{"id":"59890844-47c0-4025-8a4e-fb267b5042e2","arxiv_id":"2411.13232","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":4,"one_line_summary":"The paper proposes a two-parameter friction model f = γ v^α that reproduces high-bias I-V curves of Al/Al2O3/Al junctions and identifies 137 K as a turning point for dissipated energy fraction.","lead":"The authors fit a nonlinear friction model to current-voltage data from aluminum-oxide tunnel junctions and report that the fraction of energy lost during tunneling falls with voltage but changes non-monotonically with temperature. The finding points to a possible change in dissipative behavior around 137 K, which they attribute to electron-phonon interactions.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 137 K turning point is not statistically established: γ and α vary by only ~1% and <0.4% with no error bars, and the non-monotonicity may be an artifact of the post hoc 1.2 V split.","rationale":"I read the paper as an empirical model-fitting exercise: a two-parameter dissipation model is added above 1.2 V to repair the Simmons model, and the fitted parameters are then used to compute ΔE/E. The most vulnerable link is not the admittedly unphysical real velocity inside the barrier—that is a modeling choice—but the evidential basis for the 137 K turning point. The reader's weakest assumption about imaginary velocity is a legitimate conceptual concern, but even granting the phenomenological model, the headline claim needs error bars and robustness tests. I therefore agree partially with the reader. A bootstrap or split-point sensitivity test would settle whether the turning point is real or a fitting artifact. The reader's conditional verdict remains appropriate pending that test.","tokens_in":9538,"tokens_out":4346,"duration_ms":49466,"concrete_test":"For each temperature, bootstrap the I-V residuals (or use the fit covariance) to obtain 95% confidence intervals for γ, α, and ΔE/E at E/(φ0+εF)=0.19 and V=0.95 V. Refit with the split point varied between 1.0 and 1.4 V and recompute the ΔE/E(T) curve. If the dip at 137 K is not larger than the combined uncertainties, or moves by more than ~20 K under split variation, the critical-temperature claim is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline claim is the temperature turning point in ΔE/E at 137 K (Abstract, Fig. 7b, Fig. 8). This claim depends on the fitted dissipation parameters γ(T) and α(T) through Eq. 7. The reported variations are extremely small—γ changes by ~1.15% and α by <0.4% across 40–260 K—and no fit uncertainties, covariances, or confidence intervals are given. The fit procedure is a self-consistent alternating optimization of α and γ with no stated tolerance, and the I-V curves are split at a post hoc voltage of 1.2 V: Simmons model below, dissipation model above. A small systematic error in either model near the split, or a different choice of split voltage, could change γ(T) and α(T) enough to remove or shift the 137 K dip. Moreover, ΔE/E in Fig. 7b is extracted at a single energy fraction E/(φ0+εF)=0.19 and V=0.95 V; the existence of the turning point may depend on that choice. Since the central physical conclusion is a critical temperature, the absence of uncertainty quantification and robustness checks is load-bearing.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a phenomenological model for energy dissipation in Al/Al2O3/Al tunnel junctions by adding a nonlinear friction force f(x) = γ v(x)^α to the Simmons tunneling formula. The parameters γ and α are fitted to high-voltage I-V data at each temperature, with the barrier height and width taken from a low-voltage Simmons fit. The authors then use Eq. (7) to compute the fraction of dissipated energy ΔE/E and report that it decreases with increasing particle energy, decreases with applied voltage, and exhibits a non-monotonic temperature dependence with a turning point at T ≈ 137 K, attributed to the onset of stronger electron-phonon interactions. The paper also reports the temperature dependence of the barrier height and energy gap.","tokens_in":9843,"tokens_out":4754,"duration_ms":51265,"significance":"If the model and the identification of a critical temperature were firmly established, the work would offer a simple empirical route to estimating dissipative losses in tunnel junctions, which is relevant for nanoscale device heat management. The experimental effort is substantial: high-quality junctions, careful characterization, and barrier parameters consistent with literature values. The paper is honest about being empirical and includes comparison with the standard Simmons model. However, the central physical claim—the 137 K turning point—rests on very small variations in fitted parameters for which no uncertainties are provided, and the derivation assigns a real velocity to a tunneling electron inside a classically forbidden barrier. The model's predictive content is limited because the dissipation parameters are extracted from the same data that are subsequently used to compute the reported trends.","major_comments":[{"comment":"The temperature trends of ΔE/E in Figs. 7b and 8 are not independent predictions: they are computed via Eq. (7) from the same fitted γ(T) and α(T) that were obtained by fitting the high-voltage I-V curves. In that sense, the 'demonstration' of the 137 K turning point is a restatement of the temperature dependence of γ and α, rather than a new physical result. The authors should explicitly acknowledge this circularity and provide an out-of-sample check, such as predicting a different observable (e.g., the voltage dependence of the differential resistance above 1.2 V) that was not used in the fit.","section":"§III.D, Eq. (7)"},{"comment":"The reported relative variations of γ (~1.15%) and α (<0.4%) are extremely small, and no fit uncertainties, covariances, or confidence intervals are given for these parameters. The 137 K turning point in Fig. 7b is therefore not statistically established. The paper should propagate uncertainties from the I-V fits into γ(T), α(T), and ΔE/E, and should test the robustness of the turning point against the chosen energy fraction (E/(φ0+εF)=0.19) and applied voltage (V=0.95 V) used in Figs. 7 and 8.","section":"§III.D, Fig. 6"},{"comment":"Equation (2) assigns a real velocity v(x) = sqrt(2(V(x)-E)/m) to the tunneling electron inside the classically forbidden barrier, where E < V(x). This conflicts with the standard WKB treatment in Eq. (3), where the tunneling momentum is imaginary. Assigning a real velocity is not justified by the tunneling formalism and makes the physical interpretation of ΔE as energy dissipated along an in-barrier trajectory questionable. The authors should either provide a derivation of this effective classical model or clearly state that it is a purely phenomenological fitting device, and discuss the limitations of interpreting γ, α, and ΔE/E literally.","section":"§III.B, Eq. (2)"},{"comment":"The data are split at a post hoc voltage of 1.2 V, with Simmons model used below and the dissipation model above, but no physical criterion is given for this value, and the text later refers to 1.6 V in Fig. 5. Since γ and α are extracted from the high-voltage segment, an arbitrary split choice can bias their temperature dependence and hence the location of the 137 K turning point. The paper should show how γ(T), α(T), and ΔE/E change when the split voltage is varied (e.g., 1.1, 1.2, 1.3 V) and should reconcile the inconsistent 1.2 V/1.6 V statements.","section":"§III, Figs. 4 and 5"}],"minor_comments":[{"comment":"There are several typographical errors, including 'loosing' in the abstract, 'Simmon´s' in several places, 'Aknowledgments' in Section V, and an inverted question mark in §IV ('¿what is the effect?').","section":"Throughout"},{"comment":"The expression for capacitance is incomplete: 'C = ε??' lacks the geometric factor and the dielectric thickness. Please provide the full formula used to estimate C ~ 1×10-8 F.","section":"§II"},{"comment":"Equation (7) is labeled ΔE(s)/E but the right-hand side explicitly contains s, while the text says 'omitting s for simplicity.' Please define the notation consistently so that Eq. (7) is unambiguous.","section":"§III.B, Eq. (7)"},{"comment":"The text describes the ΔE/E versus voltage curve as showing a linear reduction, but no fitted line or correlation coefficient is provided. Adding a linear fit with slope and R² would strengthen the claim.","section":"Fig. 7c"},{"comment":"The text states that Simmons model works well 'up about 1.2 V' and later says 'for voltages below 1.6 V' in the discussion of Fig. 5. Please reconcile these threshold values or clarify that they refer to different fits.","section":"§III.B vs. Fig. 5"}],"recommendation":"major_revision","confidential_remarks":"The manuscript corresponds to a published JAP article (136, 184401, 2024) and the present version appears to be a preprint of that work. The main concern is that the headline 137 K turning point is supported by sub-percent variations in fitted parameters with no uncertainty quantification. The authors should be asked to provide error bars on γ and α, a sensitivity analysis of the split voltage, and an explicit statement that ΔE/E trends are derived from the same fitted parameters used for the I-V curves. If these additions are made, the paper could be acceptable as a phenomenological study, but the physical interpretation should be tempered accordingly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: this paper extends your earlier linear friction model to f = γv^α and uses it to fit I-V curves of Al/Al2O3/Al junctions up to near breakdown. The fits are good and the device characterization is careful. The second thing: the headline claim of a 137 K critical temperature in the dissipated energy fraction is not established by the data as presented.\n\nWhat is actually new: the nonlinear exponent α and the temperature dependence of γ and α over 40–260 K. The model is a straightforward extension of your previous work (ref. [9]), but applying it to voltages near the barrier height and extracting the dissipation fraction is a useful empirical addition. The junction quality is a real strength: the fitted area matches the optical area, the barrier width is constant, and the resistance-area product is high. The barrier height vs temperature relation and the phonon energy estimate from O'Donnell–Chen are plausible.\n\nWhere it gets soft. The physical interpretation leans heavily on assigning a real velocity v = sqrt(2(V−E)/m) to a tunneling electron inside a classically forbidden barrier. That is not justified; the WKB momentum is imaginary, and a classical friction force acting on an evanescent wave is a phenomenological stretch. The paper is honest that the model is empirical, but Eq. (7) and the dissipation fraction inherit that stretch.\n\nMore importantly, γ and α are fit to the same high-voltage I-V curves that are then used to compute ΔE/E. The trends in Figs. 7 and 8 are restatements of the fit, not independent confirmations. The variations in γ (~1.15%) and α (<0.4%) across temperature are tiny, no error bars or covariance are given, and the 1.2 V split of the data is post hoc. The 137 K dip in ΔE/E could easily be an artifact of a small systematic error near the split or a different choice of split voltage. The paper needs fit uncertainties, a statistical test for non-monotonicity, and a robustness check against split voltage before that critical temperature claim can be believed.\n\nWho gets value: people who need a practical empirical I-V fitting formula for similar large-area Al/Al2O3/Al junctions at high bias, and who are willing to treat γ and α as fitting parameters without deep physical meaning. The paper deserves a serious referee because the experimental work is careful and the empirical model may be reproducible, but it needs major revision to back the physical claims. If I were the editor, I would send it to review with a clear request for uncertainty quantification and a softened interpretation.","headline":"A useful empirical dissipation fit for tunnel junctions, but the 137 K turning point is not statistically supported.","tokens_in":10335,"tokens_out":2272,"would_cite":false,"duration_ms":23418,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["73.40.Gk","73.40.Rw","73.50.-h"],"model":"deepseek-v4-flash","headline":"This paper claims that adding a nonlinear friction force $f(x)=\\gamma v(x)^{\\alpha}$ to the Simmons tunneling model reproduces measured Al/Al2O3/Al junction I-V curves up to the barrier height, and that the dissipated-energy fraction…","keywords":["quantum tunneling","tunnel junctions","energy dissipation","nonlinear friction","Simmons model","Al/Al2O3/Al","electron-phonon interaction","I-V characteristics"],"falsifier":"Attach a sensitive thermometer or calorimeter directly to the junction and measure the heat released during the same I-V sweeps from 40 to 260 K; if the measured dissipated power shows no minimum around 137 K and no rise above it, the model's identification of a critical temperature and its friction mechanism would be contradicted regardless of the quality of the I-V fit.","tokens_in":9355,"feed_emoji":"⚡","tokens_out":8653,"duration_ms":87232,"temperature":0.7,"pith_summary":"The paper tries to establish that energy dissipation in solid-state tunnel junctions can be captured by treating a tunneling electron as losing energy to a nonlinear friction force $f=\\gamma v^{\\alpha}$ inside the barrier. With only two fitted parameters, $\\gamma$ and $\\alpha$, this extension lets the standard Simmons tunneling model fit measured Al/Al2O3/Al I-V characteristics all the way up to voltages near the barrier height, where the dissipation-free model deviates badly. If correct, the model turns ordinary current-voltage data into a probe of dissipative parameters and identifies a critical temperature, near 137 K, where electron-phonon dissipation starts to dominate over barrier-height reduction. The contribution is empirical and phenomenological rather than a microscopic derivation; its value is a compact description of high-bias tunneling with heat loss.","feed_headline":"Nonlinear friction fits tunnel-junction data up to the barrier height","feed_subtitle":"Adding $f=\\gamma v^{\\alpha}$ to Simmons' model reproduces Al/Al2O3/Al I-V curves and reveals the 137 K dissipation turnover.","key_machinery":"The load-bearing object is the energy-loss-corrected WKB transmission coefficient $T(E)=\\exp\\left[-4k_0 s\\left(\\sqrt{1+\\lambda}-\\frac{1}{3}\\lambda\\right)\\right]$, where $k_0=\\sqrt{2m(\\varepsilon_F+\\phi-E)}/\\hbar$, $\\phi=\\phi_0-eV/2$ is the voltage-reduced mean barrier height, and $\\lambda=\\gamma(2/mc^2)^{\\alpha/2}(\\varepsilon_F+\\phi-E)^{(\\alpha-1)/2}$. The nonlinear friction force $f(x)=\\gamma v^{\\alpha}$ with $v=\\sqrt{2(V(x)-E)/m}$ generates the energy loss $\\Delta E(x)$ that replaces $E$ by $E-\\Delta E(x)$ inside the exponent. This machinery extends an earlier linear-friction model to arbitrary bias by fitting only the friction coefficient $\\gamma$ and velocity exponent $\\alpha$, while keeping the barrier parameters fixed from low-voltage Simmons fits.","core_discovery":"The central claim is that the high-voltage failure of Simmons' dissipation-free model is due to energy loss during tunneling, and that this loss is well described by a velocity-dependent nonlinear friction force $f(x)=\\gamma v(x)^{\\alpha}$ acting on the electron while it is inside the barrier. Replacing the particle energy $E$ in the WKB transmission coefficient by $E-\\Delta E(x)$, with $\\Delta E(x)=\\int \\gamma v(x')^{\\alpha}\\,dx'$, and fixing the barrier height and width from low-voltage Simmons fits, the paper reproduces its measured I-V curves for voltages above 1.2 V up to $\\phi_0/e$ using only $\\gamma$ and $\\alpha$ as free parameters. From the fitted parameters the fraction $\\Delta E/E$ of incident energy dissipated across the barrier decreases as the particle energy approaches the effective barrier height and as applied voltage increases, but it has a minimum at roughly 137 K: below this temperature barrier-height reduction controls the trend, while above it growing electron-phonon interaction reverses it.","pith_inferences":["A direct measurement of heat released in the junction, rather than inference from I-V fitting, would be the cleanest test of the 137 K turnover; the paper itself does not report such a measurement.","If the classical-velocity picture is only a parameterization, then $\\gamma$ and $\\alpha$ may be reinterpreted as effective transport coefficients that absorb barrier-shape and electron-phonon effects, and an equivalent quantum dissipative model without real velocities inside the barrier could describe the same data.","Applying the same two-parameter procedure to junctions with different oxide thicknesses or barrier materials would show whether the fitted $\\alpha$ is universal or junction-specific, separating generic dissipation mechanisms from material-dependent ones.","Combining the friction model with shot-noise or infrared photon-emission measurements at large bias could connect $\\gamma$ and $\\alpha$ to the radiation emitted by junction current fluctuations, giving the phenomenological parameters a microscopic anchor."],"forward_implications":["Simmons' model without dissipation fails above about 1.2 V in these junctions; the nonlinear friction term restores an accurate fit up to voltages near the barrier height using just two fitted parameters.","At fixed temperature, the fraction of incident energy dissipated falls as the particle's energy approaches the effective barrier and as the applied voltage rises, so higher-field operation does not necessarily mean a larger fractional energy loss per tunneling electron.","Temperature acts through two competing mechanisms: barrier-height lowering reduces $\\Delta E/E$ from 40 K up to roughly 137 K, while increasing electron-phonon interactions reverse the trend above roughly 137 K, defining an optimal operating temperature for minimal relative dissipation.","The fitted parameters $\\gamma$ and $\\alpha$ stay nearly constant below about 137 K and then decline linearly with temperature, indicating that the dissipative regime of the junction changes character above this critical temperature.","Including dissipation reduces the WKB transmission coefficient relative to the dissipation-free Simmons result at the same voltage, so neglecting dissipation at high bias would bias any barrier parameters extracted from the I-V curve."],"supporting_citations":[{"why":"Supplies the dissipation-free Simmons current-voltage model and the mean-barrier-height approximation that the paper extends with friction and uses to fix $\\phi_0$ and $s$.","marker":"[10]"},{"why":"Previous linear-friction dissipative tunneling model for low voltages; this work generalizes it to $f=\\gamma v^{\\alpha}$ and to voltages up to the barrier height.","marker":"[9]"},{"why":"Gives the linear relation between the Al2O3 energy gap and barrier height used to interpret the temperature dependence of $\\phi_0$.","marker":"[19]"},{"why":"Band-gap temperature formula used to extract the average phonon frequency that supports the electron-phonon dissipation interpretation.","marker":"[24]"},{"why":"Provides the room-temperature gap of amorphous Al2O3 used to calibrate the gap-to-barrier-height relation.","marker":"[26]"},{"why":"Speed-of-sound measurement giving a phonon frequency of the same order as the fitted value, supporting the phonon interpretation.","marker":"[27]"},{"why":"Analogous sublinear drag force behavior in complex fluids used to justify allowing the velocity exponent $\\alpha$ to be less than one.","marker":"[28]"}],"fun_headline_variants":["137 K marks turnover in tunnel-junction energy loss","Tunneling electrons lose energy to nonlinear friction","Nonlinear friction inside barrier explains tunnel-junction I-V data","Tunnel-junction dissipation reverses at 137 K"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The extraction of $\\gamma$, $\\alpha$, and $\\Delta E/E$ rests on treating an electron inside a classically forbidden barrier as a particle with a real speed $v=\\sqrt{2(V(x)-E)/m}$ subject to a classical nonlinear friction force, even though the WKB momentum inside the barrier is imaginary.","fun_headline_variants_meta":{"raw":{"variants":["137 K marks turnover in tunnel-junction energy loss","Tunneling electrons lose energy to nonlinear friction","Nonlinear friction inside barrier explains tunnel-junction I-V data","Tunnel-junction dissipation reverses at 137 K"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001255,"raw_usage":{"total_tokens":5183,"prompt_tokens":1022,"completion_tokens":4161,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":638,"completion_tokens_details":{"reasoning_tokens":4097}},"tokens_in":638,"tokens_out":4161,"duration_ms":33088,"temperature":1.0,"reasoning_tokens":4097,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T16:39:53.150044+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Attach a sensitive thermometer or calorimeter directly to the junction and measure the heat released during the same I-V sweeps from 40 to 260 K; if the measured dissipated power shows no minimum around 137 K and no rise above it, the model's identification of a critical temperature and its friction mechanism would be contradicted regardless of the quality of the I-V fit.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the dissipation-free Simmons current-voltage model and the mean-barrier-height approximation that the paper extends with friction and uses to fix $\\phi_0$ and $s$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Previous linear-friction dissipative tunneling model for low voltages; this work generalizes it to $f=\\gamma v^{\\alpha}$ and to voltages up to the barrier height."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the linear relation between the Al2O3 energy gap and barrier height used to interpret the temperature dependence of $\\phi_0$."},{"cited_title":"SanGiorgio, S","cited_arxiv_id":null,"evidence_quote":"Band-gap temperature formula used to extract the average phonon frequency that supports the electron-phonon dissipation interpretation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the room-temperature gap of amorphous Al2O3 used to calibrate the gap-to-barrier-height relation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Speed-of-sound measurement giving a phonon frequency of the same order as the fitted value, supporting the phonon interpretation."},{"cited_title":"Costina and R","cited_arxiv_id":null,"evidence_quote":"Analogous sublinear drag force behavior in complex fluids used to justify allowing the velocity exponent $\\alpha$ to be less than one."}],"review_version":1}