{"id":"6b7cc967-629d-4d0e-8070-d405bd7b461f","arxiv_id":"2412.08525","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"In a strongly interacting superfluid junction, local particle loss restores thermal and spin conductances toward the universal quantized values of a non-interacting 1D Fermi gas.","lead":"Localized particle loss in a superfluid junction markedly increases thermal and spin transport, which approach the universal quantized conductance values of a non-interacting 1D Fermi gas. This suggests that engineered dissipation can switch on heat and spin currents in strongly interacting superfluids.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Spin conductance saturation is quantitatively indistinguishable from a pure-loss artifact: the zero-transport loss-only null predicts 1/tau_sigma ~ gamma_N for both dissipation mechanisms, and the S4.B dismissal is a plausibility argument, not a control.","rationale":"The paper's central claim is that local dissipation enhances both thermal and spin conductances by one to two orders of magnitude and that both approach the universal non-interacting 1D values. Reading in good faith, the thermal half is supported by several independent rests: the exponential-timescale extraction that does not rely on the phenomenological model, the physical upper bound in S4.A, the collapse across three transverse confinements (Fig. S3), and the explicit comparison showing the loss-only apparent G_T lies below the data (Fig. S4). The spin half has no equivalent control. The zero-transport loss-only null quantitatively reproduces the headline observable: for pairwise loss, symmetric per-capita reservoir loss gives Delta M(t)/N_0 proportional to e^{-gamma_N t} exactly; for spin-imbalanced loss, the prepared state Delta N_up(0) = -Delta N_down(0) yields an apparent decay rate (gamma_up+gamma_down)/2 = gamma_N. Since S4.B reports gamma_N ~ 1/tau_sigma, the extraction G_sigma = chi/(4 tau_sigma) is consistent with chi*gamma_N/4, i.e., with zero genuine spin transport. The artifact hypothesis also inherits the saturation of gamma_N with beam power and the cross-mechanism collapse, so it explains every qualitative feature of Fig. 4(c) and its inset. The authors' dismissal is a plausibility argument rather than a measurement, and their own scenario 2 predicts precisely the equality they need to exclude. The main-text sentence claiming the conductances are 'much higher than the apparent transport purely arising from atom loss' is backed for G_T but not for G_sigma, where the S4.B analysis yields equality. This is the single most load-bearing weakness of the central claim: if the control fails, the spin conductance saturation and the two-mechanism universality collapse, leaving a thermal-only result. The concern is falsifiable with data already in hand, so the appropriate disposition is the reader's CONDITIONAL verdict: require the spin-resolved reanalysis (or a displaced-beam control) before the spin claim is accepted. I therefore do not change the reader's verdict.","tokens_in":19415,"tokens_out":21621,"duration_ms":223732,"concrete_test":"Reanalyze the existing spin-resolved data (the absorption images already yield N_Lup, N_Ldown, N_Rup, N_Rdown separately; Figs. 3 and S2 show only the combined Delta M) under a zero-transport null. For pairwise-loss runs, compute the fractional polarization imbalance Delta M(t)/M_total(t): symmetric per-capita loss conserves this ratio, while genuine spin transport drives it to zero; test whether it is constant within errors. For spin-imbalanced runs, fit Delta N_up(t) and Delta N_down(t) to the loss-only rates gamma_up = 0.83 gamma_down and gamma_down (fixed by the measured N(t) and the calibrated 0.83(4) branch ratio) and test whether a common, faster transport rate is required. If the fractional imbalance is constant, or if each spin imbalance tracks its own loss rate, the reported G_sigma is a loss artifact and the saturation claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The spin half of the central claim — that G_sigma saturates at the quantum-limited value 2n_m/h (Fig. 4c) — rests on the assumption that the exponential decay of Delta M(t) (Figs. 3b,c; Eq. S3) is genuine spin transport and not atom loss. Supplement S4.B states that gamma_N is 'very close to 1/tau_sigma', and the zero-transport null model predicts exactly that equality. For pairwise loss, symmetric per-capita loss from each reservoir (the standard model for a beam centered in the channel, cf. refs. 12,13) gives M_L(t)=M_L(0)e^{-gamma_N t} and M_R(t)=M_R(0)e^{-gamma_N t}, hence Delta M(t)/N_0 = (Delta M(0)/N_0)e^{-gamma_N t} with zero transport. For spin-imbalanced loss with gamma_up = 0.83 gamma_down and the prepared state Delta N_up(0) = -Delta N_down(0), pure loss gives Delta M(t) proportional to e^{-gamma_up t}+e^{-gamma_down t}, with fitted rate approximately (gamma_up+gamma_down)/2 = gamma_N. Losses alone quantitatively reproduce the headline observable. They also reproduce the qualitative 'saturation' evidence: G_sigma = chi*gamma_N/4 is linear in the loss current I_Gamma, so the rise and flattening of Fig. 4c and the cross-mechanism collapse in the inset are inherited from the saturation of gamma_N (Fig. 4a). The S4.B dismissal (channel longer than the beam, so pairs come from the same reservoir) is an unquantified plausibility claim — no estimate of intra- vs inter-reservoir pair fraction — and it sits awkwardly with the delocalized pair correlations of a superfluid junction. More seriously, S4.B's scenario 2 predicts Delta M(t)=Delta M_0 e^{-gamma_N t}, i.e., a loss-only G_sigma equal to the reported one, contradicting the main-text assertion that the conductances are 'much higher than the apparent transport purely arising from atom loss': that statement is backed for G_T (black curves, Fig. S4) but no equivalent loss-only comparison is shown for G_sigma. The Fig.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports measurements of thermal and spin transport through a quasi-1D junction connecting two unitary Fermi superfluids, with local atom losses engineered inside the channel. The authors observe that dissipation increases the thermal and spin conductances by one to two orders of magnitude and that at strong dissipation both appear to approach the non-interacting quantized values G_T,0 = 2n_m pi^2 k_B^2 T/(3h) and G_sigma,0 = 2n_m/h. Thermal conductance is extracted from a nonlinear thermoelectric model fit and cross-checked by an independent upper bound from the diffusive timescale; spin conductance is extracted from exponential fits to the magnetization imbalance decay. Two dissipation mechanisms, spin-imbalanced and pairwise losses, are compared. The paper concludes that local dissipation can restore universal quantized heat and spin transport in a strongly interacting superfluid junction.","tokens_in":19823,"tokens_out":7178,"duration_ms":79254,"significance":"If established, this is a striking experimental result: dissipation-induced recovery of quantized transport in a strongly correlated system, with implications for open-system transport and thermoelectric control. The thermal conductance claim is supported by multiple cross-checks, including a model-independent timescale bound, a lower bound that excludes pure-loss artifacts, and collapse of the data with mode number across three transverse confinements. The comparison of two dissipation mechanisms is also a useful experimental control. However, the spin conductance claim currently rests on a single exponential decay whose fitted rate is close to the atom-loss rate, and the paper's own Supplement S4.B supplies only a plausibility argument rather than a quantitative exclusion of the loss-induced null model. The central spin claim is therefore not yet established to the standard of the abstract, although the deficiency is addressable with additional measurements.","major_comments":[{"comment":"The spin conductance is extracted from an exponential fit to Delta M(t), with G_sigma = chi/(4 tau_sigma). Supplement S4.B states that gamma_N is 'very close to 1/tau_sigma' and considers a zero-transport null model in which pure losses give Delta M(t) proportional to e^{-gamma_N t}. This null model quantitatively reproduces both the measured decay and its saturation with dissipation power: since G_sigma is proportional to 1/tau_sigma, and 1/tau_sigma is close to gamma_N, the rise and flattening of G_sigma in Fig. 4(c) are inherited from the saturation of gamma_N in Fig. 4(a). For spin-imbalanced loss the same degeneracy holds: with Gamma_up = 0.83 Gamma_down, pure loss gives a fitted decay rate near (Gamma_up + Gamma_down)/2 = gamma_N. The dismissal of the null model is a plausibility argument (the beam is smaller than the channel, so pairs come from the same reservoir) with no quantitative estimate of the intra- versus inter-reservoir pair fraction. The paper needs either a control measurement that breaks the degeneracy between transport and loss, for example a dissipation beam displaced from the channel or a spin-polarized loss calibration, or a quantitative upper bound on the cross-reservoir pair fraction that places the apparent loss-induced G_sigma below the reported values.","section":"Supplement S4.B; Eqs. (S3), (S4)"},{"comment":"The inset of Fig. 4(c) presents the cross-mechanism collapse of G_sigma versus I_Gamma as evidence that the two dissipation mechanisms produce the same spin transport. This collapse is not a diagnostic for real spin transport: in the pure-loss null model, G_sigma = chi gamma_N/4 is linear in I_Gamma (with gamma_N = I_Gamma/(N_0/2)), so any quantity proportional to gamma_N will collapse when plotted against I_Gamma. The same concern applies to the thermal conductance collapse in Fig. S3(f), although for G_T the S4.A lower-bound calculation independently excludes pure losses. Please plot the spin data together with the null-model prediction and report residuals, not only the collapse.","section":"Fig. 4(c) inset"},{"comment":"The sentence 'These conductances are much higher than the apparent transport purely arising from atom loss [42]' is supported for the thermal conductance by the black curves in S4.A, but it is not supported for the spin conductance; S4.B in fact shows that the loss-only apparent spin conductance is comparable to the reported values. The abstract and conclusion should be adjusted to state that the spin saturation is tentative pending a control measurement, or a quantitative exclusion of the null model should be added to the manuscript.","section":"Main text, Fig. 4(c) discussion"}],"minor_comments":[{"comment":"In the main text, the quantities I_N and I_S are called 'apparent currents' only in Supplement S2.C; please define this term and its relation to the measured reservoir imbalances at first use in the main text, since the closed-system model is being applied to a lossy system.","section":"Main text, Experimental setup"},{"comment":"The normalization G_T,0 = 2 n_m pi^2 k_B^2 T/(3h) uses a temperature T; please specify whether T is the initial reservoir temperature, the final temperature, or an average over the transport time, and propagate the corresponding uncertainty into the normalized conductance.","section":"S2.C and Fig. 4(c)"},{"comment":"Supplement S1.A reports that n_m decreases from about 3.1 to about 2.7 during transport; please state how this drift affects the normalized conductances in Fig. 4(c) and Fig. S3(d), or justify that it is negligible for the saturation claim.","section":"Supplement S1.A"},{"comment":"The upper bound in Eq. (S11) uses the non-dissipative value of alpha_c, while Fig. S3(a) shows that alpha_c changes with dissipation; please state this choice explicitly in the main text or in the caption of Fig. S4.","section":"Supplement S4.A, Eq. (S11)"},{"comment":"There is a typo in Supplement S3: 'more destructive to the the superfluid order' should read 'to the superfluid order'.","section":"Supplement S3"}],"recommendation":"major_revision","confidential_remarks":"The spin-loss degeneracy is the main risk to the central claim. The authors' own Supplement S4.B acknowledges gamma_N close to 1/tau_sigma, so the burden is on them to provide a control measurement or a quantitative bound. If they can supply such a control, the paper is a strong candidate for publication; the thermal side of the claim appears well supported. I do not recommend rejection at this stage because the issue is experimentally addressable within the manuscript's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. The thermal half of this paper is genuinely solid. Local particle loss raises the thermal conductance of the superfluid junction by nearly two orders of magnitude, toward the non-interacting quantized value, and the authors back it with independent checks: an upper bound from the relaxation timescale, collapse when normalized by mode number, and an explicit loss-only comparison (black curves, Fig. S4) that sits well below the data. That part is publishable as is, and it is the genuinely new observation here — dissipation opening a thermal channel in a system where pairing normally suppresses it.\n\nThe spin half does not hold up. The stress-test note is correct: a zero-transport, pure-loss null model predicts 1/tau_sigma = gamma_N, and that is precisely what the data show. S4.B admits gamma_N is 'very close to 1/tau_sigma' and then dismisses the loss-induced alternative with the argument that the beam is smaller than the channel, so pairs probably come from the same reservoir. That is an unquantified plausibility claim, not a control. The main text asserts the conductances are 'much higher than the apparent transport purely arising from atom loss,' but that statement is backed for G_T only; no equivalent loss-only curve is shown for G_sigma. Since G_sigma = chi*gamma_N/4 is linear in the loss current, the rise and saturation in Fig. 4(c) and the cross-mechanism collapse in the inset can be inherited directly from gamma_N(P_Gamma). The spin conductance saturation is quantitatively indistinguishable from a pure-loss artifact as presented.\n\nOther soft spots are minor by comparison. The offset subtraction for the strongest dissipation datasets is post hoc but disclosed. Fitting a closed-system model from Ref. [22] to a dissipative system introduces real circularity, but the thermal result survives without the model via the simple exponential relaxation fit. The reservoirs staying superfluid at the weakest dissipation is a useful check. The paper is honest about its limitations, which counts for something.\n\nWho this is for: the cold-atom mesoscopic transport community. The thermal result is worth citing on its own. The spin claim needs a control — at minimum a loss-only comparison curve like the one drawn for G_T, a direct measurement of the loss asymmetry, or a beam placed off the channel. Without that, the saturation of spin conductance is not established.\n\nRecommendation: yes, send it to serious review. The thermal finding earns referee time. But the referee must require the spin control before the headline claim is accepted, and the published version should present the spin result as conditional on that check.","headline":"The thermal conductance enhancement is real and well checked; the spin conductance saturation is quantitatively indistinguishable from a pure-loss artifact, and the paper's dismissal is a plausibility argument, not a control.","tokens_in":20446,"tokens_out":4947,"would_cite":true,"duration_ms":44178,"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":"The paper reports that local dissipation in a superfluid junction raises thermal and spin conductances by one to two orders of magnitude, saturating at the universal quantized values of a non-interacting one-dimensional Fermi gas.","keywords":["superfluid junction","quantum point contact","dissipative transport","thermal conductance","spin conductance","unitary Fermi gas","quantized conductance","open quantum systems"],"falsifier":"Measure the magnetization imbalance decay with the dissipation beam positioned just outside the channel instead of inside it; if the decay rate $\\gamma_N$ still matches $1/\\tau_\\sigma$, then loss-induced apparent spin current, not junction transport, would explain the reported spin conductance. A second check would compare the spin conductance inferred from the exponential magnetization decay with a direct measurement of spin-resolved currents through the junction.","tokens_in":19197,"feed_emoji":"⚛️","tokens_out":4941,"duration_ms":54686,"temperature":0.7,"pith_summary":"The paper reports an experiment on a one-dimensional junction connecting two strongly interacting fermionic superfluids, with a focused beam removing atoms from the junction while particle, entropy, and spin currents are measured. It claims that local dissipation, of either spin-imbalanced or pairwise type, raises thermal and spin conductances by one to two orders of magnitude, and that at strong dissipation both conductances approach the quantum-limited values of a non-interacting ballistic one-dimensional junction, $G_{T,0}=2n_m\\pi^2 k_B^2 T/(3h)$ and $G_{\\sigma,0}=2n_m/h$. If true, dissipation does not merely degrade coherent transport: it breaks the pairs that the superfluid gap protects and opens diffusive heat and spin channels that the pairing gap normally blocks. The paper's significance is that a strongly interacting, strongly dissipative system can reproduce the universal quantized transport of non-interacting fermions.","feed_headline":"Dissipation restores quantized heat and spin flow","feed_subtitle":"A loss beam breaks the pairing blockade, pushing junction conductances toward the ideal Fermi-gas limit.","key_machinery":"The central object is a quasi-one-dimensional ballistic superfluid junction connecting two reservoirs of unitary Fermi gas, with a tightly focused dissipation beam placed inside the channel. Two dissipation mechanisms are engineered with the same beam: spin-imbalanced loss that predominantly removes one spin species, and pairwise loss that removes a correlated pair. The transport coefficients are extracted from two-terminal measurements: thermal conductance from a nonlinear thermoelectric model fitted to coupled atom-number and entropy imbalances, and spin conductance from the exponential decay of a prepared magnetization imbalance. The conductances are normalized by the non-interacting quantum-limited values, so the saturation target is a direct comparison with the transverse-mode quantization of an ideal one-dimensional channel.","core_discovery":"The central claim is that local particle dissipation in a superfluid junction restores thermal and spin transport that the pairing gap suppresses, and that both conductances saturate near the quantum-limited values of an ideal non-interacting one-dimensional junction. The measured thermal conductance rises by almost two orders of magnitude and the spin conductance by more than one order, until both approach $G_{T,0}=2n_m\\pi^2 k_B^2 T/(3h)$ and $G_{\\sigma,0}=2n_m/h$, the universal limits set by the quantization of transverse modes in the channel. Qualitatively the same saturation appears for two distinct loss mechanisms, spin-imbalanced optical pumping and pairwise photoassociation, with pairwise loss slightly less destructive at equal loss current. The authors interpret this as dissipation unpairing the superfluid locally and thereby recovering conductance channels that the pairing gap had closed.","pith_inferences":["The supplement notes that the measured loss rate $\\gamma_N$ is very close to the fitted spin relaxation rate $1/\\tau_\\sigma$, and rules out loss-induced apparent spin current by a physical plausibility argument. A decisive control experiment would place the dissipation beam just outside the junction and check whether the magnetization decay rate still tracks the loss rate.","The near coincidence of both saturated conductances with the ideal-channel limits suggests measuring the ratio of reduced spin and thermal conductances as a function of dissipation strength: a universal value would strengthen the claim that the saturated channel behaves as an ideal non-interacting conductor, while a drift would expose interaction corrections.","The enhanced Seebeck response at weak dissipation suggests a testable application: applying a dissipation gradient across the reservoirs could convert an entropy imbalance into a particle current, effectively operating the junction as a dissipatively controlled thermoelectric device.","A microscopic theory of the full conductance-versus-loss curves, which the authors call for, would predict whether the saturation is governed solely by pair breaking or also by mode-specific scattering; checking that the two dissipation mechanisms collapse onto one curve when plotted against loss current would test that prediction."],"forward_implications":["At strong dissipation, both thermal and spin conductances saturate near the non-interacting ballistic values, meaning the superfluid junction conducts heat and spin like an ideal one-dimensional Fermi gas even though particle transport retains superfluid character.","Because the same saturation appears for two different loss mechanisms, the effect does not depend on how pairs are broken; to first order the overall loss rate controls the recovery of conductance.","The normalized thermal conductance collapses when plotted against transverse confinement with different occupied mode numbers, showing the saturation is tied to $n_m$, the number of transverse modes, rather than a particular trap setting.","Dissipation increases the fitted nonlinearity scale of the particle current, so the junction crosses over from a nonlinear superconducting-like contact toward a more linear transport channel.","The sharp rise of thermal and spin conductances together with the enhanced Seebeck response points toward dissipative control of thermoelectric and spin transport in a superfluid junction."],"supporting_citations":[{"why":"Supplies the phenomenological thermoelectric model, the reservoir response coefficients, and the fit and normalization procedure used to extract thermal conductance.","marker":"[22]"},{"why":"Establishes the spin-insulating baseline without dissipation and the conversion from exponential magnetization decay to spin conductance.","marker":"[19]"},{"why":"Previous dissipative quantum point contact experiment with the same spin-imbalanced loss mechanism whose particle-current suppression this work extends.","marker":"[13]"},{"why":"Quantized conductance through a dissipative atomic point contact; provides the comparison showing non-interacting dissipative channels would be strongly suppressed.","marker":"[12]"},{"why":"Created the superfluid quantum point contact platform and the nonlinear current-bias phenomenology that dissipation is shown to linearize.","marker":"[24]"},{"why":"Provide the universal quantized heat and spin conductance values used as the normalization and saturation targets.","marker":"[32–36]"},{"why":"Documents suppressed thermal conductance in the same junction class without dissipation, the baseline against which the enhancement is measured.","marker":"[25]"}],"fun_headline_variants":["Dissipation reopens quantized transport in superfluid junction","Loss beam restores ideal gas conductance in superfluid","Fermion loss saturates thermal and spin conductances","Pairing gap bypassed: dissipation drives quantum-limited flow","Dissipative superfluid junction hits ideal Fermi-gas limits"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The fitted spin conductance is only meaningful if the decay of the magnetization imbalance is real spin transport through the junction and not an artifact of atoms being lost from the reservoirs, and the authors' case for this rests on a physical plausibility argument rather than a dedicated control measurement.","fun_headline_variants_meta":{"raw":{"variants":["Dissipation reopens quantized transport in superfluid junction","Loss beam restores ideal gas conductance in superfluid","Fermion loss saturates thermal and spin conductances","Pairing gap bypassed: dissipation drives quantum-limited flow","Dissipative superfluid junction hits ideal Fermi-gas limits"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00021,"raw_usage":{"total_tokens":1360,"prompt_tokens":845,"completion_tokens":515,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":461,"completion_tokens_details":{"reasoning_tokens":434}},"tokens_in":461,"tokens_out":515,"duration_ms":5279,"temperature":1.0,"reasoning_tokens":434,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T17:44:27.555631+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the magnetization imbalance decay with the dissipation beam positioned just outside the channel instead of inside it; if the decay rate $\\gamma_N$ still matches $1/\\tau_\\sigma$, then loss-induced apparent spin current, not junction transport, would explain the reported spin conductance. A second check would compare the spin conductance inferred from the exponential magnetization decay with a direct measurement of spin-resolved currents through the junction.","supporting_citations":[{"cited_title":"Fabritius, J","cited_arxiv_id":null,"evidence_quote":"Supplies the phenomenological thermoelectric model, the reservoir response coefficients, and the fit and normalization procedure used to extract thermal conductance."},{"cited_title":"Krinner, M","cited_arxiv_id":null,"evidence_quote":"Establishes the spin-insulating baseline without dissipation and the conversion from exponential magnetization decay to spin conductance."},{"cited_title":"Huang, J","cited_arxiv_id":null,"evidence_quote":"Previous dissipative quantum point contact experiment with the same spin-imbalanced loss mechanism whose particle-current suppression this work extends."},{"cited_title":"Corman, P","cited_arxiv_id":null,"evidence_quote":"Quantized conductance through a dissipative atomic point contact; provides the comparison showing non-interacting dissipative channels would be strongly suppressed."},{"cited_title":"Husmann, M","cited_arxiv_id":null,"evidence_quote":"Documents suppressed thermal conductance in the same junction class without dissipation, the baseline against which the enhancement is measured."}],"review_version":1}