Japanese Journal of Applied Physics 49 (2010) 07HE14 BRIEF NOTE

Study of an Acoustic Field in a Microchannel

Teruyuki Kozuka, Kyuichi Yasui, Shin-ichi Hatanaka1 , Toru Tuziuti, Judy Lee, and Atsuya Towata
National Institute of Advanced Industrial Science and Technology (AIST), Nagoya 463-8560, Japan
1
The University of Electro-Communications, Chofu, Tokyo 182-8585, Japan
Received November 25, 2009; accepted February 5, 2010; published online July 20, 2010

Using a standing-wave field, it is possible to trap small objects at nodes of a sound pressure distribution. In the present study, a sound wave was
generated by a transducer outside of a microchannel, and propagated into a microchannel on a glass plate, where it generated a standing wave
field. When water containing alumina particles was injected into the microchannel, several layers of particles were formed in the sound field.
Moreover, when the ultrasound driving frequency was swept, it was possible to control the direction of the particle flow. The sound field was
numerically calculated and the experimental results are discussed. # 2010 The Japan Society of Applied Physics
DOI: 10.1143/JJAP.49.07HE14

R
ecently, noncontact micromanipulation techniques
have become necessary in micromachine technol-
ogy, biotechnology, and other fields. The radiation
pressure of ultrasound may be used for this purpose as it is
possible to trap particles at sound pressure nodes of a
standing wave field in the medium.1–3) Yasuda4,5) has studied
the manipulation of blood cells by using a standing wave
field in a microchannel with length on the order of the
wavelength. Manneberg et al.6) have studied the manipu-
lation of biological cells in a microchannel on a glass plate
irradiated with ultrasound through a metal block. Masuda
et al.7) have attempted to control a microbubble flow in a
blood vessel using acoustic radiation pressure. Yamakoshi
and Miwa8) have analyzed the behavior of a microbubble
in a standing wave field generated by two focused-type
transducers in water. In the present study, a sound wave is
generated far from a microchannel, and a standing wave
field was formed in the microchannel. Solid particles were
trapped in the sound pressure nodes of the sound field, and
the particle flow direction was controlled by changing the
frequency.
Figure 1 shows a glass plate with a microchannel. The
size of the glass plate is 50  50  5 mm3 . A microchannel
of 1  50  1 mm3 was formed at the center of the plate.
The microchannel is surrounded by three glass walls, and the
top surface is open to air. The glass plate was sandwiched
in the upper right corner of the glass plate using a holder.
Two types of experimental apparatus are shown in Fig. 2.
The apparatus in Fig. 2(a) is described as follows. A glass
plate floats on the water surface in a water tank. A circular
transducer of 15 mm is set on the bottom of the water tank
at an angle of 30 from the horizontal direction. The sound
beam is directed into the microchannel on the glass plate.
The apparatus in Fig. 2(b) is described as follows. A
rectangular transducer of 30  5 mm2 is adhered to the left
end of a glass plate using grease, and the transducer was
exposed to air.
When a suspension of alumina particles was poured into
the microchannel, the alumina particles agglomerated into
approximately 4 – 7 layers in both experimental apparatus
when the fluid was motionless. Figure 3(a) shows the
experimental result for a frequency of 4.5 MHz with the
experimental apparatus in Fig. 2(b); particles agglomerated
into six layers in the microchannel. The number of sound
pressure nodes in the standing wave field is as follows:

E-mail address: kozuka-t@aist.go.jp
07HE14-1
Glass plate Microchannel
5mm
Holding point
50mm
50mm
Fig. 1. (Color online) Glass plate with microchannel.
Glass Plate
Microchannel
PZT (Air)
Glass Plate
(Water) PZT Microchannel
30°
40~50mm
(a) (b)
Fig. 2. (Color online) Experimental apparatus. (a) Indirect type. (b)
Direct type.
N ¼ 2d f =c, where d is the distance between two opposite
walls (d ¼ 1 mm), c is the sound speed (1500 m/s), and f is
the ultrasonic frequency. As the half wavelength is 0.166 mm
when the frequency is 4.5 MHz, N is 6. The number of layers
in the experiment is consistent with the above estimation.
The ultrasound wave propagates into the microchannel
through the glass plate, and a standing wave field is formed
in the microchannel. As a result, particles were trapped at
the sound pressure nodes of the sound field. In the following,
the experimental results for the apparatus in Fig. 2(a) will
be shown, as the results were the same for both apparatus.
When the frequency of the ultrasound was changed
under this condition, the particles were spatially shifted. To
increase and decrease N by one, f must be changed to
3.75 MHz (N ¼ 5) and 5.25 MHz (N ¼ 7), respectively.
When the frequency was repeatedly swept from 3.75 to
5.25 MHz every 0.2 s, the particles moved to the right as
shown in Fig. 3(b). In contrast, the particles were shifted
to the left, as shown in Fig. 3(c), when the frequency
was swept from 5.25 to 3.75 MHz. It was confirmed that
particles agglomerated on one side of the microchannel upon
changing the frequency.
# 2010 The Japan Society of Applied Physics
Jpn. J. Appl. Phys. 49 (2010) 07HE14
Microchannel
1mm
(a) (b)
Fig. 3. Particle behavior in the straight microchannel. (a) Fixed
frequency of 4.5 MHz. (b) Frequency swept from 3.75 to 5.25 MHz every
0.2 s. (c) Frequency swept from 5.25 to 3.75 MHz every 0.2 s.
1 mm 1 mm
PZT
Micro channel
Glass plate
5mm
50mm
30mm
30°
50mm
(a) (b)
Fig. 4. (Color online) Particle behavior in the branched microchannel.
(a) Glass plate. (b) Fixed frequency of 4.5 MHz. (c) Frequency swept from
5.25 to 3.75 MHz every 0.1 s.
Next, a microchannel with a branch on the glass plate, as
shown in Fig. 4(a), was used. The transducer, which was
attached to the left side of the glass plate, was driven at
4.5 MHz and a suspension of polyethylene particles was
poured from the upper end of the microchannel at a speed of
16.7 mm/s using a pump. Figure 4(b) shows the behavior at
the branch point in the microchannel. The particles flowed
in a few layers along the microchannel. When the driving
frequency was swept from 5.25 to 3.75 MHz every 0.1 s,
it was observed that the particles were pushed to the left
side of the channel, although particles initially flowed on the
right side of the channel [Fig. 4(c)]. In other words, it was
possible to control the direction of the particle flow by
ultrasound.
The sound pressure distribution in the microchannel was
of interest. Therefore, numerical analysis was performed by
the finite element method (FEM; COMSOL Multiphysics).
The FEM calculations were performed using a two-dimen-
sional model to reduce the required memory size of the
computer used for the calculations. Although it was possible
for the sound wave to propagate from the three glass
surfaces in this study, the sound wave from the left surface,
which is nearest to the transducer, most strongly affects the
sound field. In the present study, calculations were per-
T. Kozuka et al.
High
Low
(a) (b) (c)
Fig. 5. Numerically calculated sound pressure distribution in the
branched microchannel: (a) 3.75, (b) 4.50, and (c) 5.25 MHz.
(c)
formed with the assumption that the side wall of the
microchannel uniformly vibrates. The width of the micro-
channel was 1 mm, as in the experiment. It was assumed
that there was no fluid flow in the microchannel.
Figure 5 shows the calculated sound pressure distribution
in the horizontal plane of the branched microchannel when
the left surface of the straight microchannel is uniformly
vibrating. The scale of each figure is normalized by the
maximum sound pressure. Although a standing wave field
was formed in all parts of the branched microchannel, the
sound pressure distribution is very complicated at the branch
point. The position of the particles trapped in the sound field
is different for different ultrasonic frequencies as shown in
Fig. 5. Moreover, the pressure nodes move to the left as the
ultrasonic frequency decreases. The particle trajectory is
determined by both the flow of the medium and the shifting
(c) sound field, in aggrement with the experimental observation
in Fig. 4(c).
In conclusion, a standing wave field was formed in a
microchannel with a 1  1 mm2 cross section on a glass
plate. An ultrasonic wave successfully propagated into the
microchannel on the glass plate. When a suspension of
particles was introduced into the microchannel, the particles
agglomerated into a few layers, each separated by a half
wavelength. A similar phenomenon was seen in a branched
microchannel. It was possible to control the direction of
the particle flow by changing the ultrasound frequency. A
sound field was numerically calculated by FEM under the
experimental conditions and the experimental results were
discussed.
Acknowledgment This work was supported by a Grant-in-Aid for
Scientific Research (21560254) from the Japan Society for the Promotion of
Science.
1) T. Kozuka, K. Yasui, T. Tuziuti, A. Towata, and Y. Iida: Jpn. J. Appl. Phys.
47 (2008) 4336.
2) T. Kozuka, K. Yasui, T. Tuziuti, A. Towata, and Y. Iida: Jpn. J. Appl. Phys.
48 (2009) 07GM09.
3) T. Kozuka, T. Tuziuti, H. Mitome, F. Arai, and T. Fukuda: Nihon Kikai
Gakkai Ronbunshu C 67 (2001) 1269 [in Japanese].
4) K. Yasuda: Jpn. J. Appl. Phys. 36 (1997) 3130.
5) K. Yasuda: Sens. Actuators B 64 (2000) 128.
6) O. Manneberg, S. M. Hagsater, J. Svennerbring, H. M. Hertz, J. P. Kutter,
H. Bruus, and M. Wilklund: Ultrasonics 49 (2009) 112.
7) K. Masuda, Y. Muramatsu, S. Ueda, R. Nakamoto, Y. Nakayashiki, and K.
Ishihara: Jpn. J. Appl. Phys. 48 (2009) 07GK03.
8) Y. Yamakoshi and T. Miwa: Jpn. J. Appl. Phys. 48 (2009) 07GK02.
07HE14-2 # 2010 The Japan Society of Applied Physics