{"id":"03be9e8a-039a-4ccf-9058-ca987863db00","arxiv_id":"2412.08136","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":14,"one_line_summary":"Isolated black holes accreting inside molecular clouds can accelerate protons to PeV energies, producing dark TeV-PeV gamma-ray sources and contributing to the cosmic-ray flux seen at Earth.","lead":"The paper proposes that isolated stellar-mass black holes drifting through dense molecular clouds can accelerate protons to PeV energies, producing very-high-energy gamma-ray sources that have no lower-energy counterpart. If correct, these wandering black holes would be a new class of cosmic ray accelerators and could explain the 'dark' sources recently detected by LHAASO.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'no GeV-TeV gamma-rays' claim is unquantified: pp interactions of escaping sub-PeV protons inevitably produce lower-energy photons, and the model is not compared to J0007+5659u upper limits.","rationale":"The reader's weakest-assumption (IBH velocity distribution) is a valid population-level concern, and I agree the paper's conclusions are conditional. However, I identify a more direct and immediately testable issue: the paper's claim that dark sources emit no GeV-TeV gamma-rays is not demonstrated quantitatively. Protons that escape above E_p,esc will produce secondary gamma-rays over a wide energy range via pp collisions, so the SED is not automatically dark below ~10 TeV. The manuscript shows the predicted spectrum only down to 100 GeV and does not compare it with the available upper limits for the fitted source. This matters because the distinguishing feature of LHAASO dark sources is the absence of GeV-TeV counterparts; if the model predicts detectable emission in that band, the central claim fails. My concrete test settles this without changing the overall conditional verdict, since the test could go either way and other uncertainties (velocity, source count) remain.","tokens_in":16426,"tokens_out":24108,"duration_ms":232834,"concrete_test":"Recompute the J0007+5659u gamma-ray SED from 100 MeV to 30 TeV using the exact parameters in Table 1 (M=20 Msun, n_MC=1000 cm^-3, V_k=20 km/s, R_MC=5 pc, B_MC=30 microG, d=2 kpc) and the Kelner et al. (2006) pp code with the escape spectrum from Eq. (5). Overlay the Fermi-LAT 4FGL upper limits and LHAASO-WCDA sensitivity for J0007+5659u; if the model exceeds these in the 1–100 GeV band, the dark-source explanation is falsified. If it lies below the limits, the 'no GeV-TeV counterpart' claim is confirmed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 4 states that since only sub-PeV protons escape the MAD, 'we do not expect GeV-TeV gamma-rays from the molecular cloud.' This is an overstatement: hadronic pp collisions of protons with E_p > E_p,esc produce π0-decay photons over a broad energy range extending down to ~0.1–1 GeV, not only at ~0.1 E_p. The numerical SED in Figure 3 is shown only for E > 100 GeV, and no comparison is made with Fermi-LAT or LHAASO-WCDA upper limits for J0007+5659u in the 0.1–2000 GeV band. If the predicted E^2 flux at GeV–TeV energies is above those upper limits, the source would not be 'dark' and the central explanation of LHAASO dark sources fails. This is a direct internal test of the model, independent of population-level uncertainties such as the IBH velocity distribution.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This Letter proposes that isolated stellar-mass black holes (IBHs) accreting from dense molecular gas enter a magnetically arrested disk (MAD) state in which magnetic reconnection accelerates protons to PeV energies. Protons above a characteristic escape energy E_p,esc ≈ 41 TeV (Eq. 6) leave the MAD and diffuse into the surrounding cloud, where pp collisions produce TeV–PeV gamma rays and neutrinos, while lower-energy protons are swallowed by the black hole. The model is applied to the LHAASO 'dark' source J0007+5659u and used with Galactic distributions of IBHs and molecular clouds to estimate the PeV cosmic-ray intensity at Earth, finding a 30% contribution at the knee in a reference model.","tokens_in":16705,"tokens_out":11586,"duration_ms":110126,"significance":"If correct, the scenario would identify a new class of Galactic PeVatrons and provide a natural explanation for the newly discovered LHAASO dark sources, with concrete predictions for southern-sky gamma-ray detectors and neutrino telescopes. Strengths include the internally consistent transport calculation, the anchoring of the acceleration rate and spectral index to PIC simulations (Zhang et al. 2023), the use of GRMHD-motivated MAD plasma parameters, and the explicit comparison of the model SED with LHAASO data in the >100 TeV band. However, the population-level predictions depend on several poorly constrained parameters (natal kick distribution, accretion reduction factor, total IBH number), and the dark-source match in the lower panel of Figure 3 uses parameters chosen after the fact, so the central claims require further validation and a quantitative GeV–TeV upper-limit comparison.","major_comments":[{"comment":"The claim in Section 4 that 'we do not expect GeV-TeV gamma-rays from the molecular cloud' is not supported by a quantitative comparison with the observed non-detection in the 0.1–2000 GeV band. Since the escaping proton spectrum extends continuously from E_p,esc ≈ 41 TeV (Eq. 6) to PeV energies, hadronic pp collisions produce a broad photon spectrum that reaches well below 0.1 E_p; the numerical SEDs in Figure 3 are only displayed for E > 100 GeV. The paper should show the full 0.1 GeV to 1 PeV SED for the J0007 model and overlay the Fermi-LAT and LHAASO-WCDA upper limits for J0007+5659u. If the predicted E^2 flux in the GeV–TeV band exceeds those limits, the source would not be dark and the central explanation would be inconsistent with the definition of a dark source; if it does not, the paper should state this explicitly as a test.","section":"Section 4, Figure 3"},{"comment":"The 'J0007' model in Table 1 is evidently constructed to reproduce the observed spectrum of J0007+5659u, making the bottom panel of Figure 3 a post-hoc fit rather than an a priori prediction. This circularity weakens the statement in the abstract that the scenario 'can explain' dark sources. I request that the authors either (i) treat the LHAASO measurement as an upper limit and propagate the observational uncertainties into the model parameters, or (ii) perform a forward calculation using the Galactic IBH and molecular-cloud distributions to predict the number of detectable dark sources and compare with the observed population (two sources satisfying the selection). Without one of these steps, the match to J0007+5659u is an existence proof, not an explanation.","section":"Table 1 and Section 4"},{"comment":"The estimate that IBHs contribute ≈30% of the PeV cosmic-ray flux at the knee hinges on the assumed weak natal-kick distribution (σ_v ≈ 20 km/s). Because the accretion luminosity scales as V_eff^{-3} (Eq. 1), a population containing a significant fraction of black holes with V_k > 100 km/s would be suppressed by orders of magnitude, and such high-kick BHs are inferred from X-ray binary studies and from the Galactic BH mass distribution. The paper explores only two velocity dispersions and two values of λ_w in Figure 4. I request a population calculation that adopts a realistic kick-velocity distribution (e.g., a Maxwellian with a high-velocity tail or the distribution inferred from BH X-ray binary kinematics) and reports the resulting CR intensity and its uncertainty. This is necessary to substantiate the abstract's claim of a 'significant contribution' to PeV CRs.","section":"Section 5, Eq. (1), Figure 4"},{"comment":"The maximum proton energy E_p,cut depends on the reconnection-acceleration coefficient η_rec and the diffusion enhancement factor η_diff, which are assigned fiducial values (η_rec = η_diff = 10) based on PIC simulations for η_rec and a Bohm-diffusion assumption for η_diff. These quantities are not directly measured for IBH MADs, and E_p,cut scales as η_rec^{1/2} η_diff^{1/2}; a factor of ~3 uncertainty in either parameter changes the peak gamma-ray energy by a similar factor, potentially moving the predicted >100 TeV signal below the LHAASO threshold. The paper should provide a sensitivity study over the plausible ranges of η_rec and η_diff and state the fraction of the IBH population that would still reach PeV energies under conservative choices.","section":"Section 3, Eq. (7)"}],"minor_comments":[{"comment":"The caption contains a typo: 'J0007+05659u' should be 'J0007+5659u'.","section":"Figure 2 caption"},{"comment":"There is a typo in 'hihg-sensitivity radio surveys'; it should read 'high-sensitivity'.","section":"Section 6"},{"comment":"Please verify the numerical coefficient in the expression for E_coh = e B_MC λ_coh. With λ_coh = R_MC/5, R_MC = 20 pc, and B_MC = 10 μG, the product e B λ gives approximately 0.036 EeV in Gaussian units, not 0.18 EeV; the discrepancy may indicate a different convention or a typo that could affect the diffusion timescale in Eq. (9).","section":"Equation (8)"},{"comment":"The assumption that dN/dM, dN/dV_k, and dξ/dn_MC are independent of Galactocentric radius is a strong simplification, and the paper should state whether this assumption biases the predicted CR intensity high or low.","section":"Section 5"}],"recommendation":"major_revision","confidential_remarks":"The paper is a plausible candidate for ApJL if the requested GeV–TeV upper-limit comparison and a forward source-count estimate are added. The authors should be encouraged to avoid overstating the 'no GeV–TeV gamma-rays' claim in the abstract and Section 4 until the full SED is checked. The fitting of J0007 parameters is a presentational concern; a forward source-count estimate would strengthen the letter considerably."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know before you read it. First, the core idea is genuinely new and worth taking seriously: a wandering isolated black hole in a molecular cloud, accreting in a magnetically arrested state, can accelerate protons to ~PeV via reconnection, and the escape-energy threshold (~40 TeV) naturally produces a hard TeV-PeV gamma-ray spectrum with a suppressed GeV-TeV component. That is exactly the shape that could explain the LHAASO dark sources. Second, the paper is honest enough to show that with typical parameters the model is far too faint to be detected; the match to J0007+5659u comes from an optimistic, hand-picked parameter set. So this is a plausible scenario, not an established explanation.\n\nThe novelty relative to Ioka et al. (2017) and Barkov et al. (2012) is real. Those papers did not put the IBH in a molecular cloud, did not derive the escape-threshold spectral break, and did not connect to dark sources. The transport and emission calculations are internally consistent, the microphysics is anchored to recent PIC and GRMHD work, and the paper gives concrete, falsifiable predictions: hard X-ray and infrared counterparts, neutrinos, and more sources in southern-sky surveys. The population synthesis for the PeV CR contribution is a reasonable first estimate, and the claim that IBHs could provide ~30% of the knee flux is striking even with uncertain inputs.\n\nThe soft spots are quantitative, not logical. The J0007 model is a re-fit: the Table 1 parameters are chosen after the fact to reproduce that source, so the bottom panel of Figure 3 demonstrates consistency rather than prediction. The escape-threshold argument also has an unaddressed hole. Even if only sub-PeV protons leave the MAD, pp collisions of those protons necessarily produce gamma-rays at GeV and lower energies, not just near 0.1 E_p. The paper's statement that \"we do not expect GeV-TeV gamma-rays from the molecular cloud\" is an overstatement. The SED in Figure 3 starts at 100 GeV, and there is no comparison to Fermi-LAT or LHAASO-WCDA upper limits for J0007+5659u in the 0.1-2000 GeV band. That is a direct, model-internal test, independent of population uncertainties, and it should be done before the dark-source explanation is accepted. The Galactic CR contribution also depends on several uncertain population parameters (N_IBH,tot, sigma_v, lambda_w) with no error budget, and the weak-kick assumption is indirect. These are not fatal, but they mean the paper currently shows plausibility, not proof.\n\nWho is this for? Anyone working on PeVatrons, LHAASO dark sources, or isolated black hole accretion. It deserves a serious referee and would be a useful contribution after moderate revision. My recommendation: send it out, but require the authors to confront the GeV-TeV upper-limit issue and to soften the overclaim.","headline":"A genuinely new IBH-in-molecular-cloud scenario for LHAASO dark sources and PeV CRs, but the dark-source match is hand-picked and the 'no GeV-TeV gamma-rays' claim is untested against upper limits.","tokens_in":17315,"tokens_out":5077,"would_cite":true,"duration_ms":56148,"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":"Stellar-mass isolated black holes in molecular clouds can accelerate protons to PeV energies and power LHAASO dark gamma-ray sources, potentially supplying a large share of the PeV cosmic rays observed on Earth.","keywords":["isolated black holes","PeV cosmic rays","LHAASO dark sources","magnetically arrested disk","magnetic reconnection","molecular clouds","ultrahigh-energy gamma rays","Galactic cosmic rays"],"falsifier":"Measure the space velocities of a sample of isolated stellar-mass black holes via astrometric microlensing follow-up; if a large fraction move faster than roughly 100 km/s relative to ambient molecular gas, the predicted PeV cosmic-ray contribution collapses because the accretion rate scales as $V_{\\rm eff}^{-3}$. A second decisive test is a sensitive molecular-line search toward LHAASO J0007+5659u: if no dense cloud of the required column is found, the proposed source class is ruled out.","tokens_in":2160,"feed_emoji":"🕳️","tokens_out":2778,"duration_ms":58412,"temperature":0.7,"pith_summary":"This paper proposes that isolated stellar-mass black holes (IBHs) drifting through molecular clouds are hidden PeVatrons. The black hole accretes dense gas into a magnetically arrested disk, where magnetic reconnection accelerates protons to PeV energies. Low-energy protons fall into the black hole, but protons above an escape energy of about 41 TeV leave the disk and collide with cloud gas, producing TeV-PeV gamma rays with no GeV-TeV counterpart, exactly the signature of LHAASO's \"dark\" sources. Accounting for the Galactic distributions of IBHs and molecular gas, the authors further claim these systems can make a significant contribution to the PeV cosmic rays reaching Earth, with their reference model giving about 30% of the observed proton intensity near the knee.","feed_headline":"Wandering black holes can make PeV cosmic rays","feed_subtitle":"A magnetically arrested disk lets sub-PeV protons escape and light up LHAASO's dark gamma-ray sources, plus a third of the knee flux.","key_machinery":"The central object is the magnetically arrested disk (MAD) that forms around a slowly moving IBH accreting from a molecular cloud. Magnetic reconnection, parameterized by an acceleration timescale $t_{\\rm acc}\\approx \\eta_{\\rm rec} r_L/c$ with $\\eta_{\\rm rec}=10$, produces a power-law proton spectrum with index 2. The escape energy $E_{p,\\rm esc}$ is set by balancing Bohm-like diffusive escape against infall into the black hole; this energy filter is what removes the GeV-TeV gamma-ray counterpart, because only protons above $E_{p,\\rm esc}$ reach the cloud and produce hadronic gamma rays. The proton cutoff energy $E_{p,\\rm cut}\\sim 0.48$ PeV for typical parameters scales as $M n_{MC}^{1/2} V_k^{-3/2}$, showing that low velocity, high black-hole mass, or high gas density pushes protons to PeV energies.","core_discovery":"The central claim is that an IBH embedded in a molecular cloud forms a magnetically arrested disk (MAD), and magnetic reconnection in that disk accelerates cosmic-ray protons up to PeV energies. Because diffusive escape dominates over cooling, protons above the escape energy $E_{p,\\rm esc}\\sim 41$ TeV (scaling as $M_1 n_{MC,2}^{1/2}\\lambda_{w,-1}^{1/2}V_{\\rm eff,6.3}^{-3/2}$) escape from the MAD and are injected into the surrounding cloud. These sub-PeV protons undergo proton-proton collisions with cloud gas, producing gamma rays in the TeV-PeV band while GeV-TeV gamma rays are absent, because the lower-energy protons never escape the disk. The paper applies this to LHAASO dark sources such as J0007+5659u, and demonstrates with a Galactic population model that IBHs in molecular clouds can provide a significant contribution to the PeV cosmic-ray intensity observed on Earth, with the reference model yielding about 30% of the knee flux.","pith_inferences":["I infer that the escape-energy filter is a generic spectral signature: any source class where acceleration happens in a compact region embedded in a thick target will produce gamma-ray spectra peaking above the escape energy, so future detections of a cutoff in dark-source spectra near tens of TeV would support this mechanism.","I infer that the same IBH+MAD mechanism could naturally explain the PeVatron activity in the Galactic center region, where both molecular gas and black holes are most concentrated, as the authors note in passing.","I infer that the ratio of neutrino to gamma-ray flux from a resolved dark source could distinguish this hadronic scenario from leptonic alternatives, and stacking many dark sources in the southern sky may yield a detection even if individual sources are weak.","I infer that the velocity distribution of isolated black holes is the single most informative observable to test the scenario: microlensing surveys with astrometric follow-up can measure the transverse speeds of IBHs, and if a large fraction exceed roughly 100 km/s, the predicted cosmic-ray and gamma-ray outputs would drop by orders of magnitude."],"forward_implications":["LHAASO dark sources with hard spectra above 100 TeV and no GeV-TeV counterparts can be understood as molecular clouds hosting a single IBH, with only sub-PeV protons leaking out to produce gamma rays.","The summed population of IBHs in molecular clouds can supply a sizable fraction of the PeV cosmic rays observed at Earth, with the reference model giving about 30% of the knee proton intensity, and higher values if the accretion efficiency is larger or the velocity dispersion is lower.","Future gamma-ray detectors in the southern sky (such as ALPACA and SWGO) should find additional dark sources, because molecular clouds and IBHs concentrate toward the inner Galaxy.","The scenario predicts hadronic neutrino emission accompanying the gamma rays, with a flux about an order of magnitude below current IceCube sensitivity but potentially reachable through stacking analyses or next-generation detectors.","Hard X-ray and mid-infrared follow-up of dark sources could reveal the accreting IBH, since soft X-rays and optical light are heavily absorbed by the molecular cloud."],"supporting_citations":[{"why":"Provides the LHAASO catalog containing the dark gamma-ray sources that the scenario aims to explain, including J0007+5659u.","marker":"Cao et al. 2024"},{"why":"Supplies the MAD plasma parameters and multi-wavelength emission model for quiescent X-ray binaries that the paper adopts for IBH accretion flows.","marker":"Kimura et al. 2021b"},{"why":"Establishes that the magnetic flux within the Bondi radius exceeds the MAD saturation flux, supporting MAD formation around IBHs, and previously suggested IBHs as PeVatrons.","marker":"Ioka et al. 2017"},{"why":"Provides the 3D particle-in-cell simulation results for the reconnection acceleration timescale and the power-law spectral index used in the proton transport model.","marker":"Zhang et al. 2023"},{"why":"Supports the reduction factor $\\lambda_w=0.1$ on the Bondi accretion rate based on recent GRMHD simulations.","marker":"Galishnikova et al. 2024"},{"why":"Supplies the Galactic molecular gas distribution and total gas mass used in the population-averaged cosmic-ray intensity estimate.","marker":"Nakanishi & Sofue 2016"},{"why":"Provides the volume filling factor of molecular clouds as a function of Galactic radius and the scale height of IBHs used to count IBHs inside clouds.","marker":"Tsuna et al. 2018"},{"why":"Supports the assumption that the majority of Galactic black holes receive weak natal kicks, justifying the low velocity dispersion $\\sigma_v=20$ km/s.","marker":"Repetto et al. 2017"},{"why":"Reports a microlensing-discovered isolated black hole with a low transverse velocity, providing direct evidence for at least some slow IBHs.","marker":"Sahu et al. 2022"},{"why":"Provides the hadronic interaction formalism used to compute the gamma-ray and neutrino spectra from proton-proton collisions in the molecular cloud.","marker":"Kelner et al. 2006"}],"fun_headline_variants":["Black hole disks accelerate protons to PeV energies","Isolated black holes in clouds explain LHAASO dark sources","Wandering black holes power PeV cosmic rays and dark gamma sources","How black holes in molecular clouds make the knee cosmic rays"],"cache_read_input_tokens":19328,"weakest_assumption_plain":"The scenario's output hinges on most isolated black holes moving slowly relative to molecular gas (velocity dispersion about 20 km/s), because the accretion luminosity falls as the third power of relative speed; a population with a sizable fraction of natal kicks above 100 km/s would suppress the predicted cosmic-ray and gamma-ray fluxes by orders of magnitude.","fun_headline_variants_meta":{"raw":{"variants":["Black hole disks accelerate protons to PeV energies","Isolated black holes in clouds explain LHAASO dark sources","Wandering black holes power PeV cosmic rays and dark gamma sources","How black holes in molecular clouds make the knee cosmic rays"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000643,"raw_usage":{"total_tokens":3002,"prompt_tokens":1033,"completion_tokens":1969,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":649,"completion_tokens_details":{"reasoning_tokens":1909}},"tokens_in":649,"tokens_out":1969,"duration_ms":14018,"temperature":1.0,"reasoning_tokens":1909,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T18:11:11.291164+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the space velocities of a sample of isolated stellar-mass black holes via astrometric microlensing follow-up; if a large fraction move faster than roughly 100 km/s relative to ambient molecular gas, the predicted PeV cosmic-ray contribution collapses because the accretion rate scales as $V_{\\rm eff}^{-3}$. A second decisive test is a sensitive molecular-line search toward LHAASO J0007+5659u: if no dense cloud of the required column is found, the proposed source class is ruled out.","supporting_citations":[],"review_version":1}