close

Вход

Забыли?

вход по аккаунту

?

FEDSM2003-45757

код для вставкиСкачать
Proceedings of ASME FEDSM’03
ASME 2003 Fluids Engineering Division Summer Meeting
Proceedings
of ASME
FEDSM’03
Hawaii,
United States
of America,
July 6-10,
2003
th
4 ASME_JSME Joint Fluids Engineering Conference
Honolulu, Hawaii, USA, July 6-10, 2003
FEDSM2003-45757
MICRO CHEMICAL REACTION FLOW MEASUREMENT USING PIV AND LIF TECHNIQUE
Kyosuke Shinohara
Nuclear Engineering Research Laboratory,
University of Tokyo, Tokai-mura, Ibaraki, Japan
Yasuhiko Sugii
Nuclear Engineering Research Laboratory,
University of Tokyo, Tokai-mura, Ibaraki, Japan
Induced Fluorescence) technique have been applied to macro
scale flow. Sasaki et al. [6] measured chemical reacting flow.
They measured velocity distribution using fluorescent particle
or PIV and pH distribution using fluorescent dye, quinine for
LIF. Diffusion of acetic acid jet into ammonia hydroxide was
observed. Acetic acid solution of 0.1 mol/L was used. The
acetic acid solution flew in the duct with 100 mm x 100 mm
cross-section. The velocity was the acetic acid solution was
fixed to be 5 mm/s. At the center of the duct, the small circular
nozzle with 2mm φ was set concentricly. From the nozzle,
Ammonia solution of 0.1 or 0.5 mol/L was ejected into the duct.
The injection velocity was set to be UN = 0.1 ~ 0.5 m/s, i.e.,
Re= 200 ~ 1000. The acetic acid and ammonia reacted in the
duct. The reacting factor of the above reaction was k =
108m3/(mol s). The Schumidt number was Sc= 250.
Recently, the micro PIV technique [7-9], in which a velocity
distribution with micro resolution can be measured using
microscope and pulsed laser system, has been developed. Micro
LIF technique, using temperature dependence fluorescence
dyes, has been also developed to measure a temperature
distribution caused by micro heater in micro fluidic device [10].
However, the obtained results did not show enough spatial
resolution and measurement accuracy. Furthermore, the
interaction between chemical reaction and flow field has not
been investigated.
In this study, the measurement technique of pH and velocity
distribution has been developed using micro PIV and micro LIF
technique. These techniques were applied for neutralization
reaction between acetic acid and ammonia in micro fluidic
device.
ABSTRACT
Using micro PIV technique and micro LIF technique,
velocity distributions and pH distributions of chemical reacting
flow in micro fluidic device were measured. The micro fluidic
device was Y-junction channel. In order to generate a
neutralization reaction, acetic acid and ammonia hydroxide
were introduced into each channel respectively. The results of
velocity profiles of the chemical reacting flow corresponded
closely to the theoretical profile of Poiseuille’s flow. This was
characteristic of laminar flow. The results of pH distribution of
the chemical reacting flow showed two kinds of distribution. In
one case interface was straight and mixing was not noticeable
at higher flow speed. In the other hand mixing or molecular
diffusion due to chemical reaction definitely caused at lower
flow rate. Applying both results we compared the experimental
diffusion with theoretical value. In one case the experimental
result corresponded close to theoretical value. However, it did
not agree with theory in the other case. The results indicate that
acetic acid and ammonia hydroxide mixed faster than theory in
micro scale.
Key words: chemical reaction, micro fluidic device,
molecular diffusion, micro PIV, micro LIF
INTRODUCTION
Micro fluidic systems are attracting great interest from many
research groups [1]. The miniaturization and integration of the
various chemical operations have many potential advantages
such as increased speed, efficiency, portability, and reduced
consumption. To realize µ-TAS (micro Total Analysis System),
the essential factor governing chemical reactions on microchips
is one of the key technology issues. It affects overall
performance of µ-TAS. A microunit operation, such as mixers,
reactors and so on, and integrated a complicated chemical
system on a microchip, continuous flow chemical processing
(CFCP) has been proposed [2] [3]. Chemical reaction and
molecular transport in the process were realized in and between
continues flows in multiphase network.
In order to investigate the dynamics of chemical reacting
flow, PIV (Particle Image Velocimetry) [4] and LIF [5] (Laser
EXPERIMENTAL METHOD
To generate a reaction inside micro fluidic device, Yjunction channel [11] was used as the test section. Fig.1 shows
a schematic view of the micro fluidic device, which consisted
of micro channels connected to two inlets in the left hand and
one outlet in the right hand. The micro channel was made of
quartz glass with 100 µm width and 30 µm depth. The shape of
cross section was almost rectangular with half-circle bottom.
Two channels were connected to one channel with Y shape, so
Copyright © 2003 by ASME
1
Copyright © 2003 by ASME
Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/25/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
cross-correlation method and Gaussian peak fit for sub pixel
analysis. The authors have the highly accurate PIV technique
[12], which can improve measurement accuracy and spatial
resolution for time-averaging velocity distribution, averaging
instantaneous correlation function technique.
that the two different fluids can be gathered into the one
channel.
Neutralization reaction caused by acetic acid and ammonia
solution was settled as chemical reacting flow. The reaction is
expressed as,
CH 3COOH + NH 4OH ⇒ CH 3COONH 4 + H 2O (1)
Two kinds of fluids were introduced using two micro syringes,
250 µL, pressed by syringe pumps with the constant flow rate,
3 µL/h to 200 µL/h. Since Reynolds number was in the range of
0.02 to 1.67, the flow was laminar flow. Visualized images
were taken using cooled CCD camera (1280 x 1024 pixel, 12bit
monochrome, 8 frame/s) through a microscope equipped with
filter set and an objective lens (oil immersion a magnification
M = 60, numerical aperture N.A. = 1.25, for micro PIV and M
= 40 for micro LIF, N.A. = 1.0).
Cooled CCD Camera
(1280×1024 pixels)
Microscope
Color filter
Micro syringe 250 µl
PC
Mercury lump (LIF)
Oil Immersion Objective
Lens (40x, 60x)
Microchip
Mirror
Double Pulse Nd:YAG Laser
λ=532 nm (PIV)
FIG.1 SCHEMATIC
DEVICE
VIEW
OF MICRO
FIG.2 EXPERIMENTAL SET UP FOR MICRO PIV AND
MICRO LIF
FLUIDIC
VELOCITY MEASUREMENT
160
y position [µm]
Micro PIV technique
Micro PIV technique is a quantitative method for measuring
velocity fields with micro resolution instantaneously. Fig.2
shows an experimental set up. Fluorescent particles seeded into
the test fluids, 0.4% concentration, were illuminated using the
double pulsed Nd:YAG laser beam (532 nm; green) from
downside of the microscopic stage. The fluorescent particle
with 1 µm diameter absorbs 535 nm green light and emits 575
nm orange light. The particle images were captured using CCD
camera, which can record 164 pair images, equipped with
optical filter (λ = 550 nm). Since double pulsed laser can fire
two laser beams in short time, time accuracy of images can be
enhanced using a pulse generator. Two kinds inlets flow rate set
to be 100 µL/h.
Fig.3 shows visualized image of micro channel illuminated
by halogen lamp. The observed region was 236 x 189 µm with
each pixel representing a 0.184 x 0.184 µm area. The image
was captured at 15 µm deep, center of the channel. Both
fluids started to mix at x = 20 µm. The width of one outlet
channel was same as that of two inlet channels. Since it was
illuminated by halogen lamp, lines of wall were clearly
observed. Fig.4 shows fluorescent particle image illuminated
by Nd:YAG double pulse laser. Particles appeared as bright
point sources of light randomly distributed through the channel
and flew from right to left. Background noise due to out-offocus particles and the light scattering caused by channel wall
and so on. The refraction at the channel wall was negligibly
small and the flow inside the channel, very close to wall, can be
observed clearly. A particle diameter in image was observed as
6 or 7 pixels.
Under a microscopic observation, particle diameter in image
becomes larger and particle density becomes smaller. Therefore,
the measurement accuracy was reduced using combination of
120
80
40
0
0
40
80
120
160
200
x position [µm]
FIG.3 VISUALIZED IMAGE OF MICRO CHANNEL
FOR MICRO PIV
y position [µm]
160
120
80
40
0
0
40
80
120
160
200
x position [µm]
FIG.4 FLUORESCENT PARTICLE IMAGE
Copyright © 2003 by ASME
2
Copyright © 2003 by ASME
Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/25/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
Velocity distributions
Fig.5 and Fig.6 show obtained time-averaged velocity
distributions in non-reacting case and in reacting case at flow
rate 100 µL/h. Interrogation window was taken 64 x 16 pixels
with 50% over lap. Vectors represent velocity of fluids and
direction at the point. In non-reacting ion-exchange water was
introduced into both channels. In reacting case acetic acid was
done into lower channel in image. Also, ammonia hydroxide
was done into the other.
In both cases the two-fluids were fully developed, parabolic
axial velocity distributions. Since channel width decreased half
size after the junction, the flow was accelerated and then fully
developed again. The velocity vectors very close to the wall
were measured and it was found that the wall-normal
component of the velocity vectors was close to zero. Since the
variation in time of velocity distribution was enough small,
pulsation of flow due to syringe pump was enough small. The
velocity corresponded closely to the theoretical profile of
Poiseuille’s flow.
Velocity profiles
Fig.7 and Fig.8 show velocity profiles in non-reacting case
and reacting case at x = 23.5 µm, upstream, 118 µm, and 230
µm, downstream. In the upstream, flow speed was maximized
at the center of inlet channels. Also, it was minimized at center
of outlet channel. As stream proceeded, the velocity profiles
approached parabolic shape. In the downstream its shape was
fully parabolic. Therefore, the velocity was maximized at the
center of the outlet channel. Furthermore, it was found that the
velocity close to wall was zero. At x = 118 µm, the velocity
became zero at y = 20 µm and 168 µm. At x = 230 µm, it also
did at x = 38 µm and 147 µm. The result indicates that the
channel width decreased after the junction.
x = 23.5
x = 118
x = 230
30
Flow speed [mm/s]
50 [mm/sec]
y position [µm]
160
120
20
10
80
40
0
50
100
150
200
150
200
y position[ µ m]
0
0
40
80
120
160
FIG.7 VELOCITY PROFILES
IN NON-REACTING CASE
200
x position [µm]
FIG.5 VELOCITY DISTRIBUTION
IN NON-REACTING CASE
x = 118
x = 230
x = 23.5
50 [mm/sec]
30
Flow speed [mm/s]
y position [µm]
160
120
80
20
10
40
0
0
0
40
80
120
160
200
50
100
y position[ µ m]
FIG.8 VELOCITY PROFILES
IN NON-REACTING CASE
x position [µm]
FIG.6 VELOCITY DISTRIBUTION
IN NON-REACTING CASE
Copyright © 2003 by ASME
3
Copyright © 2003 by ASME
Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/25/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
one region of the molecule to another. The displacement of
electrons results in increased potential energy of the molecule
from the ground state (S0) to the first electronic excited state
(S1). When the energy state returns to the ground state,
fluorescent light emission takes place as a radiative process.
However, there are many non-radiative processes that can
compete with the light emission and reduce the fluorescence
efficiency. They depend on the structure of the molecule in a
complicated fashion. The ratio of the total energy emitted per
quantum of energy absorbed by the molecule is called the
quantum efficiency, φ. The fluorescence energy I(Wm-3)
emitted per unit volume is defined as
(5)
I = I 0 Cφε
-2
where I0 is the incident light flux (W m ), C is the
concentration of the dye solution (m-3 kg), and φ is an
absorption coefficient (m2 kg-1). In most organic dyes, the
quantum efficiency ε is temperature dependent. The change in
fluorescence intensity is normally small, usually less than a
fraction of one per cent per K. However, the sensitivity of some
compounds such as Rhodamine B can be as high as 2% K-1. In
contrast, the absorption coefficient, e, does not have significant
temperature dependence, being less than 0.05% K-1 here. Thus,
it is possible to measure the temperature of the solution if one
can keep the incident light flux, I0, and concentration, C,
constant. As mentioned above, I0 is affected by various agencies
including convergence and divergence of the light sheet and
refraction of the light. To avoid this problem it is necessary to
normalize intensity by the intensity at the certain pH.
The fluorescence energy I (W m-3 ) emitted per unit volume
is defined as
(6)
I f (b) = I e (b) AφLεC
In reacting case velocity profiles drops around the interface.
This resulted from higher variance of velocity due to attached
particles. Since particles attached around the interface,
measurement accuracy was degraded. Also, there was
difference of max flow speed between non-reacting case and
reacting case. Two kinds of max flow speed were 25.2 mm/s
and 22.2 mm/s at x = 230 µm around the center of the channel
respectively. The result may owe to decreased flow speed due
to bubble in the upper region.
Discussion on velocity measurement
For micro PIV, the spatial resolution is limited by the
effective diameter of particle images. The diameter of
diffraction-limited point spread function in the image plane, ds
is given by
d s = 2.44M
λ
2 NA
(2)
where M is the total magnification of the microscope and NA is
the numerical aperture of the lens. Assuming a total
magnification M = 36, numerical aperture NA = 1.25 and the
wavelength of the recording light λ = 575 nm, the diameter of
the point spread function in this study is ds = 20.2 µm. The
actual image recorded on the CCD camera is the convolution of
the diffraction-limited image with the geometric image.
Approximating both the geometric and diffraction-limited
images as Gaussian functions, the resulting convolution is a
Gaussian function with an effective particle diameter de, where
2
2
(3)
d e = [ M 2 d p + d s ]1 / 2
For a magnification M = 36 and a particle diameter of dp = 1
µm, the effective particle image diameter projected onto CCD
camera is de = 6.3 µm.
In micro PIV, one is often interested in obtaining velocity
measurements with out of plane resolutions. In most PIV
applications, the thickness of a laser sheet, dz determines the
out of plane measurement domain. dz is usually chosen to be
smaller than the depth of field, δz.
Inoue and Spring [13] estimated the total depth of field as the
sum of the depth of field due to diffraction and geometric
affects
δz =
nλ
ne
+
2
MNA
NA
where If and Ie is the incident light flux(w m-2), A is fluorescent
intensity per unit, L is visualized length, C is the concentration
of the dye solution (m-3 kg), and ε is an absorption coefficient
(m2 kg-1). If I and ε is function of pH If is defined as
(7)
I f (b, PH ) = I e (b, PH ) AφLε ( PH )C
In addition strictly speaking Ie is defined as
I e (b, PH ) = I 0 e −ε ( PH )lC
(8)
where l is beam length. If If is normalized by I1f (pH = pH1) the
ratio is described as
If
I (b, PH )ε ( PH )
(9)
= e
I1 f
I1e (b, PH1 )ε ( PH1 )
and the follow equation is found:
If
e −ε ( PH ) lC
(10)
= −ε ( PH )lC
1
I1 f
e
Since l and C are constant, the ratio is dependent on only pH.
(4)
where n is the index of refraction of the immersion medium the
micro fluidic device and the objective lens, λ is the wavelength
of light in a vacuum, NA is the numerical aperture of the
objective lens, M is the total magnification of the system, and e
is the smallest resolvable distance of the image detector. In our
experiment equation (4) yields δz = 0.78 µm. The depth of field
was 2.6 % of depth of our micro channel. Therefore it was
confirmed that the evolution of velocity in z direction seldom
effected on the results.
Micro Laser Induced Fluorescence
To measure the pH distributions, fluorescent dye, Quinine,
was solved into the test fluids. The channel was illuminated
using a mercury lamp instead of the laser. Quinine absorbs 355
nm light and emits 450 nm light, whose emitted intensity is
varied depend on pH. To emphasize the fluorescent dye, two
color filters (absorption and emission) and one dichroic mirror
were used. Absorbing light around 355 nm and emission light
around 450 nm were generated using a color filter. Therefore,
the camera was captured only fluorescence of quinine. The
pH MEASUREMENT
Principle
To measure pH distribution, a fluorescent dye was used.
Fluorescence is a radiative decay process that occurs by
electronic transitions in molecules. After a fluorescent dye
molecule is exposed to an electromagnetic field, photons
entering the molecule cause displacements of electrons from
Copyright © 2003 by ASME
4
Copyright © 2003 by ASME
Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/25/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
solutions. Seven kinds of buffers pH = 4.90, 5.02, 5.90, 7.12,
7.75, 8.03, and 8.53 were taken for calibration. The test
solutions were made by mixing acetic acid and ammonia water
in the certain proportion. Intensity in the edge of image was
darker than that in the center area due to irregular illumination
by the mercury lamp. Irregular illumination was caused by
gathering into small area through an objective lens. In order to
eliminate the effect of irregular illumination, an algorithm, in
which calibration curves for each position was evaluated, was
constructed. This is described as follows:
observed region was 286 x 227 µm with each pixel representing
a 0.22 x 0.22 µm area shown in Fig.9. The observed region was
in 100 µm downstream from the region shown in Fig.3.
y position[ µ m]
200
150
x = 4×i
100
∑
'
I (i , j ) =
∑I
x = 4×( i −1) y = 4×( j −1)
y = 4× j
x = 4×i
∑
∑
(11)
(1 < i < 256,1 < j < 320)
I0
x = 4×( i −1) y = 4×( j −1)
50
0
50
100
150
200
where I is fluorescence intensity and I0 is fluorescence intensity
at the certain pH. The images were divided to windows. Its size
was 4 x 4 pixels. Therefore analyzed image size was 320 x 256.
Calibration curve was calculated in each area (4 x 4 pixels) and
many kinds of curves are obtained, so we can normalize
irregular illumination by using them.
Fig.11 shows a sample of relationship between emission
intensity and pH. The curve was estimated using a second-order
polynomial fit to the data of the form as
pH = A2 + A1 I ' + A0 I ' 2
(12)
where I’ is the fluorescence intensity normalized by its value at
pH =5.90 A2, A1, and A0 is constant. In bright region at x = 146,
y = 119 µm, normalized intensity varied dependent on pH.
However, in dark region at x = 17, y = 50 µm, it was almost
constant to pH. Therefore pH was not measurable in such dark
regions. Since quinine unites chlorine ion, in the range of less
then PH = 4.7 fluorescence intensity was degraded. Over pH =
9.0 fluorescence intensity was constant. Measurable pH range
was 4.7 to 9.0. The technique was applied to chemical reacting
flow at flow rates 3 µl/h and 200 µl/h.
250
x p o sitio n [ µ m ]
FIG.9 VISUALIZED IMAGE OF MICRO CHANNEL
FOR MICRO LIF
Recreation of property on time
Recreation of property on time was investigated. Fig.10
shows comparison of test solutions made in different time. In
both cases intensity value of the test solutions made in different
time were plotted on the same curve. It shows that test solutions
made in different time were available for a calibration
experiment.
1week after preparation
1day after preparation
1hour after preparation
250
bright region at (x,y) = (146,119)
dark region at (x,y) = (17,50)
200
4
150
100
normalized intensity
fluorescence intensity
y = 4× j
50
0
5
10
15
3
2
1
PH
FIG.10 RELATIONSHIP AMONG TEST SOLUTIONS
MADE IN DIFFERENT TIME
0
Calibration
In order to evaluate a relationship between emission intensity
and pH, calibration experiment was carried out. In order to
improve measurement accuracy the relationship between pH
and spatial-averaged intensity were evaluated using seven test
4
5
6
7
8
PH
FIG.11
A SAMPLE RELATIONSHIP BETWEEN
NORMALIZED INTENSITY AND pH
Copyright © 2003 by ASME
5
Copyright © 2003 by ASME
Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/25/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
pH distributions
Fig.12 shows an obtained pH distribution at flow rate
200 µL/h. Color represents pH. The length between the left
edge of the images and Y-junction point was 128 µm. Axial
velocity was 20.7 mm/sec at center of the image x = 175, y =
150 µm. Reynolds number was 1.67. Since in both side the
flow rate was same, the interface was observed around the
center of image. The variation of pH because of reaction was
also found. In the downstream region and upstream region,
since measurement accuracy of pH was small, the results in
these regions were not displayed. Since ammonia hydroxide
flew in the upper region, pH indicated higher values. Since
acetic acid flew, in the downer region pH indicated lower.
Fig.13 also shows an obtained pH distribution at flow rate
3 µL/h. Axial velocity was 0.31 mm/sec at center of the image
x = 175, y = 150 µm. Reynolds number was 0.02. The region
with pH = 7.0 caused by chemical reaction spread up as downer
stream. The result indicates that mixing and chemical reaction
between both fluids proceeded in the down stream.
y position[ µ m]
200
150
100
50
0
50
200
250
PH
DISCUSSION
Diffusion equation is theoretically described as follows,
∂ 2c
∂c
= −D 2
∂t
∂y
(13)
where c is concentration of hydrogen ion, t is time, and D is
diffusion coefficient. Solving the equation (13) the following
equation is obtained:
c
y −y
c( y , t ) = 1 φ ( 1
)
(y < y1)
2 Dt
2 Dt
c
y − y (y > y )
(14)
1
c( y, t ) = 2 [1 − φ ( 1
)]
2Dt
2Dt
where, c1 and c2 are initial concentration, y1 is position of
interface at t =0, φ is expressed as
φ ( x) =
200
y position[ µ m]
150
FIG.13 pH DISTRIBUTION IN CASE OF 3 µL/h
Discussion on pH measurement
The pH resolution obtained from the results shows ± 0.1 in
the range of 5.6 to 8.3 at the spatial resolution of 0.89 x 0.89
µm. It was evaluated using the algorithm. The value of 0.89
corresponded to 4 pixels in the images. Also, the value of 0.1
did to valiance of pH in 4 x 4 pixels area.
Fig.12 and Fig.13 were instantaneous distributions.
Therefore, the relationship between shutter speed of the CCD
camera and flow speed was very important. The shutter speed
was 10000 1/s (1.0 x 10-4 s) and fluids, whose flow rate was 200
µL/h, spent 9.65x10-3 s on passing from left edge to right edge
in the images. Since the shutter speed was about 1 % of the
time, the effect of shutter speed was enough small. Furthermore,
the depth of field was obtained using equation (4). Since the
depth of field was about 1.0 µm, 3.3 % of the depth of channel,
it was enough small to ignore the depth evolution of pH.
1
2π
x
∫e
−
ζ2
2
dζ
(15)
−∞
In addition the following relationship was defined,
t=
150
100
PH
50
100
150
200
x
u
(16)
where u was axial velocity in the x direction. Axial velocity u
was obtained using micro PIV technique.
Fig.14 shows concentration of hydrogen ion profiles
compared with theoretical value obtained using above
equations. Experimental profiles were obtained using obtained
pH and velocity distribution at x = 160 µm and u = 616 µm/s.
Solid lines and dash lines express the profiles at t = 0.38 s and t
= 0.32 s after start of mixing. In case of t = 0.32 s the
experimental value did not agree with theoretical value due to
difference of the interface position between experiment and
theory. In experiment interface did not exist at center of the
channel. In case of t = 0.38 s the gradient of experimental
profile was lower than theoretical gradient around the interface.
The result shows that acetic acid and ammonia mixed faster
than theoretical values. It was considered of the van der Waals
forces, electrostatic force, and so on.
50
0
100
x p o sitio n [ µ m ]
250
x p o sitio n [ µ m ]
PH
FIG.12 pH DISTRIBUTION IN CASE OF 200 µL/h
Copyright © 2003 by ASME
6
Copyright © 2003 by ASME
Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/25/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
Experiment t = 0.38 s
Theory t = 0.38 s
Experiment t = 0.32 s
Theory t = 0.32 s
[5] Sakakibara, J and Adrian R.J., 1999, ‘’Whole field
measurement of temperature in water using two-color laser
induced fluorescence,’’ Exp. Fluids, 26, 7-15
[6] Sasaki, T., 2000, “Visualization Study on Chemical
Reacting Flow using LIF” Master Thesis of Tokyo Univ.
JAPAN.
[7] Meinhart, C. D., Wereley, J. G., and Santiago, J. G., 1999,
‘’PIV measurement of a microchannel flow,’’ Exp. Fluids,
Vol.27, pp 414-419.
[8] Meinhart, C. D., Wereley, S. T., and Gray, M. H. B., 2000,
‘’Volume illumination for two-dimensional particle image
velocimetry, ‘’Meas. Sci. Technol., Vol.11, pp 809-814.
[9] Sugii,Y., Okamoto, K. and Madarame, H., Proc. 10th Int.
Symp. on Flow Visualization, 2002.
[10] Ross, D. Gaitan, M. Locascio, L. E., 2001, ‘’Temperature
Measurement in Microfluidic systems using a
Temperature-Dependent Fluorescence Dye,’’ Anal. Chem.,
73, 4117-4123.
[11] Ito, T., Uchiyama, K., Ohya, s., Kitamori, T., 2001,
“Aplication of Microchip Fabricated of Photosensitive
Glass for Thermal Lens Microscopy”, Jpn.J.Appl.phys, 40,
5469-5473
[12] Sugii, Y., Nishio, S., Okuno, T., Okamoto, K., 2000, ‘’A
highly accurate iterative PIV technique using gradient
method,’’ Meas. Sci. Technol., Vol.11, pp 1666-1673.
[13] Inoue S. and Spring KR, 1997, Video microscopy, 2nd ed.,
Oxford: Plenum Press
(×10-6)
Concentration [mol/L]
2
1
0
60
80
100
120
140
160
180
y position[µ m ]
FIG.14 CONCENTRATION OF HYDROGEN ION
PROFILES
CONCLUSION
The micro PIV technique and micro LIF technique to
measure velocity distributions and pH distributions in micro
fluidic device has been developed. Molecular diffusion was
estimated using obtained velocity and pH distribution.
Comparison with theoretical values indicated that chemical
reaction proceeded faster in the micro scale. The developed
method will allow us to enhance the researches on micro
chemical reacting flows and the results will demonstrate
advantage of µ-TAS.
ACKNOWLEDGEMENT
The authors thank Dr. Okamoto and Prof. Madarame of
university of Tokyo for discussion and Dr. Hibara, Dr. Tokeshi
and Prof. Kitamori of university of Tokyo, for discussion and
manufacturing of microchips.
REFERENCES
[1] Ho, C. M., Tai, Y. C., 1998, ‘’Micro-electro-mechanical
systems (MEMS) and fluid flows,’’ Annual Review of Fluid
Mechanics, Vol. 30, pp 579-612.
[2] Sato, K., Tokeshi, M., Kitamori, T., Sawada, T, 1999,
‘’Integration of Flow Injection Analysis and ZeptomoleLevel Detection of the Fe(II)-o-Phenanthroline Complex,’’
Anal. Sci., Vol.15, pp 641-645.
[3] Tokeshi, M., Minagawa, T., Uchiyama, K., Hibara, A., Sato,
K., Hisamato, H., and Kitamori, T., 2002, ‘’Continuous-Flow
Chemical Processing on a Microchip by Combining
Microunit Operations and a Multiphase Flow Network,’’
Anal. Chem., 74, 1565-1571.
[4] Keane, R. D., and Adrian, R. J., 1992, ‘’Theory of crosscorrelation analysis of PIV, Applied Scientific Research,’’
Vol.49, No.3, 191-215.
Copyright © 2003 by ASME
7
Copyright © 2003 by ASME
Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/25/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
Документ
Категория
Без категории
Просмотров
10
Размер файла
1 033 Кб
Теги
fedsm2003, 45757
1/--страниц
Пожаловаться на содержимое документа