Topology-Based Phase Identification of Bulk, Interface, and Confined Water Using an Edge-Conditioned Convolutional Graph Neural Network

Water plays a significant role in various physicochemical and biological processes. Understanding and identifying water phases in various systems such as bulk, interface, and confined water is crucial in improving and engineering state-of-the-art nano-devices. Various order parameters have been developed to distinguish water phases, including bond-order parameters, local structure index, and tetrahedral order parameters. These order parameters are often developed with the assumption of homogenous bulk systems, while most applications involve heterogeneous and non-bulk systems, thus, limiting their generalizability. Our study develops a methodology based on the graph neural network to distinguish water phases directly from data and to learn features instead of predefining them. We provide comparisons between baseline methods trained using conventional order-parameters as features, and a graph neural network model trained using radial distance and hydrogen-bonding information for phase classification and phase transition of water in bulk, interface, and confined systems with continuous and discontinuous phase transitions. VI, VII, VIII, and IX Ices as well as liquid water. For interface and confined systems, we use similar input as for a bulk system, but the output is a binary value indicating whether water is liquid or solid. GNNs are trained to predict whether a particular configuration of atoms is liquid-like or solid-like inside CNTs or at interface for various temperatures. We study CNT 10x10, inside which both continuous and discontinuous melting temperature of 275.0 ± 2.5 K. The behavior of GNN is monotonic within the temperature range of our study, showing an increasing number of liquid-like molecules, while RF shows a non-monotonic and inconsistent behavior with temperature increase.