Hostname: page-component-7b7567c8c-2n9c9 Total loading time: 0 Render date: 2026-11-16T13:11:10.457Z Has data issue: false hasContentIssue false

The electrostatically forced Faraday instability: theory and experiments

Published online by Cambridge University Press:  14 January 2019

Kevin Ward*
Affiliation:
University of Florida, Department of Chemical Engineering, Gainesville, FL 32611, USAUniversity of Lille, IEMN CNRS 8520, 59655, Lille, France
Satoshi Matsumoto
Affiliation:
Human Spaceflight Technology Directorate, Japan Aerospace Exploration Agency, Tsukuba, Ibaraki 305-8505, Japan
Ranga Narayanan*
Affiliation:
University of Florida, Department of Chemical Engineering, Gainesville, FL 32611, USA
*
†Email addresses for correspondence: klward3@ufl.edu, ranga@ufl.edu
†Email addresses for correspondence: klward3@ufl.edu, ranga@ufl.edu

Abstract

The onset of interfacial instability in two-fluid systems using a viscous, leaky dielectric model is studied. The instability arises as a result of resonance between the parametric frequency of an imposed electric field and the system’s natural frequency. In addition to a rigorous model that uses Floquet instability analysis, where both viscous and charge effects are considered, this study also provides convincing validating experiments. In other results, it is shown that (a) the imposition of a periodic electrostatic potential acts to counter gravity and this countering effect becomes more effective if a DC voltage is also added, (b) a critical DC voltage exists at which the interface becomes unstable such that no parametric frequency is required to completely destabilize the interface and (c) the leaky dielectric model approaches a model for a perfect dielectric/perfect conductor pair as the conductivity ratio becomes large. It is also shown via experiments that parametric resonant instability using electrostatic forcing may be reliably used to estimate interfacial tension to sufficient accuracy.

Information

Type
JFM Papers
Copyright
© 2019 Cambridge University Press 
Figure 0

Figure 1.Schematic of two-fluid electrostatically oscillated system. The normal vector points into the upper fluid, and the interface is taken to be initially flat and located at $z=0$. The voltage on the bottom electrode is oscillated with amplitude $A$ and frequency $\unicode[STIX]{x1D714}$, and the top electrode is held at a constant voltage $D$.

Figure 1

Figure 2.Scaled amplitude versus wavenumber for $D=5000$ V calculated from the inviscid perfect conductor model. Alternating subharmonic and harmonic tongues are obtained when $D\neq 0$. Fluid parameters are $\unicode[STIX]{x1D70C}_{1}=1000~\text{kg}~\text{m}^{-3}$, $\unicode[STIX]{x1D70C}_{2}=1.225~\text{kg}~\text{m}^{-3}$, $h_{1}=h_{2}=6$ mm, $\unicode[STIX]{x1D716}_{2}=1.00059$, $\unicode[STIX]{x1D6FE}=0.07275~\text{kg}~\text{s}^{-2}$ and $\unicode[STIX]{x1D714}=25$ Hz.

Figure 2

Figure 3.Scaled $A$ versus $k$ for both (a) $D=5000$ V and (b) $D=0$. Fluid parameters are $\unicode[STIX]{x1D707}_{1}=0.001~\text{kg}~\text{ms}^{-1}$, $\unicode[STIX]{x1D707}_{2}=1.983\times 10^{-5}~\text{kg}~\text{ms}^{-1}$ and other parameters equivalent to those in figure 2. Alternating subharmonic and harmonic tongues are seen when $D\neq 0$, while only harmonic tongues are observable when $D=0$. (c) Reproduction of figure 3(a) of Bandopadhyay & Hardt (2017). The current model is able to recover the previously reported results in addition to the first tongue, not reported earlier. Fluid parameters and $Ma_{i}$ are those given and defined by Bandopadhyay & Hardt (2017).

Figure 3

Figure 4.(a) Effect of $D$ on the first harmonic tongue. The tongue shifts down in critical amplitude with increasing $D$, and first reaches a critical amplitude of 0 for $D\approx 14.48$ kV at $k=301~\text{m}^{-1}$ for the fluid parameters given in figure 3(a,b). (b) Effect of frequency on the first harmonic tongue. As frequency increases, the tongue approaches the pure DC electrostatic Rayleigh–Taylor instability solution. $D$ is held at 5 kV for all tested frequencies.

Figure 4

Figure 5.Effect of varying $D$ on stability thresholds for both subharmonic (a) and harmonic (b) responses. Fluid properties are the same as those shown in figure 3(a). Increasing $D$ has the general effect of widening the instability tongues and causing a shift to higher wavenumber and lower threshold amplitude. The first harmonic tongues discussed previously are omitted for clarity.

Figure 5

Figure 6.Effect of $\unicode[STIX]{x1D70E}_{1}$ on $q_{0}$. An exponential dependence is observed for low values of $\unicode[STIX]{x1D70E}_{1}$, but upon reaching $\unicode[STIX]{x1D70E}_{1}=10^{-8}$, $q_{0}$ saturates and is no longer strongly affected by increasing $\unicode[STIX]{x1D70E}_{1}$. Fluid parameters are the same as shown previously, assuming $\unicode[STIX]{x1D70E}_{2}=3\times 10^{-15}~\text{S}~\text{m}^{-1}$ and $\unicode[STIX]{x1D716}_{1}=\unicode[STIX]{x1D716}_{2}=1.00059$. $A=D=5~\text{kV}$ and $\unicode[STIX]{x1D714}=25~\text{Hz}$.

Figure 6

Figure 7.Effect of $\unicode[STIX]{x1D70E}_{1}$ on the base state voltage profiles in the centre of the top fluid (a), at the interface (b) and in the centre of the bottom fluid (c). It is clearly seen that as $\unicode[STIX]{x1D70E}_{1}$ increases, the voltage profiles approach those of the perfect conductor model. Fluid properties and $\unicode[STIX]{x1D714}$ are equivalent to those in figure 6 and $D=0$. (d) The same properties as (c), with the exception that $\unicode[STIX]{x1D716}_{1}=10$. The base state voltage profiles saturate much more quickly with decreasing $\unicode[STIX]{x1D70E}_{1}$ if $\unicode[STIX]{x1D716}_{1}\gg \unicode[STIX]{x1D716}_{2}$.

Figure 7

Figure 8.Stability thresholds for the leaky dielectric model assuming various conductivities for the bottom fluid. It is seen that a high conductivity will yield similar results to the perfect conductor model, but lowering the conductivity quickly causes a departure from this ideality. $\unicode[STIX]{x1D714}=25$ Hz and the fluid properties are the same as those given in figure 3(b), along with $\unicode[STIX]{x1D70E}_{2}=3\times 10^{-15}~\text{S}~\text{m}^{-1}$ and (a) $\unicode[STIX]{x1D716}_{1}=2$ or (b) $\unicode[STIX]{x1D716}_{1}=50$.

Figure 8

Figure 9.Relationship between critical amplitude and forcing frequency for electrostatically (solid) and mechanically (dashed) forced systems. The fluid parameters are given in table 1, with heights of 2.5 cm each. The imposed amplitude is measured in kV for the electrostatic case and cm for the mechanical case. The graph is presented unscaled because of the difference in the physical nature of the amplitudes.

Figure 9

Table 1.Summary of fluid properties for the demonstration of critical amplitude dependence on frequency shown in figure 9. The electrical properties play no role in the mechanically forced calculation.

Figure 10

Figure 10.Assembled testing cell and experimental set-up. The geometry of choice is sealed between two conductive plates and secured with four non-conductive screws. The top and bottom plates are wired to high voltage amplifiers, and the voltages are controlled using independent signal generators. Backlighting allows for the clear visualization of the interface and any deflection using a high-speed camera.

Figure 11

Table 2.Summary of the testing geometries and operating conditions. Two tests were conducted using a cylindrical geometry without a constant voltage offset (A) and with a constant voltage offset (B).

Figure 12

Figure 11.Comparison between theoretical and experimental thresholds for Test A. The model is able to capture the experimental threshold accurately over a large frequency range. Threshold amplitudes were measured to the nearest 0.05 kV.

Figure 13

Figure 12.Comparison between theoretical and experimental thresholds for Test B. The model is able to predict the instability threshold excellently at high frequency, but interfacial disturbances arising from the DC offset cause discrepancy at lower frequencies. Threshold amplitudes were measured to the nearest 0.05 kV.

Figure 14

Figure 13.Faraday waves obtained during Test A. The frequency and amplitude of forcing are 3.5 Hz and 2.5 kV, respectively. The photograph is taken during the portion of the response cycle with the largest interfacial deflection, and the applied amplitude exceeds the critical threshold.

Figure 15

Figure 14.Interfacial tension measurement for Si–oil and water system using electrostatic Faraday instability. The solid line represents the nominal interfacial tension, and clearly provides the best fit for the experimental data ($R^{2}=0.978$). Testing properties are those given for figure 11.

Ward et al. supplementary movie

Example video of an electrostatically forced Faraday instability within an infinite geometry. The system is that described by Test A in Table 2, forced at a frequency of 4.75 Hz and an amplitude of 5 kV.

Download Ward et al. supplementary movie(Video)
Video 5.7 MB