Evaluation of water repellency of fabrics using droplet impact dynamics and machine learning

The principles of droplet impact are the foundation of the design of water-repellent solid surfaces. Although it has been known that textiles must also possess water repellency to resist the impacting raindrops during outdoor use, their water-repellent behavior based on droplet impact dynamics is still not well-understood. Here we investigated the dynamic behavior of droplet on rough woven fabrics coated with hydrophobic formulations through droplet impact test. We have also successfully integrated machine learning in this work to predict the spreading and rebound behavior of droplet, as well as to describe its overall dynamics. It was found that the droplet spreading characteristics do not provide any information on water repellency of the woven fabrics, even though the observed rebound characteristics show a wide range of water-repellent behavior among these fabrics. In addition, though high water repellency can be achieved by using fabrics with high roughness, fabric roughness is not the only factor to consider: the precise texture of the fabric surface plays a crucial role too. Our analysis revealed that the dynamic behavior of droplet during impact test was in fact strongly linked to the roll-off tendency of droplet during roll-off test.