BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 3 n. 3 | p.

093-102 | 2009 | ISSN 1982-0593

CFD SIMULATION OF MULTIPHASE FLOW IN A SIEVE TRAY OF A

DISTILLATION COLUMN
a a a a
Teleken, J. G.; Werle, L. O. 1; Marangoni, C.; Quadri, M. B.; Machado, R. A. F.
a
Federal University of Santa Catarina- Department of Chemical Engineering - Laboratory of Control of Processes -
Technological Center

ABSTRACT
Many studies on distillation columns are based on macroscopic models of mass and energy conservation.
In this work, we investigate the complex hydrodynamics of sieve trays in pilot plant distillation columns,
based on a distributed control system with heating action at intermediate points, by means of electrical
resistance heaters arranged on the surface of sieve trays (Marangoni and Machado, 2007; Werle, 2007),
using computational fluid dynamics. The main objective was to evaluate the influence of the electrical
resistance heaters on the hydrodynamics of the sieve trays. A three-dimensional mathematical
homogeneous biphasic model was implemented in the commercial code of computational fluid dynamics
(CFD), CFX 11 (AEA Technology) for numerical experimentation studies. The results show the influence of
electrical resistance heaters placed on the sieve trays on the flow patterns, although the hydrodynamics is
not affected as a whole. Consequently, this helped to provide enhanced mixing and homogenization in the
region with active bubbles, and thus the use of electrical resistance heaters could be applied in a
distributed control system approach.

KEYWORDS
distillation columns; fluid dynamics model; CFX

1
To whom all correspondence should be addressed.
Address: Federal University of Santa Catarina- Department of Chemical Engineering - Laboratory of Control of Processes -
Technological Center - University Campus - Trindade. P.O. Box 476. ZIP Code:. Florianpolis - SC. Brazil
CEP: 88040-970 | Telephone / fax number: +55 48 3721 9554 | E-mail: [email protected]

93
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 3 n. 3 | p. 093-102 | 2009 | ISSN 1982-0593

1. INTRODUCTION direction of gas flow along the height of the

dispersion to simulate the two-phase flow
The petroleum industry is one of the greatest behavior, and only the hydrodynamics of the liquid
driving sources of technological advances in the flow was obtained. The inter-phase momentum
world. This is due to the high degree of exchange (drag) coefficient was required to model
competitiveness associated with petroleum, the hydrodynamics of multiphase flow on sieve
leading to the exhaustive search for new trays. Fischer and Quarini (1998) attempted to
technologies that enable greater efficiency in the describe the three-dimensional transient gas-liquid
related processes. hydrodynamics, by assuming a constant drag
coefficient of 0.44, which was appropriate for
The development of sophisticated chemical uniform bubble flow. This drag coefficient was not
processes is associated with this globalized and appropriate for description of the hydrodynamics
competitive activity, with the application of of sieve trays operating in either the froth or spray
mathematical models that enable a better regimes. Krishna et al. (1999) and van Batten and
description of their real behavior. The process most Krishna (2000) described the hydrodynamics of
widely used in this sector is distillation and, in this sieve trays by estimating a new drag correlation
context, detailed knowledge of the mechanisms coefficient for a swarm of large bubbles on the
that occur is very important, particularly with basis of the correlation of Bennett et al. (1983) for
regard to the sophistication of the equipment and the liquid hold-up. Because this correlation over-
processes (Li et al., 2009). predicted the liquid hold-up fraction in the froth
regime, Gesit et al. (2003) used the liquid hold-up
Valve trays are widely used as phase-contact correlation of Colwell (1979), which worked well in
devices in distillation columns. The description of the froth regime, to predict the flow patterns and
the hydrodynamics of valve trays is of great hydraulics of commercial-scale sieve trays (CFD
importance in industrial practice, since the simulation of hydrodynamics of valve tray). There
separation efficiency and overall tray performance are several studies in the literature using fluid
is predicted for a given set of operating conditions, dynamics modeling for distillation columns, most
tray geometries and system properties. The notably the research by Li et al. (2009), Nikou and
hydrodynamics of valve trays has been reported in Ehsani, (2008), and Noriler et al. (2007), who, in
many publications (Kister, 1992; Szulczewska et most cases, investigated the behavior of the speed
al., 2003). However, the correlations reported for vectors and the distribution of the kinetic energy of
tray hydrodynamics have been largely empirical. liquid-gas flows in sieve trays. The model was
solved using the finite volume method with
In recent years, there has been considerable variables located in a system of generalized
academic and industrial interest in the use of coordinates (Maliska, 2004).
computational fluid dynamics (CFD) to model two-
phase flows in some chemical engineering Although CFD is becoming a powerful research
equipment. The volume-of-fluid (VOF) technique and design tool in chemical engineering there are
can be used for prior determination of multiphase no reports of simulating the hydrodynamics of
flow over structure packing. Szulczewska et al. valve trays using CFD. The main difficulty
(2003) simulated gas-liquid counter-current flow associated with this is that the valve floats as the
over a plate-type structured packing. Gu et al. valve hole gas velocity changes. When the velocity
(2004) developed a two-phase flow CFD model of the gas passing through the valve hole is higher
using the VOF method to predict the than a certain value (the critical valve hole
hydrodynamics of falling film flow over structured velocity), the valve is fully open. This means that
packing. Ataki and Bart (2006) simulated the liquid the valve lifts to the maximum extent and does not
flow structure for a structured packing element float. Under this condition, CFD can be used to
and liquid redistribution at the node of the simulate the hydrodynamics of valve trays. In this
structure packing with the VOF model. There have paper, a three-dimensional transient CFD model
also been many attempts to simulate sieve tray was developed within the two-phase Euler
hydrodynamics using CFD. Yu et al. (1998) and Liu framework for the hydrodynamics of a cylindrical
et al. (2000) neglected the variations in the fully-open valve tray. Simulations were carried out
varying the superficial gas velocity, liquid weir

94
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 3 n. 3 | p. 093-102 | 2009 | ISSN 1982-0593

loads and weir heights when the valves are fully 2.2 Momentum Equation
open. The objectives were: to propose the
development, implementation and application of a Equation 3 expresses the overall momentum
microscopic model for momentum conservation equation.
subjected to the turbulent flow of the vapor phase;
to represent the fluid dynamics of the vapor-liquid
flow in a perforated plate distillation pilot plant, (fkk vk ) + (fkk vk vk ) = fk Pk +
t (3)
with validation through correlation with the
experimental data provided in Bennett et al. [
+ fk k (v k + v kT ) + Mg.l + fkk g ]
(1983); and, finally, to examine the influence of
electrical resistance heaters placed on the tray where k and g represent the molecular viscosity
surfaces to control the distributed heating of a for the kth phase and the gravity vector,
distillation pilot plant on the fluid flow dynamics. respectively. P k is the pressure field, having the
same value for the gas end and the liquid phase.
M g,l represents the momentum transfer between
the gas and the liquid phases, and the additional
2. MATHEMATICAL MODEL momentum flux due to the velocity fluctuation
(turbulence) was incorporated into the diffusion
The model considers the gas and liquid flows in term.
a Eulerian-Eulerian framework, where the phases
are treated with transport equations. The To solve Equations 1-3, an additional equation
equations used were: mass continuity and relating the interphase momentum transfer (M g,l )
momentum equations. To solve these, it was and additional momentum fluxes is required, and
necessary to add and use the equation of this is provided by Equation 4.
momentum flux (Liu et al., 2000). It was considered
3 fgl
that the fluctuations (turbulence) reflect the Mg,l = CD v g v l (v g v l ) (4)
formation and dispersion of small swarms of 4 dg
bubbles, and that the Reynolds stresses can be
linearly related to the mean velocity gradients where dg is the bubble diameter and CD is the drag
(eddy viscosity hypothesis), as in the case of the coefficient.
relationship between the stress and strain tensors
in laminar Newtonian flow. The standard k-
turbulence model for multiphase flow was 2.3 Turbulence Equations
assumed (Nikou and Ehsani, 2008).
When applying the time average procedure in
the Navier Stokes equations, an extra term appears
2.1 Mass Continuity Equations due to the turbulent fluctuation of the velocities,
which needs to be represented by a constitutive
Equation 1 expresses the overall mass equation. This term is known as the Reynolds flux.
continuity equation. We consider that the fluctuations (turbulence)
reflects the formation and dispersion of small
swarms of bubbles, and that the Reynolds stresses
(fkk ) + (fkk vk ) = 0 (1)
t can be linearly related to the mean velocity
gradients (eddy viscosity hypothesis) as in the
where f k , k and v k represent the volume fraction, relationship between the stress and strain tensors
macroscopic density and velocity vector for the k in laminar Newtonian flow, and thus an effective
phase, respectively. The gas and liquid volume viscosity can be assumed (Equation 5), and the
fractions, f g and f l , are related through the Reynolds fluxes of a scalar are linearly related to
summation constraint indicated by Equation 2: the mean scalar gradient.

fg + fl = 1 (2) eff = + t (5)

95
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 3 n. 3 | p. 093-102 | 2009 | ISSN 1982-0593

The standard k- model is related to the In Figure 1 two perforated sieve trays of a
turbulent kinetic energy and its dissipation rate as distillation column are illustrated, which were
follows: studied in the CFD simulations.

k2 (6) 2.4 Boundary and Initial Conditions

t = C
The physical space is mapped in a cylindrical
computational space and boundary conditions are
where k is the turbulent kinetic energy and is the required for all boundaries of the physical domain.
dissipation rate of the turbulent kinetic energy. The At the inlet a uniform profile of the velocities and
conservation equation for turbulent kinetic energy turbulent properties is imposed, without any slip
and its dissipation rate can be written as: conditions on the wall for both phases. Pressure
conditions at the outlet were also applied for the

(r k ) + r U k + t k = two phases. Figure 2 shows the physical domain
t

k

used in the simulations and the configuration of
(7) the systems which were simulated. The diameter of
= r (P ) + T
(k )
the tray is 0.21 m with a height of 0.15 m. The
length of the weir is 0.16 m and the diameter of
the resistor is 0.01 m. The liquid enters the tray
(r ) + r U + t = through a half moon-shaped opening. The number
t of elements in each of the numerical meshes was
1 (8)
415164 and 1129497 elements, respectively, for
= r (C 1P C 2 ) + T
( )
the systems without and with the presence of the
k electrical resistance on the surface of the plate.

Velocity vector equations are treated as scalar

where C 1 , C 2 , C , k and 1 are the model equations, one scalar equation for each velocity
constants and T(k ) and T( ) are coefficients of component. All scalar variables are discretized and
mass transfer between phases. evaluated at the cell centers. Velocities required at
the cell faces are evaluated by applying an
A commercial CFD package, known as CFX 11, improved RhieChow interpolation algorithm (Rhie
produced by AEA Technology, was used to solve and Chow, 1983). Transport variables such as
the continuity and momentum equations for the diffusion coefficients and effective viscosities are
two-fluid mixture. This package is a finite volume evaluated and stored at the cell faces. The
solver, using body-fitted grids. The grid used is non- pressurevelocity coupling is obtained using the
staggered and discretization of the equations at the SIMPLEC algorithm (Van Doormal and Raithby,
grid is performed using a finite difference (finite 1984). No problems with numerical diffusion are
volume) method. anticipated in view of the small size of the grid and
the time steps used. A fully-implicit backward

(b)
(a)

Figure 1. Top views of modules showing the characteristics of the plate (a) without and (b) with electrical resistor.

96
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 3 n. 3 | p. 093-102 | 2009 | ISSN 1982-0593

Figure 2. 3-D multiphase physical domain (a) without and (b) with electrical resistance.

difference scheme was used for the time Figures 3a and 3c show the distribution profiles
integration. of the volumetric fraction of water in the XY plane
and XZ plane, respectively, at three different
Air, at ambient pressure, and water were used heights (1.5, 9.0 and 16.5 cm), without the
as the gas and liquid phases, respectively. At the presence of the electrical resistance on the surface
beginning of the simulation, the conditions of the trays. Figures 3b and 3d show the behavior
consisted of liquid up to the weir height, and air up of velocity vectors for air (in the XY plane) and
to the weir height, at a homogeneous temperature liquid (in the XZ plane 1.5 cm above the surface),
equal to T 0 . The velocity fields and the turbulent respectively.
properties, also considered as initial conditions for
each simulation, are shown in the next section. The It can be observed in Figure 3a that the retained
time increment used in the simulations is 0.001 s. liquid layer (hold-up) represented by the dark blue
During the simulation the volume fraction of the color (volume fraction of liquid equal to one)
liquid phase in the gasliquid dispersion in the undergoes a small increase, with the mixing of both
system is monitored and quasi-steady state is phases and foam formation because of the high gas
assumed to prevail if the value of the hold-up flow.
remains constant for sufficiently long period of
time in order to determine the time-averaged The rising of the liquid is more pronounced in
values of the various parameters. Typically, steady the region near the wall of the exit from the lower
state is achieved in about 12 s from the start of the tray, since air inlet through the holes is a control
simulations. Simulations were performed in a condition with constant velocity and normal
cluster with ten Pentium-4 computers with boundary, which forces the evolution of some
processors running at 200 MHz. A typical preferential routes, such as those observed,
simulation took around 7 days to simulate 20 s of causing part of the air to rise through the tube of
tray hydrodynamics. the liquid drop.

Figure 3b shows the behavior of the velocity

vectors of the gas phase in the XY plane. It can be
observed that there are regions of large
3. RESULTS AND DISCUSSION recirculation and turbulence near the walls. It is
also possible to observe the appearance of some
In this section, the results of simulations route preferences with higher gas phase velocity,
performed using the two physical domains studied as in the region near the fall barrier of the hold-up
and analyzed, that is, with and without the and also near the wall of the liquid drop. This
presence of electrical resistance on the tray confirms the behavior noted through observation
surfaces, are presented. of the distribution of the volume fraction of the
liquid in this plane.

97
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 3 n. 3 | p. 093-102 | 2009 | ISSN 1982-0593

Water Volume Fraction Air Superficial Velocity

(a) Volume Fraction of liquid phase in the XY plane (b) Velocity vector of gaseous phase in the XY plane
Water Volume Fraction Water Superficial Velocity

(c) Volume Fraction of liquid phase in the XZ plane at (d) Velocity vector of liquid phase in the XZ plane at a
three different heights height of 1.5 cm above the surface of the plate

Figure 3. Distribution of the liquid volumetric fraction and velocity vectors for multiphase flow
without the electrical resistance.

Figure 3c confirms the distribution of the liquid The presence of some circulation regions near
volume fraction at three different levels in the XZ the walls of the plate and also at the proximities of
plane. It can be observed that the liquid phase the liquid fall tube can be noted. This distillation
distribution at 1.5 cm above the tray represents a tray behavior presents some operational and
less homogeneous mixture of liquid and vapor efficiency problems, but it should be noted that
phases, with a region of greater concentration of this study is focused on the hydrodynamics of the
air close to the hold-up fall barrier and near the plate in order to study these phenomena in
wall of the liquid fall represented by the color particular.
green. This probably occurred due to the initial
condition at the boundary imposed on the gas To demonstrate the volume fraction distribution
entrance flow, which is constant and normal. of the liquid in the active bubbling region (-0.5 to
0.5 in X[m]), three lines were drawn on it (one in
The distribution of the liquid volume fraction at the center of the plate and two 5 cm away to the
1.5 cm above the top plate shows greater right and to the left of the central line) for both the
homogeneity of the mixture, due to the behavior of lower tray and the top tray.
the vapor flow being closer to reality, that is, in the
holes of the top plate no air entry condition was It can be observed in Figure 4 that the
imposed because the air that passes through the distribution of the liquid volume fraction in the
liquid retained in the lower tray is the same as that lower tray has a large variation in the bubbling
which enters the holes of the top tray. The region, with points of minimum fluid concentration
behavior of the velocity vector of the liquid phase closer to the wall than in the case of the hold-up
at 1.5 cm from the surface of the plate can be (close to 0.05 m). This is an undesirable factor in
observed in Figure 3d. terms of tray operation since there will be
concentration and temperature gradients on the

98
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 3 n. 3 | p. 093-102 | 2009 | ISSN 1982-0593

perforated distillation plate with an average error

of 11.25%.

In order to evaluate the influence of electrical

resistance on the surface of the plate on the
hydrodynamics, new simulations were performed
using the air and liquid flows described in the
previous case as the new physical domain and the
results are represented in Figure 6.

Figure 6a shows the distribution of the liquid

Figure 4. Volumetric fraction distribution of the liquid volume fraction and illustrates again the formation
surface of the plates in three different areas on the of a preferential route for the gas movement for
surface of the tray. the whole liquid drop. However, in the active
bubbling area the mixing occurred in a more
surface of the tray reducing its efficiency. In the top homogeneous way with lower elevation of the
tray phase mixture is considerably more hold-up. In Figure 6b, the behavior of the velocity
homogeneous, which is more appropriate for the vectors of the gas phase is observed, in the same
good operation of a perforated tray distillation. plane for which the profile of the liquid volume
fraction was demonstrated, showing circulation
The height values for the clear liquid phases zones near the walls. Again it is possible to observe
were obtained in a time period which ensured the a preferential route for the gas phase drainage
condition of a quasi-stationary state, determined through the liquid fall tube.
through the average volumetric fraction of the
liquid times the height of the computational In Figure 6c, it can be observed that at 1.5 cm
domain. These values are shown in Figure 5 and above the surface of the lower plate the mixture of
compared with those obtained by Bennett et al. phases is not homogeneous for the plate with
(1983) using an experimental correlation. electrical resistance because there are points with
different concentrations of water (or gas), as
It can be observed in Figure 5 that there is a indicated by the presence of regions with higher
small difference between the values found for the concentrations of water (strong blue) and regions
clear liquid height through simulation and those where water is almost absent (green tending
obtained by Bennett et al. (1983) through toward red). In this situation, mass transfer does
experimental correlation. The 3D homogeneous not occur in a maximized way in the bubbling
multiphase model showed a behavior that was very region, and there are concentration gradients in
similar to the experimental correlation of Bennett, this region. However, by comparing Figures 3c and
and the results obtained with the homogeneous 6c, it is clear that the mixing on the surface of the
model predict the clear liquid height in a plates with resistance is more effective than in the
case without resistance. Figure 6d shows the
behavior of the velocity vectors of the liquid phase
0.03 at 1.5 cm above the surface of the lower plate,
where it is possible to observe that circulation
Clear Liquid height (m)

0.025

0.02 occurs near the curved walls of the plate in the

0.015 region of liquid fall and in some small areas near
0.01 the electrical resistor wall.
0.005

0 Similarly, in order to demonstrate the

0.2 0.3 0.4 0.5 0.6 0.7 0.8 distribution of the liquid volume fraction in the
Superficial gas velocity (m/s)
active bubbling region (-0.05 to 0.05 in X [m]) for
Bennett Correlation CFX Models
the physical domain with the presence of the
Figure 5. Comparison of the 3-D model with the electrical resistance on the surface of the plate,
correlation of Bennett et al. (1983) for the clear three lines were drawn over this area (one in the
liquid height with a feed water velocity of 0.1 (m/s). center of the plate and two others at 5 cm to right

99
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 3 n. 3 | p. 093-102 | 2009 | ISSN 1982-0593

Water Volume Fraction Air Superficial Velocity

(a)
(a)Volume
VolumeFraction
Fractionofofliquid
liquidphase
phaseininthe
theXY
XYplane
plane (b)(b)
Volume Fraction
Volume of gaseous
Fraction of gaseousphase in the
phase XYplane
in XY planeininthe
the
ininthe
thepresence
presenceofofelectrical
electricalresistance
resistance presence
presenceofofelectrical
electricalresistance
resistance

Water Volume Fraction Water Superficial Velocity

(c)Volume
Volumefraction
fractionofofliquid
liquidphase
phaseininthe
theXZ
XZplane
planeatatthree
three (d)
(d)Vector
Vectorvelocity
velocityofofliquid
liquidphase
phaseininthe
theXZ
XZplane
planeatataa
(c)
differentheights
heightsininthe
thepresence
presenceofofelectrical
electricalresistance
resistance height
heightofof1.5
1.5cm
cmabove
abovethe thesurface
surfaceofofthe
theplate
plateininthe
the
different
presence
presenceofofelectrical
electricalresistance
resistance

Figure 7. Distribution of the liquid volumetric fraction and velocity vectors for multiphase flow
in the presence of electrical resistance.

and left of the center line) for both the lower and fraction on the plate. From the separation process
upper plates. point of view, this is beneficial because it minimizes
the possible concentration and temperature
In Figure 7, the distribution of the liquid volume gradients in this region.
fraction in the lower plate shows only a small
region within which there is a considerable
variation in the volume fraction of the liquid phase
in the bubbling region next to the wall (hold-up,
close to 0.05 m). On the top plate the mixture
between the phases was considerably more
homogeneous and thus satisfactory in terms of
proper operation of a perforated plate distillation.

Comparing the results obtained with and

without the presence of the electrical resistance on
the surface of the plates, the distribution of the
liquid volume fraction or, in other words, the
hydrodynamic flow on the plate, was not found to
be adversely affected. Analysis of Figures 4 and 7
shows that the presence of electrical resistance Figure 6. Distribution of the volumetric fraction of
even favors the mixing since there is a more the liquid surface of the plates in three different
areas on the surface of the plate in the presence of
homogeneous distribution of the liquid volume
electrical resistance.

100
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 3 n. 3 | p. 093-102 | 2009 | ISSN 1982-0593

4. CONCLUSIONS GESIT, G.; NANDAKUMAR, K.; CHUANG, K.T. CFD

modeling of flow patterns and hydraulics of
It can be concluded from the results obtained in commercial-scale sieve trays. AIChE Journal, v.
this work that the presence of electrical resistors 49, p. 910924, 2003.
on the surface of the plates influences the patterns GU, F.; LIU, C.J.; YUAN, X.G.; YU, G.C. CFD
of liquid flow over them, and this does not simulation of liquid film flow on inclined plates.
adversely affect the hydrodynamics. In fact, it Chemical Engineering Technology, v. 27, p.
favors the homogenization of phases, indicating 10991104, 2004.
that the application of this methodology to control KISTER, H.Z. Distillation Design, McGraw-Hill, New
systems with distributed heating is thus a feasible York, 1992.
option. Moreover, it improves the homogenization KRISHNA, R.; vAN BATEN, J.M.; ELLENBERGER, J.;
of phase mixing in the active bubbling region, HIGLER, A.P.; TAYLOR, R. CFD simulations of
which is very important in terms of maximizing sieve tray hydrodynamics. Chemical Engineering
mass transfer. From these results, future studies Research Design, v. 77, p. 639646, 1999.
will be carried out with mixtures of ethanol and LI, X.G.; LIU, D. X.; XU, S. M.; LI, H. CFD simulation
water in order to analyze the efficiency of the of hydrodynamics of valve tray. Chemical
separation process of the two domains, and also to Engineering and Processing: Process
evaluate the effect of electrical resistance on the Intensification, v. 48, p. 145-151, 2009.
mass transfer process in this unit. LIU, C. J. X.; YUAN, G.; YU, K. T.; ZHU, X. J. A fluid-
dynamic model for flow pattern on a distillation
tray. Chemical Engineering Science, v. 55, p.
2287-2294, 2000.
MALISKA, C. R. Transferncia de calor e mecnica
ACKNOWLEDGMENTS
dos fluidos computacional. Rio de Janeiro: Livros
Tcnicos e Cientficos, 2004. (in Portuguese)
The authors are grateful for financial support
MARANGONI, C.; MACHADO, R. A. F. Distillation
from the National Agency of Petroleum (ANP), and
tower with distributed control strategy: Feed
the Studies and Projects Fund (FINEP), by means of
temperature loads. Chemical Engineering and
the Human Resources Program of ANP for the
Technology, v.30, p.1292-1297, 2007.
Petroleum and Gas sector PRH-34-ANP/MCT.
NIKOU, M. R. K.; EHSANI, M. R. Turbulence models
application on CFD simulation of
hydrodynamics, heat and mass transfer in a
structured packing. International
5. REFERENCES Communications in Heat and Mass Transfer, v.
35, p. 1211-1219, 2008.
ATAKI, A.; BART, H. J. Experimental and CFD NORILER, D.; MEIER, H. F.; BARROS, A. A. C.; WOLF
simulation study for the wetting of a structured MACIEL, M. R. Thermal fluid dynamics analysis
packing element with liquids. Chemical of gas-liquid flow on a distillation sieve tray.
Engineering Technology, v.29, p. 336347, 2006. Chemical Engineering Journal,
BENNETT, D. L.; AGRAWAL, R.; COOK, P.J. New doi:10.1016/j.cej.2007.03.023, 2007.
pressure drop correlation for sieve tray RHIE, C.M.; CHOW, W.L. Numerical study of the
distillation columns. AIChE Journal, v. 29, p. turbulent flow past an airfoil with trailing edge
434442, 1983. separation. AIAA Journal, v. 21, p. 15251532,
COLWELL, C.L. Clear liquid height and forth density 1983.
on sieve trays, Industrial and Engineering SZULCZEWSKA,B.; ZBICINSKI, I.; GORAK, A. Liquid
Chemistry Process Design and Development, v. flow on structured packing: CFD simulation and
20, p. 298299, 1979. experimental study. Chemical Engineering
FISCHER, C.H.; QUARINI, G.L. Three-dimensional Technology., v. 26, p. 580584, 2003.
heterogeneous modeling of distillation tray van BATEN, J.M.; KRISHNA, R. Modelling sieve tray
hydraulics. Proceedings of the AIChE Annual hydraulics using computational fluid dynamics.
Meeting, November 1520, Miami Beach, USA, Chemical Engineering. Journal, v. 77, p. 143
1998. 151, 2000.

101
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 3 n. 3 | p. 093-102 | 2009 | ISSN 1982-0593

VAN DOORMAL, J.; RAITHBY, G.D. Enhancement of incompressible flows. Number. Heat Transfer,
the SIMPLE method for predicting v.7, p. 147163, 1984.

102