journal article·2026-01Open access

Design of novel full Heusler alloys Mn2YAl (Y Fe, Sc, Ni, co): A combined machine learning and DFT study of magnetic and electronic properties

Waqas Akhtar, Ahmed Ishfaq, Lu Zheng, Shanza Mubashir, Yong Liu, Nan Qu, Jingchuan Zhu

Abstract

This study presents a novel approach for predicting the magnetic and electronic properties of Full Heusler alloys (FHAs) by employing eXtreme Gradient Boosting (XGBoost), a machine learning technique, alongside Density Functional Theory (DFT) calculations. The model employs feature engineering, k-fold cross-validation, and Bayesian optimization (BO) to precisely predict the material properties derived from an extensive library of Heusler alloy compositions. we employed rigorous assessment measures, resulting in a high performance on the unseen test set with R 2 = 0.90, MSE = 0.13, and MAE = 0.21 for the magnetic and R 2 = 0.87, MSE = 0.14, and MAE = 0.26 for the electronic properties. Compared to DFT, the XGBoost model produces very precise predictions with negligible error from −12.4 to 9.6 %, thus increasing the rate at which new materials are discovered. This work demonstrates the application of machine learning (ML) in materials science and facilitates further exploration of HAs, which are characterized by their magnetic and spintronic properties.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Heusler alloys: electronic and magnetic properties · Advanced Thermoelectric Materials and Devices · Advanced Welding Techniques Analysis

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (48)

Show source-provided references
1. Model for strain and magnetization in magnetic shape-memory alloys10.1063/1.367094 ↗
2. Magnetic and martensitic phase transitions in ferromagnetic Ni–ga–Fe shape memory alloys10.1063/1.1532105 ↗
3. High figure of merit in (Ti,Zr,Hf)NiSn half-heusler alloys10.1016/j.scriptamat.2012.03.002 ↗
4. Geometric, electronic, and magnetic structure of co 2 FeSi : curie temperature and magnetic moment measurements and calculations10.1103/PhysRevB.72.184434 ↗
5. Discovery and design of Equiatomic Heusler alloys magnetic properties using machine learning with DFT approach
6. Enhanced thermoelectric performance in TiNiSn-based half-Heuslers10.1039/c2cc37121a ↗
7. Antiferromagnetic half-metals, gapless half-metals, and spin gapless semiconductors: the D03-type Heusler alloys10.1063/1.4840318 ↗
8. Search for spin gapless semiconductors: the case of inverse Heusler compounds10.1063/1.4775599 ↗
9. L2 1 and XA ordering competition in titanium-based full-Heusler alloys10.1039/C7TC03909C ↗
10. Magnetic correlations in martensitic Ni-Mn-based Heusler shape-memory alloys: neutron polarization analysis10.1103/PhysRevB.79.212401 ↗
11. Predicting elastic and electronic properties of quaternary Heusler alloys using machine learning and DFT calculations10.1088/1402-4896/ad8fe5 ↗
12. First-principles calculations of a new half-metallic Heusler alloy FeCrAs10.1016/j.jallcom.2019.153118 ↗
13. Theoretical search for half-Heusler topological insulators10.1103/PhysRevB.91.094107 ↗
14. Bainsla · 2016 · 1
15. Maity · 2020
16. Magnetic dichroism study on Mn1.8 Co1.2 Ga thin film using a combination of x-ray absorption and photoemission spectroscopy10.1088/0022-3727/48/16/164007 ↗
17. High-throughput computational design of cathode coatings for Li-ion batteries10.1038/ncomms13779 ↗
18. Finding unprecedentedly low-thermal-conductivity half-Heusler semiconductors via high-throughput materials Modeling
19. Accelerated discovery of new magnets in the Heusler alloy family10.1126/sciadv.1602241 ↗
20. Using human brain activity to guide machine learning10.1038/s41598-018-23618-6 ↗
21. Machine learning modeling of lattice constants for half-Heusler alloys
22. E · 2024
23. Machine learning for the development of new materials for a magnetic tunnel junction10.1038/s44306-025-00094-z ↗
24. Machine learning study of highly spin-polarized Heusler alloys at finite temperature
25. Mitra · 2022
26. Machine learning assisted development of Heusler alloys for high magnetic moment10.1016/j.commatsci.2025.113692 ↗
27. A machine learning approach to predict the structural and magnetic properties of Heusler alloy families10.1016/j.commatsci.2022.111836 ↗
28. https://aflow.org/, (n.d.). 2025.
29. An empirical analysis of feature engineering for predictive modeling
30. Foundational methods: descriptive statistics: bivariate and multivariate data (correlations, associations)
31. XGBoost
32. Research on a mixed gas classification algorithm based on extreme random tree10.3390/app9091728 ↗
33. ImageNet classification with deep convolutional neural networks10.1145/3065386 ↗
34. Classification of EEG spatial patterns with a tree-structured methodology: CART10.1109/TBME.1986.325684 ↗
35. Lundberg
36. A new method of locating the maximum point of an arbitrary multipeak curve in the presence of noise10.1115/1.3653121 ↗
37. Mockus · 1989 · 1
38. Tiesis
39. Efficient global optimization of expensive BlackBox functions10.1023/A:1008306431147 ↗
40. First principles methods using CASTEP10.1524/zkri.220.5.567.65075 ↗
41. Generalized gradient approximation made simple10.1103/PhysRevLett.77.3865 ↗
42. First-principles study of Ti-based X2TiH5 (X = Mg, ca, Sr) hydrides for advanced hydrogen storage applications10.1016/j.chemphys.2024.112499 ↗
43. Advanced computational screening of X2CaH4 (X = Rb and cs) for hydrogen storage applications10.1016/j.ijhydene.2024.09.321 ↗
44. Hydrogen storage potential of XNiH3 (X= Sr and Ba) compounds: a comprehensive DFT analysis10.1016/j.physleta.2024.129899 ↗
45. Relaxation of crystals with the quasi-Newton method10.1006/jcph.1996.5612 ↗
46. Error estimates for solid-state density-functional theory predictions: an overview by means of the ground-state elemental crystals10.1080/10408436.2013.772503 ↗
47. Special points for Brillouin-zone integrations10.1103/PhysRevB.13.5188 ↗
48. Slater-Pauling behavior and origin of the half-metallicity of the full-Heusler alloys10.1103/PhysRevB.66.174429 ↗