{"id":"7c9279d1-b484-402e-b48b-5354399a4f8d","arxiv_id":"2507.22249","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Muon spin relaxation and susceptibility reveal a possible new magnetic transition near 50 K in FeSn, and show that Fe0.2Co0.8Sn is a canonical spin glass driven by random Fe moments, not kagome frustration.","lead":"Using muon spin relaxation, susceptibility, and scanning tunneling microscopy, the authors find signs of a possible new magnetic phase in the kagome metal FeSn at about 50 K, far below its 376 K magnetic ordering temperature, and they show that the cobalt-rich composition Fe0.2Co0.8Sn behaves as a canonical spin glass.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 50 K feature is not yet shown to be an intrinsic bulk FeSn phase; the muon-site model is unvalidated and a cited 1971 paper may already report a spin-flip effect, so the claim should stay conditional pending a muon-free bulk check.","rationale":"The reader's weakest assumption identifies essentially the same load-bearing point: the 50 K anomaly must be intrinsic to FeSn and not a muon-specific or impurity effect. I agree with that assessment. The paper's strongest supporting evidence is the correspondence between the muSR relaxation peak and the ac susceptibility chi'' rise; this is genuine muon-free evidence and the authors' cautious wording is appropriate. However, the interpretation still depends on an unvalidated muon-site and hyperfine model, the amplitude-based bulk argument is fit-model dependent, and no direct probe of the Fe sublattice (Mossbauer, neutron diffraction, or heat capacity) is presented across T*. In addition, the manuscript cites Ligenza (1971) with a title suggestive of a spin-flip effect in FeSn without discussing it as a possible prior observation, which weakens the 'previously unidentified' qualifier. These considerations do not invalidate the measurements, but they reinforce the need for a conditional verdict pending an independent bulk probe and a literature check; no change to the reader's verdict is required.","tokens_in":22107,"tokens_out":10418,"duration_ms":134619,"concrete_test":"Perform 57Fe Mossbauer absorption spectroscopy on the same FeSn single crystal from 5 to 100 K, collecting spectra with fine temperature steps around 40-70 K. An intrinsic Fe-moment transition or spin reorientation at T* should appear as a change in the Fe hyperfine field, linewidth, relative line intensities, or as a relaxation-broadened subspectrum. If the Mossbauer spectra are temperature-independent through 50 K, the muSR 1/T1 peak and chi'' rise cannot be attributed to bulk Fe magnetism and the 'new phase' claim should be retracted; if a clear anomaly appears, T* is confirmed as intrinsic. In parallel, read Ligenza (1971) to determine whether the transition was already reported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the ~50 K anomaly is an intrinsic bulk magnetic instability of FeSn. This requires the muSR 1/T1 peak and 1/T2 rise to reflect Fe-moment dynamics rather than muon-specific effects, and the susceptibility kink to be intrinsic to FeSn rather than a minority phase. The current evidence is suggestive but stops short of excluding a muon-site or hyperfine-environment change: the DFT/muon-site calculation in Sec. III D is the only model for the two muon sites, the hyperfine contact field is transferred from bcc Fe, and the 'at least half the full amplitude' argument in Sec. III A 4 refers to fitted LF-0.4 T exponential decays, not to a directly measured magnetic volume fraction. The ac chi'' anomaly is muon-free, but it is small and could in principle share a non-muon origin such as a minority impurity phase or an instrumental background effect; no heat capacity, neutron, or Mossbauer data crossing 50 K are shown. The manuscript also cites Ligenza (1971), whose title 'A spin-flip effect in FeSn' suggests a low-temperature spin transition may already have been reported, yet Sec. IV A discusses possible origins without addressing that reference. These gaps leave the 'previously unidentified bulk phase' designation conditional.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a muon spin relaxation, magnetic susceptibility, and scanning tunneling microscopy study of the kagome metal FeSn and two Co-substituted variants, Fe0.89Co0.11Sn and Fe0.2Co0.8Sn. In FeSn the authors observe a peak in the longitudinal relaxation rate 1/T1 near 50 K in both zero-field and longitudinal-field measurements, an increase in the transverse relaxation rate 1/T2 below 50 K, and a small kink in dc susceptibility together with a rise in ac susceptibility chi'', which they present as possible signatures of a previously unidentified phase below the Neel temperature T_N ~ 376 K. They also report the muon-site occupancy problem, with about 40% of muons at a field-cancelling site, and argue from the Co-doped compound that this fraction is a muon-site effect rather than phase separation. In Fe0.2Co0.8Sn they report canonical spin-glass freezing at Tg ~ 3.5 K, with 1/T1 peaking at Tg and dying away at low temperature, and they argue that the spin-glass behavior is driven by dilution and randomness of Fe moments rather than by geometric frustration.","tokens_in":22417,"tokens_out":7231,"duration_ms":82088,"significance":"If the 50 K anomaly is intrinsic to bulk FeSn, it would constitute a new low-temperature magnetic instability in a well-studied kagome antiferromagnet and would be worth comparing with analogous muSR anomalies in CrSBr and REMn6Sn6. The paper has clear strengths: the anomaly is supported by several measurements (muSR 1/T1 in ZF and LF, 1/T2, ac and dc susceptibility), the wording is carefully hedged as 'possible signatures,' the unidentified origin is disclosed, and the muon-site calculation attempts to connect the observed 0.195 T field to a concrete structural model. The spin-glass section is particularly strong, with a clean comparison to AuFe and CuMn, a small frustration index |theta_CW|/Tg ~ 0.63, and a clear contrast with persistent-dynamics frustrated systems; those conclusions do not depend on the contested 50 K claim. The main weakness is that the bulk nature of the 50 K feature is not yet independently established: the principal muon-free evidence is a small susceptibility anomaly, and the cited 1971 paper on a spin-flip effect in FeSn is not reconciled with the 'previously unidentified' framing.","major_comments":[{"comment":"The claim that the ~50 K feature is a 'previously unidentified phase' must be reconciled with Ref. [98], Ligenza, 'A spin-flip effect in FeSn' (1971), which is cited in Sec. IV A as part of a list of Mossbauer studies showing a fully ordered volume. If that paper already reports a low-temperature spin transition in FeSn, then the novelty statement is too strong; at minimum, the authors should state what the 1971 report found and why it does not already account for the T* anomaly. As written, the reader cannot verify the 'previously unidentified' part of the central claim.","section":"Sec. IV A (Ref. [98])"},{"comment":"The argument that the LF-0.4 T spectral change involves 'at least half of the full amplitude' and therefore rules out a minority impurity phase is not sufficient for the bulk-phase conclusion. The fitted function in this section is a single exponential over the time range 0.5-10 micro-s, and the amplitude of the relaxing component is not shown to correspond to the full bulk sample; a change affecting only the ~60% high-field-site muons or a muon-site transition would produce the same observation in an intrinsic sample. To elevate the 50 K feature from 'possible signatures' to a bulk phase, the authors should provide a muon-free bulk measurement crossing 50 K, for example heat capacity, neutron diffraction, or Mossbauer spectroscopy, or explicitly restrict the claim to an anomaly of unidentified origin.","section":"Sec. III A 4"},{"comment":"The muon-site simulation is not validated as a function of temperature. The local field at the high-field site is computed from a dipolar sum using an Fe moment of 1.85 mu_B and a contact hyperfine field of -1.11 T transferred from bcc Fe; the result is then compared with the observed 0.195 T, and the authors add that a hyperfine field of -1.07 T can alternatively be estimated from the data. This internal consistency does not independently determine the muon site or exclude a low-temperature site change, which is one of the proposed muon-specific explanations for the observed anomalies. The conclusion that the 50 K anomaly is intrinsic should therefore remain provisional until the site model is checked by a complementary method or the muon-specific scenarios are explicitly ruled out.","section":"Sec. III D"}],"minor_comments":[{"comment":"The phrase 'with 1/T1 -> as T->0' is incomplete; it should read 'with 1/T1 -> 0' or state the numerical upper bound established by the data.","section":"Abstract"},{"comment":"The sentence 'In contrast, 1/T2 peak at T~250 K' should be 'In contrast, the 1/T2 peak at T~250 K'; the same paragraph would benefit from an explicit statement that this peak is attributed to muon diffusion before the 50 K anomaly is discussed in the same panel.","section":"Sec. III A 2"},{"comment":"The symbol theta_CW is used in the figure and text without being defined in the caption; define it as the Curie-Weiss temperature obtained from the high-temperature inverse susceptibility fit.","section":"Fig. 16"},{"comment":"The statement that the frozen moment size ~0.45 mu_B is 'nearly half' of the effective moment 0.81 mu_B would benefit from a citation or a short derivation, since the expected ratio in spin glasses is not obvious to all readers.","section":"Sec. III F 2"}],"recommendation":"major_revision","confidential_remarks":"The spin-glass half of the paper is solid, well contextualized, and likely publishable on its own. The 50 K anomaly is carefully hedged, and the data set is rich, but the 'previously unidentified phase' framing needs to be squared with Ref. [98] and with the absence of a muon-free bulk probe. I would not demand a new experiment as a condition of a first revision, but the authors should either provide the missing bulk check or soften the central claim to 'anomaly of unidentified origin' throughout the abstract and conclusions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a careful, honest muSR paper. The headline is a possible new 50 K anomaly in FeSn; the authors say \"possible signatures\" and they mean it. The evidence is a 1/T1 peak in ZF and LF, a rise in 1/T2, and a matching kink in dc susceptibility plus a rise in ac chi''. That is two independent bulk probes pointing at the same temperature, which is more than most such claims get. The spin-glass section on Fe0.2Co0.8Sn is the stronger half: canonical ZF/LF behavior, 1/T1 peak at Tg, dynamics dying at low temperature, frequency-dependent ac susceptibility. The contrast with persistent spin dynamics in frustrated insulators is well made, and the randomness-over-frustration argument is supported by the small |theta_CW|/Tg ratio. I buy that part.\n\nWhat is genuinely new: the 50 K anomaly has not been reported before (though a kink is visible in earlier susceptibility), and the muSR spin-glass characterization of the x=0.8 compound is new. The muon-site simulation confirms the 1980s Hartmann-Wappling report and gives a plausible picture of the field-cancelling site. The hyperfine field estimate is a consistency check, not a first-principles prediction, but it is presented as such.\n\nSoft spots. The 50 K anomaly is suggestive, not yet a phase. There is no heat capacity, neutron, or Mossbauer data crossing 50 K, and the ac chi'' anomaly is small. The \"half the full amplitude\" argument in Sec. III A 4 is reasonable but it is not a direct magnetic volume-fraction measurement. The stress-test's point that it refers to fitted exponentials is a bit too literal: the text is describing the change in the time spectra themselves. Still, a muon-free bulk probe would firm this up. Second, the paper cites Ligenza (1971) \"A spin-flip effect in FeSn\" but never discusses whether that earlier report is connected to the low-temperature behavior. That is an odd miss and should be addressed. Third, the muon-site model depends on a hyperfine field taken from bcc Fe; it works, but it is not a validated muon-site calculation. That is a minor issue since the main claims do not rest on it.\n\nWho this is for: muSR people, kagome metal experimentalists, anyone mapping the (Fe,Co)Sn phase diagram. It deserves a serious referee. The spin-glass part is close to publishable as is; the 50 K part needs the claims kept conditional, the Ligenza question answered, and ideally one non-muon bulk probe crossing 50 K. I would send it out.","headline":"Careful muSR paper with a solid spin-glass result and a suggestive 50 K anomaly that stays honestly labeled as possible; send it to referees but push for a muon-free bulk check.","tokens_in":757,"tokens_out":1000,"would_cite":true,"duration_ms":52449,"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":"Muon spin relaxation and susceptibility data point to a previously unidentified magnetic transition near 50 K inside the ordered state of the kagome metal FeSn.","keywords":["kagome metals","muon spin relaxation","FeSn","spin glass","magnetic phase transition","geometrical frustration","spin dynamics","Co substitution"],"falsifier":"Cool FeSn through 50 K and measure the iron Mossbauer spectrum or neutron spin-echo correlation; if no anomaly appears in the iron hyperfine field or bulk fluctuation spectrum near T* while the muon 1/T1 peak remains, the peak is a muon-specific artifact and the claimed phase is not intrinsic.","tokens_in":21878,"feed_emoji":"🧲","tokens_out":5937,"duration_ms":66408,"temperature":0.7,"pith_summary":"The paper tries to establish that FeSn, a metallic kagome antiferromagnet, hosts a previously unidentified magnetic instability near T* about 50 K, deep inside its Neel-ordered state, and that Co-substituted Fe0.2Co0.8Sn is a canonical spin glass whose freezing comes from random doped Fe moments rather than kagome frustration. If the 50 K feature is intrinsic, models of kagome magnetism must include an extra low-energy scale; if the spin-glass classification holds, metallic kagome materials with dilute moments behave like ordinary alloy glasses, not like frustrated insulators with persistent dynamics. The paper also clarifies which muon signals are intrinsic to the material and which are artifacts of muon diffusion and site occupancy.","feed_headline":"Hidden 50 K magnetic transition found in kagome metal FeSn","feed_subtitle":"Muon spin relaxation and susceptibility agree on a new phase below the Neel transition; doped crystals show canonical spin-glass freezing.","key_machinery":"The central tool is muon spin relaxation (muSR), where implanted spin-polarized muons precess and relax in local magnetic fields; the longitudinal relaxation rate 1/T1 tracks dynamic spin fluctuations and the transverse rate 1/T2 the static field width. A second central object is the field-cancelling muon site mu2, a high-symmetry interstitial where the dipolar fields of the ordered Fe moments cancel, producing a roughly 40 percent non-precessing signal; simulations combining dipolar and hyperfine fields reproduce the observed internal field and site occupancy, and the disappearance of the field-cancelling site in Fe0.89Co0.11Sn supports the assignment.","core_discovery":"In FeSn the muon longitudinal relaxation rate 1/T1 shows a peak near 50 K in zero and longitudinal fields, the transverse width 1/T2 increases below the same temperature, and ac and dc susceptibility show a corresponding kink; together these are presented as possible signatures of a new phase setting in well below T_N ~ 376 K. STM finds no trimerization or rotational symmetry breaking down to 7 K, so the paper does not identify the order parameter. In Fe0.2Co0.8Sn, the muon spectra freeze into a static Kubo-Toyabe form at T_g ~ 3.5 K with 1/T1 going to zero at low temperature, matching canonical dilute-alloy spin glasses CuMn and AuFe; because the Weiss temperature and frustration index are tiny, the paper concludes randomness of Fe moments, not geometric frustration, drives the glass.","pith_inferences":["If similar 1/T1 peaks are found in CrSBr and TbMn6Sn6, the 50 K type anomaly may be a generic feature of layered metallic antiferromagnets with ferromagnetically correlated triangular or kagome planes; the underlying mechanism could be interlayer coupling or moment reorientation rather than kagome frustration.","A muon-free bulk probe such as Mossbauer spectroscopy or neutron spin echo across 50 K would settle whether the peak is genuine spin fluctuation; the paper's own plan of high-resolution X-ray scattering could detect a structural component if the transition is a CDW-type instability.","The spin-glass result suggests that in metallic kagome systems, flat-band electronic correlations and dilute-moment magnetism may be largely decoupled, which could help separate charge-order and magnetic-glass searches in the same material family."],"forward_implications":["If the 50 K anomaly is a bulk phase, FeSn has two magnetic energy scales, T_N ~ 376 K and T* ~ 50 K, that any theory of its magnetic order must reproduce.","The spin-glass state of Fe0.2Co0.8Sn should be classified with canonical dilute-alloy glasses, not with geometrically frustrated quantum spin liquids, since its dynamics vanish as T goes to 0.","The 40 percent non-oscillating muon fraction in FeSn is explained by a high-symmetry field-cancelling site rather than phase separation, as supported by its disappearance in Fe0.89Co0.11Sn where sublattice disorder removes the cancellation.","Muon diffusion, not intrinsic magnetic order, accounts for the 250 K 1/T2 peak, separating muon-specific from intrinsic effects.","No trimer formation or rotational symmetry breaking appears on the bulk FeSn surface at 7 K, so the 50 K transition, if real, does not manifest as the trimer order seen in thin films."],"supporting_citations":[{"why":"Earlier muSR study of FeSn reporting the 60/40 site occupancy and field-cancelling behavior that this paper reexamines and extends.","marker":"[77]"},{"why":"Maps the (Fe,Co)Sn phase diagram including the spin-glass region and T_g scaling that the muSR results here are compared against.","marker":"[75]"},{"why":"Provides the canonical CuMn and AuFe spin-glass muSR behavior used as the template for classifying Fe0.2Co0.8Sn.","marker":"[79]"},{"why":"Supplies FeSn magnetic properties including the ordered moment size and the susceptibility derivative kink near 50 K.","marker":"[81]"},{"why":"Reports trimer formation in FeSn thin films, the claim the bulk STM measurements here test and do not reproduce.","marker":"[87]"},{"why":"Neutron study of FeSn and CoSn spin excitations that found no structural anomaly, used to argue against a CDW interpretation of the 50 K feature.","marker":"[43]"},{"why":"Shows a similar low-temperature muSR anomaly in the layered antiferromagnet CrSBr, cited as evidence that the FeSn behavior may be common to this material class.","marker":"[94]"},{"why":"Reports a similar magnetic crossover in the kagome magnet TbMn6Sn6, supporting the idea of shared physics in layered kagome antiferromagnets.","marker":"[95]"}],"fun_headline_variants":["Muon spin hints at possible phase at 50 K in FeSn kagome","Kagome FeSn: possible new magnetic phase near 50 K","Spin glass in doped kagome from Fe randomness","FeSn kagome: 50 K anomaly seen by muon spin relaxation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The 50 K phase claim rests on the assumption that the muon relaxation peak and susceptibility kink are intrinsic bulk magnetism, not muon diffusion, a muon-site change, or a minority impurity phase.","fun_headline_variants_meta":{"raw":{"variants":["Muon spin hints at possible phase at 50 K in FeSn kagome","Kagome FeSn: possible new magnetic phase near 50 K","Spin glass in doped kagome from Fe randomness","FeSn kagome: 50 K anomaly seen by muon spin relaxation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001182,"raw_usage":{"total_tokens":4984,"prompt_tokens":1146,"completion_tokens":3838,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":762,"completion_tokens_details":{"reasoning_tokens":3759}},"tokens_in":762,"tokens_out":3838,"duration_ms":28700,"temperature":1.0,"reasoning_tokens":3759,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T11:54:03.484643+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Cool FeSn through 50 K and measure the iron Mossbauer spectrum or neutron spin-echo correlation; if no anomaly appears in the iron hyperfine field or bulk fluctuation spectrum near T* while the muon 1/T1 peak remains, the peak is a muon-specific artifact and the claimed phase is not intrinsic.","supporting_citations":[{"cited_title":"Guoet al., Correlated order at the tipping point in the kagome metal CsV 3Sb5, Nature Physics , 1 (2024)","cited_arxiv_id":null,"evidence_quote":"Earlier muSR study of FeSn reporting the 60/40 site occupancy and field-cancelling behavior that this paper reexamines and extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the canonical CuMn and AuFe spin-glass muSR behavior used as the template for classifying Fe0.2Co0.8Sn."},{"cited_title":"Fanget al., Ferromagnetic helical nodal line and Kane-Mele spin-orbit coupling in kagome metal Fe 3Sn2, Physical Review B105, 035107 (2022)","cited_arxiv_id":null,"evidence_quote":"Neutron study of FeSn and CoSn spin excitations that found no structural anomaly, used to argue against a CDW interpretation of the 50 K feature."},{"cited_title":"Kakihana, K","cited_arxiv_id":null,"evidence_quote":"Shows a similar low-temperature muSR anomaly in the layered antiferromagnet CrSBr, cited as evidence that the FeSn behavior may be common to this material class."},{"cited_title":"Zhang, B","cited_arxiv_id":null,"evidence_quote":"Reports a similar magnetic crossover in the kagome magnet TbMn6Sn6, supporting the idea of shared physics in layered kagome antiferromagnets."}],"review_version":1}