Group Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Title: P1.47 - Particle classification in high viscous environment
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102023/00452
 Material Information
Title: P1.47 - Particle classification in high viscous environment Particle-Laden Flows
Series Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Physical Description: Conference Papers
Creator: Matulka, P.
Potschinski, T.
Walzel, P.
Publisher: International Conference on Multiphase Flow (ICMF)
Publication Date: June 4, 2010
 Subjects
Subject: Laminar flow, suspensions
suspensions
particle wall interaction
particle separation
 Notes
Abstract: Particle classification becomes difficult when the density between particle and fluid is low or negligible and the fluid is viscous. For such applications, a process is needed, able to separate the particles according to their size. Such applications are e. g. found in biological systems for cell separation or in the removal of gel particles from polymer melts. Particle transport in laminar tube flows at low but non zero Re numbers leads to accumulation of large particles near the tube centre and forms a particle free zone near the wall. Small particles find their position on their equilibrium radius. Downstream widening of the flow enhances segregation between large and small particles. Large particles can be collected in a centred collector tube downstream, whereas small particles follow their streamlines around the collector tube and can be removed with the remaining flow. The said particle migration is observed when the ratio of particle to tube diameter is 0.3 < d/D < 0.53 and the tube Reynolds number is in between 0.1 < Re < 40. CFD simulations reveal the shape of the streamlines in the downstream enlargement with different tube Reynolds number and collection geometries.
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: VID00452
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: P147-Matulka-ICMF2010.pdf

Full Text


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


Paper No 2596


Introduction

Traditional segregation processes apply sieves or
sedimentation. Particle separation with sieves is feasible in a
low viscous fluid environment and only with minor forces
to the particles compared to gravity or other body forces.
This separation method becomes difficult when the
viscosity is high and the strong interaction between particles,
fluid and mesh impairs efficient segregation.

Sedimentation is also an alternative for separation according
to their size due to different settling velocities of large and
small particles in the fluid. This process is applied for
example in air siffers for particle separation at cut off size
x, < 100 pLm. A low fluid viscosity facilitates the separation.
However, it is only feasible at sufficient high density
differences between particles and fluid and/or in centrifugal
fields.

The axial movement of particles in cylindrical ducts at low
duct and particle Reynolds numbers recently was
extensively analytically analyzed (Bhattacharya 2010).
However, the radial movement of particles was not yet
considered. In a duct with a circular cross section diameter
D, particles with diameters d << D in the range r > r' are
deflected towards the tube axis because of the flow
constraints close to the channel wall. Particles in the range
r non-uniform shear strain on the particle surface due to the
nonlinear velocity field. As a result, particles are collected at
the equilibrium radius r'. Larger particles d > 0.1-D separate
radially and the dispersed particles form ring zones


(Lou 2003). The position of the equilibrium radius then
depends on the ratio d/D (see Figure 1). Therefore different
particle tracks can be expected allowing for particle
separation, e.g. by separation tubes dividing the flow region
into different channels.


possible motion


parabolic
velocity profile


r/R
0 1
in a laminar tube flow with


Figure 1: Particle motion
parabolic velocity profile.


This behavior had been extensively examined in numerous
publications. First experiments have been realized by Segrd
and Silberberg (1961). In a vertical tube with small particles
in the size range of 0.07 within a tube Reynolds number between 3.2 and 173 were
examined at different flow rates. The particles were
concentrating in an annular zone. The radius of the zone


Particle classification in high viscous environment


Paul Matulka, Timo Potschinski and Peter Walzel


University of Dortmund, Department of Biochemical and Chemical Engineering, Mechanical Process Engineering
44227 Dortmund, Emil-Figge-Str. 68, Germany
!.mlu Ilalll\.ul Gbe i lu-dortmund.de


Keywords: Laminar flow, suspensions, particle wall interaction, particle separation




Abstract

Particle classification becomes difficult when the density between particle and fluid is low or negligible and the fluid is viscous.
For such applications, a process is needed, able to separate the particles according to their size. Such applications are e. g.
found in biological systems for cell separation or in the removal of gel particles from polymer melts. Particle transport in
laminar tube flows at low but non zero Re numbers leads to accumulation of large particles near the tube centre and forms a
particle free zone near the wall. Small particles find their position on their equilibrium radius. Downstream widening of the
flow enhances segregation between large and small particles. Large particles can be collected in a centred collector tube
downstream, whereas small particles follow their streamlines around the collector tube and can be removed with the remaining
flow. The said particle migration is observed when the ratio of particle to tube diameter is 0.3 < d/D < 0.53 and the tube
Reynolds number is in between 0.1 < Re < 40. CFD simulations reveal the shape of the streamlines in the downstream
enlargement with different tube Reynolds number and collection geometries.






Paper No 2596


was determined as r'/R= 0.6 (Segr6 & Silberberg 1961,
1962). Bauckhage examined the influence of the particle
size on the equilibrium radius. The maximum possible
displacement for a particle is:


rax =-T 2


A dependency of the equilibrium radius r' for Reynolds
number within the range 36 < Re < 1000 was found by
Bauckhage and described in Bauckhage (1973, 1975 and
1977) as:


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

separation according to their size and to analyze a
separation method for particles with negligible density
difference between particles and fluid with low plugging
tendency. The process as proposed is able to perform the
separation in one or more separator stages. Tube Reynolds
number 0.1 density differences between particle and flow as also
reported in Walzel et al. 2007.


Nomenclature

c volumetric concentration V,aracle/Vto (-)
d diameter of spherical particle (m)
D tube diameter (m)
r' radius of equilibrium position (m)
r radius vector, position of particle centers (m)
R R = D/2, tube radius (m)
Re tube Reynolds number (-)
Vo input feed flow rate (m3/s)
V, collector tube flow rate (m3/s)
V, overflow (m /s)
mean flow velocity in the tube (m/s)
au velocity difference between flow & particle (m/s)
No total injected number of particles (-)
N1 number of particles in the collector tube (-)
nnumber of measurements
PE PE = N,/No, partition efficiency (-)
H1 distance between separation tube discharge and
collector tube (m)
h tube length (m)

Greek letters
p density (kg/m )
9L viscosity (Pas)

Subsripts
max maximum
R radial
M middle
av average value


Experimental Facility

The experimental setup, as shown in Figure 2, consists of a
vertical acrylic glass tube with an internal diameter of 6 mm
and a total length of 2000 mm. The tube extends into a
chamber with a rectangular side length of 200 x 200 mm2.
The chamber is used as a downstream enlargement. The
level height of the liquid in the chest is 110 mm. At a
variable distance in between 8 and 38 mm from the outlet a
collector tube is mounted. The collector tube has an inner
diameter of 16 mm to provide a large cross-sectional area
with low plugging tendency. Two gear pumps are conveying
the fluid in a recycle loop while two filters were installed to
protect the pumps from the injected particles. The
experimental liquid is collected in a main tank. At a distance
of 100 mm from the entry into the vertical tube a horizontal
second tube is connected for the particle injection. This
horizontal tube can be removed and filled with a suspension
prepared in advance. To avoid amy reflux to the injector, a


1 Re
+0.064-1n36


The tube Reynolds number is defined as:


p-u-D
Re =


Equation 2 states the equilibrium radius to become smaller
for larger particles and must be zero for particles d/D = 1. I.
e., large particles must pass the tube more or less with their
center close to the centerline of the duct. In case of high
particle concentrations c > 20 %, particles also accumulate
at the tube center and the particle free zone close to the wall
still exists. Further experiments were performed by Saffman
(1956), Rubin (1977) and Matas (lr1~14. Buggisch and
Muckenfuss (2002, 2003 and 2006) investigated the
influence of solid boundary walls in shear flows of
suspensions and the particle transport. They developed a
model describing effects such as the occurrence of pseudo
wall slip or the discontinuities of the particle concentration.

In the last decade many investigations have been carried out.
Klein (1999) separated particles according to their size in
laminar flow within micro channels. The separation effect
was intensified by additional forces, caused by gravitation,
temperature gradients or a cross flow. Small particles
diffused to the centre of the flow with higher flow velocities
and large particles were pressed towards the wall and
moved with lower flow velocity. Yamada and Takagi (21I r 4
and 2005) describe a process for particle separation in a
laminar flow with a side stream to the tube. The side stream
deflects the particles against the opposite wall of the tube.
Small particles move a wider distance towards the opposite
wall compared to larger ones. The particles were separated
perpendicular to the flow direction at the channel outlet
according to their size. A downstream enlargement
enhanced the separation in the micro channel. Similar
separation processes were presented by Pamme (2007) and
Huang (21***4).

The movement of a larger number of polydisperse particles
in laminar flow is superimposed by particle interaction in a
stochastic manner. The reasons are different starting
conditions of the particles and particle-particle interactions
due to different axial velocities at different equilibrium
radii.

It was intended to apply an optimum geometry for particle






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

chamber. The small particles, not separated through the
collector tube, were collected in the chamber of the
downstream enlargement. The mixture leaves the chamber
through the overflow and transfer to the main tank.

In determination of the equilibrium radius, the particle
positions were studied with a laser-sheet-system. A 30 mW
laser with a wave-length of 658 nm was used. The laser
illuminates a plane in the acrylic-tube. An acrylic chamber
was installed around the tube which allows to take pictures
of the plane with a CCD Camera. The distance between the
tube and the laser was 160 mm, as shown in Figure 3.
34 mm


Paper No 2596


sphere valve is installed.

The experimental liquids were mixtures of glycerol and
water, which allow for varying the dynamic viscosity and
the density of the mixtures. In the present approach,
mixtures of 86 -92 wt-% glycerol were used. It was
important to obtain the particles with a density close to the
fluid. The particles were formed by dripping a 2 wt-% of
alginate, 1 wt-% powder cellulose and 0.05 wt-% to
0.15 wt-% of micro-balloons in a 0.3 molar CaCl-solution
from a capillary. In order to maintain the spherical shape
and the constant density of the particles they were kept in
the glycerol water mixtures after their formation. In order to
distinguish the particle visually according to their size, the)
were pigmented with four colours, i.e. a certain colour for
each diameter. The particle size ratio (d/D) was varied
during the experiments from 0.30 to 0.53. The standard
deviation of particle diameters in individual fractions was
less than 0.08 mm. The flow rate of the pump 1 was varied
to obtain mean velocities in a range of 0.01 0.1 m/s in the
vertical tube. The tube Reynolds number was between 0.1
and 40.

The numbers of particles injected at one stroke were
between 8 and 14 particles with the same diameter.


Side view


I 60 mm


Figure 3: Experimental set
laser-sheet-illumination


up of the


During the trials, the vertical position of the chamber was
varied from 455 mm to 1555 mm Four geometries of the
collector tube are studied and compared with a uniform
distance between outflow and collector tube of H = 19 mm,
as shown in Figure 4. The four geometries are: round-edged
(A), square-edged (B), sharp-edged inside (C) and
sharp-edged outside (D). Every geometiv has an internal
diameter of 16 mm.


Pump 2


round edged


Figure 2: Experimental set up for particle separation of
different equilibrium positions in the vertical tube. The
laser-sheet illumination on the vertical tube enables a
visualization of the particle position in the tube.


The collector tube flow was varied by the speed of Pump 2
in between 0 I \;/Vo I 0.9 The mixture flows back to the
main tank. The rest of the mixture was transferred to the


sharp edged
outside


inside


Figure 4: Collector tube geometries,
A: round-edged, B: square-edged, C: sharp-edged inside,
D: sharp-edged outside






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


Paper No 2596


ometry of the
ose, the density
be 1223 kg/m3
Iferent collector
:e ratio (d/D) of
ds number was
s were injected
nually operated
repeated with
ube geometry.
ate ratio of the
V, particles of
lIlector tube or
nation efficiency
.n the collector
Jidual particles
y of the flow in
re radius of the
Icy of particles
collector tube
ies as in Figure
eads to lowest
ineffective at a
-edged (A) and


Results and Discussion

Through our first trials, the optimal gec
collector tube was determined. For this purpe
of the glycerol-water mixture was selected to
and the viscosity was 124 mPas, and four dif
geometries as in figure 4 were tested. The siz
the particles was 0.39 and the tube Reynole
1.8. For each adjustment, 10 to 13 particles
with some of the working fluid using a mat
piston arrangement. This procedure was
several flow splits for each collector t
Depending on the flow split i.e. the flow r;
feed flow Vo to the collector tube flow '
certain size were moving either into the co
into the residual flow (overflow). The separr
is the ratio of the particle numbers found i
tube divided by the total number of indiv
passing through the device. The mean velocit
the tube is u = ir/2. R2 where R is th
tube. Figure 5 shows the separation efficient
depending on the flow rate ratio of total to
flows for the four different collector geometr
4. The sharp-edge inside geometry (C) l
gradient of the graph and the separation was
flow rate range from 0.3 to 0.6. The round
the square-edged (B) geometries yield shar
The geometry with outside sharp-edges (D)
results leading to the steep gradient of the gra




"'"1/~"d/D
~0,6 -/
0, o square-ed!
/-o round-edg

S0,2 sharp-edgl
-M-sharp-edgl

0 0,2 0,4~ 0,6
v ive H
Figure 5: Partition efficiency of particles de
flow rate ratio of total to collector tube fle
collector tube geometries

Based on the negligible density difference
particles and the fluid, and the low tube Rey
for /Vlio <0.3 all particles follow the s
the chamber of the downstream enlargement.

In addition to the experiments, numerical
performed using CFX 12.0 from ANSYS Gr
Figure 6 shows e.g. the streamlines close t
tube for the square-edged geometry (B) at a
number Re =0.5 and a flow split ratio
magnification of the streamlines at the collec
that a part of the streamlines is ending at th
the laboratory experiments it was also
particles sometimes stopped at the edge and
some time before they move to the enlargem


to the collector tube flow.


I hl1.1 34 -00 / l

Figure 6: Streamlines and velocities after the duct with the
square-edged collector tube. Velocities according to
Re = 0.5 and flow split ratio of 0.3 calculated with CFD
ANSYS.

The laminar flow within the circular duct is defined and
given by a parabolic velocity field. At the tube outlet the
flow expands to the side and a part of the flow enters into
the collector tube, as shown in Figure 6.


per separation. The main flow is divided into a central area where mainly
gives the best large particles are transported, and an outer area where the
Iph. small particles move with the flow to the chamber. It is
noticeable that the round-edged (A) and square-edged (B)
geometries produce more fluid rotation in the chamber of
down flow enlargement and the particles carried by the flow
undergo circular motion. This is the main disadvantage of
S= 0.39 two geometries because the rotating particles collide with
these particles leaving the separation tube discharge and a
clear separation process is impossible. Furthermore, some
ged particles are stopped in the separator on the surface of the
ed square-edged geometry and disturb the segregation process.
ed inside Therefore the sharp-edged outside (D) geometry was used
for further analysis of the separation process. Figure 7
ed outside
I I shows simulation results of the flow field with different
0,8 1 Reynolds number. The distance between the separation tube
discharge and the collector was 19 mm and the flow split
pending on the ratio was 0.7.
ow at different
A low Reynolds number causes oval-shaped streamlines
between the particle output from the separation tube and the
:s between the collector tube. The trajectories expand at the first moment
Inolds numbers after the separation tube and form a round flow profile.
treamlines into Thereby streamlines from the outer area of the tube flow
end in the chamber and other streamlines from the central
area of the separation tube are removed through the
collector tube. It is assumed that the particles follow the
l studies were
flow lines. A description of the trajectories with a CFD
mbH, Germany.
program is helpful for the explanation. This assumption is
o the collector
possible because the density between the particle and the
tube Reynolds
fluid is negligible, the Reynolds number is low and there is
of 0.3. The
no wall effect in that area. With increasing Reynolds
tor tube shows,
number the trajectories are concentrated and form an
eedge. During
elongated stream profile. At a Reynolds number of Re = 40,
observed that
the stream flow concentrated in the collector tube before a
Stay there for
part of it was redirected and move away into the
lent chamber or






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

p = 1223 kg/m3, 9L = 124 mPas. Two flow rates were set to
fix the tube Reynolds number at Re = 1.8 and Re = 5.9
respectively. In Figure 8 the separation characteristics of
binary particle mixtures are presented. This was done by
injecting a low concentrated suspension into the laminar
flow tube with the piston. Depending on the flow split,
particles move either into the collector tube or into the
residual flow. At low ratios, the large particles primarily end
in the collector tube and the smaller ones mainly move with
the residual flow. With increasing flow through the collector
tube an increased number of small particles also is conveyed
into the collector tube.

The figure presents the partition coefficient in terms of the
ratio of the particles entering the collector tube divided by
the total number of individual particles passing through the
device.


Paper No 2596


enlargement chamber. In our experiments the particle
motion to a large extent agrees with the simulation results of
the streamlines.


Re = 0 5


~0,6 f

~0,4- J/ 61 -Re= 1.8; dlD = 0.53
~ 02 Re= 5.9; dlD = 0.53
,2 /a * Re= 1.8; dlD = 0.31

0 0,2 0,4 . 0,6 0,8 1
V INo [-]
Figure 8: Partition efficiency of particles depending on the
flow rate ratio of collector tube flow to total flow by a
simultaneous injection of particles with two diameters.

It is visible that all particles pass through the collector tube
as soon as the flow split ratio is 0.75 for small particles and
0.62 for larger particles. Also a remarkable observation is
the coincidence of the graphs for the same Reynolds number.
With increasing Reynolds numbers the functions of partition
coefficients move to larger flow split ratios. In Figure 9, the
dependence of the partition coefficient on the flow rate ratio
is shown for three different particle diameters
simultaneously injected.


Re = 10


I as4-ool


4.010e-002


Re = 40


'1.257e-005
[m a^-1]


Figure 7: Comparison of three Reynolds number with the
sharp-edged outside collector tube geometry, Vo~ir = 0.7 .

At higher Reynolds numbers in the laboratory experimentS,
the particles first enter the collector tube before changing
their direction and later moving into the enlargement
chamber. Subsequent particles collide with them and cause a
decrease of the separation. It follows an optimal Reynolds
number for the separation process between 1 and 10.
For the next experiments, the liquid properties are given as:


O 0,2 0,4 0,6 0,8 1

Figure 9: Partition efficiency of particles depending on the
flow rate ratio by a simultaneous injection of particles with
three diameters and a constant tube Reynolds number of
Re = 3.6


veloci
5.603e-002

4.202e-0M

2.802%*002

1.4ole-on

8.334**007
[m a^-1]


Velocity
7.250e-002




3.625e-002




1.723e-006
[m e^-1)


/I


/Iln




























































VI/V
~ 0,26
0,36
-n-0,41
0 0,52



.P."'c:


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


Paper No 2596


With increasing Reynolds number, the separation
characteristic decreases and a distinct separation of particles
or a separation of particles and fluid is hardly possible.
Figure 10 shows the partition efficiency for various fluid
flow ratios. The tube Reynolds number was kept constant at
1.8. It is apparent that a higher flow split ratio increases the
partition efficiency.







-00,26
C 0,6 o/ 0,31
*0~ / 0,47
li- o, _-P 0,52
,L( 0,57

-j ^





0,3 0,4 ~d/D [-] osoe

Figure 10: Partition efficiency for a tube Reynolds number
Re = 1.8 at different flow split ratios and three particle
diameters

Next experiments were performed at tube Reynolds number
of 26 with a glycerol water mixture of 76 wt-%. The fluid
has a density of p =1197 kg/m3 and a viscosity of
9L = 27 mPas. A ternary particle mixture was injected in the
separation tube and the separation characteristics are
presented in Figure 11. With increasing Reynolds number at
Re > 10, the classification effect of the particles decreases
and becomes poor. This behaviour has been observed in the
experiments and was as well found in simulations.


Figure 12: Particle position in a laminar tube flow

The fluid for the measurement had a viscosity 9L = 370 mPas
and a density p = 1242 kg/m3. Figure 13 marks the position
of the particles for Reynolds number of 1.2 and a particle
diameter of 2.04 mm. Fifteen measurements were
performed for each position of the laser-sheet system. The
average value is described as:


n


Particles in a low Reynolds number flow slowly move
towards their equilibrium radius. The positions are however
fluctuating from the center of the tube to a radius of 2.0 mm,
but the average value for every position of the laser-sheet
system line is around the Bauckhage equilibrium radius mn
Figure 13. Within a flow distance of 1.555 m the particles
could not yet find a position on their equilibrium radius and
continuance at this radius.


S2,5 -average value

2 . ,- Bauckhage equilibrium









dimee of d/ 03





and after this distance mos of the pasrtilsfn hi
equilbriu radus cose to he ijcalculastedn Bachaerais


1


u 0,8
P
,"
'0,6
u

~0,4
.P
t:
~ 0,2


d/D [-]

Figure 11: Separation characteristic of a ternary particle
mixture with a Reynolds number of 26.

Measurements of the radial position of the particle in the
separation tube discharge were executed with a laser sheet
illumination. The data were recorded at a distance of
0.455 m and 1.555 m from the injection position. Particles
were expected to find their positions on the equilibrium
radius after an inlet distance. Figure 12 shows a photo of a
particle position in a laminar flow tube.










-average value
-- Bauckhage equilibri
I *


* . ;


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


Paper No 2596


the sharp-edged outside geometry. This has been verified by
experiments and simulations. In further measurements the
um radius length of the separation tube and the concentration influence
of particles in the fluid will be investigated. Based on the
nnt sband e tbnsse blof this knowledge to micro



Acknowledgements

We want to express our gratitude to the German Science
1,4 1,6Foundation (DFG) for the financial support.

the tube at a
e diameter of References

Bauckhage K., Zur Entmischung nicht sedimentierender
Isitions of the Suspensionen bei laminarer Rohrstrijmung, Dissertation,
distances. The Universitiit Clausthal, (1973)
inning length.
lius according Bauckhage K., The buildingMechanism of Chains of Sphere
in Poiseuille Flow a Model for the Building Mechanism of
Red Cells Chains, Biorheology Vol. 12, p. 329 334, (1975)

(5) Bauckhage K., Mdglichkeiten der Nutzung hydrodynamisch
bedingter Fest/Fltissig Entmischungen fiir Trennaufgaben

average value Bhattacharya S., Mishra C., Bhattachar a S., Analysis of
of 2.6. It can general creeping motion of a sphere inside a cylinder, J.
e value for an Fluid Mech., Vol. 642, p.295-328, (2010)
0.13 mm and
have reached Buggisch H., Muckenfuss S., Particle Concentration in
ube at that Shear Flow Nearby the Wall, Proceedings: Eurorheo
2002-01, Progress in Rheology-Theory and Applications,
p. 255-258, (2002)


S2,5

S2

S1,s




1 0 5


0, ,distancO h of the 11aser she 2
from the injection position [m]
Figure 14: Position of single particles along
Reynolds number of 2.6 and a related particl
d/D = 0.32

The standard deviation of the local radial po
particles was taken at distinct transportation c
scatter of the data is reduced with increasing r
The mean values lies close to equilibrium rad
to equation 2.





Figure 15 indicates the mean distance to the
of the particle position for a Reynolds number
be seen that the mean distance to the average~
injection distance of 1.155 m is approximately
it can be stated that the particles practically
their equilibrium position within the t
transportation distance.


Huang L. R., Cox E. C., Continuous Particle Separation
Through Deterministic Lateral Displacement, Science Vol.
304, p. 987 990, (21 rI14,

Klein T., Charakterisierung von Biopolymeren, Proteinen,
Partikeln und Kolloiden mit Feldflussfraktionierung, GIT
Laborfachzeitschrift 11, p. 1224 1228, (1999)

Lou R., Pan X. H., Laminar light particle and liquid
two-phase flows in a vertical pipe, International Journal of
Multiphase Flow 29, p. 603-620, (2003)

Matas J.-P., Morris J. F. and Guazzelli E., Inertial migration
of rigid spherical particles in poiseuille flow, J. Fluid
Mechanics, Vol. 515, p. 171 -195, (21lrl4)

Muckenfuss S., Buggisch H., Shear Flow of Suspensions
under Wall Influence: Phenomenological Theor, and Effect,
PAMM Proc. Appl. Math. Mech. 3, p.414-415, (2003)

Muckenfuss S., Wall Effects in Shear-Flowing Suspensions,
Dissertation Universitiit Karlsruhe, (2006)

Pamme N., Continuous flow separations in micro fluidic
devices, Lab on a Chip, p. 1644 1659, (2007)

Rubin G, Widerstands- und Auffriebsbeiwerte von ruhenden,


0,5


3 0,4 -

2 0,2 -
~02
o ,1 -


0,46 0,56 0,66 0,76 0,86 0,96 1,06 1,16 1,26 1,36 1,46 1,56
distance h of the laser sheet
from the injection position [m]
Figure 15: Mean radial distance to the average value at a
tube Reynolds number of 2.6 and a related particle diameter
of d/D = 0.32.

Further measurements must be performed to find the
dependence of the tube length and the equilibrium radius for
other sizes and tube Reynolds numbers.


Conclusions

Particle separation in a laminar flow with a downstream
enlargement is possible and feasible. The classification
process of particles is favored at tube Reynolds numbers
between 1 and 10 and the best geometry for the process is






Paper No 2596 7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

kugelfijrmigen Partikeln in stationiiren, wandnahen
laminaren Grenzschichten, Dissertation Universitiit
Karlsruhe, (1977)

Saffman P. G, On the motion of small spheroid particles in a
viscous liquid, J. Fluid Mechanics, p. 540 553, (1956)

Segr6 G. & Silberberg A., Radial Particle Displacements in
Poiseuille Flow of Suspensions, Nature 189, p. 209 210
(1961)

Segr6 G. & Silberberg A., Behaviour of macroscopic rigid
spheres in Poiseuille flow, Part I, J. Fluid Mechanics 14, p.
115 135, (1962)

Segr6 G. & Silberberg A., Behaviour of macroscopic rigid
spheres in Poiseuille flow, Part II, J. Fluid Mechanics 14, p.
136 157, (1962)

Takagi J., Yamada M., Continuous particle separation in a
microchannel having asymmetrically arranged multiple
branches, Lab on a Chip, p. 778 784, (2005)

Walzel P., Wiggers H. & Schaldach G, Sorting of non
settling particles in a high viscous environment,
International Conference on Multiphase Flow, Leipzig,
(2007)

Yamada M., Nakashima M., Pinched Flow Fractionation:
Continuous Size Separation of Particles Utilizing a Laminar
Flow Profile in a Pinched Microchannel, Analytical
Chemistry, p. 5465 5471, (211114)




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