 4. 3 3 .045 E + 2 3.311 E + 0 2. The slope of a straight line can be evaluated by regression analysis.3 .

both the first and the second partial derivates of p = p(V) equal zero that 2 ∂p ∂ p  ------ = 0 =  --------- . the critical pressure pc.6: Critical pressure as a function of number of C-atoms in homologous series (after GYULAY. At the point of the inflection. 1967) . and the critical temperature Tc.4 illustrates the inflection point of the critical isotherm at the critical point.    ∂V   ∂ V2 (2. 6 e -Ol no mo 5 Benzene Toluene Cyclohexane Naphtalene es f in pc [MPa] 4 iC4 iC5 3 no m r al Pa ra ffin s 2 1 0 2 4 6 8 10 12 14 C-Atoms per Mole Figure 2.9) The state of the system at this point is characterized by the critical specific volume Vc.3 Critical State and Quantities of Corresponding States Figure 2.Phase Behavior 11 2.

8 shows a pr. k).. Because of thermal decomposition.10) Figure 2.7: Critical temperature as a function of numbers of C-atoms in homologous series (after GYULAY.. with a margin of error 1 to 2%. However. for homologous series .g. it is not possible to obtain experimentally information about the critical data in case of high molecular weight. Vr phase diagram which is valid for paraffins from methane (CH4) to hexane (C6H14). the homologous series of hydrocarbons show regularities which can be used for extrapolation. .. C 2H2n(n = 1. and olefines.results in a close relation between the p. CnH2n+2. 1967) Considering the critical data pc and Tc. the critical data of homologous compounds with longer carbon chains can be extrapolated though an increasing error has to be taken into consideration.e.7 refer to the homologous series of paraffins. ---Vc T T r = ----Tc (2.6 and Figure 2.2. The “principle of corresponding states” for chemically similar substances . pc V r = -V .12 Phase Behavior 500 Naphtalene 400 300 Toluene Benzene 200 100 0 -100 -200 0 2 4 6 8 10 12 14 no rm almo Pa no ra -O ffi lef n ine s s Tc [°C] C-Atoms per Mole Figure 2.. V. T-properties of pure hydrocarbons if the state variables are substituted by the so called “reduced quantities” which are p pr = ----. The experimental data in Figure 2.

The region of an undersaturated liquid state is positioned outside the bubble point surface (low temperature). In general. the vapor pressure curve of ethane appears. T. Outside of the dew point surface (high temperature). Analogous to the pure substance. In the plane z = 1. every change in state is described by the equation of state F(p. Just as in case of 1-component systems. the critical state of binary systems is defined as the state at which the intensive properties of the phases are no more distinguishable. This is demonstrated by the example of three additional z-planes.Considering a 2-component system. The volume is equivalent to the mole volume. the state of a dry gas is given. the composition is defined by “mole fractions”. The upper broken line marks the critical points of all compositions which are possible. The mole fraction is defined as the ratio between the number of moles of a certain component and the sum of moles of all components. V.4 Binary Systems If a systems consists of more than one component. z may be the mole fraction of one (lighter) component.1) mole fractions because the sum of the mole fractions always equal 1. whereas in the plane z = 0 the one of n-heptane appears. This curve divides the envelope surface into two parts: the bubble point surface and the dew point surface. z-coordinate system in Figure 2.9. its state is also a function of composition. the critical isotherms have an inflection point according to Eq. Covering all other z-planes. A system being composed of k components is defined by the specification of (k . an envelope surface encloses the 2-phase state. .Phase Behavior 13 2.9. z) = 0. T. 2. The phase behavior of the ethane/n-heptane system is graphically illustrated by the p.

0 3.reduced volume phase diagram of paraffins with low molecular weight (after GYULAY.0 4.75 0 0 100 200 1.0 .9: Phase equilibrium surface of the binary system ethane/n-heptane (from GYULAY.0 0 0.8: Combined reduced pressure . 1967) 10 nC7 Pressure [MPa] 7. 1967) .50 5.2 Tr 6= 1 .96 0.25 C2 0.0 z 2.5 0 0.8 0.0 4 1 C 1.5 0.0 2 pr 1.00 300 Temperature [°C] Figure 2.6 0 1.0 1.0 Vr Figure 2.98 0.14 Phase Behavior 1.4 1.94 0.0 2.

00 2 90. the more flat the envelope curve will be.22 4 50. If the mixture consists of two homologous compounds with quite different volatility (in consequence of quite different molecular weights). At a given pressure. the dew point temperature of the mixture at a given pressure is always lower than that of the pure heavier component.10 also shows that Tc of a mixture lies between the critical temperatures of the pure substances. the critical data curve envelopes a very extensive temperature and pressure region. Physically. This is due to the fact that lighter molecules partially transfer their higher kinetic energy to the heavier ones by collision.25 5 29. In contrast to this. 1938) Figure 2. On the other side. The smaller the difference between the molecular weights and thus between the volatility.Phase Behavior 15 1400 4 1200 3 5 1000 Pressure [psia] 2 800 1 Composition No. . T-plane. pc of the mixture may be obviously higher than the one of the pure substances.22 3 70.10: Pressure . it can be explained by the fact that the thermal motion of the lighter molecules is obstructed by the heavier ones which exhibit more inertia. the system maintains the state of a gas phase.14 8 3. Figure 2.91 6 9.9 into the p. For example. the maximum of the critical pressure of a methane/n-decane system equals 37 MPa. [Wt%] Ethane 1 100.25 10 n-Heptane 6 600 7 400 8 9 10 200 0 100 200 300 400 500 Temperature [°F] Figure 2.78 7 6.10 shows the projection of Figure 2.27 9 1. the bubble point temperature of the mixture is always higher than that of the pure lighter component.temperature phase diagram of the binary system ethane/n-heptane (from KAY. Consequently.

This maximum is given by the intersection point of the path EF with the dotted line connecting C and C”. At some point between D’ and D”. Similar phenomena occur when the temperature is changed by an isobaric process where the constant pressure is higher than pc but lower than the cricondenbar of the system. This process is called a “retrograde condensation”. the quality line "20%" represents the states in which 20% of the system account for the liquid phase.T-diagrams. Figure 2. a condensate drops out at the dew point D’. the volume of condensate (liquid) will be at its maximum.11 also shows an isothermal decrase along the path EF where E defines the sastem to be a dry gas. the dotted line connecting point C with point C’ marks the states of the system which exhibit the highest volume percentage of condensate dropout. In Figure 2. . In Figure 2.16 Phase Behavior Figure 2. the condensate will be vaporized again. If the decrease in pressure will be continued. the path surpasses the dew point line twice. Besides the critical point. the curve enveloping the 2-phase region possesses two additional characteristic points: • C’: the point of highest pressure on the curve that is called cricondenbare. The bubble point curve and the dew point curve represent 100% and 0% liquid.like in case of the path EF -. As on Figure 2. the condensated phase has been vaporized in total. All the quality lines (isochores) converge at the critical point.11. If the constant temperature is higher than Tc but lower than the cricondentherm . so called "quality lines" are shown on p. A quality line represents a certain mole percentage being liquid or vapor in the state of phase equilibrium. • C”: the point of highest temperature on the curve that is called cricondentherm. respectively.11. Consequently.11.11 illustrates the phase behavior of a certain ethane/n-heptane system. As soon as the dew point D” has been reached.

6 0.0 -50 0 50 100 150 200 250 300 Temperature [°C] Figure 2.9683)/n-heptane It depends on the composition of the system if the cricondenbar is located on the dew point curve or on the bubble point curve.4 BPL 0.12: Mole fraction(ethane) .8 β' 1.7 [MPa] δ' δ α β 1.0 D PL z 0. As far as the system ethane/n-heptane is concerned. 1967) .6 7.11: Pressure .4 2.2 5.2 0.temperature phase diagram of the binary system ethane (z = 0.Phase Behavior 17 7 C' E D' 6 C 5 Liquid C'' Pressure [MPa] 4 Liquid and Vapor Vapor 3 20% 10% D'' 2 A 1 F B 0 0 50 100 150 Temperature [°C] Figure 2.10 elucidates that the cricondenbar is located on the bubble point curve at low mole fractions of ethane. Figure 2.temperature diagram of the binary system ethane/n-heptane (from GYULAY.0 4. 0 0.

Again the relative masses of both phases can be determined by applying the principle of the lever (see Example 2.13: Mole fraction (ethane) .8 1. 1967) The design of the corresponding p. the phase behavior of ethane/n-heptane systems is graphically illustrated in the z. A sealed container (p = 2.2 0. Determining the phase composition.18 Phase Behavior In Figure 2. dew point state is reached at point δ . The composition of the infinitesimal small liquid phase in equilibrium with the vapor phase corresponds with point δ ’.9. .Pressure diagram of the binary system ethane/n-heptane (from GYULAY. the one of the vapor phase by point A”. The mole number of ethane in the liquid phase and in the vapor phase. An example may be the composition at the point A(T = 150 oC).47)/n-heptane mixture. T = 150oC) is filled up with 100 kg of a ethane(z = 0.86 MPa). The mixture α achieves bubble point β due to an isobaric (1.T-plane corresponding to another possible projection of the surface in Figure 2.6 0. respectively. Example 2.13).4. During further increase of temperature.4 MPa) heat supply.4).13 by using the principle of lever.12. z-diagram is also possible (see Figure 2. can be evaluated from Figure 2. 9 8 7 BP Pressure [MPa] 6 5 100 °C 3 A' 2 D P 200 °C 4 15 0 °C A A'' 1 0 0.4 0.0 z Figure 2. Point β symbolizes the composition of the liquid phase which is in equilibrium with an infinitesimal small vapor phase whose composition is symbolized by the point β’. The composition of the liquid phase is given by point A’.

MC7 100 n = 0.82 The mole number of ethane in the vapor phase can now be calculated by . n = nC2 + nC7. mC7.= ---------------.785 n C2 = ----------.604 kmole The composition of the vapor phase is given by point A” in Figure 2.= 0.47 = ---------------------------------------------------------------------( 100 – mC7 ) ⁄ 30 + mC7 ⁄ 100 mC7 = 79. ( 100 – mC7 ) ⁄ 30 z = 0.215kg mC2 = 100 – 79.12 n liq ----------. MC2 30 mC7 79.485kmole .792 kmole . can be calculated: mC2 20.693 + 0. The weight of ethane. the mole weights (MC2 = 30 kg/mole.mC7.785kg Now the total mole number of the system.458 -----------n vap A A′ where nliq:total mole number in the liquid phase nvap:total mole number in the vapor phase Thus the total mole number in the vapor phase results in n vap = 0.= ---------------.= A A″ = 1.12: z = 0.693 kmole . mC2.215 = 20.215 n C7 = ----------. MC7 = 100 kg/mole) are inserted into mC 2 ⁄ M C2 z = ---------------------------------------------------------mC 2 ⁄ M C2 + mC 7 ⁄ M C7 to evaluate the weight of n-heptane in the system. is given by mC2 = 100 .Phase Behavior 19 At first.792 = 1.= 0. From Figure 2.

Thus.485 – 0.5 Multi-Component Systems 2.881 kmole . The composition of the liquid phase is given by point A’ in Figure 2.23 × 0. The mole number of ethane in the liquid phase can now be calculated by nC2 . 1 b. 2 3 d.495kmole . the sum of the perpendicular distances from any point to each side of the diagram is a constant equal to length of any of the sides. Each corner of the triangle represents one pure component.12: z = 0. The total mole number in the liquid phase results in nliq = n – n vap = 1. liq = 0.20 Phase Behavior nC2 . 1 1 L2 L1 3 L3 Const. Figure 2.203 kmole .of a point . 1 C io at tR 2 n s ta t o on 1 of A 3 D B 2 c.82 × 0. Fraction Component1 2 3 2 a. the composition .23 .672 = 0.1 Ternary Phase Diagrams It is common to illustrate the phase behavior of 3-component systems at constant pressure and temperature in so called triangular diagrams.881 = 0.5.expressed in mole fractions .604 = 0. On the basis of the equilaterality of the triangle. v a p = z × n vap = 0.14: Properties of ternary diagrams 2.

14: • For mixtures along any line parallel to a side of the diagram.15: Typical features of a ternary phase diagram . Figure 2. • Mixtures lying on any line connecting a corner with the opposite side contain a constant ratio of the component at the ends of the side.12) L2 z 2 = ------. • Mixtures of any two compositions lie on a straight line connecting the two initial points on the ternary diagram. Only the amounts of the phases change as the overall composition changes.11) Several other useful properties of the triangular diagrams are also illustrated by Figure 2.15 shows the 2-phase region for chosen p and T. LT (2. --LT where LT = L1 + L2 + L3 .= DB ---------nB DA gives the mixing ratio leading to mixture D. 1 (2. the fraction of the component of the corner opposite to that side is constant. Any overall composition on a certain tie line gives the same liquid and vapor composition being in euqilibrium. Any mixture with an overall composition lying inside the binodal curve will split into a liquid and a vapor phase. The principle of the lever finds application again and nA --. The “tie lines” connect compositions of liquid and vapor phases in equilibrium.Phase Behavior 21 in the interior of the triangle is given by L z1 = ---1 . (2..13) Vapor Region Two Phase Region Li n e ve Binod al C ur Plait Point Critical Region Liquid Region 3 Ti e 2 Figure 2. LT L3 z 3 = -----.

on the basis of a hypothetical component.22 Phase Behavior The liquid and vapor portions of the binodal curve meet at the “plait point” which represents the critical composition. It is useful to comprise the two heavier components of a ternary system and to reduce this system to a fictitious binary system.5). in any case. pc and Tc. All envelope curves are characterized by the portion of propane in the hypothetical component (propane/n-pentane) which is given by z3 C = ---------------z3 + z5 (2.T-diagram of the methane/propane/n-pentane system. also result in a single-phase system if the straight line connecting the two initial compositions does not intersect the 2-phase region. the single-phase region is splitted into three sections. As pressure increases.16 illustrates the influence of pressure on the phase behavior of a certain ternary system at constant temperature. Mixtures of a composition being located in the critical region with another one being located in the liquid or vapor region will.14) In accordance to this aspect. Figure 2. the 2-phase region shrinks.17 illustrates a corresponding application by the respective p. can be determined for any mixture of the three components (see Example 2. By drawing the tangent in the plait point on the binodal curve. The mole-% of methane are specified along the outermost envelope curve. the critical state properties. . Figure 2.

16: Triangular diagrams for the methane/propane/n-pentane system at 160 oF(71 oC) (after D OURSON et al.6 0.2 T=160°F 0. Methane p=1000 [psia] 0.2 n-Pentane 0. 1943) . Figure 2.8 0.6 0.8 0. C= 0.2 C=0 .4 0.6 T=160°F 0.4 C= 0.2 0.8 0 1.. Methane p=1500 [psia] 0.0 C=  x c3  C =   x c3 + x c5  0.4 0.6 C=0.6 0.2 n-Pentane 0.8 Propane 0.0 C= 0.2 n-Pentane  x c3  C =   x c3 + x c5  c.4 0.6 0.2 b.0 C= 0.8 0.8 Propane 0.0 C= 0.4 0.4 C=0.2 0.4 C=0. 6 C= 0 . 8 1.6 C= 0.2 a.6 0.8 0.2 0.4 0.8 C=0.4 0.6 0.6 0.6 C= 0.2 Propane C= 0.2 T=160°F 0.8 0.4 0.4  x c3  C =   x c3 + x c5  C= 0.4 0.8 0.Phase Behavior 23 Methane p=500 [psia] 0.

3 + 0. -8 -n1 = --.5 n5 = ---------.17: Critical loci of methane/propane/n-pentane systems (from KATZ et al.36 .= 0.= 0. 0. At first.1 32.45 0.1[kg kmol-1 ]) and 32.= 0.3 + 0.5 z1 = -------------------------------------. 72. .= 0.= 0..2 [kg] propane (M = 44.3 z3 = -------------------------------------.2[kg kmol-1 ]).17.5kmole . 13. the mole numbers and the respective mole fractions must be calculated.5 The hydrocarbon mixture is composed of 8 [kg] methane (M = 16[kg kmol-1 ].5 + 0.8 =0 . 6 C= 0. 44.2 0.3kmole . 0.45kmole .45 0.4 .2 n3 = ---------.45 z5 = -------------------------------------. 0 50 C %] 40 30 20 10 0 H4 1000 500 C1 0 -200 C3 C5 -100 0 100 200 300 400 Temperature [°F] Figure 2.= 0.24 Phase Behavior 3000 2500  C3  C=  C 3 + nC 5  80 75 70 [ le Mo C= C Critical Pressure [psia] 0 C= 2000 C= 0.5 + 0. 16 13. 0. 1959) Example 2.5 [kg] n-pentane (M = 72.2 60 1500 C= 1.4 0 0.24 .5 + 0. The critical data of this mixture can be evaluated by use of Figure 2.3 + 0.45 The portion of propane in propane/n-pentane is given by the hypothetical component .

no liquid phase will drop out. For example it is possible to partition the paraffinic hydrocarbons into three pseudo-components which are • methane (C1) as the light component. For the classification of natural hydrocarbon systems.according to known data.24 + 0. Not considered in this classification are changes in composition during production.4 . small amounts of ethane.5°F = 128°C and p c = 1344 psia = 9.Phase Behavior 25 z3 0.19) remains in a single-phase (gaseous) state in the reservoir during production (line 1-2). • the lighter pseudo-component including ethane to hexane (C2-C6).17 at C = 0. Figure 2.22 by p. The application of the triangular diagram is not solely confined to ternary systems. some generalization makes a description of these complex systems possible . only poor information of complex natural hydrocarbon systems has been reported until now.18 represents a hydrocarbon system whose critical temperature is significantly lower than the reservoir temperature. the considered hydrocarbon mixture is called a “dry gas”. Anyway. which occurs in the reservoir adjacent to the production well the the course of production. condensate will drop out under separator conditions. • the heavier pseudo-component including heptane and higher hydrocarbons (C7+). Dry gases contain mainly methane.24 C = ---------------. In case of an isothermal pressure decrease (full line from point 1 to 2). Even in case of both pressure and temperature decrease (dotted line). • which state will be achieved at surface conditions (separator).= --------------------------. possibly propane and some hydrocarbons of higher molecular weights. . Consequently. T-phase diagrams.27MPa . it is essential to know • if the critical temperature is lower or higher than the reservoir temperature.= 0. Anyway.4 and 40 mole percent methane: T c = 262. the system remains in the single-phase (gaseous) state. A so called “wet gas” (see Figure 2. The phase behavior of several complex and natural hydrocarbon systems are demonstrated in Figure 2.36 From Figure 2. Nevertheless.18 to Figure 2. z3 + z5 0.

The pressure at point 3 corresponds to the state in which the condensed liquid phase reaches the maximum (in mole%). 1973) Liquid Critical Point 1 2 3 Sep. 100 75 50 25 5 0 2 Sep. 1 Pressure Critical Point Liquid 75 50 25 Sep. Consequently. Systems as shown in Figure 2. the dew point of the system is reached at point 2. condensate drops out inside the reservoir. the reservoir temperature is higher than the critical one but lower than the cricondentherm. The initial conditions given by Point 1 specifies the hydrocarbon mixture as a dry gas. the amount of condensate is larger than in case of wet gases. If the pressure will decrease adjacent to the production well during production.18: Phase diagram of a dry gas (from MCC AIN.19: Phase diagram of a wet gas (from MCCAIN. Gas Pressure Mole % Liq.26 Phase Behavior In case of the system shown in Figure 2. 100 75 50 25 10 5 0 Temperature Figure 2. Gas Temperature Figure 2. 1973) . 1973) Liquid Critical Point 1 Pressure Mole % Liq. In the separator.20 are called “gas condensates”.20. 2 Gas Temperature Figure 2.20: Phase diagram of a retrograde gas condensate (from MCC AIN.

23). The separator conditions are near the bubble point curve. 100 1 Undersaturated 2 Saturated Critical Point Pressure oin tL i ne 3 75 Bu bbl e -P 50 25 0 Po in t L in e Sep.are referred to as “high shrinkage oils”.21 .22 shows a “black oil” or “low shrinkage oil”. 100 Critical Point Pressure 75 3 Sep. Temperature Figure 2.Phase Behavior 27 Liquid 1 2 Mole % Liq.21: Phase diagram of a high-shrinkage crude oil (from MCCAIN. a further pressure drop in the reservoir will lead to point 3 and thus to an increased development of the vapor phase. 50 Gas 25 Temperature Figure 2.21). Since the bubble point curve will be reached by the decrease in pressure due to production.as characterized in Figure 2. At separator conditions. Consequently. If the pressure in the neighbourhood of the production well will decrease during production to point 2. the shrinkage of the oil due to gas liberation is less pronounced than in case "white oils" (see Figure 2. about 65% of the produced hydrocarbon mixture will exist as liquid phase if the reservoir is produced at bubble point conditions. the bubble point curve is reached and the state of the system is now considered “saturated”.22: Phase diagram of a low shrinkage crude oil (from MCCAIN. The reservoir temperature is below the critical temperature. If the hydrocarbon mixture in the reservoir is a 2-phase state under initial reservoir conditions. oil and gas phase can be considered apart from one another (see Figure 2. about 85 mole% of the produced hydrocarbon mixture is in the liquid phase at separator conditions. 1973) The so called “white oils” . from the initial pressure (point 1) to the pressure 2. 1973) Figure 2. The equilibrium conditions at the initial state of the system are given by the intersection point of the dew point curve of the gas cap and the bubble point curve of the oil zone. In accordance to this fact. The initial state is characterized by point 1 at which the state of the system can be regarded as “undersaturated” liquid.The D ew - Gas . Liquid Mole % Liq.

is given by point 4 in the p. By projecting the phase surface into the p.as dotted lines . The spatial curve.24. z-plane. .28 Phase Behavior gas cap shows a “retrograde” behavior. and the one of the vapor phase. If the state of the system is represented by point 1. z-diagrams. Furthermore. splits the surface into two parts which are the dew point surface and the bubble point surface. composition I represents the separator gas while composition IV represents the corresponding separator oil of the well stream.the locations of maximum retrograde condensation are drawn. T. Outside the envelope surface. Inside the 2-phase region of system II and III. the one of composition III to a white oil. information about the composition of the system will be obtained. Gas Cap Retrograde Condensating Gas Cap Pressure C Oil Zone Pressure C pi C pi C Oil Zone Temperature Temperature Figure 2. Again an envelope surface comprises the 2-phase region in dependence on the composition. Just as in case of binary systems (see Figure 2. the system is in a single-phase state. the phase behavior of two representative mixtures of I and IV are given by the compositions II and III. z-plane. the phase behavior of natural hydrocarbon mixtures can also be illustrated in p. which connects the critical points.23: Phase diagram pairs of gas cap and oil zone In Figure 2. if the intersection point is located on the dew point curve of the gas cap between the critical point and the cricondentherm. the equilibrium composition of the liquid (x1). (y1). The system of composition II corresponds to a gas-condensate system.9). isochores of the liquid phase and .

2 0 .8 0.6 L C'IV C C'' CIII 0 .0 4 x1 0 .Phase Behavior 29 IV III II I Pd C Pb CII T1 T2 T3 y1 0.0 C'' CII C' CI CI 0.4 0 . 1967) .2 CIV IV C IV C'' II GL' C'' CI I 0 .24: Phase equilibrium surface of oil/natural gas systems (from GYULAY.4 C' CIII CIII C' CII K 1.0 III GL C1 P IV G II T1 T2 T3 I III z Figure 2.

30 Phase Behavior .

1.V). the corresponding equation of state may be written as follows: 31 . the phases can be regarded as seperate thermodynamic systems.phase consists of k components. T-data is valid in case of any chemically homogeneous phase: V = V ( p. This aspect obviously makes a mathematical consideration of these problems necessary.impossible for multi-component systems.T). On the basis of Eq.3) (3. (p. In case of a 2-phase system: n = nl i q + nv a p and therefore nVt = n liq V liq + n vap V vap where Vt: Vliq: Vvap: n: nliq: nvap: mole volume of the system mole volume of the liquid phase mole volume of the vapor phase number of moles in the system number of moles in the liquid phase number of moles in the vapor phase. The application of such diagrams enable the determination of the respective volume for certain states defined by the corresponding pressure p and temperature T. However. T ) (3. If a phase . the following relation regarding the p. (p.2) If phase equilibrium is given.on top of that . planes.Chapter 3 Equations of State The preliminary chapter included graphical illustrations of equilibrium surfaces and their normal projections into the (p. V must be defined as the volume of one mole (intensive property). V. 2. a pure graphical application of the state functions is not very practical and . (3.may be the liquid .1) In this context.z) etc.

In this case. 3.6 in discussing Eq. V. To ) +  ∂V ( T – To ) + 1  ∂ V ( T – T o ) + . or x1 x2 . …x p ) = 0 (3. Therefore some procedure bit by bit is necessary. If Vi is defined as the mole volume of component i in the phase.8 may be truncated to . It is impossible to approximate the equilibrium surface for the entire (p.6) In various cases of even practical interests. the enthalpy of the system must be generally considered additive.. T.or just the same of Eq.5) where xi is defined as the mole fraction of the component i in the (liquid) phase. 3.-------- ∂T p 2  ∂T 2  p (3. V.7) If Eq. the mole volume of the phase will result in V = ∑ xi Vi i=1 k (3. 3.into a TAYLOR-Series leads to 2  ∂V ( p – p ) + 1  ∂ V  ( p – p )2 + x V ( p. no thermal effect will take place during the mixing procedure. T ) = V ( p o .7 is valid.8) By assuming that the higher derivations are neglectable. i=1 k (3.32 Equations of State F ( p. T o ) +  ------ ...4) V = V ( p. x1 x2 . 3. T.5 at constant composition .. T) = V ( p o. 3.1 . Eq. xk)-space by one single equation of state. Put into other words: The mixing energy will be zero. the so called “constraint equation” is valid: ∑ xi = 1 . …xk ) (3.5. Considering the mole fractions. 3.1 Change of State at Low Compressibility The expansion of Eq. ------ -. This means that the enthalpy of the system is equal to the sum of the enthalpies of the single components. 3. it is obvious that V is a function of 2 + (k .1) = k + 1 variables. Regarding Eq. T.. --------- o o ∂p T 2  ∂p 2  T  2  2 V ( p. a multi-component phase behaves as an ideal mixture and the volumes are strictly additive.

only the cubic equations of state.9) (3. there may exist equations which approximate the measured values more accurately. T) = V o [ 1 – κ ( p – p o ) + α ( T – T o )] where κ : isothermal compressibility α : cubic expansion coefficient (3.Equations of State 33 1 ------1.10) Eq. ∂V ( p – po ) + -----.5. T o ) .4 and Eq. 2. In doing so. 3.10 can at best be applied for fluids in a 1-phase state. ∂V ( T – To ) V o  ∂p T V o  ∂T p where V o = V (p o.T ) = V o 1 + --.11) DER WAALS) forces are .9 may be transformed to V ( p. Anyway. the proceedings of generalization in derivating this equation of state will be elucidated.2 Equation of State of Perfect and Real Gases A gas is defined as perfect.12) (3. will be discussed in this textbook. particularly the PENG-ROBINSON equation. Eq. Of course. Experience and practice have shown that the cubic equations of state are most sufficient and beneficial for calculating the state of gases and of 2-phase systems. Then for a molar system: pV = RT where R is defined as “universal gas constant” and R = 8. 3. Considering the Eq.31434 J/mole K The compressibility factor Z is defined as pV Z = ------RT (3.-----V (p . 3. 2. the constants included in these equations are not always given. Therefore. if the intermolecular (VAN neglectable.

= 55 ×10 m .12: pc V c Zc = ----------.63 –5 55 ×10 –5 . V ideal respectively. Tc and V c into Eq.= 0. and specific volume Vc and may be determined experimentally for 1-component systems. Z is a state variable and depends on the pressure. After the volume reduction. A cylinder with volume of 0. can be evaluated by substituting the critical data pc. For real gases.5o C.79 MPa.13) Example 3.79 MPa. the pressure has been recorded as 13. temperature Tc. 3. p2 V 2 = 75 × 10 –3 –5 3 101 ×10 -----------------------. RT c (3. the factor Z equals 1.00035 m3 : 35 ×10 Z = -------------------. Zc. 3. If the gas would show an ideal behavior at 90oC and 13. Then the volume of the cylinder has been isothermally reduced to 0.075 m3 has been filled with a gas under atmospheric pressure and at a temperature of 90.13 and considering the measured volume amounting to 0. the temperature and the composition.. The critical point is defined by the pressure pc.00035 m3 (volume reduction in consequence of mercury insertion).34 Equations of State or V actual Z = ----------------.14) (3... 2 1379 ×10 –3 By use of Eq.1 Evaluation of a Z-factor from laboratory data. The critical compressibility factor. For ideal gases. the specific volume could be calculated by P1V1 = P2V2 and so p1 V 2 = V 1 ----.

3 Cubic Equations of State If the pressure of any gaseous system is low.16 cannot predict the real behavior of the system during the vaporization.16) Figure 3.to describe the continuity from gaseous to the liquid state: a  p + ------ ( V – b ) = RT .1 illustrates the VAN DER WAALS isotherms in the vicinity of the critical point. Obviously. the ideal gas equation remains sufficient to describe the volumetric behavior.15 may also be written in the following form: RT a p = ----------------. and P ENG and R OBINSON (1976). ( V – b) V2 (3. T1 = co ns t. VAN DER WAALS deducted the first equation of state which is able . The dotted section of the isotherms represents the the data which are predicted by using the VAN DER WAALS equation. SOAVE (1972).  2 V (3.. The cubic equations of state. The real behavior is shown by the straight full line BD inside the 2-phase region. 3.1: The VAN DER WAALS isotherms near the critical point .15) a and b are substance specific constants. have achieved much better results. The term a/V2 regards the pressure reduction in consequence of intermolecular attraction.Equations of State 35 3. b can be interpreted as the inherent volume of the molecules which is not available for the thermal motion of the molecules. respectively condensation. B Actual Path D Specific Volume Figure 3. Eq. 3.– -----.up to a certain degree . Pressure Critical Point Calculated from Van der Waals Equation T>T 1 2 T 2= co ns t. In the year 1873. Eq. which have been formulated by REDLICH and KWONG (1949).

– ------------. The original R EDLICH-KWONG equation included .19) – 0.22 results in (3.18) (3..21 and Eq.– ---------.23 into Eq.19 and Eq.23) (3.– b  V – -----.5 (3. Anyway. 3.= 0 .20) (3. pc 3V a c bRT c 2 2 -= --. f( T ) = T By multiplication of Eq.21) (3.– b . p  p p p At the critical point ( V = V c ) : 2 3 3 3 2 ( V – Vc ) = V – 3V c V + 3V V – V = 0 . pc pc c (3.– --------------------(V – b ) V (V + b ) (3. 3. Substitution of Eq.17) where b again is a substance specific constant. 3. 3.17 with V( V + b) ( V – b) ⁄ p and after arrangement of V corresponding to its order of power: a bRT ab 3 2 2 ------V – RT V +  . c c The comparison of Eq.36 Equations of State The R EDLICH-KWONG equation is given as RT a p = ----------------.22) ac b 3 V c = -------pc where ac = a′f ( T c ) . It is useful to write the parameter a as follows: a = a′f ( T ) where a′ is constant.24) .20 leads to RT c 3V 3 = --------. 3. 3. a is now a function of the temperature.

= ac α .42748.---- p  p  p p p  p  b 2 2 1⁄ 3 (3. Furthermore.21 and Eq.29) (3.19 and the arrangement of Z corresponding to its order of power results in 2  ZRT 3 – RT  ZRT +  a – bRT – b2 ZRT – ab = 0 . 3. The substitution of V = ZRT ⁄ p into the Eq.Equations of State 37 3 Vc 2 2 --3V = ---. 3pc pc Inserting Eq. if T = T c . 3.--------------------. in 2 3 3 3 2 b + 3b V c + 3bV c + V c = 2V c .27) (3. 3 3 ( b + Vc ) = 2V c .– 3bV c – b c b or after rearrangement.28) – 1 )V c . 3.08664.30) (3.32) Therefore: .21 can be combined to 2 – 1 RTc RT c b = -----------------------------.. 3.24: f(T) a = ac -----------. Ω a = 0.-------------------------. 3.= Ω b ----------------1⁄ 3 pc 9( 2 – 1 )pc The constants Ω b and Ω a have the following numerical values: Ω b = 0.29: ( RT c ) ( RT c ) a c = ---------------------------------. b = (2 1⁄3 (3.= Ω b --------.31) (3. f ( Tc ) It is obvious that α = 1 . 3. 3. From Eq.23 into Eq.25) (3.26) (3.18 and Eq. Eq. -.28 and Eq. 3.

2748  ------ α  T 2  r   p  (3.29 and Eq. 3.5 (3.36) (3.30 into Eq.34) r A = 4. . Since the Redlich-Kwong cubic equations of state consider these two substance-specific parameters..= 0 2 2 RT 2 ( RT ) ( RT ) ( RT ) RT 3 2 2 2 (3. Anyway.34 includes only two parameters which are pc and Tc. Tr In the original REDLICH-KWONG equation: f( T ) = T and so α = Tr – 0.38) (3.35 and Eq. Please note that the ideal gas equation contains no substance-specific parameters. 3.40) Eq. (3.– -------------. RT The substitution of Eq.– ------.33) or Z – Z + ( A – B – B )Z – AB = 0 where ap A = -------------2 ( RT) and bp B = ------.37) and pr -B = 0.5 – 0.Z – -------------.36 leads to (3. 3.------.35) 3 2 2 (3. This made the introduction of a third factor necessary. 3.8664 --.38 Equations of State bp b p ap ap bp Z – Z + -------------.39) .. 3. it has improved the calculation of PVT-properties in a fundamental way. the increasing yield of experimental data has more and more indicated that the behavior of many liquids with a non-spherical molecule structure deviates greatly from the predicted one.

The so called “acentric factor”..45) (3.41) = 1 + ( 0.44) RT c V c = 0. The equation of state from S OAVE (1972) only differs from the REDLICH-KWONG equation with respect to the definition of the factor α 0.– -------------.5 (3.47) (3. 3.43) (3.46) (3. the following terms and equations can be obtained: 3 RT 2 a 2bRT 2 a 2 . pc a = ac α .42) The weakness of all these equations ranging from the original R EDLICH-KWONG equation to all its modifications (including the SOAVE equation) is the fact of an universal unrealistic Zc factor of 1 ⁄ 3 .– 3b  V – b . has become the one with greatest acknowledgement: s ω = –  lgp + 1 at T r = 0. pc RTc b = 0.Equations of State 39 Beginning with the year 1951 (MEISSNER and SEFERIAN).– -----------------------------------------------V – b V ( V + b )+ b( V – b ) where a is given by Eq. several proposals for a third parameter have been made.48 + 1.7   r s s where p r = p ⁄ p c is the reduced boiling point pressure... 2 0. the prediction of liquid density is combined with large errors.48) . -- p  b   p p  p (3.---V –  ------. ω . Moreover.07796 --------. pc 2 (3.5 (3.– RT b – b  = 0 . Just as in case of the R EDLICH-KWONG equation.18 and b is further a substance specific constant.176ω )( 1 – T r ) . Improved approximation has been achieved with the PENG-ROBINSON equation: RT a p = -----------.457235 ----------------.307 --------.574ω – 0. ( RTc ) a c = 0.– b V +  -.

Anway.21. 3.26992ω  1 – (3. and Eq. The extension for multi-component systems requires the calculation of the respective data of the pure components and mixing rules in order to get the parameters of the mixture. 3. the calculated fluid densities are much more accurate than the ones calculated by the equations of state previously discussed. 3. 3.42 on the other side are obvious.30 and Eq. Eq. The mixing rule for the parameter b.34 to Eq.43 have been established for pure substances.17. R EDLICH-KWONG: a = i=1 ∑ xi a i k 0.16. 3. is universally defined as an arithmetic average by using b = i=1 ∑ xi bi . (3.29 to Eq.307 which is much better than 1 ⁄ 3 but still far away from reality. 3. The universal critical Z-factor of the PENG-ROBINSON equation results in 0. different mixing rules exist and are presented below.07796 ----. k (3. Eq. 3. 3.  T 2  r 3 2 2 2 3 (3.31. Tr α 0.19. S OAVE and PENG-ROBINSON.44 to Eq. Eq. A = 0.40 into Eq. 3. Eq. Eq.  + 1.38.3676  (3.5 2 . which is included in the equations of R EDLICH-KWONG.40 Equations of State Z – ( 1 – B )Z + ( A – 2B – 3B )Z – – AB – B – B = 0. 3. 3.52) Analogies between Eq.49) (3.31.. 3. 3. . and Eq.54) where ai can be calculated on the basis of the critical data of each component by using Eq.5 = 1 +  0.51) 4  1 T r / 2 .457325 ⋅    pr   ------ α . 3. 3.52 on the one side and between Eq.50) pr B = 0.53) For the temperature-dependent coefficient a .54226ω – 0.

The composition of the system. 3. . The mixing rules used by SOAVE and PENG-ROBINSON consider the binary interaction between the molecules of the components i and j.Equations of State 41 SOAVE: a = i=1 ∑ j ∑1 xix j ( a i aj ) = k k 0.52. 3. Values of the binary interaction coefficients must be obtained by fitting the equation of state to gas-liquid equilibria data for each binary mixture. Eq. 3. 3.55 and Eq.56 reduce to the form of Eq.54 if all binary interaction coefficients are zero. 3. 3. 3. respectively. They have different values for each binary pair and also take on different values for each equation of state.42. PENG-ROBINSON: a = ∑ ∑ xix j ( a i aj ) i = 1j = 1 k k 0. Another possibility of obtaining this coefficient .30 and Eq.00025 m3 contains 0.5 ( 1 – kij ) (3. Example 3.25 mole of gas.2 Calculation of the pressure by use of the PENG-ROBINSON equation of state.56.is by mean of matching the phase behavior of multi-component systems. In Eq. A laboratory cell at temperature of 100 o C with volume of 0.if no data are available .55) where ai and aj. the critical data and the acentric factors of the components are tabled below. 3.5 ( 1 – kij ) (3. Obviously. can be evaluated by Eq.48 and Eq.56) where ai is defined by Eq.55 and Eq. the terms k ij are binary interaction coefficients which are assumed to be independent of pressure and temperature. 3.

2494 . 3. Correspondingly to the calculations for methane.5 = 0.1928 MPa C1 C2 n .75 0.C4 190. the parameters of ethane and n-butane were evaluated.8018 .80 First.1999 . 3.05 Critical Data Tc.= 0.0908 0.47: α1 = 0.6 4.42 Equations of State Component Composition yi 0.= 1. the parameters b and a must be calculated for each component.88 425.= 2677 × 10 . . and from Eq.6 α 0.8018 = 0. For methane from Eq.i pc.1 = ---------------. 8 463 × 10 – 6 ( 831434 6 2 from Eq.2 3.1 × 1907 × 10 ) = 457235 × 10 ---------------------------------------------------------.4 4. 3.0115 and the reduced temperature of methane 273.52 by inserting the acentric factor for methane ω1 = 0.958 : 190.2494 × 0. They are tabled below. 3.15 Tr .46: –5 × 1907 × 10 b1 = 77796 × 10 --------------------------------------------------.60 305. i oK Acentric Factor ωi 0.48: ac .20 0. 8 463 × 10 – 6 831434 6 from Eq.0115 0.8954 1 and thus α1 = 0.

05 × 7.8260)0. must be known.245 E-5 ac.25 T = 373. 3.20(0.048 E-5 7.75 x 0.4904 E-2 0.8260)0.6045)0.75 x 0.56.5(1 .05 x 0.31434 J ⁄ K × mole .8260)0. 245E – 5 b = 3.004904) + + 0.05(0.53: b = 0.5(1 .05 x 0.5(1 .014640) + + 0.C4 bi 2.6045)0.i 0.75 x 0.677 E-5 4.0010 1.75 × 2.43 and by 0.1999)0.0000 E+0 0.= 0.1999)0.8018 1.0. 180E – 5 .0.5(1 .0. the interactive coefficients.05(1. They are given below: C1 C1 C2 n .1999)0.6045)0.75(0.6045 x 0.5(1 .2130 ai 0. 048E – 5 + 0.20 x 0.0000 E+0 0.4904 E-2 n .15 K .6045 x 0.5(1 .20 × 4.0000 E+0 a = 0.20(1.0.8260 The parameter b can be calculated by use of Eq.0.004904) + + 0.75(1.Equations of State 43 Component C1 C2 n .014640) + + 0.8260 x 1. R = 8.1464 E-1 0.0.1999 x 0.75(0. 3.20 x 0.20 x 0.3108 Now the pressure inserting can be calculated through Eq. To calculate the coefficient a by use of Eq.001 m ⁄ mole .8260 x 0.2648 E-2 0.1999 x 1.05(0.5050 αi 0.1464 E-1 C2 0. 3.000000) + + 0.5(1 . Kij.C4 0.5(1 .05 x 0.5(1 .C4 0.2648 E-2 0.6042 1.8260 x 0.000000) = 0.2494 0.002648) + + 0.000000) + + 0.0.6045 x 1.0.002648) + + 0.1999 0.67E – 5 + 0.20(0.0. 0.1999 x 0.6045 1.00025 3 V = --------------------.

457235 -----------------------------------------------.8018 . To solve this cubic equation of . 3. 3.6 ) ac = 0.47: a = 0.63 × 10 From Eq. Now the mole volume of methane at the given state variables has to be evaluated through Eq.31434 × 190.662 ×10 .44. First.248 .8018 = 0.6 K .31434 × 190. -----------------------------------------6 4.18 ×10 and –5 a = 0. and from Eq. 190. 12 Tr . pc = 4.26 MPa.328 MPa Example 3.44 Equations of State b = 3.0115 .0 kg methane (MC1 = 16 kg k/mole is hold at temperature of 305. 1 = -----------------. 6 α 0.077796 8.6 = 2. The critical data and the acentric factor of methane are known as Tc = 190. From Eq.8954 and thus α = 0. 3.5 = 0.52 by inserting the acentric factor and the reduced temperature of the system 2 305. 3.12 K and at pressure of 9. ω = 0.= 1.= 0.46: –5 b = 0.3 Calculation of the density of methane by use of the PENG-ROBINSON equation of state.48: ( 8. 601 . 4 463 × 10 From Eq. The calculation procedure results in p = 2. 2.63 MPa . 3.248 × 0.3108 . the parameters a and b have to be calculated.1988 .

p  p The calculation of r. the CARDAN equation is applied for x + rx + sx + t = 0 where 3 2 RT r = –  ------..– b . 563×10 –8 . .+ t . t = –0.  p  a 2bRT 2 s =  . 27 3 The calculation of p and q result in 3 p = – 1. 3 the formula x + rx + sx + t = 0 is reduced to 3 2 y + py + q = 0 where 3 3s – r p = ---------------3 and 2 2r rs q = -------. –8 .– -------------. s and t result in r = – 0.b – b  .4755 ×10 and –3 .– --.Equations of State 45 state.– 3b  p  p and RT 2 t = – b  a – ------. By substitution of r x = y–. 2473 ×10 s = 0. 3585 ×10 – 12 .

2 q –. the mole volume can be calculated through r x = y–. 3 3 m --. 089×10 – 12 . –5 –4 . 3350 ×10 and –4 m ⁄ mole .46 Equations of State and q = – 1.+ D .522 kg ⁄ m . -  2 3 and thus D = 1.: 3 u = 9.= 68.= --.– D 2 v = and 3 y = u + v.3272 ×10 y = 1.M ρ = ---. nV V . Based on the relationships u = 3 q –.7800 ×10 v = 5.53776 ×10 – 25 . The discriminant D is defined as 3 q 2 D =  p +  .5107 ×10 and thus –5 . V = 2.

3. 3. then: Bp Cp Z = 1 + ----------.+ -----------------.58) The second virial coefficient B can be shown to be the slope of isotherms on a plot of Z = Z ( p ) at p = 0 . When p approaches zero.58 may be neglected so that B = ZRT lim  Z – 1  .+ … ZRT ( ZRT )2 2 (3.+ --.Equations of State 47 3.60) Eq.57) The virial coefficients B. p → 0 ∂ p T (3.57 is applied for the calculation of gaseous states (e. In contrast. The BOYLE temperature is defined as the temperature above which Z-values become greater than 1 for all pressures. the third term in Eq. 3.57 according to the gas law by Z RT/p.+ -----. However. Sometimes Eq.4 Virial Equation of State All cubic equations of state mentioned above are more or less empirical or at best semi-theoretical. C etc. The pressure is expressed as 1 B C --p = RT  -.59) Since also Z becomes zero as p goes to zero. are solely a function of temperature. If V is substituted into Eq. However. de l’Hospital’s rule can be applied resulting at constant T in ∂Z B = RT lim  ------ . .… . 3. natural gas) by considering only the first two or three terms (depending on the availability of the virial coefficients). -----------p → 0 p  (3. virial coefficient vanishes at the BOYLE-temperature.g. 2 V 3  V V (3.60 implies that the 2. the form of an infinite series concerning the volume is not very appropriate for practical application. the virial equation of state has some theoretical background in statistical mechanics. they are obviously qualified for practical application.

48

Equations of State

Chapter 4

Calculation of Phase Equilibria
4.1 Mixtures
4.1.1 Definitions

In case of thermodynamical phase equilibrium of a multi-component system, all phases are - in physical terms - homogeneous mixtures. Vaporized compounds are always mixable in all relations, independent of their chemical characteristics. Therefore, only one vapor phase exists. Compounds existing in a liquid phase are only mixable in all relations if they are chemically similar. Since hydrocarbons - especially whose which belong to the same homologous series - exhibit such chemical conformity, their mixtures will be physically homogeneous, independent of the composition of the mixture. The composition of the system and its phases are specified by mole fractions. The mole fraction of the component i is given by the ratio between its mole number, ni, and the total mole number in the system (in the respective phase), n. The sum of all mole fractions equals 1. In the following, the different mole fractions are marked by zi: mole fraction of the component i in the total system xi: mole fraction of the component i in the liquid phase yi: mole fraction of the component i in the vapor phase, so that in the system in total
n i z = ----------------------- , i k n i =1 i

(4.1)

49

50

Calculation of Phase Equilibria

  n  i  x =  ----------------------  , i k   n i  i=1 liq

(4.2)

yi =

  n  i  .  ----------------------- k   n i  i=1 vap

(4.3)

If a multi-component system is in the state of thermodynamic phase equilibrium, the distribution of the component i among the vapor and the liquid phase is characterized by its distribution coefficient, the so called “K-factor”: yi K i = -- . xi (4.4)

The numerical value of Ki depends on the state of the system. Accordingly, the composition of the phases will change at any variation in state.

4.1.2

K-factors

Ideal liquid mixtures have the property of being met by RAOULT’s law so that xi =
p i -----o p i

(4.5)

where pi: partial pressure of the component i p o : tension of component i as a single-component system i xi : mole fraction of component i in the liquid phase.

The total pressure of the system adds up to p =

i= 1

∑ pi
k

=

i= 1

∑ xipoi
k

(4.6)

If a vapor phase exhibits perfect behavior, DALTON’s law can be applied and pi yi = ---p The combination of Eq. 4.5 and Eq. 4.7 results in (4.7)

where fi(liq) : fugacity of component i in a liquid mixture fi(vap) : fugacity of component i in a vapor mixture f f o i ( liq ) o i ( vap ) : fugacity of pure component i as liquid : fugacity of pure component i as vapor.. -xi f ° i (vap ) (4. it becomes evident that some other means must be obtained for the determination of K-factors if they are to be used for hydrocarbon mixtures. the definition of the K-factor is given by f° yi i (liq ) ′ K i ( id ) = -. In practice.11) (4. .= -----------------. If the phases are in equilibrium at given conditions and thus fi ( v a p ) = f i( liq) . the extrapolation of the vapor pressure curves to higher temperatures than Tc does not yield satisfactory results in most cases. Since Tc = -190. the so called “activity coefficient”.1 in terms of f/p.= xi o p i -----p (4.37 K for methane.10) respectively.12) A generalized fugacity chart for natural gases based on reduced temperatures and pressures is given in Figure 4. These fugacities serve to provide ideal equilibrium constants as defined by Eq.12. certain limitations are placed on the definition of the K-factor given above. 4. A first approach originated from DODGE and NEWTON (1937) who have shown that at any given p and T fi ( liq ) = x i f o and f i ( v a p ) = yi f o i ( vap) i ( liq ) (4.8) Anyway.Calculation of Phase Equilibria 51 K i( id ) yi -= -.9) (4. It is evident that any pure coomponent has a vapor pressure only up to its critical temperature.8 can be corrected and extrapolated to cover a wider range of conditions. On the basis of the thermodynamic concept of fugacity. 4. the vapor pressures in Eq.

the extrapolation of the fugacities on the basis of Eq.10 .2: Ideal and real K-factors of n-butane at 60[oC] In case of multi-component hydrocarbon systems.1: Fugacity of natural gases (from BROWN.1 Kid 5 4 3 2 K'id 0. attraction forces between the molecules .1 2 3 4 5 1 2 3 4 5 10 2 3 Pressure [MPa] Figure 4. 4. Anyway.covers a wider range of state conditions than the vapor pressure relations.though also limited by the critical state conditions .0 5 4 3 log K K 2 Temperature=60 °C 0. 1945) 4 3 2 n-Butane 1.2 shows that the K i( id ) -values are not sufficiently accurate to describe phase equilibria at high pressures.52 Calculation of Phase Equilibria Figure 4. ′ Figure 4.01 0.

∫  ------.5 .k) in a multi-component phase P .13) ∑ Z + 2.414 B  (4. NGHIEM..5 0.10) result in the calculation of K-factors without considering the composition. 3). As both RAOULT’s and DALTON’s laws and the fugacity relations given above (see Eq.– p dV  p RT ∞  V but also other terms which have been defined by PENG-ROBINSON (see Chap.414 B ln  ----------------------------.– i a b (4.14) (4.Calculation of Phase Equilibria 53 affect the vapor-liquid equilibria at higher p and T. 2. 4.in analogy of the “activity coefficient”.it will be discussed through Example 5. f/p .  Z – 0. K-factors can be evaluated in satisfactory accordance with the measured data by the application of the “successive solution method”. It is based on the definition of the “fugacity coefficient” of the component i (i = 1.15) . After introducing not only the PENG-ROBINSON equation of state RT a p = -----------.. P p where p is used for the total pressure of the system. 4.16) 0.. This method .11).P φi.5 k xa 2 b i = 1 i ij ----------------------------------------.P = --------------xi . the definition of the K-factor is given by φ i( liq) K i = ---------------φi ( vap ) (4.consists in an iterative process.9 and Eq.as fi.– ------------------------------------------------V – b V ( V + b ) + b ( V – b) in f 1 V RT ln .= Z – 1 – ln Z + ------. AZIZ and LI (1983) have calculated fugacity coefficients by use of b A ln φi = --i ( Z – 1 ) – ln ( Z – B ) – --------------b 2 2B where a ij is defined by a ij = ( 1 – K ij ) ( a i aj ) . it is necessary to find another solution for the determination of the K-factors.= 1 f i( liq) (see Eq. 4. If the phases are in equilibrium at given conditions and thus f i( vap ) --------------.

4.54 Calculation of Phase Equilibria (see Eq. temperature and composition is in a liquid-vapor phase equilibrium. where φi ( liq ) : fugacity coefficient of component i in the liquid phase φi ( v a p ) : fugacity coefficient of component i in the vapor phase. 4. the following must be valid (on the basis of an overall and particular material balance.. 4.18) total mole number of the system mole number of the liquid phase mole number of the vapor phase mole fraction of component i in the liquid phase mole fraction of component i in the vapor phase mole fraction of component i in the system After substitution of Eq.19 and Eq.2 Composition of Phases in Equilibrium 4. 4..2. nliq + n vap K i zi n yi = -----------------------------------nliq ⁄ K i + n vap (4.k) (Eq.19) (4.20) If the numerical value for nliq and has nvap have been chosen correctly. k ) (4.18 and after rearrangement... the following is valid: zin xi = -------------------------------.17) (4.1 Definitions If the system at a given pressure. 4.13).20) result in . 4. the calculations of xi and yi(i = 1. where n: n liq : n vap : xi : yi : zi : ( i = 1. respectively): n = nl i q + n v a p zi n = x i nliq + y i nvap . ….8 into Eq.

21 becomes ∑ xi i=1 k = i = 1 liq i ∑ --------------------------------------------) n + K (1 – n i liq k z = 1.1 Determination of the gas and liquid phase compositions at p = 14 MPa and T = 92 o C through an iterative calculation procedure. T and system composition. If n = 1. ∑ xi < 1 . FIRST ITERATION:nliq = 0.22 for iterative calculation purposes which result in the evaluation of the phase compositions at given p.210 .22) Example 5.23) Eq. 4. (4.790 nvap = 0.1 will demonstrate that it is possible to use the Eq. 4. Eq. Example 4. The actual composition of the hydrocarbon system is tabled below. 4. 4.23 elucidates that nliq must be chosen larger in the following iteration if nliq must be chosen smaller in the reverse case of ∑ xi > 1 .21) and i=1 ∑ y i = ∑ ------------------------------------p = 1 .Calculation of Phase Equilibria 55 i=1 ∑ x i= ∑ --------------------------------i = 1 n liq + n vap K i=1 k k zin (4. 4.21 and Eq. n liq ⁄ K i + n v a i=1 k k zi n (4.17 gives n vap = 1 – n liq and Eq.

17 0.876 0.031 0.5055 0.9963 C1 C2 C3 C4 C5 C6 C7+ Σ ∑ xi < 1 .246 0.0174 0.051 0.003 1.143 0.224 THIRD ITERATION: nliq = nvap = 0.3173 0.0290 0.0405 0.086 0.0000 2.0432 0.0434 0.778 0.56 Calculation of Phase Equilibria nliq must be chosen smaller in the following iteration because of Component K-Factor System zi x i = -------------------------------Composition n liq + nvap K i zi Ki nvapKi nliq +nvapKi 0.0207 0.933 0.0353 0.222 .85 1.68 0. SECOND ITERATION: nliq = nvap = 0.0324 0.036 0.0417 0.841 0.776 0.598 0.41 0.0284 0.4011 1.793 0.24 0.388 1.15 0.821 0.01 0.4404 0.

(4. pb.1.0436 0.3114 0.2 Determination of the bubble point pressure of a crude oil.0006 zi x i = -------------------------------n liq + n vap K i 0.0296 0.5145 1. n liq = n and x i= zi .5132 0. The composition of the system is given in Example 4.24) Example 4.0436 0.0487 0. linear interpolation can be applied for acceleration purposes.0134 0.0054 0. Naturally. .8896 0.0416 0.= n liq ⁄ K i + n vap ∑ i= 1 k Ki zi = ∑ Ki x i = i=1 k 1.3123 0.Calculation of Phase Equilibria 57 Iteration II Component Iteration III Iteration III zi x i = -------------------------------n liq + nvap K i 0. for a given hydrocarbon composition and for a fixed temperature. Eq.0000 C1 C2 C3 C4 C5 C6 C7+ Σ At the bubble point of the system.0327 0.0357 0. Example 4. In the vicinity of the bubble point.9999 zi yi = --------------------------------nliq ⁄ K i +nv a p 0. . Kixi are in a linear relation to pressure.22 then becomes ∑ i=1 k zi n --------------------------------------.0209 0. Ki = Ki(p) must be known.0082 1.0416 0. 4.0051 0.0358 0. Therefore.2 illustrates the iterative process of calculating the bubble point pressure.0327 0.0210 0. the amount of vapor is infinitesimal small and thus n vap = 0.

0125 0.0290 0. 4. Therefore.0094 0. the evaluation of the K-factor for heavier components is crucial.0138 3. Assumption pb = 23.020 0.0069 0.0056 0.0441 0. From Equation 4.0284 0.4404 0.678 0. However.25) In analogy of the iterative process demonstrated by Example 4.0445 1.026 1.0052 0.025 0.323 0.0100 0. This notion is mostly used to check the K-factors which have been evaluated for certain state conditions.448 0.9469 2.0104 0.1500 1.0275 0. n vap = n and yi = z i . the calculation of the dew point pressure of hydrocarbon systems by using Eq.4400 0.2 MPa yi = Kizi 0.5 MPa yi = Ki zi Ki 0.9072 2.0432 0.060 0.0235 By interpolation: pb = 23 MPa At the dew point of the system.0174 0.21: ∑ i=1 k zin -------------------------------.6720 0.0300 0. (4.4701 1. Assumption pb = 21 MPa yi = Ki zi Ki 0.3000 0.= Ki i=1 ∑ -----i = K i k y 1.0443 1.239 0.0127 0.0272 0.0519 2. Assumption pb = 22.020 0. the amount of liquid phase is infinitesimal small and thus nliq = 0.250 1. .9969 C1 C2 C3 C4 C5 C6 C7+ Σ 2.450 0.25 by summing up the y i ⁄ K i values and by variating the pressure as long as Σyi ⁄ K i = 1 .0128 0.680 0.316 0. it is theoretically possible to evaluate the dew point pressure by use of Eq.0067 0.= n liq + n vap K i ∑ i=1 k zi ----.0055 0.0062 0. 4.25 may be too inaccurate.230 0.2150 0.0000 Ki 1.8896 0.2 (evaluation of pb). zi 0.0405 0.0274 0.58 Calculation of Phase Equilibria Component System Compos.

a condensate will drop out below the dew point. Correspondingly.3: Flash and differential vaporization In evaluating hydrocarbon systems. At the same time. the vapor phase appears below bubble point pressure.as soon as the previous phase equilibrium is achieved (open system). two characteristic processes are of importance. The portions of the phases and their compositions in various states of thermodynamic equilibrium (p and V are variables) are to be evaluated. the oil phase shrinks at decreasing pressure. the liquid volume diminishes.in most cases the vapor phase . In case of an undersaturated liquid system. In case of a flash process. These are illustrated schematically in Figure 4. If the temperature of a dry gas system ranges between Tc and the cricondentherm. Such a process is called “differential vaporization (condensation)”. . The amount of the condensate will increase up to a certain value as a function of pressure decrease (retrograde condensation). the volume increases during an isothermal pressure decrease. it is of importance whether the system is signified as liquid at bubble point pressure or as a vapor at dew point pressure. further pressure decrease will now effect the vaporization of the condensate. • Preceding every further change of state. Below this point. In both cases.3: • Mass and composition of the system remain constant (closed system). mass and composition of the system are changed in removing one phase . This process is specified as the “flash vaporization (condensation)”.Calculation of Phase Equilibria 59 V L Hg Hg Hg Hg Hg flash vaporization at the bubble point Hg flash vaporization at the dew point V V L Hg Hg Hg Hg n-times differential vaporization at the dew point n-times differential vaporization at the bubble point Figure 4.

3). 4. The volume remains constant meanwhile every change in state is connected with a change of the mole number in the system and its composition (open system). The integral is then substituted by summation formulae. 4. However. because Ki is a function of p.418 . FIRST SEPARATOR: p = 3. the method of finite differences is often applied.1. Therefore.33 MPa T = 48. and yji dn vap = n liqx i – ( n l i q – dn v a p ) ( x i – d xi ) . the amount of component i will decrease by yidnv resulting from the liberation and production of an infinitesimal small amount of gas dnv.27) Eq.26) Since yi = Kixi and dnvap = dnliq.3 Determination of tank oil composition after 3-stage separation on the basis of the system composition in Example 5. This would correspond to the following reflections: In case of a saturated liquid. In case of a differential vaporization.582 nvap = 0. the following is obtained after neglection of the term dnvapdxi: K ix i dn liq = xi dn liq + n l i qdx i (4. the K-factors must always correspond to the given pressure and temperature.27 cannot be integrated.60 Calculation of Phase Equilibria The equilibria states appearing at a flash vaporization in series after one another can be calculated independent of one another by Eq.19 to Eq.21 in regard of their phase composition. some material balance has to be applied. Results have been often proved to correspond with the experimental data in a most accurate way (see Example 4. (4. 4.7 oC nliq = 0. Example 4.

100 1.15 KC7 SECOND SEPARATOR: p = 0.013* Fluids Liquid Vapor xi yi 0.8991 0.0557 0.0339 0.8132 10000 0.0021 1.000 10.1430 0.400 1.1110 0.0504 0.0095 0.0128 0.0405 0.0284 0.0281 0.003* Fluids Liquid Vapor xi yi 0.170 0.0419 0.0432 0.1207 0.039 0.0281 0.0174 0.1110 0.088 0.4011 1.230 0.44 MPa T = 48.6877 1.0485 0.0489 0.0419 0.4404 0.600 3.0560 0.0489 0.0339 0.250 0.0135 0.0000 K-Factor Ki 60.0000 C1 C2 C3 C4 C5 C6 C7+ Σ= *K7+ = 0.0288 0.0290 0.650 0.0025 0.Calculation of Phase Equilibria 61 Component Well Stream Composition zi 0.6877 0.0000 1.156 Component Feed Composition zi 0.7 oC nliq = 0.844 nvap = 0.0355 0.6530 0.0096 0.415 0.0109 0.0106 10000 C1 C2 C3 C4 C5 C6 C7+ Σ= *K7+ = 0.590 0.0000 K-Factor Ki 8.0019 0.15 KC7 .0309 0.0403 0.0485 0.

466 mol tank oil 4.0000 C1 C2 C3 C4 C5 C6 C7+ Σ= One mole reservoir crude oil produced amounts to nliq1 x nliq2 x nliq3 = 0.0403 0.052* Fluids Liquid Vapor xi yi 0.0000 K-Factor Ki 265.8132 1.0514 *K 7+ = 0.0000 0.35 1. but also on composition.15KC7 Component Feed Composition zi 0.70 0.0566 0.0444 1.7 MPa o C nliq = 0.8550 1.0007 0.0210 0.1965 0.3 5.1775 0.844 x 0.3022 0.2 Evaluation of K-Factors Using Convergence Pressures As already mentioned.0513 0.1885 0.2.0038 0.9486 nvap = 0.0355 0.0396 0.9486 = 0. Since then the collection of data has been continuously enlargend.0 46.0330 0.109 0.582 x 0.0299 0. KATZ and HACKMUTH have been the first to provide for appropriate data of gas/oil-systems.62 Calculation of Phase Equilibria TANK: p T = 0.72 0.6 14.0557 0. the Ki-factors in real systems not only depend on temperature and pressure. .0309 0.0135 0.1 = 48.

Figure 4. However.with the critical pressure of the mixture and the other one with the vapor pressure of the pure substance.just as in case of the highly volatile component .0 0.at which the vapor and the liquid phase are in equilibrium . methane shows some irregular behavior. According to observations.4: K-factors for methane-propane at T c = 100 oF (from SAGE.4 is a plot of the K-factors for the methane-propane system at the critical temperature of the mixture T c = 100 oF. the K-values have no physical meaning up to the point of apparent convergence because of the existence of only one phase beyond a certain pressure limit. .Calculation of Phase Equilibria 63 10. its K-factor will decrease. the two K-isotherms of binary system will also converge to a certain point at other temperatures than Tc. LACEY and SCHAAFSMA) (1934) The most simple relations exist in any 2-component system. Generally.3 100 200 300 500 1000 1500 log p [psia] Figure 4. Above the critical pressure of the system. the K-factors increase with increasing T at constant pressure.5 0. This point is also characterized by K = 1(y = x). In contrast to the critical isotherms. At any isothermal change in state. this convergence is only an “apparent” convergence.K > 1 for the highly volatile component methane.0 5.0 log K 1. K for the less volatile component propane will equal 1 at two different pressures. However. One corresponds . On the other side K<1 for the less volatile component propane since its concentration in the vapor phase is less than in the liquid. Its K-factor increases with temperature up to a certain limit and decreases beyond that.0 3. If the pressure increases. At any pressure . K-factors have no longer any physical meaning.

Figure 4.000 and 5. It then becomes apparent that at low pressures and temperatures.01 C10 0.001 0. the NGAA (Natural Gasoline Association of America) published a diagram series lgK = f(lg p)T in its “Equilibrium Ratio Data Book”. the isotherms may be extrapolated to K = 1 defining the pressure at this point as the convergence pressure pk of the system at the actual temperature. Note: • If the evaluation of the K-factors is based on the convergence pressure of the mixture. The differences in K-values for the two convergence pressures at pressures below 0. The (apparent) convergence pressure pk is a function of the composition of the mixture.64 Calculation of Phase Equilibria Anyway. In 1957.7 MPa (100 psia) are not significant for the lighter hydrocarbons ethane through pentane. The diagrams were based on average values of already available data from various gas condensates and gas/crude . 100 C2 C3 C4 C1 C1 10 1 C5 C6 K 0. • At a given temperature.5: Comparison of K-factors at 100 oF for 1.1 C7 C8 C9 0.5 emphasizes the dependence of the K-factors on the convergence pressure being a function of the system composition. the K-factors depend only on p and T and are independent of composition.000-psia convergence pressure (from NGAA. 1957) It has been proved that multi-component systems have properties very similar to those of the binary systems with respect to the convergence of all K-isotherms. the convergence pressure of a mixture coincides with the critical pressure of another mixing ratio of its components. the K-factors are nearly independent of composition.0001 10 2 3 45 100 2 3 45 1000 2 3 45 Pressure [psia] Figure 4.

Calculation of Phase Equilibria

65

oil-systems. This series includes the paraffinic homologous compounds - methane through decane - just as ethylene, propylene, nitrogen, carbon dioxide and hydrogen sulfide within the pressure range from 0.07 to 140 MPa and for temperatures ranging from - 148 to +260 °C. Figure 4.6 and Figure 4.7 present the respective charts for methane and hexane, respectively, in case of the convergence pressure pk = 34.5 MPa (5000 psia). The values for this convergence pressure can be used for a large number of hydrocarbon systems. Today, these data are already available for data-processing with computers. If pk is known, it is possible to choose from the measured series those K-factors , which are best suited to describe the phase equilibrium vapor/liquid of the given system. The second parameter, which has to be evaluated to proceed with the prediction of the phase and volumetric behavior of complex hydrocarbon systems, is the K-factor of the heaviest component. In conventional analyses of hydrocarbon fluids, everything heavier than hexane is grouped together and reported as the pseudo-component “C7+”. This fraction is a mixture of materials of varying volatility. However, the vapor pressure curves and the critical properties of the hydrocarbons heavier than hexane are fairly close together. Thus, it is possible to characterize the mixture “C7+” by an average set of K-values.

66

Calculation of Phase Equilibria

103 5

2 102 5
Te m pe ra tu re 50 [ °F ] 200 100 0 50 0

2 101

00 20 40 160 - 1 -1 -1 -

K = y/x

5

2 10
0

80 -1
-2 0 0

5

-2 20
-2 40

2 10-1 5

-2 60

2 10
-2

101

2

5

102

2

5

103

2

5

104

Pressure [PSIA]

Figure 4.6: K-factors for methane, 5,000 psia convergence pressure (from NGAA, 1957)

Calculation of Phase Equilibria

67

101 5
Te m pe ra t

2 100 5

ure

50

0

[° F ]

45 0 40 0

2 10-1

30 0 28 0 26 0 24 0 22 0 20 0

18 0
16 0 14 0

K = y/x

5

1 20
10 0

2 10
-2

80
60
40

5

20
0

2 10-3 5

- 20

-40

-60

2
-1 0 0

10

-4

101

2

5

102

2

5

103

-10 0

-80

2

5

104

Pressure [PSIA]

Figure 4.7: K-factors for hexane, 5,000 psia convergence pressure (from NGAA, 1957)

29) • The critical point of the component C2+ is plotted on Figure 4. • The respective composition of the liquid phase being in equilibrium are evaluated by a flash process (see Example 4.68 Calculation of Phase Equilibria 4.. the K-factors become very sensitive to the convergence pressure used and care must be . the convergence pressure pk can be read off because it coincides with the critical pressure of one of the possible mixing ratios in the “binary” system.2. As the operating pressure approaches pk.1). all other components are summarized in a second fictitious component.3 Evaluation of Convergence Pressure Several iterative methods of estimating pk have been proposed. the whole procedure has to be repeated starting with the assumption of another convergence pressure. ∑ i =k 2Wi (4.6 and Figure 4. This iterative method will be demonstrated by Example 4. • If this numerical value of the convergence pressure does not coincide with the value initially assumed.7). Wi. • With exception of the lightest component methane. if the operating pressure is considerably lower than pk.8 which shows the vapor pressure curves of homologous hydrocarbons. • The “binary” critical envelope curve of methane with this pseudo-component is drawn. The iterative process must go on until correspondence of the assumed convergence pressure and the evaluated one will be achieved.28) ∑ i =k 2 WiTc i Tpc = --------------------------------------.4. One of the most convenient methods is based on the assumption of a pseudo-binary system and consists of the following steps: • A certain convergence pressure is assumed and the K-factors of components are evaluated from the NGAA-diagrams (examples are given through Figure 4. The K-factors obtained through this method exhibit only satisfactory results. The pseudo-critical temperature of the pseudo-component C2+ in the liquid phase is evaluated in calculating the averaged mole weights in the liquid phase. through Wi = x i M i and the pseudo-critical temperature of C2+ through (4. • For the given temperature.

the pressure and temperature as well as the critical data of the components.inevitable in estimating the K-values .may lead to great errors. • The Z-factors both for the liquid and the vapor phase can be evaluated solving the equation of state (Eq. 3000. l i q K i = --------------φ i. A. • On the basis of the calculated phase compositions. charts exist only for some distinct convergence pressure (800. 3.2. • New K-values can be evaluated by inserting these values in Eq.5: • The composition of the system. the phase compositions are calculated through an iterative calculation procedure (see Example 4. 2000. The steps described above will be repeated until the K-factors converge within a certain limit.1).5 aj0. • The fugacity coefficients for both phases are evaluated by use of Eq. 4.414B  a b b 2 2B and a ij= ( 1 – K ij ) ( a i0. 1000.49) through use of the method of C ARDAN (see Example 3.– ---. the K-factors of the components of the system can be evaluated by an iterative process like the one presented through Example 4.14 and Eq. 4. and B after PENG-ROBINSON are calculated for both phases. However. 4. their acentric factors and the interactive coefficients must be initially known. Thus the interpolation . 4.5 ) . • On the basis of assumed K-factors.ln  -----------------------------. b.15: 2 ∑ k xi aij b bi i = 1 i A Z + 2 ⋅ 414B -ln φi = -. 1500.3).-----------------------------. 5000 and 10.4 Evaluation of by use of PENG-ROBINSON equation of state With help of the “successive solution method”.16: φi.  Z – 0.( Z – 1 ) – ln ( Z – B ) – -------------. the calculation procedure is repeated. • The last flash iteration results in the evaluation of the mole number in the liquid .000 psia). the EOS-parameters a. vap • On the basis of the evaluated K-values.Calculation of Phase Equilibria 69 taken in the selection of the correct value of pk.

methane for binary hydrocarbon mixtures (from WINN.8: Convergence pressure data . the mole number in the vapor phase nvap and in the evaluation of the composition of the phases. the volume and the density of the two phases can be calculated. • Using Zliq and Zvap . 900 C16 800 C7-C23 C-C 24 5 700 C10 600 C8 500 C7 C6 400 C5 iC5 C4 C* C4 iC4 C3 C* C3 Temperature [°F] 300 200 100 C2 C* C2 0 -100 200 300 400 C1 600 1000 2000 4000 6000 Convergence Pressure [psia] Figure 4. too.70 Calculation of Phase Equilibria phase. nliq. 1952) .

Thus a second iteration must be done.0194 0.2 0.9000 0.06o F is close to T c of pure n-heptane.50 721.96 K = 514.1700 0.697 2.i K Ki xi Wi=xi Mi WiTci C1 C2 C3 nC4 nC5 nC6* C7* Σ 0.4 0.0470 568.09 58.4 0.06 F 42. the convergence pressure pk can be read off on the envelope curve between methane and n-heptane at pk = 3150 psia = 21.0325 507.4 Evaluation of the convergence pressure by the iterative NGAA process described above.17 114.0292 469. This value does not coincide with the initially assumed value.72 MPa. SECOND ITERATION: pk = 20.998 1.0309 0.79 373.6973 0.0318 0.4300 0.5876 nvap = 0. The composition of the system is tabled below.6300 0.5499 0.0221 0.Calculation of Phase Equilibria 71 Example 4. and the temperature and the pressure are fixed as T = 49 and p = 15. For the given temperature of the system (T = 49 o C and 120 o F.22 Tc. respectively).4 MPa (3000 psia) .0223 0.010 1.15 86.2250 0.97 305.8 T pc = 514.5 MPa. i ---------------------------------k ∑ i = 2W i 23098.12 72.7 MPa (4000 psia) nliq = 0.07 44.050 304.2854 32.0720 0.0229 425. o C FIRST ITERATION: pk = 26.56 1101.74 1.01 *assumption: C 7+ ≈ nC 8 Tp c = ∑ i =k 2 Wi Tc.7 0.598 18541.= 540.8 0.1762 1.698 On Figure 4.698 23098.0332 369.345 4.01 o = ---------------------.6500 0.45 2054.0000 42.4124 Component zi Mi kg/kmole 30.8 0.

8800 0.6200 0.81 24038.0000 Wi = xi Mi 1. Table II presents the interaction coefficients Kij which must be known. The successive solution method is used on the basis of the composition of the system. i -----------------------------------k ∑ i=2Wi 24038.1800 0.700 0.0700 xi 0. Further refinement is impossible (K-values only evaluated for 1000 psia steps in pk).0336 0. too.038 44. The agreement with the assumption is now satisfactory.= 541.399 W i T ci 308.8: p k = 3150psia and thus pk = 21.7 K.0233 0.72 Calculation of Phase Equilibria nliq = 0. .4496 Component C1 C2 C3 nC4 nC5 nC6* C7* Σ Ki 1.738 2.2200 0.39 o = ---------------------.80 19360.424 4.0483 0.5604 nvap = 0.027 1.38 MPa and T = 338. Example 4.42 K = 514.5 Evaluation of the K-factors of the three components of a system at p = 1.00 1138.5332 0.45 379.72 MPa .55 2111.0336 0.4200 0.78 739.39 *assumption: C 7+ ≈ nC 8 k ∑ i=2WiTc.399 From Fig.0299 0. 4.010 1. the critical and reduced data as well as the acentric factors which are given below in Table I.2980 1.9 F 44.162 34.

193 0.000 0.440 0.6 pc MPa Tr pr ωi 0.C4 n .3.800 0.2: Interaction Coefficients of the Components C3 C3 n .374 0.245 0.003 0.325 3.C5 0.916 0.000 C3 n .K-factors as well as the mole numbers in the phases.001 0.767 0.198 0.52 nvap = 0.191 1.C4 0.C5 Σ 1.182 Then the flash iteration starts with the evaluation of a and b. are assumed.001 0.2 469.000 For the first iteration of a flash process.251 4.001 n .1: Composition of the System.3: FIRST ITERATION: nliq = 0.28 0.C4 n .C4 n .035 1.61 0.409 Table 4. The gas and the liquid phase compositions are calculated through an iterative calculation procedure (see Example 4.721 0. The assumed K-factors as well as the evaluated phase compositions are given in Table 4. Critical Data and Acentric Factors Component Composition Critical Data Reduced Data Acentric Factor zi C3 n .369 0.1).537 0.000 zi y i = --------------------------n L ⁄ Ki + n v 0.001 0. nliq and nvap.C5 0. At first the parameters b and a for each component have to be evaluated through .8 425.003 n .48 Component K-Factor zi xi = ----------------------nv K i + n L 0.C5 0.000 0.152 0. Table 4.Calculation of Phase Equilibria 73 Table 4.745 0.11 Tc K 369.797 0.363 3.

2.635 E-5 7. (see Eq.5 ai = ( a c α )i .) The calculation procedure is presented through Example 3.48).53 b = ∑ xi bi i=1 k k k and Eq. -----------pc (see Eq. 3.54226ω – 1. 3. (see Eq.003 E-5 ai 1.729 2.4). The results of these calculations are tabled below (see Table 4. 3.35 .47).C5 bi 5. a and b for the liquid phase as well as for the vapor phase are calculated on the basis of phase compositions evaluated through the first evaluation (see Table 4.45724 R T .74 Calculation of Phase Equilibria RT c bi = 0.5 ( 1 – K ij) .54226ω ) ( 1 – T r ) .4: First Iteration Component C3 n .0778 --------..3. 3.5 2 0. 3. pc (see Eq.56 a = ∑ ∑ i = 1j = 1 xi x j ( ai a j ) 0.46).52) and through 0. 3.37646 + 1. Then the parameters A and B are calculated after PENG-ROBINSON using Eq.C4 n . 3.247 E-5 9. 2 2 αi = 1 + ( 0.557 By use of Eq. i = 0.072 1. ac . Table 4.

b. Table 4.232 b 6. A and B for both phases are summarized in Table 4. RT The evaluated numerical values of the parameters a.5.368 E-2 2. 3 the formula x + rx + sx + t = 0 is reduced to 3 2 . The factors r.2699 0.. the CARDAN equation is applied for 3 2 2 2 3 x + rx + sx + t = 0 where 3 2 r = –( 1 – B ) . 3.2144 B 3. s and t have to be evaluated both for the liquid and the vapor phase.5: First Iteration a Liquid Phase Vapor Phase 1.551 1. s = ( A – 2B – 3B ) and 2 t = –( AB – B – B ) . By substitution of 2 3 r x = y–. To solve this cubic equation of state. 3.975 E-2 Now the Z-factors both for the liquid and the vapor phase can be evaluated solving the equation of state given by Eq.873 E-5 6.Calculation of Phase Equilibria 75 ap A = -----------2 2 R T and Eq.36 bp B = --------.072 E-5 A 0..49: Z – ( 1 – B ) Z + ( A – 2B – 3B )Z – ( AB – B – B ) = 0 .

 -  2 3 3 2 3 Contrary to Example 3. D < 0 both in case of the liquid phase and in case of the vapor phase.. 3 the Z-factors(=x) of the phases can be evaluated. and y3 into r x = y–.. respectively.– --. In defining ς = and 2 p --– --27 3 q cos ϕ = – ----. .+ t . 2ς the three solutions for the reduced form y + py + q = 0 are given by y1 = 2 3 ς cos ( ϕ ⁄ 3 ) y2 = 2 3 ς cos ( ϕ ⁄ 3 + 2π ⁄ 3 ) and 3 y3 = 2 3 ς cos ( ϕ ⁄ 3 + 4π ⁄ 3 ) .76 Calculation of Phase Equilibria y + py + q = 0 where 3s – r p = ---------------3 and 2r rs q = -------. Introducing y1.3. Thus the cubic equation of state has three real solutions. 27 3 The discriminant D is defined as 3 q 2 D =  p  +  -. y2. The solutions are summarized in Table 4.6.

---------------------. meanwhile the actual Z-factor for the vapor phase is equal to the greatest one of the three solutions for the vapor phase.ln  -----------------------------. 4. The Z-factors are introduced into Equation 4. K-factors are evaluated through Eq.2580 0.414B  b a b 2 2B k respectively. of n-butane) and φ 5 (of n-pentane) both for the liquid and the vapor phase. liq K i = --------------φi.0523 and Z vap = 0.6: First Iteration Solution f or the Z-Factors after CARDAN Z1 Liquid Phase Vapor Phase 0.1325 Z3 0. φ 4 .0526 Z2 0.( Z – 1 ) – ln ( Z – B ) – -------------. - Z – 0.8 .7852 .15 2 ∑ xi aij bi bi Z + 2 ⋅ 414B A i=1 ln φi = -.6672 As a result of the first iteration flash.7. Table 4. .Calculation of Phase Equilibria 77 Table 4.7519 φ5 0.4707 0. the actual Z-factor of the liquid phase must be identical with the smaller one of the three solutions for the liquid phase.7852 Obviously.7: First Iteration Fugacity Coefficients φ3 Liquid Phase Vapor Phase 1.2172 0.8473 φ4 0. v a p and summarized in Table 4.0523 0. Thus the first iteration results in Z l i q = 0.1831 0.16 φ i.6968 0.– ---. The results of this calculation procedure are summarized in Table 4. to evaluate the fugacity coefficients φ3 (of propane).

7453 0.11. 3 2 2 2 3 Table 4.1 (see Table 4.2742 zi x i = ----------------------n v Ki + n L 0.6260 0. b.5020 0.4 and in Table 4.1623 1.444 Component C3 n .6260 0.3358 0.0445 1. the cubic equation of state has three real solutions.C5 Σ K-Factor 1.9) On the basis of the data in Table 4.C5 K-Factor 1.8: First Iteration Components C3 n .2744 These K-factors are the basis of the second iteration which starts again with the iterative calculation procedure presented through Example 4.78 Calculation of Phase Equilibria Table 4.9: SECOND ITERATION: nliq= 0.C4 n .0001 zi y i = --------------------------n L ⁄ K i + nv 0. Correspondingly.C4 n .556 nvap= 0.4847 0. The solutions are summarized in Table 4. Again D < 0 both in case of the liquid phase and in case of the vapor phase.9 the numerical values of the parameters a. 3. Now the Z-factors both for the liquid and the vapor phase can be evaluated solving the equation of state given by Eq.2102 0.10.49 Z – ( 1 – B )Z + ( A – 2B – 3B )Z – ( AB – B – B ) = 0 through the CARDAN equation.4847 0.0000 . A and B for both phases are evaluated again and summarized in Table 4.

7812 Once more again.725 E-5 6. Table 4.414B  coefficients the fugacity φ4 (of n-butane)and φ 5 (of n-pentane) both for the liquid and the vapor phase.7812.6655 .296 E-2 2.1846 0.15 b A -ln φi = --i ( Z – 1 ) – ln ( Z – B ) – -------------b 2 2B respectively. the actual Z-factor of the liquid phase must be identical with the smallest one of the three solutions for the liquid phase.11: Second Iteration Solution for the Z-Factors.2592 0.1988 0.490 1.7511 φ5 0.7161 0.1364 Z3 0.10: Second Iteration a Liquid Phase Vapor Phase 1.997 E-2 Table 4.12: Second Iteration Fugacity Coefficients φ3 Liquid Phase Vapor Phase 1.– ---b a ∑ -----------------------------ln  Z + 2 ⋅ 414B  . The results of this calculation procedure are summarized in Table 4.4717 0.0521 and Zvap = 0.2520 0.12.0525 Z2 0.Calculation of Phase Equilibria 79 Table 4. 4.??????  Z – 0. meanwhile the actual Z-factor for the vapor phase is equal to the greatest one of the three solutions for the vapor phase. to evaluate k 2 xa i = 1 i ij b i ----------------------------------------.116 E-5 A 0.249 b 6.2173 B 3. The Z-factors are introduced into Eq. Thus the second iteration results in Zliq = 0. after CARDAN Z1 Liquid Phase Vapor Phase 0.0521 0.8477 φ4 0.

13.4769 0.1 results in the mole numbers and in the composition of the phases.7452 0.16 φi.3345 0.1609 1.562 nvap= 0. liq K i = --------------φi. Table 4. 4.6280 0.0000 zi y i = --------------------------n L ⁄ K i + nv 0.2773 .14: FINAL RESULTS: nliq = 0.438 Component K-Factor zi x i = ----------------------n v Ki + n L 0.13 Table 4.0446 0.C4 n . On the basis of the K-factors given in Table 4.2773 The K-factors from the two flash iterations exhibit satisfactory convergence.4769 0.80 Calculation of Phase Equilibria The K-factors are evaluated through Eq.C4 n .5046 0.13: Second Iteration K-Factors Components C3 n .6280 0. a final iterative calculation procedure analogous to Example 4.C5 K-Factor 1.C5 Σ 1.2101 0.9999 C3 n . v a p and given in Table 4.

e. If the pressure is low (the forces of attraction can be neglected because of the great distance of molecules) or in case of high temperatures (dominance of the kinetic energy of the molecules). the change in state can be well described by the equations of state which are also presented in Chap. in case of natural gases .1 Volume As previously discussed in Chap. The value R depends on the system of units. Below the BOYLE temperature.1) where R is the universal gas constant and n is the number of moles. 3. In the SI-system: R = 8. the state of ideal gases is defined by the following equation of state: pV = nRT (5. It has been observed that in many cases . PENG-ROBINSON).Chapter 5 Phase Properties 5. R EDLICH-KWONG.12 pV = ZnRT 81 . no intermolecular forces are acting. At any isothermal change of state. 3. 3 (VAN DER WAALS. This behavior of a gas in called a real one.31432 J / K mole.1.g. In case of real gases. In the petroleum industry. because of intermolecular forces of attraction and repulsion. In case of ideal gases. the real gases are more compressible than an ideal one. real gases behave similarly to perfect gases.the volume becomes less than half if the pressure is doubled. the decrease in volume is directly proportional to the increase in pressure.1 Natural Gases 5. the basic relation of Eq.

5) As expected. 5. ethane and propane as a function of the pressure. Figure 5. the Z-factor depends both on pressure and temperature. It is defined as the ratio of the volume of a real gas to the volume of a perfect one at identical temperature and pressure. If Eq..003473γ g -------V R ZT ZT (5.97 ρg . 5. .= 0.= ----------.4 is combined with Eq. the “specific gravity” of the gas γ g is defined by ρg M M γ g = ----------. The Z-factor is the so called “deviation factor” or “compressibility factor”.4) where the subscript “a” indicates the values for air and the superscript “o” those at standard conditions (atmospheric pressure).= ------------0 M a 28.2) The atmospheric density of the gas is often related to the atmospheric density of air at the same standard temperature.3) (5. a 0 (5. vary considerably.= -------γ g ---. which can be determined experimentally.82 Phase Properties is commonly used because of its easy applicability.1 shows the Z-factor of methane. pV = ZnRT can be transformed into ZRT pV = ----------M if m = 1[g] gas The reciprocal of the specific volume V is the density and so 1 pM ρg = --. By introducing m n = --M where m: mass M: molecular weight. Z-factors of different gases. the gas density results in Ma p 1 p ρg = --. In doing so.= ------.2. V ZRT (5.

7 are substituted in Eq. In case of one mole.00 bar T o: 273. The formations volume factor for gases.6 0. ethane and propane versus pressure at T = 140oF (from STANDING. B g. Eq. 5.7) (5. 5.Phase Properties 83 1.15 K and in USA: po: 14. 1977) 5.3 results for standard conditions in p°V ° = Z ° RT° If Eq. 5. is defined by V B g = B g ( p. 5.1.01325 bar) T o: 60.7 psia (1.6. then: (5.6) .0 oF (15 oC). T ) = -----V° where the standard conditions po and T o are generally defined in Europe: po: 1.8 pV nRT z 0.0 M etha ne e pan Pro ne ha Et 0. po and V o) at a pressure p and a temperature T. the term “formation volume factor” is widely used.4 0.3 and Eq. It expresses which volume is taken up by one cubic meter (at standard state conditions.2 Formation Volume Factor In petroleum industry.2 0 0 4 8 12 16 20 Pressure [MPa] Figure 5.1: Z-factor of methane.

C p where p C = -------------.1.1. propane and pentane as a function of reduced pressure for a series of reduced temperatures..-. ∂V = – -----. -  V ∂p T B g  ∂p  T p Z  ∂p  ∂Z For ideal gases: Z = 1. ∂p Therefore. -----. the gas compressibility is only a function of pressure and is inversely proportional to the pressure.9) 5.2 shows the excellent correspondence of the Z-factors of methane. (5. i.e.3 Compressibility On the basis of the gas law.T ) = ---. the isothermal gas compressibility results in 1 1 -----1 ∂B g . . the components of natural gases are homologous paraffins with similar molecular structures and the theorem of corresponding states can be applied. o o Z T o (5.84 Phase Properties ZT --B g ( p.4 Correlation of Z-Factor on the Basis of Reduced State Variables In general.-----cg = – -.8) 5.= 0 . Figure 5.– 1  ∂Z . --------- = .

1 r 1 CH4 C3H8 C 5H 12 C H 3 8 5 C 5H 12 T=1.6 1.4 0.2 0. Tpr and ppr.4 2. pr Figure 5.0 r CH4 C5H12 C3H8 1 3 CH4 T=1.0 2.4 0. 1949) The theorem of corresponding states can also be applied to gas mixtures through the evaluation of “pseudo-critical” data.1 0 0 0.12) (5.8 3. Then the “pseudo-reduced” data.0 T=1.9 r 3 T=1.8 2.5 r 0.2 0.2 r CH4 T=1.4 1.5 0.0 Reduced Pressure.2 2. Tpc and ppc.6 0.2: Z-factor as a function of reduced pressure for a series of reduced temperatures (from SAGE and LACEY.11) where pci: critical pressure of component i Tci: critical temperature of component i yi: mole fraction of component i in the gas phase.Phase Properties 85 1.9 0. T pc p ---p pr = ---pp c (5.13) . which are given by Tp c = i=1 ∑ y i Tc .3 r C5H12 CH4 C5H12 C3H8 C3H8 pV nRT z 0.7 CH4 C3H8 C5H12 5 T=0. can be evaluated: T T p r = -------.8 1.6 2.8 0..0 1.6 0.10) and p pc = i=1 ∑ y ip c i k (5.2 1.i k (5.3 0.

By means of these diagrams. Field Unit).depending on the feasibility of chemical analysis .3). the data of fractional distillation are not available. On the basis of these correlations. This plot can be used to evaluate the Z-factor of a natural gas on the basis of its composition and the critical data (see Example 5.5 (see also Example 5.are determined. the Z-factor can be determined even if the gas composition is not known. In the laboratory practice.86 Phase Properties where T: operating temperature p: operating pressure Note that both T and Tpc in Eq. several diagrams have been published which correlate the pseudo-critical functions of state with the gas gravity.4). These diagrams are only valid for a given quality or only for one area.1). reservoir. the engineer is able to calculate pseudo-reduced properties of natural gases for which conventional analyses are available. either given in K (Kelvin. In many cases. The respective diagram for Oklahoma City gases is given in Figure 5. the C7+ fraction is obtained as liquid residue from a fractional destillation and two properties . etc.12 must be absolute temperatures.averaged molecular weight and specific gravity . The compositions of these mixtures cannot exactly be defined. formation. Several natural gases contain considerable amounts of heavy components which will be combined .2. 5.to the pseudo components C 7+ or C9+. basin. SI unit) or in R (Rankin. Anyway. These properties have been correlated with the pseudo-critical data of this fraction (see Figure 5.) . The definition of corresponding states on the basis of the pseudo-reduced data results in a generalized plot of Z-factors for natural gases (see Figure 5.

3 0.9 15 Pseudo Reduced Pressure Figure 5.8 2.3 0.5 0.0 2.4 1.3 0.0 1..95 0.45 Compressibility Factor Z 0.9 7 8 9 10 11 12 13 14 0.5 1.0 1.9 1 . 1.6 0.Phase Properties 87 Pseudo Reduced Pressure 1.7 1.6 1 .1 1.5 15 1. 2 1.1 1.7 1 .6 1 .4 1 . 1948) Compressibility Factor Z . 1.7 1.2 1.35 1.1 Pseudo Reduced Temperature 3 . 05 1 1.2 1.8 1.1 0 1 2 3 4 5 6 7 8 1.2 5 1.0 0.3: Z-factor for natural gases (from BROWN et al.8 1.0 1.25 1.4 0.4 2 .9 2.6 2.0 0.

200 1.5: Pseudo-critical properties of Oklahoma City Gases (from MATTHEWS et al.2 Gas Gravity Figure 5. 1942) ..100 1. gr.400 Pseudo Critical Temp.1 1.] absolute 675 650 625 600 575 550 525 500 475 450 425 400 375 350 325 0.300 Sp. 1942) 700 Pseudo Critical Temperature [°R] Pseudo Critical Pressure [ib/sq in. 1. [°R] 1.4: Pseudo-critical temperatures and pressures for heptanes and heavier (from MATTHEWS et al.6 0.7 0.000 Pseudo Critical Press.0 1.8 0.5 0.88 Phase Properties 1.9 1. [psia] 500 400 300 200 100 120 140 160 180 200 220 240 Molecular Weight Figure 5.

0053 0.000 0.38 0.06 0.= 1.i yi x Tci pci yi x pci kg/kmole kg/kmole K K MPa MPa 16. Component C1 C2 C3 iC4 nC4 iC5 nC5 C6 C * 7+ Mi yi yiMi Tc.Phase Properties 89 Example 5.2 86.63 22.65 0.3 Z = 0.296 0.60 4.93  p p c  4.020 3.0023 0.1 58.982 18. .152 0.37 0.1 Evaluation of the density of a natural gas at T = 105 o C and at p = 22.1 58.355 305.0 0.476 0.8 408.98 4.0051 0.88 2.55 MPa.6 179.056 3.99 1.88 0.077 4.7 507. the critical data as well as the calculated reduced data of the components.89 202.0158 0.25 0.021 4.= 4.0086 1.0132 0.4 369.1 72.25 1.=  ------------.08 2.2 461.019 3.925 which is introduced in p∑ yM i = 1 i i ρ g = -------------------------------ZRT k where R = 8.2 114.308 0.1 425.0033 0.= ------------------------------.010 2.008 3.87 T pc 202.109 190.284 0.3 The table below summarizes the composition of the gas.31432[J°K -1mol-1].562 From Figure 5.007 3. The Z-factor is determined by using Figure 5.582 0.4 469.4 568.67 4.344 30.15 T p r = -------.166 0.2 72.01 0.9443 15.63 4.1 44.23 0.562 Σ assumption: C 7+ ≈ C 8 T 105 + 273.5 ---p pr = --p.8 4.83 4.0021 0.48 0.79 0.

γ g ---.kg ⁄ m 0. the density can be calculated through 3 0.5 as Tp c = 390R and ppc = 670psia .86 × 363. --R ZT where Ma = 28.15 ρg = 142..7018 at temperature T = 90.= 655 = 1.68 --------T pc 390 and ppr = --p.355 E .86 results from Figure 5. and R = 8.7018 × 125 E + 5 ρg = -------------------------------------------------------------------------.= 1813 = 2. Z = 0.02897 × 0.71 .97 E-3 kg/mole.90 Phase Properties 3 225.3 ρg = ----------------------------------------------------------------.65 so that .32kg ⁄ m 3 Example 5.3. Now the pseudo reduced data are calculated as T Tp r = -------. Because Ma p ρg = ------. At first the pseudo critical data are read off from Figure 5.kg ⁄ m 8.5 MPa(1813 psia).31432J/K mole.2 Evaluation of the density and the formation volume factor of an Oklahoma City Gas of the specific gravity γ g = 0. --------------pp c 670 On the basis of the pseudo-reduced data.5 E + 5 × 18.5 oC(655 R)and pressure p = 12.925 × 8.31432 × 378.31432 × 0.

00916 . 5.15 B g = 0. 12.15 K so that Z = 1. o o Z T Assuming that the gas is a perfect one at European standard conditions o p = 1bar and T o o = 273..15) 0.5 + 273.Phase Properties 91 ρ g = 97. its formation volume factor results in 0.5 273.C p where p C = -----------.774 kg/m3.T ) = ---.1 B g = -----------------------------------------------------.8 ZT --B g ( p.86 × (90..0. .---------------. The formation volume factor can be evaluated through Eq.

92

Phase Properties

ps ia

5

30

2 10
4

50

Saturation 100% HO, 47,489 Ib per mm cu ft 2

Water content Ib H2O per million cu ft total gas

5

2 103 5

2 10
2

5

2 101 5

2 100
-70 -50 -30 -10 10 30 50 70 90 110 130 150 170 190 210 230 250 270 290 320 340 360 380 400 440 480 520 560 600 640 680 720

Temperature, °F

Figure 5.6: Water content of natural gas in equilibrium with liquid water (from KATZ et al., 1959)

80 0 10 00 12 50 15 00 20 00 25 00 30 00

14 .7

10 0

20 0

40 0 50 0 60 0

3500 4000 5000 6000 8000 10,000

Phase Properties

93

5.1.5

Water Content

The conventional gas analysis does not indicate any water content. However, the natural gas in a reservoir is in thermodynamic equilibrium with the interstitial water. Therefore the gas must be assumed to be saturated with steam. The mole fraction of the steam, yw, can be calculated by the DALTON law (see Eq. 5.7). pw yw = -----p where pw: vapor pressure of water at operating temperature p: operating pressure of the system. From numerous measurements it could be deduced that the steam content can be well correlated with pressure and temperature (see Figure 5.6 and Example 5.3). On the other hand, the influence of the gas composition has been found to be insignificant. Natural gas and liquid water will combine to form solids resembling wet snow at temperatures somewhat above the crystallization temperature of pure water. These solids are called “gas hydrates”. Because these solids can form at temperatures and pressures normally encountered in natural-gas pipelines, this phenomenon is of particular interest to the petroleum industry. The conditions for the gas hydrate formation will be summarized later on. (5.14)

Example 5.3 The evaluation of the water content of a natural gas as a function of operating pressure and temperature. This can be done by correlation through Figure 5.6 FIRST STATE: p = 13.8 MPa(2000 psia) T = 93.3 oC (200
o F)

The water content at this state is evaluated through Figure 5.6 as

3.8 E – 4lb ⁄ cu ft = 6.1 E – 3kg ⁄ m .
SECOND STATE: p = 3.45 MPa (500 psia) T = 60
oC

3

(140

o F)

The water content at this state was evaluated through Figure 5.6 as

3.25 E – 4 lb ⁄ cu ft= 5.2 E – 3kg ⁄ m .
THIRD STATE: p = 2.8 MPa (400 psia)

3

94

Phase Properties

T = 60

o

C (140

o

F)

The water content at this state was evaluated through Figure 5.6 as

3.8 E – 4 lb ⁄ cu ft = 6.1 E – 3 lb ⁄ cu ft

5.1.6

Viscosity

Like all intensive physical properties, the viscosity is completely described by µ = F ( p, T, y 1 ,y 2 ,y3 …y k – 1 ) in case of natural gases and µ = F ( p, T, x 1 ,x 2 ,x3 …x k – 1 ) in case of hydrocarbon liquids. Eq. 5.15 and Eq. 5.16 simply state that the viscosity is a function of pressure, temperature and composition. LOHRENZ et.al. (1964) presented respective calculations which are useful for • determining viscosities of reservoir fluids in compositional material balance computations and • predicting the decrease in viscosity which occurs when gases dissolve in reservoir oils. The correlations presented below may be viewed as modifications of Eq. 5.15. The assumptions are practical because the composition is frequently not known. Anyway, the assumptions are sufficiently valid so that these correlations are frequently used for reservoir engineering computations. The gas viscosity µg increases with the temperature. Figure 5.7 shows the dependence of the viscosity of pure gases on the temperature at atmospheric pressure (0.1 MPa and 14.7 psi, respectively). Figure 5.8 presents the viscosity of natural gases at atmospheric pressure as a function of the averaged molecular weight. This diagram elucidates that the viscosity decreases along with increasing molecular weight. It also contains correction functions for CO 2, N2 and H2S. The theorem of corresponding states has been proved to be valid once more again. The viscosity at any state of the gaseous system can be predicted through the reduced viscosity which is defined by (5.16) (5.15)

014 0. T ) The procedure to evaluate µ at certain p and T is the following (see Example 5.9. the reduced viscosity µg.4): (5. µg.. • After the calculation of the pseudo-reduced state quantities ppr and Tpr .020 0.r .5. 1 The gas viscosity determined through this method deviates from laboratory measurements by 3% at the maximum. 5.010 0. r µ g.7: Viscosity of paraffin hydrocarbon gases natural gases at atmospheric pressure (from C ARR et al. 0.4 and Figure 5.008 0.1 has to be evaluated for the given gravity (or averaged molecular weight) of the gas and the operating temperature.012 0.006 0. the atmospheric viscosity µg. 1954) .004 50 n ge itro N ide iox nD e rbo l fid Ca Su n ge dro Hy ne tha Me tane i-Bu ne a Eth e pan Pro nane tane n-No n-Bu ane n-Oct ane n-Dec Viscosity [cP] 100 150 200 250 300 350 400 Temperature [°F] Figure 5. 1 ( p = 1.8. are determined. ppc. the pseudo-critical pressure.17) • From Figure 5.T) µg. • The actual gas viscosity can be calculated through Eq.17 where µg = µg. and the pseudo-critical temperature. T pc. • According toFigure 4.016 0. r ( p pr .022 0. T p r ) = -----------------------------------.018 lium He r Ai 0..024 0. is evaluated by use of Figure 5. This is a satisfactory solution for practical purposes.Phase Properties 95 µg (p. which is given by the ratio between the actual viscosity and the one at atmospheric pressure.

1954) 6.5 3.016 Correction odded to Vis cosity [cP] 0.014 0 1.012 0.0 2.0 0.004 0 H2S 0.0 2.0015 γ g= 2. 1954) .5 2.1 0.008 0.0020 6 0.0 3.5 40 2.15 1.0 5.0015 . µ/µ 1 1..8: Viscosity of natural gases at atmospheric pressure (from CARR et al.013 0. 1 .0 0.011 0.007 0.0 =2 γg 5 1.0010 0.5 N2 γg = 2.5 0.0 1.000) 0.0 0 0 5 10 15 Mole % HS 2 10 20 30 40 50 60 70 80 90 Molecular Weight Figure 5. 1. 0 1.5 2.6 1 .015 0.009 Correction odded to Viscosity [cP] 0 .10 1.9: Correlation of viscosity ratio with pseudo-reduced pressure and temperature (from C ARR et al.5 4.3 0.5 0. 1.0 1. Viscosity at 1 [atm] µ1 [cP] 0.005 0.4 0.0 1 .0020 0.0 CO2 Correction odded to Vis cosity [cP] 0. 1.5 1 .0020 3. 5 3.0010 0.006 0. 1 .0 4.0 2 3 4 6 8 10 2.6 1. 0 0.5 2. 8 Temp erat u re. pr Figure 5.6 0.2 0.75 0 0 1.9 0.20 duc ed 1.96 Phase Properties Gas Gravity (Air = 1.00 T r Pseudo R e Viscosity Ratio.5 5.0015 0.010 0.05 20 Pseudo Reduced Pressure.5 0.3 0 1.2 0 5 00 3.0010 0 0 5 10 15 0 0 5 10 15 Mole % N2 Mole % CO2 0.

.35 so that µg = µg. µ g.= 654.7018 .6 = 1.0122 = 0. The specific gas gravity was found as γ g = 0. . 1 = 0.5 MPa (1800 psia).0165cP and µg = 0.0122cP .7 R) and at pressure p = 12. the viscosity at atmospheric operating temperature can be read off as pressure and µg. ------------T pc 385 The ratio between the actual gas viscosity. and its viscosity at atmospheric pressure.= -----------.9: µg.0165 mPa s . can now be read off from Figure 5.5 where ppc = 670 psia and Tpc = 385oR. µg. respectively.9 F. 654. r × µg . The calculation of the pseudo-reduced data of the system result in p 1800 ---p pr = ---. 1 = 1. From Figure 5. Then the pseudo-critical data of the system are evaluated from Figure 5. r = 1.1.= 2.35 × 0.8.5 o o C (194.70 .4 Determination of the gas viscosity at temperature T = 90.69 pp c 670 and T -------.Phase Properties 97 Example 5.

5.10). Then the respective “pseudo-liquid” data of any hydrocarbon liquid at standard conditions may be evaluated from the real one at standard conditions and from the apparent data of the liberated light components.98 Phase Properties 5. they developed a method for computing the density of hydrocarbon mixtures.12 and Figure 5.13 serve this purpose. Figure 5.11 presenting the correlation of the density of a system containing methane and ethane with the density of the propanes-plus fraction. However.). This inconvenience can be avoided by attributing “apparent densities” or “apparent specific volumes” at standard conditions to the dissolved and easily volatilized components.1 Volume For practical purposes. which depend on the oil composition. It is based on Figure 5.2. it is sufficient to consider hydrocarbon liquids as ideal mixtures whose volumes can be calculated by adding up the volumes of the components. The theorem of corresponding states can be applied again to determine the properties at any other state of the system (e. a crude oil does not remain in the single-phase state at “normal” conditions if it contains any high mole fraction of light components which will be liberated at atmospheric pressure and surface temperature. the resulting pseudo-liquid density of the system containing methane and ethane has to be corrected with regard to the operating temperature and pressure. the weight -% ethane in the ethanes-plus and the weight-% methane in the system. They established that the apparent densities of the two light components methane and ethane depend on the density of the system (see Figure 5.2 Hydrocarbon Liquids 5.g. etc. can be easily determined at the standard (normal) state (atmospheric pressure and T = 60oF (15 oC)). viscosity. The interesting properties (density. Based on these data and assuming that propane and heavier components follow the rule of additive volumes. Then. STANDING and KATZ (1942) examined the properties of many crude oils. at reservoir conditions). . This calculation procedure is presented through Example 5. The correlation method after STANDING and KATZ is satisfactory for stock-tank liquids which have low concentrations of methane and ethane.

1942) .5 Apparrent Density of Methane [g/cc] 0.6 0.Heptane Methane .Phase Properties 99 0.8 0.Benzene Methane .Crystal Oil Methane .1 0.3 0.Pentane Methane .6 0.Hexane Methane .Cyclo Hexane Methane .2 Ethane .5 0.4 0.3 0. 60°F & atm. Pressure Figure 5.4 0.7 0.Heptane Ethane .Crystal Oil Methane .3 0.9 Density of System [g/cc].N-Butane Ethane .4 0.Propane Methane .10: Variation of apparent density of methane and ethane with density of the system (from STANDING and KATZ.Crude Oil 0.

1952) Density of System including Methane and Ethane [ib/cu ft] 60 Wt Eth % Etha ane n plus e in Mat er ia l .100 Phase Properties Density of Propane plus [ib/cu ft] 70 50 40 30 20 10 0 50 70 0 60 10 50 40 30 W in t % en M t ir et e h Sy an st e em 20 40 30 30 20 10 Figure 5.11: Pseudo-liquid density of systems containing methane and ethane (from STANDING.

7 [psia]. [ib/cu ft] 10 9 8 0 00 00 10 80 0 00 15 7 6 5 4 3 00 00 00 000 60 50 40 3 00 20 10 0 0 re su es Pr a si [p ] 2 1 0 25 30 35 40 45 50 55 60 65 Density at 60 [°F] and 14. [ib/cu ft] Figure 5.Phase Properties 101 Density at Pressure minus Density at 60 [°F] and 14. 1952) .7 [psia].12: Density correction for compressibility of liquids (from STANDING.

102 Phase Properties 10 Density at 60 [°F] minus Density at Temperature.55MPa ) and To b = 220 F ( 105°C ) . The bubble point is defined by pob = 3300psia ( 22. [ib/cu ft] Figure 5.13: Density correction for thermal expansion of liquids (from STANDING. [ib/cu ft] 9 8 7 6 5 4 3 24 0 22 0 20 0 18 0 160 140 12 0 Tem per atur e [°F] 2 1 0 25 100 60 30 35 40 45 50 55 60 65 Density at 60 [°F] and Pressure p. 1952) Example 5. o . The composition of the system is tabled below.5 Determination of the volume of 1 mole oil at bubble point through STANDING-KATZ correlations.

4404 0.1 72.1 58.11.5 907.= k ∑ i = 3V i k 122.69 lb ⁄ cuft ------------------0.0432 0.1271 0.0038 0.0035 0.0 0.× 100 = 1. From Figure 5.046 ---------------------------.5 lb/cu ft.650 1.× 100 = ------------------.786 1.1393 *) at standard conditions: 0.Phase Properties 103 Composition Mole Fraction zi = xi Mole Weight Mi kg/kmole 16.367 ∑ zi M i i = 1 are calculated. 15o C The density of the propane-plus can be calculated through ∑ i = 3 zi M i ---------------------------.1 MPa.0284 0.5 575.0 662.256 2. the pressure correction for pob (3300 psia) is read off as .367 507.300 ---------------------------.046 1.4001 1.2 86.0 30.05 [% ] k 123.021 = 876kg ⁄ m 3 = 54.0029 0.× 100 = 5. From Figure 5. the pseudo-liquid density of the hydrocarbon mixture at standard conditions is read off as 51.2 287.500 114.0174 0.4% k 130.0000 7.0 626.1 44.× 100 = ------------------.321 z iM i i = 2 and the weight -% methane in the entire system z1M 1 7.0405 0.1393 Then the weight -% ethane in the ethane-plus material z 2 M2 1.829 130.300 1.0020 0.12.0290 0.0 Density ziMi ρi kg/m3 Volume Vi * m3 C1 C2 C3 C4 C5 C6 C7+ Σ 0.

respectively.4 lb/cu ft.1 1.9 1.5 + 0. Cr .7 lb/cu ft so that the oil density at 3300 psia and 220oF results in ρob = 52.7 lb ⁄ cu ft and ρob = 778.4 Gas Gravity .Air = 1.0 1. Then the temperature correction for 220o F is read off from Figure 6.9 lb/cu ft so that the liquid density at 3300 psia and 60 oF results in 51. 1952) .I A 20° 30° 40 35 40° 30 50° 60° 25 20 15 0.8 0.3 1.6 0.P.7 0.14: Apparent liquid density of natural gases in various API gravity oils (from KATZ.13 as 3.0 Figure 5.5kg ⁄ m .4 – 3.104 Phase Properties 0.9 = 52. 3 Apparrent Density in Pounds per Cubic Foot of dissolved Gas at 60 [°F] and 1 Atmosphere 45 ity rav eG ud .2 1.7 = 48.

.natural gas systems. • The increase in volume because of expansion is more or less compensated by the shrinkage of the oil resulting from the transfer from reservoir to surface temperature.Phase Properties 105 Natural gases (dissolved gas) predominantly consist of methane. nm3/m3. The easily volatilized component is marked by its gravity. Therefore. This ratio is designated as Rs (solution ratio).14. Rs depends on pressure and temperature. In other words: Rs expresses how much gas will be dissolved in one cubic meter oil (volume at standard conditions) at the operating conditions. 5. the ratio is also often designated as GOR (producing gas . The apparent liquid density of the respective natural gas can be determined according to Figure 5. The amount of the dissolved gas is always indicated as the number of normal cubic meter per actual liquid volume at standard conditions. Therefore.2 Formation Volume Factor The oil volume inside the tank at surface conditions differs from the volume of the oil leaving the reservoir. • The difference between reservoir pressure and atmospheric pressure causes slight expansion of the remaining oil phase. In terms of the production. KATZ (1942) has applied the method of the apparent liquid volume also to crude oil .oil ratio).2. γ g . and the scarcely volatilized fraction is the crude oil with the density ρ o and its gravity γ o (in °API) at standard conditions. but also on the oil quality. This change in volume has three main sources: • The most important factor is the gas liberation below bubble point pressure pob.

The reciprocal value of Bo is called “shrinkage factor”. tank oil gravity and reservoir temperature to determine the pressure at which the given amount of gas would be in solution. the multiplication of the shrinkage factor with the volume inside the reservoir will result in the stock oil volume. This gas liberation results in a shrinkage of the oil volume. On the basis of many investigations. Thus. By multiplying the tank oil volume with the actual Bo. Bo can be defined as the volume of one cubic meter tank oil (oil at standard conditions) at certain operating pressure and temperature. This pressure is the bubble point pressure of the . STANDING (1947) amplified the correlation to permit the calculation of bubble point pressures and volume factors. Figure 5. the volume of the oil inside the reservoir is obtained.106 Phase Properties Bo.15 illustrates typical characteristics the relationship between Bo and p.0 0 pb Reservoir Pressure Figure 5. Bo decreases at decreasing pressure as soon as pob has been crossed. change in Bo will be ruled by the expansion of the oil.b 1. Bo. which must be taken into account. gas will be liberated. it is possible to take the solution ratio. gas gravity. As soon as the reservoir pressure falls below pob. is defined as oil volume and volume of dissolved gas at reservoir conditions B o = --------------------------------------------------------------------------------------------------------------------------------------------------------oil volume at standard conditions Correspondingly. As long as the pressure does not fall below the bubble point pressure pob during a pressure decrease from initial reservoir conditions. On the other side.15: Typical graph of formation-volume factor of oil against pressure The formation volume factor of the oil phase. The respective STANDING charts are appended at the end of this textbook: • Working with Chart 1.

Bt is defined by B t = B o + ( R si – R s )B g where Bo: volume factor of the actual oil phase Bg: volume factor of the actual gas phase Rs.13) • or by use of the Charts from STANDING (see appendix).6 Evaluation of the oil volume factor at bubble point conditions from operation data tabled below both in SI-units and Field-units. Bt.Phase Properties 107 respective hydrocarbon system. these empirical correlations (see Example 6. pob. can be estimated by use of Chart 3.14) and the STANDING diagrams (see Figure 5. (5. In most cases.18) Example 5.i: solution ratio at initial conditions Rs: solution ratio at operating conditions. The calculation is based on one cubic meter tank oil. The thermodynamic data of the system were given in Example 5.5. Bo can be evaluated • by use of the KATZ correlation (see Figure 5.12 and Figure 5. . • Charts 2 permits the evaluation of Bo at bubble point conditions if GOR represents the solution ratio.6) yield values which are comparable within 3 or 4% to measured laboratory data. • The total formation volume factor.

14 as ρg .722 = 83.640 0.2 kg/m .722 and γ o = 30o API.108 Phase Properties GOR nm 3/m3] Separator 1 Separator 2 Tank Σ 74 16 4 94 GOR cu ft/bbl 415. -------------------------------------------------------------------------------------------. the apparent liquid density of the gas can be evaluated through Figure 5.8 22.9 = 0. the initial solution ratio Rsi results in R s i = 94nm ⁄ m and 3 3 R s i = 527.1 kg Because the specific gravities are given as γg = 0.5 527. 640 + 16 × 0. respectively. a = 24. 897 + 4 × 1.897 1.225 x 0. 540 = 67.---------74 + 16 + 4 94 Now the amount of gas can be calculated through m g = R s i ςa γ g where ρ a is the air density: mg = 94 x 1.722.3 lb/cu ft and ρg . The computation of the average specific gas gravity results in γ g = 74 × 0.540 876. 3 .5 89.8 cu ft ⁄ bbl respectively.2 30 In summing up the producing gas/oil ratios GOR. a = 389.8 γg ρo kg/m 3 γo o API 0.

2 –3.7kg ⁄ m .1kg liberated gas.214 m3 apparent liquid volume of the liberated gas. as well as 876.214 and so ρ o.Phase Properties 109 Now the apparent liquid volume of the liberated gas at standard conditions (T =15 o C and atmospheric pressure) can be calculated through mg V g.2 kg tank oil and 83. The shrinkage results in . a p p = ------------.3 lb/ cu ft . From Figure 5.the pseudo-liquid density of the given hydrocarbon system at standard conditions results in 3 876. the pressure correction of this density for p = 3300 psia results in ρ o = 49.1 ρ o.4 lb/ cu ft or.= 1. in SI-units.29 .3 + 0.7 and so B o b = 1. the volume of one cubic meter oil under reservoir conditions results in 3 959. the final temperature correction for T = 220 o F results in ρ o = 50.= 790. In other words: 1. respectively.29m 743.= 0. app = --------------------------------.2 lb/ cu ft .12.288 m3 saturated reservoir liquid under reservoir conditions shrinks to 1 m3 tank oil. From Figure 5. in ρ o = 743.2kg ⁄ m 1000 + 0.3 3 V o b = ------------.13. a p p = ----------ρ g. a and so 3 83.1 V g.214 m . app = 49.9 = 50. 389.2 In summarizing 1 m3 tank oil and 0. On the basis of the calculated oil density under the conditions of the reservoir.8 = 46.2 + 83.

20) (5. p1) and (V2.48% 1.29 5. the bubble point pressure.29 – 1. is chosen as reference.9) so that ∂B 1 -----co = – -. Its isothermal compressibility is to be calculated in the same way as for the gases (see Eq. --------- -  -1. o V ∂p T B o ∂p T and after integration B o = B ob e Because C ρo = -----Bo where C is a constant. the density of an undersaturated oil phase can be calculated through ρo = ρ o be c o (p – p b ) – c o( p – p b) (5.20 and Eq. it can be calculated between any two states of the system. The compressibility of undersaturated liquids is constant within wide pressure intervals. 5. b [ 1 + c o ( p – p b ) ] . 5. Because co is very small.2. p2) by (5.21 will develop in a Taylor series.110 Phase Properties 1.23) (5.000 -----------------------------. pb.22) .× 100 = 22. ∂V = --. Similar formulae are also valid for water. (5. 5. Eq. (V1.21) In general. the terms of higher order can be disregarded so that B o ~B ob [ 1 – c o ( p – p b ) ] and ρo = ρ o.3 Compressibility of Undersaturated Liquids An undersaturated oil is a compressed liquid in the pressure range above the bubble point pressure pb. Therefore.19) .

6). cr as a function the pseudo-reduced pressure. the evaluated oil density is related to the density of water at the standard temperature.12.16).24) TRUBE (1957) has reported a method of calculating the compressibility of undersaturated liquids. the pseudo-reduced pressure and temperature can be approximately evaluated through Figure 5.Phase Properties 111 V1 – V2 co = – -------------------------V (p 1 – p2 ) where V is averaged volume between p1 and p2. • the pressure correction of this pseudo-liquid density from Figure 5. This method is based on the determination of the pseudo-critical data of the system and on the evaluation of the pseudo-reduced compressibility of the undersaturated liquid. and pseudo-reduced temperature. Tpr (see Figure 5. ρw = 1000kg ⁄ m . Then the oil compressibility can be calculated through co = c r ⁄ p pc (5. 3 .17 on the basis of • the calculation of the pseudo-liquid density of the hydrocarbon mixture at standard conditions (see procedure of Example 5. ppr.25) If the composition of the undersaturated system is not known. (5. Then.

72 0.60 0.62 0.80 0.70 0. 1957) 1300 Pseudo Critical Temperature [°R] 1200 1100 1000 900 800 700 Pseud re ratu mp e Te i cal o Crit Pseudo Critical Pressure [psia] 600 500 400 300 200 0.84 0.76 0.95 0.86 0.64 0.112 Phase Properties 10-1 Pseudo Reduced Temperature 10-2 cr 1.90 10-3 0.66 0.17: Pseudo-critical conditions of undersaturated reservoir liquids (from TRUBE.78 0.16: Pseudo-reduced compressibility for undersaturated reservoir fluids (from TRUBE.10 1.68 0.88 Pseudo Critica l Press u re Specific Gravity of Undersaturated Reservoir Liquid at Reservoir Pressure. 1957) .82 0.74 0.70 0.50 0. Corrected to 60 [°F] Figure 5.80 0.40 10 -4 1 2 3 5 7 10 20 30 50 70 100 pr Figure 5.00 0.

7 Evaluation of Boi for an under saturated oil at reservoir temperature T = 105 o C and under the initial pressure pi = 25 MPa through the reduced compressibility from TRUBE. The composition of the system was given in Example 5.5 and Bob was calculated through Example 5.84 pp c 3.5)] = 1.= ------------.15 = 1. Boi can be calculated through Eq. The compressibility calculated through of the undersaturated oil phase can be cr 0.= 6.= 2.02 –1 co = -------.Phase Properties 113 Example 5. .00292(25 .= ---------. the reduced compressibility can be read off as cr = 0.23 B o i = B ob [ 1 – co ( p i – p b ) ] and results in Boi = 1.29[1 .6).= 105 + 273.6. The computation of the pseudo-reduced temperature and pressure results in T T p r = -------.281. p pr = ---.92 E – 3MPa p pc 6.655 From Figure 5.22.84 Because B o.0.70 and p 25 ---.29 (see Example 5. b = 1.16. The critical data are tabled below.01 ------------------------------T pc 374.02 . 5.

18: Viscosity of subsurface samples of crude oil (from HOCOTT and BUCKLEY.84 D 0 500 1000 1500 2000 2500 3000 3500 Reservoir Pressure [psia] Figure 5. Gulf Coast Palo Pinto Lime.7°API Oil Gravity 1. The viscosity of the crude oil can be determined with sufficient accurancy on the basis of the known viscosity of the tank oil µ1 at the reservoir temperature T and atmospherical pressure.2. The dependence on pressure is more significant for high viscous oils than for low viscous oils.8 elucidates the respective correlation procedure. after BEAL. HOCOTT and BUCKLEY (1941) presented measurements of the viscosity of crude oils containing dissolved gases. Southwest Texas Absolute Viscosity [cP] Bubble Point A 6 5 4 3 2 1 0 Legend 100°F Reservoir Temperature 3 191 ft/bbl Gas in Solution at Bubble Point Pressure 29. the solution ratio Rsi and the undersaturated pressure in the reservoir.1 ° AP I G=0 Bubble Point .114 Phase Properties 5. . 1941.7°API G=0. Example 5. Anyway.6 5 Bubble Point 38 5ft 3 / B C Bubble Point 1 58°F 8 0 3 0ft / bbl GOR 45 . Figure 5.4 Viscosity 9 8 7 A B C D Permian. the viscosity decreases up to the bubble point because of the increasing fraction of light components. ∆p = pi pb.12 Gas Gravity bb 16 4° F lG OR 26. 1946) In general. the viscosity increases with increasing pressure. liquids decrease in viscosity with increasing temperature and increase in viscosity with increasing pressure. above the bubble point pressure. North Texas Frio Sand. West Texas Frio Sand.18 demonstrates that with increasing pressure.

50 cP all at the same temperature. 0 30 Viscosity of Gas-Saturated Oil [cP] (at Reservoir Temperature and Saturation Pressure) 30 20 0 20 10 7 5 3 2 0 60 0 80 00 10 00 120 0 14 0 40 Solution Gas/Oil Ratio [ft3/bbl] 1.0 0.0 2 3 5 7 10 20 30 50 70 100 Viscosity of Dead Oil [cP] (at Reservoir Temperature and Atmospheric Pressure) Figure 5.1 0.Phase Properties 115 100 70 50 Example: Problem: Find the gas-saturated viscosity of a crude oil having a solution gas/oil ratio of 3 600 ft/bbl and dead oil viscosity of 1.7 0.3 0.2 0 160 Example 0.5 cP on the dead oil viscosity scale (abscissa) and go up vertically to the 600 gas/oil ratio line.3 0. Procedure: Locate 1. 1959) 0 50 0 10 .58 on the gas-saturated oil viscosity scale (ordinate). Then go left horizontally to read the answer 0.5 0.5 0.7 1.19: Viscosity of gas-saturated reservoir crude oils at reservoir conditions (from CHEW and CONNALLY.

1946) .2 15 10 5 0. 50 45 P] re [c s su Pre 12 11 10 9 8 7 6 5 4 3 2 1 0 9 0.4 0.20: Prediction of crude oil viscosity above bubble point pressure (from BEAL. 0.6 25 20 0 .1 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 0 1 2 3 4 5 Undersaturated Pressure [psi] (Pressure above Bubble Point less Pressure at Bubble Point) Pressure 10 [psi] 3 Figure 5.116 Phase Properties Absolute Viscosity [cP] Crude Oil Prediction above Bubble Point Pressure Average Deviation 2.8 Viscosity [cP] e o lut A bs 30 0.5 0.7 0.3 0.7% Extrapolated 100 8 75 80 5 6 5 70 9 8 7 6 5 4 3 2 1 0 0 Detail 7 6 5 4 3 2 1 90 Absolute Viscosity of Crude Oil above Bubble Point Pressure [cP] 80 70 60 50 40 30 20 10 0 0 1000 2000 3000 4000 5000 Pressure [psi] 55 60 13 nt Poi bble t Bu de a 40 Cr u t ed tur a 5 of Gas Sa 3 ity os Vis c Detail 0 1.

Phase Properties

117

Example 6.8 Evaluation of the oil viscosity at reservoir conditions which are given below as well as the interesting production data. µob = 0.45cP .
SI-Units Reservoir Temperature Bubble Point Pressure Reservoir Pressure Initial Solution Ratio Standard Oil Viscosity T pb p Rsi µ1 105 oC 22.55 MPa 25.00 MPa 94 nm3/m3] 0.9 mPas Field Units 220
o

F

3300 psia 3600 psia 527.8 cu ft/bbl 0.9 cP

From Figure 5.19, the oil viscosity at bubble point conditions can be read off as µob = 0.45cP From Figure 5.20, the viscosity of the under saturated crude oil at p = 3300 psia can be evaluated: µoi = 0.48cP or µoi = 0.48 [ mPas ]

5.3

Brines
It has been recognized relatively late that water is - without any exception - one of the reservoir fluids. GARRISON (1935) and SCHILTHUIS (1938) have been the first who presented detailed information on the distribution of oil and water in porous rock and on the origin and occurrence of “connate water” (formation water, capillary water, interstitial water). Furthermore, they established a relationship between the water saturation and the rock permeability. Nowadays, knowledge of the chemical and physiochemical data of the respective reservoir water is of great importance for the petroleum engineer. Water analyses • give valuable information about the area of prospecting, • help to answer questions about the origin of the produced water,

118

Phase Properties

• make sure that injected water is chemically compatible with the reservoir water or with added water (plugging by precipitations must be avoided), • are indispensable for feasibility studies in case of chemical EOR processes. Last but not least, data such as density, compressibility, formation volume factor and viscosity are required for calculating the displacement process and for establishing a material balance. Additionally, this section presents an unusual phenomenon, the so-called “gas hydrate formation”, during which water and natural gas form a solid phase at temperatures above the freezing point of water.

5.3.1

Composition of Brines

Reservoir water always contains dissolved inorganic salts, above all sodium chloride, NaCl. Therefore, these waters are sometimes called “brines” although they cannot be compared with sea water because both the total salt concentration and the content of distinctions are different. The cations usually dissolved in reservoir brine are Na+, K +, Ca2+, Mg 2+ In certain cases, Ba2+, Li +, Fe2+, and sometimes Sn2+ can be found. The generally present anions are
2Cl , SO , HCO 4 3

and, in some cases, also CO 2- , NO - ,B r , J , S .
23 3

The elements are present in reservoir water within the following concentration ranges: Na, Cl Ca, SO 4 K, Sr in% in% or ppm in 100 ppm

Phase Properties

119

Al, B, Ba, Fe, Li Cr, Cu, Mn, Ni, Sn, Ti, Zr Be, Co, Ga, Ge, Pb, V, W, Zn where %: ppm: ppb:

in 1 - 100 ppm in ppb (in most cases) in ppb (in some cases)

g substance per 100 g brine mg substance per 1 kg brine mg substance per 103 kg brine

Usually micro-organisms of different species are also present in oil field brines. The origin of these organisms is not yet totally clear. They contribute to corrosion in the bore hole and to permeability reduction during water flooding. There is a wide variability of concentrations and chemical characteristics resulting from many factors. Such factors may be • the non-marine origin of a few sediments, • dilution with ground water, • enrichment because of vaporization by migrating gas, • adsorption and cation exchange with clay minerals, • dissolution of mineral salts in the migrating formation water, • the chemical reaction with components of adjacent sediments. Therefore, it is important that water from a certain horizont is characteristically distinct from all the other brines, even those from the immediate neighbourhood. Thus, water analyses are mainly used for identifying the produced water with regard to the geological formation it originates from. In doing so, a simple and easily readable representation of the great data set from a water sample is desired. The graphical method proposed by STIFF (1951) may be the simplest graphical method, has maximum utility and, hence, is the most popular. This graphical method is shown in Figure 5.21. Horizontal lines extend right and left from a centrally located vertical line. The positive ions are plotted on the the left of the vertical line, while the anions are plotted on the right. Characteristic positions are designated for the ions which are most frequently found in reservoir brines.

John Field Drach Field Scale: meq/liter Figure 5.22. 1951) The analysis data for all ions in question are indicated in ppm (milligram per liter). One of many practical applications in correlating producing formations is illustrated by Figure 5. If there is a difference between Σ meq of the positive and Σ meq of the negative ions. When the data of the water analysis are plotted on the graph and the adjacent points are connected by straight lines.120 Phase Properties Na 100 Ca 10 Mg 10 Fe 10 Cl 100 HCO3 10 SO4 10 CO3 10 Scale: meq/liter Figure 5. Although various scales can be employed. a closed “pattern” is formed as a “fingerprint” of the respective brine.21: Essential feature of the water pattern analysis system (from S TIFF. most reservoir waters may be plotted with sodium and chloride on a scale of 100 meq and a scale of 10 meq for the others. Barton Ellsworth Stoltenberg Field Stafford Rice Bloomer Field Na 100 Ca 10 Mg 10 Fe 10 Cl 100 HCO3 10 SO4 10 CO3 10 St. 1951) . the contents have to be converted to meq (milliequivalents) by dividing the respective value with the milliequivalent weight (ionic weight in mg per valence).22: Course of Arbuckle formation through Kansas shown by water patterns (from STIFF. the difference is presented as sodium. To make these values comparable with each other. The characteristic pattern of the formations is evident.

• expansion of the brine because of pressure decrease. their solubility in water is low (see Figure 5.06.0 represents the thermal contraction resulting from the cooling of the brine from the reservoir temperature to the standard temperature of 15 oC.2 Formation Volume Factor The definition of the formation volume factor of water Bw corresponds to the definition of B o. Bw increases because of the water expansion. . Figure 5.3. A further pressure reduction leads to gas liberation. If the reservoir pressure is reduced from initial pressure to the bubble point pressure.0. the initial reservoir pressure is higher than the bubble point pressure. In contrast to the solubility of natural gases in oil. Bw will reach its maximum value. Bw is always numerically small. However. the solubility decrease with increasing salinity of the brine (see Figure 5.24 shows a typical relationship between Bw and p. the change in volume may have three reasons: • liberation of dissolved gas due to pressure decrease. Additionally. Similarly to the hydrocarbon liquids. Consequently. rarely above 1. In the present case.23b). the contribution of the gas liberation to B w is low. This loss of liquid volume only balances partly the continuing expansion so that Bw further increases with further pressure drop. if the increase in volume by dissolved gas is too low to compensate the volume reduction resulting from high pressure. The difference between this value and 1. • contraction of the brine due to temperature reduction. Bw will be smaller than 1. Since the expansion due to the decrease in pressure and the contraction because of temperature decrease are also insignificant.Phase Properties 121 5.23a). Along with the reduction of the pressure to atmospheric pressure.

1944) Bw.8 0 150°F 100°F 10 20 30 -3 250°F 200°F 40 Total Solid in Brine 10 [ppm] b) Figure 5.24: Typical graph of formation volume factor of water against pressure The assumption is customary that the bubble point pressure of a gas-saturated brine is equal to the bubble point pressure of the crude oil inside the corresponding reservoir. .9 0.0 0 0 pb Reservoir Pressure Figure 5.122 Phase Properties 24 Solubility of Natural Gas in Water [ft3/bbl] 22 20 18 16 14 12 10 8 6 4 2 0 60 100 140 Pressure [psi] 5000 4500 4000 3500 3000 2500 20 00 1500 1000 50 0 180 220 260 Temperature [°F] a) Ratio: Solubility in Brine Solubility in Water Correction of Brine Salinity 1.23: Solubility of natural gas in water (from DODSON and STANDING.b 1.0 0.

23a.23 as follows: • The volume factors for pure water and pure water saturated with natural gas are read from Figure 5.00 100°F 0.23b. and corrected for salinity using Figure 5. the volume factor is computed by interpolation. for pure water is read from Figure 5. • The gas solution ratio. Naturally. 1944) Example 5.25 and Figure 5.04 1. R sw.03 200°F 1. 1. .25: Bw for pure water (dashed lines) and pure water saturated with natural gas (solid lines) as a function of pressure and temperature (from DODSON and STANDING.06 1.98 150°F 250°F 0 1000 2000 3000 4000 5000 Pressure [psia] Figure 5. • Assuming the effect of gas solubility on Bw to be linear.and Field units. gas saturated pure water has a higher volume factor than pure water at a given pressure and temperature (dashed lines).05 1. B w for a reservoir water can be computed with aid of Figure 5.02 1.9 Formation volume factor of reservoir water.01 1.99 0. The reservoir conditions as well as the salinity of the brine are given below both in SI.07 Water Formation Volume Factor [bbl/bbl] 1.Phase Properties 123 Figure 5.25 for the given reservoir data.25 presents data of Bw for pure water saturated with natural gas (solid lines).

0275 Bw for pure water saturated with natural brine can be read off from Figure 5.3 cu ft/bbl After correction of the gas solubility from Figure 5.26.124 Phase Properties SI Units Reservoir Temperature Reservoir Pressure Salinity T p 93. Rsw results in Rsw = 15.0275) 13.88 = 13. the .26: Density of brine as a function of total dissolved solids (from MCCAIN. ------------15.0275 + (1.034 . The gas solution ratio in pure water from Figure 5.000 ppm 3000 psia 30.000 ppm The evaluation of Bw for pure water from Figure 5. the density of reservoir water at standard conditions is plotted as a function of the total salt concentration. By dividing the respective density with Bw.3 o Field Units 200 o C F 20.46 cu ft/bbl Then the formation volume factor of the reservoir water results in Bw = 1.23b.3 Brine Density at 14.1.7 psia and 60°F [ib/cu ft] 76 74 72 70 68 66 64 62 60 0 5 10 15 20 25 30 Total dissolved Solids [%] Figure 5.3 x 0.46 = 1.25 results in Bw = 1. 1973) In Figure 5.033 .56 MPa 30.034.25 as Bw = 1.23a results in Rsw = 15.

4 3. . cw can be read off as cw = 3. cw.8 2. From Figure 5. of reservoir water at the From Figure 5.9.0 Compressibility of Water 6 cw 10 [bbl/bbl psi] 3.2 3.1 1.1 E .8 3.27b).Phase Properties 125 corresponding density under reservoir conditions can be received.10 Evaluation of the compressibility reservoir data of Example 5.0 2.6 2. 4. can be evaluated without considering the gas in solution. including effects of gas in solution (from DODSON and STANDING. 1944) Example 5. At a given pressure and temperature.Water Ratio [ft /bbl] b) Figure 5.27a at 3000 psia and 200 oF.27a the coefficient of the isothermal water compressibility.2 1.27: The isothermal coefficient of compressibility of pure water.0 0 5 10 15 3 20 25 Gas . the effect of gas in solution is to increase cw (see Figure 5.6 3.6 psi –1 .4 60 100 140 180 220 260  1  ∂V  c w = −    V  ∂P T ] sia 0 [p 100 00 20 0 0 3000 40 0 0 50 0 600 Temperature [°F] a) Ratio: Solution Compressibility Water Compressibility Correction for Gas in Solution 1.3 1.

3 Viscosity Regarding the viscosity of oil field brines. 5.3 × 0. no corresponding data have been published until now.46 cu ft / bbl . The viscosity should rise with pressure and increase significantly with increasing salt concentration. –1 .126 Phase Properties From Figure 5.23a.10 Pa respectively. the compressibility of the reservoir brine results in cw = 1. Correction of gas solubility from Figure 5. .3.6 psi or –1 . The fact that low molecular paraffins are dissolved in water under reservoir conditions should furthermore lead to a considerable reduction of the brine viscosity.3 cu ft bbl –1 . cw = 5.28 shows the water viscosity as a function of the temperature.23b: R w = 15. only few data are available. After evaluating the effect of the dissolved gas from Figure 5.12 × 3.88 = 13.28.1 E – 6 = 3. Figure 5. It is recommended to use this graph for estimating the viscosity of reservoir waters without conisdering both the pressure and the salinity of the brine. These dependencies can be observed in Figure 5.23b. the gas solubility in pure water results in R sw = 15. Unfortunately.03 E .47 E .

The lattice of the hydrate crystal is mainly formed by water molecules while the hydrocarbon molecules occupy vacancies within the lattice.3. This process takes place at temperatures slightly higher than the freezing point of pure water.8 Saline Water (60000 [ppm]) at 14.4 Natural Gas Hydrates Natural gas and liquid water can form solid material which remind of wet snow.6 1. the pressure decreases in the production line.2 [psia] 14. the condensation of water from the gas. 1950) 5. .0 0.7 [psia] 7100 [psia] Vapor Pressure 1.Phase Properties 127 2.2 0 0 50 100 150 200 250 300 350 Temperature [°F] Figure 5. therefore. Gas hydrates behave rather like dissolutions of gases in ice than like chemical compounds.4 0. Therefore. the hydrate formation is rather of physical than of chemical nature.0 1. For example.6 0. The water lattice seems to be similar to ice. such as across a choke or in a separator. This can cause a reduction in the temperature of the gas (adiabatic process) and. because the formation enthalpy is comparable.8 0.4 Dynamic Viscosity [cP] 1.2 1.28: The viscosity of water at oil field temperature and pressure (from VAN WINGEN. The conditions necessary for hydrate formation could be established. The phenomenon is of special interest for petroleum industry. These solid substances are called “gas hydrates”.

especially in case of high gas gravity and temperature. it has to be taken into account that electrolytes present in the water will lead to decreasing temperatures at which hydrate formation will occur. a meta-stable equilibrium between water and hydrocarbon gas may occur (under conditions at which generally hydrate formation takes place).128 Phase Properties Although gas hydrates can be compared rather to solid solutions than to chemical compounds. The connection Q1 . Anyway.30 presents the hydrate formation conditions for natural gases of different gravities. water and natural gas are in equilibrium). gas hydrate.29 shows a part of the phase diagram of a mixture of water and a light hydrocarbon.31. Figure 5. even if liquid water is present. the results should be applied with caution because of the large discrepancies of the few published experimental data and the correlation present. This may be caused by the crystal structure. The ratio of hydrocarbon to water depends mainly on the size of gas molecules. Figure 5. The curves can be used to estimate the conditions of hydrate formation. The quadruple point Q1 is at approximately 0 oC (ice. Seed crystals can immediately induce the hydrate formation. a certain number of water molecules is associated to each gas molecule. 1973) . However. Hydrate portion of the phase diagram for a typical mixture of water and a light hydrocarbon (from MCC AIN.Q2 represents the equilibrium natural gas-water-hydrate. This line is of special interest. The point Q2 is a “quadruple point” at which four phases are in equilibrium. water-soluble substances are often added to the system to inhibit or suppress the hydrate formation. The presence of liquid water is the most important prerequisite for hydrate formation. The effect of the concentration of various inhibitors on the hydrate-formation temperatures of a natural gas is plotted in Figure 5. Additionally. Thus.

0 . 8 0.Phase Properties 129 Hydrocarbon Liquid + Water Hydrate + Water Q2 C Hydrate + Ice Pressure Water + Hydrocarbon Gas Q1 Ice + Hydrocarbon Gas Temperature Figure 5. 1. 1973) 6000 5000 4000 3000 Pressure for Hydrate Formation [psia] 2000 1500 1000 800 600 500 400 300 200 150 100 80 50 60 30 6 0.29: Hydrate portion of the phase diagram for a typical mixture of water and a light hydrocarbon (from MCCAIN. 1945) . 0 9 ty vi ra G M ne ha et as G 40 30 40 50 60 70 80 90 Temperature [°F] Figure 5.30: Pressure-temperature curves for predicting hydrate formation (from KATZ. 7 0.

1959) Temperature calculations have been carried out for typical natural gases. This has been done to evaluate diagrams by which the maximum adiabatic pressure reduction without gas hydrate formation can be estimated.11 Evaluation of the maximum of adiabatic expansion of a natural gas without any danger of hydrate formation.63 Gravity Gas (5-21) 4 Sodium Chloride on Methane on 0..31: Depression of hydrate formation temperature by inhibitors (from KATZ et al.555 Gas Gravity (5-43) 18 Depression of Hydrate Formation Temperature [°F] 16 14 12 10 1 2 3 4 8 6 4 2 0 0 4 8 12 16 20 24 Weight per cent Inhibitor Figure 5.68 Gravity Gas (5-43) 3 Calcium Chloride on 0.32.30. Example 5. Problem: Up to which pressure can this gas be expanded . and the results have been combined with the hydrate formation conditions given in Figure 5.130 Phase Properties 20 1 Methanol on 0. One of these diagrams is given by Figure 5.63 Gravity Gas (5-21) 2 Sodium Chloride on 0.

It follows a horizontal movement to the 160 o F isotherm. The intersection of the 800 psia line with the dashed line results in 65 o F (18.80 0. .32 was entered at pi = 3000 psia. .5 195 190 180 170 160 150 140 13 0 1 20 110 100 Initail Pressure [psia] 2000 1500 1000 800 600 500 400 300 50 90 80 70 60 200 150 40 100 100 150 200 300 400 600 1000 2000 3000 Final Pressure [psia] Figure 5.The next step consists of the vertical movement to the abscissa. [°F] 6000 200 5000 4000 3000 1 97.80 pi 20.The ordinate of Figure 5.32: Permissible expansion of 0.8 gravity gas without hydrate formation (from KATZ. 10000 8000 Initial Temp.3 o C) after expansion to 800 psia (5. 800 psia is the lowest final pressure which will preclude hydrate formation.68 [MPa] 3000 psia Ti 71[oC] 160 oF formation and what will the hydrate . .Phase Properties 131 SI-Units Gas Gravity Initial Pressure Initial Temperature adiabatically without temperature be? Solution steps: Field-Units γ g 0.52 MPa). 1945).

132 Phase Properties .

Chapter 6 pVT-Measurements Accurate crude oil data are necessary to control the production of a hydrocarbon reservoir efficiently.2 General Criteria Based on the general remarks about the types and behavior of hydrocarbon systems. the determination of its type. Therefore. It is obvious that .1 Sampling 6. 6.1. and its composition is assured to supply data for • geological and reservoir engineering evaluation and forecasting. it can be generally stated that the best time of sampling is at the beginning of production. volumetric and phase behavior. They are of importance for further calculations of the processes which take place inside a reservoir. conclusions can be drawn about general sampling criteria.1 Objectives The objectives of sampling are to receive samples from a suitable place in the production wells or surface facilities.except in case of dry and wet gases . The samples should represent the system in the reservoir under its initial conditions. 6. Only then.1.characteristic samples can be received only in the early life of the reservoir. 133 . • laboratory studies concerning enhanced oil recovery (EOR) methods modifying PVT properties (and the viscosity) of the system.

In every case. the production gas/liquid ratio. The wells should be allocated at characteristic sections of the reservoir. if the total composition of the system entering the well and leaving it at the wellhead remains constant.1 represent average characteristics which can be used as some rule of thumb. It is clear. There are certain approximations for selecting sampling methods which consider some of the properties of the produced fluids. .134 pVT-Measurements To meet all requirements. that the behavior of any system also depends on reservoir conditions which are missing on the tabulated data. Representative samples cannot be received from wells which are perforated at the water/oil or gas/oil contact and are producing water and/or non-equilibrium gas. To determine the fluid properties along with the depth of the pay zone (where monophase system exists). steady state conditions must be maintained. Samples can be taken from the flowing wells (subsurface sampling) or from the equilibrium phases of the separators (separator samples). 6. however. GOR. There are restrictions which must always be considered: • Considering the gravitational effect. • Reliable sampling conditions in the flowing wells and at the wellhead only exist. preliminary screening must be used cautiously. a minimum of one sample per 30 m can be recommended. • No characteristic fluid can be received within the flowing wells. Therefore. should remain constant during a long period of time. Number of samples will depend on the dip of the structure as well as on the pay thickness. it is clear that the sampling in closed wells could easily lead to misleading data. where a multiphase system exists or is developing. the careful selection of wells for sampling is very important. The third possibility is to get representative samples from the wellhead (wellhead sampling).3 Sampling Methods There are three methods of sampling. The method used depends on the type of system to be sampled. Moreover. Favorable conditions at steady state flow are indicated by stabilization of the pressure and temperature at the wellhead and in the separator. Values summarized in Table 6.1.

oil. In a wide region of flow rates.800 > 800 11000 . only those separators can be used where construction ensures highly efficient separation of the entering phases. Separators must contain pressure gauges and thermometers as well as devices for the accurate measurement of the gas volumes (liberated and/or delivered). the bottles and the connecting tube are filled up with an indifferent liquid (mercury or water from the separator or from other sources) to maintain pressure and to . The upper valve is connected with the sampling valve of the separator by a flexible tubing which is resistant to pressure and corrosion by phases sampled. To determine the phase boundary during the time of sampling. When highly paraffinic systems are produced exhibiting a cloud point in the vicinity of the separator temperature. The continuous monitoring of the concentration of some key components in the gas phase is recommended. and water exist. that no separator sampling is needed for dry gases and for saturated and undersaturated oils. but steady state conditions must exist. Sometimes. These bottles are in vertical position during sampling. Each phase can be sampled through sampling valves built in the separator at that part where pure gas. separators should be heated. and of water. Samples of the equilibrium phases should be take using specially designed bottles with valves at their extreme ends. This procedure should be considered to control the development and maintenance of steady-state flow.780 780 . Note.1: Screening sampling methods Type Density of the produced liquid [kg/m3 ] < 740 740 . For sampling. Sampling of equilibrium phases from separators can be used for all systems.20000 1000 . if no free water enters the production well and if there is a considerable length along which a1-phase oil. no liquid droplets should depart the system with the gas.pVT-Measurements 135 On the basis of the general remarks on the phase behavior of different systems in their Table 6. the following guidelines can be applied in screening sampling methods.13000 200 .1500 < 200 Gas/liquid ratio [m3 /m3 ] Dry gas Wet gas Gas/condensate system High shrinkage oils Low shrinkage oils initial state and on the changes in their behavior in the depression zone and in the wells. liquid hydrocarbons. separators often contain high pressure windows for visual control.

The sample is displaced by injection of the inert liquid into the sampler using a volumetric pump or .by gravitational effects. . the flow of a 1-phase hydrocarbon fluid must be assured. Whenever the temperature in the separator is lower than the ambient temperature. Sampling at the wellhead Sampling at the wellhead is a suitable process in case of the production of dry gas or undersaturated oil at wellhead conditions. It is then displaced by the sampled phase (gas or liquid). The protection of the samples in the shipping bottles and their correct marking for identification are important as it has already been mentioned. Bottom hole sampling Bottom hole sampling can be used when only the phase to be sampled is present at the perforated zone. After reaching the desired depth (usually with open valves to allow the flushing of the sampler by the flowing fluids). The valves are operated mechanically or electrically from the surface or within the sampler at the predetermined depth or time. additional 10% of mercury (or brine) should be removed by opening the lower valve. The previous remarks referring to the protection and identification of the shipping bottles are also valid for wellhead samples.is filled up with an inert liquid (e. the valves are closed.in a simple manner . The valves in the contains must be protected by steel caps during shipping. The method can be controlled by sampling at different depths and by analyzing the water content of samples. mercury).for systems which exist in a 1-phase state up to the wellhead.in general . samples can be taken above the water level. However. In doing so. After closing the valves into he container and in the separator and after dismounting the container.g. At the surface. Bottom hole samplers are specially designed devices containing one valve or two valves at their extreme ends. If water is also produced with the hydrocarbons but remains at the bottom under steady-state conditions. No sampling within the wells is needed for dry gases or . The container is connected with a valve at the wellhead and flushed with the fluid produced.136 pVT-Measurements exclude air. Data for identification must be delivered with the sample. the use of mercury (or brine) is recommended in sampling the liquid phase. The created gas cap will prevent uncontrolable pressure increase during shipping.together with the shipping bottle . They are run into the well through a sealed tube mounted to the christmas tree by a mechanically driven cable. about 90% of mercury (or brine) has to be displaced by the separator liquid. the sample should immediately be transferred to containers through a flexible steel tube which .

dissolved gases may be liberated from the water having a composition which could largely differ from that of the equilibrium hydrocarbon system. Afterwards. A similar approach can also be used for gas-condensate systems.1. Whenever water production cannot be avoided and the volume of the produced water is high.if the liquid is in equilibrium with other phase(s) . . pressure and temperature are increased up to the reservoir conditions. Sterile samplers and shipping bottles but also anaerobic conditions (in the course of sample transfer) must be guaranteed. As for liquids. the ratio of the phases entering the wells does not characterize the initially-in-place fluids (bubble point oil system). As a result. especially if water is also present. it is important • to prevent any loss . a similar procedure as above should be used to restore the composition of the hydrocarbon phase. primarily those of oils.especially in their lighter fractions -. This procedure is often necessary in case of reservoirs containing only water from which water will be produced for industrial or public supply. However. In this case. it is only possible to take samples from the equilibrium phases of the separator and to recombine them under separator conditions corresponding to their volume ratio. The basic concept of this method is that . the samples must be preserved in their original state. However. no processes in the reservoir/well system should take place which may result in a difference between the overall composition of the reservoir fluid and the produced fluid. In this case.pVT-Measurements 137 6. Specially designed sampling is needed to determine the microbiological contamination of the produced fluid(s).4 Special Problems There are sometimes problems arising from low permeability oil reservoirs (wells being produced with high depression). • to avoid the contact with air because oxidation processes could drastically modify the composition of hydrocarbon liquids.the composition of the oil phase is not influenced by the contacted volume of the other phase(s) being in equilibrium with the oil.

equipments and methods have been developed for the petroleum industry. Some known volume of an inert liquid is injected into the sampling bottle(s).1: Scheme of PVT equipments Samples to be studied are transferred to the P V T cell.2 Experimental Determination of the Volumetric and Phase Behavior To determine the volumetric and phase behavior of naturally occurring systems under reservoir/well/surface conditions. while the same liquid is withdrawn from the cell initially filled with it.138 pVT-Measurements 6.2.1 Equipment Gasometer Vg G Control unit Separator L Analysis pVT Cell Sample Hg p Gauge Volumetric Pump Figure 6. The selection criteria for the inert liquid are: • It must exhibit low volatility under (i) ambient conditions and also (ii) under temperatures and pressures which represent the working conditions. Sample . 6. A general scheme of the set of equipments used for PVT studies shown in Figure 6. • It should not react with the fluids contacted.1.

the accuracy of measurements depends on the resolution of the measuring instruments. the proportions of gas and liquid to be injected into the PVT cell must be equal to those measured into the field separator at the time of sampling. On the other hand. The cell has to be a heat.and temperature. Moreover. larger cell volumes are used to study (i) gases. liberated gas or condensate in the shipping bottle has to be dissolved or evaporated in advance. The cell is mounted into a thermostatic bath which ensures from detrimental temperature inconstancy in the cell.pVT-Measurements 139 • Its physical and chemical properties must be well known. In this case.2. Sometimes only one pump may be used for this purpose. if the liquid transported by the volumetric pump(s) is separated from the samples. Although many different PVT cells are specified.resistant windows built into the cell body. Nevertheless. relatively smaller ones are used to investigate characteristic oil systems. considering the change in volume of the system within the pressure range applied in the PVT studies. . 6. No restriction is needed. however. the samples will be transferred and measured by a piston built into the containers and the PVT cell. the cell is agitated or stirred inside. Therefore. where the volumetric and phase behavior can directly be observed through one or more pressure. The sample being transferred into the cell must be in a 1-phase state. Although mercury is widely used. if the inert liquid leaves the cell through a suitable operated valve at the bottom of the cell. Double-acting or two single-acting volumetric pump(s) are needed for this process. the relative error can be decreased by a reasonable increase of the sample volume studied. Referring to separator samples. sometimes low melting point alloys are used. This valve controls the rate and volume of the out-flowing inert liquid.2 PVT-Cells Evaluation of the PVT behavior is carried out on the system transferred into the PVT cell. The inside of the cell is lighted from outside by transmitted or reflected (from the inner wall of the cell) light. However. • windowed PVT cells. (ii) gas-condensate systems. The electrically heated (or cooled) medium circulating around the cell could be air or liquid. (iii) critical oils. “micro PVT cells” are not used in practice. To obtain the desired rapid establishment of the equilibrium conditions at any given pressure and/or temperature. Therefore.and pressure-resistant vessel which is equipped with valves made of non-corrosive materials. and (iv) water. two basic types can be distinguished: • blind PVT cells (without any possibility of visual observation of the system to be measured).

2: Blind PVT cell Blind PVT cells are usually used to study systems in which no phase transition takes place or phase transition can easily be determined using the isothermal p . oil systems.2. A characteristic blind cell is shown in Figure 6.140 pVT-Measurements Valve Thermometer Valve Steel Body Thermocouple Cell Bath Heat Isolator Jacket Heating Wire Hg Valve Figure 6.3). By opening .001[m3] which is large enough to be used for gas and oil studies as well. In V1 Out Pressure Gauge V2 Figure 6.V relationship (dry gases. water).3: PVT cell (after BURNETT) A special PVT cell for gases was developed by B URNETT (see Figure 6. Its volume is about 0. The chamber possessing the smaller volume contains the gas under given pressure. It contains two or more chambers with different volumes which are separated by valves.

A PVT cell with variable volume is presented in Figure 6.g. wet gases under well or separator conditions.5. No volumetric pumps are needed in this case. . The volume corresponding to that motion can be read on the calibrated stem of the plungers. the change in pressure can be measured. The volumes of the liquid phase can be determined by the angle of rotating the cell until liquid phases appear in the window.4.pVT-Measurements 141 successively the valves of the evacuate other chambers with known volume. It is also required whenever phase transition cannot exactly be determined by isothermal p .V relationships (volatile oils or partially miscible liquid/liquid systems with or without free gas phase). Windowed P V T cells Windowed PVT cells are usually used to evaluate systems where the knowledge of the volumes of coexisting phases is necessary (e. The condensed phase can be moved to the window by simultaneous lifting (or sinking) the two built-in plungers.4: PVT cell (after DEAN-POETTMAN) A typical one-windowed cell is shown in Figure 6. gas condensate systems). Gas Liquid Hg Windows Gas Liquid Hg Figure 6.

for Operation of the Piston 2 Piston with joist Packing 3 Hole and Valve for Introduction of Gasoline 4 Hole and Valve for Introduction and Sampling of Gas 5 Spiral for stirring Gas.6) is concerned. The scheme of this cell is shown in Figure 6. pressure can be kept at the desired constant value by the simultaneous lifting of the plunger. The Liquid is stirred by a Spiral Coil Wound around the axis 6 Sapphire Viewing Window 7 Viewing Window 8 Lighting Window 9 Liquid Volume Measurement Chamber 10 Magnetic Agitator 10 Figure 6. the condensed liquid is collected and can be lifted up to the windows which are built into the middle part. Meanwhile. In the lower part.6: PVT cell (after S LOAN) . This can be done by injecting inert fluid. 1 2 3 5 4 6 7 8 9 1 Orifice for Introduction of Mercury through the Upper Cap.) As far as S LOAN’s windowed cell (see Figure 6.7.5: Variable volume cell (after VELOKIVSKIY et al.142 pVT-Measurements Window Figure 6. It serves also for the measurement of the volume of phases. there are three sections. In the upper part. Three windows are built into the body of the most generally used cell as shown in Figure 6.8. a plunger is moved by the fluid injected or withdrawn by a volumetric pump. Another type contains one capillary glass tube within the cell through which the inner volume can be lighted and observed using an optical system.

8: RUSKA cell .pVT-Measurements 143 Capillary (Glass) Figure 6.7: PVT cell (after WELLS-ROOF) Windows Figure 6.

They are filled with the inert liquid used in the volumetric pumps. Pressure in the system is measured by gauges containing BOURDON-tubes. 6. The transported volume is proportional to the change in volume of the system studied.the so called positive displacement pumps . the latter sometimes with variable speed. In practice.4 Auxiliary Equipment The PVT cell.2. There are also gauges in which an electric signal is proportional to the pressure measured.3 Volumetric Pumps Volumetric pumps . A special type of volumetric pumps is the double-acting one: Two plungers are working simultaneously. Its general construction does not differ from the pump which is mechanically driven. valves are mounted in the network of tubes.are used to measure the volume of the inert liquid injected to and withdrawn from the cell respectively. A scheme of a hand-operated volumetric pump is shown in Figure 6. The pressure can be increased or decreased.2. a container is used which is also interconnected . In order to store the inert liquid (mercury). but in opposite directions.144 pVT-Measurements Figure 6. also dead weight gauges can be used for this purpose and for the calibration of the B OURDON-tube-gauges as well. hand. the inert liquid is displaced by a plunger from a pressure resistant and thermostatically controlled cylinder. These gauges are also built into the tube network. However. and other parts of the PVT system are interconnected by pressure-resistant stainless steel tubes. The movement of the connecting screw-spindle can be read on a calibrated rod and on a dial to increase the accuracy.and mechanically operated pumps are used.9.9: RUSKA volumetric mercury pump 6. the volumetric pump(s). In order to separate single parts of the equipment. In the generally used types.

In this case the flash process simulates the whole course of production. The temperature of the volumetric pump(s).simulates the behavior of the system far inside the reservoir where its composition is not influenced by the production. For this purpose gasometers are used. measurements should be carried out in a pressure region which correctly characterizes the phase transition and the 2-phase (or multiphase) behavior of the system. At each pressure step.carried out at reservoir temperature . Therefore. They are distinguishable by considering the change and the direction of change in the composition of the system. temperature. .pVT-Measurements 145 through valves with the different parts of the PVT system. 6. A scheme of the flash process has been presented in Figure 4.3.3 Methods Basically. in the PVT cells. 6. the volumes of the system and that of the existing phases are measured and the development of phases is observed. a laboratory separator must be inserted between the PVT cell and the gasometer.1 Flash Process In the course of flash processes. Whenever pressure. its volume must be measured. The lowest pressure applied is determined by the system studied. For systems in which no phase transition takes place. Measurements are usually repeated at other temperatures.3. If the gas phase liberated in the PVT cell will be displaced. pressure is gradually increased or decreased at constant temperature under equilibrium conditions. the lowest pressure selected could be the atmospheric one. too. the process can be automated and computerized. However. and in the thermostate around the cell is usually measured by thermocouples (may be controlled by thermometers). gas could condense under ambient conditions or at pre-selected pressures. there are three methods generally used in PVT studies. covering some range between reservoir and surface temperatures (in the separator or in the stock tanks). and the volume in the pumps are measured electrically. Measurements start at or above the reservoir pressure. Whenever phase transition occurs. In this case the flash process .

when • in the initial state of a gas condensate system. further information on the volume (and composition) of phases separated can be given. The differential process is repeatedly carried out. Both are constant volume processes. the liquid phase exists beyond the restriction of retrograde condensation. In the course of the differential study .3. the values of characteristic parameters will differ from those determined by the flash process. This can be done under initial conditions (gas cycling) or by starting from a given depleted state (gas injection with pressure increase). The development or disappearance of phases is observed. • partial or total evaporation of the immobile liquid condensate is needed. p is re-established by injecting a measured volume of the pre-selected gas. In contrast to the flash process during which the composition is maintained and the volume is varied.2 Differential Process The differential process describes the change in the volumetric and phase behavior of the system at reservoir temperature and under equilibrium conditions. There are two subgroups of the reverse differential process. the composition of the latter is also determined. In the course of a stepwise process. In depleted (gas condensate or oil) reservoirs. an isothermal pressure increase is realized under constant volume conditions. To follow the change in composition of the system and to characterize the displaced phase (gas). If the effect of pressure maintenance by gas injection is being studied after each step of the differential measurement.3 -. the differential process is carried out at constant volume and by variation of the system composition.146 pVT-Measurements 6. As a result of the gradually changing composition. where any decrease in the reservoir pressure results from the stepwise production of one of the equilibrium phases (usually liberated or free gas). usually down to the lowest possible reservoir (separator) pressure. In the course of gas injection (using a pre-selected gas). the effect of increasing the pressure by pre-selected gas can also be studied.3 Reverse Differential Process The reverse differential process is used to study the effect of any gas injection process on the volumetric and phase behavior of systems. the phase volumes are measured and phase transition(s) are observed at reservoir temperature. This method is used. the volumes of the system and those of the existing phases are measured. except the starting point representing the intact system under initial reservoir conditions. The highest pressure to be .3. By use of a laboratory separator in which pressure and temperature correspond to those during the field separation process.also explained in Figure 4. 6.

pVT-Measurements 147 reached corresponds to the fracturing pressure of the formation.it is about 10% higher than the initial reservoir pressure. The effect of gas injection is often the combination of the methods discussed above. or . it is obviously no need to continue the process in a higher pressure region.if this information is not available . If some 1-phase system develops below this pressure. .