{"id":"ceebc37b-aab8-4be5-b3e4-f3cc1bc11d3d","arxiv_id":"1908.04111","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":3,"one_line_summary":"The authors claim to observe (hc/2e)/m periodic magnetoresistance oscillations in an aluminum ring for m up to 20 and attribute them to multiple Andreev reflection, but the evidence rests on Fourier harmonics and qualitative fits.","lead":"This paper reports that a tiny superconducting aluminum ring produces electrical-resistance oscillations with periods that are fractions of the usual magnetic flux quantum when pushed with high current at low temperature. It suggests that repeated electron reflections at superconducting boundaries multiply the effective charge of Cooper pairs, but the supporting data analysis is mostly curve fitting.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Fourier peaks at m f0 are exactly what a non-sinusoidal Phi0-periodic V(B) produces; without a quantitative baseline the fractional-period claim collapses.","rationale":"The reader's weakest assumption is the same as the one I regard as load-bearing: the Fourier peaks at m f0 are the expected harmonic content of a non-sinusoidal waveform with the fundamental Little-Parks period, and the paper never supplies a quantitative baseline. I therefore agree with the reader's identification of the problem. I also note the paper's self-stated limitations: the fitting is explicitly qualitative, the cause of the two dissipative states is deferred, and the supporting theory is admitted to be valid only for shorter samples. These limitations reinforce but are not the core issue. The core issue is that a finite Fourier sum at frequencies m f0, with arbitrary amplitudes and phases, is mathematically identical to an anharmonic periodic function of period dB0; hence the observation of many harmonics cannot discriminate between the two interpretations. The paper's phrase 'not a consequence of the inharmonicity' is a bare assertion, and its replacement of a statistical test by hand-picked piecewise fits from a very small set of field windows makes the evidence especially vulnerable. The central claim could still be true; the problem is that the analysis as presented does not establish it. Because the reader's verdict is REJECT and my concern supports that verdict without adding a different one, the verdict is unchanged. A single quantitative baseline, synthesizing the expected harmonic spectrum from the measured V(I) and the Phi0-periodic Little-Parks critical current, would settle the question.","tokens_in":15173,"tokens_out":6048,"duration_ms":65684,"concrete_test":"Using the raw V1(B) trace (curve 3 of Fig. 3), the low-current Little-Parks R(B) data (inset of Fig. 2), and the measured V(I) characteristic (right inset of Fig. 1), synthesize the expected V_synth(B) at I_dc = 9.8 uA and T = 1.282 K from a standard current-biased ring model in which I_c(B) is strictly periodic with period Phi0 and the resistance is set by the measured V(I) nonlinearity. Compute the FFT of V_synth(B) over the same 6-12 G window and compare the amplitudes and phases of the m f0 peaks with Fig. 4. If the synthesized harmonic spectrum matches the observed peaks within the reproducibility of the repeated experimental traces, the fractional-period claim is unnecessary and should be withdrawn; if it cannot reproduce them, the fractional-period interpretation gains concrete support.","verdict_should_be":"UNCHANGED","load_bearing_attack":"FFT peaks at integer multiples m f0 are the Fourier decomposition of any non-sinusoidal waveform with fundamental period dB0, so the spectra in Figs. 4-6 cannot by themselves establish independent (hc/2e)/m periods. The paper's assertion that 'inharmonicity of oscillations is believed to make a very low contribution to the higher harmonics' is never quantified; no control is given for the harmonics generated by the measured nonlinear V(I) characteristic or by a finite FFT window of length one or three fundamental periods. The piecewise fits in Figs. 5-8 do not resolve the ambiguity: they are sums of sinusoids at m f0 with phases restricted to multiples of pi/4, and the text explicitly says the fitting is qualitative and 'has nothing to do with the theoretical description.' Because both descriptions are mathematically the same finite Fourier sum, the observation of nearly equal harmonic amplitudes is exactly the signature of an anharmonic Phi0-periodic oscillation, not evidence of fractional periods. The subsequent inference from Phi0/m to effective charge 2em is definitional and would be sound only if the fractional-period observation survived a baseline test.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports magnetoresistance measurements on a superconducting aluminum mesoscopic ring at currents above the critical current and temperatures slightly below Tc. The authors claim to observe quantum oscillations with periods Phi0/m for integer m up to 20, which they interpret as an effective superconducting charge e* = 2em produced by multiple Andreev reflection in a phase-slip center or SNS junction. Evidence for the fractional periods is drawn from Fourier spectra of short field intervals (Figs. 4–6) and from piecewise sinusoidal fits of the form s + p sum_k a_k sin(2π m f0 B + phi_k) (Figs. 5–8). The paper also reports hysteresis in V(B) and two dissipative states, but the central novelty is the fractional-period claim.","tokens_in":15327,"tokens_out":4633,"duration_ms":47799,"significance":"If the fractional-period claim were correct, this would be a striking nonequilibrium effect, extending Little-Parks physics to strongly driven superconducting rings and implying a Cooper-pair charge enhanced by a factor m. The data and analysis, however, do not rule out the standard interpretation that the spectral peaks at multiples of f0 are ordinary higher harmonics of a non-sinusoidal waveform with the fundamental Little-Parks period. The paper provides no quantitative baseline for the harmonic content, no statistical tests, and the fitting is explicitly qualitative. The claim as presented is therefore not supported, and the effective-charge interpretation is definitional rather than a consequence of an independently established fractional-period observation.","major_comments":[{"comment":"The FFT spectra shown in Figs. 4–6 contain peaks at frequencies f_m = m f0. For any periodic signal with fundamental period dB0 = 1/f0, a nonsinusoidal waveform produces exactly such peaks as its Fourier harmonics. The statement in the text that \"Inharmonicity of oscillations is believed to make a very low contribution to the higher harmonics\" is an assertion with no quantitative support. No control is presented for the harmonic content generated by the measured nonlinear V(I) characteristic, and no estimate is given for spectral leakage from the finite field window, which has length dB0 or 3 dB0. Without such a baseline, these spectra cannot distinguish independent (hc/2e)/m oscillations from the ordinary Fourier decomposition of a distorted Phi0-periodic oscillation. This ambiguity is load-bearing because the entire fractional-period claim rests on these peaks.","section":"Results and Discussion, Fourier analysis (Figs. 4–6)"},{"comment":"The fitting functions V2, V3, Fk, and Ak are sums of sinusoids at frequencies m f0 with phases restricted to multiples of pi/4. The text states explicitly that \"The fitting has nothing to do with the theoretical description of the V1(B) oscillations.\" Since any periodic function can be approximated by such a finite Fourier sum, the visual agreement between the fits and the data does not provide additional evidence for independent fractional periods. No residuals, goodness-of-fit metrics, or comparison against the null model of a single anharmonic Phi0-periodic waveform are presented. The conclusion that the data \"really has a certain set of different oscillation periods\" is therefore not justified by the fitting procedure.","section":"Results and Discussion, fitting of V1(B) (Figs. 5–8)"},{"comment":"The central evidence is based on one measured curve (V1, curve 3 of Fig. 3) and a small number of field intervals. Other measurements are described in the text as \"not presented here\" or \"not shown,\" and no raw traces with error bars, run-to-run statistics, or significance tests are given. Given the harmonic ambiguity, the reproducibility of the claimed peaks is essential. The sparse, hand-picked data presentation does not allow the reader to assess whether the spectral features are robust or merely noise and windowing artifacts.","section":"Results and Discussion, experimental data (Figs. 2–3)"},{"comment":"The inference from a period Phi0/m to an effective charge e* = 2em is definitional through the fluxoid quantization relation Phi0 = hc/e*. If a true fractional period were established, this inference would follow, but it would not independently confirm the physical mechanism. Because the evidence for the fractional periods is not established (comments above), the central claim of an increased Cooper-pair charge is unsupported. The multiple Andreev reflection mechanism is offered only as a possible cause; no model links the observed set of m values to the MAR process, and the estimate 2Δ/eV ≈ 10 is an order-of-magnitude ratio that does not predict the observed distribution or amplitudes of harmonics.","section":"Conclusion, interpretation"}],"minor_comments":[{"comment":"The notation dB0 and dB1,2 is introduced without a clear definition of the subscript convention; the authors should define these quantities explicitly at first use to avoid confusion between the fundamental period and the field-window length.","section":"Results and Discussion, first paragraph after Fig. 1"},{"comment":"The formula \"Δ(T) = 3.07 kTc Δ(0)(1 − T/Tc)^1/2\" appears dimensionally inconsistent and is likely a typographical error; the authors should replace it with the standard BCS temperature dependence, for example Δ(T) ≈ 1.76 kTc (1 − T/Tc)^1/2 near Tc.","section":"Results and Discussion, gap expression"},{"comment":"Reference [7] contains a typographical error: \"John Willey and Sons\" should be \"John Wiley and Sons.\"","section":"References"},{"comment":"The phrase \"modified to the sum of harmonic (hc/2e)/m periodic oscillations\" is ambiguous; it should clarify whether the fundamental period itself changes or whether additional periodic components appear on top of the Phi0-periodic signal.","section":"Abstract and Conclusion"},{"comment":"The caption of Fig. 2 does not adequately distinguish the curves labeled 1a, 1b, and 2; in particular, the relationship between the inset curves and the main panel should be stated more explicitly.","section":"Fig. 2 caption"}],"recommendation":"reject","confidential_remarks":"The manuscript reports a striking effect but the evidence is undermined by the absence of a harmonic baseline and by the explicitly qualitative nature of the fits. I would encourage the authors, if they resubmit, to provide raw data with error bars, a quantitative null-hypothesis test (e.g., fitting an anharmonic Phi0-periodic waveform and comparing residuals with the rich harmonic model), and a clear statement of how many independent measurements support the fractional-period claim."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: the paper claims to have seen (hc/2e)/m magnetoresistance oscillations in a single aluminum ring, with m up to 20, and reads that as an effective Cooper-pair charge of 2em. That is a big claim, and the evidence doesn't support it. The FFT spectra show peaks at integer multiples of the fundamental frequency f0, which is exactly what you get from any non-sinusoidal oscillation with the ordinary Little-Parks period. The authors dismiss the harmonic-distortion explanation with one sentence—'inharmonicity of oscillations is believed to make a very low contribution'—and never quantify it. There's no control, no synthetic signal built from the measured V(I) nonlinearity, no comparison against a simply distorted Phi0-periodic waveform. The fits in Figs. 5-8 are sums of sinusoids at m f0 with phases restricted to multiples of pi/4; that's just a truncated Fourier series of a periodic function, so it can't resolve the ambiguity. The text itself says the fitting is qualitative and 'has nothing to do with the theoretical description.' So the central observation, period Phi0/m, is not established.\n\nWhat is genuinely new here is the experimental situation: a dirty Al ring (radius ~2 micron, wall ~0.27 micron) at low temperature and high dc current, showing reproducible V(B) oscillations with strong higher harmonics and a field-sweep hysteresis. That is worth knowing about. The authors are also honest about several limitations: the NMR theory they cite is valid only for short samples, the cause of the two dissipative states is deferred, and they admit the fits are not a theoretical description. The sample characterization is reasonably careful, and the citation list covers the relevant hc/4e literature.\n\nWhere the paper falls short is in the leap from 'peaks at m f0' to 'independent fractional periods.' That leap is definitional if the observation were real—flux period hc/e* would imply e*=2em—but the observation is the very thing in question. The MAR interpretation is speculative and tied to the data only through a rough ratio 2Delta/eVn ~10.\n\nFor a specialist in mesoscopic superconductivity, this is a useful cautionary tale about FFT interpretation. It is not a demonstration of fractional flux quantization. But I wouldn't desk-reject it either: the claim is important, the data may contain something real, and the authors are not hiding their tracks. A serious referee should send it back and ask for the raw traces, error bars, and a quantitative baseline comparing the observed harmonic amplitudes against those of an ordinary anharmonic Phi0-periodic signal. Until then, the conclusion is unsupported.","headline":"Claims of (hc/2e)/m oscillations up to m=20 rest on Fourier peaks that are indistinguishable from harmonics of an anharmonic Phi0-periodic signal; the paper lacks the baseline needed to support its central claim.","tokens_in":15963,"tokens_out":3649,"would_cite":false,"duration_ms":36194,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["74.25.N-","74.78.Na","74.45.+c"],"model":"deepseek-v4-flash","headline":"Under high current and low temperature, an aluminum ring develops magnetoresistance oscillations with periods as short as (hc/2e)/20, which the authors attribute to multiple Andreev reflections multiplying the effective Cooper-pair charge.","keywords":["fractional flux period oscillations","Little-Parks effect","multiple Andreev reflection","superconducting aluminum ring","phase slip center","effective Cooper-pair charge","negative magnetoresistance","mesoscopic superconductivity"],"falsifier":"Compute the FFT of a model waveform that is strictly periodic with period $\\Phi_0$ but has the sharply peaked shape of the low-current Little-Parks curve, using the same field window and sampling; if the model produces comparable peaks at $m f_0$ for $m=2\\text{–}20$, then the data do not require fractional periods.","tokens_in":14853,"feed_emoji":"🧲","tokens_out":12192,"duration_ms":116833,"temperature":0.7,"pith_summary":"The paper reports an experimental claim about a single mesoscopic aluminum ring: when the temperature is lowered and the dc current is raised, the usual Little-Parks magnetoresistance oscillations with period $\\Phi_0 = hc/2e$ are replaced by a voltage waveform whose frequency spectrum contains comparable peaks at $m$ times the fundamental frequency, for $m$ up to 20. The authors read each peak as an independent oscillation with period $\\Phi_0/m = (hc/2e)/m$, and interpret the shorter period as an increase in the effective charge of the superconducting pairs from $2e$ to $2em$. They propose multiple Andreev reflection inside a phase-slip center or SNS junction formed in the ring as the mechanism that carries $m$ pairs per event. If the interpretation is right, a homogeneous ring without fabricated weak links can exhibit fractional flux-period oscillations under nonequilibrium bias, a regime that has not been described theoretically. The paper also notes that the harmonic content from waveform distortion is believed to be very low, but gives no quantitative baseline.","feed_headline":"Superconducting ring shows flux oscillations at 1/20th usual period","feed_subtitle":"High current gives an aluminum ring voltage periods down to (hc/2e)/20, hinting Andreev reflections multiply pair charge","key_machinery":"The load-bearing object is the Fourier spectrum of the magnetoresistance waveform. By choosing field windows with $dB_{1,2}=j\\,dB_0$ ($j=1$ or 3), the authors force spectral-window artifacts to land at known multiples of the fundamental frequency, so peaks at $f_m=m f_0$ can be assigned to physical periods $\\Phi_0/m$. The explanation they attach to those peaks is multiple Andreev reflection (MAR): in a diffusive normal region between superconducting banks, a quasiparticle bounces back and forth as electron and hole, and after $n$ reflections it transfers $m=n/2$ (or $n/2-1$) Cooper pairs, making the supercurrent respond as if the pair charge were $2em$. The mechanism is supported by the condition that the inelastic scattering length $2\\lambda_Q$ is much larger than the normal-region length $2\\xi(T)$, so many reflections can occur before decoherence.","core_discovery":"The central claim is that the measured $V(B)$ curves, recorded at $I_{dc}=7.5\\text{–}11\\,\\mu\\text{A}$ and $T\\approx1.28\\,\\text{K}$ in fields of order 6–12 G, contain real oscillations at the fractional flux periods $\\Phi_0/m$ for $m=2$ up to 20. The evidence is a set of FFT spectra taken over field windows deliberately chosen to be exact integer multiples of the fundamental period $dB_0=\\Phi_0/S_{\\rm eff}$; in those spectra the peaks at $f_m=m f_0$ have amplitudes close to that of $f_0$. The authors fit short segments of the waveform with sums of sines at these frequencies and conclude that the ring is not merely showing a distorted $\\Phi_0$-periodic response, but a superposition of fractional-period components. They state that this is the first study of such $(hc/2e)/m$ oscillations and that the period reduction by a factor $m$ can be interpreted as an effective Cooper-pair charge $e^*=2em$ generated by multiple Andreev reflections in a diffusive phase-slip center or SNS junction.","pith_inferences":["A decisive check the paper does not perform is to compare the measured harmonic amplitudes against those of a distorted but strictly $\\Phi_0$-periodic waveform; if a realistic saw-toothed Little-Parks curve reproduces the near-flat spectrum up to $m=20$, the fractional-period interpretation would not be required.","If MAR is the cause, the distribution of $m$ values should shift systematically when $I_{dc}$ changes, because the instantaneous voltage sets the number of allowed reflections; measuring the same ring across a current ramp would give a direct test.","The claimed effect sits naturally next to the earlier $hc/4e$ observations in hybrid systems; a useful extension would be to look for temperature and field boundaries where the system switches between pure $\\Phi_0$, $\\Phi_0/2$, and higher-order periodicity."],"forward_implications":["If the fractional periods are real, flux quantization in a driven superconducting ring is not locked to a single quantum $\\Phi_0$; the circulating current can respond at periods $\\Phi_0/2$ through $\\Phi_0/20$.","The same field-window FFT procedure can be applied to other rings, cylinders, or SNS devices to search for $(hc/2e)/m$ oscillations in different materials and geometries.","The MAR mechanism implies the dominant values of $m$ and the maximum $m$ should depend on bias current and temperature through the ratio $2\\Delta(T,B)/(eV)$, making the oscillation spectrum electrically tunable.","The observation in a homogeneous ring, with no intentional weak links, means a current-induced phase-slip center or SNS junction can act as the Andreev reflector, extending MAR physics from fabricated junctions to self-formed nonequilibrium regions."],"supporting_citations":[{"why":"Supplies the baseline Little-Parks effect that the paper claims is modified to fractional periods.","marker":"[2]"},{"why":"Reports earlier anomalous magnetoresistance and loss of hc/2e periodicity in superconducting loops, the effect this work extends.","marker":"[3]"},{"why":"Reports hc/4e oscillations in hybrid SNS rings attributed to multiple Andreev reflection, the closest precedent for fractionally periodic oscillations.","marker":"[6]"},{"why":"Gives the hot-quasiparticle theory used to explain the negative-magnetoresistance regions where the fractional oscillations appear.","marker":"[14]"},{"why":"Foundational multiple-Andreev-reflection theory for SNS junctions that supplies the m-pair transfer mechanism.","marker":"[15]"},{"why":"Extends multiple-Andreev-reflection theory to diffusive junctions and gives the inelastic-length condition used to justify many reflections.","marker":"[16]"},{"why":"Adds the diffusive MAR theory used to argue that many reflections survive in disordered aluminum.","marker":"[17]"},{"why":"Earlier measurement by the same group of MAR subharmonic voltage plateaus with up to 32 reflections, supporting a large m.","marker":"[18]"}],"fun_headline_variants":["Superconducting ring reveals flux oscillations at 1/20th period","Fractional flux periods down to 1/20th observed in ring","Andreev reflections yield 1/20th flux quantum oscillations","Ring shows flux periods shrunk by factor 20 via Andreev","First sighting of (hc/2e)/m oscillations in a ring"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central claim depends on treating the evenly spaced peaks in the frequency plot as independent physical oscillation periods; if those peaks are only the ordinary overtones of a non-sinusoidal $\\Phi_0$-periodic signal, the fractional-period interpretation collapses.","fun_headline_variants_meta":{"raw":{"variants":["Superconducting ring reveals flux oscillations at 1/20th period","Fractional flux periods down to 1/20th observed in ring","Andreev reflections yield 1/20th flux quantum oscillations","Ring shows flux periods shrunk by factor 20 via Andreev","First sighting of (hc/2e)/m oscillations in a ring"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000165,"raw_usage":{"total_tokens":1201,"prompt_tokens":847,"completion_tokens":354,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":463,"completion_tokens_details":{"reasoning_tokens":258}},"tokens_in":463,"tokens_out":354,"duration_ms":3687,"temperature":1.0,"reasoning_tokens":258,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:52:10.733632+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the FFT of a model waveform that is strictly periodic with period $\\Phi_0$ but has the sharply peaked shape of the low-current Little-Parks curve, using the same field window and sampling; if the model produces comparable peaks at $m f_0$ for $m=2\\text{–}20$, then the data do not require fractional periods.","supporting_citations":[{"cited_title":"Santhanam, C","cited_arxiv_id":null,"evidence_quote":"Reports earlier anomalous magnetoresistance and loss of hc/2e periodicity in superconducting loops, the effect this work extends."},{"cited_title":"The values of f0 found from the period of the Little-Parks type oscillations (the inset of Fig","cited_arxiv_id":null,"evidence_quote":"Reports hc/4e oscillations in hybrid SNS rings attributed to multiple Andreev reflection, the closest precedent for fractionally periodic oscillations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the hot-quasiparticle theory used to explain the negative-magnetoresistance regions where the fractional oscillations appear."},{"cited_title":"Vloeberghs, V","cited_arxiv_id":null,"evidence_quote":"Foundational multiple-Andreev-reflection theory for SNS junctions that supplies the m-pair transfer mechanism."},{"cited_title":"Zadorozhny and Y","cited_arxiv_id":null,"evidence_quote":"Extends multiple-Andreev-reflection theory to diffusive junctions and gives the inelastic-length condition used to justify many reflections."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Adds the diffusive MAR theory used to argue that many reflections survive in disordered aluminum."}],"review_version":1}