{"id":"1d2eddab-96e2-4c27-a997-b0c89da0d948","arxiv_id":"2501.01871","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Short gas puffs in ADITYA-U lengthen sawtooth period by flattening density, suppressing trapped electron modes, and delaying the critical temperature gradient that triggers the crash.","lead":"Short gas puffs in the ADITYA-U tokamak double the period of sawtooth temperature oscillations by flattening the density profile, suppressing core turbulence, and slowing heat outflow. This offers a simple, heating-free method for sawtooth control in small fusion devices.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Most load-bearing concern: the 'critical gradient ~0.11 at rho~0.2' threshold is derived from only three time points per cycle and SXR foil-ratio Te profiles, so the puff/no-puff invariance could be an averaging artifact rather than a real crash trigger. The GTC/TEM step is secondary: Fig.","rationale":"Dolui et al. report a reproducible experimental effect: short gas puffs roughly double the sawtooth ramp time (Figs. 2-3), with density flattening outside rho ~ 0.3 and core Te peaking (Fig. 4). The strongest claim, however, is the sharper one: a critical 1/L_Te ~ 0.11 at rho ~ 0.2, independent of puffing, triggers the crash. That claim supports the interpretation of gas puffing as merely delaying approach to a fixed threshold. The weakest link in this chain is not the gyrokinetic simulation—Fig. 8's heat-pulse recovery is direct evidence of reduced heat transport, and the TEM explanation is not needed to establish the delay—but the empirical threshold itself. The manuscript gives no ECE or Thomson scattering Te measurement, the SXR inversion radius rho ~ 0.2 is at the edge of the SXR core, and the gradient is sampled at only three time points per cycle. These are exactly the conditions under which a 'universal' threshold can be produced by smoothing and averaging rather than by plasma physics. A re-analysis at full time resolution with per-cycle crash-time alignment, as described above, would settle this. If the threshold survives, the paper's central claim is materially strengthened; if it does not, the observed sawtooth delay remains but the universal trigger claim and the 'delay to threshold' explanation would need revision. The reader's CONDITIONAL verdict remains appropriate; I would attach the condition to releasing the SXR/Te data and the time-resolved threshold analysis rather than only to turbulence measurements.","tokens_in":8461,"tokens_out":7562,"duration_ms":83090,"concrete_test":"Reconstruct Te(r) for every sawtooth cycle in the puff and no-puff phases from the raw SXR tomography data at the full available time resolution, using two independent inversion algorithms and varying the spline smoothing; at each crash, take the last time sample before the crash and compute 1/L_Te at rho = 0.2. Then compare the full pre-crash distributions with a two-sample test (e.g., Welch's t or Mann-Whitney). If the puff and no-puff distributions are not consistent with a common mean (p < 0.05) or the scatter exceeds the error bars in Fig. 7(b), the 'irrespective of gas puff' threshold claim is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central assertion (Sec. III E and Conclusion) is that a universal critical temperature gradient, 1/L_Te ~ 0.11 at rho ~ 0.2, triggers every sawtooth crash, with or without gas puff. The supporting evidence, Fig. 7(b), is thin in two ways. First, the text states each sawtooth period is sampled at only 'three different time intervals from the beginning to the end of the swt period', so the pre-crash value of 1/L_Te is not measured at the actual crash time. The claim that the gradient 'attains its maximum value just before the crash' is an interpolation, not a direct observation, and averaging over tens of cycles can hide cycle-to-cycle differences between puff and no-puff cases. Second, Te(r) is reconstructed from SXR foil-ratio/tomography with no ECE or Thomson scattering; at rho ~ 0.2, near the edge of the SXR-emitting core, the local gradient is highly sensitive to the tomographic inversion and to spline smoothing (Fig. 4 shows spline fits). If the apparent 0.11 threshold shifts or separates once higher-time-resolution, per-cycle profiles are analyzed, the strongest claim—a universal trigger condition—fails. I do not think the GTC/TEM link is the most load-bearing weakness: the reduced heat transport after the puff is already directly evidenced by the roughly two-times slower heat-pulse recovery in Fig. 8, so the TEM microphysics is an add-on rather than the keystone of the threshold claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports experiments on the ADITYA-U tokamak in which short gas puffs extend the sawtooth period by roughly a factor of two. The authors propose a mechanism: the gas puff flattens the density profile outside the core, suppressing trapped electron modes (TEM), which reduces core heat transport and delays the recovery of a critical electron temperature gradient at the inversion radius (rho ~ 0.2). The central new claim, stated in Sec. III E and the Conclusion, is that sawtooth crashes occur whenever the inverse temperature scale length reaches 1/L_Te ~ 0.11 at rho ~ 0.2, regardless of whether a gas puff was applied. Supporting evidence includes the measured period enhancement, the modified density and temperature profiles, GTC gyrokinetic simulations of TEM suppression, and a wavelet analysis of SXR fluctuations.","tokens_in":8977,"tokens_out":2698,"duration_ms":27108,"significance":"If the universal critical-gradient claim is correct, the paper identifies a simple, auxiliary-heating-free control knob for sawtooth stabilization in small and medium tokamaks, and it provides a concrete trigger condition that could be tested on other devices. The experimental observation of sawtooth-period enhancement is direct, reproducible over many discharges, and clearly documented. The paper also attempts to connect edge density perturbations to core turbulence and transport through gyrokinetic simulations, which is a valuable addition. However, the strongest claim—a universal 1/L_Te ~ 0.11 crash threshold—rests on limited profile data and an associated interpolation, so its significance is currently conditional on stronger evidence.","major_comments":[{"comment":"The universal critical-gradient claim is supported only by Fig. 7(b), where each sawtooth period is sampled at 'three different time intervals from the beginning to the end of the swt period' and each point is an average over tens of cycles. The pre-crash value of 1/L_Te is therefore an interpolation, not a measurement at the crash time. This is load-bearing because the central conclusion asserts that crashes occur at 1/L_Te ~ 0.11 in both puff and no-puff cases; cycle-to-cycle scatter within the averaged tens of cycles could hide a systematic offset between the two cases. The authors should show per-cycle or at least higher-time-resolution Te profiles at multiple times just before each crash, and demonstrate that the inferred threshold is stable without averaging.","section":"Sec. III E (Fig. 7)"},{"comment":"The Te profiles in Fig. 7(a) are reconstructed from SXR foil-ratio measurements and shown as spline fits, with no independent ECE or Thomson scattering validation. At rho ~ 0.2, near the edge of the SXR-emitting core, the local gradient 1/L_Te is highly sensitive to the tomographic inversion and to spline smoothing. The apparent convergence of puff and no-puff cases onto 1/L_Te ~ 0.11 could be an artifact of the reconstruction and smoothing procedure rather than a physical trigger. The authors should quantify the sensitivity of the inferred threshold to the inversion and spline parameters, or present an independent temperature diagnostic.","section":"Sec. III E (Fig. 7) and Sec. II (SXR diagnostics)"},{"comment":"The wavelet analysis subtracts the contribution of a 'fluctuation-free swt cycle of similar amplitude and timescale generated synthetically' before concluding that broadband turbulence develops before the crash. This subtraction is an ad hoc assumption; if the synthetic cycle does not faithfully represent the non-turbulent component of the SXR signal, the apparent pre-crash broadband feature may be an artifact. The authors should validate the subtraction procedure—for example, by testing it on synthetic signals with known turbulence content or by showing that the result is insensitive to details of the synthetic cycle—before using this feature as evidence for a turbulence-driven trigger.","section":"Sec. III E (Fig. 9(b))"}],"minor_comments":[{"comment":"The text contains typos such as 'ADITY A-U' with an inserted space and a duplicated phrase in Ref. [5] ('maintaining good confinement and maintain- ing good confinement').","section":"Throughout"},{"comment":"The GTC simulations are described as identifying TEM suppression, but no direct fluctuation measurements are shown to confirm that the simulated turbulence change occurs in the experiment. Since the paper already has the direct heat-pulse recovery evidence in Fig. 8, this limitation should be stated explicitly so readers do not mistake the simulation for a measurement.","section":"Sec. III C"},{"comment":"The statement that 1/L_Te 'attains its maximum value just before the crash' is an interpolation from only three time intervals; the wording should be softened to reflect that the maximum is inferred, not directly observed at the crash instant.","section":"Sec. III E"},{"comment":"The discussion of the Kadomtsev model and the m = 1 mode is interesting but somewhat disconnected from the critical-gradient claim; a brief explanation of how the precursor mode interacts with the temperature gradient threshold would improve clarity.","section":"Sec. III D"}],"recommendation":"major_revision","confidential_remarks":"The paper appears already published as Phys. Rev. Research 7, 033161 (2025), so the review is retrospective. The main concern is the thinness of the evidence for the central universal-threshold claim; the period-doubling observation itself is solid. If the authors can supply per-cycle gradient analysis and sensitivity checks for the SXR-based Te profiles, the claim would be much more convincing."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The core result here is real: on ADITYA-U, short fuel gas puffs reproducibly stretch the sawtooth ramp from ~600 microseconds to 1–1.5 ms, across many discharges, and the effect scales with puff size. That alone is a useful, simple actuator for sawtooth control in small/medium tokamaks that lack auxiliary heating. The paper also does something right that's easy to miss: Fig. 8 directly shows the SXR heat pulse returning to pre-crash levels in ~500 microseconds without puff and ~1000 with puff, so the reduced heat transport after the puff is measured, not just simulated.\n\nThe soft spot is the critical-gradient claim. The 1/L_Te ~ 0.11 threshold at rho ~ 0.2 is built from only three time points per sawtooth cycle, averaged over tens of cycles, and the Te profiles come from SXR foil-ratio tomography rather than ECE or Thomson scattering. At rho ~ 0.2, near the edge of the SXR-emitting core, the local gradient is quite sensitive to the tomographic inversion and the spline fit. So the invariance of the threshold with and without puff might be an averaging artifact. That's the load-bearing concern. The GTC/TEM step is secondary—the heat-pulse data already support reduced transport, so even if the TEM identification is imperfect, the overall mechanism doesn't collapse.\n\nI'd like to see per-cycle gradient evolution at higher time resolution, ideally with a second Te diagnostic, and some uncertainty quantification on the threshold. The data being unavailable on request only makes that check harder. None of this kills the paper; the period enhancement is direct and well-documented. But the universal trigger condition should be framed as a hypothesis, not a conclusion, until more data back it up.\n\nWho's this for? Tokamak experimentalists working on sawteeth and anyone looking for cheap actuators on small devices. It's a solid PRR-style experimental paper with a plausible mechanism and one overreaching claim. I'd send it to a referee, and if I were the referee, I'd push for a revision that either supplies the per-cycle gradient data or downgrades the threshold to a working hypothesis.","headline":"Solid new result on gas-puff sawtooth stabilization, but the universal critical-gradient trigger needs stronger diagnostic support.","tokens_in":9461,"tokens_out":2111,"would_cite":true,"duration_ms":20664,"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":"Short gas puffs delay tokamak sawtooth crashes by flattening the density profile and slowing the temperature gradient's climb to a fixed threshold.","keywords":["sawtooth instability","tokamak","gas puff injection","trapped electron mode","cold pulse propagation","critical temperature gradient","sawtooth control","core turbulence"],"falsifier":"A turbulence diagnostic with electron-scale resolution viewing $\\rho\\approx0.2$ in the same discharges that shows no drop in fluctuation amplitude after the gas puff while the sawtooth period still doubles would falsify the TEM-suppression mechanism. Collecting many crashes and finding any where the crash occurs with $1/L_{T_e}$ clearly different from about $0.11$ at the inversion radius would falsify the claimed threshold.","tokens_in":8310,"feed_emoji":"⚡","tokens_out":10878,"duration_ms":100401,"temperature":0.7,"pith_summary":"This paper reports that a short puff of fuel gas at the edge of a small ohmically heated tokamak roughly doubles the time between sawtooth crashes, and it traces the physical route: the puff flattens the radial density profile, which suppresses trapped-electron-mode turbulence in the core, reducing heat transport and letting the central electron temperature rise to a more peaked profile. The crash fires only when the temperature gradient near the inversion radius ($\\rho\\sim0.2$) crosses a threshold value ($1/L_{T_e}\\sim0.11$), so the reduced transport delays the moment the gradient reaches that threshold. The paper proposes this as a simpler alternative to electron cyclotron heating for controlling sawtooth behavior in tokamaks without auxiliary heating. This matters because sawtooth crashes redistribute core heat and can seed other instabilities, and a gas valve is an easy actuator to study them with.","feed_headline":"Gas puffs double the wait between tokamak sawtooth crashes","feed_subtitle":"Flattening the density profile quiets core turbulence, so the temperature gradient takes longer to hit the crash trigger.","key_machinery":"The load-bearing object is the inverse electron temperature gradient at the sawtooth inversion radius, $1/L_{T_e}=-(1/T_e)(dT_e/dr)$ evaluated near $\\rho\\sim0.2$; the paper treats it as the quantity that crosses the same threshold at every crash, with or without a puff. The second mechanism is the gas-puff cold pulse: edge gas raises density in the outer mid-radius and flattens the density profile, stabilizing the trapped electron mode (TEM), a microinstability driven by particles trapped on the low-field side of the magnetic well. Gyrokinetic simulations using the reconstructed equilibrium profiles show the TEM fluctuation pattern contracting and weakening in the core after the puff, which lowers the heat diffusivity. With heat transport reduced, Ohmic heating takes longer to rebuild the steep core gradient, and wavelet analysis of soft-x-ray emission ties the sharp rise of broadband core temperature turbulence at that gradient to the crash itself.","core_discovery":"The central claim is that sawtooth crashes in these discharges are not triggered by a fixed time or by the precursor oscillation alone, but by the local electron temperature gradient at the inversion radius reaching a critical value, about $1/L_{T_e}\\sim0.11$ at $\\rho\\sim0.2$, whether or not a gas puff was applied. The gas puff acts upstream: it raises density outside the core, flattening the density profile, which suppresses trapped electron modes and lowers electron heat transport. The core then heats more slowly after each crash, so the ramp phase lasts roughly twice as long before the gradient threshold is reached. The authors support the turbulence link with gyrokinetic simulations showing suppressed core turbulence after the puff and with wavelet analysis showing broadband core temperature fluctuations rising sharply just before the crash. They deliberately compare their observations to sawtooth modification by electron cyclotron heating and conclude that edge gas puffing is a workable, simpler actuator for sawtooth control.","pith_inferences":["Editorial inference: If the critical-gradient rule holds beyond this machine, the density-profile effect could be exploited in reactor-relevant devices that cannot rely on central heating, though puff size and timing would need rescaling.","Editorial inference: The mechanism predicts a directly observable signature that the paper does not show, namely that electron-scale fluctuation amplitude in the core should fall within about a millisecond of the puff and recover over the next few sawtooth cycles, mirroring the ramp-time recovery.","Editorial inference: Modulated gas puffs might serve not just to delay crashes but to deliberately pace them, which would be useful for studying sawtooth effects on impurity transport and neoclassical tearing modes."],"forward_implications":["Sawtooth pacing and stabilization become available to small and medium tokamaks that have no auxiliary heating; only a gas valve and a density-profile response are needed.","The sawtooth period can be tuned by gas-puff size, since the ramp-time increase grows with the amount of injected fuel until the discharge degrades.","The threshold picture predicts that any actuator that slows core heat transport, such as gas puffs or electron cyclotron heating, should delay sawtooth crashes through the same channel.","The sharp rise of broadband core temperature fluctuations just before the crash supports a turbulence-triggered crash mechanism rather than a purely resistive MHD process."],"supporting_citations":[{"why":"Establishes the gas-puff cold-pulse propagation in this tokamak and supplies the injection and diagnostic specifics used to interpret the density and temperature changes.","marker":"[14]"},{"why":"Provides the transport model that predicts density-profile flattening stabilizes trapped electron modes and reduces electron heat transport.","marker":"[15]"},{"why":"Offers direct experimental evidence from edge cooling that core temperature fluctuations are suppressed, supporting the TEM link.","marker":"[24]"},{"why":"Documents core electron temperature turbulence during sawtooth oscillations in another tokamak, motivating the turbulence-triggered crash picture.","marker":"[13]"},{"why":"Supplies the gyrokinetic simulation framework and previous identification of trapped electron modes in these discharges, used to show post-puff TEM suppression.","marker":"[25]"},{"why":"Show an earlier observation of sawtooth modification by electron cyclotron heating, the behavior the gas-puff results are compared with.","marker":"[17]"}],"fun_headline_variants":["Gas puff doubles sawtooth period without auxiliary heating","Short gas pulse lengthens tokamak sawtooth crashes by suppressing turbulence","Sawtooth slowdown via gas puff: a simpler control knob","Gas puffs quiet core turbulence, stretching sawtooth ramp time"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The causal chain assumes the gyrokinetic simulations faithfully capture the suppression of trapped-electron turbulence after the gas puff, because no direct turbulence measurement is made; if the simulated turbulence change does not match the real plasma, the link from density flattening to slower heat transport is unverified, even though the measured period enhancement stands on its own.","fun_headline_variants_meta":{"raw":{"variants":["Gas puff doubles sawtooth period without auxiliary heating","Short gas pulse lengthens tokamak sawtooth crashes by suppressing turbulence","Sawtooth slowdown via gas puff: a simpler control knob","Gas puffs quiet core turbulence, stretching sawtooth ramp time"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000413,"raw_usage":{"total_tokens":2105,"prompt_tokens":887,"completion_tokens":1218,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":503,"completion_tokens_details":{"reasoning_tokens":1144}},"tokens_in":503,"tokens_out":1218,"duration_ms":9224,"temperature":1.0,"reasoning_tokens":1144,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:17:36.641011+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A turbulence diagnostic with electron-scale resolution viewing $\\rho\\approx0.2$ in the same discharges that shows no drop in fluctuation amplitude after the gas puff while the sawtooth period still doubles would falsify the TEM-suppression mechanism. Collecting many crashes and finding any where the crash occurs with $1/L_{T_e}$ clearly different from about $0.11$ at the inversion radius would falsify the claimed threshold.","supporting_citations":[{"cited_title":"Macwan et al","cited_arxiv_id":null,"evidence_quote":"Establishes the gas-puff cold-pulse propagation in this tokamak and supplies the injection and diagnostic specifics used to interpret the density and temperature changes."},{"cited_title":"Rodriguez-Fernandez, C","cited_arxiv_id":null,"evidence_quote":"Provides the transport model that predicts density-profile flattening stabilizes trapped electron modes and reduces electron heat transport."},{"cited_title":"Wang et al., Core electron temperature turbulence and trans- port during sawtooth oscillations in the DIII-D tokamak, Nucl","cited_arxiv_id":null,"evidence_quote":"Documents core electron temperature turbulence during sawtooth oscillations in another tokamak, motivating the turbulence-triggered crash picture."},{"cited_title":"Singh, D","cited_arxiv_id":null,"evidence_quote":"Supplies the gyrokinetic simulation framework and previous identification of trapped electron modes in these discharges, used to show post-puff TEM suppression."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Show an earlier observation of sawtooth modification by electron cyclotron heating, the behavior the gas-puff results are compared with."}],"review_version":1}