{"id":"5df52e4b-050b-4fac-b87e-4dcdf581ea6c","arxiv_id":"2607.13793","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Reduced-mantle super-Earths can transiently re-inflate late in life, with bulk density drops up to ~60%, as escape-driven pressure loss turns dissolved water into a light H2 atmosphere.","lead":"A simulation study of close-in rocky super-Earths finds that planets with chemically reduced mantles can temporarily inflate late in their lives: as photoevaporation strips a heavy CO atmosphere, water dissolved in the magma ocean degasses as light H2, puffing the planet up before final loss. If real, the effect would give observers a way to read a planet's deep geochemistry from its radius and atmosphere.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Fixed mantle fO2 is the pivotal untested assumption: H-escape oxidation may push the mantle past ΔIW+2 before the CO envelope is stripped, suppressing the H2O→H2 transition that drives reflation.","rationale":"The reader's weakest assumption—fixed mantle fO2—is exactly the load-bearing concern. The paper's central claim is that reflation arises from a redox-controlled compositional transition, but the model holds the redox state constant while hydrogen escape necessarily removes H and leaves O, which should drive oxidation. The paper acknowledges this in Section 4.3, but its defense ('provided this condition is met') is untested and not derived from the model. Because the mechanism is defined by the timing of the CO-stripping relative to the redox window, a self-consistent fO2 evolution could eliminate the reflation peak entirely or move it outside the quoted parameter range. The concrete test of adding Fe/FeO bookkeeping would directly settle this. Other limitations, such as the uncertain escape efficiency η, are handled by a parameter study and are less central because the paper does not claim a precise η value; however, fixed fO2 is an internal omission rather than an external parameter uncertainty. Therefore the reader's conditional verdict is appropriate, and no change to that verdict is needed.","tokens_in":22975,"tokens_out":8254,"duration_ms":86616,"concrete_test":"Modify the strongest reflation simulation (ΔIW=0, Hocean=20, η=1e-4, a=0.01 AU; Section 4.2) to include a time-dependent redox bookkeeping: track the mantle Fe-metal/FeO ratio, consume O from H2O dissociation as H2 escapes, and update ΔIW each timestep using an ideal-solution FeO activity model. Record the time when ΔIW exceeds +2 and compare with the time when the CO volume mixing ratio drops below 0.5. If oxidation precedes CO stripping, the reflation peak in Figure 2 is suppressed; if not, the fixed-fO2 assumption is benign for the mechanism.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reflation mechanism requires that the mantle remain reduced (fO2 ≲ ΔIW+2) until the initial CO-dominated envelope is stripped and dissolved H2O degasses as H2. Section 4.3 explicitly fixes ΔIW for each simulation and acknowledges that hydrogen escape can oxidize the mantle, citing Kasting et al. (1993), Katyal et al. (2020), and Cherubim et al. (2025). The paper's response—'provided this condition is met within the relevant evolutionary window, the qualitative compositional transition is preserved'—is a supposition, not a calculation. The oxidant source is internal to the model: every H atom lost as H2 leaves behind an O atom from H2O dissociation, which must be accommodated by the mantle's Fe/FeO buffer. Whether this shifts fO2 from IW to IW+2 before the CO column is stripped depends on the buffering capacity of Fe metal and FeO, which is not tracked (Section 2.2 explicitly treats oxygen as not conserved across simulations). If oxidation occurs first, the late-stage outgassing would produce H2O/CO2 rather than H2, the atmospheric MMW would not drop, and the reflation peak in Figure 2 would be suppressed. This is not a matter of external consensus; it is an internal sensitivity: the central claim requires a condition the model does not simulate.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper uses the proteus coupled interior-atmosphere evolution model to simulate close-in irradiated super-Earths with secondary, outgassed atmospheres undergoing hydrodynamic escape. It proposes a mechanism called 'reflation': for geochemically reduced mantles near the iron-wüstite buffer, an initially CO-dominated secondary atmosphere is stripped by escape; the accompanying pressure drop degasses H2O dissolved in the magma ocean, which is converted to H2, lowering the atmospheric mean molecular weight and transiently inflating the radius / decreasing the bulk density by up to ~60% before complete atmospheric erosion. Oxidized mantles instead outgas heavy CO2/SO2-rich atmospheres and evolve by monotonic deflation. A parameter study maps reflation occurrence to ΔIW ≈ 0, intermediate escape efficiencies η ≈ 10^-3 to 10^-4, semi-major axes a ≲ 0.05 AU, and initial hydrogen inventories ≳ 5 Earth oceans. The authors discuss observational population-level signatures and provide an unusually explicit limitations section.","tokens_in":23289,"tokens_out":11339,"duration_ms":146221,"significance":"If the mechanism is robust, reflation is a qualitatively novel, non-monotonic evolutionary pathway for super-Earths and a potential tracer of deep-mantle redox, with falsifiable population-level predictions (an excess of puffy super-Earths near the radius valley and correlated atmospheric composition anomalies). The paper's strengths are its broad parameter grid, the explicit crossover-flux check supporting unfractionated escape in the reflation phase, and the candid treatment of model limitations. However, the central quantitative claim is conditional on assumptions -- most importantly fixed mantle fO2 with non-conserved oxygen -- that are acknowledged but not quantified. The paper therefore reads as a solid proof-of-concept rather than a definitive predictive claim.","major_comments":[{"comment":"The reflation mechanism requires that the mantle remain reduced (fO2 ≲ ΔIW+2) at the time the initial CO-dominated envelope is stripped. In the model, ΔIW is held fixed throughout each simulation and oxygen is explicitly not conserved across simulations (§2.2), so the oxidant released by H escape (each H lost as H2 leaves an O atom from H2O) cannot feedback on mantle redox. The statement in §4.3 that 'provided this condition is met within the relevant evolutionary window, the qualitative compositional transition is preserved' is an assertion, not a calculation: the Fe/FeO buffering capacity is not tracked. Since Fig. 3 boundaries and Fig. 1 onset times are derived under this assumption, the central quantitative claim is conditional on an untested redox trajectory. Please add a simple oxygen bookkeeping estimate (integrated H escape versus accessible FeO/Fe-metal reservoir) or couple a re","section":"§4.3 and §2.2"},{"comment":"The f/fc test is computed only for the H2-dominated inflated phase, which is the endpoint of the reflation process. The preceding CO-dominated stripping phase sets the timing and feasibility of the H2O→H2 transition; if diffusive fractionation removes H2 preferentially during that phase, the H2 reservoir may be depleted before it can re-inflate the atmosphere. The statement that fractionation would 'modulate rather than erase' the mechanism is plausible but not derived from the presented diagnostic. Please extend the crossover estimate to the CO-dominated phase, or justify explicitly why the H2-dominated criterion is the bounding case.","section":"§4.3, crossover-flux paragraph"}],"minor_comments":[{"comment":"'fO2 between ΔIW±0 and ΔIW+2' should read 'ΔIW = 0 to +2'.","section":"§4.2"},{"comment":"'inspired by the physical characteristics of to the ultra-short period super-Earth' has a grammatical duplication ('of to').","section":"§2.4"},{"comment":"The Hu et al. (2024) Nature entry is duplicated in the reference list.","section":"References"},{"comment":"The x-axis tick labels appear to omit the 5 and 6 decade markers in the log-time axis; please check the figure rendering.","section":"Figure 2"},{"comment":"'IsoF ATE' appears to be a formatting/name error; presumably this should be 'IsoFATE'.","section":"§4.3"},{"comment":"The color scale would benefit from an explicit label stating that lower values of min(ρ_bulk/ρ_bulk,0) correspond to stronger reflation; the caption partly says this but the colorbar itself is not labeled in the figure.","section":"Figure 3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a transparent and well-scoped proof-of-concept. The central issue is the fixed-fO2 / non-conserved-oxygen assumption; a quantitative sensitivity test on mantle redox evolution would materially strengthen the claim. I do not see grounds for rejection: the mechanism is physically coherent, the parameter coverage is broad, and the authors are candid about limitations. A revision that addresses the redox-evolution concern and the partial escape-fractionation diagnostic would make the paper suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Briefly: this paper shows that in highly irradiated, geochemically reduced super-Earths, the coupled feedback between redox-controlled outgassing and hydrodynamic escape can produce a late, transient atmospheric re-inflation. An initially CO-dominated secondary atmosphere is stripped, dissolved H2O degasses and converts to H2, the mean molecular weight drops, and bulk densities decrease by up to ~60% before final erosion. That is genuinely new — the cited escape models assume monotonic stripping with fixed composition. The reflation behavior is an emergent output of the equations, not a fitted target, and the authors run a broad parameter grid. The crossover-flux check in §4.3 gives real support for the unfractionated-escape assumption in the regime that matters. The limitations section is candid.\n\nThe main soft spot is exactly the one the stress-test flags: mantle fO2 is held fixed for each simulation. The mechanism requires the mantle to still be reduced when the CO envelope is stripped, and hydrogen escape is itself an oxidizing process — every H lost as H2 leaves an O behind that has to be accommodated by the Fe/FeO buffer. The authors acknowledge this and say the qualitative transition is preserved 'provided this condition is met,' but that is a supposition, not a calculation. I don't think it kills the mechanism — the buffering capacity of a magma ocean with Fe metal could plausibly absorb the oxidant — but it is the difference between a mechanism paper and a predictive one. The quantitative boundaries in Figure 3 and the observability estimates should be treated as conditional until self-consistent redox evolution is included.\n\nOther soft spots are minor. The escape efficiency η is held constant and is genuinely unconstrained for high-MMW secondary atmospheres; the authors say so, and their chosen range is defensible. Species-dependent fractionation is not modeled, but the crossover argument suggests it won't erase the effect. The paper doesn't release run artifacts; for a grid study that is annoying but not disqualifying. Also fix the abstract's 'water inventories' — the parameter is hydrogen.\n\nWho should read it: modelers of radius-valley evolution and anyone planning PLATO/Ariel population surveys. Bottom line: the mechanism is credible, the writing is clear, and the central claim holds up as a qualitative result. The paper deserves a serious referee. Send it to peer review, with a request for the authors to add either a simple redox-evolution sensitivity test or a quantitative buffering argument. If the next version shows reflation survives oxidation, I'd be happy to cite it.","headline":"A credible new mechanism for transient super-Earth inflation, but the fixed mantle redox assumption needs testing before quantitative predictions are trusted.","tokens_in":23806,"tokens_out":3110,"would_cite":true,"duration_ms":30345,"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":"Geochemically-reduced super-Earths can transiently re-inflate their atmospheres late in their evolution, temporarily cutting bulk density by up to ~60% before final erosion, a signature of deep mantle redox.","keywords":["super-Earths","atmospheric escape","redox geochemistry","magma ocean","outgassing","radius valley","mean molecular weight","planetary evolution"],"falsifier":"A concrete test: use a self-consistent model that couples mantle oxidation to hydrogen escape. If the mantle oxygen fugacity rises above +2 relative to the iron-wüstite buffer before the CO envelope is stripped enough for H2O to degas as H2, reflation disappears. Observationally, a large unbiased survey of old (1–5 Gyr) close-in super-Earths should reveal a population of low-density planets (bulk density roughly 40–60% of a bare-rock model) with H2/H2O-dominated secondary atmospheres; the absence of such objects—or a candidate caught in the CO-dominated pre-transition state that never drops in","tokens_in":22832,"feed_emoji":"🪐","tokens_out":11015,"duration_ms":93933,"temperature":0.7,"pith_summary":"This paper argues that a super-Earth's deep mantle redox state can reverse the usual one-way story of atmospheric stripping. Using coupled interior-atmosphere simulations of highly irradiated planets, the authors show that reduced mantles near the iron-wüstite buffer initially outgas heavy CO-dominated atmospheres; as hydrodynamic escape strips this envelope, the falling surface pressure degasses water stored in the underlying magma ocean, and reduced chemistry converts it to light H2. The resulting drop in mean molecular weight inflates the atmosphere and can lower the measured bulk density by up to ~60%, hundreds of millions to billions of years after formation, before the envelope is finally lost. Oxidized, Earth-like mantles instead buffer a heavy CO2/SO2 atmosphere and deflate monotonically. If correct, reflation turns the time evolution of transit radius and density into a probe of a planet's interior geochemistry and volatile budget, with population-level signatures near the radius valley.","feed_headline":"Late re-inflation can cut super-Earth densities by 60%","feed_subtitle":"Coupled simulations tie late puffy episodes to deep-mantle redox, a population-level probe of exoplanet interiors.","key_machinery":"The central mechanism is the redox-dependent competition between volatile solubility in the magma ocean and gas-phase speciation, coupled to energy-limited hydrodynamic escape. The 'iron-wüstite buffer' is a reference oxygen fugacity that separates reducing from oxidizing mantle conditions. At reducing conditions near that buffer, carbon is outgassed as heavy CO (mean molecular weight ~28 g/mol) while hydrogen stays dissolved in the melt as H2O; when hydrodynamic escape driven by stellar XUV radiation removes the CO-rich envelope, the falling surface pressure shifts the solubility equilibrium, releasing the dissolved water, which reduced gas-phase chemistry converts to light H2 (~2 g/mol). T","core_discovery":"On the paper's own terms, the central discovery is that the redox-sensitivity of mantle outgassing—not just the starting envelope—controls whether irradiated super-Earth atmospheres shrink monotonically or experience a transient re-inflation. In simulations with reduced mantles (oxygen fugacity near the iron-wüstite buffer), the secondary atmosphere begins CO-dominated while hydrogen remains dissolved as H2O in the magma ocean; energy-limited hydrodynamic escape strips the CO, and the resultant drop in surface pressure shifts solubility equilibria so that stored water degasses and equilibrates to H2. The atmosphere's mean molecular weight falls from ~28 to ~2 g/mol, the scale height grows, a","pith_inferences":["If reflation operates, the radius valley may not be a sharp monotonic step but could contain a sub-population of reflated planets transiting through it, so demographic fits should allow for non-monotonic density tracks.","The paper fixes mantle oxygen fugacity through each run; if hydrogen escape progressively oxidizes the mantle, reflation could be suppressed or postponed. Searching for reflating super-Earths around old stars therefore indirectly constrains the rate of redox evolution via hydrogen loss.","The same solubility–escape feedback could extend to sub-Neptunes that have lost their primordial hydrogen envelopes but retain molten interiors; such planets might reflate later and at larger radii, though stronger gravity would damp the effect.","A testable extension: in multi-planet systems with close-in super-Earths, reflating and non-reflating siblings around the same star should differ systematically in density and atmospheric composition, isolating interior redox from stellar irradiation effects."],"forward_implications":["Super-Earth bulk densities can evolve non-monotonically over Gyr timescales, so a single measured density may correspond to very different interior states depending on epoch.","A population of low-density 'puffy' super-Earths at ages of hundreds of Myr to Gyr, unexplained by standard photoevaporation models, would point to reduced-mantle reflation rather than primordial H/He envelopes.","Reflating planets should show a spectroscopic signature: a transition from CO-dominated to H2/H2O-enriched atmospheric composition, ending in sulfur-rich gas before final erosion.","Planets with oxidized mantles should instead show stable heavy CO2/SO2 atmospheres and monotonically increasing density, offering a clean dichotomy for observations to test.","Correlating atmospheric composition, bulk density, irradiation, and stellar age across a large survey could reveal population-level reflation signatures and thereby constrain deep mantle redox states."],"fun_headline_variants":["Redox-driven reflation can slash super-Earth density by 60%","Mantle redox flips super-Earth atmospheres to puffy phase","Super-Earth density plummet linked to mantle redox reflation","Super-Earth reflation reveals deep mantle redox state"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that each simulation holds the mantle's oxygen fugacity fixed; reflation requires the mantle to still be reduced (within about 2 log units of the iron-wüstite buffer) at the moment the initial CO-dominated atmosphere is stripped, whereas in reality hydrogen escape could oxidize the mantle before that transition, shifting or suppressing the reflation.","fun_headline_variants_meta":{"raw":{"variants":["Redox-driven reflation can slash super-Earth density by 60%","Mantle redox flips super-Earth atmospheres to puffy phase","Super-Earth density plummet linked to mantle redox reflation","Super-Earth reflation reveals deep mantle redox state"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001542,"raw_usage":{"total_tokens":6049,"prompt_tokens":835,"completion_tokens":5214,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":579,"completion_tokens_details":{"reasoning_tokens":5152}},"tokens_in":579,"tokens_out":5214,"duration_ms":34542,"temperature":1.0,"reasoning_tokens":5152,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T03:41:17.177957+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete test: use a self-consistent model that couples mantle oxidation to hydrogen escape. If the mantle oxygen fugacity rises above +2 relative to the iron-wüstite buffer before the CO envelope is stripped enough for H2O to degas as H2, reflation disappears. Observationally, a large unbiased survey of old (1–5 Gyr) close-in super-Earths should reveal a population of low-density planets (bulk density roughly 40–60% of a bare-rock model) with H2/H2O-dominated secondary atmospheres; the absence of such objects—or a candidate caught in the CO-dominated pre-transition state that never drops in","supporting_citations":[],"review_version":1}