{"id":"b8f27f27-1dfa-4532-ad7f-feb0e5b460f0","arxiv_id":"1908.10253","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"A second-order near-axis expansion is used to rapidly construct quasisymmetric stellarator configurations with record-low symmetry-breaking and the first numerical confirmation of the 1/A^3 scaling.","lead":"This paper presents a fast method to design quasisymmetric stellarator magnetic fields using a higher-order expansion about the magnetic axis. The method constructs boundary shapes that achieve exceptionally small symmetry-breaking fields at high aspect ratio, which could speed up the search for practical fusion device designs.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Reported 1e-7 T symmetry-breaking amplitudes lack resolution-convergence evidence; the record claim and high-aspect-ratio scaling points may rest on a numerical noise floor.","rationale":"The central claim is that the construction achieves the ideal 1/A^3 scaling and record-low symmetry-breaking. The numerical evidence for the scaling is strongest where the residual amplitudes are far above any plausible noise floor, but the paper explicitly stakes the record on the A=320 point, where the amplitudes are about 2e-7 T. Without resolution studies, the reader cannot distinguish a true physical amplitude from a numerical floor. The paper's own admission of convergence problems for the QH case at high A makes this a live risk, not a hypothetical. The proposed test directly checks convergence at the two highest aspect ratios; if the amplitudes move, the record claim must be withdrawn or qualified and the scaling demonstration should be quoted only over the converged range. If the amplitudes are stable, the concern is resolved and the record claim stands. The Appendix B 'generally linearly independent' statement is a secondary gap; it does not affect the empirical demonstration but should be clarified in revision.","tokens_in":36323,"tokens_out":23060,"duration_ms":215468,"concrete_test":"Recompute the A=320 and A=160 quasi-axisymmetric equilibria (Sec. 5.2) with VMEC resolution increased by a factor of two or more in both radial (e.g., Ns=99 vs 49) and poloidal/toroidal mode numbers, and compute the Boozer spectrum with BOOZ_XFORM on a finer grid. If the largest single symmetry-breaking mode amplitude or Stot changes by more than 50%, or if the amplitudes rise above 2e-7 T, the record claim and the high-A scaling points are not converged. For additional confidence, cross-check with an independent equilibrium solver such as DESC or SPEC on the same boundary.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline quantitative results—the smallest-ever-reported symmetry-breaking amplitudes (<2e-7 T) and the clean 1/A^3 scaling down to that level—depend on VMEC and BOOZ_XFORM resolving relative amplitudes of about 1e-7 in a 1 T field at aspect ratio 320. No convergence tests with respect to VMEC radial/poloidal resolution or BOOZ_XFORM grid size are reported. The authors acknowledge convergence difficulty for the quasi-helical case at high A (Sec. 5.4), so the reliability of the quasi-axisymmetric A=320 point is not established. If those tiny amplitudes are dominated by numerical noise, the 'smallest ever reported' claim is unsupported, and the claimed demonstration of 1/A^3 scaling over the full range reduces to a shorter interval. A secondary concern is the unproven 'generally linearly independent' assertion for the 6x6 linear system in Appendix B, which is necessary for the finite-a boundary correction to yield the desired B through O((r/R)^2). This gap affects the generality of the method, though not the empirical demonstrations.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper extends the Garren–Boozer near-axis expansion for quasisymmetric stellarators from O(r/R) to O((r/R)^2), deriving a streamlined set of equations (Appendix A) and a finite-minor-radius boundary construction that includes a carefully chosen subset of O((r/R)^3) terms (Section 3, Appendix B). The construction is solved numerically in milliseconds, and the resulting boundary shapes are fed into VMEC/BOOZ_XFORM to verify that the computed equilibria have the desired Boozer field-strength spectrum. The paper reports the first numerical demonstrations of the predicted 1/A^3 scaling of quasisymmetry-breaking modes, examples of quasi-axisymmetric, quasi-helically symmetric, tokamak-stellarator hybrid, and non-stellarator-symmetric configurations, and a claim of the smallest-ever symmetry-breaking amplitudes (<2e-7 T) in a strongly nonaxisymmetric vacuum equilibrium.","tokens_in":36589,"tokens_out":7051,"duration_ms":73854,"significance":"If the results are taken at face value, this is a substantial advance: it provides a fast, direct, analytically grounded method for generating high-precision quasisymmetric stellarator geometries, gives the first numerical evidence for the Garren–Boozer 1/A^3 scaling, and demonstrates quasisymmetry quality far beyond what conventional optimization has reported. The manuscript's strengths include a detailed and largely self-contained analytic derivation in Appendix A, a nontrivial finite-a correction in Section 3 with supporting analysis in Appendix B, and VMEC/BOOZ_XFORM checks spanning quasi-axisymmetry, quasi-helical symmetry, finite pressure/current, and broken stellarator symmetry. The method's speed and analytic output are genuinely useful for stellarator design and for initializing conventional optimization.","major_comments":[{"comment":"The abstract's central claims—the first numerical demonstration of the 1/A^3 scaling and the 'smallest ever reported' symmetry-breaking amplitudes (<2e-7 T)—rest on VMEC/BOOZ_XFORM resolving relative amplitudes of order 10^-7 in a 1 T field at A=320, but no resolution-convergence study is reported. The authors note in §5.4 that converged VMEC values are difficult to obtain for very small symmetry-breaking modes at the highest aspect ratios, so it is not established that the A=320 quasi-axisymmetric point is above the numerical noise floor. Please provide convergence tests with respect to VMEC radial/poloidal resolution and BOOZ_XFORM grids, state the estimated numerical uncertainty of each point in Fig. 8, and qualify the record/scaling claims accordingly.","section":"§5.2, Fig. 8; Eq. (5.4)"},{"comment":"The construction is described as achieving quasisymmetry 'fully' through O((r/R)^2), yet the paper states in §4.1 that exact constancy of B20 is not proved and that only minimization of its toroidal variation is performed. Residual φ-dependence of B20 contributes a symmetry-breaking term scaling as 1/A^2 (seen for Config 2 in Fig. 4 at large A), so the asymptotic scaling is not the claimed 1/A^3 unless B20 is exactly constant. Either prove existence/construct a solution with constant B20, or reframe the Garren–Boozer scaling claim as a finite-aspect-ratio result over the demonstrated range.","section":"§4.1, §5.2, Fig. 4"},{"comment":"The proof that the finite-a boundary correction yields the desired field strength through O((r/R)^2) depends on the assertion that the six homogeneous linear equations for {ξ0, ξs, ξc, γ0, γs, γc} are 'generally linearly independent.' No proof or explicit condition is given, and a singular or near-singular system would invalidate the uniqueness step. Please supply a proof or a numerical verification of the determinant at the operating points used in Section 5, or state the conditions under which the system is nonsingular.","section":"Appendix B, paragraph following (B27)–(B28)"},{"comment":"The 'smallest ever reported' superlative is not supported by a systematic quantitative comparison with previously published quasisymmetric configurations. Please cite the previous best published symmetry-breaking amplitudes using the same measure (e.g., Eq. (5.4)) and give the comparison criterion, or soften the claim to 'smaller than the values we have traced in the literature.'","section":"Abstract; §5.2; §6"}],"minor_comments":[{"comment":"The quasi-helically symmetric configuration is reported to be very sensitive to rounding of the axis coefficients from four to three significant digits; please quantify this sensitivity (for example, the resulting change in X3/Y3 and in the boundary shape) and discuss its implications for reproducibility of the numerical examples.","section":"§5.4"},{"comment":"The section title 'Quasi-axisymmetry fully through O((r/R)^2)' conflicts with the text, which notes a 'small remaining toroidal variation' of B20; the terminology should be adjusted to reflect that the configuration is optimized rather than exactly quasisymmetric to that order.","section":"§5.2"},{"comment":"The scaling plot would be much more informative with estimated error bars or shaded uncertainty bands derived from the requested convergence studies; currently the reader cannot distinguish genuine scaling from numerical floor effects.","section":"Fig. 8"},{"comment":"Please define the symbol N used in the mode-selection condition n ≠ N m explicitly at the point of first use in Section 5, since N denotes the quasisymmetry helicity earlier in the paper.","section":"Eq. (5.2)"}],"recommendation":"major_revision","confidential_remarks":"This is a strong and likely influential paper, and the analytic derivation appears sound. The main risk is overclaiming numerical precision and asymptotic scaling from results that lack convergence evidence and exact B20 constancy. I would be happy to support publication after the authors add resolution-convergence tests, provide a systematic comparison for the 'smallest ever reported' claim, and either prove or clearly delimit the generality of the Appendix B linear-independence step. The paper is well within the scope of J. Plasma Phys."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this paper actually delivers what it promises. Landreman and Sengupta solve the O((r/R)^2) Garren-Boozer equations that were derived but never solved, and they verify the result with VMEC/BOOZ_XFORM across a range of configurations. The 1/A^3 scaling of quasisymmetry-breaking is demonstrated numerically for the first time, and the evidence is convincing. This is a real step for fast stellarator design. What is new and good: the streamlined derivation in Appendix A is a genuine contribution, not just housekeeping. The finite-minor-radius boundary correction in Section 3, retaining a specially chosen piece of the (X3,Y3) shape, is clever and apparently necessary; without it the achieved field strength misses at O((r/R)^2). The numerical checks are thorough: B20 extracted from VMEC converges to the constructed function as A increases, in multiple settings including finite pressure, finite current, and non-stellarator-symmetric cases. The quasi-helical example is especially demanding, and it also tracks the predicted scaling. The method is orders of magnitude faster than conventional optimization, and the paper is honest about the limitations of the construction at finite aspect ratio. Soft spots, in proportion. The stress-test note about resolution convergence is a fair hit. The 'smallest ever reported' claim (<2e-7 T at A=320) rests on VMEC and BOOZ_XFORM resolving mode amplitudes around 1e-7 relative to a 1 T field. No resolution scan is shown for the A=320 QA point, and the authors themselves note convergence difficulties for the QH case at high A. I would not call this fatal, since the 1/A^3 scaling is visible over a wide range even if the last point is noisy, but the record claim should be softened or backed with a convergence study. The unproven 'generally linearly independent' assertion for the 6x6 system in Appendix B is a smaller issue: it is a gap in the proof of generality, not in the empirical demonstrations. Also, exact constancy of B20 is not proved; the authors say so themselves, which is fine, but the claim that it can be made arbitrarily small is heuristic. No code or data is released, which is a minor reproducibility complaint for a paper that would benefit enormously from it. Who this is for: plasma physicists working on stellarator design, near-axis expansions, or quasisymmetry theory. If that is your area, read it. If not, the scaling result is still a nice piece of applied mathematics. Recommendation: send it to serious peer review. The central argument holds up; the paper deserves referee time. The referees should push for either a resolution-convergence check on the A=320 point or a more modest wording of the record claim, and a clean statement of the linear-independence assumption in Appendix B.","headline":"Genuinely new O((r/R)^2) near-axis construction with strong numerical verification; the record-low quasisymmetry-breaking claim needs a resolution-convergence check before it can be trusted, but the central scaling result holds.","tokens_in":852,"tokens_out":859,"would_cite":true,"duration_ms":23959,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["52.55.Hc","52.30.Cv"],"model":"deepseek-v4-flash","headline":"This paper claims that a second-order near-axis construction with a corrected finite-radius boundary builds quasisymmetric stellarator equilibria whose symmetry-breaking modes follow the ideal $1/A^3$ scaling, reaching amplitudes below…","keywords":["quasisymmetry","stellarator","near-axis expansion","aspect-ratio scaling","magnetohydrodynamic equilibrium","Boozer coordinates","stellarator optimization"],"falsifier":"For an axis shape not used in the paper, construct the boundary at several aspect ratios and compute the full three-dimensional equilibrium inside it; if the total quasisymmetry-breaking measure does not fall as the cube of the inverse aspect ratio, or if the individual $m=0$ and $m>0$ measures deviate from the predicted $1/A^2$, $1/A^3$, and $1/A^4$ scalings, the central scaling claim is wrong. A more elementary check is whether $[B_{m=0}(\\phi,r=a)-B(\\phi,r=0)]/a^2$ stops converging to the predicted $B_{20}(\\phi)$ as $A$ grows.","tokens_in":36147,"feed_emoji":"🧲","tokens_out":10387,"duration_ms":95695,"temperature":0.7,"pith_summary":"Quasisymmetric stellarators confine plasma using a magnetic field whose strength has a continuous symmetry, even though the field itself does not, avoiding the large currents that make tokamaks unstable. This paper aims to show that such configurations can be built directly, without iterating a three-dimensional equilibrium code, by solving the near-axis expansion through second order in $r/R$ and applying a corrected boundary construction. If the construction works, it replaces an expensive optimization search with an analytic procedure that runs in milliseconds and provides the first numerical confirmation that symmetry-breaking errors scale as the cube of the inverse aspect ratio. The reported symmetry-breaking amplitudes, below $2\\times 10^{-7}$ T in a strongly nonaxisymmetric configuration, would mean quasisymmetry can be made essentially exact at high aspect ratio.","feed_headline":"Near-axis method hits 1/A^3 quasisymmetry scaling","feed_subtitle":"Boundary surfaces computed in milliseconds yield symmetry-breaking modes below 2e-7 tesla.","key_machinery":"The central object is the near-axis expansion about the magnetic axis: the position vector is written in the Frenet-Serret frame as $\\mathbf{r} = \\mathbf{r}_0(\\phi) + X(r,\\vartheta,\\phi)\\mathbf{n} + Y(r,\\vartheta,\\phi)\\mathbf{b} + Z(r,\\vartheta,\\phi)\\mathbf{t}$, with $X$, $Y$, $Z$, and the field strength $B$ expanded in powers of $r/R$ and the equilibrium and quasisymmetry conditions imposed order by order in Boozer coordinates $(\\theta,\\phi)$. The load-bearing mechanism is the finite-radius boundary correction of Section 3: substituting a finite minor radius $a$ into the truncated second-order expansion gives the leading field strength a spurious $a^2$ contribution, and including $X_3$, $Y_3$ proportional to $X_1$, $Y_1$ with the coefficient $\\lambda = -Q B_0/(2 s_G \\bar B)$ cancels it, so the true equilibrium inside the constructed boundary matches the target field through $O((r/R)^2)$.","core_discovery":"The central claim is that a finite-aspect-ratio quasisymmetric stellarator can be generated from the second-order near-axis expansion by substituting a finite minor radius $a$ and adding a specific third-order correction: $X_3$ and $Y_3$ are set to $\\lambda$ times $X_1$ and $Y_1$ with $\\lambda = -Q B_0/(2 s_G \\bar B)$, which removes the spurious $a^2$ term in the leading field strength. With this corrected boundary, independent three-dimensional MHD equilibrium computations reproduce the intended Boozer-coordinate field strength through $O((r/R)^2)$, and the symmetry-breaking measures fall with aspect ratio as $1/A^3$ for $m>0$ modes, $1/A^4$ for on-axis mirror modes, and eventually $1/A^2$ where the residual toroidal variation of $B_{20}$ dominates. This is the first numerical demonstration of the predicted ideal scaling: quasisymmetry can be realized to arbitrary precision at sufficiently high aspect ratio, and the paper reports symmetry-breaking amplitudes below $2\\times 10^{-7}$ T in a strongly nonaxisymmetric configuration.","pith_inferences":["One extension the authors leave implicit is that the same finite-radius correction logic should apply to omnigenous targets, since Section 3 is formulated for an arbitrary desired field strength, not only quasisymmetry.","If the $1/A^3$ scaling is universal, the near-axis construction becomes a generator for mapping the solution space of stellarators rather than finding isolated examples, allowing systematic exploration of input parameters.","Deliberately allowing $B_0$ to vary toroidally and canceling it with $B_{20}$ at one radius would move quasisymmetry off-axis, which the paper notes as a possible advantage for fast-particle confinement but does not test.","The extreme sensitivity of the quasi-helical example to rounding of the axis-shape coefficients suggests that precision quasisymmetry at moderate aspect ratio requires fine-tuning; quantifying that sensitivity would help predict where the construction is practical."],"forward_implications":["Any axis shape with nonvanishing curvature can, in principle, be made quasisymmetric to arbitrary precision by choosing a sufficiently large aspect ratio.","Because the construction takes milliseconds per configuration, it enables rapid scans over families of stellarators and provides initial conditions for conventional optimization codes.","Second-order terms introduce triangularity and Shafranov shift, so generated shapes resemble previously optimized devices without fitting to them.","A vacuum configuration with $\\iota > 0.4$ can have symmetry-breaking mode amplitudes below $2\\times 10^{-7}$ T, and at a 5 T on-axis field the largest errors are about the size of the Earth's magnetic field.","The construction also works with nonzero pressure and current, producing a tokamak-stellarator hybrid with finite $\\beta$."],"supporting_citations":[{"why":"Derives the second-order quasisymmetry equations and the third-order obstruction that implies the $1/A^3$ scaling.","marker":"Garren & Boozer (1991a)"},{"why":"Provides the near-axis expansion of the magnetic field strength and equilibrium used throughout the construction.","marker":"Garren & Boozer (1991b)"},{"why":"The first-order near-axis construction that this paper extends to second order and to corrected finite-radius boundaries.","marker":"Landreman, Sengupta & Plunk (2019)"},{"why":"Supplies earlier direct-construction theory and the cylindrical-coordinate transformation groundwork.","marker":"Landreman & Sengupta (2018)"},{"why":"The VMEC equilibrium code used to compute full three-dimensional equilibria inside the constructed boundaries.","marker":"Hirshman & Whitson (1983)"},{"why":"The BOOZ_XFORM code used to transform VMEC results to Boozer coordinates so symmetry-breaking modes can be measured.","marker":"Sanchez et al. (2000)"}],"fun_headline_variants":["Record quasisymmetry: 2e-7 T breaking at finite aspect ratio","First proof of 1/A^3 quasisymmetry scaling in stellarators","Finite-aspect-ratio stellarators with quasisymmetry, no MHD code","Near-axis method achieves record precision without MHD code"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the truncated near-axis expansion, evaluated at a finite minor radius and corrected by the required third-order shape term, produces a boundary whose true plasma equilibrium realizes the intended field strength to the claimed order; the paper verifies this for several examples but does not prove it for arbitrary inputs.","fun_headline_variants_meta":{"raw":{"variants":["Record quasisymmetry: 2e-7 T breaking at finite aspect ratio","First proof of 1/A^3 quasisymmetry scaling in stellarators","Finite-aspect-ratio stellarators with quasisymmetry, no MHD code","Near-axis method achieves record precision without MHD code"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000963,"raw_usage":{"total_tokens":4166,"prompt_tokens":1076,"completion_tokens":3090,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":692,"completion_tokens_details":{"reasoning_tokens":3006}},"tokens_in":692,"tokens_out":3090,"duration_ms":21084,"temperature":1.0,"reasoning_tokens":3006,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:48:30.675300+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For an axis shape not used in the paper, construct the boundary at several aspect ratios and compute the full three-dimensional equilibrium inside it; if the total quasisymmetry-breaking measure does not fall as the cube of the inverse aspect ratio, or if the individual $m=0$ and $m>0$ measures deviate from the predicted $1/A^2$, $1/A^3$, and $1/A^4$ scalings, the central scaling claim is wrong. A more elementary check is whether $[B_{m=0}(\\phi,r=a)-B(\\phi,r=0)]/a^2$ stops converging to the predicted $B_{20}(\\phi)$ as $A$ grows.","supporting_citations":[{"cited_title":"Plasma Phys","cited_arxiv_id":null,"evidence_quote":"The first-order near-axis construction that this paper extends to second order and to corrected finite-radius boundaries."}],"review_version":1}