Jumps, somersaults, and symmetry breaking in Leidenfrost drops

When a droplet of water impacts a heated surface, the drop may be observed to bounce. Recently is has been found that small quantities ($\sim 100$ ppm) of polymer additives such as polyethylene oxide can significantly increase the maximum bouncing height of drops. This effect has been explained in terms of the reduction of energy dissipation caused by polymer additives during the drop retraction and rebound, resulting in higher mechanical energy available for bouncing. Here we demonstrate, by comparing three types of fluids (Newtonian, shear-thinning, and viscoelastic), that the total kinetic energy carried by low-viscosity Newtonian drops during retraction is partly transformed into rotational kinetic energy rather than dissipated when compared with high-viscosity or non-Newtonian drops. We also show that non-Newtonian effects play little role in the energy distribution during drop impact, while the main effect is due to the symmetry break observed during the retraction of low-viscosity drops.

Figure 1

Flow curves of the model fluids measured with an MCR302 rheometer (Anton Paar) equipped with a cone-plate geometry (75 mm diameter; $2^{\circ }$ angle): $◯$, 100 ppm XG; $▲$, 80 ppm PAA, $\square$, 400 ppm XG, and $●$, 300 ppm PAA. Solid lines represent the Carreau-Yasuda model fit curves of the average values of measured XG and PAA data. Dashed lines and dot-dashed lines indicate the constant viscosities of GLY with infinite-shear viscosity and GLY with zero-shear viscosity, respectively.

Figure 2

Maximum bouncing height of the drops normalized with respect to the equilibrium drop diameter of model fluids as a function of the impact Weber number: (a) $◯$, 100 ppm XG; $◊$, 80 ppm PAA; $▲$, GLY with infinite-shear viscosity, and $■$, GLY with zero-shear viscosity. (b) $◯$, 400 ppm XG; $◊$, 300 ppm PAA; $▲$, GLY with infinite-shear viscosity, and $■$, GLY with zero-shear viscosity.

Figure 3

(a) Rebound morphology of Leidenfrost drops of different model fluids at $\text{We}\approx 70$. The time between two consecutive images is 5 ms. (b) Spreading and retracting morphology of Leidenfrost drops of different model fluids at $\text{We}\approx 70$. The time between two consecutive images is 1 ms.

Figure 4

Maximum bouncing height (modified data for GLY with infinite-shear viscosity) of the drops normalized with respect to the equilibrium drop diameter of model fluids as a function of the impact Weber number: (a) $◯$, 100 ppm XG; $◊$, 80 ppm PAA; $▲$, GLY with infinite-shear viscosity, and $■$, GLY with zero-shear viscosity. (b) $◯$, 400 ppm XG; $◊$, 300 ppm PAA; $▲$, GLY with infinite-shear viscosity, and $■$, GLY with zero-shear viscosity. Error bars represent the mean square value of the errors on the center of mass height and on the virtual lengths obtained from Eq. (4).