{"id":"60942280-4370-44d0-a879-17c5a48475b2","arxiv_id":"2606.25242","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Site-selective NMR/NQR shows low-energy Mn pyrochlore excitations grow below 10 K, identifying intrinsic frustration-driven heavy-fermion formation in YMn2Zn20-based compounds.","lead":"Nuclear magnetic resonance measurements on a manganese pyrochlore compound show that the heavy-electron behavior comes from the manganese lattice itself, not from impurity manganese atoms. The study supports a picture where geometrical frustration, rather than antiferromagnetic quantum criticality, generates the heavy-fermion state.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The '20× smaller 1/T1T ⇒ weaker exchange' step rests on an unmeasured 55Mn hyperfine coupling; the paper's support for comparable Ahf is logically insufficient.","rationale":"The reader's weakest-assumption is exactly the step I find load-bearing. The central claim is two-fold: (1) the heavy-fermion state is intrinsic to the Mn pyrochlore network, and (2) it is frustration-related rather than AFM quantum critical. The site-selective NQR data substantially support (1): the 16d signal is narrow, the 115In site shows a similar temperature dependence, and the low-T saturation of 1/T1T is a genuine Fermi-liquid marker. However, claim (2) depends on interpreting the absolute scale of 1/T1T relative to YMn2 to conclude that exchange is weakened. This interpretation requires Ahf comparability, which is neither measured nor convincingly argued. The paper admits the Knight shift-χ route is blocked by excess Mn. No other independent evidence ties the magnitude of 1/T1T to the strength of magnetic fluctuations across the two compounds. Hence, the most load-bearing concern is the unmeasured hyperfine coupling. I would keep the reader's conditional verdict: the reported data are credible and the intrinsic nature is reasonably established, but the mechanistic conclusion should not be accepted without a direct or calculated Ahf determination.","tokens_in":9100,"tokens_out":5684,"duration_ms":65327,"concrete_test":"Measure 55Mn NMR Knight shift K(T) at high field (7–10 T) in YMn2.11Zn17.53In2.36, and simultaneously use 115In NMR shift as a local reference. Fit the bulk susceptibility to a sum of an excess-Mn localized Curie–Weiss term and an intrinsic temperature-dependent term; then construct K vs χ_intrinsic(T) for the 16d Mn site. The slope gives Ahf. Compare with Ahf(YMn2) from Ref. 29. If Ahf is within ~30% of the YMn2 value, the 20× comparison stands; if it is smaller by a factor ≳2, the paper's conclusion is inconclusive. A DFT calculation of the 55Mn hyperfine field in both structures would provide an independent test.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In §3.2 the authors conclude from the ~20 times smaller 1/T1T that magnetic interactions in YMn2.11Zn17.53In2.36 are 'substantially weakened' relative to YMn2, and this is then used in §3.3 to argue against AFM quantum-critical fluctuations and for frustration-induced excitations. The argument requires that the 55Mn hyperfine coupling Ahf is comparable in the two compounds. The paper explicitly states that a quantitative Knight shift-χ estimate of Ahf is difficult because excess Mn contributes to bulk susceptibility, and the only offered justification is that 1/T1T at the Mn site is larger than at the In site. That comparison is not informative for the cross-compound value: on-site 3d hyperfine couplings routinely dominate transferred couplings at nonmagnetic sites, whatever their absolute size. If Ahf in the pyrochlore compound were, say, a factor 3–5 smaller than in YMn2, the observed twentyfold reduction in 1/T1T would no longer imply suppressed magnetic fluctuations; the intrinsic fluctuation spectrum could be comparable or even enhanced. The low-temperature saturation of 1/T1T remains a legitimate Fermi-liquid signature and speaks against a divergent QC contribution, but the positive attribution of the heavy-fermion mass to frustration-induced magnetic excitations, as opposed to, e.g., disorder, relies on the unverified magnitude comparison.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"55Mn, 115In, and 27Al NMR/NQR measurements are reported for In- and Al-substituted YMn2Zn20, a candidate d-electron heavy-fermion system with a Mn pyrochlore lattice. In YMn2.11Zn17.53In2.36 the spectra show narrow lines consistent with selective In occupation of Zn 16c sites, while YMn2.06Zn12.23Al7.71 shows broad lines attributed to random Al occupation of several Zn sites. Zero-field 55Mn-NQR measurements show that 1/T1T is roughly constant at high temperature, increases below about 10 K, and tends to level off at the lowest temperatures; a similar enhancement is seen at the 115In site and in the Al-substituted sample. The authors argue that the enhancement is intrinsic to the 16d Mn pyrochlore network rather than caused by excess Mn impurity spins, because it is observed at the Mn site itself and does not follow Curie behavior. The absolute 1/T1T is about 20 times smaller than in YMn2, which is interpreted as a weakening of magnetic exchange due to the enlarged Mn-Mn distance. From the saturating, non-divergent 1/T1T and the small absolute value, they conclude that the heavy-fermion state is better described by frustration-induced low-energy magnetic excitations than by AFM quantum-critical fluctuations.","tokens_in":9503,"tokens_out":13829,"duration_ms":133821,"significance":"The observation of a saturating 1/T1T enhancement at zero field at two nuclear sites, with spectral assignments that are internally consistent between NMR fits and NQR frequencies, is a solid experimental result; the site-selective isolation of the 16d Mn response from excess-Mn contributions is a genuine advance for this family. If the quantitative comparison with YMn2 were secured, the paper would provide a rare d-electron example where a heavy-fermion state is attributed to geometrical frustration rather than proximity to an antiferromagnetic quantum-critical point, which is of clear interest to the strongly-correlated-electron community. The main caveat is that the '20 times smaller' comparison and the resulting conclusion about weakened exchange depend on an unmeasured hyperfine-coupling normalization, a limitation the authors acknowledge. The intrinsic character of the low-energy excitations and the Fermi-liquid-like saturation are nevertheless well supported by the present data.","major_comments":[{"comment":"The claim that 1/T1T is about 20 times smaller than in YMn2 implies substantially weakened magnetic interactions assumes comparable 55Mn hyperfine couplings Ahf in the two compounds. The paper explicitly states (§3.2) that a quantitative Knight shift-chi estimate is difficult because excess Mn impurities contribute to the bulk susceptibility. The offered justification, that 1/T1T at the Mn site is larger than at the In site, does not constrain the cross-compound value of Ahf: an on-site 3d hyperfine coupling routinely dominates the transferred coupling at a nonmagnetic site regardless of its absolute magnitude. If Ahf in the pyrochlore compound were a factor 3-5 smaller, the 20x reduction would no longer establish suppression of magnetic fluctuations. This comparison is used in §3.3 and the abstract/conclusion to argue against AFM quantum criticality and for frustration-induced heavy-fer","section":"§3.2, §3.3, abstract/conclusion"},{"comment":"The comparison of structural disorder between In- and Al-substituted samples is not fully controlled. The In-substituted sample has x=2.36 while the Al-substituted sample has x=7.71; the broad 55Mn and 27Al lines in Figs. 2(d)-(e) could reflect the larger substitution level rather than an intrinsic difference in site selectivity. The authors themselves attribute the broadening to 'the relatively large amount of Al substitution' (text following Fig. 2), which is consistent with this reading. Ref. 22 supports distinct site occupancies, but the present NMR/NQR spectra alone do not establish that In is intrinsically more selective than Al. Either restrict this claim to the studied concentrations or examine an Al-substituted sample with comparable x.","section":"§3.1, Fig. 2(d)-(e)"}],"minor_comments":[{"comment":"Specify the temperature and pressure conditions for the '20 times smaller' comparison. YMn2 1/T1T is strongly T-dependent, and the legend indicates YMn2 at 0.33 GPa while Y0.96Lu0.04Mn2 appears without a pressure label; clarify whether the factor refers to a fixed temperature or to the peak of the enhancement.","section":"§3.2, Fig. 4"},{"comment":"Typo: 'temepratures' should read 'temperatures' in the last sentence of §3.2.","section":"§3.2"},{"comment":"State explicitly that for I=5/2 the ±1/2↔±3/2 and ±3/2↔±5/2 NQR lines occur at νzz and 2νzz, respectively, so that the consistency between the NMR fit (νzz=4.7 MHz) and the NQR peaks (4.78 and 9.55 MHz) is transparent to the reader.","section":"§3.1"},{"comment":"The Al-substituted 1/T1T data shown in Fig. 4 deserve more than the single sentence given. Please state whether the enhancement also saturates at low temperatures and how the absolute value compares with the In-substituted sample.","section":"§3.2"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope for JPSJ and the experimental core is solid. The main risk is the unverified cross-compound hyperfine normalization behind the 20x comparison; the authors are transparent about the difficulty but the abstract and conclusion are worded more strongly than the evidence allows. If they can supply any independent constraint on Ahf or reframe the claim as consistent with weakened exchange, the paper should be acceptable. The In-vs-Al site-selectivity claim would also benefit from a matched-concentration control, though this is secondary to the central physics. No concerns about novelty disclosure or citation practices."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here is the short version: the paper's solid contribution is site-selective 55Mn-NQR/NMR data showing that the low-temperature 1/T1T enhancement belongs to the 16d pyrochlore Mn sites, not to excess-Mn impurities. That is genuinely new and experimentally clean. The bigger interpretive claim — frustration-driven heavy fermion rather than AFM quantum criticality — is real but rests on a comparison with YMn2 that depends on an unmeasured hyperfine coupling, and I think the stress-test note lands there correctly.\n\nThe authors exploit In substitution's site selectivity at the 16c Zn site to get sharp NQR lines with νzz = 4.7 MHz (Mn) and 9.95 MHz (In), versus the much broader, more disordered Al-substituted case. 1/T1T rises below 10 K at both the Mn and In sites and then levels off at low temperature. The saturation is a clear Fermi-liquid signature and speaks against a divergent quantum-critical contribution. Seeing the same enhancement at two different nuclear sites is good evidence that the low-energy excitations are intrinsic to the Mn sublattice rather than impurity spin-glass behavior.\n\nThe soft spot is the one the stress-test identifies. In §3.2 the authors conclude that the ~20x smaller 1/T1T relative to YMn2 means substantially weakened magnetic interactions. That requires comparable 55Mn hyperfine couplings. They know this; they explicitly say a quantitative Knight shift–χ estimate is difficult because excess Mn pollutes the bulk susceptibility. Their fallback — that the Mn-site rate is larger than the In-site rate — only shows that on-site hyperfine dominates transferred hyperfine at the Mn nucleus. It does not calibrate the absolute on-site coupling against YMn2. If Ahf were a factor of three or five smaller, the 20x reduction would no longer imply suppressed fluctuations; the intrinsic spectrum could be comparable or even enhanced. So the cross-compound magnitude argument is underdetermined.\n\nA secondary, smaller gripe: the YMn2 comparison points in Fig. 4 are from YMn2 under 0.33 GPa and Y0.96Lu0.04Mn2, not pure ambient YMn2, so the hyperfine coupling could shift there too. That does not weaken the Fermi-liquid saturation argument, but it adds another layer to the 20x claim.\n\nThe data and fitting are standard, internally consistent, and free of circularity. The citation pattern is fine — prior work from the same group on structure and thermodynamics is appropriately cited, and the comparison data are external.\n\nWho this is for: experimentalists working on frustrated d-electron heavy fermions and on NMR/NQR studies of pyrochlore systems. It deserves a serious referee; the referee should push on the hyperfine-coupling normalization and ask for either a direct Knight-shift measurement or a more explicit argument for comparable Ahf before the strong mechanistic conclusion is accepted.","headline":"Site-selective NMR/NQR gives a credible intrinsic-Mn relaxation signature in YMn2Zn20, but the frustration-vs-quantum-critical readout leans on an unmeasured hyperfine-coupling comparison.","tokens_in":9955,"tokens_out":3587,"would_cite":true,"duration_ms":40963,"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":"The heavy-fermion state in YMn2Zn20 originates from the frustrated Mn pyrochlore network itself, as shown by site-selective NMR/NQR measurements.","keywords":["heavy fermion","pyrochlore lattice","geometric frustration","NMR","NQR","spin-lattice relaxation","YMn2Zn20","d-electron strongly correlated system"],"falsifier":"Measure the 55Mn Knight shift as a function of bulk susceptibility on samples whose excess Mn content is known and subtracted, to extract the hyperfine coupling Ahf; if Ahf is found to be much smaller than in YMn2, the 20-fold reduction in 1/T1T would not establish suppression of magnetic fluctuations.","tokens_in":9037,"feed_emoji":"🧲","tokens_out":4306,"duration_ms":42772,"temperature":0.7,"pith_summary":"This paper reports NMR and NQR measurements on the d-electron heavy-fermion candidates YMn2+δZn20-xInx and YMn2+δZn20-xAlx, with the aim of deciding whether the large electronic specific heat comes from the intrinsic Mn pyrochlore sublattice or from excess Mn impurities sitting on Zn sites. By isolating the resonance of the 16d pyrochlore Mn sites, the authors observe a clear enhancement of the spin-lattice relaxation rate divided by temperature (1/T1T) below 10 K, which tracks the specific-heat enhancement and saturates at low temperature. The relaxation enhancement is also seen at the In site, and its magnitude is about 20 times smaller than in YMn2, where antiferromagnetic correlations are strong. The authors conclude that the heavy-fermion state is intrinsic to the Mn pyrochlore network and is driven by frustration-induced low-energy magnetic excitations rather than by conventional antiferromagnetic quantum-critical fluctuations. If correct, this makes the compound a rare d-electron example where mass enhancement arises from geometrical frustration alone.","feed_headline":"NMR pins heavy-fermion state to Mn pyrochlore network","feed_subtitle":"Site-selective NMR/NQR shows the mass enhancement is intrinsic to the frustrated Mn sublattice, not impurity spins.","key_machinery":"The central probe is the nuclear spin-lattice relaxation rate 1/T1 measured by 55Mn NQR at zero field on the 16d pyrochlore Mn sites, along with supporting 115In and 27Al NQR/NMR. In a metal, 1/T1T is proportional to the imaginary part of the dynamical spin susceptibility summed over q and is therefore a direct measure of low-energy magnetic fluctuations. Site-selectivity lets the authors separate intrinsic pyrochlore Mn dynamics from excess Mn impurity spins. The comparison quantity is the same 1/T1T measured in YMn2, whose short Mn-Mn distance gives strong antiferromagnetic correlations.","core_discovery":"The central claim is that the heavy-fermion behavior of YMn2Zn20-based compounds is an intrinsic property of the Mn 16d pyrochlore sublattice. Using 55Mn NQR as a local probe at zero field, the paper isolates the intrinsic Mn response from the excess Mn impurity contribution that complicates bulk measurements. The measured 1/T1T is roughly temperature-independent at high temperature (Korringa behavior), rises below 10 K, and then levels off at low temperature, consistent with a Fermi-liquid ground state. The same enhancement appears at the 115In site, confirming it is not a site-specific artifact. Comparing with YMn2, the absolute 1/T1T is about 20 times smaller, which the authors attribute","pith_inferences":["If the frustration-driven scenario is correct, pressure tuning of the Mn-Mn distance should move the system toward a critical point, possibly revealing magnetically ordered or superconducting ground states; the authors hint at pressure studies.","The In-versus-Al difference suggests that substituting elements that selectively occupy specific Zn sites is a strategy to control disorder in cage compounds; a testable extension is to map the 1/T1T enhancement against In concentration.","The paper leaves open the quantitative value of the 55Mn hyperfine coupling; a Knight-shift measurement on a sample free of excess Mn would directly test the 20x suppression interpretation.","Similar low-energy enhancement may be visible in other d-electron pyrochlore systems with enlarged B-B spacing, suggesting a general route to frustration-driven heavy fermions."],"forward_implications":["The heavy-fermion state with γ≈280 mJ K−2 mol−1 is intrinsic to the Mn pyrochlore network; excess Mn impurities do not cause it.","The 1/T1T enhancement below 10 K that saturates at low T indicates a Fermi-liquid ground state, not divergent quantum-critical fluctuations.","The ~20x smaller 1/T1T versus YMn2 quantitatively shows weakened magnetic exchange from the enlarged Mn-Mn distance.","In substitution at the Zn 16c sites preserves a homogeneous Mn environment, making In-substituted samples the cleaner platform for studying this physics.","The similar temperature dependence seen at the In site supports a network-intrinsic origin for the low-energy excitations."],"fun_headline_variants":["NMR/NQR isolates intrinsic heavy-fermion response in Mn pyrochlore","Frustrated Mn lattice, not excess atoms, generates heavy fermions","Heavy-fermion state traced to Mn pyrochlore sublattice","Site-selective NQR reveals Mn sublattice heavy-fermion origin","Low-energy Mn excitations drive heavy-fermion formation"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The argument that the 20-fold smaller 1/T1T reflects weakened magnetic exchange rather than a smaller hyperfine coupling assumes that the 55Mn hyperfine coupling in YMn2Zn20 is comparable to that in YMn2; the paper cannot directly measure this because excess Mn impurities contaminate the bulk susceptibility used in Knight-shift analysis.","fun_headline_variants_meta":{"raw":{"variants":["NMR/NQR isolates intrinsic heavy-fermion response in Mn pyrochlore","Frustrated Mn lattice, not excess atoms, generates heavy fermions","Heavy-fermion state traced to Mn pyrochlore sublattice","Site-selective NQR reveals Mn sublattice heavy-fermion origin","Low-energy Mn excitations drive heavy-fermion formation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000285,"raw_usage":{"total_tokens":1565,"prompt_tokens":843,"completion_tokens":722,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":587,"completion_tokens_details":{"reasoning_tokens":627}},"tokens_in":587,"tokens_out":722,"duration_ms":7552,"temperature":1.0,"reasoning_tokens":627,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T02:02:30.902990+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the 55Mn Knight shift as a function of bulk susceptibility on samples whose excess Mn content is known and subtracted, to extract the hyperfine coupling Ahf; if Ahf is found to be much smaller than in YMn2, the 20-fold reduction in 1/T1T would not establish suppression of magnetic fluctuations.","supporting_citations":[],"review_version":2}