SPE 93517

Tuning EOS Using Molecular Thermodynamics To Construct Asphaltene Deposition

Envelope (ADE)
M.A. Fahim, Kuwait U., and S.I. Andersen, IVC-SEP, Technical U. of Denmark

Copyright 2005, Society of Petroleum Engineers Inc.

Molecular Thermodynamics approach
This paper was prepared for presentation at the 14th SPE Middle East Oil & Gas Show and
Conference held in Bahrain International Exhibition Centre, Bahrain, 12–15 March 2005.
Crude oil is considered to be containing asphaltenes
This paper was selected for presentation by an SPE Program Committee following review of
information contained in a proposal submitted by the author(s). Contents of the paper, as
and resins in continous media of other hydrocarbon.
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to Asphaltenes are represented by attractive spheres that can
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at associate with themselves. Resins are represented by attractive
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
sphere chains that can associate with asphaltenes, but not with
for commercial purposes without the written consent of the Society of Petroleum Engineers is themselves. Asphaltene molecules and resin chains are
prohibited. Permission to reproduce in print is restricted to a proposal of not more than 300
words; illustrations may not be copied. The proposal must contain conspicuous dissolved in a continuous medium that affects dispersion-force
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
interactions among asphaltenes and resin chains. The
Helmholtz energy of the system is given by an expression
Abstract based on random-phase-approximation theory . Statistical-
associated-fluid theory within the McMillan-Mayer
The objective of this work to calculate ADE of description of liquid solutions. The Helmholtz energy A
asphaltene from molecular thermodynamics. Aphaltenes and contains five contributions as given by Chapman et al , 1993
resins are represented by pseudo-pure components, and all
other components in a crude oil are presented by a continuous
medium that affects attractions among asphaltene and resin A= Aid+ Ahs + Avdw + Aassoc + Achain (1)
molecules. Model parameters are evaluated systematically
from average properties of asphaltenes and resins in crude oils, Where N is the number of association cites and superscript id
and from dispersion-force properties of the oil medium. Given stands for an ideal-gas mixture of asphaltenes and resin chains
the composition of the medium, and asphaltene and resin at system temperature, volume, and molecular number
concentrations, the molecular-thermodynamic, calculated densities; hs stands for the contribution for hard-sphere
results for the effects of oil composition and pressure on repulsion when resin chains are disconnected; vdw stands for
asphaltene precipitation are in good agreement with the contribution from van der Waals dispersion-force
experimental measurements for different reservoir fluids, for attractions; assoc stands for the contribution from asphaltene-
several crude oils. The absolute average error was 9.8% asphaltene and asphaltene-resin associations; and chain stands
for the entropic contribution to A from connectivity of resin
Introduction chains

Asphaltenes are heavy, black materials dissolved in In this work only asphaltene-asphaltene and
most petroleum. When petroleum is produced or transferred, asphaltene- resin associations will be considered assuming that
asphaltenes can precipitate, plugging-up equipment and each asphaltene molecule has N identical association sites, and
thereby shutting down operations. Every effort is made by oil each resin chain has only one association site. Asphaltene can
companies to avoid asphaltene precipitation because, when it associate with each other, but resin chains can only associate
occurs, the cost of repair can be very large. It is important to with asphaltenes. The Helmholtz energy due to association is
establish the conditions (temperature, pressure, composition) given by Chap man et al., 1990
which make asphaltene does not precipitate, so it requires to
have information about onset pressure to prevent the A assoc ⎛ 1− x α ⎞ ⎛ 1 − x β ⎞ (2)
= NX A ⎜ ln x α + ⎟ + X R ⎜ ln x β + ⎟
precipitation. Because of deviation of Equation of state, it can RT ⎝ 2 ⎠ ⎝ 2 ⎠
not used to predict the asphaltene onset pressure , the where XA and XR are, respectively, mole fractions of
existence of model to predict the asphaltene onset point is the asphaltene molecules and resin chains on a medium-free basis;
major task. This work is directed at establishing a reliable a xα is the fraction of association sites at asphaltene molecules
model based on molecular-thermodynamic theory which
that are not bonded; xβ is the fraction of association sites at
corrects the deviation of EOS.
resin chains that are not bonded. Fractions xα and xβ are given
2 SPE 93517

by: Z = ZSRK + Zassoc (10)

This pressure and compressibility factor are also called PCPA

and ZCPA , referring to Cubic EOS plus association (CPA).
( )
−1
x α = 1 + N ρ A ∆αα x α + ρ R ∆αβ x β (3)
The flash calculation are carried using ZCPA. With this value of
the total volume and Xα mole fraction of asphaltene free site,
Xβ mole fraction of resin site the pressure contribution from
( )
−1 association can be calculated from equation (7).
x β = 1 + N ρ A ∆αβ x α (4)
Experimental data for asphaltene onset ptrssures were
where ρA , ρΒ are as before the number densities of asphaltene obtained from different literature sources Jamaluddin 2004,
molecules and resin chains in the oil medium; ∆αα, ∆αβ are Bruce and Nghiem 2004 , Burke et al 1990 , Behar et al 2003
given by ,and Sorensen 2004. Four set of data at upper, saturation and
lower pressure with different temperature were used to the
∆αα = g 11hs (σ 11 )[Exp (E αα / kT ) − 1]σ 113 κ αα (5) association model by tuning the parameter (E/kT) αα and
(E/kT) ββ ghsij. This two values can be measured
experimentally by micro-Calorimeter technique developed by
Andersen et al (2004). It was found that ghsij, the contact value
∆αβ = g 12hs (σ 12 )[Exp (E αβ / kT ) − 1]σ 123 κ αβ (6) of hard sphere pair has a value ranging 1.0 – 1.2 as reported by
Wu , 2000 and a value of 1.2 was used. If ghsij was used as
where ghsij (σij) is the contact value of the pair correlation fitting parameter , it was found that it converge to a value of
function in the sphere mixture of asphaltene and resin about unity. The objective is to minimize the error in
segments καα and καβ are parameters related to the square-well calculating the onset pressure as :
widths for the association potentials. To reduce the number of
unknown parameters, it can be assumed that they are identical ⎛ P exp − P calc ⎞
constants for associations between asphaltene and asphaltene, error = ⎜⎜ ⎟⎟ * 100 (11)
⎝ P exp ⎠
and between asphaltene and resin chain.

To apply this equations to asphaltene precipitation at reservoir by changing the adjustable parameters (E/kT) αα and (E/kT) ββ
conditions, it is needed first to establish a reliable method to
estimate required model parameters. The association Discussion
contribution to pressure is given by expression
For quantitative applications, molecular parameters
Passoc ∂ ⎛ Aassoc ⎞ (6) must be estimated independently or from a few experimental
=− ⎜ ⎟
RT ∂V ⎝ RT ⎠ data for the particular system of interest. The model
parameters of this model include the diameter of the
The pressure contribution from association is calculated from asphaltene molecule segment σA and the diameter of resin
equation (2) and (6) as: segment σR , asphaltene-asphaltene association energy
(E/kT)αα , the asphaltene-resin association energy (E/kT)αβ ,
the number of association sites of each asphaltene molecule N
Passoc 1⎛ ⎛ 1 − xα ⎞ ⎛ 1 − xβ ⎞ ⎞ (7)
= − ⎜⎜ N X A ⎜ ln xα + ⎟ + X R ⎜⎜ ln χ β + ⎟⎟ ⎟
⎟ , and the volume parameter kα. Also the segment densities of
RT V⎝ ⎝ 2 ⎠ ⎝ 2 ⎠⎠ pure asphaltenes and pure resins should be known. These
properties are not readily obtained from typical asphaltene or
where V is the total volume, x α mole fraction of asphaltene crude-oil characterizations. For phase-equilibrium
free site, x β mole fraction of resin site. The total pressure P is calculations, We can use the average molecular weights and
given as a sum of the contributions from the SRK part and the densities for pure asphaltenes and resins. . As discussed by
association part. Speight (1991, 1994), the average molecular weights of
asphaltenes and resin chains are about 2,000 and 800 ,
P = PSRK + Passoc (8) respectively. The mass density of a typical asphaltene is about
1200 kg/m3, close to that of the heaviest hydrocarbons, and the
Where PSRK is give by : mass density of a typical resin is about 1,000 kg/m3 Speight,
(1991) From the mass densities and molecular weights, the
RT a number densities of pure asphaltenes and resin chains can be
P= − (9) calculated. Using the molecular weight of asphaltene and mass
V − b V (V + b ) density , the diameter of an asphaltene molecule segment lies
in a narrow range, 1.43 -1.58 nm. A value of σ=1.5 nm is
The total volume V, is calculated at pressure P presented by used in this calculation.
equation (8). The compressibility factor in this case is :
SPE 93517 3

Because resin is a liquid at ambient conditions, its

packing fraction should be less than that of a hard-sphere solid 700
Exp. upper ADE Exp saturation
at the fluid solid phase transition (0.55) . it can be assumed
600 Exp. lower ADE Model upper ADE
that the diameter of a resin segment is about 0.5 nm, and the
Model saturation Model lower ADE
number of segments per chain is about 10. Table 1 shows
values of parameter used in this study. 500

Pressure, atm
400
Table 1: Parameter used in the study
Parameter Asphaltene Resin 300
Density* , kg/m3 1200 1000
Volume parameter, kα 0.05 0.05 200
Number of segments 1 10
Segment diameter, nm 1.5 0.5 100

* experimental value is used when available 0

350 360 370 380 390 400
Different crudes from different literature sources were used in
this work and comparison of calculated onset pressure by the Temperature, K
model with experimental values were satisfactory and the
(AAE) was about 9.8% with maximum error of 44% and Figure 2: Comparison of ADE obtained by molecular
minimum error of 0.4%. the detailed result shown in table 4. thermodynamic model with experimental ( Crude E)
Figure 1 2 , and 3 show a comparison of calculated ADE four
samples with experimental data. Table 2 show properties of
crudes oil used in generating ADE. 600
Exp. upper ADE Exp saturation
Exp. lower ADE Model upper ADE
Model saturation Model lower ADE
500 500
Exp. upper ADE Exp saturation
Exp. lower ADE Model upper ADE
Pressure, atm

Model saturation Model lower ADE 400

400

300
Pressure, atm

300
200

200
100
360 370 380 390 400
Temperature, K
100 Figure 3: Comparison of ADE obtained by molecular
350 375 400 425 450 thermodynamic model with experimental ( Crude S)
Temperature, K

Figure 1: Comparison of ADE obtained by molecular It is noticed that from the data presented in table 4 that the
thermodynamic model with experimental ( Crude F) value of asphaltene self association (E/kT)αα is ranging
between 3-20 and that of asphaltene-resin interacting is
ranging between 6-30. If we use an average value of 10 and 16
For calculation of mole fraction of asphaltene and volume of respectively, the model become completely predictive. Table
the crude at different temperature and pressure flash 5 shows summary of error analysis for onset pressure
calculation was used. (equations 8 through 128 ) . This calculation when using (E/kT)αα and (E/kT)αβ as fitting
equations are highly non – linear and Newton’s raphson was parameters and when using average values Work is underway
used to solve the equations.. to measure asphaltene/resin interactive energies for a broad
data base of crude oils.

Conclusion

The molecular thermodynamic model discussed here

has been tested with experimental data for a few reservoir
fluids to describe the effects of pressure and oil composition
4 SPE 93517

on asphaltene precipitation. This model can be applied to characterization most often used is to stop at C7+. The
identify the operation conditions at the onset of asphaltene reservoir fluid compositional analysis is usually carried out by
precipitation, asphaltenes and resins. The parameters related to defining the pure hydrocarbon components (C1 - C6), non-
association interactions, N , E/kT and Masph, are confined hydrocarbons (N2 , H2S, CO2), and grouping components
within physically meaningful lower and upper limits of onset heavier than heptane or decane into one components or limited
pressure . However, because calculated results are often number of pseudo-components. Because EOS accuracy
sensitive to these parameters, and because asphaltenes and depends on number and characterization of the plus-fraction,
resins in crude oils differ from one crude oil to another, in based on molecular weight and density of C7+ fraction , this
typical cases, they must be correlated with some experimental fraction is divided into 10 or 12 pseudocomponents.,
data for the particular crude oil of interest. It is expected if the
interactive energies are measured by micro-calorimeter , the In this study Katz (1983) method was used and the resulting
model will be completely predictive. relationship can be conveniently expressed mathematically by
the following expression:
Acknowledgement

M. A. Fahim would like to thank Kuwait University Zn = 1.38205 Z7+ e-0.25903 n (I-1)
for granting him the sabbatical year 2003/2004 where he spent
part of it at Denmark Technical University. where
Z7+ = mole fraction of C7+
Nomenclature n = number of carbon atoms
Zn = mole fraction of pseudo-component with n
Aassoc Association part of correction carbon atoms
P Pressure
R Gas constant
T Temperature The molecular weight and density of last fraction (n+) is
N Number of chain calculated as follows:
M Molecular weight n
xα mole fraction of asphaltene free site MWn+ = [ Z7+ MW7+ - ∑ (Zn . MWi)]/Zn+ (I-2)
xβ mole fraction of resin free site i=7
XA mole fraction of asphaltene
XR mole fraction of resin n
E Association Energy SGn+=Zn+ MWn+ /(Z7+ MW7+ / SG7+) - ∑(Zn . MWi/Zi) (I-3)
σ Mean diameter i=7
κ Volume parameter
Once the mole fraction and the molecular weight and gravity
ρΑ Density of asphaltne
of each subtraction are determined, critical properties and
ρR Density of resin
acentric factor are calculated
φiL Fugacity coefficient in liquid phase
To use the EOS, it is required to characterize the
φiV Fugacity coefficient in vapor phase
subtractions by estimating their critical properties, The
φis Fugacity coefficient in solid phase Pedersen et al. (1989) correlation is used to calculate critical
ghsij Contact of the pair correlation function properties and acentric factor for the pseudo-fractions forming
heptane plus using the molecular weight (MW) and density
Κi Equilibrium ratio
(SG) of the subtraction as follows:
z Feed fraction
Z Compressibility factor
Pc Critical pressure Tc = 1.6312x102. SG+ 8.6052*101- ln(MW) + 4.3475x10-
1
Tc Critical pressure .MW-1.8774x10-3/MW (I-4)
ω Acentric factor
lnPc = -1.3408x10-1 + 2.5019. SG + 2.0846x102/MW -
a First SRK parameter
3.9872x103/MW2 (I-5)
b Second SRK parameter
The acentric factor is calculated from the following equations:
Appendix
m = 0.48 + 1.574ω - 0.176ω2 (I-6)
Since Petroleum fluids consist of thousands of
different hydrocarbon constituents. The diversity of the where
chemical structure of the individual components increases
with carbon number; it is therefore impractical to analyze for m = 7.4310x10 -1 + 4.8122x10 -3 . MW +9.6707x10-3. SG -
all the C7+ components. A standard composition analysis 3.7184x10-6. MW2 (I-7)
SPE 93517 5

where Tc is in K, Pc in atm, MW is in g/mol, and SG is in Jamalddin, A.K.M. , Nighswander, J.N. SPE 80261, 2003
g/cm3
Katz, D., “JPT, June 1983, pp. 1205-1214
Oil n gas mixture is a complicated system with
special characterization. However one can assume that there is Merino-Garcia , D. and Andersen , S. I (2003); 4th
no heavy hydrocarbon with high molecular weight such as international conference on petroleum Phase behaviour ,
asphaltene in vapor phase. The following material balance can Norway
be written :
Pedeson, K. S. Fredenslund, Aa., ThomassenP., 1989. Gulf
n n
z i K iVL Publishing, Huston.
∑1 i ∑i V ( K VL − 1) − S + 1 = 1
x V
= (I-8)
i Sorensen , C. ; (2003) M-Sc thesis, Technecal university of
Denmark.
n n
zi
∑x
1
i
L
=∑
i V (K VL
− 1) − S ( K iSL − 1) + 1
=1 (I-9)
Speight, J. G., The Chemistry and Technology of Petroleum,
i
Dekker, . New York 1991.
K nSL z n (I-10) Speight, J. G. . Arab. J. Sci. Eng., 19, 335 1994 .
x nS =
1 − V + S ( K nSL − 1)
Wu, J., J. M. Prausnitz, and A. Firoozabadi, AIChE J., 44,.
Where 1188 1998 .

φiL Wu J. and John M. Prausnitz , A. Firoozabadi January 2000

K iVL = (I-11)
φiV Vol. 46, No. 1 AIChE
φiL (I-12)
K iSL =
φiS

Initial values of equilibrium ration for each component in the

mixture at specified temperature and pressure can predicted
from wilson’s equation as initial guess for calculation

Pci ⎡ T ⎤
K iVL = exp ⎢5.37(1 + ω i )(1 − ci )⎥ (I-13)
P ⎣ T ⎦

φiL and φiV can be calculated from SRK - EOS and for φiS
can be evaluated from solid phase fugacity at reference point.

References

Behar E., Mougin P. , and Pina A. Oil and Gas Science

Technology, Vol 58 (2003), No 6, pp 637-646

Burce F.K. , Nghiem L.X. , SPE64465, 2000

Burke, N. E., R. E. Hobbs, and S. F. Kashou J. Pet. Technol.,

42, 1440. 1990

Chapman , W., Gubbins, K., Jackson, G. and Radosz, M.

(1989); Fluid Phase Equilibria 52, 31-38

Chapman , W., Gubbins, K., Jackson, G. and Radosz, M.

(1990) ; Ind Eng Chem Res. 29, 1709-1721

Jamalddin, A.K.M. , Nighswander, J.N. SPE 87292, 2000

6 SPE 93517

Table 2: Properties of crude oils used in generating ADE

S.G MWT Asp. Res Crude T Pupper Psat Plow

Crude C7% C7+ C7+ Wt% Wt% MWt (K) (atm) (atm) (atm)
E1 26.88 0.865 228.1 1.3 7.0 82.5 372.0 466.8 219.1 121.4
E2 26.88 0.865 228.1 1.3 7.0 82.5 378.0 448.1 223.0 128.3
E3 26.88 0.865 228.1 1.3 7.0 82.5 384.0 437.2 223.0 133.2
E4 26.88 0.865 228.1 1.3 7.0 82.5 389.0 423.4 223.0 133.2
F1 32.29 0.877 248.8 1.4 10.4 103.1 361.0 367.3 169.8 105.6
F2 32.29 0.877 248.8 1.4 10.4 103.1 383.0 275.5 181.3 185.5
F3 32.29 0.877 248.8 1.4 10.4 103.1 400.0 248.3 197.4 158.9
F4 32.29 0.877 248.8 1.4 10.4 103.1 422.0 258.5 207.3 200.3
K1 28.11 0.852 209.5 0.5 11.3 81.8 322.0 590.8 173.7 94.0
K2 28.11 0.852 209.5 0.5 11.3 81.8 339.0 528.0 194.7 86.9
K3 28.11 0.852 209.5 0.5 11.3 81.8 355.0 490.0 201.4 93.3
K4 28.11 0.852 209.5 0.5 11.3 81.8 389.0 484.1 221.1 104.1
K5 28.11 0.852 209.5 0.5 11.3 81.8 496.0 470.0 240.1 80.0
S1 26.88 0.865 228.1 1.3 18.8 82.0 372.0 460.3 216.3 -
S2 26.88 0.865 228.1 1.3 18.8 82.0 377.0 442.4 220.5 129.2
S3 26.88 0.865 228.1 1.3 18.8 82.0 383.0 431.1 220.0 -
S4 26.88 0.865 228.1 1.3 18.8 82.0 389.0 416.9 220.9 131.6

Table 3: Properties of crude oils used in testing model

S.G MWT Asp. Res Crude T Pupper Psat Plow

Crude C7% C7+ C7+ Wt% Wt% MWt (K) (atm) (atm) (atm)
A 47.96 0.959 329.0 15.8 8.3 171.3 373.0 351.7 198.4 125.3
B 26.61 0.853 215.9 0.5 0.9 78.4 384.0 543.0 - -
C 30.29 0.926 295 3.5 18.0 403 409.6 284.2 182.6
D 39.74 0.846 223.2 0.9 1.5 109.8 386.0 228.3 133.2 88.8
H 30.68 0.877 248.8 1.4 10.4 99.3 422.0 374.0 272.1 -
I 24.17 0.875 368.9 4.6 7.6 103.0 361.0 361.6 289.9 260.8
J 30.41 0.846 204.8 0.5 6.0 84.4 386.0 510.2 205.3 79.0
L 27.94 0.846 205.2 0.3 7.4 76.0 383.0 421.4 225.0 125.3
M 29.88 0.832 209.8 0.5 1.1 86.8 391.0 255.0 160.3 106.6
N 32.69 0.8772 239.0 6.1 18.0 122.4 322.0 491 149.7 55.0
O 50.92 0.852 210.7 1.0 4.5 126.0 316.0 151.4 65.8 -
P 43.33 0.832 208.7 0.9 4.6 110.8 327.0 258.5 82.6 -
Q 29.92 0.833 213.1 0.5 6.4 84.2 327.0 350.4 200.7 -
R 66.68 0.902 281.0 7.8 7.5 202.4 376.0 205.1 - -
T 24.17 0.875 368.9 4.6 13.5 109.6 361.0 355.0 286.1 257.2
U 41.39 0.874 245.5 3.3 2.3 123.2 386.0 268.6 116.8 -
V 73.56 0.860 230.3 1.5 2.1 193.7 383.0 258.6 169.8 128.3
W 42.19 0.874 245.5 3.3 7.7 125.2 386.0 270.0 126.3 68.2
 SPE 93517

Table 4: Calculation result of different crude for molecular Thermodynamic method

T PEXP PSRK Passoc Pcalc

Crude Masp MRes (E/kT)αα (E/kT)αβ Xα Xβ XA XR %ErrA %ErrB %ErrC
(K) (atm) (atm) (atm) (atm)
A 373.0 351.0 555.0 3500 900 17.0 25.0 0.00202 3.2E-06 0.00322 0.00756 -131.0 424.0 -20.8 -41.7 -58.1
B 384.0 543.0 331.0 3500 850 20.0 30.0 0.00089 5.0E-08 0.00040 0.00090 19.4 350.4 35.5 37.9 39.0
C 403.0 450.0 490.0 2500 900 9.0 11.0 0.15012 0.07497 0.00330 0.01410 -26.1 463.9 -3.1 1.2 -8.9
D 386.0 228.0 170.0 3500 850 20.0 25.0 0.00100 0.00001 0.00060 0.00150 19.1 189.1 17.1 22.4 25.4
E1 372.0 466.0 493.0 2000 800 10.0 17.0 0.10443 0.00036 0.00087 0.00526 -14.0 479.0 -2.8 -3.0 -5.8
E2 378.0 448.0 475.0 2000 800 9.0 15.0 0.16594 0.00167 0.00087 0.00526 -10.7 464.3 -3.6 -3.2 -6.0
E3 383.0 437.0 454.0 2000 800 9.0 13.0 0.17414 0.01062 0.00087 0.00526 -8.4 445.6 -2.0 -1.0 -3.9
E4 389.0 423.0 440.0 2000 800 9.0 15.0 0.16602 0.00167 0.00087 0.00526 -10.4 429.6 -1.6 -1.1 -4.0
E5 361.0 364.0 430.0 2000 800 15.0 17.0 0.00836 0.00359 0.00300 0.00410 -54.4 375.6 -3.2 -10.5 -18.1
F1 361.0 367.0 404.0 2200 850 15.0 17.0 0.00857 0.00389 0.00100 0.01040 -25.8 378.2 -3.1 -4.5 -10.1
F2 386.0 275.0 304.0 2200 850 15.0 17.0 0.00814 0.00419 0.00100 0.01040 -21.8 282.2 -2.6 -4.3 -10.5
F3 400.0 248.0 271.0 2200 850 13.0 15.0 0.02317 0.00833 0.00100 0.01040 -16.3 254.7 -2.7 -2.6 -9.3
F4 422.0 258.0 300.0 2200 850 17.0 19.0 0.00330 0.00124 0.00100 0.01040 -26.6 273.4 -6.0 -9.7 -16.3
H 422.0 374.0 347.0 3500 820 11.0 13.0 0.04934 0.02331 0.00100 0.01040 19.7 366.7 2.0 0.7 7.2
I 361.0 361.0 379.0 2500 800 10.0 12.0 0.09384 0.04552 0.00035 0.01130 -8.6 370.4 -2.6 -0.1 -5.0
J 386.0 510.0 550.0 3500 830 11.0 12.0 0.04947 0.06076 0.00030 0.00180 -5.5 544.5 -6.8 -6.6 -7.8
K1 322.0 590.0 275.0 3900 830 3.0 25.0 0.88830 0.00000 0.00035 0.01130 51.4 326.4 44.7 49.2 53.4
K2 339.0 528.0 258.0 3900 830 3.0 28.0 0.88845 0.00000 0.00035 0.01130 57.2 315.2 40.3 46.7 51.1
K3 355.0 490.0 256.0 3900 830 4.0 30.0 0.77038 0.00000 0.00035 0.01130 61.6 317.6 35.2 43.6 47.8
K4 389.0 484.0 247.0 3900 830 4.0 26.0 0.77574 0.00000 0.00035 0.01130 53.2 300.2 38.0 44.1 49.0
K5 496.0 470.0 296.0 3900 830 3.0 25.0 0.89051 0.00000 0.00035 0.01130 56.0 352.0 25.1 31.4 37.0
L 383.0 421.0 428.0 2350 800 10.0 12.0 0.09651 0.04701 0.00030 0.00240 -3.5 431.5 -2.5 -3.0 -1.7
M 391.0 255.0 192.0 3500 850 12.0 15.0 0.03056 0.00519 0.00040 0.00110 6.3 198.3 22.3 22.8 24.7
N 322.0 491.0 620.0 2800 850 9.0 11.0 0.14243 0.07089 0.01280 0.03390 -115.6 504.4 -2.7 10.3 -26.3
O 316.0 151.0 190.0 2700 820 10.0 11.0 0.09064 0.11058 0.00550 0.00750 -33.1 156.9 -3.9 -0.5 -25.8
P 327.0 258.0 288.0 3000 800 13.0 17.0 0.01947 0.00129 0.00066 0.00460 -14.8 273.2 -5.9 -7.3 -11.6
Q 327.0 350.0 375.0 2500 800 10.0 12.0 0.09407 0.04542 0.00034 0.00640 -6.1 368.9 -5.4 -3.9 -7.1
R 376.0 205.0 178.0 2500 820 9.0 11.0 0.15017 0.07495 0.00550 0.00750 26.0 204.0 0.5 -5.4 13.2
7
 8

Table 4: Calculation result of different crude for molecular Thermodynamic method (Continue)

T PEXP PSRK Passoc Pcalc

Crude Masp MRes (E/kT)αα (E/kT)αβ Xα Xβ XA XR %ErrA %ErrB %ErrC
(K) (atm) (atm) (atm) (atm)
S1 372.0 460.0 423.0 3000 820 9.0 13.0 0.13799 0.00985 0.00090 0.01880 24.1 447.1 2.8 0.1 8.0
S2 377.0 442.0 420.0 3000 820 9.0 11.0 0.13833 0.06833 0.00090 0.01880 15.9 435.9 1.4 -3.1 5.0
S3 383.0 431.0 416.0 3000 820 7.0 9.0 0.35540 0.20191 0.00090 0.01880 7.4 423.4 1.8 -4.2 3.5
S4 389.0 416.0 413.0 3000 820 4.0 6.0 0.82803 0.68614 0.00090 0.01880 1.2 414.2 0.4 -7.1 0.7
T 361.0 355.0 558.0 3500 800 20.0 25.0 0.00085 0.00001 0.00320 0.01350 -144.0 414.0 -16.6 -42.0 -57.2
U 386.0 268.0 317.0 3000 850 15.0 17.0 0.00744 0.00337 0.00230 0.00230 -42.2 274.8 -2.5 -10.5 -18.3
V 383.0 258.0 266.0 2400 800 9.0 11.0 0.15255 0.07670 0.00110 0.00210 -5.8 260.2 -0.9 0.3 -3.1
W 386.0 270.0 300.0 2500 800 11.0 13.0 0.05819 0.02752 0.00230 0.00770 -22.3 277.7 -2.9 -2.5 -11.1
* A using variable value of (E/kT)αα and (E/kT)αβ
∗ B using average value of (E/kT)αα and (E/kT)αβ
* C using EOS with no tuning

Table 5 : Comparison of error in onset pressure for different cases:

Variable Average SRK

(E/kT)αα and (E/kT)αβ (E/kT)αα and (E/kT)αβ No corection or tuning
AAE 9.8% 11.8% 19.2%
MAX 44% 49% 58.1%
MIN 0.2% 0.48% 0.72%
SPE 93517