7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
Numerical Investigation of Bubble Growth in Subcooled Flow Boiling
Dewen Yuan, Liangming Pan*, Deqi Chen
School of Power Engineering, Chongqing University, Chongqing 400044, China
Email: cneng~cqu.edu.cn
Keywords: bubble growth; subcooled flow boiling: VOF model: UDF subroutine: Numerical simulation
Abstract: Bubble growth in subcooled flow boiling is a quite complicated process. It includes not only microlayer
evaporation on heating wall, but also involves top condensation when the bubble emerges to subcooled bulk. In present paper,
a bubble growth model for subcooled flow boiling is proposed to simulate bubble dynamics with CFD method. A microlayer
evaporation model at the bubble bottom and a heat and mass transfer model on bubble interface are used. User Defined
Function (UDF) interface in Fluent platform is applied to realize this model. The volume of fluid (VOF) multiphase flow
model combined with phase change model is used to simulate bubble growing. The effects of bulk velocity, liquid subcooling,
wall superheating and vaporliquid contact angle are considered in present model. The results reveal that bubble growth
process in subcooled flow boiling is a dynamic result of evaporation and condensation. With increasing wall superheating, the
bubble growth rate will be accelerated; however, with higher liquid subcooling, the bubble growth rate will decrease. The
simulation also reveals that the evolution of bubble shape is a complex process. During bubble growing, the pressure inside
the bubble decreases and distributes uniform inside the vapor bubble; and there will be a flow liquid region of having
minimum pressure in the vicinity of a growing bubble. The eddy can be seen on both side of a growing bubble and
accelerates contraction of the vaporliquid interface at bottom of bubble. With the distance from the heating wall increasing,
the flow liquid temperature decreases. In superheated liquid region, the temperature gradient which does not change with
bubble growth time is very great, while the temperature of subcooled liquid region which is far from the heating wall
increases with bubble growing.
Introduction
In order to understand and optimize the process of boiling
heat transfer, it is vital to understand the mechanism of
bubble nucleation, growth and departure. As a complex
process, the investigation of model for bubble formation and
disappearance has gotten only limited success. In recent
years, some investigations about single bubble growth
model for nucleate boiling have been performed.
Mei and Chen et al. (1995) performed a numerical analysis
about bubble growth in saturated heterogeneous boiling.
They considered the bubble growth is determined by
simultaneous energy transfer between vapor bubble and
liquid microlayer or heating wall. Finite different solutions
for temperature fields of microlayer and heating wall were
obtained on expanding coordinates as the bubble growing.
Samuel (1998) presented a numerical analysis about
axissymmetric vapor bubble growth. An interface tracking
method in conjunction with a finite volume method on a
moving unstructured mesh was used. The control volume
continuity, momentum and energy equations were modified
at the presence of a phase interface to include surface
tension and discontinuous pressure and velocity.
Heo and Koshizuka et al (2002) showed a numerical study
about the growth of bubble in transient pool boiling through
moving particle semiimplicit with mesh less advection by
flowdirectional local grid (MPSMAFL) method. The
growth process of a bubble with different initial radii was
calculated with high heat flux and high subcooling
condition.
Li and Yan et al (2003) provided a new understanding about
the interfacial transport characteristics of inviscid spherical
bubble with different geometric parameters, rising in a
stagnant hot or bisolution liquid. The flow and temperature
fields around bubbles and similarly sized rigid spheroids
were computed. The interfacial transport characteristics of
spherical bubbles were studied numerically.
Liao and Mei et al (21***0 developed a physical model for
vapor bubble growth in saturated nucleate boiling that
includes heat transfer through the microlayer and the bulk
superheated liquid surrounding the bubble as well. Both
asymptotic and numerical solutions for the liquid temperature
field surrounding a hemispherical bubble indicated that there
is a thin unsteady thermal boundary layer existed adjacent to
the bubble dome. During the early stages of bubble growth,
heat transfer to the bubble dome through the unsteady
thermal boundary layer constitutes a substantial contribution
to vapor bubble growth,
Genske and Stephan (2006) considered that the region around
a single growing bubble can be subdivided into three
subregions: microregion, bubble area, and surrounding
liquid. Their results showed that the flow pattern in the liquid
around a growing vapor bubble is not only determined by the
movement of the bubble surface, but also the vapor flow flied
inside the bubble. In regions away from the bubble, heat
conduction would be the dominant factor. Velocity and
temperature field, heat flux, bubble contour, and departure
diameter were calculated for different fluids by the authors.
Mukherjee and Kandlikar (2007) presented a static contact
angle model and a dynamic contact angle model based on
contact line velocity and the sign of the contact line velocity
which have been used at the base of a vapor bubble growing
on a heated wall. The complete NavierStokes equation was
solved and the liquidvapor interface was captured through
the LevelSet technique. The effects of dynamic contact angle
on bubble dynamics and vapor volume growth rate were
compared with results obtained with the static contact angle
model.
Hazi and Markus (2009) used the latticeBoltzmann approach
to simulate heterogeneous boiling on a horizontal plate in
stagnant and slowly flowing fluid. They found that the bubble
departure diameter is proportional to gl/2 and the release
frequency scales with g3/4 in a stagnant fluid, where g is the
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
gravitational acceleration. Simulation results showed no
dependence between the bubble departure diameter and the
static contact angle, but the bubble release frequency
increased exponentially with the latter.
Numerical researches about bubble growth mostly based on
pool boiling. However, the character of bubble growth in
subcooled flow boiling is quite different from that of the
pool boiling. The bubble growth process in subcooled flow
boiling is affected by the thermal characteristics of heating
wall and mainstream liquid, etc. Due to the existence of the
mainstream liquid velocity, the motion characteristic of
bubble interface and the flow field around the bubble are
more complex than that of pool boiling. Thus, distortion and
deformation of the bubble may occur which will lead to
characterized features during the growth,
In present paper, based on the characteristic of bubble
growth process in subcooling flow boiling, a bubble growth
model is proposed and a numerical simulation is performed
to understand the process of bubble growth. The simulation
results well reflect bubble growth process as a dynamic
result of evaporation and condensation which agree the
experimental results very well, and the relationship between
the vaporliquid interface motion and velocity field inside
the bubble also presented here.
Nomenclature
E energy
F surface tension force
hf, latent heat, kJkg :
the net mass flow rate per unit area,
G kgm s'
M mass per mole, kgmol :
p pressure, Pa:
q heat flux, kW m ;
r ~bubble radius, m:
R universal gas constant, Jmol K :
S the mass source term:
t time, s:
T temperature, K
u velocity, ms~'
x, y coordinate axis:
microlayer thickness under the bubble
8 bottom, m:
a surface tension coefficient, N/m
coefficient of the vaporliquid
interface.
oc. volume fraction:
p density, kgm
thermal conductivity of liquid,
?L Wm K1
AT the temperature difference, K.
Subscripts
1 liquid:
v ~vapor:
w wall;
saturated
evaporation
condensation:
ith fluid component
2 Numerical simulation methods
The VOF formulation relies on the fact that two or more
fluids (or phases) are not interpenetrating. For each additional
phase that was added to model, a variable is introduced: the
volume fraction of the phase in the computational cell. In
each control volume, the volume fractions of all phases sum
to unity. The fields for all variables and properties are shared
by the phases and represent volumeaveraged values, as long
as the volume fraction of each of the phases is known at each
location. This paper investigates the bubble growth process
by VOF model (Fluent Inc., 2005, Hirt and Nichols, 1981)
and UDF interface of the CFD software Fluent.
2.1 VOF model
Generally, a VOF algorithm (Hirt and Nichols, 1981) solves
the problem of updating the phase volume fraction field
given the fixed grid, the velocity field and the phase volume
fraction are determined in previous time step. In
twodimensional problem, the interface is considered to be a
continuous and piecewise smooth line. The problem is
reduced to the reconstruction of an approximation of the
interface in each cell, knowing only the volume fraction of
each phase in the cell itself and in the neighboring cells.
During all simulation cases in present work, a piecewise
linear interface calculation (PLIC) (Yongs, 1984) interface
reconstruction method has been used for interpolation in a
cell. In the existing CFD code, this scheme is the most
accurate one and it is applicable for general unstructured
meshes as used here, this interpolation scheme assumes that
the interface between two fluids has a linear slope within
each cell and this linear shape is used for the calculation of
the advection of the fluid through the cell interfaces.
In VOF model, the kth fluid's volume fraction in the cell is
denoted as ak then the following three conditions are
possible:
ak= 0: the cell does not contain the kth fluid;
ak = 1: the cell is full of the kth fluid:
0
kth fluid and one or more other fluids.
In present model, there are two phases: vapor and
liquid. When ac, 1, it represents vapor bubble region; when
ai= 1, it is liquid region. When 0
bubble interface exiting in this region. Based on the local
value of ak, the appropriate properties and variables will be
assigned to each control volume within the domain.
(1) Continuity equation
In VOF model, the volume fraction of each fluid ack is
calculated by tracking the interface between different phases
throughout the solution domain. Tracking of the interfaces
between different phases present in the system is
accomplished by solving continuity equations of the phase
volume fraction for phases. A physical interpretation can be
given by various terms in this continuity equation .For the
vapor phase, this equation has the following form:
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
6a Sm
.'+uVa =
at p4 (1)
where S,, is the mass source term of vapor in the bubble
growth process, which can be estimated by Eq. (8) and
(10) through the UDF interface.
(2) Momentum equations
S(pu) + V (puu) = Vp+ V [(Vu +Vu )] +pg +F
(2)
Where F is the surface tension force, which can be
estimated through CSF (continuum surface force) model
(Brackbill, et al., 1992) as following:
= P zVax
(3) Energy equation
t(pE)+ V (u(pE + p)) = V (iVT)+ Sq 4
(4)
The physical properties of p and ?L (effective thermal
conductivity) are shared by both of the two phases. Source
term SEq COntains contributions from all other volumetric
heat sources.
(4) Vaporliquid interfacial density
The properties appearing in the transport equations
are determined by the presence of the component phases in
each control volume (Fluent Inc., 2005). In a twophase
system, for example, if the phases are represented by the
vapor and liquid, and if the volume fraction of the second
phase has been tracked, the density in each cell would be
given by:
p = alpt + (1 a)p, (5)
2.2 UDF procedures
The User Defined Functions (UDF) interface of the
commercial CFD code of Fluent (Fluent Inc., 2005) is
employed to simulate the heat and mass transfer of the
phase change. UDF subroutine allows us to customize
Fluent to fit particular modeling needs. UDF is a function
that user program can be dynamically loaded with the
Fluent solver to enhance the standard features of the code.
They are defined by DEFINE macros which are supplied
by Fluent Inc. They can access data from the Fluent solver
using predefined macros and functions. Present UDF
programs are based on following described bubble
generation theories of microlayer evaporation to simulate
the bubble growth.
3 Bubble growth physical models in subcooled
flow boiling
3.1 Bubbles growth process in subcooled flow
boiling
The channel is vertically arranged, and the fluid flows
vertically from bottom in ydirection. The temperature of
the mainstream region T1 is lower than that of the liquid
saturation temperature Ts corresponding to local pressure.
Tw is the temperature of singleside heated wall which is
higher than Ts. There is a temperature boundary layer close
to heating wall. In this region, the liquid temperature is
higher than saturation temperature but lower than the
Ssuperheated liquid layer
T1 >Ts
subcooled liquid region
( T1
t; vapor bubble
Y IMicrolayer region
heakting wall
flow direction
Fig.1 Schematic diagram of bubble growth model
heating wall temperature, i.e, T, < T1 < Tw. There is a liquid
microlayer region near the bottom of bubble which
temperature is higher than T,. (3 is the contact angle of the
bubble with the heating wall. The bubble on the heating wall
grows up with growing times goes by to, tl, t2.
At the initial time, to, bubble radius is relatively small.
The whole bubble is submerged in superheating liquid layer.
The heat and mass supplied for bubble growth are derived
from the liquid microlayer evaporation at the bottom of
bubble and superheating liquid around bubble surface. As
the bubble diameter increasing, at the time of tl, the top of
bubble pass through the superheating boundary layer, and
enters the subcooled mainstream region. Consequently, the
condensation will occur at the top of the bubble. At this
period, as the mass into the bubble caused by evaporation at
the bottom and the interface exceeds the condensation mass
at the bubble top, the bubble continues to grow up. At the
time of t2, the bubble diameter is greatly larger. Because a
large portion of bubble top region is submerged in
subcooled mainstream liquid, condensation effect becomes
significant and bubble growth rate will be gradually slowed.
When the condensation mass of vapor is approximately
equivalent to the mass adding into the bubble by
evaporation at the bottom and the interface, the bubble
growth rate will slow down to zero and bubble diameter
remains unchanged.
3.2 Bubble growth model
Based on above analysis, there are four assumptions in
current model: i) there is a very small initial bubble existed
on heating wall which radii will be estimated through
classic bubble nucleation theory; ii) the vapor in the
growing bubble is ideal gas; iii) the temperature at the initial
bubble growth time is equal to saturation temperature Ts
corresponding to the working pressure; iv) the pressure in
the initial bubble is based on YongLaplace equation:
P, =P; +
r (6)
where, o is surface tension coefficient, N/m.
3.2.1 Microlayer liquid evaporation model on the bottom of
growing bubble
Experimental results (Tong and Tang, 2001) showed that
there is a liquid microlayer exists at the bottom of bubble
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
vaporliquid interface region can be defined by:
y'm~hip(12)
It should be aware that m is the mass source term of the
continuity equation and the heat flux q is the energy source
term of the energy equation according to VOF model,
Wm3
3.3 Geometry and boundary conditions
Fig.2 Schematic of the geometry structure
In present work, a single bubble growth process has
been simulated in subcooled flow boiling in a narrow
channel. Due to the symmetrical characteristic in the wide
side of the bubbles, it can be simplified to a
threedimensional geometry structure Fig.2 (a). In the
simulation geometry structure model, the vertical
rectangular narrow channel has a inlet section of 2 mmx2
mm, and length of L=10 mm.
Based on the assumption of the bubble growth model,
there is a small bubble on heating wall at the initial stage of
the simulation. The diameter of the small bubble is
calculated by the Han and Griffith correlation equation (Han
and Giffith, 1965).
s(T, T, ) 12 I(T, rlr TrT
r.=3 (Tw T, ) 1 3(Tw T, )Z h. p,
In this paper, rmin is ranged from 20 pLm to 50 pLm in the
subcooling flow boiling.
Mesh around the bubble has been refined to obtain the
details of bubble growth, According to mesh sensitivity
testing, the number of meshes 120000 is suitable for
simulation. The heating wall temperature keeps constant
while the other wall is adiabatic. The inlet of the channel is
set as velocityinlet condition, and the outlet is set as
pressureoutlet condition. The viscous model is the laminar
model.
4 Results and Discussion
4.1 Bubble growth curve
The varying bubble diameters with time in different working
conditions are simulated. Based on the numerical results,
analysis about bubble growth curve in different simulation
condition with p=0.1020.132 MPa ATw=110 K, ATi816 K,
u=0.100.60 ms ', (3=850 has been performed.
Fig.4 depicts the impacts of wall superheating on bubble
growth. With other working conditions remained unchanged,
the bubble growth rate would increase with raising wall
superheated. According to microlayer evaporation model,
higher wall superheating lead to themore microlayer
evaporation intensity. Thus, the bubble growth rate would
during bubble growth, and it is very important to the process.
The convectional heat transfer through this layer can be
ignored due to its small thickness. Thus, the heat flux
through the microlayer liquid can be estimated as:
T, T
6 (7)
where, ?L is the thermal conductivity of the liquid,
Wm2K ; Tw is the temperature of the heating wall, K; T,
is the vapor temperature of growing bubble, K; 8 is the
microlayer thickness under the bubble bottom, m.
The mass transferring from the micro liquid layer to
vapor bubble can be estimated by:
G = q / h (8)
where hf, is the latent heat, kJ/kg; q is the heat flux through
the microlayer, kWm respectively. G is the net mass
flow rate per unit area, kgm2S,
Because Fluent is the CFD soft ware which based
on the FVM(Finite Volume Method),the mass source
which is added to the continuity equation is the mass flow
rate per volume, kgm3S1
cell1 (9)
Where, Aell, is the area of a face in the jth cell, Veli3 is the
volume of the jth cell. m is the mass source to vapor bubble
volume fraction in the jth cell of vapor bubble.
3.2.2 Interfacial bubble heat and mass transfer
model
Vaporliquid interface is quite important to bubble growth.
When the radius of bubble is relatively small, the whole
bubble is submerged in superheated liquid layer. Heat will
transfer to bubble from the superheated liquids through the
bubble interface, and the diameter will be increased
consequently. When the diameter of bubble increases to a
certain scale, the bubble interfacial area will be divided into
two parts; one is submerged in the superheated liquid layer
where the heat would be transferred from the superheated
liquid to the bubble; the other one is submerged in
subcooled liquid where the vapor in the top of bubble would
be condensed and the heat would be transferred from bubble
to the subcooled liquid through the interfacial areas.
Under the conditions of thermodynamic equilibrium, the
mass transfer mechanism representing evaporation
coefficient will be not distinguished from the ones of the
condensation coefficient; and ones often have the
simplification of oeGo, (Mareka and Straub, 2001). For
a curved interface, an analogous equation can be deduced
(Marek, 1996, Marek and Straub, 2001):
G=4e M ,T)P, (10)
43geV27cR J~J
When G>0, the net mass from liquid will enter into the
bubble through the vaporliquid interface; when G <0, the
net mass from bubble will enter into the liquid through the
interface.
thz = a, G J
cellJ(11)
m is the mass transfer rate proportional to vapor bubble
volume fraction in the jth cell of vaporliquid interface
region.
The heat flux due to the phase change in the jth cell of
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
large amount of heat and mass is transferred from the
microlayer to the bubble. The bubble grows fast as a
hemisphere. The value of bubble diameter increases from
0.2 mm to 0.54 mm. After 0.4 ms, the pressure inside
bubble decreases quickly, bubble growth in the stage is the
1.2
p =0.13 MPa
1 u =0.25 ms'
ATw=8 K
0 .8
S0.6
S12 K
0.2 e 16 K
O 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
Time (ms)
Fig.6 The subcooling of mainstream impact on bubble growth
become higher.
p=0.13 MPa
u=0.15 m~s
0. Tf=12 K
0.4 l K 6 4
3 K M 1K
00.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Time (ms)
Fig.4 the wall superheating impact on bubble
growth
p=0.12 MPa
ATw=8 K
0.6Ay=5K
~04%0.1(m/s)
S0.2 er 0.3 (m/s)
&0.6 (m/s)
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Time (ms)
Fig.5 The velocity of mainstream impact on bubble growth
Fig.5 shows the impacts of the liquid subcooling on bubble
growth. With increasing liquid subcooling, bubble growth
rate would slow down significantly at the later bubble
growth stage. It can be explained by that the condensation
intensity will be enhanced at the bubble top, which lead to
a decline in bubble growth rate. It also indicates that the
bubble growth process is a dynamic consequence resulted
from evaporation and condensation.
Fig.6 depicts the effect of the mainstream velocity on
bubble growth. Under the same conditions, the velocity of
mainstream has a mild impact on the bubble growth
comparing with the mainstream subcooling. It can be
explained that with thinner superheated liquid layer under
highspeed condition, bubble growth rate would become
slower than that of low speed circumstance.
4.2 Bubble shape evolution during bubble growth and
detaclunent process
The evolution of bubble shape in subcooling flow boiling
would be impacted by a couple of factors. In this paper, the
section through the central of bubble in XY plane is
selected, in order to make a better analysis of bubble shape
and physical filed around growing bubble.
Figure shows the evolution process.with p=0.102 MPa,
AT,,=10K,ATfll5K,u=0.20 ms ', B=850. Comparing with
Fig.6, the numerical results agree well with typical
consecutive image of bubble evolution in visualization
experiment of subcooling flow boiling. Figure 8 is the
curve of bubble diameter in the time of bubble growth and
detaclunent.
As shown in Fig.7, there is a growing bubble at the
nucleation site on the heating wall. The nucleation site
coordinate is z=x=0.0 m, y=0.005 m. At the initial time of
bubble growth (0.10.4 ms), due to high pressure in the
bubble, bubble growth is in the stage of inertial control. A
ow~n
1 6on.=

bln
n..
.
IEOson.. In70...
.. r
59nun em 05um
., ~ n .. ~ Wo
. x x .. x .
Fig.7 The evolution of bubble shape in bubble growth and
detaclunent processing working condition p=0.102 MPa,
AT =10 K, ATfl5 K ,u=0.20 ms ',B=850
o.s
bubble growth period bubble departure period
0.7
0.6 bubble departue
0. fomnuclearnsit
0.4
bubble departure
,0.3 fom heating wall
0.2
0.1
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8
Time (ms)
Fig.8 The bubble diameter in bubble growth and
detaclunent process on working condition p=0.102
MPa,AT,,=10 K ATf l5 K,u=0.25 ms ',=850
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
1.2 ms, the bubble growth is in the thermal diffusion control
stage. The bubble growth rate is relatively low and stable.
The pressure inside the bubble fluctuates from 1500 Pa to
2000 Pa. From 1.2 ms to 2.6 ms, bubble is in detaching.
Because of the intensive condensation of a departing bubble
caused by subcooled liquid, the bubble diameter
dramatically decreases and the pressure inside bubble drops
rapidly also.
2500 *gague pressure
2250 \ lnsidebubble
S2000
1750
S1500
S1250
750
500
002040608 1 12141618 2 222426
Time (ms)
Fig.10 presents the pressure evolution inside a bubble
during bubble growth and detachment
4.4 Velocity field of growing bubble
Figure 11 the velocity vector field at t=1.0 ms, t=2.8 ms.
According to the analysis of the pressure filed around a
bubble, eddies is formed by the pressure discrepancy around
the bubble. As shown at = 1.0 ms, eddies can be observed on
both sides of the growing bubble. The eddy accelerates
contraction of the vaporliquid interface at bottom of bubble,
which leads to bubble shape changing from ellipsoid to
approximate sphere. As the eddies transfer subcooled liquid
to the heating wall on both sides of bubble, the
microconvection in the vicinity of bubble is enhanced,
which would increase heat transfer significantly. At t=2.8
ms, the difference of the fluid velocity on both sides of the
departing bubble is quite evident, which accelerates bubble
departing from the heating wall.
o is
* 2ms  4ms
0 6ms * 8ms
o as oms *1 2ms
O 0 0001 0 0002 m00003 0 0004 0 0005
Fig.12 the velocity cuve along the centerline of a
bubble at different bubble growth time.
It shows that along the centerline of bubble, from iner side
of the bubble to the mainstream liquid region around the
thermal diffusion control. As the condensation intensity
gradually enhances at the top of bubble, bubble growth rate
becomes much slower than that of former. From 0.6 ms to
0.8 ms, bubble expands slowly in hemisphere shape.
However, with increased bubble diameter, the force
balance acting on the growing bubble has been broken. As
the result, the shape of bubble evolves from a
hemispherical to an ellipse at the time of t= 1.0 ms. At
t=1.2 ms, bubble diameter increases to the maximum value
0.71 mm. After that Bubble begins to detach from
nucleation point. From 1.2 ms to 2.6 ms, the shape of
bubble evolves from ellipse to spherical and bubble
diameter decreased gradually. Bubble remains as a roughly
sphere until it lifts off from the heating wall.
4.3 Pressure distribution evolution during bubble growth
0 0056F
0 0056 ~
xim
Fig.9 the pressure distribution of bubble growth at 0.4
ms and 1.0 ms.
As shown in Fig.9, in the bubble growth process, the
pressure are identical in different region inside vapor
bubble. The pressure inside the bubble is higher than that
of the flow liquid around the bubble. Farther away from
the bubble, the pressure of the flow liquid becomes much
higher. At t=0.4 ms, the value of gauge pressure inside
bubble is 1930 Pa. Away from the top of bubble, there is a
flow liquid region with the minimum pressure, which
gauge pressure value is 1880 Pa. At t=1.0 ms, the value of
gauge pressure inside bubble is 1539 Pa. The minimum
pressure of flow liquid region is at the two sides of the
growing bubble.
Fig.10 presents the pressure evolution inside a bubble
during bubble growth and detachment. As a whole trend,
the pressure inside the bubble is decreased continuously.
From 0.1 ms to 0.6 ms, the bubble growth is in the inertial
control stage. Bubble expands rapidly lead to the
dramatically pressure drop inside bubble. From 0.6 ms to
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
Conclusion
In this paper, a bubble growth model has been proposed for
single bubble in subcooled flow boiling, and a numerical
simulation of bubble growth process is realized by VOF
model and UDF interface through CFD software Fluent.
The factors such as velocity of mainstream, subcooling of
mainstream, the superheating of heating wall and
vaporliquid contact angle have been considered to
understand the mechanism of bubble growth. The bubble
growth curve also has been obtained: the simulation results
agree the experimental results with relative error within
+25%. The results can be concluded as following:
1. The process of bubble growth in subcooled flow
boiling is a dynamic result of evaporation and condensation.
With increasing wall superheating, the rate of the bubble
growth would be accelerated; with higher mainstream
liquid subcooling, the growth rate of later stage would be
easier decreased.
2. The evolution of bubble shape is a quite
complicated process. At the initial time of bubble growth, a
large amount of heat and mass transfers through the
microlayer into the bubble and the bubble grows fast and
expands rapidly as a hemisphere. As the condensation
gradually enhances at the top of bubble, bubble growth rate
becomes much slower than that of before. With increasing
bubble diameter, the change of forces balance acting on a
growing bubble has taken place. Thus, the shape of bubble
evolves from hemispherical to ellipse.
3. During bubble growing, the pressure inside the
bubble will be decreased and it is identical in different
region inside vapor bubble, and it will be higher than that of
the flow liquid around the bubble. Farther away from the
bubble, the pressure of the flow liquid becomes much higher
and a flow liquid region with the mnummum pressure will
exist in the vicinity of growing bubble.
4. Few eddies will exist on both sides of the growing
bubble. The eddies accelerate contraction of the
vaporliquid interface at the bottom of the bubble, which
result to bubble shape changing with bubble growth and
detachment process and enhances the of microconvection
heat transfer in the vicinity of bubble. With increase of
bubble growth time, the velocity inside the bubble gradually
reduces.
5. During bubble growing, a superheated flow liquid
region will be presented near the heating wall. In this region,
the temperature gradient which does not change with the
growth time is very large. With the increase of the distance
from the heating wall, the flow liquid temperature decreases.
Because of the convection heat transfer, temperature of the
subcooled liquid region far from the heating wall increases
with the bubble growth process.
Acknowledgements
The authors are grateful for the support of the National
Science Foundation of China (No.50406012) and the
support of the Researching Foundation of the Laboratory of
Bubble Physics and Nature Circulation of China under
contract of 9140C37101020802.
References
Brackbill,J. U., Kothe,D. B. and Zemach,C., 1992. A
bubble, the velocity increases gradually and at the
vaporliquid interface the velocity fluction appear. From
0.2 ms to 1.2 ms, with increasing of bubble growth time,
the velocity inside the bubble gradually reduces.
4.5 Temperature field distribution of going bubble
0.00481C
Xlm
Fig.13 is the temperature field distribution around a
growing bubble at t=0.6 ms and t=1.0 ms.
As shown in the figure, with increasing distance from
the heating wall, flow liquid temperature decreases. It can
be seen that there is a superheated flow liquid region near
the heating wall. In this region, the temperature gradient is
very high. The coordinate xaxis ranges from 0 pLm to 40
pLm, the flow liquid temperature ranges from 383 K to 373
K. Furthermore, at t=0.6 ms and t=1.0 ms, the flow liquid
temperature gradient in this region does not change during
bubble growth. A part of heat and mass transferring to the
growing bubble derives from this superheated flow liquid
region around growing bubble. Far from the heating wall,
the coordinate xaxis coordinate is larger than 40 pLm, the
flow liquid is subcooled flow liquid, of which the
temperature is lower than the saturation temperature.
Because of the convection heat transfer in the vicinity of
growing bubble, the subcooled liquid temperature increase
during bubble growth. At t=0.6 ms, x=0.00012 m, the
isothermal line is below 360 K, while t=1.0 ms, the
isothermal line is 362 K, temperature increases more than
2 K
Liquid temperature of vaporliquid interface close to
heating wall is higher than that of vapor temperature. Far
from the heating wall, the flow liquid temperature of the
vaporliquid interface is below the growing vapor bubble
temperature. Heat is transfer from vapor bubble to the
subcooled flow liquid, and consequently the condensation
will take place.
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
Marek, R. and Straub, J., 2001. The origin of
thermocapillary convection in subcooled nucleate pool
boiling. International Joumnal of Heat and Mass Transfer 44,
619632.
Marek, R., 1996. Einfluj3 thermokapillarer Konvektion und
inerter Gase beim Blasensieden in unterkuihlter Fluti ssigkeit,
Ph.D. Thesis, Technical University of Munich.
Marek, R. and Straub, J., 2001. Analysis of the evaporation
coefficient and the condensation coefficient of water.
International Joumnal of Heat and Mass Transfer 44, 3953
Mei, R., Chen, W. and Klausner, J.F., 1995.Vapor bubble
growth in heterogeneous boilingI. Formulation.
International Joumnal of Heat and Mass Transfer. 38,
909919.
Mukherjee, A. and Kandlikar,S.G, 2007. Numerical study
of single bubbles with dynamic contact angle during
nucleate pool boiling, Intemnational Joumnal of Heat and
Mass Transfer. 50, 127138.
Nukiyama, S., 1934. The maximum and minimum values of
the heat transmitted from metal to boiling water under
atmospheric pressure, Joumnal of Japan Society of
Mechanical Engineers. 37, 367374.
Prodanovic, V, Fraser, D. and Salcudean, M., 2002.
Bubble behavior in subcooled flow boiling of water at low
pressures and low flow rates, Intemnational Joumnal of
Multiphase Flow, 28, 119
Samuel W.J. Welch, 1998. Direct simulation of vapor
bubble growth, Intemnational Joumnal of Heat and Mass
Transfer. 41, 16551666.
Siedel, S., Cioulachtjian, S. and Bonjour, J., 2008.
Experimental analysis of bubble growth, departure and
interactions during pool boiling on artificial nucleation sites.
Experimental Thermal and Fluid Science 32, 15041511.
Situ, R., Hibiki,T. and Ishii, M., 2005. Bubble liftoff size in
forced convective subcooled boiling flow.Intemnational
Journal of Heat and Mass Transfer, 48, 55365548.
Thomncroft, G.E., Klausner, J.F. and Mei, R., 1998. An
experimental investigation of bubble growth and
detachment in vertical upflow and downflow boiling,
International Joumnal of Heat and Mass Transfer 41,
38573871.
Tong, L.S. and Tang, Y.S., 2001. Boiling heat transfer and
two phase flow(Second edition). pp.4546.
Yongs, D.L., 1984. Numerical simulation of turbulent
mixing by RayleighTaylor instability, Physics. D. 12, 3244.
continuum method for modeling surface tension, Joumnal of
Computational Physics. 2, 335354.
Brian, K. Mori., and Baines, W. Douglas, 2001. Bubble
departure from cavities. International Joumnal of Heat and
Mass Transfer 44, 771783
Chen, D.Q., Pan, L.M. and Yuan, D.W., 2009. An
Experimental Investigation of Bubble Growth in Water
Subcooled Flow Boiling in a Vertical Narrow Rectangular
Channel, submitted to Intemnational Joumnal of Heat and
Mass Transfer.
Eames, I. W., Marr N. J. and Sabir, H., 1997. The
evaporation coefficient of water, a review. International
Journal of Heat and Mass Transfer 12, 2963 2973
Fluent Inc., 2005. Fluent 6.2 User's Guide, Fluent Inc,
Lebanon, USA
Genske, P. and Stephan, K., 2006. Numerical simulation of
heat transfer during growth of single vapor bubbles in
nucleate boiling. International Joumnal of Thermal Sciences.
45, 299309.
Han,C.Y. and Giffith, P., 1965. The mechanism of Heat
Transfer in Nucleate Pool Boiling ,Part 1 Bubble
Initiation ,Growth and Departure Intemnational Joumnal of
Heat and Mass Transfer 8, 887895.
Hazi, G. and Markus, A., 2009. On the bubble departure
diameter and release frequency based on numerical
simulation results, Intemnational Joumnal of Heat and Mass
Transfer. 52, 14721480.
Heo, S., Koshizuka,S. and Oka, Y., 2002. Numerical
analysis of boiling on high heatflux and high subcooling
condition using MPSMAFL. International Joumnal of Heat
and Mass Transfer. 45, 26332642.
Hirt, C.W. and Nichols, B.D., 1981. Volume of fluid
method for the dynamics of free boundaries, Joumnal of
Computational Physics. 39, 201225.
Lee, H.C., and Byung D.O., 2003. Single bubble growth in
saturated pool boiling on a constant wall temperature
surface. International Joumnal of Multiphase Flow 29,
18571874
Li, W.Z., Yan, Y.Y. and Smith J.M., 2003. A numerical
study of the interfacial transport characteristics outside
spherical bubbles and solids. International Joumnal of
Multiphase Flow. 29, 435460.
Liao, J., Mei, R. and Klausner, J.F., 2004. The influence of
the bulk liquid thermal boundary layer on saturated
nucleate boiling. International Joumnal of Heat and Fluid
Flow. 25, 196208.