Improving Binary Droplet Collision Model Prediction of the Bouncing Regime*

In this work, we experimentally investigate binary droplet collisions of identical droplets of 2%, 4%, and 8% of hydroxypropyl methylcellulose (HPMC) solutions in water. Monodisperse nozzles were used to generate the droplets, while the collisions events were captured using a high-speed camera. The collisions outcomes - namely bouncing, coalescence, reflexive separation, and stretching separation - were mapped in the parameter space of Weber number and the impact parameter for the three solutions. These studies were consistent with published studies and confirmed that existing models poorly predict the boundary of the bouncing regime at low impact parameter. We attribute this to modelling assumptions that are related to the considered kinetic energy and the surface area at the maximum deformation of the droplets. Therefore, a new model was developed by redefining the surface area at the moment of the maximum deformation of the droplets and making it a function of the impact parameter. The new model shows a striking improvement in the prediction of the bouncing boundary, where the main absolute error was reduced by a factor of 9 compared to the existing models.