Вы находитесь на странице: 1из 43

Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour

1
Upstream Process
Engineering Course
1. Hydrocarbon Phase Behaviour

Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
2
Reservoir Barrel
METHANE &
ETHANE
CARBON DIOXIDE
HYDROGEN SULPHIDE
SAND
WATER
LOWEST
BOILING POINT
HIGHEST
BOILING POINT
gases
naphtha
gasoline
kerosene
diesel oils
Lubricating
oils
Fuel oil
residue
Propane and butane gas for lighter fuel
& camping stoves
Chemicals for medicines, plastics, paints,
cosmetics & clothing materials
Petrol for vehicles
Jet fuel and paraffin
Diesel fuel
Machine oil, waxes and polishes
Fuel for ships and central heating
Bitumen for road surfaces and roofing
materials
SUBSTANCE USES
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
3
Reservoir Fluid Types
Four general types of reservoir
fluids
Black Oil - Forties
Volatile Oil - Gyda
Condensate - Bruce
Gas - SNS Villages
Typical P-T diagrams showing the position of
the critical point for all four reservoir fluid types
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
4
Phase Equilibrium Definitions
Phase Diagram
A record of the effects of temperature, pressure and composition on the kinds and numbers of
phases that can exist in equilibrium with each other
Bubble Point
The point at which the first infinitesimally small vapour bubble appears in a liquid system. The
bubble point curve on a phase diagram represents 0% vapour
Dew Point
The point at which the first infinitesimally small droplet of condensation forms in a gaseous
system. The dew point curve on a phase diagram represents 0% liquid
Phase Envelope
The area on a pressure-temperature phase diagram for a mixture enclosed by the bubble and
dew point curves. This area represents the set of conditions for the mixture were vapour and
liquid phases co-exist in equilibrium.
Cricondenbar
The maximum pressure at which vapour and liquid can co-exist in equilibrium
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
5
Phase Equilibrium Definitions

Cricondentherm
The maximum temperature at which vapour and liquid can co-exist in equilibrium
Critical Point
At the critical point, liquid and vapour phases of a fluid have identical physical properties
Quality Lines
Lines through the two-phase region showing a constant percentage of liquid and vapour
Retrograde
The name given to phase behaviour above the critical temperature and pressure were vapour
and liquid phases coexist and the amount of vaporisation or condensation changes with pressure
and temperature in the opposite direction to normal behaviour.
Equation of State
An equation which describes the relationship between pressure, temperature and molar volume
of any homogenous fluid at equilibrium
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
6
Phase Equilibrium Definitions

Fugacity
A thermodynamic concept arising from the consideration of the change in Gibbs free energy
with changes in pressure and temperature for non-ideal mixtures and substances. Three types of
fugacity can be defined: pure component fugacity, mixture fugacity , partial fugacity . Fugacity
must have the same units as pressure by definition but will only be equal to pressure under ideal
conditions (e.g. low pressure gases).
Binary Interaction Coefficient
A constant which accounts for the deviation from ideality for component pairs in mixtures.
These are specific to one equation of state as they are calculated by the regression of measured
data. The coefficients used in a liquid activity method are essentially regression constants used
to calculate the liquid activity coefficient
Liquid activity coefficient
The ratio of the partial fugacity of a component in a mixture to its pure component fugacity at
the same physical conditions, divided by its mole fraction in the mixture.
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
7
HYSYS EOS Set Up
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
8
Black Oil
Component Mol %
Methane (CH
4
) 48.83
Ethane (C
2
H
6
) 2.75
Propane (C
3
H
8
) 1.93
Butanes (C
4
H
1 0
) 1.6
Pentanes (C
5
H
1 2
) 1.15
Hexane (C
6
H
1 4
) 1.59
C7+ 42.15
Molecular Weight of C7+ 225
GOR (scf /bbl) 625
Oil Gravity ( API ) 34.3
Colour Black
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
9
Volatile Oil
Component Mol %
Methane (CH
4
) 64.36
Ethane (C
2
H
6
) 7.52
Propane (C
3
H
8
) 4.74
Butanes (C
4
H
1 0
) 4.12
Pentanes (C
5
H
1 2
) 2.97
Hexane (C
6
H
1 4
) 1.38
C7+ 14.91
Molecular Weight of C7+ 181
GOR (scf /bbl) 2000
Oil Gravity ( API ) 50.1
Colour Brown
HYSYS is not
flawless
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
10
Gas Condensate
Component Mol %
Methane (CH
4
) 87.07
Ethane (C
2
H
6
) 4.39
Propane (C
3
H
8
) 2.29
Butanes (C
4
H
1 0
) 1.74
Pentanes (C
5
H
1 2
) 0.83
Hexane (C
6
H
1 4
) 0.6
C7+ 3.8
Molecular Weight of C7+ 112
GOR (scf /bbl) 18200
Oil Gravity ( API ) 60.8
Colour Straw
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
11
Dry Gas
Component Mol %
Methane (CH
4
) 95.85
Ethane (C
2
H
6
) 2.67
Propane (C
3
H
8
) 0.34
Butanes (C
4
H
1 0
) 0.52
Pentanes (C
5
H
1 2
) 0.08
Hexane (C
6
H
1 4
) 0.12
C7+ 0.42
Molecular Weight of C7+ 157
GOR (scf /bbl) 105000
Oil Gravity ( API ) -54.7
Colour White
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
12
Fluid Sampling
Drill Stem Tests
Carried out on appraisal wells to provide sub-surface
information and data for facilities design
Laboratory and field measurements made
Basis for verifying simulation models
Extended Well Test
Testing over many months
Sub-surface uncertainty reduced
Opportunities to undertake extensive facilities testing
Environmental implications require to be evaluated
Topsides Samples
Regular samples are taken to maintain product quality
and ensure levels of chemical treatment are sufficient
Pressurised samples are taken of crude at normal
operating conditions for laboratory analysis and
atmospheric samples
Bottom Hole Samples
Capture pressurised reservoir
sample e.g asphaltene analysis
QA for drilling fluid
contamination
For fast track development
sometimes only sample
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
13
Reservoir fluids consist of a mixture of mainly hydrocarbon components and
water. Other commonly found components of reservoir fluids include carbon
dioxide(CO
2
), hydrogen (H
2
), hydrogen sulphide (H
2
S) and nitrogen (N
2
).
Common hydrocarbon components include the following classes of compound.
Alkanes or Paraffins (both straight chain and isomeric)
Cycloparaffins
Aromatics
Naphthalenics
It is not practical to model every component of a reservoir fluid due to the large
numbers of components that are present. Generally, for acceptable phase
behaviour prediction, it is sufficient to specify the mole fractions of the main light
end paraffins, typically C1 to C10 for black oils. If dealing with gas condensate
systems, however, more rigorous compositional data may be required, particularly
if modelling retrograde behaviour.

Fluid Analysis
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
14
Crude Oil Composition Schematic
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
15
For the heavier components in a reservoir fluid mix one or more pseudo-
components (or hypothetical components) can be used.
Pseudo-Components are a mixture of many components with different properties
but modelled as one component with generalised physical properties.
Pseudo-Components are generally defined either by the critical properties or by
average molecular weight, specific gravity and normal boiling point. For most
computer simulations a minimum of two of these must be specified. Acentricity
factors and binary interaction parameters can also be entered; most simulation
packages will estimate these if not supplied by the user. The more information
that is given on a pseudo component, however, the more accurate the phase
prediction can be.
The pseudo component facility can also be used to enter data into a simulation for
a component not contained in the component library of the simulation package
being used.
Pseudo-Components
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
16
Heavy End Fluid Characterisation
Prediction of heavy end physical properties is
important to good prediction of phase
behaviour; incomplete or inaccurate data can
radically affect phase calculations. For
example the prediction of bubble and dew
points.
Standard fractional distillation tests can be
used to produce boiling point curves and
physical properties so that the heavy ends can
be split into a number of pseudo components.
Some simulation packages have the facility to
take in distillation test data directly and to
produce the pseudo-components
automatically.
Some tuning may be required to match the
properties of the modelled reservoir fluids to
well test data.
Stabilised Crude Boiling Curve
0
10
20
30
40
50
60
70
80
90
9
0
1
2
0
1
5
0
1
8
0
2
1
0
2
4
0
2
7
0
3
0
0
3
3
0
3
6
0
3
9
0
4
4
0
5
0
0
TBP Cut Deg C
T
o
t
a
l

D
i
s
t
i
l
l
a
t
e

W
t
%
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
17
Pseudo-Components
Component Analysis
0
10
20
30
40
50
C
O
2
C
2
n
C
4
C
6
C
9
C
1
2
C
1
5
C
1
8
C
2
1
C
2
4
C
2
7
C
3
0
C
3
3
C
3
6
+
E
t
h
y
l
b
e
n
z
e
n
e
Component
M
o
l

%
43 Components simplify for
simulation
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
18
Pseudo-Components
C7 plus Components
0
1
2
3
4
5
6
7
8
C
7
C
9
C
1
1
C
1
3
C
1
5
C
1
7
C
1
9
C
2
1
C
2
3
C
2
5
C
2
7
C
2
9
C
3
1
C
3
3
C
3
5
B
e
n
z
e
n
e
E
t
h
y
l
b
e
n
z
e
n
e
Component
M
o
l

%
Pseudo Component Simplification
0
2
4
6
8
10
12
14
16
P
s
e
u
d
o

C
7
P
s
e
u
d
o

C
8
P
s
e
u
d
o

C
9
P
s
e
u
d
o

C
1
3
P
s
e
u
d
o

C
1
7
P
s
e
u
d
o

C
1
9
P
s
e
u
d
o

C
2
3
P
s
e
u
d
o

C
2
7
P
s
e
u
d
o

C
3
1
P
s
e
u
d
o

C
3
7
P
s
e
u
d
o

C
4
5
P
s
e
u
d
o

C
6
2
M
o
l

%
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
19
Fluid Modelling
The choice of thermodynamic model is critical when evaluating phase behaviour
Fluid phase behaviour and properties can be predicted using a either a Black Oil model
or a Compositional model
Black Oil models are generally used for multiphase modelling in flowlines and
wellbores. They predict phase behaviour and bulk properties for varying conditions by
relating the fluid volumetric properties at the surface (e.g. B
o
, GOR, Oil Gravity and Gas
Gravity) to the down-hole conditions. No detailed knowledge of fluid composition is
required. Black oil models are not recommended for gas condensate systems particularly
where there is retrograde condensation.
Compositional models are used for rigorous heat and mass balance work. They use a
property package and a composition specified in terms of individual components which
are available in the package database or component library. If required, pseudo-
components can be specified for non-library components or for boiling point cuts. Binary
interaction parameters for components pairs can be used to give improved VLE property
predictions.
Modern simulation packages generally have several of property packages available for
selection including both equations of state and activity models.
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
20
Property Package Selection
Selection of the property package must take into account the components and the
operating conditions
For most purposes, Peng Robinson (PR), Soave-Redlich-Kwong (SRK), or a modified
version of these methods, will provide sufficiently accurate modelling for oil and gas
field applications.
The PR and SRK equations of state give accurate modelling for systems containing up
to 5% N
2
, CO
2
or H
2
S. For systems with greater than 5% N
2
, CO
2
or H
2
S these
equations of state are still recommended if the system does not include free water. It
may be advisable, however, to utilise user defined binary interaction parameters if
available. This will depend upon the simulation/flash package being used and the quality
of the data in its components library. It is recommended to consult the user guide (or
help desk) for the particular package being used.
Polar compounds can be problematic for EOS methods - methanol, glycol and water -
empirical correction factors are often required to improve accuracy. Most simulation
packages offer specialised property packages to handle such systems which utilise a
combination of an EOS to predict vapour phase fugacity coefficients and an activity
coefficient model for the liquid phase.

Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
21
Fugacity
Fugacity is a conceptual term which is related to Gibbs free energy, temperature,
volume and pressure, the units of fugacity are the same as pressure
Although fugacity has the same units as pressure it will only be equal to pressure under
ideal conditions
The fugacity coefficient for a pure component is defined as:



For a component in a gas mixture, the fugacity coefficient is:



Where y
i
is the mole fraction of component y in the vapour phase

P
fi
i = |
P y
f
i
i
i = |
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
22
Equilibrium Ratios
The vapour-liquid equilibrium constant (K) is defined as the mole fraction of any
component in the vapour phase divided by the mole fraction of the same component in
the liquid phase
K
i
= y
i
/ x
i


An equation of state can be used to calculate accurate K values when the vapour and
liquid phases co-exist in equilibrium
For equilibrium, the fugacity of a component in the vapour phase must be equal to the
fugacity of the component in the liquid phase, when this criterion is met:
K
i
= |
iL
/ |
iv


Use of Equilibrium Ratios is key to the preparation of oil and gas processing heat and
mass balances
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
23
Application of K Values
The basic application of K values is the calculation of dew points, bubble points and the
vapour-liquid behaviour inside the phase envelope (flash calculations)
Bubble Point Calculation
E K
i
x
i
= E y
i
= 1.0
Assume a temperature for the known pressure (or assume pressure if temperature is known)
Find K
i
at pressure and temperature known and assumed
Multiply K
i
by the corresponding x
i
(eqn. 1)
If summation of values is 1.0, then pressure and temperature is correct, if not repeat until E K
i
x
i

= 1.0, within accuracy limits
Dew Point Calculation
E (y
i
/K
i
) = E x
i
= 1.0
The steps involved in a dew point calculation are the same as for the bubble point calculation, using eqn. 2 in
place of eqn. 1
A dew point calculation is less exact than a bubble point calculation, especially for lean
gases containing a small amount of heavy ends

Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
24
Flash Calculation
Flash calculations determine the amount of vapour and liquid in a two phase
system
F = V + L (Overall)
F z
i
= V y
i
+ L x
i
(For each component)
Where
F = mols of total feed
V = mols of gas leaving system for F mols of feed
L = mols of liquid leaving system for F mols of feed
z
i
= mols of component i in the feed stream per mol total feed
y
i
= mol fraction of component i in the gas stream (V)
x
i
= mol fraction of component i in the liquid stream (L)
Let F = 1 and substitute y
i
= K
i
x
i

and
if Ex
i
or Ey
i
= 1.0 then a solution has been obtained
0 . 1
) (
=
+
=

i
i
i
VK L
z
x
0 . 1
) (
=
+
=

i
i
i
K
V
L
z
y
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
25
Soave-Redlich-Kwong
SRK is robust and well proven, ideal for use in oil & gas and refinery
applications. It is suitable for cryogenic conditions and pressures up to 300
bara.
For the highest accuracy binary interaction parameters for component pairs
should be used. Correlations can be used to generate binary interaction
parameters for pseudo-components.
Results for mixtures of hydrogen and hydrocarbons are good; those for
aromatics less so but these can be improved using appropriate binary
interaction parameters.
Can handle up to 25% H
2
S with acceptable degree of accuracy if binary
interaction parameters are used.
Generally not considered suitable where accuracy is required for systems with
polar compounds.
Liquid compressibility predictions are accurate enough for fugacity
calculations but not for accurate liquid densities (can be 10 - 20% low).

Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
26
Soave-Redlich-Kwong
Phase behaviour in the critical region can be predicted although calculations
are somewhat unstable at the critical point itself.
SRK is not considered to be accurate for heavy end VLE prediction and
vacuum systems
The variables a and b are derived from the Van-der-Waals equation, which
was based on a model where molecules were represented by hard spheres
which behaved in a classical and predictable fashion
The parameter b represents the hard-sphere volume of the molecules
The parameter a represents the intermolecular attraction

Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
27
The temperature dependant term a/\T of the
Redlich-Kwong equation was replaced by a
function a(T, e) involving the temperature
and acentric factor by Soave

Fugacity Coefficient
ln | = z - 1 - ln (z - B) - A/B ln (1 + B/z)
Mixtures
ao = EEy
i
y
j
(ao)
ij
b = Ey
i
b
i
A = EEy
i
y
j
A
ij
B = Ey
i
B
i
(ao)
ij
= (1 - k
ij
)[(ao
i
)(ao)
j
]
0.5
Soave-Redlich-Kwong
Standard Form

Polynomial Compressibility Form
z
3
- z
2
+ (A - B-

B
2
)

z - AB = 0
Parameters
A = (aoP)/(R
2
T
2
) = 0.42748oP
r
/T
r
2

B = (b P)/(RT) = 0.08664P
r
/T
r

a = O
a
R
2
T
c
2
/P
c
= 0.42748R
2
T
c
2
/P
c
b = O
b
RT
c
/P
c
= 0.08664RT
c
/P
c

b) V(V
a
b) (V
RT
P
+

=
2
5 . 0
172 . 0 574 . 1 48 . 0
where ) 1 ( 1
e e + =
+ =
m
Tr m
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
28
Peng-Robinson
Peng-Robinson is robust and well proven, ideal for use in oil & gas and
refinery applications. It is generally similar in performance and applicability to
SRK except as follows.
It is generally accurate over a wider range of conditions.
It is more accurate around the critical point.
Liquid phase density prediction is more accurate.
Improved PR packages can be used for heavy end VLE prediction and vacuum
distillation
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
29
Peng-Robinson
Validity is roughly the same as S-
R-K, but is more accurate near the
critical point

Mixtures
ao = EEy
i
y
j
(ao)
ij
b = Ey
i
b
i
(ao)
ij
= (1 - k
ij
) [(ao )
i
(ao )
j
]
0.5

A = EEy
i
y
j
A
ij

B = Ey
i
B
i
A
ij
= (1 - k
ij
)(A
i
A
j
)
0.5
k
ii
= 0
Standard Form

Polynomial Compressibility Form
z
3
- (1- B)z
2
+ (A - 3B
2
- 2B)z - (AB - B
2
- B
3
) = 0
Parameters
a= 0.45724 [(R
2
T
c
2
)/P
c
]
b = 0.07780 RT
c
/P
c

o = |1 + (0.37464 + 1.54226e - 0.26992e
2
) (1-Tr
0.5
)]
2

A = (aoP)/(R
2
T
2
) = 0.45724oP
r
/T
r
2

B = (bP)/(RT) = 0.07780P
r
/T
r
Fugacity Coefficient
ln | = z - 1 - ln (z - B) - [A/2(2B)
0.5
] ln [(z+2.414B)/(z-0.414B)]

)
b
- 2bV
V
(
a
b) (V
RT
P
2 2
+

=
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
30
Notation
Definitions
a Attraction parameter
b Residual volume parameter
O SRK Parameter (numerical co-
efficient)
A Derived parameter
B Derived parameter
z = PV/RT, compressibility factor
a
i
=
^
f
i
/f
0
, activity of a species in a
mixture
^
f
i
Partial fugacity of species
i
k
ij
Binary Interaction Parameter
(values of kij for around 100 pairs are
available)
a
ij
Cross-parameter


Greek Symbols
o Parameter
oij Volatility of species i relative to that of
species j
| = f
i
/P, Fugacity Coefficient
e Acentric Factor
Molar Density
= a
i
/x
i
, Activity Coefficient of species i
Subscripts
c Value of any variable at the critical
point
r Reduced Value
pc Pseudo-critical value
Superscripts
0 property of a standard state, e.g. f
0

Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
31
Accentric Factor
The acentric factor describes the
change in the intermolecular attraction
component, a with temperature
In the S-R-K equation of state, the
variation of factor a of the component
i with temperature is given by:
a
i
= a
ci
o
i
(T)
where a
ci
is the value of a derived
from the critical values
a
ci
= 0.427 R
2
T
ci
2
/ P
ci

and o
i
(T) = (1+m
i
(1-T
ri
0.5
))
2
The variable m
i
is a function of the
acentric factor e
i


e
i
= 0.480 + 1.574 e
i
- 0.172 e
i
2
Compound Ac ent ric Fac t or
Nit rogen 0. 039
Carbon Diox ide 0. 239
Met hane 0. 011
Et hane 0. 099
But ane 0. 119
Hex ane 0. 299
Oc t ane 0. 398
Dec ane 0. 489
The table shows typical acentric factors,
the value increases with the size of the
molecule and polarity
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
32
Binary Interaction Coefficients
The cubic Equations of State were originally developed for pure substances
and have been subsequently adapted for mixtures
This adaptation introduced a measure of the polar and other interactions
between pairs of dissimilar molecules
For a component in a mixture:


where z
i
and z
j
are mol fractions of components i and j respectfully
and

where k
ij
is the binary interaction coefficient
Binary Interaction Coefficients represent a flexible way of modelling the ideal
EOS to match the non-ideal reality of many mixtures

= ij j i j i a z z a
) 1 ( ) ( ij j i ij k a a a =
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
33
EOS Bubble Pressure Routine
Use critical and
accentric values to
calculate P-R
parameters for
single components.
Initialise
interaction
parameters.
Use mixing
rules to
calculate mixed
P-R
parameters.
Calculate Z for
vapour and
liquid.
Calculate
fugacity.
Iterate on the vapour phase
composition until the vapour
phase species fugacity is
equal to the liquid phase
species fugacity.
Input feed
composition, guess
vapour composition
and bubble pressure.
Check vapour and liquid
mole fractions sum to 1. If
not iterate on pressure.
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
34
EOS Flash Calculation
Given
z
j
, p, T
Assume
K
j

Perform Flash
Calculations
Calculate

L
j

Calculate
K
j

Calculate
C
v
j

Solution
Test for
Convergence
y
j
,
n
v
X
j,
n
L

Yes
No
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
35
Example
Calculation of the saturation pressure of n-Pentane at 100 C using an Equation of State
At equilibrium, the fugacity of the liquid and vapour phases is equal
The experimentally determined vapour pressure is 5.86 atm
Critical properties of n-Pentane are:
T
c
= 469.7 K, P
c
= 33.25 atm, e = 0.251
Using S-R-K :



Calculate A & B


Substitute values of A & B into the polynomial compressibility equation:

Solve to find Z for liquid and vapour phases
Calculate the fugacity of the vapour and liquid phases:
ln | = z - 1 - ln (z - B) - A/B ln (1 + B/z)
|
V
= f
V
/P
V and
|
L
= f
L
/P
V

At equilibrium, f
V
= f
L
therefore |
V =
|
L
2
172 . 0 574 . 1 48 . 0 where ) 1 ( 1 e e o + = + = m Tr m
( ) ( ) ( ) 094 . 1
6 . 469
373
1 251 . 0 172 . 0 251 . 0 574 . 1 48 . 0 1
2
=
|
|
.
|

\
|
+ + = o
( )( ) P T Tc P P A c 0224 . 0 / / 42747 . 0
2
= = o
( )( ) P T Tc P P B c 0033 . 0 / / 08664 . 0 = =
0 ) (
2 2 3
= + AB Z B B A Z Z



P V apour Li qui d V apour Li qui d Rat i o
4 0. 8943 0. 0184 3. 665 5. 196
5 0. 8647 0. 0229 4. 4774 5. 2287 0. 8563
6 0. 831 0. 0275 5. 2482 5. 2512 0. 9994
7 0. 7955 0. 0321 5. 9775 5. 2807 1. 132
z f
By interpolation of the above table, the predicted vapour pressure is 6.01 atm
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
36
E.o.S Comparison: Peng Robinson v
Zudkevitch Joffee
9 barg
54 C
1.1 barg
53 C
50 mbd stabilised crude
PR ZJ
MW 18.53 18.43
Q (mmscfd) 15.0 14.6
C1 (mol%) 88.57 88.94
C2 (mol%) 7.17 6.87
C3 (mol%) 0.13 0.12
PR ZJ
MW 22.09 21.54
Q (mmscfd) 0.9 1.2
C1 (mol%) 74.75 76.34
C2 (mol%) 16.33 15.80
C3 (mol%) 0.45 0.41
Reservoir
Fluids
GOR PR ZJ
1st Stage (scf/bbl) 299 292
2nd Stage (scf/bbl) 19 24
Overall (scf/bbl) 318 316
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
37
Phase Behaviour Pipeline Gas
Components Gas A
C1 86.5473
C2 7.30256
C3 2.01471
IC4 0.475166
NC4 0.566198
IC5 0.213075
NC5 0.230081
C6 0.120042
C7 0.100035
C8 0.060021
N2 0.880308
CO2 1.49052
SUM 100
Molar weight 19.1
HC dewpoint 23.5 C @
(Cricondentherm) 40.1 bara
HC dewpoint 101.2 bara
(Cricondenbar) @ -16.4
C
Gullfaks Fluid from SOW
0
10
20
30
40
50
60
70
80
90
100
110
120
130
-40 -35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35
Temperature (C)
P
r
e
s
s
u
r
e

(
b
a
r
)
BWRS
PR
RKS
Cricondentherm
Cricondenbar
Application of BWRS erroneous Pipeline two-phase
Minimum
Pipeline
Operating
Pressure
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
38
High Pressure Compressor Simulation
At pressure above ca 250 Bar P-R and SRK
may not accurately represent the system
thermodynamics
Other EOS options should be considered
Lee Kessler Plocker - LKP
P-R with Lee Kessler Plocker Enthalpy PR-LK
Corresponding States Method - Infochem
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
39
Limitations of Phase Behaviour Prediction
All phase behaviour methods assume an equilibrium is reached, this is an idealised
situation and in reality equilibrium is not always achieved.
EOS can not be used to predict water/hydrocarbon separation
Phase behaviour prediction deals with mass transfer effects and therefore does not take
into account bulk fluid flow. In practice there is almost inevitably some bulk transfer of
fluids between the phases: emulsions may be formed due to high shear rates across
control valves and the addition of corrosion inhibitor which holds water in the oil phase,
oil will carry over into the oil phase, liquid droplets carry over into the gas phase.
Prediction methods are generally only valid over a particular range of conditions and for
particular components.
EOS do not accurately predict phase behaviour at the critical point.
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
40
Polar Fluids
For the majority of calculations, simple treatment of hydrocarbon/water behaviour is
sufficient, however for systems containing large amounts of polar fluids (water,
methanol or glycol) more rigorous treatment is required.
For polar systems it is generally recommended to use an activity model to predict liquid
phase behaviour. Although work is being undertaken to extent the validity of EOS
methods into the liquid phase (e.g. Peng-Robinson-Stryjek-Vera [PRSV])
For water/oil mixtures the accuracy of calculations depends on the availability and
accuracy of the binary interaction data available, at low pressures NRTL will give
accurate results providing all the data required is available.
For methanol/water mixtures, NRTL provides accurate results at pressures below 1000
kPa.
Many of the commercially available simulators provide methods for such systems.
Expert advice is usually available from their technical support.
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
41
Activity Models
The most common liquid activity methods are:
NRTL - Non-random two liquid method
Useful for highly non-ideal systems where very sharp separations are required
Valid for both liquid-liquid and vapour-liquid-liquid systems, excellent for aezotropic column
Accurate liquid density data is essential
Should not be used if the temperature of any component is above the critical temperature
Should not be used for pressures above 1500 kPa
UNIQUAC
This method utilises phase equilibrium data to predict phase data for systems where no such
information is available
Activity coefficients for mixtures are related to interactions between structural groups and
corrections are then applied to take into account differences in molecular size and molecular
interactions
UNIQUAC is an approximate method and should not be used for detailed design
Works best at pressures below 500 kPa and temperatures between 300 K and 425 K
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
42
PVT Matching
The Equation of State model is generally accurate for predicting phase behaviour for
normal oil systems. Physical property estimates such as density require careful
evaluation. For example, the P-R EOS underpredicts by 5-10% the density of light
hydrocarbons (C1 to C5) and overpredicts the heavier ones
Since all volumetric oil properties, FVF and GOR for example are derived from specific
gravity, it is important that density is modelled accurately. Also many items of
equipment will require accurate prediction of physical properties - pumps, heat
exchangers, compressors etc.
The liquid density derived from EOS methods is very sensitive to property and
compositional changes hence matching to experimentally determined results is often
undertaken
PVT matching involves manipulating the boiling point and associated Tc, Pc and
Acentric Factor of the pseudo-components. Matching requires great care and guidance
should be sought if unfamiliar with the techniques
Condensates can be much more difficult to match than oils as the use of Binary
Interaction Parameters is not as effective in this area
Upstream Process Engineering Course Prepared by Genesis Oil and Gas Consultants Ltd Hydrocarbon Phase Behaviour
43
Crude Assays
An Assay is a detailed fluid analysis of
crude oil, sampled under strictly
controlled conditions
Assay reports include the quantities and
properties of distillates used for crude
valuation purposes, the quantity of
waxes, heavy metals, salt and sulphur
and physical properties of the fluid such
as viscosity
Assay samples are taken monthly for the
purposes of crude valuation purposes
and annually for crudes being exported
via a pipeline shared with other fields
SPR CRUDE OIL COMPREHENSIVE ANALYSIS
Sample ID BAYOU CHOCTAW SWEET Date of Assay 18/09/2000

Crude
Specific Gravity, 60/60 F 0.8447 Ni, ppm 3.50 RVP, psi @ 100 F 4.62
API Gravity 36.0 V, ppm 5.49 Acid number, mg KOH/g 0.084
Sulfur, Wt. % 0.36 Fe, ppm 0.844 Mercaptan Sulfur, ppm 7.021
Nitrogen, Wt. % 0.114 H2S Sulfur, ppm 0
Micro Car. Res., Wt. % 2.22 Org. Cl, ppm 0.3 Viscosity: 77 F 6.874 cSt
Pour Point, F 31 UOP "K" 11.94 100 F 4.623 cSt
Fraction Gas 1 2 3 4 5 6 Residuum Residuum
C2 - C5 - 175 - 250 - 375 - 530 - 650 -
Cut Temp. C4 175 F 250 F 375 F 530 F 650 F 1050 F 650 F+ 1050 F+

Vol. % 1.7 7.3 8.1 14.2 16.3 10.0 31.8 42.4 10.7
Vol. Sum % 1.7 9.0 17.1 31.3 47.6 57.6 89.3 100.0 100.0
Wt. % 1.2 5.8 7.1 13.1 15.9 10.1 34.3 47.0 12.7
Wt. Sum % 1.2 7.0 14.1 27.1 43.0 53.1 87.4 100.0 100.0
Specific Gravity, 60/60 F 0.6730 0.7396 0.7763 0.8240 0.8526 0.9116 0.9349 1.004
API Gravity 78.8 59.8 50.8 40.2 34.5 23.7 19.9 9.4
Sulfur, Wt. % 0.0043 0.0040 0.0123 0.07 0.21 0.57 0.69 1.04
Molecular Weight 97 111 136 184 246 425
Hydrogen, Wt. % 15.89 14.65 na 12.99 10.61
Mercaptan Sulfur, ppm 14.6 10.1 22.5 17.4
H2S Sulfur, ppm 0.03 0.8 0.7 0.02
Organic Cl, ppm 2.1 0.5 0.5 0.6
Research Octane Number* 68.4 61.1 42.3
Motor Octane Number* 66.5 58.6 40.0
Flash Point, F 77 171 246 303
Aniline Point, F 122.4 144.1 164.3 193.3
Acid Number, mg KOH/g 0.03 0.10
Cetane Index 47.1 53.2
Diesel Index 62.2 58.0 56.6
Naphthalenes, Vol. % 4.42 8.20
Smoke point, mm 20.3 16.8
Nitrogen, Wt. % 0.0015 0.006 0.108 0.240 0.603
Viscosity, cSt 77 F 2.473
100 F 1.951 4.795
130 F 3.312 37.03 95.3
180 F 14.22 28.28 5671
210 F 1722
275 F 249.3
Freezing Point, F -28.54
Cloud Point, F 24.0 106
Pour Point, F 19.9 102 75
Ni, ppm 7.539 26.2
V, ppm 11.81 41.0
Fe, ppm 3.856 13.87
Micro Car. Res., Wt. % 5.07 18.17
* = calculated from gas chromatographic data