Take Off of Small Leidenfrost Droplets

We put in evidence the unexpected behavior of Leidenfrost droplets at the later stage of their evaporation. We predict and observe that, below a critical size $R_l$, the droplets spontaneously take off due to the breakdown of the lubrication regime. We establish the theoretical relation between the droplet radius and its elevation. We predict that the vapor layer thickness increases when the droplets become smaller. A satisfactory agreement is found between the model and the experimental results performed on droplets of water and of ethanol.

Figure 1

Experimental set-up. A cloud of submillimetric droplets is rained toward a silicon wafer inserted in a copper bloc kept at a controlled temperature $T_1$. The droplet of radius $R$ is at temperature $T_0$ and at a distance $h$ from the substrate.

Figure 2

Top: Pictures of a Leidenfrost droplet and its reflected image on the silicon wafer. Bottom: Height of the droplet as a function of time. After being dropped, it first bounces and lands on the substrate. The droplet then stands on the substrate in the lubrication regime. It finally takes off from the substrate.

Figure 3

Dimensionless height of the droplets as a function of their radii. The full line is a best fit to Eq. (6). Note that a minority of droplets fall back on the substrate. This is certainly due to dust contamination. The same data are represented in the inset but in log-log. The dotted line indicates the value of $R_l$ for water Leidenfrost droplets on a substrate kept at $T=400°\mathrm{C}$.

Figure 4

Dimensionless height of ethanol droplets as a function of $R$. The substrate is kept at $T=400°\mathrm{C}$. The same data are represented in the inset but in log-log. The full line corresponds to the scaling law predicted in Eq. (6).