DFT-BASED MODELING AND AI-DRIVEN OPTIMIZATION OF SUSTAINABLE ENERGY MATERIALS
Journal of Artificial Intelligence and Technical Sciences · 26.06.2026 · 1 (S1) · 18
DOI: 10.66366/aits.2026.7 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Artificial Intelligence and Technical Sciences, online erschienen am 26.06.2026. Verfasst von Rustam Rustamov und 3 weiteren Autor:innen.
Abstract (Originalsprache)
A comprehensive first-principles density functional theory (DFT) investigation is performed to systematically examine the structural, electronic, and magnetic properties of transition metal (TM = Cr, Mn, Fe, Co, V, Cu) doped aluminum nitride (AlN) nanosheets. Calculations are carried out using the Atomistix ToolKit (ATK) with the PBEsol generalized gradient approximation functional and PseudoDojo pseudopotentials, employing a 1×3×5 supercell of 120 atoms. TM dopants are substituted at Al sites with concentrations of 2.08%, 4.16%, and 6.25%, and spin-polarized total energy comparisons between ferromagnetic (FM) and antiferromagnetic (AFM) configurations are performed to identify the magnetic ground states. Pristine AlN nanosheets are found to be non-magnetic wide-bandgap semiconductors with a calculated bandgap of approximately 4.05 eV. Upon TM substitution at Al sites, spin degeneracy is lifted, inducing pronounced spin polarization and substantial modifications to the electronic structure near the Fermi level. Strong hybridization between TM-3d and N-2p orbitals governs the resulting magnetic behavior across all dopant types. Cr- and Cu-doped AlN nanosheets exhibit half-metallicity with 100% spin polarization, where one spin channel is metallic while the other remains semiconducting, rendering them highly promising candidates for spintronic device applications. V- and Fe-doped nanosheets retain semiconducting character in both spin channels but display reduced bandgaps and weak ferromagnetic ordering. Mn-doped AlN nanosheets favor AFM coupling with minimal density-of-states variation near the Fermi level. Binding energy calculations confirm structural stability for all TM dopants, with Cr yielding the highest binding energy (−3.74 eV). The DFT-generated dataset of electronic and magnetic descriptors also serves as the foundation for a prospective AI-assisted optimization framework aimed at accelerating the discovery of low-dimensional materials with tailored spintronic functionalities.
Zitieren
APA
Rustamov, R., Ismayilova, N., Guliyev, J., & Ismayil, M. A. (2026). DFT-BASED MODELING AND AI-DRIVEN OPTIMIZATION OF SUSTAINABLE ENERGY MATERIALS. Journal of Artificial Intelligence and Technical Sciences, 1(S1), 18. https://doi.org/10.66366/aits.2026.7
Vancouver
Rustamov R, Ismayilova N, Guliyev J, Ismayil MA. DFT-BASED MODELING AND AI-DRIVEN OPTIMIZATION OF SUSTAINABLE ENERGY MATERIALS. Journal of Artificial Intelligence and Technical Sciences. 2026;1(S1):18. doi:10.66366/aits.2026.7
BibTeX
@article{Rustamov2026,
author = {Rustamov, Rustam and Ismayilova, Narmin and Guliyev, Jeyhun and Ismayil, Muhammad Ali},
title = {DFT-BASED MODELING AND AI-DRIVEN OPTIMIZATION OF SUSTAINABLE ENERGY MATERIALS},
journal = {Journal of Artificial Intelligence and Technical Sciences},
year = {2026},
volume = {1},
number = {S1},
pages = {18},
doi = {10.66366/aits.2026.7}
}
EUVIMED