Pith. sign in

REVIEW 3 major objections 6 minor 22 references

Semiconductor nanofilms as thermal phonon polarizers: competing effects of scattering selection rules and boundary mode conversion

T0 review · 3 major / 6 minor · reviewed 2026-07-07 · glm-5.2

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

Pith's one-line read Nanofilms polarize heat into transverse phonons, halving InP conductivity

desk verdict New mechanism for mode-polarized heat currents in nanofilms via competition between selection rules and boundary mode conversion; isotropic projection is the load-bearing approximation that needs checking read the letter →

arxiv 2607.05296 v1 pith:XT7RHM7Y submitted 2026-07-06 cond-mat.mtrl-sci cond-mat.mes-hall

Vasumathy Ravishankar , Navaneetha K. Ravichandran This is my paper · ORCID
classification cond-mat.mtrl-scicond-mat.mes-hall
keywords phononrulesselectionfilmsnanoscaleconversionheatmode
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 argues that two mechanisms known to govern phonon heat transport in bulk semiconductors — intrinsic three-phonon scattering selection rules and extrinsic boundary mode conversion — compete directly when a semiconductor is made thin enough (10–500 nm), and that this competition produces a heat current dominated by one phonon polarization. In bulk indium phosphide (InP) at cryogenic temperatures (≤100 K), a selection rule called AAA # 2 (arising from the large velocity gap between longitudinal acoustic (LA) and transverse acoustic (TA) branches) weakens the intrinsic scattering of LA phonons, making them the dominant heat carriers. But when those same LA phonons strike the boundaries of a nanofilm, they convert into TA modes with high probability, while the reverse TA→LA conversion is suppressed by a Snell's-law critical-angle effect. The net result is that the LA contribution is drained away, the heat current becomes TA-polarized, and the thermal conductivity drops to roughly 50% of the bulk value for 10–100 nm InP films at 100 K under specular boundaries with mode conversion. The paper establishes this using first-principles variance-reduced Monte Carlo solutions of the steady-state Peierls-Boltzmann equation, comparing InP (where the LA/TA velocity gap is large) against boron phosphide (BP, where acoustic branches are bunched and the effect is weak). The central object carrying the argument is the mode-conversion probability at the boundary, which factorises into a material term D(ν,p)v(ν,p) (density of states times group velocity, evaluated per polarization at fixed frequency) and a geometric term cos θ_out set by Snell's law. Both terms favour LA→TA over TA→LA, and both are stronger in InP than in BP because the larger velocity separation in InP amplifies the contrast between polarizations. The paper also shows a size effect: as film thickness shrinks, boundary mode conversion precedes intrinsic scattering for an increasing fraction of phonons, causing κ/κ_b to saturate at a thickness-independent, temperature-independent limit set by the relative LA and TA contributions to bulk κ.

What carries the argument

The argument rests on three linked components. First, the AAA # 2 selection rule: in materials where LA and TA dispersions are well separated (InP, InAs, InSb), three-phonon all-acoustic scattering of low-frequency LA phonons is suppressed, amplifying their bulk κ contribution. Second, the boundary mode-conversion probability [Eq. 2]: elastic reflection at a film boundary redistributes phonons across polarizations at fixed frequency, with relative probabilities proportional to D(ν,p')v_⊥(ν,p'), where D is the density of states and v_⊥ is the group velocity component normal to the boundary. Because D_λ v_λ is larger for TA than LA, LA→TA conversion is favoured and TA→LA is disfavoured. Third,

What would settle it

Measure the in-plane thermal conductivity of 10–100 nm InP films at 100 K with specular boundaries. If κ does not drop to approximately 50% of bulk, or if the spectral contribution of LA phonons is not preferentially suppressed relative to TA, the mode-conversion polarization mechanism is not operating as predicted. Alternatively, if polarized phonon spectroscopy (e.g., inelastic neutron or Raman-based probes adapted for thin films) shows no TA-polarized non-equilibrium phonon population in the films, the central claim is falsified.

Watch

Extended reading notes

Core claim

The paper's central discovery is that phonon scattering selection rules, which amplify LA phonon heat transport in bulk InP, are overridden at nanoscale film boundaries by mode conversion that preferentially depopulates LA phonons into TA modes. This produces a TA-polarized non-equilibrium heat current and suppresses thermal conductivity to ~50% of bulk for 10–100 nm InP films at 100 K. The asymmetry in mode conversion arises from two compounding factors: the material factor D_λ v_λ is larger for TA than LA (because the large velocity gap gives TA phonons higher density of states and comparable or higher velocities at fixed frequency), and Snell's law restricts TA→LA conversion to phononsInc

Load-bearing premise

The load-bearing premise is that the anisotropic, wave-vector-dependent group velocities and relaxation times of real crystals can be accurately collapsed into a single isotropic representation per polarization branch. If this isotropic projection distorts the relative magnitudes of the density-of-states–velocity product D_λ v_λ between LA and TA branches — or the wave-vector ratio q(ν,TA)/q(ν,LA) that sets the Snell's-law critical angle for mode conversion — then the asymmy

Editorial extensions

If this is right

  • Nanofilms of InP and related III-V semiconductors could serve as thermal phonon polarizers, producing heat currents with controlled polarization that enable selective coupling to electrons, defects, or strain fields — useful for engineering hot-carrier lifetimes and defect-insensitive thermal properties.
  • The predicted κ suppression to ~50% of bulk for 10–100 nm InP films at 100 K is directly measurable with existing thin-film thermal conductivity techniques (time-domain thermoreflectance, suspended microbridge) on films already fabricable at sub-10-μm thickness.
  • The principle extends to any material where a large LA/TA velocity gap activates the AAA # 2 selection rule, suggesting a materials-design criterion: large acoustic-branch separation produces strong thermal polarization in nanofilms.
  • The temperature-independent saturation of κ/κ_b at small thickness (driven by the T-independent relative LA/TA contributions to bulk κ between 50–100 K) provides a clean experimental signature: below a critical thickness, further thinning or cooling should not change the normalized conductivity.
  • The finding that diffuse boundary scattering erases all mode-conversion effects (because direction randomisation dominates) means the polarizing effect requires specular or near-specular boundaries — a constraint on surface preparation for any device application.
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, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. This manuscript predicts that phonon scattering selection rules and boundary mode conversion compete in nanoscale semiconductor films to produce mode-polarized heat currents at cryogenic temperatures (≤100 K). Using first-principles phonon properties and a variance-reduced Monte Carlo (VRMC) solution of the steady-state Peierls-Boltzmann equation, the authors show that in InP nanofilms, the AAA # 2 selection rule amplifies LA phonon contributions to bulk κ, while boundary mode conversion at specular surfaces preferentially depopulates LA phonons, yielding a TA-polarized heat current with κ suppressed to ~50% of bulk at 10–100 nm. The effect is weaker in BP, where TA phonons dominate. The relaxation time approximation is justified by comparison to full iterative solutions (within ~12%). The central physical mechanism — that the D_λ·v_λ ratio and Snell's law geometric factor both favor LA→TA conversion — is clearly articulated and supported by Figs. 1–3.

Significance. The paper identifies a previously unexplored mechanism for generating mode-polarized thermal phonon currents in nanoscale films, with potential implications for symmetry-selective phonon engineering. Strengths include: (1) no fitted free parameters — all phonon properties derive from first-principles DFT calculations (Ref. [2]); (2) the RTA is validated against full iterative bulk solutions; (3) the VRMC method is well-established; (4) the prediction is falsifiable — the κ/κ_b ratio and mode polarization are specific quantitative predictions testable in InP films with controlled surface specularity. The claim that the effect is material-dependent (InP vs. BP) and temperature-dependent (Fig. 4) adds further testability. However, the significance of the quantitative predictions rests on the isotropic projection approximation, which is not directly validated for the specific ratios that drive the mode-conversion asymmetry.

major comments (3)
  1. Paragraph following Eq. 1: The isotropic projection of wave-vector-dependent group velocities and relaxation times (citing Ref. [16]) is the load-bearing approximation for the central claim. The mode-conversion probabilities in Eq. 2 depend on the ratio D(ν,TA)·v(ν,TA) / D(ν,LA)·v(ν,LA), and the Snell's law critical angle depends on q(ν,TA)/q(ν,LA). If the isotropic projection distorts these ratios differently for InP vs. BP — or differently from the directional values along the transport direction — the predicted ~50% κ suppression in InP could be an artifact. The manuscript provides no direct comparison between isotropic-projected and full anisotropic D_λ·v_λ ratios. A figure or table comparing the projected vs. anisotropic ratios for both materials, at least along representative high-symmetry directions, would substantially strengthen the claim. Without this, the reader cannot assesss
  2. Eq. 2 and the Snell's law geometric factor: The mode-conversion model assumes elastic scattering (fixed ν) and flux conservation. The manuscript states that evanescent/surface LA fields for θ_in > θ_c are localized to ~nm-thick regions and do not contribute to heat current, but this is stated without calculation or citation. For THz phonons at 100 K, the penetration depth and energy storage in these evanescent modes could be non-negligible relative to a 10 nm film. A brief estimate or reference supporting the 'localized to ~nm' claim would strengthen this argument, particularly since the smallest film thicknesses studied (d=10 nm) are comparable to this length scale.
  3. The T-independent small-d limit of κ/κ_b (Fig. 4 and accompanying text) is attributed to T-independent relative LA/TA contributions to κ_b (~40% LA, ~60% TA) between 50–100 K. However, this explanation is qualitative. A more direct demonstration — e.g., showing that the mode-converted phonon distribution from Eq. 2 is indeed T-independent when expressed in terms of the normalized spectral contributions — would make this claim more convincing. As stated, the reader must take the T-independence on faith rather than seeing it derived.
minor comments (6)
  1. Fig. 2(a): The inset schematic showing incidence/reflection angles is small and difficult to read. Enlarging or separating it into its own panel would improve clarity.
  2. The phrase 'first-principles calculations' is used throughout, but the phonon relaxation times are taken from Ref. [2] (Ravichandran & Broido, PRX 2020). While this is legitimate, a brief note clarifying that the first-principles inputs are from prior published DFT work, with the present contribution being the VRMC simulation and mode-conversion model, would help the reader.
  3. The discussion of evanescent modes (end of the paragraph containing the Snell's law analysis) would benefit from a quantitative estimate or a reference. As written, the claim that these modes are 'localized to ~nm-thick regions' is unsupported.
  4. Supplementary figures are referenced (S1, S2, S3) but not provided in the main text. While standard, ensuring these are included in review materials is essential for full assessment.
  5. The abstract states 'cryogenic temperatures (≤100 K).' The choice of 100 K as the upper bound is reasonable but could be briefly justified — e.g., noting that above this temperature, intrinsic scattering rates increase and mask the selection-rule effect.
  6. Ref. [9] appears to be a 2026 Advanced Materials article; if this is an in-press or preprint reference, the authors should verify publication status and update the citation.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for a careful and constructive report. The referee correctly identifies that the isotropic projection approximation is load-bearing for our central quantitative claim, and we agree that direct validation of this approximation for the specific ratios driving mode-conversion asymmetry is needed. We will add a comparison between isotropic-projected and full anisotropic D_λ·v_λ ratios for both materials. We will also add a quantitative estimate of evanescent mode penetration depth and a more direct demonstration of the T-independence of the small-d limit. One comment regarding the T-independence derivation we address with a partial revision, as the full analytical derivation is complex but a numerical demonstration is feasible and will be added.

read point-by-point responses

  1. Referee: Paragraph following Eq. 1: The isotropic projection of wave-vector-dependent group velocities and relaxation times is the load-bearing approximation. No direct comparison between isotropic-projected and full anisotropic D_λ·v_λ ratios is provided.

    Authors: The referee is correct that the isotropic projection approximation is load-bearing for the mode-conversion probabilities in Eq. 2, and that direct validation of this approximation for the specific D_λ·v_λ ratios is absent from the current manuscript. We agree this is a gap. In the revised manuscript, we will add a figure or table comparing the isotropic-projected D_λ·v_λ ratios against full anisotropic calculations along representative high-symmetry directions (Γ→X, Γ→K, Γ→L) for both InP and BP. This will allow readers to directly assess whether the isotropic projection distorts the LA/TA contrast differently between the two materials. We note that the isotropic projection has been previously validated for predicting κ in thin films of cubic crystals (Refs. [17–19]), and the RTA bulk κ values we report agree with full iterative solutions to within ~12%, which provides indirect evidence that the projection does not severely distort the relevant phonon properties. However, we agree that the mode-conversion ratios specifically warrant direct validation, and we will provide it. revision: yes

  2. Referee: Eq. 2 and Snell's law: The claim that evanescent/surface LA fields for θ_in > θ_c are localized to ~nm-thick regions is stated without calculation or citation. For THz phonons at 100 K in a 10 nm film, penetration depth could be non-negligible.

    Authors: The referee raises a valid point. The statement about evanescent mode localization is currently made without supporting calculation. We will add a brief quantitative estimate in the revised manuscript. The penetration depth of an evanescent LA field can be estimated as δ = 1/√(q_LA² - q_parallel²), where q_parallel = q_TA sin θ_in is the conserved parallel momentum component and q_LA is the LA wave vector magnitude at the same frequency. For THz phonons in InP (q ~ 0.1–1 nm⁻¹), when θ_in modestly exceeds θ_c, the evanescent decay length is on the order of ~1–5 nm. We will include this estimate explicitly and cite relevant literature on evanescent phonon fields at interfaces. We acknowledge that for the smallest film thicknesses studied (d = 10 nm), this length scale is not entirely negligible, and we will discuss this caveat: the fraction of phonons with θ_in significantly exceeding θ_c is small, and the energy stored in evanescent modes is further reduced by the D_λ·v_λ weighting, but the limitation at d = 10 nm will be noted honestly. revision: yes

  3. Referee: The T-independent small-d limit of κ/κ_b is attributed to T-independent relative LA/TA contributions to κ_b, but this explanation is qualitative. A more direct demonstration is needed.

    Authors: The referee is right that the current explanation is qualitative and that a more direct demonstration would strengthen the claim. The T-independence of the small-d limit follows from two facts: (1) at small d, the phonon distribution is governed entirely by the mode-conversion distribution from Eq. 2, which is T-independent when expressed in terms of normalized spectral contributions, and (2) the relative LA/TA contributions to κ_b are approximately T-independent between 50–100 K. In the revised manuscript, we will add a figure showing the mode-converted phonon distribution from Eq. 2 expressed in terms of normalized spectral contributions, demonstrating its T-independence directly. We will also show the LA and TA fractional contributions to κ_b as a function of T to make the second point quantitative rather than qualitative. We note that a full analytical derivation of the T-independence from Eq. 2 is non-trivial because the mode-conversion distribution depends on the equilibrium energy distribution e^d_λ, which is T-dependent; the T-independence of the normalized ratios emerges from the cancellation of T-dependent factors in the ratio D_λ v_λ e^d_λ / Σ_p D_λ v_λ e^d_λ. We will include this analytical argument alongside the numerical demonstration. revision: partial

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: self-citations provide independent first-principles inputs, not fitted-to-target parameters

full rationale

The paper's derivation chain uses first-principles DFT phonon properties (group velocities, DOS, relaxation times from Ref [2]), standard physics (flux conservation Eq. 2, Snell's law), and a variance-reduced Monte Carlo method to predict nanofilm thermal conductivity. The load-bearing self-citation is Ref [2] (Ravichandran & Broido, PRX 2020), which provides phonon-phonon scattering rates and selection rules derived from DFT — these are independent of the present paper's target (nanofilm κ). No parameter is fitted to a subset of data and then 'predicted' on related data. The D_λ·v_λ ratios that drive mode conversion asymmetry (Fig. 3) are computed from first-principles dispersions, not fitted. The isotropic projection (Ref [16], by different authors Hua & Minnich) is an approximation with known limitations but is not circular. The saturation behavior at small d is a derived size effect, not a fit. The minor score of 2 reflects that Ref [2] is load-bearing and co-authored by a present author, but it is a published, externally verifiable first-principles calculation that does not define its outputs in terms of the present paper's results. The reader's concern about the isotropic projection distorting D_λ·v_λ ratios is a correctness risk, not a circularity issue.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new particles, forces, fields, or entities. All physical quantities (phonon modes, group velocities, density of states, relaxation times) are standard.

free parameters (1)
  • None fitted = N/A
    The paper does not fit any parameters to the thin-film κ data. Phonon relaxation times are from first-principles DFT calculations (Ref [2]). Boundary mode conversion probabilities follow from Eq. 2 (flux conservation). The only inputs are material choice, film thickness, temperature, and boundary specularity type (specular vs. diffuse).
assumptions (5)
  • domain assumption Relaxation time approximation (RTA) is sufficient for these materials at 100 K
    Stated in the paragraph following Eq. 1: 'it is sufficient to solve the SSPBE under the relaxation time approximation (RTA).' Justified by comparing RTA bulk κ to full iterative solution: 2589 vs 2906 W/mK (BP), 337 vs 347 W/mK (InP). The ~12% difference for BP is non-trivial but does not change the qualitative mechanism.
  • domain assumption Isotropic projection of anisotropic phonon properties is valid for cubic crystals
    Paragraph following Eq. 1: 'wave vector-dependent group velocities and relaxation times are projected into an isotropic form as described in Ref. [16] - an approximation that has been successfully used to predict heat flow in thin films of cubic crystals [17-19].' This is a standard approximation but affects the mode conversion probabilities through the relative magnitudes of D_λ and v_λ across polarizations.
  • domain assumption Boundary mode conversion is elastic (frequency-conserving)
    Eq. 2 sums over polarizations 'at a fixed frequency ν to represent elastic scattering.' This assumes no energy exchange at the boundary, which may not hold for rough or chemically modified surfaces.
  • standard math Matthiessen's rule for combining three-phonon, four-phonon, and isotope scattering rates
    Stated in the paragraph following Eq. 1: 'τ_λ is the phonon relaxation time calculated using the Matthiessen's rule with contributions from three-phonon, four-phonon and phonon-isotope interactions.' Standard approximation in phonon transport; known to break down when scattering channels are strongly correlated.
  • domain assumption Snell's law applies to phonon mode conversion at film boundaries
    Used to determine reflection angles: 'sin θ_out = q(ν,p)/q(ν,p') sin θ_in.' This treats phonon reflection analogously to electromagnetic wave refraction, assuming coherent specular reflection. Validity depends on boundary quality at the phonon wavelength scale.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Semiconductor nanofilms as thermal phonon polarizers: competing effects of scattering selection rules and boundary mode conversion." pith.science (2026). https://pith.science/paper/XT7RHM7Y

@misc{pith2026260705296,
  author       = {Pith},
  title        = {Pith review of: Semiconductor nanofilms as thermal phonon polarizers: competing effects of scattering selection rules and boundary mode conversion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XT7RHM7Y}},
  note         = {Machine review of arXiv:2607.05296}
}
abstract

Phonon scattering selection rules are known to control heat flow through bulk solids. Here we show that these selection rules also modulate heat flow through nanoscale semiconductor films, although through a previously-unexplored mechanism. Using first-principles calculations, we expose a competition between these selection rules and phonon mode conversion at boundaries of nanoscale films, that drives mode-polarized heat currents at cryogenic temperatures ($\le$ 100 K). This polarizing effect is stronger in materials like indium phosphide, where selection rules based on large velocity differences between phonon branches amplifies the longitudinal acoustic (LA) phonon contribution to thermal conductivity by restricting their intrinsic scattering events, while boundary mode conversion in nanoscale films suppresses it by depopulating the LA phonons. The resulting transverse-polarized non-equilibrium phonons will enable symmetry-selective engineering of phonon coupling to electrons, strains and defects in nanoscale films, that is difficult to achieve in bulk solids.

Figures

Figures reproduced from arXiv: 2607.05296 by the authors.

Figure 1
Figure 1. FIG. 1. Acoustic phonon dispersions of InP and BP along [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Effect of mode conversion on the [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. The normalized material factor [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Thickness-dependent [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 22 canonical work pages

  1. [2]

    Ravichandran and David Broido

    Navaneetha K. Ravichandran and David Broido. Phonon-Phonon Interactions in Strongly Bonded Solids: Selection Rules and Higher-Order Processes.Physical Review X, 10(2):021063, 2020

    work page 2020

  2. [16]

    Chengyun Hua and Austin J. Minnich. Analytical Green’s function of the multidimensional frequency- dependent phonon Boltzmann equation.Physical Review B, 90(21):214306, 2014

    work page 2014

  3. [1]

    Lindsay, D

    L. Lindsay, D. A. Broido, and T. L. Reinecke. First- Principles Determination of Ultrahigh Thermal Conduc- tivity of Boron Arsenide: A Competitor for Diamond? Physical Review Letters, 111(2):025901, 2013

    work page 2013

  4. [3]

    Fei Tian, Bai Song, Xi Chen, Navaneetha K. Ravichan- dran, Yinchuan Lv, Ke Chen, Sean Sullivan, Jaehyun Kim, Yuanyuan Zhou, Te-Huan Liu, Miguel Goni, Zhi- wei Ding, Jingying Sun, Geethal Amila Gamage Udala- matta Gamage, Haoran Sun, Hamidreza Ziyaee, Shuyuan Huyan, Liangzi Deng, Jianshi Zhou, Aaron J. Schmidt, Shuo Chen, Ching-Wu Chu, Pinshane Y. Huang, D...

    work page 2018

  5. [4]

    Huang, David G

    Sheng Li, Qiye Zheng, Yinchuan Lv, Xiaoyuan Liu, Xiqu Wang, Pinshane Y. Huang, David G. Cahill, and Bing Lv. High thermal conductivity in cubic boron arsenide crystals.Science, 361(6402):579–581, 2018

    work page 2018

  6. [5]

    Experimental observation of high thermal conductivity in boron arsenide.Science, 361(6402):575–578, 2018

    Joon Sang Kang, Man Li, Huan Wu, Huuduy Nguyen, and Yongjie Hu. Experimental observation of high thermal conductivity in boron arsenide.Science, 361(6402):575–578, 2018

    work page 2018

  7. [6]

    Ravichandran and David Broido

    Navaneetha K. Ravichandran and David Broido. Non- monotonic pressure dependence of the thermal con- ductivity of boron arsenide.Nature Communications, 10(1):1–8, 2019

    work page 2019

  8. [7]

    Anoma- lous thermal transport under high pressure in boron ar- senide.Nature, 612(7940):459–464, 2022

    Suixuan Li, Zihao Qin, Huan Wu, Man Li, Martin Kunz, Ahmet Alatas, Abby Kavner, and Yongjie Hu. Anoma- lous thermal transport under high pressure in boron ar- senide.Nature, 612(7940):459–464, 2022

    work page 2022

Show all 22 references
  1. [8]

    Ravichandran and David Broido

    Navaneetha K. Ravichandran and David Broido. Expos- ing the hidden influence of selection rules on phonon- phonon scattering by pressure and temperature tuning. Nature Communications, 12(1):3473, 2021

    2021

  2. [9]

    Yulin Zhu, Aswin L. N. Kondusamy, Ke Chen, Pawan Koirala, Hanlin Wu, Mahammed S. Patel, Evan R. Glaser, Sam White, James C. Culbertson, John L. Lyons, Navaneetha K. Ravichandran, Mohamed Zetati, Xinglu Wang, Rafik Addou, Robert M. Wallace, Songrui Hou, Sam Vaziri, Xinyu Bao, D...

    2026

  3. [10]

    Ashis Kundu, Xiaolong Yang, Jinlong Ma, Tianli Feng, Jes´ us Carrete, Xiulin Ruan, Georg K. H. Madsen, and Wu Li. Ultrahigh Thermal Conductivity ofθ-Phase Tan- talum Nitride.Physical Review Letters, 126(11):115901, 2021

    2021

  4. [11]

    Kelly, Mary H

    Suixuan Li, Chuanjin Su, Zihao Qin, Ahmet Alatas, Mar- tin Kunz, Takahiro Yamada, Shelly D. Kelly, Mary H. Upton, Anthony Gironda, Jiyong Zhao, Bora Kalkan, Wanli Yang, Toshihiro Aoki, and Yongjie Hu. Metal- licθ-phase tantalum nitride has a thermal conductivity triple that of...

    2026

  5. [12]

    Glen A. Slack. The Thermal Conductivity of Nonmetal- lic Crystals. In Henry Ehrenreich, Frederick Seitz, and David Turnbull, editors,Solid State Physics, volume 34, pages 1–71. Academic Press, 1979

    1979

  6. [13]

    P´ eraud and Nicolas G

    Jean-Philippe M. P´ eraud and Nicolas G. Hadjiconstanti- nou. Efficient simulation of multidimensional phonon transport using energy-based variance-reduced Monte Carlo formulations.Physical Review B, 84(20):205331, 2011

    2011

  7. [14]

    P´ eraud and Nicolas G

    Jean-Philippe M. P´ eraud and Nicolas G. Hadjiconstanti- nou. An alternative approach to efficient simulation of micro/nanoscale phonon transport.Applied Physics Let- ters, 101(15):153114–153114–4, 2012

    2012

  8. [15]

    Ravichandran and Austin J

    Navaneetha K. Ravichandran and Austin J. Minnich. Co- herent and incoherent thermal transport in nanomeshes. Physical Review B, 89(20):205432, 2014

    2014

  9. [17]

    A. J. Minnich. Determining Phonon Mean Free Paths from Observations of Quasiballistic Thermal Transport. Physical Review Letters, 109(20):205901, 2012

    2012

  10. [18]

    Ravichandran, Hang Zhang, and Austin J

    Navaneetha K. Ravichandran, Hang Zhang, and Austin J. Minnich. Spectrally Resolved Specular Reflec- tions of Thermal Phonons from Atomically Rough Sur- faces.Physical Review X, 8(4):041004, 2018

    2018

  11. [19]

    Transient Hydrodynamic Lattice Cooling by Pi- cosecond Laser Irradiation of Graphite.Physical Review Letters, 127(8):085901, 2021

    Jihoon Jeong, Xun Li, Sangyeop Lee, Li Shi, and Yaguo Wang. Transient Hydrodynamic Lattice Cooling by Pi- cosecond Laser Irradiation of Graphite.Physical Review Letters, 127(8):085901, 2021

    2021

  12. [20]

    Ravichandran and Austin J

    Navaneetha K. Ravichandran and Austin J. Minnich. Role of thermalizing and nonthermalizing walls in phonon heat conduction along thin films.Physical Re- view B, 93(3):035314, 2016

    2016

  13. [21]

    Ghossoub, Krishna Valavala, Jun Ma, Manjunath C

    Dhruv Gelda, Marc G. Ghossoub, Krishna Valavala, Jun Ma, Manjunath C. Rajagopal, and Sanjiv Sinha. Specu- larity of longitudinal acoustic phonons at rough surfaces. Physical Review B, 97(4):045429, 2018

    2018

  14. [22]

    Layer-resolved release of epitaxial layers in III-V heterostructure via a buffer-free mechanical separation 6 technique.Science Advances, 8(3):eabl6406, 2022

    Honghwi Park, Heungsup Won, Changhee Lim, Yux- uan Zhang, Won Seok Han, Sung-Bum Bae, Chang-Ju Lee, Yeho Noh, Junyeong Lee, Jonghyung Lee, Sungh- wan Jung, Muhan Choi, Sunghwan Lee, and Hongsik Park. Layer-resolved release of epitaxial layers in III-V heterostructure via a buf...

    2022

Pith tools

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