REVIEW 2 major objections 6 minor 45 references
Phononic frictional losses of a particle crossing a crystal: linear-response theory
T0 review · 2 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash
Pith's one-line read This paper claims that the phononic friction on a point particle sliding through a harmonic crystal can be written as an explicit integral over phonon modes, and that this analytic force quantitatively matches molecular-dynamics…
desk verdict A careful, genuinely useful extension of phononic friction theory from 1D to 2D and 3D, with honest internal validation; the one-phonon truncation is the only unquantified step and is likely harmless at the parameters used. 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 engine is the imaginary part of the retarded density-density response function $\chi^R_{nn}(Q,Q+G_\perp,\omega)$ of the harmonic crystal, evaluated in the one-phonon approximation—keeping only single-phonon exchanges in the density correlations. Its imaginary part, Eq. (C30), feeds the generalized power expression Eq. (18); the resonance condition $\omega_\lambda(Q) = Q\cdot v_{SL}$ selects the phonon modes that absorb energy, while the conservative-crystal version carries a denominator $|v_{SL} - v_\lambda(Q)|$ whose zeros produce shock-wave singularities when the slider speed matches a phonon group velocity. Lattice symmetries reduce the $Q$ integration to one octant (3D) or quadrant (2D), and a finite phonon lifetime $\gamma$ broadens the delta functions into Lorentzians, making the integral regular and allowing all $Q$ points to contribute. The slider–crystal coupling enters through the Fourier transform of a regularized Lennard-Jones potential, chosen so that $\tilde V(q)$ decays rapidly and the integral converges.
What would settle it
Numerically evaluate the exact retarded density-density response of the same harmonic crystal without the one-phonon approximation—for instance by evaluating Eq. (C18) with the full factor $e^{\phi_j^r}\sin(\phi_j^i)$ instead of $\phi_j^i$—insert it into Eq. (18), and compare with Eq. (1). A sizable discrepancy in the $Q$ regions where the integral is concentrated would falsify the approximation; alternatively, repeat the MD comparison at coupling $\varepsilon$ large enough that $F/\varepsilon^2$ deviates from the analytic curve, indicating multi-phonon effects.
Extended reading notes
Core claim
On the paper's own terms, the central discovery is Eq. (1): for a slider moving along a high-symmetry channel of a simple-cubic harmonic crystal, the friction force is a sum over reciprocal-lattice vectors perpendicular to the velocity and an integral over one octant of reciprocal space of $\tilde V(|Q|)\tilde V(|Q+G_\perp|)$ times polarization projections $[Q\cdot\epsilon_\lambda(Q)]\,[(Q+G_\perp)\cdot\epsilon_\lambda(Q)]$, Debye–Waller factors, and a double-Lorentzian weight $L(Q,v_{SL},\gamma)$ that peaks at the resonance $\omega_\lambda(Q)=Q_x v_{SL}$. The 2D analogue is Eq. (41). The friction is second order in the coupling strength $\varepsilon$, and in the $T=0$ weak-coupling limit the Planck constant drops out, leaving a purely classical expression. The paper validates both formulas by direct comparison with molecular-dynamics simulations in the same parameter regime: the analytic curves follow the simulated friction across the velocity range, including the positions, shapes, and branch decomposition of the friction peaks. In the dissipationless limit only phonons satisfying the resonance condition contribute, and friction peaks are enhanced where the slider speed matches a phonon group velocity (the shock-wave condition); in 3D, unlike in the earlier 1D model, resonant channels remain open at every speed, so friction never vanishes.
Load-bearing premise
Everything hinges on the one-phonon approximation in Appendix C, which assumes the slider loses energy by creating one phonon at a time; if multi-phonon events dominate the wave vectors that contribute most to the integral, the predicted friction would be wrong even in a perfect harmonic crystal.
Editorial extensions
If this is right
- The analytic formula gives friction at any speed by evaluating one integral, avoiding the cost and finite-size drift of molecular-dynamics friction measurements, especially in 3D.
- For a conservative 3D crystal, friction stays nonzero at all slider speeds, in contrast with the 1D chain where supersonic sliding becomes frictionless.
- Friction-peak speeds are determined solely by the phonon dispersion, so changing the slider–crystal interaction alters the magnitude and the relative longitudinal/transverse contributions but not where the peaks occur.
- The approximate expression Eq. (33) yields the velocity asymptotics: friction grows linearly at low speed, peaks near the scale $d\omega_{ph}$, and decays as $v_{SL}^{-3}$ at high speed.
- The same framework directly describes energy loss of neutral particles channeled through crystals, with the caveat that very short-range interactions make the required numerical integration harder.
Reading between the lines
- Beyond the paper, the same response-function route could be extended to semi-infinite crystals; if the bulk structure survives, surface-friction experiments might show resonance and shock-wave peaks shifted by Rayleigh-wave dispersion rather than bulk phonons.
- Because peak positions depend only on the phonon dispersion, measured friction-peak velocities could be used as a phonon-spectroscopy probe of a crystal, an inverse use the authors do not discuss.
- One could test the one-phonon assumption directly by computing the exact density-density response of Eq. (C18) numerically; any large disagreement at the dominant $Q$ values would show where multi-phonon corrections enter.
- If a Dyson-like resummation extends this approach to stronger coupling, the first corrections to friction should appear at order $\varepsilon^3$ and shift the resonance peaks, a concrete target for future simulation.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper presents a linear-response theory of the kinetic friction experienced by a weak point slider moving at constant velocity through a 2D square or 3D simple-cubic harmonic crystal. Starting from the exact retarded density-density response of the lattice, the authors make a one-phonon approximation, incorporate zero-temperature Debye-Waller factors, and introduce a phenomenological phonon damping gamma to obtain explicit analytic formulas for the friction force as a function of slider velocity: Eq. (1) in 3D and Eq. (41) in 2D. The formulas decompose the friction into a sum over perpendicular reciprocal-lattice vectors and a Q-space integral over products of the interaction potential Fourier transform, Debye-Waller factors, polarization projections, and a velocity-dependent Lorentzian line shape, with the conservative-crystal limit reducing to resonant surface integrals satisfying omega_lambda(Q)=Q·v_SL. The authors evaluate these expressions numerically with publicly available code and compare them with Langevin molecular dynamics simulations of the same harmonic-lattice model with the same parameters and no fitted quantities; they report close agreement in magnitudes, velocity dependence, and peak positions across three decades of velocity, and they provide physical understanding of branch-resolved contributions, umklapp-enabled resonance at all speeds, and shock-wave peaks. An order-of-magnitude approximation is also derived and tested.
Significance. If the results hold, this is a welcome extension of the authors' earlier 1D analytic friction work to 2D and 3D channeling geometries, and it is one of the few analytic, fitting-free benchmarks for nanoscale phononic friction. The strongest assets are the care and completeness of the derivation (full appendices for phonon symmetries, response functions, Fourier transforms, and the MD protocol); the consistency of the approximations with the simulation model; the explicit falsifiable predictions (peak positions set by the phonon dispersion and independent of the slider potential shape, as verified in Figs. 9 and 11; epsilon-squared scaling verified in Fig. 6); the convergence checks of the Q integration (Nq=400 vs 800); and the public availability of the code [14]. The paper also gives physical insight into the resonance condition omega=Q·v_SL, branch-resolved friction, shock-wave singularities, and the qualitative difference from the 1D supersonic cutoff. The scope is explicitly limited to weak coupling, harmonic crystals, and commensurate channeling directions, which is appropriate for the claimed setting.
major comments (2)
- [Appendix C, Eq. (C19); Secs. IV-V] The truncation e^{phi_r} sin(phi_i) ≈ phi_i in Eq. (C19) is the only approximation in Eq. (1) that the MD comparison does not share, so the agreement cannot localize its error budget; the paper asserts its validity only qualitatively ('appropriate for not too large wave vector Q'). The authors should quantify the discarded multi-phonon terms for the parameter regime used: with zero-point displacement about 4 pm and dominant resonant |Q| of order 1-8 a^-1, |Q·u| is of order 0.01-0.07, making the relative correction from phi_r and phi_i^3/6 in the exact expression (C18) of order 10^-4 to 10^-3, hence negligible compared with the plotted discrepancies. Adding this estimate (and, if feasible, evaluating (C18) directly on the same Q grid) would close the uncertainty budget of the quantitative claim; as written, that claim carries no stated error from this step. Note also that the classical T=0 MD cannot probe the quantum multi-phonon sector of the response, so the estimate is needed as the justification for the truncation in this regime, and the same small parameter underlies the setting of the Debye-Waller factors to unity in Sec. V.
- [Sec. V, Figs. 4 and 6] The convergence statement for the 3D integrals is incomplete. The Nq=800 curve in Fig. 4 retains oscillations with amplitudes of order 0.2-0.5 in F/epsilon^2 in the range v_SL ≈ 1-2 a(K/m)^(1/2), and the text attributes the residual deviations between theory and MD to 'imperfect convergence of the integration grid' without a quantitative bound. Because the central claim is quantitative agreement with MD, the paper should provide a convergence measure (e.g., a comparison at Nq=1600, the difference between the Nq=400 and Nq=800 results as a function of v_SL, or local error bars on the theory curve) and state the resulting uncertainty in the quoted F(v_SL). Without this, the 3D agreement is presented without a numerical error bar.
minor comments (6)
- [Sec. V text; captions of Figs. 4 and 5] The initial positions for the two 3D channeling lines are written as x0 = a/2 (ex + ey) and x0 = a/2 ex in the text of Sec. V and in the figure captions; these should read a/2 (ey + ez) and a/2 ey to agree with Sec. II and with the phase factors (16)-(17).
- [Fig. 10] The caption quotes the velocity 0.2545 a(K/m)^(1/2) for peak 2 while the text of Sec. VI.A quotes 0.2535 a(K/m)^(1/2); the two values should be harmonized.
- [Appendix D] Because the MD uses a smoothly truncated slider-atom potential (cutoff 5a) while the theory uses the untruncated Fourier transform (8)/(9), a sentence estimating the neglected tail (of order 10^-4 epsilon for the adopted parameters) would remove a small residual ambiguity in the comparison.
- [Sec. V] A quantitative statement of the agreement, e.g., the maximum or RMS relative deviation between the Nq=800 curve and the MD points in Figs. 4-5, would make the 'remarkable accord' claim more precise than the visual comparison alone.
- [Sec. VI.A] The small deviations between theory and MD at the lowest velocities in Fig. 9 are acknowledged but not explained; a brief remark on whether they reflect the onset of the first-Born constant-velocity assumption would be useful.
- [Sec. IV.B] In the sentence following Eq. (25), the claim that a resonant surface exists for every speed relies on the unrestricted (umklapp) Q integration; this is correct, but it is worth stating explicitly for readers comparing with the 1D supersonic cutoff.
Circularity Check
No significant circularity: the friction formula is derived from the model Hamiltonian and compared, not fitted, to independent MD simulations.
full rationale
The paper's central claim is the analytic friction formula Eq. (1) (and its 2D analogue Eq. (41)), obtained from the Hamiltonian of Sec. II through linear-response theory. The derivation chain is self-contained: the dissipated power is expressed in terms of the retarded density-density response in Eq. (11), simplified by symmetry to Eq. (18), and then evaluated under the one-phonon approximation of Eq. (C19) to reach Eqs. (C26) and (C30), which feed the final formulas. At no point is the target friction force assumed as an input, nor is any parameter fitted to the simulated friction curves. The model parameters (m, K, K', a, sigma, d, epsilon, gamma) are chosen a priori and used consistently in both the analytic evaluation and the MD simulations, so the MD comparison is an independent numerical check of the same physical model, not a circular validation. The one-phonon truncation in Eq. (C19) is an approximation whose quantitative accuracy is not independently bounded in the paper, but that is a correctness or robustness concern, not circularity: the derivation does not redefine the target in terms of itself. Self-citations to the earlier 1D work (Refs. [10,11]) provide motivation and an analogous framework, but the 2D/3D response functions and final integrals are derived explicitly in Appendices C-F and do not depend on an unverified self-cited uniqueness or ansatz. Therefore no circular step meeting the evidence standard is present, and the appropriate score is 0.
Assumptions & free parameters
free parameters (5)
- sigma =
0.55 a
- d =
0.75 a (0.9 a in Fig. 11)
- gamma =
0.2 (K/m)^{1/2} in 3D; 0.05 (K/m)^{1/2} in 2D
- epsilon =
5e-4 K a^2 (MD), analytic results reported per epsilon^2
- K' =
K/2
assumptions (6)
- domain assumption Harmonic crystal with first- and second-neighbor springs (K and K'=K/2) on a square or simple-cubic lattice.
- domain assumption First Born approximation: the slider moves at constant velocity, x_SL(t) = x0 + v_SL t.
- domain assumption One-phonon approximation: e^{phi_j^r} sin(phi_j^i) approx phi_j^i.
- ad hoc to paper Phenomenological dissipation: a uniform imaginary part i*gamma/2 is added to all phonon frequencies.
- domain assumption Debye-Waller factors are set to unity at T=0.
- domain assumption Infinite perfect crystal, no surface effects, channeling along a high-symmetry (100) direction.
Cite this review
Pith. "Pith review of Phononic frictional losses of a particle crossing a crystal: linear-response theory." pith.science (2026). https://pith.science/paper/HVQDGEQM
@misc{pith2026250108239,
author = {Pith},
title = {Pith review of: Phononic frictional losses of a particle crossing a crystal: linear-response theory},
year = {2026},
howpublished = {\url{https://pith.science/paper/HVQDGEQM}},
note = {Machine review of arXiv:2501.08239}
}
read the original abstract
We address weak-coupling frictional sliding with phononic dissipation by means of analytic many-body techniques. Our model consists of a particle (the "slider") moving through a two- or three-dimensional crystal and interacting weakly with its atoms, and therefore exciting phonons. By means of linear-response theory we obtain explicit expressions for the friction force slowing down the slider as a function of its speed, and compare them to the friction obtained by simulations, demonstrating a remarkable accord.
Figures
Figures from the paper (6 more)
Reference graph
Works this paper leans on
-
[14]
E. Panizon, G. E. Santoro, E. Tosatti, G. Riva, and N. Manini, Analytic understanding and control of dy- namical friction, Phys. Rev. B 97, 104104 (2018)
work page 2018
-
[1]
Conservative crystal We start from a regular conservative crystal. We ex- press the retarded LR function χR nn(x, x′, t− t′) ≡ −i ℏ θ(t − t′)⟨[n(x, t− t′), n(x′, 0)]⟩ (C1) in its Fourier representation: χR nn(Q, Q + G⊥, ω) (C2) = lim V →∞ − i ℏ Z dt θ(t)eiωt 1 V Z V d3x Z V d3x′ × e−iQ·xei(Q+G⊥)·x′ ⟨[n(x, t), n(x′, 0)]⟩ = lim V →∞ − i ℏV Z dt θ(t)eiωt × ⟨...
-
[2]
At the closer approach distance, the repulsive region r < σ= 0 .55 a of the in- teraction potential, Fig. 2, is involved: as a result, the slider-crystal collisions excite phonons more effectively than the smoother attractive tail. A. An approximate expression Since the evaluation of Eq. (1) requires a relatively cumbersome integration, a simpler approxim...
-
[3]
A. Einstein, ¨Uber die von der molekularkinetischen the- orie der w¨ arme geforderte bewegung von in ruhenden fl¨ ussigkeiten suspendierten teilchen, Ann. Phys.322, 549 (1905)
work page 1905
-
[4]
atomic form fac- tor of the slider
Dissipative crystal The next step is the introduction of a finite phonon lifetime. For this purpose, we add a small imaginary part to the frequency: ω → ω + iγ/2. This modification introduces a uniform exponential decay with rate γ/2 to all phonon modes, corresponding to a lifetime 2 /γ. With this phenomenological modification, we rewrite the re- tarded o...
-
[5]
Y. Tao, Z. Wei, Y. Dong, Z. Duan, Y. Kan, Y. Zhang, and Y. Chen, Phononic dynamics in sliding friction, Phys. Rev. B 108, 214313 (2023)
work page 2023
- [6]
-
[7]
L. Landau and E. Lifshitz, Electrodynamics of Continu- ous Media (Pergamon, Oxford, 1984) p. Ch. 14
work page 1984
Show all 45 references
-
[8]
Callen and T
H. Callen and T. Welton, Irreversibility and generalized 23 noise, Phys. Rev. 83, 34 (1951)
1951
-
[9]
Kubo, The fluctuation-dissipation theorem, Rep
R. Kubo, The fluctuation-dissipation theorem, Rep. Prog. Phys. 29, 255 (1966)
1966
-
[10]
E. D. Smith, M. O. Robbins, and M. Cieplak, Friction on adsorbed monolayers, Phys. Rev. B 54, 8252 (1996)
1996
-
[11]
Panizon, G
E. Panizon, G. Santoro, E. Tosatti, G. Riva, and N. Manini, Erratum: Analytic understanding and con- trol of dynamical friction, Phys. Rev. B 109, 219904(E) (2024)
2024
-
[12]
Pines, Elementary excitations in solids (Perseus, Reading, 1999)
D. Pines, Elementary excitations in solids (Perseus, Reading, 1999)
1999
-
[13]
Guerra, U
R. Guerra, U. Tartaglino, A. Vanossi, and E. Tosatti, Ballistic nanofriction, Nat. Mater. 9, 634 (2010)
2010
-
[15]
= 1√ 2 a K m 1/2# , (A5) v2D L = a K + K ′ m 1/2
We verified that, thanks to the rapid decay of the Fourier transform of the potential and the smoothness of the integrand function, the evaluated friction force does not change visibly if these truncation limits and the grid fineness are improved. Figure 9 compares the frictio...
-
[16]
Gemmell, Channeling and related effects in the motion of charged particles through crystals, Rev
D. Gemmell, Channeling and related effects in the motion of charged particles through crystals, Rev. Mod. Phys. 46, 129 (1974)
1974
-
[17]
G. F. Giuliani and G. Vignale, Quantum Theory of the Electron Liquid (Cambridge Univ. Press, 2005)
2005
-
[18]
unimi.it/dataset.xhtml?persistentId=doi: 10.13130/RD_UNIMI/ZV3MH3 (2025)
Support data and computer codes are publicly avail- able at the following URL: https://dataverse. unimi.it/dataset.xhtml?persistentId=doi: 10.13130/RD_UNIMI/ZV3MH3 (2025)
2025 doi
-
[19]
van Hove, The occurrence of singularities in the elastic frequency distribution of a crystal, Phys
L. van Hove, The occurrence of singularities in the elastic frequency distribution of a crystal, Phys. Rev. 89, 1189 (1953)
1953
-
[20]
Commonly channeled charged particles [7, 12, 40, 41] dis- sipate more energy through the excitation of electrons than through the generation of phonons that is addressed by the present theory
-
[21]
S. V. Kozhevnikov, Y. N. Khaidukov, F. Ott, and F. Radu, Channeling of neutrons in the potential well of a planar waveguide, J. Exp. Theor. Phys. 126, 592 (2018)
2018
-
[22]
V. G. Baryshevskii and A. M. Zaitseva, Orientational ef- fects due to traveling of fast neutrons and hard γ-quanta through crystals, phys. stat. sol. b 157, 129 (1990)
1990
-
[23]
Korotchenko and Y
K. Korotchenko and Y. P. Kunashenko, Neutrons planar channeling in crystals, Nuovo Cimento 34, 537 (2011)
2011
-
[24]
B´ ee,Quasielastic Neutron Scattering (Hilger, Bristol, 1988)
M. B´ ee,Quasielastic Neutron Scattering (Hilger, Bristol, 1988)
1988
-
[25]
S. A. Adelman and J. D. Doll, Generalized Langevin equation approach for atom/solid-surface scattering: General formulation for classical scattering off harmonic solids, J. Chem. Phys. 64, 2375 (1976)
1976
-
[26]
O. M. Braun, M. Peyrard, V. Bortolani, A. Franchini, and A. Vanossi, Transition from smooth sliding to stick- slip motion in a single frictional contact, Phys. Rev. E 72, 056116 (2005)
2005
-
[27]
Kajita, H
S. Kajita, H. Washizu, and T. Ohmori, Approach of semi- infinite dynamic lattice Green’s function and energy dis- sipation due to phonons in solid friction between com- mensurate surfaces, Phys. Rev. B 82, 115424 (2010)
2010
-
[28]
Kajita, Green’s function nonequilibrium molecular dy- namics method for solid surfaces and interfaces, Phys
S. Kajita, Green’s function nonequilibrium molecular dy- namics method for solid surfaces and interfaces, Phys. Rev. E 94, 033301 (2016)
2016
-
[29]
M. Lee, R. L. C. Vink, C. A. Volkert, and M. Kr¨ uger, Noncontact friction: Role of phonon damping and its nonuniversality, Phys. Rev. B 104, 174309 (2021)
2021
-
[30]
Y. Tao, Y. Dong, Y. Kan, Z. Wei, Y. Zhang, C. Liu, and Y. Chen, Phononic doppler effect in sliding friction, Phys. Rev. B 109, 205410 (2024)
2024
-
[31]
Y. Tao, C. Sun, S. Huang, Y. Dong, Y. Kan, Z. Wei, Y. Zhang, Z. Ni, and Y. Chen, Phononic perspective of anisotropic friction: Universal laws from the frictional phonon spectrum, Phys. Rev. E 111, 055506 (2025)
2025
-
[32]
Kajita, A
S. Kajita, A. Pacini, G. Losi, N. Kikkawa, and M. C. Righi, Accurate multiscale simulation of frictional inter- faces by quantum mechanics/Green’s function molecular dynamics, J. Chem. Theory Comput. 19, 5176 (2023)
2023
-
[33]
Y. Dong, Q. Li, and A. Martini, Molecular dynamics sim- ulation of atomic friction: A review and guide, J. Vac. Sci. Technol. A 31, 030801 (2013)
2013
-
[34]
Vanossi, N
A. Vanossi, N. Manini, M. Urbakh, S. Zapperi, and E. Tosatti, Modeling friction: from nanoscale to mesoscale, Rev. Mod. Phys. 85, 529 (2013)
2013
-
[35]
Kamenev, Field Theory of Non-Equilibrium Systems (Cambridge Univ
A. Kamenev, Field Theory of Non-Equilibrium Systems (Cambridge Univ. Press, Cambridge, 2011)
2011
-
[36]
Mermin, A short simple evaluation of expressions of the Debye-Waller form, J
N. Mermin, A short simple evaluation of expressions of the Debye-Waller form, J. Math. Phys. 7, 1038 (1966)
1966
-
[37]
Ashcroft and M
N. Ashcroft and M. Mermin, Solid State Physics (Holt- Saunders, Philadelphia, 1976)
1976
-
[38]
Y. Sang, M. Dub´ e, and M. Grant, Thermal effects on atomic friction, Phys. Rev. Lett. 87, 174301 (2001)
2001
-
[39]
O. M. Dudko, A. Filippov, J. Klafter, and M. Urbakh, Dynamic force spectroscopy: a Fokker Planck approach, Chem. Phys. Lett. 352, 499 (2002)
2002
-
[40]
Szlufarska, M
I. Szlufarska, M. Chandross, and R. Carpick, Recent ad- vances in single-asperity nanotribology, J. Phys. D 41, 123001 (2008)
2008
-
[41]
Brukman, G
M. Brukman, G. Gao, R. Nemanich, and J. Harrison, Temperature dependence of single-asperity diamond- diamond friction elucidated using AFM and MD simu- lations, J. Phys. Chem. C 112, 9358 (2008)
2008
-
[42]
Steiner, R
P. Steiner, R. Roth, E. Gnecco, A. Baratoff, S. Maier, T. Glatzel, and E. Meyer, Two-dimensional simulation of superlubricity on NaCl and highly oriented pyrolytic graphite, Phys. Rev. B 79, 045414 (2009)
2009
-
[43]
F. W. J. Olver, A. B. O. Daalhuis, D. W. Lozier, B. I. Schneider, R. F. Boisvert, C. W. Clark, B. R. Miller, B. V. Saunders, H. S. Cohl, and M. A. McClain, eds., NIST Digital Library of Mathematical Functions (Re- lease 1.1.11, 2023)
2023
-
[44]
L. R. Logan, C. S. Murthy, and G. R. Srinivasan, Elec- tronic energy loss for helium channeling in silicon, Phys. Rev. A 46, 5754 (1992)
1992
-
[45]
Scandale, M
W. Scandale, M. D’Andrea, L. Esposito, M. Garat- tini, E. Guillermain, D. Mirarchi, S. Montesano, A. Na- tochii, R. Rossi, G. I. Smirnov, V. Zhovkovska, F. Mur- tas, F. Galluccio, F. Addesa, F. Iacoangeli, Z. Islam, Y. Gavrikov, Y. Ivanov, M. Koznov, M. Malkov, L. Mal- yarenko...
2021
Reviewed August 10, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.