front page - tuprints - Technische Universität Darmstadt

In the experiments with sessile liquid film (Shibkov et al. [135, 137]), or sessile
supercooled drop (Bauerecker et al. [12] and Jung et al. [64]), the the nucleation
site was singular, and a clear propagating front of the ice dendrite was identified,
as shown in Figure 2.23 in Chapter 2. In the case of a drop impact, the ice dendrite
was unrecognizable, and the nucleation took place uniformly across the entire drop
surface. At the measured 3 K supercooling, the ice dendrite growth has a speed of
2mm/s (read from Figure 2.22 in Chapter 2). The first stage of solidification would
take 650 ms to finish the φ1.5 mm drop. However in the case of drop impact,
the duration was significantly shorter. At the achieved low supercooling of 3 K to
5 K, the nucleation process is known to be primarily governed by thermal diffusion
[72, 73], and the growth rate of the ice dendrite must be the same for the
both cases of with and without liquid flow. The difference between the two cases
is the initial nucleation site, i.e. when, where and how much they formed. In the
presence of strong liquid flow, numerous initial nucleation sites formed simultaneously.
The ice dendrite initiated from each singular nucleation site met each other
and stopped their growth, forming an overall uniform nucleation process inside
the drop. Furthermore, the liquid flow promoted the contact of the individual ice
dendrites, as well as the contact of the nucleation sites with the supercooled liquid.
Consequently, the duration of the first stage of solidification was much shorter than
that in a sessile liquid.
Multiple recordings as Figure 4.7 were acquired. The duration of the nucleation
varied greatly, from 4.17 ms to 18.07 ms. Since the measured temperature of the
impinging drop varied merely from −2 ◦ C to −3 ◦ C, corresponding to 1mm/s to
2mm/s growth rate of the ice dendrite, this variation was caused by the random
formation of the initial nucleation sites. This randomness in the experiment might
be attributed to the spatial heterogeneity of wettability of the SHS, which influenced
the dynamic motion of the drop, and further influenced the formation of the
nucleation sites. The wettability of the SHS was not perfectly reproducible.
Occasionally, nucleation occurred right upon impact. Figure 4.8 illustrates a
typical total rebounding of the supercooled drop on a SHS at 5 ◦ C, and the nonspherical
shape of the drop head emerged in the first image after impact. The drop
head recovered the spherical shape at the 0.45 ms, suggesting that the ice structure
was so fragile that the deformation of the liquid could easily break it. The
same effect can be recognized in Figure 4.6, where drop easily deformed despite
the formation of the ice/water structure. This is not surprising because only a tiny
fraction of the drop solidified during the heterogeneous nucleation at the achieved
low supercooling.
4.2. Impact of Supercooled Drop on Superhydrophobic Surfaces 109