Ab initio approaches to high-entropy alloys: a comparison of CPA, SQS, and supercell methods

Karabin, Mariia; Mondal, Wasim Raja; Östlin, Andreas; Ho, Wai-Ga D.; Dobrosavljević, Vladimir; Tam, Ka-Ming; Terletska, Hanna; Chioncel, Liviu; Wang, Yan; Eisenbach, Markus

doi:10.1007/s10853-022-07186-9

Cited by 17 publications

(6 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For small-sized to medium-sized systems, typically containing hundreds of atoms, the SQS model is generally considered computationally efficient. However, as the size of the system increases to thousands of atoms or more, the computational cost of using the SQS model becomes prohibitive even for existing US-DOE supercomputing facilities because the number of needed special quasi-random structures increases exponentially with the system size [47].…”

Section: Related Workmentioning

confidence: 99%

Transferring predictions of formation energy across lattices of increasing size*

Lupo Pasini,

Karabin,

Eisenbach

2024

Mach. Learn.: Sci. Technol.

Self Cite

View full text Add to dashboard Cite

In this study, we show the transferability of graph convolutional neural network (GCNN) predictions of the formation energy of the nickel-platinum (NiPt) solid solution alloy across atomic structures of increasing sizes. The original dataset was generated with the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) using the second nearest-neighbor modified embedded-atom method (2NN MEAM) empirical interatomic potential. Geometry optimization was performed on the initially randomly generated face centered cubic (FCC) crystal structures and the formation energy has been calculated at each step of the geometry optimization, with configurations spanning the whole compositional range. Using data from various steps of the geometry optimization, we first trained our open-source, scalable implementation of GCNN called HydraGNN on a lattice of 256 atoms, which accounts well for the short-range interactions. Using this data, we predicted the formation energy for lattices of 864 atoms and 2,048 atoms, which resulted in lower-than-expected accuracy due to the long-range interactions present in these larger lattices. We accounted for the long-range interactions by including a small amount of training data representative for those two larger sizes, whereupon the predictions of HydraGNN scaled linearly with the size of the lattice. Therefore, our strategy ensured scalability while reducing significantly the computational cost of training on larger lattice sizes.

show abstract

Section: Related Workmentioning

confidence: 99%

Transferring predictions of formation energy across lattices of increasing size*

Lupo Pasini,

Karabin,

Eisenbach

2024

Mach. Learn.: Sci. Technol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…[35] The Special Quasirandom Structure (SQS) method is widely employed and highly effective because it provides the most accurate approximation of average local correlation. [36,37] Our calculations were performed using the Vienna Ab initio Simulation Package (VASP) [38] with projector-augmented wave pseudopotentials (PAW) [39] and the Perdew-Burke-Ernzerhof (PBE) version of the generalized gradient approximation (GGA) for exchange-correlation. [40] The total density of states (DOS) of FeCoNi, FeCoNiAl, Fe-CoNiTi, and FeCoNiAlTi is shown in Figure 2.…”

Section: Enhanced Interband Transition In Feconialti Heamentioning

confidence: 99%

Enhanced Interband Transition in FeCoNiAlTi High‐Entropy Alloy for Ultrafast Nonlinear Optical Applications

Liu,

Li,

et al. 2023

Advanced Optical Materials

View full text Add to dashboard Cite

High‐entropy alloys (HEAs) possess superior mechanical, thermal, and electronic properties, which attract great interest in the fields of photothermal, electrocatalysis, and magnetic applications. Herein, the nonlinear optical properties of FeCoNiAlTi HEA with a large specific surface area for the first time is investigated. Owing to the strong d–d interaction, the interband transition near the Fermi level could be enhanced in HEAs, leading to broadband optical absorption. Excitation with the picosecond laser source, the FeCoNiAlTi HEA possesses saturable absorption with a large modulation depth of 7.61% and 30.47% at 1 and 1.5 µm, respectively. Furthermore, at the higher excitation level at 1.5 µm, the saturable absorber exhibits a typical reverse saturable absorption property. Subsequently, the FeCoNiAlTi HEA on the tapered fiber first serves as the saturable absorber in the Yb‐doped fiber laser and the Er‐doped fiber laser, leading to the noise‐like pulses at 1 µm, and the stable Q‐switching, conventional soliton, and dissipative soliton resonance mode‐locking operations at 1.5 µm. This work confirms that the enhanced saturable absorption in HEAs can endow the high‐performance ultrashort pulse generation.

show abstract

Section: Related Workmentioning

confidence: 99%

Transferable prediction of formation energy across lattices of increasing size

Pasini

Karabin

Eisenbach

2023

Preprint

View full text Add to dashboard Cite

In this study, we show the transferability of graph convolutional neural network (GCNN) predictions of the formation energy of the nickel-platinum (NiPt) solid solution alloy across atomic structures of increasing sizes. The original dataset was generated with the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) using the second nearest-neighbor modified embedded-atom method (2NN MEAM) empirical interatomic potential. Geometry optimization was performed on the initially randomly generated face centered cubic (FCC) crystal structures and the formation energy has been calculated at each step of the geometry optimization, with configurations spanning the whole compositional range. Using data from various steps of geometry optimization, we first trained the GCNN on a lattice of 256 atoms, which accounts well for the short-range interactions. Using this data, we predicted the formation energy for lattices of 864 atoms and 2,048 atoms, which resulted in lower-than-expected accuracy due to the long-range interactions present in these larger lattices. We accounted for the long-range interactions by including a small amount of training data representative for those two larger sizes, whereupon the predictions of the GCNN scaled linearly with the size of the lattice. Therefore, our strategy ensured scalability while reducing significantly the computational cost of training on larger lattice sizes.

show abstract

Ab initio approaches to high-entropy alloys: a comparison of CPA, SQS, and supercell methods

Cited by 17 publications

References 44 publications

Transferring predictions of formation energy across lattices of increasing size*

Transferring predictions of formation energy across lattices of increasing size*

Enhanced Interband Transition in FeCoNiAlTi High‐Entropy Alloy for Ultrafast Nonlinear Optical Applications

Transferable prediction of formation energy across lattices of increasing size

Contact Info

Product

Resources

About