Thermally Assisted Heterogeneous Cavitation through Gas Supersaturation

We demonstrate that besides gaseous pockets also a gas supersaturated spot on a substrate can be a nucleus for cavitation. The supersaturation is achieved by either a formerly dissolved bubble or by heating locally the surface below the boiling temperature. The experiments are conducted in a thin film of water; one side of the water film is in contact with a gold coated substrate that is heated by a continuous laser through plasmonic heating. For nucleation of a bubble, the pressure at the heated spot is reduced by a transient rarefaction wave. The experimental findings suggest that the local gas supersaturation is responsible to nucleate cavitation and thus connects the phase transitions of cavitation and boiling. Additionally, the pressure waves in the liquid gap are studied numerically.

Figure 1

Experimental setup for cavitation inception by a focused pulsed laser in a water-filled gap. The latter consists of a gold-coated microscope slide and a cover slip. Between both glass plates a $4–10\mu \mathrm{m}$ thick water layer is sandwiched. The pulsed laser is focused onto the gold layer, creating a plasma and transverse Lamb-type wave. As it pushes the glass plates apart, the pressure in its wake is positive ($+p$) and in its front negative ($-p$), nucleating secondary cavitation bubbles. The visual wave is caused by the second spatial derivative of the pressure gradients. A second laser (CW) used for local heating is focused into the gap at a slightly different position compared to the pulsed laser. The setup allows a simultaneous generation of cavitation bubbles and observation with a high-speed camera. Illumination is provided from the opposite side with a femtosecond laser. Two different cases are studied to act as cavitation nucleus: (1) a vapor bubble, created by localized boiling with sufficient time to dissolve, and (2) a locally heated spot during the passage of a rarefaction wave. For comparison a preexisting vapor bubble is studied as well [13].

Figure 2

Nucleation of a bubble by a rarefaction wave in a $\approx 4\mu \mathrm{m}$ gap. The lower right shows part of the primary cavitation bubble. (a) A vapor bubble is created and has dissolved at $t=-4\mathrm{ms}$ at the blue cross. Upon the passage of the tension wave, a cavitation bubble is nucleated ($t=60\mathrm{ns}$) at the location where the vapor bubble dissolved. $t=0$ corresponds to the time when the laser pulse reaches the glass slide. (b) A spot is heated (blue cross) below the saturation temperature. The heated spot acts as a cavitation nucleus once the rarefaction wave passes.

Figure 3

Nucleation of a cavitation bubble from a dissolved bubble by a rarefaction wave (a) in a $\sim 10\mu \mathrm{m}$ gap, (b),(c) in a $\sim 4\mu \mathrm{m}$ gap. A CW laser is focused for $20–40\mu \mathrm{s}$ at the position with the blue cross. It is switched off 1 ms before the high-speed recording has started. (c) Detail image of (b) with higher temporal resolution. The red circles (diameter: $25\mu \mathrm{m}$) mark the region which is void of random cavitation.

Figure 4

Numerical simulations (a) in the $10\mu \mathrm{m}$ gap [13]. The bubble boundary is indicated with a white line. ${\sigma }_{xy}$ is component of the stress tensor, i.e., the shear stress corresponding to a shear wave with vertical deflection ($y$ direction) and horizontal propagation direction ($x$ direction). (b) shows a magnified view where the amplitude of deformation is multiplied by a factor of 200. (c) Position of bubble boundary (radius) and the Lamb wave over time with experimental values. (d) Pressure over time at a distance of $110\mu \mathrm{m}$ from the bubble center. The duration of the tension wave is estimated from the time between minimum and maximum pressure, which are $\approx 41\mathrm{ns}$ in the thin gap and 29 ns in the thick gap.

Figure 5

(a) Temperature in the water on the axis of rotation above the center of the heated spot at a height of $0.1\mu \mathrm{m}$ (red), $0.5\mu \mathrm{m}$ (magenta), and $1\mu \mathrm{m}$ (blue) above the gold layer. The green line shows the assumed heating power. The inset represents the simulation setup. (b) Temperature at the end of heating ($t^{*}=10\mu \mathrm{s}$) and (c) $4\mu \mathrm{s}$ after the heating has stopped. The dashed line in (b) and (c) represents the position of the gold layer, water is simulated above and glass below. The orange and red lines represent the temperature isoclines of 50 and 100 °C, respectively.