{"id":"a81bb956-8c8b-46a8-b618-86733a743dc0","arxiv_id":"2607.14722","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Reducing current in the coil nearest the straight section of HSX or W7-X creates giant edge islands and unpaired X-points far outside the previously studied configuration space.","lead":"Simulations show that cutting current in one specific coil in two stellarators—HSX and W7-X—can grow edge magnetic islands far beyond previously reported sizes, sometimes leaving X-points with no closed island. This reveals existing fusion machines can explore extreme divertor regimes without new hardware.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'exceeds all released W7-X configurations' claim rests on a finite-time separatrix tracing (d_min/d_max, §2.1) whose convergence is never checked; if the trace length is too short, the quantitative island-size ordering may be wrong.","rationale":"The reader's weakest assumption (coil-force feasibility) concerns the experimental prospect, not the central computational claim. The most load-bearing element of the central claim is the quantitative demonstration that the new configurations are larger than all released W7-X configurations (§4.3, Fig. 10). That demonstration depends entirely on the d_min/d_max metric, which is computed from finite-time manifold tracing with a manually chosen number of periods (§2.1). This choice is a free parameter with no sensitivity analysis. A concrete convergence test can settle whether the quantitative ordering is stable. If it is stable, the concern is resolved and the quantitative claim stands; if not, the paper's novelty rests on an unreliable metric. This is a sharper concern than coil forces because it targets the paper's main quantitative result rather than its speculative experimental outlook. The verdict remains CONDITIONAL in either case: the qualitative discovery of large islands and unpaired X-points is still supported by Poincaré plots, so no rejection is warranted, but the strength of the 'new class' claim should be conditioned on the metric robustness.","tokens_in":16604,"tokens_out":11039,"duration_ms":111140,"concrete_test":"Recompute d_min and d_max for YFK and for a representative released configuration such as CYM using, e.g., 2, 4, 6, and 8 field periods in the separatrix trace (and with the same root-finding parameters). If either metric changes by more than 10–15% between consecutive trace lengths, or if the ordering YFK > CYM reverses, then the quantitative 'exceeds all released configurations' claim is not robust and should be restated qualitatively.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's quantitative novelty—that the new islands are 'spatially larger than previously reported' and that YFK has d_max ≈ 50 cm versus <30 cm for all released configurations (Figs. 5, 10)—is computed from d_min/d_max defined in §2.1. The inner separatrix is found by tracing X-point eigenvectors 'for a finite number of field periods, selected so that the inner separatrix is well-sampled but before the chaotic lobes become warped.' This stopping point is chosen manually per configuration and is not varied or justified. For a homoclinic tangle, the sampled part of the manifold depends strongly on the number of periods traced; d_min is a minimum over the sampled segment, so tracing further can uncover closer approaches and reduce d_min, while d_max can grow. The effect need not be the same for the proposed giant-island configurations (which have small turnstile lobes by design, §4.4) and for the released configurations, so the claimed factor-of-two margin against the released set may be an artifact of the chosen trace length. No convergence study or error bar is reported for the central quantitative metric. If d_min/d_max are not converged, the ordering in Fig. 10 and the statement that YFK is 'approximately double the maximum of released configurations' (p.15) could be wrong, weakening the 'new class' claim even though the topological existence of large islands remains plausible.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports computational magnetic configurations in HSX and W7-X with 'giant' edge islands, obtained by reducing the current in one coil (main coil 6 in HSX, non-planar coil 5 in W7-X) near the straight-section symmetry plane. The authors characterize the island size with d_min and d_max metrics, show Poincaré plots and connection lengths, and use the EMC3-Lite heat-diffusion model to estimate PFC heat loads for selected HSX and W7-X configurations. The central claim is modest: such configurations exist computationally, that they are spatially larger than previously reported configurations, and that in the most extreme cases the island O-points leave the domain of the field-line map, leaving 'unpaired' X-points. The paper explicitly acknowledges that coil-force analysis and full divertor-performance evaluation are outside its scope.","tokens_in":16949,"tokens_out":6023,"duration_ms":53312,"significance":"If the quantitative claims are robust, the paper establishes a new, larger class of edge topologies in two existing modular-coil stellarators, which is relevant for island-divertor physics and for assessing the flexibility of existing machines. The main strengths are the transparent methodology (explicit coil-current tables, use of an established X/O-point root-finding scheme, vacuum-field analysis), the absence of circular fitting, and the clearly stated limitations. However, the central quantitative novelty — that the new islands exceed all released W7-X configurations in d_min/d_max — rests on a finite-time separatrix trace whose convergence is not demonstrated. Also, the existence of 'unpaired X-points' is asserted but not documented by explicit root-finding results. These points need to be addressed before the paper can be accepted as a rigorous quantitative study.","major_comments":[{"comment":"The d_min/d_max calculation traces X-point manifolds 'for a finite number of field periods, selected so that the inner separatrix is well-sampled but before the chaotic lobes become warped' (§2.1). No convergence study or sensitivity to the trace length is reported. Since d_min is the minimum over the sampled manifold, a longer trace could lower d_min, and d_max could increase; the effect may differ between the proposed giant-island configurations and the released ones. The claim that YFK has d_max ≈ 50 cm versus <30 cm for all released configurations (approximately double, §4.3) is load-bearing for the 'spatially larger' and 'far beyond existing configuration space' conclusions. Please provide a trace-length convergence study, report the chosen stopping points, and give error bars or a sensitivity range for the values in Fig. 10.","section":"§2.1, §4.3, Fig. 10"},{"comment":"The abstract and title emphasize 'unpaired X-points', and the text states that in the most extreme cases 'the island O-points leave the domain of the field line map' and the divertor legs of the remaining X-points 'do not close around the island'. However, the paper does not show explicit X/O-point existence data (e.g., a table or plot of the root-finding results as the coil current is scanned, or the number of X/O-points found in the extreme configurations). The Poincaré plots are suggestive, but they do not quantitatively demonstrate the disappearance of the O-point while an X-point persists. Please add a direct characterization of the X/O-point set for the extreme cases, or revise the wording to match the evidence actually presented.","section":"Abstract, §2, §3, §4"}],"minor_comments":[{"comment":"Typo: 'confugration' should be 'configuration'.","section":"Figure 6 caption"},{"comment":"Typo: 'escpaing' should be 'escaping'.","section":"Figure 7 caption"},{"comment":"The sentence 'The and connection length and Poincaré data is shown...' has a missing word or extra 'and'.","section":"Section 3"},{"comment":"Typo: 'axillary' should be 'auxiliary' (e.g., 'the axillary coil currents').","section":"Section 1 and passim"},{"comment":"Inconsistency: §2 states that I_PC_A = I_PC_B = 0.25 selects ι=5/6 and -0.23 selects ι=5/4, but Table 2 labels DMM with [0.25,0.25] as 'low iota' with edge ι=5/5, and FTM with [-0.23,-0.23] as 'high iota' with edge ι=5/5. Please correct the edge-iota values or the coil-current assignments.","section":"Table 2"},{"comment":"The text says the analysis is for four configurations SVM, NUL, WNJ, YFK, but Figure 8 caption labels panels g-i as TDK, while the text refers to those panels as WNJ. Also, TDK appears in Table 2 but is not mentioned as one of the four selected configurations. Please clarify which configuration is shown and reconcile the numbering.","section":"Figure 8, §4.2"},{"comment":"The EMC3-Lite heat-load results are for a single set of transport parameters for each device (e.g., HSX: Te=Ti=50 eV, ne=1e18 m^-3, χ=1 m^2/s; W7-X: T=100 eV, ne=1e19 m^-3, χ=3 m^2/s). No sensitivity scan is shown. Since the abstract describes configurations as 'promising for PFC heat loads', please at least add a caveat about how sensitive the heat-load fractions and peak values are to these assumptions, or perform a small parameter variation.","section":"§4.2, Figs. 7,9"},{"comment":"Typo: 'these studies provide a examples' should be 'provide examples'.","section":"§5"}],"recommendation":"major_revision","confidential_remarks":"The paper is a credible computational topology study, and the qualitative existence of large islands is likely correct. The main barrier is that the quantitative comparison with released configurations (the 'exceeds all' claim) depends on a non-converged-looking metric, and the 'unpaired X-point' phenomenon is not directly documented with root-finding results. Both are fixable with additional analysis. The coil-force and transport-model limitations are appropriately acknowledged by the authors, so I would not reject on those grounds. I recommend major revision, not because the central idea is flawed, but because the quantitative strength of the claim needs to be backed by convergence evidence and explicit X/O-point data."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my read. The genuinely new thing is the observation that, in both HSX and W7-X, reducing the current in the coil nearest the symmetry-plane straight section (main coil 6 / non-planar coil 5) blows up the edge island chain to sizes far beyond the previously documented envelope, and in the limit produces unpaired X-points whose legs do not close around an O-point. That is a concrete, previously unreported result, and it opens a plausible experimental knob for studying island divertors and non-resonant divertor physics in existing devices. The vacuum-field analysis is clean: X/O-point finding via the field line map, Poincare plots and manifold tracing, and the presentation of connection lengths is informative. I also give the authors credit for keeping the claim modest—they say over and over that coil forces, MHD, and neoclassical transport are not assessed, and that giant islands are not claimed to be favourable. That is the right tone.\n\nNow the soft spots. The stress-test worry about the island-size metric is real and I think it is the key technical issue. d_min and d_max are defined by tracing X-point manifolds for a 'finite number of field periods, selected so that the inner separatrix is well-sampled but before the chaotic lobes become warped.' That selection is manual, and there is no convergence test or sensitivity scan. The central quantitative claim—that YFK is about twice the d_max of any released W7-X configuration—could shift if the trace length is not long enough. The margin might still hold, but the paper does not demonstrate it. I would not call the existence of the giant islands into question; the Poincare plots make them visible. But the factor-of-two comparison against the database needs a robustness check before it supports a 'new class' statement.\n\nThe heat-load fractions come from EMC3-Lite with single assumed transport parameters and no uncertainty analysis. That is a minor issue given the paper's scope, but it matters if the configurations are advertised as 'promising for PFC heat loads.' And the experimental feasibility rests on coil forces that are explicitly not analyzed; the authors say so, which is honest, but it means the practical payoff is unresolved. No code or data is shipped, so independent reproduction is harder.\n\nWho is this for? People working on stellarator edge topology, island divertors, and non-resonant divertor designs. It's a computational discovery paper, not a transport or experimental validation paper. I think it deserves a serious referee: the result is novel, the vacuum-field analysis is sound, and the soft spots are addressable in revision. My recommendation is to send it to review, with a request that the authors add a convergence study for d_min/d_max and a small sensitivity scan for the EMC3-Lite inputs.","headline":"Genuinely new computational finding, but the central island-size metric needs a convergence check before the 'new class' claim is taken at face value.","tokens_in":17456,"tokens_out":3145,"would_cite":true,"duration_ms":30480,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["52.55.Hc"],"model":"deepseek-v4-flash","headline":"By reducing current in a single coil near the straight section, both HSX and W7-X can be pushed into a regime where edge islands grow far beyond previously reported sizes, sometimes so large that the island O-points leave the field line map","keywords":["giant islands","island divertor","unpaired X-points","stellarator","HSX","W7-X","coil current variation","edge magnetic topology"],"falsifier":"A concrete falsifier would be a coil-force structural analysis showing that the reduced currents required for giant islands (e.g., I_main6 = 0 or I_NPC5 = 0.1) produce forces exceeding the engineering limits of the coils or supports. Alternatively, an experiment on HSX with I_main6 = 0 that fails to show the predicted increase in d_min and d_max would directly contradict the vacuum-field prediction.","tokens_in":16476,"feed_emoji":"🧲","tokens_out":5435,"duration_ms":42672,"temperature":0.7,"pith_summary":"This paper claims that a single, simple control knob—the current in the coil nearest the symmetry plane of the straight section (main coil 6 in HSX, non-planar coil 5 in W7-X)—can drive the edge magnetic structure into a new regime of 'giant' islands, far larger than any configuration previously reported for these machines. In the most extreme cases, the island O-points leave the field line map entirely and the divertor legs of the remaining X-points do not close around the island, a topology that departs from the ordinary island divertor. The authors quantify island size with two metrics, d_min and d_max, and show that the largest examples exceed those of all released W7-X configurations. Using the EMC3-Lite heat diffusion model, they identify W7-X configurations in which most of the power lands on the divertor plates, suggesting these regimes could be promising for plasma-facing component heat loads. The paper establishes that such regimes are obtainable and can in principle be studied experimentally, though coil-force and stability analyses are left for future work.","feed_headline":"Weaken one coil, get giant islands in two stellarators","feed_subtitle":"Simply reducing one coil's current creates edge islands twice as large as any released W7-X configuration.","key_machinery":"The central object is the magnetic island chain at a low-order rational surface, whose size is quantified by two metrics: d_min, the minimum distance from the island's inner separatrix to its O-point, and d_max, the maximum distance; these bound the radial extent of the divertor legs. The controlling mechanism is the reduction of current in a single coil located at the symmetry plane of the straight section, which pushes the O-point outward and enlarges the island, eventually removing it from the field line map. X- and O-points are located by root-finding on the field line map and classified by the trace of its Jacobian, and the separatrix is traced from the X-point eigenvectors. The EMC3-Li","core_discovery":"The central discovery is that reducing the current in a single symmetry-plane coil—I_main6 in HSX and I_NPC5 in W7-X—is a sufficient (though not necessary) condition for producing giant island chains, with rotational transform 4/3 or 4/4 for HSX and 5/6, 5/5, or 5/4 for W7-X. As the current is lowered, the island O-points move outward and the islands grow monotonically, as measured by d_min and d_max. In the most extreme cases, the O-points leave the domain of the field line map, leaving 'unpaired' X-points whose divertor legs do not encircle an island. These configurations represent a new class of edge magnetic topologies for existing machines, and they demonstrate that the operational spac","pith_inferences":["A testable extension would be to reproduce the single-coil reduction in other low-shear modular-coil stellarators; if the same growth of d_min and d_max appears, the mechanism is generic rather than device-specific.","The island-size metrics d_min and d_max capture only radial extent; a fuller topological characterization—for example, connection-length distributions or turnstile flux—might reveal additional structure in the unpaired X-point regime.","The vacuum-field calculations here ignore finite-beta and plasma-response effects; a natural next step is to recompute these configurations with equilibrium codes to see whether the giant islands survive plasma pressure.","The authors' speculation that a toroidally extended low-current gap produces giant islands could be turned into a design rule for future coil optimization, where the current distribution is deliberately shaped to leave such a gap."],"forward_implications":["A single coil current becomes a practical control knob for tuning edge magnetic topology in two existing stellarators, without any change to coil geometry.","The giant-island configurations produce long, approximately straight divertor legs, which may aid in the design of tight baffling and geometric closure of the divertor.","The unpaired X-point topology represents a regime beyond the ordinary island divertor, expanding the menu of edge configurations available for studying divertor physics.","The d_min and d_max values of the proposed W7-X configurations exceed all released W7-X configurations, indicating unexplored operational space in currently approved machines.","For the selected W7-X configurations, EMC3-Lite simulations show that 77–94% of the exhaust power lands on the divertor plates, suggesting these regimes could be compatible with PFC heat-load requirements."],"fun_headline_variants":["Single coil current drop yields giant islands in HSX and W7-X","One coil tweak creates giant edge islands in two stellarators","Cut one coil's current, grow giant islands in HSX and W7-X","Reducing one coil current triggers giant islands in HSX and W7-X"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The experimental promise rests on the assumption that the non-standard coil currents—such as I_main6 = 0 in HSX or I_NPC5 = 0.1 in W7-X—are operationally achievable; the authors note that coil forces analysis is not explored and safe operation is a topic of ongoing research.","fun_headline_variants_meta":{"raw":{"variants":["Single coil current drop yields giant islands in HSX and W7-X","One coil tweak creates giant edge islands in two stellarators","Cut one coil's current, grow giant islands in HSX and W7-X","Reducing one coil current triggers giant islands in HSX and W7-X"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000211,"raw_usage":{"total_tokens":1330,"prompt_tokens":902,"completion_tokens":428,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":646,"completion_tokens_details":{"reasoning_tokens":348}},"tokens_in":646,"tokens_out":428,"duration_ms":3642,"temperature":1.0,"reasoning_tokens":348,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T01:12:53.356260+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete falsifier would be a coil-force structural analysis showing that the reduced currents required for giant islands (e.g., I_main6 = 0 or I_NPC5 = 0.1) produce forces exceeding the engineering limits of the coils or supports. Alternatively, an experiment on HSX with I_main6 = 0 that fails to show the predicted increase in d_min and d_max would directly contradict the vacuum-field prediction.","supporting_citations":[],"review_version":1}