{"id":"2fc29096-f982-45ba-9d5c-8670767bd660","arxiv_id":"1908.03223","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"The heat produced and absorbed during an action potential can be explained by the release and restoration of electrical energy stored in the cell membrane, provided the membrane carries a bias of negative surface charges on its inner side.","lead":"A new model calculates how much heat a nerve cell releases when it fires, and it shows that an imbalance of electric charge on the two sides of the cell membrane can explain the measured warmth. The work offers a physical explanation for a thermal signature that has accompanied nerve recordings for nearly a century.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim treats the action potential as a reversible, work-free release of membrane field energy (Eq. 20), while the process is actually driven by dissipative ion currents; a Hodgkin-Huxley energy-balance check is needed before assigning primacy to the electrostatic mechanism.","rationale":"The stress-test pass identifies the same weakest assumption as the reader: the model applies equilibrium thermodynamics to a fundamentally dissipative, out-of-equilibrium process. The paper flags this explicitly in the Discussion and in Appendix A, which only addresses double-layer relaxation rather than the energy balance of the ion currents that drive depolarization. The electrostatic free-energy derivation and entropy terms are coherent, and the surface-charge bias is drawn from a defensible literature range, so I see no reason to reject the paper. However, the central attribution claim is conditional: it requires that nonreversible, non-electrostatic contributions to the measured heat be negligible. That condition is not established, and it is exactly the kind of assumption that a Hodgkin-Huxley energy-balance calculation can test. Because the reader already assigned CONDITIONAL, the verdict need not change.","tokens_in":15349,"tokens_out":8741,"duration_ms":104322,"concrete_test":"Run a standard Hodgkin-Huxley simulation of a single action potential at 0°C (rest -70 mV, peak +20 mV) and compute per unit membrane area: (i) the dissipative heat from ion currents, e.g. the integral of I_Na(V-E_Na)+I_K(V-E_K)+I_L(V-E_L) over the action potential, and (ii) the paper's predicted ΔU for the same waveform using its chosen surface-charge bias (σ_i=-0.1 C m^-2, σ_o=-0.05 C m^-2). Compare the magnitude of the initial heat phase in (i) with the predicted 40-70 µJ m^-2. If the ionic Joule heat is comparable to or larger than the predicted capacitive heat, the reversible electrostatic mechanism cannot be the primary heat source; if it is negligible, the central concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing premise is Eq. (20), ΔU_el = Q, which follows from asserting that no electrical work is done on the membrane domain because the bulk solutions are electroneutral. In an action potential the membrane capacitance is charged and discharged by ion currents through channels; these currents move charge across the domain boundary and dissipate energy as heat. The same first-law accounting in Appendix E includes an explicit electrical-work term V dq (Eqs. 43-44), and the Discussion concedes the model 'assumes complete reversibility: all internal energy changes are first converted into heat, and then all heat is converted back' and 'cannot discard the possibility of other sources of heat.' Section VIIC connects field energy to ionic-current dissipation, but its Eq. (33) treats a linear dielectric charged from zero field, not a gated, concentration-gradient-driven cycle. If dissipative Joule heating from Na+ and K+ currents is comparable to the capacitive energy change, the measured heat cannot be attributed primarily to reversible membrane field energy, even if the electrostatic free-energy derivation is internally correct. The numerical agreement also depends on a surface-charge bias chosen within a plausible range, but the conceptual primacy claim stands or falls on this energy-balance assumption.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper revisits the classical Condenser Theory of nervous conduction, in which the heat produced and absorbed during an action potential is attributed to the reversible release and restoration of electrical energy stored across the cell membrane. The authors build an electrostatic model of the membrane with surface charges and diffuse layers, solve Poisson-Boltzmann equations at equilibrium, and derive expressions for the electric free energy and entropy changes of the membrane and diffuse layers. They obtain a revised membrane free-energy contribution F_m = (1/2)c_m φ_t V_m and entropy corrections proportional to temperature derivatives of the permittivity and capacitance. Using typical physiological parameters, they predict a heat release of 40–70 µJ m^-2 for a depolarization from -70 mV to +20 mV, which reaches the lower part of the experimental range (60–180 µJ m^-2) only when a surface-charge bias σ_i - σ_o = -0.05 to -0.1 C m^-2 is assumed. Based on this, the abstract and Discussion conclude that the change in membrane electrical energy is the primary mechanism of heat production and absorption by neurons.","tokens_in":15588,"tokens_out":11154,"duration_ms":122050,"significance":"If the central claim could be established, the paper would make a substantial contribution to nerve biophysics by providing a thermodynamic basis for the heat of nervous conduction and by resolving the quantitative failure of the simple parallel-plate capacitor model. The work has real strengths: the electrostatic derivation in the appendices is careful and internally consistent; the paper supplies a derivation of the previously asserted entropy correction; and it identifies surface-charge asymmetry as a physically plausible contributor to the heat magnitude. However, the quantitative conclusion is not parameter-free, and the load-bearing assumption that the action potential is a reversible, work-free process is not validated. The paper is best read as a promising equilibrium-thermodynamic framework, not as a definitive proof that membrane field energy is the primary heat source.","major_comments":[{"comment":"The identification Q = ΔU_el follows from the assertion that no electrical work is done on the membrane domain because the bulk solutions are electroneutral. This premise is not valid during an action potential: the potential change is driven by Na+ and K+ currents through voltage-gated channels, which transport charge across the boundary of the membrane domain and dissipate energy as Joule heat. The first law written in Appendix E (Eqs. (43)-(44)) explicitly includes the electrical-work term V dq, so work cannot be set to zero merely because the bulk reservoirs are electroneutral. The Discussion itself concedes that the model 'assumes complete reversibility' and 'cannot discard the possibility of other sources of heat.' To support the abstract's 'primary mechanism' conclusion, the authors need to supply a quantitative energy balance that includes dissipative ion-current heating (for example, a Hodgkin-Huxley-based calculation) and shows that this contribution is small compared with the capacitive energy change.","section":"II.B, Eq. (20)"},{"comment":"The predicted heat of 40–70 µJ m^-2 is obtained for surface-charge biases in the range 0 to -0.1 C m^-2, and the experimental lower bound of 60 µJ m^-2 is reached only for the larger biases. Because the cited literature range for surface charge densities is very broad (-0.002 to -0.37 C m^-2), the bias is effectively a free parameter in the quantitative comparison. The authors should report the prediction across the full plausible range of surface-charge bias and action-potential amplitude, with uncertainty propagation, rather than presenting the chosen bias as the condition under which agreement occurs. Without this, the match to experiment is partially circular and cannot serve as evidence for the 'primary mechanism' claim.","section":"III.C and Discussion, final paragraph"},{"comment":"The Ampere-law derivation equates heat with field energy for a linear dielectric charged from zero field. This is not a model of an action-potential cycle, which starts at V_m = -70 mV, overshoots to positive values, and returns to -70 mV; for a linear lossless dielectric the net field-energy change over a complete cycle is zero. The argument therefore does not by itself demonstrate that gated ion-current dissipation is negligible, and it does not remove the need for the dynamic energy balance requested in the first major comment.","section":"VIIC, Eq. (33)"},{"comment":"The entropy correction is derived for a membrane-only capacitor whose stored free energy is (1/2) C V^2, but in the main text it is applied to F_m = (1/2)c_m φ_t V_m. In the global decomposition of Eqs. (38)-(41), F_m is a mathematical cross-term that emerges after integrating by parts and shifting the potential reference, not the physical field energy of the membrane alone, which is (1/2)c_m φ_t^2. The authors should justify why the entropy coefficient T/c_m ∂c_m/∂T should multiply F_m rather than (1/2)c_m φ_t^2; without this justification, the substantial entropy contribution shown in Fig. 3 is not rigorously established.","section":"II.D.2 and Appendix E, Eq. (26)"}],"minor_comments":[{"comment":"The value ρ = 5.8 g cm^-3 for nerve tissue appears to be a typo or an unstated definition of dry density; typical soft-tissue density is close to 1 g cm^-3. Please correct or clarify, and verify the numerical value of τ_h, although the qualitative conclusion of fast thermal equilibration is unaffected.","section":"VII.B, Eq. (29)"},{"comment":"The table lists 1/ε ∂ε/∂T as -0.43 %/K at 0°C, while the text quotes T/ε ∂ε/∂T = -1.17; these are consistent, but the distinction between the two forms should be stated explicitly to avoid confusion.","section":"Table I and Section II.D.1"},{"comment":"The captions do not state the fixed values of ionic concentrations, resting potential, and temperature used for all curves; adding these details would make the figures self-contained.","section":"Figs. 2-4"},{"comment":"The discussion of the entropy factor first argues against a positive T/c_m ∂c_m/∂T based on ∂ε/∂T of fatty acids and then adopts a positive value of +0.3%/K based on dimensional changes. The logical relation between these two lines of evidence should be stated more explicitly, since the apparent contradiction is resolved only by Eq. (27).","section":"I.D"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is well organized and the appendices show careful derivations, but the abstract overstates the conclusion relative to the reversibility assumption that the authors themselves concede in the Discussion. I recommend that the revised version be evaluated with particular attention to the requested energy-balance comparison with dissipative ion currents, as this determines whether the 'primary mechanism' claim can be supported."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper is better than the abstract makes it look. It does two concrete things: it derives a corrected expression for the membrane's electric free energy, F_m^el = 1/2 c_m φ_t V_m, from a Poisson-Boltzmann model with surface charges and diffuse layers, and it gives a first-law derivation of the entropy relation T ΔS = (T/c_m) ∂c_m/∂T ΔF that earlier authors simply assumed. Both derivations are careful, and the appendices are internally consistent. That alone is a useful correction to the condenser-theory literature, which had been stuck with ad hoc formulas.\n\nThe quantitative claim is more fragile. The paper predicts 40–70 μJ/m² of heat for a typical action potential, reaching the experimental 60–180 μJ/m², but this match depends on the surface charge bias σ_i − σ_o being around −0.05 to −0.1 C/m². The bias is not derived; it is chosen within a literature range. The bias does have independent support from lipid asymmetry and from Plaksin et al., so this is partial circularity, not a free fit, but it is still a fitted parameter in the demonstration.\n\nThe bigger issue is Eq. (20), where ΔU_el = Q follows from setting electrical work to zero because the bulk solutions are electroneutral. In an action potential, the membrane is charged and discharged by Na+ and K+ currents through channels. Those currents move charge across the domain boundary and dissipate energy. The first-law accounting in Appendix E explicitly includes a V dq term, so the neglect of work for the whole membrane domain is not obviously justified. The authors concede this in the Discussion (\"assumes complete reversibility\", \"cannot discard the possibility of other sources of heat\"), and they hedge that the electrical mechanism is the \"most prominent\" rather than the only one. But the abstract's claim of primacy is stronger than the model establishes. A Hodgkin-Huxley energy-balance check—comparing the capacitive energy change against the resistive Joule heating over the action potential cycle—would be the natural way to close this gap.\n\nThat said, the reversibility assumption is not crazy: the measured heat is reversible in the sense that production is followed by absorption, and the capacitor charging/discharging picture is the leading candidate in the literature. The paper just does not prove it. I would not desk reject this. It deserves a serious referee. The electrostatic derivation is publishable, the entropy derivation fills a gap, and the energy-balance concern is addressable in revision. My recommendation: send it to review, and ask for a direct engagement with the dissipative-current objection and a moderated central claim.\n\nWho is this for? Membrane biophysicists and anyone working on thermodynamic signatures of nerve activity. Also people doing double-layer thermodynamics in capacitive systems, since the entropy-of-dipoles contribution is a nice addition.","headline":"A careful electrostatic derivation that fixes the condenser theory's free-energy formula and entropy correction, but the quantitative match to heat measurements relies on a surface-charge bias and a reversibility assumption yet to be validated.","tokens_in":16112,"tokens_out":3190,"would_cite":true,"duration_ms":31455,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper argues that the heat produced and reabsorbed during nervous conduction is primarily the reversible release and restoration of the membrane's stored electrical energy, and that a corrected electrostatic model with asymmetric…","keywords":["action potential","nerve heat","condenser theory","membrane capacitance","surface charge asymmetry","electrical double layer","Poisson-Boltzmann model","reversible thermodynamics"],"falsifier":"Record the membrane potential intracellularly while measuring the heat produced per unit membrane area in the same nerve fiber, and compare the measured heat to the model's prediction: roughly 60 $\\mu$J m$^{-2}$ for a -70 mV to +20 mV action potential with a -0.05 C m$^{-2}$ surface-charge bias. If the observed heat is systematically larger, delayed, or poorly correlated with the reversible free-energy trace, the electrical-energy mechanism is not primary.","tokens_in":15175,"feed_emoji":"⚡","tokens_out":7287,"duration_ms":76703,"temperature":0.7,"pith_summary":"This paper seeks to explain a long-standing puzzle: why a nerve releases a small burst of heat and then reabsorbs a similar amount as an action potential passes. The authors revive the old condenser theory, which attributes this heat to electrical energy stored across the cell membrane, but they fix its quantitative failure by modeling the membrane as a charged lipid bilayer with diffuse ion layers on both sides. Their corrected expression for the membrane's electric free energy, together with entropy changes in water and in the lipid bilayer, predicts heat in the range of experimental values when the inside surface is about 0.05 C m$^{-2}$ more negative than the outside. The paper concludes that the change in the membrane's electrical energy is the primary mechanism of heat production and absorption during nervous conduction.","feed_headline":"Nerve firing heat is reversible membrane electric energy","feed_subtitle":"A corrected condenser theory with asymmetric surface charges reproduces the measured heat range.","key_machinery":"The machinery is a one-dimensional Poisson-Boltzmann model of the cell membrane as a charged lipid bilayer flanked by electrical double layers, with surface charge densities $\\sigma_i$ (inside) and $\\sigma_o$ (outside) entering through electrostatic boundary conditions. The key derived identity is $q = -c_m \\phi_t$, linking the capacitive charge to the transmembrane potential $\\phi_t$, and the central free-energy expression is $F_m^{el} = \\frac{1}{2} c_m \\phi_t V_m$. Entropy changes are computed from the temperature dependence of the water permittivity, $T\\Delta S^{DL}_{el} = \\frac{T}{\\varepsilon}\\frac{\\partial \\varepsilon}{\\partial T} \\Delta F^{DL}_{el}$, and of the membrane capacitance, $T\\Delta S^{m}_{el} = \\frac{T}{c_m}\\frac{\\partial c_m}{\\partial T} \\Delta F^{m}_{el}$, with the membrane term traced to changes in bilayer thickness and area rather than dielectric permittivity.","core_discovery":"The central claim is that the heat produced and reabsorbed by neurons during an action potential is primarily the reversible change in the electrical energy stored in the membrane and its adjacent diffuse layers. The authors derive the electric free energy of the membrane as $F_m^{el} = \\frac{1}{2} c_m \\phi_t V_m$, where $\\phi_t$ is the potential difference across the membrane surfaces and $V_m$ is the bulk membrane potential; this replaces the earlier parallel-plate formula $\\frac{1}{2} c_m V_m^2$, which underestimates the heat, and also corrects the overestimate from using $\\frac{1}{2} c_m \\phi_t^2$. They add entropy terms from the temperature dependence of the dielectric permittivity of water and of the membrane capacitance, and show that a realistic surface-charge bias, with the internal side more negative than the external side, increases the predicted heat substantially. For a typical depolarization from -70 mV to +20 mV with a bias of -0.05 C m$^{-2}$, the predicted heat is about 60 $\\mu$J m$^{-2}$, inside the experimentally measured range of 60-180 $\\mu$J m$^{-2}$. If the action potential starts from a more negative resting potential, the model predicts up to 150 $\\mu$J m$^{-2}$.","pith_inferences":["If the capacitive mechanism is right, a high-resolution thermal measurement combined with an intracellular voltage trace could serve as a non-invasive probe of the membrane's surface-charge asymmetry; the paper does not develop this inverse use.","The same free-energy expression should apply to artificial lipid bilayers and to capacitive energy-storage devices with surface charges and dipole entropy, where the entropy of dipoles is often neglected; the paper gestures toward these engineering connections but leaves quantitative predictions for those systems unstated.","If simultaneous heat and voltage recordings later reveal a heat component that lags the membrane potential or grows with ionic current, that would point to Joule heating in ion channels or electromechanical dissipation as a second contributor; the present model assumes such effects are negligible.","A direct testable extension is to measure the heat per unit area and the membrane potential time course on the same small nerve fiber; the model predicts a specific heat magnitude for a given action potential and surface-charge bias, so deviations would sharpen where the equilibrium assumption fails."],"forward_implications":["If the mechanism is correct, the thermal signature of a nerve impulse is a reversible capacitive effect: heat released during depolarization is reabsorbed during repolarization, leaving a net heat of roughly zero over a full action potential.","The predicted heat depends strongly on the asymmetry of surface charge between the two sides of the membrane; equal negative surface charges on both sides contribute almost nothing, while an internal bias of -0.05 C m$^{-2}$ brings the heat into the measured range.","Larger action potentials, starting from resting potentials as negative as -100 mV, yield predicted heats up to 150 $\\mu$J m$^{-2}$, covering the upper end of the experimental range.","Entropy changes in the diffuse layers and in the membrane are comparable in magnitude but opposite in sign, so any accurate heat calculation must include both; they partially offset each other.","With a surface-charge bias, the membrane can continue to release heat even after the membrane potential overshoots to positive values, which changes the shape of the heat trace and removes the 'notch' seen when the charges are symmetric."],"supporting_citations":[{"why":"Supplies the experimental heat values and the temperature time-course evidence supporting the condenser theory.","marker":"[5]"},{"why":"Proposed the transmembrane-potential version of the condenser theory and the entropy factor that this paper rederives.","marker":"[13]"},{"why":"Provides the coupled Poisson-Boltzmann model of the charged lipid bilayer that the paper extends.","marker":"[20]"},{"why":"Gives the surface-charge-bias estimate and the +0.3%/C temperature dependence of membrane capacitance used as model inputs.","marker":"[18]"},{"why":"Supplies the dielectric entropy relation for polarized media that the paper adapts to water and to the membrane.","marker":"[24]"},{"why":"Serves as the source for ion-channel physiology, resting potential, and surface-charge density ranges in excitable membranes.","marker":"[9]"},{"why":"Provides the standard resting potential (-70 mV) and cellular biophysics parameters used in the calculations.","marker":"[16]"},{"why":"Supplies the action-potential depolarization to +20 mV used in the heat estimate.","marker":"[33]"}],"fun_headline_variants":["Neuron heat traced to asymmetric membrane charge bias","Membrane charge bias drives neuron heat signature","Asymmetric membrane charges set neuron heat output","Reversible membrane charge shift explains neuron heat","Neuron heat burst: membrane bias sets the magnitude"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the membrane and its surrounding ion layers are in equilibrium throughout the action potential, so every change in the electric field's energy is converted into heat and back again without other work or dissipation; if the opening and closing of ion channels injects significant Joule heat or mechanical work, that premise breaks and the predicted heat no longer matches the measured one.","fun_headline_variants_meta":{"raw":{"variants":["Neuron heat traced to asymmetric membrane charge bias","Membrane charge bias drives neuron heat signature","Asymmetric membrane charges set neuron heat output","Reversible membrane charge shift explains neuron heat","Neuron heat burst: membrane bias sets the magnitude"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000662,"raw_usage":{"total_tokens":3091,"prompt_tokens":1079,"completion_tokens":2012,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":695,"completion_tokens_details":{"reasoning_tokens":1953}},"tokens_in":695,"tokens_out":2012,"duration_ms":18465,"temperature":1.0,"reasoning_tokens":1953,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:20:45.898333+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Record the membrane potential intracellularly while measuring the heat produced per unit membrane area in the same nerve fiber, and compare the measured heat to the model's prediction: roughly 60 $\\mu$J m$^{-2}$ for a -70 mV to +20 mV action potential with a -0.05 C m$^{-2}$ surface-charge bias. If the observed heat is systematically larger, delayed, or poorly correlated with the reversible free-energy trace, the electrical-energy mechanism is not primary.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the experimental heat values and the temperature time-course evidence supporting the condenser theory."},{"cited_title":"membrane domain","cited_arxiv_id":null,"evidence_quote":"Proposed the transmembrane-potential version of the condenser theory and the entropy factor that this paper rederives."},{"cited_title":"Ritchie, Progress in biophysics and molecular biology 26, 147 (1973)","cited_arxiv_id":null,"evidence_quote":"Provides the coupled Poisson-Boltzmann model of the charged lipid bilayer that the paper extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the surface-charge-bias estimate and the +0.3%/C temperature dependence of membrane capacitance used as model inputs."},{"cited_title":"Plaksin, E","cited_arxiv_id":null,"evidence_quote":"Supplies the dielectric entropy relation for polarized media that the paper adapts to water and to the membrane."},{"cited_title":"Neher and B","cited_arxiv_id":null,"evidence_quote":"Serves as the source for ion-channel physiology, resting potential, and surface-charge density ranges in excitable membranes."},{"cited_title":"Hille,Ion channels of excitable membranes, Vol","cited_arxiv_id":null,"evidence_quote":"Provides the standard resting potential (-70 mV) and cellular biophysics parameters used in the calculations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the action-potential depolarization to +20 mV used in the heat estimate."}],"review_version":1}