{"id":"b5462539-5e4d-40c8-a6ab-e7c0f88a3b86","arxiv_id":"1908.08415","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"The paper provides accurate equilibrium MD thermal conductivity data for the one-component plasma from Γ=0.1 to 180 and shows Landau-Spitzer theory holds only for Γ≲0.3, with no model accurate for Γ≳10.","lead":"Equilibrium molecular dynamics simulations map the thermal conductivity of the one-component plasma across a wide range of coupling strengths, from weak to strong. The new data resolve earlier discrepancies between simulation methods and show that standard theoretical models fail for strongly coupled plasmas.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Production-phase ensemble is unspecified; if the Nosé-Hoover thermostat remains coupled during data collection, the heat-flux ACF could be damped and all reported conductivities biased.","rationale":"The reader's verdict is CONDITIONAL, with the unspecified production ensemble as the weakest assumption. I agree that this is the most load-bearing concern because it directly affects the validity of all reported simulation data. If the thermostat remained coupled, the heat flux autocorrelation function could be damped, systematically biasing the thermal conductivity at every Γ, including the new low-Γ data that support the Landau-Spitzer test and the conclusion that no theory works for Γ>10. The concern is not that the authors necessarily made this mistake—the good agreement with non-equilibrium MD for Γ≥2 suggests they likely turned the thermostat off—but that the manuscript does not state it, making the results non-reproducible and the central claim unverifiable from the text. The other issues raised by the reader (the '~2% accuracy' claim versus actual low-Γ error bars, and the qualitative nature of Landau-Spitzer agreement) are real but secondary; they would weaken specific wording but not invalidate the overall conclusions. Therefore, the verdict should remain CONDITIONAL, pending clarification of the production ensemble. A concrete check is to re-run a few cases with explicit NVE and NVT production and compare; this settles the concern directly.","tokens_in":14309,"tokens_out":15631,"duration_ms":150009,"concrete_test":"Request or reconstruct the LAMMPS input scripts and re-run at least two representative cases, e.g., Γ=10 and Γ=60, with (a) NVE production after NVT equilibration and (b) continuous NVT with the same Nosé-Hoover parameters. Compare the resulting thermal conductivity and the heat-flux ACF envelope to those reported in Table II and Fig. 2. If the NVT-production result differs from the NVE result by more than the reported statistical error (about 2-3%), the missing ensemble specification is a critical flaw; if the two agree within error, the concern is resolved and only a textual clarification is needed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section IIA describes only the equilibration phase with a Nosé-Hoover thermostat to achieve the desired temperature; it never states whether the thermostat is turned off during the production run. The Green-Kubo relation in Eq. (1) is derived for Hamiltonian dynamics, and a Nosé-Hoover thermostat introduces a non-Hamiltonian friction term that can couple to the slow, long-lived heat-flux oscillations that the paper itself identifies as critical for Γ≥10 (e.g., ACFs persisting hundreds of ω_p^{-1} in Fig. 2). If the thermostat coupling time is comparable to these correlation times, the ACF would be artificially damped, biasing λ* at all Γ. The agreement with non-equilibrium MD (Fig. 1) is reassuring but does not fully resolve the issue for the low-Γ regime where no independent simulation data exist, and the new low-Γ data are the basis for the Landau-Spitzer test. Because the manuscript gives no explicit statement of the production ensemble, the central claim of 'accurate data' is not reproducible and could rest on an unverified dynamical bias.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports equilibrium molecular dynamics (MD) simulations of the thermal conductivity of the one-component plasma (OCP) using the Green-Kubo formalism, covering Γ from 0.1 to 180. The authors use long heat-flux time series (up to 6×10^5 ω_p^{-1}), analyze statistical noise via cross-correlations and the Zwanzig-Ailawadi estimate, decompose the conductivity into kinetic, potential, virial, and cross contributions, and compare the total conductivity against Landau-Spitzer, Lee-More, Tanaka-Ichimaru, and Effective Potential Theory (EPT). They conclude that Landau-Spitzer agrees with simulations for Γ≲0.3, that EPT and Tanaka-Ichimaru are accurate up to Γ≈7, and that none of the models captures the regime Γ≳10, where potential and virial heat-flux components dominate. The paper also provides analytic fits to the MD data.","tokens_in":14584,"tokens_out":4894,"duration_ms":46358,"significance":"If the reported data are reliable, this is a valuable benchmark: it is the first equilibrium MD dataset extending to Γ<1, provides the first MD test of Landau-Spitzer thermal conductivity, and quantitatively documents the failure of binary-collision-based theories at strong coupling. The component decomposition (kinetic/potential/virial) is a useful diagnostic for future theory development. The error analysis is thoughtful, using six independent time series, cross-correlation noise estimators, and explicit comparison with the predicted 1/τ scaling. The agreement with prior non-equilibrium MD for Γ≥2 partially mitigates concerns about the equilibrium method. However, the central claim of 'accurate data' is weakened by an unspecified production-phase ensemble and by an accuracy statement that is not consistent with the tabulated statistical uncertainties.","major_comments":[{"comment":"The manuscript specifies a Nosé-Hoover thermostat only for the equilibration phase and never states whether the thermostat is turned off during the production runs. Because the Green-Kubo relation in Eq. (1) is derived for Hamiltonian dynamics, and because the paper itself emphasizes heat-flux autocorrelation oscillations persisting for hundreds of ω_p^{-1} at Γ≥10, coupling a thermostat during production could damp those oscillations and bias the reported conductivities. The agreement with non-equilibrium MD in Fig. 1 for Γ≥2 is reassuring but does not validate the new Γ<0.5 data used for the Landau-Spitzer test. Please state explicitly the ensemble used during data collection (e.g., NVE or NVT) and, if NVT is used, justify its consistency with the Green-Kubo formula.","section":"Sec. IIA, Eq. (1)"},{"comment":"The abstract's claim that 6×10^5 ω_p^{-1} time series allow '≲2% accuracy' is not supported by the Table II numbers if those are standard deviations of the six independent time series: the relative standard deviations at Γ=0.5, 1.0, 2.0, and 5.0 are about 3.7%, 3.5%, 4.0%, and 2.7%, respectively. If the tabulated values are instead standard errors of the mean (standard deviation divided by sqrt(6)), the text should say so explicitly and the abstract should state which quantity corresponds to 'accuracy.' Please reconcile the abstract with the tabulated uncertainties.","section":"Abstract and Table II"},{"comment":"The choice of the averaging window [100,150] ω_p^{-1} for Γ≥0.5 is justified only by the qualitative statement that alternate windows do not change results 'significantly.' Since the cumulative integral still exhibits oscillations at the ~5% level at t=100 ω_p^{-1}, a quantitative sensitivity analysis (e.g., a table or figure showing λ* for several averaging windows) is needed to support the claimed accuracy. In addition, for Γ<0.5 a different saturation-based method is used; the crossover between methods and its effect on the reported values should be documented.","section":"Sec. IIC"}],"minor_comments":[{"comment":"The notation ⟨...⟩ is used both for the ensemble average in Eq. (1) and for the finite-time average in Eq. (3); please distinguish the two, for example by using a subscript for the time-average.","section":"Sec. IIB, Eq. (4)"},{"comment":"Reference [41] lists the first author as 'Z. Donk' instead of 'Z. Donkó' and omits the publication year; please correct the citation.","section":"Reference [41]"},{"comment":"The table caption states 'standard deviation' but the text says the six values were 'averaged ... with their standard deviation.' Please specify whether the reported error is the standard deviation of the six values or the standard error of the mean, and use consistent terminology throughout.","section":"Sec. IIC, Table II"},{"comment":"In the discussion of the Landau-Spitzer limit, the text says the kinetic fit is 'proportional to C/ln(DΓ^{-3/2})' at low Γ, but the full expression in Eq. (7) also contains the prefactor Γ^{-5/2}; please correct this limiting-form description.","section":"Sec. IIIA and Eq. (11)"},{"comment":"The word 'neagitive' appears in the sentence 'reaches its maximum neagitive value'; this should be 'negative.'","section":"Sec. IIC"},{"comment":"The text states that the fit of Eq. (9) passes through all markers 'with one exception'; please identify which data point is the exception, since this is not otherwise evident from Fig. 4.","section":"Sec. IIC, Eq. (9)"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope for a plasma physics journal and does not appear to have citation or novelty issues. The main technical concern is the unspecified production ensemble, which is a reproducibility issue rather than a fundamental flaw; I expect the authors can address it by stating the ensemble and justifying the use of the Green-Kubo relation. The accuracy statement in the abstract should also be made consistent with the reported statistics. After these revisions, the paper would be a solid contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis is a strong computational paper that gives the OCP community what it needed: a consistent set of equilibrium MD thermal conductivities from Γ=0.1 to 180, with an explicit treatment of the long-time ACF noise that tripped up earlier simulations. The low-Γ data are genuinely new, and they allow a direct check of Landau-Spitzer. The component decomposition (kinetic/potential/virial) is useful, and the comparison of EPT, Tanaka-Ichimaru, and Lee-More will be cited.\n\nThe error analysis is the best part. The Zwanzig-Ailawadi scaling for the ACF noise is verified with cross-correlations, and the choice of 6×10^5 ω_p^{-1} runs is justified. I believe the reported conductivities for Γ≥2, where they overlap with non-equilibrium MD, are reliable. The good agreement with Donko-Hartmann is reassuring.\n\nThe soft spots are mostly reporting issues, but one is real. Section IIA describes only the Nose-Hoover equilibration; it never states whether the thermostat is switched off during production. If it stays on, the heat-flux ACF is damped and all conductivities—especially the long-oscillation ones at high Γ—are biased. The agreement with NEMD suggests they did turn it off, but the manuscript must say so. This is not a fatal flaw, but it is a reproducibility issue that has to be fixed.\n\nThe other issue is the '~2% accuracy' claim in the abstract. Table II shows standard deviations of 4-8% for Γ<0.5 (e.g., 306±26 at Γ=0.1). That is not 2%. The claim may refer to a different measure, but as written it overstates precision. The Landau-Spitzer test also shows only qualitative agreement (roughly 17% high), which is fine as a first test, but 'agree' is too strong.\n\nThe central negative result—none of the analytic models work for Γ≳10—is robust. The potential and virial components clearly dominate there, and no kinetic theory can capture that.\n\nThis paper deserves a serious referee. The data will be used by others, so the ensemble ambiguity must be resolved before publication. I'd send it back to the authors for a clarification revision rather than reject.","headline":"A valuable, mostly reliable OCP thermal conductivity benchmark that advances the field, but the unspecified production ensemble and overstated 2% accuracy claim need fixing before publication.","tokens_in":15148,"tokens_out":2265,"would_cite":true,"duration_ms":23340,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["52.25.Fi","52.65.Yy"],"model":"deepseek-v4-flash","headline":"New simulations give accurate one-component plasma conductivity, and no theory works at strong coupling.","keywords":["thermal conductivity","one-component plasma","Green-Kubo","molecular dynamics","Coulomb coupling","heat flux autocorrelation","strongly coupled plasma","Landau-Spitzer theory"],"falsifier":"Re-run the Γ = 30 simulation with the thermostat explicitly disabled during production and compare the heat-flux autocorrelation and integrated conductivity with the reported value; if the conductivity shifts by more than the stated error bars, the dataset is biased. Alternatively, an independent non-equilibrium MD calculation at Γ = 30 that reproduces the reported λ* would support the claim, while disagreement would falsify it.","tokens_in":14079,"feed_emoji":"🔥","tokens_out":10068,"duration_ms":87986,"temperature":0.7,"pith_summary":"Using equilibrium molecular dynamics and the Green-Kubo formalism, this paper produces thermal conductivity data for the one-component plasma (OCP) across $0.1\\le\\Gamma\\le180$, resolving a long-standing 30% discrepancy between equilibrium and non-equilibrium simulation results. The authors show that the heat-flux autocorrelation function decays slowly, with oscillations persisting for hundreds of plasma periods, so prior equilibrium runs were too short and their noise levels contaminated the conductivity. The new long ($6\\times10^5\\,\\omega_p^{-1}$) simulations give the first accurate conductivities for $\\Gamma\\lesssim1$, which serves as the first computational test of Landau-Spitzer theory: agreement holds only for $\\Gamma\\lesssim0.3$. Extensions to strong coupling are tested, and although EPT and Tanaka-Ichimaru work reasonably up to $\\Gamma\\approx10$, none of the examined theories reproduces the $\\Gamma\\gtrsim10$ regime, where the heat flux is dominated by potential-energy and virial contributions rather than kinetic transport.","feed_headline":"Simulations settle plasma heat flow; theory fails past Γ=10","feed_subtitle":"Long equilibrium runs reconcile old discrepancies and show Landau-Spitzer works only for Γ under 0.3.","key_machinery":"The machinery is the Green-Kubo relation with a component-resolved heat-flux autocorrelation. The heat flux $\\mathbf{j}$ is split into a kinetic part (particle mass flow carrying kinetic energy), a potential part (potential energy transported by mass flow), and a virial part (stress coupled to mass flow). The paper computes the autocorrelation of each part and of the cross terms, showing for the first time in the OCP that at $\\Gamma\\gtrsim10$ the potential and virial terms oscillate in phase and dominate the transport. A second piece of machinery is the noise-control protocol: because the ACF tail obeys the $1/\\tau$ noise scaling of Zwanzig-Ailawadi, the authors use $6\\times10^5\\,\\omega_p^{-1}$ time series and average the cumulative integral over a window where oscillations have decayed but noise has not yet built up.","core_discovery":"The central discovery is that the OCP heat-flux autocorrelation function (ACF) has a slowly decaying oscillatory envelope, predicted to scale as $1/\\sqrt{t}$, so the cumulative Green-Kubo integral is still oscillating by about 5% at $t=100\\,\\omega_p^{-1}$ and only converges after hundreds of plasma periods. Prior equilibrium MD runs of roughly $400$ and $8000\\,\\omega_p^{-1}$ were too short, which explains the discrepancies with non-equilibrium MD. The authors run $6\\times10^5\\,\\omega_p^{-1}$ trajectories, verify noise levels against cross-correlations of different heat-flux components, and average the cumulative integral over $[100,150]\\,\\omega_p^{-1}$. The resulting dimensionless conductivity $\\lambda^*=\\lambda/(n\\omega_p k_B a^2)$ agrees with non-equilibrium data for $\\Gamma\\ge2$ and extends down to $\\Gamma=0.1$. Decomposing the heat flux into kinetic, potential, and virial parts shows that the kinetic term dominates up to $\\Gamma\\approx5$, while for $\\Gamma\\gtrsim10$ the potential, virial, and their cross term are largest; no theory that treats only kinetic transport can capture this regime.","pith_inferences":["The in-phase oscillations of the potential and virial ACF components and the opposite-signed cross term suggest that a conserved quantity, such as an enthalpy current, may be exchanged between the two channels; a mode-coupling or hydrodynamic description might reproduce this cancellation and extend theory past $\\Gamma\\approx10$.","The $1/\\sqrt{t}$ noise envelope implies that even longer runs, or variance-reduction estimators such as cepstral analysis, could push the accuracy below 2% and extend the dataset to lower $\\Gamma$ or to two-component plasmas without prohibitive cost.","A direct test of the unstated thermostat protocol, whether the Nose-Hoover thermostat is removed during production, would determine whether the reported conductivities are biased; this check is not addressed in the paper but is a necessary reproducibility test.","The different ranges of validity of EPT for diffusion, shear viscosity, and thermal conductivity suggest that each transport coefficient has a different sensitivity to many-body correlations, which could guide which coefficient is best for benchmarking new strong-coupling theories."],"forward_implications":["The new dataset interpolates from weak coupling to solidification at $\\Gamma\\approx170$ and provides a fitted formula (Eqs. 7–9) for use in hydrodynamic simulations of dense plasmas.","Landau-Spitzer theory is confirmed only for $\\Gamma\\lesssim0.3$; using it beyond that coupling overestimates the thermal conductivity.","The Lee-More model's ad hoc replacement of the Debye length by the ion-sphere radius is shown to be inadequate for $\\Gamma\\gtrsim1$, a relevant caveat for high-energy-density simulations that use this model.","Existing theories such as EPT can describe the kinetic component of the heat flux to fairly strong coupling, but none describes the potential-virial dominated regime, indicating a need for transport theories that include multi-particle dynamics.","The component decomposition reveals that at $\\Gamma\\approx9$ the conductivity reaches a minimum where the kinetic term becomes smaller than the sum of potential and virial terms, providing a benchmark for future theories."],"supporting_citations":[{"why":"Non-equilibrium MD data that the new equilibrium results are benchmarked against and agree with for Γ≥2.","marker":"[10]"},{"why":"Early equilibrium MD results whose discrepancy with non-equilibrium data this work attributes to short time series.","marker":"[23]"},{"why":"Later equilibrium MD results also affected by finite-time noise; used as a comparison point.","marker":"[24]"},{"why":"Provides the finite-time noise scaling (1/τ) used to estimate the required simulation length.","marker":"[39]"},{"why":"The Landau-Spitzer theory tested against the new weak-coupling data.","marker":"[31]"},{"why":"Tanaka-Ichimaru theory, whose generalized Coulomb logarithm is tested up to Γ≈10.","marker":"[20]"},{"why":"Effective Potential Theory, tested against the MD data and shown to match only the kinetic part at strong coupling.","marker":"[21]"},{"why":"Lee-More dense-plasma conductivity model shown to fail for Γ≳1.","marker":"[32]"},{"why":"Enskog-corrected EPT that matches the MD data at moderate coupling (Γ≲7).","marker":"[44]"}],"fun_headline_variants":["Plasma heat flow: theories fail for Γ>10, long runs needed","Short MD runs misled plasma thermal conductivity","Simulations settle plasma heat conduction; theory holds only for weak coupling","Plasma heat flow breaks all theories for Γ≥10","OCP heat flux needs 600,000-period MD runs to converge"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper never states whether the Nose-Hoover thermostat used during equilibration is turned off during the production runs; if the thermostat remains coupled, it could damp the heat-flux autocorrelation and bias all reported conductivities.","fun_headline_variants_meta":{"raw":{"variants":["Plasma heat flow: theories fail for Γ>10, long runs needed","Short MD runs misled plasma thermal conductivity","Simulations settle plasma heat conduction; theory holds only for weak coupling","Plasma heat flow breaks all theories for Γ≥10","OCP heat flux needs 600,000-period MD runs to converge"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000948,"raw_usage":{"total_tokens":4073,"prompt_tokens":996,"completion_tokens":3077,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":612,"completion_tokens_details":{"reasoning_tokens":2990}},"tokens_in":612,"tokens_out":3077,"duration_ms":20690,"temperature":1.0,"reasoning_tokens":2990,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:42:12.474377+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the Γ = 30 simulation with the thermostat explicitly disabled during production and compare the heat-flux autocorrelation and integrated conductivity with the reported value; if the conductivity shifts by more than the stated error bars, the dataset is biased. Alternatively, an independent non-equilibrium MD calculation at Γ = 30 that reproduces the reported λ* would support the claim, while disagreement would falsify it.","supporting_citations":[{"cited_title":"Donk´ o and P","cited_arxiv_id":null,"evidence_quote":"Non-equilibrium MD data that the new equilibrium results are benchmarked against and agree with for Γ≥2."},{"cited_title":"Bernu, P","cited_arxiv_id":null,"evidence_quote":"Early equilibrium MD results whose discrepancy with non-equilibrium data this work attributes to short time series."},{"cited_title":"Salin and J.-M","cited_arxiv_id":null,"evidence_quote":"Later equilibrium MD results also affected by finite-time noise; used as a comparison point."},{"cited_title":"Zwanzig and N","cited_arxiv_id":null,"evidence_quote":"Provides the finite-time noise scaling (1/τ) used to estimate the required simulation length."},{"cited_title":"Spitzer and R","cited_arxiv_id":null,"evidence_quote":"The Landau-Spitzer theory tested against the new weak-coupling data."},{"cited_title":"Tanaka and S","cited_arxiv_id":null,"evidence_quote":"Tanaka-Ichimaru theory, whose generalized Coulomb logarithm is tested up to Γ≈10."},{"cited_title":"An Effective Potential Theory for Transport Coefficients Across Coupling Regimes","cited_arxiv_id":"1303.3202","evidence_quote":"Effective Potential Theory, tested against the MD data and shown to match only the kinetic part at strong coupling."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Lee-More dense-plasma conductivity model shown to fail for Γ≳1."},{"cited_title":"Modified Enskog Kinetic Theory for Strongly Coupled Plasmas","cited_arxiv_id":"1506.03112","evidence_quote":"Enskog-corrected EPT that matches the MD data at moderate coupling (Γ≲7)."}],"review_version":1}