Pith. sign in

REVIEW 4 major objections 4 minor 31 references

Broad Spectral Tuning of Ultra-Low Loss Polaritons in a van der Waals Crystal by Intercalation

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Pith's one-line read Intercalating sodium into α-V2O5 shifts its phonon-polariton band by about 30 cm−1 while lifetimes stay in the picosecond range.

desk verdict Intercalation shifts the Reststrahlen band of α-V2O5 by ~30 cm-1 while preserving picosecond phonon polariton lifetimes—a real result, but the lifetime comparison is not yet airtight. read the letter →

arxiv 2501.08705 v1 pith:7EWJ7IJP submitted 2025-01-15 physics.optics cond-mat.mtrl-sci

classification physics.opticscond-mat.mtrl-sci
keywords phononpolaritonsvanderWaalscrystalsintercalationReststrahlenbandalpha-V2O5s-SNOMnanoFTIRmid-infrarednanophotonics
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 establishes that intercalating sodium atoms into the layered semiconductor α-V2O5 is a workable route to spectrally shift its Reststrahlen bands, and therefore the phonon polaritons inside them, without destroying their low-loss character. If correct, it gives a concrete method for placing phonon-polariton devices at frequencies the parent crystal cannot reach, simply by choosing the intercalant and its concentration. The paper supports the claim with near-field images and nanoFTIR line scans showing the first band moving roughly 30 cm−1 red, with ab initio phonon calculations that reproduce the shift, and lifetime measurements of 4±1 ps in the intercalated crystal versus 6±1 ps in pristine α-V2O5.

What carries the argument

The central object is the Na-intercalated van der Waals crystal α'-(Na)V2O5, in which sodium atoms sit between the weakly bound layers and modify the phonon modes and effective charges, shifting the Reststrahlen bands. The experimental machinery is s-SNOM and nanoFTIR polariton interferometry, where tip-launched and antenna-launched polaritons interfere with back-reflected fields; the fringe period gives the polariton wavelength and dispersion, while the fringe decay, fitted with an exponentially damped sine wave corrected by a 1/√x geometric spreading factor, gives the propagation length. The propagation length is converted into a lifetime using the measured group velocity via τ = L/vg.

What would settle it

Measure the same α'-(Na)V2O5 flakes at several thicknesses and frequencies with a different technique that extracts the complex permittivity directly, such as FTIR ellipsometry, and compute the lifetime from that permittivity; if the resulting lifetime falls well below the picosecond range or differs markedly from pristine α-V2O5, the central claim fails.

Watch

Extended reading notes

Core claim

The paper demonstrates that intercalating Na atoms between the van der Waals layers of α-V2O5 red-shifts the first Reststrahlen band by about 30 cm−1, roughly 60% of the initial band width, and that the phonon polaritons supported in the resulting α'-(Na)V2O5 crystal remain anisotropic and ultra-low-loss, with a measured lifetime of 4±1 ps compared with 6±1 ps in pristine α-V2O5. It further shows, through first-principles phonon calculations feeding a Lorentz-oscillator permittivity model, that the other two Reststrahlen bands are also modified by intercalation, explaining why they become nearly invisible in the measurements. The central conclusion is that intercalation can be used as a broad spectral tuning mechanism for phonon polaritons while preserving their low-loss propagation.

Load-bearing premise

The central claim leans on converting near-field fringe decay into a lifetime through a chosen damping parameter and a damped-sine model; if that conversion is not right, the 4±1 ps value and the 'intercalation preserves low losses' conclusion do not follow.

Editorial extensions

If this is right

  • α-V2O5 supports anisotropic phonon polaritons in three Reststrahlen bands, with elliptic in-plane propagation in RB1 and hyperbolic behavior in RB2 and RB3.
  • Intercalating Na atoms red-shifts RB1 by roughly 30 cm−1, moving the polariton dispersion to lower frequencies, with ab initio calculations giving a similar larger shift of about 50 cm−1.
  • Phonon polaritons in α'-(Na)V2O5 remain anisotropic and low-loss, with a measured lifetime of 4±1 ps and quality factor Q≈2.5, close to the pristine values of 6±1 ps and Q≈3.5.
  • If the method generalizes to other ions and layered hosts, the spectral position of phonon polaritons could be chosen by intercalation chemistry rather than by choosing a different material.

Reading between the lines

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

  • The same intercalation route could be tried on other polar van der Waals crystals, such as α-MoO3, where the variety of intercalant ions and concentrations may tune the Reststrahlen bands; the paper states this as an expectation, not a demonstrated result.
  • The 30 cm−1 experimental shift versus roughly 50 cm−1 in calculation leaves room for testing how sodium concentration, staging, or strain adjusts the TO–LO splitting, which could refine the model.
  • Because α-V2O5 can also be electrically doped, combining intercalation with electrostatic gating could turn the static spectral shift into a dynamically switchable polariton frequency; this is an implication beyond the paper's demonstrated static tuning.
  • A direct time-domain measurement of the polariton decay, rather than fringe-decay fitting, would be the cleanest check of whether the 4±1 ps lifetime is intrinsic or partly an artifact of the extraction model.
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

4 major / 4 minor

Summary. The paper reports that Na intercalation of the van der Waals semiconductor α-V2O5 (forming α'-(Na)V2O5) shifts the mid-infrared Reststrahlen band RB1 to lower frequencies by about 30 cm^-1 (roughly 60% of the initial RB width), as directly observed by nanoFTIR line scans and s-SNOM dispersion measurements. The authors also report that phonon polaritons in the intercalated crystal remain ultra-low-loss, with a lifetime of 4 ± 1 ps inferred from antenna-launched fringe decay, compared with 6 ± 1 ps in pristine α-V2O5. Ab initio phonon calculations reproduce the direction of the RB shift and are used to explain the disappearance of RB2 and RB3 in the intercalated material.

Significance. If the central claims hold, the paper demonstrates a broadly useful tuning mechanism for phonon polaritons in van der Waals crystals, addressing a recognized limitation of these materials. The strengths of the paper are the direct near-field observation of the RB shift, the use of nanoFTIR to map the spectral position of the bands, the mapping of in-plane anisotropic propagation, and the ab initio calculations that corroborate the shift direction. However, the 'ultra-low losses preserved' claim is less strongly supported: it rests on a single lifetime comparison between flakes of different thickness measured at different frequencies, and the lifetime extraction uses a damping parameter that is itself tuned to reproduce the experimental lifetimes. The RB-shift result is more robust, but the quantitative comparison between the measured (~30 cm^-1) and calculated (~50 cm^-1) shifts needs tightening.

major comments (4)
  1. [Fig. 5 and 'Finally, we study and compare...' paragraph] The claim that intercalation preserves ultra-low losses rests on a single comparison between a 130-nm-thick α-V2O5 flake measured at 1010 cm^-1 and a 107-nm-thick α'-(Na)V2O5 flake measured at 973 cm^-1. For slab phonon polaritons, the fringe decay length and the conversion τ = L/v_g depend on slab thickness (through confinement and substrate losses) and on frequency through v_g(ω). The 2 ps difference with overlapping error bars therefore does not isolate the effect of Na intercalation; a larger intercalation-induced loss increase could be hidden by the thickness and frequency offsets. Please provide like-for-like measurements (same thickness and frequency, or a quantitative model that corrects for these differences) before claiming that intercalation preserves ultra-low losses.
  2. [Methods, 'Dielectric function of α-V2O5'] The lifetime extraction is partly circular. The Methods state that γ_z = 1.5 cm^-1 is adopted 'to better fit the experimental lifetime values', and the group velocities used in τ = L/v_g are obtained from the same Lorentz-oscillator permittivity. Thus the reported 4 ± 1 ps and 6 ± 1 ps lifetimes are not an independent measurement of loss. Please compute the group velocity from the unadjusted permittivity (γ_z = 2.0 cm^-1) or directly from the measured dispersion, and report whether the conclusion changes.
  3. [Fig. 4 and 'To better understand the polaritonic effects...' paragraph] The ab initio calculation gives a center-to-center RB'1 shift of about 50 cm^-1, while the measured shift is about 30 cm^-1. The text calls this 'good agreement', but no uncertainty is given for the measured RB edges and the 20 cm^-1 discrepancy is not discussed. Since the tunability claim is quantified by this shift, please provide a quantitative comparison with error bars and a brief discussion of possible sources of the discrepancy (e.g., Na concentration, anharmonicity, or approximations in the Lorentz model).
  4. [Fig. 3 and 'We note that the nanoFTIR image in Fig. 3b...' paragraph] The absence of RB'2 and RB'3 in the intercalated crystal is explained after the fact by the calculated narrowing and spectral shifts of these bands. Because the same experimental data are used to infer both the shift of RB'1 and the absence of the other bands, an independent check is needed to rule out that the missing bands are due to the specific flake thickness, the limited spectral window, or an inhomogeneous intercalation. A wider-range nanoFTIR scan or measurements on a second flake of different thickness would strengthen this point.
minor comments (4)
  1. [Fig. 3 caption and main text] There is a figure-callout inconsistency: the text refers to 'Figure 3b' for the nanoFTIR scans of the intercalated flake, but in the figure caption panel b is the pristine α-V2O5 scan and panel d is the α'-(Na)V2O5 scan. Please correct the callouts.
  2. [Methods, 'Dielectric function of α-V2O5'] The displayed Drude-Lorentz equation is poorly typeset, with superscripts and subscripts garbled. Please ensure the equation is rendered with clear indices for the three axes.
  3. [Abstract and introduction] The phrases 'ultra-low losses' and 'ultra-long lifetime' are used for lifetimes of 4–6 ps that are inferred through a model with a tuned damping parameter; consider softening the wording or adding a caveat that the lifetimes are model-derived.
  4. [Full text, 'similar thicknesses'] In the lifetime-comparison paragraph, the two flakes are described as having 'similar thicknesses of 130 and 107 nm'. A 23 nm difference is not negligible for slab polaritons; please either use the same-thickness data or justify the approximation quantitatively.

Circularity Check

1 steps flagged · score 2.0 of 10

Minor self-citation and post-hoc γ_z calibration; central RB-shift claim is independent.

  1. other [Methods, 'Dielectric function of α-V2O5' (γ_z value)]
    "The values of γ_a were taken from ref. [15]. γ_z = 1.5 cm-1 is also considered (Supplementary Information) to better fit the experimental lifetime values and in analogy to ref [33]."

    The Lorentz damping γ_z sets the phonon loss in the model, and the text explicitly states that this value was chosen to match the same experimental lifetime values that support the 'ultra-low losses preserved' claim. The supporting citation, ref [33], is a same-group preprint, so this choice is not anchored to an independent external measurement. This is a post-hoc calibration rather than a prediction, and it is not load-bearing for the independent RB-shift result; the reported lifetimes are derived from measured fringe decay and measured dispersion slopes, so the circularity is minor and localized.

full rationale

The paper's central RB-shift claim is an independent near-field measurement (nanoFTIR and s-SNOM dispersions), and the ab initio phonon calculation provides an independent first-principles trend (predicted ~50 cm-1 shift vs ~30 cm-1 measured); neither reduces to a fitted output. The lifetime comparison τ=L/vg is built from measured fringe decay lengths and measured dispersion group velocities, so the reported 4±1 ps and 6±1 ps are not literally computed from the Lorentz model. The only noticeable circular flavour is the Methods statement that γ_z = 1.5 cm-1 was adopted to better fit the experimental lifetime values, with ref [33] (same-group preprint) as analogy; since γ_z is the loss parameter of the model, this is a post-hoc calibration rather than an independent test of low loss. However, it is not load-bearing for the central tunability claim, and the lifetime numbers themselves are experimental. The different flake thicknesses and probing frequencies (130 nm at 1010 cm-1 vs 107 nm at 973 cm-1) are a correctness/confound concern, not circularity. Overall score 2.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central observation is experimental, so the free-parameter count is moderate. The main non-experimental load-bearing items are the Lorentz-oscillator model and the fringe-decay model used to convert measurements into dispersions and lifetimes; neither is independently benchmarked against a different measurement.

free parameters (2)
  • TO/LO phonon frequencies of the Drude-Lorentz permittivity = ω_TO^x=765, ω_LO^x=952, ω_TO^y=506, ω_LO^y=842, ω_TO^z=976, ω_LO^z=1037 cm-1
    Adjusted by fitting FTIR and s-SNOM measurements with transfer-matrix calculations (Methods). These parameters control the Reststrahlen band positions used to interpret near-field data.
  • Damping constants γ_a = γ_x=40, γ_y=19, γ_z=2.0 or 1.5 cm-1
    Taken from ref. [15], with γ_z=1.5 cm-1 also considered to better fit experimental lifetime values. This directly affects the extracted PhP lifetimes.
assumptions (4)
  • domain assumption The optical response of α-V2O5 and α'-(Na)V2O5 is described by a local Lorentz-oscillator dielectric tensor with principal axes along the crystallographic axes.
    Used to convert near-field polariton fringes into permittivities and dispersions; invoked in Methods and Fig. 1c and Fig. 4b.
  • domain assumption Phonon modes and effective charges from ab initio calculations, combined with the Lorentz model, give the permittivity of the intercalated crystal.
    Used to explain the disappearance of RB'2 and RB'3 and to predict the RB'1 shift in Fig. 4b; the comparison with experiment is only qualitative.
  • domain assumption The decay of s-SNOM fringes follows an exponentially damped sinusoid with a 1/sqrt(x) geometrical spreading factor.
    Underlies the extracted propagation lengths and lifetimes in Fig. 5; if this model is wrong, the lifetime values change.
  • domain assumption The spin-Peierls transition near 35 K confirms that the grown crystals are adequately intercalated α'-(Na)V2O5.
    Used in Methods to verify sample identity; it does not directly characterize the local Na distribution in each measured flake.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Broad Spectral Tuning of Ultra-Low Loss Polaritons in a van der Waals Crystal by Intercalation." pith.science (2026). https://pith.science/paper/7EWJ7IJP

@misc{pith2026250108705,
  author       = {Pith},
  title        = {Pith review of: Broad Spectral Tuning of Ultra-Low Loss Polaritons in a van der Waals Crystal by Intercalation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7EWJ7IJP}},
  note         = {Machine review of arXiv:2501.08705}
}
abstract

Phonon polaritons (PhPs) -- light coupled to lattice vibrations -- in polar van der Waals (vdW) crystals are promising candidates for controlling the flow of energy at the nanoscale due to their strong field confinement, anisotropic propagation, and ultra-long lifetime in the picosecond range \cite{ref1,ref2,ref3,ref4,ref5}. However, the lack of tunability in their narrow and material-specific spectral range -- the Reststrahlen Band (RB) -- severely limits their technological implementation. Here, we demonstrate that the intercalation of Na atoms in the vdW semiconductor $\alpha$-V$_2$O$_5$ enables a broad spectral shift of RBs, and that the PhPs excited exhibit ultra-low losses (lifetime of $4 \pm 1$~ps), similar to PhPs in the non-intercalated crystal (lifetime of $6 \pm 1$ ps). We expect our intercalation method to be applicable to other vdW crystals, opening the door for the use of PhPs in broad spectral bands in the mid-infrared domain.

Figures

Figures reproduced from arXiv: 2501.08705 by the authors.

Figure 2
Figure 2. Real-space imaging of a -V2O5 flake. a-c. Near-field amplitude images s3 of an -V2O5 flake with thickness d = 105 nm at incident frequencies ω0 = 1031 cm−1 , 1026 cm−1 , and 1020 cm−1 , respectively. d-f Profiles along the [100] (green lines) and [001] (blue lines) directions, extracted from the near-field amplitude images in a-c, respectively [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Real-space nano-spectroscopy of -V2O5 and intercalated ’-(Na)V2O5 flakes. a. Illustration of the orthorhombic lattice structure of layered -V2O5 (red spheres, oxygen atoms; blue atoms, vanadium atoms; blue pyramids, polyhedral defined by the oxygen atoms). The orthorhombic structure is based on bilayers of distorted VO5 square pyramids stacked along the [010] direction - with interlayer distance c = 0.44 nm - via… view at source ↗
Figure 4
Figure 4. PhPs dispersion and ab initio permittivity in -V2O5 and intercalated ’- (Na)V2O5 crystals. a. Dispersion of PhPs along the [100] and [001] directions in a -V2O5 (full symbols) flake with thickness d = 105 nm and a ’-(Na)V2O5 (empty symbols) flake with thickness d = 190 nm. Dashed and continuous horizontal lines mark the approximate TO and LO phonon modes in -V2O5 (TO, 980 cm−1 ; LO, 1040 cm−1 ), and ’-(Na)V2O5… view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: Anisotropy and lifetimes of PhPs in -V2O5 and intercalated ’-(Na)V2O5 flakes. a,b. Near-field amplitude images s3 of -V2O5 and ’-(Na)V2O5 flakes with thicknesses d = 130 nm, and d = 107 nm at illuminating frequencies ω0 = 1010 cm−1 (RB1) and ω0 = 973 cm−1 (RB’1), r…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

31 extracted references · 30 canonical work pages

  1. [1]

    Dai, S. et al. Tunable Phonon Polaritons in Atomically Thin van der Waals Crystals of Boron Nitride. Science 343, 1125-1129 (2014)

  2. [2]

    Ma, W. et al. In-plane anisotropic and ultra-low-loss polaritons in a natural van der Waals crystal. Nature 562, 557–562 (2018)

  3. [3]

    Giles, A. J. et al. Ultralow-loss polaritons in isotopically pure boron nitride. Nat. Mater. 17, 134 (2017)

  4. [4]

    Zheng, Z., et. al. Highly Confined and Tunable Hyperbolic Phonon Polaritons in Van Der Waals Semicon ducting Transition Metal Oxides . Adv. Mat. 30(13):1705318 (2018)

  5. [5]

    Zheng, Z. et al. A mid-infrared biaxial hyperbolic van der Waals crystal. Sci. Adv. 5, eaav8690 (2019)

  6. [6]

    Low, T. et al. Polaritons in layered two-dimensional materials. Nat. Mater. 16, 182 (2016)

  7. [7]

    N., Fogler, M

    Basov, D. N., Fogler, M. M. & García de Abajo, F. J. Polaritons in van der Waals materials. Science 354, 6309, aag1992 (2016)

  8. [8]

    Li, P. et al. Reversible optical switching of highly confined phonon -polaritons with an ultrathin phase-change material. Nat. Mater. 15, 870 (2016)

Show all 31 references
  1. [9]

    Sukimura, H. et al. Highly confined and switchable mid -infrared surface phonon polariton resonances of planar circular cavities with a phase change material. Nano Lett. 19, 2549–2554 (2019)

  2. [10]

    Caldwell, J. D. et al. Photonics with hexagonal boron nitride. Nat. Rev. Mater. 4, 552-567 (2019)

  3. [11]

    Dunkelberger, A. D. et al. Active tuning of surface phonon polariton resonances via carrier photoinjection. Nat. Photonics 12, 50–56 (2018)

  4. [12]

    Ratchford, D. C. et al. Controlling the Infrared Dielectric Function through Atomic - Scale Heterostructures. ACS Nano 13, 6730–6741 (2019)

  5. [13]

    & Lambrecht, W

    Bhandari, C. & Lambrecht, W. R. L. Phonons and related spectra in bulk and monolayer V2O5. Phys. Rev. B 89, 045109 (2014). 9

  6. [14]

    Sucharitakul, S. et al. V2O5: A 2D van der Waals Oxide with Strong In -Plane Electrical and Optical Anisotropy. ACS Appl. Mater. Interfaces 9, 23949 –23956 (2017)

  7. [15]

    & Vennik, J

    Clauws, P. & Vennik, J. Lattice Vibrations of V 2O5. Determination of TO and LO Frequencies from Infrared Reflection and Transmission. Phys. status solidi 76, 707–713 (1976)

  8. [16]

    Gomez-Diaz, J. S. & Alù, A. Flatland Optics with Hyperbolic Metasurfaces. ACS Photonics 3, 2211–2224 (2016)

  9. [17]

    S., Tymchenko, M

    Gomez-Diaz, J. S., Tymchenko, M. & Alù, A. Hyperbolic Plasmons and Topological Transitions Over Uniaxial Metasurfaces. Phys. Rev. Lett. 114, 233901 (2015)

  10. [18]

    Chen, J. et al. Optical nano-imaging of gate -tunable graphene plasmons. Nature 487, 77 (2012)

  11. [19]

    Fei, Z. et al. Gate-tuning of graphene plasmons revealed by infrared nano-imaging. Nature 487, 82 (2012)

  12. [20]

    Autore, M. et. al. Boron nitride nanoresonators for phonon -enhanced molecular vibrational spectroscopy at the strong coupling limit . Light: Science & Applications 14, 17172 (2018)

  13. [21]

    & Hillenbrand, R

    Huth, F., Schnell, M., Wittborn, J., Ocelic, N. & Hillenbrand, R. Infrared - spectroscopic nanoimaging with a thermal source. Nat. Mater. 10, 352 (2011)

  14. [22]

    S., Catlow, C

    Braithwaite, J. S., Catlow, C. R. A., Gale, J. D. & Harding, J. H. Lithium Intercalation into Vanadium Pentoxide: a Theoretical Study. Chem. Mater. 11, 1990 –1998 (1999)

  15. [23]

    Liu, J., Xia, H., Xue, D. & Lu, L. Double -Shelled Nanocapsules of V 2O5-Based Composites as High -Performance Anode and Cathode Materials for Li Ion Batteries. J. Am. Chem. Soc. 131, 12086–12087 (2009)

  16. [24]

    Xiong, F. et al. Li Intercalation in MoS 2: In Situ Observation of Its Dynamics and Tuning Optical and Electrical Properties. Nano Lett. 15, 6777–6784 (2015)

  17. [25]

    Cha, Judy.J., et. al. Two-Dimensional Chalcogenide Nanoplates as Tunable Metamaterials via Chemical Intercalation. Nano Lett. 13, 5913–5918 (2013)

  18. [26]

    Zhang, R., Waters, J., Geim, A. K. & Grigorieva, I. V. Intercalant -independent transition temperature in superconducting black phosphorus. Nat. Commun. 8, 15036 (2017)

  19. [27]

    Pons-Valencia, P., et. al. Launching of hyperbolic phonon-polaritons in h-BN slabs by resonant metal plasmonic antennas. Nat. Commun. 10, 3242 (2019)

  20. [28]

    Woessner, A. et al. Highly confined low-loss plasmons in graphene –boron nitride heterostructures. Nat. Mater. 14, 421 (2014)

  21. [29]

    Heterointerface effects in the electrointercalation of van der 10 Waals heterostuctures

    Kwabena Bediako, D. Heterointerface effects in the electrointercalation of van der 10 Waals heterostuctures. Nature 558, 425-429 (2018)

  22. [30]

    Talwar, N. T. Direct evidence of LO phonon -plasmons coupled modes in n -GaN. Appl. Phys. Lett. 97, 051902 (2010). ACKNOWLEDGEMENTS J.T-G and G.Á -P acknowledge support through the Severo Ochoa Program from the Goverment of the Principality of Asturias (PA-18-PF-BP17-126 and P...

  23. [100]

    Solid horizontal lines mark the approximate transversal phonon modes in ’-(Na)V2O5 (TO, 950 cm−1), defining RB’1

    and [001] directions of a ’-(Na)V2O5 flake s howing the near -field amplitude s3 (normalized to the near-field amplitude on Au, s3,Au) as a function of distance between tip and flake edge. Solid horizontal lines mark the approximate transversal phonon modes in ’-(Na)V2O5 (TO...

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.