Neural Network Prediction of Micrometer-Scale Equivalent Contact Angle Mapping: From Microforce Measurements to Local Wettability Characterization

The microscopic structural heterogeneity of superhydrophobic surfaces may lead to variations in wettability on the micrometer scale. By combining high-precision adhesion force measurements with neural networks, the variations and wettability states of superhydrophobic surfaces at the micrometer scale can be accurately assessed. Based on the droplet cantilever probe technique, the interaction forces between droplets and superhydrophobic surfaces were quantitatively measured. A neural network model was designed and trained using experimental data, enabling nonlinear mapping between the adhesion force and macroscopic contact angle, thereby generating a micrometer-scale equivalent contact angle distribution map. The experiment included surfaces with different microstructures to validate the universality of the method. Compared with traditional wettability characterization methods, the force-neural network fusion method (F-NNFM) could reveal the microscopic variations and states of wettability on superhydrophobic surfaces with a spatial resolution of 5 μm. The neural network model successfully correlated the nonlinear relationship between the adhesion force and contact angle ( R > 0.9, with errors <5°). The concept of the “micrometer-scale equivalent contact angle distribution map” extended the characterization of macroscopic wetting states to microscopic scale studies.