7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
Vibrationinduced attraction of a particle from a wall in microgravity
Mehrrad Saadatmanda, Masahiro Kawajia'* and Howard H. Hub
'Dept. of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, ON M5S 3E5, Canada
bDept. of Mechanical Engineering and Applied Mechanics, University of Pennsylvania, Philadelphia, PA, USA
kawaji@me.ccny.cuny.edu, mehrrad.saadatmand@utoronto.ca and hhu@seas.upenn.edu
Keywords: Microgravity; Vibration; ParticleWall Interaction; Gjitter, Particle Dynamics, Particle Motion,
Hydrodynamics, Vibration Frequency and Amplitude, Attraction Force, Repulsion Force
Abstract
The effects of small vibrations on a particle oscillating near a solid wall in a fluid cell, relevant to material processing such as
crystal growth in space, have been investigated by direct numerical simulations. Simulations have been conducted for a solid
particle suspended in a fluid cell filled with a fluid of 1 cSt viscosity, vibrating sinusoidally in a horizontal direction. The
simulations revealed the existence of a vibrationinduced force attracting the particle towards the nearest cell wall which
varied with the cell vibration frequency. The predicted flow patterns around the particle revealed that different velocity and
pressure distributions would be induced by the particle motion. In particular, the flow in the gap between the particle and the
nearest wall was predicted to accelerate and pressure to decrease in accordance with Bernoulli's principle, which would result
in the attraction force.
Introduction
The space platforms with near zero gravity environment
have been considered an ideal place for the production of
high quality protein and semiconductor crystals from
solutions or melts, by eliminating the sedimentation effect
and buoyancyinduced convective motion of the fluid
experienced on Earth. Ideally, in the absence of gravity,
diffusive mass transfer would dominate the crystal growth
which can result in the formation of crystals of better
morphologies or properties. Motivated by this idea, there
have been many experiments conducted aboard space
platforms to produce, for example, protein crystals of better
quality. The results, however, have shown unexpected
deviations from the ideal diffusioncontrolled crystal growth
in fluid cells under microgravity as crystals have been
observed to move inside the fluid cells (Chayen et al., 1997,
Lorber et al., 2000).
Although the space platforms are under microgravity, the
fluid cells still experience forces due to accelerations that
alter the nearly weightless condition (Ostrach, 1982). The
predominant component of theses forces occurs from
random accelerations called gjitter (Knabe and Eilers,
1982). The unexpected movements of the crystals observed
in space experiments can be related to gjitter and can
adversely affect the growth and quality of crystals (Chayen
et al., 1997, Lorber et al., 2000, Kawaji et al., 2003,
SimicStefani et al., 2006). In an orbiting spacecraft, gjitter
occurs in random directions with different frequencies and
acceleration levels. These accelerations are produced by
mechanical vibrations caused by the operation of equipment
such as pumps, fans and motors, crew motion, spacecraft
maneuvers, atmospheric drag and the earth's gravity
gradient (Knabe and Eilers, 1982, Jules et al., 2004, Shafie
et al., 2007).
Many studies have been conducted on the motion of a
particle in the past that confirm the vibrations can induce a
hydrodynamic force on a particle suspended in a fluid of
different density. Stokes (1851), Boussinesq (1885), and
Basset (1888) derived expressions for the forces affecting a
particle submerged in a fluid subjected to harmonic and
arbitrary motions. More recently, an equation of motion for
a small rigid sphere in a nonuniform flow was developed
by Maxey and Riley (1982). Chelomey (1983) studied the
behavior of solid particles immersed in a liquidfilled
container subjected to vibrations and found that particles
heavier than the surrounding fluid would emerge at the top
while particles lighter than the fluid would move
downwards. He attributed this paradoxical behavior to the
effects of nonuniform forces developed during the
vibrations. Coimbra and Rangel (2001) developed an
analytical model for the periodic motion of a small particle
in a viscous fluid to analyze the motion of particles induced
by vibrations. In their analytical model they combined
virtual mass, Stokes drag, and history forces. Later, their
analysis was confirmed experimentally by Coimbra et al.
(2004) and L'Esperance et al. (2005).
Recently, Hassan et al. (2006a, 2006b) and Hassan and
Kawaji (2007a, 2007b) investigated the vibrationinduced
motions of a particle in a fluid cell for different density
particles in water. Under the vibration conditions they
examined, water could be considered inviscid. Moreover,
Hassan et al. (2006c) studied both analytically and
experimentally, the effects of vibrations of a fluid cell on the
motion of a particle suspended by a wire in a fluid cell
under normal gravity. Their study showed the occurrence of
an attraction force on the particle when the particle
oscillated in an inviscid fluid near the cell wall. To mimic
the vibrations that produce gjitter, the cell was oscillating
with a sinusoidal motion of known amplitude and
frequency.
When the particle was oscillating far away from the cell
walls, the vibrations had no effect on the mean particle
position. However, placing the particle close to one of the
cell walls resulted in a clear shift of the particle's mean
position towards that wall. They studied oscillations both
normal and parallel to the nearest wall. For both cases the
drift in the mean particle position increased with the cell
vibration frequency and amplitude and decreased with an
increase in the distance between the particle and the nearest
wall. Hassan et al. (2006c) also obtained analytical relations
for the shift in the mean position of the particle towards the
nearest cell wall when the particle was oscillating either in
parallel with the nearest cell wall or normal to it.
For example, for a particle oscillating in parallel with the
nearest cell wall, the drift towards the wall, XD is
XD 9 L a2 2 P PL L
32 g (2ps +P)2 H(1
which is the drift towards the wall for the particle oscillating
in parallel with it. In Equation (1), L is the distance from the
suspension point to the center of mass of the particle, g is
the gravitational acceleration, a is the cell amplitude, m is
the angular frequency (= 27f, where f is the frequency in
Hz), ps and pL are particle and liquid densities
respectively, Ro is the particle radius and H is the initial
distance between the stationary particle and the nearest cell
wall. Figure 1 shows the experimental data and a plot of
Equation (1) for the particle drift versus cell vibration
frequency for a steel particle in water. Unfortunately,
Hassan et al. (2006c) did not include any investigations into
the mechanisms responsible for the attraction of the particle
to the nearest wall.
Recently, SimicStefani et al. (2006) conducted direct
numerical simulations to investigate the effect of small
vibrations on particles such as protein crystals growing in
protein crystal growth (PCG) cells under zero gravity. Their
study revealed the effect of six parameters on the crystal
motion; frequency and amplitude of the vibration, crystal
diameter and density, and fluid viscosity and density.
Moreover, they observed formation of a vortex ring around
a moving particle that changed the direction of its rotation
based on the particletofluid density ratio. However,
SimicStefani et al. (2006) did not investigate the effect of
the nearest wall on the particle oscillation in their study.
The objective of this study is to numerically investigate the
mechanisms that cause attraction of the particle to the
nearest wall in a fluid cell. To achieve this objective, direct
numerical simulations were conducted for a particle in a
fluid cell under zero gravity. The fluid cell was subjected to
sinusoidal vibrations of known amplitude and frequency
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
under zero gravity to investigate the flow patterns around
the particle that cause the attraction force.
 Theoretical Eq.
o Exprimestal
810
go
I so/
140t
AL
20 Z
a 5 10 15 20 25
Frequency (HZ)
Figure 1 Variations of the particle drift with the cell
frequency for a steel particle in water: cell amplitude = 1.0
mm (parallel to the wall), wire length plus particle's radius,
L = 76 mm, distance between the centre of the particle and
wall before cell vibration, H = 7.0 mm, particle diameter =
12.7 mm (from Hassan et al., 2006c).
Nomenclature
a cell amplitude (m)
A cell vibration amplitude
f frequency (Hz)
f body force per unit mass
F hydrodynamic force
g gravitational acceleration (ms2)
G external body force
gs steady residual acceleration
H initial distance between the stationary particle
and the nearest cell wall
I moment of inertia matrix
L distance from the suspension point to the center
of mass of the particle (m)
n unit normal vector on the surface of the particle
pointing into the particle
Ro particle radius (m)
t time (s)
T, moment acting on the particles
u velocity of fluid
i test function for the fluid velocity
V translational velocity
V velocity
XD particle drift towards the cell wall (m)
X position of the center of mass of the sphere
x position of a given point on the surface of the
particle.
Greek letters
p liquid density (kgm3)
C stress tensor
c angular frequency (= 27iJ) (rads1)
2 Computational domain
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
8d2 Boundaries of the domain
Subsripts
f fluid
i particle
L fluid
0 fluid
S particle
Numerical Simulation
The Direct Numerical Simulation code, Partflow3d,
developed by Hu et al. (2001) is used in this work to
simulate the motion of a solid particle in a fluid cell. The
description of the code given below is mainly taken from
Hu et al. (2001) which the interested readers are referred to
for more detailed information. The code is based on a finite
element technique. Since the movements of the solid
particles and surrounding fluid are coupled, the numerical
code solves both the fluid and particle governing equations
combined into a single equation. The code considers the
motion of N rigid solid particles in an incompressible fluid.
It also considers a fluid domain, d20(t), at a given time t,
and domain of particles d, (t), where the index i (= 1, 2,
..., N) represents different solid particles.
Partflow3d uses a mesh composed of unstructured
tetrahedron elements that is suitable for the irregularity and
movement of the domain. The nodes on the particle surface
are moving with the particle and therefore the code solves a
moving boundary problem. To accommodate this issue, the
mesh is regenerated at each time step. The code is written in
3D and is based on an Arbitrary LagrangianEulerian
technique, which uses the standard Galerkin finite element
method. At each time step, the positions of the particle and
the mesh are updated explicitly, while the velocities of the
particle and fluid are determined implicitly.
Governing Equations
The governing equations for the motion of the fluid are the
conservation of mass,
Vu = 0, (2)
and the conservation of momentum,
Du
p/ = p/f+V.o, (3)
where u is the velocity vector, pf is the density of the fluid,
f is the body force per unit mass, a is the stress tensor
Du
and is given by,
Dt
Du cu
D + (u.V)u. (4)
Dt O.
For the rigid particles, the governing equations are
Newton's second law for the translational motion,
dV
m, V =G,+F,=G, a.ndS, (5)
at, (t)
and the Euler equations for the rotation,
d
dt
I, +,) XI x,)
dt ( ,
T, (xX,)x(an)dS,
S0,(t)
where m, is the mass, and I, is the moment of inertia matrix
of the ith particle, V, is the translational velocity of the
particles, o, is the angular velocity of the particles, G, is
the external body force, n is the unit normal vector on the
surface of the particle pointing into the particle, X, is the
particle's centroid and d.2 (t) denotes the boundaries of
the particles domain. Equations (5) and (6) also show that
the hydrodynamic force F, and the moment acting on the
particles T, are obtained by integrating the fluid stress over
the particles' surface.
In the specific use of Partflow3d for this work, a sinusoidal
motion is applied to the fluid cell containing the particle.
The position and the acceleration of the fluid cell are
sinusoidal functions of time in a fixed frame of reference.
The cell acceleration therefore, can be specified through the
body force f which can be defined as,
f =g Awt sin(ot)n, (7)
where g, represents steady residual accelerations, A is the
cell vibration amplitude (half of peaktopeak amplitude),
and co is the angular frequency (=22iJ). The residual
acceleration was assumed to be zero in this work.
The nonslip boundary condition on the particle surface was
considered as,
u = + o,(x X,), (8)
where u is the velocity of fluid, V, is the velocity of particle,
X, is position of the center of mass of the sphere, and x is
the position of a given point on the surface of the particle.
The nonslip boundary condition for the boundaries of the
computational domain (cell walls) was
u = 0. (9)
Combined fluidsolid formulation
In order to use the finiteelement method, Hu et al. (2001)
derived a weak formulation that incorporates both the fluid
and particle equations of motion, Equations (3), (5), and (6).
This was achieved by multiplying Equation (3) by the test
function for the fluid velocity, u and integrating over the
fluid domain at a time instant t. The result is
S Dt f ud& + f ( Vii)d2
f J( n)idS = 0. (10)
I
For the particles, the test function will be replaced by
Equation (8). By using equations of motion for the particles,
Equations (5) and (6), the following relation would be
derived
 (a n)idS
02
= (a.n).[V, + x(xX,)]dS
an
=V, (cn)dS ,. f(xX,)x(cn)dS
an an ,
=V, ( dVG, +) d (11)
Sdt ) dt
Substituting Equation (11) into (10) will result in the
combined fluidparticle momentum equation
SDu
p D fDt udQ + Ja: D[u]d
+ V, G,I
I I IN dt
+ d(I) 0. (12)
I<
Similarly, the weak formulation for the mass conservation,
Equation (2), can also be obtained by multiplying its
corresponding test function and integrating over the fluid
domain.
As it is clear from Equation (12), for the fluidparticle
system, the hydrodynamic forces and moments acting on the
particle do not explicitly appear in the formulation. This is
due to considering the fluid and particles as one system
which turns the hydrodynamic forces to internal forces. This
reduces the computational time, and is an advantage when it
comes to compute the fluid and particles' behaviors
numerically.
Results and Discussion
In this work, the vibrationinduced motion of a single
spherical particle in a liquidfilled cell was considered. The
particle is initially placed near one of the vertical cell walls
as shown in Figure 2. As the cell is vibrated in the
zdirection with a specified amplitude and frequency, the
particle would be induced to vibrate in the zdirection,
parallel to the nearest wall, due to a density difference
between the particle and surrounding liquid. It is important
to note that the particle was fixed in ydirection in the
simulations in order to maintain an average for the
hydrodynamic force. This helped better studying the force
acting on the particle by preventing the particle from
moving towards the wall.
For the present 3dimensional simulations, a fluid cell with
internal dimensions of 140 mm height and 50 mm x 50 mm
cross section under zero gravity environment was
considered. A spherical stainless steel particle with a
diameter of 12.7 mm was considered to be placed very close
to one of the cell walls (Figure 2) with its centroid at x = 7.6
cm, y = 4.3 cm, and z = 2.5 cm, so that the distance between
the edge of the particle and the nearest cell wall was 650
microns.
To understand the effect of the nearest wall on the
hydrodynamic forces acting on the particle vibrating near a
fluid cell wall, direct numerical simulations were conducted
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
for a liquid with a viscosity of 1 cSt and density of 1,000
kg/m3. The fluid cell was subjected to sinusoidal vibrations
of 1.0 mm amplitude (half of peaktopeak) and two
different frequencies of 7 and 15 Hz. All the numerical
results presented in this work were obtained with a
sufficiently small mesh that the results would not vary more
than 1% with further mesh refinement.
All the simulations were conducted in 3D, and the
predicted velocity and pressure fields extracted from the
horizontal zy plane that cuts through the centroid of the
Figure 2 Schematic of the fluid cell and particle showing
the horizontal plane for flow visualization.
particle as shown in Figures 2 and 3 will be presented below
to reveal the flow pattern around the particle.
The flow pattern around the moving particle will be
described according to the diagram shown in Figure 3.
There are four different zones of interest close to the
particle: the front and back zones which lie in front and
back of the moving particle, the wall zone between the
particle and the nearest wall, and the fluid zone on the
opposite side.

Particle Back Zone
Fluid ZoneWall Zone
 Front Zone
Particle Movement Direction
Cell Wall
Figure 3 A view of the particle and fluid motion in the
horizontal plane passing through the centroid of the particle.
C""C~,,
cLo
In one cycle, the particle moves from z = +A to A and back
to +A, where A is the particle amplitude. To discuss the
results of simulations of the particle movement in the fluid
cell, let us consider five different positions of the particle in
each cycle:
Position 1: particle is moving in z direction with its
maximum velocity,
Position 2: particle is halfway towards its end position (z
= A/2),
Position 3: particle reaches the end position (z = A) and
its velocity is zero,
Position 4: particle is moving in +z direction and is
halfway to its maximum velocity,
Position 5: particle is moving in +z direction and reaches
its maximum velocity.
VibrationInduced Particle and Fluid Motion
In the following figures, the units of velocity and pressure
are cm/s and g/cm.s2, respectively. The flow patterns
predicted are discussed below for the particle positions 1 to
5 for the cell vibration amplitude of 1.0 mm and two
different vibration frequencies of 15 Hz and 7 Hz.
Figures 4 and 5 show the particle at Position 1 for cell
vibration frequencies of 15 and 7 Hz, respectively. As it is
clear from these figures, the pressure in the wall zone and
the fluid zone are lower than the other areas around the
particle for both cases. However, the low pressure area
around the wall zone is larger than the fluid zone and also
the fluid velocity is the largest in the wall zone. This clearly
indicates that the particle would be attracted towards the
nearest wall based on the Bernoulli's principle.
 N,
N
.
20
*40
40
S* 1 i20
140
Figure 4 Position 1: Velocity and pressure distributions
around the particle for cell amplitude of 1.0 mm and
frequency of 15 Hz.
Another interesting observation from these figures is that
the flow direction in the wall zone is opposite to the
direction of the moving particle. This interesting
phenomenon occurs possibly due to the low viscosity of the
fluid. As the particle moves in a given direction, the liquid
between the particle and the nearest wall is squeezed and
directed towards the opposite direction. If the fluid were
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
highly viscous, the shear stress would be much greater and
the liquid may not be forced to flow in the direction
opposite to the particle.
Figure 5 Position 1: Velocity and pressure distributions
around the particle for cell amplitude of 1.0 mm and
frequency of 7 Hz.
Figures 6 and 7 show the particle at Position 2 for cell
vibration frequencies of 15 and 7 Hz, respectively. As
shown, the pressures in the front zone and the wall zone are
still lower than the other parts of the fluid which again
indicates the particle attraction towards the nearest wall.
p
140
120
100
8o
680
100
12
40
120
110
Figure 6 Position 2: Velocity and pressure distributions
around the particle for cell amplitude of 1.0 mm and
frequency of 15 Hz.
At Position 3 (Figs. 8 and 9), the particle stops and changes
its direction. Because the fluid is still moving in the z
direction, the pressure in the back zone becomes higher than
in the front zone. When the particle reverses its direction
and moves in the +z direction (Position 4 as shown in
Figures 10 and 11), the pressure in the front zone is still
higher than in the back zone (considering that the back and
front zones are now switched). However, the pressure in the
wall zone is still low and at the same time the flow velocity
in the wall zone is the highest in the fluid. This suggests that
the particle should still be attracted to the wall.
t, .
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
Figure 7 Position 2: Velocity and pressure distributions
around the particle for cell amplitude of 1.0 mm and
frequency of 7 Hz.
z
p,
leo
120
1o0
80
20
160
s
Figure 8 Position 3: Velocity and pressure distributions
around the particle for cell amplitude of 1.0 mm and
frequency of 15 Hz.
20
80
20
20
Figure 9 Position 3: Velocity and pressure distributions
around the particle for cell amplitude of 1.0 mm and
frequency of 7 Hz.
p ,~
Ii
Figure 10 Position 4: '
around the particle for
frequency of 15 Hz.
Z
LY
p
140
120
100
so
80
60
40
20
0
20
40
80
100
120
140
1 1
velocity and pressure distributions
cell amplitude of 1.0 mm and
. V
.4
L,
p
160
120
n
so
00
6O
20
0
2a
40
60
80
100
120
140
160
Figure 11 Position 4: Velocity and pressure distributions
around the particle for cell amplitude of 1.0 mm and
frequency of 7 Hz.
Finally, at Position 5 (Figures 12 and 13) when the particle
reaches its highest velocity, the wall zone still has the
lowest pressure and the highest velocity in the fluid. At the
same time, the pressure in the front zone is lower than in the
back zone. The pressure and the velocity predictions again
suggest that the particle should be attracted to the wall.
Studying the flow pattern around the particle for a 1 cSt
fluid suggests that the attraction force on a particle
oscillating in the vicinity of a wall in a fluid cell is governed
by the Bernoulli's principle. The fluid cell vibration induces
the particle motion which in turn results in the fluid motion.
The velocity and pressure in the gap between the oscillating
particle and the nearest wall reach the highest and lowest
values, respectively, compared to those in other regions.
Consequently, the particle is attracted towards the nearest
wall, as theoretically predicted by Hassan et al. (2006c) for
an inviscid fluidfilled cell.
Figure 12 Position 5: Velocity and pressure distributions
around the particle for the cell amplitude of 1 mm and
frequency of 15 Hz.
Figure 13 Position 5: Velocity and pressure distributions
around the particle for cell amplitude of 1.0 mm and
frequency of 7 Hz.
Attraction force exerted on the particle
The simulation results point to the existence of an attraction
force on the particle towards the nearest cell wall for a cell
filled with a 1 cSt fluid. Figures 14 and 15 show the
hydrodynamic forces exerted on the particle in the
ydirection for a cell vibration amplitude of 1.0 mm and
frequencies of 7 and 15 Hz, respectively. As these figures
show, an attraction force on the particle towards the nearest
wall is generated. The averaged force over one period of
oscillation is calculated to be equal to 1.97 dynes for 7 Hz
and 9.89 dynes for 15 Hz frequency.
The main contribution to the attraction force is related to the
maximum velocity of the particle. Note that the negative
velocities in Figures 14 and 15 indicate the particle moving
in z direction. Therefore, both the positive and negative
peaks in the particle velocity in both figures correspond to
the maximum magnitude of the particle velocity. One can
clearly note from Figures 14 and 15 that the maximum
attraction force occurs when the particle velocity reaches a
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
maximum magnitude for the 1 cSt liquid.
S Force at ydirection
V........... Velocity at zdirection
6,
o 2
0iN AP'I
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Time, s
Figure 14 Hydrodynamic force exerted on the particle in
ydirection and particle velocity in the direction of cell
vibration for cell vibration amplitude of 1.0 mm and
frequency of 7 Hz.
50
 Force at ydirection
40 ................ Velocity at zdirection
30
20
4 10
Z10
20
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Time, s
Figure 15 Hydrodynamic force exerted on the particle in
ydirection and particle velocity in the direction of cell
vibration for cell vibration amplitude of 1.0 mm and
frequency of 15 Hz.
Conclusions
Direct numerical simulations on a spherical particle
suspended in a fluid cell were performed in order to
understand the mechanisms leading to attraction of the
particle to the closest wall in a fluid cell. The particle's
motion was induced by small sinusoidal vibrations of
known amplitude and frequency under zero gravity
environment. The focus of the work was on the
hydrodynamic forces affecting the particle. The simulations,
performed for a fluid of 1 cSt viscosity and 1000 kg/m3
density, revealed an attraction force towards the nearest cell
wall for this fluid.
Studying the pressure distribution around the particle
showed that for the 1 cSt fluid, the pressure between the
particle and the nearest wall is lower than in other regions
inside the fluid cell. Moreover, the velocity distributions
showed that for this fluid, the highest velocity in the fluid
cell occurs between the particle and the nearest wall. Based
on these observations, it is clear that the hydrodynamic
attractive force on the particle is related to the Bernoulli's
principle of reduced pressure in high velocity zones.
Acknowledgements
The authors would like to thank the Canadian Space Agency
for an SSEP grant to financially support this work.
References
Basset, A. B.: A treatise on hydrodynamics. Deighton, Bell
and Co Press, Cambridge, UK 21, 260284 (1888)
Boussinesq, J.: Sur la resistance qu'oppose un fluide ind6fini
en repos, sans pesanteur, au movement varin d'une sphere
solide qu'il mouille sur toute sa surface, quand les vitesses
restent bien continues et assez faibles pour que leurs carries et
products soient n6gligeables. C. R. Acad. Sci. Paris, Vol. 100,
935937 (1885)
Chayen, N. E., Snell, E. H., Helliwell, J. R. & Zagalsky, P.
F.: CCD video observation of microgravity crystallization:
apocrustacyanin Cl. Journal of Crystal Growth, Vol. 171,
219225 (1997)
Chelomey V. N.: Paradoxes in mechanics caused by
vibrations. Akad. Nauk. SSSR., Vol. 270, 6267 (1983)
Coimbra, C. F. M. & Rangel, R. H.: Spherical particle motion
in harmonic stokes flows. AIAA Journal, Vol. 39, 16731682
(2001)
Coimbra, C. F. M. & Rangel, R. H.: Spherical particle motion
in harmonic stokes flows. AIAA Journal, Vol. 39, 16731682
(2001)
Coimbra, C. F. M., L'Esperance, D. & Lambert, R. A.: An
experimental study on stationary history effects in
highfrequency stokes flows. Journal of Fluid Mechanics,
Vol. 504, 353363 (2004)
H.H. Hu, N.A. Patankar, & M.Y. Zhu.: Direct Numerical
Simulations of FluidSolid Systems Using the Arbitrary
LagrangianEulerian Technique. J. Comput. Phys., Vol. 169,
427462 (2001)
Hassan, S. & Kawaji, M.: Vibrationinduced particle drift in
a fluid cell under microgravity. Microgravity Science and
Technology, Vol. 19, 109112 (2007b)
Hassan, S. & Kawaji, M.: Wall effects on vibration induced
particle motion in a fluid cell in space. AIAA journal, Vol.
45, 20902092 (2007a)
Hassan, S., Kawaji, M., Lyubimova, T. P. & Lyubimov, D.
V.: Effects of vibrations on particle motion near a wall:
existence of attraction force. Int. J. Multiphase Flow, Vol. 32,
10271054 (2006c)
Hassan, S., Lyubimova, T. P., Lyubimov, D. V. & Kawaji,
M.: Motion of a sphere suspended in a vibrating liquidfilled
container. J. Appl. Mech., Vol. 73, 7278 (2006a)
Hassan, S., Lyubimova, T.P., Lyubimov, D.V. & Kawaji,
M.: The Effects of Vibrations on Particle Motion in a
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
Semiinfinite Fluid Cell. ASME J. Applied Mechanics, Vol.
73, 610621 (2006b)
Jules, K., McPherson, K., Hrovat, K., Kelly, E. & Reckart,
T.: A status report on the characterization of the microgravity
environment of the international space station. Acta
Astronautica, Vol. 55, 335364 (2004)
Kawaji, M., Gamache, O., Hwang, D.H., Ichikawa, N.,
Viola, J.P. & Sygusch, J.: Investigation of Marangoni and
natural convection during protein crystal growth. J. Crystal
Growth, Vol. 258, 420430 (2003)
Knabe, W. & Eilers, D.: Lowgravity environment in
spacelab. Acta Astronautica 9, Vol. 187198 (1982)
L'Esperance, D., Coimbra, C. F. M. & Trolinger, J. D.:
Experimental verification of fractional history effects on the
viscous dynamics of small spherical particles. Experiments
in Fluids, Vol. 38, 112116 (2005)
Lorber, B., Ng, J. D., Lautenschlager, P. & Giege, R.:
Growth kinetics and motion of thaumatin crystals during
USML2 and LMS microgravity missions and comparison
with earth controls. Journal of Crystal Growth, Vol. 208,
665677 (2000)
Maxey, M. & Riley, J.: Equation for a Small Rigid Sphere in
a Nonuniform Flow. Phys. Fluids, Vol. 26, 883889 (1982)
Ostrach, S.: Lowgravity fluid flows. Ann, Rev. Fluid Mech.,
Vol. 14, 313345 (1982)
Shafie, S., Amin, N. & Pop, I.: gJitter free convection flow
in the stagnationpoint region of a threedimensional body.
Mechanics Research Communications, Vol. 34 (2), 115 
122 (2007)
SimicStefani, S., Kawaji, M. & Hu, H. H.: Gjitterinduced
motion of a protein crystal under microgravity. J. Crystal
Growth, Vol. 294, 373384 (2006)
Stokes, C.G.: Mathematical and Physical Papers. Vol. 3, 1
(1851)
