PSI G - Equation of State Tutorial

29-03-01 1
Equation of State Tutorial
Jerry L. Modisette
14 September 2000
Introduction
As the art of pipeline flow simulation has advanced, the rigor and, we trust, the accuracy
of the many elements making up a flow model have increased. Steady-state models have
given way to transient solutions, fluid properties are calculated and tracked along the
pipeline, and configurations are represented in more detail. Isothermal models have been
replaced with solutions of the energy flow equation supported by real-fluid
thermodynamics and ground heat flow models. Accurate fluid properties and
thermodynamics require accurate equations of state.
There are many more equations of state than could be reasonably discussed in a single
paper. This tutorial reviews current practice using equations of state in the simulation of
fluid flow in pipelines, starting with fundamental considerations, following with a
discussion of several ancient & modern equations of state, and concluding by discussing
whats reasonable to use. Reasonable is a subjective word, and the decisions as to which
equations to consider and which to use for what are based on the authors experience in
simulating the flow of gases, liquids, supercritical fluids, two-phase systems, on
preferences arising from that experience, and on externally imposed requirements.
Whats an Equation of State?
An equation of state is a relationship between state variables, such that specification of
two state variables permits the calculation of the other state variables. There are many
state variables; usually in fluid dynamics we talk about pressure, temperature, and density
because these variables appear in the equations of motion.
Examples of equations of state are, for gases, the ideal gas law:
M
RT
P

= (1)
where P is the pressure, psia
is the density, lb/cu ft
R is the gas constant, psia-cu ft/deg R-pound-mole
T is the absolute temperature, deg R
M is the molecular weight
PSI G - Equation of State Tutorial
29-03-01 2
For liquids, we have the bulk modulus equation:
( )
!
!
"
#
$
$
%
&
+ =

o
o
P P
1 (2)
where is the bulk modulus, psi
and the subscript o indicates a reference condition.
We also have for liquids a thermal expansion equation:
( ) [ ]
o o
T T = 1 (3)
where is the thermal expansion coefficient, deg
-1
These ideal gas and liquid equations work reasonably well over limited temperature and
pressure ranges for many substances. However, pipelines commonly operate outside
these ranges and may move substances that are not ideal under any conditions. We will
examine the limitations on these ideal equations of state, and present some equations of
state with wider validity.

A Little Scientific Background
One of the first equations of state for gases was Boyles Law:
P
P
V
V
o
o
= (4)
where V is the gas volume,
P is the pressure,
and the subscript, o, refers to initial or standard conditions.
Boyles law expresses the observation that the volume of a gas decreases as the pressure
increases. Boyles law is only part of an equation of state, since it only involves two
variables. The second part is Charles Law:
o o
T
T
V
V
= (5)
where T is the absolute temperature.
Charles law expresses the observation that gases expand as they are heated.
Charles law contains a couple of significant concepts:
Absolute Temperature Actually, the original form of Charles law had the
temperature replaced by the temperature plus a constant. The constant was later
determined to be absolute zero in whatever temperature units were being used.
Zero Volume Molecules Since the volume goes to zero at absolute zero (according to
Charles law) the molecules must have zero volume. The correctness, or lack thereof of
this concept will become important as we examine equations of state.
PSI G - Equation of State Tutorial
29-03-01 3
The combination of Boyles and Charles laws, plus another conceptual jump, gives a form
of the ideal gas law:
nRT PV = (6)
where n is the number of moles of gas
Equation (6) is the formulation of the ideal gas law favored by chemists. It becomes
equation (1) by defining density as:
V
nM
= (7)

What Do We Want from an Equation of State?
In pipeline flow simulations we use equations of state for the following:
! Determine the density from the temperature & pressure for:
Linepack calculations
Flowmeter calibration
Pressure drop calculations
! Determine thermodynamic variables for;
Thermal modeling
Compressor calculations
Vapor-liquid equilibrium
These uses imply certain characteristics of an effective equation of state:
! Accuracy (<0.1% for custody transfer flow meters)
! Applicable over wide temperature and pressure ranges
! Applicable over wide range of compositions
! Rigorous (for thermodynamics; Not quite the same thing as accuracy)
! Works for liquids too
! Easy to use !!!
There is usually a contradiction between the last characteristic and the others.
PSI G - Equation of State Tutorial
29-03-01 4
How does the Ideal Gas Law Stack Up?
The ideal gas law was originally developed in the form of equation (6), rather than
equation (1) because it is easier to measure gas volumes than gas densities, and because
chemists tend to think in terms of moles rather than masses. Although the ideal gas law
was originally derived from Boyles law and Charles law, it can also be obtained from the
kinetic theory of gases.
Getting the ideal gas law from the kinetic theory of gases requires a couple of assumptions
which give us some insight into the physics:
The gas molecules occupy no volume, which was already implied by Boyles law.
There are no forces between the molecules except at the instant of collision.
For a gas at atmospheric (standard) conditions, these two assumptions are nearly
satisfied. There is so much space between the molecules that the forces between the
molecules are only significant when they are very close compared to the average distance
between them, and the volume is nearly all comprised of empty space. One thing I learned
in freshman chemistry was that a gram-mole of gas occupies 22,400 cubic centimeters at
atmospheric conditions. If the gas is water, a gram-mole is 18 grams. The density of liquid
water is 1 gram/cubic centimeter, so the volume of the liquid is 18 cubic centimeters.
Since the molecules of a liquid nearly fill up all the space, the molecules of a mole of
gaseous water at atmospheric conditions occupy less than 1/1000 of the total gas volume.
The average distance between these molecules is the cube root of 22,400 divided by
Avogadros number, 6.02e
23
, which works out to be a bit over 10
-6
centimeters. This is a
rather small distance, but the range of the intermolecular forces is about 10
-8
centimeters.
Therefore, for a gas under standard conditions, the distance between the molecules is
nearly always beyond the reach of the intermolecular forces.
Looking at the ideal gas law in the context of the list of desirable characteristics, we have:
! Accuracy Good at low densities (generally low pressures and/or high temperatures)
! Temperature and pressure ranges Not good near dew point or critical point
! Composition range All composition effects are in the molecular weight
! Rigor Not so good: The Joule-Thompson effect, which is caused by intermolecular
forces, is zero for ideal gases.
! Liquids Hopeless: For a liquid, the basic assumptions dont even come close. The
molecules of a liquid are always in contact with one another, and are always under the
influence of forces exerted by their neighboring molecules. This is why liquids are very
difficult to compress.
! Easy to use !!! A winner!
The pressure limitation alone is enough to drive a requirement for better equations of
state for pipelines.

PSI G - Equation of State Tutorial
29-03-01 5
Somewhat Better Gas Laws
Van der Waals
We will consider the Van der Waals equation of state for the following reasons:
! Its historical interest
! It is based an a better physical understanding of how gases work
! It provides a useful improvement for some gases and conditions
! Some of the modern, more accurate equations of state (SRK and PR) are based on it
It was observed early on that the gas law didnt quite work for higher pressures and lower
temperatures. Van der Waals determined that the effect of the long range forces and the
volume occupied by the molecules was approximately accounted for by the equation:
( ) nRT nb V
V
a n
P =
!
!
"
#
$
$
%
&
+
2
2
(8)
where: b a, are constants, known appropriately as the Van der Waals constants.
It should be fairly obvious that b is a volume and is, approximately, the volume of a mole.
The term nb V is the free volume the molecules have to run around in.
It may not be quite so obvious that
2
2
V
a n
corresponds to an attractive force between
molecules. This attractive force makes the pressure less than it would be for an ideal gas,
hence the positive sign.
The Van der Waals equation works reasonably well for pressures below 200 psi and
temperatures above 0 deg F, provided that the gas isnt close to condensing. Its being
used successfully right now for a low pressure HCl gas pipeline.
I was amused to see a note in a handbook saying: It is known that a and b vary to some
extent with temperature.! This comment obviously dates from a time when high
pressures and/or very low temperatures didnt often occur.
The Van der Waals equation can be expressed in terms of density by substituting
V
nM
= from equation (7): nRT nb
nM
nM
a n
P =
!
!
"
#
$
$
%
&

!
!
!
!
!
"
#
$
$
$
$
$
%
&
!
!
"
#
$
$
%
&
+

2
2

(9)
Which, with a little manipulation, simplifies to:
RT b
M
M
a
P =
!
!
"
#
$
$
%
&

!
!
"
#
$
$
%
&
+

2
2
(9a)
PSI G - Equation of State Tutorial
29-03-01 6
The Universal Gas Law
The universal gas law is:
M
RT z
P

= (10)
where z is the compressibility, dimensionless.
Note: We sometimes talk about the super-compressibility,
v
F z is related to the super-
compressibility, by the equation:
( )
2
1
v
F
Z = .
The wonderful thing about the universal gas law is that it will describe any gas; you just
need to know the value of z . The not-so-wonderful thing is that determining z can be
quite a chore, and z varies with pressure and temperature. The universal gas law does not
solve the problem of obtaining an accurate equation of state over a large range of
pressures and temperatures. It recasts the problem into the determination of z .
There are a couple of advantages to the universal gas law:
One is that the value of z is a measure of how far the gas is from ideality. At
atmospheric conditions, the value of z is typically around 0.99. Under pipeline
conditions, the value is typically around 0.9. For condensed hydrocarbons, the value is
typically less than 0.5.
Another advantage is that the universal gas law can used in model and thermodynamic
calculations based on the ideal gas law, with z and the thermodynamic quantities
themselves determined from a better equation of state in a separate process.
The Gas Constant
We have discussed several equations of state in which the gas constant, R , appears. The
gas constant is a universal constant of nature, whose value depends on the units used.
Note that the units of the gas constant are always energy/deg-mole, which happens to be
the same as those of molar specific heat.
One handbook gives values of the gas constant for 84 sets of units. The table below gives a
few values, sometimes used in pipelines.
Gas Constant, R Pressure
Units
Volume
Units
Temperature Units Moles
10.7335 psia cu ft Deg R lb-moles
1545 psfa cu ft Deg R lb-moles
8.314 pascals cu m kelvins g-moles
8314.4 pascals cu m kelvins kg-moles
83,144 bar cu m kelvins kg-moles

PSI G - Equation of State Tutorial
29-03-01 7
The Virial Equation of State
The virial equation of state is:
nRT
V
T C
V
T B
PV
'
(
)
*
+
,
+ + + = ...
) ( ) (
1
2
(11)
This equation is rather obviously similar to a Taylors series expansion in
V
1
. Again the
problem has been reformulated into finding the temperature dependent coefficients , ,C B
etc. These coefficients themselves may be expanded in power series in temperature. The
number of coefficients tends to get out of hand. About 20 years ago an equation of state of
the virial type was published which had 28 constant coefficients, which was still
inaccurate as the gas approached condensation. However, we shall see even more
coefficients.
PSI G - Equation of State Tutorial
29-03-01 8
What Makes a Good Equation of State?
A good equation of state should be simple, accurate, and should cover a wide range of
pressures, temperatures, and compositions. Such an equation of state does not exist.
There are equations of state that are accurate over large ranges of pressure and
temperature, even near the dew point or critical conditions. They are not simple! And
changing the composition means changing the equation of state, which is another
involved process.
For natural gas pipelines, we want equations of state that are accurate over the conditions
and compositions over which gas pipelines operate. Furthermore, we would like to be able
to handle composition changes in a straightforward way.
Special Pipeline Equations of State
For natural gas transmission pipelines in the United States, the gas quality, or the
composition, is usually kept within a range that permits the use of equations of state of
moderate complexity. Examples of such equations of state are NX-19 (AGA-3) and the
AGA-8 gross characterization method.
For pipelines there are two somewhat different applications, which may affect the choice
of equation of state:
Custody Transfer Although physical accuracy is important, the legal and financial
aspects of custody transfer applications impose a primary requirement that the parties
agree on the equation of state to be used. In the past this requirement has led to the
use of standard equations or procedures even when there were known inaccuracies.
Fortunately, the current AGA recommended equations are accurate as over the range
of gas quality, pressures, and temperatures used for transmission pipelines in the U.S.
Simulation Accurate simulations require equations of state that are physically
accurate. For such fluids as LPGs and ethylene, and for gas gathering operations
equations of state with a wider range of applicability are required to support accurate
simulations.
There have been a series of special pipeline equations of state developed over the years:
NX-19, Sarum, and the two AGA-8 equations. These equations of state have progressively
improved in physical accuracy over the range of pressures, temperatures, and
compositions occurring in U.S> transmission pipelines. They have tended to be almost
purely empirical
PSI G - Equation of State Tutorial
29-03-01 9
The AGA-8 Equation of State
There are actually two AGA-8 equations of state, called the detail characterization method
and the gross characterization method.
The AGA-8 detail characterization method equation of state is:
( ) ( )
'
'
(
)
*
*
+
,
+ + =
- -
=

58
13
*
18
13
*
exp 1
n
k
n
b k
n n n
u
n
n
u
n
n n n n n
D c D D k c b T C T C D Bd dRT P (12)
where d is the molar density of the gas
B is the second virial coefficient ( ) (T C in equation (11))

c
D

= is the reduced density

*
n
C are coefficients which are functions of the composition

n n n n
k c b u , , , are constants.
The AGA-8 gross characterization method equation of state is:
[ ]
2
1 d C d B dRT P
mix mix
+ + = (13)
where
mix mix
C B , are the second & third virial coefficients, respectively. The mix subscript
indicates that the calculation of the virial coefficients for the mixture does not consider
the detailed composition, but considers hydrocarbons collectively (as an equivalent
hydrocarbon) plus terms for CO
2
and nitrogen.
The AGA Transmission Measurement Committee Report No. 8: Compressibility Factors
for Natural Gas and Other Related Hydrocarbon Gases (1992) gives constants,
recommended pure gas parameters, and mixing rules for equations (12) and (13).
NX-19 Equation of State
The NX-19 equation of state was used for many years for transmission gas pipelines until
it was replaced by better equations. It is more of a procedure than what we normally think
of as an equation. The procedure is found in the 1962 AGA Manual for the Determination
of the Supercompressibility Factors for Natural Gas: PAR Research Projecy NX-19
Extension of Supercompressibility Tables. This paper compares NX-19 results with
other equations of state, primarily so that those still using NX-19 can get an idea of the
discrepancies.
PSI G - Equation of State Tutorial
29-03-01 10
Real-Fluid Equations of State
Real-Fluid means that the equation of state has no idealizations such as are used in the
ideal gas law, the bulk modulus equation, the thermal expansion equation, or even the
Van der Waals equation. Since no one has been able to develop an equation of state valid
near critical or dew points from first principles, this means that such equations are largely
based on empirical fits to data.
In principle, given enough data and a willingness to do a lot of uninspiring but difficult
work, the accuracy of an empirical fit, and the complexity of the system described are
unlimited. For example, the USCGS puts out an empirical fit based on a spherical
harmonic expansion of the Earths magnetic field that covers the entire earth, including
the effects of iron ore bodies, etc. There are 600 terms in the equation.
The AGA-8 detail characterization method uses an equation of state with 58 coefficients,
which are determined from 58 sets of 10 parameters, plus several parameters for each
component, plus a mixing parameter for each pair of components. (Theyve almost caught
up with the USCGS!) The equation still doesnt work for liquids, or near critical
conditions.
We will consider three widely used equations of state that do work reasonably well near
the dew point, and for both liquids and gases: Soave-Redlich-Kwong (SRK), Peng-
Robinson (PR), and Benedict-Webb-Rubin-Starling (BWRS). In addition to covering a
wide range of conditions, these equations also can be expressed in generalized forms with
mixing rules that permit the calculation of the coefficients for different compositions. For
this reason these equations, and AGA-8 are sometimes called compositional equations of
state.
SRK and PR, along with the Van der Waals equation, are called cubic equations of state,
because expansion of the equations into a polynomial results in the highest order terms in
density (or specific volume) being cubic, or third power terms. BWRS adds fifth & sixth
power and exponential density terms.
The cubic equations are all of the form:
B AV V
a
b V
RT
P
+ +
+

=
2
(14)
If V is the molar volume and A and B are zero, (14) becomes the Van der Waals
equation, (8). For the more general equations, A b a , , and B are functions of temperature
which must be determined from empirical fits to the data.
PSI G - Equation of State Tutorial
29-03-01 11
The Soave-Redlich-Kwong (SRK) Equation of State
The SRK equation is a modification by Soave of the Redlich-Kwong (RK) equation, which
had been widely used for chemical equilibrium calculations. The SRK equation produces
better liquid densities, although BWRS is even better. For the SRK equation, B in
equation (14) becomes zero, and b A = . a and b are given by:
( )( ) [ ]
2
5 . 0 2
2 2
1 176 . 0 574 . 1 48 . 0 1
42748 . 0
r
c
c
T
P
T R
a + + = (15)
c
c
P
RT
b
08664 . 0
= (16)
where is a measure of the gas molecules deviation from spherical symmetry called
the Pitzer acentric factor

c
r
T
T
T = is the reduced temperature
and the subscript, c , refers to critical conditions.
The numerical constants and the
r
T dependence are selected so as to produce fits to
hydrocarbon vapor pressures.
From equation (16) the Van der Waals interpretation of b as the volume of the molecules
says that the molecules occupy about 1/12 of the gas volume at critical conditions.
With a and b from (15) and (16), the SRK equation is:
( )( ) [ ]
V
P
RT
V
T
P
T R
P
RT
V
RT
P
c
c
r
c
c
c
c
08664 . 0
1 176 . 0 574 . 1 48 . 0 1
42748 . 0
08664 . 0
2
2
5 . 0 2
2 2
+
+ +
+

=

(17)
Many workers have produced variations on the SRK equation of state involving fits (by
adjusting the numerical constants) to other sets of data, and even modifying the form of
the temperature dependence. These variations will not be addressed here.
Equation (15) may be expressed in terms of density with

M
V = :
( )( ) [ ]
c
c
r
c
c
c
c
MP
RT
T
P M
T R
MP
RT
M
RT
P

08664 . 0
1
1 176 . 0 574 . 1 48 . 0 1
42748 . 0
08664 . 0
1
2
5 . 0 2
2
2 2 2
+
+ +
+

= (17a)
PSI G - Equation of State Tutorial
29-03-01 12
The Peng-Robinson (PR) Equation of State
In the PR equation of state, b A 2 = ,
2
b B = ,
( )( ) [ ]
2
5 . 0 2
2 2
1 26992 . 0 54226 . 1 37464 . 0 1
45724 . 0
r
c
c
T
P
T R
a + + = (18)
and
c
c
P
RT
b
07780 . 0
= (19)
so that the PR equation is:
( )( ) [ ]
2
2
2
5 . 0 2
2 2
07780 . 0 07780 . 0
2
1 26992 . 0 54226 . 1 37464 . 0 1
45724 . 0
07780 . 0
!
!
"
#
$
$
%
&
+
+ +
+

=
c
c
c
c
r
c
c
c
c
P
RT
V
P
RT
V
T
P
T R
P
RT
V
RT
P

(20)
where, again, the numerical constants were obtained by fits to hydrocarbon vapor
pressure data.
In terms of density, equation (20) becomes:
( )( ) [ ]
2
2
5 . 0 2
2
2 2 2
07780 . 0 07780 . 0
2 1
1 26992 . 0 54226 . 1 37464 . 0 1
45724 . 0
07780 . 0
1
!
!
"
#
$
$
%
&
+
+ +
+

=
c
c
c
c
r
c
c
c
c
MP
RT
MP
RT
T
P M
T R
MP
RT
M
RT
P

(20a)
Limitation on Cubic Equations of State
Van der Waals, SRK, and PR have the common problem that they are not adequate at
high densities, either for liquids or supercritical fluids. The reason lies in the nature of the
intermolecular force. At distances greater than the typical separation between molecules
in a liquid, the force is attractive, and changes as rather low power of the distance between
molecules. Once the molecules are in contact, the force becomes repulsive, and increases
exponentially as the distance decreases. This is another way that saying that the molecules
are incompressible. Pushing the molecules closer together requires deforming the
molecules, which they resist strongly.
Representing the rapid increase in the repulsive intermolecular force with distance
requires a dependence on density which is stronger than a cubic relationship.
One way to get the stronger dependence is to carry the virial equation (11) out to more
terms. The problem is that many terms are required. In addition to the complexity, a large
number of coefficients to be evaluated from the data can make the resulting function less
general.
PSI G - Equation of State Tutorial
29-03-01 13
The BWRS equation avoids this problem by adding exponential terms, in addition to
higher order density terms.
PSI G - Equation of State Tutorial
29-03-01 14
The Benedict-Webb-Rubin-Starling (BWRS) Equation of State
The BWRS equation is a modification of an equation of state first published by Benedict,
Webb & Rubin (BWR) in 1940. The density dependence of BWR is retained in BWRS; the
temperature dependence of the coefficients is changed. In 1973 Starling wrote a book,
Fluid Thermodynamic Properties of Light Petroleum Systems, Gulf Publishing
Co. with BWRS coefficients for light hydrocarbons, mixing rules for determining
coefficients for mixtures of hydrocarbon gases, and with procedures for determining
thermodynamic quantities and for vapor/liquid equilibrium calculations.
Probably because of its ability to cover both liquids and gases and the availability of
coefficients and mixing rules for many hydrocarbons in one place, BWRS is probably the
most widely used equation of state for simulation of pipelines with high density
hydrocarbons, or with condensation. In one application on a liquid ethane pipeline, the
ethane flashed into a gas in part of the pipeline due to an upset condition. The real-time
model, using the BWRS equation of state, continued to simulate the flow in the part
liquid, part gas pipeline and even detected a leak test that happened to be under way
during the upset.
Simplicity is not among the good qualities of the BWRS equation of state. The form of the
equation is:
!
!
"
#
$
$
%
&

!
!
"
#
$
$
%
&
+ + !
"
#
$
%
&
+ + !
"
#
$
%
&
+
!
!
"
#
$
$
%
&
+ + =
2
2
2
2
2 3
3
6
6
3
3
2
2
4 3 2
exp 1
M M T M
c
M
T
d
a
M
T
d
a bRT
M T
E
T
D
T
C
A RT B
M
RT
P
o o o
o o

Where the eleven coefficients, , , , , , , , ,
,
d c b a E D C B A
o o o o o
and , must be determined
empirically. Values of the coefficients may be determined either by fitting data for the gas
of interest, or by calculation from the composition using mixing rules, critical properties
of the pure compounds, and the generalized coefficients provided by Starling.
Once the coefficients are known for the fluid of interest, all state variables can be
calculated from two known state variables. Unfortunately, many flow models determine
pressure and temperature from the flow equations, leaving the density to be determined
from the equation of state. Since BWRS is implicit in density, an iterative density
calculation is required. Profilers run on flow models using BWRS usually show that the
model is spending most of its time in the density calculation routine.
PSI G - Equation of State Tutorial
29-03-01 15
Equations of State for Mixtures
The SRK and PR equation as given are for pure compounds. Mixing rules for determining
the effective critical properties for mixtures are given in The Properties of Gases and
Liquids by Reid, Prausnitz, and Poling.
Starling (1973) provides detailed mixing rules for determining coefficients for the BWRS
equation.
Results from Selected Equations of State
Equations of state are most easily compared in terms of the compressibility, z. The
fractional difference in the calculated standard volumes is the same as the fractional
difference in z for two equations. In Figure 1, Van der Waals, NX-19, AGA-8 (detailed
characterization) and BWRS are compared for pure methane over a range of pressures at
0, 60, and 120 deg F.

Figure 1a Methane - 0 deg F
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
0 500 1000 1500 2000
Pressure, psia
z
-
f
a
c
t
o
r
Van der Waals
NX-19
AGA-8
BWRS
PSI G - Equation of State Tutorial
29-03-01 16



It can be seen that NX-19 and AGA-8 agree well over the entire range. BWRS agrees with
the NX-19 and AGA-8 up to about 1500 psia. As expected, Van der Waals agrees with the
Figure 1b Methane, 60 deg F
0.80
0.82
0.84
0.86
0.88
0.90
0.92
0.94
0.96
0.98
1.00
0 500 1000 1500 2000
Pressure, psia
z
-
f
a
c
t
o
r
Van der Waals
NX-19
BWRS
AGA-8
Figure 1c Methane - 120 deg F
0.86
0.88
0.90
0.92
0.94
0.96
0.98
1.00
0 500 1000 1500 2000
Pressure, psia
z
-
f
a
c
t
o
rVan der Waals
NX-19
AGA-8
BWRS
PSI G - Equation of State Tutorial
29-03-01 17
other equations at low pressures. At high pressures, Van der Waals is off because of the
inadequacy of a single parameter in representing the effect of the intermolecular forces.
At this point its worthwhile to discuss the source of the NX-19, AGA-8 and BWRS
coefficients. NX-19 and AGA-8 are empirical fits to a large number of
pressure/density/temperature measurements for gases of pipeline quality. The significant
meaning of pipeline quality for equation of state is that the heavy hydrocarbon and
water contents are limited.
The BWRS coefficients used herein are primarily based on empirical fits to vapor-liquid
equilibrium data for pairs of components over a large range of compositions. Generally,
PR and SRK coefficients are also based on vapor-liquid equilibrium data.
In view of the data upon which they are based, it is no surprise that NX-19 and AGA-8 do
not work very well for gases near condensation. It is, perhaps, more of a surprise that
BWRS works so well for gases far from condensation.
Effects of Composition
Effects of Composition
Generally, heavier hydrocarbon molecules are larger and have a greater affinity for one
another. Both of these factors make heavier hydrocarbons deviate more from an ideal gas.
The AGA-8 report gives the compositions of several natural gases for reference purposes.
Table 1 gives the compositions for some for these gases, with components heavier than
butane added to the n-butane mole percent.
Component Gulf Coast Amarillo Ekofisk High CO2-N2
Methane 96.5222 90.6724 85.9063 81.2110
Ethane 1.8186 4.5279 8.4919 4.3030
Propane 0.4596 0.8280 2.3015 0.8950
i Butane 0.0977 0.1037 0.3486 0.1510
n Butane 0.2468 0.2720 0.4495 0.1530
N2 0.2595 3.1284 1.0068 5.7020
CO2 0.5956 0.4676 1.4954 7.5850

Figure 2 shows results from AGA-8 and BWRS for these gases. The SRK and PR give
results similar to BWRS.

PSI G - Equation of State Tutorial
29-03-01 18

The general trend is lower compressibilities (deviations from ideality) as the proportion of
heavier components increase. This result is to be expected, because the heavier molecules
have a greater attraction for one another. Because the C5+ components were lumped in
with n-butane, the figures slightly under estimate this effect.
Figure 2a AGA-8 Compositions
0.70
0.75
0.80
0.85
0.90
0.95
1.00
0 500 1,000 1,500 2,000
Pressure, psia
z
-
f
a
c
t
o
r
Methane
Gulf Coast
Amarillo
Ekofisk
High CO2-N2
Figure 2b BWRS Compositions
0.70
0.75
0.80
0.85
0.90
0.95
1.00
0 500 1,000 1,500 2,000
Pressure, psia
z
-
f
a
c
t
o
r
Methane
Gulf Coast
Amarillo
Ekofisk
High CO2-N2
PSI G - Equation of State Tutorial
29-03-01 19
Condensation
A major reason for using BWRS, PR, or SRK is their applicability near condensation
conditions. BWRS results for propane are shown in Figure 3.
It can be seen that the BWRS equation gives results through the condensation region.
However, a
Caveat is in order, as can be seen in Figure 3b. The results are nearly the same as in 3a,
with the addition of points within the density range 4-30 lb/cu ft. The behavior of the
curve is strange, even going to a negative pressure. This is because these densities are not
possible for propane at this temperature (60 deg F). However, the equation of state, which
Figure 3a BWRS Propane Liquifaction
0.00
0.20
0.40
0.60
0.80
1.00
0 500 1,000 1,500 2,000
Pressure, psia
z
-
f
a
c
t
o
r
0.00
8.00
16.00
24.00
32.00
40.00
D
e
n
s
i
t
y
,

l
b
/
c
u

f
t
Z Density
Figure 3b BWRS Propane Liquifaction
0.00
0.20
0.40
0.60
0.80
1.00
-100 400 900 1,400 1,900
Pressure, psia
z
-
f
a
c
t
o
r
0.00
8.00
16.00
24.00
32.00
40.00
D
e
n
s
i
t
y
,

l
b
/
c
u

f
t
Z Density
PSI G - Equation of State Tutorial
29-03-01 20
is an empirical fit, will calculate a pressure for any density given it. Care must be taken to
reject invalid densities.
Conclusions
Ideal equations of state are generally inadequate for custody transfer conversions or for
accurate pipeline flow simulations. Accurate equations of state are available; they tend to
be complex. There are two types of state equations, differing in how the coefficients are
fitted to data. The equations recommended by AGA (NX-19, AGA-8) are fitted directly to
pressure/temperature/density data over the range of conditions and compositions
commonly found in U.S. transmission pipelines. SRK, PR, and BWRS are fitted to
vapor/liquid equilibrium data.
Generally, the AGA equations are more accurate for normal transmission pipeline
conditions and compositions. SRK, PR, and BWRS cover a wider range of conditions.
They are expected to be more accurate for gathering systems operating close to
condensation.
For a transmission pipeline using an AGA equation for flow meter conversions and SRK,
PR, or BWRS for a simulation model, the differences between the equations should not be
significant, in view of the uncertainties in other pipeline parameters, in particular, the
temperature profile along the pipeline.
SRK, PR, and BWRS are valid for liquid hydrocarbons as well as gases, and can be used to
determine vapor/liquid equilibrium. In the condensation region, for pure compounds care
is needed not to force non-physical densities intermediate between gases and liquids. For
mixtures in the condensation region, the mixture compensation itself may be non-
physical. That is, no single fluid exists at that composition under those conditions. The
fluid separates into a liquid and a gas, either of which has the mixture composition. In
such a case, the vapor/liquid equilibrium must be determined, and two-phase flow must
be taken into account.
All of the available real-fluid equations are implicit in density, so that calculation of the
density from the pressure & temperature requires iterative methods. This fact argues for
formulating the flow equations in terms of density rather than pressure. However, at
boundaries with measured pressures, iterative calculations of boundary densities will still
be required. Perhaps someone will take on the labor of re-formulating one of the
equations of state in with density as the dependent variable.
PSI G - Equation of State Tutorial
29-03-01 21
Biography

Jerry L. Modisette, Ph. D.
Dr. Modisette is an independent pipeline software consultant, currently working with
Energy Solutions International (formerly LICEnergy and Wright-Logue Associates).
Dr. Modisette started his professional career with NASA, where he worked on boundary
layer diffusion, solid rocket design, and astrophysics. During the Apollo program he was
Chief of the Space Physics division at NASA MSC in Houston where he was responsible
for protection of the astronauts from space radiation hazards.
In 1969 Dr. Modisette became a professor of physics, associate dean, & research director
at Houston Baptist University. In 1971, while in academia, he began development of
pipeline technology, beginning with leak detectors based on rarefaction waves, and
culminating with the first real-time pipeline simulator installed in 1978. He also worked
in other fields, patenting inventions for measurement of properties of drilling fluids,
generation of energy from ocean waves, vapor recovery, and refrigeration.
In 1980 he left the university to develop pipeline applications software full time. He was
one of the founders and the provider of technology for a series of companies, CRC
Bethany International, Real Time Systems, and Advanced Pipeline Technologies (APT).
During this time he developed much of the simulation and applications technology that
form the basis of the industry today. He also trained many of the people who are now
leading consultants, technologists, or managers in the industry.
In 1992 he founded Modisette Associates, Inc., which took over the business of APT.
Modisette Associates operated successfully, doing major pipeline applications projects in
North America. These projects included replacing the model and leak detection system he
had installed on a Canadian pipeline in 1978 with updated technology. This pipeline,
running a Modisette real-time model from 1978 to the present, is the longest record of
continuous operation of a pipeline model in the world.
In 1998, Modisette Associates was acquired by LICENERGY, Inc., where Dr, Modisette
was Chief Scientist until January 2000.
Other developments by Dr. Modisette include:
! The first transient pipeline model with an accurate thermal model. (1979)
! The first transient two-phase flow model, featuring physically based, continuous flow
regime transitions, and rigorous vapor-liquid equilibrium calculations. (1982)
! The first real-time model to calculate the mixing by turbulent diffusion of products at a
batch interface. (1978)
Education:
B. S. (Mathematics) Louisiana Polytechnic Institute, 1956
M. S. (Physics) Virginia Polytechnic Institute, 1960
Ph. D. (Space Science) Rice University, 1967