{"id":"5b626f7a-ee46-4758-9012-b021a0ab8e8c","arxiv_id":"1908.04886","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Bulk viscous electron flow has positive magnetoresistance for arbitrary inhomogeneity in one-dimensional periodic models and in weakly inhomogeneous ballistic-to-hydrodynamic crossover calculations, unlike narrow channels.","lead":"This paper argues that in a clean metal where electrons flow like a fluid, applying a magnetic field always makes the bulk electrical resistance larger, not smaller. That means a rise in resistance under a small magnetic field can serve as a simple experimental fingerprint of viscous electron flow.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'regardless of strength' claim rests on a 1D-periodic exact solution; the paper's own admitted 2D gap is the load-bearing missing step.","rationale":"The reader's weakest assumption identifies the same load-bearing gap: the exact hydrodynamic result applies only to 1D periodic inhomogeneity, while the abstract's 'regardless of the strength of inhomogeneity' asserts a 2D conclusion. My reading of Sec. 2.2 confirms that the derivation of rho_yy in Eq. (22c) and the positive-sign estimate in Eqs. (25)-(27) rely on n(y) only, and the paper's own words ('The one shortcoming in our argument is...') concede the 2D extension is left unresolved. Similarly, in Sec. 3.3 the statement 'we have not demonstrated that magnetoresistance is positive across the ballistic-to-hydrodynamic crossover for large amplitude inhomogeneity' appears verbatim, and the subsequent 'we see no reason' is an expectation, not a proof. I have no separate concern beyond this, and I agree with the reader's CONDITIONAL verdict: the claim is credible and internally consistent for the 1D case, but the 'always positive' phrasing overstates the derived support. The concrete test I propose would settle whether a 2D counterexample exists; until then, the condition should be treated as plausible but unestablished.","tokens_in":10279,"tokens_out":4479,"duration_ms":44181,"concrete_test":"Perform a 2D finite-element solution of the linearized hydrodynamic equations (1) with periodic checkerboard density n(x,y)=n1 in circular/square domains of diameter w and background n2 (n1<<n2), with transitions of width a, using eta(n,B) and eta_H(n,B) from Eq. (4); impose a uniform current and compute the effective longitudinal magnetoresistivity rho_xx(B) with Hall voltage conditions appropriate for bulk. Scan n2/n1 in {5,50}, w/a in {10,100}, and B up to omega_c tau_ee ~ 1. If any set gives rho_xx(B)-rho_xx(0)<0, the 'always positive' claim is falsified; if all give positive values, the concern is mitigated but the general claim still requires an analytic argument for 2D.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract claims that 'regardless of the strength of inhomogeneity, bulk magnetoresistance is always positive in the hydrodynamic regime,' but the exact solution in Sec. 2.2 (Eqs. (21)-(27)) treats media inhomogeneous in only one direction (n=n(y), Fig. 1); the full 2D problem is not solved. The paper states this explicitly: 'The one shortcoming in our argument is... that the system was only inhomogeneous in one of the two directions... We leave a final resolution of the two-dimensional transport problem to elsewhere.' The heuristic argument that 2D 'would not qualitatively change' rests on the claim that Hall voltages prevent flow along zero-resistance contours, but in 2D current can detour around density domains, altering the spatial pattern of the Hall field and possibly changing the sign of the B^2 coefficient. Additionally, Sec. 3.3 concedes that positive magnetoresistance across the ballistic-to-hydrodynamic crossover for large-amplitude 2D inhomogeneity 'has not been demonstrated' ('we see no reason for this to not be the case'). Thus the load-bearing condition—that the 1D periodic solution controls the sign in arbitrary 2D density landscapes—is exactly the support the paper identifies as missing.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript addresses the sign of magnetoresistance in two-dimensional electron fluids in which electron-electron collisions dominate. In the hydrodynamic section, the authors solve the linearized hydrodynamic equations for a periodic medium inhomogeneous in one direction, obtaining exact expressions for the conductivity and resistivity (Eqs. (21)-(22c)), and use a two-density cartoon to estimate the small- and large-field behavior. They find that the small-B derivative of rho_yy can be negative only for parameter regimes they argue are not physical, and they conclude that bulk magnetoresistance is positive regardless of inhomogeneity strength. In the kinetic-theory section, they compute the density spectral weight for weakly inhomogeneous media in a solvable Boltzmann model and in a model with a hierarchy of odd-harmonic lifetimes, finding positive magnetoresistance across the ballistic-to-hydrodynamic crossover and showing that the bulk Gurzhi resistance minimum is destroyed by a small magnetic field. The abstract draws the strong conclusions that negative magnetoresistance is not a signature of viscous bulk flow and that magnetic-field sensitivity is a diagnostic for the Gurzhi effect.","tokens_in":10459,"tokens_out":9431,"duration_ms":93354,"significance":"The result, if it held in full generality, would resolve a recent controversy: it would refute the idea that negative magnetoresistance is a signature of viscous flow in a bulk crystal, and it would propose a practical experimental test (extreme sensitivity of a bulk resistance minimum to a magnetic field) for the Gurzhi effect. The paper has real strengths: the derivation of Eq. (22c) is algebraic and self-contained; the B=0 limit reproduces Ref. [15] and the weak-field kinetic-theory limit reproduces Refs. [20,21]; the spectral-weight computation places the magnetic field inside the propagator rather than treating it perturbatively; and the qualitative conclusions are checked with more than one collision model. The proposed magnetic-field diagnostic is falsifiable and would be useful to experimental groups. However, the headline generality of the claim is not supported by the calculations, as detailed in the major comments.","major_comments":[{"comment":"The exact result behind the abstract's 'regardless of the strength of inhomogeneity' claim is derived only for media inhomogeneous in one direction, n=n(y), with the transport coefficients in Eqs. (21)-(22c). The authors explicitly state that 'the one shortcoming in our argument is... that the system was only inhomogeneous in one of the two directions' and leave a final 2D resolution to elsewhere. In a 2D inhomogeneous medium, current may detour around density domains, so the Hall-field pattern and the signs of the various B^2 cross-terms are not fixed by the 1D solution; the one-sentence heuristic about local Hall effects is not a calculation. Moreover, in the 1D geometry rho_xx=0 (Eq. (22a)), so the 'bulk' magnetoresistance is assessed through a specific orientation (rho_yy) that has no direct 2D analogue. The central claim therefore needs either a 2D calculation or a restriction to unidirectional inhomogeneity.","section":"Sec. 2.2 (Eqs. (21)-(22c))"},{"comment":"The crossover calculation is explicitly restricted to weakly inhomogeneous media (Sec. 3.1), yet the text concedes that positive magnetoresistance for large-amplitude inhomogeneity 'has not been demonstrated' and only asserts 'we see no reason for this to not be the case.' Because the abstract and conclusion present positive magnetoresistance and the magnetic-field test for the Gurzhi effect as general statements, this leaves another load-bearing gap. This sentence should be backed by a numerical or analytic estimate, or the generality of the claim should be reduced.","section":"Sec. 3.3 (crossover, after Fig. 3)"},{"comment":"The exact expression (22c) is positive semidefinite as a transport coefficient, but it does not by itself prove that d(rho_yy)/d(B^2) > 0 for all profiles: the small-B estimate (24) contains a negative contribution, and the conclusion that magnetoresistance is positive rests on the scaling condition (25) combined with hydrodynamic inequalities and the assumption z >= 1. Thus the hydrodynamic section establishes a useful class of models, but the claim 'always positive' is not a theorem; the authors should state the precise sufficient conditions (e.g., the regime of validity of the estimates in Eqs. (23)) or prove monotonicity directly for periodic n(y).","section":"Eqs. (22c)-(27)"}],"minor_comments":[{"comment":"The reference to 'Figure 4' in the sentence after Eq. (50) appears to be a typo; the first panel showing rho as a function of xi/l_ee at fixed l_B/xi is Figure 2, while Figure 4 is first used in Section 3.4.","section":"Sec. 3.3 (after Eq. (50))"},{"comment":"The author string of Ref. [6] lists 'E. V. Levinson' twice; the first occurrence appears to be a typographical error, presumably for G. M. Gusev.","section":"Reference [6]"},{"comment":"The symbol omega_c is used in Eq. (4) but is not defined until Eq. (36) in Sec. 3.2; it should be defined at first use, for example by omega_c = v_F B / p_F.","section":"Sec. 2.1 (Eq. (4))"}],"recommendation":"major_revision","confidential_remarks":"The paper is a useful contribution and the kinetic-theory calculation is valuable, but the submitted version's abstract and conclusion overstate the domain of validity of the central result. The load-bearing gap is the 2D inhomogeneity problem, which the authors themselves identify as open. I would ask the editor to require a revised version that either supplies the 2D hydrodynamic calculation or substantially narrows the claim; a heuristic paragraph is not enough to support 'regardless of the strength of inhomogeneity.'"},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things about this paper. First, the core physics claim — that bulk magnetoresistance in an electron fluid is positive in the weakly inhomogeneous regime across the ballistic-to-hydrodynamic crossover — is solid and useful, and the extreme sensitivity of the bulk Gurzhi effect to a magnetic field is a genuinely simple experimental diagnostic. Second, the more aggressive part of the abstract, 'regardless of the strength of inhomogeneity,' is not actually proven. The exact hydrodynamic solution covers one-dimensional periodic inhomogeneity only; the extension to two dimensions is a stated expectation, not a result.\n\nWhat is new: Eq. (22c) is a clean closed-form resistivity for a 1D periodic density profile at arbitrary inhomogeneity strength (within the constant-viscosity hydrodynamics they set up), and the kinetic-theory calculation with multiple collision operators extends the perturbative Refs [20,21] to finite B, reproducing Ref [15] at B=0. The numerical results show that a bulk resistance minimum from viscous effects should vanish in a modest magnetic field, which separates it from Kondo-like minima. That is a valuable prediction.\n\nThe paper is admirably honest. It explicitly states the one-dimensional shortcoming and says the 2D problem is left to future work. The Sec. 3.3 sentence that positive magnetoresistance across the crossover for large-amplitude 2D inhomogeneity 'has not been demonstrated' is easy to miss but is there. My main complaint is that the abstract and conclusion do not carry that caveat; they say 'always positive' without the qualifier. The derivation also treats η and ηH as constants in the exact solution, then uses position-dependent η in the cartoon estimates — a modeling inconsistency that is not harmful for the qualitative conclusion but does weaken the 'exact' label.\n\nThe kinetic-theory model has one free cutoff (nmax), but the qualitative results are robust to the choice of collision operator, so that is a minor point. The citation pattern is fine; the paper builds on [20,21] and [15] and credits them.\n\nBottom line: this is a paper worth refereeing, not desk-rejecting. I would ask for the global claim to be softened to match the 1D-periodic proof, or for a 2D numerical demonstration. For the subfield of electron hydrodynamics, it is a useful contribution. I'd bring it to the group and would cite it if I worked on magnetotransport.","headline":"A solid, honest extension of the perturbative results, with a useful experimental diagnostic, but the 'always positive' claim outruns the 1D-periodic proof.","tokens_in":11035,"tokens_out":4531,"would_cite":true,"duration_ms":46704,"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 argues that in bulk crystals, electron hydrodynamic flow always produces positive magnetoresistance, no matter how strong the density inhomogeneity, so negative magnetoresistance cannot be a signature of viscosity.","keywords":["electron hydrodynamics","magnetoresistance","viscous electron flow","Gurzhi effect","inhomogeneous media","kinetic theory","Fermi liquid","ballistic-to-hydrodynamic crossover"],"falsifier":"Measure $\\rho(B)$ at low temperature in a bulk two-dimensional electron system whose width is much larger than the viscous mean free path and whose density is deliberately made inhomogeneous; a decrease of resistance as $B^2$ grows—$\\partial\\rho/\\partial(B^2)<0$—would contradict the central claim. Alternatively, solve the hydrodynamic equations on a fully two-dimensional periodic density landscape and search numerically for a regime with negative magnetoresistance, which the paper leaves open.","tokens_in":10020,"feed_emoji":"🧲","tokens_out":9939,"duration_ms":93573,"temperature":0.7,"pith_summary":"In sufficiently clean metals, electrons can flow like a viscous fluid, and physicists have disagreed about whether that flow makes the electrical resistance rise or fall when a magnetic field is applied. This paper argues that in a bulk crystal the answer is unambiguous: magnetoresistance is always positive in the hydrodynamic regime, no matter how strong and irregular the density inhomogeneity, and the same positive sign persists across the ballistic-to-hydrodynamic crossover in weakly disordered samples. The negative magnetoresistance seen in clean systems is a narrow-channel phenomenon, coming from a Hall voltage that builds up across the channel but cannot survive in a continuous medium. The consequence, if the paper is right, is that the magnetic-field sensitivity of a resistance minimum becomes a practical test: a bulk resistance minimum that vanishes at modest fields is a viscous effect, not an impurity-scattering one. The authors flag that their exact calculation treats density variations along only one direction and leave the fully two-dimensional problem open.","feed_headline":"Viscous electron flow always gives positive bulk magnetoresistance","feed_subtitle":"In a small magnetic field the bulk viscous resistance minimum vanishes, giving a clean experimental test","key_machinery":"The central object in the hydrodynamic half is the linearized set of charge, energy, and momentum conservation equations for a Fermi liquid in a magnetic field, with local density $n(y)$, shear viscosity $\\eta$, and Hall viscosity $\\eta_H$ varying periodically in one spatial direction. In a two-dimensional isotropic Fermi liquid these viscosities obey $\\eta(B)=\\eta_0/(1+(2\\omega_c\\tau_{ee})^2)$ and $\\eta_H(B)=2\\omega_c\\tau_{ee}\\eta(B)$, which ties the magnetic-field response to the viscous mean free path $\\ell_{ee}=v_F\\tau_{ee}$. The key identity is the exact formula for the longitudinal resistivity $\\rho_{yy}$ as a sum of positive-semidefinite terms involving $(\\partial_y(1/n))^2$, $(\\Psi-\\langle n\\rangle^{-1}B(\\eta_H/\\eta)\\partial_y(1/n))^2$, and thermal-conductivity corrections, where $\\Psi$ is the periodic antiderivative of $n-\\langle n\\rangle$. In the kinetic-theory half the load-bearing object is the charge-density spectral weight $A_{nn}(k)$; a projection trick reduces the Boltzmann-equation inversion to a small matrix whose entries are Bessel functions $J_{n-\\tilde{n}}(k\\ell_B)$, so the resistivity across the ballistic-to-hydrodynamic crossover can be evaluated from the disorder spectrum $|\\mu(k)|^2$.","core_discovery":"On its own terms, the paper claims that in a bulk (quasi-)two-dimensional electron fluid the longitudinal magnetoresistance is positive for any magnetic field, independent of the strength of inhomogeneity, and that this also holds for weakly disordered metals near the onset of viscous flow. The mechanism is the continuity of the electrochemical potential in a continuous medium: the local Hall voltage that would appear in a narrow channel cannot simply terminate, so a field-dependent circulating current $J_y$ flows between regions of different density, and its dissipation adds positive contributions to the bulk resistivity $\\rho_{yy}$. Solving the hydrodynamic equations exactly for a periodic one-dimensional density profile yields a positive-semidefinite conductivity tensor and an explicit formula for $\\rho_{yy}$; estimating the terms with sharp density domains and Fermi-liquid viscosity $\\eta(B)=\\eta_0/(1+(2\\omega_c\\tau_{ee})^2)$ shows the magnetoresistance is positive at both small and large $B$. In the kinetic-theory calculation near the ballistic-to-hydrodynamic crossover, the resistivity rises rapidly with magnetic field regardless of the microscopic collision integral, and the zero-field non-monotonic temperature dependence of the resistivity, the bulk analogue of the Gurzhi effect, disappears once $\\ell_B$ is even modestly smaller than the inhomogeneity scale.","pith_inferences":["If the 'always positive' claim is right, then previously reported negative magnetoresistance in wide samples must originate elsewhere—for example, in current-path geometry, contacts, or two-dimensional percolation around density domains—and a direct narrow-versus-wide comparison on the same crystal would separate those mechanisms.","Because $\\ell_{ee}\\sim T^{-2}$ in a Fermi liquid, the magnetic-field scale at which the resistivity starts rising should itself track $T^{-2}$; fitting $\\rho(B)$ across temperatures could extract the temperature-dependent viscosity without needing a zero-field subtraction.","The authors' 1D shortcoming is directly testable: numerically solving the same hydrodynamic equations on a genuinely two-dimensional periodic density landscape would show whether a 2D current path can thread around density domains and lower the resistance, which is the case left open in the paper.","A designed periodic density modulation (for instance, a lithographic superlattice) with a tunable field could provide a controlled check of the predicted scaling $\\rho_{yy}\\sim (B w)^2/\\eta(n_1,B)$ at small $B$."],"forward_implications":["In a bulk sample, a resistance minimum whose depth is destroyed by a modest magnetic field is evidence for viscous electron flow, and can be separated from impurity-scattering minima that are far less field-sensitive.","Negative magnetoresistance in a hydrodynamic conductor is a geometry effect; it should only be expected in narrow channels, not in bulk transport measurements.","The magnetoresistance remains positive even when the microscopic collision operator is changed, so the sign is a robust property of momentum-conserving electron-electron scattering rather than a quirk of the relaxation-time approximation.","Near the ballistic-to-hydrodynamic crossover, the positive magnetoresistance appears already at small fields, so magnetotransport can diagnose the onset of viscous flow before full hydrodynamics is established.","The field strength at which $\\rho(B)$ starts to climb is set by the ratio of the inhomogeneity length scale to the viscous mean free path, so measurements of magnetoresistance could be used to estimate $\\ell_{ee}(T)$."],"supporting_citations":[{"why":"Proposed the narrow-channel hydrodynamic mechanism that produces negative magnetoresistance and that the paper must explain away.","marker":"[17]"},{"why":"Reported negative magnetoresistance in narrow graphene channels, the confined geometry the paper agrees can show this sign.","marker":"[19]"},{"why":"Gave a perturbative argument that bulk viscous magnetoresistance is positive; this paper extends it beyond weak inhomogeneity.","marker":"[20]"},{"why":"Similarly argued positive bulk magnetoresistance at weak inhomogeneity; combined with [20] it sets the baseline the paper generalizes.","marker":"[21]"},{"why":"Supplied the kinetic-theory framework and the zero-field bulk viscous resistance minimum that the paper recomputes with a magnetic field.","marker":"[15]"},{"why":"Supplied the general formula expressing resistivity in weakly inhomogeneous media in terms of the charge-density spectral weight.","marker":"[25]"},{"why":"Provided the block-inversion and Bessel-function trick used to evaluate the spectral weight exactly in the kinetic model.","marker":"[11]"},{"why":"Identified the hierarchy of excitation lifetimes in 2D Fermi gases that motivates the more sophisticated collision operators.","marker":"[30]"}],"fun_headline_variants":["Bulk electron fluid magnetoresistance is always positive","Even with inhomogeneity, viscous electron flow gives positive magnetoresistance","Viscous bulk magnetoresistance: positive sign, robust to inhomogeneity","Small field kills bulk Gurzhi effect, gives positive magnetoresistance","Always positive magnetoresistance in viscous electron fluids"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument's load-bearing premise is that transport in a bulk crystal with density variations in both directions behaves like the exactly solved case with variations in only one direction; the authors state that the fully two-dimensional problem is left to future work, so if two-dimensional flow paths can short-circuit the field-dependent dissipation, the 'always positive' conclusion could fail.","fun_headline_variants_meta":{"raw":{"variants":["Bulk electron fluid magnetoresistance is always positive","Even with inhomogeneity, viscous electron flow gives positive magnetoresistance","Viscous bulk magnetoresistance: positive sign, robust to inhomogeneity","Small field kills bulk Gurzhi effect, gives positive magnetoresistance","Always positive magnetoresistance in viscous electron fluids"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001569,"raw_usage":{"total_tokens":6251,"prompt_tokens":919,"completion_tokens":5332,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":535,"completion_tokens_details":{"reasoning_tokens":5245}},"tokens_in":535,"tokens_out":5332,"duration_ms":32594,"temperature":1.0,"reasoning_tokens":5245,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:30:32.388845+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure $\\rho(B)$ at low temperature in a bulk two-dimensional electron system whose width is much larger than the viscous mean free path and whose density is deliberately made inhomogeneous; a decrease of resistance as $B^2$ grows—$\\partial\\rho/\\partial(B^2)<0$—would contradict the central claim. Alternatively, solve the hydrodynamic equations on a fully two-dimensional periodic density landscape and search numerically for a regime with negative magnetoresistance, which the paper leaves open.","supporting_citations":[{"cited_title":"Negative magnetoresistance in viscous flow of two-dimensional electrons","cited_arxiv_id":"1603.04587","evidence_quote":"Proposed the narrow-channel hydrodynamic mechanism that produces negative magnetoresistance and that the paper must explain away."},{"cited_title":"Measuring Hall Viscosity of Graphene's Electron Fluid","cited_arxiv_id":"1806.01606","evidence_quote":"Reported negative magnetoresistance in narrow graphene channels, the confined geometry the paper agrees can show this sign."},{"cited_title":"Viscous magnetoresistance of correlated electron liquids","cited_arxiv_id":"1612.09275","evidence_quote":"Gave a perturbative argument that bulk viscous magnetoresistance is positive; this paper extends it beyond weak inhomogeneity."},{"cited_title":"Hydrodynamic flows of non-Fermi liquids: magnetotransport and bilayer drag","cited_arxiv_id":"1706.03775","evidence_quote":"Similarly argued positive bulk magnetoresistance at weak inhomogeneity; combined with [20] it sets the baseline the paper generalizes."},{"cited_title":"Kinetic theory of transport for inhomogeneous electron fluids","cited_arxiv_id":"1706.04621","evidence_quote":"Supplied the kinetic-theory framework and the zero-field bulk viscous resistance minimum that the paper recomputes with a magnetic field."},{"cited_title":"Memory matrix theory of magnetotransport in strange metals","cited_arxiv_id":"1502.04704","evidence_quote":"Supplied the general formula expressing resistivity in weakly inhomogeneous media in terms of the charge-density spectral weight."},{"cited_title":"The Hierarchy of Excitation Lifetimes in Two-Dimensional Fermi Gases","cited_arxiv_id":"1905.03751","evidence_quote":"Identified the hierarchy of excitation lifetimes in 2D Fermi gases that motivates the more sophisticated collision operators."}],"review_version":1}