As the liquid pressure drops below the vapor pressure, liquid-to-gas phase change occurs, and vapor filled cavities emerge within the liquid phase. Once the vapor-filled cavities are subjected to higher liquid pressures again, they collapse violently and generate shockwaves and local high temperature (>5000 K) and pressure (>500 bar) values, which in turn results in erosion on the nearby solid surfaces and harm turbomachinery. Due to this harmful nature of cavitation, it has been historically seen as an unwanted phenomenon in many industries. Although these concerns still hold today, researchers showed that the intense conditions created during the collapse of cavitation bubbles can also be used to intensify several processes, including wastewater treatment, hydrogen production through electrolysis, and biofuel production, by employing cavitation assisted reactor technologies. Modelling such devices is proven to be computationally challenging due to their complex geometries and significantly different time and length scales encountered in a cavitating flow. Thus, the current state-of-the-art in modelling of cavitating flows consists of studies that emphasize either the dynamics of a single bubble under cavitation conditions or the accurate representation of the geometry and the flow field, while over-simplifying the other. The existing Eulerian cavitation closures in CFD software follows the latter approach by neglecting several key characteristics of the phenomenon, such as the impact of shockwaves or the bubble chemistry following the dissociation of water molecules. Therefore, although these models are computationally inexpensive, they fail to meet the needs of the modelling efforts for the emerging cavitation-assisted reactor technologies, where especially the collapse conditions and the resulting bubble chemistry plays a significant role.
This work will start by discussing the drawbacks of the current CFD methods for modelling cavitation when applied to cavitation-intensified reactors. Next, results from a recent study solving single bubble dynamics equations on Lagrangian cavitation bubbles tracked on a Eulerian flow field are discussed. Such an approach allows cavitation collapse conditions to be predicted while considering the detailed flow behavior of the cavitation reactor. However, several challenges were encountered with this approach, especially when it comes to coupling the effect of the bubble volume change to the Eulerian pressure field. In addition, results from a recent study focusing on single bubble dynamics are shown to illustrate how simple algebraic expressions can be obtained for predicting the reactive species formed under bubble collapse. These expressions could be introduced into CFD models as closures to include the outcome of cavitation bubble collapses without resolving the short timescales of the collapses, greatly reducing the computational cost compared to the aforementioned Lagrangian approach. In conclusion, possibilities are discussed for including cavitation collapse effects in Eulerian cavitation models to facilitate improved usage of CFD models to design cavitation-intensified reactors.