Group Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Title: 16.6.1 - Bubble and liquid-phase motion in a decaying turbulence field of oscillating-grid turbulence
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102023/00406
 Material Information
Title: 16.6.1 - Bubble and liquid-phase motion in a decaying turbulence field of oscillating-grid turbulence Fluid Structure Interactions
Series Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Physical Description: Conference Papers
Creator: Imaizumi, R.
Saito, T.
Publisher: International Conference on Multiphase Flow (ICMF)
Publication Date: June 4, 2010
 Subjects
Subject: oscillating-grid
bubble-liquid interaction
decaying turbulence
bubble swarm
PIV
 Notes
Abstract: Quantitatively clarifying the complex flow structure of bubbly flows is necessary for optimal design and control of industrial plants (e.g. Heat exchanger, Chemical reactor, Bubble column and so on). In the flows, there are two-way interactions between bubbles and liquid-phase turbulence. These interactions are dominant factors affecting the flow structure. Hence the objective of this study is to clarify the turbulence modulation induced by interaction between a bubble swarm and ambient liquid-phase turbulence. In the present study, we employed oscillating-grid turbulence and a bubble launcher equipped with an audio speaker in order to generate both ideal and comprehensible turbulence and controlled bubbles. At first, we generated homogeneous isotropic turbulence by using the oscillating-grid (oscillating frequency: 4 Hz, stroke: 40 mm) in cylindrical acrylic pipe (height: 600 mm, inner diameter: 149 mm) filled with ion-exchanged and degassed water. The liquid-phase motion shifted into decaying turbulence after stopping the oscillating-grid in arbitrary time. A few moments later, a bubble swarm was launched into the decaying turbulence using the bubble launcher. We measured the bubble motion by visualization and the liquid phase motion by a PIV system with LIF method equipped with high-speed video camera. The experiments were carried out at three main conditions as follows. The first one was performed under the decaying turbulence without a bubble swarm (Condition-O). The second one was performed under turbulence induced by the bubble swarm which is launched in the stagnant water (Condition-B). The third one was performed under the decaying turbulence with the bubble swarm (Condition-OB). First, the modulation of the bubble trajectory is discussed. When the bubbles are launched into the decaying turbulence field, the existence region of bubbles was expanded. On the other hand, the vertical component of bubble velocity under Condition-B and Condition-OB were hardly changed. Second, the modulation of the decaying rate of turbulence in the flow field is discussed. The decay rates under Condition-OB became larger than that under Condition-O. We consider this phenomenon is caused by the characteristic bubble wake.
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: VID00406
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 1661-Imaizumi-ICMF2010.pdf

Full Text

Paper No 7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010


Bubble and liquid-phase motion in a decaying turbulence field
of oscillating-grid turbulence


Ryota Imaizumi* and Takayuki Saitot

Graduate School of Engineering, Shizuoka University,
3-5-1 Johoku, Naka-ku, Hamamatsu, Shizuoka 432-8561, Japan
t Graduate School of Science and Technology, Shizuoka University,
3-5-1 Johoku, Naka-ku, Hamamatsu, Shizuoka 432-8561, Japan
E-mail: ttsaito@ipc.shizuoka.ac.jp


Keywords: Oscillating-grid, Bubble-liquid interaction, Decaying turbulence, Bubble swarm, PIV


Abstract

Quantitatively clarifying the complex flow structure of bubbly flows is necessary for optimal design and control of
industrial plants (e.g. Heat exchanger, Chemical reactor, Bubble column and so on). In the flows, there are two-way
interactions between bubbles and liquid-phase turbulence. These interactions are dominant factors affecting the flow structure.
Hence the objective of this study is to clarify the turbulence modulation induced by interaction between a bubble swarm and
ambient liquid-phase turbulence. In the present study, we employed oscillating-grid turbulence and a bubble launcher
equipped with an audio speaker in order to generate both ideal and comprehensible turbulence and controlled bubbles. At first,
we generated homogeneous isotropic turbulence by using the oscillating-grid (oscillating frequency: 4 Hz, stroke: 40 mm) in
cylindrical acrylic pipe (height: 600 mm, inner diameter: 149 mm) filled with ion-exchanged and degassed water. The
liquid-phase motion shifted into decaying turbulence after stopping the oscillating-grid in arbitrary time. A few moments later,
a bubble swarm was launched into the decaying turbulence using the bubble launcher. We measured the bubble motion by
visualization and the liquid phase motion by a PIV system with LIF method equipped with high-speed video camera. The
experiments were carried out at three main conditions as follows. The first one was performed under the decaying turbulence
without a bubble swarm (Condition-O). The second one was performed under turbulence induced by the bubble swarm which
is launched in the stagnant water (Condition-B). The third one was performed under the decaying turbulence with the bubble
swarm (Condition-OB). First, the modulation of the bubble trajectory is discussed. When the bubbles are launched into the
decaying turbulence field, the existence region of bubbles was expanded. On the other hand, the vertical component of bubble
velocity under Condition-B and Condition-OB were hardly changed. Second, the modulation of the decaying rate of
turbulence in the flow field is discussed. The decay rates under Condition-OB became larger than that under Condition-O. We
consider this phenomenon is caused by the characteristic bubble wake.


1. Introduction

Gas-liquid two-phase flows are encountered in many
industrial plants such as chemical reactors, heat exchangers
and GLAD system (e.g. Saito et al., 1999, 2000 and 2001).
Hence, it is very important for such industrial applications
to clarify the flow characteristics in order to improve their
efficiency and safety. However, the flows have multi-scale
phenomena and complex structure. In the flows, the
interaction between the bubble motion and the liquid
motion is one of key mechanisms. Both motions affect
each other. The turbulence modulation resulting from the
interaction should be experimentally and comprehensibly
clarified.
The authors have been investigating the interaction
between bubble swarms and liquid-phase flow. In order to
clearly extract the interaction between bubbles and
liquid-phase turbulence and systematically clarify the
phenomena, isotropic turbulence without mean flow is the
most appropriate. In various studies of homogeneous


isotropic turbulence an oscillating-grid was employed.
Thompson and Turner (1975) quantified oscillating-grid
turbulence in a rest water column. Hopfinger and Toly
(1976) researched on homogeneous isotropic turbulence
and proposed various equations. Hopfinger, et al. (1982)
applied the oscillating-grid to a rotating tank experiment,
and observed the interaction between the vortex produced
by the rotation and the homogeneous isotropic turbulence.
Qi Zhou and Nian-Sheng Cheng (2009) discussed a
two-way interaction between settling behavior of various
types of small solid particles and the ambient turbulence
field generated by the oscillating-grid. They reported that
the reduction of the settling velocity cannot be simply
correlated to turbulence intensity; however, the fluctuation
of the settling velocity in the oscillating-grid turbulence is
significant as compared to the still water case. Morikawa
and Saito (2008) reported the turbulence modulation
induced by the interaction between the flow induced by
bubble swarms and the ambient isotropic turbulence of
liquid-phase by using LDA. In this study, completely






Paper No


controlled bubble swarms were launched in decay process
of the homogeneous isotropic turbulence. The bubble
motion and turbulence modulation affected by
bubble-liquid interaction were investigated. The
liquid-phase motion was obtained by PIV analysis and the
bubble motion was measured by visualization.

2. Nomenclature

d grid bar thickness (mm)
f frequency of oscillating-grid (Hz)
M mesh size (mm)
S stroke of oscillating-grid (mm)
w vertical velocity (mms-')
u horizontal velocity (mms ')
Re Reynolds number (-)
D bubble diameter (mm)
We Weber number (-)
r curvature radius (mm)
Asp aspect ratio (-)

Greek letters
K curvature (-)

Subscripts
b bubble
mean mean
eq equivalent
rms root mean square
R right
L left

3. Experimental Setup
3.1 Experimental setup
The liquid motion was obtained by PIV analysis and the
bubble motion was measured by visualization. The
experimental setup used in this study is illustrated in Figs.
1 and 2. The vessel was made of acrylic pipe (149 mm in
inner diameter and 600 mm in height) covered with a
rectangular acrylic water jacket removing the influences of
refraction and deformation of the image. The origin of the
coordinate system (x, y, z) was set at the center of the
bottom of the vessel. The vessel was filled with purified
water deionizedd, filtrated and degassed tap water) up to a
depth of 580 mm. We generated homogeneous isotropic
turbulence in the liquid phase by using an oscillating grid
in the stagnant water. The grid was tightly bolted at the end
of the oscillating bar. The grid (the mesh size M = 18 mm,
thickness of the grid bar d = 4 mm) was driven by
computer-controlled servomotor (the frequency f = 4 Hz,
the stroke S = 40 mm). The neutral position of the grid
oscillation was at z = 392 mm. The neutral position was
redefined as the new origin of z'. At this point, z' axis was
redefined downward.
The bubble swarms were launched using a bubble
launch device equipped with three hypodermic needles,
hence uniform bubbles were repeatedly formed and
launched at controllable launch timing (Morikawa and
Saito, 2008; Saito, et al., 2010; Kariyasaki and Ousaka,
2001; Sanada, et al., 2005).
Figure 1 shows the experimental setup used in the


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

PIV/LIF experiment to obtain liquid-phase motion. A
CW-YAG laser (wavelength 532 upm), and laser
fluorescence particles (diameter 50 upm, excitation
wavelength 532 nm and emission wavelength 570 nm)
completely dispersed in the water were used A CW-YAG
laser (wavelength 532 upm), and laser fluorescence particles
(diameter 50 pm, excitation wavelength 532 nm and
emission wavelength 570 nm) completely dispersed in the
water were used.
The laser beam illuminated the measurement area as a
laser-light sheet through rod lens; the measurement area
was positioned at the center-plane of the cylindrical vessel.
The measurement area was filmed by a high-speed video
camera (FAST CAM SA1.1-4SH, Photron). The
measurement area was x = 24.8 24.8 mm, y = 0 mm
and z' = 100 149.7 mm. The spatial resolution was 1024
x 1024 pixel2 (48.5 pm/pixel), and the frame rate was 250
fps. In order to extract the contour of the bubbles, a
ring-shaped continuous red-LED array (wavelength 660
nm) was placed at the opposite side of the camera.
The PIV measurements were performed under the
following five experimental conditions; Condition-O:
liquid-phase decaying oscillating turbulence formed after
90 sec oscillation of the grid and stopped oscillation,
Condition-B: only the bubble swarm launched in stagnant
water, and Condition-OB: the bubble swarm launched into
the decaying turbulence. A single bubble swarm consisted
of three bubbles. We set up two conditions at launch of the
bubble swarm in the present experiments. Under
Condition-B(1) and -OB(1), the single bubble swarm is
launched just one time. Under Condition-B(3) and -OB(3),
the single bubble swarm is launched three times
consistently, and the launch frequency is 8Hz.
Figure 2 shows the experimental setup for visualization
of each bubble motion under the experimental conditions:
Condition-B(1), Condition-OB(1), a single bubble
launched in the stagnant water and a single bubble
launched into the decaying turbulence. The measurement
area was captured by two high-speed video cameras
(Phantom v9.0, Vision research), i.e. the measurements
were performed on x-z plane and y-z plane.
The schematic diagrams of these experimental
conditions are shown in Fig. 3. From these results, the
bubble-liquid interactions were quantitatively discussed.

3.2 Bubble launch device
We used the bubble launch device as shown in Fig. 4.
The device consists of two pressure controllers, a function
generator, a power amplifier and three audio speakers.
Each audio speaker pushed pure air in the tube connecting
the speaker cone and hypodermic needle (0.65 mm in outer
diameter, 0.40 mm in inner diameter) under a strictly
controlled air flow rate using these pressure controllers. It
precisely and repeatedly formed uniform bubbles at each
hypodermic needle. The bubble launch interval was
arbitrarily controlled by function generator. The edges of
all hypodermic needles were set toward x-direction. These
needles were able to repeatedly release the bubbles of the
same initial shape. In addition, the trajectories were highly
controlled (Saito, et al., 2010). The average equivalent





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


diameter of bubbles launched from this device was 2.9
mm.


0 OB(1)


, .. 1 V-






OB(3)


f Water jacket

y ~Cylindrical vessel
Grid
(a) Controller, (b) Servomotor, (c) Grid, (d) cylindrical
vessel, (e) Rectangular water jacket, (f) LED light, (g)
YAG laser, (h) Rod lens, (i) High-speed video camera,
(j) Hypodermic needles, (k) Bubble launch device
Fig. 1: Experimental setup for PIV measurement.




(a) ..... (b)
(c)
(d)
S


(f)
z(w)



y (v)
(g) (h)



(a) Controller, (b) Servomotor, (c) Grid, (d) cylindrical
vessel, (e) Rectangular waterjacket, (f) Flat light sources,
(g) High-speed video cameras, (h) Hypodermic needles,
(i) Bubble launching device
Fig. 2: Experimental setup for visualization of bubbles.


time [sec]


0 1
Grid worked. 't


Grid is stopped. Bubbles are launched.

Fig. 3: Experimental conditions.

- Air (
- - -Signal I

S(b)
(b)


a) Function generator, (b) Power amplifier, (c) Audio
speakers, (d) Pressure controller No. 1, (e) Pressure
controller No. 2, (f) Cylinder of pure air
Fig. 4: Bubble launch device.

3.3 PIV Measurement
In PIV measurement of the gas-liquid two-phase flows,
excessive amount of the tracer particles affect the bubble
motion due to surface contamination. Hence, we should
carefully decide number density of the tracer particles in
order to measure both original bubble motion and liquid
motion. However, small number density of the tracer
particles induces an increase in error vectors, and
inevitable limitation of spatial resolution. For the purpose
solving this problem, in the present study, we employed
the recursive cross correlation method for PIV algorism.
The initial interrogation area, the downsized final
interrogation area, and the overlap of the interrogation area
were set arbitrary to achieve the optimal PIV condition.
Under Condition-B and -OB, the scattered light from the
bubble interfaces increases error vectors in the


Paper No


(g)


B(1)



B(1)


------ ............ -- ........---- ....... I-------- ..... ... ... ...... .....








liquid-phase velocity field. In order to reduce the error
vectors, we employed the threshold processing called "2o
processing" (mean vector +/- double of standard deviation
of the vectors) well-known in LDA measurement of
gas-liquid turbulent flows (Mudde and Saito 2001). The
standard deviation was calculated from the PIV dataset. We
analyzed four sets of PIV-image dataset (consecutive 5456
images). From time-series velocity data, datasets at each
coordinates, instantaneous velocities and averaged velocity
at each point were calculated.

4. Results and Discussion
4.1 Homogeneous isotropic turbulence
The profiles of the ratio of w,.m and u,, which are
calculated from PIV results under oscillating grid are
shown in Fig. 5. The target area of the PIV measurement
was positioned at the centerline of the vessel (i.e.
measurement area of x = -12 mm 12mm). The obtained
data were averaged in overall measurement time. The ratio
ranges from 0.7 to 1.3 at areas h3 and h4 as shown in Fig.
4. This result well agrees with the result of LDV
measurement by Morikawa and Saito (2008). Hopfinger
and Toly (1976) discussed homogeneous isotropy under
oscillating-grid turbulence and described that the
homogeneous isotropy is satisfied in the range in which the
ratio of wrmandum, is 0.7 1.3. Therefore, homogeneous
isotropic turbulence is considered to be formed
satisfactorily in the present oscillating-grid.


tI



hi h3 h5



h2 h4 h6
24mm

1 35mm

MS"


hi 20mm

E 60mm

E 100mm
h3

140mm

180mm
h5
220mm
h6 55
255mm


Fig. 5: Ratio of w,, and Urm, obtained by PIV result.

4.2 Bubble motion
4.2.1 Trajectory of bubbles in a swarm
The bubble shape examined in this study is categorized
into a type of oblate ellipsoidal. Shadow images of bubble
were filmed by high-speed video cameras. The spatial
resolution was 1200 x 1632 pixel2 (215 pim/pixel). The
frame rate was 500 fps. In total, 30 bubbles (3 bubbles x
10 times) under Condition-B(1) and -OB(1) are analyzed.
The average equivalent diameter of bubble was 2.9 mm.
When the bubbles were launched from the hypodermic
needles in the stagnant water, they moved in 2-dimensional


Paper No


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

zigzag motion and reproducible trajectory in the initial
region (Sone and Saito, 2008). The edge of hypodermic
needle was set toward x-direction. Hence the trajectories of
the bubbles observed from y-z plane became rectilinear.
The trajectories of the bubbles (member bubbles of the
swarm) gravity-center are shown in Fig. 6. The raw
shadow images were captured from both x-z plane and y-z
plane. The trajectories under Condition-OB(1) were almost
the same as those under Condition-B(1) in x-z plane. The
trajectories under Condition-B(1) were nearly rectilinear in
y-z plane. On the other hand, those under Condition-OB(1)
were different in y-z plane. The existence region of the
bubbles under Condition-OB(1) was expanded in
y-direction. From this result, the bubbles were considered
to be stimulated from 2-dimensional zigzag motion to
3-dimensional motion by the effect of the ambient
liquid-phase turbulence.
The average vertical-velocity component of the bubbles
at the gravity center is obtained from the results. The
velocity component under Condition-B(1) and that under
Condition-OB(1) are 276 mm/s and 283 mm/s,
respectively.
The properties of the bubbles are listed in Table 1.
Poorte et al. (2002) reported that the reduction in average
rising velocity of the bubbles depends on the structure of
the turbulence and the rising velocities of small bubbles
(0.3 1.9 mm in diameter) decrease by up to 35% in grid
turbulence compared with the quiescent conditions. They
described that the ambient turbulence causes the decrease
in bubble velocity in their other report (1997). The relative
velocity between the bubbles and the surrounding liquid
induces the lift force; as a result the bubbles are
transported toward the region of the largest relative
velocity. The drag force and the lift force increase
downwardly. Accordingly, the velocity of the bubbles
decreases. In this study, the bubble motion under
Condition-OB(1) was shifted to 3-dimensional motion,
therefore the path length under Condition-OB(1) is
considered to be longer than that under Condition-B(1).
The vertical velocity component of the bubble under
Condition-OB(1) was about 2.5% larger than that under
Condition-B(1). Taking the other velocity components into
account, the bubble velocity under Condition-OB(1)
increases by 7 to 8 %. The equivalent bubble diameter
between Condition-B(1) and -OB(1) are completely the
same. Hence the buoyancy forces of bubble under
Condition-B(1) and -OB(1) are also the same. The size of
the bubble examined in this study is larger than that of
Poorte. Hence, the shape deformation of the bubble in this
study is larger than that of Poorte. This phenomenon is
considered to be caused by variation of the drag force
owing to the bubble shape deformation and the interaction
between the bubble shape and the ambient liquid-phase
turbulence.

Table 1: The properties of the bubbles.
Condition De [mm] wb [mm/s] Reb [-] We[-]
B(1) 2.9 276 790 3.07
OB(1) 2.9 283 810 3.22


0.520
0.440
0.360
0.280
0.200
Wrm/U





Paper No


350


300


S250


200


150
-40


350


300


250


200


-20 0
x [mm]
(a) x-z plane


20 40


-40 -20 0 20 40
y [mm]
(b) y-z plane
Fig. 6: Trajectory of gravity-center of the bubbles in the
swarm. (z > 150)

4.2.2 Single bubble motion in the decaying
turbulence
As described in the previous section, the transition from
2D zigzag motion to 3D motion was stimulated by the
effects of the decaying turbulence. However, the grid
oscillation and/or the flow induced by it might affect
launch condition (initial condition) of the bubbles.
Tomiyama, et al. (2002) reported that the terminal velocity
and bubble trajectory of single bubble chain depend on the
initial shape deformation of a bubble. It is essential to
clarify the difference of initial shape deformation in order
to describe the shift of bubble motion.
We performed an experiment of a single bubble
launched into the decaying turbulence. The shadow images
of the bubbles were captured by high-speed video cameras.
The spatial resolution was 1024 x 1024 pixel2 (11.4
pm/pixel). The frame rate was 1000 fps.
The trajectory of the bubble gravity-center is plotted in
Fig. 7. In an immediate region of the bubble launch (just
above the hypodermic needle), trajectories under
Condition-B and Condition-OB were the same in x-z plane
and y-z plane. Miyamoto and Saito (2005) evaluated shape
deformation of a bubble using its curvature. They
characterized the curvature of bubble as explained in Fig. 8.
The curvature is calculated by equation 1;


Deq
2r


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

We performed the same analysis for deep understanding
of the shift of the bubble motion. Figure 9 shows the
curvature of the single bubble. The fluctuations of the
curvature under Condition-B and -OB are also almost the
same. The fluctuation of the curvature observed in x-z
plane was larger than that in y-z plane. The right and left
curvatures periodically fluctuated in x-z plane.
The phase shifting between right and left curvatures was
180 degrees. On the other hand, the fluctuations observed
in y-z plane were almost the same, and their amplitudes
were very smaller than those in x-z plane. In addition, no
phase shifting was observed. From these results, the shape
deformation of bubble is hardly changed in the initial
region (region near the hypodermic needle). In other words,
the bubble launch condition is not affected by the decaying
turbulence. Hence, the stimulation of the transition from
the 2D zigzag motion to the 3D motion is considered to be
caused by the interaction between the bubble and ambient
liquid-phase turbulence.


80

78

I76

74

72


80

78

I76

74

72


-4 -2 0
x [mm]
(a) x-z plane


2 4


- -B
--B
--OB





t
y


Fig. 7: Trajectory of center-of-gravity of single bubble.
(near the hypodermic needle)







Fig. 8: Bubble shape characterized via two fitting circles.


-4 -2 0
y [mm]
(b) y-z plane


2 4






Paper No


6
6 KR of-B
5 - R aof -OB f,
... A of-B/ ~
4 ... of -OB *

I- I
:3 I

2 '

1 -


70 80 90 100 110
Elapsed time [msec]
(a) x-z plane


6

5

4

t3

2

1

a


70 80 90 1
Elapsed time [msec]


(b) y-z plane
Fig. 9: The curvature of single bubble.
(near the hypodermic needle)


4.3 Liquid-phase motion (Standard deviation of
liquid-phase velocity)
The time-series liquid-phase velocity data were obtained
from the PIV measurement. The standard deviations were
calculated using the data sets. First, the liquid-phase mean
velocities were calculated by equation 2. Second, the
differences between the time-series liquid-phase velocities
and the liquid-phase mean velocities were calculated by
equation 3. Finally, the standard deviations of the
liquid-phase velocities were calculated by equation 4. The
standard deviations of the liquid-phase vertical velocity
and horizontal velocity components were calculated by the
following equations:


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

The standard deviations of the liquid-phase horizontal
component are plotted in Fig. 10. These were obtained
from spatial and time average. This spatial and time
average values were calculated from data sets of 250
frames.
When the bubble swarms passed through the
measurement area, the peak intensity under
Condition-OB(3) (continuous bubble swarms, three
continuous bubble swarms) is much larger than that of the
single bubble swarm. The former intensity is larger than
that of the latter in overall. The standard deviations (u,.ns
and wr,) under Condition-OB(3), -OB(1) at t = 4 sec take
the peak value, in this order. However, at t = 10 sec, the
standard deviation under Condition-O is the highest,
followed by Condition-OB(1) and -OB(3). Moreover, from
the results of Condition-OB(1) and -OB(3), the decay-rate
of the turbulence depends on the number of bubbles. From
these results, we consider that the decay-rate of the
turbulence was increased by the bubble swarms.
wms is larger than u,, before the bubble swarms passed
through the measurement area. Figure 11 shows vertical
and horizontal liquid-phase mean velocities (Umean and
Wmean) as well as u,,s and Wr,., under each condition. mean
has positive value (about 9 10 mm/s) in the initial region.
The mean flow is considered to be caused by an outer
frame of the grid and a central shaft of the grid. However,
the mean flow is much smaller than the bubble vertical
velocity (less than 3%). Consequently, the mean flow did
not affect the bubble motion.
The changes of uirm and w,,~ under Condition-O were
gentler than those of Umean and Wmean under Condition-O.
Moreover, u,,~ and Wr,. under Condition-O were almost
constant in all time. Hence, the homogeneous isotropic
turbulence generated by the oscillating-grid was also kept
in the decaying turbulence. The liquid-phase vertical mean
velocity Wmean decreased with time. On the other hand, the
liquid-phase horizontal mean velocity Umean was hardly
decreased. The timing of the increase in the liquid-phase
mean velocity is earlier than that of the standard deviation.
These phenomena are related to the wake of the bubble.
Hence, we need to research how the wake of the bubble is
shed in the decaying turbulence.
When the bubble swarm was launched in the
measurement area, u,,, wrs and Wmean increased. On the
other hand, Umean hardly changed when the bubble swarm
was launched in the measurement area. Moreover,
increment of wrmwas bigger than that of u,,. From these
results, we consider that the bubble effect on the vertical
direction is stronger than that on the horizontal direction.


5. Conclusions


For the purposes of clarifying the bubble-liquid interaction,
the modulation of the bubble motion and the liquid-phase
turbulence modulation induced by the bubble swarm were
experimentally investigated. We employed the
oscillating-grid decaying turbulence and the in-house
bubble launch device. The PIV measurement of the
liquid-phase and the visualization of the bubble motion


W1ms


KR of -B
- KRf -OB
S... A of-B
.. L of-OB



-I0


V.






Paper No


Elapsed time [sec]
Fig. 10: The standard deviations of the liquid-phase.


- Wrms
" Wmean
- uM
. rms
-"" mean


- I


10r

5




O (


0 10 2(
Elapsed time [sec]
(a) Condition-O


10
WrMS


5.,
0~7 Im


3 10 2(
Elapsed time [sec]
(b) Condition-OB(1)


10


or


I'\
I'


- Wrms
" Wmean
- ur
.... rm
"" umean


- 0 -


Elapsed time [sec]
(c) Condition-OB(3)


Fig. 11: Standard deviation and mean velocity of
liquid-phase.


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

were carried out. From these result, the bubble-liquid
interactions were discussed.
From the results of the visualization of the bubble
swarm, the bubble motion was shifted due to the decaying
turbulence. The bubble motion in the stagnant water was
2-dimensional zigzag motion. On the other hand, the
bubble motion in the oscillating-grid decaying turbulence
was accelerated from 2-dimensional motion to the
3-dimensional motion. Moreover, the bubble rising
velocity increased in the decaying turbulence. From the
results of the visualization of the single bubble, the bubble
trajectory and the curvature of the bubble in each condition
was not changed near the hypodermic needle. Hence, the
transition of the bubble motion was caused by the
interaction between the bubble and ambient liquid-phase
turbulence.
From the PIV result, the liquid-phase mean velocity and
the standard deviation of the liquid-phase were calculated.
When the bubble swarm was launched in the decaying
turbulence, the vertical mean velocity and the standard
deviation increased. The standard deviation under
Condition-OB(1) and -OB(3) were lower than that under
Condition-O at t = 10 sec.



Acknowledgements
The present study was promoted and financially
supported by Category "A" of the Grants-in-Aid for
Scientific Research, Japan Society for the Promotion of
Science. We thank JSPS for its help in the financial
support.

References
Saito, T, Kajishima, T., Tsuchiya, K., Kosugi, S., "Mass
transfer and structure of bubbly flows in a system of CO2
disposal into the ocean by a gas lift column," Chemical
Engineering Science, Vol. 54, pp 4945-4951 (1999).

Saito, T, Kajishima, T, Nagaosa, R., "CO2 sequestration at
sea by gas-lift system of shallow injection and deep
releasing," Environmental Science & Technology, Vol. 34,
pp 4140-4145 (2000).

Mudde, R. F., Saito, T., "Hydrodynamical similarities
between bubble column and bubbly pipe flow," Journal of
FluidMechanics, Vol. 437, pp 203-228 (2001).

Thompson, S. M., Turner, J. S., "Mixing across an
interface due to turbulence generated by an oscillating
grid," Journal of Fluid Mechanics, Vol. 67, pp 349-368
(1975).

Hopfinger, E. J., Toly, J. A., "Spatially decaying turbulence
and its relation to mixing across density interfaces,"
Journal ofFluid Mechanics, Vol. 78, pp 155-175 (1976).

Hopfinger, E.J., Brow, F.K., Gagen, Y., "Turbulence and
waves in a rotating tank," Journal ofFluidMechanics, Vol.
125, pp 505-535 (1982).


)






n1

il






Paper No


Qi, Zhou., Nian-Sheng, Cheng., "Experimental
investigation of single particle settling in turbulence
generated by oscillating grid," Chemical Engineering
Journal, Vol. 149, pp 289-300 (2009).

Morikawa, K., Urano, S., Sanada, T., Saito, T.,
"Turbulence modulation induced by bubble swarm in
oscillating-grid turbulence," Journal of Power and Energy
Systems, Vol. 2, No.l, pp 330-339 (2008).

Saito, T., Sakakibara, K., Miyamoto, Y, Yamada, M., "A
study of surfactant effects on the liquid-phase motion
around a zigzagging-ascent bubble using a recursive
cross-correlation PIV," Chemical Engineering Journal, Vol.
158, pp 39-50 (2010).

Kariyasaki, A., Ousaka, A., "Control of bubble size and
detaching frequency from a submerged nozzle by means of
pressure oscillation induced in the gas flow," 6th World
Congress of Chemical Engineering, Melbourne, Australia
(2001).

Sanada, T., Watanabe, M., Fukano, T., Kariyasaki, A.,
"Behavior of a single coherent gas bubble chain and
surrounding liquid jet flow Structure," Chemical
Engineering Science, Vol. 60, pp 4888-4902 (2005).


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

Sakakibara, K., Yamada, M., Miyamoto, Y, Saito, T.,
"Measurement of the surrounding liquid motion of a single
rising bubble using a Dual-Camera PIV system," Flow
Measurement and Instrumentation, 18, pp 211-215 (2007).

Daiji, S., Kazuya, S., Masato, Y, Sanada, T., Saito, T.,
"Bubble Motion and its surrounding liquid motion through
the collision of a pair of bubbles," Journal of Power and
Energy Systems, Vol. 2, No.l, pp 306-317 (2008).

Poorte, R. E. G, Biesheuvel, A., E\plciimcnl on the
motion of gas bubbles in turbulence generated by an active
grid," Journal of Fluid Mechanics, Vol. 461, pp 127-154
(2002).

Spelt, P. D. M., Biesheuvel, A., "On the motion of gas
bubbles in homogeneous isotropic turbulence," Journal of
FluidMechanics, Vol. 336, pp 221-244 (1997).

Tomiyama, A., Celata, GP., Hosokawa, S., Yoshida, S.,
"Terminal velocity of single bubbles in surface tension
force dominant regime," International Journal of
Mulutiphase Flow, Vol. 461, pp 127-154 (2002).

Miyamoto, Y, Saito, T., "Relationship between Interface
motion of an isolated bubble and its zigzagging motion of
the center of gravity," Transactions of the Japan Society of
Mechanical Engineers series B, Vol. 71, 1307-1313, in
Japanese (2005).




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