REVIEW 3 major objections 5 minor 8 references
Large tuning of the optical properties of nanoscale NdNiO3 via electron doping
T0 review · 3 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash
Pith's one-line read Electron doping switches NdNiO3 from a metal to a 3–4 eV band-gap insulator with low optical loss.
desk verdict First measured optical constants of electron-doped NdNiO3, with a caveat on the two-layer model; worth a serious referee. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central mechanism is a two-layer ellipsometric model: a top fully Li-doped insulating layer with an opened band gap sits over a bottom undoped metallic layer, and the pair is fit to variable-angle spectroscopic ellipsometry data using Drude and Lorentz or Gaussian oscillators. Depth-resolved ToF-SIMS measurements showing an exponentially decaying lithium concentration justify the model, since prior first-principles work indicates the gap opens sharply only near full electron doping. The model is what extracts the refractive index and extinction coefficient of each phase together with the layer thicknesses, and Tauc plots of the extracted absorption are then used to estimate the band gap.
What would settle it
Intercalate a film thin enough, roughly below 10 nanometers, that ToF-SIMS shows lithium throughout the full thickness, then measure its ellipsometric Psi and $\Delta$ and fit n and kappa directly with a single layer; if those optical constants differ from the two-layer values reported here, the abrupt-interface assumption is the source of bias. Equivalently, fit the existing data with a graded-index multilayer constrained to the measured SIMS profile and check whether the fit improves and the top-layer constants shift.
Extended reading notes
Core claim
The paper's central claim is that field-driven lithium intercalation switches nanoscale NdNiO3 from a metallic state with no optical band gap into an insulating state with a gap of roughly 3 to 4 eV, and that this switch produces a large, low-loss change in the complex refractive index across 0.3 to 2.5 micrometers. From ellipsometric fits, the undoped phase is metallic, with the extinction coefficient rising with wavelength, while the Li-doped phase has $n \approx 2$, $\kappa < 0.1$ from 0.4 to 0.7 micrometers, and $\kappa < 0.02$ from 0.7 to 2.5 micrometers. Because time-of-flight secondary-ion mass spectrometry shows that the lithium profile is a gradient, the authors extract these values with a two-layer model rather than assuming uniform doping, and they verify the doped film's transparency visually on transparent LaAlO3 substrates. Tauc analysis places the gap at 2.9 to 3.2 eV assuming an indirect transition or 3.7 to 4.3 eV assuming a direct transition, which the paper summarizes as roughly 3 to 4 eV.
Load-bearing premise
The load-bearing assumption is that the exponential lithium concentration gradient can be modeled as two abrupt layers, a fully doped insulator on top and an undoped metal below, with no partially doped region in between; if intermediate doping states matter, the extracted refractive indices, extinction coefficients, and 3 to 4 eV band gap could be biased.
Editorial extensions
If this is right
- If the extracted constants are correct, NdNiO3 becomes a visible and near-infrared optical phase-change material whose refractive index changes substantially on lithium intercalation while the doped phase stays low-loss.
- The reported gap of roughly 3 to 4 eV is consistent with several first-principles predictions for hydrogen- or lithium-doped rare-earth nickelates, suggesting the electronic mechanism is not specific to one ionic species.
- Because the doping is field-driven and persists at room temperature, NdNiO3-based optical devices could be tuned electrically without heating and without continuous power.
- The two-layer analysis strategy should extend to other nanoscale materials with inhomogeneous ion or electron concentration profiles, where uniform-doping fits would fail.
- The low extinction coefficient of the doped phase across the near infrared opens a route to electrically switchable optical components at telecommunication wavelengths.
Reading between the lines
- Editorial inference: if the central claim holds, a partially doped film is effectively a graded-index layer, so controlling intercalation time or voltage could yield continuously tunable refractive index rather than only the two endpoint states.
- Editorial inference: the combination of large refractive-index contrast and $\kappa < 0.02$ in the near infrared suggests a testable extension to electrically switchable phase shifters or tunable resonators, for example by measuring the resonance shift of a doped NdNiO3 metasurface.
- Editorial inference: the reproducible exponential lithium profile could itself be treated as a programmable gradient-index coating, with its depth shape set by doping time and voltage.
- Editorial inference: the gap estimate depends on choosing an indirect or direct Tauc exponent, so a measurement on a uniformly doped film using angle-resolved photoemission or direct absorption would settle which assignment is correct.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript reports synthesis of crystalline NdNiO3 films, reversible Li-ion intercalation from a liquid electrolyte, and variable-angle spectroscopic ellipsometry (VASE) characterization of both undoped and Li-doped regions over 0.3-2.5 um wavelength. Because ToF-SIMS shows a Li concentration gradient, the authors use a two-layer model (fully doped top layer / undoped or lightly doped bottom layer) to extract the complex refractive indices of the two phases. They report that the undoped film is metallic, while the Li-doped insulator has n near 2, kappa below 0.1 in the visible and below 0.02 in the near-infrared, and Tauc analysis yields an optical gap of roughly 3-4 eV. The same qualitative behavior is reproduced on two substrates (Nb:STO and LAO).
Significance. The qualitative finding is credible and potentially valuable: electron doping of NdNiO3 produces a transparent insulating state, and the reported index contrast between the metallic and insulating phases is large. The paper is transparent about the doping gradient and provides detailed oscillator parameters in Table S1, which aids reproducibility. If the quantitative n, kappa, and band-gap values hold, this work demonstrates electrically controlled, non-volatile, low-loss refractive-index tuning in a perovskite nickelate. However, the central quantitative claims rest on the two-layer ellipsometric model and on a single-Lorentz-oscillator representation of the doped layer, so the uncertainty in those values needs to be addressed before the results can be taken at face value.
major comments (3)
- [Section 2 (Fig. 2e and Note S1)] The two-layer model for the Nb:STO sample fixes the bottom layer as undoped NdNiO3 and fits the top layer as fully doped, while the ToF-SIMS profile in Fig. 2e shows an exponential Li decay. The paper itself acknowledges in the Fig. S1 note that intermediate doping states are not modeled. Because Ref. 23 reports a gap of only about 1.4 eV at half doping in H-doped SmNiO3, partially doped material is not optically equivalent to undoped metallic NdNiO3. Consequently the fitted top-layer n and kappa, and the Tauc gap derived from them, can partially absorb the missing intermediate-doping region. Please quantify this systematic error by fitting a graded-layer or multi-layer profile (e.g., 3-5 layers with intermediate oscillator sets) and reporting the resulting range of n, kappa, and gap values.
- [Table S1 and Fig. 4] The doped layer is represented by a single Lorentz oscillator, and the Tauc plots in Fig. 4 are computed from the absorption coefficient of that same fitted oscillator. A Lorentzian model imposes a specific absorption tail, so the linear extrapolation in the (alpha h nu)^0.5 plot may be largely determined by the oscillator center energy and broadening rather than by an independently measured electronic band edge. Please demonstrate robustness of the 2.9-3.2 eV (indirect) and 3.7-4.3 eV (direct) estimates by refitting with at least one alternative dispersion model (e.g., a Tauc-Lorentz model or a numerical inversion with Kramers-Kronig consistency) and by showing how much the extracted intercept changes.
- [Sections 2 and 3 (Note S1)] No figure-of-merit or confidence intervals are reported for the ellipsometric fits. The top-layer thickness and its optical constants are fitted simultaneously for the Nb:STO sample, and the LAO analysis only varies the bottom-layer thickness from 1 to 5 nm. The consistency between Nb:STO and LAO is a useful internal check, but both samples are analyzed with the same two-layer approximation, so it does not isolate the systematic model error. Please provide typical ellipsometric MSE values, parameter correlation information, or an explicit covariance estimate for the reported n and kappa.
minor comments (5)
- [Figure S1c] The panel label reads "60 s doping," but the text and Fig. S1b describe samples doped for 120 and 180 s; please correct the label.
- [Figure 4c] The legend entries "Ref. [23] [23][16]" are difficult to parse; please separate the references and align them clearly with the plotted symbols.
- [Note S1 and Table S1] The LAO substrate oscillator parameters list Br in eV while also giving wavenumbers in parentheses; please state the conversion used and keep units consistent throughout the table.
- [Page 5, undoped NdNiO3 fitting] The text says the single-layer model shows good agreement with the data, but no quantitative fit-quality metric is given; please add the MSE or equivalent for the undoped fits as well.
- [Tauc plot discussion (Section 4)] The argument that the material differs from Co3O4 and CoxFe3-xO4 because absorption is substantial only below 400 nm is not by itself a test of whether the relevant transition is from the valence band maximum to the conduction band minimum; please support this claim with a comparison to the calculated band structure or at least state it as a model assumption.
Circularity Check
No significant circularity: the optical constants are fitted to raw ellipsometry and the Tauc band gap is a post-hoc transform of those fitted constants; cited prior theory is external and independent.
full rationale
The derivation chain is: ToF-SIMS establishes a Li gradient; a two-layer ellipsometric model (top fully doped insulator, bottom undoped metal) is fitted directly to measured Ψ and Δ spectra; the fitted top-layer n and κ are then converted to α and analyzed with Tauc plots to estimate Eg. No fitted parameter is renamed as a prediction: Eg is not one of the fitted oscillator parameters, and the Tauc extrapolation is an independent post-processing step applied to the measured complex refractive index. The two-layer approximation is a modeling assumption, but it is justified by the measured gradient plus an external first-principles calculation (ref. 23, Yamauchi & Hamada) that shows a steep gap opening near full doping; that citation is not self-referential and does not embed the paper's own result. Refs. 14 and 18 do overlap with co-author Ramanathan, but they are used only as background and for comparison of SmNiO3 band gaps, not as load-bearing support for the NdNiO3 extraction. The undoped optical constants are measured directly on an adjacent region, and the LAO sample provides a consistency check with a transparent substrate. The acknowledged simplifications (ignoring intermediate doping states, SIMS counts vs. sputter time, no uniqueness analysis) are correctness and robustness concerns, not circularity: the reported n, κ, and gap are not equal to the model inputs by construction. No circular step can be exhibited with a specific equation or fitted-parameter renaming, so the appropriate score is 0.
Assumptions & free parameters
free parameters (5)
- Undoped NdNiO3 oscillator parameters (Drude and Lorentz) =
Drude strength 9.9825 eV, broadening 1.0518 eV; Lorentz parameters in Table S1
- Doped NdNiO3 oscillator parameters (single Lorentz) =
Center 4.3961 to 4.4246 eV, amplitude about 3.3 to 4.4 eV; Table S1
- Two-layer thicknesses (Nb:STO sample) =
20.7 nm doped / 27.3 nm undoped
- Two-layer thicknesses (LAO sample) =
47.5 nm doped / 3 nm lightly doped
- Substrate oscillator parameters (Nb:STO and LAO) =
Table S1 entries
assumptions (4)
- domain assumption Drude-Lorentz-Gaussian-Cauchy oscillator models can represent the dielectric functions of NdNiO3, Nb:STO, and LAO over 0.3 to 2.5 micrometers.
- domain assumption The doped film can be approximated as two uniform layers (fully doped top, undoped or lightly doped bottom) despite the continuous Li gradient measured by ToF-SIMS.
- domain assumption The Tauc plot method is valid for Li-doped NdNiO3.
- domain assumption The band gap opens only near full electron doping, so the top layer has a fully opened gap and the bottom layer has minimal gap.
Cite this review
Pith. "Pith review of Large tuning of the optical properties of nanoscale NdNiO3 via electron doping." pith.science (2026). https://pith.science/paper/JL5DUDBY
@misc{pith2026241115679,
author = {Pith},
title = {Pith review of: Large tuning of the optical properties of nanoscale NdNiO3 via electron doping},
year = {2026},
howpublished = {\url{https://pith.science/paper/JL5DUDBY}},
note = {Machine review of arXiv:2411.15679}
}
read the original abstract
We synthesized crystalline films of neodymium nickel oxide (NdNiO3), a perovskite quantum material, switched the films from a metal phase (intrinsic) into an insulator phase (electron-doped) by field-driven lithium-ion intercalation, and characterized their structural and optical properties. Time-of-flight secondary-ion mass spectrometry (ToF-SIMS) showed that the intercalation process resulted in a gradient of the dopant concentration along the thickness direction of the films, turning the films into insulator-metal bilayers. We used variable-angle spectroscopic ellipsometry to measure the complex refractive indices of the metallic and insulating phases of NdNiO3. The insulator phase has a refractive index of n ~ 2 and low absorption in the visible and near infrared, and analysis of the complex refractive indices indicated that the band gap of the insulating phase is roughly 3-4 eV. Electrical control of the optical band gap, with corresponding large changes to the optical refractive indices, creates new opportunities for tunable optics.
Figures
Figures from the paper (1 more)
Reference graph
Works this paper leans on
-
[1]
(1) Ko, J. H.; Yoo, Y . J.; Lee, Y .; Jeong, H.-H.; Song, Y . M. A review of tunable photonics: Optically active materials and applications from visible to terahertz. IScience 2022, 25 (8). (2) Veeder, T.; Dehzangi, A.; Ramanathan, S.; Kats, M.; Yu, N.; Hu, J.; Roberts, C.; Polking, M.; Tibbetts, K.; Majumdar, A.; et al. Accelerating discovery of tunable ...
work page 2022
-
[3]
Physical Review Letters 2011, 107 (17), 176401. (13) Shahsafi, A.; Roney, P.; Zhou, Y .; Zhang, Z.; Xiao, Y .; Wan, C.; Wambold, R.; Salman, J.; Yu, Z.; Li, J.; et al. Temperature-independent thermal radiation. Proceedings of the National Academy of Sciences 2019, 116 (52), 26402-26406. (14) Shi, J.; Zhou, Y .; Ramanathan, S. Colossal resistance switching...
work page 2011
-
[5]
The refractive indices of the Li -doped NdNiO 3 on a LaAlO 3 substrate obtained from the ellipsometry results (Figure 3e in the main text), with different thickness combinations of the top doped and bottom lightly doped NdNiO3. For the ellipsometry data analysis, we enforced the thickness of the bottom lightly doped layer from 1 to 5 nm, with intervals of...
-
[6]
(24) Klein, J.; Kampermann, L.; Mockenhaupt, B.; Behrens, M.; Strunk, J.; Bacher, G
Physical Review B 2023, 108 (4), 045108. (24) Klein, J.; Kampermann, L.; Mockenhaupt, B.; Behrens, M.; Strunk, J.; Bacher, G. Limitations of the Tauc plot method. Advanced Functional Materials 2023, 33 (47), 2304523. 1 Supporting information: Large tuning of the optical properties of nanoscale NdNiO3 via electron doping Yeonghoon Jin1,#, Teng Qu2,#, Siddh...
work page 2023
-
[7]
(a) Complex refractive index of the Li-doped NdNiO3 with doping times of 120 and 180 s
Optical properties of two Li-doped NdNiO 3 films on Nb:STO (0.5% Nb- doped SrTiO3) with different doping times. (a) Complex refractive index of the Li-doped NdNiO3 with doping times of 120 and 180 s. (b) Illustrations of the samples with the thickness estimated by ellipsometry. Doping was conducted on one chip but at different spots, and therefore the ent...
work page 2000
-
[2676]
(20) Zhang, H.-T.; Park, T. J.; Islam, A. N.; Tran, D. S.; Manna, S.; Wang, Q.; Mondal, S.; Yu, H.; Banik, S.; Cheng, S.; et al. Reconfigurable perovskite nickelate electronics for artificial intelligence. Science 2022, 375 (6580), 533-539. (21) Hauser, A. J.; Mikheev, E.; Moreno, N. E.; Hwang, J.; Zhang, J. Y .; Stemmer, S. Correlation between stoichiome...
work page 2022
-
[4279]
Vanadium dioxide for dynamically tunable photonics
(9) Ko, B.; Badloe, T.; Rho, J. Vanadium dioxide for dynamically tunable photonics. ChemNanoMat 2021, 7 (7), 713-727. (10) Lacorre, P.; Torrance, J.; Pannetier, J.; Nazzal, A.; Wang, P.; Huang, T. Synthesis, crystal structure, and properties of metallic PrNiO3: Comparison with metallic NdNiO3 and semiconducting SmNiO3. Journal of Solid State Chemistry 199...
work page 2021
-
[4860]
(15) Cui, Y .; Liu, X.; Fan, W.; Ren, J.; Gao, Y . Metal–insulator transition in RNiO3 (R= Pr, Nd, Sm, Gd, Tb, Dy, Ho, Er) induced by Li doping: A first-principles study. Journal of Applied Physics 2021, 129 (23). (16) Yoo, P.; Liao, P. First principles study on hydrogen doping induced metal-to-insulator transition in rare earth nickelates RNiO3 (R= Pr, N...
work page 2021
Reviewed August 12, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.