High-Entropy Shape Memory Alloy Knowledge base: Bridging Data and Discovery

Shape Memory Alloys (SMAs) face limitations in thermal stability and operational temperature, which restrict their use in high-temperature applications. Leveraging High Entropy Alloy (HEA) design strategies for multi-principal element SMAs offers a pathway to enhance thermal stability and actuation performance in extreme conditions. The empirical rules based on parameters such as entropy of mixing, atomic size difference, enthalpy of mixing, valence electron configuration, and electronegativity difference have been made for High Entropy Alloys. Establishing similar guidelines for High Entropy Shape Memory Alloys (HESMAs) can accelerate discovery by focusing on promising regions within the vast multicomponent compositional space.

This study introduces a database tool for HESMAs, which archives and explores alloy compositions while computing atomistic and thermodynamic features from chemical data. These features are correlated with transformation temperatures and hysteresis behavior using ML. The tool includes an interactive user interface that allows users to select elements, properties, and filters, and visualize results. It provides a structured data for both newcomers and experts to design and explore new HESMA compositions with targeted properties.

The tool is also Machine Learning (ML)–ready, automatically generating HEA descriptors suitable as ML inputs. Its functionality is demonstrated using Ni-Ti-Cu-Co-Hf-Zr based HESMAs. Predicted compositions from the tool were experimentally validated and matched well with expectations. Overall, this work represents a significant step toward data-driven materials discovery for high-performance shape memory alloys through integrated computation, visualization, and machine learning capabilities.