Thermodynamic study of reconstructed crystal surfaces.

Rubbo, Marco; Bruno, Marco; Prencipe, Mauro

doi:10.1016/j.susc.2014.10.015

Cited by 4 publications

(5 citation statements)

References 23 publications

(32 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this work, we are interested in the neural network prediction of DFT calculations in low coordination environments rather than providing detailed study of LiF surfaces, which has been the work of previous studies. 32 Thus, optimization and testing of various minimum energy structures were not performed. We therefore note that the surface energies calculated here should not be taken as precise determination of values that should be compared to experimental values.…”

Section: Resultsmentioning

confidence: 99%

“…We note that this differs from other proposed (111) surfaces, such as the octopolar reconstructed surface where 1/4 of the atoms in the outermost layer and 3/4 of the atoms on the subsequent layer are removed. 32 The surface energies calculated from DFT are compared to the neural network in Table 1 for the (100), (110), (111) Li , and (111) F surfaces. The DFT-calculated surface energies ranged from 34.6 to 67.2 meV/Å 2 , representing a broad range of values.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Modeling LiF and FLiBe Molten Salts with Robust Neural Network Interatomic Potential

Lam

Ballinger

et al. 2021

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

Lithium-based molten salts have attracted significant attention due to their applications in energy storage, advanced fission reactors, and fusion devices. Lithium fluorides and particularly 66.6%LiF–33.3%BeF2 (Flibe) are of considerable interest in nuclear systems, as they show an excellent combination of favorable heat transfer, neutron moderation, and transmutation characteristics. For nuclear salts, the range of possible local structures, compositions, and thermodynamic conditions presents significant challenges in atomistic modeling. In this work, we demonstrate that atom-centered neural network interatomic potentials (NNIPs) provide a fast method for performing molecular dynamics of molten salts that is as accurate as ab initio molecular dynamics. For LiF, these potentials are able to accurately reproduce ab initio interactions of dimers, crystalline solids under deformation, crystalline LiF near the melting point, and liquid LiF at high temperatures. For Flibe, NNIPs accurately predict the structures and dynamics at normal operating conditions, high-temperature–pressure conditions, and in the crystalline solid phase. Furthermore, we show that NNIP-based molecular dynamics of molten salts are scalable to reach long time scales (e.g., nanosecond) and large system sizes (e.g., 105 atoms) while maintaining ab initio density functional theory accuracy and providing more than 3 orders of magnitude of computational speedup for calculating structure and transport properties.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Modeling LiF and FLiBe Molten Salts with Robust Neural Network Interatomic Potential

Lam

Ballinger

et al. 2021

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

show abstract

“…According to the terminology universally accepted concerning epitaxy, when the lattice constants of phases A and B match, that is when (1 × 1)-A ≡ (1 × 1)-B, the interface is said to be coherent; when a relation such as (m × n)-A ≡ (k × s)-B exists, with m, n, k and s integers (and the supercell parameters are not too long on the lattice length scale), the interface is commensurate; otherwise it is incommensurate. Moreover, as widely discussed in some recent papers, [2][3][4][5][6][7][8][9] crystal faces (hkl) A and (h′k′l′) B can show several surface terminations (e.g., different structures for the same surface). If the number of the surface terminations is p and r for (hkl) A and (h′k′l′) B , respectively, the number of possible interface configurations can be very high, p × r.…”

Section: Introductionmentioning

confidence: 82%

Epitaxy: a methodological approach to the study of an old phenomenon

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Here, there are two possible surfaces: one with Li-termination (111)Li and one with F-termination (111)F as shown in Figure 2b) and 2c). We note that this differs from other proposed (111) surfaces, such as the octopolar reconstructed surface where ¼ of the atoms in the outermost layer and ¾ of the atoms on the subsequent layer are removed [19]. The surface energies calculated from DFT are compared to neural network in Table 1 for the (100), ( 110) , (111)Li and (111)F surfaces.…”

Section: Pair Interactions Bulk Eos and Surfacesmentioning

confidence: 99%

Modeling LiF and FLiBe Molten Salts with Robust Neural Network Interatomic Potentials

Lam

Li²,

Ballinger³

et al. 2021

Preprint

View full text Add to dashboard Cite

Lithium-based molten salts have attracted significant attention due to their applications in energy storage, advanced fission reactors and fusion devices. Lithium fluorides and particularly 66.6%LiF-33.3¾F2 (Flibe) are of considerable interest in nuclear systems, as they show an excellent combination of desirable heat-transfer and neutron-absorption characteristics. For nuclear salts, the range of possible local structures, compositions, and thermodynamic conditions presents significant challenges in atomistic modeling. In this work, we demonstrate that atom-centered neural network interatomic potentials (NNIP) provide a fast and accurate method for performing molecular dynamics of molten salts. For LiF, these potentials are able to accurately model dimer interactions, crystalline solids under deformation, semi-crystalline LiF near the melting point and liquid LiF at high temperatures. For Flibe, NNIPs accurately predicts the structures and dynamics at normal operating conditions, high temperature-pressure conditions, and in the crystalline solid phase. Furthermore, we show that NNIP-based molecular dynamics of molten salts are scalable to reach long timescales (e.g., nanosecond) and large system sizes (e.g., 105 atoms), while maintaining ab initio accuracy and providing more than three orders of magnitude of computational speedup for calculating structure and transport properties.

show abstract

Thermodynamic study of reconstructed crystal surfaces.

Cited by 4 publications

References 23 publications

Modeling LiF and FLiBe Molten Salts with Robust Neural Network Interatomic Potential

Modeling LiF and FLiBe Molten Salts with Robust Neural Network Interatomic Potential

Epitaxy: a methodological approach to the study of an old phenomenon

Modeling LiF and FLiBe Molten Salts with Robust Neural Network Interatomic Potentials

Contact Info

Product

Resources

About