Hostname: page-component-7b7567c8c-2n9c9 Total loading time: 0 Render date: 2026-11-17T01:09:02.563Z Has data issue: false hasContentIssue false

Strong-shock-driven Richtmyer–Meshkov instability at quasi-single-mode interfaces

Published online by Cambridge University Press:  08 May 2026

Wei Cai
Affiliation:
State Key Laboratory of High-Temperature Gas Dynamics, School of Engineering Sciences, University of Science and Technology of China, Hefei 230026, PR China
He Wang*
Affiliation:
State Key Laboratory of High-Temperature Gas Dynamics, School of Engineering Sciences, University of Science and Technology of China, Hefei 230026, PR China
Ting Si
Affiliation:
State Key Laboratory of High-Temperature Gas Dynamics, School of Engineering Sciences, University of Science and Technology of China, Hefei 230026, PR China
*
Corresponding author: He Wang, ustchewang@ustc.edu.cn

Abstract

Richtmyer–Meshkov instability (RMI) at quasi-single-mode interfaces subjected to strong shocks with Mach number exceeding 3.0 is investigated through shock-tube experiments. To reveal the influence of higher-order initial modes, one single-mode and four quasi-single-mode interfaces with varying modal compositions are examined. The Richtmyer theory is experimentally verified for the first time to accurately capture the single-mode interface evolution from start-up to the linear stage. In contrast, it fails for quasi-single-mode scenarios, highlighting the significant effect of higher-order modes. By incorporating the influence of higher-order modes via linear superposition, a linear model for quasi-single-mode interface evolution is proposed, whose validation indicates negligible modal coupling in early evolution. For the nonlinear period, a representative model for weak-shock-driven RMI at a single-mode interface performs poorly, as it does not consider the influence of higher-order modes and key mechanisms such as shock proximity, secondary compression and spike acceleration. Based on the present findings, an empirical nonlinear model with favourable predictive capability is developed. Furthermore, by matching the new linear and nonlinear models, a complete description of the amplitude evolution of single-mode and quasi-single-mode interfaces from start-up to nonlinear stages is achieved, offering new insight into modelling strong-shock-driven RMI.

Information

Type
JFM Papers
Copyright
© The Author(s), 2026. Published by Cambridge University Press
Figure 0

Figure 1.Sketch of the shock-tube facility capable of generating a planar shock with $M_s \gt 3.0$ while maintaining a ‘clean’ flow field.

Figure 1

Figure 2.Schematics of (a) interface formation method and (b) high-speed shadowgraph imaging system.

Figure 2

Figure 3.Schematics of initial interface configurations, where $a^-$ is initial amplitude, and $\lambda$ is perturbation wavelength.

Figure 3

Table 1.Coefficients of initial modal components for single-mode and quasi-single-mode interfaces.

Figure 4

Table 2.Significant parameters for five experiments with various initial interface configurations: $ \textit{MF}$ is the mass fraction of $\mathrm{SF_6}$ downstream of the initial interface; $A = (\rho _2 - \rho _1)/(\rho _2 + \rho _1)$ is the post-shock Atwood number, with $\rho _1$ and $\rho _2$ denoting the densities of gases upstream and downstream of the post-shock interface, respectively; $T$ is the ambient temperature; $M_s$ is the Mach number of the incident shock; $v_t$ is the velocity of the transmitted shock; $\Delta u$ is the interface velocity jump; $\dot {a}_{l}^{e}$ and $\dot {a}_{l}^{\textit{imp}}$ are the linear amplitude growth rates obtained from experiments and predicted by the impulsive model, respectively. Units of parameters representing velocity (or amplitude growth rate) and temperature are m s−1 and K, respectively.

Figure 5

Figure 4.Experimental shadowgraphs of the evolution of single-mode and quasi-single-mode interfaces accelerated by strong shock waves: (a) single-mode interface; (b) circular-arc interface; (c) V-shaped interface; (d) trapezoidal interface; (e) square-waveform interface. Here, $a$ is interface amplitude, and numbers represent the time in $\unicode {x03BC}\mathrm{s}$.

Figure 6

Figure 5.Sketches of the shock waves and interface morphology at the early and late stages for trapezoidal and square-waveform cases, along with the perturbed velocity field in the region behind the transmitted shock, obtained by subtracting the background one-dimensional velocity in the area between transmitted and reflected shocks from the particle velocity field. The colour of the arrow represents the magnitude of the velocity: the darker the colour, the greater the velocity. Red, blue and black lines denote the shocked interface, transmitted shock and transverse wave, respectively.

Figure 7

Figure 6.Dimensionless amplitude evolution of single-mode and quasi-single-mode interfaces from the start-up to linear stages. Symbols indicate experimental data; solid and dashed lines indicate predictions from Richtmyer theory and the mR model, respectively, where the mR model denotes the model obtained by extending the Richtmyer theory through linear superposition. Note that the predictions of the Richtmyer theory and mR model coincide for the single-mode case.

Figure 8

Figure 7.Temporal amplitude variations obtained from experiments (symbols) and predicted by the ZG model (dashed lines): (a) overall interface; (b) bubble and spike.

Figure 9

Figure 8.Temporal amplitude variations obtained from experiments (symbols) and predicted by the $\varepsilon$-ZG model (dash-dotted line): (a) overall interface; (b) bubble and spike.

Figure 10

Figure 9.Comparison of the experimental measurements and matched model predictions for the complete amplitude evolution from the early-time start-up to the late-time nonlinear stage.

Figure 11

Figure 10.Comparison of the experimental measurements and matched model predictions for the amplitude evolution of (a) the bubble and (b) the spike from the early-time start-up to the late-time nonlinear stage.

Figure 12

Table 3.Key parameters of the repeated runs for each interface configuration: $ \textit{MF}$ is the mass fraction of $\mathrm{SF_6}$ downstream of the initial interface; $A = (\rho _2 - \rho _1)/(\rho _2 + \rho _1)$ is the post-shock Atwood number, with $\rho _1$ and $\rho _2$ denoting the densities of gases upstream and downstream of the post-shock interface, respectively; $T$ is the ambient temperature; $M_s$ is the Mach number of the incident shock; $v_t$ is the velocity of the transmitted shock; $\Delta u$ is the interface velocity jump; $\dot {a}_{l}^{e}$ is the experimental linear amplitude growth rate. Units of parameters representing velocity (or amplitude growth rate) and temperature are m s−1 and K, respectively.

Figure 13

Figure 11.Image preprocessing procedures for Fourier analysis, illustrated with an example frame from the V-shaped case at $t=103$$\unicode {x03BC}$s: (a) original experimental image spanning one perturbation wavelength; (b) binarised image; (c) interface contour in physical coordinates.