{"id":"bb041276-25d4-4354-85cb-fa37bd3889f0","arxiv_id":"2502.03646","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Ab initio GW+DMFT calculations show that orbital-selective Kondo coherence produces coexisting 3D and quasi-2D Gamma-centered Fermi surfaces in UTe2 at low temperature.","lead":"A computational study of the candidate topological superconductor UTe2 reports that it hosts two coexisting Fermi surfaces centered at the Gamma point, one three-dimensional and one quasi-two-dimensional, which emerge as the material cools to 25 K. The result could explain why different experiments see different Fermi surface shapes and supports the presence of the 3D Fermi surface needed for topological superconductivity.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim rests on an undefined Fermi-surface criterion: constant contours of orbital-projected A(k, omega=0) at an unspecified isovalue can create or destroy apparent 3D/2D sheets without any change in quasiparticle poles.","rationale":"The paper's strongest claim is a statement about the topology and dimensionality of the Fermi surface, so the operational definition of 'Fermi surface' is load-bearing. The manuscript never states the isovalue or defines a criterion based on poles of the Green's function. All figures that support the headline, Figs. 1c-1e and 2a-2c, are color maps of orbital-projected A(k,0). In an interacting Kondo system, this quantity is nonzero and smoothly varying across the Brillouin zone, so any closed contour is a choice of threshold, not necessarily a Fermi surface. The temperature dependence of the apparent shape can be dominated by the temperature dependence of the spectral weight scale: the same isovalue picks out different parts of the spectral function at 900 K and 25 K. The reader's weakest assumption identifies exactly this threshold issue, and the paper's own statement that the U-6d surface is visible only below 0.06 A(eV) confirms that the result is cutoff-dependent. A concrete pole-tracking test would settle it. Because the requested clarification is feasible and the central claim depends on it, the reader's CONDITIONAL verdict is appropriate; no change is needed.","tokens_in":10117,"tokens_out":3597,"duration_ms":34372,"concrete_test":"Replot the orbital-projected spectral function as A(k,omega) and extract the quasiparticle pole omega_k at each k (or the maximum of A(k,omega) near omega=0) for T=25 K, 100 K, 300 K, and 900 K. Define the Fermi surface as the set of k where a pole crosses omega=0 (equivalently, where Re G^{-1}(k,0)=0). Then compare the pole-derived Gamma-centered sheets with the color-scale contours in Figs. 1 and 2, varying the isovalue from 0.01 to 0.1 A(eV) in factor-of-2 steps. If the apparent 3D-to-quasi-2D transformation disappears or the Gamma-centered sheets are absent in the pole criterion, the central claim fails; if the sheets persist at all thresholds and match the pole crossings, the visualization concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim depends on identifying Fermi surfaces as constant-intensity contours of the orbital-projected spectral weight A(k, omega=0) at an unspecified isovalue. In an interacting Kondo lattice, A(k,0) is finite and broad for all k; it does not vanish on one side of a Fermi surface. A fixed color-scale cutoff can therefore transform a broad, momentum-dependent hump into a closed pocket or an open cylinder depending only on the chosen threshold, without any actual reconstruction of the quasiparticle Fermi surface. The paper's evidence is almost entirely of this kind: Figs. 1c-1e and 2a-2c show color maps of A(k,0), and the text states that weak features are 'visible only at a lower range of spectral weight (<0.06 A(eV))'. No quasiparticle pole position, no E(k) crossing omega=0, and no Luttinger count is presented. The claim that the Te2-5p sheet 'transforms from the closed 3D one to the open 2D one' upon cooling is inferred from the same isovalue at two temperatures, although the spectral weight itself changes by an order of magnitude with temperature. Because the weakly weighted Gamma-centered features are invisible at high isovalues, the coexistence of 3D and quasi-2D sheets could be a visualization artifact rather than a property of the Green's function. The 25 K sign-problem limitation makes the low-temperature 'should exist' statement an extrapolation, but the more fundamental issue is the undefined Fermi-surface criterion.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports LQSGW+DMFT calculations for the heavy-fermion superconductor UTe2, including spin-orbit coupling, with all interaction parameters and the double-counting energy computed from first principles using an experimental lattice constant. The central claim is that, upon cooling to 25 K, orbital-selective Kondo coherence produces two coexisting Γ-centered Fermi surfaces: a quasi-2D cylinder from Te2-5p states and a 3D dumbbell from U-6d states, on top of a non-Γ-centered quasi-2D cylindrical sheet. This coexistence is proposed to explain the conflicting quantum-oscillation and ARPES reports of 2D vs. 3D Fermi surfaces and to support the existence of a 3D Fermi surface required for topological superconductivity. The evidence consists primarily of orbital-projected spectral weight A(k,ω=0) maps at different temperatures, together with spectral functions showing enhanced Kondo coherence along selected directions.","tokens_in":10457,"tokens_out":4747,"duration_ms":41548,"significance":"If the central claim is correct, the paper resolves a live experimental controversy in UTe2 and provides a concrete microscopic mechanism (orbital-selective Kondo coherence) for the apparent temperature and probe dependence of the Fermi surface. The main strengths are that the calculation is genuinely first-principles: no correlation parameters are fitted, the Coulomb tensor and double-counting are computed, and the LQSGW+DMFT machinery is state of the art. The paper also correctly emphasizes that the Γ-centered sheets carry weak spectral weight, which is consistent with their being invisible in some probes. However, the significance is contingent on the Fermi surface identification being reliable, and the current evidence is not yet conclusive.","major_comments":[{"comment":"The Fermi surfaces are identified exclusively from constant-intensity contours of the orbital-projected spectral weight A(k,ω=0) at an unspecified isovalue. For an interacting Kondo lattice, A(k,0) is finite for all k, so any closed or open contour can be generated by choosing the threshold; the text's statement that the U-6d surface is 'visible only at a lower range of spectral weight (<0.06 A(eV))' (Results, 'Enhanced 3D Fermi surface of U-6 d') makes the threshold dependence explicit. The paper should provide direct evidence that these contours correspond to quasiparticle Fermi surfaces: e.g., the locus of poles of the spectral function, the momentum-distribution discontinuity, or a Luttinger count. Without this, the coexistence claim is not established.","section":"Results, 'Transition from 3D to quasi-2D Fermi surface of Te2-5p' and Fig. 2a-c"},{"comment":"The claimed 3D-to-quasi-2D transformation upon cooling is based on comparing contours of A(k,0) at T=900 K and 25 K using the same color scale, but the spectral weight itself changes by roughly an order of magnitude between these temperatures (e.g., 'weak and dispersive' at 900 K vs. 'more increased' at 25 K). A fixed isovalue will therefore produce different pocket sizes and even different topology even if the quasiparticle band structure is unchanged. The authors should demonstrate, for at least one representative k-path, that the peak in A(k,ω) at ω=0 crosses the Fermi level or that the quasiparticle pole evolves continuously as T is lowered.","section":"Fig. 1d-e and Fig. 2b-c"},{"comment":"The central claim is about 'low temperatures' (experimental dHvA and STM temperatures are below 1 K), but the calculation is limited to T ≥ 25 K by the sign problem, as stated in the Discussion. The sentence 'as the temperature decreases below 25 K, the Kondo coherence should become stronger' is an extrapolation rather than a result. The paper should either soften the low-temperature claim (e.g., in the title and abstract) or provide a concrete physical argument—such as an estimate of the Kondo scale from the computed self-energy—that the 25 K topology persists at experimental temperatures. As written, the key prediction '3D Fermi surface should exist at low temperature' rests on an untested extrapolation.","section":"Discussion, last two paragraphs"}],"minor_comments":[{"comment":"The isovalue(s) used to define the Fermi surface contours are never stated in the captions or in the text; please add them, or state explicitly that the contours are guides drawn by hand.","section":"Figs. 1-2 captions"},{"comment":"The caption appears to mislabel panels: it refers to cross sections at T=25 K (a) and T=900 K (b), but the figure contains panels (c), (d), and (e).","section":"Fig. 1 caption"},{"comment":"The sentence 'The The LQSGW+DMFT approach' contains a duplicated article.","section":"Methods, section A"},{"comment":"References [3] and [32] are the same paper (Miao et al., PRL 124, 076401), and [10] and [31] are the same paper (Xu et al., PRL 123, 217002); please consolidate these duplicates.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper is technically sophisticated and the first-principles setup is commendable, but the central claim currently rests on an unspecified spectral-weight contour criterion. I would need to see either a pole-based Fermi surface definition or a clear justification for why the chosen isovalue faithfully tracks the quasiparticle Fermi surface before endorsing the coexistence scenario. The extrapolation below 25 K is a secondary but real concern. The duplicate references and figure caption issues are easy to fix."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nQuick take: this is a serious LQSGW+DMFT calculation on UTe2 that goes beyond earlier DFT/DFT+U/DFT+DMFT work by predicting a temperature-driven coexistence of two Gamma-centered Fermi surfaces with different dimensionality, driven by orbital-selective Kondo scattering. If the Fermi-surface identification holds up, it would reconcile contradictory quantum oscillation and ARPES measurements and preserve the 3D sheet argued to be needed for topological superconductivity. That is a genuinely new result, not a fitting exercise: interactions, double counting, and the Coulomb tensor are computed from first principles; only the lattice constants come from experiment.\n\nWhat the paper does well: the method is state of the art and the authors are appropriately careful about the Kondo scale, the role of spin-orbit coupling, and the sign-problem limitation at low T. The comparison with and without SOC is a nice diagnostic. The writing is clear and the relation to previous experiments and calculations is laid out honestly.\n\nSoft spots, in order of importance. First, the central claim rests on identifying Fermi surfaces as constant-intensity contours of orbital-projected A(k, omega=0) at an isovalue that is never specified. In a Kondo lattice, A(k,0) is broad and finite everywhere; contour levels can create or destroy apparent pockets without any change in quasiparticle poles. The paper does not show a quasiparticle pole crossing omega=0, a Luttinger count, or a robustness scan over isovalue. The text mentions that the U-6d sheet is visible only below 0.06 eV states, which underlines the arbitrariness. Second, the predicted evolution below 25 K is an extrapolation, as the authors openly state; that is a limit of the method, not a sin, but it means the \"should exist\" language is doing work. Third, the data availability statement says \"available upon reasonable request\" with no input files or parameters; for a claim this visual, that is a reproducibility gap.\n\nOverall, the calculation is credible and the narrative is coherent, but the evidence for the specific Fermi-surface topology is weaker than the title suggests. The paper deserves a serious referee, but a referee should push for a quantitative definition of the Fermi surface and isovalue robustness checks before publication.\n\nRecommendation: send to peer review with a request for a defined Fermi-surface criterion; do not desk-reject.","headline":"A serious LQSGW+DMFT calculation that makes a plausible but under-specified claim about coexisting 3D and quasi-2D Fermi surfaces in UTe2; it deserves peer review, but the Fermi-surface definition needs to be made quantitative.","tokens_in":10953,"tokens_out":1886,"would_cite":false,"duration_ms":17149,"reading_group":"maybe","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 claims that low-temperature UTe2 hosts both a quasi-2D and a 3D Fermi surface centered at the Brillouin-zone center, produced by orbital-selective Kondo scattering.","keywords":["UTe2","Kondo effect","Fermi surface","heavy fermion","dynamical mean-field theory","topological superconductivity","orbital selectivity","spin-orbit coupling"],"falsifier":"Recompute the orbital-projected spectral weight at several isovalues and locate the poles of $A(\\mathbf{k},\\omega=0)$: if the $\\Gamma$-centered Te2-5p cylinder and U-6d dumbbell disappear at slightly higher cutoffs or appear only as shoulders rather than peaks, the coexistence is a visualization artifact, whereas if the poles persist and a sub-25 K quantum-oscillation measurement resolves the closed 3D pocket, the claim stands.","tokens_in":9913,"feed_emoji":"⚛️","tokens_out":17021,"duration_ms":136685,"temperature":0.7,"pith_summary":"UTe$_2$ is a heavy-fermion superconductor whose Fermi surface is experimentally contested: some probes see a three-dimensional sheet, others only quasi-two-dimensional cylinders. This paper argues that both observations are correct because the material has two orbital-dependent Fermi surfaces centered at the Brillouin-zone center $\\Gamma$ that behave differently as the temperature falls. Using a first-principles many-body method that includes the Kondo effect, the authors find that a Te2-5p sheet starts as a closed 3D pocket at high temperature and becomes a quasi-2D cylinder at 25 K, while a U-6d sheet remains 3D with a dumbbell shape and gains spectral weight on cooling. The mechanism is orbital-selective Kondo scattering: coherent Kondo hybridization redistributes spectral weight differently for different orbitals, and without spin-orbit coupling the quasi-2D sheet does not form. If correct, the result reconciles the conflicting measurements and keeps topological superconductivity viable, since the 3D $\\Gamma$-centered sheet can support odd-parity pairing; the authors note they cannot compute below 25 K due to the sign problem, so the full low-temperature picture is an extrapolation.","feed_headline":"Cooling turns one UTe2 Fermi surface from 3D to quasi-2D","feed_subtitle":"This coexistence explains conflicting experiments and keeps topological superconductivity in play.","key_machinery":"The load-bearing object is the orbital-projected spectral weight $A(\\mathbf{k},\\omega=0)$ computed by LQSGW+DMFT, a first-principles scheme that combines a self-consistent quasiparticle GW calculation with dynamical mean-field theory for the local U-5f and U-6d self-energies, with spin-orbit coupling included and experimental lattice constants as the only external input. The transformation mechanism is the orbital-selective Kondo effect: the c-axis Kondo hybridization between Te2-5p states of $|j=3/2, j_z=\\pm3/2\\rangle$ and U-5f states of $|j=5/2, j_z=\\pm5/2\\rangle$ becomes coherent upon cooling and pushes the Te2-5p spectral weight into a quasi-2D cylinder, while U-6d/U-5f hybridization behaves differently in the $\\Gamma$–$Y$ direction and produces the dumbbell 3D sheet. In the calculation without spin-orbit coupling, the $\\Gamma$-centered Te2-5p sheet does not appear, so spin-orbit coupling is identified as a necessary ingredient for the quasi-2D reconstruction.","core_discovery":"Stated in the Results, the central claim is that ab-initio many-body calculations that explicitly include the Kondo effect reveal the coexistence of 3D and quasi-2D Fermi surfaces centered at the $\\Gamma$ point at low temperatures, due to the orbital selective Kondo effect. Concretely, the Te2-5p state with $|j=3/2, j_z=-3/2\\rangle$ forms a 3D oblate pocket at high temperature and turns into a quasi-2D cylindrical Fermi surface by 25 K, while the U-6d state with the same quantum numbers forms a 3D dumbbell-shaped sheet centered at $\\Gamma$ that becomes sharper on cooling. A non-$\\Gamma$-centered quasi-2D cylindrical sheet, mostly from weakly correlated U-6d and Te-5p bands, persists at both temperatures. The authors read this as a Fermi-surface reconstruction driven by Kondo coherence that is orbital- and direction-selective: stronger hybridization along $\\Gamma$–Z for Te2-5p makes that sheet cylindrical, while stronger hybridization along $\\Gamma$–Y for U-6d gives the dumbbell shape, and the result is presented as the microscopic explanation for why angle-resolved photoemission and quantum oscillation experiments report different dimensionalities.","pith_inferences":["Beyond the paper, the same data could be analyzed for quasiparticle poles rather than spectral-weight isosurfaces; if the $\\Gamma$-centered contours are only incoherent shoulders, the claimed 3D-to-quasi-2D transformation would reduce to a plotting artifact rather than a true Fermi-surface topology change.","Beyond the paper, the orbital-selective mechanism predicts that other Kondo lattices with multiple conduction orbitals should show temperature-dependent Fermi-surface reconstruction that depends on which orbital a probe couples to, so orbital-sensitive experiments such as resonant ARPES and mass-resolved quantum oscillations could test the general picture.","Beyond the paper, if the U-6d 3D sheet is the topological carrier, its weak spectral weight implies that the topological response is tunable by anything that shifts Kondo coherence, such as pressure, disorder, or magnetic field, so the superconducting topological phase may be more fragile than a band-structure calculation would suggest.","Beyond the paper, extending the same calculation to the superconducting channel would show whether the coexistence of a quasi-2D Te2-5p cylinder and a 3D U-6d dumbbell at $\\Gamma$ stabilizes odd-parity pairing or changes the topological invariant; this is a direct, testable extension of the authors' call for further study."],"forward_implications":["The conflicting experimental reports are not contradictory: low-temperature quantum-oscillation and magneto-conductance measurements can be dominated by the non-$\\Gamma$-centered quasi-2D cylindrical sheet, while temperature-dependent and orbital-selective probes can resolve the $\\Gamma$-centered 3D and quasi-2D sheets.","Topological superconductivity in UTe$_2$ is not ruled out by reports of quasi-2D Fermi surfaces, because the calculation finds a clear $\\Gamma$-centered 3D U-6d Fermi surface at low temperature that can host the required odd-parity pairing.","The appearance of the quasi-2D Te2-5p sheet is a Kondo-coherence effect: below roughly 50 K, coherent hybridization along the $c$ axis turns a closed 3D pocket into a quasi-2D cylinder, with no structural transition involved.","Spin-orbit coupling is essential to this reconstruction: in the calculation without spin-orbit coupling, the $\\Gamma$-centered Te2-5p quasi-2D sheet does not form.","Below 25 K, the two $\\Gamma$-centered sheets are expected to gain spectral weight and quasiparticle lifetime as Kondo coherence strengthens, making them more observable even though they remain weaker than the non-$\\Gamma$-centered cylinder."],"supporting_citations":[{"why":"Establishes the orbital-selective Kondo mechanism and the roughly 500 K Kondo scale, including the c-axis Te2-5p/U-5f hybridization central to the reconstruction.","marker":"[14]"},{"why":"Supplies the LQSGW+DMFT method used to compute all temperature-dependent Fermi surfaces and spectral functions.","marker":"[25]"},{"why":"Provides quantum-oscillation evidence for coexisting 3D and 2D Fermi-surface pockets that the paper explains with two $\\Gamma$-centered sheets.","marker":"[4]"},{"why":"Reports a low-temperature measurement finding only a quasi-2D Fermi surface, the conflicting observation the calculation is designed to reconcile.","marker":"[5]"},{"why":"Reports high-field oscillation measurements that see only quasi-2D Fermi-surface sheets, another conflicting observation.","marker":"[6]"},{"why":"Presents an ARPES study reporting a quasi-2D Fermi surface, one of the experimental results the orbital-selective picture must match.","marker":"[3]"},{"why":"Gives a prior correlated calculation with a $\\Gamma$-centered 3D Fermi surface connected to topological superconductivity, against which the new U-6d dumbbell is compared.","marker":"[8]"},{"why":"Gives a DFT+U calculation producing only non-$\\Gamma$-centered quasi-2D cylindrical sheets, which the persistent cylindrical sheet in this paper matches.","marker":"[10]"}],"fun_headline_variants":["UTe2's dual Fermi surfaces resolve conflicting experiments","Orbital-selective Kondo gives UTe2 both 3D and quasi-2D Fermi surfaces","At 25 K, UTe2 splits Fermi surfaces into 3D and quasi-2D","Orbital-selective Kondo flips one UTe2 Fermi surface to quasi-2D","Why UTe2 shows both 2D and 3D Fermi surfaces"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the contours drawn from the orbital-projected spectral weight $A(\\mathbf{k},\\omega=0)$ at a chosen isovalue are genuine quasiparticle Fermi surfaces rather than artifacts of the visualization cutoff, a premise the paper does not establish because it never specifies the threshold, the $\\Gamma$-centered features carry weak spectral weight, and temperatures below 25 K are inaccessible due to the sign problem.","fun_headline_variants_meta":{"raw":{"variants":["UTe2's dual Fermi surfaces resolve conflicting experiments","Orbital-selective Kondo gives UTe2 both 3D and quasi-2D Fermi surfaces","At 25 K, UTe2 splits Fermi surfaces into 3D and quasi-2D","Orbital-selective Kondo flips one UTe2 Fermi surface to quasi-2D","Why UTe2 shows both 2D and 3D Fermi surfaces"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000431,"raw_usage":{"total_tokens":2248,"prompt_tokens":1038,"completion_tokens":1210,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":654,"completion_tokens_details":{"reasoning_tokens":1097}},"tokens_in":654,"tokens_out":1210,"duration_ms":9819,"temperature":1.0,"reasoning_tokens":1097,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T04:14:04.514673+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the orbital-projected spectral weight at several isovalues and locate the poles of $A(\\mathbf{k},\\omega=0)$: if the $\\Gamma$-centered Te2-5p cylinder and U-6d dumbbell disappear at slightly higher cutoffs or appear only as shoulders rather than peaks, the coexistence is a visualization artifact, whereas if the poles persist and a sub-25 K quantum-oscillation measurement resolves the closed 3D pocket, the claim stands.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the orbital-selective Kondo mechanism and the roughly 500 K Kondo scale, including the c-axis Te2-5p/U-5f hybridization central to the reconstruction."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the LQSGW+DMFT method used to compute all temperature-dependent Fermi surfaces and spectral functions."},{"cited_title":"Broyles, Z","cited_arxiv_id":null,"evidence_quote":"Provides quantum-oscillation evidence for coexisting 3D and 2D Fermi-surface pockets that the paper explains with two $\\Gamma$-centered sheets."},{"cited_title":"Eaton, T","cited_arxiv_id":null,"evidence_quote":"Reports a low-temperature measurement finding only a quasi-2D Fermi surface, the conflicting observation the calculation is designed to reconcile."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports high-field oscillation measurements that see only quasi-2D Fermi-surface sheets, another conflicting observation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives a prior correlated calculation with a $\\Gamma$-centered 3D Fermi surface connected to topological superconductivity, against which the new U-6d dumbbell is compared."}],"review_version":1}