Effect of stand-off distance and spatial resolution on the pressure impact of near-wall vapor bubble collapses

Theresa Trummler, Steffen J. Schmidt, Nikolaus A. AdamsInternational Journal of Multiphase Flow 2021 · DOI · arXiv

Abstract

The destructive potential of cavitation bubbles near solid boundaries is a critical concern in hydraulic engineering and medical applications. This study utilizes high-fidelity 3D compressible Navier-Stokes simulations to investigate the influence of stand-off distance ($S/R_0$) and grid resolution on the pressure impacts generated during bubble collapse. We identify a fundamental shift in the peak pressure mechanism: wall-attached bubbles are dominated by vapor torus collapse, while detached bubbles are primarily impacted by micro-jet formation. Furthermore, a rigorous grid convergence study reveals that while macroscopic bubble dynamics converge at moderate resolutions, the accurate prediction of peak wall pressures requires extremely fine grids (up to 320 cells per bubble radius), highlighting a significant challenge for quantitative erosion modeling.

Introduction

Cavitation—the rapid formation and collapse of vapor bubbles in a liquid—is a phenomenon with significant implications across various fields. In engineering, it causes severe erosion on ship propellers and fuel injectors. Conversely, in medicine, it is harnessed for therapeutic purposes such as lithotripsy. The "damage potential" of a collapsing bubble is highly sensitive to its proximity to a boundary, typically quantified by the stand-off distance $S/R_0$, where $S$ is the distance from the wall to the bubble center and $R_0$ is the initial radius.

Schematic of bubble configurations at different stand-off distances ($S/R_0$).

As shown in Figure 1, bubbles can be partially "buried" ($S/R_0 < 0$), attached ($0 < S/R_0 < 1$), or detached ($S/R_0 > 1$). Understanding how these configurations dictate the collapse mechanism—specifically the roles of micro-jet impact versus shock wave emission—is essential for predicting material damage.

Methodology

The researchers employed a fully compressible 3D Navier-Stokes solver to simulate vapor bubble collapses under a high driving pressure of 100 bar. The numerical framework utilized a thermodynamic equilibrium model (homogeneous mixture) to handle phase changes without explicit interface tracking. The liquid phase was modeled using a modified Tait equation of state (EOS), while a logarithmic EOS was applied to the two-phase mixture.

A key focus of the study was the effect of spatial resolution. Simulations were conducted across six grid densities, ranging from 20 to 320 cells per initial bubble radius ($c/R_0$). The highest resolution case involved approximately 670 million cells, representing one of the most computationally intensive 3D cavitation studies to date.

Results and Discussion

Mechanism Shift with Stand-off Distance

The simulations reveal that the primary source of peak wall pressure shifts as the bubble moves away from the surface. For attached bubbles ($S/R_0 \leq 0.85$), the collapse of the residual vapor torus generates the maximum pressure impact. In contrast, for detached bubbles ($S/R_0 > 1$), the high-speed micro-jet piercing the bubble and impacting the distal side becomes the dominant mechanism.

Temporal evolution of bubble collapse (a), pressure-time history (b), volume change (c), and peak pressure distribution (d).

As illustrated in Figure 6, the collapse process involves a rapid reduction in volume (c) followed by the emission of a high-pressure wave (b). For detached bubbles, micro-jets reach velocities 10–15 times the characteristic velocity, consistent with "water hammer" pressure models. The highest wall pressure occurs when the bubble is perfectly attached ($S/R_0 = 0$), reaching values over 2600 times the characteristic pressure ($p^*$).

The Resolution Trap

A critical finding of this study is the extreme sensitivity of peak pressure to grid resolution. While "main features"—such as the general bubble shape and collapse timing—converge at relatively coarse grids, peak wall pressures do not. For wall-attached bubbles, the predicted peak pressure increases almost inversely with cell size. This lack of convergence implies that quantitative erosion predictions are highly dependent on the chosen grid, necessitating resolutions of at least 160–320 $c/R_0$ for reliable results.

Conclusion

This study clarifies the dual nature of cavitation impact mechanisms based on stand-off distance and underscores the computational challenges in predicting peak pressures. The transition from torus-driven to jet-driven impacts provides a clearer framework for understanding cavitation erosion. Future work will extend these findings to lower driving pressures and incorporate surface tension effects to better align with atmospheric laboratory conditions and micro-bubble dynamics.

Summary updated

BibTeX
@article{trummler2021effect,
  author = {Theresa Trummler and Steffen J. Schmidt and Nikolaus A. Adams},
  title = {{Effect of stand-off distance and spatial resolution on the pressure impact of near-wall vapor bubble collapses}},
  journal = {International Journal of Multiphase Flow},
  year = {2021},
  doi = {10.1016/j.ijmultiphaseflow.2021.103618},
}
Figure