Accurate and robust static hydrophobic contact angle measurements using machine learning

Recent years has seen a surge in new approaches to material discovery. However, in order to validate and characterise these novel materials that are being discovered at unprecedented speeds, methods of experimental characterisation must match the pace of discovery. This is only possible through the use of high throughput autonomous approaches. % Automation addresses sources of human error To address this need, and to overcome the shortcomings in present methods, we present a machine learning (ML) approach to static contact angle measurement that is more accurate and faster than current best practice methods. The ML model was trained on a large data-set (> 7.2 million) generated via solutions of the Young-Laplace equation where the contact angle is known a priori, removing all sources of error from human input. The data-set included the effects of surface roughness, gravity, the size of drop relative to the image, reflections of the drop on the surface, and contact angles from 110° to 180°. The presented ML model (valid for contact angles >110°), in combination with a new automated image and contour processing approach, is shown to be more accurate than other methods when benchmarked against an experimental data-set, with an estimated error of 1 degree. The ML model is also two orders of magnitude faster at predicting contact angles than Young-Laplace fitting (the current best practice approach). The accuracy and speed of the presented approach provides a viable pathway towards robust and reproducible high-throughput contact angle analysis. The open-source software Conan-ML is provided for use and development of new approaches to goniometry.