Pith. sign in

REVIEW 3 major objections 4 minor 14 references

Optimal transition states for polaron hopping transport without supercells

T0 review · 3 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read This paper shows that polaron hopping barriers, transition states, and adiabatic mobilities can be computed without supercells by combining variational polaron equations with the string method, and that the resulting rutile TiO2 electron mo

desk verdict A genuinely new string-method-plus-variational-polaron combination with a promising TiO2 [001] result; the [110]/[111] channels and the unwritten metric assumption both need referee attention. read the letter →

arxiv 2607.18096 v2 pith:THM4DI3E submitted 2026-07-20 cond-mat.mtrl-sci

Vasilii Vasilchenko , Matteo Giantomassi , Samuel Poncé , Xavier Gonze This is my paper · ORCID
classification cond-mat.mtrl-sci
keywords polaronhoppingsmallminimum-energypathstringmethodvariationalequationsrutileTiO2LiFadiabatictransport
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

Polaron hopping controls charge transport in many materials where electrons get trapped by their own lattice distortion, but standard supercell DFT calculations are expensive and unreliable for such transitions. This paper introduces a method that skips supercells entirely: it works in reciprocal space with a variational polaron energy, and uses the simplified string method to find the minimum-energy path between two polaron states. The path yields the transition state, the activation barrier, an effective attempt frequency, and through transition-state theory, the adiabatic hopping rate and mobility. Applied to rutile TiO2, the computed electron-polaron mobility along the most favourable [001] direction matches measured mobilities, while the Boltzmann transport equation overestimates them because it assumes delocalized band carriers. In LiF, the method uncovers rotation and multi-step hopping mechanisms for the hole polaron.

What carries the argument

The central object is the variational polaron energy functional Epol[A,B] depending on electronic coefficients A and phonon deformation coefficients B, whose gradients provide forces on the polaron potential energy surface. The simplified string method evolves a discretized path between two polaron configurations using only the phonon gradient, with the electronic coefficients re-optimized for each image at fixed deformation; equal-arc-length reparametrization keeps the path smooth. This gives minimum-energy paths and transition states without real-space supercells.

What would settle it

Compute the [110] and [111] minimum-energy paths on k/q meshes denser than 11×11×22 (or include the delocalized transition paths shown in the paper in the transition-state rate) and check whether activation energies of about 81 meV and 78 meV and the mobilities remain stable; alternatively, measure direction-resolved electron mobility in oriented rutile TiO2 single crystals between 200 and 400 K and compare with the predicted [001]/[110]/[111] anisotropy.

Watch

Extended reading notes

Core claim

The central claim is that the minimum-energy paths for polaron transfer, and therefore the activated barriers and hopping mobilities, can be obtained variationally in reciprocal space without constructing supercells. The authors combine the ab initio polaron equations with the simplified string method, representing the path by phonon deformation fields that determine the optimal electronic polaron via constrained minimization. This yields optimized adiabatic transition states with lower barriers than linear interpolation: for rutile TiO2 the [001] barrier drops from 12 meV to 4 meV, and the resulting room-temperature mobility of 0.446 cm2/V·s agrees with experiment, whereas iterative Boltzma

Load-bearing premise

The load-bearing premise is that a polaron that spreads out across the whole simulation cell at the transition state is not a valid adiabatic hopping event and can be discarded; for the [110] and [111] directions in rutile TiO2, the reported anisotropic mobilities change if that premise fails, and the convergence of the localized path with cell size is not fully demonstrated beyond the two largest meshes used.

Editorial extensions

If this is right

  • In rutile TiO2 the adiabatic hopping mobility computed from optimized paths agrees with experiment, while band-like Boltzmann transport substantially overestimates it, indicating that self-trapped carriers demand a hopping description.
  • Optimizing the energy path matters: linear interpolation overestimates activation barriers (e.g., TiO2 [001] from 12 meV to 4 meV) and would underestimate mobilities.
  • The method resolves anisotropic and multi-step hopping mechanisms: for the LiF hole polaron, inter-site translation through a two-centre transition state (barrier about 438 meV) is rate-limiting, while on-site rotation is almost barrier-less (20 meV).
  • The approach yields direction-resolved mobilities, identifying [001] as the dominant electron-polaron channel in rutile TiO2, with [110] and [111] channels requiring larger simulation meshes to remain localized.
  • The effective attempt frequency can be computed as a phonon-weighted average at the initial polaron configuration, rather than resorting to a single longitudinal-optical mode.

Reading between the lines

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

  • Because the framework avoids supercells and uses reciprocal-space gradients, it could be extended to all-coupling variational polaron theories, potentially covering weak- and intermediate-coupling regimes where the present strong-coupling adiabatic approximation is less reliable.
  • The reported [110] and [111] mobilities rest on discarding transition states where the polaron delocalizes across the simulation cell; a natural test is to include those delocalized paths in a rate theory or compare with diagrammatic Monte Carlo at lower temperatures.
  • The crossover temperature T* = ΔEa/kB is about 45 K for the TiO2 [001] channel, predicting a measurable negative-to-positive temperature dependence of mobility that oriented single-crystal experiments could confirm.
  • The same variational string method could be applied to hole polarons, exciton polarons, or defect-bound polarons in other oxides and halides, where the formalism only needs the electronic eigenvalues, phonon dispersions, and electron-phonon matrix elements.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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 introduces a supercell-free approach to polaron hopping transport by combining the variational ab initio polaron equations with the simplified string method. The string method is used to optimize paths in the space of phonon coefficients B between self-trapped polaron states; along the path, the electronic coefficients are relaxed, and the resulting minimum-energy paths (MEPs) are used to extract activation energies, attempt frequencies, adiabatic transfer rates, and hopping mobilities. Applications to LiF (hole and electron polarons) and rutile TiO2 (electron polaron) are presented. The authors report multi-step hopping in LiF, near-barrierless rotation of the LiF hole polaron, anisotropic MEPs in TiO2, and a [001] electron mobility in rutile that agrees with experiment, while iBTE overestimates it.

Significance. The methodological proposal is significant and timely: it extends a variational reciprocal-space polaron formalism to transition-state search, avoiding supercell artifacts and providing explicit gradients. The paper includes code availability in ABINIT, materials-cloud data, and comparisons with MLMD, DMC, and experiment. If the transition paths are valid, the approach offers a scalable route to polaron-hopping mobility. However, two issues are load-bearing: the string method is implemented in B space with an unphysical Euclidean metric, and the [110]/[111] TiO2 results depend on discarding delocalized transition states without demonstrated convergence. For these reasons, the central quantitative claims are not yet fully supported.

major comments (3)
  1. [IV.B–IV.C, Eqs. (3), (10), (13)–(14)] The string method is formulated and implemented in B space using the Euclidean norm for both the force evolution and the reparametrization. From Eq. (3), the physical mass-weighted normal coordinate is Q_{qν} ∝ B_{qν}/√ω_{qν}; the lattice kinetic energy is T = 1/2 Σ|Q̇|² = 1/2 Σ|Ḃ|²/ω_{qν}. The intrinsic reaction coordinate therefore uses the metric Σ|δB|²/ω_{qν}, not Σ|δB|². Eq. (10) should be written in mass-weighted coordinates, and the arc length in Eq. (13) should be correspondingly weighted. As written, the optimized curve is not the minimum-energy path on the physical pPES; intermediate images and multi-step mechanisms (e.g., Fig. 2b) are not guaranteed to have physical meaning. A converged maximum of the curve would still be a stationary point of E_pol, but the paper does not verify this, and the metric used may direct the string to a different saddle. This concern applies to al
  2. [II.C, Figs. 6–7, Table I] For the [110] and [111] channels in rutile TiO2, the localized MEPs are reported only for n=9 and n=11, respectively; smaller BvK meshes yield delocalized transition states and are discarded as outside adiabatic hopping theory. No systematic convergence of the localized MEP with mesh size is shown, and for [111] the parameters are taken from a single mesh without thermodynamic-limit extrapolation. Because the anisotropic mobility comparison in Fig. 8 and the room-temperature values in Table I use these channels, the exclusion of delocalized paths is load-bearing. The authors should (i) define a quantitative localization criterion for the transition state, (ii) report the energy and barrier of the delocalized paths (they appear in Fig. 6 but their barriers are not given), and (iii) test convergence of the localized MEP with increasing n, or explain why the n=9/n=11 results are converged.
  3. [II.C, Fig. 7] Activation energies are extrapolated to the thermodynamic limit by linear fits in N_p^{-1/3}; no uncertainty estimates or justification for the linear dependence are provided. For the [001] channel the extrapolated barrier is 3.86 meV, of the order of the scatter expected from such fits and much less than k_BT at room temperature. The agreement of the [001] mobility with experiment could therefore be sensitive to the extrapolation procedure. Please provide error bars or a more detailed convergence analysis, and clarify how many mesh points enter each extrapolation.
minor comments (4)
  1. [IV.C] The number of images N and the dependence of the converged barrier on N are not reported; please provide these convergence details for reproducibility.
  2. [II.C] The sentence 'The initial linearly-interpolated energy barriers, agree' contains a grammatical error and should be rewritten. Also, the text should clarify which polaron-density isosurface is used for the delocalized transition states in Fig. 6.
  3. [Data availability] The Materials Cloud Archive entry is marked 'draft'; a stable DOI should be provided in the published version.
  4. [IV.E, Eqs. (19)–(22)] The reorganization energy and electronic coupling are defined but not used in the reported rates and mobilities; the role of these quantities in the adiabatic regime should be stated explicitly, or their reporting should be justified.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the MEPs, barriers, and mobilities are computed from the variational polaron functional and string-method equations, with experiment used only as an external comparison.

full rationale

The derivation chain is self-contained. The polaron energy functional (Eq. 1) and its gradients (Eqs. 4-5) are stated explicitly, and the string-method evolution (Eqs. 10-14) is applied to these gradients without any fitted parameter. Activation energies and attempt frequencies are outputs of the same variational calculation; the experimental mobilities in Fig. 8 are used only as a posteriori benchmarks, not as constraints. The comparisons with DMC and MLMD are likewise independent external results. Self-citations [34] and [39] provide the underlying variational formalism and ground-state polaron solutions, but the present paper restates the functional and computes all inputs (ε_nk, ω_qν, g_mnν) from standard DFT/DFPT, so those citations are background rather than load-bearing. The main caveats—the use of an unweighted L2 metric in B-space rather than the 1/ω-mass-weighted metric, the exclusion of delocalized transition states for the [110] and [111] directions in Sec. II.C, and the model-dependent definition of νeff—are physical/correctness concerns, not circular reductions: none of these steps makes a predicted quantity equal to an input by construction. Hence no circularity score above 0 is warranted.

Assumptions & free parameters 2 free parameters · 6 assumptions · 0 invented entities

The paper introduces no new physical entities. Its central claim rests on the variational polaron energy functional (from prior work by the same group) and on several modeling choices: strong-coupling adiabatic approximation, linear EPC, DFT XC functionals, linear extrapolation in N_p^{-1/3}, and the discarding of delocalized transition states. The largest burden is the last choice, which is needed for the anisotropic TiO2 mobilities.

free parameters (2)
  • Effective attempt frequency (ν_eff) = e.g., 15.69 THz (LiF), 13.08 THz (LiF e-), ~3.5-15.7 THz (TiO2 channels)
    Defined by Eq. (24) as a phonon-weighted average of |B0,qν|^2 ω_qν. This modeling choice scales the mobilities linearly; the authors could have used ν_LO instead.
  • Thermodynamic-limit extrapolation intercepts = reported ΔE_a values (e.g., 437.76 meV LiF h+, 3.86 meV TiO2 [001])
    The final activation energies are obtained by a linear fit of computed values vs N_p^{-1/3}; the slope and intercept are fitted to computed data points.
assumptions (6)
  • domain assumption Strong-coupling adiabatic (Landau-Pekar) approximation
    The central energy functional (Eq. 1) rests on this approximation; lattice fluctuations are neglected and the charge adjusts instantaneously to the lattice (Sec. IV.A). The authors acknowledge it limits validity for weakly bound polarons (e.g., LiF electron).
  • domain assumption Linear electron-phonon coupling truncation
    The variational polaron equations treat the deformation to first order; second-order EPC and anharmonicity are neglected (Discussion).
  • domain assumption DFT exchange-correlation functionals (PBE for LiF, LDA for TiO2)
    The electronic structure and electron-phonon matrix elements come from these functionals (Sec. IV.F); LDA is used because PBE gives imaginary phonons in rutile.
  • ad hoc to paper Locality of adiabatic transition states
    The paper discards MEPs in which the polaron becomes delocalized at the transition state because they fall outside adiabatic hopping theory (Sec. II.C). This selection is essential for the reported [110]/[111] anisotropies.
  • ad hoc to paper Linear extrapolation of energies in N_p^{-1/3}
    Activation and binding energies are linearly extrapolated to the thermodynamic limit using N_p^{-1/3} (Figs. 3,5,7); this functional form is not derived and lacks confidence intervals.
  • domain assumption Adiabatic Arrhenius rate expression
    The hopping rate is k = ν exp(-ΔEa/kBT) from MEHAM theory, assuming adiabatic transfer and neglecting tunneling (Sec. IV.E).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Optimal transition states for polaron hopping transport without supercells." pith.science (2026). https://pith.science/paper/THM4DI3E

@misc{pith2026260718096,
  author       = {Pith},
  title        = {Pith review of: Optimal transition states for polaron hopping transport without supercells},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/THM4DI3E}},
  note         = {Machine review of arXiv:2607.18096}
}
abstract

Polaron formation localizes charge carriers and drives a crossover from band-like to hopping transport in materials. Hopping dynamics can be obtained from DFT supercell calculations of transition states, but these suffer from polaron self-interaction, spurious electrostatics, and poor scaling with polaron size. We introduce a supercell-free framework for ab initio polaron hopping transport based on the ab initio polaron equations formalism, its variational formulation, and the string method. The approach optimizes transition states between self-trapped polaron states directly in reciprocal space and provides the polaron configurations along the path, enabling evaluation of adiabatic hopping rates and mobilities. We apply the method to LiF and rutile TiO$_2$, revealing multi-step and anisotropic hopping mechanisms. In rutile TiO$_2$, the computed electron-polaron mobility agrees with experiment, whereas band-like Boltzmann transport substantially overestimates the mobility. Our results establish a scalable route to first-principles polaron-hopping dynamics in materials in which charge motion is governed by self-trapping.

Figures

Figures reproduced from arXiv: 2607.18096 by the authors.

Figure 1
Figure 1. FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 8
Figure 8. Figure 8: The reference experiments are based on resistivity [PITH_FULL_IMAGE:figures/full_fig_p006_8.png]
Figure 6
Figure 6. Figure 6: FIG. 6 [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7 [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8 [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 11
Figure 11. Figure 11: The initial and final hopping states are treated [PITH_FULL_IMAGE:figures/full_fig_p010_11.png]
Figure 11
Figure 11. Figure 11: shows the adiabatic and diabatic transition paths. By construction, however, the variational po￾laron equations formalism assumes the adiabatic strong￾coupling approximation. Therefore, only adiabatic tran￾sitions are described consistently within the present ap￾proac…
Figure 10
Figure 10. Figure 10: FIG. 10 [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references

  1. [1]

    orientation y z Intermediate state y z Transition state

  2. [110]

    orientation x y Final state

  3. [111]

    We also compute the mobility using the iterative Boltz- mann transport equation (iBTE) framework 15, as imple- mented in theABINITsoftware package 50–52

    directions,µ p approaches the measured values at higher temperatures,T >300 K, indicating thermally activated hopping. We also compute the mobility using the iterative Boltz- mann transport equation (iBTE) framework 15, as imple- mented in theABINITsoftware package 50–52. The re- sulting values, shown in Fig. 8, overestimate both the hopping mobilities an...

  4. [4]

    orientation x y Transition state (rotation)

  5. [8]

    orientation x y Start Hopping End Reaction coordinate (-) 0.0 0.2 0.4 0.6 0.8Energy (meV) Ea c LiF (e ) Hopping Initial state x y Transition state x y Final state x y FIG. 2.Polaron dynamics in LiF.Rowsa,bandccorrespond to: hole polaron on-site [001]→[010] rotation; hole polaron hopping along the [110] direction accompanied by on-site [001]→[100] rotation...

  6. [9]

    However, due to the low activation barrier, the mobility in this direc- tion switches from positive to negative temperature de- pendence at ∆E [001] a /kB ≈45 K

    transfer is the most favourable, with a transfer rate ofk p = 13.54 THz, which is one order of magnitude higher than those of the other transitions. However, due to the low activation barrier, the mobility in this direc- tion switches from positive to negative temperature de- pendence at ∆E [001] a /kB ≈45 K. At room temperature, its value isµ p = 0.446 c...

  7. [10]

    orientation y z Start Hopping End Reaction coordinate (-) 0.0 0.2 0.4 0.6 0.8 Energy (eV) Ea b LiF (h+ ) Hopping Initial state

  8. [11]

    orientation y z Intermediate state y z Final state

Show all 14 references
  1. [12]

    A similar behaviour is observed for the other directions, where the mobility increases with temperature

    direction follows the same temperature trend in our calculations and in DMC, although the latter gives slightly larger values. A similar behaviour is observed for the other directions, where the mobility increases with temperature. The discrepancy with DMC may be attributed to...

  2. [13]

    hop 3.86 2.918 45 13.54 4.46×10 −1 1.28 TiO2 (e−) [110] hop 80.86 15.72 6.428 938 0.69 1.10×10 −1 2.95×10 −1

  3. [14]

    It can also help to assess when conventional Boltzmann transport becomes inappropri- ate

    hop 78.01 3.529 905 0.77 3.31×10 −2 8.90×10 −2 fers a way to compute direction-resolved hopping chan- nels, compare competing transitions, and identify rate- limiting processes. It can also help to assess when conventional Boltzmann transport becomes inappropri- ate. Because t...

  4. [15]

    be- yond quasiparticles

    directions, at leastn= 9 andn= 11, respectively, are required to obtain localized transfer. Second and third panels show the computed adiabatic transfer ratesk p and mobilities µp. 150 200 250 300 350 400 T (K) 10 1 101 103 (cm2/V s) TiO2 (e ), mobility p [001] p [110] p [111]...

    2021

  5. [100]

    orientation x y Transition state (hopping)

  6. [101]

    orientation x y Intermediate state

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.