Fluid Phase Equilibria 457 (2018) 62e73

Contents lists available at ScienceDirect

Fluid Phase Equilibria

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / fl u i d

Prediction of dew points and liquid dropouts of gas condensate

mixtures
Nefeli Novak a, Vasiliki Louli a, Stathis Skouras b, Epaminondas Voutsas a, *
a
Laboratory of Thermodynamics and Transport Phenomena, School of Chemical Engineering, National Technical University of Athens, 9, Heroon
Polytechniou Str., Zografou Campus, 15780 Athens, Greece
b
Statoil ASA, Research & Technology Center, Trondheim, Norway

a r t i c l e i n f o a b s t r a c t

Article history: Thermodynamic modeling of reservoir fluids behavior is of great importance for the oil and gas industry.
Received 5 August 2017 For design and safe operation purposes, it is necessary to have an accurate, simple and robust model for
Received in revised form the prediction of the phase equilibrium of reservoir fluids. In this work, the UMR-PRU model is applied in
4 October 2017
the prediction of dew points and liquid dropouts of gas condensate mixtures, and is compared with the
Accepted 24 October 2017
widely used cubic equations of state Soave-Redlich-Kwong (SRK) and Peng-Robinson (PR) as well as the
non-cubic Perturbed Chain Statistical Associating Fluid Theory (PC-SAFT) equation of state. The evalu-
ation of the models was based both on synthetic and real gas condensate mixtures. For real gas con-
Keywords:
Gas condensates
densates, a characterization method for the C7þ fraction based on the one proposed by Pedersen is
Dew point developed for UMR-PRU. The results reveal that UMR-PRU is a sufficiently accurate model for dew point
Liquid dropout and liquid dropout predictions of gas condensates.
Cubic equation of state © 2017 Elsevier B.V. All rights reserved.
UMR-PRU

1. Introduction differentiate from one another either in the attractive parameter or

Accurate knowledge of the phase behavior of gas condensate
mixtures is essential for the unobstructed gas processing and
optimization as well as for gas transportation. More specifically,
accurate dew point and liquid dropout data are required in order to
ensure that the fluid is either transported in single phase, i.e. above
the dew point pressure at a specific temperature, or that the liquid
produced due to condensation does not drastically change the
mixture composition or damage the infrastructure. PVT experi-
ments like saturation points and liquid dropout measurements,
otherwise referred to as constant composition expansion experi-
ments (CCE), are effectively used to design an efficient extraction
process without operational problems and loss of valuable product.
The same data can also be utilized for flow assurance through the
pipeline network.
Equations of state (EoSs) are routinely used for the prediction of
phase behavior and thermodynamic properties of petroleum fluids.
Cubic EoSs, like Peng-Robinson (PR) [1] and Soave-Redlich-Kwong
(SRK) [2], are still very popular for such calculations. Many modi-
fications have been proposed creating families of EoSs that
* Corresponding author.
E-mail address: evoutsas@chemeng.ntua.gr (E. Voutsas).
the co-volume parameter. Modifications of the attractive term
parameter aim to improve vapor pressure predictions of the con-
densable components [3e8] and supercritical behavior of the light
ones [6,9e11] while the modifications of co-volume parameter aim
to the improvement of density predictions [7,12]. Towards the
improvement of the density predictions, the use of a volume
translation is a simple and effective method that does not alter the
equilibrium conditions [13e15]. Extension of a cubic EoS to mix-
tures requires mixing rules for the attractive-term and co-volume
parameters. A widely employed method to extend the cubic EoS
to multicomponent mixtures is via the Van der Waals one-fluid
(vdW1f) mixing rules, i.e. quadratic composition dependency for
both parameters coupled with the geometric mean combining rule
for the cross attractive-term parameter and the arithmetic mean
rule for the cross co-volume parameter. However, it has been
shown [16] that cubic EoSs coupled with the vdW1f mixing and
combining rules fail for asymmetric -with respect to the size of the
molecules involved-mixtures. So, apart from improving the EoS for
pure component predictions, the use of advanced mixing rules,
such as those derived by combining a cubic EoS and an excess Gibbs
energy model, like UNIQUAC, NRTL, Wilson and UNIFAC, has been
known to improve EoS predictions, even for mixtures containing
non-polar components, such as CCS [17] and natural gas mixtures

https://doi.org/10.1016/j.fluid.2017.10.024
0378-3812/© 2017 Elsevier B.V. All rights reserved.
 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73
[18e21].
Except for the classical two-parameter cubic EoSs, three-
parameter cubic EoSs [22e26] have also been used for petroleum
fluid modeling, due to their improved density predictions [27,28].
Three-parameter EoSs, like Patel-Teja and Patel-Teja-Valderrama,
have shown to be more accurate than PR and SRK for density and
compressibility factor predictions [28,29]. Dew point predictions
however are not necessarily improved with the addition of a third
parameter [28,29].
Another class of equations of state, of non-cubic form, based on
the statistical thermodynamic principles of SAFT (Statistical Asso-
ciating Fluid Theory), has gained popularity in the industry after the
integration of the derived EoS in commercial process simulators.
For example, the Perturbed-Chain SAFT EoS (PC-SAFT) is widely
used for predicting asphaltene precipitation [30e39]. Recently, Yan
et al. [39] compared dew point predictions of PC-SAFT with those
obtained by PR and SRK and found that while all models are similar
for natural gas dew points, for gas condensates PC-SAFT yields
better results. Surprisingly, for mixture liquid densities, PC-SAFT
was found to yield on average similar results with the untrans-
lated PR, although it is clearly superior for pure components [40].
Gas condensate mixtures consist of numerous different com-
ponents, which makes a complete component analysis practically
impossible. The most common case of compositional analysis of
reservoir fluids is to perform detailed analysis up to n-hexane,
while the heavy end of the fluid is described by a single plus frac-
tion (C7þ). EoS predictions for real gas condensates are greatly
influenced by the characterization method of the C7þ fraction, so a
good coupling of an EoS with a characterization method is very
important for accurate results [28,41e43]. The expected un-
certainties of EoSs in gas condensate PVT and saturation point
predictions vary from one author to another, depending on the
characterization method and the physical properties used for the
pseudocomponents. The general conclusion that most researchers
agree on is that traditional cubic EoSs do not yield accurate phase
equilibrium predictions, unless the fluid is somehow matched to
experimental data [29]. These data, such as PVT experiments,
saturation points etc., can be used to manipulate the fluid proper-
ties, like the MW of the plus fraction [44], or the binary interaction
parameters [45] in order to match the model predictions with the
experimental data. This has also been done indirectly by calculating
EoS dependent pseudo-critical properties for the plus fraction
regressed on PVT data [46].
Most characterization methods comprise of three steps: split-
ting the C7þ fraction into single carbon number (SCN) fractions,
assigning physical properties to each SNC according to suitable
correlations based on combinations of molecular weight (MW),
boiling point (Tb) and density or specific gravity (SG), and, finally,
lumping them together again into multiple carbon number (MCN)
fractions for computational reasons without losing the information
of the SCN approach [47e49]. Detailed review of physical property
correlations is available in the literature [50e52]. The selection of a
specific correlation for physical properties of the SCN fractions has a
significant effect on the EoS predictions but no correlation is
consistently better than the others [53]. Several characterization
methods have been proposed in the literature for cubic EoSs
[51,54e58], the most popular being the ones proposed by Whitson
[58] and Pedersen [51]. The same approach has been also proposed
for PC-SAFT EoS, however, the determination of the pseudo-
components parameters is still under investigation [39,40,59,60].
In previous publications, it has been shown that the UMR-PRU
model is superior to SRK, PR and PC-SAFT in dew point pre-
dictions of both synthetic and real natural gases [18e20,61]. In this
study, the UMR-PRU model is further applied in the prediction of
dew points and liquid dropouts of synthetic gas condensates and is
63
compared with PR, SRK and PC-SAFT. Furthermore, the perfor-
mance of UMR-PRU model in the prediction of dew points and
liquid dropouts of real gas condensates combined with a properly
adapted Pedersen characterization method is examined and
compared with PR and SRK EoSs.
2. Thermodynamic models
All models considered in this study have been thoroughly pre-
sented in the literature. SRK [2] and PR [62] have been applied in
this work coupled with the vdW1f mixing and combining rules
using temperature independent binary interaction parameters
(BIPs) retrieved either from ASPEN-HYSYS process simulator
databank or determined in this work. The UMR-PRU model [61,63]
has been applied using the UNIFAC interaction parameters pre-
sented by Louli et al. [61]. Pure component critical temperatures
and pressures as well as acentric factors, needed for SRK, PR and
UMR-PRU, were taken from the DIPPR data compilation [64].
For PC-SAFT [65,66], the three pure component parameters for
non-associating compounds: the segment number (m), the
segment diameter (s), and the segment energy parameter (ε=k),
were taken from Gross and Sadowski [66] and Ting et al. [37],
except for n-tetracosane and 2,2,4,4,6,8,8- heptamethyl-nonane,
which are not available in the literature. The parameters for these
components were determined in this study by fitting pure
component vapor pressure and saturated liquid density data and
are tabulated in Table 1. PC-SAFT has been extended to mixtures by
introducing mixture segment diameter and segment energy pa-
rameters derived from the van der Waals one fluid theory [66].
Similarly to cubic EoSs, PC-SAFT has been applied using tempera-
ture independent binary interaction parameters in the dispersion
term of the EoS taken either from the literature or determined in
this work.
3. Results and discussion
3.1. Dew point predictions of synthetic gas condensate (SGC)
mixtures
The database of synthetic gas condensate (SGC) mixtures that
was developed for the evaluation of the models in dew point pre-
dictions, consists of eight mixtures that contain hydrocarbon (HC)
molecules of diverse chemical structure such as straight chain al-
kanes, branched alkanes and aromatic hydrocarbons, at various
compositions. Moreover, the experimental dew point data cover a
wide range of temperatures and pressures. The compositions of the
SGCs are given in Table 2.
Initially the effect of the binary interaction parameters (BIPs) in
the dew point predictions of gas condensate mixtures was exam-
ined. A typical example is presented with PR in Fig. 1, where the
effect of using different sets of BIPs is showcased, dividing the HCs
into two categories. The first category includes HCs lighter than n-
decane, hereby denoted as HCset1, and the second, HCs at least as
heavy as n-decane, denoted as HCset2. The following scenarios
Table 1
PC-SAFT pure component parameters regressed in this work. Experimental vapor
pressure (Ps) and saturated liquid molar volume (Vl) data were taken from DIPPR
[64].
Compound T-range (K) m s (Å) ε/k (K) %DPs %DVl
n-Tetracosane 325e804 9.7400 3.9582 254.36 2.6 3.3
2,2,4,4,6,8,8- 200e690 5.0666 4.2832 265.35 0.8 2.0
heptamethyl-nonane
PNDP abs ðPis;exp Pis:calc Þ PNDP abs ðVl;iexp Vl;icalc Þ
%DP s ¼ NDP
100
i¼1 Pis;exp
.%DVl ¼ NDP
100
i¼1 Vl;iexp
.
64 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73

Table 2
Molar compositions of SGC mixtures.

Compounds SGC 1 SGC 2 SGC 3 SGC 4 SGC 5 SGC 6 SGC 7 SGC 8

[67] [67] [67] [68] [68] [69] [69] [70]

Methane 0.7905 0.7905 0.7905 0.7581 0.7472 0.8232 0.8205 0.7297

Ethane 0.1132 0.1097
Propane 0.0871 0.0895 0.1771
n-Butane 0.0488 0.0442
i-Pentane 0.0023 0.0023 0.0023
n-Pentane 0.1999 0.1999 0.1999 0.0505 0.0500
n-Hexane 0.0023 0.0023 0.0023 0.0458

n-Octane 0.0268 0.0387

Toluene 0.0199 0.0272
n-Decane 0.0198 0.0199
n-Tridecane 0.0133
n-Hexadecane 0.0230 0.0330 0.0194 0.0201
n-Tetracosane 0.0050 0.0020
Phenanthrene 0.0102
Naphthalene 0.0050 0.0030
1-methyl-Naphthalene 0.0254
2,2,4,4,6,8,8- 0.0087
heptamethyl-nonane

Table 4
Average absolute deviations in saturation point pressure for synthetic gas conden-
sate mixtures. BIPs for SRK, PR and PC-SAFT were found in literature with the
exception of those for CH4/heavy HCs which were calculated in this work.

NDPa SRK PR PC-SAFT UMR-PRU

AADPb AADPb AADPb AADPb

SGC 1 8 10.2 15.1 6.9 4.2

SGC 2 12 12.6 10.4 7.9 13.1
SGC 3 7 5.6 10.6 5.9 3.7
SGC 4 12 69.3 58.8 31.2 2.4
SGC 5 7 5.9 2.0 15.9 2.4
SGC 6 2 3.1 0.9 15.9 11.4
SGC 7 4 6.2 2.0 13.0 6.7
SGC 8 3 22.4 18.3 26.3 8.9
Average 55 17.9 16.2 15.3 6.2
a
NDP: Number of experimental data points.
b 100
PNDP abs ðPiexp Picalc Þ
AADP ¼ NDP i¼1 Piexp
.

Fig. 1. Phase envelope prediction with PR EoS for SGC 7. Diamonds: experimental data;
solid line: non-zero BIPs for all binaries; short-dashed lines: non-zero BIPs only be-
tween methane and HCset2; long-dashed line: non-zero BIPs only between methane
and all HCs; dotted line: all BIPs were set equal to zero. The BIPs used for CH4/HCset2
binaries are the ones calculated in this work.
were examined: (a) non-zero BIPs for all binaries, (b) non-zero BIPs
only between methane and heavy hydrocarbons (HCset2), (c) non-
zero BIPs only between methane and all HCs, and (d) BIPs equal to
zero for all binaries. It was shown that the BIPs that mainly affect
dew point predictions are the ones between methane and heavy
hydrocarbons. Actually, the use of interaction parameters only be-
tween methane with the heavier hydrocarbons yields practically

Table 3
VLE predictions (kij ¼ 0) and correlations for binary CH4/HCset2 systems using SRK, PR and PC-SAFT EoSs.

CH4 with Reference NDPa T range P range %AADPb (kij ¼ 0) Optimum kij values %AADPb (opt. kij) %AADPb
(K) (bar)
SRK PR PC-SAFT SRK PR PC-SAFT SRK PR PC-SAFT UMR-PRU

n-Decane [75,103,104] 115 303e423 1e362 8.2 9.4 13.4 0.0349 0.0388 0.0213 3.4 3.5 4.0 5.9
n-Tetracosane [77,78] 122 330e440 20e971 21.9 25.3 27.1 0.0465 0.0556 0.0383 10.0 9.9 2.5 16.0
n-Hexadecane [79e81] 92 290e360 21e703 23.4 25.4 24.1 0.0558 0.0592 0.0319 5.4 5.3 3.7 9.5
1-methyl-Napthalene [82] 7 464 21e251 18.5 21.8 27.7 0.0938 0.1009 0.0662 1.4 1.4 1.3 13.8
Phenanthrene [83e85] 60 380e473 178e960 48.3 51.6 44.23 0.2004 0.2025 0.0854 7.6 7.5 4.2 13.3
Naphthalene [85] 12 373e423 19e87 25.0 29.8 26.1 0.1110 0.1220 0.0537 2.9 2.9 2.7 7.1
n-Tridecanec e e e e e e e 0.0330 0.0364 0.0233 e e e e
2,2,4,4,6,8,8
heptamethyl-nonanec e e e e e e e 0.0558 0.0592 0.0319 e e e e
Average 22.3 24.8 25.1 6.4 6.4 3.4 11.0
a
NDP: Number of experimental data points.
b 100
PNDP abs ðPiexp Picalc Þ
%AADP ¼ NDP i¼1 Piexp
, where P is the bubble point pressure.
c
No experimental vapor-liquid equilibrium data are available.
 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73 65

Fig. 2. Phase envelopes for synthetic gas condensates mixtures. Diamond markers: experimental data; Blue line: PR; Black line: PC-SAFT; Green line: SRK; Red line: UMR-PRU. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
66 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73

the same dew point predictions with those obtained when BIPs for
all pairs containing methane or the complete set of BIPs are used.
Although, Fig. 1 presents results only for PR, the same stands for
SRK and PC-SAFT.
Due to the sensitivity of the EoSs to the BIPs of methane with the
heavier HCs, the corresponding BIPs were recalculated in this work
while the rest were taken from the literature. Specifically, for
methane with HCset1, the binary interaction parameters, although
not important for dew point predictions, were taken from the
HYSYS® process simulator database for PR and SRK. For PC-SAFT,
the corresponding binary interaction parameters were found in
the literature [71,72]. For methane with hydrocarbons belonging to
HCset2, binary interaction coefficients were determined by fitting
binary vapor-liquid equilibrium experimental data. The values of
the BIPs are presented in Table 3 along with the deviations in
bubble point pressures. For two mixtures, methane with n-tride-
cane and methane with 2,2,4,4,6,8,8-heptamethyl-nonane, no
experimental VLE data were found in the literature. For n-tride-
cane, the mean value of the BIPs of CH4/n-dodecane and CH4/n-
tetradecane was used, which were fitted to VLE data [73e75]. For
2,2,4,4,6,8,8-heptamethyl-nonane the value of its isomer, hex-
adecane, was used instead. All BIPs used for PR, SRK and PC-SAFT
are given in the supplementary tables S1 to S3.
Bubble point pressure predictions with EoSs, i.e. all kij's were set
equal to zero, get progressively worse with increasing carbon
number. In most cases SRK yields the best predictions. As far as the
correlation ability of the models is concerned, PR and SRK yield
similar results, inferior to PC-SAFT, which in accordance with pre-
vious studies exhibits good correlation ability independently of
chain length and molecule complexity [76]. UMR-PRU offers a very
good compromise between accuracy and predictive ability, since its
group interaction parameters between CH4 and CH2 groups were
fitted to VLE data for alkanes up to hexadecane, and it is reasonable
to expect that the UMR model cannot compete with EoS specific
binary mixture correlations [77e85].
Moreover, although as already mentioned BIPs between hy-
drocarbons other than those containing methane are not important
and they could set equal to zero, in this work BIPs for these binaries
were also used. For PR and SRK they were taken from the HYSYS®
process simulator database and for PC-SAFT from the literature.
The extension from binary phase equilibrium to phase equilib-
rium in multicomponent gas condensate mixtures is not an easy
task. The regression of binary interaction parameters at conditions
where gas condensates are present is not always possible due to the
lack of VLE experimental data. Usually the available experimental
VLE data for methane mixtures with heavy HCs are limited to
relatively low pressures unlike the high pressures encountered in
gas condensate phase equilibria [86]. Furthermore, other
limitations may be imposed by different equilibrium types like
liquid-liquid splits [87] and/or solid precipitation [78,81,83,88e90].
Deviations in saturation point pressure are presented in Table 4,
while Fig. 2 presents graphically the results in PT diagrams. The
introduction in PR, SRK and PC-SAFT of non-zero BIPs shifts the dew
point branch to higher pressures (Fig. 1), which becomes more
pronounced at high molar fractions of heavy HC, and usually leads
to better dew point predictions than using zero BIPs. Only for
mixtures 2, 4 and 8 the introduction of non-zero BIPs leads to worse
results than using zero BIPs.
SGC 1 that contains the highest molecular weight HC (n-tetra-
cosane) in this work is better described by UMR-PRU followed by
PC-SAFT, while cubic EoSs underestimate the experimental data.
For SGC2 that is actually the same gas as SGC1 but contains
naphthalene instead of n-tetracosane, although all models can
capture the change of aromaticity by predicting a narrower phase
envelope than that of SGC1, they overestimate the experimental
data with PC-SAFT EoS yielding the best results. For SGC3 that
contains both n-tetracosane and naphthalene, SRK, PC-SAFT and
especially UMR-PRU predict more accurately the dew points than
for SGC2. However, both cubic EoSs, particularly PR, underestimate
the experimental data of SGC3. For SGC4 that contains phenan-
threne, UMR-PRU gives very good results, unlike PC-SAFT and cubic
EoSs that yield poor predictions. In the case of SGC5 that contains
the same components as SGC4, but the aromatic phenanthrene that
is replaced by n-hexadecane, UMR-PRU still gives very good results.
Satisfactory results are also obtained by SRK and especially PR, in
contrast to PC-SAFT that highly overpredicts the dew points. SGCs 6
and 7 are practically identical. For these mixtures, cubic EoSs are
more accurate as both UMR-PRU and especially PC-SAFT over-
predict the experimental dew points. Finally, for SGC 8, UMR-PRU
yields very good predictions, while cubic EoSs and, especially, PC-
SAFT overshoot the phase envelope.
To summarize, PC-SAFT systematically overpredicts the dew
point branch of asymmetric gas condensates especially when n-
hexadecane is present. Cubic EoSs fail to describe gas condensate
mixtures that contain polyaromatic components. UMR-PRU does
not present any systematic deviations for specific type of mole-
cules, composition or complexity and exhibits the lowest overall
deviation in saturation point predictions.
3.2. Liquid dropout predictions for SGC mixtures
The database used for the evaluation of the models in liquid
dropout predictions of synthetic gas condensates is presented in
Table 5. Unfortunately, such experimental data are scarce, so the
database is less diverse than that developed for dew points.
Methane molar concentrations ranges from 0.63 up to 0.82 and

Table 5
Molar compositions of VPL mixtures.

SGC VPL 1 VPL 2 VPL 3 VPL 4 VPL 5 VPL 6 VPL 7 VPL 8 VPL 9 VPL 10

[91] [91] [91] [91] [91] [91] [91] [69] [69] [69]

T (K) 366.5 338.7 366.5 366.5 338.7 310.9 394.3 333.2 353.2 383.2

Nitrogen e e e e 0.49 12.74 15.19 e e e

Carbon Dioxide e e e e 0.21 15.06 0.19 e e e
Methane 76.35 80.69 81.59 80.97 81.34 54.90 63.88 82.05 82.05 82.05
Ethane 5.68 6.02 5.95 5.66 6.01 3.59 4.22 e e e
Propane 2.95 3.04 2.98 3.06 3.00 2.58 3.08 8.95 8.95 8.95
n-Butane e e e e e e e e e e
n-Pentane 5.29 3.56 3.39 4.57 4.10 2.56 3.09 5 5 5
n-Heptane 3.73 2.65 2.49 3.30 2.78 1.73 2.13 e e e
n-Octane e e e e e e e e e e
n-Decane 3.04 2.02 1.73 2.44 2.08 1.35 1.59 1.99 1.99 1.99
Toluene 2.96 2.02 1.87 e e 5.48 6.64 e e e
n-Hexadecane e e e e e e e 2.01 2.01 2.01
 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73 67

temperatures from 310.9 to 394.3 K.
The liquid dropout on a volume basis is defined as following:
100,ð1  bÞvliq
VpL% ¼
ð1  bÞvliq þ b,vgas
vliq is the liquid phase molar volume, vgas is the vapor phase molar
volume and b ¼ V/(V þ L), where V corresponds to the moles of the
vapor phase and L to the moles of the liquid phase at equilibrium.
Since liquid molar volume is involved in the calculations, volume
translation should be used with cubic EoSs and UMR-PRU. How-
ever, volume translation does not significantly affect the liquid
dropout predictions, as its effect is almost canceled out due to the
(1) definition of the VpL%. Typical examples with and without Pene-
loux volume translation [15] are shown in Fig. 3 for PR and UMR-
PRU. The results for both models are almost identical with and
without volume translation, and, therefore, no volume translation
was used in the rest of this work.

Fig. 3. Effect of volume translation on liquid dropout predictions. Diamond markers: experimental data; Blue line: PR; Red line: UMR-PRU; Solid line: vpl% without volume
translation; Cross markers: vpl% with Peneloux volume translation. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of
this article.)

Fig. 4. Liquid dropout curves for synthetic gas condensates mixtures. Diamond markers: experimental data; Blue line: PR; Black line: PC-SAFT; Green line: SRK; Red line: UMR-PRU.
(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
68 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73

VpL results obtained from isothermal flash calculations are
presented graphically in Fig. 4. The predictions for PR, PC-SAFT and
UMR-PRU are very satisfactory for the low-pressure part of the VpL
curve up to the maximum VpL, with no model having an advantage
over the others. PC-SAFT yields the lowest maximum VpL pre-
dictions between the models, and SRK the highest. The worst re-
sults are obtained in most cases by SRK, which was expected since
SRK yields worse liquid molar volume predictions for hydrocarbon
mixtures than the other models [40]. For lower temperatures, that
are usually closer to the critical point of the mixtures, deviations
from the experimental data are more pronounced both at
maximum liquid dropout and at the high pressure part of the curve,
since the predicted upper dew-point pressure lead to substantial
error in the VpL (VPL1, VPL2, VPL6 and VPL8). Moreover, the effect
of temperature on liquid dropouts (VPL 8 to 10) is qualitatively
captured by all models, with higher maximum dropouts as tem-
perature decreases.
comparison because no well-established characterization method
exists for this model.
Pedersen's method uses as input variables the molecular weight
and density of the C7þ fraction and consists of two steps, splitting
and lumping. At first, the plus fraction is split into single carbon
number fractions (SCN), from C7 up to C80. Eq. (2) assumes a linear
relationship between the carbon number of the SCN fractions
(CSCN) and the natural logarithm of the SCN mole fraction (zSCN)
within the Cþ fraction, and each SCN is assigned to a molecular
weight (MW) according to Eq. (4). In this equation, the constant 14
accounts for the molecular weight increment from each carbon
atom that is accompanied from two hydrogen atoms and the
term 4 accounts for the presence of aromatic structures in the
reservoir fluid. Constants A and B are determined by simultaneous
solving of Eqs. (2), (5) and (6). In this procedure components with
molar fractions smaller than 108 are ignored.

CSCN ¼ A þ B,ln zSCN (2)

3.3. Real gas condensate (RGC) mixtures

For the evaluation of the performance of UMR-PRU model

against cubic EoSs in real gas condensates, a database has been rSCN ¼ C þ D,ln CSCN (3)
developed that consists of 28 RGCs for which experimental dew
points and/or liquid dropouts are available. The composition and
properties of the aforementioned RGCs are presented in Table 6. MWSCN ¼ 14*CSCN  4 (4)
In order to apply the models in real fluids, a characterization
method has to be employed for its unidentified heavy end fraction.
The characterization method proposed by Pedersen [44,49,51,100]
is utilized for the cubic EoSs and the UMR-PRU model. For UMR- X
80

PRU, the method has been properly adapted, taking into account ZCþ ¼ zSCN (5)
SCN¼7
that UMR-PRU treats compounds of known chemical structure,
since it utilizes UNIFAC. PC-SAFT has not been included in this

Table 6
Molar composition and C7þ properties of RGC systems.

Real Gas Condensate 1 2 3 4 5 6 7 8 9 10 11 12 13 14

[92] [27] [27] [41] [41] [93] [93] [93] [93] [93] [94] [95] [95] [95]

Nitrogen 3.912 0.600 0.440 0.436 0.749 0.550 0.530 0.520 0.520 0.510 2.170 0.610 0.570 0.590
Carbon Dioxide 0.750 5.560 5.470 0.839 0.524 3.110 2.980 2.960 2.930 2.890 0.340 0.460 0.370 0.420
Methane 70.203 85.190 85.480 95.792 96.003 75.570 72.890 72.460 71.740 70.930 70.640 68.640 59.390 64.430
Ethane 9.220 3.570 3.500 2.309 2.101 9.100 8.960 8.930 8.880 8.850 10.760 13.900 13.780 13.820
Propane 2.759 1.540 1.460 0.211 0.202 2.780 2.810 2.810 2.820 2.830 4.940 6.890 7.580 7.240
i-Butane 0.662 0.370 0.370 0.056 0.048 0.540 0.550 0.560 0.560 0.560 e 0.660 0.790 0.730
n-Butane 0.981 0.570 0.580 0.072 0.058 0.600 0.620 0.620 0.630 0.640 3.020 2.660 3.280 2.980
i-Pentane 0.402 0.270 0.280 0.024 0.026 0.160 0.170 0.170 0.170 0.180 e 0.620 0.840 0.740
n-Pentane 0.422 0.230 0.230 0.019 0.023 0.470 0.620 0.650 0.690 0.730 1.350 0.940 1.310 1.130
n-Hexane 0.816 e e 0.029 0.038 0.460 0.620 0.640 0.690 0.730 0.900 1.140 1.800 1.480

C7þ 9.873 2.100 2.140 0.210 0.228 6.670 9.250 9.670 10.360 11.150 5.880 3.480 10.290 6.440
MW(Cþ) (g/mol) 192.80 129.81 143.60 156.08 151.99 193.50 193.50 193.50 193.50 193.50 153.00 152.30 177.10 166.40
density at 15  C (kg/m3) 803.0 772.8 785.5 804.4 799.4 807.3 807.3 807.3 807.3 807.3 809.3 776.3 800.2 790.6

Real Gas Condensate 15 16 17 18 19 20 21 22 23 24 25 26 27 28

[96] [48] [48] [97] [97] [97] [97] [97] [97] [98] [55] [55] [55] [99]

Nitrogen 2.440 2.260 2.010 0.280 1.150 0.950 0.770 5.820 0.100 3.340 0.221 4.141 1.688 2.120
Carbon Dioxide 0.080 5.670 5.620 e 0.320 e e e e 0.600 7.806 0.238 0.815 0.230
Methane 82.100 45.740 46.790 86.860 80.610 74.060 89.800 56.850 94.510 74.160 71.052 91.246 84.629 73.150
Ethane 5.780 11.470 12.650 2.370 4.600 7.720 3.810 11.090 1.680 7.900 10.565 2.711 5.898 9.390
Propane 2.870 7.590 5.870 1.500 2.380 4.850 1.510 8.580 0.920 4.150 4.823 0.609 2.875 4.860
i-Butane 0.560 e e e e e e e e 0.710 0.657 0.137 0.574 1.020
n-Butane 1.230 6.380 6.040 1.330 1.900 3.100 0.800 4.810 0.610 1.440 1.613 0.231 0.820 1.780
i-Pentane 0.520 e e e e e e e e 0.530 0.383 0.058 0.271 0.600
n-Pentane 0.600 4.310 3.920 0.860 1.160 1.750 0.400 2.210 0.310 0.660 0.592 0.075 0.236 0.490
n-Hexane 0.720 5.920 4.780 1.000 0.880 1.510 0.450 1.520 0.320 0.810 0.549 0.072 0.247 0.930

C7þ 3.100 10.660 12.320 5.800 7.000 6.060 2.460 9.120 1.550 5.690 1.738 0.482 1.946 5.430
MW(Cþ) (g/mol) 132.00 148.00 148.00 178.00 194.00 130.000 158.00 157.00 127.00 151.02 130.25 139.18 153.55 162.00
density at 15  C (kg/m3) 773.4 803.8 803.8 867.4 823.3 779.9 793.0 830.5 762.1 808.5 761.2 795.7 795.0 795.7
 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73 69

used in the UMR-PRU model, the correlations of Retzekas et al.

P
80
[101] were employed (Eqs (8)e(10)), which have been especially
zSCN ,MWSCN
developed for heavy hydrocarbons. Acentric factors of SCN are
MWCþ ¼ SCN¼7 (6)
P
80 calculated with the Ambrose-Walton method [102], which is
zSCN described in Eqs (11)e(14).
SCN¼7
Since the number of SCN pseudocomponents is far too large to
use with any EoS for real time applications, the next step is the
P
80
lumping procedure, where some of the SCN fractions are grouped
zSCN ,MWSCN
(lumped) together into multiple carbon number fractions (MCN) to
rCþ ¼ SCN¼7 (7)
P
80
zSCN ,MWSCN
reduce the number of pseudocomponents. Pedersen recommends a
rSCN weight-based grouping where each MCN contains approximately
SCN¼7
the same weight amount [49]. For all models, 12 pseudocompo-
The density of the plus fraction is assumed to be a linear rela- nents have been used [28,42]. The critical properties and acentric
tionship between the density of the single carbon number fractions factors for MCN fraction are calculated by Eq. (15).
(rSCN) and the natural logarithm of the corresponding carbon
number (CSCN) (Eq. (3)). Eqs. (3) and (7) are simultaneously solved
TbSCN ¼ 63:839016 r0:438991
SCN MW0:415147
SCN (8)
to generate the density of each SCN by fitting constants C and D. As
a starting point for the calculation, the density of the heaviest
measured component should be used. Since no experimentally TcSCN ¼ 18:3908r0:3702 0:5949
SCN TbSCN (9)
determined densities apart from that of the C7þ fraction are usually
available, we have chosen to initialize all density calculations by
2:2909
using the value of Katz and Firoozabadi for the SCN with 6 carbon PcSCN ¼ 5:103: 107 r2:3981
SCN TbSCN (10)
atoms that is equal to 685 kg/m3 [45].
After the calculation of z, MW and density for each SCN, physical
properties are assigned to them: true boiling point (Tb), critical tSCN ¼ 1  TbSCN=Tc (11)
SCN
temperature (Tc), critical pressure (Pc) and acentric factor (u).
Pedersen proposes the use of critical properties regressed from PVT
data separately for PR and SRK [98]. These correlations are designed 5:97616tSCN þ1:29874t1:5 2:5 5
SCN 0:60394tSCN 1:06841tSCN
f 0;SCN ¼
to enhance the performance of PR and SRK for gas condensate TbSCN=Tc
SCN
mixtures and for this reason they are used in this work for the
(12)
corresponding EoS. However, they are EoS specific and cannot be
applied to another EoS. For the calculation of critical properties

Fig. 5. Saturation pressure cross plots for UMR-PRU, PR, SRK and Elsharkawy correlation for real gas condensate mixtures.
70 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73

Fig. 6. Liquid dropout curves for real gas condensate mixtures. Diamond markers: experimental data; Blue line: PR; Green line: SRK; Red line: UMR-PRU. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)
 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73
Fig. 6. (continued).
5:03365tSCN þ1:11505t1:5 2:5 5
SCN 5:41217tSCN 7:46628tSCN
f 1;SCN ¼
TbSCN=Tc
SCN
lnðPcSCN =1:01325Þ þ f 0;SCN
uSCN ¼ 
f 1;SCN
P
SCNend
ZSCN PropertySCN
PropertyMCN ¼ SCNstart
P
SCNend
ZSCN
SCNstart
For UMR-PRU, a third step is also required in the characteriza-
tion procedure, that is the assignment of each pseudocomponent to
a compound with specific chemical structure. The structure that is
selected is that of the normal alkane with the closest molecular
weight to that of the MCN pseudocomponent.
The dew point predictions for the RGC systems with SRK and PR
coupled with the original Pedersen method with zero binary
interaction parameters as suggested by Pedersen [49], and the
UMR-PRU model coupled with the adapted Pedersen method
described above, are presented in cross plots of dew point pressure
in Fig. 5. Predictions obtained with the correlation proposed by
Elsharkawy [42], which is based on reservoir temperature and
compositions, are also presented in the same figure for comparison
purposes. UMR-PRU leads to very satisfactory dew point pre-
dictions as shown from the low average absolute deviation (AAD)
and the high correlation coefficient (R2). The AAD in dew point
pressure is 6.8% for the UMR-PRU model, 9.0% for SRK, 10.0% for PR
EoS and 10.3% for the Elsharkwawy correlation. Furthermore, UMR-
PRU model gives the highest R2 as compared to the other methods.
Typical liquid dropout predictions for real gas condensate
mixtures are illustrated in Fig. 6. Experimental data for some of
these mixtures are available at various temperatures. For the
application of the models the characterization scheme for the C7þ
fraction presented above has been followed. As in the case of SGC
71
mixtures, PR and UMR-PRU give very satisfactory results with
UMR-PRU being overall better than PR. On the other hand, SRK
overpredicts the experimental data in most cases. The experi-
mental liquid dropout values increase with decreasing temperature
(13) and increasing C7þ fraction, a trend that is captured correctly by all
models.
(14) 4. Conclusions
The UMR-PRU model has been evaluated in the prediction of
dew points and liquid dropouts of gas condensate mixtures, and
has been compared with PC-SAFT, SRK and PR. In the evaluation,
(15) both synthetic and real gas condensate mixtures have been
considered. For real fluids, the characterization method of Pedersen
et al. has been properly modified so as to be applicable to the UMR-
PRU model.
For synthetic gas condensates, it was found that PC-SAFT over-
predicts in most cases the experimental dew point data, cubic EoSs
fail to describe mixtures that contain polyaromatic components,
while UMR-PRU yields the lowest overall deviation from the
experimental data. For liquid dropouts of synthetic gas conden-
sates, the predictions of PR, PC-SAFT and UMR-PRU are very satis-
factory for the low-pressure part of the liquid dropout curve up to
the maximum liquid dropout, while poor results are obtained in
most cases by SRK. For the high-pressure part of the curve high
errors are observed especially for the near-critical mixtures.
For real gas condensate mixtures, UMR-PRU leads to very
satisfactory dew point predictions exhibiting the lowest deviations
from the experimental data. For liquid dropouts, PR and UMR-PRU
give the best results with UMR-PRU being overall better than PR,
while SRK in most cases overpredicts the experimental data.
Acknowledgements
This work has been financed by Gassco and Statoil. Gassco is the
operator for the Norwegian gas transport system and of several gas
processing and receiving terminals in Norway and in Europe. Statoil
is an energy company with substantial international activities and
the operator of a large part of Norwegian oil and gas production.
72 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73

List of symbols [12] A. Haghtalab, P. Mahmoodi, S.H. Mazloumi, A modified PengeRobinson

equation of state for phase equilibrium calculation of liquefied, synthetic
natural gas, and gas condensate mixtures, Can. J. Chem. Eng. 89 (6) (2011)
AAD average absolute deviation 1376e1387.
BIP Binary Interaction Parameter [13] A. Peneloux, E. Rauzy, R. Fre ze, A consistent correction for Redlich-Kwong-
SCN single carbon number Soave volumes, Fluid Phase Equil. 8 (1) (1982) 7e23.
[14] J. Ahlers, T. Yamaguchi, J. Gmehling, Development of a universal group
HC Hydrocarbon contribution equation of state. 5. Prediction of the solubility of high-boiling
HCset1 HC less heavy than n-decane compounds in supercritical gases with the group contribution equation of
HCset2 HC at least as heavy as n-decane state volume-translated PengRobinson, Ind. Eng. Chem. Res. 43 (20) (2004)
6569e6576.
m Number of segments per chain [15] J.-N. Jaubert, et al., Note on the properties altered by application of a
MCN multiple carbon number Penelouxetype volume translation to an equation of state, Fluid Phase Equil.
MW molecular weight 419 (2016) 88e95.
[16] G.M. Kontogeorgis, et al., A novel method for investigating the repulsive and
NDP Number of data Points attractive parts of cubic equations of state and the combining rules used with
P Pressure (bar) the vdW-1f theory, Chem. Eng. Sci. 53 (3) (1998) 541e552.
RGC real gas condensates [17] S. Lasala, et al., VLE properties of CO2 e based binary systems containing N2,
O2 and Ar: experimental measurements and modelling results with
SCN single carbon number advanced cubic equations of state, Fluid Phase Equil. 428 (2016) 18e31.
SG specific gravity [18] V. Louli, et al., Application of the UMR-PRU model to multicomponent sys-
SGC synthetic gas condensates tems: prediction of the phase behavior of synthetic natural gas and oil sys-
tems, Fluid Phase Equil. 261 (1e2) (2007) 351e358.
T Absolute temperature (K)
[19] E. Skylogianni, et al., Measurement and prediction of dew points of six
VpL% volume percent liquid natural gases, Fluid Phase Equil. 424 (2016) 8e15.
Z Mole fraction [20] V. Louli, et al., Measurement and prediction of dew point curves of natural
gas mixtures, Fluid Phase Equil. 334 (2012) 1e9.
[21] N. Spiliotis, et al., Application of the LCVM model to multicomponent sys-
Greek letters tems: extension of the UNIFAC interaction parameter table and prediction of
u Acentric factor the phase behavior of synthetic gas condensate and oil systems, Fluid Phase
Equil. 101 (0) (1994) 187e210.
s segment diameter (A)
[22] N.C. Patel, A.S. Teja, A new cubic equation of state for fluids and fluid mix-
ε/k segment energy parameter (K) tures, Chem. Eng. Sci. 37 (3) (1982) 463e473.
r density (kg/m3) [23] J.O. Valderrama, A generalized Patel-Teja equation of state for polar and
nonpolar fluids and their mixtures, J. Chem. Eng. Jpn. 23 (1) (1990) 87e91.
[24] F. Esmaeilzadeh, M. Roshanfekr, A new cubic equation of state for reservoir
Subscripts fluids, Fluid Phase Equil. 239 (1) (2006) 83e90.
b Boiling [25] M. Cismondi, J. Mollerup, Development and application of a three-parameter
RKePR equation of state, Fluid Phase Equil. 232 (1) (2005) 74e89.
c Critical
[26] T.-M. Guo, L. Du, A three-parameter cubic equation of state for reservoir
Cþ plus fraction fluids, Fluid Phase Equil. 52 (1989) 47e57.
exp Experimental value [27] M. Bonyadi, F. Esmaeilzadeh, Prediction of gas condensate properties by
EsmaeilzadeheRoshanfekr equation of state, Fluid Phase Equil. 260 (2)
calc Predicted value
(2007) 326e334.
[28] K. Nasrifar, O. Bolland, M. Moshfeghian, Predicting natural gas dew points
Appendix A. Supplementary data from 15 equations of state, Energy & Fuels 19 (2) (2005) 561e572.
[29] T.H. Ahmed, Comparative study of eight equations of state for predicting
hydrocarbon volumetric phase behavior, SPE Reserv. Eng. 3 (01) (1988)
Supplementary data related to this article can be found at 337e348.
https://doi.org/10.1016/j.fluid.2017.10.024. [30] R. Bagherinia, M. Assareh, F. Feyzi, An improved thermodynamic model for
Wax precipitation using a UNIQUAC þ PC-SAFT approach, Fluid Phase Equil.
425 (2016) 21e30.
References [31] S.R. Panuganti, et al., SAFT model for upstream asphaltene applications, Fluid
Phase Equil. 359 (2013) 2e16.
[1] D.-Y. Peng, D.B. Robinson, A new two-constant equation of state, Ind. Eng. [32] S.R. Panuganti, et al., PC-SAFT characterization of crude oils and modeling of
Chem. Fundam. 15 (1) (1976) 59e64. asphaltene phase behavior, Fuel 93 (2012) 658e669.
[2] G. Soave, Equilibrium constants from a modified Redlich-Kwong equation of [33] F.M. Vargas, et al., Modeling asphaltene phase behavior in crude oil systems
state, Chem. Eng. Sci. 27 (6) (1972) 1197e1203. using the perturbed chain form of the statistical associating fluid theory (PC-
[3] M.S. Graboski, T.E. Daubert, A modified Soave equation of state for phase SAFT) equation of state, Energy & Fuels 23 (3) (2009) 1140e1146.
equilibrium calculations. 1. Hydrocarbon systems, Ind. Eng. Chem. Process [34] F.M. Vargas, et al., Development of a general method for modeling asphal-
Des. Dev. 17 (4) (1978) 443e448. tene stability, Energy & Fuels 23 (3) (2009) 1147e1154.
[4] P.M. Mathias, T.W. Copeman, Extension of the Peng-Robinson equation of [35] D.L. Gonzalez, et al., Modeling study of CO2-induced asphaltene precipita-
state to complex mixtures: evaluation of the various forms of the local tion, Energy & Fuels 22 (2) (2007) 757e762.
composition concept, Fluid Phase Equil. 13 (0) (1983) 91e108. [36] D.L. Gonzalez, et al., Modeling of asphaltene precipitation due to changes in
[5] R. Stryjek, J.H. Vera, PRSV: an improved Pengdrobinson equation of state for composition using the perturbed chain statistical associating fluid theory
pure compounds and mixtures, Can. J. Chem. Eng. 64 (2) (1986) 323e333. equation of state, Energy & fuels 21 (3) (2007) 1231e1242.
[6] C.H. Twu, J.E. Coon, J.R. Cunningham, A new generalized alpha function for a [37] D.P. Ting, et al., Applications of the SAFT equation of state to asphaltene
cubic equation of state Part 1. Peng-Robinson equation, Fluid Phase Equil. phase behavior, in: Asphaltenes, Heavy Oils and Petroleomics, 2006.
105 (1) (1995) 49e59. [38] D.L. Gonzalez, et al., Prediction of asphaltene instability under gas injection
[7] K. Nasrifar, M. Moshfeghian, Vaporeliquid equilibria of LNG and gas with the PC-SAFT equation of state, Energy & Fuels 19 (4) (2005) 1230e1234.
condensate mixtures by the NasrifareMoshfeghian equation of state, Fluid [39] K. Schou Pedersen, et al., Modeling of EOR PVT Data Using PC-SAFT Equation,
Phase Equil. 200 (1) (2002) 203e216. Society of Petroleum Engineers, 2012.
[8] H. Saffari, A. Zahedi, A new alpha-function for the peng-robinson equation of [40] W. Yan, F. Varzandeh, E.H. Stenby, PVT modeling of reservoir fluids using PC-
state: application to natural gas, Chin. J. Chem. Eng. 21 (10) (2013) SAFT EoS and Soave-BWR EoS, Fluid Phase Equil. 386 (2015) 96e124.
1155e1161. [41] H. Liu, et al., Phase behavior and compressibility factor of two China gas
[9] A. Danesh, et al., Improving predictions of equation of state by modifying its condensate samples at pressures up to 95MPa, Fluid Phase Equil. 337 (2013)
parameters for super critical components of hydrocarbon reservoir fluids, 363e369.
Fluid Phase Equil. 112 (1) (1995) 45e61. [42] A.M. Elsharkawy, Predicting the dew point pressure for gas condensate
[10] K. Nasrifar, O. Bolland, Square-well potential and a new a function for the reservoirs: empirical models and equations of state, Fluid Phase Equil. 193
SoaveRedlichKwong equation of state, Ind. Eng. Chem. Res. 43 (21) (2004) (1e2) (2002) 147e165.
6901e6909. [43] D. Galatro, F. Marín-Cordero, Considerations for the dew point calculation in
€ter, T.W. de Loos, J. de Swaan Arons, Improved modeling of the phase
[11] E. Flo rich natural gas, J. Nat. Gas Sci. Eng. 18 (0) (2014) 112e119.
behavior of asymmetric hydrocarbon mixtures with the Peng-Robinson [44] K.S. Pedersen, A.L. Blilie, K.K. Meisingset, PVT calculations on petroleum
equation of state using a different temperature dependency of the param- reservoir fluids using measured and estimated compositional data for the
eter a, Ind. Eng. Chem. Res. 37 (5) (1998) 1651e1662. plus fraction, Ind. Eng. Chem. Res. 31 (5) (1992) 1378e1384.
 N. Novak et al. / Fluid Phase Equilibria 457 (2018) 62e73
[45] D.L. Katz, A. Firoozabadi, in: Predicting Phase Behavior of Condensate/Crude-
oil Systems Using Methane Interaction Coefficients, 1978.
[46] K.S. Pedersen, A. Fredenslund, P. Thomassen, Properties of Oils and Natural
Gases, vol. 5, Gulf Pub Co, 1989.
[47] C.H. Whitson, Characterizing Hydrocarbon Plus Fractions, 1983.
[48] K.H. Coats, Simulation of gas condensate reservoir performance, J. Petrol.
Technol. 37 (10) (1985) 1,870e1,886.
[49] K.S. Pedersen, P. Thomassen, A. Fredenslund, Thermodynamics of petroleum
mixtures containing heavy hydrocarbons. 3. Efficient flash calculation pro-
cedures using the SRK equation of state, Ind. Eng. Chem. Process Des. Dev. 24
(4) (1985) 948e954.
[50] C.H. Whitson, Effect of C7þ properties on equation of state predictions, Soc.
Pet. Eng. J. 24 (6) (1984) 685.
[51] K.S. Pedersen, P. Thomassen, A. Fredenslund, Thermodynamics of petroleum
mixtures containing heavy hydrocarbons. 1. Phase envelope calculations by
use of the Soave-Redlich-Kwong equation of state, Ind. Eng. Chem. Process
Des. Dev. 23 (1) (1984) 163e170.
[52] H.A. Aladwani, M.R. Riazi, Some guidelines for choosing a characterization
method for petroleum fractions in process simulators, Chem. Eng. Res. Des.
83 (2) (2005) 160e166.
[53] C.H. Whitson, Effect of C7þ properties on equation-of-state predictions, Soc.
Pet. Eng. J. 24 (6) (1984) 685.
[54] C. Li, et al., Prediction of the dew point pressure for gas condensate using a
modified PengeRobinson equation of state and a four-coefficient molar
distribution function, J. Nat. Gas Sci. Eng 27 (2015) 967e978.
[55] K. Folsta, G. Camargo, R. Espo sito, Gas condensate characterization from
chromatogram areas and retention times, Fluid Phase Equil. 292 (1) (2010)
87e95.
[56] A. Shariati, C.J. Peters, M. Moshfeghian, Further evaluation of the shar-
iatiepetersemoshfeghian C7þ characterization method, Fluid Phase Equil.
179 (1e2) (2001) 23e41.
[57] M. Sportisse, A. Barreau, P. Ungerer, Modeling of gas condensates properties
using continuous distribution functions for the characterisation of the heavy
fraction, Fluid Phase Equil. 139 (1e2) (1997) 255e276.
[58] C.H. Whitson, Society of Petroleum Engineers of AIME Report 11200, vol.
1982, 1982.
[59] K. Schou Pedersen, C. Hasdbjerg, PC-SAFT Equation of State Applied to Pe-
troleum Reservoir Fluids, Society of Petroleum Engineers, 2007.
[60] X. Liang, et al., On petroleum fluid characterization with the PC-SAFT
equation of state, Fluid Phase Equil. 375 (2014) 254e268.
[61] E. Voutsas, et al., Thermodynamic property calculations with the universal
mixing rule for EoS/GE models: results with the PengeRobinson EoS and a
UNIFAC model, Fluid Phase Equil. 241 (1e2) (2006) 216e228.
[62] D.B. Robinson, D.Y. Peng, The Characterization of the Heptanes and Heavier
Fractions for the GPA Peng-Robinson Programs, Gas Processors Association
Research Report RR-28, 1978.
[63] E. Voutsas, K. Magoulas, D. Tassios, Universal mixing rule for cubic equations
of state applicable to symmetric and asymmetric Systems: results with the
PengRobinson equation of state, Ind. Eng. Chem. Res. 43 (19) (2004)
6238e6246.
[64] T.E. Daubert, R.P. Danner, Physical and Thermodynamic Properties of Pure
Chemicals, Data Compilation, Hemisphere, New York, 1994.
[65] J. Gross, G. Sadowski, Application of perturbation theory to a hard-chain
reference fluid: an equation of state for square-well chains, Fluid Phase
Equil. 168 (2) (2000) 183e199.
[66] J. Gross, G. Sadowski, Perturbed-chain SAFT: an equation of state based on a
perturbation theory for chain molecules, Ind. Eng. Chem. Res. 40 (4) (2001)
1244e1260.
[67] M.R. Jensen, et al., Crystallisation of heavy hydrocarbons from three syn-
thetic condensate gases at high pressure, Fluid Phase Equil. 208 (1e2) (2003)
247e260.
[68] P. Ungerer, et al., High pressure-high temperature reservoir fluids: investi-
gation of synthetic condensate gases containing a solid hydrocarbon, Fluid
Phase Equil. 111 (2) (1995) 287e311.
[69] F. Gozalpour, et al., Vapoureliquid equilibrium volume and density mea-
surements of a five-component gas condensate at 278.15e383.15 K, Fluid
Phase Equil. 206 (1e2) (2003) 95e104.
[70] F. Gozalpour, et al., Viscosity and density data for a six-component model
fluid at temperatures to 423 K and pressures to 140 MPa, High. Temp. High.
Press. 34 (4) (2002) 483e486.
[71] P. Arce, M. Aznar, Modeling of critical lines and regions for binary and
ternary mixtures using non-cubic and cubic equations of state, J. Supercrit.
Fluids 42 (1) (2007) 1e26.
[72] D.N. Justo-García, et al., Calculation of critical points for multicomponent
mixtures containing hydrocarbon and nonhydrocarbon components with
the PC-SAFT equation of state, Fluid Phase Equil. 265 (1e2) (2008) 192e204.
[73] H. Nourozieh, M. Kariznovi, J. Abedi, Vaporeliquid equilibrium measurement
and thermodynamic modeling of binary systems (methane þ n-tetradecane),
Fluid Phase Equil. 318 (2012) 96e101.
[74] M.P.W.M. Rijkers, et al., Measurements on the phase behavior of binary
mixtures for modeling the condensation behavior of natural gas: Part II. The
system methane þ dodecane, Fluid Phase Equil. 72 (1992) 309e324.
[75] S. Srivastan, et al., Solubility of methane in hexane, decane, and dodecane at
73
temperatures from 311 to 423 K and pressures to 10.4 MPa, J. Chem. Eng.
Data 37 (4) (1992) 516e520.
[76] P.D. Ting, et al., Phase equilibrium modeling of mixtures of long-chain and
short-chain alkanes using PengeRobinson and SAFT, Fluid Phase Equil. 206
(1e2) (2003) 267e286.
[77] J.F. Arnaud, et al., Excess volumes and saturation pressures for the system
methane þ n-tetracosane at 374 K. Representation by improved EOS mixing
rules, Fluid Phase Equil. 124 (1e2) (1996) 177e207.
[78] E. Flo€ter, T.W. De Loos, J. de Swaan Arons, High pressure solid-fluid and
vapour-liquid equilibria in the system (methaneþ tetracosane), Fluid phase
Equil. 127 (1) (1997) 129e146.
[79] M.P.W.M. Rijkers, C.J. Peters, J. de Swaan Arons, Measurements on the phase
behavior of binary mixtures for modeling the condensation behavior of
natural gas: Part III. The system methane þ hexadecane, Fluid Phase Equil. 85
(1993) 335e345.
[80] H.-M. Lin, H.M. Sebastian, K.-C. Chao, Gas-liquid equilibrium in hydrogen þ
n-hexadecane and methane þ n-hexadecane at elevated temperatures and
pressures, J. Chem. Eng. Data 25 (3) (1980) 252e254.
[81] M. Glaser, et al., Phase equilibria of (methane þ n-hexadecane) and (p, Vm, T)
of n-hexadecane, J. Chem. Thermodyn. 17 (9) (1985) 803e815.
[82] H.M. Sebastian, et al., Gas-liquid equilibrium in binary mixtures of methane
with tetralin, diphenylmethane, and 1-methylnaphthalene, J. Chem. Eng.
Data 24 (2) (1979) 149e152.
[83] E. Flo€ter, et al., High-pressure phase equilibria in the systems methaneþ
phenanthrene and methaneþ 1-phenyldodecane up to 400 MPa, Fluid Phase
Equil. 134 (1) (1997) 1e19.
[84] P.V. Malone, R. Kobayashi, Light gas solubility in phenanthrene: the hydro-
gendphenanthrene and methanedphenanthrene systems, Fluid Phase
Equil. 55 (1) (1990) 193e205.
[85] N.A. Darwish, K.A. Gasem, R.L.J. Robinson, Solubility of methane in benzene,
naphthalene, phenanthrene, and pyrene at temperatures from 323 to 433 K
and pressures to 11.3 MPa, J. Chem. Eng. Data 39 (4) (1994) 781e784.
[86] S. Stamataki, D. Tassios, Performance of cubic EOS at high pressures, Oil Gas
Sci. Technol. 53 (3) (1998) 367e377.
[87] J.M. Prausnitz, R.N. Lichtenthaler, E.G. de Azevedo, Molecular Thermody-
namics of Fluid-phase Equilibria, Pearson Education, 1998.
[88] S.-P. Hong, K.A. Green, K.D. Luks, Phase equilibria of the mixtures methane þ
n-hexane þ n-hexatriacontane, methane þ toluene þ naphthalene, and
methane þ n-hexane þ naphthalene, Fluid Phase Equil. 87 (2) (1993)
255e272.
[89] E. Flo€ter, T.W. de Loos, J. de Swaan Arons, Hyperbaric reservoir fluids: high-
pressure phase behavior of asymmetric methane þn-alkane systems, Int. J.
Thermophys. 16 (1) (1995) 185e194.
[90] J.J.B. Machado, T.W. de Loos, High pressure solidefluid and vapoureliquid
equilibria in model hyperbaric fluids: the system methane þ tetracosane þ
triacontane, Fluid Phase Equil. 226 (2004) 83e90.
[91] L. Yarborough, Vapor-liquid equilibrium data for multicomponent mixtures
containing hydrocarbon and nonhydrocarbon components, J. Chem. Eng.
Data 17 (2) (1972) 129e133.
[92] T. Yang, W.D. Chen, T.M. Guo, Phase behavior of a near-critical reservoir fluid
mixture, Fluid Phase Equil. 128 (1e2) (1997) 183e197.
[93] W. Yan, et al., A systematic experimental study on the phase behavior of
complex fluid mixtures up to near-critical region, Fluid Phase Equil. 190
(1e2) (2001) 159e178.
[94] K. Coats, G. Smart, Application of a regression-based EOS PVT program to
laboratory data, SPE Reserv. Eng. 1 (03) (1986) 277e299.
[95] C.W. Donohoe, R.D. Buchanan Jr., Economic evaluation of cycling gas-
condensate reservoirs with nitrogen, J. Petrol. Technol. 33 (02) (1981)
263e270.
[96] A. Firoozabadi, Y. Hekim, D.L. Katz, Reservoir depletion calculations for gas
condensates using extended analyses in the peng-robinson equation of state,
Can. J. Chem. Eng. 56 (5) (1978) 610e615.
[97] S.M.J. Majidi, et al., Evolving an accurate model based on machine learning
approach for prediction of dew-point pressure in gas condensate reservoirs,
Chem. Eng. Res. Des. 92 (5) (2014) 891e902.
[98] K.S. Pedersen, P.L. Christensen, J.A. Shaikh, Phase Behavior of Petroleum
Reservoir Fluids, CRC Press, 2012.
[99] J.L. Vogel, L. Yarborough, The Effect of Nitrogen on the Phase Behavior and
Physical Properties of Reservoir Fluids, Society of Petroleum Engineers, 1980.
[100] K.S. Pedersen, P. Thomassen, A. Fredenslund, Thermodynamics of petroleum
mixtures containing heavy hydrocarbons. 2. Flash and PVT calculations with
the SRK equation of state, Ind. Eng. Chem. Process Des. Dev. 23 (3) (1984)
566e573.
[101] E. Retzekas, et al., Prediction of physical properties of hydrocarbons, petro-
leum, and coal liquid fractions, Ind. Eng. Chem. Res. 41 (6) (2002)
1695e1702.
[102] B.E. Poling, et al., The Properties of Gases and Liquids vol. 5, McGraw-Hill
New York, 2001.
[103] H.-M. Lin, et al., Gas-liquid equilibrium in binary mixtures of methane with
N-decane, benzene, and toluene, J. Chem. Eng. Data 24 (2) (1979) 146e149.
[104] M. Rijkers, et al., Measurements on the phase behavior of binary hydrocar-
bon mixtures for modelling the condensation behavior of natural gas: Part I.
The system methaneþ decane, Fluid phase Equil. 71 (1e2) (1992) 143e168.