Sei sulla pagina 1di 6

Modeling of the Effect of Ultrasonic

Frequency and Amplitude on Acoustic


Streaming

Young Ki Lee, Jeong IL Youn, and Young Jig Kim

Abstract properties of castings and interfere with the interactions of


The application of ultrasound to casting process can solids and liquids [4].
improve the quality of the casting through the effects of In general, the effective influence of high power ultra-
grain refinement, degassing, wetting, deagglomeration sound on the liquid metal is attributed to the occurrence of
and dispersion. For treating large volume of liquid metal, mechanically violent cavitation bubbles, which activate and
Computational Fluid Dynamics (CFD) study has recently increase the crystallization nuclei in the melt [5–9]. Besides
developed an analysis tool capable to model acoustic cavitation, another phenomenon occurs during ultrasonic
streaming and cavitation in liquid metal. In general, the injection into fluids is acoustic streaming that is defined as a
cavitation zone is a very small region in fluid, acoustic steady fluid motion caused by the attenuation of ultrasonic
streaming is main factor for scale-up of ultrasound waves in fluids [10].
process because that is a non-linear physical effect which Acoustic streaming acts as a stirring mechanism, causing
can assist in effective dispersion of cavitation bubble with an increased part of the liquid volume to be treated by the
propagation of the ultrasonic wave in liquid. But previous local cavitation zone [5].
studies of CFD have limited theoretical model revision to For the application of ultrasonic treatment in the casting
accurately predict cavitation phenomenon. Therefore, this process it is important to be able to predict the treatment time
study is focused on the effect of ultrasonic injection of arbitrary volumes a priori. Heretofore ultrasonic treatment
conditions and fluid characteristics on acoustic streaming. times have often been determined experimentally for certain
The acoustic streaming velocity in fluid was measured specific geometries, but there is no general applicable model
using Particle Image Velocimetry (PIV) and CFD mod- for the prediction of the treatment time has been developed
eling was performed to predict the acoustic streaming and yet. To be able to predict the ultrasonic treatment time of a
treatment time in specific volume of fluid by ultrasonic volume it is first required to know the acoustic streaming in
injection. fluid. Acoustic streaming in fluid is generally governed by
the ultrasonic intensity, and the ultrasonic intensity is
determined by Eq. (1).
 
Keywords
Ultrasound Acoustic streaming Particle image
1
velocimetry CFD modeling I ¼ qcð2pfAÞ2 ð1Þ
2
where q is the density of propagation medium, c is the speed
of ultrasound in the propagation medium, f is the ultrasonic
Introduction frequency, and A is the ultrasonic amplitude [3].
Therefore, this study is focused on the effect of ultrasonic
Over the past years, many studies have been published on frequency, amplitude and fluid density on acoustic stream-
the application of ultrasonic vibrations to liquid metal pro- ing. The acoustic streaming velocity in fluid was measured
cessing [1–3]. Ultrasonic processing can significantly affect using Particle Image Velocimetry (PIV) and CFD modeling
heat and mass transfer in liquids, modify the structure and was performed to predict the acoustic streaming and ultra-
sonic treatment time in fluid by ultrasonic injection.
Y. K. Lee  J. I. Youn  Y. J. Kim (&)
School of Advanced Materials Science and Engineering,
Sungkyunkwan University, Suwon, 16419, Korea
e-mail: yjk1122@skku.edu

© The Minerals, Metals & Materials Society 2019 1573


C. Chesonis (ed.), Light Metals 2019, The Minerals, Metals & Materials Series,
https://doi.org/10.1007/978-3-030-05864-7_199
1574 Y. K. Lee et al.

Fig. 1 Captured image of


moving reflectors in the distilled
water mixed with ZnCl2 powder;
a before and b just after ultrasonic
injection

(gauge pressure = 0 kPa). The boundaries except


Experimental Procedure velocity-inlet, pressure-outlet and are set to be wall bound-
ary, and the heat transfer between wall boundary and fluid is
In an experimental environment, the opaque molten metal is ignored.
not accessible with PIV technique for flow velocity mea- The governing equations contain continuous and
surement. Many liquid properties related to fluid flow are momentum (N–S) equations, turbulence and turbulence
similar to water and aluminum [5]. Therefore, in this study, dissipation equations are also considered because turbulence
distilled water was used instead of liquid metal. A small flow may form during ultrasonic treatment [11].
amount of reflectors was added to the water bath as a tracer Governing equation that the convective motion can be
as shown in Fig. 1. After starting the ultrasonic injection the modeled is used the Navier–Stokes equations as shown in
flow inside the vessel, movement of the tracer particles was Eq. (2) [11–14].
taken by a high speed video camera. Then, the captured
images were processed by a PIV software to determine an @q
Mass : þ rðqmÞ ¼ Sm
averaged vector field of acoustic streaming. @t
To predict the effect of fluid characteristics on the @
Momentum : ðqmÞ þ rðqmmÞ ¼ rp þ qg þ F
acoustic streaming, the density was controlled as 1, 1.2, 1.4, @t !
1.6, and 1.8 g/cm3 by mixing with ZnCl2 powder and dis- @ X
Enegy : ðqEÞ þ rðmðqðE þ pÞÞ ¼ r hj J j þ Sh
tilled water. And to predict the effect of ultrasonic injection @t j
conditions on the acoustic streaming, the resonant frequency
was controlled as 15, 20, and 25 kHz, and the amplitude was ð2Þ
controlled as 20, 28, 38, 48, 66 lm. Turbulence model is used the standard k–e model that is
essentially a high Reynolds number model and assumes the
existence of isotropic turbulence and the spectral equilibrium
Computation Fluid Dynamics (CFD) Modeling [11, 14]. The turbulence kinetic energy and its rate of dis-
sipation are obtained from transport equation. The k–e
CFD modeling was performed by ANSYS Fluent software to model constants Cl, Ce1, Ce2, re and rk were set as 0.09,
predict the acoustic streaming and ultrasonic treatment time in 1.44, 1.92, 1.3 and 1.0, respectively [15].
the fluid. Figure 2 shows the three-dimensional computational The time-dependent interaction of the sonotrode and the
geometry and boundary conditions used in present study. fluid is modeled using the acoustic equation proposed by
The boundary of the sonotrode tip from which ultrasound Ffowcs-Williams and Hawkings. The Ffowcs-Williams and
is introduced into the fluid is set to be the velocity inlet Hawkings (FW-H) equation is essentially an inhomogeneous
boundary. The measured flow velocities by PIV technique wave equation that can be derived by manipulating the
were used as velocity inlet conditions for CFD modeling. continuity equation and the Navier-Stokes equations. [11].
The outlet boundary is set to be free-pressure outlet, the The Ffowcs-Williams and Hawkings acoustic equation is
pressure of the outlet is considered as one atmosphere. given below Eq. (3).
Modeling of the Effect of Ultrasonic Frequency … 1575

Fig. 2 a Schematic diagram of


geometry, b Boundary conditions
(z axial symmetric)

1 @ 2 p0
 r 2 p0
a20 @t2
@2 x
¼ fTij Hðf Þg  f½Pij nj þ qui ðun  vn Þdðf Þg
@xi @xj @xi
@
þ f½q0 vn þ qðun  vn Þdðf Þg
@t
ð3Þ

p′ is the sound pressure at the far field (p′ = p − p0). f = 0


denotes a mathematical surface introduced to embed the
exterior flow problem (f > 0) in an unbounded space, which
facilitates the use of generalized function theory and the
free-space Green function to obtain the solution. The surface
(f = 0) corresponds to the source (emission) surface, and can
be made coincident with a body (impermeable) surface or a
permeable surface off the body surface. is the unit normal Fig. 3 Variation of acoustic streaming velocity with fluid density at
vector pointing toward the exterior region (f > 0), a0 is the various amplitudes
far-field sound speed, and is the Lighthill stress tensor [15].
Figure 4 shows the acoustic streaming velocities with fluid
density at various ultrasonic resonant frequency. As shown in
Result and Discussion Fig. 4, the acoustic streaming velocity decreases with
increasing fluid density similar to Fig. 3, but the ultrasonic
Figure 3 shows the acoustic streaming velocities directly frequency did not significantly affect the flow velocities.
under the sonotrode obtained from the measurement by PIV The CFD modeling results are shown in Fig. 5, which
system with fluid density at various ultrasonic amplitude. shows the contour plots of velocity magnitude with ultra-
The acoustic streaming velocities at same ultrasonic injec- sonic injection conditions and fluid densities. As shown in
tion amplitude are inversely proportional to fluid density. Fig. 5a–c, the flow velocities directly under the sonotrode
For example, when the ultrasonic amplitude is 66 µm, the increased linearly with 0.28, 0.30, and 0.35 m/s as the
acoustic streaming velocity decreased with 0.47, 0.38, 0.32, ultrasonic amplitude increases with 38, 48, 66 lm. But as
0.30 m/s as the fluid density. The gradients of acoustic shown in Fig. 5d–f, the ultrasonic frequency did not sig-
streaming velocity reduction with increasing density are nificantly affect the flow velocities. The acoustic streaming
similar for each ultrasonic amplitude. velocities with increasing ultrasonic frequency as 15, 20,
In a fluid of the same density, acoustic streaming velocity 25 kHz are similar to 0.20 m/s. As shown in Fig. 5g–i, as
increases linearly with increasing ultrasonic amplitude. the density of the fluid increased at 1, 1,8, 2.4 g/cm3
1576 Y. K. Lee et al.

especially, the flow velocity decreased sharply at 0.35, 0.17,


and 0.13 m/s.
To predict the effective energy range and ultrasonic
treatment time of the fluid according to the ultrasonic
injection time, Dense Discrete Particle Model (DDPM)
simulation was performed. The content of particles added in
the fluid was determined by result of erosion quantification
experiment. Through the previous studies of ultrasonic de-
gassing in Al alloys as shown in Figs. 6 and 7, ultrasonic
treatment is completed when the DDPM concentration is
more than 80% in the area of 0.19 kg/m3. Figures 8 and 9
show the ultrasonic effective energy range of ultrasound and
completion of ultrasonic treatment time according to the
ultrasonic injection time. The effective energy range of the
ultrasound increased rapidly with the increase of the ultra-
sonic amplitude, however the change of the effective energy
Fig. 4 Variation of acoustic streaming velocity with fluid density at
various frequencies
range according to the ultrasonic frequency hardly occurred.

Fig. 5 Contour plots of velocity


magnitude; a 15 kHz 38 lm,
1 g/cm3, b 15 kHz 48 lm,
1 g/cm3, c 15 kHz 66 lm,
1 g/cm3, d 15 kHz 20 lm,
1 g/cm3, e 20 kHz 38 lm,
1 g/cm3, f 25 kHz 20 lm,
1 g/cm3, g 15 kHz 66 lm,
1 g/cm3, h 15 kHz 66 lm,
1.8 g/cm3 and i 15 kHz 66 lm,
2.4 g/cm3
Modeling of the Effect of Ultrasonic Frequency … 1577

Fig. 6 Cross sections of RPT


specimens and hydrogen
concentration of A356 Al alloys
with ultrasonic injection time

Fig. 7 Hydrogen concentration of A356 Al alloys with ultrasonic Fig. 8 Effective energy range of ultrasound with ultrasonic injection
injection time time at various amplitudes
1578 Y. K. Lee et al.

2. O.V. Abramov (1994) Ultrasound in Liquid and Solid Metals,


CRC Press, Boca Raton
3. G.I. Eskin, D.G. Eskin (2015) Ultrasonic Treatment of Light Alloy
Melts. CRC Press, New York
4. L. Nastac (2014) Numerical modeling of fluid flow and solidifi-
cation characteristics of ultrasonically processed A356 alloys.
ISIJ. INT. 54(8):1830–1835
5. M.C. Schenker, M.J.B.M. Pourquie, D.G. Eskin, B.J. Boersma
(2013) PIV quantification of the flow induced by an ultrasonic
horn and numerical modeling of the flow and related processing
times. Ultrason. Sonochem. 20:502–509
6. S. Komarerv, K. Oda, Y. Ishiwata, N. Dezhkunov (2013)
Characterization of Acoustic Cavitation in Water and Molten
Aluminum Alloy. Ufirason. Sonothem. 20(2):754–761
7. I. Tzanakis, G.S.B Lebon, D.G Eskin, K.A. Pericleous (2016)
Characterisation of the Ultrasonic Acoustic Spectrum and Pressure
Field in Aluminium Melts with Advanced Cavitometer. J. Mater.
Process. Technol, 229:582–586
8. I. Tzanakis, G.S.B Lebon, D.G Eskin, K.A. Pericleous (2016)
Investigation of the Factors Influencing Cavitation Intensity during
Fig. 9 Effective energy range of ultrasound with ultrasonic injection the Ultrasonic Treatment of Molten Aluminum. Mater. Des.
time at various frequency 90:979–983
9. H. Huang, D. Shu, Y. Fu, J. Wang, B. Sun (2014) Synchrotron
Radiation X-ray Imaging of Cavitaticn Bubbles in Al-Cu Alloy
Melt. Ultrason. Sonochem. 21(4):1275–1278
Conclusion 10. Y. Ishiwata et al. (2012) Investigation of acoustic streaming in
Aluminum melts exposed to high-intensity ultrasonic irradiation.
Paper presented at the 13th ICAA, Pittsburgh, Pennsylvania, 3–7
The acoustic streaming velocities were quantified with June 2012
variation of fluid density and ultrasonic amplitude and fre- 11. G. Wang, P. Croaker, M. Dargusch, D. McGuckin, D. StJohn
quency. Based on the quantified results of acoustic streaming (2017) Simulation of convective flow and thermal conditions
during ultrasonic treatment of an Al-2Cu alloy. Comput. Mater.
velocity, CFD modeling was performed to predict the Sci. 134:116–125
acoustic streaming in the fluid. Through CFD modeling, we 12. J. Kang, X. Zhang, S. Wang, J. Ma, T. Huang (2015) The
predicted the variation of the acoustic streaming velocity in comparison of ultrasonic effects in different metal melts Ultrason-
the fluid and the variation of effective range according to the ics 57:11–15
13. I. Senocak, W. Shyy (2002) J. Comput. Phys. 176(2):363–383
application time. Ultrasonic amplitude and density of fluid 14. B. Sajjadi, A.A.A. Raman, S. Ibrahim (2015) Influence of
have a great effect on acoustic streaming, but ultrasound ultrasound power on acoustic streaming and micro-bubbles
frequency has no effect. formations in a low frequency sono-reactor: Mathematical and
3D computational simulation Ultrason. Sonochem. 24:193–203
15. T. Kumaresan, A. Kumar, A.B. Pandit, J. B. Joshi (2007)
Modeling Flow Pattern Induced by Ultrasound: The Influence of
References Modeling Approach and Turbulence Models Ind. Eng. Chem. Res
46:2936–2950
16. ANSYS Inc (2013) ANSYS Fluent Theory Guide. SAS IP Inc,
1. J. Campbell (1981), Effect of Vibration During Solidification.
Canonsburg
Intern. Met. Rev. 26:71–108

Potrebbero piacerti anche