{"id":"5a7b47f4-3340-4d1d-b0e2-aea9e159cf82","arxiv_id":"1908.02343","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Josephson junctions made with Nb leads and a SmB6 topological Kondo insulator weak link show supercurrent, Shapiro steps, and hysteretic magnetic diffraction patterns attributed to Kondo-breakdown surface ferromagnetism.","lead":"Physicists built superconducting junctions on the surface of samarium hexaboride, a material whose interior is an insulator, and saw supercurrent flow through its surface with unusual magnetic-field patterns. The patterns suggest that surface magnetism and superconductivity can coexist, potentially useful for future topological quantum devices.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The SmB6 surface-ferromagnetism interpretation of the anomalous Ic(B) patterns is not directly tested; without a magnetization measurement or a nonmagnetic control, the central claim remains an inference.","rationale":"I read the paper in good faith. The transport observations are solid: induced supercurrent, Shapiro steps, and a Fraunhofer-like response in at least one junction. The problem is not the data but the interpretation. The anomalous diffraction/hysteresis is assigned to surface ferromagnetism from Kondo breakdown, yet no independent evidence for such magnetism is provided, and the text itself enumerates conventional alternatives that are not ruled out. The Kondo-breakdown framework is imported from theory, not fit to the data. Since the reader's CONDITIONAL verdict already captures this weakness and my independent read reaches the same conclusion, no verdict change is needed.","tokens_in":10519,"tokens_out":7052,"duration_ms":81479,"concrete_test":"Perform scanning SQUID microscopy (or a micro-Hall magnetometry scan) on the same patterned SmB6 junction at base temperature after field cycling through the hysteresis loop used in Fig. 3, and look for a remanent, hysteretic local magnetic-field profile localized at the junction that correlates with the direction of the last field sweep. Observing such a profile would directly support Kondo-breakdown surface ferromagnetism; its absence would falsify the central interpretation, since the proposed mechanism requires ferromagnetic domains generating flux inside the junction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is the identification of the anomalous critical-current diffraction patterns (Figs. 3(b)-(d)) and normal-state magnetoresistance hysteresis (Figs. 4(b)-(c)) with ferromagnetic SmB6 surface states produced by Kondo breakdown. The paper contains no direct measurement of surface magnetization or of stray flux from such domains; the connection is made by invoking the Kondo-breakdown framework of refs [38,39], one of which is coauthored by a member of the experimental team. The warm-up-to-20-K test excludes trapped flux in the magnet and in Nb electrodes, but it does not exclude the mechanisms the authors themselves list: geometric asymmetry, self-field effects from nonuniform current, magnetocaloric effects, or magnetic impurity scattering. The same self-acknowledged ambiguity applies to the normal-state butterfly hysteresis, where the text concedes that magnetocaloric effects and magnetic impurity scattering must be considered. No quantitative comparison links the size or direction of the observed ±5 mT central-peak shift (or the direction of the hysteresis) to a particular ferromagnetic domain configuration, domain-wall flux, or Kondo-breakdown parameter. Therefore, the central claim of coexistence of surface ferromagnetism with proximity-induced superconductivity is an inference from transport data alone; if the anomalies were reproduced by a nonmagnetic control junction or by a self-field/asymmetry model, the unconventional interpretation would be reduced to unexplained deviations.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports fabrication and low-temperature transport measurements of lateral Josephson junctions in which superconducting Nb leads are coupled to the (001)-polished surface of SmB6 single crystals. The authors demonstrate a supercurrent that persists up to 3-6 K, observe Shapiro steps under microwave irradiation with voltage spacing consistent with hf/2e, and measure critical-current diffraction patterns in an out-of-plane magnetic field. One junction shows a Fraunhofer-like pattern; two other junctions show missing side lobes, a central peak shifted by approximately ±5 mT, and hysteresis with field sweep direction. A fourth junction shows a nonmonotonic temperature dependence of Ic and hysteretic, sweep-rate-dependent normal-state magnetoresistance. The authors attribute these anomalies to ferromagnetism on the SmB6 surface arising from Kondo breakdown, coexisting with proximity-induced superconductivity.","tokens_in":10804,"tokens_out":5521,"duration_ms":58149,"significance":"The direct observation of a Josephson supercurrent and Shapiro steps in Nb/SmB6/Nb junctions is a useful experimental contribution that supports the promise of topological Kondo insulators as clean weak links with an insulating bulk. If the ferromagnetism interpretation were firmly established, the result would point to a new coexistence regime of surface magnetism and proximity superconductivity. However, the paper's central claim is currently an inference from transport data alone. The manuscript itself lists several alternative mechanisms—geometric asymmetry, self-field effects, magnetocaloric effects, and magnetic impurity scattering—and does not quantitatively rule them out. The significance of the paper as a claim about unconventional Josephson behavior is therefore contingent on additional evidence that is not provided.","major_comments":[{"comment":"The attribution of the shifted central peak and sweep-direction hysteresis to ferromagnetic surface states is not uniquely supported by the presented data. The manuscript explicitly lists geometric asymmetry, self-field effects, magnetocaloric effects, and magnetic impurity scattering as possible causes, but provides no quantitative estimate or control experiment to exclude them. The warm-up-to-20-K test only rules out trapped flux in the Nb electrodes and the magnet; it does not address the other mechanisms. Without a direct magnetization measurement, a nonmagnetic control junction, or a quantitative model that links the ±5 mT shift and the hysteresis direction to a specific ferromagnetic domain configuration, the coexistence claim remains an unsupported inference.","section":"Fig. 3(b)-(d) and the 'magnetic field response' paragraph"},{"comment":"The theoretical Fraunhofer overlay is not sufficiently specified. If the effective junction area used to compute the flux period is a fitted parameter, the 'reasonable agreement' between the data and the Fraunhofer curve does little to establish that junction-3 is conventional; if it is measured, the geometric dimensions and any field-focusing corrections should be stated. This comparison underlies the contrast between conventional and anomalous devices and needs to be transparent for the reader to assess the significance of the deviations.","section":"Fig. 3(a) inset"},{"comment":"Only four junctions are reported, and each anomalous diffraction pattern or hysteresis curve is shown for a single device without repeated sweeps, error bars, or sample-to-sample statistics. Since the text itself notes uncontrolled variations in polishing and nanofabrication as possible causes of surface modification, it is not possible to distinguish intrinsic SmB6 surface behavior from device-specific artifacts without either multiple devices showing consistent anomalies or a control junction with a nonmagnetic weak link.","section":"Device statistics and reproducibility"},{"comment":"The nonmonotonic Ic(T) with an increase between 300 mK and 1.6 K is presented as 'could be consistent with' Kondo breakdown and thermal activation of trivial carriers, but no model, calculation, or independent measurement is given to support this assignment. Similarly, the normal-state butterfly hysteresis is acknowledged to be possibly due to magnetocaloric effects or magnetic impurity scattering; the connection of these data to the supercurrent anomalies is not established by any quantitative comparison. These interpretive steps are load-bearing for the claim that Kondo-breakdown ferromagnetism coexists with the induced supercurrent.","section":"Fig. 4 and the 'temperature dependence' paragraph"}],"minor_comments":[{"comment":"The name 'Shaphiro steps' appears repeatedly and should be corrected to 'Shapiro steps.'","section":"Throughout"},{"comment":"There is a spacing error in 'Kond o' in the title; this should be corrected.","section":"Title and abstract"},{"comment":"The voltage criterion used to define Ic from the IV characteristics is not stated; this should be specified for reproducibility.","section":"Methods/Measurement details"},{"comment":"The claim of 2π periodicity rests on observing the fundamental Shapiro spacing hf/2e; explicitly showing the absence of half-integer steps (i.e., no features at hf/e) would strengthen this conclusion.","section":"Fig. 2 and AC Josephson effect"},{"comment":"The phrase 'There has been an evidence of helical nature of such states' should be reworded, for example to 'Evidence for the helical nature of such states has been reported.'","section":"Introduction"},{"comment":"The statement that the work will advance understanding of 'topologically nontrivial superconductors' overreaches, as no evidence for topological superconductivity or Majorana modes is presented; the claim should be toned down to match the demonstrated results.","section":"Conclusion"}],"recommendation":"major_revision","confidential_remarks":"The interpretive framework relies heavily on refs [38,39]; ref [39] is coauthored by one of the present authors (P. Ghaemi). This is not inherently inappropriate, but the self-referential support should be acknowledged, and the framework should ideally be tested against alternative models. In addition, the relationship to prior SmB6 Josephson-junction work (e.g., ref [33]) should be clarified in the introduction, since the novelty claim depends on precisely what was demonstrated there. The paper's strengths are the clean supercurrent and Shapiro-step observations; the revision should focus on either directly supporting the ferromagnetism interpretation or substantially reframing the claims as a report of anomalous transport phenomena without a definitive microscopic explanation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Punchline: the paper reports something real and new—proximity supercurrent in Nb/SmB6/Nb junctions, with Shapiro steps at the correct 2e spacing and a Fraunhofer-like pattern in one junction—and it documents anomalies in the magnetic diffraction pattern (hysteresis, missing side lobes, shifted central peak) that I have not seen in earlier SmB6 Josephson work. The soft spot is the interpretation: attributing these anomalies to surface ferromagnetism from Kondo breakdown is an inference that the data do not directly support.\n\nWhat the paper does well: four junctions of different lengths, careful microwave measurements at multiple frequencies, a warm-up test that excludes trapped flux in the Nb leads, and an honest discussion of alternative mechanisms. The supercurrent persisting to several kelvin is a useful data point for the field. The authors also show opposite shift directions for two junctions, which argues against a single trivial asymmetry and in favor of something sample-specific. That is genuine new information.\n\nWhere it is soft: the central claim of coexisting surface ferromagnetism and proximity superconductivity rests on connecting the anomalous diffraction patterns and the normal-state butterfly magnetoresistance to Kondo-breakdown ferromagnetism. There is no magnetization measurement, no control junction without SmB6, and no quantitative link between the observed ±5 mT shift or hysteresis direction and any specific domain configuration. The paper itself lists geometric asymmetry, self-field effects, magnetocaloric effects, and magnetic impurity scattering as possible explanations for parts of what is seen, and the warm-up test does not rule those out. The reliance on the authors' own theoretical framework (refs [38,39], one coauthored by a team member) makes the interpretation more of a proposal than an independent test. The anomalous temperature dependence of Ic is also explained post hoc with the same framework. So the 'unconventional' label is really a hypothesis.\n\nCitation pattern: not a problem in itself—the prior SmB6 Josephson paper is cited and the new observations go beyond it—but the self-referential nature of the interpretation should be flagged in any review.\n\nWho should read this: experimentalists working on SmB6, topological insulator junctions, and proximity effects. It deserves a serious referee because the experiment is nontrivial and the observations are new. A good referee should ask for a nonmagnetic control, a magnetization probe, or a quantitative analysis of self-field/asymmetry before the ferromagnetism claim is accepted. As it stands, I would accept it as a report of anomalous Josephson behavior in SmB6 weak links, with the ferromagnetism claim framed as a proposal rather than a conclusion.","headline":"Solid experiment, credible supercurrent and Shapiro steps in SmB6 weak links, but the Kondo-breakdown ferromagnetism interpretation is a hypothesis that the data do not directly test.","tokens_in":11351,"tokens_out":3428,"would_cite":true,"duration_ms":35081,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["85.25.Dq","74.45.+c","74.90.+n"],"model":"deepseek-v4-flash","headline":"The paper reports that niobium leads on a samarium hexaboride crystal form a Josephson junction whose supercurrent flows through topological surface states and whose shifted, hysteretic magnetic diffraction patterns signal surface…","keywords":["topological Kondo insulator","samarium hexaboride (SmB6)","Josephson junction","proximity-induced superconductivity","Fraunhofer diffraction pattern","Shapiro steps","surface ferromagnetism","Kondo breakdown"],"falsifier":"A decisive check: place a polished SmB6 crystal of the same provenance in a SQUID magnetometer or under a magnetic force microscope and cycle the out-of-plane field through ±150 mT at base temperature. The Kondo-breakdown picture predicts a hysteretic remanent surface moment large enough to shift the junction's central diffraction peak by a few mT; if no such surface magnetization is resolved, the anomalous Fraunhofer patterns must instead be assigned to geometric asymmetry, self-fields, or flux redistribution.","tokens_in":10323,"feed_emoji":"🧲","tokens_out":11774,"duration_ms":116118,"temperature":0.7,"pith_summary":"The paper reports Josephson junctions made by placing two superconducting niobium leads close together on the surface of a samarium hexaboride (SmB6) crystal, and argues that the supercurrent flows through the crystal's topological surface states rather than through a metallic bulk. The key observations are a robust critical current that persists to several kelvin, Shapiro steps under microwave radiation, and magnetic diffraction patterns that deviate strongly from the standard Fraunhofer pattern: missing side lobes, a central peak shifted away from zero field, and hysteresis that depends on field sweep direction. The authors interpret these deviations as evidence that the SmB6 surface develops ferromagnetic domains at low temperature, produced by a Kondo breakdown that frees f-electron moments near the surface. If correct, SmB6 becomes a platform where surface ferromagnetism and proximity-induced superconductivity coexist, which matters for proposals that put Majorana bound states at the interface between superconductors and topological surfaces.","feed_headline":"SmB6 junction supercurrent shows ferromagnetic fingerprints","feed_subtitle":"Missing side lobes, a shifted peak, and hysteresis point to surface magnetism beside the supercurrent.","key_machinery":"The carrying object is the lateral Nb–SmB6–Nb junction: two sputtered niobium electrodes separated by 50–200 nm on a polished SmB6 crystal, with an in-situ argon mill applied immediately before deposition to obtain a clean interface. Its function is to force the proximity supercurrent through the SmB6 surface rather than through a bulk channel, making the junction a phase-sensitive interferometer for anything living on that surface. The interpretive machinery is the Kondo-breakdown picture of the SmB6 surface, which supplies the ferromagnetic flux: near the crystal boundary, reduced screening lowers the Kondo temperature so that f-electron moments break out of singlets, order ferromagnetically, and imprint extra flux on the junction. The diffraction pattern is then read as a balance between applied field and internal magnetization, which naturally explains both the shifted central peak and the sweep-direction hysteresis.","core_discovery":"The central claim, on the paper's own terms, is that the low-temperature supercurrent in these Nb/SmB6/Nb junctions is carried by the topological surface states of SmB6, and that the junctions' response to out-of-plane magnetic field reveals the coexistence of surface magnetism and induced superconductivity. Concretely, the devices show critical currents up to 3–6 K, Shapiro steps at voltage spacing $hf/2e$ for microwave frequency $f$, and magnetic diffraction patterns that deviate from the standard Fraunhofer form $I_c(\\Phi)=I_c(0)|\\sin(\\pi\\Phi/\\Phi_0)/(\\pi\\Phi/\\Phi_0)|$: the side lobes can be absent, the central maximum can sit at $\\pm 5$ mT rather than zero, and the critical current at zero field depends on the direction of the field sweep. Because warming to 20 K and recooling do not restore the zero-field critical current, the paper rules out trapped vortices as the cause and instead attributes the extra flux to ferromagnetic surface states. The proposed microscopic origin is Kondo breakdown: at the outermost layers, the reduced screening of f-electron moments frees them from Kondo singlets, and these moments form ferromagnetic domains that generate the flux. The same Kondo-breakdown picture is used to explain the anomalous rise of critical current between 300 mK and 1.6 K and the butterfly-shaped, sweep-rate-dependent magnetoresistance with minima near $\\pm 24$ mT.","pith_inferences":["Extending beyond the paper: a control Nb/Cu/Nb junction with the same lithographic geometry and field protocol would isolate the SmB6 surface contribution from geometric asymmetry and self-field effects.","Extending beyond the paper: if Kondo breakdown controls the surface magnetism, the hysteresis and peak shift should vary systematically with surface termination and polishing quality, since the polar (001) surface and polish-induced disorder directly set the breakdown layer thickness.","Extending beyond the paper: direct probes of surface magnetization on the same crystals—spin-polarized scanning tunneling microscopy, anomalous Hall effect, or magnetic torque—could confirm the ferromagnetic domains and turn the junction into a calibrated local magnetometer."],"forward_implications":["Because the supercurrent in SmB6 junctions survives to 3–6 K, topological-surface weak links become usable in a temperature range far more accessible than the sub-1 K operation typical of Nb/Bi2Se3 junctions.","The offset and hysteresis of the magnetic diffraction pattern give a junction-based readout of surface magnetization: the applied field needed to maximize $I_c$ cancels the flux from the ferromagnetic surface.","The data support a three-regime temperature picture of Kondo breakdown—surface singlet transport at the base, extra trivial-carrier supercurrent between ~0.3 K and ~1.6 K, and supercurrent suppression above ~1.6 K by freed moments.","If the surface is truly ferromagnetic while proximity-superconducting, SmB6 junctions are a setting in which magnetism and superconductivity meet at a topological surface, one of the ingredients proposed for topological superconductivity and Majorana bound states."],"supporting_citations":[{"why":"Provides the Kondo-breakdown theory that frees f-electron moments at the SmB6 surface, the paper's proposed origin of the ferromagnetic flux.","marker":"[38, 39]"},{"why":"Establishes prior evidence for ferromagnetic domain walls in SmB6 surface states, cited to connect the observed hysteretic magnetotransport to ferromagnetic surface domains.","marker":"[27]"},{"why":"Shows perfect Andreev reflection in an Au-SmB6/YB6 structure attributed to topological surface states, supporting the premise that the SmB6 surface alone can carry the proximity supercurrent.","marker":"[30]"},{"why":"Documents proximity-induced superconductivity in SmB6 with a high critical temperature, providing the comparison that the observed 3–6 K supercurrent extends.","marker":"[33]"},{"why":"Supplies the precedent of anomalous, hysteretic Josephson diffraction patterns from multiple dynamical domains in Sr2RuO4, used to argue the shifted central peak indicates magnetic flux rather than artifact.","marker":"[35]"},{"why":"Shows that superconductor-ferromagnet-superconductor junctions produce similar anomalous Fraunhofer patterns with maximal supercurrent at nonzero field, the closest analogue for the SmB6 data.","marker":"[36]"}],"fun_headline_variants":["Topological Kondo insulator weak links host exotic Josephson supercurrent","SmB6 junctions reveal surface magnetism alongside supercurrent","Anomalous Fraunhofer pattern hints at ferromagnetic topological surface","Nb-SmB6-Nb devices show Shapiro steps and magnetic hysteresis","Kondo breakdown drives ferromagnetic order in superconducting junctions"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the anomalous diffraction patterns and hysteresis are caused by ferromagnetic surface states generated by Kondo breakdown, since the paper offers no direct magnetization measurement and does not exclude the geometric, self-field, flux-trapping, magnetocaloric, or impurity-scattering alternatives it lists.","fun_headline_variants_meta":{"raw":{"variants":["Topological Kondo insulator weak links host exotic Josephson supercurrent","SmB6 junctions reveal surface magnetism alongside supercurrent","Anomalous Fraunhofer pattern hints at ferromagnetic topological surface","Nb-SmB6-Nb devices show Shapiro steps and magnetic hysteresis","Kondo breakdown drives ferromagnetic order in superconducting junctions"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000196,"raw_usage":{"total_tokens":1435,"prompt_tokens":1091,"completion_tokens":344,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":707,"completion_tokens_details":{"reasoning_tokens":258}},"tokens_in":707,"tokens_out":344,"duration_ms":3644,"temperature":1.0,"reasoning_tokens":258,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:47:07.966291+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive check: place a polished SmB6 crystal of the same provenance in a SQUID magnetometer or under a magnetic force microscope and cycle the out-of-plane field through ±150 mT at base temperature. The Kondo-breakdown picture predicts a hysteretic remanent surface moment large enough to shift the junction's central diffraction peak by a few mT; if no such surface magnetization is resolved, the anomalous Fraunhofer patterns must instead be assigned to geometric asymmetry, self-fields, or flux redistribution.","supporting_citations":[{"cited_title":"Nakajima, P","cited_arxiv_id":null,"evidence_quote":"Establishes prior evidence for ferromagnetic domain walls in SmB6 surface states, cited to connect the observed hysteretic magnetotransport to ferromagnetic surface domains."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows perfect Andreev reflection in an Au-SmB6/YB6 structure attributed to topological surface states, supporting the premise that the SmB6 surface alone can carry the proximity supercurrent."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Documents proximity-induced superconductivity in SmB6 with a high critical temperature, providing the comparison that the observed 3–6 K supercurrent extends."},{"cited_title":"Kidwingira, J","cited_arxiv_id":null,"evidence_quote":"Supplies the precedent of anomalous, hysteretic Josephson diffraction patterns from multiple dynamical domains in Sr2RuO4, used to argue the shifted central peak indicates magnetic flux rather than artifact."}],"review_version":1}