An equation of state is a thermodynamic equation, which provides a mathematical relationship between two or more state functions associated with the matter, such as its temperature, pressure, volume, or internal energy. One of the simplest equations of state for this purpose is the ideal gas law, which is roughly accurate for gases at low pressures and high temperatures. However, this equation becomes increasingly inaccurate at higher pressures and lower temperatures, and fails to predict condensation from a gas to a liquid.
Introduced in 1949, the Redlich-Kwong equation of state [87] was a considerable improvement over other equations of that time. It is an analytic cubic equation of state and is still of interest primarily due to its relatively simple form. The original form is
(7–88) |
| where | |
| P = absolute pressure (Pa) | |
| R = universal gas constant | |
| V = specific molar volume ( | |
| T = temperature (K) | |
|
| |
|
|
Many investigators have attempted to improve the accuracy of the Redlich-Kwong equation. ANSYS Fluent has adopted the original form of the Redlich-Kwong equation, as well as the following modified forms:
The Soave-Redlich-Kwong [112] equation is a three-parameter equation of state, which can be applied
for vapor, supercritical, and liquid property predictions. It has
found wide acceptance mainly in the oil and gas industry and requires
knowledge of the critical temperature, critical pressure, and acentric
factor. It was developed by replacing the Redlich-Kwong attractive
coefficient defined as with a two-parameter form
, where
is the acentric factor.
The Peng-Robinson equation [82] is a three-parameter equation of state, also requiring the critical temperature, critical pressure and acentric factor parameters. It is thought to perform as well as the Soave-Redlich-Kwong equation, with an advantage in the prediction of the liquid densities.
The Aungier-Redlich-Kwong [8] equation provides improved predictions for vapor and supercritical fluids near the critical point, as well as for materials with a negative value of the acentric factor. The Aungier modified form is a four parameter equation and requires the critical specific volume in addition to the critical temperature, critical pressure and acentric factor.
The following limitations exist in ANSYS Fluent for all cubic equation of state models:
Pressure-inlets, mass flow-inlets, and pressure-outlets are the only inflow and outflow boundaries available for use with the real gas models.
Non-reflecting boundary conditions are not compatible with the real gas models when using the density-based solver. If your model requires NRBC and a real gas model, you must use the pressure-based solver.
The cubic equation of state real gas models are compatible with the Eulerian multiphase models.
Note: Note the following restrictions:
Only one phase can be a real gas.
The cubic equation of state models are compatible
with the Lagrangian Dispersed Phase Models. If you are modeling droplet
or multicomponent particles, note that the current formulation does
not take into account the near-critical point phenomena, which means
that accurate results can be obtained for droplet temperatures below . See Using the Cubic Equation of State Models with the Lagrangian
Dispersed Phase Models for
more details.
The real gas models are not available with the premixed, inert, and composition PDF transport combustion models.
The cubic equation of state real gas models can be used with the following species models:
Species transport. Chemical reactions can be modeled with the finite rate and eddy dissipation models. Note that the Dimension Reduction model is not available with the real gas models.
Non-premixed model and partially-premixed model. Note that the Compressibility Effects option must be enabled in the Species Model dialog box in order for the real-gas models to be available. In addition the following restrictions apply when the non-premixed model is used together with a cubic equation of state real-gas model:
The empirical stream options cannot be used, as for the empirical species the critical properties cannot be defined.
Condensed species such as h2o<l> and c<s> are not supported and should be excluded from the PDF table, so ensure that you add all condensed species for your system in the excluded species list prior to the PDF table generation.
The laminar flame speeds for the partially-premixed model are assumed to be the same as for ideal-gases.
When the cubic equation of state models are applied for subcritical conditions near or under the phase dome, the two-phase flow is not modeled and either vapor or liquid state must be selected. Subcritical and supercritical states can co-exist in the same simulation.
The general form of pressure P for the cubic equation of state models is written as [8] :
(7–89) |
| where | |
|
P = absolute pressure (Pa) | |
|
V = specific molar volume ( | |
|
T = temperature (K) | |
|
R = universal gas constant |
The coefficients ,
,
,
, and
are
given for each equation of state as functions of the critical temperature
, critical pressure
, acentric factor
and critical specific volume
. Note that the attractive coefficient
also has a temperature
dependence, which varies for each equation of state model, and is
commonly written as
.
Redlich-Kwong Equation [92]:
(7–90) |
(7–91) |
(7–92) |
The parameter is set equal to
, while
and
are set to 0. The Redlich-Kwong
equation is the simplest of the cubic equations of state in ANSYS Fluent and
requires two parameters only, the critical temperature
and the critical pressure
.
Soave-Redlich-Kwong Equation [112]:
(7–93) |
(7–94) |
and
are given by Equation 7–91 and Equation 7–92 respectively. As in the original Redlich-Kwong
equation the parameter
is set equal to
, while
and
are set to 0. The Soave-Redlich-Kwong
requires three parameters, the critical temperature
, the critical pressure
, and the acentric factor
.
Peng-Robinson Equation [82]:
(7–95) |
(7–96) |
The function is given by Equation 7–93,
with n provided in Equation 7–97 as follows:
(7–97) |
In the Peng-Robinson equation is set equal to
,
is equal to
, and
is set to 0.
Similar to the Soave-Redlich-Kwong equation, the Peng-Robinson
equation is a three-parameter equation and requires the critical temperature , the critical pressure
, and the acentric factor
.
Aungier-Redlich-Kwong Equation [8]:
(7–98) |
(7–99) |
and
are given by Equation 7–91 and Equation 7–92, respectively. As in the original Redlich-Kwong
equation, the parameter
is set equal to
and
is set to 0. Parameter
is given by:
(7–100) |
The Aungier-Redlich-Kwong equation requires four parameters,
namely the critical temperature , the critical pressure
, the critical specific volume
, and the acentric factor
.
Enthalpy, entropy, and specific heat are computed in terms of
the relevant ideal gas properties and the departure functions. The
departure function of
any conceptual property
is defined as [87]
(7–101) |
where is the value of the property as computed
from the ideal gas relations. The departure function
can be derived from basic thermodynamic relations
and the equation of state.
Following the above definition, the enthalpy for the equations of state
models is given by the following equations [8]:
(7–102) |
(7–103) |
(7–104) |
| where, | |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
See Equation 7–89 for a description of other coefficients.
Similarly, the departure internal energy can
be shown to be
(7–105) |
| where | |
|
| |
|
| |
|
| |
|
| |
|
|
The specific heat can be computed from the ideal
specific heat
and the departure specific
heat at constant volume
as follows:
(7–106) |
(7–107) |
| where | |
|
| |
|
| |
|
| |
|
| |
|
|
In Equation 7–107 is computed
by differentiating the equation of departure internal energy (Equation 7–105) with respect to
, and the partial derivatives
of the specific volume are computed by differentiating Equation 7–89 appropriately.
The entropy is computed by
(7–108) |
(7–109) |
| where | |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
is given by Equation 7–104 and
,
, and
are computed using Equation 7–90 – Equation 7–100, depending on the equation of state model. Note that the pressure
term in Equation 7–108 cancels out as both
and
are evaluated at the reference pressure.
Equations describing real-gas properties require the knowledge
of the critical constants for pure components and mixtures. These
consist of the critical temperature (), critical
pressure (
), critical specific volume
(
), and the acentric factor (
).
Several critical constants for fluid materials in the ANSYS Fluent property
database propdb.scm have been compiled from
a variety of sources available in the open literature [87], [78], [79], [107], [108], [113], [7].
For those fluid materials, for which the critical properties
have not been found in the open literature, these have been estimated
using the commercially available software CRANIUM by Molecular Knowledge
Systems Inc. [1] : http://www.molknow.com/Cranium/cranium.htm
Critical property values for many hydrocarbon and nitrogenous radical species have been obtained from Tang and Brezinsky [118]. Where the critical properties for the radicals were not available in the literature, these were estimated using a modification of the Joback method [87]. This assumes that the radical site constitutes a distinct group with zero group contribution and utilizes the group contribution values for stable species.
The critical properties of coal volatiles have been estimated
assuming that the volatiles can be approximated by a mixture of CO, ,
,
, and
in such
a way, that the atom composition and the net calorific value of the
volatiles is similar to that of the assumed mixture. The critical
properties of the lignite and biomass volatiles have been assumed
equal to those of formaldehyde. The critical properties of diesel
and jet-a fuels have been set equal to those of decane. The critical
properties of kerosene have been set equal to those of dodecane.
For the computation of properties in real-gas mixtures, ANSYS Fluent follows
the so-called pseudocritical method. According to this method, the
behavior and properties of a real gas mixture will be the same as
that of a pure component, to which appropriate critical constants
are assigned. These mixture critical constants are functions of the
mixture composition and pure component critical properties, and are
sometimes called pseudocritical constants, because their values are
generally expected to be different from the true mixture critical
constants that may be determined experimentally. However for computational
purposes they are the appropriate critical constant values for the
mixture. According to the pseudocritical method, ANSYS Fluent applies Equation 7–89– Equation 7–109 also for mixtures, where the critical
temperature , critical pressure
, critical
specific volume
, and acentric factor
, are replaced by the
corresponding mixture critical constants, critical temperature
, critical pressure
,
critical specific volume
, and acentric
factor
.
The following options are available in ANSYS Fluent for the calculation of the mixture pseudocritical constants:
The simplest rule for computing the pseudocritical
constants for a real
gas mixture is the mole fraction average [87]. This method is recommended for mixtures where
the pure component critical properties for all components are not
very different:
(7–110) |
where
= mixture
pseudocritical constant (temperature, pressure, specific volume or
acentric factor)
= critical
constant of component i (temperature, pressure, specific volume or
acentric factor)
= mole fraction of component
i
= number of components in mixture
An alternative approach is based on the one-fluid
van der Waals mixing rules as expressed in [92]. According to this approach, in order to apply the equation of
state models to mixtures, the coefficients and
in Equation 7–89 are replaced by composition-dependent expressions as follows:
(7–111) |
(7–112) |
where
= mole fraction of species
i
With the appropriate expressions from Equation 7–90– Equation 7–100 for each equation of state, and assuming a mixture acentric factor for the evaluation
of parameter
in Equation 7–94, Equation 7–97, and Equation 7–99, the mixing rules Equation 7–111 and Equation 7–112 can be rearranged to yield direct
expressions of the mixture critical properties as functions of the
mole fractions and the component critical properties [8].
The resulting expressions for
the mixture critical specific temperature are as follows:
Soave-Redlich-Kwong and Peng-Robinson models
(7–113) |
Redlich-Kwong model and Aungier-Redlich-Kwong model:
(7–114) |
where for the Aungier-Redlich-Kwong model is obtained from Equation 7–99 as function of the mixture acentric factor
. For the Redlich-Kwong
model
.
The mixture critical specific
pressure is computed as:
(7–115) |
The mixture critical specific volume is
computed as:
(7–116) |
The notation for Equation 7–113 to Equation 7–116 is
=
mixture pseudocritical temperature (K)
= mixture pseudocritical pressure (Pa)
= mixture
pseudocritical molar volume (
/kmol)
= critical
temperature for component i (K)
=
critical pressure for component i (Pa)
= critical molar volume for component i (
/kmol)
= mole fraction
for component i
For single or multicomponent flows, you will enable the cubic equation of state real gas models by selecting real-gas-soave-redlich-kwong, real-gas-peng-robinson, real-gas-aungier-redlich-kwong, or real-gas-redlich-kwong from the Density drop-down list in the Create/Edit Materials dialog box.
Setup →
Materials → Create/Edit...
The required inputs for the cubic equation of state real gas models for single component flow and mixtures are described below.
Single Component Flow
When any of the cubic equation of state models is enabled, enter the following material properties in the dialog box:
ideal specific heat
molecular weight
standard state entropy
reference temperature
critical temperature
critical pressure
critical specific volume
acentric factor
Important: Your inputs for the specific heat in the Materials dialog box will now be used to compute the ideal property functions ,
, and
in Equation 7–102, Equation 7–106, and Equation 7–108, respectively. In ANSYS Fluent the
departure properties will be computed and added to the ideal part,
to yield the real gas specific heat, enthalpy, and entropy.
Mixtures
When one of the cubic equation of state models is selected from the Density drop-down list, specify the following material properties for the mixture material in the dialog box:
ideal specific heat
critical temperature
critical pressure
critical specific volume
acentric factor
You also need to enter the following material properties for each of the mixture components in the dialog box:
molecular weight
standard state entropy
reference temperature
When you are modeling a real-gas mixture, the following methods are available for the mixture critical constants:
constant: defines a constant critical temperature, critical pressure, critical specific volume, or acentric factor for the mixture material.
mole-weighted-mixing-law: applies Equation 7–110 for critical temperature, critical pressure, critical specific volume, or acentric factor for the mixture material.
one-fluid-van-der-waals-mixing-law applies Equation 7–113 and Equation 7–114 for the mixture critical temperature (depending on the real-gas model), Equation 7–115 for the mixture critical pressure, and Equation 7–116 for the mixture molar volume.
Ensure to click the button so that all the above mentioned properties are uploaded and are visible in the interface.
If you have selected mixing-law for the mixture ideal specific heat you will also need to enter the ideal specific heat values for the individual mixture components. If you have not selected constant as the option for any of the critical properties, you must enter the corresponding pure component critical properties for the mixture components.
If the operating conditions in your model are in the subcritical
regime, select Vapor or Liquid as the Real Gas Phase in the Operating
Conditions dialog box (Figure 7.40: The Operating Conditions for a Real Gas Phase). Alternatively, you can use
the define/operating-conditions/set-phase text command. Note that the default is Vapor and therefore Vapor is assumed if no changes
are made to the Real Gas Phase settings. If the
operating conditions in your model are entirely in the supercritical
regime, this setting will have no effect.
In addition, you may specify the phase for a specific fluid zone by typing the following text command at the ANSYS Fluent console prompt:
>define boundary-conditions modify-zones change-zone-phaseSelect a name/id from fluid zones list [(liquid-zone fluid-1)]liquid-zoneSet zone real-gas phase: -1:use global setting 0:liquid 1:vapor [-1]0
The flow modeling of real-gas flow is more complex and challenging than simple ideal-gas flow. Therefore, the solution might converge at a slower rate with real-gas flow than when running ideal-gas flow. It is recommended that you first attempt to converge your solution using first-order discretization then switch to second-order discretizations and re-iterate to convergence.
Special considerations apply if the operating regime in your
model is fully or partly subcritical, where the physical state may
be vapor, liquid, or a vapor/liquid mixture. In the cubic EOS real-gas
model, the phase state is determined by the selection of the cubic
root to calculate the molar-volume. This is illustrated in Figure 7.41: The PV Diagram for the Cubic Equation of State Real Gas Model, which shows a pressure versus
molar volume (PV) diagram, with a subcritical isotherm ABFED at temperature
T calculated from a cubic EOS. Curve C1CC2 represents the critical
isotherm and curve ACD the saturation dome. Points A, F, and D represent
the three roots of the EOS at the saturation pressure Ps, where A
corresponds to the molar volume of saturated liquid, D to the molar
volume of saturated vapor and point F does not have a physical significance.
Points A1 and D1 correspond to the liquid and vapor molar volumes
at pressure P1, which is lower than the saturation pressure. The liquid
and vapor molar volumes at pressure P2, which is higher than the saturation
pressure, are marked as A2 and D2 respectively. Points B and E, where
the partial derivatives of pressure with respect to volume are
0, are called spinodal points. The loci of these points for all subcritical
temperatures, the spinodal curve, is shown with the dashed curve BCE
and sets the boundary beyond which the equation of state is no longer
valid, because the local derivative of pressure with respect to volume
becomes positive. State points inside the dome, up to the spinodal
curve, are called “metastable” because normally they
only exist temporarily in small local regions until phase change occurs.
It is important to realize that the current implementation of the cubic equations of state real gas model does not determine the saturation conditions and does not model the two-phase flow where liquid and vapor coexist. In the subcritical regime and for conditions where the cubic EOS has three roots, the phase state selection is controlled by your input of Vapor or Liquid for Real Gas Phase in the Operating Conditions dialog box (Figure 7.40: The Operating Conditions for a Real Gas Phase). The default setting is Vapor, which corresponds to the state points to the right of the vapor spinodal. With reference to Figure 7.41: The PV Diagram for the Cubic Equation of State Real Gas Model, for the operating point at temperature T and pressure P1, if the Real Gas Phase is set to Vapor, the molar volume at point D1, which corresponds to superheated vapor, will be selected in the calculation. On the other hand, if you have selected Liquid, the molar volume of point A1 will be taken, and the corresponding state will be metastable superheated liquid. Similarly for pressure P2, which is higher than the saturation pressure Ps, if you select Liquid for Real Gas Phase, the liquid molar volume at A2 will be computed, and if you select Vapor the molar volume will correspond to the subcooled vapor state point D2.
In case the calculations fall inside the saturation dome, the solver will not limit the solution, but if conditions are predicted that extend beyond the vapor spinodal curve a warning will be issued:
temperature is below the spinodal point in 12 cells on zone 3.DE27233
Cubic equations of state real gas models are not available for
two-phase flows but can be applied to conditions of supercritical
pressure and subcritical temperature
, where the fluid is in the liquid state,
and the supercritical liquid co-exists with gas and supercritical
fluid in the same simulation. In those cases, phase change may take
place, without any of the flow conditions falling inside the saturation
dome.
For multicomponent simulations, when Diffusion Energy
Source is enabled in the Species Model dialog box, the species energy diffusion is by default suppressed
in the supercritical liquid regime and across the gas-liquid boundary.
The diffusion energy source can be included in the liquid regime using
the define/models/species/liquid-energy-diffusion? text command.
If your simulation contains a liquid stream, the appropriate modeling approach in the various operating condition regimes is as follows:
For a liquid phase does not exist.
In the region , you can define the flow streams directly in the boundary conditions by setting the appropriate pressure and temperature and the properties will be computed directly by the cubic EOS.
For or
, a stream can exist in liquid phase when its temperature
. If you would like to model phase change in this regime, the following recommendations apply:
The DPM droplet and multicomponent models are adequate
and recommended for the conditions away from the critical point. This
regime can be defined as , where the liquid physical properties can be assumed
independent of pressure.
The region is characterized by near-critical-point
phenomena, such as strong liquid density and specific heat dependence
on both temperature and pressure. The DPM model can also be used in
this regime, but you should be cautious, as it will not take into
consideration the near-critical-point behavior. In addition, the applicability
of the evaporation and boiling rate equations is questionable in this
regime.
If the pressure in your model is above the critical point, it is recommended to use the compressible-liquid method for Density of the droplet material.
The temperature gives an indication of the maximum droplet temperature for applicability of the DPM models for each droplet material. These temperature limits are listed in Table 7.2: Temperature Limits for Droplet Materials in ANSYS Fluent Database prodb.scm for many of the droplet materials in ANSYS Fluent's
propdb.scm materials database. For supercritical pressures (), the DPM models can still be used provided the droplet temperatures remain below the limit
.
Table 7.2: Temperature Limits for Droplet Materials in ANSYS Fluent Database prodb.scm
| Material | Tc (K) | Pc (MPa) | Normal BP (K) | Tlim (K) |
| Argon | 150.86 | 4.89 | 87.30 | 135.77 |
| Benzene | 562.00 | 4.89 | 353.00 | 505.80 |
| Helium | 5.30 | 0.23 | 4.20 | 4.77 |
| Hydrogen | 32.98 | 1.30 | 20.40 | 29.68 |
| Methyl-alcohol | 512.00 | 8.10 | 338.00 | 460.80 |
| Heptanes | 540.00 | 2.74 | 371.00 | 486.80 |
| Hexane | 507.00 | 3.02 | 342.00 | 456.30 |
| Octane | 569.00 | 2.49 | 399.00 | 512.10 |
| Pentane | 470.00 | 3.37 | 309.00 | 423.00 |
| Nitrogen | 126.00 | 3.40 | 77.40 | 113.40 |
| Oxygen | 154.00 | 5.04 | 90.20 | 138.60 |
| Toluene | 592.00 | 4.11 | 384.00 | 532.80 |
| Water | 647.00 | 22.00 | 373.00 | 582.30 |
For high pressure simulations the boiling point will be different from the normal boiling point, and for varying pressure applications the boiling point will vary with the droplet location in the domain.
When a cubic equation of state real gas model is enabled in a simulation that includes evaporating droplet particles, the boiling point is calculated from the vapor pressure data directly, as the temperature where the saturation vapor pressure equals the domain pressure. In addition, the latent heat of the evaporating or boiling droplet will vary with the droplet temperature.
The latent heat at temperature is given by
(7–117) |
where is the normal boiling point and
is the latent heat
at the normal boiling point.
In the Create/Edit Materials dialog box you must enter the Normal Boiling Point (NBP) and the Latent Heat at NBP for the droplet-particle Material Type. These inputs will be used for calculating the Latent Heat at the reference temperature (see Equation 16–441) and the latent heat in the droplet energy balance during vaporization and boiling according to Equation 7–117.
Important: The constant property option is disabled for the Saturation Vapor Pressure property when a real gas model is used in the simulation. Also, ensure to enter the appropriate droplet saturation vapor pressure data to cover the complete pressure/temperature range in your model.
Finally when a cubic equation of state real-gas model is enabled in a simulation with droplet models, the condition for switching from the vaporization to the boiling law will be
(7–118) |
where is the saturation vapor pressure and P is the domain pressure. If
while in the boiling law, the model will switch back to the vaporization law. Under supercritical pressure conditions (
), the vaporization models are applicable at droplet temperatures below the critical point, and the switching condition from vaporization to boiling
is never met because the vapor pressure curve is defined only up to the critical point.
All postprocessing functions properly report and display the current thermodynamic and transport properties of the selected real gas model. The thermodynamic and transport properties controlled by the cubic equations of state real gas model include the following:
density
enthalpy
entropy
sound speed
specific heat
any quantities that are derived from the properties listed above (for example, total quantities, ratio of specific heats)
In addition to the properties listed above, you can also report
compressibility factor
reduced temperature
reduced pressure
spinodal temperature
subcritical condition
If you are modeling a real-gas mixture you can report the composition-dependent mixture critical properties
critical temperature
critical pressure
critical specific volume
acentric factor