SIMULATIONS OF THE KETTLE REBOILER SHELL SIDE

THERMAL-HYDRAULICS WITH DIFFERENT TWO-PHASE

FLOW MODELS
by

Milada PEZO, Vladimir D. STEVANOVI], and @arko STEVANOVI]

Original scientific paper
UDC: 532.529:66.011
BIBLID: 0354-9836, 10 (2006), 2, 127-140

A computational fluid dynamics approach is presented for the simulation

and analyses of the kettle reboiler shell side thermal-hydraulics with two
different models of two-phase flow – the mixture and two fluid model. The
mixture model is based on solving one momentum equation for two-phase
mixture flow and a closure law for the calculation of the slip between gas
and liquid phase velocities. In the two fluid modeling approach the momen-
tum balance is formed for each phase, while the gas-liquid interaction due
to momentum exchange at the interface surface is predicted with an empiri-
cal correlation for the interface friction coefficient. In both approaches the
two-phase flow is observed as two inter-penetrating continua. The models
are solved for the two-dimensional geometry of the kettle reboiler shell side
vertical cross section. The computational fluid dynamics numerical method
based on the SIMPLE type algorithm is applied. The results of both liquid
and vapour velocity fields and void fraction are presented for each model-
ing approach. The calculated void fraction distributions are compared with
available experimental data. The differences in the modeling approaches
and obtained results are discussed. The main finding is that the void frac-
tion distribution and two-phase flow field strongly depends on the modeling
of the slip between liquid and gas phase velocity in mixture model or on the
interface friction model in two fluid model. The better agreement of the nu-
merically predicted void fraction with the experimental data is obtained
with the two fluid model and an interfacial friction model developed for the
conditions of two-phase flows in large volumes of kettle reboilers or differ-
ent designs of steam generators.
Key words: kettle reboiler, thermal-hydraulics, numerical simulation,
computational fluid dynamics

Introduction

Shell and tube heat exchangers are among the most widely used types of heat
exchangers. Various shell and tube heat exchangers are designed for vapour generation
on the shell side. They are widely applied in chemical, process, and energy power indus-
try, in refregirations and air-conditioning equipments, and they are applied such as
reboilers, steam generators, and evaporators. It has been estimated that more than 50% of

127
 THERMAL SCIENCE: Vol. 10 (2006), No. 2, pp. 127-140

all heat exchangers employed in process industries are used to boil fluids and involve
two-phase flow on the shell side [1]. In process industry they are known as reboilers,
while kettle reboilers are one of the most common reboiler types [2]. Also, some develop-
ments of horizontal steam generators for nuclear power plants are based on the kettle
reboiler design [3].
A typical design of the kettle reboiler applied in the process industry is shown in
fig. 1. The evaporating fluid flows on the shell side, across a horizontal tube bundle. The
heat is transferred to the boiling two-phase mixture from a hot fluid that circulates inside
the tubes. The liquid level is controlled by a weir, so that the bundle is always submerged
in liquid. The gap between the bundle and the shell allows internal recirculation of liquid.
The liquid enters the bundle at its bottom only. The mass velocity of fluid across the bun-
dle is increased by the recirculation of liquid, affecting the global heat transfer coeffi-
cient.

Figure 1. Schematic view

of the kettle reboiler

Previous investigations of the kettle reboiler shell side thermal-hydraulics have

been performed with experimental and analytical models of various levels of complexity
regarding the multidimensionality and thermal-hydraulic complexity of boiling two-
-phase flow conditions. Void fractions and pressure drops in one-dimensional upward
two-phase flows across the tube bundles with various tubes’ arrangements are presented
in [4, 5]. One dimensional investigations were performed first, because the process of
evaporation is complex and the vertical flow across the tube bundle is dominant. But,
reboiler shell side thermal-hydraulics is strongly influence by multidimensional effects.
It was shown that the vertical pressure change is not constant along the bundle, causing a
lateral pressure change, which must be satisfied by a lateral flow [6]. In [7] a more realis-
tic two-dimensional investigation is presented with information how variation of heat
flux, weir height, bundle size and pressure affect the kettle reboiler processes. Two-di-
mensional numerical models of the kettle reboiler shell side thermal-hydraulics are pre-
sented in [8-10]. The main findings in performed researches are: (a) the homogeneous
model of two-phase flow provides too high values of the void fraction which means that
the gas and liquid phase velocity slip should be taken into account in a kettle reboiler de-

128
Pezo, M., Stevanovi}, V. D., Stevanovi}, Z.: Simulations of the Kettle Reboiler Shell ...

sign or analyses procedures, and (b) the liquid phase circulation is organized on the
reboiler shell side, where the intensity of circulation influences the heat transfer
coefficient and the void fraction distribution.
This paper presents the possibilities of two commonly applied two-phase flow
models for the prediction of the kettle reboiler shell side thermal-hydraulics. These are
(a) the mixture model of two-phase flow with the application of the closure law for the
prediction of the gas and liquid phase slip velocity (for instance presented in [11]), and
(b) the two fluid model of two-phase flow with the closure law for the prediction of the
gas and liquid phase interfacial friction (applied in [10]).

Modelling approaches

Several assumptions are introduced in modeling the kettle reboiler shell side
thermal-hydraulics:
(a) The shell side flow in the slab of the kettle reboiler vertical cross section (as presented
with the left scheme in fig. 1) is two-dimensional;
(b) Steady-state conditions are modelled;
(c) The shell side two-phase mixture is saturated;
(d) The surface tension at the gas-liquid interface is neglected, as it is not important for
bulk two-phase flow phenomena. Hence, pressure is the same for both phases within
the numerical control volume;
(e) Flow governing equations are written in the non-viscous form, while the turbulent
viscosity effects are taken into account indirectly through friction coefficients for the
tube bundles flow resistance and two-phase interfacial friction force; and
(f) The porous medium concept is used in the simulation of two-phase flow within tube
bundles. In the applied mixture model it is assumed that the porous media is 100%
open to the fluid flow, while in the two-fluid model the real volumes of liquid and gas
phase are taken into account. In both models, the resistance of the tube bundle to the
two-phase flow is taken into account through the appropriate volumetric force.

Mixture model

The mixture model solves the continuity equation for the mixture, the momen-
tum equation for the mixture, and the volume fraction equation for the secondary phases,
as well as algebraic expressions for the relative velocities (if the phases are moving at dif-
ferent velocities) [11]. The energy equation for the mixture is also a constituent of the
mixture model, but here it is not included since the saturated liquid and two-phase flows
on the reboiler shell side are considered.
The continuity equation for the two-phase mixture is:
r
Ñ × ( r m vm ) = 0 (1)

129
 THERMAL SCIENCE: Vol. 10 (2006), No. 2, pp. 127-140

r
where v m is the mass-averaged velocity:
r
å ak rk v k
r k=fg
vm = (2)
rm

and r m is the mixture density:

r m = å ak rk (3)
k=fg

a k is the volume fraction of phase k (liquid phase k = f or gas phase k = g).

The momentum equation for the mixture can be obtained by summing the indi-
vidual momentum equations for both phases:
r r r r r r
Ñ × ( r m v m v m ) = -Ñp + r m g + Ñ × (a g r g v dr g v dr g ) - Fwm (4)
r
where v dr g is the drift velocity of the dispersed vapour phase:
r r r
v dr g = v g - v m (5)

Introducing eq. (2) in eq. (5) the following expression for the vapour drift veloc-
ity is derived in which the slip velocity between the vapor and liquid phase figures (also
referred to as the relative velocity):
r r r
v fg = vg - v f (6)

The slip velocity is calculated with the following semi-empirical correlation:

r ( r m - r g ) d g2 r
v fg = a (7)
18m f f drag
r
where dg is the bubble diameter and a is the mixture acceleration. The drag coefficient
fdrag is calculated with the empirical relations of Schiller and Naumann:

. Re 0. 678 , Re £ 1000
ì1 + 015
f drag = í (8)
î 00183
. Re, Re > 1000
r
and the acceleration a is of the form:
r r
r r r r ¶ v m r Dv m
a = g - (v m × Ñ )v m - =g - (9)
¶t Dt
r
The volumetric force of tube bundle resistance to two-phase mixture flow F wm
is calculated as:
r 2 rm vm 2 r
e e
Fwm = å V e (10)
e =1 2 De

130
Pezo, M., Stevanovi}, V. D., Stevanovi}, Z.: Simulations of the Kettle Reboiler Shell ...

r
where e is a unit vector in the direction of the Cartesian coordinate axis, and De is the
width of the computational cell. The coefficient of the local pressure drop in e direction Ve
is calculated according to [12].
The vapour volume fraction is calculated from the continuity equation:
r r
Ñ × (a g r g v m ) = -Ñ × (a g r g v dr g ) (11)

Two fluid model

Here applied two-fluid model of two-phase flow consists of the continuity and
momentum equations for each phase and corresponding closure laws.
Mass conservation equations for liquid and vapour phase are:

¶ (a f r f ) r
+ Ñ × (a f r f Uf ) = - Ge (12)
¶t
¶ (a g r g ) r
+ Ñ × (a g r g U g ) = Ge (13)
¶t
Liquid and vapour momentum conservation equations:
r
¶ (a f r f U f ) r r r r r r
+ Ñ × (a f r f U f U f ) = -a f Ñp + a f r f g + Fg f - Fw f - Ge U f i (14)
¶t
r
¶ (a g r g U g ) r r r r r r
+ Ñ × (a g r g U g U g ) = -a g Ñp + a g r g g + Fg f - Fw g - Ge U f i (15)
¶t

Evaporation rate is calculated as:

q cell
Ge = aw (16)
hf g
where
¢¢
q cell = Acell q cell (17)

when heat flux at the wall of the shell tube is defined, or:

q cell = h tp Acell DTsat (18)

when the temperature of the tube wall is defined.

The heat transfer coefficient calculations in the tube bundle are defined by the
Chen type of correlation [13]:

h tp = Fhconv + Sh nb (19)

131
 THERMAL SCIENCE: Vol. 10 (2006), No. 2, pp. 127-140

or
. k
0023
h tp = +
D Re . 8 Pr 0. 4 F
0
(20)
k f0. 79 c p0.f45 r 0f . 49
+000122
. (T w - Tsat ) 0. 24 ( p w - p) 0. 75 S
s 0. 5 m 0f. 29 h f0g. 24 r g0. 24

where applied coefficients are defined as:

p w = psat (21)

-1
F = 10
. for X TT £ 010
.
-1 -1
(22)
F = 235
. ( X TT . ) 0. 736
+ 0213 for XTT > 010
.

. (Re F 1. 25 )1.14 ]-1

S = [1 + 012 for Re F 1. 25 < 325
.
S = [1 + 0.42(Re F 1. 25 ) 0. 78 ] . £ Re F 1. 25 < 700
for 325 . (23)
S = 01
. for Re F 1. 25 ³ 700 .
and Martinelli parameter:
0. 9 0. 5 0.1
æ1-a ö æ rg ö æ m f ö
X TT =ç ÷ ç ÷ ç ÷ (24)
è a ø ç r ÷ çm ÷
è f ø è g ø
and Reynolds number:
G (1 - X ) Dhy
Re = (25)
mf

The wall drag term can be expressed as a pressure drop due to the tube bundle
friction. The pressure drop due to the two-phase mixture flow around tubes in a bundle is
determined by taking into account the separate contribution of the each phase to the total
pressure drop. The pressure drop of phase k in e direction is defined as:
r r k u$ k2e r
ak e
F wk = V k e (1 - a w ) (26)
2 a f + a g De

where V k e is pressure loss coefficient in y direction and u$ k e , k = f , g is the maximum

velocity of the phase in e direction.
The interfacial friction terms have influence on void fraction and relative veloc-
ity of the phases [14]:
r 3 Cd i r r r r
Fg f = r f a × Ug - U f ( U g - U f ) (27)
4 Db

132
Pezo, M., Stevanovi}, V. D., Stevanovi}, Z.: Simulations of the Kettle Reboiler Shell ...

where Cdi is the interfacial drag coefficient and Db is the diameter of the bubble. For
bubble flow j £ 0.3 [14]:
2
gDr ìï1 + 7 1767
. f ( j) 6 üï
C d i = 0267
. Db í ý (28)
s ïî 1867 . f ( j) ïþ

where
f ( j) = (1 - j)1. 5 (29)
and
ag
j= (30)
a f + ag

For churn-turbulent flows (j > 0.3), a new correlation is proposed based on the
results proposed in [15]:

gDr
C d i = 1487
. Db (1 - j) 3 (1 - 0.75j) 2 (31)
s

The functional dependence of the ratio Cdi/Db on the void fraction is given in fig.
2. For transitional flow patterns, there is a rapid decrease of Cdi/Db ratio. This could be at-
tributed to the decrease of the two-phase flow interfacial area concentration.
According to the applied porous media approach, the control volume within the
tube bundle is occupied with tubes and a free area filled with single phase or two-phase
mixture. Tube bundle is represented as porous area, with the porous factor:
D2
p
y =1 - 4 (32)
a2

The presented models are solved with the control volume based numerical
method and code presented in [16]. Details about the solving procedure are presented in
[17].

Figure 2. Ratio of interfacial friction

coefficient Cdi and particle diameter Dp vs.
void fraction for two-phase flow across a tube
bundle

133
 THERMAL SCIENCE: Vol. 10 (2006), No. 2, pp. 127-140

Boundary conditions

Boundary conditions include the line of symmetry, the shell wall, and the inlet flows
and outlet recirculating flow. At the line of symmetry gradients of all variables are con-
stant. Saturated liquid inflow was specified at the bottom center of the heat exchanger,
simulating the feed flow in an actual kettle reboiler. The flow rate was specified using the
overflow and the vaporization rate. At the outflow boundary in an actual kettle reboiler,
the vapour separates from the liquid and leaves the reboiler. In the process, most of the
liquid recirculates within the reboiler, but some of
the liquid recirculates within the evaporator, and
some of the liquid is carried over the weir.
Recirculation of the liquid is simulated by defin-
ing the velocity gradient at the outlet surface.
Boundary conditions are presented in fig. 3 for
two fluid model velocity components V along ver-
tical coordinate and W component along horizon-
tal coordinate. The reduction of these boundary
conditions to the mixture model is straightfor-
ward. In the mixture model the void fraction at the
free surface of the two-phase mixture is equal to 1,
while no change of the two-phase mixture veloc-
ity components is assumed as in case of the
vapour velocity component in the two fluid model
Figure 3. Boundary conditions (fig. 3).

Results and discussion

Developed models of two-phase flow on the ket-

tle reboiler shell side are applied to the numerical
simulation and analyses of the experimental condi-
tions presented in [18]. Geometric parameters re-
quired to specify the heat exchanger are shown in
fig. 4 and tab. 1. Experiments were performed under
atmospheric pressure.
The presented reboiler geometry is discretized
with two-dimensional control volumes, fig. 5.
Results of the numerical simulations performed
with two fluid model and mixture model are pre-
sented in figs. 6, 7, and 8. In those pictures total
two-phase flow mass flux vectors were calculated
Figure 4. Geometry of the kettle as:
r r r
reboiler experimental test section G = a f r f U f + ag rg U g (34)
[18]

134
Pezo, M., Stevanovi}, V. D., Stevanovi}, Z.: Simulations of the Kettle Reboiler Shell ...

Table 1. Dimensions and parameters of the

experimental kettle reboiler [18]

Tube diameter 2r 19 mm
Pitch h 25.4 mm
Shell radius R 0.368 m
Weir height from the
0.210 m
center of shell b
Bundle center/shell
0.114 m
radius offset a
Number of tubes 241
Tubes arrangement In-line square
Working fluid Refrigerant R113
Figure 5. Meshes used for modeling

The results in fig. 6 clearly show the

formation of one circulation centre at the
boundary of the upper half of the tube
bundle (the bundle is depicted with
dashed line). Strong downward flow ex-
ists in the downcomer between the bundle
and the shell wall. This natural circula-
tion is governed by the density dif-
ferences between the mixture in the bun-
dle-shell downcomer (lower void fraction
and higher density) and the boiling
two-phase mixture in the bundle (higher
void and lower density). The reference
mass flux vectors of 1400 kg/m2s for two
fluid model results and 1700 kg/m2s in
case of mixture model indicate almost the
Figure 6. Mass flux vector for liquid for two
same intensity of the overall circulation fluid model (left) and mixture model (right)
around and through the bundle. But, the
mixture model predicts less intensive lat-
eral flow from the downcomer towards the tube bundle. According to the mixture model,
the two-dimensional character of the flow between the downcomer and the tube bundle is
suppressed; both the bundle and downcomer flows are vertical in opposite directions.
The two-phase flow model predicts lower void fractions than the mixture model,
fig. 7. This is due to the more intensive liquid downward flow in the bundle-shell
downcomer and liquid penetration from the downcomer to the bundle, as presented in the
corresponding left picture in fig. 7. The thick full line in both pictures in fig. 7 represents

135
 THERMAL SCIENCE: Vol. 10 (2006), No. 2, pp. 127-140

the lower boundary of the region of higher

void as observed in the experimental investi-
gation. The region above the full thick line is
characterized as frothy. In right side of fig. 7
a very rapid increase in void is calculated
above the dividing line with void values
above 0.8 at the top of the bundle. It is hardly
to expect that the high values of void frac-
tion can be achieved under considered heat
flux of 20 kW/ m2.
Figure 8 shows the horizontal void frac-
tion distributions at the half of tube bundle
height calculated with the Scharge et al. em-
pirical correlation as reported in [9] and pre-
Figure 7. Void fraction contour plot for
two-phase model (left) and mixture model dicted with here presented two fluid model
(right) with comparation with Cornwells and mixture model. The mixture model does
experiments [15] not provide satisfactory agreement with the
two fluid model and the referent data of
Schrage, especially in the inner areas of the
bundle close to the axis of symmetry. The
two fluid model provides much better agreement with the referent void fraction data.
As it is presented with the results of the numerical simulation, the mixture model
is less suitable for the simulation and analyses of the kettle reboiler shell side thermal-hy-
draulics. This is attributed to the fact that the mixture model is inherently not suitable for
the modelling of vapour and liquid phase separation in the large two-phase volumes with
liquid phase recirculation, especially at the two-phase mixture free-surface (so called
swell level).

Figure 8. Comparison of the

horizontal void fraction distri-
butions at the half of the tube
bundle height calculated with
the Scharge et al. model and re-
ported in [9] and predicted with
here presented two fluid model
and mixture model

136
Pezo, M., Stevanovi}, V. D., Stevanovi}, Z.: Simulations of the Kettle Reboiler Shell ...

Conclusions

The mixture and two fluid models are applied to the simulation of the kettle
reboiler shell side thermal-hydraulics. Results obtained with the mixture, based on the
two-phase mixture continuity and momentum equation and corresponding semi-empiri-
cal closure laws for the vapor and liquid phase slip velocity, does not provide reliable re-
sults of the kettle reboiler shell side thermal-hydraulics. The lateral two-phase flow from
the downcomer to the tube bundle is suppressed and too high values of the void fraction
are obtained. On the other hand, more complex two fluid model, based on the mass and
momentum balance equations for each phase and corresponding closure laws for the in-
terface momentum exchange due to the interface friction, provides much better agree-
ment with the referent results of the void fraction distribution. Also, two fluid model pro-
vides more plausible results of the two-phase flow structure on the kettle reboiler shell
side than the mixture model. The inability of the mixture model to predict correctly the
shell side thermal-hydraulics is due to the solving of one momentum equation for the
two-phase mixture, and corresponding limitations in modelling the vapour and liquid
phase separation within the large volume of the kettle reboiler shell side with liquid phase
circulation.

Nomenclature

A
r – area of tube bundle cell, [m]
a – mixture acceleration, [ms–2]
aW – specific area of tube bundle cell, [m–1]
Cdi – interfacial drag coefficient, [-]
cp – specific heat at constant pressure, [kJkg–1K–1]
D – tube diameter, [m]
Db – diameter bubble, [m]
Dhy – hydraulic equivalent diameter, [m]
dg – bubble diameter, [m]
r
e – unit vector in the direction of the Cartesian coordinate axis, [–]
De
r – width of the computational cell in y or z direction, [m]
F – source term, [Nm–1]
Fwm – volumetric force of tube bundle resistance to two-phase mixture, [kgm–2s–2]
fdrag – drag coefficient, [-]
G – mass flux, [kgm–2s–1]
g – gravitational acceleration, [ms–2]
h – heat transfer coefficient, [Wm–2K–1]
k – thermal conductivity, [kWm–1K–1]
Pr – Prandtl number, (= mcp/k) [-]
p – pressure, [Pa]
q – heat flux, [Wm–2]
Re – Reynolds number, (= rUDhy/m), [-]
T – temperature, [K]
t – time, [s]
U – velocity of the phase (components Uky = Vk, Ukz = Wk), [ms–1]

137
 THERMAL SCIENCE: Vol. 10 (2006), No. 2, pp. 127-140

v – velocity, [ms–1]
r
v dr – drift velocity, [ms–1]
r
vm – mass-averaged velocity, [ms–1]
r
v fg – relative velocity, [ms–1]
X – flow quality, [–]
XTT – Martinelli parameter, [–]

Greec letters

a – volume fraction in the control volume, [–]

Ge – evaporation rate, [–]
V – pressure loss coefficient, [Nm–1]
s – surface tension, [Nm–1]
m – viscosity, [Pas]
r – density, [kgm–3]
j – vapour volume fraction (void) in two-phase flow, [–]
y – porosity, [–]

Subscripts

conv – convection
r
e – e direction
f – liquid
g – gas
i – interface parameter
in – inlet
k – phase
m – mixture
nb – neigbour cell
p – particle
sat – saturation
tp – tube bundle
w - wall
y - pertain to y direction
z - pertain to z direction

References

[1] Dowlati, R., Kawaji, M., Chan, A. M. C., Pitch-to-Diameter Effect on Two-Phase Flow
Across an in-Line Tube Bundle, AIChE J., 36 (1990), pp. 765-772
[2] Whalley, P. B., Hewitt, G. F., Reboilers in: Multiphase Science and Technology (Eds. G. F.
Hewitt, J. M. Delhaye, N. Zuber), vol. 2, Hemisphere Publ. Co. 1986, pp. 275-331
[3] Tujikura, Y., Oshibe, T., Kijima, K., Tabuchi, K., Development of Passive Safety Systems
for Next Generation PWR in Japan, Nucl. Eng. Des., 201 (2000), pp. 61-70
[4] Dowlati, R., Chan, A. M. C., Kawaji, M., Hydrodynamics of Two-Phase Flow Across Hori-
zontal Rod Bundles, J. Fluids Eng., 114 (1992), pp. 450-456
[5] Dowlati, R., Kawaji, M., Chan, A. M. C., Two-Phase Crossflow and Boiling Heat Transfer in
Horizontal Tube Bundles, J. Heat Transfer, 118 (1996), pp. 124-131

138
Pezo, M., Stevanovi}, V. D., Stevanovi}, Z.: Simulations of the Kettle Reboiler Shell ...

[6] King, M. P., Jensen, M. K., Local Heat Transfer and Flow Pattern Distributions in a Kettle
Reboiler, Two-Phase Flow Modelling and Experimentation, (1995), pp. 1289-1296
[7] Gebbie, J. G., Jensen, M. K., Void Fraction Distributions in a Kettle Reboiler, Exp. Therm.
Fluid Sci., 14 (1997), pp. 297-311
[8] Burnside, B. M., 2-D Kettle Reboiler Circulation Model, Int. J. Heat Fluid Flow, 20 (1999),
pp. 437-445
[9] Edwards, D. P., Jensen, M. K., A Two-Dimensional Numerical Model of Two-Phase Heat
Transfer and Fluid Flow in a Kettle Reboiler, HTD-159, Phase Change Heat Transfer,
ASME, 1991
[10] Pezo, M., Stevanovi}, V. D., Stevanovi}, Z., A Two-Dimensional Model of the Kettle
Reboiler Shell Side Thermal-Hydraulics (accepted for publication), International Journal of
Heat and Mass Transfer, (2005), available on-line at www.sciencedirect.com
[11] Sanyal, J., Vasquez, S., Roy, S., Dudukovi}, M. P., Numerical Simulation of Gas-Liquid Dy-
namics in Cylindrical Bubble Column Reactor, Chemical Engineering Science, 54 (1999),
pp. 5071-5083
[12] Rasohin, N.G., Nuclear Power Plants Steam Generators (in Russian), Atomizdat, Moscow,
1980
[13] Groeneveld, D. C., Snoek, C. W., A Comprehensive Examination of Heat Transfer Correla-
tions Suitable for Reactor Safety Analysis, in: Multiphase Science and Technology (Eds. G.
F. Hewitt, J. M. Delhaye, N. Zuber), vol. 2, Hemisphere Publ. Co., 1986, pp. 189-193
[14] Ishii, M., Zuber, N., Drag Coefficient and Relative Velocity in Bubbly, Droplet or Particulate
Flows, AIChE J., 25 (1979), pp. 843-855
[15] Rousseau, J. C., Houdayer, G., Advanced Safety Code CATHARE Summary of Verification
Studies on Separate Effects Experiments, Proceedings, 2nd International Topical Meeting on
Nuclear Reactor Thermal Hydraulic–NURETH 2, Santa Barbara, USA, 1983
[16] Spalding, D. B., Developments in the IPSA Procedure for Numerical Computation of
Multiphase-Flow Phenomena with Interphase Slip, Unequal Temperatures, etc, in: Numeri-
cal Properties and Methologies in Heat Transfer (Ed. T. M. Shin), Hemisphere Publ. Co.,
1983, pp. 421-436
[17] Pezo, M., Two-Dimensional Numerical Simulation and Analyses of the Kettle Reboiler Shell
Side Thermal-Hydraulics, M. Sc. thesis, University of Belgrade, 2004
[18] Cornwell, K., Duffin, N. W., Schuller, R. B., An Experimental Study of the Effects of Fluid
Flow on Boiling within a Kettle Reboiler Tube Bundle, ASME Paper No. 80-HT-45, 1980

139
 THERMAL SCIENCE: Vol. 10 (2006), No. 2, pp. 127-140

Authors' addresses:

M. Pezo, @. Stevanovi}
VIN^A Institute of Nuclear Sciences,
Department of Thermal Engineering and Energy
P. O. Box 522, 11001 Belgrade
Serbia

V. D. Stevanovi}
Faculty of Mechanical Engineering, University of Belgrade
16, Kraljice Marije, 11000 Belgrade
Serbia

Corresponding author (V. D. Stevanovi}):

E-mail: estevavl@eunet.yu

Paper submitted: January 18, 2006

Paper revised: June 10, 2006
Paper accepted: June 15, 2006

140