Pith. sign in

REVIEW 3 major objections 6 minor 46 references

Mechanical loss in amorphous solids: spatial correlations, interacting transitions, and annealed thermodynamic pathways

T0 review · 3 major / 6 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read Mechanical loss in amorphous solids arises from a connected network of interacting atomic transitions, not from an independent superposition of two-level systems, and annealing lowers loss by flattening the distribution of inherent-structur

desk verdict Genuinely new spatial-clustering and loop-asymmetry evidence supports the network description, but the thermal-search sampling completeness is the soft spot that needs a convergence check before the quantitative loss claims can be trusted. read the letter →

arxiv 2607.17434 v1 pith:4G5L4YQT submitted 2026-07-19 cond-mat.mtrl-sci

Steven Blaber , Jörg Rottler This is my paper · ORCID
classification cond-mat.mtrl-sci PACS 62.40.+i61.43.-j63.50.Lm
keywords mechanicallossamorphoussolidstwo-levelsystemsenergylandscapeconnectednetworkannealingmoleculardynamicsgravitationalwavedetectors
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 makes a specific claim: the standard description of low-frequency mechanical loss in amorphous solids as a collection of independent two-level systems (TLS) is systematically incomplete. Using molecular simulations of amorphous silicon and titania, the authors show that atomic transitions between energy minima cluster in space and modify each other's barriers, so the correct minimal description is a connected network of states whose relaxation is governed by collective eigenmodes, not single transitions. In that network description, slower quenching narrows the spread of energy minima and, opposite to the TLS prediction, lowers room-temperature mechanical loss. The paper identifies a 'mountain' energy profile along the dominant relaxation paths as the structural cause of the loss, and shows that artificially flattening the landscape removes it. A sympathetic reader would care because this reframes how loss spectra and annealing trends in mirror coatings for gravitational-wave detectors should be interpreted.

What carries the argument

The load-bearing object is the connected-network loss formula (Eq. 5), Q^{-1}_CN = (β N γ0^2 / 4 V_C) g^T M D(ωτ) M^T g, where M diagonalizes the symmetrized master-equation rate matrix S constructed from Arrhenius transition rates between inherent-structure minima, g is the probability-weighted strain coupling vector, and D(ωτ) is the diagonal spectral factor ωτ/(1+ω²τ²). This formula turns the energy landscape into a graph of nodes (minima) and edges (transitions) and expresses loss as a sum over collective relaxation eigenmodes. Two analysis tools carry the argument: the participation ratio of the dominant loss eigenvector, which tells how many states actually participate, and the Steiner

What would settle it

Reconstruct the transition network of a-Si with at least 10 times the number of thermal-search trajectories or with an independent accelerated sampling method; if the predicted Q^{-1} at 300 K and 1000 Hz changes by more than the reported standard error, the finite-sampling assumption is violated. A complementary check is computing the radial distribution of transition locations in a much larger sample: the claimed clustering peak at about 1 Å must persist.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that the transitions responsible for mechanical loss in amorphous solids are spatially clustered (a radial distribution peak roughly 60 times the random-gas value at about 1 Å) and energetically coupled: four-state loops that should be perfectly symmetric under an independent-TLS model show energy deviations up to about 0.25 eV at separations below about 4 Å. Dissipation is therefore carried by eigenmodes of the master-equation rate matrix, and the connected-network loss formula (Eq. 5) generalizes the TLS result. When the energy landscape is annealed, either by slower quenching or by an artificial narrowing parameter α, the distribution of

Load-bearing premise

The central claim rests on the assumption that the transition network found by finite thermal searches (600 K, 200 ps, 100–200 trajectories per sample) faithfully represents the energy landscape that governs loss at 300 K and 1000 Hz — if rare, slow, or high-barrier transitions are missed, the network connectivity, path lengths, and predicted Q^{-1} are biased.

Editorial extensions

If this is right

  • Better-annealed amorphous coatings should show lower room-temperature mechanical loss even when the number of active defects is unchanged, because the relevant control parameter is the width of the inherent-structure energy distribution, not the barrier distribution alone.
  • An experimental loss spectrum that shows a higher-temperature peak after annealing would be a direct signature of the network mechanism, since the TLS model predicts the opposite trend for a tightened energy landscape.
  • Even the most TLS-like structural motifs — four-state cycles — carry interaction corrections of order 0.25 eV, so any quantitative interpretation of loss data in terms of independent TLS is unreliable in this regime.
  • The mountain energy profile along dominant eigenmodes provides a concrete structural target: loss can be reduced by eliminating high-lying minima in the middle of relaxation paths, for example by thermal or energetic processing that flattens the landscape.
  • For a-TiO2 specifically, the dominant room-temperature loss mode is a small cluster of states exchanging with a diffuse background through an energetic bottleneck; the same mode is absent after artificial annealing, so annealing protocols that raise the lowest-energy barriers should suppress this channel.

Reading between the lines

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

  • Editorial inference: If the roughly 4 Å clustering length survives in more accurate potentials and larger samples, it suggests a universal structural correlate, probably linked to dynamical heterogeneity, so interaction corrections should be expected in essentially all amorphous solids, including ultrastable glasses, though at lower defect densities.
  • The connected-network formula is directly transferable to dielectric loss in superconducting qubits, where interacting-defect models are already standard; computing the network's dielectric analog may yield quantitative predictions for qubit coherence fluctuations.
  • A testable extension is to run the same analysis on samples prepared by vapor deposition or with controlled hydrogen content, where experiments already show reduced loss; the model predicts the loss reduction should track the narrowing of the energy-minimum distribution, which can be checked by simulated or measured calorimetric signatures.
  • The paper's own order-of-magnitude estimate implies that only the most defect-poor experimental amorphous silicon samples fall below the interaction threshold (n ≈ 0.3 nm^-3); typical sputtered coatings, including those used in gravitational-wave detectors, are likely deep in the interacting regime, so the independent-TLS interpretation of their loss spectra may need revision.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. Blaber and Rottler use molecular dynamics simulations of amorphous Si and TiO₂ to argue that thermally activated structural transitions form a spatially clustered, interacting network, so the standard independent-TLS description of low-frequency mechanical loss is incomplete. They construct transition networks from 600 K thermal-search trajectories, measure spatial clustering via g(r), quantify deviations from independent TLS around four-state loops (Fig. 5), and study how quench rate and an artificial energy-landscape flattening parameter α affect the loss Q⁻¹ computed from the connected-network formula of Ref. [26]. They find that slower quench and α→0 reduce room-temperature loss in the network model while the TLS model predicts the opposite, and they analyze dominant relaxation paths to explain the mechanism.

Significance. If correct, the paper would strengthen the case that mirror-coating mechanical loss in gravitational-wave detectors cannot be captured by an ensemble of independent two-level systems and that a network-level description with interactions is needed. The paper's strengths are its direct computation of the spatial distribution of transition locations in two different amorphous materials, the explicit identification of cubic/loop structures, and the attempt to connect annealing trends to network topology. The analytical loss formula is, however, taken from the authors' own Ref. [26]; the new contribution is mainly the spatial-correlation analysis, the loop-asymmetry metric, and the annealing/path analysis. These are valuable and falsifiable in principle, but the load-bearing claims currently rest on the completeness/representativeness of the thermal-search network and on an artificial annealing parameter.

major comments (3)
  1. [Sec. III (sampling)] The network on which all subsequent analysis is based is built from 600 K, 200 ps thermal-search trajectories sampled every 100 fs (100 per a-Si sample, 200 per a-TiO₂). At this temperature, Arrhenius rates strongly favor low barriers, and because each trajectory continues from the newly found minimum, the sampled edges are a path-biased subset of the full landscape. No convergence check (e.g., number of unique transitions vs simulation length, comparison with activation-relaxation technique or other saddle searches, or sensitivity to number of seeds) is reported. Consequently Figs. 2, 5, 8 and 9 characterize the sampled subset, not necessarily the full transition network. Please add such checks or substantially weaken the abstract's 'require a network description' claim.
  2. [Fig. 5, Sec. III] The key quantitative evidence for strong interactions excludes the most supportive data point: 'a data point in our a-Si sample at a quenchrate of 10^10 K/s with an energy barrier∼4 eV and a ΔE_max∼4 eV at Δr_min^rms∼1 Å has been excluded from the average.' Since this is precisely a short-distance, large-deviation event, its exclusion without stated criteria biases the interaction claim downward. Please report the mean including this point and justify the exclusion; the Discussion's 'up to ∼0.25 eV' also appears inconsistent with the excluded point.
  3. [Eq. (7), Sec. IV] The annealing mechanism claim is supported by an artificial parameter α that linearly rescales energy minima and redefines barriers while keeping their average fixed. This is not a physical annealing protocol. The match α=0.8 to a 10^10 K/s quench in Fig. 6 is made on the energy distribution only, while the pronounced loss reduction in Fig. 8 comes from pushing α→0, far outside the range realized by the quench-rate comparison. The actual 10^10 vs 10^11 comparison is suggestive but limited. Please calibrate α against independent slow-annealing simulations or otherwise demonstrate that the α deformation preserves the network correlations relevant for loss; otherwise the 'annealed thermodynamic pathways' conclusion is not established.
minor comments (6)
  1. [Abstract and Sec. V] The language 'strongly interacting transitions' is stronger than the evidence: loop energy asymmetries are consistent with interactions but do not directly measure interaction forces. Please phrase as 'consistent with interactions'.
  2. [Throughout] Typos and formatting: 'disorderd' (Intro), 'nudged elestic band' (Sec. III), 'quenchrates' (Sec. III), and inconsistent spacing in '10 10K/s' and 'a-Si anda-TiO2'.
  3. [Fig. 9] One data point at 180 K for α=0 is excluded ('Steiner tree length of 479 nodes, PR of 474, two node length of 40') without justification. Report it or justify the exclusion.
  4. [Sec. III] The notation Δr_min^rms is introduced via an expression for Δr_min^2; clarify whether the plotted quantity is the square root of that expression.
  5. [Fig. 6] The match between α=0.8 and the 10^10 K/s quench is asserted but no quantitative metric (e.g., Kolmogorov-Smirnov distance) is given.
  6. [Eq. (3)] The meaning of V_ij should be stated explicitly (barrier relative to what reference energy) so that the reader can connect Eq. (3) with the redefinition in Eq. (7).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: load-bearing claims are supported by new simulation measurements, and the self-cited loss formula is a derived result rather than a fitted input.

full rationale

The paper does not satisfy the quoted-reduction test for circularity. The central network-loss formula Eq. (5) is taken from the authors' own Ref. [26], but it is presented as a result of the stated master equation, detailed-balance rate matrix, Arrhenius rates, and linear response (Eqs. (2)-(4)); it is not a redefinition of the data or of the target claim. The abstract's claim that transitions 'require a general network description' is supported by new, independent measurements in this paper: clustering of transition locations (Fig. 2), four-state-loop asymmetry and its distance dependence (Fig. 5), cubic network structures with nine real-space locations (Fig. 4), and eigenvector path-length analysis (Figs. 9-11). None of these is fitted to Q^-1 or to an external loss curve. The artificial annealing parameter alpha is defined by Eq. (7) and alpha=0.8 is calibrated to match the 10^10 K/s energy-minimum distribution, not to experimental loss data; the alpha->0 trend is a model extrapolation, so it is self-consistency rather than a fitted prediction. The thermal-search sampling protocol (600 K, 200 ps) could miss rare or high-barrier transitions, and the paper itself acknowledges needing larger systems and better-annealed samples in Section V; however, this is a representativeness/correctness limitation, not a circular reduction of the output to the input. No uniqueness theorem, ansatz-by-citation, or renaming step is invoked as the load-bearing argument.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

No new physical entities are postulated. The network, Steiner trees, mountain structures, and the alpha transformation are analysis constructs or parameters, not new forces, particles, or dimensions.

free parameters (4)
  • k0 (bare transition rate) = assumed equal for all transitions
    Section II: 'we assume equal bare transition rates k_ij = k0 for all transitions in our numerical calculations'; sets the absolute rate scale and therefore the position and height of loss peaks; cited to Ref [31] but not measured here.
  • alpha (artificial annealing parameter) = alpha=0.8 used to match 10^10 quench; alpha=0 gives delta-function energy distribution
    Eq. (7) scales inherent-structure energies while holding average barrier fixed; central tool for attributing loss reduction to energy-minima flattening; the choice alpha=0.8 in Fig. 6 is tuned to mimic a slower quench.
  • gamma0 (strain coupling prefactor) = not specified in this paper
    Appears in Eq. (5) Q^-1_CN = beta N gamma0^2/(4 V_C) ...; inherited from Ref [26], controls absolute loss magnitude; comparisons here are mostly relative or trend-based.
  • transition filtering thresholds (participation ratio, max atomic displacement) = not reported
    Section III: 'filter them based on participation ratio and maximum atomic displacement to remove unlikely candidates'; exact thresholds affect which transitions enter the network and hence all derived quantities.
assumptions (6)
  • domain assumption The master equation (2) with Arrhenius transition rates (3) and detailed balance describes the dynamics of the energy-landscape network.
    Section II; standard for this class of models but an idealization of atomistic dynamics.
  • domain assumption Equal bare transition rate k0 for all transitions.
    Section II; made for simplicity and justified by a prior a-Si study [31], not revalidated for a-TiO2.
  • ad hoc to paper Equation (5) for Q^-1_CN from Ref [26] is correct within linear response.
    The central loss formula is taken from the authors' own previous paper and not re-derived or independently benchmarked here.
  • domain assumption Thermal search trajectories at 600 K for 200 ps with 100/200 trajectories per sample capture a representative subset of the transition network.
    Section III; finite sampling may miss rare/high-barrier transitions, affecting connectivity, path lengths, and loss predictions.
  • domain assumption The interatomic potentials (Tersoff for a-Si, Buckingham for a-TiO2) capture the important energy barriers.
    Section III; different potentials can change defect populations (cf. [24]), so barrier statistics may be potential-dependent.
  • ad hoc to paper Artificial annealing transformation Eq. (7) preserves the physics relevant to mechanical loss while changing only the spread of energy minima.
    Eq. (7) defines E^alpha=alpha E and adjusts barriers to keep V_avg fixed; this is a constructed transformation, not a physically realized annealing protocol, so conclusions about annealing mechanisms rest on its validity.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Mechanical loss in amorphous solids: spatial correlations, interacting transitions, and annealed thermodynamic pathways." pith.science (2026). https://pith.science/paper/4G5L4YQT

@misc{pith2026260717434,
  author       = {Pith},
  title        = {Pith review of: Mechanical loss in amorphous solids: spatial correlations, interacting transitions, and annealed thermodynamic pathways},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4G5L4YQT}},
  note         = {Machine review of arXiv:2607.17434}
}
read the original abstract

The disordered and defect-rich structure of amorphous solids forms heterogeneous, high-dimensional energy landscapes. Such an energy landscape can be described by a discrete-state network of transitions between stable energy minima. Under low-frequency mechanical oscillations, defect-mediated, thermally activated transitions provide a microscopic mechanism for mechanical dissipation that are the dominant cause of mechanical loss in the mirror coatings of ground based gravitational waves detectors. Using molecular simulations, we find spatially correlated and strongly interacting transitions that require a general network description instead of a superposition of independent two-level systems as traditionally assumed. An annealing study combined with an analysis of dominant relaxation paths in the energy landscape reveals novel mechanisms for reducing room temperature mechanical loss.

Figures

Figures reproduced from arXiv: 2607.17434 by the authors.

Figure 2
Figure 2. FIG. 2. Radial distribution function of the mean transition [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 1
Figure 1. FIG. 1. Top: network of transitions involved in every tran [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 4
Figure 4. FIG. 4. Cubic network structure present in the full network [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (6 more)
Figure 5
Figure 5. Figure 5: FIG. 5. The maximum energy deviation from the TLS model [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6. Summary of the impact of quench rate on the energy [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7. Fraction of the network that is connected (number of [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. Impact of the energy landscape on the mechanical [PITH_FULL_IMAGE:figures/full_fig_p006_8.png]
Figure 10
Figure 10. Figure 10: FIG. 10. Energy landscape (left) along the backbone [PITH_FULL_IMAGE:figures/full_fig_p007_10.png]
Figure 11
Figure 11. Figure 11: FIG. 11. Energy landscape (left) along the backbone [PITH_FULL_IMAGE:figures/full_fig_p008_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 2 linked inside Pith

  1. [26]

    Blaber, D

    S. Blaber, D. Bruns, and J. Rottler, Connected network model for the mechanical loss of amorphous materials, Proc. Natl. Acad. Sci. U.S.A123, e2513161123 (2026)

    2026

  2. [1]

    J. Steinlechner, Development of mirror coatings for gravitational-wave detectors, Philosophical Transactions of the Royal Society A: Mathematical, Physical and En- gineering Sciences376, 20170282 (2018)

    2018

  3. [2]

    Vajente, L

    G. Vajente, L. Yang, A. Davenport, M. Fazio, A. Ananyeva, L. Zhang, G. Billingsley, K. Prasai, A. Markosyan, R. Bassiri,et al., Low mechanical loss tio 2: Geo 2 coatings for reduced thermal noise in grav- itational wave interferometers, Physical Review Letters 127, 071101 (2021)

    2021

  4. [3]

    Aspelmeyer, T

    M. Aspelmeyer, T. J. Kippenberg, and F. Marquardt, Cavity optomechanics, Reviews of Modern Physics86, 1391 (2014)

    2014

  5. [4]

    Milkus and A

    R. Milkus and A. Zaccone, Atomic-scale origin of dy- namic viscoelastic response and creep in disordered solids, Phys. Rev. E95, 023001 (2017)

    2017

  6. [5]

    Zaccone,Theory of Disordered Solids

    A. Zaccone,Theory of Disordered Solids. From Atom- istic Dynamics to Mechanical, Vibrational, and Thermal Properties(Springer, 2023)

    2023

  7. [6]

    Damart, A

    T. Damart, A. Tanguy, and D. Rodney, Theory of har- monic dissipation in disordered solids, Physical Review B95, 054203 (2017)

    2017

  8. [7]

    Damart and D

    T. Damart and D. Rodney, Atomistic study of two- level systems in amorphous silica, Physical Review B97, 014201 (2018)

    2018

Show all 46 references
  1. [8]

    W. A. Phillips, Two-level states in glasses, Reports on Progress in Physics50, 1657 (1987)

    1987

  2. [9]

    X. Liu, D. R. Queen, T. H. Metcalf, J. E. Karel, and F. Hellman, Hydrogen-free amorphous silicon with no tunneling states, Phys. Rev. Lett.113, 025503 (2014)

    2014

  3. [10]

    Steinlechner, I

    J. Steinlechner, I. W. Martin, J. Hough, C. Kr¨ uger, S. Rowan, and R. Schnabel, Thermal noise reduction and absorption optimization via multimaterial coatings, Physical Review D91, 042001 (2015)

    2015

  4. [11]

    Molina-Ruiz, R

    M. Molina-Ruiz, R. Zhou, A. Markosyan, R. Bassiri, M. Fejer, A. Ananyeva, S. Tait, G. Vajente, A. Dav- enport, C. Menoni,et al., Revealing the role of hy- drogen in reducing optical absorption and mechani- cal loss in magnetron-sputtered amorphous silicon for gravitational-wav...

    2025

  5. [12]

    I. W. Martin, E. Chalkley, R. Nawrodt, H. Armandula, R. Bassiri, C. Comtet, M. Fejer, A. Gretarsson, G. Harry, D. Heinert,et al., Comparison of the temperature depen- dence of the mechanical dissipation in thin films of ta2o5 and ta2o5 doped with tio2, Classical and Quantum Gr...

    2009

  6. [13]

    I. W. Martin, R. Bassiri, R. Nawrodt, M. Fejer, A. Gre- tarsson, E. Gustafson, G. Harry, J. Hough, I. MacLaren, S. Penn,et al., Effect of heat treatment on mechani- cal dissipation in ta2o5 coatings, Classical and Quantum Gravity27, 225020 (2010)

    2010

  7. [14]

    I. W. Martin, R. Nawrodt, K. Craig, C. Schwarz, R. Bassiri, G. Harry, J. Hough, S. Penn, S. Reid, R. Ro- bie,et al., Low temperature mechanical dissipation of an ion-beam sputtered silica film, Classical and Quantum Gravity31, 035019 (2014)

    2014

  8. [15]

    Kuo, H.-W

    L.-C. Kuo, H.-W. Pan, C.-L. Chang, and S. Chao, Low cryogenic mechanical loss composite silica thin film for low thermal noise dielectric mirror coatings, Optics let- ters44, 247 (2019)

    2019

  9. [16]

    Khadka, A

    S. Khadka, A. Markosyan, K. Prasai, A. Dana, L. Yang, S. Tait, I. Martin, C. Menoni, M. Fejer, and R. Bassiri, Cryogenic mechanical loss of amorphous germania and titania-doped germania thin films, Classical and Quan- tum Gravity40, 205002 (2023)

    2023

  10. [17]

    Molina-Ruiz, Y

    M. Molina-Ruiz, Y. Rosen, H. Jacks, M. Abernathy, T. Metcalf, X. Liu, J. DuBois, and F. Hellman, Origin of mechanical and dielectric losses from two-level systems in amorphous silicon, Physical Review Materials5, 035601 (2021)

    2021

  11. [18]

    Molina-Ruiz, A

    M. Molina-Ruiz, A. Markosyan, R. Bassiri, M. Fe- jer, M. Abernathy, T. Metcalf, X. Liu, G. Vajente, A. Ananyeva, and F. Hellman, Hydrogen-induced ul- tralow optical absorption and mechanical loss in amor- phous silicon for gravitational-wave detectors, Physical Review Letters1...

    2023

  12. [19]

    Puosi, F

    F. Puosi, F. Fidecaro, S. Capaccioli, D. Pisignano, and D. Leporini, In silico broadband mechanical spec- troscopy of amorphous tantala, Physical Review Re- search1, 033121 (2019)

    2019

  13. [20]

    J. P. Trinastic, R. Hamdan, C. Billman, and H.-P. Cheng, Molecular dynamics modeling of mechanical loss in amor- phous tantala and titania-doped tantala, Physical Review B93, 014105 (2016)

    2016

  14. [21]

    Prasai, J

    K. Prasai, J. Jiang, A. Mishkin, B. Shyam, S. Angelova, I. Birney, D. A. Drabold, M. Fazio, E. K. Gustafson, G. Harry,et al., High precision detection of change in in- termediate range order of amorphous zirconia-doped tan- tala thin films due to annealing, Physical review let...

    2019

  15. [22]

    Jiang, A

    J. Jiang, A. S. Mishkin, K. Prasai, R. Zhang, M. Yazback, R. Bassiri, M. M. Fejer, and H.-P. Cheng, Analysis of two-level systems and mechanical loss in amorphous zro2- doped ta2o5 by non-cage-breaking and cage-breaking transitions, The Journal of Chemical Physics154(2021)

    2021

  16. [23]

    L´ evesque, S

    C. L´ evesque, S. Roorda, F. Schiettekatte, and N. Mousseau, Internal mechanical dissipation mecha- nisms in amorphous silicon, Physical Review Materials 6, 123604 (2022)

    2022

  17. [24]

    Girard, C

    R. Girard, C. L´ evesque, N. Mousseau, and F. Schiet- tekatte, Activation-relaxation technique study of two- level system impact on internal dissipation using dft- based moment tensor potential, Physical Review Materi- als10, 053601 (2026)

    2026

  18. [25]

    Prasai, J

    K. Prasai, J. Jiang, K. Lee, L. Yang, M. Chicoine, S. Khadka, A. Markosyan, A. Mehta, C. Menoni, S. Pa- tel,et al., Atomic structure of amorphous optical coatings of tio 2 doped geo 2 by grazing-incidence total x-ray scat- tering measurements, arXiv preprint arXiv:2605.27462 (2026)

    2026 arXiv

  19. [27]

    Levin, Internal thermal noise in the ligo test masses: A direct approach, Physical Review D57, 659 (1998)

    Y. Levin, Internal thermal noise in the ligo test masses: A direct approach, Physical Review D57, 659 (1998)

    1998

  20. [28]

    Gardiner,Handbook of Stochastic Methods for Physics, Chemistry and the Natural Sciences, 2nd ed

    C. Gardiner,Handbook of Stochastic Methods for Physics, Chemistry and the Natural Sciences, 2nd ed. (Springer, 1985)

    1985

  21. [29]

    Angelani, G

    L. Angelani, G. Parisi, G. Ruocco, and G. Viliani, Con- nected network of minima as a model glass: Long time dynamics, Physical Review Letters81, 4648 (1998)

    1998

  22. [30]

    Banerjee and C

    S. Banerjee and C. Dasgupta, Characterization of the dynamics of glass-forming liquids from the properties of the potential energy landscape, Physical Review E—Statistical, Nonlinear, and Soft Matter Physics85, 021501 (2012)

    2012

  23. [31]

    Valiquette and N

    F. Valiquette and N. Mousseau, Energy landscape of re- laxed amorphous silicon, Phys. Rev. B68, 125209 (2003)

    2003

  24. [32]

    Tersoff, Modeling solid-state chemistry: Interatomic potentials for multicomponent systems, Physical review B39, 5566 (1989)

    J. Tersoff, Modeling solid-state chemistry: Interatomic potentials for multicomponent systems, Physical review B39, 5566 (1989)

    1989

  25. [33]

    Matsui and M

    M. Matsui and M. Akaogi, Molecular dynamics simula- tion of the structural and physical properties of the four polymorphs of tio2, Molecular Simulation6, 239 (1991)

    1991

  26. [34]

    A. P. Thompson, H. M. Aktulga, R. Berger, D. S. Bolin- tineanu, W. M. Brown, P. S. Crozier, P. J. In’t Veld, A. Kohlmeyer, S. G. Moore, T. D. Nguyen,et al., Lammps-a flexible simulation tool for particle-based ma- terials modeling at the atomic, meso, and continuum scales, Com...

    2022

  27. [35]

    Khomenko, C

    D. Khomenko, C. Scalliet, L. Berthier, D. R. Reichman, and F. Zamponi, Depletion of two-level systems in ultra- stable computer-generated glasses, Physical review let- ters124, 225901 (2020)

    2020

  28. [36]

    Luckabauer, T

    M. Luckabauer, T. Hayashi, H. Kato, and T. Ichitsubo, Decreasing activation energy of fast relaxation processes in a metallic glass during aging, Physical Review B99, 140202 (2019)

    2019

  29. [37]

    Prasai, R

    K. Prasai, R. Bassiri, H.-P. Cheng, and M. M. Fejer, Annealing-induced changes in the atomic structure of amorphous silica, germania, and tantala using acceler- ated molecular dynamics, physica status solidi (b)258, 2000519 (2021)

    2021

  30. [38]

    Queen, X

    D. Queen, X. Liu, J. Karel, H. Jacks, T. Metcalf, and F. Hellman, Two-level systems in evaporated amorphous silicon, Journal of Non-Crystalline Solids426, 19 (2015)

    2015

  31. [39]

    C. Y. Clare, Interacting defect model of glasses: Why do phonons go so far?, Physical review letters63, 1160 (1989)

    1989

  32. [40]

    H. M. Carruzzo, E. R. Grannan, and C. Y. Clare, Nonequilibrium dielectric behavior in glasses at low tem- peratures: Evidence for interacting defects, Physical Re- view B50, 6685 (1994)

    1994

  33. [41]

    J. M. Martinis, K. B. Cooper, R. McDermott, M. Stef- fen, M. Ansmann, K. D. Osborn, K. Cicak, S. Oh, D. P. Pappas, R. W. Simmonds, and C. C. Yu, Decoherence in josephson qubits from dielectric loss, Phys. Rev. Lett. 11 95, 210503 (2005)

    2005

  34. [42]

    P. V. Klimov, J. Kelly, Z. Chen, M. Neeley, A. Megrant, B. Burkett, R. Barends, K. Arya, B. Chiaro, Y. Chen, et al., Fluctuations of energy-relaxation times in super- conducting qubits, Physical review letters121, 090502 (2018)

    2018

  35. [43]

    M¨ uller, J

    C. M¨ uller, J. H. Cole, and J. Lisenfeld, Towards under- standing two-level-systems in amorphous solids: insights from quantum circuits, Reports on Progress in Physics 82, 124501 (2019)

    2019

  36. [44]

    B´ ejanin, C

    J. B´ ejanin, C. Earnest, A. Sharafeldin, and M. Mariantoni, Interacting defects generate stochastic fluctuations in superconducting qubits, Physical Review B104, 094106 (2021)

    2021

  37. [45]

    Q. Wang, S. M. G´ omez, J. S. Salcedo-Gallo, R. Leibovitz, J. Freeman, S. A. Agnew, S. Bedkihal, W. J. Scheideler, and M. Fitzpatrick, Evidence of memory effects in the dy- namics of two-level system defect ensembles using broad- band, cryogenic transient dielectric spectrosco...

    2025 arXiv

  38. [46]

    Connected Network Model for the Mechanical Loss of Amorphous Materials

    Q. Wang, J. S. Salcedo-Gallo, S. Bedkihal, T. Xia, M. W. Olszewski, V. Fatemi, and M. Fitzpatrick, Spectroscopy and coherent control of two-level system defect ensembles using a broadband 3d waveguide, Materials for Quantum Technology5, 045201 (2025). 1 Supplemental Material f...

    2025

Pith tools

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