Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Systematic generation of electron models for Second-Principles Density Functional Theory Methods

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A quasi-automated, symmetry-enforced pipeline builds SPDFT electron models that reproduce DFT band structures on geometries outside the training set, for SrTiO3 and LiF.

desk verdict A genuinely useful automation of SPDFT electronic-model construction with solid neutral-geometry validation, but the abstract's charge-state fidelity claim is tested nowhere. read the letter →

arxiv 2505.22056 v1 pith:GDC7MN2X submitted 2025-05-28 cond-mat.mtrl-sci cond-mat.other

Nayara Carral-Sainz , Toraya Fernández-Ruiz , Jorge Íñiguez , Javier Junquera , Pablo Garcia-Fernandez This is my paper
classification cond-mat.mtrl-scicond-mat.other
keywords second-principlesDFTWannierfunctionselectron-latticecouplingelectron-electroninteractionsspacegroupsymmetrySrTiO3LiFpolaronsandexcitons
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper attempts to turn second-principles DFT (SPDFT) model construction into a quasi-automated, reproducible procedure: a user supplies a reference crystal structure and a handful of physical thresholds, and the pipeline returns an electronic Hamiltonian that tracks DFT band structures at a far lower computational cost. The key division of labor is that one-electron hoppings and electron-lattice couplings are computed directly from first-principles data, while the effective electron-electron response parameters are fitted to a small, standardized training set of doped and spin-polarized supercells. Space-group symmetry is enforced on every tensor, which removes spurious parameter differences and reduces the number of required DFT calculations by orders of magnitude. If the method performs as claimed, large-scale simulations of polarons and excitons in materials such as SrTiO3 and LiF become practical.

What carries the argument

The central object is the SPDFT energy functional of Eq. (9), written in a basis of localized Wannier orbitals (orbitals extracted from the DFT wavefunctions): the one-electron hopping matrix $\gamma_{ab}$, the linear and quadratic electron-lattice couplings $\mathbf{f}_{ab,\lambda}$ and $\overleftrightarrow{g}_{ab,\lambda\upsilon}$, and the effective electron-electron integrals $U_{ab,a'b'}$ and $I_{ab,a'b'}$. Space-group operations of the reference geometry are applied to every tensor, grouping symmetry-equivalent parameters and letting one DFT calculation stand for all symmetry-equivalent distortions; the resulting symmetry constraints are enforced with Lagrange multipliers during the electron-electron fit. The electron-electron integrals are obtained by solving the linear fitting equations (32)-(33) against Hamiltonians from the training set, and their influence is screened by a spatial cutoff $\delta_{ree}$ and a goal-function threshold $\delta_\Theta$ that retain only the integrals that matter. This machinery converts the old fitting problem into a mostly computed, symmetry-consistent, and systematically refinable model.

What would settle it

Train a SrTiO3 model using only the paper's stated doping range, −0.3 to +0.3 e with spin polarizations up to 0.3 e, then compute the DFT Hamiltonian for a held-out +0.5 e doped supercell and compare the model's valence and t2g bands near the gap. If the error approaches the LDA gap of about 1.7 eV, the fitted $U$ and $I$ integrals have failed to transfer to an unseen charge state, and the central claim of high-fidelity reproduction across charge states is falsified.

Watch

Extended reading notes

Core claim

The paper claims that a largely automated protocol can construct SPDFT electronic models that reproduce DFT reference data across atomic configurations and charge states. The protocol works in a basis of localized Wannier orbitals, separates valence and conduction manifolds during wannierization so the reference density stays well defined, and enforces the space group of the reference structure on all Hamiltonian, electron-lattice, and electron-electron parameters. One-electron and electron-lattice terms are computed directly (hoppings from wannierized Hamiltonians; couplings by finite differences), while the electron-electron integrals $U_{ab,a'b'}$ and $I_{ab,a'b'}$ — effective responses of the Hamiltonian to changes in density and spin — are fitted to a standardized training set that includes hole/electron doping and spin-polarized configurations. For SrTiO3 and LiF, the resulting models keep deviations in the valence and lower conduction bands small even for random atomic displacements up to 0.17 Å, with errors much smaller than typical band separations, and the training set shrinks to a few hundred single-point calculations.

Load-bearing premise

The load-bearing premise is that the electron-electron interaction parameters fitted to a small set of doped and spin-polarized supercell calculations transfer unchanged to other geometries, larger supercells, and charge states outside that training set.

Editorial extensions

If this is right

  • Materials with strong electron-lattice coupling, such as transition-metal perovskites, can be simulated in supercells large enough for polaron formation and hopping without a self-consistent DFT loop at every step.
  • The real-time time-dependent SPDFT extension can compute optical spectra with explicit electron-hole interactions, making exciton binding and dynamics in wide-gap insulators like LiF directly accessible.
  • Because symmetry turns tens of thousands of candidate DFT calculations into a few hundred, systematic model generation for new materials becomes affordable, and the models become comparable across different systems and users.
  • The explicit hierarchy of cutoffs ($\delta_{rh}$, $\delta_{rel}$, $\delta_f$, $\delta_g$, $\delta_{ree}$, $\delta_\Theta$) gives a principled route to refinement: tighten a cutoff and the model improves in a controlled way.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper tests random geometries within the same charge-state range as the training set, so a natural extension is to hold out entire charge states or spin configurations; the fitted $U$ and $I$ parameters either transfer, which would confirm the claim, or they do not, which would set the method's actual domain of validity.
  • Because the fitted electron-electron integrals are effective density responses rather than bare Coulomb integrals, their values could be read across materials as a screening descriptor; the paper does not discuss this, but the model files it produces would permit such a comparison.
  • The LDA-based SrTiO3 model did not localize electron polarons, and the paper attributes this to self-interaction error; a direct test of the pipeline is to rebuild it from hybrid-functional data, where localization should emerge if the fitted parameters carry the needed physics.
  • For metals, where the reference density matrix must be non-diagonal and geometry-dependent, the subtraction procedure that separates electron-electron from electron-lattice effects is more delicate and remains largely untested; applying the pipeline to a simple metal would reveal whether quasi-automation survives metallic occupations.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The manuscript presents an improved, largely automated workflow for constructing the electronic part of second-principles DFT (SPDFT) models. The authors generalize the reference electron density to allow non-diagonal density matrices, reformulate linear and quadratic electron-lattice couplings using absolute atomic displacements with acoustic-sum-rule constraints, enforce space-group symmetry to reduce the number of required DFT calculations, and combine direct computation of one-electron and electron-lattice parameters with fitting of the electron-electron parameters U and I to a standardized training set of doped and spin-polarized DFT Hamiltonians. The workflow is applied to SrTiO3 and LiF; the authors report low goal-function values on random-displacement test sets, band-structure error bars at d = 0.17 Å, and a check against an alternative SrTiO3 phase. The central claim is that the resulting models reproduce DFT reference data with high fidelity across various atomic configurations and charge states.

Significance. If the geometry-transferability results hold, this is a substantial practical advance: the symmetry-based reduction in the number of first-principles calculations (e.g., from 57,840 to 15 for the SrTiO3 electron-lattice data at δrel = 2.0 Å) makes SPDFT model construction feasible for a wider class of materials, and the explicit goal-function metric provides a transparent quality assessment. The validation on held-out random geometries, the test on the antiferrodistortive phase of SrTiO3, and the careful documentation of cutoff choices are genuine strengths. The main weakness is that the charge-state half of the abstract's claim is not validated by any held-out test; because U and I are fitted parameters, their transferability to unseen doping levels, spin polarizations, and combined distortion-doping configurations is load-bearing for the promised polaron/exciton applications.

major comments (3)
  1. [Sec. IV C 2, Sec. IV D, Abstract] The abstract claims high fidelity 'across various atomic configurations and charge states,' but the charge-state part is not tested. The U and I parameters are fitted on the Sec. IV C 2 training set (charges from -0.3 to +0.3 e in 0.1 e steps, with fully spin-polarized variants, plus neutral magnetizations for LiF), while the Sec. IV D validation consists of random neutral geometries and alternative phases. No held-out doping level, held-out magnetization, or doped distorted geometry is reported. Since U and I carry the entire density/spin response of the model, the claim that the models reproduce DFT for arbitrary charge states is unsupported. Please add a cross-validation where at least one doping level or magnetization is excluded from the fit and the model error on that held-out configuration is reported, or compare model vs DFT for a doped distorted geometry (e.g., the electron-doped SrTiO3 configuration discussed in Sec. VI A).
  2. [Sec. VI A, Figs. 5-6; Sec. VI B, Figs. 11-12] The same random-displacement test set appears to be used both for selecting the cutoffs δrel, δf, and δg and for reporting the final model errors. Fig. 5(a) is used to choose δrel, Fig. 5(b) to choose δf and δg, and Fig. 6 then reports the errors of the selected model on the same kind of test set; the analogous procedure is used for LiF in Figs. 11 and 12. This makes the reported errors an in-sample estimate of model performance after hyperparameter tuning. Please either reserve a separate final test set not used in any cutoff selection, or report the sensitivity of the final error to the cutoff choices so that the reader can assess the potential optimism bias.
  3. [Sec. VI A, polaron paragraph] The only application that would exercise the fitted U/I response outside the training set is the attempted electron doping of SrTiO3, but the outcome is reported only as a failure to localize charge, attributed to the LDA functional. No comparison of the second-principles model against DFT for that doped configuration is provided, so this episode neither validates nor invalidates the charge-state transferability of U/I. Please clarify whether such a comparison was performed, or state explicitly that it was not and that the charge-state transferability remains to be established.
minor comments (4)
  1. [Figs. 5 and 11 captions] The captions call Θ the 'average error per calculation,' but the text (e.g., Sec. VI B) describes Θ as a sum over Hamiltonian matrix-element deviations (≈1 eV2 summed over 11,964 terms). Please clarify the normalization used in the plots.
  2. [Sec. III D] The final sentence, 'More details will be given in a forthcoming simulation,' is an incomplete placeholder and should be replaced by a reference or removed.
  3. [Sec. I and Sec. VII] The statement that the LiF model has 'already proven capable' of describing optical properties such as excitons is not demonstrated in this manuscript; either show the data or refer explicitly to the forthcoming publication.
  4. [Eq. (29)] Equation (29) appears to contain a typographical error: the first equality writes h^{DFT}_{a'b'}(u_{λ'}) on both sides, while the intended relation should connect the Hamiltonian at the symmetry-related geometry to the Hamiltonian at u_λ. Please correct the indices.

Circularity Check

1 steps flagged · score 6.0 of 10

Charge-state fidelity is a least-squares restatement: the U/I electron-electron parameters are fit to doped and spin-polarized DFT Hamiltonians, and the abstract's 'various charge states' claim is evidenced only by that training objective; only atomic-configuration transfer is tested out-of-sample.

  1. fitted input called prediction [Abstract; Sec. IV C 2 (Eqs. 30-32, 34-35); Sec. IV D; Sec. VI A (Fig. 8)]
    "In both cases, the resulting models reproduce DFT reference data with high fidelity across various atomic configurations and charge states. / Electron-electron parameters are determined by fitting to a training set of DFT-calculated Hamiltonians for a representative ensemble of electronic configurations {A}."

    The U and I parameters are the entire charge/spin response of the model (Eq. 10). They are obtained by minimizing Θ = Σ_ab Σ_A [h_model − h_DFT]^2 (Eqs. 30-32) over doped and spin-polarized configurations, with total charge varied from −0.3 to +0.3 e in steps of 0.1 e and, for LiF, neutral magnetizations up to 2.0 μB. The paper's only evidence for the 'charge states' part of the abstract claim is Fig. 8, which plots this same training objective as U/I terms are added. The validation protocol (Sec. IV D) generates test geometries only by random atomic displacements; no held-out doping level, no held-out magnetization, and no doped distorted geometry is reported.

full rationale

The paper's derivation of the electronic model is mostly self-contained: one-electron hoppings are computed from wannier90, electron-lattice couplings are obtained by finite differences of DFT Hamiltonians with acoustic-sum-rule enforcement, and the validation on randomly displaced geometries (Sec. IV D, Figs. 5-7, 11-12) is a genuine held-out test against external DFT data. The symmetry reduction and the linear-equation fitting procedure are internally consistent. The one circular element is the charge-state part of the central claim: U/I electron-electron parameters are fitted to a training set of doped and spin-polarized DFT Hamiltonians, and the abstract's statement that the models 'reproduce DFT reference data ... across various ... charge states' is supported only by the fitting objective, not by any held-out charge-state test. The paper itself acknowledges the limitation in Sec. I ('it does not include a comprehensive validation in more complex or application-driven scenarios') and in Sec. VII ('analogous procedures to compute electron-electron interaction terms—such as the Uab,a′b′—are not yet available'), which confirms that U/I are fit-only quantities. This is partial circularity: the atomic-configuration prediction is independent and well tested, while the charge-state reproduction reduces by construction to the least-squares fit. Self-citations to Refs. [23], [31], and [45] are normal prior-work support and are not load-bearing circularity here, because the central geometry validation is against external DFT data computed for this paper.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central claim depends on the SPDFT energy expression and lattice Taylor expansion inherited from prior work (Refs. [23], [31], [45]), on standard Wannier-function machinery, and on the correctness of the DFT reference calculations. The genuinely new content is the parameter-extraction pipeline, whose main user-controlled free parameters are the spatial and pruning cutoffs and the fitted U/I integrals. The transferability of these fitted integrals and the validity of the second-order electron-lattice expansion are the key unproved premises.

free parameters (6)
  • Hamiltonian cutoff delta_rh = 8.0 Å for both SrTiO3 and LiF
    User-selected; chosen from C(r) saturation; directly controls model size and accuracy (Sec. VI A, Fig. 4).
  • Electron-lattice coupling cutoff delta_rel = 4.0 Å (SrTiO3), 3.0 Å (LiF)
    User-selected; chosen to include full unit-cell interactions while limiting cost; tuned against test-set error (Sec. VI A, Fig. 5).
  • Electron-lattice pruning thresholds delta_f and delta_g = 0.1 eV/Å and 0.1 eV/Å^2
    User-selected; terms below threshold are discarded; tested values range 0.01-0.5 (Sec. IV C 3, Fig. 5b).
  • Electron-electron fitting threshold delta_Theta = 0.2 eV^2
    User-selected; decides which U/I groups enter the model; set from goalfunction decomposition (Sec. VI A, Fig. 8).
  • Electron-electron integrals U_ab,a'b' and I_ab,a'b' = not tabulated; determined by fitting (Eqs. 32-35)
    Fitted to DFT Hamiltonians for doped and spin-polarized configurations; they are the central fitted parameters of the model (Sec. IV C 2).
  • Finite-difference step delta_x for e-l couplings = not stated in manuscript
    Required to compute f and g in Eqs. (36)-(37); the unreported value affects parameter accuracy and reproducibility.
assumptions (6)
  • domain assumption Born-Oppenheimer approximation: electrons follow nuclei adiabatically
    Used throughout Sec. II to define lattice and electronic expansions; standard in SPDFT.
  • domain assumption The deformation density delta_n is small enough for a low-order expansion of the DFT energy (Eq. 7)
    Sec. II, Eqs. (2)-(8); if violated, the second-order energy expression fails.
  • domain assumption No bond-topology transformations occur; Taylor expansion around the RAG is valid
    Sec. II final paragraph: method explicitly excludes bond breaking, melting, and ionic diffusion.
  • standard math Wannier functions transform under space-group operations as the atomic basis orbitals do
    Sec. IV B, Eqs. (22)-(29); used to symmetrize parameters and reduce the training set; relies on group representation properties.
  • ad hoc to paper Displacements of atoms farther apart than delta_rel are uncorrelated for electron-lattice coupling
    Sec. IV C 3, Eq. (38); assumed to justify the spatial cutoff; no convergence proof beyond the two test materials.
  • domain assumption The DFT functional used (LDA for SrTiO3, PBE for LiF) provides a valid reference for the target physics
    Sec. V; the method inherits all DFT errors; acknowledged in the failure to obtain polarons in LDA.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Systematic generation of electron models for Second-Principles Density Functional Theory Methods." pith.science (2026). https://pith.science/paper/GDC7MN2X

@misc{pith2026250522056,
  author       = {Pith},
  title        = {Pith review of: Systematic generation of electron models for Second-Principles Density Functional Theory Methods},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GDC7MN2X}},
  note         = {Machine review of arXiv:2505.22056}
}
abstract

We present a systematic, quasi-automated methodology for generating electronic models in the framework of second-principles density functional theory (SPDFT). This approach enables the construction of accurate and computationally efficient models by deriving all necessary parameters from first-principles calculations on a carefully designed training set. A key feature of our method is the enforcement of space group symmetries, which reduces both the number of independent parameters and the required computational effort. The formalism includes improved treatments of one-electron Hamiltonians, electron-lattice coupling-through both linear and quadratic terms-and electron-electron interactions, enabling accurate modeling of structural and electronic responses. We apply the methodology to SrTiO$_{3}$ and LiF, materials representative of transition-metal perovskites and wide-band-gap insulators, respectively. In both cases, the resulting models reproduce DFT reference data with high fidelity across various atomic configurations and charge states. Our results validate the robustness of the approach and highlight its potential for simulating complex phenomena such as polarons and excitons. This work lays the foundation for extending SPDFT to real-time simulations of optoelectronic properties and further integration with machine-learning methods.

Figures

Figures reproduced from arXiv: 2505.22056 by the authors.

Figure 1
Figure 1. FIG. 1: Atomic and Wannier function positions. (a) [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2: (a) Atoms of an equispaced one-dimensional lin [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3: (a) Electronic band structure of SrTiO [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: FIG. 4: (Color online) Black solid line (left axis): Dec [PITH_FULL_IMAGE:figures/full_fig_p013_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5: (a) Goalfunction Θ for SrTiO [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6: (Color online) Electronic band structures of SrTiO [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7: (Color online) Electronic band structure of SrTiO [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]
Figure 9
Figure 9. Figure 9: FIG. 9: (Color online) (a) Electronic band structure of [PITH_FULL_IMAGE:figures/full_fig_p017_9.png]
Figure 10
Figure 10. Figure 10: FIG. 10: Logarithmic plot of the [PITH_FULL_IMAGE:figures/full_fig_p018_10.png]
Figure 11
Figure 11. Figure 11: FIG. 11: (Color online) Goalfunction Θ for LiF, measuring the average error per calculation in [PITH_FULL_IMAGE:figures/full_fig_p019_11.png]
Figure 12
Figure 12. Figure 12: FIG. 12: (Color online) Band structures illustrating the statistical error of the second-principles model across a test [PITH_FULL_IMAGE:figures/full_fig_p019_12.png]
Figure 13
Figure 13. Figure 13: FIG. 13: (Color online) (a) Decomposition of the goal [PITH_FULL_IMAGE:figures/full_fig_p020_13.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Extension of Second-Principles Density Functional Theory into the time domain

    cond-mat.mtrl-sci 2025-07 conditional novelty 6.0 of 10

    A real-time, real-space extension of second-principles DFT that propagates the density matrix and yields optical spectra for large metallic and insulating systems.

Reference graph

Works this paper leans on

69 extracted references · 49 canonical work pages · cited by 1 Pith paper

  1. [1]

    However, enforcing the symmetry condition of Eq

    One-electron parameters The hopping matrix elements, γab, are computed using wannier90 from a nonmagnetic ground-state DFT cal- culation at the RAG. However, enforcing the symmetry condition of Eq. (23) requires high numerical precision in the DFT setup, such as dense k-point meshes, which is computationally demanding. An alternative approach, implemented...

    work page

  2. [2]

    Electron-electron parameters, training set and fitting procedure Electron-electron parameters are determined by fitting to a training set of DFT-calculated Hamiltonians for a representative ensemble of electronic configurations {A}. The target is to minimize the deviation between the DFT matrix elements, hDFT ab (A), and the second-principles model predic...

    work page

  3. [3]

    A way of making Europe

    Electron-lattice terms Electron-lattice coupling terms are computed us- ing finite-difference approximations applied to the one- electron Hamiltonian obtained from density functional theory (DFT). Specifically, the x-component of the linear electron-lattice coupling vector ⃗f fd ab,λ is estimated using the central difference formula f fd,x ab,λ ≈ − hDFT a...

    arXiv

  4. [4]

    R. M. Martin, Electronic Structure: Basic Theory and Practical Methods (Cambridge University Press, 2004)

    2004

  5. [5]

    Kohanoff, Electronic Structure Calculations for Solids and Molecules: Theory and Computational Methods (Cambridge University Press, 2006)

    J. Kohanoff, Electronic Structure Calculations for Solids and Molecules: Theory and Computational Methods (Cambridge University Press, 2006)

    2006

  6. [6]

    R. G. Parr and W. Yang, Density F unctional T heory of Atoms and M olecules (Oxford University Press, 1989)

    1989

  7. [7]

    R. M. Dreizler and E. K. U. Gross, Density Functional Theory: An Approach to the Quantum Many-Body Prob- lem (Springer, 1990)

    work page 1990

  8. [8]

    Szabo and N

    A. Szabo and N. Ostlund, Modern Quantum Chemistry: Introduction to Advanced Electronic Structure Theory , Dover Books on Chemistry (Dover Publications, 1996)

    work page 1996

Show all 69 references
  1. [9]

    D. R. Bowler and T. Miyazaki, Calculations for millions of atoms with density functional theory: linear scal- ing shows its potential, J. Phys.: Condens. Matter 22, 074207 (2010)

    2010

  2. [10]

    P. W. Anderson, More is different, Science 177, 393 (1972)

    1972

  3. [11]

    H. Y. Hwang, Y. Iwasa, M. Kawasaki, B. Keimer, N. Na- gaosa, and Y. Tokura, Emergent phenomena at oxide interfaces, Nat. Mater. 11, 103 (2012), (and references therein)

    2012

  4. [12]

    Nagaosa and Y

    N. Nagaosa and Y. Tokura, Emergent electromagnetism in solids, Phys. Scr. 2012, 014020 (2012)

    2012

  5. [13]

    Ohtomo and H

    A. Ohtomo and H. Y. Hwang, A high-mobility electron gas at the LaAlO 3/SrTiO3 heterointerface, Nature 427, 423 (2004)

    2004

  6. [14]

    Nagaosa and Y

    N. Nagaosa and Y. Tokura, Topological properties and dynamics of magnetic skyrmions, Nat. Nanotechnol. 8, 899 (2013)

    2013

  7. [15]

    S. Das, Y. L. Tang, Z. Hong, M. A. P. Gon¸ calves, M. R. McCarter, C. Klewe, K. X. Nguyen, F. G´ omez-Ortiz, P. Shafer, E. Arenholz, V. A. Stoica, S.-L. Hsu, B. Wang, C. Ophus, J. F. Liu, C. T. Nelson, S. Saremi, B. Prasad, A. B. Mei, D. G. Schlom, J.´I˜ niguez, P. Garc ´ ıa-F...

    2019

  8. [16]

    Junquera, Y

    J. Junquera, Y. Nahas, S. Prokhorenko, L. Bellaiche, J. ´I˜ niguez, D. G. Schlom, L.-Q. Chen, S. Salahuddin, D. A. Muller, L. W. Martin, and R. Ramesh, Topological phases in polar oxide nanostructures, Rev. Mod. Phys. 95, 025001 (2023)

    2023

  9. [17]

    Binder and D

    K. Binder and D. W. Heermann, Monte Carlo Simulation in Statistical Physics. An Introduction (Springer, 2019)

    2019

  10. [18]

    Schnakenberg, Polaronic impurity hopping conduction, Phys

    J. Schnakenberg, Polaronic impurity hopping conduction, Phys. Stat. Sol. (b) 28, 623 (1968)

    1968

  11. [19]

    Asadi, A

    K. Asadi, A. J. Kronemeijer, T. Cramer, L. Jan An- ton Koster, P. W. M. Blom, and D. M. de Leeuw, Polaron hopping mediated by nuclear tunnelling in semiconduct- ing polymers at high carrier density, Nat. Commun. 4, 1710 (2013)

    2013

  12. [20]

    T. G. Castner and W. K¨ anzig, The electronic structure of v-centers, J. Phys. Chem. Solids 3, 178 (1957)

    1957

  13. [21]

    K¨ anzig and T

    W. K¨ anzig and T. O. Woodruff, The electronic structure of an h-center, J. Phys. Chem. Solids 9, 70 (1959)

    1959

  14. [22]

    J. M. Garc ´ ıa-Lastra, J. S. G. Myrdal, R. Christensen, K. S. Thygesen, and T. Vegge, DFT + U study of pola- ronic conduction in Li 2O2 and Li2CO3: Implications for Li–Air batteries, J. Phys. Chem. C 117, 5568 (2013)

    2013

  15. [23]

    S. P. Ong, Y. Mo, and G. Ceder, Low hole polaron migra- tion barrier in lithium peroxide, Phys. Rev. B 85, 081105 (2012)

    2012

  16. [24]

    Q. Guo, Y. Liu, M. Liu, H. Zhang, X. Qian, J. Yang, J. Wang, W. Xue, Q. Zhao, X. Xu, W. Ma, Z. Tang, Y. Li, and Z. Bo, Enhancing the performance of organic solar cells by prolonging the lifetime of photogenerated excitons, Adv. Mater. 32, 2003164 (2020)

    2020

  17. [25]

    Q. Guo, X. Zhao, B. Song, J. Luo, and J. Tang, Light emission of self-trapped excitons in inorganic metal halides for optoelectronic applications, Adv. Mater. 34, 2201008 (2022)

    2022

  18. [26]

    Garc ´ ıa-Fern´ andez, J

    P. Garc ´ ıa-Fern´ andez, J. C. Wojde l, J. ´I˜ niguez, and J. Junquera, Second-principles method for materials sim- ulations including electron and lattice degrees of freedom, Phys. Rev. B 93, 195137 (2016)

    2016

  19. [27]

    Schmidt, M

    J. Schmidt, M. R. G. Marques, S. Botti, and M. A. L. Marques, Recent advances and applications of machine learning in solid-state materials science, Npj Comput. Mater. 5, 83 (2019)

    2019

  20. [28]

    Z. J. Baum, X. Yu, P. Ayala, Y. Zhao, S. P. Watkins, and Q. Zhou, Artificial intelligence in chemistry: Current trends and future directions, J. Chem. Inf. Model. 61, 3197 (2021)

    2021

  21. [29]

    H. Li, Z. Wang, N. Zou, M. Ye, R. Xu, X. Gong, W. Duan, and Y. Xu, Deep-learning density functional theory hamiltonian for efficient ab initio electronic- structure calculation, Nat. Comput. Sci. 2, 367 (2022)

    2022

  22. [30]

    com/mzjb/DeepH-pack, accessed: 1 April 2025

    Deeph (deep hamiltonian) method, https://github. com/mzjb/DeepH-pack, accessed: 1 April 2025

    2025

  23. [31]

    Y. Zhuo, A. Mansouri T., and J. Brgoch, Predicting the band gaps of inorganic solids by machine learning, J. Phys. Chem. Lett. 9, 1668 (2018)

    2018

  24. [32]

    J. D. Evans and F.-X. Coudert, Predicting the mechani- cal properties of zeolite frameworks by machine learning, Chem. Mater. 29, 7833 (2017)

    2017

  25. [33]

    Marzari, A

    N. Marzari, A. A. Mostofi, J. R. Yates, I. Souza, and D. Vanderbilt, Maximally localized Wannier functions: Theory and applications, Rev. Mod. Phys. 84, 1419 (2012)

    2012

  26. [34]

    J. C. Wojde l, P. Hermet, M. P. Ljungberg, P. Ghosez, and J. ´I˜ niguez, First-principles model potentials for lattice- dynamical studies: general methodology and example of application to ferroic perovskite oxides, J. Phys.: Con- dens. Matter 25, 305401 (2013)

    2013

  27. [35]

    Porezag, T

    D. Porezag, T. Frauenheim, T. K¨ ohler, G. Seifert, and R. Kaschner, Construction of tight-binding-like potentials on the basis of density-functional theory: Application to carbon, Phys. Rev. B 51, 12947 (1995)

    1995

  28. [36]

    W. M. C. Foulkes and R. Haydock, Tight-binding models and density-functional theory, Phys. Rev. B 39, 12520 22 (1989)

    1989

  29. [37]

    Elstner, D

    M. Elstner, D. Porezag, G. Jungnickel, J. Elsner, M. Haugk, T. Frauenheim, S. Suhai, and G. Seifert, Self-consistent-charge density-functional tight-binding method for simulations of complex materials properties, Phys. Rev. B 58, 7260 (1998)

    1998

  30. [38]

    Gonze, B

    X. Gonze, B. Amadon, G. Antonius, F. Arnardi, L. Baguet, J.-M. Beuken, J. Bieder, F. Bottin, J. Bouchet, E. Bousquet, N. Brouwer, F. Bruneval, G. Brunin, T. Cavignac, J.-B. Charraud, W. Chen, M. Cˆ ot´ e, S. Cottenier, J. Denier, G. Geneste, P. Ghosez, M. Giantomassi, Y. Gille...

    2020

  31. [39]

    Fern´ andez-Ruiz, N

    T. Fern´ andez-Ruiz, N. Carral-Sainz, J. ´I˜ niguez, J. Jun- quera, and P. Garcia-Fernandez, to be sent to Phys. Rev. B (2025)

    2025

  32. [40]

    J. M. Soler, E. Artacho, J. D. Gale, A. Garc ´ ıa, J. Jun- quera, P. Ordej´ on, and D. S´ anchez-Portal, Thesiesta method for ab initio order- N materials simulation, J. Phys.: Condens. Matter 14, 2745 (2002)

    2002

  33. [41]

    A. A. Mostofi, J. R. Yates, Y.-S. Lee, I. Souza, D. Van- derbilt, and N. Marzari, wannier90: A tool for obtaining maximally-localised wannier functions, Comput. Phys. Commun. 178, 685 (2008)

    2008

  34. [42]

    Pizzi, V

    G. Pizzi, V. Vitale, R. Arita, S. Bl¨ ugel, F. Freimuth, G. G´ eranton, M. Gibertini, D. Gresch, C. Johnson, T. Koretsune, J. Iba˜ nez-Azpiroz, H. Lee, J.-M. Lihm, D. Marchand, A. Marrazzo, Y. Mokrousov, J. I. Mustafa, Y. Nohara, Y. Nomura, L. Paulatto, S. Ponc´ e, T. Pon- wei...

    2020

  35. [43]

    Souza, N

    I. Souza, N. Marzari, and D. Vanderbilt, Maximally local- ized wannier functions for entangled energy bands, Phys. Rev. B 65, 035109 (2001)

    2001

  36. [44]

    Marzari, A

    N. Marzari, A. A. Mostofi, J. R. Yates, I. Souza, and D. Vanderbilt, Maximally localized wannier functions: Theory and applications, Rev. Mod. Phys. 84, 1419 (2012)

    2012

  37. [45]

    Garc ´ ıa, N

    A. Garc ´ ıa, N. Papior, A. Akhtar, E. Artacho, V. Blum, E. Bosoni, P. Brandimarte, M. Brandbyge, J. I. Cerd´ a, F. Corsetti, R. Cuadrado, V. Dikan, J. Ferrer, J. Gale, P. Garc ´ ıa-Fern´ andez, V. M. Garc ´ ıa-Su´ arez, S. Garc ´ ıa, G. Huhs, S. Illera, R. Koryt´ ar, P. Koval...

    2020

  38. [46]

    M. I. Aroyo, J. M. P´ erez-Mato, D. Orobengoa, E. Tasci, G. De La Flor, and A. Kirov, Crystallography online: Bil- bao crystallographic server, Bulg. Chem. Commun. 43, 183 (2011)

    2011

  39. [47]

    M. I. Aroyo, A. Kirov, C. Capillas, J. M. P´ erez-Mato, and H. Wondratschek, Bilbao Crystallographic Server. II. Representations of crystallographic point groups and space groups, Acta Cryst. A62, 115 (2006)

    2006

  40. [48]

    Escorihuela-Sayalero, J

    C. Escorihuela-Sayalero, J. C. Wojde l, and J.´I˜ niguez, Ef- ficient systematic scheme to construct second-principles lattice dynamical models, Phys. Rev. B 95, 094115 (2017)

    2017

  41. [49]

    Kaduk, T

    B. Kaduk, T. Kowalczyk, and T. Van Voorhis, Con- strained density functional theory, Chem. Rev. 112, 321 (2012)

    2012

  42. [50]

    Garc ´ ıa, N

    A. Garc ´ ıa, N. Papior, A. Akhtar, E. Artacho, V. Blum, E. Bosoni, P. Brandimarte, M. Brandbyge, J. I. Cerd´ a, F. Corsetti, R. Cuadrado, V. Dikan, J. Ferrer, J. Gale, P. Garc ´ ıa-Fern´ andez, V. M. Garc ´ ıa-Su´ arez, S. Garc ´ ıa, G. Huhs, S. Illera, R. Koryt´ ar, P. Koval...

    2020

  43. [51]

    D. M. Ceperley and B. J. Alder, Ground state of the electron gas by a stochastic method, Phys. Rev. Lett. 45, 566 (1980)

    1980

  44. [52]

    J. P. Perdew and A. Zunger, Self-interaction correction to density-functional approximations for many-electron systems, Phys. Rev. B 23, 5048 (1981)

    1981

  45. [53]

    J. P. Perdew, K. Burke, and M. Ernzerhof, Generalized gradient approximation made simple, Phys. Rev. Lett. 77, 3865 (1996)

    1996

  46. [54]

    Junquera, O

    J. Junquera, O. Paz, D. S´ anchez-Portal, and E. Artacho, Numerical atomic orbitals for linear-scaling calculations, Phys. Rev. B 64, 235111 (2001)

    2001

  47. [55]

    D. R. Hamann, M. Schl¨ uter, and C. Chiang, Norm- conserving pseudopotentials, Phys. Rev. Lett. 43, 1494 (1979)

    1979

  48. [56]

    Troullier and J

    N. Troullier and J. L. Martins, Efficient pseudopotentials for plane-wave calculations, Physical review B 43, 1993 (1991)

    1991

  49. [57]

    Kleinman and D

    L. Kleinman and D. M. Bylander, Efficacious form for model pseudopotentials, Phys. Rev. Lett. 48, 1425 (1982)

    1982

  50. [58]

    H. J. Monkhorst and J. D. Pack, Special points for Brillouin-zone integrations, Phys. Rev. B 13, 5188 (1976)

    1976

  51. [59]

    Marzari and D

    N. Marzari and D. Vanderbilt, Maximally localized gen- eralized Wannier functions for composite energy bands, Phys. Rev. B 56, 12847 (1997)

    1997

  52. [60]

    Zhong and D

    W. Zhong and D. Vanderbilt, Effect of quantum fluc- tuations on structural phase transitions in SrTiO 3 and BaTiO3, Phys. Rev. B 53, 5047 (1996)

    1996

  53. [61]

    Y. L. Li, S. Choudhury, J. H. Haeni, M. D. Biegal- ski, A. Vasudevarao, A. Sharan, H. Z. Ma, J. Levy, V. Gopalan, S. Trolier-McKinstry, D. G. Schlom, Q. X. Jia, and L. Q. Chen, Phase transitions and domain struc- tures in strained pseudocubic (100) SrTiO 3 thin films, Phys. Re...

    2006

  54. [62]

    Block and A

    D. Block and A. Wasiela, Self-trapped exciton in alkali fluorides: ODMR study, Solid State Commun. 28, 455 (1978)

    1978

  55. [63]

    K. S. Song and H. T. Williams, Self-trapped excitons (Springer, 1996). 23

    1996

  56. [64]

    Williams and K

    R. Williams and K. Song, The self-trapped exciton, J. Phys. Chem. Solids 51, 679 (1990)

    1990

  57. [65]

    S´ anchez, C

    F. S´ anchez, C. Ocal, and J. Fontcuberta, Tailored sur- faces of perovskite oxide substrates for conducted growth of thin films, Chem. Soc. Rev. 43, 2272 (2014)

    2014

  58. [66]

    Kawasaki, A

    M. Kawasaki, A. Ohtomo, T. Arakane, K. Takahashi, M. Yoshimoto, and H. Koinuma, Atomic control of SrTiO3 surface for perfect epitaxy of perovskite oxides, App. Surf. Sci. 107, 102 (1996)

    1996

  59. [67]

    van Benthem, C

    K. van Benthem, C. Els¨ asser, and R. H. French, Bulk electronic structure of SrTiO 3: Experiment and theory, J. Appl. Phys. 90, 6156 (2001)

    2001

  60. [68]

    A. V. Krukau, O. A. Vydrov, A. F. Izmaylov, and G. E. Scuseria, Influence of the exchange screening parame- ter on the performance of screened hybrid functionals, J. Chem. Phys. 125, 224106 (2006)

    2006

  61. [69]

    At higher energies, beyond the usual range of optical/UV experiments, the bands differ

    The low-lying conduction bands obtained in siesta are a good match for those of a more accurate plane-wave code. At higher energies, beyond the usual range of optical/UV experiments, the bands differ. In particular, the ones ob- tained with siesta become disentangled. This for...

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.