Group Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Title: 11.7.4 - Binary water droplet collision study in presence of solid aerosols in air
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102023/00296
 Material Information
Title: 11.7.4 - Binary water droplet collision study in presence of solid aerosols in air Collision, Agglomeration and Breakup
Series Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Physical Description: Conference Papers
Creator: Foissac, A.
Malet, J.
Mimouni, S.
Feuillebois, F.
Publisher: International Conference on Multiphase Flow (ICMF)
Publication Date: June 4, 2010
 Subjects
Subject: water droplet
collision
spray
PWR
coalescence
bouncing
aerosol
surface tension
 Notes
Abstract: During the course of a hypothetical severe accident in a nuclear Pressurized Water Reactor (PWR), hydrogen may be produced by the reactor core oxidation and distributed into the containment. Spray systems are used in order to limit overpressure, to enhance the gas mixing in order to avoid hydrogen accumulation, and to wash out fission products (aerosols). The efficiency of these spray systems may depend on the evolution of the droplet size distribution in the containment. Collisions between drops can modify this distribution. Binary droplet collision can lead to coalescence, bouncing or splashing into tiny droplets. Furthermore, Qian & Law (1997) have shown that the collision outcomes depend on ambient conditions, like the gas pressure and mixture composition. Rabe et al. (2009) have confirmed this dependence for typical nuclear reactor conditions. The objectives of the present work are to obtain a full description of water droplet binary collision regimes under atmospheric conditions and to study by comparison the influence of the presence of aerosols in the surrounding gas. For that purpose, aerosols are simulated by latex particles with a geometric mean diameter of 0.1-0.5 μm. Accurate measurements exhibit for the first time ever a bouncing regime with water under atmospheric conditions. It is also observed that transitions between regimes are not affected by the presence of latex aerosols at a concentration of 105 part/cm3.
General Note: The International Conference on Multiphase Flow (ICMF) first was held in Tsukuba, Japan in 1991 and the second ICMF took place in Kyoto, Japan in 1995. During this conference, it was decided to establish an International Governing Board which oversees the major aspects of the conference and makes decisions about future conference locations. Due to the great importance of the field, it was furthermore decided to hold the conference every three years successively in Asia including Australia, Europe including Africa, Russia and the Near East and America. Hence, ICMF 1998 was held in Lyon, France, ICMF 2001 in New Orleans, USA, ICMF 2004 in Yokohama, Japan, and ICMF 2007 in Leipzig, Germany. ICMF-2010 is devoted to all aspects of Multiphase Flow. Researchers from all over the world gathered in order to introduce their recent advances in the field and thereby promote the exchange of new ideas, results and techniques. The conference is a key event in Multiphase Flow and supports the advancement of science in this very important field. The major research topics relevant for the conference are as follows: Bio-Fluid Dynamics; Boiling; Bubbly Flows; Cavitation; Colloidal and Suspension Dynamics; Collision, Agglomeration and Breakup; Computational Techniques for Multiphase Flows; Droplet Flows; Environmental and Geophysical Flows; Experimental Methods for Multiphase Flows; Fluidized and Circulating Fluidized Beds; Fluid Structure Interactions; Granular Media; Industrial Applications; Instabilities; Interfacial Flows; Micro and Nano-Scale Multiphase Flows; Microgravity in Two-Phase Flow; Multiphase Flows with Heat and Mass Transfer; Non-Newtonian Multiphase Flows; Particle-Laden Flows; Particle, Bubble and Drop Dynamics; Reactive Multiphase Flows
 Record Information
Bibliographic ID: UF00102023
Volume ID: VID00296
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 1174-Foissac-ICMF2010.pdf

Full Text

7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010


Binary water droplet collision study in presence of solid aerosols in air


Arnaud Foissac1, Jeanne Malet1, St6phane Mimouni2 and Frangois Feuillebois3

SInstitut de Radioprotection et de Siret6 Nucl6aire DSU/SERAC
Laboratoires d'Etudes et de Mod6lisation en A6rodispersion et Confinement
BP 68, F-91192 Gif-sur-Yvette Cedex, France

2 Electricity de France R&D Division
M6canique des Fluides, Energies et Environnement
6, quai Watier, F-78400 Chatou, France

3 LIMSI-CNRS
BP 133, F-91403 Orsay Cedex, France

arnaud.foissac @ irsn.fr


Keywords: Water droplet, collision, spray, PWR, coalescence, bouncing, aerosol, surface tension




Abstract

During the course of a hypothetical severe accident in a nuclear Pressurized Water Reactor (PWR), hydrogen may be produced
by the reactor core oxidation and distributed into the containment. Spray systems are used in order to limit overpressure, to
enhance the gas mixing in order to avoid hydrogen accumulation, and to wash out fission products (aerosols). The efficiency of
these spray systems may depend on the evolution of the droplet size distribution in the containment. Collisions between drops
can modify this distribution. Binary droplet collision can lead to coalescence, bouncing or splashing into tiny droplets.
Furthermore, Qian & Law (1997) have shown that the collision outcomes depend on ambient conditions, like the gas pressure
and mixture composition. Rabe et al. (2009) have confirmed this dependence for typical nuclear reactor conditions.
The objectives of the present work are to obtain a full description of water droplet binary collision regimes under atmospheric
conditions and to study by comparison the influence of the presence of aerosols in the surrounding gas. For that purpose, aerosols
are simulated by latex particles with a geometric mean diameter of 0.1-0.5 pm. Accurate measurements exhibit for the first time
ever a bouncing regime with water under atmospheric conditions. It is also observed that transitions between regimes are not
affected by the presence of latex aerosols at a concentration of 105 part/cm3.


Introduction

One of the main contributors to the containment early failure
during a PWR severe accident is associated to the presence of
hydrogen within the containment building. The hydrogen
produced by the reactor core oxidation and released from the
reactor coolant system to the containment could mix or
accumulate in different parts of the containment. If the
composition of hydrogen-steam-air mixture reaches a certain
threshold, combustion could occur. In order to prevent this
risk, spray systems are disposed at the top of the containment.
They are used to limit overpressure, to enhance the gas
mixing and avoid hydrogen accumulation, and to wash out
fission products and structure materials which can be
released. The spray systems efficiency may depend on the
evolution of the droplet size and velocity distributions during
their fall. The spray is provided by nozzles attached at
approximately 50 cm intervals at the top of the reactor vessel.
These nozzles are generally used with water at a pressure


drop AP of 3.5 bars each one, producing a mass flow-rate of
approximately 1 kg/s. A single nozzle creates a hollow cone
swirling spray of 60 angle. Collisions between drops from
adjacent nozzles can modify the cloud size and velocity
distributions. Four main different regimes have been pointed
out: bouncing, coalescence, reflexive and stretching
separations. For ambient conditions with water droplets,
coalescence, reflexive separation and stretching separation
have already been studied (Ashgriz & Poo 1990). Roth et al.
(2007) showed that two other regimes can exist at high
Weber number We: stretching with digitations and splashing.
These two regimes lead to a strong fragmentation of initial
droplets. However, bouncing regime with water droplets
under atmospheric conditions has never been identified
accurately. The first part of this paper is concerned with a
characterization of this bouncing regime.
The various collision outcomes depend on ambient
conditions, like the gas pressure and mixture composition, as
shown by Qian & Law (1997). Rabe et al. (2008) confirmed









this dependency under typical gaseous reactor conditions.
However, during their fall in the containment, droplet
trajectories meet fission products and structure material
aerosols. These aerosols when collected by droplets may
modify liquid properties, like surface tension coefficient or
viscosity. Collision outcomes may then change. As a result,
the presence of solid aerosols in the containment could
influence the cloud size. The second part of this study is
therefore concerned with the influence of aerosols on droplet
collision outcomes. The influence of possible variations in
droplet temperature is not considered in this paper.
The first objective of the present work is to obtain a full
description of the water droplet binary collision regimes in
atmospheric conditions, especially, including bouncing.
Results will be compared to existing models developed by
Ashgriz & Poo (1990), Rabe et al. (2010) and Estrade (1999).
The second objective will focus on the influence on the
binary collision outcome of the presence of solid aerosols
which are representative of fission products in the
surrounding gas.


Nomenclature


Diameter (m)
Effective kinetic energy (J)
Surface energy (J)
Dimensionless impact parameter
Droplet mass (kg)
Relative velocity (m.s-1)
Velocity relative to the centre of mass (m.s-1)
C I.i,,sI.i Weber number
Symmetric Weber number
Dimensional impact parameter (m)



Diameter ratio
Density (kg.m3)
Surface tension (N.m1)


Ambient
Atmospheric
Critical
Kinetic
Large droplet
Liquid
Small droplet


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

sprays are generally performed for engine applications, i.e.
for a range of droplet sizes much lower than the one obtained
with nuclear reactor spray nozzles. Thus, it seems adequate
to start with the interaction between two drops in this size
range, with the outlook to apply these results to the
interaction of sprays.
In most studies, four binary collision regimes can be pointed
out: bouncing, coalescence, reflexive separation and
stretching separation (see Figure 1).


1 9.


0*






li
Bouncing Coalescence


*0 0
.. *








Reflexive Streching
Separation Separation


Figure 1. Examples of different collision outcomes:
bouncing, coalescence, reflexive separation and stretching
separation.


The collision process is characterized by three parameters:
the Weber number We, the impact parameter I and the
diameter ratio A:

We Piq (1)


2x
I=
d, +d,

A =d


where di is the large droplet diameter, d, the small droplet
diameter, ur the droplet relative velocity, Pliq the liquid density,
a the surface tension coefficient and x the dimensional impact
parameter which represents the distance from the centre of
the large droplet to the relative velocity vector originating
from the centre of the small droplet when at contact (see
Figure 2).


Background

Few experimental studies (Kim et al. 2009, Ko & Ryou 2005)
concern the interaction of two sprays and the analysis of such
results is rather difficult: when two sprays impinge, the size
distribution of the spray can change locally due to inertia
effects (migration of smaller size droplet in a different way
than the larger droplets) as well as droplet interactions. It is
then difficult to distinguish which of the phenomenon
involves locally the variation of the size distribution. If such
effects are coupled, it is then not obvious to validate a droplet
interaction model. Furthermore, studies of the interaction of


Figure 2. Geometric impact parameter x for droplet
collision.


d
Eeff
Esrf
I
m
Ur

We
Wesym
X
x

Greek
letters
A
P
a

Subsripts
amb
atm
c
kin
I
liq
s









The Weber number We looks appropriate when considering
two equal size droplets or for a droplet and a flat interface.
But for intermediate situations of unequal droplet sizes, there
is no reason to omit properties of the large droplet. To avoid
this restriction, Rabe et al. (2010) give another expression of
the Weber number, called the 'symmetric Weber number'
Wesym. Since the Weber number is the ratio of a kinetic energy
to a surface energy, Rabe et al. (2010) choose these energies
as the sum of kinetic energies of incoming droplets and the
sum of their surface energies:
1 |v l [2 1 2 (4)
2n \\v +-2m,v,\\ (4)
Wesym + axd
aird + oVid/

where ms and m; are the mass of the small and large droplets,
v, and v, the velocities relative to the centre of mass of the
incoming droplets. This equation can then be simplified:
2
Weym = We A (5)
(I + A)(1 + A)

When using the symmetric Weber number Wesym, all
boundaries between collision regime become independent of
the diameter ratio A. For equal size droplets, We, becomes
simply 1/48 of the classical Weber number based on the small
droplet size and the relative velocity.

In PWR spray systems, it is difficult to estimate the value of
the symmetric Weber number We,,. Using the characteristics
of spray nozzle, two regions inside the containment reactor
can be identified: an upper region where sprays interact
directly, with high relative velocity between droplets, leading
to high Wesym of 10-20, and a lower region where droplet
velocities have strongly decreased, leading to a small Wesym.
In the upper part, regimes of fragmentation, as splashing,
stretching with digitations, stretching separation or reflexive
separation may occur (Roth et al. 2007). In the lower region,
bouncing and coalescence should become preponderant.

Transitions between coalescence and separation regimes
have been studied by Ashgriz & Poo (1990) or Rabe et al.
(2010). But, for water droplets, bouncing has never been
accurately characterized under atmospheric conditions.
Therefore, no equation exists for the transition to bouncing
with water droplets. Estrade (1999) proposed an equation for
the transition to bouncing, using experimental data on
ethanol droplets. This equation will be detailed in a following
part, and adjusted on our experimental results.

Many studies, like those of Klaseboer et al. (2000), Chen &
Lee (1999), or Leal (2004), deal with the influence of the
presence of surfactants at the interface between gas and
liquid on binary collision outcomes. They show that these
surfactants can change the way of expulsing the thin gas film
between two approaching droplets. In presence of surfactants,
this gas film is trapped and droplets cannot become enough
close from each other to establish contact and lead to
coalescence. Therefore, adding surfactants entails a decrease
of coalescence, and gives advantage to bouncing. Moreover,
surfactants decrease the surface tension coefficient a. If the
surface tension coefficient is modified, then the Symmetric
Weber number will have a new value as well, since it is


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

inversely proportional to the surface tension coefficient. The
addition of surfactants can also have an influence on
viscosity and density of droplets.

In our study, solid particles replace surfactants. Indeed,
Kissane (2008) gives an interpretation of the experimental
programs FTPO and FTP1 realized on the PHEBUS facility.
PHEBUS-FP is one of the most important research project in
the severe accident field, because this facility allows to study
formation, transport and deposition of fission products and
structure material aerosols across the primary cooling system
and the containment. FTPO and FTP 1 programs are different
by their scenario. Results show that it is impossible to give a
simple typical proportion of aerosols rejected in the
containment. They can be volatile, soluble, or insoluble.
Fission products are in general volatile or soluble, whereas
elements from material structure, mainly represented in
terms of concentration, are insoluble. Kissane (2008)
proposes to represent the aerosol population into the
containment as a log-normal equation with an aerodynamic
mass median diameter between 1.5 and 2 pm with a
geometric standard deviation of 2. Geometric mean diameter
can be approximated at 0.5 pm. Particle concentration in the
containment is estimated between 105 and 106 particles per
cm3.
Molecular properties of such particles are different from
surfactant ones. Surfactants are constituted with a
hydrophobic part and a hydrophilic one, whereas solid
particles do not have a polarity. Furthermore, dimensions of
solid particles that we consider can vary, as far as we know,
from some micrometers size to millimeters. Nanometric
particles have not been measured, whereas a surfactant is of
nanometric size. Does the surface tension coefficient a, or
any other droplet characteristic, change with the addition of
solid particles as we observe when surfactants are added to
water? This question has no answer at the moment, even if
Okubo (1995) compares surfactants as solid particles. In
particular, he shows that surface tension coefficient can
decrease with the addition of solid particles, when the solid
particle concentration reaches a high value which leads to a
crystal-like structured suspension. Under this critical solid
particle concentration value, surface tension coefficient a is
constant.


Experimental Facility

To investigate droplet collision, an experimental method
widely used in many studies like the ones of Ashgriz & Poo
(1990), Estrade (1999), Rabe et al. (2010), or Roth et al.
(2007), consists in producing two calibrated droplet streams
with converging trajectories. Binary droplet collision could
thus be periodically obtained and recorded. Droplet streams
generators are constituted by a steel tube ended by an iridium
plate with a calibrated hole. A piezoelectric cell surrounds
the upward extremity of the tube and allows mechanical
perturbation of the liquid jet according to electrical signal
modulation. Instability also growing along the jet, in relation
with Rayleigh theory, leaves to the break-up of the water
filament and to the development of droplets with same
characteristics (velocity, diameter, direction). The
experimental set-up (see Figure 3) allows the production of
water droplets with diameters from 200 to 700 pm. Droplet









velocities used are between 1 and 19 m.s-1, and droplet
stream collision angles between 10 and 95. Collision
observation is performed by two cameras which record
respectively the front and the lateral views as shown on
Figure 4. The whole experiment is confined in a 1.1 m3
ventilated box.


o Pressuiseddeonised water supply
Particle filter
SGate
hssflow meter
0 Dropletgenerator


STension Amphfier Ivicrometc positonmg system
SFrequencygeneator Xein Lamp
SPulse generator CCDCamera
Optical table 0 Pictue analysis computer


Figure 3. Schematic description of the experimental
set-up for the investigation of binary droplet collision (Rabe
et al. 2008).


rmirm






j


f011


sparative
bem _


Ifiroboscepic
/i ba

I ZtW-


E

1 [


nefVew


Figure 4: Periodic droplet collision recording system.



A shadowgraphic process is used to record a large number of
collisions by lighting the scene with a very short stroboscopic
flash (about 150 ns). Picture sequences are collected and
transmitted to a computer which is in charge of image
processing. An image sequence treatment process, developed
for this purpose in JAVA with the ImageJ software, is then
applied to calculate droplet velocities, impact parameter and
Weber number. This experimental postprocessing is well
adapted to achieve a large amount of data which provide a
complete mapping of droplet collision outcomes in a large
field of Weber number. An effort had also been carried out to
evaluate the uncertainties of measurements for each collision.
It has been assess to reach a maximum of 12 % for Weber
number and 10 % for impact parameter, especially due to the
pixel size of the Sony XCD-SD90 CCD camera.

Experimental campaign has been carried out in order to


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

evaluate the occurrence of many collision outcomes in
presence of solid particles in the surrounding gas. Latex
insoluble particles with a geometric mean diameter of
0.1-0.5 pm have been injected in the box using a TSI
atomizer. Pressurized air enters into the atomizer and creates
a depression that leads to atomization liquid solution with
solid latex particles inside. Silica gel is used to dry the
atomised particles and only keep the latex particles. A
concentration of approximately 105 particles per cm3 has
been reached in the ventilated box using this method. The
concentration has been measured by a Condensation Nucleus
Counter 3025. Particles cross a butanol vapor that condenses
on them, allowing its optical detection.

Finally, particle solutions have been characterized by the
measurement of the surface tension and kinematic viscosity.
Measurements have been carried out with a tensiometer
K10ST using the ring method. Kinematic viscosity
measurements have been carried out with a Viscoclock. This
method estimates the time for a liquid to go through a
capillary.


Results in a clean environment under ambient
gaseous condition

A first experimental campaign has been carried out in a clean
environment, under ambient gaseous condition (see Figure 5).
Particle concentration has been measured at approximately
10 particles per cm3 of gas in the ventilated box, what is 1000
times less than the typical aerosol concentration in the
atmosphere. Droplets diameter were about 300 pm, and the
range of Symmetric Weber number studied has been fixed
from 0.2 to 3 (10 to 150 for the classical Weber number).


1
0,9
0,8
'-, 0,7
0,6
10,5
0,4
0,3
0,2
0,1

0


0,5 1 1,5 2
Symmetric Weber number We


Figure 5. Experimental results of different outcomes for
collisions of water droplets of same diameter (d=300 pim,
A= 1, Pant, Tamb) obtained in a clean environment.


Accuracy of the experimental results is very good and
transitions between regimes are well defined.

Ashgriz & Poo (1990) proposed equations to model the
transition between coalescence and stretching separation on
the first hand, and between coalescence and reflexive









separation on the second one. Equations of these transition
curves are:


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

Curves proposed by Ashgriz & Poo (1990) and Rabe et al.
(2010) have been plotted on the Figure 6.


Wecoal stretch


(7(1 +A2)23 -4(1+A2))(1+A3)
4(72 + A61)(1 + A2)


(1+dA3)[3(1 +A)(1-I)(A3Ds +(,1)]1/2
Wecoa/ref 3(1+ A2)[(1+ A3) -(1- 2)(, +A3, )


Where l, 2', s( and are function of (I,A):


I
0.9
0.8
S0.7
|0.6 |-- -
0.5 n-- *. -


l 0.4
0.3


771 = 2(1 )2 (1 -2)1/2 1
72 = 2(A- )2 (A- 2)11/2 -A3

I= -I(1+A)
2


1 (2A- 2 )(A+)
Z. = 4
s 2 -
S (3A Z)
4A


1 (2-_-2)(1 +-)
4L
1= 2
-(3 -)
4


d,
for w >



2

for C(>-
2

for w
2


with z = (1 I)(1 + A). co is the width of the overlapping

interaction region : o = (1 I).



Rabe & al. (2010) proposed simple formulae expressing the
boundaries of collision outcomes using the properties of the
Symmetric Weber number.
For the transition between coalescence and reflexive
separation, Rabe et al. (2010) considers the competition
between an efficient kinetic energy Eeff and the surface
energy Es5,. Experimental results allow Rabe et al. (2010) to
determine the final equation for the transition between
reflexive separation and coalescence:


SStretching separation
Reflexive separation

' .. .:-

b r.- : -.


-z '-*-^- *? *




0 0.5 1 1.5 2 2.5 3
Symmetric Weber number Weym


Figure 6. Experimental results of different outcomes for
collisions of water droplets of same diameter (d=300 pm,
A=1, Patm, Tamb) obtained in a clean environment and
transition curves by Ashgriz & Poo (1990) and Rabe et al.
(2010).

Transition between coalescence and reflexive separation is
well represented by the Ashgriz & Poo equation, whereas
transition between coalescence and stretching separation is
underestimated. Transition curves proposed by Rabe et al.
(2010) are in good agreement with our experiments.

Estrade (1999) proposes an expression for the transition to
bouncing. He assumes that the criterion for bouncing is that
the droplet initial kinetic energy of deformation does not
exceed the energy required to produce a limit deformation.

Based on an energy balance as the approach of Rabe et al.
(2010), Estrade (1999) proposed an equation for the
transition to bouncing:


A(1+ A )(4F- -12)
coal boun _(1 2)


I = 3.59 1 0.45
We y


For the transition between coalescence and stretching
separation, Rabe et al. (2010) uses the same approach. It
provides an expression for the critical Impact Parameter at
the transition between stretching separation and coalescence:


W,trech + 8Westrech Wesym Westrech
4We


From Rabe et al. (2010) experimental results, this expression
gives a correct representation of the boundary curve between
the two regimes for Wetrch = 0.53.


with = + +1=
K:2c = 2/2/3 + 2 +



fraction of volume interaction.


and X is the


Estrade (1999) estimates that the parameter pc is equal to
0.265 for ethanol droplets of same diameter.

This curve proposed by Estrade (1999) is shown on the
Figure 7, for water droplets of approximately 300 pim, at Patm,
in a clean environment. Transition to bouncing is well
represented by the Estrade equation when (pc is equal to
0.458848. The position of the curve is very dependent on the
value of this parameter.






7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010


Sx Stretching separation
0.9 Reflexive separation

0.7 ---- --
o-.7 ._ ._' ,_- -

S0.6 $ 4 ; -
0.4 -- - - --




0.2
0.1


0.5 1 1.5 2
Symmetric Weber number We ym


2.5 3


Figure 7. Experimental results of different outcomes for
collisions of water droplets of same diameter (d=300 pm,
A=1, Patm, Tamb) obtained in a clean environment and
Estrade equation for the transition to bouncing.


Results with solid particles in the surrounding gas

The presence of solid particles in water may have an
influence on the droplets characteristics, as the surface
tension coefficient or viscosity. Measurements of surface
tension a has been realized with a latex particle suspension
(see Figure 8). Surface tension a remains constant until the
maximum tested concentration of 10 g.11. It means that the
critical concentration to make decrease the surface tension
coefficient presented by Okubo (1995) is not reached in our
case.


90

- 85

80

75

0 70

65

S60

m 55


Op0 2,00 4O0 6,00 8,00 10,00
Latex particle concentration (g.11)

Figure 8. Surface tension of a latex particle solution.



Figure 9 shows that the kinematic viscosity of a latex particle
solution does not change significantly between 0 and 10 g.11.
In the case of a PWR severe accident, a maximal
concentration of 1 g.1-1 is expected. Even if at this
concentration, no influence on the surface tension and the
kinematic viscosity has been demonstrated, binary collision
experiments are necessary to evaluate whether latex particles
can however have an effect. For example, as for surfactants,
the film drainage between colliding drops can be affected by
the presence of solid particles at the air-water interface.


-
-- 1,10


S1,08 -


* 1,06
C-
w I
o 1,04


. 1,02


1,00
0,00


2,00 4,00 6,00 8.00
Latex particle concentration (g.1-1)


Figure 9. Kinematic viscosity of a latex particle solution.


For the experimental campaign in a dirty environment, latex
particles have been injected in the surrounding gas of the
collision, with a concentration of approximately
105 particles per cm3, what is 10 times more than the typical
aerosol concentration in the atmosphere (see Figure 10).


300000

250000

2 200000 collision
.S measurements
I 150000

S100000

50000


0 2000 4000 6000 8000 10000
Time (s)

Figure 10. Evolution of latex particle concentration in the
ventilated box during the dirty environment test.


During their course, droplets can collect latex aerosols before
the collision. Results of the binary collision experiment in
presence of latex particles are presented on Figure 11.


x Stretching separation
0,9 -- Reflexive separation
0, -- -
*0,7*
0,6

|065 - -- -i -

0,3
0,2 :


0, ---
0 0,5 1 1,5 2 2,5 3
Symmetric Weber number We y


Figure 11. Experimental results of different outcomes for
collisions of water droplets of same diameter (d=300 pm,
A= 1, Patm, Tamb) obtained in a dirty environment
([latex] gas=105 particles per cm3).


hU


U-. 1









Even if coalescence domain seems to be reduced (see
Figure 12), it is difficult to conclude to an effect of the latex
particles on collision issue. The collection of solid aerosols
may fix the air-water interface, and may make harder the
droplet surface. According to the uncertainties on the
Symmetric Weber number and on the Impact Parameter of
about 10 %, the influence of the solid particles can be masked.
Indeed, if water solid particle concentration reaches a very
high value, droplet can be considered as a hard sphere,
entailing a decrease of the coalescence domain and an
increase of the bouncing domain.

Moreover, distance between the droplet generator and the
collision point is about 10 cm, what is very few to collect
enough aerosols. In PWR containment, droplet's path can be
very long, and can reach many meters, which leaves enough
time to collect more aerosols than in our experiment.


S- Coalesence %Vithout Aerosots
o Coalescence Wtho Aerosols
0.9
0. OCoalescence thAerosols
0.8


0.7



0.2
0.1

0 0.5 1 1.5 2 2.5 3
Symmetric Weber number We,

Figure 12. Experimental results of coalescence regime for
collisions of water droplets of same diameter (d=300 pm,
A=1, Patm, Tamb) with and without aerosols.


Finally, droplets from a same generator follow each other
very closely as seen on the Figure 1. This configuration does
not facilitate the collection of aerosols and is not
representative of the droplet cloud inside PWR containment.
To conclude to a real effect of the presence of solid aerosols,
test with higher concentration of particles should be
performed. Solid particles could also directly be put in the
water supply to control precisely the solid particle
concentration. Unfortunately, the presence of solid particles
in the water did not allow us to create stable droplets with the
piezoelectric generators, so that these experiments did not
lead to significant results.


Conclusions

Binary collision outcomes, with or without solid particles in
the surrounding gas, have been compared, thanks to a regime
map depending on the impact parameter and the Symmetric
Weber number. Transition curves proposed by Ashgriz & Poo
(1990) and Rabe et al. (2010), in a clean environment, have
been compared to our experimental results. Models
developed by Rabe et al. (2010) are in a good agreement with
our results, whereas Ashgriz & Poo (1990) transition to


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

stretching underestimates the experimental results, as it has
been noticed by Rabe et al. (2010).
For the first time, bouncing regime is shown to exist with
water droplets under atmospheric conditions. Accuracy of
the setup and post-processing allows us to work on the
transition to bouncing. The Estrade semi-empirical equation
for the transition to bouncing can be fitted well with our
experimental data. However, this equation shows a strong
sensibility to experimental parameters. Therefore, a more
physical interpretation of the phenomenon, inspired by the
approach of Rabe et al. (2010) for the other regimes, could
lead to a simpler equation based on less parameters, and
consider the dependency of the bouncing regime with the gas
pressure and the temperature.
These first experiments of binary collision in presence of
solid particles have not shown a strong influence compared
to the clean case. Even if a small decrease of the coalescence
domain exists, this difference is within the error interval. It is
thus difficult to conclude to an effect of the solid particles in
gas on the collision outcome. However, if an effect really
exists, these experimental results show that the collision
outcomes are not so dependent with the presence of solid
particles, since neither the surface tension coefficient nor the
kinematic viscosity are modified. Experiments with higher
solid particle concentration could allow us to point out an
effect.


Acknowledgements

The authors would like to thank J. Alengry and D. Birraux, in
internship from Paris XI University, for the technical support
and discussions on this subject.


References

Ashgriz, N. & Poo, J.Y. Coalescence and separation in binary
collisions of liquid drops. Journal of drop size and velocity
distributions, Vol. 221, 183-204 (1990)

Chen, L.H. & Lee, Y.H. The effects of a surfactant on the
mass transfer in spray-tower extraction model. Chemical
Engineering Journal, Vol. 73, 77-81 (1999)

Estrade, J.P. Experimental investigation of dynamic binary
collision of ethanol droplets-a model for droplet coalescence
and bouncing, International Journal of Heat and Fluid Flow,
Vol. 20, 486-491 (1999)

Kim, S., Lee, D.P. & Lee, C.S. Modeling of binary droplet
collisions for application to inter-impingement sprays.
International Journal of Multiphase Flow, Vol. 35, 533-549
(2009)

Kissane, M.P. On the nature of aerosols produced during a
severe accident of a water-cooled nuclear reactor. Nuclear
Engineering and Design, Vol. 238, 2792-2800 (2008)

Ko, GH. & Ryou H.S. Droplet collision processes in an
inter-spray impingement system. Journal of Aerosol Science,
Vol. 36, 1300-1321 (2005)






7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

Klaseboer, E., Chevallier, J.P., Gourdon, C. & Masbernat, O.
Film drainage between colliding drops at constant approach
velocity: experiments and modelling. Journal of Colloid and
Interface Science, Vol. 229, 274-285 (2000)

Leal, L.G. Flow induced coalescence of drops in a viscous
fluid. Physics of Fluids, Vol. 16, 1833-1851 (2004)

Okubo, T. Surface tension of structured colloidal suspensions
of polystyrene and silica spheres at the air-water interface.
Journal of Colloid and Interface Science, Vol. 171, 55-62
(1995)

Qian, J. & Law, C.K. Regimes of coalescence and separation
in droplet collision. Journal of Fluid Mechanics, Vol. 331,
59-80 (1997)

Rabe, C., Berthoumieu, P., Malet, J. & Feuillebois, F. Binary
water droplet collision under various gaseous conditions
typical for nuclear reactors. ILASS-Europe, Como, Italy
(2008)

Rabe, C., Malet, J. & Feuillebois, F. On the influence of
droplet coalescence in spray systems for containment safety.
13th International Topical Meeting on Nuclear Reactor
Thermal Hydraulics, Kanazawa city, Japan (2009)

Rabe, C., Malet, J. & Feuillebois, F Experimental
investigation of water droplet binary collisions and
description of outcomes with a symmetric Weber number.
Physics of Fluid (2010)

Roth, N., Rabe, C., Weigand, B., Feuillebois, F., & Malet, J.
Droplet collision outcomes at high Weber number.
ILASS-Europe, Mugla, Turkey (2007)




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs