Abstract
Shape memory alloys (SMAs) are desirable candidates for elastocaloric effect materials, but they all suffer from large thermal hysteresis (T-hys). This study analyzes multicomponent TiNi-based SMAs dataset by machine learning (ML) to explore new SMAs with narrow T-hys. The second-largest eigenvalue lambda(2) of the stretch trans-formation matrix U is added to the original dataset to guide the ML process as a feature. Firstly, lambda(2) is obtained by first-principles calculations combined with ML. XGBoost Regressor (XGBR) combined with Leave-One-Out Cross -Validation (LOO-CV) is selected from four algorithms for modeling with the highest coefficient of determination R-2 of 0.87. The introduction of lambda(2) improves the performance of the model. The dataset is divided into 15 groups based on different doping elements (such as Hf, Cu, Zr, etc.), among which TiNiCu is the most predictive component with the R-2 of 0.89. Over 500 TiNiCu components are randomly generated and predicted T-hys. Based on the contour maps created from the prediction results, it is found that T-hys is likely to decrease with the in-crease of Cu doping in general, and minimum T-hys occurs when the Cu is about 15 at. %, which is consistent with the existing experimental results. Eventually, a potential Thys minimum (1.2 K) region of TixNiyCuz (58.3%<= x <= 58.5%, 26.5% <= y <= 27%, 14.8% <= z <= 15.3%, x +y +z =100%) SMA composition is predicted. Our study not only provides a potential selection of narrow T-hys TiNi-based SMAs but also indicates combining of XGBoost and DFT calculation is an effective strategy for materials design.
NSTL
Elsevier