{"id":"e4eb78d1-5656-4b1b-b436-f5e77656ac6c","arxiv_id":"2412.18674","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Optimizing stellarators against trapped-electron instability suppresses TEM turbulence but leaves universal instabilities that drive substantial heat flux, which a moderate rise in beta can greatly reduce.","lead":"This paper uses numerical optimization to design two stellarator magnetic shapes in which trapped-electron turbulence is suppressed, then shows a different instability takes over and drives large heat loss. It matters because stellarator designers need fast optimization targets, and this work reveals that suppressing one instability can simply expose another, which can be controlled by raising plasma pressure.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Optimized configurations are local p'=0 equilibria at a single surface; global embedding could alter shear, curvature, or trapping and undo the TEM suppression and UI dominance.","rationale":"Both the reader and this pass identify the same load-bearing assumption: the optimized equilibria are local, p'=0, single-surface constructions, and the paper's conclusions about TEM suppression and UI dominance are only shown in that local geometry. The mode-identification procedure is internally consistent and cross-checked (cross phases, geometry-response tests, and eigenvalue behavior), and the converged density-gradient nonlinear runs give credible support for the UI-driven heat-flux claim. The additional limitations—unconverged omega_Te=4 nonlinear runs, the unexplained beta=5e-4 PT heat-flux spike, and the lack of released code/data—are real but secondary; they do not threaten the main density-gradient result. The local-to-global embedding, however, is the premise on which the entire optimization story rests. A global equilibrium could have different shear, curvature, or magnetic wells, all of which directly enter the AE metric and the gyrokinetic geometry. Without a global check, the configurations remain mathematical local solutions rather than demonstrated stellarator designs. This justifies the CONDITIONAL verdict.","tokens_in":22877,"tokens_out":12094,"duration_ms":117212,"concrete_test":"Obtain the optimized NT and PT surface shapes from the authors and run a global VMEC equilibrium with each as the plasma boundary (or as a fixed internal surface at s=0.5), with p'=0 or a small-beta profile. At the global s=0.5 surface, recompute f_TEM from Eq. (16) and repeat the GENE kx=0 linear scans at omega_n=4, omega_Te=0. If f_TEM rises above the HSX value (0.3702) or the fastest-growing mode at k_y about 0.1-0.5 is identified as a TEM rather than a UI, the central claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim depends on the optimized local equilibria remaining representative of a realizable stellarator. The objective function (Eq. 8) is minimized with p' fixed to zero at a single flux surface s=0.5, and the available-energy metric f_TEM (Eq. 16) is built from bounce-averaged drift frequencies (Eqs. 25-26) and curvatures (Eqs. 28-29) that are valid only for p'=0. Nothing demonstrates that these surfaces can be extended to a global equilibrium with nested flux surfaces; in a global solution, force balance couples the surface shape to the pressure and current profiles, so shear, integrated local shear Lambda, curvature components, and magnetic-well structure can all change. Because TEM suppression is achieved by reducing the trapped-electron free energy, and the UI branches are identified through their response to the same geometry, even a moderate change in the well structure at s=0.5 could restore TEM instability or shift the dominant mode. The paper reports no global-equilibrium check (e.g., VMEC) and ships no equilibrium files, so the local-to-global transferability is entirely unverified.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports two local 3D stellarator equilibria, one with helically rotating negative triangularity (NT) and one with positive triangularity (PT), obtained by optimizing a local MHD equilibrium for quasihelical symmetry and the available energy of trapped electrons. The authors verify TEM suppression with linear and nonlinear gyrokinetic simulations: in the density-gradient-only scenario (ωn=4, ωTe=0), the dominant low-ky instabilities in the optimized configurations are identified as toroidal universal instabilities (NT) and slab universal instabilities (PT), which drive substantial electrostatic heat flux, while HSX remains TEM-dominated. A β scan shows that moderate β reduces or nearly suppresses this UI-driven heat flux, with electromagnetic heat flux remaining subdominant. The paper concludes that future optimizations targeting electrostatic drift-wave turbulence should consider UIs as well as TEMs.","tokens_in":23057,"tokens_out":5138,"duration_ms":53008,"significance":"If the result holds, this is a valuable demonstration that a fast, geometry-based available-energy objective can be used in stellarator optimization to suppress TEM-driven turbulence, and it usefully identifies UIs as a possible unintended consequence of such optimization. The paper's strengths are its converged nonlinear gyrokinetic simulations for the ωn=4, ωTe=0 scenario, the careful mode-identification procedure based on electron cross phases and controlled geometry modifications (constant-B and slab-like limits), and the clear, falsifiable prediction that UIs dominate at low ky and are stabilized by moderate β. The available-energy target is taken from prior published work, so there is no circularity in the suppression claim. The main caveats are that the optimized equilibria are local, single-flux-surface, p'=0 equilibria, and that all linear identification and β-scan results use only kx=0 modes.","major_comments":[{"comment":"The optimization is performed on local 3D equilibria with p' fixed to zero at a single flux surface s=0.5 (Eq. 8 and Sec. II.1), and the available-energy target fTEM relies on curvature expressions Eqs. (28)–(29) that are valid only for p'=0. The manuscript never demonstrates that these optimized surfaces can be embedded in a global equilibrium with nested flux surfaces. In a global solution, force balance couples the surface shape to pressure and current profiles, so global shear, integrated local shear Λ, curvature components, and the magnetic-well structure can all change; since TEM suppression is achieved by modifying exactly these quantities, the reported suppression and UI dominance may not persist. The authors should either add a global-equilibrium check (e.g., with VMEC) on the same surface, or explicitly re-scope all claims from 'configurations' to 'local flux-tube equilibria' throughout the abstract and conclusions.","section":"Sec. II.1–II.2"},{"comment":"All linear eigenvalue results, including the mode-identification procedure and the β scan, are restricted to kx=0; the text states 'Only modes centered at kx=0 were computed.' The nonlinear cross-phase histograms in Figs. 8 and 9 show discrepancies from the linear kx=0 cross phases at 0.4 ≤ ky ≤ 0.7, which the authors attribute to 'clusters of subdominant UIs or UIs at finite kx.' This leaves open the possibility that finite-kx modes are the true dominant instabilities at low ky, which would affect both the UI-dominance claim and the β-stabilization results in Sec. V. The authors should justify the kx=0 restriction by presenting kx spectra for representative ky values, or scan kx for at least a few ky values.","section":"Sec. III and Sec. V"}],"minor_comments":[{"comment":"The abstract and conclusions describe the optimized objects as 'configurations' without noting that they are local, single-flux-surface equilibria; adding a qualifier would prevent over-interpretation.","section":"Abstract and Sec. II.2"},{"comment":"There are several typos: 'configuartion' in Sec. II.2, 'hare' for 'are' in Sec. III.1, 'decease' for 'decrease' in Sec. V, and 'occurrs' in Appendix A.1.","section":"Sec. II.2, Sec. III.1, Sec. V, Appendix A.1"},{"comment":"The caption states 'with ωn = 0, ωT e = 4', but the Sec. V text states the scan uses ωn = 4 and ωTe = 0; this should be corrected.","section":"Fig. 12 caption"},{"comment":"The Fig. 24 caption says the 'NT configuration' although the surrounding text describes the PT equilibrium; also, in Appendix A.1 the growth-rate discussion for the PT case is attributed to Fig. 20, but Fig. 20 displays the NT case. Please re-check the figure/table numbering and cross-references.","section":"Fig. 24 caption and Appendix A.1 text"},{"comment":"The nonlinear simulations for the ωn=0, ωTe=4 scenario are stated to be not converged; the claim that there are 'no significant fluctuations observed at scales with ky ≤ 2' should therefore be presented more cautiously or supported with a resolution check.","section":"Sec. IV"}],"recommendation":"major_revision","confidential_remarks":"The central result is plausible and the local gyrokinetic evidence is solid, but the lack of a global-equilibrium embedding test is a real risk for the paper's central claim, and the kx=0 restriction is a second load-bearing gap. I would ask the authors to add a VMEC or equivalent check, or to explicitly limit all claims to local equilibria, and to address the kx dependence of the dominant modes. Depositing the equilibrium input files would also improve reproducibility."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a credible, well-scoped computational demonstration that optimizing local quasihelically symmetric equilibria against the trapped-electron available-energy metric actually suppresses TEMs in gyrokinetic simulations, and it shows, maybe more importantly, that universal instabilities then take over and that beta can control them. It is not a global stellarator design, and the authors do not pretend otherwise. If you judge it as a proof-of-principle, it mostly holds.\n\nThe genuinely new parts are the two optimized equilibria, the NT-vs-PT comparison (negative triangularity gives no tokamak-like benefit once AE is reduced), and the UI/beta story. The AE metric and local-equilibrium optimization are prior work from the same program, so novelty is incremental but real.\n\nThe paper does several things carefully. Mode identification uses cross phases plus controlled geometry alterations (constant-B, slab-like), which is the right way to separate TEMs from UIs. The main nonlinear runs (omega_n=4) are stated to be converged, and the comparison against HSX gives a useful baseline. They also disclose their own weak spots, which I respect: the omega_Te nonlinear runs are unconverged, and the PT beta=5e-4 flux spike is unexplained.\n\nSoft spots, in order of weight. First, the equilibria are local, p'=0, single-surface shapes. Nothing demonstrates that they can be embedded in a global equilibrium without changing shear, curvature, or well structure enough to undo the TEM suppression. This is the biggest caveat and it is not addressed. Second, linear runs are kx=0 only; finite-kx and subdominant modes could matter, although the nonlinear cross-phase analysis partially covers this. Third, the ETG/temperature-gradient story rests on unconverged nonlinear runs, so it should be treated as preliminary. Fourth, no code, data, or equilibrium files are shipped, which limits reproducibility for a paper whose main artifact is two geometries.\n\nThese are real but not fatal for a proof-of-principle. The central claim—that available-energy optimization can suppress TEMs in local QHS equilibria and that UIs then dominate—is supported by the evidence presented. I'd send it to peer review and ask the authors to add a global-embedding check, finite-kx linear runs, and a better story on the PT beta spike.","headline":"A credible proof-of-principle that available-energy optimization suppresses TEMs in local QHS equilibria; the local-to-global embedding question is the main caveat, not the UI or beta results.","tokens_in":23698,"tokens_out":2155,"would_cite":true,"duration_ms":20747,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["52.35.Ra","52.55.Hc","52.65.Tt"],"model":"deepseek-v4-flash","headline":"Two optimized 3D stellarator shapes suppress trapped-electron-mode turbulence, exposing universal instabilities that a modest rise in beta reduces.","keywords":["stellarator","trapped electron mode","quasihelical symmetry","available energy","universal instability","gyrokinetic turbulence","beta stabilization","helical triangularity"],"falsifier":"Embed the optimized local flux-surface shapes in a global equilibrium and repeat the linear and nonlinear gyrokinetic scans at $\\omega_n=4$, $\\omega_{T_e}=0$; if low-$k_y$ trapped-electron modes with TEM cross-phases reappear with growth rates near the reference configuration, or if raising $\\beta$ to $4\\times10^{-3}$ fails to reduce the heat flux, the local-optimization claim is not robust.","tokens_in":22600,"feed_emoji":"🌀","tokens_out":9393,"duration_ms":80699,"temperature":0.7,"pith_summary":"The paper reports two three-dimensionally shaped stellarator configurations, one with negative and one with positive helical triangularity, generated by optimizing a local equilibrium for quasihelical symmetry and for the available energy of trapped electrons. In gyrokinetic simulations the trapped electron mode (TEM) is suppressed in both, and the sign of the triangularity no longer matters the way it does in tokamaks. With a strong density gradient and no temperature gradient, the dominant low-$k_y$ instabilities become universal instabilities, which drive substantial heat flux; a moderate increase in plasma beta halves that heat flux in the negative-triangularity case and nearly eliminates it in the positive-triangularity case.","feed_headline":"Shaping halts TEM turbulence; beta then halves its replacement heat","feed_subtitle":"Optimized 3D shapes suppress trapped-electron modes, leaving universal instabilities that a modest beta increase tames.","key_machinery":"The load-bearing object is the available-energy metric $f_{\\mathrm{TEM}}$, an integral over trapped-electron pitch angles and wells of the free energy available to collisionless trapped-electron fluctuations, built from the normalized bounce-averaged drift frequencies $\\hat{\\omega}_x$, $\\hat{\\omega}_y$ and the diamagnetic drift frequency $\\hat{\\omega}_*^T$. Because it depends only on geometry and background gradients, it is cheap enough to serve as an optimization target, and it is correlated with nonlinear TEM heat flux through $Q \\propto f_{\\mathrm{TEM}}^{3/2}$. The optimization minimizes $f_{\\mathrm{TEM}}$ alongside a local quasisymmetry measure and penalty terms for shear, rotational transform, parallel current, aspect ratio, and flux-surface regularity, within a local three-dimensional MHD equilibrium with pressure gradient held at zero.","core_discovery":"The central claim is that minimizing the available energy of trapped electrons, together with a local measure of quasisymmetry, is sufficient to produce quasihelically symmetric stellarator equilibria whose TEM-driven turbulence is suppressed in gyrokinetic simulations. The two optimized equilibria cut the available-energy metric $f_{\\mathrm{TEM}}$ by more than an order of magnitude relative to the starting equilibria and below the level of the reference quasihelically symmetric configuration, even though the two flux-surface shapes keep opposite signs of helical triangularity. Linear mode identification, based on cross-phases and on artificial removal of particle trapping and curvature, then shows that at $\\omega_n = 4$, $\\omega_{T_e} = 0$ the low-$k_y$ instabilities in these configurations are toroidal universal instabilities (negative triangularity) and slab universal instabilities (positive triangularity), not trapped electron modes. Nonlinear simulations show that universal instabilities can drive electron heat flux more than an order of magnitude larger than the TEM flux in the reference configuration, and that raising $\\beta$ to a few times $10^{-3}$ halves (negative triangularity) or nearly suppresses (positive triangularity) that flux while electromagnetic heat flux remains at least an order of magnitude smaller.","pith_inferences":["A natural extension is to build a complementary available-energy-style metric for the passing-electron density-gradient free energy that drives universal instabilities, and minimize it together with the TEM target; the paper leaves that combination unexplored.","Because the optimization was performed on a single flux surface with $p'=0$, the practical payoff depends on whether the optimized shape survives embedding in a global equilibrium; that is a testable next step not carried out here.","The beta dependence suggests an operational design rule: in quasihelically symmetric devices with strong density gradients, operating near a few times $10^{-3}$ in beta may suppress both TEMs and UIs, while too-low beta leaves UI transport and too-high beta risks electromagnetic modes."],"forward_implications":["Available-energy optimization can be used as a fast, geometry-only proxy for TEM stability in stellarator design, replacing much more expensive gyrokinetic objective evaluations.","In these reduced-TEM equilibria, the sign of helical triangularity (negative vs positive) is not the controlling factor for TEM suppression or turbulence; available energy is.","At low beta and strong density gradient, the instability that limits confinement in such optimized configurations may be the universal instability rather than a TEM, so reduced TEM activity alone does not guarantee low transport.","A moderate increase in plasma beta (a few times $10^{-3}$) can substantially reduce UI-driven electrostatic heat flux without introducing significant electromagnetic heat flux, although in the positive-triangularity case beta near $10^{-2}$ excites a kinetic ballooning mode.","Future stellarator optimizations aimed at electrostatic drift-wave transport should include a UI target, or operate at sufficiently high beta, when density gradients are large."],"supporting_citations":[{"why":"Supplies the available-energy metric $f_{\\mathrm{TEM}}$ used as the TEM-optimization target, including its scaling with heat flux.","marker":"[47]"},{"why":"Provides the local three-dimensional MHD equilibrium model in which the flux surfaces are optimized and simulated.","marker":"[44]"},{"why":"Defines the reference quasihelically symmetric configuration and its TEM-driven turbulence baseline against which the reduced-TEM equilibria are compared.","marker":"[23]"},{"why":"Contributes the cross-phase and geometry-modification procedure used to identify universal instabilities and distinguish them from trapped electron modes.","marker":"[33]"},{"why":"Supplies the local quasisymmetry measure combined with available energy in the objective function.","marker":"[50]"},{"why":"Provides nonlinear reference heat-flux and cross-phase data used for the comparison with the reduced-TEM configurations.","marker":"[53]"}],"fun_headline_variants":["Optimized 3D shapes kill TEMs, but universal instabilities lurk","Negative triangularity fails to tame stellarator turbulence like tokamaks","Beta boost tames universal instabilities in optimized stellarators","TEM-free stellarators still heat leak via universal instabilities","3D shaping suppresses TEMs, then beta halves leftover flux"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that suppressing the available energy of trapped electrons on one local flux surface with zero pressure gradient, at one chosen gradient pair, is enough to suppress TEM-driven turbulence in a physically realized quasihelically symmetric stellarator; global equilibrium effects could change trapping, shear, or curvature and undo the suppression.","fun_headline_variants_meta":{"raw":{"variants":["Optimized 3D shapes kill TEMs, but universal instabilities lurk","Negative triangularity fails to tame stellarator turbulence like tokamaks","Beta boost tames universal instabilities in optimized stellarators","TEM-free stellarators still heat leak via universal instabilities","3D shaping suppresses TEMs, then beta halves leftover flux"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000358,"raw_usage":{"total_tokens":1987,"prompt_tokens":1042,"completion_tokens":945,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":658,"completion_tokens_details":{"reasoning_tokens":856}},"tokens_in":658,"tokens_out":945,"duration_ms":6761,"temperature":1.0,"reasoning_tokens":856,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T04:34:37.428041+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Embed the optimized local flux-surface shapes in a global equilibrium and repeat the linear and nonlinear gyrokinetic scans at $\\omega_n=4$, $\\omega_{T_e}=0$; if low-$k_y$ trapped-electron modes with TEM cross-phases reappear with growth rates near the reference configuration, or if raising $\\beta$ to $4\\times10^{-3}$ fails to reduce the heat flux, the local-optimization claim is not robust.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the available-energy metric $f_{\\mathrm{TEM}}$ used as the TEM-optimization target, including its scaling with heat flux."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the reference quasihelically symmetric configuration and its TEM-driven turbulence baseline against which the reduced-TEM equilibria are compared."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides nonlinear reference heat-flux and cross-phase data used for the comparison with the reduced-TEM configurations."}],"review_version":1}