Are you sure?
This action might not be possible to undo. Are you sure you want to continue?
Chapter 3 Equations of State
The simplest way to derive the Helmholtz function of a fluid is to directly integrate the
equation of state with respect to volume (Sadus, 1992a, 1994). An equation of state can be
applied to either vapour liquid or supercrit ical phenomena without any conceptual
difficulties. Therefore, in addition to liquidliquid and vapour liquid properties, it is also
possible to determine transitions between these phenomena from the same inputs. All of the
physical properties of the fluid except ideal gas are also simultaneously calculated. Many
equations of state have been proposed in the literature with either an empirical, semi
empirical or theoretical basis. Comprehensive reviews can be found in the works of Martin
(1979), Gubbins (1983), Anderko (1990), Sandler (1994), Economou and Donohue (1996),
Wei and Sadus (2000) and Sengers et al. (2000).
The van der Waals equation of state (1873) was the first equation to predict vapourliquid
coexistence. Later, the RedlichKwong equation of state (Redlich and Kwong, 1949)
improved the accuracy of the van der Waals equation by proposing a temperature
dependence for the attractive term. Soave (1972) and Peng and Robinson (1976) proposed
additional modifications of the RedlichKwong equation to more accurately predict the
vapour pressure, liquid density, and equilibria ratios. Guggenheim (1965) and Carnahan and
Starling (1969) modified the repulsive term of van der Waals equation of state and obtained
more accurate expressions for hard sphere sys tems. Christoforakos and Franck (1986)
modified both the attractive and repulsive terms of van der Waals equation of state. Boublik
(1981) extended the CarnahanStarling hard sphere term to obtain an accurate equation for
hard convex geometries.
51
In addition to modeling small and simple molecules, considerable emphasis has been placed
on modeling long and convex molecules. Based on theory of Prigogine (1957) and Flory
(1965), an equation for molecules treated as chains of segments, which is called Perturbed
HardChainTheory (PHCT) was constructed by Beret and Prausnitz (1975) and Donohue
and Prausnitz (1978). To reduce the mathematical complexity of PerturbedHardChain
Theory, Kim et al. (1986) developed a simplified version of the theory by replacing the
complex attractive part by a simpler expression. At the almost same time, Vimalchand and
Donohue (1985) obtained a fairly accurate multipolar mixture calculation by using the
Perturbed Anisotropic Chain theory, and Ikonomou and Donohue (1986) extended the
Perturbed Anisotropic Chain Theory to the Associated Perturbed Anisotropic Chain Theory
by taking into account the existence of hydrogen bonding.
Wertheim (1987) proposed a thermodynamic perturbation theory (TPT), which
accommodates hardchain molecules. Chapman et al. (1988) generalized the TPT model to
obtain the compressibility factor of a hardchain of segments. Ghonasagi and Chapman
(1994) and Chang and Sandler (1994) modified TPT for the hardsphere chain by
incorporating structural information for the diatomic fluid (TPTD). Sadus (1995) derived
the simplified thermodynamic perturbation theory–dimer (STPTD) equation from TPTD.
Sadus (1999b) later developed STPTD to the empirical simplified thermodynamic
perturbation theorydimer (ESTPTD) equation, and tested the accuracy of the equation
against simulation data for hardsphere chains containing up to 201 hardsphere segments.
Sadus (1999 a) also derived an equation of state for hard convex body chains from the TPT
of hard sphere chains (Wertheim, 1987; Chapman et al., 1988).
52
Jin et al. (1993), Povodyrev et al. (1996) and Kiselev (1997) developed theoretical crossover
equations of state for pure fluids and binary mixtures which incorporate the scaling laws
asymptotically close to the critical point and which are transformed into the regular classical
expansion far away from the critical point. Kiselev (1998) used the modified PatelTeja
cubic equation of state (Patel and Teja, 1982) as a starting point to propose a general
procedure for transforming any classical equation of state into a crossover equation of state.
Wyczalkowska et al. (1999) also developed a global crossover equation of state, and Kiselev
et al. (2001) extended the crossover modification of the statisticalassociating fluid theory
(Chapman et al., 1988, 1990) equation of state. Wyczalkowska et al. (2000) demonstrated
that both pure H
2
O and D
2
O obey a universal scaled crossover equation of state in the critical
region which satisfies corresponding state for the nearcritical behaviour of the
thermodynamic properties. Abdulkadirova et al. (2002) extended the pure component to
binary mixtures (H
2
O + D
2
O) using a crossover equation of state.
In this chapter we will review equations of state for high pressure with particular emphasis
on critical points calculations. A crossover theory for equations of state is also discussed.
53
3.1 Equations of State for High Pressure Equilibria and Critical
Phenomena
3.1.1 Cubic Equations of State
Cubic equations of state are equations, which when expanded have volume terms raised to
the first, second, and third power. Most commonly encountered phase equilibrium
calculations, such as vapourliquid equilibria, involve only two phases for which a cubic
equation is suitable. Cubic equations have the advantage that the three values of volume can
be obtained analytically without the need for an iterative solution procedure.
The van der Waals equation of state (1873) is the simplest cubic equation of state for fluid
phase equilibria. It can be regarded as a “hard sphere term + attractive term” equation of
state composed from the contribution of repulsive and attractive intermolecular interactions
(Sadus, 1994). The van der Waals equation was the first equation capable of representing
vapour liquid coexistence. The pressure (p) is related to the temperature (T), ideal gas
constant (R) and molar volume (V) via:
2
V
a
b V
RT
p −
−
· (3.1)
It has two pure component parameters a and b. The parameter a is a measure of the attractive
forces between the molecules, and b is related to the size of the molecules. Since the van der
Waals equation of state is cubic in volume, three volumes exist for any given temperature
and pressure. Usually, p
c
, T
c
and V
c
are known and Eq. 2.1 (Chapter 2) may be
simultaneously solved to provide solutions for the parameters a and b:
54
c
c
p
T R
a
64
27
2 2
· (3.2)
c
c
p
RT
b
8
· (3.3)
Van Konynenburg and Scott (1980) successfully demonstrated that most of the critical
equilibria exhibited by binary mixtures could be qualitatively predicted by the van der Waals
equation of state, but it is rarely sufficiently accurate for critical properties and phase
equilibria calculations. For example, the critical compressibility factor of all fluids including
pure component and binary mixtures predicted by van der Waals equation is 0.375 (Eq. 3.2
and Eq. 3.2 are used in Eq. 2.1 at the critical point), whereas the real value for different
hydrocarbons varies from 0.24 to 0.29. Many modifications at the van der Waals equation of
state have been proposed to address this deficiency. Five examples of cubic equations of
state based on the van der Waals equation are listed in Table 3.1.
55
Table 3.1 Examples of Improved Cubic Equations of State
Reference Equations of state Eq.
RedlichKwong (1949)
5 . 0
) ( T b V V
a
b V
RT
p
+
−
−
· (3.4)
SoaveRedlichKwong (1972)
) (
) (
b V V
T a
b V
RT
p
+
−
−
· (3.5)
PengRobinson (1976)
) ( ) (
) (
b V b b V V
T a
b V
RT
p
− + +
−
−
· (3.6)
StryjekVeraPengRobinson
(1986)
) ( ) (
) , (
b V b b V V
T a
b V
RT
p
− + +
−
−
·
θ
(3.7)
PatelTeja (1982)
) ( ) (
) (
b V c b V V
T a
b V
RT
p
− + +
−
−
· (3.8)
Eq. (3.4) was proposed as an empirical modification of the van der Waals equation to make
the attractive term temperaturedependent. The parameters a and b are usually expressed as
c
c
p
T R
a
5 . 2 2
4278 . 0 · (3.9)
c
c
p
RT
b 0867 . 0 · (3.10)
Spear et al. (1969) demonstrated that the RedlichKwong equation of state could be used to
reliably calculate the vapour liquid critical properties of binary mixtures. Chueh and
Prausnitz (1967a, b) also showed that the RedlichKwong equation can be adapted to predict
both vapour and liquid properties. Deiters and Schneider (1976) and Baker and Luks (1980)
56
have successfully applied the RedlichKwong equation to the high pressure phase equilibria
of binary mixtures. Deiters and Pegg (1989) used the RedlichKwong equation with
quadratic mixing rules to calculate phase diagrams for binary fluid mixtures and to classify
them according to the global phase diagram. Polishuk et al. (1999) predicted the closed loop
critical behaviour between type III and V with the RedlichKwong equation of state.
Soave (1972) suggested to replace the term
5 . 0
T
a
with a more general temperaturedependent
term a(T) (Eq. (3.5) in the Table 3.1), where
2 5 . 0
2 2
]} ) ( 1 [ 1 ){ ( 4274 . 0 ) (
c c
c
T
T
m
p
T R
T a − + · (3.11)
2
176 . 0 57 . 1 480 . 0 ω ω − + · m (3.12)
c
c
p
RT
b 08664 . 0 · (3.13)
and ω is the acentric factor. Soave (1972) calculated the vapour pressures of a number of
hydrocarbons and several binary systems with the SoaveRedlichKwong equation, and
compared the results of the calculations with experimental data. In contrast to the original
RedlichKwong equation, Soave’s modification fitted the experimental (vapourliquid) curve
well and it was able to predict the phase behaviour of mixtures in the critical region. Elliott
and Daubert (1987) improved the accuracy of the calculated critical properties of 95 binary
57
systems containing hydrocarbons. Zheng et al. (1999) also used the SoaveRedlichKwong
equation to calculate the phase equilibria of systems containing methane, carbon dioxide and
nitrogen, and compared the results of calculation with experimental data.
Peng and Robinson (1976) redefined a(T) as
2 5 . 0
2 2
]} ) ( 1 [ 1 { 45724 . 0 ) (
c c
c
T
T
k
p
T R
T a − + · (3.14)
2
26922 . 0 5422 . 1 37464 . 0 ω ω − + · k (3.15)
c
c
p
RT
b 07780 . 0 · (3.16)
The Peng and Robinson equation of state was proposed as Eq. (3.6), it slightly improves the
prediction of liquid volumes and predicts a critical compressibility factor of 307 . 0 ·
c
Z (the
critical compressibility factor of the RedlichKwong equation is 0.333, and that van der
Waals is 0.375). Peng and Robinson (1977) gave examples of the use of the PengRobinson
equation for predicting the vapour pressure and volumetric behaviour of pure component and
binary mixtures. In general, Eq. (3.6) performed as well as or better than Eq. (3.5). Han et al.
(1988) reported that the PengRobinson equation of state was superior for predicting vapour
liquid equilibrium in hydrogen and nitrogen containing mixtures. Harstad et al. (1997)
showed that the PengRobinson equation of state could be used to obtain a relatively
accurate, noniterative and computationally efficient correlation of highpressure fluid
58
mixtures used in gas turbines and rocket engines. Zhao and Olesik (1999) used the Peng
Robinson equation of state to calculate methanol and CHF
3
binary mixtures. The results
showed that the PengRobinson equation of state with two temperatureindependent binary
parameters was capable of representing the experimental data over the entire temperature
range with an average relative deviation within 6%.
Stryjek and Vera (1986) proposed the following modified temperature functionality for the
PengRobinson equation of state to extend the range of applicability to polar components:
)} 7 . 0 ]( ) ( 1 [ { 45724 . 0 ) (
5 . 0
1
2 2
c c c
c
T
T
T
T
k k
p
T R
T a − + + · (3.17)
3 2
0196554 . 0 17131848 . 0 4897153 . 1 378893 . 0 ω ω ω + − + · k (3.18)
c
c
p
RT
b 0778 . 0 · (3.19)
where
1
k is an adjustable parameter characteristic of each pure component. The Stryjek
VeraPengRobinson equation of state (Stryjek and Vera, 1986) was proposed as Eq. (3.7).
In addition, it must be noticed that optimization of
1
k values was performed using a
particular set of critical constants (Stryjek and Vera, 1986). The correct application of
StryjekVeraPengRobinson equation of state requires usage of this set.
59
The PengRobinson and SoaveRedlickKwong equations are widely used in industry
(Sadus, 1994). The advantages of these equations are they are easy to use and that they often
accurately represent the relation between temperature, pressure, and phase compositions in
binary and multicomponent systems. These equations only require the critical properties and
acentric factor for the generalized parameters. Little computer resources are required and
those lead to good phase equilibrium correlation. However, the success of these
modifications is restricted to the estimation of phase equilibria pressure. The calculated
saturated liquid volumes are not improved and they are invariably higher than the measured
data.
Patel and Teja (1982) proposed a third constant c into the van der Waals equation’s attraction
term, resulting in Eq. (3.8). The equation of state allows for adjustment of the critical
compressibility factor Z
c
instead of predicting a fixed value. This added flexibility improves
saturation property predictions for polar fluids while maintaining a simple form that doesn’t
require a large amount of pure component or mixture experimental data for accuracy.
As with the van der Waals equation of state, the parameters a and b in the PatelTeja
equation are evaluated using the conditions at the critical point given by Eq. (2.1). The
condition for the third parameter c is:
c
c c
c
RT
V p
· η (3.20)
60
Instead of having a fixed value of the critical compressibility factor fixed by parameters a
and b, the value is the arbitrary, substance specific, empirical parameter
c
η .
c
η is
determined by minimizing the errors between experimental saturated liquid volume and
those calculated by the equation of state. Patel and Teja (1982) calculated
c
η for 38
substances. They also correlated
c
η to the acentric factor ω , for nonpolar substances
yielding the following equation.
2
0211947 . 0 076799 . 0 329032 . 0 ω ω η + − ·
c
(3.21)
Satisfying the conditions of Eq. (2.1) and Eq. (3.20) yields the following equations for the
parameters a, b, and c:
) ( ) (
2 2
T
p
T R
T a
c
c
a
α Ω · (3.22)
c
c
b
P
RT
b Ω · (3.23)
c
c
c
p
RT
c Ω · (3.24)
where,
c c
η 3 1− · Ω (3.25)
61
c b b c c a
η η η 3 1 ) 2 1 ( 3 3
2 2
− + Ω + Ω − + · Ω (3.26)
and
b
Ω is the smallest positive root of the following cubic equation:
0 3 ) 3 2 (
3 2 2 3
· − Ω + Ω − + Ω
c b c b c b
η η ζ (3.27)
In the Eq. (3.22), the term ) (T α is given by:
} ] ) ( 1 [ 1 { ) (
5 . 0
F
T
T
T
c
− + · α (3.28)
F is a substance specific empirical parameter also determined by minimizing errors between
experimental saturated vapour pressures and those calculated by the equation of state. Patel
and Teja (1982) calculated F for the same 38 substances and also correlated it to the acentric
factor for nonpolar substances:
2
295937 . 0 30982 . 1 452413 . 0 ω ω − + · F (3.29)
Smith (1995) gave an expression for the compressibility factor Z,
0 ) ( ) 2 ( ) 1 (
2 2 2 3
· − + + + − − − − + − + AB BC C B Z A C B B BC Z C Z (3.30)
62
where
2
) ( RT
ap
A · (3.31)
RT
bp
B · (3.32)
RT
cp
C · (3.33)
Eq, (3.30) will yield one to three positive real roots depending upon the number of phases in
the system. If a vapour and liquid phase are present, the smallest root represents the liquid’s
compressibility while the largest root is that of the vapour.
Palenchar et al. (1986) have reported a rare, direct comparison of the quality of prediction of
the critical locus using different cubic equations of state. In addition to the RedlichKwong,
PengRobinson and SoaveRedlichKwong equations, calculations for equations of state
proposed by Teja and Patel (1982) and Adachi et al. (1983) were reported for the carbon
dioxide + methane, butane, decane or water; methane + hexane or heptane; and propane +
hexane and water + benzene binary mixtures. They concluded that the critical temperatures
and pressures of type I systems could be adequately predicted although there was a
substantial discrepancy between theory and experiment for the critical volume. The Peng
Robinson and SoaveRedlichKwong equations yielded a more accurate representation of
mixtures exhibiting a discontinuity in critical equilibria than the Patel and Teja equation.
63
Polishuk et al. (1999) compared the critical lines predicted by RedlichKwong, Peng
Robinson, SoaveRedlichKwong, StryjekVeraPengRobinson, PatelTeja and Trebble
Bishnoi (1987) equations of state using classical mixing rules with the experimental data
available for the mixtures of methane + alkanes up to and including octane. The results show
that different cubic equation of state can yield similar semiquantitative and even quantitative
predictions with the optimised values.
3.1.2 Noncubic Equations of State
The development of cubic equations of state has been focused on obtaining an improved
empirical representation of attractive interactions. Improvements to cubic equations
emphasize obtaining better agreement with experiment at low temperatures and pressures for
phase equilibria. At high pressures and temperatures, repulsive interactions can be expected
to be the dominant influence determining the properties of the fluid. There are many accurate
representations which have been developed for the repulsive interactions of hard spheres and
which were incorporated into equation of state. In general, the effect of using an accurate
model of repulsion is to generate a noncubic equation of state.
Normally, noncubic equations of state are of two broad types. One type is obtained by
simply modifying the repulsive term of the van der Waals equation (hard sphere term).
Another type is obtained by modifying both attractive and repulsive terms or combining an
accurate hard sphere model with an empirical temperature dependent attractive contribution.
64
Many improvements to the simple van der Waals hard sphere repulsive term have been
reported (Mulero et al., 2001). Some important hard sphere expressions are summarised in
Table 3.2. A comprehensive review of hard sphere terms is given by Mulero et al. (2001).
Table 3.2 Summary of hard sphere expressions of modification of repulsive term of van der
Waals equation. Where y = b/(4V) is a temperature independent packing fraction of spherical
molecules.
Reference Hard sphere expression Eq.
ReissFrischLebowitz (1959)
3
2
) 1 (
1
y
y y
−
+ +
(3.34)
Thiele (1963)
2
2
) 1 (
3 2 1
y
y y
−
+ +
(3.35)
Guggenheim (1965)
4
) 1 (
1
y −
(3.36)
CarnahanStarliing (1969)
3
3 2
) 1 (
1
y
y y y
−
− + +
(3.37)
Scott (1971)
b V
b V
−
+
(3.38)
Boublik (1981)
3
3 2 2 2
) 1 (
) 1 3 3 ( ) 2 3 ( 1
y
y y y
−
− + − + − + α α α α
(3.39)
Kolafa and Nezbeda (1994)
3
4 3 2
) 1 (
) (
3
2
1
y
y y y y
−
+ − + +
(3.40)
MalijevskyVeverka (1999)
) 3076 . 0 5979 . 0 056 . 0 1 ( ) 1 (
3262 . 0 6539 . 1 056 . 1 1
3 2 3
3 2
y y y y
y y y
+ + + −
+ + +
(3.41)
The Guggenheim equation of state (1965) is a more accurate hard sphere model which has
been extensively used for the calculation of critical equilibra:
65
2 4
) 1 ( V
a
y V
RT
p −
−
· (3.42)
Both a and b can be obtained by using critical properties ( , / 4963 . 0
2 2
c c
p T R a ·
c c
p RT b / 18727 . 0 · ), and at critical point, the compressibility factor
3611 . 0 ) /( · ·
c c c c
RT V p Z . The Guggenheim equation has been used to predict the critical
properties of a diverse range of binary mixtures (Hicks and Young, 1976; Hurle et al.,
1977a,b; Hicks et al. 1977, 1978; Toczylkin and Young, 1977, 1980a, b, c; Waterson and
Young, 1978; Semmens et al., 1980; Sadus and Young, 1985a, b; Sadus, 1992a, 1994; and
Wang et al. 2000). Despite the diversity of the systems studied, good results were
consistently reported for the vapourliquid critical locus. The critical liquidliquid line of
type II binary mixtures was also represented adequately. In contrast, calculations involving
type III equilibria are typically only semiquantitative (Christou et al., 1986) because of the
added difficulty of predicting the transition between vapourliquid and liquidliquid
behaviour.
Carnahan and Starling (1969) obtained an accurate representation of hard sphere interactions.
The CarnahanStarlingvan der Waals (CSvdW) equation is formed by coupling this hard
sphere term with the van der Waals attractive term (Carnahan and Starling, 1972):
66
2 3
3 2
) 1 (
) 1 (
V
a
y V
y y y RT
p −
−
− + +
· (3.43)
The covolume (
c
p RT b / 18727 . 0 · ) and attractive (
c c
p T R a / 4963 . 0
2 2
· ) equation of state
parameters are related to the critical properties, and critical compressibility factor
35896 . 0 ·
c
Z .
Sadus (1993) has demonstrated the CSvdW equation of state can be used to predict type III
equilibria of nonpolar mixtures with considerable accuracy. Yelash and Kraska (1998) found
several types of closed loop liquidliquid immiscibility behaviour in binary mixtures based
on CSvdW equation of state.
Boublik (1981) has generalised the CarnahanStarling hard sphere potential for molecules of
arbitrary geometry via the introduction of a nonsphericity parameter (a). Svejda and Kohler
(1983) employed the Boublik expression in conjunction with Kihara’s (1963) concept of a
hard convex body (HCB) to obtain a generalised van der Waals equation of state:
2 3
3 2 2 2
) 1 (
] ) 1 3 3 ( ) 2 3 ( 1 [
V
a
y V
y y y RT
p −
−
− + − + − +
·
α α α α
(3.44)
67
In Eq. (3.44), α is the deviation from spherical geometry which is obtained by considering
the mean radius R
*
, surface area S
*
and volume V
*
of a convex body:
*
* *
3V
S R
· α (3.45)
In the limiting case of a hard sphere, 1 · α , and Eq. (3.44) becomes identical to the
CarnahanStarling equation Eq. (3.43).
Eq. (3.44) has been used for the calculation of the vapourliquid critical properties of binary
mixtures containing nonspherical molecules (Sadus et al., 1988; Christou et al. 1991). The
results obtained were slightly better than could be obtained from similar calculations using
the Guggenheim equation of state. Sadus (1993) proposed an alternative procedure for
obtaining the equation of state parameters. Eq. (3.44) in conjunction with this modified
procedure can be used to predict type III critical equilibria of nonpolar binary mixtures with
a good degree of accuracy.
There are many other nonspherical or hard chain equations of state which have not been used
for the calculation of critical equilibria. Vimalchand and Donohue (1989) have reviewed
several hard chain equations of state. Van Pelt et al. (1992) have applied a simplified version
68
of the perturbed hard chain theory (SPHCT) equation of state proposed by Beret and
Prausnitz (1975) to binary critical equilibria.
The SPHCT equation of state has the following form:
YV v V
Y RTzcv
V
RT
c p
* 2
*
3
2
) 1 (
2
4 ) 1 (
+
−
,
`
.

−
− + ·
η
η
η (3.46)
where 1 ) 2 / exp( − · ckT q Y ε and V v / 7405 . 0
*
· η , where v
*
is the molecular volume
parameter, ε is the intermolecular potential energy per unit area; c is 1/3 of the total number
of external degrees of freedom of a molecule; q is the number of segments per molecule and
z is the closepacked coordination number. In common with other equations of state, good
results were obtained for the critical pressures and temperatures of many binary mixtures.
The calculations cannot be expected to yield accurate critical volumes for binary mixtures
because the critical volume of the components was used as an adjustable parameter to
optimise the agreement of theory and experiment for the critical pressures and temperatures
of the pure components. Eq. (3.46) was successfully used (van Pelt et al., 1995) to
qualitatively reproduce a variety of binary mixture phenomena including type VI behaviour.
Recently, development of shape factor model equations of state were reported by Huber and
Ely (1994), and EstelaUribe and Trusler (1998).
69
Christoforakos and Franck (1986) and Heilig and Franck (1989) have developed equations
specifically for high pressure equilibria which also incorporate the CarnahanStarling hard
sphere term. The Christoforakos Franck equation of state introduces a temperature
dependent covolume term
3 . 0
) / ( 4 T T b
c
· β into the repulsive term:
2
3
3
3 3 2 2
) 1 ) / )(exp( 1 ( 4
) / 1 (
/ / / 1
V
kT RT
V
V V V
V
RT
p
− −
−
−
− + +
·
ε λ β
β
β β β
(3.47)
The attractive term was obtained by examining the virial coefficient of gases in terms of a
squarewell potential. The ε parameter reflects the depth of the potential, whereas λ is the
relative width of the well. The equation has been successfully applied to the high pressure
phase behaviour of some binary aqueous mixtures (Christoforakos and Franck, 1986; Franck,
1987). In particular, the type III critical locus of binary mixtures of H
2
O + Ar, Xe, N
2
, CH
4
or carbon dioxide is predicted accurately.
The HeiligFranck (1989) equation of state uses the same form of repulsive term of Eq.
(3.47), but it has a different attractive term which is obtained by applying the Padé
approximation of the virial coefficients of a squarewell fluid:
) / ( ) / 1 (
) / / / 1 (
3
3 3 2 2
B C V V
RTB
V V
V V V RT
p
+
−
−
− + +
·
β
β β β
(3.48)
70
where B and C are the second and third virial coefficients of a squarewell fluid respectively.
Accurate calculations of the critical properties of binary mixtures have been reported (Heilig
and Franck, 1989, 1990). Mather et al. (1993) have demonstrated that HeiligFranck equation
of state can accurately predict type III critical loci of H
2
O + Ne and H
2
O + Kr mixtures.
Deiters (1981a, b) has adopted a semiempirical approach to construct an equation with three
adjustable parameters a, b and c.
b
T R
T aI
RT
cc p
eff
eff
2
3 2
0
] 1 ) / 1 [exp( ]} ) 1 /( 2 4 [ 1 {
ρ
ρ
η η η − − − − + · (3.49)
where c c V b 06911 . 0 , 6887 . 0 , 7404 . 0 , /
0
− · · · · λ ρ η ρ and y a cT T
eff
/ ) / ( λρ + · .
Parameter I is a hard sphere constant, and the values of y range from 1 for a dilute gas and
between 0.34 and 0.46 for dense fluids. In Eq. (3.49), the Carnahan Starling expression for
hard spheres has been adjusted to accurately fit the experimental vapour pressure data for Ar
via parameter c
0
, and deviations from spherical geometry are reflected in values of c > 1. The
equation can be used to accurately predict the vapour liquid properties of binary mixtures
(Deiters, 1982; Deiters and Swaid, 1984; Calado et al., 1981). Calculations of the vapour
liquid critical locus of binary mixtures (Mainwaring et al., 1988a, b) indicate that good
results can only be obtained for mixtures of molecules of similar size. The equation has not
71
been extensively tested for noncontinuous critical equilibria, but type III critical properties of
H
2
O + benzene and H
2
O + He can be adequately predicted (Mainwaring et al., 1988a).
3.1.3 Thermodynamic Perturbation Theory Equations of State
The hard chain equations of state are potentially important for developing accurate equations
of state for macromolecules. Historically, attention has been focused on using accurate hard
sphere models to represent the repulsive interactions of geometrically simple molecules.
Progress has been achieved in depicting the properties of hard nonspherical bodies (Sadus,
1995).
Boublik (1981) used Kihara’s (1963) concept of a hard convex body (HCB) as the basis of
an equation of state for nonspherical hard bodies. Flory’s concept of excluded volume has
been incorporated (Dickman and Hall, 1986) in a reasonably accurate hard chain equation of
state. Chiew (1990) developed a rigorous theory for representing hard chains, and Song et al.
(1994) formulated the theory into a useful equation of state for polymers. Wertheim (1987)
proposed a thermodynamic perturbation theory (TPT) which accommodates hardchain
molecules. Chapman et al. (1988) generalised Wertheim’s TPT model to obtain the equation
of state for the compressibility factor of a hard chain of m segments. The form of the
compressibility factor is
72
,
`
.

∂
∂
+ − − ·
y
g
y m mZ Z
HS HS HC
) ( ln
1 ) 1 (
σ
(3.50)
where ) (σ
HS
g is the hard sphere sitesite correlation function at contact, σ is the hard
sphere diameter, 6 /
3
ρσ π m y · is the packing fraction, and ρ is the number density. The
compressibility factor of hard sphere can be accurately determined from the Carnahan
Starling equation. For the CarnahanStarling equation, the sitesite correlation function is
3
) 1 ( 2
2
) (
y
y
g
HS
−
−
· σ (3.51)
Ghonasgi and Chapman (1994) modified TPT for the hard sphere chain by incorporating
structural information for the diatomic fluid. The compressibility factor of a hard chain can
be determined from the hard sphere compressibility factor and the sitesite correlation
function at contact of both hard spheres (
HS
g ) and hard dimers (
HD
g )
,
`
.

∂
∂
+ − −
,
`
.

∂
∂
+ − ·
y
g
y
m
y
g
y
m
mZ Z
HD HS HS m
) ( ln
1 ) 1
2
(
) ( ln
1
2
σ σ
(3.52)
Chiew (1991) obtained the sitesite correlation result for hard dimers:
73
2
) 1 ( 2
2 1
) (
y
y
g
HD
−
+
· σ (3.53)
Chang and Sandler (1994) proposed TPTD1 and TPTD2. The form of TPTD1 can be
expressed as
,
`
.

+ −
+
+ − −
,
`
.

− −
−
+ −
,
`
.

−
− + +
·
) 2 1 )( 1 (
) 2 ( 2
1 ) 1
2
(
) 2 )( 1 (
) 2 5 (
1
2 ) 1 (
1
3
3 2
y y
y y m
y y
y y m
y
y y y
m Z
m
(3.54)
TPTD2 can be represented as
,
`
.

+ − −
− −
+ −
−
,
`
.

− −
−
+ −
,
`
.

−
− + +
·
) 414 . 0 534 . 0 )( 2 )( 1 (
) 414 . 0 24 . 0 498 . 3 (
1 ) 1
2
(
) 2 )( 1 (
) 2 5 (
1
2 ) 1 (
1
2
3
3 2
y y y
y y y m
y y
y y m
y
y y y
m Z
m
(3.55)
Starting from the points of Ghonasgi and Chapman (1994) and Chang and Sandler (1994)
above, Sadus (1995) proposed that, in general,
) ( c y g g
HS HD
+ · α (3.56)
74
where a and c are the constants for a straight line and the values can be obtained by fitting
the molecular simulation data for
HS
g and
HD
g , and obtained a new equation of state, called
the simplified thermodynamic perturbed theorydimer (STPTD) equation of state. The
general form of the STPTD equation of state for pure hard sphere chains is
) ( 2
) 2 ( ln
) 1 ( ) 1 ( 1
c y
y m
y
g
y m Z m Z
HS HS m
+
−
+
∂
∂
− + − + ·
α
α
(3.57)
Sadus (1995) applied the STPTD equation to the prediction of both the compressibility
factors of 4, 8, 16, 51 and 201mer hard chains and the second virial coefficients of up to
128mer chains. Comparison with molecular simulation data indicated that the STPTD
equation generally predicts both the compressibility factors and the second virial coefficient
more accurately than other equations of state (Chiew equation, GFD, TPTD1, TPTD2).
By using some elements of the one fluid theory, Sadus (1996) extended the STPTD
equation to hard sphere chain mixtures with no additional equation of state parameters
required. The compressibility factor predicted by the STPTD equation of state was
compared with molecular simulation data for several hard sphere chain mixtures containing
components with either identical or dissimilar hard sphere segments. Good agreement with
75
simulation data was obtained when the ratio of hardsphere segment diameters for the
component chains is less than 2.
Recently, Sadus (1999a) proposed an empirical simplified thermodynamic perturbed theory
dimer (ESTPTD) equation of state, which is a simplification of STPTD equation. In the
equation of state an empirical relationship (Shah et al., 1994) for Z
HS
is used,
) 1 (
) ( 1
0
1 0
y k
y k k
Z
HS
−
+ +
· (3.58)
where 2864 . 1
0
· k and 8225 . 2
1
· k are empirically evaluated constants (Shah et al., 1994).
The addition of Eq. (3.58) dimer properties into Eq. (3.57), the resulting ESTPT equation
(Sadus, 1999a) is
c y
m y
y k k k
y k
y k
y k
m
y k
y k y k k
m Z
m
+
−
+
,
`
.

− +
−
−
− +
,
`
.

−
− +
+ ·
α
α
2
) 2 (
1
2
) 1 (
) 1 (
) (
1
2
0 1 0
2
0
0
0
2
0
2 2
0 1 0
(3.59)
Sadus (1999b) extended the hard sphere chain equation to hard convex body chain equation
of state (HCBC). The compressibility factor of a hardconvexchain of m segments is
obtained as
76
,
`
.

∂
∂
+ − − ·
y
g
y m mZ Z
HCB
HCB HCBC
) ( ln
1 ) 1 (
σ
(3.60)
where ) (σ
HCB
g is the hard convex body sitesite correlation function at contact, the symbol
σ in Eq. (3.60) represents the breadth of the HCB. The compressibility factor of a hard
convex body can be accurately determined from the HCB equation. For the HCB equation,
the sitesite correlation function is
,
`
.

−
−
−
+ · 1
) 1 (
2 / 1
1
3
y
y
S g
R HCB
(3.61)
where
R
S is the ratio of the actual surface area of the HCB
HCB
S to the surface area of hard
spheres ) (equiv S
HS
occupying a diameter equivalent to the HCB diameter:
) (equiv S
S
S
HS
HCB
R
· (3.62)
Sadus (1999b) discussed the details on how to obtain
R
S , and gave the general form of
packing fraction for pure component,
77
,
`
.

− − − + − − −
−
+ − −
−
− + − + − +
·
3 2
2
3
3 2 2 2
) 1 ( ) 3 3 ( ) 5 . 2 3 ( 1 )( 1 (
5 . 2
1 ) 1 (
) 1 (
) 1 3 3 ( ) 2 3 ( 1 (
y S y S y S y
y S y S
m
y
y y y m
Z
R R R
R R
HCBC
α α α α
(3.63)
Sadus (1999b) examined the effect of molecular shape on the compressibility factor and the
second virial coefficient for chain molecule consisting of spheres, prolate spherocylinders,
oblate spherocylinders and doublecones, and found that these properties are very sensitive to
segment shape.
3.2 Mixing rules
Most models of the fluid state such as conformal solution theory, perturbation models and
equations of state are initially developed for pure substances. Therefore, directly extending
an existing pure fluid model to multicomponent equilibria is an important task (Sadus,
1992a). This is most commonly achieved by mixing rules and combining rules which relate
the properties of the pure components to the properties of the mixtures. The discussion will
be limited to the extension of parameters a and b. These two parameters have a real physical
significance and are common to many realistic equations of state.
The simplest possible mixing rule is a linear average of the equation of state parameters:
78
∑
·
i
i i
a x a (3.64)
∑
·
i
i i
b x b (3.65)
Eq. (3.64) is sometimes employed (Han et al., 1988) because of its simplicity, but Eq. (3.64)
is rarely used because it does not account for the important role of unlike interactions in
binary fluids. Consequently, employing both Eq. (3.64) and Eq. (3.65) would lead to the poor
agreement of theory with experiment.
The van der Waals onefluid prescriptions are the most widely used mixing rules:
∑∑
·
i j
ij j i
a x x a (3.66)
∑∑
·
i j
ij j i
b x x b (3.67)
where a
ii
and b
ii
are the equation of state parameters for pure components and cross
parameters a
ij
and b
ij
(i ? j ) are determined by an appropriate combining rule.
79
Eq. (3.66) and Eq. (3.67) are based on the implicit assumption that the radial distribution
function of the component molecules is identical, and they both explicitly contain a
contribution from interactions between dissimilar molecules. A comparison ( Harismiadis et
al., 1991) with computer simulation has concluded that the van der Waals mixing rules are
reliable for mixtures exhibiting up to an eightfold difference in the size of the component
molecules. The performance of the van der Waals mixing rules has also been thoroughly
tested for several equations of state by Han et al. (1988). They used the van der Waals
mixing rule to obtain parameter a and a linear mixing rule to obtain parameter b. The results
showed that most of equations of state with the van der Waals mixing rules were capable of
representing vapourliquid equilibria with only one binary adjustable parameter for obtaining
a
ij
. Eq. (3.65) and Eq. (3.66) were adequate for mixtures of nonpolar and slightly polar
compounds (Peng and Robinson, 1976; Han et al., 1988). Voros and Tassios (1993)
compared six mixing rules (the one and two parameters van der Waals mixing rules; the
pressure and densitydependent mixing rules; two mixing rules based on excess Gibbs
energy models: MHV2 and WongSandler) and concluded that the van der Waals mixing
rules give the best results for nonpolar systems. For the systems which contained strongly
polar substances such as alcohol, water and acetone, the van der Waals mixing rule did not
yield reasonable vapourliquid equilibrium results. Anderko (1990) gave some examples of
the failure of the van der Waals mixing rules for strongly nonideal mixtures. There is some
evidence (Deiters, 1987) that the following mixing rules are more approprate at high
pressure:
80
∑∑
·
i j
ij ij j i
x x
γ
γ
σ ε σ ε
0 0
(3.68)
∑∑
·
i j
ij j i
x x
γ γ
σ σ
0
(3.69)
where a and b are proportional to
3
0 0
σ ε and
3
0
σ , respectively.
0
ε and
0
σ are the
characteristic attractive energy and collision diameter. The parameter γ is obtained from
comparison with experimental data, and is typically assigned a value of either 2 or 2.4. Some
attempts have been made to estimate it from theoretical considerations. In terms of the
conformal parameters (recall
3
ij ij
g h · ) the above relationships can be represented as:
∑∑
·
i j
ij ij j i
g f x x fg
γ
γ
(3.70)
∑∑
·
i j
ij j i
g x x g
γ γ
(3.71)
Sadus (1989) used conformal solution theory to derive an alternative to the conventional
procedure for obtaining parameter a of equation of state. Instead of proposing an average of
pure component parameter data, the parameter a for the mixture is calculated directly.
81
Consequently, a is a function of composition only via the conformal parameters and the
contribution from the combinatorial entropy of mixing. The parameter a is obtained by
taking the positive root of the following qua dratic equation:
0 )] /( )]} 1 ( /[ 1 ) / ( / {
} ) / ( ) / {( [ ) ( )}] 1 ( /{ 1 ) / ( / [ 2
) ) / ( / / ( 2 )] / / ( [
)] ) / ( / 2 / 2 ) / ( / 2 [
' ' 2 ' ' ' 2
2 ' 2 ' 2 2 2 ' ' ' 6
2 ' ' ' ' ' 3 ' ' ' ' 2
2 ' ' ' ' ' 2 ' ' ' 6 2
· − − − − +
− − − + + − +
− + + − − +
+ − − + −
θ φ φ θ φ
φ θ
θ φ θ φ
θ
h h x x h h h h
h h f f RTV x x h h h h
h h fh h f h h f f h h aRTV
h h fh h f h h f f f f a
B A B
A
A B
(3.72)
where superscripts
'
and
' '
denote successive differentiation of the conformal parameters,
and θ and φ are characteristic of the equation of state. The main advantage of Eq. (3.72) is
that the parameter a can be calculated directly from the critical properties of pure
components without using combining rules for the contribution of unlike interactions. Sadus
(1992a, b) has applied the above equation to the calculation of the vapourliquid critical
properties of a wide range of binary mixtures. The agreement was generally very good in
view of the fact that no adjustable parameters were used to arbitrarily optimise the agreement
between theory and experiment.
Huron and Vidal (1979) have developed mixing rules from the excess thermodynamic
properties of mixing. For example, in the case of the van der Waals equation and assuming a
simple model of the energy of the fluid, the following relationship (Mollerup, 1986) is
obtained:
82
∑ ∑
− − + − ·
i
i i i
E
i i i i
b y y b x RTV V G y b a x V a }] ) 1 /( ) 1 ( ln{ [ ) / ( (3.73)
where b V y / · ,
i i i
b V y / · and G
E
is the excess Gibbs function. The above approach is
mainly limited to low pressure equilibria but it may be useful for predicting critical solution
temperature phenomena. Another approach, which is discussed elsewhere (Hoheisel and
Kohler, 1984), is the concept of local composition.
Many of the above density dependent prescriptions are undoubtedly more accurate than the
van der Waals model at high densities. But for high pressure phenomena, there is insufficient
data to make an assessment. Some computer simulation of the high pressure behaviour of
binary fluids would be useful in this instance. It is clear that the increased complexity is a
disadvantage for high pressure calculations, particularly if reliable theoretical estimates of
parameters such as γ and g
ij
(Eq. 3.71) cannot be obtained and empirical, methods must be
adopted. High pressure phenomena also occur over a range of densities. The van der Waals
prescriptions are certainly appropriate for vapourliquid phenomena but other rules may be
beneficial for dense liquidliquid critical phenomena. The analysis of type III phe nomena,
where there is a transition between vapour liquid and liquid like densities, could also benefit.
However it is difficult to isolate the effect of mixing rules on the phase behaviour of fluids
because most equation of state calculations rely on additional combining rule parameters for
their accuracy. It appears unlikely that any of the above prescription could eliminate this
83
feature. Thus, the overall effect may be to increase the complexity of the calculations without
necessarily improving the accuracy and genuine predictive capability.
A different approach has been adopted by Boublik (1970) and Mansoori et al. (1971). They
proposed a direct extension of the CarnahanStarling hard sphere term,
3 3 2 3 2 2 3
) 1 /( ] ) / ( } 1 ) / 3 ( ) / 3 {( } 2 / 3 {( 1 [ y y F E y F DE F E y F DE Z − − + − + − + · (3.74)
where
V F N y
A
6 / π · (3.75)
∑
·
i
i i
x F
3
σ (3.76)
∑
·
i
i i
x E
2
σ (3.77)
and
∑
·
i
i i
x D σ (3.78)
84
At low densities, the compressibility factor predicted by Eq. (3.71) is identical to the
CarnahanStarling equation with the onefluid model. However, at moderate and high
densities, there is a significant deviation, particularly if molecules of dissimilar size are
involved. Therefore, the analysis of vapourliquid properties is unlikely to be affected, but in
common with other density dependent mixing rules, it may be beneficial for some dense
fluid critical phenomena. Dimitrelis and Prausnitz (1986) have concluded that the equation is
particularly advantageous for the prediction of vapour liquid equilibria if the molecular size
ratio
3
2 1
) / ( σ σ exceeds 2.
The temperature dependence of the equation of state parameters is typically obtained by
scaling the values obtained at the vapour liquid critical point to lower temperatures and
pressures. The critical values are adequate for high pressure equilibria. The exception is
liquidliquid phenomena for which the vapour liquid values are intuitively inadequate. The
following equation (Christoforakos and Franck, 1986) is acceptable:
3 . 0
2 1 1 2
) / )( ( ) ( T T T b T b · (3.79)
It should be noted that the temperature dependence of the parameter a is normally reflected
in a lower than normal ? value (see Eq. 3.84) and that the agreement with experiment is often
very good without formally introducing any temperature dependence in the b term.
85
Wong and Sandler (1992) used the Helmholtz function to develop mixing rules to satisfy the
second virial condition. For the mixture parameters of an equation of state, a and b are
∑
∞
+ ·
i
E
i
i
i
C
A
b
a
x b a ) ( (3.80)
∑
∑∑
− +
−
·
∞
i i
i
i
E
i j
ij j i
RT b
a
x
RT
A
RT
a
b x x
b
1
) (
(3.81)
where C is a constant dependent on the equation of state selected and
E
A
∞
is the excess
Helmholtz function at infinite pressure, and
)] ( ) [(
2
) 1 (
) (
RT
a
b
RT
a
b
k
RT
a
b
j
j
i
i
ij
ij
− + −
−
· − (3.82)
where
ij
k is a binary interaction parameter.
Wong and Sandler (1992) tested the Eq. (3.80) and Eq. (3.81), and reported that they were
accurate in describing both simple and complex phase behaviour of binary systems for
86
diverse systems. Wong et al. (1992) demonstrated that the WongSandler mixing rules can
be used for highly nonideal mixtures. Huang and Sandler (1993) compared the Wong
Sandler mixing rules for nine binary systems. They demonstrated that the mixing rules can
be used to make high pressure vapourliquid equilibrium predictions from low pressure data.
Orbey and Sandler (1994) used the WongSandler mixing rule to correlate the vapourliquid
equilibria of various polymer + solvent and solvent + long chain hydrocarbon mixtures. They
concluded that the WongSandler mixing rule can correlate the solvent partial pressure in
concentrated polymer solutions with high accuracy over a range of temperatures and
pressures with temperatureindependent parameters.
To go smoothly from activity coefficientlike behaviour to the van der Waals one fluid
mixing rule, Orbey and Sandler (1995) slightly reformulated the WongSandler mixing rules
by rewriting the cross second virial term given in Eq. (3.82) as
RT
k a a b b
RT
a
b
ij j i
j i
ij
) 1 (
2
) (
−
−
+
· − (3.83)
Orbey and Sandler (1995) tested five binary systems, and showed that this new mixing rule
was capable of both correlating and predicting the vapourliquid equilibrium of various
complex binary mixtures accurately over wide ranges of temperature and pressure. Castier
and Sandler (1997a, b) performed critical point calculations in binary systems utilizing cubic
equations of state combined with the WongSandler mixing rules, and investigated the
87
influence of the mixing rules on the shape of the calculated critical phase diagrams. The
results showed that the combination can obtain many different types of critical phase
diagrams. Rudolph et al. (2000) used the PengRobinson equation of state with the Wong
Sandler mixing rule to investigate binary mixtures of water plus surfactant systems
containing C
4
E
1
, C
6
E
3
, C
7
E
3
and C
7
E
5
. They found that the model fitted the experimentally
found type of phase behaviour quite correctly. Comparison and evaluation for various
mixing rules can be found in the works of Knudsen et al. (1996), Michelsen and Heidemann
(1996), Wang et al (1996), Orbey and Sandler (1996), Twu et al. (1998), Wei and Sadus
(1999b) and Saghafi and Moshfeghian (2000).
3.3 Combining Rules
Any realistic mixture prescriptions will invariably contain contributions from interactions
between unlike molecules. This means that the cross term
ij
a and
ij
b ( j i ≠ ) must be
evaluated. The most widely used combining rule for the
ij
a term was first proposed by van
der Waals:
jj ii ij ij
a a a ξ · (3.84)
Alternatively, the following combining rule for critical calculations can be used,
88
jj ii
jj ii
ij ij ij
b b
a a
b a ξ · (3.85)
The
ij
ξ term (also commonly defined as 1k
ij
) is frequently introduced into combining rules
to optimise the agreement between theory and experiment. The
ij
ξ parameter is frequently
interpreted as reflecting the strength of intermolecular interaction. This interpretation is
supported by the fact that values of
ij
ξ obtained from the analysis of the critical properties of
many binary mixtures consistently decline with increasing size difference between the
component molecules. Mixtures of molecules of similar size, but different molecular
interactions, are associated with different interaction parameters. Nonetheless, the
requirement for ?
ij
values to optimise agreement between theory and experiment is also
testimony to the inadequacy of the combining rule, mixture prescriptions and the equation of
state. It has been frequently documented (Mainwaring et al. 1988b; Elliot and Daubert, 1987)
that different ?
ij
values are required to obtain optimal agreement between theory and
experiment for vapourliquid and liquidliquid critical equilibria, respectively.
The most widely used combining rules for
ij
b are the simple arithmetic rules:
2
jj ii
ij ij
b b
b
+
· ζ (3.86)
89
and the Lorentz combining rule (Hicks and Young, 1975; Sadus, 1992a, 1994):
8
) (
3 3 / 1 3 / 1
jj ii
ij ij
b b
b
+
·ζ (3.87)
Both the simple arithmetic and Lorentz combining rules yield nearly identical results for
mixtures of molecules of similar size, but the discrepancy increases substantially for
mixtures of molecules of very dissimilar size. The adjustable parameter
ij
ζ is used to
optimise the agreement between theory and experiment. It is generally found that the
parameter
ij
ζ is not required for the accurate prediction of vapourliquid critical properties
(Sadus, 1994). Similarly, good agreement between theory and experiment for type II liquid
liquid equilibria can usually be obtained by solely adjusting the
ij
ξ term and setting
ij
ζ = 1.
However, obtaining an optimal
ij
ζ parameter is essential for the accurate representation of
type III behaviour. It is found (Sadus, 1993) that the arithmetic combining rule substantially
overestimates the twophase region, whereas the Lorentz rule is more accurate but it extends
the onephase region. Sadus (1993) has examined type III behaviour of nonpolar mixtures
using both arithmetic and Lorentz combining rules. Good agreement between theory and
experiment could be obtained by determining both the optimal
ij
ξ and
ij
ζ parameters.
Sadus (1993) proposed an alternative combining rule by taking a 2:1 geometric average of
the Lorentz and arithmetic rules without the parameter
ij
ζ , that is,
90
3 / 1 2 3 / 1 3 / 1 3 / 1
) ( ) )}( 2 ( 4 / 1 {
jj ii jj ii ij
b b b b b + + · (3.88)
Sadus (1993) reported that the new combining rule (Eq. 3.88) is generally more accurate that
either the simple arithmetic or Lorentz combining rules.
3.4 Crossover Equation of State
Chen et al. (1990) and Jin et al. (1993) introduced a theory that incorporates scaling laws
asymptotically close to the critical point and that are transformed into the regular classical
expansion far away from the critical point, and this lead to the crossover equations of state.
Although the original crossover equation of state gives an accurate representation of the
thermodynamic properties of fluids in a wide region around the critical point of pure
components, it cannot be extrapolated to low densities (Kiselev, 1998).
Fox (1983) proposed an approach to the incorporation of scaling laws in classical equations
of state. The idea of the approach consists in the renormalization of the temperature and the
density in the equations of state. The result is that the temperature and the density become
nonanalytic scaling functions of the dimensionless distance to the critical point. Kiselev
(1998) developed a cubic crossover equation of state based on the nonclassical critical
phenomena for pure fluids which incorporates the scaling laws asymptotically close to the
91
critical point and which is transformed into the original cubic equation of state far away from
the critical point.
3.4.1 Classical Expression for the Critical part
In the critical region, the van der Waals equation of state corresponds to the meanfield, or
Landau theory of critical phenomena (Laudau and Lifshitz, 1980; Patashinskii and
Pokrovskii, 1979). The main assumption of the classical theory of critical phenomena is that
the critical part of the Helmhotz function, A ∆ of the system can be represented by a Taylor
expansion in the powers of the order parameter (Kiselev, 1998)
∑∑
∆ · ∆ ∆
i j
j i
ij
a A η τ η τ ) , ( (3.89)
where 1 / − ·
c
T T τ is the dimensionless deviation of the temperature T from the critical
temperature T
c
,
ij
a are the systemdependent coefficients, and η ∆ is an order parameter
(Landau and Lifshitz, 1980). In the critical region 1   << τ and   η ∆ <<1, therefore, there are
two main terms in Eq. (3.89)
4
04
2
12
) , ( η η τ η τ ∆ + ∆ · ∆ ∆ a a A (3.90)
that correspond to the critical phenomena (Landau and Lifshitz, 1980). Eq. (3.90) is valid
only in the temperature region 1   << << τ Gi (Gi is the Ginzburg number, Landau and
Lifshitz, 1980) where the longscale fluctuations in the order parameter are negligible
92
(Laudau and Lifshitz, 1980; Patashinskii and Pokrovskii, 1979). If Gi <<   τ , the singular
part of the thermodynamic potential of a system becomes a nonanalytical scaling function of
the τ and η ∆ .
A theoretical approach for constructing a crossover expression for the thermodynamic
potential of a system in the critical region based on the renormalizationgroup calculations of
Nicoll and Bhattachafjee (1981) and Nicoll and Albright (1986) has been developed by Chen
et al. (1990). Using this approach, the critical fluctuations close to the critical point result is:
) ( ) , (
2
2
4
04
2
2
2 2
12
1 1 1
τ η η τ η τ
β γ β γ α
K Y a Y Y a A − ∆ + ∆ · ∆ ∆
∆
−
∆
−
∆
−
(3.91)
where, β α, and γ are critical exponents, ) (
2
τ K is the kernel term, which provides the
correct scaling behaviour of the isochoric specific heat asymptotically close to the critical
point (Kiselev, 1998),
) 1 (
2
1
) (
1
2
20
2
− ·
∆
−
α
τ τ Y a K (3.92)
where, Y is a crossover function to be determined from the set of coupled algebraic equations
by Jin et al. (1993).
Eq. (3.91) is equivalent to Eq. (3.90) with the replacement of the dimensionless temperature
τ and the order parameter η ∆ by the renormalized values (Kiselev, 1998):
93
,
1
2
~
∆
−
·
α
τ τ Y
1
4
2
~
∆
−
∆ · ∆
β γ
η η Y (3.93)
and adding the kernel term ) (
2
τ K . As one moves further from the critical point,   τ and
  η ∆ increase, more terms in Eq. (3.89) and Eq. (3.90) may be considered.
3.4.2 Crossover Expression
It is wellknown the Helmholtz energy per mole (Sandler, 1999) that can be described as
∫
+ − ·
ideal
A pdV V T A ) , ( (3.94)
where
ideal
A corresponds to the temperature dependent function of the idealgas part of the
Helmholtz energy.
For a classical equation of state ) , ( T V p , can be written in the dimensionless form
RT
PV
p
c
·
~
(3.95)
The dimensionless Helmholtz energy can be represented in the form
94
) ( ) ( ) , (
1 ) , (
) , ( 0
~
0
~ ~
~
~
T u T P V V T A
RT
A
dV
V
p RT
RT RT
V T A
V T A
ideal
c
+ ∆ − ∆ ∆ ∆ · + − · ·
∫
(3.96)
where ) , (
~
V T A ∆ ∆ ∆ corresponds to the critical part of the Helmholtz function, 1 − · ∆
r
T T
(
c r
T T T / · ) is the dimensionless deviation of the temperature from the critical temperature
Tc, and 1 / − · ∆
c
V V V is also the dimensionless order parameter, RT V T V p T p
c c
/ ) , ( ) (
0
~
·
is the dimensionless pressure at the critical isochore
c
V V · and 0
~
u is relation to employed
equation of state.
T ∆ and V ∆ are needed to renormalized in the critical part of the classical Helmholtz
function ) , (
~
V T A ∆ ∆ ∆ according to Eq. (3.95) by Kiselev (1998). The critical temperature
and pressure of the equations of state can be taken from available measurements or
predictions, while the critical density is usually found as fitting parameter of the model,
chosen to the best description of the vapourliquid equilibrium surface far away from the
critical point. Since any classical equation of state does not reproduce the thermodynamic
surface of a fluid in the critical region, the critical density found by this method does not
coincide with the real critical density of the system (Kiselev, 1998). The additional terms to
take into account the difference between the classical critical temperature T
c
and critical
volume V
c
, and the real critical parameters T
Rc
and V
Rc
were introduced in Eq. (3.94) by
Kiselev (1998). The new forms of dimensionless temperature
~
τ and the order parameter
η ∆
95
1 1
3
) 2 ( 2
2
~
) 1 (
∆
−
∆
−
∆ + + ·
α α
τ τ τ τ Y Y
c
(3.97)
1 1
2
2
4
2
~
) 1 (
∆
−
∆
−
∆ ∆ + + ∆ · ∆
α β γ
η η η η Y Y
c
(3.98)
where the order parameter 1 / − · ∆
c
V V η , while
Rc c c
T T / ∆ · ∆τ and
Rc c c
V V / ∆ · ∆η are
the dimensionless shifts of the critical temperature and the critical volume, respectively.
Parameters T
Rc
and V
Rc
are the real critical temperature and volume of fluid. The crossover
function Y in Eqs. (3.97) and (3.98) can be written in the parametric form as
1
2
2
]
1
) 1 (
[ ) (
∆
+ +
+
·
q q
q q
q Y (3.99)
and the parametric variable q can be found from the solution of the equation
) ( ) (
1
2 1
2
2
2 1 2 2
q Y
Gi
d d
b
Gi
q
LM
∆
−
+ + ∆
+ ·
β
β
τ τ η τ
(3.100)
359 . 1 ) 2 1 ( / ) 2 (
2
· − − · β γ β γ
LM
b is the linearmodel parameter (Kiselev and Sengers,
1993). The terms τ ∝ and
2
τ ∝ correspond to a projection of the rectilinear diameter of the
coexistence curve in the temperaturedensity variables ) 1 ( 2 / ) ( τ ρ ρ ρ ρ d
c L V d
+ · + · on
the temperaturevolume projection ) 1 ( / 1
2
2 1
τ τ ρ d d V V
c d d
+ + ≅ · .
96
To complete the transformation of the classical Helmholtz function into the crossover form,
we must replace the classical dimensionless temperature T ∆ and the volume V ∆ in Eq.
(3.96) with the renormalized values
~
τ and
~
η ∆ , and the kernel term ). (
2
τ K The crossover
expression for the Helmholtz function is
) ( ) ( ) ( ) , ( ) , (
2
0
~
0
~ ~ ~ ~ ~
τ η τ K T u T p V A V T A − + ∆ − ∆ ∆ · (3.101)
3.4.3 Crossover Equation of State
A crossover equation of state incorporates the scaling laws asymptotically close to the
critical point and is transformed into the original equation of state far away from the critical
point. The general form of the crossover equation of state obtained from the crossover
expression in Eq. (3.102):
]
]
]
]
]
,
`
.

∆ ∂
∂
+ +
,
`
.

∆ ∂
∆ ∂
− ·
T
Rc
c
T
Rc
c
c
K
V
V
T p
A
V
V
V
RT
p
η
τ
η
) (
) (
2
0
~
~
(3.102)
For the PatelTeja crossover equation of state (Kiselev, 1998), the value of
0
~
p is:
3 2 1
0
~
) (
1
) (
b b Z
T
T
T
b
T p
c
a
c
α Ω
− · (3.103)
97
and the critical part
~
A ∆ is given by
3 2
~ ~
1
~
3
~
2
~
~
1
~
~ ~ ~
) (
1 /
1 /
ln
) (
1 ln ) , (
b b Z T
T
b
b
b
T
T
b
A
c
a
c
a
c
η τ α
η
η
η
τ α
η
η τ
∆ Ω
−
∆
+
,
`
.

+ ∆
+ ∆
Ω
Ω
+
,
`
.

+
∆
− · ∆ ∆ (3.104)
where
c b a
and Ω Ω Ω , , are functions of the critical compressibility factor
c
Z (Eq. (3.25) 
Eq. (3.27)), and Ω
2 2
6
c c b b
Ω + Ω Ω + Ω · Ω (3.105)
The definition of ) (T α can be found in Eq. (3.28) and Eq. (3.29). Parameters b
1
, b
2
and b
3
in
Eq. (3.104) and Eq. (3.105) are
c
c b
c
c b
c
b
Z
b
Z
b
Z
b
2
1
2
1
1
3
2
1
Ω + Ω + Ω
+ ·
Ω − Ω + Ω
+ ·
Ω
− ·
(3.106)
Remember that 0
~
u in Eq. (3.101) with PatelTeja equation of state is
98
ideal
a c
A
b
b T
T
T
b T u
~
3
2
1
0
~
) ln(
) (
ln ) ( +
Ω
Ω
+ − ·
α
(3.107)
and
RT
A
A
ideal
ideal ·
~
(3.108)
Kiselev (1998) used the modified PatelTeja cubic equation of state (Patel and Teja, 1982) to
compare with experimental data for pure CO
2
, water, and refrigerants R32 and R125 in the
one and twophase regions. He showed that the crossover PatelTeja equation of state yields
a much better representation of the thermodynamic properties of pure fluids, especially in the
critical region and for vapourliquid equilibrium, than the original PatelTeja equation of
state.
Kiselev and Ely (1999) also developed a crossover modification of the statistical associating
fluid theory equation of state for macromolecular chain fluids. The crossover statistical
associating fluid theory equation of state has the same form as the cubic one of Eq. (3.103),
but ) (
0
~
T p and
~
A ∆ have the different forms because of different classical equation of state,
) , ( ) (
0
~
c
c
V T p
RT
V
T p · (3.109)
and
) ( ) 1 ln( ) 0 , ( ) , ( ) , (
~
0
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
τ η η τ η τ η τ p A A A r r ∆ + + ∆ − − ∆ · ∆ ∆ (3.110)
where r A
~
is the dimensionless residual part of the Helmholtz energy.
99
Kiselev and Ely (1999) compared the crossover statistical associating fluid theory and
original equation of state with experimental singlephase data for ntriacontane and n
tetracontane. The results show that the crossover statistical associating fluid theory equation
of state reproduces the saturated pressure data in the entire temperature range from the triple
point to the critical temperature with an average absolute deviation of about 3.8%, the
saturated liquid densities with an average absolute deviation of about 1.5%, the saturated
vapour densities with an average absolute deviation of about 3.4%, and gives a much better
representation of the experimental values of pressure and the liquid density in the critical
region.
Wyczalkowska et al. (1999) also developed a global crossover equation of state to show how
the density fluctuations affect all thermodynamic properties more and more significantly
when the fluid approaches the critical point.
References
Abdulkadirova, K. S., Wyczalkowska, A. K., Anisimov, M. A. and Sengers, J. V. (2002).
Thermodynamic Properties of Mixtures of H
2
O and D
2
O in the Critical Region, J. Chem.
Phys., 116, 45974610.
Adachi, Y., Lu, B. C. Y. and Sugie, H. (1983). ThreeParameter Equation of State, Fluid
Phase Equilib., 13, 133142.
Anderko, A. (1990). EquationofState Methods for the Modelling of Phase Equilibria, Fluid
Phase Equilib., 61, 145225.
100
Baker, L. E. and Luks, K. D. (1980). Critical Point and Saturation Pressure Calculations for
Multipoint Systems, Soc. Pet. Eng. J., 20, 1524.
Beret S. and Prausnitz, J. M. (1975). Perturbed HardChain Theory: An Equation of State for
Fluids Containing Small or Large Molecules, AIChE J., 26, 11231132.
Boublik, T. (1970). HardSphere Equation of State, J. Chem. Phys., 53, 471473.
Boublik, T. (1981). Statistical Thermodynamics of Nonspherical Molecule Fluids. Ber.
Bunsenges, Phys. Chem. 85, 10381041.
Calado, J. C. G., Deiters, U. and Streett, W. B. (1981). LiquidVapour Equilibrium in the
Krypton + Methane System, J. Chem. Soc. Farad I., 77, 25032513.
Carnahan, N. F. and Starling, K. E. (1969). Equation of State for Nonattracting Rigid Sheres,
J. Chem. Phys., 51, 635636.
Castier, M. and Sandler, S. I. (1997a). Critical Points with the WongSandler Mixing Rule. I.
Calculations with the van der Waals Equation of State, Chem. Eng. Sci., 52, 33933399.
Castier, M. and Sandler, S. I. (1997b). Critical Points with the WongSandler Mixing Rule.
II. Calculations with a Modified PengRobinson Equation of State, Chem. Eng. Sci., 52,
35793588.
Chang, J. and Sandler, S. I. (1994). An Equation of State for the HardSphere Chain Fluid:
Theory and Monte Carlo Simulation, Chem. Eng. Sci., 49, 27772791.
Chapman, W. G., Jackson, G. and Gubbins, K. E. (1988). Phase Equilibria of Associating
Fluid: Chain Molecules with Multiple Bonding Sites, Mol. Phys., 65, 10571079.
Chapman, W. G., Gubbins, Gubbins, K. E., Jackson, G. and Radosz, M. (1990). New
Reference Equation of State for Associating Liquids, Ind. Eng. Chem. Res., 29, 1709
1721.
101
Chen, Z. Y., Albright, P. C. and Sengers, J. V. (1990). Crossover from Singular Critical to
Regular Classical Thermodynamic Behavior of Fluids, Phys. Rev. A 41, 31613177.
Chiew, Y. C. (1990). PercusYevick IntegralEquation Theory for Athermal HardSphere
Chains: I. Equations of State, Mol. Phys., 70, 129143.
Chiew, Y. C. (1991). PercusYevick IntegralEquation Theory for Athermal HardSphere
Chains: II. Average Intermolecular Correlation Functions, Mol. Phys., 73, 359373.
Christoforakos, M. and Franck, E. U. (1986). An Equation of State for Binary Fluid Mixtures
to High Temperatures and High Pressures. Ber. Bunsenges. Phys. Chem., 90,780789.
Christou, G., Morrow, T. Sadus, R. J. and Young, C. L. (1986). Phase Behaviour of
Fluorocarbon + Hydrocarbon Mixtures: Interpretation of Type II and Type III Behaviour
in Terms of a ‘Hard Sphere + Attractive Term’ Equation of State, Fluid Phase Equilib.,
25, 263272.
Christou, G., Young, C. L. and Svejda, P. (1991). GasLiquid Critical Temperatures of
Binary Mixtures of Polar Compounds + Hydrocarbons or + Other Polar Compounds.
Ber. Bunsenges. Phys. Chem., 95, 510514.
Chueh, P. L. and Prausnitz, J. M. (1967a). VapourLiquid Equilibria at High Pressures.
VapourPhase Fugacity Coefficients in Nonpolar and QuantumGas Mixtures, Ind. Eng.
Chem. Fundam., 6, 492498.
Chueh, P. L. and Prausnitz, J. M. (1967b). Vapour Liquid Equilibria at High Pressures.
Calculation of Partial Molar Volumes in Nonpolar Liquid Mixtures, AIChE J., 13, 1099
1107.
Deiters, U. K. and Schneider, G. M. (1976). Fluid Mixtures at High Pressures. Computer
Calculations of the Phase Equilibrium and the Critical Phenomena in Fluid Binary
102
Mixtures from the RedlichKwong Equation of State, Ber. Bunsenges. Phys. Chem, 80,
13161321.
Deiters, U. (1982). A New Semiempirical Equation of State for Fluid: III. Application to
Phase Equilibria in Binary Mixtures, Chem. Eng. Sci., 37, 855861.
Deiters, U. K. and Swaid, I. (1984). Calculation of FluidFluid and SolidFluid Phase
Equilibria in Binary Mixtures at High Pressures, Ber. Bunsenges. Phys. Chem., 88, 791
796.
Deiters, U. K. (1987). DensityDependent Mixing Rules for the Calculation of Fluid Phase
Equilibria at High Pressures, Fluid Phase Equilib., 33, 267293.
Deiters, U. K. and Pegg, I. L., (1989). Systematic Investigation of the Phase Behaviour in
Binary Fluid Mixtures. I. Calculations Based on the RedlichKwong Equation of State, J.
Chem. Phys. 90, 66326641.
Dickman, R. and Hall, C. K. (1986). Equation of State for Chain Molecules: Continuous
Space Analog of Flory Theory, J. Chem. Phys., 85, 41084115.
Dimitrelis, D. and Prausnitz, J. M. (1986). Comparison of two Hard Sphere Reference
Systems for Perturbation Theories for Mixtures, Fluid Phase Equilib., 31, 121.
Donohue, M. D. and Prausnitz J. M. (1978). Perturbed Hard Chain Theory for Fluid
Mixtures: Thermodynamic Properties for Mixtures in Natural Gas and Petroleum
Technology, AIChE J., 24, 849860.
Economou, I. G. and Donohue, M. D. (1996). Equations of State for Hydrogen Bonding
Systems, Fluid Phase Equilib., 116, 518529.
Elliott, J. R. and Daubert, T. E. (1987). Evaluation of an Equation of State Method for
Calculating the Critical Properties of Mixtures, Ind. Eng. Chem. Res., 26, 16861691.
103
EstelaUribe, J. F. and Trusler, J. P. M. (1998). Shape Factors for Light Hydrocarbons, Fluid
Phase Equilib., 151, 225234.
Flory, P. I. (1965). Statistical Thermodynamics of Liquid Mixtures, J. Am. Chem. Soc., 87,
18331838.
Fox, J. F. (1983). Method for Construction of Nonclassical Equation of State, Fluid Phase
Equilib., 14, 4553.
Franck, E. U. (1987). Fluid at High Pressures and Temperatures, J. Chem. Thermodyn., 19,
225244.
Ghonasgi, D. and Chapman, W. G. (1994). A New Equation of State for Hard Chain
Molecules, J. Chem. Phys., 100, 66336639.
Gubbins, K. E., Shing, K. S. and Streett, W. B. (1983). Fluid Phase Equilibria: Experiment,
Computer Simulation, and Theory, J. Phys. Chem., 87, 45734585.
Guggenheim, E. A. (1965). Variations on van der Waals’ Equation of State for High
Densities, Mol. Phys., 9, 199200.
Han, S. J., Lin, H. M. and Chao, K. C. (1988). VapourLiquid Equilibrium of Molecular
Fluid Mixtures by Equation of State, Chem. Eng. Sci., 43, 23272367.
Harismiadis, V. I., Koutras, N. K., Tassios, D. P. and Panagiotopoulos, A. Z. (1991). How
Good is Conformal Solutions Theory for Phase Equilibrium Predictions?, Fluid Phase
Equilib., 65, 118.
Harstad, K. G., Miller, R. S. and Bellan, J. (1997). Efficient HighPressure State Equations,
AIChE J., 43, 16051610.
Heilig, M. and Franck, E. U. (1989). Calculation of Thermodynamic Properties of Binary
Fluid Mixtures to High Temperatures and High Pressures, Ber. Bunsenges. Phys. Chem.,
93, 898905.
104
Heilig, M. and Franck, E. U. (1990). Phase Equilibria of Multicomponent Fluid Systems to
High Pressures and Temperatures, Ber. Bunsenges. Phys. Chem., 94, 2735.
Hicks, C. P. and Young C. L. (1975). The GasLiquid Critical Properties of Binary Mixtures,
Chem. Rev., 75, 119175.
Hicks, C. P. and Young, C. L. (1976). Critical Properties of Binary Mixtures, J. Chem. Soc.
Faraday I, 72, 122133.
Hicks, C. P., Hurle, R. L. and Young, C. L. (1977). Theoretical Prediction of Phase
Behaviour at High Temperatures and Pressures for NonPolar Mixtures: IV. Comparison
with Experimental Results for Carbon Dioxide + nAlkane mixtures, J. Chem. Soc.
Faraday II, 73, 18841888.
Hicks, C. P., Hurle, R. L., Toczylkin, L. S. and Young, C. L. (1978). Upper Critical Solution
Temperatures of Hydrocarbon + Fluorocarbon Mixtures, Aust. J. Chem., 31, 1925.
Hoheisel, H. and Kohler, F. (1984). Local Composition in Liquid Mixtures, Fluid Phase
Equilib., 16, 1324.
Huang, H. and Sandler, S. I. (1993). Prediction of VapourLiquid Equilibria at High Pressure
Using Activity Coefficient Parameters Obtained from Low Pressure Data: A
Comparison of two Equation of State Mixing Rules, Ind. Eng. Chem. Res., 32, 1498
1503.
Huber M.L. and Ely, J. F. (1994). A Predictive Extended Corresponding States Model for
Pure and Mixed Refrigerants Including an Equation of State for R134a, Int. J. Refrig.
17, 1831.
Hurle, R. L., Toczylkin, L. S. and Young, C. L. (1977a). Theoretical Prediction of Phase
Behaviour at High Temperatures and Pressures for NonPolar Mixtures: II. Comparison
105
with Upper Critical Solution Temperatures for Perfluoromethylcyclohexane +
Hydrocarbons, J. Chem. Soc. Faraday II, 73, 613617.
Hurle, R. L., Toczylkin, L. S. and Young, C. L. (1977b). Theoretical Prediction of Phase
Beha viour at High Temperatures and Pressures for NonPolar Mixtures: III. GasGas
Immiscibility, J. Chem. Soc. Faraday II, 73, 618622.
Huron, M. J. and Vidal, J. (1979). New Mixing Rules in Simple Equations of State for
Representing VapourLiquid Equilibria of Strongly NonIdeal Mixtures, Fluid Phase
Equilib., 3, 255271.
Ikonomou, G. D. and Donohue, M. D. (1986). Thermodynamics of HydrogenBonded
Molecules: The Associated Perturbed Anisotropic Chain Theory, AIChE J. 32, 1716
1725.
Jin, G. X., Tang, S. and Sengers, J. V. (1993). Global Thermodynamic Behavior of Fluid
Mixtures in the Critical Region, Phys. Rev. E 47, 388402.
Kihara, T. (1963). Convex Molecules in Gaseous and Crystalline States, Adv. Chem. Phys.,
V, 147188.
Kiselev, S. B. (1997). Prediction of the Thermodynamic Properties and the Phase Behavior
of Binary Mixtures in the Extended Critical Region, Fluid Phase Equilib., 128, 128.
Kiselev, S. B. (1998). Cubic Crossover Equation of State, Fluid Phase Equilib., 147, 723.
Kiselev S. B., Ely, J. F., Adidharma, H. and Radosz, M. (2001). A Crossover Equation of
State for Associating Fluids, Fluid Phase Equilib., 183, 5364.
Kim, C. H., Vimalchand, P., Donohue, M. D. and Sandler, S. I. (1986). Local Composition
Model for Chainlike Molecules: A New Simplified Version of the Perturbed Hard Chain
Theory, AIChE J., 32, 17261733.
106
Knudsen, K., Stenby, E. H. and Fredenslund, A. (1996). A Comprehensive Comparison of
Mixing Rules for Calculation of Phase Equilibria in Complex Systems, Fluid Phase
Equilib., 82, 361368.
Kolafa, J. and Nezbeda, I. (1994), LennardJones Fluid: An Accurate Analytic and
TheoreticallyBased Equation of State, Fluid Phase Equilib., 100, 134.
Landau, L. D. and Lifshitz, E. M. (1980). Statistical Physics, 3
rd
Ed., Pergamon, New York.
Mainwaring, D. E., Sadus, R. J. and Young, C. L. (1988a). Deiters’s Equation of State and
Critical Phenomena, Chem. Eng. Sci., 43, 459466.
Mainwaring, D. E., Sadus, R. J. and Young, C. L. (1988b). Prediction of Binary and Ternary
Critical Properties Using Deiters’ Equation, Hard Sphere and Hard Convex Body
Equations of State, Fluid Phase Equilib., 42, 85103.
Malijevsky, A. and Veverka, J. (1999). New Equations of State for Pure and Binary Hard
Sphere Fluids, Phys. Chem. Chem. Phys., 1, 42674270.
Mansoori, G. A., Carnahan, N. F., Starling, K. E. and Leland, T. W. Jr. (1971). Equilibrium
Thermodynamic Properties of the Mixture of Hard Spheres, J. Chem. Phys., 54, 1523
1525.
Martin, J. J. (1979). Cubic Equations of StateWhich?, Ind. Eng. Chem. Fundam, 18, 8197.
Mather, A. E., Sadus, R. J., and Franck, E. U. (1993). Phase Equilibria in the Water +
Krypton and Water + Neon Systems to 250 MPa and 700K, J. Chem. Thermodyn., 25,
771779.
Michelsen, M. L. and Heidemann, R. A. (1996). Some Properties of Equation of State
Mixing Rules Derived from Excess Gibbs Energy Expressions, Ind. Eng. Chem. Res.,
35, 278287.
107
Mollerup, J. (1986). A Note on the Derivation of Mixing Rules from Excess Gibbs Energy
Models, Fluid Phase Equilib., 25, 323327.
Mulero, A., Galan, C. and Cuadros, F. (2001). Equations of State for Hard Spheres. A review
of Accuracy and Applications, Phys. Chem. Chem. Phys., 3, 49914999.
Nicoll, J. F. and Bhattachafjee, J. K. (1981). Crossover Functions by RenormalizationGroup
Matching: O ( epsilon
2
) Results, Phys. Rev. B, 23, 389401.
Nicoll, J. F. and Albright, P. C. (1986). Background Fluctuations and Wegner Corrections,
Phys. Rev. B, 34, 19911996.
Orbey, N. and Sandler S. I. (1994). VapourLiquid Equilibrium of Polymer Solutions Using
a Cubic Equation of State, AIChE J., 40, 12031209.
Orbey, N. and Sandler S. I. (1995). Reformulation of WongSandler Mixing Rule for Cubic
Equations of State, AIChE J., 41, 683690.
Orbey, N. and Sandler S. I. (1996). Analysis of Excess Free Energy Based Equations of State
Models, AIChE J., 42, 23272334.
Palenchar, R. M., Erickson, D. D. and Leland, T. W. (1986). Prediction of Binary Critical
Loci by Cubic Equations of State, ACS Symp. Ser., 300, 132155.
Patashinskii, A. Z. and Pokrovskii, V. L. (1979). Fluctuation Theory of Phase Transitions,
Pergamon, New York.
Patel, N. C. and Teja, A. S. (1982). A New Cubic Equation of State for Fluids and Fluid
Mixtures, Chem. Eng. Sci., 37, 463473.
Peng, D. Y. and Robinson, D. B. (1976). A New TwoConstant Equation of State, Ind. Eng.
Chem. Fundam., 15, 5964.
108
Peng, D. Y. and Robinson, D. B. (1977). A Rigorous Method for Predicting the Critical
Properties of Multicomponent Systems from an Equation of State, AIChE J., 23, 137
144.
Polishuk I., Wisniak J. and Segura H., (1999). Prediction of the Critical Locus in Binary
Mixtures Using Equation of State. I. Cubic Equation of State, Classical Mixing Rules,
Mixtures of MethaneAlkanes, Fluid Phase Equilib., 164, 1347.
Prigogine, I. (1957). The Molecular Theory of Solutions, NorthHolland, Amsterdam.
Povodyrev, A. A., Jin, G. X., Kiselev, S. B. and Sengers, J. V. (1996). Crossover Equation of
State for the Thermodynamic Properties of Mixtures of Methane and Ethane in the
Critical Region, Int. J. Thermophysics., 17, 909944.
Redlich, O. and Kwong, J. N. S. (1949). On the Thermodynamics of Solutions: V. An
Equation of State: Fugacities of Gaseous Solutions, Chem. Rev., 44, 233244.
Reiss, N. R. , Frisch, H. L. and Lebowitz, J. L. L. (1959). Statistical Mechanics of Rigid
Spheres, J. Chem. Phys., 31, 369380.
Rudolph, E. S. J., Langeveld, J. H., de Loos, Th. W. and de Swaan Arons, J. (2000). Phase
Behaviour of Water Plus Nonionic Surfactant Systems: Experiments and Modeling
Applying the PengRobinson Equation of State with the WongSandler Mixing Rules and
the UNIQUAC gEModel, Fluid Phase Equilib., 173, 8196.
Sadus, R. J. and Young, C. L. (1985a). Phase Behaviour of Carbon Dioxide and
Hydrocarbon Mixtures, Aust. J. Chem., 38, 17391743.
Sadus, R. J. and Young, C. L. (1985b). Phase Behaviour of Binary nAlkanenitrile and n
Alkanes and Nitromethane and Alkane Mixtures, Fluid Phase Equilib., 22, 225229.
109
Sadus, R. J., Young, C. L. and Svejda, P. (1988). Application of Hard Convex Body and
Hard Sphere Equation of State to the Critical Properties of Binary Mixtures, Fluid Phase
Equilib., 39, 8999.
Sadus, R. J. (1989). Attractive Forces in Binary Mixtures: a Priori Prediction of the Gas
Liquid Critical Properties of Binary Mixtures, J. Phys. Chem., 93, 37873790.
Sadus, R. J. (1992a). High Pressure Phase Behaviour of Multicomponent Fluid Mixtures,
Elsevier, Amsterdam.
Sadus, R. J. (1992b). Novel Critical Transitions in Ternary Fluids Mixtures, J. Phys. Chem.,
96, 51975202.
Sadus, R. J. (1993). Influence of Combining Rules and Molecular Shape on the High
Pressure Phase Equilibria of Binary Fluid Mixtures, J. Phys. Chem., 97, 19851992.
Sadus, R. J. (1994) Calculating Critical Transitions of Fluid Mixtures: Theory Vs.
Experiment, AIChE J., 40, 13761403.
Sadus R. J. (1995). Equations of State for HardSphere Chains, J. Phys. Chem., 99, 12363
12366.
Sadus R. J. (1996). A Simplified Thermodynamic Perturbation TheoryDimer Equation of
State for Mixtures of HardSphere Chains, Macromol., 29, 72127216.
Sadus, R. J. (1999a). Simple Equation of State for HardSphere Chains, AIChE J., 45, 2454
2457.
Sadus, R. J. (1999b). An Equation of State for Hard Convex Body Chains, Mol. Phys., 73,
12791284.
Saghafi, A. and Moshfeghian, M. (2000). Evaluation of VaporLiquid Equilibrium of Ethane
Binary Systems Using UNIQUACBased Mixing Rules and its Extension to
Multicomponent System, Fluid Phase Equilib., 169, 3147.
110
Sandler, S. I. (1994). Models for Thermodynamic and Phase Equilibria Calculations, Marcel
Dekker, New York.
Sandler, S. I. (1999). Chemical and Engineering Thermodynamics, 3
rd
Ed., John Wiley &
Sons, New York.
Scott, R. L. (1956). Corresponding State Treatment of Nonelectrolyte Solutions, J. Chem.
Phys., 25, 193.
Scott, R. L., (1971). Eyring, H., Henderson, D. and Jost, W. (Eds.), In Physical Chemistry,
An Advanced Treatise, Vol. 8A: Liquid State, Academic Press, New York.
Scott, R. L. (1972). The Thermodynamics of Critical Phenomena in Fluid Mixtures, Ber.
Bunsenges, Phys. Chem., 76, 296308.
Semmens, J., Waterson, S. D. and Young, C. L. (1980). Upper Critical Solution
Temperatures of PerfluoronHexane + Alkane, Sulfur Hexafluoride + Siloxane and
Perfluoropropane + Siloxane Mixtures, Aust. J. Chem., 33, 19871992.
Sengers, J. V., Kayser, R. F., Peters, C. J. and White Jr, H. J. (Eds.) (2000). Equations of
State for Fluids and Fluid Mixtures, Elsevier, Amsterdam.
Shah, V. M., Bienkowski, P. R. and Cochran, H. D. (1994). Generalized Quartic Equation of
State for Pure Nonpolar Fluids, AIChE J., 40, 152159.
Smith, V. S. (1995). SolidFluid Equilibria in Natural Gas Systems, Ph.D. Thesis, Georgia
Institute of Technology.
Soave, G. (1972). Equilibrium Constants From a Modified RedlichKwong Equation of
State, Chem. Eng. Sci., 27, 11971203.
Song, Y., Lambert, S. M. and Prausnitz, J. M. (1994). LiquidLiquid Phase Diagrams for
Binary Polymer Solutions from a Pertubed HardSphereChain Equation of State, Chem.
Eng. Sci., 49, 27652775.
111
Spear, R. R., Robinson, Jr. R. L. and Chao, K. C. (1969). Critical States of Mixtures and
Equations of State, Ind. Eng. Chem. Fundam., 8, 28.
Stryjek, R. and Vera, J. H. (1986). PRSV: An Improved PengRobinson Equation of State for
Pure Compounds and Mixtures, Can, J. Chem. Eng., 64, 323340.
Svejda, P. and Kohler, F. (1983). A Generalized van der Waals Equation of State: I.
Treatment of Molecular Shape in Terms of the Boublik Nezbeda Equation, Ber.
Bunsenges. Phys. Chem., 87, 672680.
Thiele, E. (1963). Equation of State for Hard Spheres, J. Chem. Phys., 39, 474479.
Toczylkin, L. S. and Young, C. L. (1977). Comparison of the Values of Interaction
Parameter from Upper Critical Solution Temperatures of Acetone + Alkanes with Values
Obtained from GasLiquid Critical Temperatures, Aust. J. Chem., 30, 767770.
Toczylkin, L. S. and Young, C. L. (1980a). Gasliquid Critical Temperatures of Mixtures
Containing Electron Donors. Part 1. Ether Mixtures, J. Chem. Thermodyn., 12, 355364.
Toczylkin, L. S. and Young, C. L. (1980b). GasLiquid Critical Temperatures of Mixtures
Containing Electron Donors. Part 2. Amine Mixtures, J. Chem. Thermodyn., 12, 365370.
Toczylkin, L. S. and Young, C. L. (1980c). Upper Critical Solution Temperatures of
Hydrocarbon + Fluorocarbon Mixtures: Mixtures with Specific Interactions, J. Chem.
Thermodyn., 33, 465469.
Trebble, M. A. and Bishnoi, P. R. (1987). Development of a New FourParameter Cubic
Equation of State, Fluid Phase Equilib., 35, 121.
Twu, C. H., Coon, J. E. and Bluck, D. (1998). A ZeroPressure Cubic Equation of State
Mixing Rule for Predicting High Pressure Phase Equilibria Using Infinite Dilution
Activity Coefficients at Low Temperature, Fluid Phase Equilib., 150151, 181189.
112
Van Konynenburg, P. H. and Scott, R. L. (1980). Critical Lines and Phase Equilibria in
Binary van der Waals Mixtures, Philos. Trans. Roy. Soc. London A, 298, 495540.
Van Pelt, A., Peters, C. J. and de Swaan Arons, J. (1992). Application of the Simplified
PerturbedHardChain Theory for Pure Components Near the Critical Point, Fluid Phase
Equilib., 74, 6783.
Vilmalchand, P. and Donohue, M. D. (1985). Thermodynamics of Quadrupolar Molecules:
The PerturbedAnisotropicChain Theory, Ind. Eng. Chem. Fundam., 24, 246257.
Vilmalchand, P. and Donohue, M. D. (1989). Comparison of Equations of State for Chain
Molecules, J. Phys. Chem., 93, 43554360.
Voros, N. G. and Tassios, D. P. (1993). VapourLiquid Equilibria in Nonpolar/Weakly Polar
Systems with Different Types of Mixing Rules, Fluid Phase Equilib., 91, 129.
Wang, J. L., Wu, G. W. and Sadus, R. J. (2000). ClosedLoop LiquidLiquid Equilibria and
the Global Phase Behaviour of Binary Mixtures Involving HardSphere + van der Waals
Interactions, Mol. Phys., 98, 715723.
Wang, W., Qu, Y., Twu, C. H. and Coon, J. E. (1996). Comprehensive Comparison and
Evaluation of the Mixing Rules of WS, MHV2, and Twu et al., Fluid Phase Equilib.,
116, 488494.
Waterson, S. D. and Young, C. L. (1978). GasLiquid Critical Temperatures of Mixtures
Containing an OrganoSilicon Compound, Aust. J. Chem., 31, 957962.
Wei, Y. S. and Sadus, R. J. (2000). Equations of State for the Calculation of FluidPhase
Equilibria, AIChE J., 46, 169195.
Wertheim, M. S. (1987). Thermodynamic Perturbation Theory of Polymerization, J. Chem.
Phys., 87, 73237331.
113
Wong, D. S. H., Orbey, H. and Sandler, S. I. (1992). Equation of State Mixing Rule for
Nonideal Mixtures Using Available Activity Coefficent Model Parameters and that
Allows Extrapolation over Large Ranges of Temperature and Pressure, Ind. Eng. Chem.
Res., 31, 20332039.
Wong, D. S. H. and Sandler, S. I. (1992). A Theoretically Correct Mixing Rule for Cubic
Equations of State, AIChE J., 38, 671680.
Wyczalkowska, A. K., Anisimov, M. A. and Sengers, J. V. (1999). Global Crossover
Equation of State of a van der Waals Fluid, Fluid Phase Equilib., 160, 523535.
Wyczalkowska, A. K., Abdulkadirova, K. S., Anisimov, M. A. and Sengers, J. V. (2000).
Thermodynamic Properties of H
2
O and D
2
O in the Critical Region, J. Chem. Phys. 113,
49855002.
Yelash, L. V. and Kraska, T. (1998). ClosedLoops of LiquidLiquid Immiscibility in Binary
Mixtures of Equal Sized Molecules Predicted with a Simple Theoretical Equation of
State, Ber. Bunsenges, Phys. Chem., 102, 213223.
Zhao, J. and Olesik, S. V. (1999). Phase Diagram Studies of MethanolCHF3 and Methanol
H2OCHF3 Mixtures, Fluid Phase Equilib., 154, 261284.
Zheng, D. Q. , Ma, W. D., Wei, R. and Gao, T. M. (1999). Solubility Study of Methane,
Carbon Dioxide and Nitrogen in Ethylene Glycol at Elevated Temperatures and
Pressures, Fluid Phase Equilib., 155, 277286.
114
In addition to modeling small and simple molecules, considerable emphasis has been placed on modeling long and convex molecules. Based on theory of Prigogine (1957) and Flory (1965), an equation for molecules treated as chains of segments, which is called PerturbedHardChainTheory (PHCT) was constructed by Beret and Prausnitz (1975) and Donohue and Prausnitz (1978). To reduce the mathematical complexity of PerturbedHardChainTheory, Kim et al. (1986) developed a simplified version of the theory by replacing the complex attractive part by a simpler expression. At the almost same time, Vimalchand and Donohue (1985) obtained a fairly accurate multipolar mixture calculation by using the Perturbed Anisotropic Chain theory, and Ikonomou and Donohue (1986) extended the Perturbed Anisotropic Chain Theory to the Associated Perturbed Anisotropic Chain Theory by taking into account the existence of hydrogen bonding.
Wertheim
(1987)
proposed
a
thermodynamic
perturbation
theory
(TPT),
which
accommodates hardchain molecules. Chapman et al. (1988) generalized the TPT model to obtain the compressibility factor of a hardchain of segments. Ghonasagi and Chapman (1994) and Chang and Sandler (1994) modified TPT for the hardsphere chain by incorporating structural information for the diatomic fluid (TPTD). Sadus (1995) derived the simplified thermodynamic perturbation theory–dimer (STPTD) equation from TPTD. Sadus (1999b) later developed STPTD to the empirical simplified thermodynamic perturbation theorydimer (ESTPTD) equation, and tested the accuracy of the equation against simulation data for hardsphere chains containing up to 201 hardsphere segments. Sadus (1999 a) also derived an equation of state for hard convex body chains from the TPT of hard sphere chains (Wertheim, 1987; Chapman et al., 1988). 51
Jin et al. (1993), Povodyrev et al. (1996) and Kiselev (1997) developed theoretical crossover equations of state for pure fluids and binary mixtures which incorporate the scaling laws asymptotically close to the critical point and which are transformed into the regular classical expansion far away from the critical point. Kiselev (1998) used the modified PatelTeja cubic equation of state (Patel and Teja, 1982) as a starting point to propose a general procedure for transforming any classical equation of state into a crossover equation of state. Wyczalkowska et al. (1999) also developed a global crossover equation of state, and Kiselev et al. (2001) extended the crossover modification of the statisticalassociating fluid theory (Chapman et al., 1988, 1990) equation of state. Wyczalkowska et al. (2000) demonstrated that both pure H2 O and D2 O obey a universal scaled crossover equation of state in the critical region which satisfies corresponding state for the nearcritical behaviour of the thermodynamic properties. Abdulkadirova et al. (2002) extended the pure component to binary mixtures (H2 O + D2 O) using a crossover equation of state.
In this chapter we will review equations of state for high pressure with particular emphasis on critical points calculations. A crossover theory for equations of state is also discussed.
52
and third power. which when expanded have volume terms raised to the first. Usually. such as vapourliquid equilibria.3. The parameter a is a measure of the attractive forces between the molecules. The van der Waals equation was the first equation capable of representing vapour liquid coexistence. The pressure ( ) is related to the temperature (T). 2. The van der Waals equation of state (1873) is the simplest cubic equation of state for fluid phase equilibria. and b is related to the size of the molecules.1 Cubic Equations of State Cubic equations of state are equations. Tc and Vc are known and Eq. 1994).1. Most commonly encountered phase equilibrium calculations. three volumes exist for any given temperature and pressure.1 Equations of State for High Pressure Equilibria and Critical Phenomena 3. second. involve only two phases for which a cubic equation is suitable. Cubic equations have the advantage that the three values of volume can be obtained analytically without the need for an iterative solution procedure. It can be regarded as a “hard sphere term + attractive term” equation of state composed from the contribution of repulsive and attractive intermolecular interactions (Sadus.1) It has two pure component parameters a and b. ideal gas p constant (R) and molar volume (V) via: p= RT a − 2 V −b V (3. Since the van der Waals equation of state is cubic in volume. pc.1 (Chapter 2) may be simultaneously solved to provide solutions for the parameters a and b: 53 .
2.2 and Eq. the critical compressibility factor of all fluids including pure component and binary mixtures predicted by van der Waals equation is 0.24 to 0. whereas the real value for different hydrocarbons varies from 0.1. Many modifications at the van der Waals equation of state have been proposed to address this deficiency.2 are used in Eq.3) Van Konynenburg and Scott (1980) successfully demonstrated that most of the critical equilibria exhibited by binary mixtures could be qualitatively predicted by the van der Waals equation of state. For example.2) b= RTc 8 pc (3.29. 3. 54 . 3.1 at the critical point). but it is rarely sufficiently accurate for critical properties and phase equilibria calculations.a= 27 R 2 Tc2 64 pc (3.375 (Eq. Five examples of cubic equations of state based on the van der Waals equation are listed in Table 3.
0867 RTc pc (3.5 RT a(T ) − V − b V (V + b) RT a (T ) − V − b V (V + b) + b(V − b) (3. The parameters a and b are usually expressed as a = 0. (3.4) p= p= p= RT a − V − b V (V + b)T 0 .5) (3.5 pc (3.10) Spear et al.Table 3.4278 R 2Tc2.7) p= p= RT a (θ . b) also showed that the RedlichKwong equation can be adapted to predict both vapour and liquid properties.6) (3. Deiters and Schneider (1976) and Baker and Luks (1980) 55 .4) was proposed as an empirical modification of the van der Waals equation to make the attractive term temperaturedependent.T ) − V − b V (V + b) + b(V − b) RT a(T ) − V − b V (V + b) + c(V − b) (3.8) Eq. (1969) demonstrated that the RedlichKwong equation of state could be used to reliably calculate the vapour liquid critical properties of binary mixtures.1 Examples of Improved Cubic Equations of State Reference RedlichKwong (1949) SoaveRedlichKwong (1972) PengRobinson (1976) StryjekVeraPengRobinson (1986) PatelTeja (1982) Equations of state Eq. Chueh and Prausnitz (1967a.9) b = 0. (3.
where a T 0 . Soave (1972) calculated the vapour pressures of a number of hydrocarbons and several binary systems with the SoaveRedlichKwong equation. (3.have successfully applied the RedlichKwong equation to the high pressure phase equilibria of binary mixtures.11) m = 0.08664 RTc pc (3. Elliott and Daubert (1987) improved the accuracy of the calculated critical properties of 95 binary 56 .13) and ω is the acentric factor.5 with a more general temperaturedependent a (T ) = 0. Polishuk et al.5) in the Table 3.480 + 1. Soave’s modification fitted the experimental (vapourliquid) curve well and it was able to predict the phase behaviour of mixtures in the critical region.1).57ω − 0.176ω 2 (3. In contrast to the original RedlichKwong equation.12) b = 0.4274 ( R 2 Tc2 T ){1 + m[1 − ( ) 0 . Soave (1972) suggested to replace the term term a(T) (Eq.5 ]}2 pc Tc (3. Deiters and Pegg (1989) used the RedlichKwong equation with quadratic mixing rules to calculate phase diagrams for binary fluid mixtures and to classify them according to the global phase diagram. (1999) predicted the closed loop critical behaviour between type III and V with the Redlich Kwong equation of state. and compared the results of the calculations with experimental data.
37464 + 1 . and compared the results of calculation with experimental data.6) performed as well as or better than Eq. noniterative and computationally efficient correlation of highpressure fluid 57 . and that van der Waals is 0. (3. Peng and Robinson (1977) gave examples of the use of the PengRobinson equation for predicting the vapour pressure and volumetric behaviour of pure component and binary mixtures. Peng and Robinson (1976) redefined a(T) as a (T ) = 0.333.systems containing hydrocarbons. carbon dioxide and nitrogen. Han et al.5).16) The Peng and Robinson equation of state was proposed as Eq.307 (the critical compressibility factor of the RedlichKwong equation is 0. (1997) showed that the PengRobinson equation of state could be used to obtain a relatively accurate. Harstad et al. it slightly improves the prediction of liquid volumes and predicts a critical compressibility factor of Z c = 0.5422 ω − 0.15) b = 0. (1999) also used the SoaveRedlichKwong equation to calculate the phase equilibria of systems containing methane. Zheng et al.07780 RTc pc (3.5 ]} 2 pc Tc (3. Eq.375). (3. (3.14) k = 0 .26922 ω 2 (3. (1988) reported that the PengRobinson equation of state was superior for predicting vapourliquid equilibrium in hydrogen and nitrogen containing mixtures. In general.6).45724 R 2 Tc2 T {1 + k[1 − ( ) 0 .
7). Zhao and Olesik (1999) used the PengRobinson equation of state to calculate methanol and CHF3 binary mixtures. The correct application of StryjekVeraPengRobinson equation of state requires usage of this set.4897153ω − 0.18) b = 0. 1986).5 ](0. Stryjek and Vera (1986) proposed the follo wing modified temperature functionality for the PengRobinson equation of state to extend the range of applicability to polar components: a (T ) = 0.45724 R 2Tc2 T T {k + k1[1 + ( ) 0. In addition.17131848ω 2 + 0.0196554ω 3 (3. it must be noticed that optimization of k1 values was performed using a particular set of critical constants (Stryjek and Vera.378893 + 1. The StryjekVeraPengRobinson equation of state (Stryjek and Vera. 58 .17) k = 0. 1986) was proposed as Eq. (3.0778 RT c pc (3. The results showed that the PengRobinson equation of state with two temperatureindependent binary parameters was capable of representing the experimental data over the entire temperature range with an average relative deviation within 6%.mixtures used in gas turbines and rocket engines.7 − )} pc Tc Tc (3.19) where k1 is an adjustable parameter characteristic of each pure component.
The equation of state allows for adjustment of the critical compressibility factor Zc instead of predicting a fixed value.1). These equations only require the critical properties and acentric factor for the generalized parameters. 1994). The calculated saturated liquid volumes are not improved and they are invariably higher than the measured data. the success of these modifications is restricted to the estimation of phase equilibria pressure. the parameters a and b in the PatelTeja equation are evaluated using the conditions at the critical point given by Eq. pressure. However. (2. This added flexibility improves saturation property predictions for polar fluids while maintaining a simple form that doesn’t require a large amount of pure component or mixture experimental data for accuracy. resulting in Eq. and phase compositions in binary and multicomponent systems. Patel and Teja (1982) proposed a third constant c into the van der Waals equation’s attraction term. The advantages of these equations are they are easy to use and that they often accurately represent the relation between temperature.8). Little computer resources are required and those lead to good phase equilibrium correlation. The condition for the third parameter c is: ηc = p cVc RT c (3. As with the van der Waals equation of state.20) 59 . (3.The PengRobinson and SoaveRedlickKwong equations are widely used in industry (Sadus.
23) c = Ωc RT c pc (3.24) where. for nonpolar substances yielding the following equation. substance specific.Instead of having a fixed value of the critical compressibility factor fixed by parameters a and b. Patel and Teja (1982) calculated η c for 38 substances.22) b = Ωb RTc Pc (3. empirical parameter η c .1) and Eq. (2.329032 − 0. (3. the value is the arbitrary.076799 ω + 0.25) . They also correlated η c to the acentric factor ω .21) Satisfying the conditions of Eq. Ω c = 1 − 3ηc 60 (3. η c = 0. b.20) yields the following equations for the parameters a.0211947 ω 2 (3. and c: a (T ) = Ω a R 2Tc2 α (T ) pc (3. η c is determined by minimizing the errors between experimental saturated liquid volume and those calculated by the equation of state.
5 ) ]F } Tc (3. (3. Patel and Teja (1982) calculated F for the same 38 substances and also correlated it to the acentric factor for nonpolar substances: F = 0 .29) Smith (1995) gave an expression for the compressibility factor Z.27) In the Eq.28) F is a substance specific empirical parameter also determined by minimizing errors between experimental saturated vapour pressures and those calculated by the equation of state.295937 ω 2 (3.2 Ω a = 3η c2 + 3(1 − 2η c )Ω b + Ωb + 1 − 3η c (3. Z 3 + (C − 1) Z 2 + ( −2 BC − B 2 − B − C + A) Z + ( B 2 C + BC − AB ) = 0 (3.30982 ω − 0 .22). the term α (T ) is given by: α (T ) = {1 + [1 − ( T 0 .30) 61 .26) and Ω b is the smallest positive root of the following cubic equation: Ω 3 + ( 2 − 3ζ c )Ω 2 + 3η c2 Ω b − η c3 = 0 b b (3.452413 + 1 .
(1986) have reported a rare. Palenchar et al. They concluded that the critical temperatures and pressures of type I systems could be ade quately predicted although there was a substantial discrepancy between theory and experiment for the critical volume.32) C= cp RT (3. If a vapour and liquid phase are present. calculations for equations of state proposed by Teja and Patel (1982) and Adachi et al. (3. the smallest root represents the liquid’s compressibility while the largest root is that of the vapour. PengRobinson and Soave RedlichKwong equations. methane + hexane or heptane. 62 . decane or water. and propane + hexane and water + benzene binary mixtures.33) Eq.31) B= bp RT (3.where A= ap ( RT ) 2 (3. direct comparison of the quality of prediction of the critical locus using different cubic equations of state. (1983) were reported for the carbon dioxide + methane. The PengRobinson and Soave RedlichKwong equations yielded a more accurate representation of mixtures exhibiting a discontinuity in critical equilibria than the Patel and Teja equation. In addition to the RedlichKwong. butane.30) will yield one to three positive real roots depending upon the number of phases in the system.
noncubic equations of state are of two broad types. (1999) compared the critical lines predicted by RedlichKwong. In general. One type is obtained by simply modifying the repulsive term of the van der Waals equation (hard sphere term). the effect of using an accurate model of repulsion is to generate a noncubic equation of state.1. At high pressures and temperatures. Normally. There are many accurate representations which have been developed for the repulsive interactions of hard spheres and which were incorporated into equation of state.2 Noncubic Equations of State The development of cubic equations of state has been focused on obtaining an improved empirical representation of attractive interactions. Improvements to cubic equations emphasize obtaining better agreement with experiment at low temperatures and pressures for phase equilibria. 3. repulsive interactions can be expected to be the dominant influence determining the properties of the fluid.Polishuk et al. SoaveRedlichKwong. PatelTeja and TrebbleBishnoi (1987) equations of state using classical mixing rules with the experimental data available for the mixtures of methane + alkanes up to and including octane. StryjekVeraPengRobinson. PengRobinson. 63 . Another type is obtained by modifying both attractive and repulsive terms or combining an accurate hard sphere model with an empirical temperature dependent attractive contribution. The results show that different cubic equation of state can yield similar semiquantitative and even quantitative predictions with the optimised values.
5979 y 2 + 0. (2001). (3.3262 y 3 (1 − y) 3 (1 + 0.3076 y 3 ) (3.Many improvements to the simple van der Waals hard sphere repulsive term have been reported (Mulero et al.6539 y 2 + 0.2 Summary of hard sphere expressions of modification of repulsive term of van der Waals equation. A comprehensive review of hard sphere terms is given by Mulero et al. Where y = b/(4V) is a temperature independent packing fraction of spherical molecules.. Some important hard sphere expressions are summarised in Table 3.39) Kolafa and Nezbeda (1994) (3.056 y + 0.40) MalijevskyVeverka (1999) 1 + 1.2.34) Thiele (1963) (3.38) Boublik (1981) 1 + (3α − 2) y + (3α 2 − 3α + 1) y 2 − α 2 y 3 (1 − y ) 3 1+ y + y2 − 2 3 (y + y4 ) 3 (1 − y ) 3 (3.056 y + 1. Table 3. 2001).37) Scott (1971) (3. Reference ReissFrischLebowitz (1959) Hard sphere expression 1+ y + y 2 (1 − y) 3 1+ 2 y + 3y2 (1 − y ) 2 1 (1 − y ) 4 Eq.35) Guggenheim (1965) (3.36) CarnahanStarliing (1969) 1+ y + y 2 − y3 (1 − y ) 3 V +b V −b (3.41) The Guggenheim equation of state (1965) is a more accurate hard sphere model which has been extensively used for the calculation of critical equilibra: 64 .
Hicks et al. Despite the diversity of the systems studied.. 1980a. calculations involving type III equilibria are typically only semiquantitative (Christou et al. In contrast. 1978. 1972): 65 .. Sadus and Young. 1977. b. Sadus.18727 RTc / pc ). Toczylkin and Young. 1994. 1986) because of the added difficulty of predicting the transition between vapourliquid and liquidliquid behaviour. 1977. 2000). Hurle et al. 1985a. 1977a.4963 R 2 Tc2 / pc . 1992a. 1976. good results were consistently reported for the vapourliquid critical locus. The CarnahanStarlingvan der Waals (CSvdW) equation is formed by coupling this hard sphere term with the van der Waals attractive term (Carnahan and Starling. 1980. The critical liquidliquid line of type II binary mixtures was also represented adequately. the compressibility factor Z c = pcVc /( RT c ) = 0.b. Waterson and Young. b. Carnahan and Starling (1969) obtained an accurate representation of hard sphere interactions..42) Both a and b can be obtained by using critical properties ( a = 0. and Wang et al. and at critical point. The Guggenheim equation has been used to predict the critical properties of a diverse range of binary mixtures (Hicks and Young. 1978. Semmens et al.p= RT a − 2 4 V (1 − y ) V (3. c.3611 . b = 0.
Boublik (1981) has generalised the CarnahanStarling hard sphere potential for molecules of arbitrary geometry via the introduction of a nonsphericity parameter (a). and critical compressibility factor Z c = 0.18727 RT / pc ) and attractive (a = 0.p= RT (1 + y + y 2 − y 3 ) a − 2 V (1 − y) 3 V (3.4963 R 2 Tc2 / pc ) equation of state parameters are related to the critical properties.43) The covolume ( b = 0. Svejda and Kohler (1983) employed the Boublik expression in conjunction with Kihara’s (1963) concept of a hard convex body (HCB) to obtain a generalised van der Waals equation of state: p= RT [1 + (3α − 2) y + (3α 2 − 3α + 1) y 2 − α 2 y 3 ] a − 2 V (1 − y ) 3 V (3. Sadus (1993) has demonstrated the CSvdW equation of state can be used to predict type III equilibria of nonpolar mixtures with considerable accuracy.35896 .44) 66 . Yelash and Kraska (1998) found several types of closed loop liquidliquid immiscibility behaviour in binary mixtures based on CSvdW equation of state.
1991). Van Pelt et al. Eq. surface area S* and volume V* of a convex body: α= R* S * 3V * (3. Christou et al. α = 1 .In Eq. There are many other nonspherical or hard chain equations of state which have not been used for the calculation of critical equilibria.44). α is the deviation from spherical geometry which is obtained by considering the mean radius R* . (3. The results obtained were slightly better than could be obtained from similar calculations using the Guggenheim equation of state. Sadus (1993) proposed an alternative procedure for obtaining the equation of state parameters. (3.44) in conjunction with this modified procedure can be used to predict type III critical equilibria of nonpolar binary mixtures with a good degree of accuracy. (3. and Eq.45) In the limiting case of a hard sphere. (3.43). (3.44) has been used for the calculation of the vapourliquid critical properties of binary mixtures containing nonspherical molecules (Sadus et al.44) becomes identical to the CarnahanStarling equation Eq. Eq. (1992) have applied a simplified version 67 .. 1988. Vimalchand and Donohue (1989) have reviewed several hard chain equations of state.
7405 v * / V . The calculations cannot be expected to yield accurate critical volumes for binary mixtures because the critical volume of the components was used as an adjustable parameter to optimise the agreement of theory and experiment for the critical pressures and temperatures of the pure components. and EstelaUribe and Trusler (1998). c is 1/3 of the total number of external degrees of freedom of a molecule.46) was successfully used (van Pelt et al.. The SPHCT equation of state has the following form: p = (1 + c) RT V 2η 2 RTzcv*Y 4η − − (1 − η ) 3 V 2 + v*YV (3. good results were obtained for the critical pressures and temperatures of many binary mixtures. ε is the intermolecular potential energy per unit area. Eq. (3.46) where Y = exp( ε q / 2ckT) − 1 and η = 0. 1995) to qualitatively reproduce a variety of binary mixture phenomena including type VI behaviour. 68 . Recently.of the perturbed hard chain theory (SPHCT) equation of state proposed by Beret and Prausnitz (1975) to binary critical equilibria. where v* is the molecular volume parameter. In common with other equations of state. q is the number of segments per molecule and z is the closepacked coordination number. development of shape factor model equations of state were reported by Huber and Ely (1994).
The equation has been successfully applied to the high pressure phase behaviour of some binary aqueous mixtures (Christoforakos and Franck.47) The attractive term was obtained by examining the virial coefficient of gases in terms of a squarewell potential. (3. CH4 or carbon dioxide is predicted accurately. Xe.48) 69 .47). Franck. The ε parameter reflects the depth of the potential. whereas λ is the relative width of the well. The Christoforakos Franck equation of state introduces a temperature dependent covolume term 4 β = b(Tc / T ) 0.Christoforakos and Franck (1986) and Heilig and Franck (1989) have developed equations specifically for high pressure equilibria which also incorporate the CarnahanStarling hard sphere term. The HeiligFranck (1989) equation of state uses the same form of repulsive term of Eq. 1986. In particular. but it has a different attractive term which is obtained by applying the Padé approximation of the virial coefficients of a squarewell fluid: p= RT (1 + β / V + β 2 / V 2 − β 3 / V 3 ) RTB − 3 V (1 − β / V ) V (V + C / B ) (3. 1987). the type III critical locus of binary mixtures of H2 O + Ar. N2.3 into the repulsive term: p= RT 1 + β / V + β 2 / V 2 − β 3 / V 3 4RT β ( λ3 − 1)(exp( ε / kT ) − 1) − V (1 − β / V ) 3 V2 (3.
49). 1988a. c0 = 0. Calado et al. 1982. Parameter I is a hard sphere constant.. 1984. 1990). b and c.06911 c and Teff = ( cT / a + λρ ) / y .46 for dense fluids. 1989. The equation has not 70 . λ = −0..7404 ρ . Calculations of the vapourliquid critical locus of binary mixtures (Mainwaring et al. η = 0.6887 . and deviations from spherical geometry are reflected in values of c > 1. and the values of y range from 1 for a dilute gas and between 0.34 and 0. (3. b) indicate that good results can only be obtained for mixtures of molecules of similar size.where B and C are the second and third virial coefficients of a squarewell fluid respectively. b) has adopted a semiempirical approach to construct an equation with three adjustable parameters a. Mather et al. (1993) have demonstrated that HeiligFranck equation of state can accurately predict type III critical loci of H2 O + Ne and H 2 O + Kr mixtures. Deiters and Swaid. Deiters (1981a. Rρ 2Teff RT p = {1 + cc0 [4η − 2η /(1 − η ) ]} − aI [exp(1 / Teff ) − 1] ρ b 2 3 (3. In Eq. Accurate calculations of the critical properties of binary mixtures have been reported (Heilig and Franck.49) where ρ = b / V . the Carnahan Starling expression for hard spheres has been adjusted to accurately fit the experimental vapour pressure data for Ar via parameter c0 . 1981). The equation can be used to accurately predict the vapour liquid properties of binary mixtures (Deiters.
3 Thermodynamic Perturbation Theory Equations of State The hard chain equations of state are potentially important for developing accurate equations of state for macromolecules. Progress has been achieved in depicting the properties of hard nonspherical bodies (Sadus. Chiew (1990) developed a rigorous theory for representing hard chains. 1988a). Flory’s concept of excluded volume has been incorporated (Dickman and Hall. Historically..been extensively tested for noncontinuous critical equilibria. 3. (1994) formulated the theory into a useful equation of state for polymers. and Song et al. Boublik (1981) used Kihara’s (1963) concept of a hard convex body (HCB) as the basis of an equation of state for nonspherical hard bodies.1. attention has been focused on using accurate hard sphere models to represent the repulsive interactions of geometrically simple molecules. but type III critical properties of H2 O + benzene and H2O + He can be adequately predicted (Mainwaring et al. 1995). (1988) generalised Wertheim’s TPT model to obtain the equation of state for the compressibility factor of a hard chain of m segments. The form of the compressibility factor is 71 . 1986) in a reasonably accurate hard chain equation of state. Wertheim (1987) proposed a thermodynamic perturbation theory (TPT) which accommodates hardchain molecules. Chapman et al.
50) where g HS (σ ) is the hard sphere sitesite correlation function at contact.Z HC = mZ HS ∂ ln g HS (σ ) − (m − 1)1 + y ∂y (3. y = π m ρσ 3 / 6 is the packing fraction.52) Chiew (1991) obtained the sitesite correlation result for hard dimers: 72 . For the CarnahanStarling equation. The compressibility factor of hard sphere can be accurately determined from the CarnahanStarling equation. and ρ is the number density.51) Ghonasgi and Chapman (1994) modified TPT for the hard sphere chain by incorporating structural information for the diatomic fluid. The compressibility factor of a hard chain can be determined from the hard sphere compressibility factor and the sitesite correlation function at contact of both hard spheres ( g HS ) and hard dimers ( g HD ) Z m = mZ HS − m ∂ ln g HS (σ ) m ∂ ln g HD (σ ) 1 + y − ( − 1) 1 + y 2 ∂y 2 ∂y (3. the sitesite correlation function is g HS (σ ) = 2− y 2(1 − y ) 3 (3. σ is the hard sphere diameter.
Sadus (1995) proposed that.534 + 0.24 y − 0.53) Chang and Sandler (1994) proposed TPTD1 and TPTD2.414 y) 2 (3.55) Starting from the points of Ghonasgi and Chapman (1994) and Chang and Sandler (1994) above.54) TPTD2 can be represented as 1+ y + y2 − y3 m y (5 − 2 y ) − 1 + Z m = m 3 2 (1 − y)( 2 − y) − (1 − y ) m y (3.498 − 0. The form of TPTD1 can be expressed as 1+ y + y 2 − y3 m y (5 − 2 y ) m 2 y (2 + y ) Z m = m − 1 + 3 2 (1 − y)( 2 − y ) − ( 2 − 1)1 + (1 − y)(1 + 2 y) (1 − y ) (3. in general.56) 73 . g HD = g HS (αy + c ) (3.414 y 2 ) ( − 1)1 + (1 − y)( 2 − y)( 0.g HD (σ ) = 1 + 2y 2(1 − y ) 2 (3.
GFD. called the simplified thermodynamic perturbed theorydimer (STPTD) equation of state. 51. 16. 8. Sadus (1996) extended the STPTD equation to hard sphere chain mixtures with no additional equation of state parameters required. Comparison with molecular simulation data indicated that the STPTD equation generally predicts both the compressibility factors and the second virial coefficient more accurately than other equations of state (Chiew equation.57) Sadus (1995) applied the STPTD equation to the prediction of both the compressibility factors of 4. The general form of the STPTD equation of state for pure hard sphere chains is Z m = 1 + m( Z HS − 1) + (1 − m) y ∂ ln g HS α ( 2 − m) y + ∂y 2(αy + c ) (3. TPTD1. By using some elements of the one fluid theory. Good agreement with 74 .and 201mer hard chains and the second virial coefficients of up to 128mer chains. TPTD2).where a and c are the constants for a straight line and the values can be obtained by fitting the molecular simulation data for g HS and g HD . The compressibility factor predicted by the STPTD equation of state was compared with molecular simulation data for several hard sphere chain mixtures containing components with either identical or dissimilar hard sphere segments. and obtained a new equation of state.
the resulting ESTPT equation (Sadus. 1994) for ZHS is used.. In the equation of state an empirical relationship (Shah et al.58) dimer properties into Eq. Sadus (1999a) proposed an empirical simplified thermodynamic perturbed theorydimer (ESTPTD) equation of state.57). 1999a) is 2 ( k + k1 ) y − k 0 y 2 2 k0 y αy (2 − m) k 02 y + (1 − m ) Z m = 1 + m 0 − 2 2 1 − k y k + k − k y + 2αy + c (1 − k 0 y) 0 0 1 0 (3. 1 + (k 0 + k1 ) y (1 − k0 y) Z HS = (3.. The compressibility factor of a hardconvexchain of m segments is obtained as 75 . The addition of Eq.2864 and k1 = 2. Recently. (3.58) where k 0 = 1.simulation data was obtained when the ratio of hardsphere segment diameters for the component chains is less than 2. which is a simplification of STPTD equation. (3.8225 are empirically evaluated constants (Shah et al. 1994).59) Sadus (1999b) extended the hard sphere chain equation to hard convex body chain equation of state (HCBC).
60) represents the breadth of the HCB. 76 . For the HCB equation. The compressibility factor of a hard convex body can be accurately determined from the HCB equation.61) where SR is the ratio of the actual surface area of the HCB S HCB to the surface area of hard spheres S HS (equiv) occupying a diameter equivalent to the HCB diameter: SR = S HCB S HS (equiv) (3. (3. and gave the general form of packing fraction for pure component.60) where g HCB (σ ) is the hard convex body sitesite correlation function at contact. ∂ ln g HCB (σ ) Z HCBC = mZ HCB − ( m − 1) 1 + y ∂y (3. the sitesite correlation function is 1− y / 2 g HCB = 1 + S R (1 − y ) 3 − 1 (3. the symbol σ in Eq.62) Sadus (1999b) discussed the details on how to obtain SR .
5S R y − S R y 2 − ( m − 1)1 + (1 − y)(1 − (3 − 2. directly extending an existing pure fluid model to multicomponent equilibria is an important task (Sadus. oblate spherocylinders and doublecones. and found that these properties are ve ry sensitive to segment shape. 3. perturbation models and equations of state are initially developed for pure substances.2 Mixing rules Most models of the fluid state such as conformal solution theory. This is most commonly achieved by mixing rules and combining rules which relate the properties of the pure components to the properties of the mixtures. prolate spherocylinders. The simplest possible mixing rule is a linear average of the equation of state parameters: 77 .63) Sadus (1999b) examined the effect of molecular shape on the compressibility factor and the second virial coefficient for chain molecule consisting of spheres.5S ) y + (3 − 3S ) y 2 − (1 − S ) y 3 R R R (3. The discussion will be limited to the extension of parameters a and b. 1992a).Z HCBC = m(1 + (3α − 2) y + (3α 2 − 3α + 1) y 2 − α 2 y 3 (1 − y ) 3 2. These two parameters have a real physical significance and are common to many realistic equations of state. Therefore.
64) and Eq. (3.66) b = ∑ ∑ xi x j bij i j (3. 78 .64) is sometimes employed (Han et al. Consequently.a = ∑ xi a i i (3. 1988) because of its simplicity. but Eq. The van der Waals onefluid prescriptions are the most widely used mixing rules: a= ∑∑ x x i i j j a ij (3.. (3. employing both Eq.65) would lead to the poor agreement of theory with experiment.67) where aii and bii are the equation of state parameters for pure components and cross parameters aij and bij (i ? j) are determined by an appropriate combining rule.65) Eq. (3.64) b = ∑ xi b i i (3. (3.64) is rarely used because it does not account for the important role of unlike interactions in binary fluids.
and two parameters van der Waals mixing rules.66) and Eq.. Voros and Tassios (1993) compared six mixing rules (the one. Anderko (1990) gave some examples of the failure of the van der Waals mixing rules for strongly nonideal mixtures. (3. 1976. 1987) that the following mixing rules are more approprate at high pressure: 79 . (1988).and densitydependent mixing rules. the pressure.Eq. A comparison ( Harismiadis et al. (3. Eq. There is some evidence (Deiters. and they both explicitly contain a contribution from interactions between dissimilar molecules. two mixing rules based on excess Gibbs energy models: MHV2 and WongSandler) and concluded that the van der Waals mixing rules give the best results for nonpolar systems.65) and Eq. the van der Waals mixing rule did not yield reasonable vapourliquid equilibrium results. 1991) with computer simulation has concluded that the van der Waals mixing rules are reliable for mixtures exhibiting up to an eightfold difference in the size of the component molecules. water and acetone. They used the van der Waals mixing rule to obtain parameter a and a linear mixing rule to obtain parameter b. (3. Han et al. For the systems which contained strongly polar substances such as alcohol.. The performance of the van der Waals mixing rules has also been thoroughly tested for several equations of state by Han et al.66) were adequate for mixtures of nonpolar and slightly polar compounds (Peng and Robinson. (3. The results showed that most of equations of state with the van der Waals mixing rules were capable of representing vapourliquid equilibria with only one binary adjustable parameter for obtaining aij.67) are based on the implicit assumption that the radial distribution function of the component molecules is identical. 1988).
The parameter γ is obtained from comparison with experimental data.69) 3 3 where a and b are proportional to ε 0 σ 0 and σ 0 .ε 0σ 0 = γ ∑∑x x ε i j i j ij σ ij γ (3. Some attempts have been made to estimate it from theoretical considerations.70) g γ = ∑∑ xi x j g ij i j γ (3. the parameter a for the mixture is calculated directly. ε 0 and σ 0 are the characteristic attractive energy and collision diameter.68) σ 0 = ∑∑ x i x jσ ij γ i j γ (3. respectively. and is typically assigned a value of either 2 or 2. In terms of the 3 conformal parameters (recall hij = g ij ) the above relationships can be represented as: fg γ = ∑ ∑ xi x j f ij g ij i j γ (3.4. Instead of proposing an average of pure component parameter data. 80 .71) Sadus (1989) used conformal solution theory to derive an alternative to the conventional procedure for obtaining parameter a of equation of state.
in the case of the van der Waals equation and assuming a simple model of the energy of the fluid.72) + φ Bθ 2 {h '' / h − (h ' / h) 2 − 1 /[ x (1 − x )]} − φ Aφ B h '' /( hθ )] = 0 where superscripts ' and '' denote successive differentiation of the conformal parameters. For example. a is a function of composition only via the conformal parameters and the contribution from the combinatorial entropy of mixing. The main advantage of Eq. The agreement was generally very good in view of the fact that no adjustable parameters were used to arbitrarily optimise the agreement between theory and experiment. Huron and Vidal (1979) have developed mixing rules from the excess thermodynamic properties of mixing. (3.72) is that the parameter a can be calculated directly from the critical properties of pure components without using combining rules for the contribution of unlike interactions. 1986) is obtained: 81 . The parameter a is obtained by taking the positive root of the following qua dratic equation: a 2θ 6 [−2 f '' / f + ( f ' / f ) 2 − 2h'' / h − 2 f ' h ' / fh + ( h ' / h) 2 )] + aRTV[φ Bθ 2 (−h '' / h − f '' / f )] + 2φ Aθ 3 (h '' / h + f ' h ' / fh − ( h ' / h) 2 ) 2 + 2θ 6 [ −h'' / h + ( h ' / h) 2 + 1 /{ x(1 − x)}] − ( RTV ) 2 [φ A {( f ' / f ) 2 − (h ' / h) 2 } (3. Sadus (1992a.Consequently. the following relationship (Mollerup. and θ and φ are characteristic of the equation of state. b) has applied the above equation to the calculation of the vapourliquid critical properties of a wide range of binary mixtures.
yi = Vi / bi and GE is the excess Gibbs function. The above approach is mainly limited to low pressure equilibria but it may be useful for predicting critical solution temperature phenomena.a = V ∑ xi ( ai / bi ) y i − G EV + RTV [∑ xi ln{bi ( yi − 1) /( y − 1)b}] i (3. 3. Many of the above density dependent prescriptions are undoubtedly more accurate than the van der Waals model at high densities. could also benefit. However it is difficult to isolate the effect of mixing rules on the phase behaviour of fluids because most equation of state calculations rely on additional combining rule parameters for their accuracy. Another approach.73) where y = V / b . there is insufficient data to make an assessment. methods must be ij adopted. But for high pressure phenomena. It appears unlikely that any of the above prescription could eliminate this 82 . Some computer simulation of the high pressure behaviour of binary fluids would be useful in this instance. It is clear that the increased complexity is a disadvantage for high pressure calculations. 1984). The analysis of type III phe nomena. High pressure phenomena also occur over a range of densities. particularly if reliable theoretical estimates of parameters such as γ and g (Eq. is the concept of local composition.71) cannot be obtained and empirical. which is discussed elsewhere (Hoheisel and Kohler. The van der Waals prescriptions are certainly appropriate for vapourliquid phenomena but other rules may be beneficial for dense liquidliquid critical phenomena. where there is a transition between vapour liquid and liquid like densities.
feature. (1971).75) F = ∑ x iσ i3 i (3.76) E = ∑ xi σ i2 i (3.74) where y = π N A F / 6V (3. Z = [1 + {( 3DE / F − 2} y + {( 3E 3 / F 2 ) − (3DE / F ) + 1} y 2 − ( E 3 / F 2 ) y 3 ] /(1 − y) 3 (3.78) 83 . They proposed a direct extension of the CarnahanStarling hard sphere term. the overall effect may be to increase the complexity of the calculations without necessarily improving the accuracy and genuine predictive capability. Thus. A different approach has been adopted by Boublik (1970) and Mansoori et al.77) and D = ∑ xi σ i i (3.
there is a significant deviation. the compressibility factor predicted by Eq. particularly if molecules of dissimilar size are involved.3 (3. it may be beneficial for some dense fluid critical phenomena.71) is identical to the CarnahanStarling equation with the onefluid model. The following equation (Christoforakos and Franck. but in common with other density dependent mixing rules. The temperature dependence of the equation of state parameters is typically obtained by scaling the values obtained at the vapour liquid critical point to lower temperatures and pressures.At low densities. However. The exception is liquidliquid phenomena for which the vapour liquid values are intuitively inadequate.79) It should be noted that the temperature dependence of the parameter a is normally reflected in a lower than normal ? value (see Eq. (3. the analysis of vapourliquid properties is unlikely to be affected.84) and that the agreement with experiment is often very good without formally introducing any temperature dependence in the b term. at moderate and high densities. 84 . The critical values are adequate for high pressure equilibria. Therefore. 3. Dimitrelis and Prausnitz (1986) have concluded that the equation is particularly advantageous for the prediction of vapour liquid equilibria if the molecular size ratio (σ 1 / σ 2 ) 3 exceeds 2. 1986) is acceptable: b(T 2 ) = b(T1 )(T1 / T 2 ) 0.
(3. and (b − (1 − k ij ) aj a a ) ij = [(bi − i ) + (b j − )] RT 2 RT RT (3. (3. For the mixture parameters of an equation of state. Wong and Sandler (1992) tested the Eq. and reported that they were accurate in describing both simple and complex phase behaviour of binary systems for 85 .80) and Eq.81). a and b are E ai A∞ a = b(∑ xi + ) bi C i (3.81) E where C is a constant dependent on the equation of state selected and A∞ is the excess Helmholtz function at infinite pressure.80) b= ∑∑ x x i i j j (b − a ) ij RT AE a 1 + ∞ − ∑ xi i RT bi RT i (3.Wong and Sandler (1992) used the Helmholtz function to develop mixing rules to satisfy the second virial condition.82) where k ij is a binary interaction parameter.
and investigated the 86 . Orbey and Sandler (1995) slightly reformulated the WongSandler mixing rules by rewriting the cross second virial term given in Eq. To go smoothly from activity coefficientlike behaviour to the van der Waals one fluid mixing rule. (1992) demonstrated that the WongSandler mixing rules can be used for highly nonideal mixtures. Castier and Sandler (1997a. They concluded that the WongSandler mixing rule can correlate the solvent partial pressure in concentrated polymer solutions with high accuracy over a range of temperatures and pressures with temperatureindependent parameters. b) performed critical point calculations in binary systems utilizing cubic equations of state combined with the WongSandler mixing rules. (3.diverse systems. Wong et al.82) as (b − bi + b j a ) ij = − RT 2 ai a j (1 − k ij ) RT (3. They demonstrated that the mixing rules can be used to make high pressure vapourliquid equilibrium predictions from low pressure data. Orbey and Sandler (1994) used the WongSandler mixing rule to correlate the vapourliquid equilibria of various polymer + solvent and solvent + long chain hydrocarbon mixtures. Huang and Sandler (1993) compared the WongSandler mixing rules for nine binary systems.83) Orbey and Sandler (1995) tested five binary systems. and showed that this new mixing rule was capable of both correlating and predicting the vapourliquid equilibrium of various complex binary mixtures accurately over wide ranges of temperature and pressure.
influence of the mixing rules on the shape of the calculated critical phase diagrams. Michelsen and Heidemann (1996). 3. C6 E3 . Comparison and evaluation for various mixing rules can be found in the works of Knudsen et al. They found that the model fitted the experimentally found type of phase behaviour quite correctly. The results showed that the combination can obtain many different types of critical phase diagrams. Wei and Sadus (1999b) and Saghafi and Moshfeghian (2000). 87 .3 Combining Rules Any realistic mixture prescriptions will invariably contain contributions from interactions between unlike molecules. (1996). (2000) used the PengRobinson equation of state with the WongSandler mixing rule to investigate binary mixtures of water plus surfactant systems containing C4 E1 . The most widely used combining rule for the a ij term was first proposed by van der Waals: a ij = ξ ij aii a jj (3. the following combining rule for critical calculations can be used. Orbey and Sandler (1996). C7 E3 and C7 E5 .84) Alternatively. (1998). Wang et al (1996). This means that the cross term a ij and bij ( i ≠ j ) must be evaluated. Twu et al. Rudolph et al.
are associated with different interaction parameters. mixture prescriptions and the equation of state. but different molecular interactions. the requirement for ? values to optimise agreement between theory and experiment is also ij testimony to the inadequacy of the combining rule. The ξ ij parameter is frequently interpreted as reflecting the strength of intermolecular interaction. Elliot and Daubert. This interpretation is supported by the fact that values of ξ ij obtained from the analysis of the critical properties of many binary mixtures consistently decline with increasing size difference between the component molecules. It has been frequently documented (Mainwaring et al.85) The ξ ij term (also commonly defined as 1 ij ) is frequently introduced into combining rules k to optimise the agreement between theory and experiment.aij = ξ ij bij aii a jj bii b jj (3. Nonetheless. 1987) that different ?ij values are required to obtain optimal agreement betwee n theory and experiment for vapourliquid and liquid liquid critical equilibria. 1988b. respectively. Mixtures of molecules of similar size.86) 88 . The most widely used combining rules for bij are the simple arithmetic rules: bij = ζ ij bii + b jj 2 (3.
Good agreement between the ory and experiment could be obtained by determining both the optimal ξ ij and ζ ij parameters. 89 . The adjustable parameter ζ ij is used to optimise the agreement between theory and experiment. 1994): bij = ζ ij (b1 / 3 + b1 / 3 ) 3 ii jj 8 (3. but the discrepancy increases substantially for mixtures of molecules of very dissimilar size. Sadus (1993) has examined type III behaviour of nonpolar mixtures using both arithmetic and Lorentz combining rules. that is. obtaining an optimal ζ ij parameter is essential for the accurate representation of type III behaviour. Sadus (1993) proposed an alternative combining rule by taking a 2:1 geometric average of the Lorentz and arithmetic rules without the parameter ζ ij . It is generally found that the parameter ζ ij is not required for the accurate prediction of vapourliquid critical properties (Sadus. good agreement between theory and experiment for type II liquidliquid equilibria can usually be obtained by solely adjusting the ξ ij term and setting ζ ij = 1. Sadus.and the Lorentz combining rule (Hicks and Young. It is found (Sadus. 1994). 1993) that the arithmetic combining rule substantially overestimates the twophase region.87) Both the simple arithmetic and Lorentz combining rules yield nearly identical results for mixtures of molecules of similar size. However. 1975. Similarly. 1992a. whereas the Lorentz rule is more accurate but it extends the onephase region.
it cannot be extrapolated to low densities (Kiselev. The idea of the approach consists in the renormalization of the temperature and the density in the equations of state. 3.4 Crossover Equation of State Chen et al. The result is that the temperature and the density become nonanalytic scaling functions of the dimensionless distance to the critical point. (1993) introduced a theory that incorporates scaling laws asymptotically close to the critical point and that are transformed into the regular classical expansion far away from the critical point. Kiselev (1998) developed a cubic crossover equation of state based on the nonclassical critical phenomena for pure fluids which incorporates the scaling laws asymptotically close to the 90 . Although the original crossover equation of state gives an accurate representation of the thermodynamic properties of fluids in a wide region around the critical point of pure components.88) Sadus (1993) reported that the new combining rule (Eq. 1998). 3. Fox (1983) proposed an approach to the incorporation of scaling laws in classical equations of state. and this lead to the crossover equations of state.88) is generally more accurate that either the simple arithmetic or Lorentz combining rules.1 bij = {1 / 4(21 / 3 )}(bii / 3 + b1/ 3 ) 2 (bii + b jj )1/ 3 jj (3. (1990) and Jin et al.
therefore. (3.1 Classical Expression for the Critical part In the critical region. In the critical region  τ << 1 and  ∆η  <<1. Eq.89) where τ = T / Tc − 1 is the dimensionless deviation of the temperature T from the critical temperature Tc . or Landau theory of critical phenomena (Laudau and Lifshitz. aij are the systemdependent coefficients. ∆η ) = a12τ∆η 2 + a04∆η 4 (3. the van der Waals equation of state corresponds to the meanfield. 1980. Patashinskii and Pokrovskii.89) ∆A(τ . 3. 1979). ∆η ) = ∑∑ a τ ij i j i ∆η j (3. The main assumption of the classical theory of critical phenomena is that the critical part of the Helmhotz function. and ∆η is an order parameter (Landau and Lifshitz.4. (3. there are two main terms in Eq. 1998) ∆ A(τ . 1980) where the longscale fluctuations in the order parameter are negligible 91 .90) is valid only in the temperature region Gi <<  τ << 1 (Gi is the Ginzburg number. ∆ A of the system can be represented by a Taylor expansion in the powers of the order parameter (Kiselev.90) that correspond to the critical phenomena (Landau and Lifshitz. Landau and Lifshitz.critical point and which is transformed into the original cubic equation of state far away from the critical point. 1980). 1980).
Y is a crossover function to be determined from the set of coupled algebraic equations by Jin et al. (1993).90) with the replacement of the dimensionless temperature τ and the order parameter ∆ η by the renormalized values (Kiselev.91) is equivalent to Eq. which provides the correct scaling behaviour of the isochoric specific heat asymptotically close to the critical point (Kiselev. K (τ 2 ) is the kernel term. Patashinskii and Pokrovskii. If  τ << Gi . 1979).91) where. 1998).(Laudau and Lifshitz. (3. (3. α . ∆η ) = a12τY − α 2 ∆1 ∆η Y 2 γ −2 β 2∆1 + a04∆η Y 4 γ −2 β ∆1 − K (τ 2 ) (3. the singular part of the thermodynamic potential of a system becomes a nonanalytical scaling function of the τ and ∆η . the critical fluctuations close to the critical point result is: ∆A(τ .92) where. 1998): 92 . − 1 K (τ ) = a 20τ 2 (Y ∆1 − 1) 2 2 α (3. A theoretical approach for constructing a crossover expression for the thermodynamic potential of a system in the critical region based on the renormalizationgroup calculations of Nicoll and Bhattachafjee (1981) and Nicoll and Albright (1986) has been developed by Chen et al. 1980. (1990). Eq. β and γ are critical exponents. Using this approach.
(3.4. As one moves further from the critical point.90) may be considered. more terms in Eq.V ) = −∫ pdV + Aideal (3. 3.  τ  and  ∆η  increase.94) where Aideal corresponds to the temperature dependent function of the idealgas part of the Helmholtz energy.95) The dimensionless Helmholtz energy can be represented in the form 93 . (3. T ) .2 Crossover Expression It is wellknown the Helmholtz energy per mole (Sandler.89) and Eq. 1999) that can be described as A(T . For a classical equation of state p(V . can be written in the dimensionless form p= ~ PVc RT (3. ∆ η = ∆η Y ~ γ − 2β 4∆1 (3.τ = τY ~ − α 2∆1 .93) and adding the kernel term K (τ 2 ) .
p 0 (T ) = p(Vc . while the critical density is usually found as fitting parameter of the model. The new forms of dimensionless temperature τ ~ ~ and the order parameter ∆η 94 . The critical temperature and pressure of the equations of state can be taken from available measurements or predictions. the critical density found by this method does not coincide with the real critical density of the system (Kiselev. The additional terms to take into account the difference between the classical critical temperature Tc and critical volume Vc .96) where ∆ A (∆ T . (3.94) by Rc Kiselev (1998). ∆V ) corresponds to the critical part of the Helmholtz function. and ∆ V = V / Vc − 1 is also the dimensionless order parameter. and the real critical parameters T and VRc were introduced in Eq. (3. ∆V ) according to Eq. ∆ T = Tr − 1 ( T r = T / Tc ) is the dimensionless deviation of the temperature from the critical temperature Tc . T )Vc / RT is the dimensionless pressure at the critical isochore V = Vc and u 0 is relation to employed equation of state.V ) = =− RT RT ~ ~ ~ ~ Aideal RT p ∫ Vc dV + RT = ∆ A(∆T . 1998). ~ ~ ~ ∆ T and ∆ V are needed to renormalized in the critical part of the classical Helmholtz function ∆ A (∆ T .95) by Kiselev (1998).~ A(T . ∆V ) − ∆V P 0 (T ) + u 0 (T ) (3. V ) 1 A(T . chosen to the best description of the vapourliquid equilibrium surface far away from the critical point. Since any classical equation of state does not reproduce the thermodynamic surface of a fluid in the critical region.
τ =τ Y ~ − α 2 ∆1 + (1 + τ ) ∆τ c Y 2 ( 2 −α ) 3∆1 (3.359 is the linearmodel parameter (Kiselev and Sengers. 95 .98) can be written in the parametric form as q (1 + q) 2 ∆ 1 Y ( q) = [ ] 1+ q + q 2 (3.98) where the order parameter ∆ η = V / Vc − 1 . respectively.97) and (3. while ∆τ c = ∆ Tc / T Rc and ∆η c = ∆V c / V Rc are the dimensionless shifts of the critical temperature and the critical volume. The terms ∝ τ and ∝ τ 2 correspond to a projection of the rectilinear diameter of the coexistence curve in the temperaturedensity variables ρ d = ( ρV + ρ L ) / 2 = ρ c (1 + dτ ) on the temperaturevolume projection V d = 1 / ρ d ≅ Vc (1 + d1τ + d 2τ 2 ) . 1993). Parameters T and VRc are the real critical temperature and volume of fluid. (3.97) ∆ η = ∆ ηY ~ γ − 2β 4 ∆1 + (1 + ∆η )∆ η cY 2 −α 2 ∆1 (3.99) and the parametric variable q can be found from the solution of the equation τ ∆η + d1τ + d 2τ 2 2 2 q = + bLM ( ) Y Gi Gi β 2 1− 2 β ∆1 (q) (3.100) 2 bLM = (γ − 2 β ) / γ (1 − 2β ) = 1. The crossover Rc function Y in Eqs.
The general form of the crossover equation of state obtained from the crossover expression in Eq. the value of p 0 is: p 0 (T ) = ~ 1 Tc Ω aα (T ) − b1 T Z c b2b3 (3. (3.To complete the transformation of the classical Helmholtz function into the crossover form.103) 96 .V ) = ∆ A(τ .96) with the renormalized values τ and ∆ η . (3. ∆ η ) − ∆ V p 0 (T ) + u 0 (T ) − K (τ 2 ) ~ ~ ~ ~ ~ ~ (3.4.102) ~ For the PatelTeja crossover equation of state (Kiselev. and the kernel term K (τ 2 ).102): RT Vc p= − Vc V Rc ~ ~ ∂∆ A + p (T ) + Vc 0 ∂∆ η V Rc T ∂K (τ 2 ) ∂∆ η T (3. The crossover expression for the Helmholtz function is ~ ~ A(T . 1998).101) 3.3 Crossover Equation of State A crossover equation of state incorporates the scaling laws asymptotically close to the critical point and is transformed into the original equation of state far away from the critical point. we must replace the classical dimensionless temperature ∆T and the volume ∆V in Eq.
104) and Eq. ∆η ) = − ln T ~ Ω T Z c b2 b3 b1 ∆η/ b + 1 b1 3 ~ ~ ~ ~ ~ ~ (3. and Ω c are functions of the critical compressibility factor Z c (Eq.101) with PatelTeja equation of state is 97 .29).105) are b1 = 1 − Ωb Zc Ωb + Ωc − Ω 2Zc Ωb + Ωc + Ω 2Z c b2 = 1 + (3. and Ω Ω = Ω 2 + 6Ω b Ω c + Ω 2 b c (3. (3.106) b3 = 1 + ~ Remember that u 0 in Eq. (3.25) Eq. Ω b . b2 and b3 in Eq.104) where Ω a . (3.28) and Eq. (3.105) The definition of α (T ) can be found in Eq.27)). Parameters b1 . (3.and the critical part ∆ A is given by ~ ~ ~ ~ ∆ η + 1 + Tc Ω aα (τ ) ln ∆ η/ b2 + 1 + ∆η − Tc Ω aα (τ )∆η ∆ A (τ . (3. (3.
1982) to compare with experimental data for pure CO 2 .and twophase regions.Vc ) RT (3.u 0 (T ) = − ln b1 + ~ ~ Tc Ω aα (T ) b2 ln( ) + A ideal T Ω b3 (3. and refrigerants R32 and R125 in the one. water.0) − ln( ∆ η + 1) + ∆ η p 0 (τ ) ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ (3. The crossover statistical associating fluid theory equation of state has the same form as the cubic one of Eq. but p 0 (T ) and ∆ A have the different forms because of different classical equation of state. ∆ η ) − A r (τ . 98 . Kiselev and Ely (1999) also developed a crossover modification of the statistical associating fluid theory equation of state for macromolecular chain fluids.107) and Aideal = ~ Aideal RT (3.108) Kiselev (1998) used the modified PatelTeja cubic equation of state (Patel and Teja.103). ~ ~ p 0 (T ) = and ~ Vc p(T . than the original PatelTeja equation of state.109) ∆ A (τ . He showed that the crossover PatelTeja equation of state yields a much better representation of the thermodynamic properties of pure fluids.110) where Ar is the dimensionless residual part of the Helmholtz energy. (3. especially in the critical region and for vapourliquid equilibrium. ∆ η ) = A r (τ .
Kiselev and Ely (1999) compared the crossover statistical associating fluid theory and original equation of state with experimental single phase data for ntriacontane and ntetracontane. The results show that the crossover statistical associating fluid theory equation of state reproduces the saturated pressure data in the entire temperature range from the triple point to the critical temperature with an average absolute deviation of about 3.8%, the saturated liquid densities with an average absolute deviation of about 1.5%, the saturated vapour densities with an average absolute deviation of about 3.4%, and gives a much be tter representation of the experimental values of pressure and the liquid density in the critical region.
Wyczalkowska et al. (1999) also developed a global crossover equation of state to show how the density fluctuations affect all thermodynamic properties more and more significantly when the fluid approaches the critical point.
References
Abdulkadirova, K. S., Wyczalkowska, A. K., Anisimov, M. A. and Sengers, J. V. (2002). Thermodynamic Properties of Mixtures of H2 O and D2 O in the Critical Region, J. Chem. Phys., 116 , 45974610. Adachi, Y., Lu, B. C. Y. and Sugie, H. (1983). ThreeParameter Equation of State, Fluid Phase Equilib., 13, 133142. Anderko, A. (1990). EquationofState Methods for the Modelling of Phase Equilibria, Fluid Phase Equilib., 61, 145225.
99
Baker, L. E. and Luks, K. D. (1980). Critical Point and Saturation Pressure Calculations for Multipoint Systems, Soc. Pet. Eng. J., 20, 1524. Beret S. and Prausnitz, J. M. (1975). Perturbed HardChain Theory: An Equation of State for Fluids Containing Small or Large Molecules, AIChE J., 26, 11231132. Boublik, T. (1970). HardSphere Equation of State, J. Chem. Phys., 53, 471473. Boublik, T. (1981). Statistical Thermodynamics of Nonspherical Molecule Fluids. Ber. Bunsenges, Phys. Chem. 85, 10381041. Calado, J. C. G., Deiters, U. and Streett, W. B. (1981). LiquidVapour Equilibrium in the Krypton + Methane System, J. Chem. Soc. Farad I., 77, 25032513. Carnahan, N. F. and Starling, K. E. (1969). Equation of State for Nonattracting Rigid Sheres, J. Chem. Phys., 51, 635636. Castier, M. and Sandler, S. I. (1997a). Critical Points with the WongSandler Mixing Rule. I. Calculations with the van der Waals Equation of State, Chem. Eng. Sci., 52, 33933399. Castier, M. and Sandler, S. I. (1997b). Critical Points with the WongSandler Mixing Rule. II. Calculations with a Modified PengRobinson Equation of State, Chem. Eng. Sci., 52, 35793588. Chang, J. and Sandler, S. I. (1994). An Equation of State for the HardSphere Chain Fluid: Theory and Monte Carlo Simulation, Chem. Eng. Sci., 49, 27772791. Chapman, W. G., Jackson, G. and Gubbins, K. E. (1988). Phase Equilibria of Associating Fluid: Chain Molecules with Multiple Bonding Sites, Mol. Phys., 65, 10571079. Chapman, W. G., Gubbins, Gubbins, K. E., Jackson, G. and Radosz, M. (1990). New Reference Equation of State for Associating Liquids, Ind. Eng. Chem. Res., 29, 17091721.
100
Chen, Z. Y., Albright, P. C. and Sengers, J. V. (1990). Crossover from Singular Critical to Regular Classical Thermodynamic Behavior of Fluids, Phys. Rev. A 41, 31613177. Chiew, Y. C. (1990). PercusYevick IntegralEquation Theory for Athermal HardSphere Chains: I. Equations of State, Mol. Phys., 70, 129143. Chiew, Y. C. (1991). PercusYevick IntegralEquation Theory for Athermal HardSphere Chains: II. Average Intermolecular Correlation Functions, Mol. Phys., 73, 359373. Christoforakos, M. and Franck, E. U. (1986). An Equation of State for Binary Fluid Mixtures to High Temperatures and High Pressures. Ber. Bunsenges. Phys. Chem., 90,780789. Christou, G., Morrow, T. Sadus, R. J. and Young, C. L. (1986). Phase Behaviour of Fluorocarbon + Hydrocarbon Mixtures: Interpretation of Type II and Type III Behaviour in Terms of a ‘Hard Sphere + Attractive Term’ Equation of State, Fluid Phase Equilib., 25, 263272. Christou, G., Young, C. L. and Svejda, P. (1991). Gas Liquid Critical Temperatures of Binary Mixtures of Polar Compounds + Hydrocarbons or + Other Polar Compounds. Ber. Bunsenges. Phys. Chem., 95, 510514. Chueh, P. L. and Prausnitz, J. M. (1967a). VapourLiquid Equilibria at High Pressures. VapourPhase Fugacity Coefficients in Nonpolar and QuantumGas Mixtures, Ind. Eng. Chem. Fundam., 6, 492498. Chueh, P. L. and Prausnitz, J. M. (1967b). Vapour Liquid Equilibria at High Pressures. Calculation of Partial Molar Volumes in Nonpolar Liquid Mixtures, AIChE J., 13, 10991107. Deiters, U. K. and Schneider, G. M. (1976). Fluid Mixtures at High Pressures. Computer Calculations of the Phase Equilibrium and the Critical Phenomena in Fluid Binary
101
and Hall. Eng. Dickman. E. Deiters. Phys. 33. (1982). and Donohue. Phys. R. 37. D. M. Systematic Investigation of the Phase Behaviour in Binary Fluid Mixtures. M.. 80. Phys. and Prausnitz J. J.Mixtures from the RedlichKwong Equation of State. Res. Chem. K. Sci. 121. Deiters. D. K. (1986). Calculations Based on the RedlichKwong Equation of State. 855861. Ind.. (1987). 66326641. D. 26.. Equation of State for Chain Molecules: ContinuousSpace Analog of Flory Theory. Comparison of two Hard Sphere Reference Systems for Perturbation Theories for Mixtures. and Prausnitz. Ber. and Pegg. Calculation of FluidFluid and SolidFluid Phase Equilibria in Binary Mixtures at High Pressures. Chem. J. 16861691. A New Semiempirical Equation of State for Fluid: III. Ber. (1984). 267293. Deiters. Chem. and Daubert.. U. Dimitrelis. C. U. 85. (1996). I. J. I. Eng. 116 . T. M. Chem. U. DensityDependent Mixing Rules for the Calculation of Fluid Phase Equilibria at High Pressures. U. (1987). Perturbed Hard Chain Theory for Fluid Mixtures: Thermodynamic Properties for Mixtures in Natural Gas and Petroleum Technology. R. J.. Equations of State for Hydrogen Bonding Systems. K. (1989). 24. (1986). Chem. 849860. K. I. Fluid Phase Equilib. (1978). Fluid Phase Equilib. 518529. 31. 102 . Chem. G. I.. 791796. 41084115. Economou. AIChE J. Application to Phase Equilibria in Binary Mixtures. Bunsenges.. Donohue.. 88. Fluid Phase Equilib. Phys. L. and Swaid. 90. Elliott.. Deiters. Bunsenges. 13161321. Evaluation of an Equation of State Method for Calculating the Critical Properties of Mixtures. M.
U. Sci. K. I. K. Computer Simulation. J. Variations on van der Waals’ Equation of State for High Densities. 19. and Streett. (1989).. and Bellan. Calculation of Thermodynamic Properties of Binary Fluid Mixtures to High Temperatures and High Pressures. VapourLiquid Equilibrium of Molecular Fluid Mixtures by Equation of State. E. Fluid Phase Equilib. Fluid Phase Equilib. Chem. S. Phys. 65. 225244. Fluid Phase Equilibria: Experiment. C. G. F. J. S. 45734585.. 14. Thermodyn. Chem. P. and Chao. Ber. How Good is Conformal Solutions Theory for Phase Equilibrium Predictions?. M. Miller..EstelaUribe.. Guggenheim. Koutras. Chem. 43. 199200. N. K. F. J.. Heilig. (1965). 9. Fluid at High Pressures and Temperatures. 100 . (1994). Efficient HighPressure State Equations.. and Trusler. (1965). (1987). 23272367. Chem. 118. K. 18331838. Phys. Fluid Phase Equilib. Phys. Fox.. 66336639.. E.. and Panagiotopoulos. 93. Phys. R. J. Tassios. W. A New Equation of State for Hard Chain Molecules. 103 . W. Am. Harismiadis. AIChE J. Lin.. V. Franck. 87. D. J. U. J. 151. Statistical Thermodynamics of Liquid Mixtures. P. Chem. and Franck. Method for Construction of Nonclassical Equation of State. (1991). B. 225234. and Theory. Flory. Han. (1983).. 16051610.. A. Harstad. Z. Ghonasgi. (1997). P. Bunsenges. G. I. (1998). Soc. K.. Mol. and Chapman. (1983). H. J. Shape Factors for Light Hydrocarbons. D.. S.. 4553. 898905. Gubbins. E. Shing. E. M.. 43. Chem. J. A. (1988). J. M. Eng. 87.
Rev. Prediction of VapourLiquid Equilibria at High Pressure Using Activity Coefficient Parameters Obtained from Low Pressure Data: A Comparison of two Equation of State Mixing Rules. Toczylkin. P. 2735. Chem.. P. L. H. R. C. J. (1976). F. Chem. (1990). Soc. and Young. L. Faraday I. J. Phys. Huang. (1984). C. (1978). C.. 32. P. (1975). U. C. C. Comparison 104 .. Local Composition in Liquid Mixtures. C. 31. Hicks. A Predictive Extended Corresponding States Model for Pure and Mixed Refrigerants Including an Equation of State for R134a. M. Eng. Upper Critical Solution Temperatures of Hydrocarbon + Fluorocarbon Mixtures.. L.. The GasLiquid Critical Properties of Binary Mixtures. (1977a). Chem. J. Chem. Chem. Aust. R. Fluid Phase Equilib. 73. Huber M. Ind. and Young C... S. 1324. L. 94. J. and Young. and Young. E. and Ely. and Young.. and Franck. 119175. Res.Heilig. Phase Equilibria of Multicomponent Fluid Systems to High Pressures and Temperatures. L. Hicks. Hoheisel. 75. (1977). R. Ber.L. Theoretical Prediction of Phase Behaviour at High Temperatures and Pressures for NonPolar Mixtures: II. Bunsenges. F. Theoretical Prediction of Phase Behaviour at High Temperatures and Pressures for NonPolar Mixtures: IV. L. L. Hurle. L. J. P. 1925. Hicks. 16. Soc. Hurle. 122133. 18841888. 1831. Comparison with Experimental Results for Carbon Dioxide + nAlkane mixtures. L. S. 17.. and Sandler. 72. Int. S. Toczylkin. and Kohler. C. Chem. I. (1994). 14981503. Hurle. Refrig. Faraday II. L. Critical Properties of Binary Mixtures. C. Hicks. (1993). H.
Huron. 73. Kiselev.. 613617. Donohue.. Toczylkin. 388402. Soc. and Vidal. 723. Chem.. (1986). Adidharma. Cubic Crossover Equation of State. (1977b). 255271. (1993). H. J. J. D. New Mixing Rules in Simple Equations of State for Representing VapourLiquid Equilibria of Strongly NonIdeal Mixtures. 32. Fluid Phase Equilib. Fluid Phase Equilib. 32. A Crossover Equation of State for Associating Fluids. 3. S. Jin. Kiselev.. and Radosz. (1986). (1998).. J. M. 183 . 5364. Fluid Phase Equilib. C. R. P. B. and Sengers. and Sandler. 618622. and Young. Theoretical Prediction of Phase Beha viour at High Temperatures and Pressures for NonPolar Mixtures: III. H.. Kiselev S. X. Local Composition Model for Chainlike Molecules: A New Simplified Version of the Perturbed Hard Chain Theory. (1963). E 47. L. Ely. Global Thermodynamic Behavior of Fluid Mixtures in the Critical Region. 105 . V. G. 73. Faraday II. L. S. M. (1997). (2001). C. S.. Convex Molecules in Gaseous and Crystalline States. J. 147188. Vimalchand. D. Hurle. V. Faraday II... Kihara. 128. S. L.with Upper Critical Solution Temperatures for Perfluoromethylcyclohexane + Hydrocarbons. 128. GasGas Immiscibility. 147. D. S. and Donohue. B. T. Kim. Prediction of the Thermodynamic Properties and the Phase Behavior of Binary Mixtures in the Extended Critical Region. Tang. F. Adv. 17261733.. Fluid Phase Equilib. AIChE J. I. Chem. Phys. J. G.. Phys.. B. AIChE J. Rev. Chem. Ikonomou. (1979). J. 17161725. Thermodynamics of HydrogenBonded Molecules: The Associated Perturbed Anisotropic Chain Theory. M. M. Soc.
J. and Franck. Res. K. Eng. Ind. and Fredenslund. Chem. Starling. M. Chem. E. Fluid Phase Equilib. 134. Michelsen. Stenby.. A. R. 82. I. H. Eng.Knudsen. F. (1996). 459466. R. Sadus. Ind. A. 25. Mainwaring. Prediction of Binary and Ternary Critical Properties Using De iters’ Equation. Phase Equilibria in the Water + Krypton and Water + Neon Systems to 250 MPa and 700K. (1994). Chem. E. Sadus... Mainwaring. 54. N. D. L. and Nezbeda. A. 15231525. (1996). 18. C. New York. and Lifshitz. 771779. Chem. (1979).... 100. J. 35. (1993). Pergamon. Hard Sphere and Hard Convex Body Equations of State. Sadus. Fundam. and Young. LennardJones Fluid: An Accurate Analytic and TheoreticallyBased Equation of State. (1971). R. Phys.. Thermodyn. Statistical Physics. New Equations of State for Pure and Binary HardSphere Fluids. R. (1988a). E. 8197. L. E. D. and Veverka. Fluid Phase Equilib. T. 361368. C.... 106 . (1999).. L. and Young. Jr. A. D. 85103.. (1980). Malijevsky. J. Chem. A Comprehensive Comparison of Mixing Rules for Calculation of Phase Equilibria in Complex Systems. K.. Chem. J. E. Phys. Cubic Equations of StateWhich?. Mansoori. Deiters’s Equation of State and Critical Phenomena. Kolafa. Equilibrium Thermodynamic Properties of the Mixture of Hard Spheres. Carnahan. J. J. J. Landau. Mather. J. Eng. Some Properties of Equation of State Mixing Rules Derived from Excess Gibbs Energy Expressions. 42674270. E. 278287. Fluid Phase Equilib. Martin. 43. Phys. 42. Sci. A. L.. (1988b).. M. E.. and Leland. U. 1. Chem. J. G. and Heidemann. W. 3rd Ed.
Rev. Orbey. 42. (1996). Rev. 25.. ACS Symp. I.. Equations of State for Hard Spheres. Mulero. and Pokrovskii.. 23. N. A Note on the Derivation of Mixing Rules from Excess Gibbs Energy Models. T. J. (1986). Fluid Phase Equilib. and Leland. Erickson. 300. D. B. R. Fluctuation Theory of Phase Transitions. 107 . and Albright. (1982). Pergamon. J. C. and Bhattachafjee. Chem.. 3. (1976). Orbey. 463473. Chem. Chem. A New Cubic Equation of State for Fluids and Fluid Mixtures. Y. (1981). 19911996. Ser. (1995). Phys.. Eng. J. 23272334. Eng. Patel. M. and Sandler S. I. AIChE J.. Nicoll. Reformulation of WongSandler Mixing Rule for Cubic Equations of State. Galan. 40. D. 132155. K. D. VapourLiquid Equilibrium of Polymer Solutions Using a Cubic Equation of State. C. Patashinskii. 389401. A New TwoConstant Equation of State. A review of Accuracy and Applications. Ind. Chem. and Sandler S. D. and Robinson. 41. New York. Nicoll. 49914999. A. Phys. 34. (1979). Orbey.Mollerup. B. A. Z. (1994). 683690. N. C. Background Fluctuations and Wegner Corrections. V. J. P. Peng. S. Phys. and Sandler S. AIChE J. and Cuadros. Palenchar. Crossover Functions by RenormalizationGroup Matching: O ( epsilon 2) Results. I. (1986). B. A. (2001). Fundam. 37. 323327. Analysis of Excess Free Energy Based Equations of State Models... N. F. 15. F. W. N.. Prediction of Binary Critical Loci by Cubic Equations of State. F. AIChE J. and Teja. L.. Sci. 5964. 12031209. (1986). Phys.
108 . and Young.. V. S. Chem. L. 137144. 44. Wisniak J. R. J. and de Swaan Arons. J. S.. Chem. Y. R. .. Prediction of the Critical Locus in Binary Mixtures Using Equation of State. N. Cubic Equation of State. Frisch.. Reiss. Phase Behaviour of Water Plus Nonionic Surfactant Systems: Experiments and Modeling Applying the PengRobinson Equation of State with the WongSandler Mixing Rules and the UNIQUAC gEModel. Int. (1977). Amsterdam. Crossover Equation of State for the Thermodynamic Properties of Mixtures of Methane and Ethane in the Critical Region. H. W. Th. R. (1949). L. Redlich. Rev. An Equation of State: Fugacities of Gaseous Solutions. 909944. and Kwong. 17. Kiselev. Phys. The Molecular Theory of Solutions. de Loos. B. Fluid Phase Equilib. J. J. J. 1347.. Polishuk I. Prigogine. AIChE J. (2000). J.. Mixtures of Methane Alkanes.. Sadus. Sadus. A Rigorous Method for Predicting the Critical Properties of Multicomponent Systems from an Equation of State. J. and Lebowitz. N. E. (1999). and Segura H. J. 31. Aust.. (1957). G. Rudolph. 22. (1985b). O. On the Thermodynamics of Solutions: V. L. Fluid Phase Equilib. NorthHolland. J. 38. Statistical Mechanics of Rigid Spheres. Povodyrev... Langeveld. and Young. Jin. 369380. 17391743. D. Phase Behaviour of Carbon Dioxide and Hydrocarbon Mixtures. (1985a)... H. L.Peng. 164. Classical Mixing Rules. C. A. Phase Behaviour of Binary nAlkanenitrile and n Alkanes and Nitromethane and Alkane Mixtures. 225229. (1959). S. 8196. X. (1996).. 173. J. 23.. 233244. I. Thermophysics. I. and Sengers. Chem. C. B. A. J. Fluid Phase Equilib. L. and Robinson. D.
J. P. J. Sadus.. An Equation of State for Hard Convex Body Chains.. 51975202. Chem. J. AIChE J. J. 29. J. 73. (1992b). J. (1989). R. Application of Hard Convex Body and Hard Sphere Equation of State to the Critical Properties of Binary Mixtures. Sadus. R. (1994) Calculating Critical Transitions of Fluid Mixtures: Theory Vs. Sadus.. (1996). Sadus. 12791284. 1236312366. 169. (1999b). Macromol. J. 24542457. 19851992. and Moshfeghian. Equations of State for HardSphere Chains. J. Phys. R. 39.. 40. Sadus. A Simplified Thermodynamic Perturbation TheoryDimer Equation of State for Mixtures of HardSphere Chains. 72127216.. Simple Equation of State for HardSphere Chains. Phys. R. A. M. J. 45.. Influence of Combining Rules and Molecular Shape on the High Pressure Phase Equilibria of Binary Fluid Mixtures. (1992a). 13761403. J. Sadus R.. Elsevier. Sadus R. J. Amsterdam. (1988). Chem. High Pressure Phase Behaviour of Multicomponent Fluid Mixtures.. R. 99. Fluid Phase Equilib. Chem. and Svejda. 37873790. Saghafi. Phys. Evaluation of VaporLiquid Equilibrium of Ethane Binary Systems Using UNIQUACBased Mixing Rules and its Extension to Multicomponent System. R. L. (1999a). (1995). Young. Sadus. Experiment. J.. Phys.. Sadus. Chem.. J. (2000). R. (1993).Sadus. 97. C. Phys. Novel Critical Transitions in Ternary Fluids Mixtures. Attractive Forces in Binary Mixtures: a Priori Prediction of the GasLiquid Critical Properties of Binary Mixtures. AIChE J. 93. Fluid Phase Equilib. R. 3147. 109 . 8999. J. Mol. 96.
Scott. (1999). New York. M. AIChE J. Marcel Dekker. Phys. Eng. Chem.. S. New York. G. Thesis. Sci. L.. D. The Thermodynamics of Critical Phenomena in Fluid Mixtures. Equilibrium Constants From a Modified RedlichKwong Equation of State.. 49. Equations of State for Fluids and Fluid Mixtures. Eyring. Kayser. Sci. L. P. W. 19871992.. C. 27. Chem. (1956). Bunsenges . (1971). S. 3 Ed. 296308. V. 40. 76. rd Sandler. R.D.Sandler. J. R. SolidFluid Equilibria in Natural Gas Systems. Elsevier. J. Smith. V. J. Henderson. and Young. L. Vol. Y. Shah. Sulfur Hexafluoride + Siloxane and Perfluoropropane + Siloxane Mixtures. Song.. Peters. R. Chemical and Engineering Thermodynamics. and Cochran. C. R. Scott.. 152159. S.. Upper Critical Solution Temperatures of PerfluoronHexane + Alkane. Scott. Lambert. J. 27652775. and Prausnitz. (1972). M. Waterson. Chem. Phys. Generalized Quartic Equation of State for Pure Nonpolar Fluids. Ph. S. Soave. (1972). 11971203. 25. and Jost. J.. An Advanced Treatise. In Physical Chemistry. I. H.. 193. L. R.. J. Ber. Liquid Liquid Phase Diagrams for Binary Polymer Solutions from a Pertubed HardSphereChain Equation of State. (1994). Georgia Institute of Technology. H.. Corresponding State Treatment of Nonelectrolyte Solutions.. (Eds. and White Jr. Eng. Bienkowski. I.. Academic Press. (1995). (1994). 8A: Liquid State. V.). D. 33. M. Aust. Sengers. F. Chem.) (2000). J. D. S. (1994). (1980). Models for Thermodynamic and Phase Equilibria Calculations. Semmens.. John Wiley & Sons. Amsterdam. (Eds. 110 . Chem. H. New York.
E. Gasliquid Critical Temperatures of Mixtures Containing Electron Donors. Chem. 30. C. Eng. Trebble. Chem. 474479. Thermodyn.. and Bishnoi.. (1998). H. 111 . D. C. Chem. (1980c). L. Comparison of the Values of Interaction Parameter from Upper Critical Solution Temperatures of Acetone + Alkanes with Values Obtained from GasLiquid Critical Temperatures. 365370. S. A. J. 181189. C. Thermodyn. Coon. Toczylkin. P. Development of a New FourParameter Cubic Equation of State. R. 33. (1969). 355364. Bunsenges. M.Spear. Phys.. R. L.. C. and Young. (1980a). Part 1. J. Thermodyn. C. L. Chem. and Young. J. 672680. F. Fluid Phase Equilib. and Bluck. 323340. S. J. (1963). (1983). A ZeroPressure Cubic Equation of State Mixing Rule for Predicting High Pressure Phase Equilibria Using Infinite Dilution Activity Coefficients at Low Temperature. S. (1986). 87. L. PRSV: An Improved PengRobinson Equation of State for Pure Compounds and Mixtures. (1980b). 64. Stryjek. Amine Mixtures.. Part 2. Chem. 35. K. Chem. 28. Treatme nt of Molecular Shape in Terms of the Boublik Nezbeda Equation. 121. 8. 12. Equation of State for Hard Spheres. Upper Critical Solution Temperatures of Hydrocarbon + Fluorocarbon Mixtures: Mixtures with Specific Interactions. Svejda. R. Chem. Jr. R. Ind. 12. 39.. (1987). Aust. L. L. Ber. and Vera.. Ether Mixtures. H. Chem. L. Critical States of Mixtures and Equations of State. 767770. Can. L. Robinson. S. and Kohler. R.... and Young. Toczylkin. Phys.. Fluid Phase Equilib. A Generalized van der Waals Equation of State: I. J. J. Fundam. L. and Young. Twu. E. C. 150151. J. 465469. J. Thiele. Eng. GasLiquid Critical Temperatures of Mixtures Containing Electron Donors. P. Toczylkin. and Chao. Toczylkin.. (1977).
M. Mol. Phys. D. and Twu et al. Roy. Critical Lines and Phase Equilibria in Binary van der Waals Mixtures. 24. P. W. (1996).. 87. and Scott. Comparison of Equations of State for Chain Molecules. 31.. Fluid Phase Equilib. Qu. J. G. (1987). 246257. S. Y. ClosedLoop LiquidLiquid Equilibria and the Global Phase Behaviour of Binary Mixtures Involving HardSphere + van der Waals Interactions. Phys. P. MHV2. Wu. 116. Chem. C. 93. Voros. and Donohue. Philos... Thermodynamic Perturbation Theory of Polymerization. R. 298. S. E. 129. L. D. Aust. Fluid Phase Equilib. 43554360. and de Swaan Arons. Thermodynamics of Quadrupolar Molecules: The PerturbedAnisotropicChain Theory. Comprehensive Comparison and Evaluation of the Mixing Rules of WS. 91. P. J. H. S. C. Fundam. L. (2000). Phys. W. Van Pelt. VapourLiquid Equilibria in Nonpolar/Weakly Polar Systems with Different Types of Mixing Rules. (1978). 98. J. (1980). M. Fluid Phase Equilib. C. Soc. (1989). (1992)... M. 112 . Wang. and Tassios. Vilmalchand. J. 495540. 957962. 46. 74. D. N. London A. (1985). 73237331. P. Chem. Eng. 488494. and Coon. Wei. Wang.. L. (1993). Peters. 715723. and Sadus. D. and Young. Twu. 169195.. Wertheim.. J. R. and Donohue.Van Konynenburg. Equations of State for the Calculation of FluidPhase Equilibria. (2000). Chem. Chem.. Ind.. J.. H. G. Application of the SimplifiedPerturbedHardChain Theory for Pure Components Near the Critical Point. GasLiquid Critical Temperatures of Mixtures Containing an OrganoSilicon Compound. Y. Waterson. J. AIChE J. Trans. Vilmalchand. R. A... J. 6783. and Sadus. J.
D.Wong.. 102 . and Olesik.. M. . A. A. Fluid Phase Equilib. Fluid Phase Equilib . Equation of State Mixing Rule for Nonideal Mixtures Using Available Activity Coefficent Model Parameters and that Allows Extrapolation over Large Ranges of Temperature and Pressure. and Sandler. and Kraska.. (1992). J.. ClosedLoops of LiquidLiquid Immiscibility in Binary Mixtures of Equal Sized Molecules Predicted with a Simple Theoretical Equation of State. 160. (1992). M. 277286. S. Chem. Yelash. and S engers. 261284. (2000). (1999).. (1999).. 213223. Abdulkadirova. Anisimov. A. Fluid Phase Equilib. Zhao. Bunsenges. K. T. S. and Sengers. and Sandler. D. J. 113. Q. Phys. Res. Carbon Dioxide and Nitrogen in Ethylene Glycol at Elevated Temperatures and Pressures. S. Chem. 154. Wyczalkowska. Phase Diagram Studies of MethanolCHF3 and MethanolH2OCHF3 Mixtures. H. L. I. M. 49855002. 523535. Global Crossover Equation of State of a van der Waals Fluid. R. D. Ber. Chem. Phys. J. 20332039. 671680. I. 31. J. Anisimov. and Gao. A Theoretically Correct Mixing Rule for Cubic Equations of State. AIChE J. S.. Thermodynamic Properties of H2 O and D2 O in the Critical Region. K. Orbey. K.. 38. Ind. (1999). A. V. D. S. S. H. H.... Ma. V. W. Zheng. V. 113 . Wyczalkowska. Wong. 155 . T. V. (1998). Wei. Eng. Solubility Study of Methane.
114 .
This action might not be possible to undo. Are you sure you want to continue?
Use one of your book credits to continue reading from where you left off, or restart the preview.