About Phenomenology and Modeling of Dropwise Condensation

International audienceThe modeling of dropwise condensation remains, at the present time, difficult. Indeed, in order to predict the heat transfers in this regime, it is necessary to know the heat flux which crosses each drop according to its size as well as the dropsize distribution on the surface considering radii that can vary from a few nanometers to several centimeters. As this distribution is a function of the life cycle of each drop, an overview of a drop's lifecycle has been first done in order to better understand the phenomena underlying dropwise condensation. Particular attention has been paid on the drop growth rate modeling. Secondly, a description of the drop-size distribution models has been done. Due to the very large number of drops, very fast dynamics and the difference in drop-sizes, only two types of modeling are available. The first approach is based on a semi-empirical law to model the distribution of the largest drops (i.e., those with a radius greater than a few microns) together with a population balance for the size distribution of the smallest drops. The second approach consists in the following of all the drops along time in order to determine the stationary dropsize distribution. This approach already succeed to predict the size distribution of the big drops many times. Based of these overviews, an individual-based modeling has been developed and computed, focusing on the behavior of the smallest droplets. A comparison of the results obtained with this model with the ones obtained with classical population-balanced approach has then be realized. Discrepancies of several orders of magnitude have been found on drop-size distribution. This important difference is attributed to one of the hypothesis of population balance modeling, i.e., the hypothesis of constant renewal rate whatever the drop radius. The impact of such deviations in the drop-size distribution on global heat transfer has then be quantified. In most of the configurations studied, the population balance approach predicts global heat fluxes about 30% higher compared to the individual-based model's ones. Finally, a parametric study has been done considering three parameters that can be potentially controlled in experimental works: advancing contact angle, nucleation sites density, and departure radius