{"id":"a5a068d2-241b-4585-a31c-963f16077bf0","arxiv_id":"2505.06604","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"New Ge isotope data for carbonaceous and enstatite chondrites give a bulk silicate Earth carbonaceous-chondrite fraction of 0.64±0.16, larger than the Zn fraction of 0.29±0.07, indicating late delivery of volatile-rich carbonaceous bodies.","lead":"Germanium isotope measurements on 23 chondrites show that Earth's moderately volatile elements record a larger contribution from carbonaceous chondrite-like material than zinc isotopes do, pointing to a late, volatile-rich addition. This supports models in which a few Moon-sized carbonaceous embryos supplied volatile elements to the growing Earth.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The late-accretion claim hinges on unmeasured Ge metal-silicate isotope fractionation during core formation; the Si-based correction of -0.3 to -0.5‰ could reduce the inferred CC fraction from 0.64 to ~0.4, erasing the distinction from Zn.","rationale":"The paper provides a valuable new dataset (first Ge isotope data for enstatite and most carbonaceous chondrites) and a coherent interpretation linking the BSE's Ge isotope composition to late-stage delivery of volatile-rich CC material. The measurements appear careful, the data tables are complete, and the observed systematic variations among chondrites are internally consistent with two-component mixing models. However, the central geodynamic conclusion—that the Ge vs Zn CC fraction difference records late accretion of volatile-rich CC bodies—rests critically on the assumption that core formation did not significantly fractionate Ge isotopes. Section 5.1 explicitly identifies this as an unconstrained parameter: no Ge metal-silicate experiments exist, and the Si-based analogue yields a correction of -0.3 to -0.5‰, which the paper itself labels 'not ideal.' Applying even the lower end of this correction moves the pre-core BSE from 0.60‰ to about 0.30‰, which lowers the calculated CC fraction from 0.64 to about 0.40, statistically indistinguishable from the Zn value of 0.29±0.07. The manuscript's qualitative escape clause—that the CC endmember might be a mix of CI and other CC chondrites—does not remove the need for an experimentally or theoretically determined Ge metal-silicate fractionation factor. Because this single unmeasured parameter determines whether the paper's flagship 2:1-vs-1:2 ratio exists, the headline claim should be regarded as conditional pending direct constraints on Ge isotope behavior during core formation. My read matches the reader's weakest-assumption assessment, and the appropriate verdict remains CONDITIONAL.","tokens_in":25880,"tokens_out":4196,"duration_ms":38968,"concrete_test":"Determine the equilibrium Ge isotope fractionation between liquid Fe metal and silicate melt under core-formation conditions. Concretely: (1) conduct high-pressure experiments (e.g., 1-10 GPa, 2000-3500 K) with a Ge double spike, measuring Δ74Ge_metal-silicate, or (2) compute reduced partition function ratios for Ge in Fe-metal and silicate liquids using first-principles methods (DFT/MD). If the result is |Δ74Ge| < 0.1‰, the core-formation correction is negligible and the 0.64 CC fraction stands. If it is -0.3 to -0.5‰ as the Si analogue suggests, recompute f_CC using the corrected pre-core BSE composition; if f_CC drops to ≤0.4, the Ge-Zn difference is no longer resolvable and the late-stage volatile-rich CC interpretation is unsupported.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The inference that the BSE's Ge isotope composition records a CC fraction of 0.64±0.16 that is larger than Zn's 0.29±0.07—the basis for the late-stage volatile-rich CC accretion claim—assumes that core formation did not fractionate Ge isotopes. Section 5.1 acknowledges that no Ge metal-silicate partitioning experiments exist and that the Si analogue is 'not ideal' (Si lithophile vs Ge siderophile), yet applies a Δ30Si metal-silicate ≈ -0.4 to -0.6‰ at 3500 K, mass-scales it to Δ74Ge ≈ -0.3 to -0.5‰, and then proceeds with the measured BSE value of 0.60±0.02. If a correction of -0.3‰ applies, the pre-core BSE δ74/70Ge is ~0.30‰, and solving equation (1) with EC (-0.17‰) and CI (1.00‰) endmembers yields f_CC ≈ 0.40; at -0.5‰ it is ~0.23. Both values overlap the Zn-based CC fraction of 0.29±0.07, eliminating the 2:1 vs 1:2 distinction that is the paper's central evidence for late volatile-rich CC delivery. The paper notes this possibility but does not quantify its effect on the headline claim, and the subsequent argument that a lower CC fraction could be offset by invoking additional CC sources is not a substitute for knowing the actual fractionation factor.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents the first mass-dependent Ge isotope data (δ74/70Ge) for carbonaceous and enstatite chondrites, measured by double-spike MC-ICP-MS with full analytical validation. It finds that Ge isotope and concentration variations among carbonaceous chondrites correlate with matrix mass fraction and with other moderately volatile elements, consistent with two-component mixing of a heavy, CI-like matrix and a light, non-matrix (chondrule-rich) component. Enstatite and ordinary chondrites define a narrow, relatively light range. Using a BSE value of δ74/70Ge = 0.60 ± 0.02 and Eq. (1) with enstatite and CI endmembers, the authors infer a CC-derived Ge fraction of 0.64 ± 0.16, larger than the Zn-based value of 0.29 ± 0.07. They interpret this difference as evidence for late-stage accretion of volatile-rich CC bodies to Earth, consistent with Moon-sized embryo delivery models.","tokens_in":26167,"tokens_out":6138,"duration_ms":54663,"significance":"The analytical contribution is strong: the double-spike method, full data tables, reproducibility on standards, and agreement with previous terrestrial reference materials give confidence in the new chondrite data. If the geochemical interpretation holds, the paper offers a new siderophile, moderately volatile tracer for reconstructing Earth's accretion, complementing nucleosynthetic isotope tracers. The paper also provides a coherent, falsifiable framework linking Ge isotope systematics in chondrites to mixing, and it presents a quantitative prediction for the CC fraction of Ge versus Zn that can be tested by future experiments. The central limitation is that the headline interpretation depends on an unmeasured core-formation isotope effect, and the paper's own uncertainty analysis admits a correction large enough to erase the key Ge-Zn distinction.","major_comments":[{"comment":"The central claim of a ~2:1 CC:NC ratio for Ge, distinct from Zn, uses the measured BSE δ74/70Ge of 0.60 ± 0.02 without applying the core-formation correction estimated in Section 5.1. Section 5.1 states that the pre-core BSE may have been up to ~0.3–0.5‰ lighter than today's value. Substituting δ74/70Ge = 0.30‰ and 0.10‰ into Eq. (1), with EC = –0.17‰ and CI = 1.00‰, yields f_CC ≈ 0.40 and 0.23, respectively. Both values overlap with the Zn-based f_CC of 0.29 ± 0.07, eliminating the 2:1 versus 1:2 distinction that is the basis for the late-stage volatile-rich CC delivery claim. The paper's qualitative statement that a lower CC fraction could be offset by invoking additional CC sources with lower δ74/70Ge is not quantified and is not a substitute for constraining the actual Ge metal–silicate fractionation factor. Please provide a quantitative sensitivity analysis of f_CC to the core-formation correction and condition the conclusions on this uncertainty.","section":"Section 5.1 and 5.2 (Eq. 1, Fig. 7)"},{"comment":"The use of Si as an analogue for Ge is acknowledged as 'not ideal,' yet the derived correction range of –0.3 to –0.5‰ is subsequently used as if it were a plausible bound on the true effect. Because the entire late-accretion interpretation hinges on this correction being small, the authors should either (a) obtain or cite direct experimental metal–silicate Ge isotope fractionation data, (b) provide a theoretical estimate based on Ge bonding and coordination, or (c) explicitly state that the late-accretion conclusion remains unevaluated until such data become available. As written, the conclusions overstate the certainty of the preferred model.","section":"Section 5.1"},{"comment":"The x95 value for Ge (~0.4) is model-dependent, being calculated assuming single-stage core formation and a fixed equilibration factor k = 0.2. The degree of agreement between the measured Ge CC fraction and the Zn-fitted model curve depends on this assumed value. The paper should show how f_CC(Ge) and the comparison to Zn change for plausible variations in k and D, rather than adopting a single set of parameters. Without such a sensitivity test, the consistency shown in Fig. 7 is less compelling than implied.","section":"Section 5.2, Fig. 7"}],"minor_comments":[{"comment":"The reported non-matrix component δ74/70Ge value of –2.62!\"... contains garbled formatting in the manuscript; please render the value and its error properly (e.g., –2.62 ± 0.24/–0.28‰) so that the regression output is clear.","section":"Section 4.1"},{"comment":"The two basalts analyzed in this study (BHVO-2: 0.53 ± 0.03; BCR-2: 0.58 ± 0.07) give a mean δ74/70Ge of ~0.54 ± 0.03, which is slightly lower than the adopted BSE value of 0.60 ± 0.02. Please clarify whether this offset is analytically significant and justify the use of the literature BSE value in Eq. (1), or discuss the effect of using 0.54 instead of 0.60 on the derived CC fraction.","section":"Table 1 and Section 5.2"},{"comment":"The supplementary argument that the BSE's Mo is mixed NC-CC relies substantially on a single IAB iron meteorite analysis (Campo del Cielo) and a proposed non-exponential mass fractionation artifact. This argument is presented as a definitive demonstration, but it is contentious and based on limited data; consider framing it more cautiously as an assessment of the existing literature.","section":"Supplementary Information, Mo discussion"}],"recommendation":"major_revision","confidential_remarks":"The paper is analytically strong and the chondrite dataset is valuable, but the headline accretionary conclusion currently rests on an unquantified core-formation correction that the authors themselves estimate could erase the Ge-Zn distinction. The requested sensitivity analysis is feasible within the paper's scope and would make the central claim defensible. I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Hi —\n\nBottom line: the analytical work is genuinely good, and the dataset is a real contribution. The first Ge isotope measurements for enstatite and non-CB carbonaceous chondrites are careful and reproducible, with full tables and a clean double-spike setup. The two-component mixing interpretation follows the established pattern for Zn, Te, and other moderately volatile elements, and the correlation with matrix mass fraction is convincing. Notably, Ge shows a much larger per-amu fractionation between matrix and non-matrix components than other MVEs (~0.9‰/amu), which is interesting and deserves attention. The enstatite chondrite data are also valuable: they show Ge is fairly homogeneous across petrologic types, consistent with Ge residing in metal rather than sulfides.\n\nSo what's the catch? The central claim — that the BSE's Ge isotope composition records a ~2:1 CC-to-NC mixture, distinct from Zn's ~1:2, and that this signals late volatile-rich CC delivery — depends on the assumption that core formation did not fractionate Ge isotopes. Section 5.1 is admirably frank about the absence of Ge metal-silicate partitioning experiments, but then applies a mass-scaled Si analogue of -0.3 to -0.5‰. If even the smaller correction applies, the BSE's pre-core δ74/70Ge drops to ~0.3‰, and the inferred CC fraction falls to ~0.4, right on top of the Zn value. The paper mentions this possibility but does not propagate it into the headline numbers; the abstract and conclusions still advertise a ~2:1 ratio. That is the load-bearing weak point.\n\nThe endmember choice (EC and CI) is defensible and the scenario discussion is thoughtful. There is some circularity in comparing the Ge-derived fraction to a model curve fitted to Zn, but the Ge value itself is measured, not fitted, so that's a moderate concern at most. The supplementary argument defending the BSE's mixed Mo heritage is clever but leans heavily on reinterpreting a single Campo del Cielo analysis, which is a bit thin.\n\nWho is this for? Cosmochemists and accretion modelers will want the data; the interpretation is a good discussion piece but not a settled result. I'd send it to peer review — the dataset deserves scrutiny and the authors have earned the chance to fix the core-formation propagation. My recommendation: accept with a revision that quantifies how the -0.3 to -0.5‰ uncertainty affects the CC fraction and the Ge–Zn comparison. If they present the corrected range openly, the paper would be much stronger.","headline":"Solid new Ge isotope data for chondrites, but the late-accretion story rests on an unmeasured core-formation fractionation that could erase the Ge–Zn distinction.","tokens_in":26770,"tokens_out":2414,"would_cite":true,"duration_ms":25734,"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 germanium isotopes in Earth's mantle record a late influx of volatile-rich carbonaceous material delivered by Moon-sized embryos.","keywords":["germanium isotopes","moderately volatile elements","bulk silicate Earth","late accretion","carbonaceous chondrites","enstatite chondrites","core formation","planetary accretion"],"falsifier":"Measure the equilibrium germanium isotope fractionation between liquid metal and liquid silicate at about 3500 K, the temperature inferred from Ge partitioning during core formation. If $\\Delta^{74}\\mathrm{Ge}_{\\mathrm{metal-silicate}}$ is near zero, the paper's preferred carbonaceous fraction of $0.64\\pm0.16$ stands; if it approaches $-0.5$‰, the pre-core BSE would be too light to require a carbonaceous fraction much larger than zinc's, and the claimed late volatile-rich influx would not be resolvable. A second, independent check would be finding any non-carbonaceous chondrite with $\\delta^{74/70}\\mathrm{Ge} > 0.60$, which would remove the need for a carbonaceous endmember altogether.","tokens_in":25597,"feed_emoji":"🌍","tokens_out":14523,"duration_ms":113634,"temperature":0.7,"pith_summary":"The paper uses new mass-dependent germanium isotope measurements ($\\delta^{74/70}\\mathrm{Ge}$) of carbonaceous and enstatite chondrites to identify the source of Earth's moderately volatile elements (elements condensing between roughly 1250 and 650 K). It finds that the germanium isotope composition of the bulk silicate Earth (BSE), $0.60\\pm0.02$‰, is indistinguishable from a roughly 2:1 mixture of CI-chondrite-like and enstatite-chondrite-like germanium, meaning about 64% of the BSE's Ge came from carbonaceous (outer Solar System) material. Because germanium partitions strongly into metal, early-delivered Ge would have been removed to the core, so this signature records only the later stages of accretion. The carbonaceous fraction recorded by Ge is more than twice that recorded by zinc, which the authors interpret as the late-stage delivery of volatile-rich carbonaceous bodies to Earth, probably a few Moon-sized embryos. The result matters because it connects the volatile inventory of Earth to the dynamical timing of accretion.","feed_headline":"Ge isotopes trace a late carbonaceous influx to Earth","feed_subtitle":"The siderophile clock says volatile-rich carbonaceous bodies arrived late, likely as Moon-sized embryos.","key_machinery":"The central tracer is the mass-dependent germanium isotope ratio $\\delta^{74/70}\\mathrm{Ge}$, reported in permil relative to a reference standard and measured with a $^{70}\\mathrm{Ge}$–$^{73}\\mathrm{Ge}$ double spike. The argument is carried by two mixing relations: within carbonaceous chondrites, $\\delta^{74/70}\\mathrm{Ge}$ and Ge concentration correlate with matrix mass fraction and define a volatile-rich, isotopically heavy CI-like matrix endmember and a volatile-poor, isotopically light chondrule/non-matrix endmember; and at the scale of the whole Earth, the BSE value is treated as a mixture of enstatite-chondrite (non-carbonaceous, NC) and CI-chondrite (carbonaceous, CC) endmembers using the same lever-rule formula previously applied to nucleosynthetic isotope anomalies. The timing information comes from the preference of Ge for metal, quantified by the parameter $x_{95}$ (the fraction of Earth's mass after which the last 95% of an element was added to the mantle), which is about 0.4 for Ge, so the Ge isotope signature is windowed to late accretion. The paper's potential core-formation correction, estimated by scaling experimentally determined silicon metal-silicate isotope fractionation to germanium, is the main auxiliary input that sets the systematic uncertainty.","core_discovery":"In its preferred model, the paper claims that the germanium isotope composition of the bulk silicate Earth (BSE) is a two-endmember mixture, with the non-carbonaceous endmember having the $\\delta^{74/70}\\mathrm{Ge}$ of enstatite chondrites ($-0.17\\pm0.42$‰) and the carbonaceous endmember having the composition of CI chondrites, the most primitive volatile-rich meteorites ($1.00\\pm0.04$‰). Applying a simple mass balance to the measured BSE value of $0.60\\pm0.02$‰ yields a carbonaceous fraction of $0.64\\pm0.16$ for Ge, compared with $0.29\\pm0.07$ for Zn from nucleosynthetic isotope anomalies. The paper argues this offset is exactly what is expected if volatile-rich carbonaceous bodies were added late in Earth's accretion: siderophile Ge delivered early was stripped into the core, so the Ge in the present-day mantle is dominated by late additions, whereas zinc, which prefers silicates, preserves the whole accretion history. It further argues that the Ge isotope systematics among carbonaceous chondrites are the product of mixing between volatile-rich, isotopically heavy matrix and volatile-poor, isotopically light chondrules, with Ge showing the largest per-amu isotope fractionation of any moderately volatile element yet measured. The paper concludes that the Moon-sized embryos invoked by dynamical models to deliver carbonaceous material to Earth must themselves have been volatile-rich, either because they were undifferentiated, too large to degas, or built from previously volatile-rich objects.","pith_inferences":["If future metal-silicate experiments show that Ge isotopes fractionate as strongly as the silicon analogue suggests ($-0.3$ to $-0.5$‰), the inferred carbonaceous fraction for Ge would fall to roughly 0.4 or below, bringing Ge into agreement with Zn and undercutting the late-addition claim; the paper's scenario is therefore directly testable by experiment.","The same $\\delta^{74/70}\\mathrm{Ge}$ tracer could be applied to Mars and Vesta to ask whether their siderophile volatile inventories also require volatile-rich late embryos; the main obstacle would be establishing the core-formation correction for each body.","Because the Ge–Te isotope correlation in chondrites is tight, measuring mass-dependent Te isotopes in additional BSE samples with better precision could decide whether the late veneer was carbonaceous or non-carbonaceous, a question the current data leave open.","The model predicts a stochastic relationship between planet mass and Ge isotope composition among terrestrial planets; a survey of Martian meteorites and lunar impact-melt rocks could test the Moon-sized-embryo delivery mechanism."],"forward_implications":["If the preferred model is right, Earth's late-stage accretion involved volatile-rich carbonaceous bodies, and those bodies were Moon-sized or larger, since smaller planetesimals would have been accreted earlier and would have made Mars too carbonaceous-rich.","The carbonaceous fraction recorded by Ge (about 0.64) should be higher than the carbonaceous fraction recorded by lithophile volatile elements such as Zn (about 0.29), providing a testable fingerprint of late addition rather than early mixing.","The BSE's position on the chondrite $\\delta^{74/70}\\mathrm{Ge}$–$\\delta^{128/126}\\mathrm{Te}$ mixing line implies a similar carbonaceous fraction for Te (about 0.57), though the large uncertainty currently prevents using Te isotopes to constrain the late veneer.","The Ge isotope spread among carbonaceous chondrites places germanium on the same matrix-chondrule mixing trend as Zn, Te, Rb, and Cd, so the same volatility-driven fractionation process that shapes other moderately volatile elements also shaped Ge, with Ge showing the largest per-amu effect."],"supporting_citations":[{"why":"Supplies the dynamical model in which carbonaceous material is delivered late through Moon-sized embryos and the x95 framework connecting an element's siderophile character to the recorded CC fraction.","marker":"Nimmo et al., 2024"},{"why":"Establishes the x95 concept that lets siderophile elements record only the later stages of accretion, the backbone of the timing argument.","marker":"Dauphas, 2017"},{"why":"Provides the ordinary-chondrite Ge isotope data used as non-carbonaceous baselines and shows Ge isotopes are not reset by parent-body metamorphism.","marker":"Florin et al., 2020"},{"why":"Supplies matrix mass fractions, Te isotope data, and the two-component mixing approach used to interpret carbonaceous chondrite Ge variations.","marker":"Hellmann et al., 2020"},{"why":"Reports the nucleosynthetic Zn isotope anomalies from which the roughly 1:2 non-carbonaceous to carbonaceous mixing ratio for Zn is taken.","marker":"Steller et al., 2022"},{"why":"Provides the metal-silicate partition coefficients for Ge used to estimate the core-formation temperature and the x95 value for Ge.","marker":"Righter et al., 2011"},{"why":"Provides the terrestrial igneous-rock Ge isotope compilation that defines the BSE value of 0.60±0.02‰.","marker":"Rouxel and Luais, 2017"},{"why":"Provides the experimentally determined Si metal-silicate isotope fractionation used as the analogue for the potential Ge core-formation correction.","marker":"Hin et al., 2014"},{"why":"Supplies the Mo isotopic evidence for late carbonaceous addition that supports the mixed non-carbonaceous/carbonaceous heritage assumed for Earth's late accretion.","marker":"Budde et al., 2019"}],"fun_headline_variants":["Ge isotope clock dates late carbonaceous Earth influx","Siderophile Ge shows volatile-rich embryos added late","Ge isotopes reveal Moon-sized volatile-rich impactors","Earth's Ge from late, volatile-rich carbonaceous mix","Ge isotope mix pins volatile-rich late accretion to Earth"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that germanium isotopes were not substantially fractionated when Earth's core formed, because no metal-silicate Ge isotope experiments exist; the authors use silicon as an analogue and acknowledge that the comparison is 'not ideal,' with a possible correction of $-0.3$ to $-0.5$‰ that would lower the inferred carbonaceous fraction from 0.64 to about 0.4 or less.","fun_headline_variants_meta":{"raw":{"variants":["Ge isotope clock dates late carbonaceous Earth influx","Siderophile Ge shows volatile-rich embryos added late","Ge isotopes reveal Moon-sized volatile-rich impactors","Earth's Ge from late, volatile-rich carbonaceous mix","Ge isotope mix pins volatile-rich late accretion to Earth"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000711,"raw_usage":{"total_tokens":3271,"prompt_tokens":1090,"completion_tokens":2181,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":706,"completion_tokens_details":{"reasoning_tokens":2106}},"tokens_in":706,"tokens_out":2181,"duration_ms":15900,"temperature":1.0,"reasoning_tokens":2106,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T22:38:13.411747+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the equilibrium germanium isotope fractionation between liquid metal and liquid silicate at about 3500 K, the temperature inferred from Ge partitioning during core formation. If $\\Delta^{74}\\mathrm{Ge}_{\\mathrm{metal-silicate}}$ is near zero, the paper's preferred carbonaceous fraction of $0.64\\pm0.16$ stands; if it approaches $-0.5$‰, the pre-core BSE would be too light to require a carbonaceous fraction much larger than zinc's, and the claimed late volatile-rich influx would not be resolvable. A second, independent check would be finding any non-carbonaceous chondrite with $\\delta^{74/70}\\mathrm{Ge} > 0.60$, which would remove the need for a carbonaceous endmember altogether.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the experimentally determined Si metal-silicate isotope fractionation used as the analogue for the potential Ge core-formation correction."}],"review_version":1}