{"id":"e4d1ad0c-f9e5-46d7-9369-f4425b10c9e4","arxiv_id":"2502.01322","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A fiber-laser source combining gain-managed nonlinear amplification and resonant dispersive-wave emission generates tunable sub-20 fs visible pulses with up to 39 nJ energy and 2.2 MW peak power.","lead":"This paper describes a compact fiber laser system that produces very short visible light pulses, under 20 femtoseconds, whose color can be tuned from violet to red. The source reaches over 2 million watts of peak power at a high repetition rate, which could make advanced microscopy and spectroscopy cheaper and more practical outside specialized laser labs.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 13 fs / 2.2 MW headline rests on an SHG-FROG trace with only 40 delay steps over ±500 fs (≈25 fs step), too coarse to resolve a 13 fs pulse; the retrieval is likely underdetermined.","rationale":"The reader's weakest assumption was that the SHG-FROG retrieval accurately reconstructs the temporal profile. I agree that this is the most load-bearing point, but I identify a more specific and testable deficiency: the 512×40 grid with a ±500 fs delay range gives a ~25 fs delay step, which is marginal or insufficient for a 13 fs pulse. A low FROG error alone cannot rescue an under-sampled trace, and no independent verification is reported. This strengthens the existing CONDITIONAL verdict rather than changing it. I also note secondary overreach in the abstract's tuning claim: sub-20 fs is demonstrated at only three wavelengths (520, 576, 660 nm), and beyond 700 nm the RDW is described as merging into a supercontinuum rather than cleanly separating; however, the FROG sampling issue is the single most load-bearing concern because it directly underpins the headline 13 fs and 2.2 MW numbers.","tokens_in":7453,"tokens_out":6822,"duration_ms":69767,"concrete_test":"Re-measure the compressed RDW at 40 bar with an SHG-FROG using at least 256 delay steps over a ±200 fs window (delay step <2 fs), or with an independent technique such as SPIDER or a two-photon-absorption autocorrelation. Compare the retrieved FWHM and main-pulse energy fraction with Fig. 4(d); if the retrieved duration is not within a few fs of 13 fs, or if the trace cannot be fit with the reported 0.8% error, the headline duration and peak power are not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claims—13 fs duration and 2.2 MW peak power—depend entirely on the SHG-FROG retrieval shown in Fig. 4. The paper states that the retrieval used a 512×40 grid, and the displayed delay axis spans −500 to +500 fs, implying a delay step of roughly 25 fs. A 13 fs pulse has an SHG-FROG intensity autocorrelation width of order 18–25 fs, so the central feature is sampled by at most one or two delay points. FROG retrieval from such a coarsely sampled trace is underdetermined: many different pulse shapes can fit the sampled trace, and the reported 0.8% retrieval error does not establish uniqueness or accuracy. No independent measurement (e.g., autocorrelation, SPIDER, or a finer-delay FROG) is provided. A second, related issue is that the 10 µm type-I BBO crystal's phase-matching bandwidth and the wavelength calibration are not analyzed, so systematic errors in the retrieved duration cannot be excluded. Because the 13 fs value is multiplied by the 39 nJ energy to obtain 2.2 MW, any error in the retrieved pulse duration directly changes the headline peak power. Thus the load-bearing assumption—that the FROG retrieval faithfully reconstructs the true temporal profile—is not adequately supported by the presented data.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a compact fiber-laser source that combines gain-managed nonlinear amplification (GMNA) with resonant dispersive-wave (RDW) emission in a gas-filled antiresonant hollow-core fiber. The authors claim sub-20 fs pulses tunable from approximately 400 nm to beyond 700 nm, with energies up to 39 nJ and peak powers up to 2.2 MW at 4.8 MHz repetition rate. The system comprises a NALM oscillator, a pre-amplifier, a GMNA stage, and an RDW stage, with argon pressure used to tune the RDW wavelength. Temporal characterization is performed with all-reflection SHG-FROG, and spectra are calibrated with NIST-traceable lamps. The paper also reports stability measurements over tens of hours and a relative intensity noise of 0.8%.","tokens_in":7688,"tokens_out":9285,"duration_ms":82104,"significance":"If the quantitative claims are fully supported, this is a valuable demonstration: a relatively simple and compact fiber-based source reaching few-femtosecond-scale, tunable visible pulses with megawatt peak power would offer a credible alternative to more complex Ti:sapphire and optical parametric amplifier systems for applications such as multiphoton imaging and ultrafast spectroscopy. The manuscript has notable strengths: the RDW tuning data are presented over a wide range, the spectral calibration is described, the FROG retrieval error is reported as 0.8%, and stability data are included. However, the central pulse-duration and peak-power claims currently rest on a FROG measurement whose delay sampling appears too coarse to resolve a 13 fs pulse, and no uncertainty analysis is provided for the headline numbers. These issues must be addressed before the claims can be considered fully established.","major_comments":[{"comment":"The headline 13 fs duration and 2.2 MW peak power rest entirely on the SHG-FROG retrieval in Fig. 4, but the trace is sampled on a 512×40 grid over a −500 to +500 fs delay window, i.e. approximately 25 fs steps. For a 13 fs pulse the SHG-FROG temporal feature has an intensity-autocorrelation width of only about 18–25 fs, so the central feature is sampled by at most one or two delay points. A 0.8% retrieval error on such a coarse grid does not establish uniqueness; many different pulse shapes can fit the same sparsely sampled trace. The authors should provide a finer-delay FROG measurement or an independent temporal characterization (e.g., autocorrelation or SPIDER), demonstrate retrieval stability with respect to grid spacing, and report the resulting uncertainty on the retrieved duration and peak power, including the effect of the 10 µm type-I BBO phase-matching bandwidth and the background-subtraction procedure.","section":"Fig. 4 and accompanying SHG-FROG text"},{"comment":"The abstract claims 'sub-20 femtosecond pulses tunable from 400 nm to beyond 700 nm,' but temporal characterization is reported only at 520, 576, and 660 nm (Fig. 3b–d) and for the compressed pulse in Fig. 4. No pulse duration is measured across the rest of the tuning range, and the text itself states that beyond 700 nm the RDW 'does not cleanly separate from the pump pulse, instead it forms part of a supercontinuum.' The sub-20 fs tunability claim should be restricted to the wavelengths actually characterized, or additional temporal measurements across the full range should be provided.","section":"Abstract and Fig. 3"},{"comment":"The manuscript reports 39 nJ and 2.2 MW without uncertainties. The RDW energy is extracted from calibrated spectra and 'validated using a power meter,' but no calibration uncertainty, reproducibility, or systematic-error analysis is given. The peak power depends on both the FROG-retrieved 13 fs duration and the assumed 81% energy fraction in the main peak, so these sources of uncertainty should be propagated into the headline values or the dominant systematic errors should be stated explicitly.","section":"Energy and peak-power reporting"}],"minor_comments":[{"comment":"The text says an '850 nm long-pass filter' is used to separate the visible RDW from the pump; a long-pass filter at 850 nm would reject the 400–700 nm RDW and transmit the ~1030 nm pump. If a short-pass filter was intended, please correct this.","section":"RDW compression description"},{"comment":"The central wavelength and gas pressure of the compressed pulse characterized in Fig. 4 are not stated in the text; please give these values to allow direct comparison with the 520, 576, and 660 nm measurements in Fig. 3.","section":"Fig. 4 caption and text"},{"comment":"The statements 'more than 75% of the pulse energy concentrated in the main pulse' (GMNA) and 'more than 81% of the energy located within the main peak' (RDW) need a precise definition of how the main pulse is separated from the pedestal or background, since this directly affects the peak-power estimate.","section":"Main-pulse energy fraction definitions"},{"comment":"Please clarify whether the spectra in Fig. 3(a) are individually normalized or shown on a common intensity scale; the separate energy axis makes the current plotting convention ambiguous.","section":"Fig. 3(a) normalization"},{"comment":"Please clarify the FROG grid convention: if the delay axis has only 40 points over ±500 fs, how is the 512-point axis distributed, and what delay step was actually used? This is important for assessing the temporal resolution of the retrieval.","section":"FROG grid convention"}],"recommendation":"major_revision","confidential_remarks":"The main technical risk is the SHG-FROG sampling: the 512×40 grid with a ~25 fs delay step appears too coarse to resolve a 13 fs pulse, and the reported 0.8% retrieval error does not by itself address uniqueness. If the authors can supply a finer-delay FROG trace or an independent temporal measurement (with uncertainties), the central claims would become credible. The paper's architecture and spectral tuning data are otherwise compelling, and I see no evidence of circularity or invented entities; the concerns are purely about experimental support for the headline numbers."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real demonstration that a gain-managed nonlinear amplifier can drive resonant dispersive-wave emission in a gas-filled hollow-core fiber and produce tunable visible pulses with tens of nanojoules and few-tens-of-femtoseconds duration. The setup is compact and the tuning scan from 400-700 nm is a useful data point for people building practical ultrafast sources. It deserves a serious referee. But the 13 fs/2.2 MW headline is more fragile than the text suggests, and the core result overlaps with the authors' own Laser Congress 2024 paper.\n\nWhat the paper does well: the energy-vs-pressure data are credible, with calibrated spectra cross-checked against a power meter; the 0.8% RIN stability is a nice practical result; and chirped-mirror compression after spectral filtering is a sensible way to get the RDW pulse short at the output. The 13% conversion efficiency at 540 nm is respectable.\n\nThe soft spots are mostly about the FROG characterization. The stress-test concern is valid. The trace in Fig. 4 was retrieved on a 512x40 grid, and the delay axis spans -500 to +500 fs. That gives roughly 25 fs steps. A 13 fs pulse has an SHG-FROG delay width of order 18-25 fs, so the central feature is sampled by at most two points. The 0.8% retrieval error only means the algorithm found a self-consistent solution; it does not establish uniqueness. No independent pulse-width measurement is given. Since the 2.2 MW peak power is computed from the retrieved 13 fs duration, an error in the FROG reconstruction directly changes the headline claim. The authors also report no uncertainties on the 13 fs, 39 nJ, or 2.2 MW values.\n\nThe abstract's claim of 'sub-20 fs tunable from 400 nm to beyond 700 nm' is stronger than what was measured. Only 520, 576, and 660 nm were characterized with FROG; the rest of the tuning range is inferred from spectra, and beyond 700 nm the RDW merges into the pump continuum, so 'tunable' there is doing extra work. None of this invalidates the demonstration, but it should be tightened.\n\nOn novelty: ref 38 is the same group's conference paper reporting megawatt-scale sub-15 fs visible pulses from the same method. The present letter adds more wavelength coverage, FROG traces at three wavelengths, and stability data. That is a meaningful extension, but it is not a first demonstration.\n\nWho this is for: groups in ultrafast spectroscopy or multiphoton imaging who want a fiber-based alternative to Ti:sapphire/OPA. A referee should ask for a finer FROG delay grid or an independent autocorrelation/SPIDER measurement, and explicit error bars. With those additions, the paper would be a good journal article. I would send it to peer review.","headline":"A credible but incremental fiber-based route to tunable visible femtosecond pulses; the 13 fs headline rests on a FROG grid too coarse to trust without more evidence.","tokens_in":8293,"tokens_out":3777,"would_cite":true,"duration_ms":33959,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["42.65.Wi","42.65.Re"],"model":"deepseek-v4-flash","headline":"A fiber oscillator plus gas-filled hollow-core fiber produces tunable sub-20 fs visible pulses with up to 39 nJ energy and 2.2 MW peak power.","keywords":["ultrafast laser pulses","resonant dispersive-wave emission","gain-managed nonlinear amplification","hollow-core antiresonant fiber","fiber laser","visible pulse generation","pulse compression","frequency-resolved optical gating"],"falsifier":"Run an independent autocorrelation or direct spectral-phase measurement on the compressed RDW pulse at 40 bar and check whether the retrieved 13 fs FWHM and 2.2 MW peak power are reproduced; also measure the BBO crystal's phase-matching acceptance across the full RDW spectrum to ensure sidebands are not being filtered.","tokens_in":7224,"feed_emoji":"⚡","tokens_out":5522,"duration_ms":45269,"temperature":0.7,"pith_summary":"This paper sets out to show that a fiber-laser-based system can replace the bulky, expensive Ti:sapphire and optical-parametric-amplifier chains normally used for tunable ultrafast visible pulses. The authors combine a gain-managed nonlinear amplifier (GMNA) with resonant dispersive-wave (RDW) emission in a gas-filled hollow-core antiresonant fiber, and report sub-20 fs pulses tunable from about 400 nm to beyond 700 nm, with energies up to 39 nJ and peak powers exceeding 2 MW at 4.8 MHz. If correct, this turns a technique previously tied to amplified solid-state pump lasers into a compact, modest-cost source suitable for multiphoton imaging and ultrafast spectroscopy. The central result is that the shortest measured pulse is 13 fs, with more than 81% of its energy in the main peak.","feed_headline":"Fiber laser produces tunable 13 fs visible pulses at 2 MW","feed_subtitle":"A compact fiber system delivers megawatt, sub-20 fs pulses across 400–700 nm without complex pump lasers.","key_machinery":"Two mechanisms carry the argument. Gain-managed nonlinear amplification is an over-extended self-similar amplifier: a narrowband low-energy seed is amplified in a Yb-doped fiber whose longitudinally varying, gain-saturated nonlinear dynamics broaden the spectrum into a near-linear chirp, so a simple grating pair compresses it to 38 fs at ~10 MW. Resonant dispersive-wave emission is the phase-matched transfer of energy from a soliton to a linear (dispersive) wave in the presence of high-order dispersion; tuning the argon pressure shifts the phase-matching wavelength, and the antiresonant hollow-core fiber's transmission band is chosen so no high-loss resonance sits between pump and RDW, raising conversion efficiency to 13%.","core_discovery":"The paper demonstrates that resonant dispersive-wave emission can be pumped directly by a fiber oscillator plus GMNA rather than by an amplified Ti:sapphire or Yb system. A 38 fs, ~10 MW compressed GMNA pulse at 520 nJ drives a 35 cm antiresonant hollow-core fiber filled with argon; varying the gas pressure from 20 to 75 bar tunes the RDW from roughly 400 nm to beyond 700 nm. At 40 bar, spectral filtering and chirped-mirror compression yield a retrieved 13 fs FWHM pulse with 39 nJ energy and an estimated 2.2 MW peak power, and pulses at 520, 576, and 660 nm are all measured under 20 fs. The RDW carries up to 13% of the coupled pump energy, exits in a fundamental-mode beam, and the system runs for tens of hours with 0.8% relative intensity noise. The authors argue this gives comparable or better pulse duration, tunability, and peak power than Ti:sapphire or OPA sources at a fraction of the complexity.","pith_inferences":["A natural next test is to pump the same architecture with a Mamyshev oscillator or standard step-index gain fiber, trading some peak power for a substantially lower-cost, fully fiber-integrated system.","Frequency-doubling the GMNA pump to the green, as the authors note, could drive deep-ultraviolet RDW emission below 400 nm with the same compact footprint, extending the source's range.","If the FROG retrieval is confirmed by an independent autocorrelation, the architecture becomes a credible candidate for a commercial turnkey source; if not, the headline 13 fs and 2.2 MW numbers would need revision.","Filling the fiber with a Raman-active gas could add soliton self-frequency shifting, letting one fiber-based device span from the deep UV to the infrared, but that extension is speculative without demonstration."],"forward_implications":["A single fiber-based source can deliver sub-20 fs pulses at megawatt peak power across the visible, a regime previously requiring Ti:sapphire amplifiers or optical parametric amplifiers.","Tuning is achieved by changing gas pressure in the hollow-core fiber, which is simpler and faster than replacing nonlinear crystals or adjusting OPA stages.","At the demonstrated 4.8 MHz repetition rate and ~39 nJ energies, the source is directly relevant to multiphoton microscopy and time-resolved visible spectroscopy.","Because the GMNA output is 38 fs and ~10 MW, it can drive RDW generation without a separate temporal compression stage.","The documented tens-of-hours stability and 0.8% relative intensity noise make the system plausible for extended hands-off operation in application labs."],"supporting_citations":[{"why":"Establishes the GMNA regime: amplification with longitudinal asymmetric gain and self-phase modulation yields high-energy pulses compressible to sub-40 fs without pre-chirping.","marker":"[23]"},{"why":"Demonstrates resonant dispersive-wave generation in gas-filled hollow-core fiber, the phase-matching process the paper uses for visible light.","marker":"[9]"},{"why":"Shows pressure tuning of RDW emission, the mechanism used here to tune from ~400 nm to beyond 700 nm.","marker":"[10]"},{"why":"Provides the broader RDW source framework and efficiency behavior that the paper extends with a fiber-laser pump.","marker":"[14]"},{"why":"Reports few-femtosecond RDW pulses in antiresonant fibers, giving the comparison for why the 13 fs pulse is bandwidth-limited at a smaller frequency shift.","marker":"[16]"},{"why":"Shows deep-ultraviolet RDW generation at low pump energies, cited as the route to extending this system below 400 nm via second-harmonic pumping.","marker":"[19]"},{"why":"Presents the Mamyshev oscillator, cited as a possible simpler replacement for the oscillator and GMNA stages in future compact versions.","marker":"[20]"}],"fun_headline_variants":["Fiber laser emits 13 fs pulses at 2 MW peak","Compact fiber laser: 13 fs, 2 MW, tunable visible","No complex pump: fiber laser hits 13 fs at 2 MW","13 fs, 2 MW, 400–700 nm from a fiber laser","Megawatt fiber pulses: 13 fs across 400–700 nm"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The quoted 13 fs duration and 2.2 MW peak power rest entirely on the SHG-FROG retrieval, which reports only a 0.8% retrieval error and no independent autocorrelation or uncertainty analysis, so a bias in the BBO crystal phase-matching bandwidth, spectral calibration, or background subtraction would change the headline numbers.","fun_headline_variants_meta":{"raw":{"variants":["Fiber laser emits 13 fs pulses at 2 MW peak","Compact fiber laser: 13 fs, 2 MW, tunable visible","No complex pump: fiber laser hits 13 fs at 2 MW","13 fs, 2 MW, 400–700 nm from a fiber laser","Megawatt fiber pulses: 13 fs across 400–700 nm"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000869,"raw_usage":{"total_tokens":3746,"prompt_tokens":909,"completion_tokens":2837,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":525,"completion_tokens_details":{"reasoning_tokens":2739}},"tokens_in":525,"tokens_out":2837,"duration_ms":17990,"temperature":1.0,"reasoning_tokens":2739,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T15:40:29.357888+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run an independent autocorrelation or direct spectral-phase measurement on the compressed RDW pulse at 40 bar and check whether the retrieved 13 fs FWHM and 2.2 MW peak power are reproduced; also measure the BBO crystal's phase-matching acceptance across the full RDW spectrum to ensure sidebands are not being filtered.","supporting_citations":[{"cited_title":"Sidorenko, W","cited_arxiv_id":null,"evidence_quote":"Establishes the GMNA regime: amplification with longitudinal asymmetric gain and self-phase modulation yields high-energy pulses compressible to sub-40 fs without pre-chirping."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates resonant dispersive-wave generation in gas-filled hollow-core fiber, the phase-matching process the paper uses for visible light."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows pressure tuning of RDW emission, the mechanism used here to tune from ~400 nm to beyond 700 nm."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the broader RDW source framework and efficiency behavior that the paper extends with a fiber-laser pump."},{"cited_title":"Brahms, D","cited_arxiv_id":null,"evidence_quote":"Reports few-femtosecond RDW pulses in antiresonant fibers, giving the comparison for why the 13 fs pulse is bandwidth-limited at a smaller frequency shift."},{"cited_title":"Sabbah, K","cited_arxiv_id":null,"evidence_quote":"Shows deep-ultraviolet RDW generation at low pump energies, cited as the route to extending this system below 400 nm via second-harmonic pumping."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Presents the Mamyshev oscillator, cited as a possible simpler replacement for the oscillator and GMNA stages in future compact versions."}],"review_version":1}