Hostname: page-component-7b7567c8c-tnkh4 Total loading time: 0 Render date: 2026-11-16T02:03:58.671Z Has data issue: false hasContentIssue false

Divergent Richtmyer–Meshkov instability on a heavy gas layer

Published online by Cambridge University Press:  24 March 2023

Duo Zhang
Affiliation:
Advanced Propulsion Laboratory, Department of Modern Mechanics, University of Science and Technology of China, Hefei 230026, PR China
Juchun Ding*
Affiliation:
Advanced Propulsion Laboratory, Department of Modern Mechanics, University of Science and Technology of China, Hefei 230026, PR China
Ting Si
Affiliation:
Advanced Propulsion Laboratory, Department of Modern Mechanics, University of Science and Technology of China, Hefei 230026, PR China
Xisheng Luo*
Affiliation:
Advanced Propulsion Laboratory, Department of Modern Mechanics, University of Science and Technology of China, Hefei 230026, PR China
*
†Email addresses for correspondence: djc@ustc.edu.cn; xluo@ustc.edu.cn
†Email addresses for correspondence: djc@ustc.edu.cn; xluo@ustc.edu.cn

Abstract

Experiments on divergent Richtmyer–Meshkov (RM) instability at a heavy gas layer are performed in a specially designed shock tube. A novel soap-film technique is extended to generate gas layers with controllable thicknesses and shapes. An unperturbed gas layer is first examined and its two interfaces are found to move uniformly at the early stage and be decelerated later. A general one-dimensional theory applicable to an arbitrary-thickness layer is established, which gives a good prediction of the layer motion in divergent geometry. Then, six kinds of perturbed SF$_6$ layers with various thicknesses and shapes surrounded by air are examined. At the early stage, the amplitude growths of the inner interface for various-thickness layers collapse quite well and also can be predicted by the Bell model for cylindrical RM instability at a single interface, which indicates a negligible interface coupling effect. Later, a rarefaction wave accelerates the inner interface, causing a dramatic rise in the growth rate. It is found that a thicker gas layer will result in a larger extent that the rarefaction wave can promote the instability growth. A modified Bell model accounting for both Rayleigh–Taylor (RT) instability and interface stretching caused by a rarefaction wave is established, which well reproduces the quick instability growth. At late stages, reverberating waves inside the layer are negligibly weak such that the inner interface growth is dominated by RM instability and RT stability. The major factors driving the outer interface development are a compression wave and interface coupling. A new interface coupling phenomenon existing uniquely in divergent geometry caused by the gradual thinning of the gas layer is observed and also modelled.

Information

Type
JFM Papers
Copyright
© The Author(s), 2023. Published by Cambridge University Press
Figure 0

Figure 1.(a) Sketch of the curved part of the divergent shock tube, (b) an enlarged view of section II of the interface-formation device and (c) a drawing of the interface-formation device.

Figure 1

Figure 2.(a) Sequences of schlieren images showing the movements of the unperturbed gas layers of different thickness ($d_0$) after the impact of a cylindrically divergent shock and (b) sketches of the wave patterns and interface morphologies at typical moments for the $d_0 = 20$ mm layer. The lines left behind at the initial interface locations in panel (a), which are marked by white dot lines, indicate the protruding filaments. Here, II$_2$ is the initial outer interface; SI$_1$, the shocked inner interface; SI$_2$, the shocked outer interface; TS$_j$, $j$th transmitted shock; RS$_j$, $j$th reflected shock; RW$_1$, the rarefaction wave reflected from the outer interface; and CW$_1$, the compression wave reflected from the inner interface. The unit of numbers is $\mathrm {\mu }$s.

Figure 2

Figure 3.Trajectories of the inner and outer interfaces at (a) early and (b) late stages for unperturbed gas layers with different thicknesses ($d_0=10$ mm, 20 mm, 30 mm). A wave diagram is also included in panel (a). The solid line in panel (a) refers to the prediction of 1-D theory of Liang & Luo (2021a). ICS, incident cylindrical shock. The other symbols are the same as those in figure 2.

Figure 3

Table 1.Detailed parameters corresponding to the initial conditions for each case. Note: $a_{01}$ and $n$ are the initial amplitude and azimuthal mode number of the inner interface, respectively; $Ma$ refers to the Mach number of incident cylindrical shock; $d_0$ to the layer thickness; $\lambda$ to the wavelength of the inner interface; mfra(SF$_6$) to the mass fraction of SF$_6$ inside the layer; $A$ to the Atwood number; $A^+$ to the post-shock Atwood number; $\gamma$ to the specific heat ratio inside the layer and $c$ to the post-shock sound speed inside the layer.

Figure 4

Figure 4.Schlieren images showing the developments of wave patterns and interfacial morphologies for all cases. ICS denotes the incident cylindrical shock and II$_1$ is the initial first interface. The other symbols are the same as those in figure 2. The numbers are in $\mathrm {\mu }$s.

Figure 5

Figure 5.Sketches of the wave patterns and interface morphologies at typical moments for case 2. Here, $\Delta V_2$ ($\Delta V_2^*$) is the velocity of the outer interface after the impact of TS$_1$ (CW$_1$). The other symbols are the same as those in figure 4. The numbers are in $\mathrm {\mu }$s.

Figure 6

Figure 6.Normalized variations of the amplitude of the inner interface versus time for different cases. SI refers to the perturbation growth of a single air/SF$_6$ interface from Li et al. (2020). The dashed line refers to the prediction of Bell model, the dash–dotted line to the prediction of the Bell-RT model, and the solid line to the prediction of the Bell-RT-m model that considers the interface stretching and RT instability caused by RW$_1$.

Figure 7

Table 2.The post-shock parameters for six cases. Note: $t_{i}^{+}$ refers to the time just after the shock crosses the inner ($i=1$) or outer ($i=2$) interface; $a_{i}^{+}$ to the corresponding amplitude at $t_{i}^{+}$; $\Delta V_{i}$ to the post-shock velocity of the interface; $\dot {a}_{0}$ to the growth rate of the inner interface at $t_{i}^{+}$.

Figure 8

Figure 7.Sketches showing the interface stretching process by rarefaction wave RW$_1$.

Figure 9

Table 3.The relevant parameters for cases 1–6. Note: $a_{1}^{*}$ is the inner interface amplitude just before the arrival of RW$_1$; $d^{*}$ is the layer thickness at the time when TS$_1$ encounters the outer interface; $L$ is the width between the head and tail of RW$_1$ when it arrives at the inner interface; $\Delta V_1^*$ is the velocity of the inner interface after the impact of RW$_1$; $\Delta t_{1}$ ($\Delta t_{2}$) is the time of the first (second) stage for RW$_1$ interacting with the perturbed inner interface; $g$ is the acceleration of the inner interface caused by RW$_1$; and $\dot {a}_{s}$ is the average growth rate caused by interface stretching.

Figure 10

Figure 8.Variations of the amplitude of the outer interface versus time. The solid line refers to the model prediction considering the effect of CW$_1$ and interface coupling, the dashed line to model prediction considering interface coupling effect and the dash–dotted line to the prediction of (3.27).

Figure 11

Table 4.The relevant parameters for the interaction between CW$_1$ and the outer interface. Note: $a_2^*$ is the perturbation amplitude of the outer interface just before the arrival of CW$_1$; $\Delta t_{cw}$ is the duration for the perturbed CW$_1$ passing through the outer interface; $\beta$ is the coupling angle; and $\dot {a}_1^{cou}$ ($\dot {a}_2^{cou}$) represents the growth rate of the inner (outer) interface.

Figure 12

Figure 9.Variations of $c_1$ and $c_2$ with the layer thickness.

Figure 13

Figure 10.Schematic diagrams showing the gas layer deformation for unperturbed and perturbed cases.