Plasma–liquid interactions represent a growing interdisciplinary area of research involving plasma science, fluid dynamics, heat and mass transfer, photolysis, multiphase chemistry and aerosol science. This review provides an assessment of the state-of-the-art of this multidisciplinary area and identifies the key research challenges. The developments in diagnostics, modeling and further extensions of cross section and reaction rate databases that are necessary to address these challenges are discussed. The review focusses on non-equilibrium plasmas.

A model of single-bubble sonoluminescence (SBSL) is constructed. In the model, the temperature is assumed to be spatially uniform inside the bubble except at the thermal boundary layer near the bubble wall even at the strong collapse based on the theoretical results of Kwak and Na [Phys. Rev. Lett. 77, 4454 (1996)]. In the model, the effect of the kinetic energy of gases inside the bubble is taken into account, which heats up the whole bubble when gases stop their motions at the end of the strong collapse. In the model, a bubble in water containing air is assumed to consist mainly of argon based on the hypothesis of Lohse et al. [Phys. Rev. Lett. 78, 1359 (1997)]. Numerical calculations under a SBSL condition reveal that the kinetic energy of gases heats up the whole bubble considerably. It is also clarified that vapor molecules $\left({\mathrm{H}}_2\mathrm{O}\right)$ undergo chemical reactions in the heated interior of the bubble at the collapse and that chemical reactions decrease the temperature inside the bubble considerably. It is suggested that SBSL originates in thermal radiation from the whole bubble rather than a local point (the bubble center) heated by a converging spherical shock wave widely suggested in the previous theories of SBSL.

There are two main mysteries in bulk nanobubbles which are cavitation nuclei. One is the mechanism of stability of a bulk nanobubble. The other is the problem whether OH radicals are produced from bulk nanobubbles without a dynamic stimulus. For the former problem, several proposed models are briefly reviewed. The dynamic equilibrium model is discussed in details that a bulk nanobubble is stabilized by a partial coverage of the bubble surface by a hydrophobic material. The TEM images of bulk nanobubbles seem to support the dynamic equilibrium model. For the latter problem, numerical simulations of dissolution of an air nanobubble are reviewed, which suggest that no OH radical is produced from a dissolving nanobubble. A possible role of H2O2 generated during bulk nanobubble production using hydrodynamic cavitation is briefly discussed in relation to the experimental results of "OH radical" detection.

Computer simulations of bubble oscillations are performed under conditions of multibubble sonoluminescence (MBSL) in water for various ultrasonic frequencies. The range of the ambient bubble radius for sonoluminescing bubbles narrows as the ultrasonic frequency increases; at 20 kHz it is 0.1–100 μm while at 1 MHz it is 0.1–3 μm. At 1 MHz, any sonoluminescing bubble disintegrates into a mass of smaller bubbles in a few or a few tens of acoustic cycles, while at 20 kHz and 140 kHz some sonoluminescing bubbles are shape stable. The mechanism of the light emission also depends on the ultrasonic frequency. As the ultrasonic frequency increases, the amount of water vapor trapped inside bubbles at the collapse decreases. As a result, MBSL originates mainly in plasma emissions at 1 MHz while it originates in chemiluminescence of OH radicals and plasma emissions at 20 kHz.