{"id":"fa5866e2-fc69-4e4b-a366-c2dd312ca0b4","arxiv_id":"2412.06367","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Two molecular dynamics techniques bracket the melting point of CaO near 2940 to 3066 K and yield a high-pressure melting curve up to 30 GPa that shows the overheating ratio grows with pressure.","lead":"Molecular dynamics simulations give the melting point of calcium oxide (CaO) as about 2940 to 3066 K at ambient pressure, depending on the simulation method. The study also maps how the melting temperature of CaO rises with pressure up to 30 GPa, which matters for modeling planetary interiors and high-temperature ceramics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Short-range unphysical attraction in the BMH potential may bias the high-pressure Tf values and the central overheating-ratio trend; the paper's own 30 GPa limit is the load-bearing weak point.","rationale":"The paper's strongest claim is the direct high-pressure melting curve and the pressure-dependent overheating ratio, and the load-bearing premise is that the BMH potential remains physically reliable up to 30 GPa. The reader's weakest_assumption correctly identifies this. The internal inconsistency between the text and Table 3 (16-24% vs 19.1-31.8% overheating) is real but secondary, since the qualitative trend is present in both; the ~126 K difference between the two ambient Tf determinations is within about 2 sigma and does not by itself invalidate the pressure trend. The potential issue is more fundamental because it affects every high-pressure point in both Tf and Ts, and the authors themselves state that the short-range behavior becomes 'very serious' at high pressure and that the 30 GPa limit is set by the lack of repulsion. The paper does have independent support: the ambient enthalpy of fusion agrees with NIST-JANAF and Alvares et al., thermal expansion matches experiment, and the ambient Tf falls within the experimental spread. Those checks, however, do not extend to 30 GPa. A modified-potential rerun or AIMD coexistence at selected pressures would settle whether the central trend is an artifact. Because this is a targeted concern rather than a demonstrated error, the CONDITIONAL verdict remains appropriate.","tokens_in":15450,"tokens_out":4262,"duration_ms":45681,"concrete_test":"Recompute Tf and Ts at 0, 15, and 30 GPa using the same BMH potential augmented with a steep repulsive 24-6 Lennard-Jones term on Ca-O pairs—the 'accepted solution' noted in Sec. 2.1—tuned to remove the unphysical attractive well while reproducing the ambient lattice parameter and thermal expansion. If either Tf shifts by more than the reported ~50 K uncertainty or the overheating ratio changes by more than ~2 percentage points relative to Table 3, the pressure-dependence claim is not robust to the known short-range defect.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that Ts(P)/Tf(P) increases with pressure, so assuming a constant overheating ratio is invalid—requires that both the two-phase Tf and the homogeneous Ts be accurate at 30 GPa. The paper's own Sec. 2.1 states that the BMH potential is 'attractive at very short distances for the cation-anion pairs,' that this well is 'unphysical,' and that 'whenever a Ca-O pair gets too close, the molecular dynamics run is spoiled'; Sec. 3.3 says the 30 GPa ceiling exists precisely because of the 'lack of a short-range repulsive barrier.' A potential with an unphysical short-range well can bias the solid-liquid free-energy difference under compression: if the liquid samples the spurious attractive well, or is prevented from sampling physically correct close Ca-O contacts, Tf(P) is shifted relative to the ambient anchor. The reported high-pressure curve is anchored to the two-phase coexistence value Tf = 2940 K rather than the void-melting 3066 K, and no independent high-pressure benchmark is given. Even if the qualitative trend survives, a bias of ~100-200 K at 30 GPa would materially change the quantitative overheating ratios and the fitted Ts(P)-Tf(P) separation, which are the paper's main physical conclusion.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript reports classical and ab initio molecular dynamics simulations of the melting behavior of CaO. At ambient pressure, the void-nucleated melting technique gives Tf = 3066 ± 12 K and the two-phase coexistence technique gives Tf = 2940 ± 65 K. Using classical MD with the two-phase method, the authors compute the high-pressure melting curve and the thermal instability curve up to 30 GPa, and they report that the overheating ratio η = Ts/Tf − 1 increases with pressure, from 19.1% at ambient pressure to 31.8% at 30 GPa in Table 3. This is used to argue that assuming a constant overheating ratio, as in earlier work by Sun et al., is not justified. The manuscript also reports an enthalpy of fusion of 80.37 kJ/mol from classical MD, in good agreement with tabulated values and with prior simulations.","tokens_in":15641,"tokens_out":4495,"duration_ms":47129,"significance":"If the reported results are correct, the paper provides a useful direct melting curve for CaO up to 30 GPa without imposing a constant Clapeyron slope or a constant overheating ratio, and it identifies a pressure-dependent overheating ratio that bears on extrapolation strategies for refractory oxides. The paper has notable strengths: it compares two independent melting techniques, validates the caloric curve against NIST-JANAF and earlier simulations, and combines classical and ab initio MD. The enthalpy of fusion result is robust and well aligned with literature values. However, the central quantitative claims currently rest on a numerical inconsistency between the text and Table 3, and on a classical potential whose short-range unphysical attraction is acknowledged by the authors; these issues need to be resolved before the main conclusions can be fully accepted.","major_comments":[{"comment":"The text states that the overheating ratio increases 'from about 16% at ambient pressure to more than 24% at P=30 GPa,' but Table 3 lists 19.1% at 0 GPa and 31.8% at 30 GPa for the same quantities. Since the pressure dependence of η is the main physical conclusion of the paper, this discrepancy must be reconciled. In addition, Eq. (2) does not reproduce the tabulated values at P = 0: the fit gives Tf = 2926.2 K and Ts = 3480.6 K, whereas Table 3 reports 2940 K and 3503 K. Please provide the residuals or goodness-of-fit for Eq. (2) and explain the anchoring choice.","section":"Sec. 3.3 and Table 3"},{"comment":"The BMH potential is described as 'attractive at very short distances for the cation-anion pairs,' with an 'unphysical' attractive well, and the paper states that the 30 GPa ceiling exists because of the 'lack of a short-range repulsive barrier.' Since both the melting curve Tf(P) and the thermal instability curve Ts(P) are computed with this potential, a short-range artifact that is sampled differently by the solid and liquid under compression could bias both quantities and hence the reported increase of η with pressure. The potential was validated for liquid structure and enthalpy at ambient pressure, but no high-pressure validation is provided. Please quantify the spurious close Ca–O contact populations at high pressure, or repeat key points (e.g., 15 and 30 GPa) with an added repulsive short-range term or with ab initio MD, and report whether the η(P) trend is preserved.","section":"Sec. 2.1 and Sec. 3.3"},{"comment":"The two ambient-pressure melting temperatures, 3066 ± 12 K (void-nucleated) and 2940 ± 65 K (two-phase), differ by 126 K, which is roughly 1.9 combined standard errors. Calling these values 'consistent' is too strong without a quantitative reconciliation, especially because the paper chooses 2940 K as the zero-pressure anchor for the high-pressure melting curve while using 3066 K for the enthalpy-of-fusion calculation and for comparison with laser-heating experiments. An offset of this size propagates directly into the reported Tf(P) curve. Please justify the choice of the two-phase value as the anchor and discuss the systematic error this introduces.","section":"Sec. 3.1.1, Sec. 3.1.2, and Sec. 3.3"}],"minor_comments":[{"comment":"The word 'Claperyon' appears in the abstract and should be 'Clapeyron.'","section":"Abstract"},{"comment":"The phrase 'Contrary to what is reported in Ref.,13' contains a stray comma before the reference number; please correct the citation format.","section":"Sec. 3.1.2"},{"comment":"The sentence 'This can be explained by empirical laws described in which have the same form of the Equations 2 and 3 above' is incomplete; a reference or derivation appears to be missing.","section":"Sec. 3.3"},{"comment":"Table 2 lists the AIMD values at 3000 K and 3200 K under 'enthalpy of fusion,' while the text calls these values 'solution enthalpy' at temperatures other than the melting temperature; please clarify the distinction and relabel the table entries accordingly.","section":"Sec. 3.2 and Table 2"},{"comment":"Please specify how the uncertainty of ±65 K for the two-phase Tf was obtained; the description of the NPH runs does not indicate whether this is a standard deviation of time averages, a fit uncertainty, or an estimate of finite-size effects.","section":"Sec. 3.1.2"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope for a condensed-matter/materials modeling journal, and the central idea of computing the CaO melting curve without constant-overheating or constant-Clapeyron assumptions is valuable. The main blocking issues are the numerical inconsistency between the text and Table 3 for η, the unquantified effect of the unphysical short-range attraction on the high-pressure results, and the unresolved 126 K discrepancy between the two ambient-pressure melting temperatures. These are addressable with additional analysis or targeted simulations, so I do not recommend rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper delivers a direct two-phase coexistence melting curve for CaO up to 30 GPa and argues that the overheating ratio is pressure-dependent, not constant. That is a genuinely useful result for geochemistry and for people modeling CaO-bearing systems. The enthalpy of fusion (80.37 kJ/mol) matches NIST-JANAF and Alvares et al. well, and the caloric curves look carefully done. The authors also deserve credit for using the same BMH potential as Alvares et al. and explicitly comparing the void-nucleated and two-phase methods; that makes the method comparison meaningful.\n\nThe soft spots are real but addressable. First, the text says the overheating ratio goes from about 16% at ambient pressure to more than 24% at 30 GPa, but Table 3 lists 19.1% to 31.8%. That is not a rounding difference, and it directly affects the paper's main claim. Second, the two ambient-pressure melting temperatures (3066 ± 12 K and 2940 ± 65 K) differ by about 126 K, roughly two combined standard deviations; calling them \"consistent\" without any reconciliation is too glib. The reader deserves a comment on whether this reflects a systematic bias in one method or a statistical fluke. Third, the BMH potential has an unphysical short-range attractive well, which the authors acknowledge and work around by stopping at 30 GPa. This is a legitimate worry for the high-pressure part of the curve: if the liquid samples the spurious well under compression, the pressure dependence of Tf could shift by a few tens to a hundred kelvin, and the quantitative overheating ratios would follow. The qualitative trend (Ts rising faster than Tf) is supported by the Lennard-Jones result they cite, so I am not calling the central claim wrong, but the quantitative numbers in Table 3 should be treated with caution until the potential issue is discussed more thoroughly. The \"first time without assuming a Clapeyron slope\" claim also deserves a softer phrasing, since Wang et al. (Ref. 13) computed a melting curve with different potentials.\n\nNet: this is a solid, citable contribution that needs a careful revision on the numbers and a frank discussion of the potential's short-range behavior. I would send it to peer review; a good referee can push the authors to clean up the inconsistencies and test the sensitivity of the pressure trend.","headline":"Useful direct melting curve for CaO to 30 GPa, but internal number mismatches and a known short-range potential flaw need fixing before I'd trust the quantitative overheating ratios.","tokens_in":16305,"tokens_out":1336,"would_cite":true,"duration_ms":15511,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A molecular dynamics study computes the melting curve of calcium oxide up to 30 GPa without assuming a constant overheating ratio, and finds the ratio grows with pressure.","keywords":["calcium oxide","melting curve","molecular dynamics","two-phase coexistence","overheating ratio","Born-Meyer-Huggins potential","high pressure","enthalpy of fusion"],"falsifier":"Measure the CaO melting temperature in a laser-heated diamond anvil cell at 10, 20, and 30 GPa; the paper predicts 3779 ± 45 K, 4224 ± 48 K, and 4444 ± 52 K. If the measured values fall outside roughly the quoted uncertainties, the potential-based melting curve is falsified; alternatively, a free-energy (thermodynamic integration) calculation with the same potential or an ab initio potential would show whether the slope is an artifact.","tokens_in":15118,"feed_emoji":"🔥","tokens_out":8259,"duration_ms":68234,"temperature":0.7,"pith_summary":"Calcium oxide (CaO, lime) is a key component of planetary mantles and refractory ceramics, yet its melting point is experimentally uncertain by hundreds of kelvin because the material is reactive, volatile, and hard to contain at high temperature. This paper uses classical molecular dynamics to compute the melting temperature of CaO at ambient pressure with two independent techniques, obtaining 3066 ± 12 K by void-nucleated melting and 2940 ± 65 K by solid-liquid two-phase coexistence, and then extends the two-phase method to produce the first direct melting curve up to 30 GPa without assuming a constant Clapeyron slope or a constant overheating ratio. The central new result is that the thermal instability temperature Ts of the perfect crystal rises with pressure faster than the equilibrium melting temperature Tf, so the overheating ratio grows from about 19% at ambient pressure to about 32% at 30 GPa. That finding matters because previous high-pressure estimates of the CaO melting curve were obtained by scaling Ts by a fixed factor, an assumption the paper shows to be unjustified. If correct, the paper provides a direct computational melting curve for a geochemically central oxide and a cautionary result for similar scaling approaches.","feed_headline":"CaO melting curve computed directly up to 30 GPa","feed_subtitle":"For lime (CaO), the overheating ratio grows with pressure, so constant-scaling melting estimates miss the mark.","key_machinery":"The argument is carried by classical molecular dynamics with the Born-Meyer-Huggins potential (ionic charges ±1.2e) combined with the two-phase solid-liquid coexistence technique. In that technique a supercell is half solid and half liquid, and after equilibration in the NPT ensemble the system is run in the isobaric-isenthalpic (NPH) ensemble: if the temperature drifts upward the liquid is recrystallizing, if it drifts downward the solid is melting, and the plateau temperature where neither happens is Tf. For the ambient-pressure cross-check the paper also uses void-nucleated melting, where a spherical cavity of increasing size is carved into the crystal until the apparent melting temperature plateaus. The high-pressure curve is obtained by repeating the coexistence simulation at 5 GPa intervals up to 30 GPa, with empirical fits $T_f(P) = 235.241 P^{0.558} + 2926.195$ K and $T_s(P) = 300.369 P^{0.618} + 3480.601$ K.","core_discovery":"The paper's central claim is that the melting curve of CaO can be computed directly, without empirical input about the Clapeyron slope or the overheating ratio, and that when this is done the melting curve deviates from the constant-scaling picture. At ambient pressure the two-phase coexistence method yields Tf = 2940 ± 65 K and the void-nucleated method yields 3066 ± 12 K, bracketing recent laser-heating measurements; the melting temperature then rises to 3432 ± 43 K at 5 GPa and 4444 ± 52 K at 30 GPa. The thermal instability temperature Ts, measured on defect-free crystals, rises from 3503 K to 5858 K over the same pressure range, so the overheating ratio η = Ts/Tf − 1 increases from 19.1% to 31.8%. This pressure dependence directly contradicts the assumption, used in the only previous high-pressure melting curve of CaO, that Ts can be scaled by a constant factor to obtain Tf. The paper therefore positions its two-phase coexistence calculation as the first direct determination of the CaO melting curve beyond ambient pressure.","pith_inferences":["Editorial inference: The same constant-scaling shortcut that this paper invalidates for CaO is used elsewhere for other refractory oxides, so if the pressure dependence of the overheating ratio is generic, those melting curves may need revisiting.","Editorial inference: Because the paper's own potential is acknowledged to become unphysical at very short Ca-O distances, an immediate test would be to recompute the 30 GPa point with a potential that adds a steep repulsive wall and see whether the predicted flattening of the curve survives.","Editorial inference: The ambient-pressure bracket of 2940–3066 K could be sharpened with longer isobaric-isenthalpic runs, which would also separate the two recent laser-heating measurements more decisively."],"forward_implications":["The CaO melting curve is available from 0 to 30 GPa as a direct simulation output, so thermodynamic assessments of CaO-bearing planetary and ceramic systems no longer need a Clapeyron slope borrowed from low-pressure experiments.","The overheating ratio of CaO rises from about 19% at ambient pressure to about 32% at 30 GPa, so scaling the thermal instability temperature by a fixed factor underestimates the melting temperature by an amount that grows with pressure.","The ambient-pressure agreement between the two-phase (2940 ± 65 K) and void-nucleated (3066 ± 12 K) techniques indicates the computed melting temperature is not an artifact of a single nucleation protocol.","The classical-MD enthalpy of fusion, 80.37 kJ/mol, matches the assessed tabulated value of 79.5 kJ/mol, supporting the thermodynamic consistency of the potential used for the melting curve."],"supporting_citations":[{"why":"Supplies the Born-Meyer-Huggins potential and the previous void-melting computational reference whose values this work reproduces and refines.","marker":"[12]"},{"why":"The only previous high-pressure melting curve of CaO, obtained by scaling the thermal instability temperature by a constant factor; the target the paper shows to be unjustified.","marker":"[15]"},{"why":"Laser-heating measurement that fixes the upper recent experimental melting point used for comparison.","marker":"[4]"},{"why":"Laser-pulse melting measurement that fixes the lower recent experimental melting point used for comparison.","marker":"[8]"},{"why":"Assessed thermochemical tables that supply the reference enthalpy of fusion and the ambient melting temperature used in the constant-scaling approach.","marker":"[16]"},{"why":"Prior Lennard-Jones simulation study reporting the same pressure-dependent divergence between melting and overheating temperatures, used to rationalize the CaO result.","marker":"[63]"}],"fun_headline_variants":["CaO melting curve up to 30 GPa from two MD methods","Direct CaO melting curve challenges constant overheating ratio","CaO melting: overheating ratio grows with pressure, simulations show","First direct high-pressure CaO melting curve from simulations","CaO melting curve: two methods agree, scaling assumption fails"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the Born-Meyer-Huggins potential with ionic charges ±1.2e stays accurate enough for computing melting at compressions up to 30 GPa, even though at very short Ca-O distances the potential becomes attractive and unphysical and can spoil the simulation; the paper itself treats 30 GPa as the upper limit for this reason.","fun_headline_variants_meta":{"raw":{"variants":["CaO melting curve up to 30 GPa from two MD methods","Direct CaO melting curve challenges constant overheating ratio","CaO melting: overheating ratio grows with pressure, simulations show","First direct high-pressure CaO melting curve from simulations","CaO melting curve: two methods agree, scaling assumption fails"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.0008,"raw_usage":{"total_tokens":3586,"prompt_tokens":1082,"completion_tokens":2504,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":698,"completion_tokens_details":{"reasoning_tokens":2421}},"tokens_in":698,"tokens_out":2504,"duration_ms":16781,"temperature":1.0,"reasoning_tokens":2421,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T19:44:21.476290+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the CaO melting temperature in a laser-heated diamond anvil cell at 10, 20, and 30 GPa; the paper predicts 3779 ± 45 K, 4224 ± 48 K, and 4444 ± 52 K. If the measured values fall outside roughly the quoted uncertainties, the potential-based melting curve is falsified; alternatively, a free-energy (thermodynamic integration) calculation with the same potential or an ab initio potential would show whether the slope is an artifact.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Born-Meyer-Huggins potential and the previous void-melting computational reference whose values this work reproduces and refines."},{"cited_title":"The high-pressure melting curve of CaO","cited_arxiv_id":null,"evidence_quote":"The only previous high-pressure melting curve of CaO, obtained by scaling the thermal instability temperature by a constant factor; the target the paper shows to be unjustified."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Laser-heating measurement that fixes the upper recent experimental melting point used for comparison."},{"cited_title":"Laser‐pulse melting of calcium oxide and some peculiarities of its high‐temperature behavior","cited_arxiv_id":null,"evidence_quote":"Laser-pulse melting measurement that fixes the lower recent experimental melting point used for comparison."},{"cited_title":"NIST-JANAF Thermochemical Tables, 4th Edition; American Institute of Physics, 1998","cited_arxiv_id":null,"evidence_quote":"Assessed thermochemical tables that supply the reference enthalpy of fusion and the ambient melting temperature used in the constant-scaling approach."}],"review_version":1}