Wettability of Chemically Heterogeneous Clay Surfaces: Correlation between Surface Defects and Contact Angles as Revealed by Machine Learning.

Quantifying the wettability of clay surfaces and how it changes in the presence of gas mixtures is crucial for designing geo-energy applications such as underground hydrogen storage and carbon capture and sequestration. While computational studies exist for the wettability of atomically perfect mineral substrates, actual minerals possess heterogeneities. This study employs molecular dynamics simulations to examine the impact of surface defects on the wettability of kaolinite surfaces exposed to hydrogen, methane, and carbon dioxide. The results show that siloxane surfaces become more hydrophilic as defect densities increase and that the gases can strongly affect wettability. Carbon dioxide, in particular, shows stronger adsorption on heterogeneous surfaces than hydrogen and methane. As a consequence, carbon dioxide can strongly affect wettability. Additionally, our results show that higher salt concentrations reduce water contact angle, which is important because salt is likely present in the subsurface. A machine learning classification algorithm is applied to interpret the results and develop predictive capabilities. Our findings highlight the importance of surface defects on wettability, which is essential for designing geological repositories for geo-energy applications ranging from enhanced gas recovery to carbon sequestration and intermittent hydrogen storage.