Machine-learning Methods to Predict Wetting Properties of Iron-Based Composites

We used three different methods of statistical data analysis to establish correlations between the water contact angle (CA) on ductile iron as a function of composition, roughness (grit size), elapsed time between sample preparation and CA measurement, and droplet size. The three methods are the Linear Regression Analysis (LRA), Artificial Neural Network (ANN) model, and the multivariate Polynomial Regression Analysis. It is established that the size of the water droplet is statistically insignificant, while correlations with the other three parameters were found. The surface roughness is the most important predictor of the CA. A low coefficient of determination of the linear regression indicates that the correlation is non-linear. The ANN model showed much stronger predictive potential than the LRA. We discuss the correlation with experimental values of the contact angle and physical mechanisms behind the observed trends. It is particularly promising that the ANN can be trained to predict the wetting characteristics. The application of machine learning methods to synthesize new materials and coatings with desired surface properties, such as self-cleaning, is a technology, which may become a part of the emergent “Triboinformatics” field, related to the application of the machine learning methods to surface science and engineering.